Antiviral medications for preventing cytomegalovirus disease in solid organ transplant recipients

  • Conclusions changed
  • Review
  • Intervention

Authors

  • Elisabeth M Hodson,

    Corresponding author
    1. The Children's Hospital at Westmead, Centre for Kidney Research, Westmead, NSW, Australia
    2. The University of Sydney, Sydney School of Public Health, Sydney, NSW, Australia
    Search for more papers by this author
  • Maleeka Ladhani,

    1. The Children's Hospital at Westmead, Centre for Kidney Research, Westmead, NSW, Australia
    Search for more papers by this author
  • Angela C Webster,

    1. The University of Sydney, Sydney School of Public Health, Sydney, NSW, Australia
    2. The University of Sydney at Westmead, Centre for Transplant and Renal Research, Westmead Millennium Institute, Westmead, NSW, Australia
    3. The Children's Hospital at Westmead, Cochrane Renal Group, Centre for Kidney Research, Westmead, NSW, Australia
    Search for more papers by this author
  • Giovanni FM Strippoli,

    1. The University of Sydney, Sydney School of Public Health, Sydney, NSW, Australia
    2. The Children's Hospital at Westmead, Cochrane Renal Group, Centre for Kidney Research, Westmead, NSW, Australia
    3. University of Bari, Department of Emergency and Organ Transplantation, Bari, Italy
    4. Mario Negri Sud Consortium, Department of Clinical Pharmacology and Epidemiology, Santa Maria Imbaro, Italy
    5. Diaverum, Medical-Scientific Office, Lund, Sweden
    Search for more papers by this author
  • Jonathan C Craig

    1. The University of Sydney, Sydney School of Public Health, Sydney, NSW, Australia
    2. The Children's Hospital at Westmead, Cochrane Renal Group, Centre for Kidney Research, Westmead, NSW, Australia
    Search for more papers by this author

Abstract

Background

The risk of cytomegalovirus (CMV) infection in solid organ transplant recipients has resulted in the frequent use of prophylaxis with the aim of preventing the clinical syndrome associated with CMV infection. This is an update of a review first published in 2005 and updated in 2008.

Objectives

To determine the benefits and harms of antiviral medications to prevent CMV disease and all-cause mortality in solid organ transplant recipients.

Search methods

We searched MEDLINE, EMBASE and the Cochrane Central Registry of Controlled Trials (CENTRAL) in The Cochrane Library to February 2004 for the first version of this review. The Cochrane Renal Group's specialised register was searched to February 2007 and to July 2011 for the first and current updates of the review without language restriction.

Selection criteria

We included randomised controlled trials (RCTs) and quasi-RCTs comparing antiviral medications with placebo or no treatment, comparing different antiviral medications and comparing different regimens of the same antiviral medications in recipients of any solid organ transplant. Studies examining pre-emptive therapy were excluded.

Data collection and analysis

Two authors independently assessed study eligibility, risk of bias and extracted data. Results were reported as risk ratios (RR) or risk differences (RD) with 95% confidence intervals (CI) for dichotomous outcomes and by mean difference (MD) with 95% CI for continuous outcomes. Statistical analyses were performed using the random-effects model. Subgroup analysis and univariate meta-regression were performed using restricted maximum-likelihood to estimate the between study variance. Multivariate meta-regression was performed to investigate whether the results were altered after allowing for differences in drugs used, organ transplanted, and recipient CMV serostatus at the time of transplantation.

Main results

We identified 37 studies (4342 participants). Risk of bias attributes were poorly performed or reported with low risk of bias reported for sequence generation, allocation concealment, blinding and selective outcome reporting in 25% or fewer studies.

Prophylaxis with aciclovir, ganciclovir or valaciclovir compared with placebo or no treatment significantly reduced the risk for CMV disease (19 studies; RR 0.42, 95% CI 0.34 to 0.52), CMV infection (17 studies; RR 0.61, 95% CI 0.48 to 0.77), and all-cause mortality (17 studies; RR 0.63, 95% CI 0.43 to 0.92) primarily due to reduced mortality from CMV disease (7 studies; RR 0.26, 95% CI 0.08 to 0.78). Prophylaxis reduced the risk of herpes simplex and herpes zoster disease, bacterial and protozoal infections but not fungal infection, acute rejection or graft loss.

Meta-regression showed no significant difference in the relative benefit of treatment (risk of CMV disease or all-cause mortality) by organ transplanted or CMV serostatus; no conclusions were possible for CMV negative recipients of negative organs.

Neurological dysfunction was more common with ganciclovir and valaciclovir compared with placebo/no treatment. In direct comparison studies, ganciclovir was more effective than aciclovir in preventing CMV disease (7 studies; RR 0.37, 95% CI 0.23 to 0.60) and leucopenia was more common with aciclovir. Valganciclovir and IV ganciclovir were as effective as oral ganciclovir. The efficacy and adverse effects of valganciclovir/ganciclovir did not differ from valaciclovir in three small studies. Extended duration prophylaxis significantly reduced the risk of CMV disease compared with three months therapy (2 studies; RR 0.20, 95% CI 0.12 to 0.35). Leucopenia was more common with extended duration prophylaxis but severe treatment associated adverse effects did not differ between extended and three month durations of treatment.

Authors' conclusions

Prophylaxis with antiviral medications reduces CMV disease and CMV-associated mortality in solid organ transplant recipients. These data suggest that antiviral prophylaxis should be used routinely in CMV positive recipients and in CMV negative recipients of CMV positive organ transplants.

Resumo

Medicamentos antivirais para a prevenção de citomegalovirose em pacientes que receberam transplante de órgãos sólidos

Introdução

O risco de infecção por citomegalovírus (CMV) nos pacientes que recebem transplantes de órgãos sólidos levou à utilização frequente de medicamentos profiláticos com objetivo de prevenir a síndrome associada com a infecção por CMV. Esta é uma atualização de uma revisão publicada pela primeira vez em 2005 e atualizada em 2008.

Objetivos

Avaliar os benefícios e riscos de medicamentos antivirais para prevenir a doença por CMV (citomegalovirose) e a mortalidade por todas as causas em receptores de transplante de órgãos sólidos.

Métodos de busca

Pesquisamos o MEDLINE, o EMBASE e o Cochrane Central Registry of Controlled Trials (CENTRAL) na Cochrane Library até fevereiro de 2004 para a primeira versão desta revisão. Pesquisamos o registro especializado Cochrane Renal Group até fevereiro de 2007, e até julho de 2011 para a primeira e esta segunda atualização da revisão. Não houve exclusão de idiomas.

Critério de seleção

Incluímos ensaios randomizados controlados (ECR) e ensaios quasi-randomizados comparando medicamentos antivirais versus placebo ou nenhum tratamento, comparando diferentes medicações antivirais e comparando diferentes regimes dos mesmos medicamentos antivirais em pacientes que receberam qualquer transplante de órgão sólido. Excluímos os estudos que avaliaram outras terapias preventivas.

Coleta dos dados e análises

Dois revisores avaliaram independentemente a elegibilidade do estudo, o risco de viés e extraíram os dados. Os resultados foram apresentados como risco relativo (RR) ou diferenças de risco (RD) com intervalo de confiança de 95% (95% CI) para desfechos dicotômicos e pela diferença média (MD) com 95% CI para desfechos contínuos. Realizamos as análises estatísticas utilizando o modelo de efeitos aleatórios. Realizamos a análise de subgrupos e meta-regressão univariada utilizando verossimilhança restrita para estimar a variância entre estudos. Realizamos meta-regressão multivariada para investigar se os resultados eram alterados depois de se considerarem as diferenças de drogas utilizadas, órgão transplantado e a presença de anticorpos anti-CMV (status sorológico) do receptor no momento do transplante.

Principais resultados

Identificamos 37 estudos (4.342 participantes). Os parâmetros de risco de viés foram mal avaliados ou relatados na maioria dos estudos. Menos de 25% deles foram classificados como tendo baixo risco de viés para a geração da sequência de randomização, sigilo de alocação, cegamento e relato de desfecho seletivo.

A profilaxia com aciclovir, ganciclovir ou valaciclovir em comparação com placebo ou nenhum tratamento reduziu significativamente o risco de infecção por CMV (19 estudos; RR 0,42, 95% CI 0,34-0,52), infecção por CMV (17 estudos; RR 0,61, 95% CI 0,48-0,77 ), mortalidade por todas as causas (17 estudos; RR 0,63, 95% CI 0,43-0,92), principalmente devido à redução da mortalidade por citomegalovirose (7 estudos; RR 0,26, 95% CI 0,08-0,78). A profilaxia reduziu o risco de herpes simplex e herpes zoster, infecções bacterianas e por protozoários, mas não a infecção fúngica, rejeição aguda ou a perda do enxerto.

A meta-regressão não evidenciou diferença significativa no benefício relativo do tratamento (risco de citomegalovirose ou mortalidade por qualquer causa) por órgão transplantado ou status sorológico; nenhuma conclusão foi possível para pacientes negativos para o CMV que receberam órgãos negativos.

Disfunção neurológica foi mais comum nos pacientes que receberam ganciclovir e valaciclovir em comparação com placebo/nenhum tratamento. Nos estudos de comparação direta, o ganciclovir foi mais eficaz do que o aciclovir na prevenção da citomegalovirose (7 estudos; RR 0,37, 95% CI 0,23-0,60) e a leucopenia foi mais comum com o aciclovir. O valganciclovir e o ganciclovir IV foram tão eficazes como o ganciclovir oral. De acordo com três estudos pequenos, a eficácia e os efeitos adversos do valganciclovir/ganciclovir não diferiram do valaciclovir. A profilaxia de longo prazo reduziu significativamente o risco de citomegalovirose em comparação com três meses de profilaxia (2 estudos; RR 0,20, 95% CI 0,12-0,35). A leucopenia foi mais comum com a profilaxia de longo prazo, mas os efeitos adversos graves associados ao tratamento não diferiram entre os tratamentos de longo prazo e de três meses.

Conclusão dos autores

A profilaxia com drogas antivirais reduz a citomegalovirose e a mortalidade associada ao CMV em pacientes que receberam transplante de órgãos sólidos. Esses dados sugerem que a profilaxia com antivirais deve ser utilizada rotineiramente em pacientes CMV-positivos e em pacientes CMV-negativos que recebem transplantes de órgãos de pacientes CMV-positivos.

Notas de tradução

Tradução do Centro Cochrane do Brasil (Michele Palmeira da Silva).

Plain language summary

Antiviral drugs used as protective and preventive therapy reduce CMV disease and CMV-associated deaths in solid organ transplant recipients

Cytomegalovirus (CMV; a herpes virus) is the most common type of virus detected in people who have received solid organ transplants (kidney, heart, liver, lung and pancreas). CMV disease is a major cause of illness and death during the first six to 12 months after transplantation. Two main strategies to prevent CMV disease have been adopted: protection and prevention (prophylaxis) of viral infections for all organ recipients using antiviral drugs, or 'pre-emptive therapy' of organ recipients, who develop evidence of CMV infection during routine screening.

We looked at the benefits and harms of antiviral prophylaxis to prevent CMV disease in people who are solid organ transplant recipients. The evidence we found shows that some antiviral drugs (ganciclovir, valaciclovir and aciclovir) reduced the risk of CMV disease, death due to CMV disease, clinical disease caused by herpes simplex and herpes zoster viruses, bacterial infections and protozoal infections.

For CMV disease and death, the relative benefits of aciclovir, ganciclovir and valaciclovir appear consistent across recipients of heart, kidney and liver transplants. These benefits occur in both CMV positive transplant recipients and CMV negative transplant recipients of CMV positive donor organs, with or without the inclusion of antilymphocyte antibody therapy, and the benefits were seen at all measured time points. We found that ganciclovir is more effective than aciclovir and as effective as valganciclovir, which is currently the most commonly used antiviral drug to prevent CMV disease in transplant recipients.

Extended duration of prophylaxis was found to be more effective than three months of therapy in kidney and lung transplant recipients. More studies are needed to determine the optimum duration and dosage of antiviral drugs for all solid organ transplant recipients.

Resumo para leigos

Medicamentos antivirais utilizados como terapia protetora e preventiva para reduzir doenças e mortes associadas ao citomegalovírus em receptores de transplante de órgãos sólidos

O citomegalovírus (CMV, um herpesvírus) é o tipo mais comum de vírus detectado em pessoas que receberam transplantes de órgãos sólidos (rim, coração, fígado, pulmão e pâncreas). A doença causada pelo CMV é uma das principais causas de enfermidade e morte durante os primeiros seis a doze meses após o transplante. Existem duas estratégias principais para prevenir a doença por CMV: proteção e prevenção (profilaxia) de infecções virais para todos os receptores de órgãos usando drogas antivirais ou a "terapia preventiva" de receptores de órgãos que desenvolvem sinais de infecção por CMV durante exames de rotina.

Nós analisamos os benefícios e malefícios da profilaxia antiviral para prevenir a doença por CMV em pessoas receptoras de transplantes de órgãos sólidos. A evidência que encontramos mostra que algumas drogas antivirais (ganciclovir, valaciclovir e aciclovir) reduziram o risco de infecção por CMV, de morte devido à doença por CMV, de doenças causadas por herpes simplex e herpes zoster, de infecções bacterianas e de infecções por protozoários.

O uso de aciclovir, valaciclovir e ganciclovir parecem reduzir de forma consistente o risco de doença por CMV e de morte em todos as pessoas que receberam de transplante de coração, rim e fígado. Esses benefícios ocorrem tanto nos transplantados que já têm anticorpos contra o CMV como naqueles sem anticorpos que receberam um órgão de um doador com anticorpos para CMV. Esses benefícios também valem para as pessoas que receberam anticorpos antilinfocíticos, e os benefícios foram observados em todos os momentos avaliados. Observamos que o ganciclovir é mais eficaz do que o aciclovir e tão eficaz quanto o valganciclovir, que é atualmente a droga antiviral mais comumente usada para prevenir a doença por CMV em receptores de transplante.

O uso de profilaxia de longa duração foi mais eficaz do que o uso de medicamentos por apenas três meses nos pacientes com transplante renal e pulmonar. Mais estudos são necessários para se saber qual é a duração e qual é a dosagem ideal dos medicamentos antivirais para todos os receptores de transplantes de órgãos sólidos.

Notas de tradução

Tradução do Centro Cochrane do Brasil (Michele Palmeira da Silva).

Laički sažetak

Antivirusni lijekovi korišteni kao zaštitna i preventivna terapija smanjuju CMV bolest i CMV povezanu smrtnost u pacijenata s transplantiranim čvrstim organima

Citomegalovirus (CMV, herpes virus) je najčešći tip virusa otkriven u ljudi kojima su transplantirani tzv. čvrsti (solidni) organi (bubreg, srce, jetra, pluća i gušterača). CMV bolest je glavni uzrok infekcije i smrti u prvih 6 do 12 mjeseci nakon transplantacije. Postoje dvije glavne strategije za prevenciju CMV bolesti: zaštita i prevencija (profilaksa) virusnih infekcija za sve primatelje organa koristeći antivirusne lijekove, ili “preemptivna terapija” primatelja organa kod kojih se uoče simptomi CMV infekcije tijekom rutinskog pregleda.

U ovom Cochrane sustavnom pregledu analizirana je korist i šteta od antivirusne profilakse za sprječavanje CMV bolesti u ljudi s transplantiranim čvrstim organima. Pronađeni dokazi pokazuju kako neki antivirusni lijekovi (ganciklovir, valaciklovir i aciklovir) smanjuju rizik od CMV bolesti i smrti, bolesti uzrokovane herpes simplex i herpes zoster virusima, bakterijskih infekcija te infekcija jednostaničnim praživotinjama.

Za CMV bolest i smrtnost, relativna korist aciklovira, ganciklovira i valaciklovira se pokazala jednakom kod primatelja srca, bubrega i jetre. Korist je uočena i kod CMV pozitivnih primatelja i kod CMV negativnih primatelja CMV pozitivnog doniranog organa, sa ili bez uključivanja antilimfocitne terapije antitijelima, a koristan učinak uočen je u svim vremenskim točkama mjerenja rezultata. Otkriveno je da je ganciklovir učinkovitiji od aciklovira te jednako učinkovit kao valganciklovir, koji se trenutno najčešće koristi kao antivirusni lijek za sprječavanje CMV bolesti nakon transplantacije organa.

Produljeno trajanje profilakse se pokazalo učinkovitijim nego samo tri mjeseca terapije kod primatelja bubrega i pluća. Potrebno je provesti daljnja istraživanja kako bi se moglo odrediti optimalno trajanje i doza antivirusnih lijekova nakon svih transplantacija čvrstih organa.

Bilješke prijevoda

Hrvatski Cochrane
Prevela: Andrea Gelemanović
Ovaj sažetak preveden je u okviru volonterskog projekta prevođenja Cochrane sažetaka. Uključite se u projekt i pomozite nam u prevođenju brojnih preostalih Cochrane sažetaka koji su još uvijek dostupni samo na engleskom jeziku. Kontakt: cochrane_croatia@mefst.hr

Summary of findings(Explanation)

Summary of findings for the main comparison. Antiviral prophylaxis versus placebo/no treatment compared with use for preventing cytomegalovirus disease in solid organ transplant recipients
  1. ₑOnly 7/19 studies reported on this outcome. Small numbers of events.
    ²Few studies and events.

Antiviral prophylaxis versus placebo/no treatment compared with use for preventing cytomegalovirus disease in solid organ transplant recipients
Patient or population: solid organ transplant recipients
Settings: tertiary hospitals
Intervention: antiviral prophylaxis versus placebo/no treatment
OutcomesIllustrative comparative risks* (95% CI)Relative effect
(95% CI)
No of Participants
(studies)
Quality of the evidence
(GRADE)
Comments
Assumed riskCorresponding risk
 Antiviral prophylaxis versus placebo/no treatment
CMV disease and CMV infection in all treated patients: all symptomatic CMV diseaseStudy populationRR 0.42
(0.34 to 0.52)
1981
(19 studies)
⊕⊕⊕⊕
high
 
299 per 1000126 per 1000
(102 to 156)
Moderate
357 per 1000150 per 1000
(121 to 186)
CMV disease for different organ transplants: Kidney transplant recipientsStudy populationRR 0.42
(0.31 to 0.57)
1132
(11 studies)
⊕⊕⊕⊕
high
 
297 per 1000125 per 1000
(92 to 169)
Moderate
400 per 1000168 per 1000
(124 to 228)
CMV disease for different organ transplants: Liver transplant recipientsStudy populationRR 0.49
(0.29 to 0.84)
616
(5 studies)
⊕⊕⊕⊝
moderate
 
262 per 1000128 per 1000
(76 to 220)
Moderate
306 per 1000150 per 1000
(89 to 257)
CMV disease for different organ transplants: Heart transplant recipientsStudy populationRR 0.39
(0.25 to 0.63)
232
(3 studies)
⊕⊕⊕⊝
moderate
 
412 per 1000161 per 1000
(103 to 260)
Moderate
425 per 1000166 per 1000
(106 to 268)
Death associated with CMV diseaseStudy populationRR 0.26
(0.08 to 0.78)
1300
(7 studies)
⊕⊕⊕⊝
moderate²
 
23 per 10006 per 1000
(2 to 18)
Moderate
39 per 100010 per 1000
(3 to 30)
All-cause mortality according to antiviral medicationStudy populationRR 0.63
(0.43 to 0.92)
1838
(17 studies)
⊕⊕⊕⊕
high
 
71 per 100045 per 1000
(30 to 65)
Moderate
45 per 100028 per 1000
(19 to 41)
Graft loss: all medicationsStudy populationRR 0.74
(0.47 to 1.17)
825
(10 studies)
⊕⊕⊕⊝
moderate²
 
93 per 100069 per 1000
(44 to 109)
Moderate
117 per 100087 per 1000
(55 to 137)
Acute rejection: all medicationsStudy populationRR 0.9
(0.78 to 1.05)
1420
(13 studies)
⊕⊕⊕⊕
high
 
468 per 1000421 per 1000
(365 to 491)
Moderate
500 per 1000450 per 1000
(390 to 525)
Herpes simplex and Herpes zoster infection: all medicationsStudy populationRR 0.27
(0.19 to 0.4)
1483
(9 studies)
⊕⊕⊕⊕
high
 
281 per 100076 per 1000
(53 to 113)
Moderate
260 per 100070 per 1000
(49 to 104)
*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI)
CI: Confidence interval; RR: Risk ratio
GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate
Very low quality: We are very uncertain about the estimate

Summary of findings 2 Ganciclovir versus aciclovir for preventing cytomegalovirus disease in solid organ transplant recipients

Summary of findings 2. Ganciclovir versus aciclovir for preventing cytomegalovirus disease in solid organ transplant recipients
  1. ₑSmall number of events in limited number of studies.

Ganciclovir versus aciclovir for preventing cytomegalovirus disease in solid organ transplant recipients
Patient or population: solid organ transplant recipients
Settings: tertiary hospitals
Intervention: ganciclovir versus aciclovir
OutcomesIllustrative comparative risks* (95% CI)Relative effect
(95% CI)
No of Participants
(studies)
Quality of the evidence
(GRADE)
Comments
Assumed riskCorresponding risk
ControlGanciclovir versus aciclovir
CMV disease and CMV infection in all treated patients: CMV disease in all patientsStudy populationRR 0.37
(0.23 to 0.6)
1113
(7 studies)
⊕⊕⊕⊕
high
 
177 per 100066 per 1000
(41 to 106)
Moderate
226 per 100084 per 1000
(52 to 136)
Death associated with CMV diseaseStudy populationRR 0.33
(0.07 to 1.58)
832
(6 studies)
⊕⊕⊕⊝
moderate
 
10 per 10003 per 1000
(1 to 15)
Moderate
9 per 10003 per 1000
(1 to 14)
All-cause mortalityStudy populationRR 1.13
(0.82 to 1.58)
1138
(8 studies)
⊕⊕⊕⊝
moderate
 
103 per 1000117 per 1000
(85 to 163)
Moderate
109 per 1000123 per 1000
(89 to 172)
Acute rejectionStudy populationRR 0.98
(0.87 to 1.1)
1009
(6 studies)
⊕⊕⊕⊕
high
 
491 per 1000481 per 1000
(427 to 540)
Moderate
517 per 1000507 per 1000
(450 to 569)
Graft lossStudy populationRR 0.55
(0.27 to 1.13)
268
(3 studies)
⊕⊕⊝⊝
low
 
148 per 100081 per 1000
(40 to 167)
Moderate
167 per 100092 per 1000
(45 to 189)
Other viral infectionsStudy populationRR 0.81
(0.32 to 2.01)
740
(4 studies)
⊕⊕⊕⊝
moderate
 
35 per 100028 per 1000
(11 to 70)
Moderate
44 per 100036 per 1000
(14 to 88)
Invasive fungal infectionsStudy populationRR 0.67
(0.4 to 1.1)
401
(3 studies)
⊕⊕⊝⊝
low
 
149 per 1000100 per 1000
(60 to 164)
Moderate
51 per 100034 per 1000
(20 to 56)
*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI)
CI: Confidence interval; RR: Risk ratio
GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate
Very low quality: We are very uncertain about the estimate

Summary of findings 3 Valaciclovir versus ganciclovir or valganciclovir for preventing cytomegalovirus disease in solid organ transplant recipients

Summary of findings 3. Valaciclovir versus ganciclovir or valganciclovir for preventing cytomegalovirus disease in solid organ transplant recipients
  1. ₑSmall numbers of patients.

Valaciclovir versus ganciclovir or valganciclovir for preventing cytomegalovirus disease in solid organ transplant recipients
Patient or population: solid organ transplant recipients
Settings: known or unknown
Intervention: valaciclovir versus ganciclovir or valganciclovir
OutcomesIllustrative comparative risks* (95% CI)Relative effect
(95% CI)
No of Participants
(studies)
Quality of the evidence
(GRADE)
Comments
Assumed riskCorresponding risk
ControlValaciclovir versus ganciclovir or valganciclovir
CMV disease and CMV infection in all treated patients: CMV diseaseStudy populationRR 0.74
(0.15 to 3.75)
188
(3 studies)
⊕⊕⊝⊝
low
 
32 per 100024 per 1000
(5 to 120)
Moderate
25 per 100019 per 1000
(4 to 94)
All-cause mortalityStudy populationRR 1.03
(0.15 to 6.9)
154
(2 studies)
⊕⊕⊝⊝
low
 
26 per 100027 per 1000
(4 to 182)
Moderate
28 per 100029 per 1000
(4 to 193)
Acute rejectionStudy populationRR 0.91
(0.22 to 3.73)
188
(3 studies)
⊕⊕⊝⊝
low
 
181 per 1000165 per 1000
(40 to 675)
Moderate
125 per 1000114 per 1000
(27 to 466)
Graft lossStudy populationRR 1.34
(0.23 to 7.86)
107
(2 studies)
⊕⊕⊝⊝
low
 
73 per 100097 per 1000
(17 to 572)
Moderate
56 per 100075 per 1000
(13 to 440)
*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI)
CI: Confidence interval; RR: Risk ratio
GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate
Very low quality: We are very uncertain about the estimate

Summary of findings 4 Extended duration compared with 3 months of valganciclovir compared with use for preventing cytomegalovirus disease in solid organ transplant recipients

Summary of findings 4. Extended duration compared with 3 months of valganciclovir compared with use for preventing cytomegalovirus disease in solid organ transplant recipients
  1. ₑConsiderable heterogeneity between studies.

Extended duration compared with 3 months of valganciclovir compared with use for preventing cytomegalovirus disease in solid organ transplant recipients
Patient or population: solid organ transplant recipients
Settings: known or unknown
Intervention: extended duration compared with three months of valganciclovir
OutcomesIllustrative comparative risks* (95% CI)Relative effect
(95% CI)
No of Participants
(studies)
Quality of the evidence
(GRADE)
Comments
Assumed riskCorresponding risk
 Extended duration compared with three months of valganciclovir
CMV disease at end of treatmentStudy populationRR 0.2
(0.12 to 0.35)
454
(2 studies)
⊕⊕⊕⊕
high
 
314 per 100063 per 1000
(38 to 110)
Moderate
316 per 100063 per 1000
(38 to 111)
CMV syndromeStudy populationRR 0.4
(0.27 to 0.6)
454
(2 studies)
⊕⊕⊕⊕
high
 
310 per 1000124 per 1000
(84 to 186)
Moderate
272 per 1000109 per 1000
(73 to 163)
CMV invasive disease: Number at 12 monthsStudy populationRR 0.23
(0.01 to 3.5)
454
(2 studies)
⊕⊕⊝⊝
low
 
66 per 100015 per 1000
(1 to 229)
Moderate
109 per 100025 per 1000
(1 to 381)
CMV infection at end of treatmentStudy populationRR 0.27
(0.1 to 0.71)
454
(2 studies)
⊕⊕⊕⊕
high
 
502 per 1000136 per 1000
(50 to 357)
Moderate
542 per 1000146 per 1000
(54 to 385)
Biopsy-proven acute rejection at 12 monthsStudy populationRR 0.99
(0.42 to 2.37)
454
(2 studies)
⊕⊕⊝⊝
low
 
183 per 1000182 per 1000
(77 to 435)
Moderate
192 per 1000190 per 1000
(81 to 455)
Opportunistic infectionsStudy populationRR 0.71
(0.33 to 1.57)
456
(2 studies)
⊕⊕⊝⊝
low
 
343 per 1000244 per 1000
(113 to 539)
Moderate
399 per 1000283 per 1000
(132 to 626)
Total treatment related adverse effectsStudy populationSee comment456
(2 studies)
⊕⊕⊕⊕
high
Risks were calculated from pooled risk differences
426 per 1000503 per 1000
(418 to 588)
Moderate
353 per 1000417 per 1000
(346 to 487)
*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval; RR: Risk ratio
GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate
Very low quality: We are very uncertain about the estimate

Background

Description of the condition

Cytomegalovirus (CMV) is the most common virus pathogen in solid organ transplant recipients being a major cause of morbidity and mortality during the first six months post-transplant (Fishman 1998; Rubin 2000). The overall incidence of symptomatic CMV disease in the transplant population ranges from 30% to 50% with the incidence and severity being highest among lung recipients (Linden 2000). Approximately 50% of deaths following lung transplantation are attributed to infection (Michaels 2000). Like all herpes viruses, CMV has the propensity to establish lifelong latency infection in the host after the initial infection has resolved. Therefore, a solid organ recipient may be infected either by exogenous virus or by reactivation of latent virus if they were CMV positive pre-transplant. Those at highest risk of symptomatic CMV disease are CMV seronegative patients who receive organs from CMV seropositive donors, and CMV seropositive patients on heavily immunosuppressive regimens (Fishman 1998; Rubin 2000). CMV may manifest as a non-specific illness characterised by fever, mononucleosis, leucopenia and thrombocytopenia, or as a variety of clinical syndromes including pneumonitis, hepatitis, encephalitis and focal gastrointestinal disease. In addition, CMV infection causes morbidity in organ recipients through indirect effects on their immune response (Rubin 1989), and is associated with increased risk of allograft injury and rejection (Grattan 1989; Keenan 1991), opportunistic infections (Fishman 1995; Hadley 1995; Van den Berg 1996) and late onset malignancies such as Epstein-Barr virus lymphoproliferative disease (Basgoz 1995).

Description of the intervention

Two main strategies to prevent CMV disease have been adopted: universal prophylaxis of organ recipients with antiviral agents and/or immunoglobulins, or pre-emptive therapy of organ recipients, who develop evidence of asymptomatic CMV infection during screening (Rubin 1989). Antiviral medications may be given intravenously (ganciclovir, aciclovir, immunoglobulins) but are now more commonly administered once daily orally with the availability of the longer acting oral preparations valganciclovir and valaciclovir. Prophylaxis is usually administered for three to six months during the time that patients are most at risk of CMV infection and disease. Pre-emptive therapy relies upon monitoring for CMV infection by pp65 antigenaemia assay or for CMV DNA using quantitative polymerase chain reaction (PCR) with administration of antiviral therapy when CMV infection is diagnosed (Emery 2000).

How the intervention might work

This review examines the use of prophylaxis to prevent CMV infection and CMV disease. Prophylaxis is usually administered for the first three to six months after transplant when the recipient is at highest risk of CMV infection. Prevention of CMV disease should reduce the associated morbidity and mortality. In addition, prophylaxis may reduce the indirect effects of CMV infection including opportunistic infections, acute rejection and graft loss.

Why it is important to do this review

There remains a lack of consensus on the merits of the various CMV prophylaxis protocols available (Fishman 1998; Humar 2009). Universal prophylaxis exposes all solid organ transplant recipients to the adverse effects of medications, particularly haematological effects (leucopenia, neutropenia, increased risk of infection) with valganciclovir, and neurological effects with valaciclovir. However, based on epidemiological studies many recipients do not develop disease without prophylaxis (Humar 2009). Thus, prophylaxis among kidney transplant recipients has commonly been limited to CMV negative recipients of CMV positive kidneys and to recipients receiving antibodies to lymphocyte antigens. Prophylaxis may also be associated with an increased risk of late onset CMV disease occurring after discontinuation of prophylaxis and with the development of resistant organisms (Humar 2009). A systematic review was therefore required to assess the benefits and harms of antiviral prophylaxis in solid organ transplants.

A meta-analysis of prophylactic treatment versus placebo/no treatment was originally published in The Cochrane Database of Systematic Reviews (Couchoud 1998a). When this review was updated in 2008, more recent articles comparing prophylaxis with antiviral medications (including aciclovir, ganciclovir, valaciclovir, valganciclovir) were included. This review also included studies comparing one prophylactic antiviral medication with another. We have examined the effect of prophylaxis with antiviral agents in recipients of solid organ transplant recipients on CMV disease, all CMV infection, the incidence of acute rejection, graft loss, opportunistic infections and death. We have compared the treatment effect of each regimen among different solid organs and different risk groups. Finally, the review evaluated potential harms caused by antiviral medications, namely nephrotoxicity, bone marrow suppression, and emergence of resistant CMV strains. Other reviews have evaluated pre-emptive therapy on detection of CMV viraemia (Strippoli 2006a; Strippoli 2006b) and the use of other agents (immunoglobulins, vaccines, interferon) alone or in combination with antiviral medications (Hodson 2007). The review was originally published in 2005 and was updated in 2008. It is now updated in 2013.

The Cochrane review Pre-emptive therapy for cytomegalovirus viraemia to prevent cytomegalovirus disease in solid organ transplant recipients (Owers 2013) has been updated concomitantly with this review. Pre-emptive therapy compared with placebo/no specific therapy reduced the risk of CMV disease by 70% (6 studies; 288 participants). While there was no significant difference in the prevention of CMV disease with pre-emptive therapy compared with prophylaxis (7 studies; 753 participants), there was some imprecision of results and significant heterogeneity among studies limiting the applicability of these data to patient management.

Objectives

This review aimed to assess the benefits and harms of antiviral medications for preventing symptomatic CMV disease in solid organ transplant recipients of all ages, irrespective of CMV serostatus prior to transplantation. The secondary aims were to evaluate the efficacy of antiviral medications in preventing all CMV infection (symptomatic and asymptomatic where CMV is detected only by laboratory investigation) and in decreasing the incidence of acute rejection, graft loss, death (all-cause mortality and mortality due to CMV disease), opportunistic infections and to evaluate the harms of each antiviral medication.

The review compared studies of antiviral medications with placebo/no treatment and explored comparisons between two or more antiviral agents and/or two different doses or durations of the same antiviral agent. Thirdly, it has compared the treatment effect of each regimen between different solid organs and finally, among the different risk groups (i.e. pre-existent CMV serostatus and/or level of immunosuppression).

Methods

Criteria for considering studies for this review

Types of studies

All randomised controlled trials (RCTs) and quasi-RCTs (RCTs in which allocation to treatment was obtained by alternation, use of alternate medical records, date of birth or other predictable method) were included.

Types of participants

Participants of all ages, irrespective of CMV serostatus prior to transplantation, who have undergone at least one solid organ transplant (kidney, liver, lung, heart, pancreas). Bone marrow and other cellular transplants were excluded.

Types of interventions

Interventions included antiviral medications (aciclovir, ganciclovir, valaciclovir, valganciclovir). Comparisons were made between antiviral medications and placebo/no treatment, two different antiviral medications, or two varying doses or durations of an antiviral medication.

Studies of pre-emptive treatment (i.e. treatment on detection of CMV viraemia), immunoglobulin alone or with antiviral medications, vaccines or interferon were excluded. Treatment regimens for symptomatic CMV disease were excluded as these are the subject of other reviews (Strippoli 2006a; Strippoli 2006b; Hodson 2007).

Types of outcome measures

Primary outcomes

The primary outcome measures were the incidence of CMV disease (documented CMV infection with clinical symptoms) and all-cause mortality. The definition of symptomatic CMV disease used was that defined by the study investigators. This was usually the diagnosis of CMV infection in association with one or more of the following: CMV syndrome (temperature of 38°C or more with no other documented source in association with one or more of atypical lymphocytosis, leucopenia or thrombocytopenia), pneumonitis, focal gastrointestinal disease, liver function abnormality, or encephalitis.

Secondary outcomes

Secondary outcomes included the incidence all CMV infection (symptomatic and asymptomatic); acute rejection; graft loss; death due to CMV disease; opportunistic infections; time to CMV disease; and harms (including nephrotoxicity, bone marrow suppression, emergence of resistant CMV strains, late onset of CMV disease). All outcomes were recorded as present/absent except time to the development of CMV disease.

The definition of CMV infection used was that defined by the study investigators. This was usually the isolation of CMV from a cultured specimen from any site, or positive histopathology or CMV antigen detection in a tissue specimen, or the presence of CMV pp65 antigenaemia, or an elevation in CMV viral load as detected by qualitative or quantitative PCR (as defined by the investigator).

Graft loss was defined as the need for dialysis for kidney transplantation or retransplantation for other organs during the follow-up period of the study. Acute rejection was defined as used by the individual authors. This was either biopsy proven or clinical, defined by rise in creatinine levels with respect to kidney transplants or response to rejection treatment.

Search methods for identification of studies

Initial search

A systematic and comprehensive literature search was carried out to identify eligible RCTs (Appendix 1). There was no language restriction. We searched:

  • The Cochrane Renal Group's specialised register (February 2004).

  • The Cochrane Central Register of Controlled Trials (CENTRAL in The Cochrane Library Issue 1, 2004).

  • MEDLINE (1966 to February 2004) using the optimally sensitive search strategy developed for identification of RCTs (Dickersin 1994).

  • EMBASE (1980 to February 2004) using the optimally sensitive search strategy developed for identification of RCTs (Lefebvre 1996).

The Trials Search Coordinator ensured that all relevant studies had been identified. Additional studies were located through article reference lists and from abstracts from international meetings.

Review update 2008

For this update the Cochrane Renal Group's specialised register and CENTRAL was searched to February 2007. CENTRAL and the Renal Group's specialised register contain the handsearched results of conference proceedings from general and speciality meetings. This is an ongoing activity across the Cochrane Collaboration and is both retrospective and prospective (Master List 2007). Please refer to The Cochrane Renal Group's Module in The Cochrane Library for the complete list of nephrology conference proceedings searched.

Electronic searches

For the current update (2013) we searched the Cochrane Renal Group's specialised register to July 2011 through contact with the Trials Search Co-ordinator using search terms relevant to this review. The Cochrane Renal Group’s specialised register contains studies identified from the following sources.

  • Quarterly searches of the Cochrane Central Register of Controlled Trials CENTRAL.

  • Weekly searches of MEDLINE OVID SP.

  • Handsearching of renal-related journals and the proceedings of major renal conferences.

  • Searching of the current year of EMBASE OVID SP.

  • Weekly current awareness alerts for selected renal journals.

  • Searches of the International Clinical Trials Register (ICTRP) Search Portal and ClinicalTrials.gov.

Studies contained in the specialised register are identified through search strategies for CENTRAL, MEDLINE, and EMBASE based on the scope of the Cochrane Renal Group. Details of these strategies as well as a list of handsearched journals, conference proceedings and current awareness alerts are available in the specialised register section of information about the Cochrane Renal Group.

Appendix 1 presents terms used in search strategies for this review.

Searching other resources

  1. Reference lists of nephrology textbooks, review articles and relevant studies.

  2. Letters seeking information about unpublished or incomplete studies to investigators known to be involved in previous studies.

Data collection and analysis

Selection of studies

Two authors independently screened titles and abstracts retrieved from the searches and identified those studies that met the inclusion criteria. This process favoured over-selection in order to include all relevant studies. The full article was retrieved if uncertainty existed or when the abstract was not available. Any disagreement with article selection was resolved through discussion and consultation.

Data extraction and management

Two authors independently extracted data from eligible studies using standardised data extraction forms. Studies reported in foreign language journals were translated before data extraction. Participant characteristics (number, age, sex, comorbidities), interventions (type of treatment, dose, duration, co-interventions) and primary and secondary outcome measures were recorded. Authors were contacted to obtain missing information on allocation concealment. Any discrepancies in data extraction were resolved in discussion. Where results of a study were published more than once, the most complete data were extracted from all sources and used in the analysis only once.

Assessment of risk of bias in included studies

The following items were independently assessed by two authors using the risk of bias assessment tool (Higgins 2011; Appendix 2).

  • Was there adequate sequence generation (selection bias)?

  • Was allocation adequately concealed (selection bias)?

  • Was knowledge of the allocated interventions adequately prevented during the study (detection bias)?

    • Participants and personnel

    • Outcome assessors

  • Were incomplete outcome data adequately addressed (attrition bias)?

  • Are reports of the study free of suggestion of selective outcome reporting (reporting bias)?

  • Was the study apparently free of other problems that could put it at a risk of bias?

Measures of treatment effect

Dichotomous outcomes (CMV disease, all-cause mortality) were expressed as risk ratios (RR) with 95% confidence intervals (CI). Risk differences (RD) with 95% confidence intervals were calculated for adverse effects. Continuous outcomes were calculated as mean differences (MD) with 95% CI.

Unit of analysis issues

If available, data for the first period of cross-over studies were to be included in meta-analyses; otherwise, cross-over studies were reported in the text only.

Dealing with missing data

Study authors were contacted for information on sequence generation, allocation concealments and for missing data. Where missing data were few and not thought likely to influence results, the available data were analysed.

Assessment of heterogeneity

Heterogeneity was analysed using a Chi² test on N-1 degrees of freedom, with an alpha of 0.05 used for statistical significance and with the I² test (Higgins 2003). I² values of 25%, 50% and 75% correspond to low, medium and high levels of heterogeneity.

Assessment of reporting biases

The updated review included all studies identified in the Cochrane Renal Group's specialised register, which is updated regularly with published and unpublished reports identified in congress proceedings. This reduces the risk of publication bias. All reports of a single study were reviewed to ensure that all outcomes were reported to reduce the risk of selection bias.

Data synthesis

Data were pooled using a random-effects model to calculate a summary estimate of effect.

Subgroup analysis and investigation of heterogeneity

To explore clinical differences among studies that might be expected to influence the magnitude of the treatment effect for the primary outcomes of CMV disease and all-cause mortality, subgroup analysis and univariate meta-regression was performed using STATA software (StataCorp LP, Texas, USA) using restricted maximum-likelihood to estimate the between-study variance. The potential sources of variability defined a priori were organ transplanted, antiviral medication used, use of immunosuppressive regimen including antibody therapy, treatment duration, donor/recipient CMV status at transplant, the time from transplant that the outcomes were measured, and methodological quality. Multivariate meta-regression was performed to investigate whether the results were altered after allowing for the differences in drug used, organ transplanted and recipient CMV serostatus at the time of transplantation.

Sensitivity analysis

Where a study's results differed considerably from other studies in a meta-analysis, exclusion of the study was investigated to determine whether this altered the result of the meta-analysis.

Results

Description of studies

Results of the search

In the original search in February 2004, 1120 reports were initially identified from the literature search (Figure 1). The titles were screened and 927 articles were excluded. The remaining 193 abstracts or full text reports were screened and 32 studies were included.

Figure 1.

Study flow diagram.

In the 2008 update, two new studies (two reports) were included, and three additional reports of already included studies were identified. One study was excluded because the intervention was ineligible for inclusion.

A further search in July 2011 identified six new potentially eligible studies (17 reports) and 18 new excluded studies (31 reports). There were also 29 additional reports of 13 already included studies and four additional reports of three studies, which had already been excluded. Of the six potentially eligible studies, three studies (14 reports) were included (2VAL Study 2010 Kidney; IMPACT 2010 Kidney; Palmer 2010 Lung), two were excluded after full text review (Said 2007; Pescovitz 2009) and one was an ongoing study (Villano 2010).

Included studies

In the original review published in 2005, 19 studies compared aciclovir (Balfour 1989 Kidney; Barkholt 1999 Liver; Gavalda 1997 Liver; Kletzmayr 1996 Kidney; Rostaing 1994 Kidney; Saliba 1993 Liver), ganciclovir (Ahsan 1997 Kidney; Brennan 1997 Kidney; Cohen 1993 Liver; Conti 1995 Kidney; Gane 1997 Liver; Hibberd 1995 Kidney; Leray 1995 Kidney; Macdonald 1995 Heart; Merigan 1992 Heart; Pouteil-Noble 1996 Kidney; Rondeau 1993 Kidney) or valaciclovir (Egan 2002 Heart; Lowance 1999 Kidney) with placebo or no treatment. Fifteen of these 19 studies excluded CMV negative recipients of CMV negative donors. Eleven studies compared different antiviral medications (Badley 1997 Liver, Duncan 1993 Lung, Flechner 1998 Kidney, Green 1997 Liver, Martin 1994 Liver; Nakazato 1993 Liver; Paya 2004 All; Reischig 2005 Kidney; Rubin 2002 All; Winston 2003 Liver; Winston 1995 Liver); and two studies (Hertz 1998 Heart/lung; Winston 2004 Liver) compared different regimens of ganciclovir administration. Recipients of transplants other than heart, kidney and liver were not included in studies comparing treatment with placebo or no treatment and were investigated in only three comparison studies. All identified studies were published in English language. Among studies comparing antiviral medications with placebo/no treatment, no significant publication bias could be demonstrated on funnel plot (Figure 2). There were too few studies comparing ganciclovir and aciclovir to subject the data to a funnel plot. The 2005 review included 32 studies (3737 participants) (Figure 1).

Figure 2.

Funnel plot of 19 trials comparing antiviral medications with placebo or no treatment

In the 2008 update, five additional publications were included. These were an abstract of an included study (Ahsan 1997 Kidney); one publication reported the full results of an included study, and an additional publication assessed one outcome from that study (Reischig 2005 Kidney); and two new studies (Nafar 2005 Kidney; Pavlopoulou 2005 Kidney). Pavlopoulou 2005 Kidney compared valaciclovir with ganciclovir and Nafar 2005 Kidney compared oral with IV ganciclovir. The 2008 update included 34 studies (3850 participants).

In the 2013 update, three additional studies were included (2VAL Study 2010 Kidney; IMPACT 2010 Kidney; Palmer 2010 Lung). 2VAL Study 2010 Kidney compared valaciclovir with valganciclovir, but only preliminary results at four months were available; IMPACT 2010 Kidney compared 200 days of oral valganciclovir with 100 days in kidney transplant recipients; and Palmer 2010 Lung compared 12 months of oral valganciclovir with three months in lung transplant recipients. The 2012 update included 37 studies (4342 participants).

Green 1997 Liver specifically included children; the inclusion criteria for the Paya 2004 All and Rubin 2002 All studies indicated that children aged over 12 years could be included; however, the youngest participant in the Rubin 2002 All study was aged 20 years, and the average participant age in the Paya 2004 All study was 45 years.

Excluded studies

In the 2005 review, we excluded 47 studies after full text review: four were systematic reviews; 10 were narrative reviews; 12 involved ineligible interventions; and 21 were not RCTs.

In the 2008 update, one study was excluded because it compared pre-emptive therapy with prophylaxis (Khoury 2006).

In the 2013 update, 19 additional studies (34 reports) were excluded after reviewing abstracts: six were not RCTs and 13 studies involved an ineligible intervention. We excluded two studies after full text review: Pescovitz 2009 was a pharmacokinetic study and Said 2007 was a sequential study. We also identified four additional reports of three studies that had previously been excluded.

Risk of bias in included studies

Allocation

The risk of bias was low for sequence generation in 12 studies (2VAL Study 2010 Kidney; Ahsan 1997 Kidney; Badley 1997 Liver; Balfour 1989 Kidney; Cohen 1993 Liver; Egan 2002 Heart; Flechner 1998 Kidney; Macdonald 1995 Heart; Martin 1994 Liver; Palmer 2010 Lung; Paya 2004 All; Reischig 2005 Kidney); high in one study (Brennan 1997 Kidney); and unclear in the remaining studies.

Of 19 studies comparing prophylaxis with placebo or no treatment, the risk of bias was low for allocation concealment in four (21%) studies (Cohen 1993 Liver; Egan 2002 Heart;; Pouteil-Noble 1996 Kidney; Saliba 1993 Liver); high in one study (Brennan 1997 Kidney); and the information was unclear in 14 studies. Of 13 studies comparing different medications, allocation concealment bias was low in six studies (2VAL Study 2010 Kidney; Badley 1997 Liver; Flechner 1998 Kidney; Paya 2004 All; Reischig 2005 Kidney; Rubin 2002 All); and information was not available for seven studies. Of the remaining studies, allocation concealment bias was low in two studies (IMPACT 2010 Kidney; Palmer 2010 Lung) but Information on allocation concealment was not available for three (Hertz 1998 Heart/lung; Nafar 2005 Kidney; Winston 2004 Liver).

Blinding

Performance bias was assessed as low risk in 10 studies (27%), including seven that compared prophylaxis with placebo (Balfour 1989 Kidney; Barkholt 1999 Liver; Gane 1997 Liver; Lowance 1999 Kidney; Macdonald 1995 Heart; Merigan 1992 Heart; Pouteil-Noble 1996 Kidney); one study comparing different antiviral agents (Paya 2004 All); and two studies comparing different durations of the same medication (IMPACT 2010 Kidney; Palmer 2010 Lung). The risk of bias was unclear for blinding of participants and investigators in one study (Egan 2002 Heart); and the remaining studies were assessed as being at high risk of performance bias.

The risk of detection bias was low in nine studies (24%) (Balfour 1989 Kidney; Barkholt 1999 Liver; Gane 1997 Liver; IMPACT 2010 Kidney; Lowance 1999 Kidney; Macdonald 1995 Heart; Merigan 1992 Heart; Palmer 2010 Lung; Paya 2004 All; Pouteil-Noble 1996 Kidney); unclear in one study (Egan 2002 Heart); and the remaining studies were judged to be at high risk of detection bias.

Incomplete outcome data

We identified 34 studies (92%) that were considered to be at low risk of attrition bias. Of these, 19 studies compared prophylaxis with placebo/no treatment (Ahsan 1997 Kidney; Balfour 1989 Kidney; Barkholt 1999 Liver; Brennan 1997 Kidney; Cohen 1993 Liver; Conti 1995 Kidney; Egan 2002 Heart; Gane 1997 Liver; Gavalda 1997 Liver; Hibberd 1995 Kidney; Kletzmayr 1996 Kidney; Lowance 1999 Kidney; Macdonald 1995 Heart; Merigan 1992 Heart; Pouteil-Noble 1996 Kidney; Rondeau 1993 Kidney; Rostaing 1994 Kidney; Saliba 1993 Liver; Winston 1995 Liver); 10 compared different antiviral medications (Badley 1997 Liver; Duncan 1993 Lung; Flechner 1998 Kidney; Green 1997 Liver; Martin 1994 Liver; Nakazato 1993 Liver; Pavlopoulou 2005 Kidney; Paya 2004 All; Rubin 2002 All; Winston 2003 Liver); and five compared different regimens of ganciclovir (Hertz 1998 Heart/lung; Nafar 2005 Kidney; Winston 2004 Liver) or of valganciclovir (IMPACT 2010 Kidney; Palmer 2010 Lung). In two studies, it was unclear whether attrition bias existed (2VAL Study 2010 Kidney; Leray 1995 Kidney). The remaining study was considered to be at high risk of attrition bias (Nafar 2005 Kidney).

Selective reporting

No protocols were available. Studies were considered to be at low risk of bias if they reported all the expected outcomes (CMV disease, CMV infection, acute rejection, graft loss, death, opportunistic infections, adverse effects). Seven studies were considered to be at low risk of bias (Balfour 1989 Kidney; Barkholt 1999 Liver; Egan 2002 Heart; Gane 1997 Liver; IMPACT 2010 Kidney; Paya 2004 All; Winston 1995 Liver). Four studies were considered to be at unclear risk of bias (2VAL Study 2010 Kidney; Leray 1995 Kidney; Pouteil-Noble 1996 Kidney; Saliba 1993 Liver). The remaining 26 studies were considered to be at high risk of bias because they failed to report adequately on one or more outcomes.

Other potential sources of bias

Five studies were considered at low risk of bias as they reported funding from government or university sources (2VAL Study 2010 Kidney; Badley 1997 Liver; Balfour 1989 Kidney; Reischig 2005 Kidney; Rondeau 1993 Kidney). Thirteen studies were considered to be at high risk of bias because they reported pharmaceutical sponsorship (Barkholt 1999 Liver; Brennan 1997 Kidney; Egan 2002 Heart; Gane 1997 Liver; Hibberd 1995 Kidney; IMPACT 2010 Kidney; Lowance 1999 Kidney; Merigan 1992 Heart; Nakazato 1993 Liver; Palmer 2010 Lung; Paya 2004 All; Rubin 2002 All; Winston 2003 Liver; Winston 1995 Liver). In the remaining 19 studies it was unclear whether pharmaceutical sponsorship existed or what impact it had on the conduct of the study.

Effects of interventions

See: Summary of findings for the main comparison Antiviral prophylaxis versus placebo/no treatment compared with use for preventing cytomegalovirus disease in solid organ transplant recipients; Summary of findings 2 Ganciclovir versus aciclovir for preventing cytomegalovirus disease in solid organ transplant recipients; Summary of findings 3 Valaciclovir versus ganciclovir or valganciclovir for preventing cytomegalovirus disease in solid organ transplant recipients; Summary of findings 4 Extended duration compared with 3 months of valganciclovir compared with use for preventing cytomegalovirus disease in solid organ transplant recipients

Antiviral medication versus placebo/no treatment

We identified 19 studies (1981 patients) that compared antiviral medications with placebo or no treatment. Six studies administered aciclovir (Balfour 1989 Kidney; Barkholt 1999 Liver; Gavalda 1997 Liver; Kletzmayr 1996 Kidney; Rostaing 1994 Kidney; Saliba 1993 Liver); 11 studies administered ganciclovir (Ahsan 1997 Kidney; Brennan 1997 Kidney; Cohen 1993 Liver; Conti 1995 Kidney; Gane 1997 Liver; Hibberd 1995 Kidney; Leray 1995 Kidney; Macdonald 1995 Heart; Merigan 1992 Heart; Pouteil-Noble 1996 Kidney; Rondeau 1993 Kidney); and two studies administered valaciclovir (Egan 2002 Heart; Lowance 1999 Kidney).

CMV disease and CMV infection

The average risk of CMV disease was 30% (range 11% to 72%). Prophylaxis with all agents significantly reduced the risk for CMV disease overall (Analysis 1.1.1 (19 studies, 1981 participants): RR 0.42, 95% CI 0.34 to 0.52; I² = 13%), CMV syndrome (Analysis 1.1.2 (11 studies, 1570 participants): RR 0.41, 95% CI 0.29 to 0.57; I² = 0%) and CMV invasive organ disease (Analysis 1.1.3 (12 studies, 1628 participants): RR 0.34, 95% CI 0.21 to 0.55; I² = 35%) compared with placebo or no treatment. No significant heterogeneity between studies was detected in the effect of prophylaxis on CMV disease, syndrome and invasive organ disease.

Figure 3 shows the cumulative meta-analysis demonstrating changes over time for CMV disease. There was a consistent reduction in CMV disease with antiviral prophylaxis from the first study in 1989 with the relative risk remaining stable from 1996 but with a progressive narrowing in confidence intervals.

Figure 3.

CMV disease: cumulative meta-analysis showing change over time

Time to onset of CMV disease was reported in 11 studies. Prophylaxis significantly increased the time from transplant to the onset of CMV disease in nine studies. Different methods of reporting prevented these data being combined in a meta-analysis.

The average risk of CMV infection in the placebo/no treatment arms of all studies was 49% (range 36% to 100%). Prophylaxis significantly reduced CMV infection (Analysis 1.1.4 (17 studies, 1786 participants): RR 0.61, 95% CI 0.48 to 0.77; I² = 76%). Considerable heterogeneity existed between studies for CMV infection with no explanation apparent, but the summary estimates for individual studies favoured prophylaxis in 15/17 studies.

Subgroup analyses for CMV disease

Subgroup analyses according to antibody status, antiviral medications, organ transplanted, treatment duration, use of antilymphocyte therapy, time to outcome assessment, study quality and other aspects of study design did not demonstrate any differences in treatment effects. Multivariate meta-regression showed no significant difference in CMV disease after allowing for potential confounding or effect-modification by prophylactic drug used, organ transplanted or recipient serostatus in CMV positive recipients and CMV negative recipients of CMV positive donors. (See Table 1).

Table 1. Potential sources of variability: CMV disease and all-cause mortality
  1. ¹Studies in "positive" group included those in which recipients were positive for CMV with donor positive or negative for CMV.
    ²Studies in "negative" group included those in which CMV negative recipients received CMV positive organs.
    ³Studies in which recipients were CMV positive and the donors CMV positive (positive group) or negative (CMV negative group).
    ªGanciclovir studies only.

VariableCMV diseaseAll-cause mortality
Number of studiesRR (95% CI)P for interactionNumber of studiesRR (95% CI)P for interaction

Antiviral medication

  1. Aciclovir

  2. Ganciclovir

  3. Valaciclovir

  1. 6

  2. 11

  3. 2

  1. 0.45 (0.29 to 0.69)

  2. 0.44 (0.34 to 0.58)

  3. 0.30 (0.19 to 0.49)

0.43
  1. 5

  2. 10

  3. 2

  1. 0.67 (0.38 to 1.20)

  2. 0.69 (0.29 to 1.65)

  3. 0.50 (0.22 to 1.15)

0.85

Time to outcome assessment

  1. 3 to 6 months

  2. 9 to 12 months

  1. 11

  2. 8

  1. 0.46 (0.36 to 0.58)

  2. 0.36 (0.22 to 0.58)

0.37
  1. 7

  2. 10

  1. 0.63 (0.40 to 0.97)

  2. 0.64 (0.31 to 1.33)

0.83

Recipient CMV status

  1. Positive (donor +ve or -ve)¹

  2. Negative (donor +ve)²

  1. 13

  2. 10

  1. 0.34 (0.24 to 0.50)

  2. 0.52 (0.37 to 0.74)

0.12
  1. 7

  2. 4

  1. 0.59 (0.30 to 1.18)

  2. 1.42 (0.44 to 4.66)

0.23

Donor CMV status³

  1. Positive (recipients all +ve)

  2. Negative (recipients all +ve)

  1. 5

  2. 5

  1. 0.18 (0.09 to 0.36)

  2. 0.33 (0.11 to 0.95)

0.37
  1. No data

  2. No data

  1. No data

  2. No data

No data

Organ transplanted

  1. Kidney

  2. Liver

  3. Heart

  1. 11

  2. 5

  3. 3

  1. 0.42 (0.31 to 0.57)

  2. 0.49 (0.29 to 0.84)

  3. 0.39 (0.25 to 0.63)

0.93
  1. 10

  2. 4

  3. 3

  1. 0.49 (0.24 to 1.00)

  2. 0.64 (0.39 to 1.00)

  3. 1.82 (0.39 to 8.51)

0.13

Antibody therapy

  1. Yes

  2. No

  1. 11

  2. 6

  1. 0.43 (0.33 to 0.55)

  2. 0.47 (0.29 to 0.76)

0.74
  1. 10

  2. 5

  1. 0.81 (0.33 to 2.01)

  2. 0.63 (0.39 to 1.00)

0.93

Treatment durationª

  1. 6 weeks or less

  2. More than 6 weeks

  1. 7

  2. 4

  1. 0.49 (0.36 to 0.68)

  2. 0.33 (0.21 to 0.53)

0.72
  1. 6

  2. 4

  1. 0.91 (0.17 to 4.92)

  2. 0.62 (0.30 to 1.30)

0.15

Allocation concealment

  1. Adequate

  2. Unclear or inadequate

  1. 4

  2. 15

  1. 0.50 (0.31 to 0.79)

  2. 0.41 (0.33 to 0.51)

0.64
  1. 3

  2. 14

  1. 0.26 (0.06 to 1.20)

  2. 0.67 (0.45 to 0.99)

0.88

Blinding

  1. Yes

  2. No

  1. 5

  2. 14

  1. 0.35 (0.25 to 0.48)

  2. 0.47 (0.37 to 0.59)

0.18
  1. 5

  2. 12

  1. 0.62 (0.39 to 0.98)

  2. 0.65 (0.33 to 1.27)

0.97

Intention to treat

  1. Yes

  2. No

  1. 10

  2. 9

  1. 0.38 (0.30 to 0.48)

  2. 0.47 (0.33 to 0.68)

0.37
  1. 9

  2. 8

  1. 0.62 (0.40 to 0.98)

  2. 0.65 (0.32 to 1.29)

0.57
CMV disease in patients stratified by antibody status

Subgroup analysis revealed that treatment efficacy in CMV disease did not vary significantly according to recipient serostatus. Medication significantly reduced the risk of CMV disease (Analysis 1.2.1 (13 studies, 1348 participants): RR 0.34, 95% CI 0.24 to 0.50; I² = 24%) in CMV positive recipients (donor positive or negative). Medication significantly reduced the risk of CMV disease (Analysis 1.2.2 (10 studies, 423 participants): RR 0.52, 95% CI 0.37 to 0.73; I² = 27%) in CMV negative recipients of CMV positive organs.

Subgroup analysis showed that treatment efficacy did not vary in CMV positive recipients if they received a CMV positive organ (Analysis 1.2.4 (5 studies, 276 participants): RR 0.19, 95% CI 0.09 to 0.37; I² = 0%) or CMV negative organ (Analysis 1.2.5 (5 studies, 160 participants): RR 0.32, 95% CI 0.11 to 0.95; I² = 0%).

Insufficient data (Analysis 1.2.3; 4 studies, 38 participants, 2 events) were available to determine the efficacy of prophylaxis on CMV disease in CMV negative recipients of CMV negative donors.

CMV disease in all patients stratified by antiviral medication

The treatment efficacy did not vary according to antiviral medication used on subgroup analysis. When analysed separately aciclovir (Analysis 1.3.1 (6 studies, 421 participants): RR 0.45, 95% CI 0.29 to 0.69; I² = 8%), ganciclovir (Analysis 1.3.2 (11 studies, 917 participants): RR 0.44, 95% CI 0.34 to 0.58; I² = 23%) and valaciclovir (Analysis 1.3.3 (2 studies, 643 participants): RR 0.30, 95% CI 0.19 to 0.49; I² = 0%) significantly reduced the risk for CMV disease compared with placebo or no treatment.

CMV disease in all patients stratified by transplanted organ

The treatment efficacy on CMV disease did not vary according to organ transplanted. Prophylaxis was effective in reducing the risk of CMV disease in kidney (Analysis 1.4.1 (11 studies, 1132 participants): RR 0.42, 95% CI 0.31 to 0.57; I² = 27%), liver (Analysis 1.4.2 (5 studies, 616 participants): RR 0.49, 95% CI 0.29 to 0.84; I² = 57%) and heart transplant recipients (Analysis 1.4.3 (3 studies, 232 participants): RR 0.39, 95% CI 0.25 to 0.63; I² = 0%).

CMV disease in ganciclovir treated patients stratified by treatment duration

In ganciclovir studies, duration of treatment was arbitrarily divided into fewer than six weeks and six weeks or more. There was no difference in treatment efficacy (Analysis 1.5). Effect of duration could not be assessed for other medications, which were generally administered for three months.

CMV disease in patients stratified for the use of antilymphocyte antibody

Subgroup analysis showed no difference in treatment efficacy against CMV disease if the immunosuppressive regimen did (Analysis 1.6.1 (11 studies, 666 participants): RR 0.43, 95% CI 0.33 to 0.55; I² = 0%) or did not (Analysis 1.7.1 (6 studies, 649 participants): RR 0.47, 95% CI 0.29 to 0.76; I² = 47%) include an antilymphocyte antibody given during prophylaxis for induction or rejection.

CMV-related death or other causes

In seven studies that reported the number of deaths due to CMV disease, the average mortality rates in the placebo/no treatment arms due to CMV disease and to non-CMV causes were 2.3% (range 0.3% to 7.4%) and 5.7% (0% to 15.6%) respectively. Prophylaxis significantly reduced the risk of death due to CMV disease (Analysis 1.8.1 (7 studies, 1300 patients): RR 0.26, 95% CI 0.08 to 0.78; I² = 0%) but not the risk from non-CMV causes (Analysis 1.8.2 (7 studies, 1300 patients): RR 0.71, 95% CI 0.44 to 1.17; I² = 0%).

All-cause mortality

The average all-cause mortality rate reported at one year or less post-transplant in the placebo/no treatment arms of all studies was 7.1% (range 0% to 37%). Prophylaxis significantly reduced all-cause mortality (Analysis 1.9 (17 studies, 1838 participants): RR 0.63, 95% CI 0.43 to 0.92; I² = 0%).

Figure 4 shows the cumulative meta-analyses demonstrating change over time for all-cause mortality. While the relative risk remained stable, the confidence intervals narrowed progressively with evidence for a significant reduction in all-cause mortality becoming evident with the addition of the Lowance 1999 Kidney study.

Figure 4.

All-cause mortality cumulative meta-analysis showing change over time

Subgroup analyses for all-cause mortality

Subgroup analyses according to CMV status, antiviral medications, organ transplanted, treatment duration, use of antilymphocyte therapy, time to outcome assessment, study quality and other aspects of study design did not demonstrate any differences in all-cause mortality. Multivariate meta-regression showed no significant difference in all-cause mortality after allowing for potential confounding or effect-modification by prophylactic drug used, organ transplanted or recipient serostatus in CMV positive recipients and CMV negative recipients of CMV positive donors. (See Table 1).

All-cause mortality stratified by CMV status

No differences in all-cause mortality were seen with CMV positive recipients (Analysis 1.10.1 (7 studies, 738 participants): RR 0.59, 95% CI 0.30 to 1.18; I² = 2%) or CMV negative recipients of CMV positive organs (Analysis 1.10.2 (4 studies, 288 participants): RR 1.42 95% CI 0.44 to 4.66; I² = 0%) on subgroup analysis. Data were not available to determine if the effects of antiviral medications on all-cause mortality differed between CMV positive recipients of CMV negative and CMV positive recipients of CMV positive organs.

All-cause mortality stratified by transplanted organ

All-cause mortality was reduced (Analysis 1.11 (17 studies, 1838 participants): RR 0.63, 95% CI 0.43 to 0.92; I² = 0%). However, the reduction could not be demonstrated for individual organs because of the small numbers of events and patients for individual organs.

All-cause mortality in ganciclovir treated patients stratified by treatment duration

There was no difference in all-cause mortality among studies evaluating ganciclovir for six weeks or less or more than six weeks (Analysis 1.12).

All-cause mortality in studies stratified according to use of antilymphocyte therapy

There was no difference in all-cause mortality whether or not antibody therapy was administered (Analysis 1.6.2; Analysis 1.7.2).

Additional outcomes

For graft loss, acute rejection, invasive fungal infection and post-transplant lymphoproliferative disease (PTLD) there was no significant difference between antiviral prophylaxis and placebo or no treatment (Analysis 1.7.1; Analysis 1.13.2; Analysis 1.13.4; Analysis 1.13.6). The risk of acute rejection did not differ on subgroup analysis between studies using biopsy diagnosis (Analysis 1.14.1 (5 studies, 827 participants): RR 0.97, 95% CI 0.71 to 1.32; I² = 62%) and those using clinical criteria (Analysis 1.14.2 (8 studies, 599 participants): RR 0.91, 95% CI 0.76 to 1.08; I² = 14%). In one study using valaciclovir with subgroups pre-specified according to CMV serostatus, prophylaxis significantly reduced the risk of acute rejection in CMV negative recipients of CMV positive kidneys (Lowance 1999 Kidney) (Analysis 1.15.1 (208 participants): RR 0.51, 95% CI 0.35 to 0.74) compared with CMV positive recipients (Analysis 1.15.2 (408 participants): RR 0.84, 95% CI 0.63 to 1.10) (test of interaction χ² = 4.33; P = 0.04). This difference is responsible for the heterogeneity demonstrated between valaciclovir studies for acute rejection (Analysis 1.15.3 (2 studies, 643 participants): RR 0.81, 95% CI 0.55 to 1.19; I² = 85%).

Prophylaxis with aciclovir, ganciclovir or valaciclovir reduced the risk for clinical disease caused by herpes simplex and herpes zoster (Analysis 1.13.3 (9 studies, 1483 participants): RR 0.27, 95% CI 0.19 to 0.40; I² = 27%). Combining the studies of different medications showed that bacterial (Analysis 1.13.5 (3 studies, 175 participants): RR 0.65, 95% CI 0.44 to 0.96; I² = 0%) and protozoal infections (Analysis 1.13.7 (2 studies, 114 participants): RR 0.31, 95% CI 0.10 to 0.99; I² = 0%) were significantly reduced by prophylaxis.

There were 16 studies that reported data on adverse effects of medications. Except for six placebo-controlled studies, we could not determine baseline adjusted effects of medications on leucopenia, kidney function and neurological dysfunction as the numbers of patients with these abnormalities were not reported for the no treatment groups. In placebo-controlled studies, valaciclovir significantly increased the risk for hallucinations (8.5% compared with 0.97%) (Analysis 1.16.9 (1 study, 616 participants): RR 8.78, 95% CI 2.69 to 28.71). There was no significant difference in neurological dysfunction with aciclovir (Analysis 1.16.3). No significant differences were identified for leucopenia (Analysis 1.16.1; Analysis 1.16.4; Analysis 1.16.7) or reduced kidney function (Analysis 1.16.2; Analysis 1.16.5; Analysis 1.16.8) with any medication (See Table 2).

Table 2. Summary of outcomes for antiviral medication versus placebo/no treatment
  1. ªPlacebo-controlled RCTs only.
    ^Heterogeneity of study results present.

OutcomeAciclovir
Studies; RR (95% CI)
Ganciclovir
Studies; RR (95% CI)
Valaciclovir
Studies; RR (95% CI)
All medications
Studies; RR (95% CI)
Acute rejection4; 1.03 (0.78 to 1.36)7; 0.92 (0.70 to 1.21)2; 0.81 (0.51 to 1.28)^13; 0.90 (0.78 to 1.17)
Graft loss4; 0.77 (0.35 to 1.68)6; 0.73 (0.41 to 1.28)No data10; 0.74 (0.47 to 1.17)
Herpes simplex or zoster infections3; 0.30 (0.14 to 0.62)4; 0.25 (0.08 to 0.78)2; 0.28 (0.20 to 0.40)9; 0.27 (0.19 to 0.40)
Post-transplant lymphoproliferative disease1; 2.90 (0.12 to 68.2)1; 0.34 (0.01 to 8.33)No data2; 1.01 (0.11 to 9.51)
Bacterial infections1; 0.67 (0.33 to 1.38)1; 0.72 (0.44 to 1.17)1; 0.27 (0.07 to 1.05)3; 0.65 (0.44 to 0.96)
Fungal infections1; 1.30 (0.31 to 5.39)2; 0.28 (0.07 to 1.12)No data3; 0.58 (0.19 to 1.73)
Protozoal infectionsNo data2; 0.31 (0.01 to 0.99)No data2; 0.31 (0.01 to 0.99)
LeucopeniaªNo data3; 0.99 (0.37 to 2.65)1; 1.05 (0.62 to 1.78) 
Creatinine > 200 µmol/Lª2; 1.14 (0.27 to 4.70)3; 2.36 (0.91 to 6.15)No data 
Hallucinationsª1; 10.6 (0.62 to 183.3)1; 1.59 (0.98 to 2.58)1; 8.78 (2.69 to 28.7) 
Subgroup analyses by methodological quality for CMV disease and all-cause mortality

Subgroup analysis, stratifying studies by methodological quality and aspects of study design, specified a priori, showed that treatment efficacy to reduce CMV disease and all-cause mortality did not vary significantly among studies.

  • Study publication date: Studies were arbitrarily divided into those published before 1997 and those published in or after 1997. There was no difference in treatment efficacy.

  • Study quality: Studies were divided according to quality assessment (adequate allocation concealment or other, blinding or no blinding, intention to treat analysis carried out or not). On subgroup analysis, no differences in treatment efficacy for CMV disease or all-cause mortality were detected for allocation concealment (Analysis 2.1; Analysis 3.1) blinding (Analysis 2.2; Analysis 3.2) or intention-to-treat analysis (Analysis 2.3; Analysis 3.3).

  • Time of outcome assessment: There was no difference in treatment efficacy for CMV disease and all-cause mortality if outcome was assessed at three to six months or nine to 12 months (Analysis 2.4; Analysis 3.4).

Ganciclovir versus aciclovir

Eight studies compared ganciclovir with aciclovir (Badley 1997 Liver; Duncan 1993 Lung; Flechner 1998 Kidney; Martin 1994 Liver; Nakazato 1993 Liver; Rubin 2002 All; Winston 1995 Liver; Winston 2003 Liver).

CMV disease and CMV infection

In head-to-head studies, ganciclovir was more effective than aciclovir in preventing CMV disease in all recipients (Analysis 4.1.1 (7 studies, 1113 participants): RR 0.37, 95% CI 0.23 to 0.60; I² = 33%), in CMV positive recipients (Analysis 4.2.1 (5 studies, 722 participants): RR 0.27, 95% CI 0.13 to 0.55; I² = 7%) and in CMV negative recipients of CMV positive organs (Analysis 4.3.1 (5 studies, 246 participants): RR 0.64, 95% CI 0.41 to 0.99; I² = 0%). There were insufficient data in CMV negative recipients of CMV negative donors to determine if a difference in efficacy exists (Analysis 4.4).

On subgroup analysis, no differences in efficacy could be demonstrated between studies in which the participants received ganciclovir for three months (Analysis 4.1.5 (4 studies, 703 participants): RR 0.28, 95% CI 0.09 to 0.82; I² = 62%) and those in which the participants received ganciclovir followed by aciclovir (Analysis 4.1.6 (3 studies, 410 participants): RR 0.38, 95% CI 0.22 to 0.64; I² = 0%). Subgroup analysis demonstrated the efficacy of antiviral medication was not dependent on the organ transplanted for either CMV disease (Analysis 4.5.1; Analysis 4.5.2; Analysis 4.5.3) or CMV infection (Analysis 4.5.4; Analysis 4.5.5; Analysis 4.5.6).

Ganciclovir was more effective than aciclovir in reducing CMV infection (Analysis 4.1.4 (6 studies, 815 participants): RR 0.44; 95% CI 0.28 to 0.67; I² = 73%) in all recipients and in CMV positive recipients (Analysis 4.2.2 (5 studies, 522 participants): RR 0.30, 95% CI 0.16 to 0.58; I² = 70%) but not in CMV negative recipients of CMV positive organs (Analysis 4.3.4 (4 studies, 228 participants): RR 0.63, 95% CI 0.36 to 1.09; I² = 58%) but there was significant heterogeneity among the studies.

All-cause mortality

There were no significant differences in the risk of death due to CMV disease (Analysis 4.6.1 (6 studies, 832 participants): RR 0.33, 95% CI 0.07 to 1.58; I² = 0%) or all-cause mortality (Analysis 4.6.2 (8 studies, 1138 participants): RR 1.13, 95% CI 0.82 to 1.58; I² = 0%).

Additional outcomes

No significant differences were reported for acute rejection (Analysis 4.7.1); graft loss (Analysis 4.7.2); other viral infections (Analysis 4.7.3); fungal infections (Analysis 4.7.4); bacterial infections (Analysis 4.7.5); protozoal infections (Analysis 4.7.6); or obliterative bronchiolitis in lung transplant recipients (Analysis 4.7.7). Three studies or fewer provided outcomes for graft loss, obliterative bronchiolitis and for opportunistic infections other than other viral infections.

Leucopenia was significantly more common with ganciclovir compared with aciclovir (Analysis 4.7.8 (6 studies, 955 participants): RR 3.28, 95% CI 1.48 to 7.25; I² = 0%) but no significant differences were demonstrated for kidney (Analysis 4.7.9) or neurological dysfunction (Analysis 4.7.10).

Ganciclovir/aciclovir versus ganciclovir

One study (Green 1997 Liver) compared ganciclovir given for 14 days followed by aciclovir to one year with ganciclovir for 14 days in 48 children, who had received liver transplants. No significant differences in efficacy were demonstrated for CMV disease (Analysis 5.1.1), CMV infection (Analysis 5.1.2), all-cause mortality (Analysis 5.2.1) or Epstein-Barr virus infections (Analysis 5.3.1).

Valganciclovir versus ganciclovir

One study (Paya 2004 All) compared valganciclovir with ganciclovir in CMV negative recipients of CMV positive organs and included patients receiving kidney, liver, heart and combined kidney-pancreas transplants.

CMV disease and CMV infection

Valganciclovir and ganciclovir were not significantly different in the prevention of CMV disease at six months (Analysis 6.1.1) or one year post-transplant (Analysis 6.1.2). Similarly there were no significant differences at six months and one year in the prevention of CMV syndrome (Analysis 6.1.3; Analysis 6.1.4) and CMV invasive organ disease (Analysis 6.1.5; Analysis 6.1.6). Subgroup analysis showed that, at six months, valganciclovir was significantly more effective than ganciclovir in kidney transplant recipients (Analysis 6.1.8 (120 participants): RR 0.27, 95% CI 0.01 to 0.75) compared with liver, heart or kidney-pancreas transplant recipients (Analysis 6.1.7; Analysis 6.1.9; Analysis 6.1.10) (test of interaction Chi² = 6.34; P = 0.01).

There were no significant differences at six months (Analysis 6.1.11) and one year (Analysis 6.1.12) in the prevention of CMV infection.

All-cause mortality

No significant differences were detected between medications in death due to CMV disease (Analysis 6.2.1) or all-cause mortality (Analysis 6.2.2).

Additional outcomes

No significant differences were detected in acute rejection, graft loss and opportunistic infections (Analysis 6.3.1; Analysis 6.3.2; Analysis 6.3.3). Neutrophil counts below 1000/mm³ occurred in 13% of patients treated with valganciclovir compared with 8% treated with ganciclovir but the difference was not significant (Analysis 6.3.7). No differences were detected in cessation of medications due to neutropenia, anaemia, thrombocytopenia or tremor (Analysis 6.3.4; Analysis 6.3.5; Analysis 6.3.6; Analysis 6.3.7; Analysis 6.3.8).

Valaciclovir versus ganciclovir/valganciclovir

Three studies compared valaciclovir with ganciclovir (Pavlopoulou 2005 Kidney; Reischig 2005 Kidney) or with valganciclovir (2VAL Study 2010 Kidney) in kidney transplant recipients.

CMV disease and CMV infection

The risk of CMV disease (Analysis 7.1.1) and CMV infection (Analysis 7.1.2) did not differ significantly with valaciclovir compared with ganciclovir or valganciclovir prophylaxis. There was no significant difference in the risk of CMV disease (Analysis 7.1.3) and CMV infection (Analysis 7.1.4) in CMV positive recipients of CMV positive or negative transplants or of the risk of CMV disease (Analysis 7.1.5) and CMV infection (Analysis 7.1.6) in CMV negative recipients of CMV positive organs.

All-cause mortality

No significant differences were detected in all-cause mortality (Analysis 7.2.1).

Additional outcomes

The risk of acute rejection did not differ significantly with valaciclovir compared with ganciclovir (Analysis 7.3.1 (3 studies, 188 participants): RR 0.91; 95% CI 0.22 to 3.73; I² = 64%). However, there was significant heterogeneity among the three studies with Reischig 2005 Kidney reporting a significantly reduced risk for acute rejection with valaciclovir (seen in participants with delayed graft function), while 2VAL Study 2010 Kidney showed a trend towards a higher risk of rejection with valaciclovir. No difference in the risk of graft loss was detected (Analysis 7.3.3).

No differences were detected in the risk of leucopenia, thrombocytopenia, anaemia, neurological dysfunction or need to reduce or cease study medications (Analysis 7.3.3; Analysis 7.3.4; Analysis 7.3.5; Analysis 7.3.6; Analysis 7.3.7). No differences were detected in the risk for other herpes infections (Analysis 7.3.8). Non-viral infections were increased in patients treated with valaciclovir in one study (Analysis 7.3.9 (83 participants): RR 0.59, 95% CI 0.44 to 0.80) due to the increase in urinary tract infections in that group.

Kidney function

Kidney function at the end of the study did not differ significantly with valaciclovir compared with ganciclovir or valganciclovir (Analysis 7.4.1; Analysis 7.4.2).

Prophylaxis with different regimens of ganciclovir

Hertz 1998 Heart/lung compared daily with thrice weekly IV ganciclovir in heart-lung transplant recipients. Winston 2004 Liver and Nafar 2005 Kidney compared oral with IV ganciclovir.

Daily versus thrice weekly ganciclovir

No significant differences were detected in CMV disease, CMV syndrome, CMV invasive tissue disease or CMV infection (Analysis 8.1.1; Analysis 8.1.2; Analysis 8.1.3; Analysis 8.1.4). In addition, no differences in all-cause mortality and death due to CMV disease (Analysis 8.1.5; Analysis 8.1.6) or in bacteraemia, bronchiolitis obliterans or leucopenia (Analysis 8.1.7; Analysis 8.1.8; Analysis 8.1.9) were detected.

Oral versus IV ganciclovir

No significant differences were detected in CMV disease, CMV syndrome, CMV invasive tissue disease or CMV infection (Analysis 8.2.1; Analysis 8.2.2; Analysis 8.2.3; Analysis 8.2.4). In addition, no differences in all-cause mortality, acute rejection or graft loss (Analysis 8.2.5; Analysis 8.2.6; Analysis 8.2.7) or in leucopenia and the need to cease medications due to leucopenia (Analysis 8.2.8; Analysis 8.2.9) were detected.

Prophylaxis with extended durations of valganciclovir

Two studies compared extended durations of valganciclovir. One study compared 200 days with 100 days in kidney transplant recipients (IMPACT 2010 Kidney) and the other study compared one year with three months in lung transplant recipients (Palmer 2010 Lung). Data included in meta-analyses from Palmer 2010 Lung were taken from percentages reported in the study as the authors were not able to provide the original data.

CMV disease and CMV infection

The risk of CMV disease was significantly reduced at the end of treatment (Analysis 9.1.1 (2 studies, 454 participants): RR 0.20, 95% CI 0.12 to 0.35; I² = 0%), at 9 months (Analysis 9.1.2 (1 study, 318 participants): RR 0.39, 95% CI 0.25 to 0.60), 12 months (Analysis 9.1.3 (1 study, 318 participants): RR 0.44, 95% CI 0.29 to 0.66) and 24 months (Analysis 9.1.4 (1 study, 318 participants): RR 0.55, 95% CI 0.38 to 0.79). The number of patients with CMV syndrome (Analysis 9.2 (2 studies, 454 participants): RR 0.27, 95% CI 0.10 to 0.71/; I² = 12%) was also significantly reduced. The risk for CMV invasive disease was higher in lung transplant recipients compared with kidney transplant recipients at 12 months so there was considerable heterogeneity in the analysis (Analysis 9.3.1 (2 studies, 454 participants): RR 0.17, 95% CI 0.03 to 1.34; I² = 44%). There were few episodes of CMV invasive disease in kidney transplant recipients and the numbers did not differ at 24 months (Analysis 9.3.2).

The risk of CMV infection was significantly reduced at the end of treatment (Analysis 9.4.1 (2 studies, 454 participants): RR 0.27, 95% CI 0.10 to 0.71; I² = 82%), at 9 months (Analysis 9.4.2 (1 study, 318 participants): RR 0.27, 95% CI 0.10 to 0.71) and at 12 months (Analysis 9.4.3 (1 study, 318 participants): RR 0.73, 95% CI 0.57 to 0.95).

Other outcomes

There were no significant differences in all-cause mortality at 12 (Analysis 9.5.1) and 24 months (Analysis 9.5.2), in graft loss at 12 (Analysis 9.6.1) and 24 months (Analysis 9.6.2), in biopsy proven acute rejection at < 100 days (Analysis 9.7.1), 12 months (Analysis 9.7.2) and 24 months (Analysis 9.7.3) and in post-transplant diabetes mellitus (Analysis 9.8.2). There was considerable heterogeneity (I² = 87%) in the analysis of opportunistic infections (Analysis 9.8.1 (2 studies, 454 participants): RR 0.71, 95% CI 0.33 to 1.57) since IMPACT 2010 Kidney reported that opportunistic infections were significantly less common among patients treated with extended duration valganciclovir while Palmer 2010 Lung found no difference (Analysis 9.8.1 (318 participants): RR 0.48, 95% CI 0.30 to 0.77).

Adverse effects

Total treatment related adverse effects (Analysis 9.9.1) and serious treatment related adverse effects (Analysis 9.9.2) did not differ significantly between treatment groups. Leucopenia was significantly more common (Analysis 9.9.3 (1 study, 320 participants): RD 0.12, 95% CI 0.01 to 0.22) and significantly more likely to result in treatment termination (Analysis 9.9.4 (1 study, 320 participants): RD 0.04, 95% CI 0.00 to 0.07) in patients treated for 200 days compared with those treated for 100 days in the IMPACT 2010 Kidney. Termination for any treatment-related adverse effect did not differ significantly in the Palmer 2010 Lung study (Analysis 9.9.5). While the number of hospitalisations did not differ overall or for all adverse effects (Analysis 9.9.7) among treatment groups, there were significantly fewer hospitalisations for CMV disease in patients treated for 200 days (Analysis 9.9.6 (1 study, 418 total hospitalisations): RD -0.10, 95% CI -0.17 to -0.04) in the IMPACT 2010 Kidney. There was no significant increase in CMV mutants, which confer ganciclovir resistance, in participants with positive viral load who were treated for an extended duration compared with those treated for 100 days or three months (Analysis 9.9.8).

Discussion

Summary of main results

Antiviral agents compared with placebo/no specific treatment

This systematic review found that the antiviral agents, ganciclovir, valaciclovir and aciclovir, improve outcomes for solid organ transplant recipients far beyond the primary indication for use. In addition to reducing the risk of CMV disease by 60%, these agents reduced all-cause mortality by 40%, predominantly due to reduced mortality from CMV disease, as well as reducing clinical disease caused by herpes simplex and herpes zoster (70%), bacterial infections (35%), and protozoal infections (70%). The relative benefits of aciclovir, ganciclovir and valaciclovir in relation to CMV disease and mortality appeared to be consistent among recipients of heart, kidney and liver transplants. These benefits occurred in both CMV positive recipients and CMV negative recipients of CMV positive organs, irrespective of whether immunosuppression included antilymphocyte antibody therapy, and were not dependent on the time of outcome assessment. Although there were no placebo-controlled RCTs of valganciclovir, a study (Paya 2004 All) comparing valganciclovir (the prodrug of ganciclovir) and ganciclovir demonstrated no significant differences in the risk for CMV disease, all-cause mortality and other outcomes, indicating that outcomes demonstrated in this systematic review in placebo/no treatment studies can be extrapolated to valganciclovir.

There was no clear reduction in graft loss or acute rejection, although a small but clinically important benefit has not been excluded. The summary relative risk for both outcomes favours antiviral agents but the 95% confidence intervals were relatively wide and consistent with there being no effect. The exception was in a predefined subgroup in a single study (Lowance 1999 Kidney) in which CMV prophylaxis reduced the risk for biopsy-proven acute rejection in CMV negative recipients of CMV positive kidney transplants by 50%.

Based on data from a single large study (Lowance 1999 Kidney) valaciclovir significantly increased the risk for hallucinations. There was no significant increase in adverse effects with aciclovir or ganciclovir, although the 95% CIs were wide. Very few studies adequately reported harms so that significant differences in adverse effects between medication and placebo could be excluded. It is possible that other differences in side effect profiles exist between agents but have not been demonstrated.

Relative efficacy of antiviral medications

Having demonstrated that antiviral agents as a drug class reduce all-cause mortality and CMV disease, we then sought to determine which antiviral regimen was the most beneficial. Indirect comparisons demonstrated no difference between antiviral agents administered. In head-to-head studies ganciclovir was significantly more effective than aciclovir in preventing CMV disease, demonstrating the importance of assessing the comparative effects of drugs in direct comparison studies. This difference may be explained by differences in duration of therapy in the indirect studies. Aciclovir was administered for 84 days or more but ganciclovir was given for shorter durations (9 to 42 days) in seven of the 11 included ganciclovir studies. Hence, agent and duration was evaluated rather than agent alone, as in direct comparison studies.

One large study (Paya 2004 All) demonstrated no significant difference in efficacy between ganciclovir and its prodrug, valganciclovir. Although three small studies demonstrated no difference in efficacy to prevent CMV disease among ganciclovir or valganciclovir and valaciclovir (2VAL Study 2010 Kidney; Pavlopoulou 2005 Kidney; Reischig 2005 Kidney), the wide confidence intervals of the summary estimate (RR 0.74, 95% CI 0.15 to 3.75) indicate that a significant difference in efficacy cannot be excluded. Based on existing study data, aciclovir is inferior to ganciclovir, and no clear superiority has been demonstrated between ganciclovir and valganciclovir or between valaciclovir and ganciclovir/valganciclovir.

Prophylaxis with extended durations of valganciclovir

Extended prophylaxis with valganciclovir resulted in significant reductions in the risks of CMV disease, CMV infection and opportunistic infections but no significant differences in other outcomes (acute rejection, all-cause mortality, graft loss). Leucopenia was more common with extended duration of prophylaxis, but hospitalisations due to CMV disease were reduced.

Overall completeness and applicability of evidence

Antiviral agents compared with placebo/no specific treatment

Our major findings, that CMV antiviral prophylaxis prevents CMV disease and all-cause mortality, irrespective of organ transplanted and CMV serostatus, are strengthened by two features of the data; the consistency of these findings across all studies and the finding that almost all eligible studies reported both major outcomes of interest (lack of outcome reporting bias). We identified 19 eligible studies and the summary estimate favours antiviral medication for the outcome 'prevention of CMV disease' in 18 studies. Similarly, 17 studies contributed data to the all-cause mortality outcome. With fewer events, the play of chance would be expected to be greater, but only two studies (Macdonald 1995 Heart; Merigan 1992 Heart) had point estimates suggesting increased mortality from CMV prophylaxis. Unlike the outcome of CMV disease, no individual study demonstrated a significant reduction in all-cause mortality with antiviral medication. This was evident only from the meta-analytic estimate. The overall I² was 12.6% for CMV disease and 0% for all-cause mortality demonstrating very low heterogeneity beyond chance, despite the clear differences in patient groups (Characteristics of included studies). Supporting this contention, as shown in Table 1, no pre-defined potential source of variability for the effects of antiviral medication was significant, including standard quality items for study conduct and reporting such as allocation concealment, blinding and intention-to-treat. We cannot exclude a difference in the magnitude of the effect of antiviral medication in solid organ transplant recipients. However, any difference is likely to be clinically unimportant since data from 19 studies and about 2000 patients were insufficient to demonstrate any difference. In addition, the remarkable consistency in results across all studies suggests any undetected difference would be in magnitude, and not direction of effect.

The data were relatively sparse in four areas, and further research is still needed. For the outcome of all-cause mortality in heart transplant recipients, there are few relevant studies (2), patients (205) and events (4) making the effects of antiviral medications on heart transplant recipients very uncertain. Both studies had higher death rates in the active arms but the 95% confidence intervals were very wide, results are consistent with other patient groups (liver and kidney), and the likely pathway for benefit - reduction in CMV disease - is evident in this patient group.

Second, there were very scant data in the seronegative donor to seronegative recipient group, even though this group is frequently given antiviral agents to prevent CMV disease (Baliga 2004). These patients are almost exclusively not enrolled in studies, because of low event rates. However, there are no studies examining the efficacy of antiviral agents to prevent de novo CMV disease in such CMV seronegative patients.

Third, our conclusions on the other benefits of antiviral medications and the adverse effects of these drugs (Table 2) must be considered more cautiously for reasons of imprecision of summary estimates and that many eligible studies did not report these outcomes. Therefore, these results may be biased. The direction of bias cannot be determined without obtaining additional data from the authors regarding these outcomes.

Fourth, only one study specifically addressed children (Green 1997 Liver). This is despite that children commonly receive prophylaxis with antiviral agents since they are at a high risk of CMV disease because many are CMV seronegative and receive organs from CMV seropositive donors. Information on the efficacy of prophylaxis with antiviral agents from RCTs of adult transplant recipients has been extrapolated to children. Non randomised studies suggest valganciclovir is effective and tolerated in children (Camacho-Gonzalez 2011).

Relative efficacy of antiviral medications

The data clearly demonstrated that ganciclovir was superior to aciclovir in preventing CMV disease, and aciclovir is no longer used for prophylaxis. A single large study indicated no significant differences between oral ganciclovir and oral valganciclovir. Clinical practice data from this study have been extrapolated to indicate that oral valganciclovir can substitute for oral ganciclovir and valganciclovir is now generally the preferred agent for prophylaxis. Oral ganciclovir is no longer marketed.

Limited data (3 studies, 188 patients) meant that it remains unclear whether there are any differences in efficacy between valganciclovir/ganciclovir and valaciclovir in preventing CMV disease. The full results of all included patients are awaited for the 2VAL Study 2010 Kidney to determine whether any differences in efficacy exist between valganciclovir and valaciclovir. The available studies comparing valganciclovir/ganciclovir with valaciclovir have only enrolled kidney transplant recipients and it is unclear whether the data can be extrapolated to other transplanted organs.

Prophylaxis with extended durations of valganciclovir

Two studies in kidney (318 recipient CMV positive, donor CMV negative participants) and lung transplant recipients (136 donor CMV positive/recipient CMV negative; and donor CMV positive or negative/recipient CMV positive participants) have demonstrated that extended durations of prophylaxis with valganciclovir resulted in a lower risk of CMV disease and infection. Neither study identified an increase in CMV mutations resistant to therapy, but study numbers were likely to be too small to demonstrate any difference. Both studies reported few cases of CMV disease occurring after the end of the extended period of prophylaxis. Further data are required to demonstrate whether the benefits of extended prophylaxis in other organ transplants justify the increased costs and adverse effects.

Quality of the evidence

This review now contains 37 studies. Most studies, including those recently published, did not provide sufficient information to determine whether sequence generation and allocation concealment were at a low risk of bias (Figure 5; Figure 6). It is a matter of concern that there was no blinding of participants, investigators and outcome assessors in almost 75% of studies. The primary outcome of CMV syndrome is a clinical diagnosis supported by laboratory diagnosis of CMV infection and other information. Therefore, it is possible that CMV syndrome was misdiagnosed in some participants. Studies that lack adequate allocation concealment and blinding may overestimate treatment effects (Moher 1998; Schultz 1995).

Figure 5.

Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies

Figure 6.

Risk of bias summary: review authors' judgements about each risk of bias item for each included study

The overall quality of the evidence for studies comparing antiviral medications with placebo or no specific treatment was considered high for some outcomes (CMV disease, all-cause mortality, acute rejection, CMV disease in kidney transplant recipients). It was considered moderate for mortality due to CMV disease, CMV disease in liver or heart transplants and graft loss because of limited numbers of studies reporting these outcomes (Summary of findings for the main comparison).

The overall quality of the evidence for studies comparing ganciclovir and aciclovir was considered high for CMV disease in all patients and for acute rejection. It was considered moderate for all-cause mortality, mortality due to CMV disease and other viral infections, and low for other fungal infections and graft loss because of the limited number of events in the studies in which these outcomes were reported (Summary of findings 2).

The overall quality of the evidence for studies comparing ganciclovir/valganciclovir with aciclovir/valaciclovir was considered low because of the small number of studies with few participants (Summary of findings 3).

The overall quality of the evidence for studies comparing extended duration with three months of therapy was considered high for CMV disease, CMV syndrome, CMV infection and total adverse reactions. It was considered low for invasive CMV disease, acute rejection and opportunistic infections because of the heterogeneity between studies (Summary of findings 4).

Potential biases in the review process

The literature search was updated to July 2011. Although 29 additional reports of 13 studies, which had been included in previous versions of the review, were identified, these reports did not provide additional data for the review. It is possible that further reports of studies have been added to the Cochrane Renal Group's Specialised Register since the last search. Preliminary data from one study (2VAL Study 2010 Kidney) have been included in meta-analyses. It is possible that when full recruitment and follow-up are available, different results may be obtained.

About half the studies did not report all important outcomes so there is a risk of selection bias. In particular, there were limited data on death due to CMV disease, on graft loss and on other infections.

Agreements and disagreements with other studies or reviews

The results of this review confirm and expand the findings of three previous systematic reviews (Couchoud 1998b; Couchoud 1998a; Fiddian 2002; Gourishankar 2001), which included 12, 10 and 9 studies respectively comparing antiviral medications with placebo or no treatment for prevention of CMV disease. All found that prophylaxis reduced the risk for CMV disease in solid organ transplant recipients. One review (Couchoud 1998b; Couchoud 1998a) found no effect on mortality (10 studies; RR 0.69, 95% CI 0.41 to 1.18) and a second (Fiddian 2002), which included two studies using immunoglobulin and antiviral agents, found that prophylaxis with aciclovir or valaciclovir significantly reduced all-cause mortality (1321 patients; OR 0.60, 95% CI 0.40 to 0.90). Similarly, a more recent systematic review (Kalil 2005) including 11 studies, found that prophylaxis with antiviral medications compared with placebo or no specific treatment significantly reduced CMV disease, all-cause mortality and opportunistic infections with similar degrees of benefit to those found in our review, although inclusion criteria differed in the two reviews. Eight studies of prophylaxis included in our review were excluded from the analyses of universal prophylaxis in the review by Kalil 2005. The two reviews differed in that our review showed no significant reduction of acute rejection with antiviral prophylaxis but Kalil 2005 identified a significant reduction in acute rejection with treatment (OR 0.72, 95% CI 0.57 to 0.91) using a fixed-effect model for the analysis. However, there was some heterogeneity in the analyses of acute rejection in both reviews. Further analyses using a random-effects model identified that both reviews found no significant differences in the risk of acute rejection between antiviral therapy and placebo/no specific treatment. Both reviews found a significant reduction in acute rejection using a fixed-effect model.

Our systematic review differs from previous reviews in that comparisons of different antiviral medications were included so that conclusions on the comparative effects of agents can be made. In addition, our review included a detailed exploration of potential heterogeneity. The finding of a reduction in all-cause mortality is largely explained by a reduced mortality due to CMV disease, although a reduction in mortality due to other causes cannot be totally excluded. The latter is biologically plausible because CMV disease leads to an increase in other opportunistic infections in heart and liver transplant recipients (George 1997; Valentine 1999). This is suggestive of a mechanism whereby the prevention of CMV disease may prevent other infective complications that contribute to overall mortality.

Both prophylaxis and pre-emptive therapy significantly reduce CMV disease compared with placebo or no specific therapy in solid organ transplant recipients. However, the available evidence base for prevention of CMV disease with prophylaxis compared with placebo/no specific therapy (19 studies, 1981 participants) is large and of high quality (GRADE) compared with the low quality data (6 studies, 288 participants) supporting pre-emptive therapy (Owers 2013). Further studies are required to determine the relative efficacies, adverse effects and costs of pre-emptive therapy and prophylaxis because currently available data (7 studies, 753 participants), while showing no significant differences in efficacy though a lower risk of leucopenia with pre-emptive therapy, demonstrated considerable heterogeneity among studies thus limiting the applicability of these data to patient management.

Authors' conclusions

Implications for practice

This systematic review has shown that prophylaxis of CMV positive recipients and CMV negative recipients of CMV positive organs with antiviral medications given for three months post solid organ transplantation reduces the risk of CMV disease and all-cause mortality and may well reduce the risk of other opportunistic infections. What are the implications of this study to clinical practice? Previous treatment guidelines (Jassal 1998; Van der Bij 2001) recommended CMV prophylaxis for all recipients of solid organ transplants who received immunosuppression with antilymphocyte antibody products and for CMV negative recipients of CMV positive organs. In liver and heart transplant recipients, prophylaxis was also recommended for all CMV positive recipients of solid organ transplants because of the higher risk for CMV disease. Prophylaxis was not generally recommended for CMV positive kidney transplant recipients or for donor negative/recipient negative recipients (Jassal 1998) based on the low incidence of CMV disease in these groups. Our data suggested that these recommendations for use were too narrow because the benefits for patient survival and the constant relative benefits for CMV disease, irrespective of CMV serostatus, had not been recognised previously.

Recent guidelines recommend that all kidney transplant recipients except donor negative/recipient negative recipients should receive antiviral prophylaxis for at least three months post-transplant (KDIGO 2009). Similarly, guidelines from the AST Infectious Diseases Community of Practice (Humar 2009) recommend antiviral prophylaxis for both CMV seropositive recipients and for CMV seronegative recipients of CMV seropositive donors of any solid organ transplant. Consensus guidelines from the Infectious Disease Section of the Transplantation Society (Kotton 2010) recommended antiviral prophylaxis for CMV seronegative recipients of CMV positive donor organs. These guidelines considered that either prophylaxis or pre-emptive therapies could be used in CMV positive recipients but noted the lack of data on pre-emptive therapy in subpopulations including lung and small bowel transplants.

The absolute effects of antiviral medications on the prevention of CMV disease and all-cause mortality are shown quantitatively in groups of patients at different baseline risk for these outcomes (Table 3). The primary determinants for CMV disease are organ transplanted and serostatus whereas organ transplanted is the most important determinant for all-cause mortality. Table 3 shows that benefit exceeds harm for all but the lowest risk groups assuming equal importance of the outcomes. However, given that the clinical importance of all-cause mortality and CMV disease are significantly greater than the adverse effects of medications, most patients and clinicians, when provided with this information, are likely to use CMV prophylaxis with antiviral medications across all risk categories, except in the seronegative donor and recipient groups for whom there are few data.

Table 3. Effects of antiviral medication on CMV disease and all-cause mortality
  1. ¹Data from references.
    ²Calculated from summary estimates of RR (0.42 for prevention of CMV disease, 0.63 for all-cause mortality).
    ³Based on proportion of patients, treated with valaciclovir, who developed hallucinations.
    ªDonor positive or negative for CMV; recipient negative.
    ^Donor positive recipient negative for CMV.

Recipient groupWithout prophylaxis¹With prophylaxis²Number preventedNumber with harms³

CMV disease

  1. Kidneyª

  2. Kidneyª; liver^; heartª

  3. Liver, heartª; all^, antibody therapy included in immunosuppressive regimen

  1. 7/100

  2. 28/100

  3. 59/100

  1. 3/100

  2. 12/100

  3. 25/100

  1. 4/100

  2. 16/100

  3. 39/100

  1. 7/100

  2. 7/100

  3. 7/100

All-cause mortality

  1. Kidney

  2. Liver

  3. Heart or lung

  1. 6/100

  2. 20/100

  3. 24/100

  1. 4/100

  2. 13/100

  3. 15/100

  1. 2/100

  2. 7/100

  3. 9/100

  1. 7/100

  2. 7/100

  3. 7/100

Two RCTs (IMPACT 2010 Kidney; Palmer 2010 Lung) have now demonstrated that extended duration prophylaxis with valganciclovir in CMV seropositive donor/CMV negative recipients of kidney and lung transplants and seropositive recipients of lung transplants reduces the risk of CMV disease compared with three months of therapy suggesting that extended duration prophylaxis should be considered in patients at higher risk of CMV disease (Humar 2009).

Implications for research

There are no data from RCTs on the efficacy of prophylaxis compared with placebo in lung transplants and few data in heart transplants. However, such studies are no longer ethical based on the demonstration of efficacy in other organ transplants. Future studies may be required in the seronegative donor-recipient group depending on the prevalence of CMV disease in this group with newer and more potent immunosuppressive regimens. Further studies are required to determine optimum duration and dosage of medications in different organ transplants. Currently valganciclovir is most commonly used for prophylaxis. It remains possible that smaller doses than currently recommended may be effective for prophylaxis as demonstrated for IV ganciclovir (Hertz 1998 Heart/lung).

Further studies are required to evaluate the comparative effects, including harms, of antiviral medications in clinical use at present or in the future. More information is required on the efficacy of prophylaxis with different regimens of immunosuppressive regimens used for prevention and treatment of rejection in different organ transplants.

Overall, prophylaxis did not significantly reduce the risk for acute rejection or graft loss. Further information is required to determine whether prophylaxis can reduce the risk for rejection in particular groups of patients, whether it affects the number or severity of rejection episodes, and whether it reduces graft loss at time periods beyond one year.

Adequately powered and well-designed RCTs are required to determine the relative efficacies, adverse effects and costs of universal prophylaxis in comparison with pre-emptive therapies particularly in transplant populations at lower risk of CMV disease.

Acknowledgements

  • Mr Peter Barclay, Dr Cheryl Jones, Ms Kathy Kable and Ms Dushanythi Vimalachandra who contributed to the original iteration of this review (Hodson 2005b). They contributed to the design, quality assessment, data collection, entry, analysis and interpretation, and writing of the review.

  • This review has also been published in the Lancet (Hodson 2005c).

  • The authors would like to thank all authors who responded to our enquiries about their studies.

  • The authors wish to thank Ms Narelle Willis, Managing Editor of the Cochrane Renal Group, and Ms Ruth Mitchell and Ms Gail Higgins, Trials Search Co-ordinators of the Cochrane Renal Group, for their help with this study.

  • The authors wish to thank Dr Cécile Couchoud, who wrote the initial systematic review Cytomegalovirus prophylaxis with antiviral agents for solid organ transplantation (Couchoud 1998a).

Data and analyses

Download statistical data

Comparison 1. Antiviral prophylaxis versus placebo/no treatment
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 CMV disease and CMV infection in all treated patients19 Risk Ratio (M-H, Random, 95% CI)Subtotals only
1.1 All symptomatic CMV disease191981Risk Ratio (M-H, Random, 95% CI)0.42 [0.34, 0.52]
1.2 CMV syndrome111570Risk Ratio (M-H, Random, 95% CI)0.41 [0.29, 0.57]
1.3 CMV organ involvement121628Risk Ratio (M-H, Random, 95% CI)0.34 [0.21, 0.55]
1.4 Total CMV infection171786Risk Ratio (M-H, Random, 95% CI)0.61 [0.48, 0.77]
2 All symptomatic CMV disease stratified by antibody status17 Risk Ratio (M-H, Random, 95% CI)Subtotals only
2.1 CMV antibody +ve recipients131348Risk Ratio (M-H, Random, 95% CI)0.34 [0.24, 0.50]
2.2 CMV +ve donor / CMV -ve recipient10423Risk Ratio (M-H, Random, 95% CI)0.52 [0.37, 0.73]
2.3 CMV -ve donor / CMV -ve recipient438Risk Ratio (M-H, Random, 95% CI)1.0 [0.09, 11.03]
2.4 CMV +ve donor / CMV +ve recipient5276Risk Ratio (M-H, Random, 95% CI)0.19 [0.09, 0.37]
2.5 CMV -ve donor / CMV +ve recipient5160Risk Ratio (M-H, Random, 95% CI)0.32 [0.11, 0.95]
3 CMV disease in all patients by antiviral medication191981Risk Ratio (M-H, Random, 95% CI)0.42 [0.34, 0.52]
3.1 Aciclovir6421Risk Ratio (M-H, Random, 95% CI)0.45 [0.29, 0.69]
3.2 Ganciclovir11917Risk Ratio (M-H, Random, 95% CI)0.44 [0.34, 0.58]
3.3 Valaciclovir2643Risk Ratio (M-H, Random, 95% CI)0.30 [0.19, 0.49]
4 CMV disease for different organ transplants191980Risk Ratio (M-H, Random, 95% CI)0.44 [0.35, 0.55]
4.1 Kidney transplant recipients111132Risk Ratio (M-H, Random, 95% CI)0.42 [0.31, 0.57]
4.2 Liver transplant recipients5616Risk Ratio (M-H, Random, 95% CI)0.49 [0.29, 0.84]
4.3 Heart transplant recipients3232Risk Ratio (M-H, Random, 95% CI)0.39 [0.25, 0.63]
5 CMV disease and ganciclovir duration11 Risk Ratio (M-H, Random, 95% CI)Subtotals only
5.1 Six weeks or less7478Risk Ratio (M-H, Random, 95% CI)0.49 [0.36, 0.68]
5.2 More than 6 weeks4439Risk Ratio (M-H, Random, 95% CI)0.33 [0.21, 0.53]
6 ATG therapy and antiviral efficacy11 Risk Ratio (M-H, Random, 95% CI)Subtotals only
6.1 CMV disease in all treated patients11666Risk Ratio (M-H, Random, 95% CI)0.43 [0.33, 0.55]
6.2 All-cause mortality10643Risk Ratio (M-H, Random, 95% CI)0.82 [0.33, 2.02]
7 Immunosuppression without ATG induction and antiviral efficacy6 Risk Ratio (M-H, Random, 95% CI)Subtotals only
7.1 CMV disease in all treated patients6649Risk Ratio (M-H, Random, 95% CI)0.47 [0.29, 0.76]
7.2 All-cause mortality5529Risk Ratio (M-H, Random, 95% CI)0.63 [0.39, 1.00]
8 Mortality due to CMV disease or other causes7 Risk Ratio (M-H, Random, 95% CI)Subtotals only
8.1 CMV disease71300Risk Ratio (M-H, Random, 95% CI)0.26 [0.08, 0.78]
8.2 Other causes71300Risk Ratio (M-H, Random, 95% CI)0.71 [0.44, 1.17]
9 All-cause mortality according to antiviral medication171838Risk Ratio (M-H, Random, 95% CI)0.63 [0.43, 0.92]
9.1 Aciclovir5301Risk Ratio (M-H, Random, 95% CI)0.67 [0.38, 1.20]
9.2 Ganciclovir10894Risk Ratio (M-H, Random, 95% CI)0.69 [0.29, 1.65]
9.3 Valaciclovir2643Risk Ratio (M-H, Random, 95% CI)0.50 [0.22, 1.15]
10 All-cause mortality according to CMV status91026Risk Ratio (M-H, Random, 95% CI)0.74 [0.41, 1.32]
10.1 CMV +ve recipients7738Risk Ratio (M-H, Random, 95% CI)0.59 [0.30, 1.18]
10.2 CMV -ve recipients of CMV +ve organs4288Risk Ratio (M-H, Random, 95% CI)1.42 [0.44, 4.66]
11 All-cause mortality for different organ transplants171838Risk Ratio (M-H, Random, 95% CI)0.63 [0.43, 0.92]
11.1 Kidney transplant recipients101109Risk Ratio (M-H, Random, 95% CI)0.49 [0.24, 1.00]
11.2 Liver transplant patients4497Risk Ratio (M-H, Random, 95% CI)0.63 [0.39, 1.00]
11.3 Heart transplant recipients3232Risk Ratio (M-H, Random, 95% CI)1.82 [0.39, 8.51]
12 All-cause mortality and ganciclovir duration10 Risk Ratio (M-H, Random, 95% CI)Subtotals only
12.1 Six weeks or less6455Risk Ratio (M-H, Random, 95% CI)0.91 [0.17, 4.92]
12.2 More than 6 weeks4439Risk Ratio (M-H, Random, 95% CI)0.62 [0.30, 1.30]
13 Additional outcomes - all medications16 Risk Ratio (M-H, Random, 95% CI)Subtotals only
13.1 Graft loss10825Risk Ratio (M-H, Random, 95% CI)0.74 [0.47, 1.17]
13.2 Acute rejection131420Risk Ratio (M-H, Random, 95% CI)0.90 [0.78, 1.05]
13.3 Herpes simplex and H. zoster infection91483Risk Ratio (M-H, Random, 95% CI)0.27 [0.19, 0.40]
13.4 Invasive fungal infection3189Risk Ratio (M-H, Random, 95% CI)0.58 [0.19, 1.73]
13.5 Bacterial infection3174Risk Ratio (M-H, Random, 95% CI)0.65 [0.44, 0.96]
13.6 EBV-associated PTLD2359Risk Ratio (M-H, Random, 95% CI)1.01 [0.11, 9.51]
13.7 Protozoal infections2114Risk Ratio (M-H, Random, 95% CI)0.31 [0.10, 0.99]
14 Acute rejection according to method of diagnosis131420Risk Ratio (M-H, Random, 95% CI)0.90 [0.78, 1.05]
14.1 Biopsy-proven acute rejection5821Risk Ratio (M-H, Random, 95% CI)0.97 [0.71, 1.32]
14.2 Clinical diagnosis of acute rejection or method not stated8599Risk Ratio (M-H, Random, 95% CI)0.91 [0.76, 1.08]
15 Valaciclovir - additional outcomes2 Risk Ratio (M-H, Random, 95% CI)Subtotals only
15.1 Acute rejection in donor CMV +ve / recipient CMV -ve grafts1208Risk Ratio (M-H, Random, 95% CI)0.51 [0.35, 0.74]
15.2 Acute rejection in CMV +ve recipients1408Risk Ratio (M-H, Random, 95% CI)0.84 [0.63, 1.10]
15.3 Total with acute rejection2643Risk Ratio (M-H, Random, 95% CI)0.81 [0.55, 1.19]
16 Adverse effects6 Risk Ratio (M-H, Random, 95% CI)Subtotals only
16.1 Leucopenia with aciclovir00Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
16.2 Kidney dysfunction with aciclovir2159Risk Ratio (M-H, Random, 95% CI)1.14 [0.27, 4.70]
16.3 Neurological dysfunction with aciclovir155Risk Ratio (M-H, Random, 95% CI)10.62 [0.62, 183.26]
16.4 Leucopenia with ganciclovir3509Risk Ratio (M-H, Random, 95% CI)0.99 [0.37, 2.65]
16.5 Kidney dysfunction with ganciclovir3509Risk Ratio (M-H, Random, 95% CI)2.36 [0.91, 6.15]
16.6 Neurological dysfunction with ganciclovir3509Risk Ratio (M-H, Random, 95% CI)1.59 [0.98, 2.58]
16.7 Leucopenia with valaciclovir1616Risk Ratio (M-H, Random, 95% CI)1.05 [0.62, 1.78]
16.8 Kidney dysfunction with valaciclovir00Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
16.9 Neurological dysfunction with valaciclovir1616Risk Ratio (M-H, Random, 95% CI)8.78 [2.69, 28.71]
Analysis 1.1.

Comparison 1 Antiviral prophylaxis versus placebo/no treatment, Outcome 1 CMV disease and CMV infection in all treated patients.

Analysis 1.2.

Comparison 1 Antiviral prophylaxis versus placebo/no treatment, Outcome 2 All symptomatic CMV disease stratified by antibody status.

Analysis 1.3.

Comparison 1 Antiviral prophylaxis versus placebo/no treatment, Outcome 3 CMV disease in all patients by antiviral medication.

Analysis 1.4.

Comparison 1 Antiviral prophylaxis versus placebo/no treatment, Outcome 4 CMV disease for different organ transplants.

Analysis 1.5.

Comparison 1 Antiviral prophylaxis versus placebo/no treatment, Outcome 5 CMV disease and ganciclovir duration.

Analysis 1.6.

Comparison 1 Antiviral prophylaxis versus placebo/no treatment, Outcome 6 ATG therapy and antiviral efficacy.

Analysis 1.7.

Comparison 1 Antiviral prophylaxis versus placebo/no treatment, Outcome 7 Immunosuppression without ATG induction and antiviral efficacy.

Analysis 1.8.

Comparison 1 Antiviral prophylaxis versus placebo/no treatment, Outcome 8 Mortality due to CMV disease or other causes.

Analysis 1.9.

Comparison 1 Antiviral prophylaxis versus placebo/no treatment, Outcome 9 All-cause mortality according to antiviral medication.

Analysis 1.10.

Comparison 1 Antiviral prophylaxis versus placebo/no treatment, Outcome 10 All-cause mortality according to CMV status.

Analysis 1.11.

Comparison 1 Antiviral prophylaxis versus placebo/no treatment, Outcome 11 All-cause mortality for different organ transplants.

Analysis 1.12.

Comparison 1 Antiviral prophylaxis versus placebo/no treatment, Outcome 12 All-cause mortality and ganciclovir duration.

Analysis 1.13.

Comparison 1 Antiviral prophylaxis versus placebo/no treatment, Outcome 13 Additional outcomes - all medications.

Analysis 1.14.

Comparison 1 Antiviral prophylaxis versus placebo/no treatment, Outcome 14 Acute rejection according to method of diagnosis.

Analysis 1.15.

Comparison 1 Antiviral prophylaxis versus placebo/no treatment, Outcome 15 Valaciclovir - additional outcomes.

Analysis 1.16.

Comparison 1 Antiviral prophylaxis versus placebo/no treatment, Outcome 16 Adverse effects.

Comparison 2. Effect of methodological quality on CMV disease in studies of prophylaxis versus placebo/no treatment
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Allocation concealment19 Risk Ratio (M-H, Random, 95% CI)Subtotals only
1.1 Adequate4262Risk Ratio (M-H, Random, 95% CI)0.50 [0.31, 0.79]
1.2 Inadequate/unclear151719Risk Ratio (M-H, Random, 95% CI)0.41 [0.33, 0.51]
2 Blinding of participants/investigators19 Risk Ratio (M-H, Random, 95% CI)Subtotals only
2.1 Blinding51135Risk Ratio (M-H, Random, 95% CI)0.35 [0.25, 0.48]
2.2 No blinding14846Risk Ratio (M-H, Random, 95% CI)0.47 [0.37, 0.59]
3 Intention-to-treat analysis (ITT)19 Risk Ratio (M-H, Random, 95% CI)Subtotals only
3.1 ITT undertaken101569Risk Ratio (M-H, Random, 95% CI)0.38 [0.30, 0.48]
3.2 ITT not undertaken9412Risk Ratio (M-H, Random, 95% CI)0.47 [0.33, 0.68]
4 CMV disease by time of outcome assessment or trial publication date19 Risk Ratio (M-H, Random, 95% CI)Subtotals only
4.1 Outcome at 9-12 months81277Risk Ratio (M-H, Random, 95% CI)0.36 [0.22, 0.58]
4.2 Outcome at 3-6 months11704Risk Ratio (M-H, Random, 95% CI)0.46 [0.36, 0.58]
4.3 Trials published before 199712821Risk Ratio (M-H, Random, 95% CI)0.48 [0.37, 0.63]
4.4 Trials published in 1997 and later71160Risk Ratio (M-H, Random, 95% CI)0.32 [0.24, 0.44]
Analysis 2.1.

Comparison 2 Effect of methodological quality on CMV disease in studies of prophylaxis versus placebo/no treatment, Outcome 1 Allocation concealment.

Analysis 2.2.

Comparison 2 Effect of methodological quality on CMV disease in studies of prophylaxis versus placebo/no treatment, Outcome 2 Blinding of participants/investigators.

Analysis 2.3.

Comparison 2 Effect of methodological quality on CMV disease in studies of prophylaxis versus placebo/no treatment, Outcome 3 Intention-to-treat analysis (ITT).

Analysis 2.4.

Comparison 2 Effect of methodological quality on CMV disease in studies of prophylaxis versus placebo/no treatment, Outcome 4 CMV disease by time of outcome assessment or trial publication date.

Comparison 3. Effect of methodological quality on all-cause mortality in studies of prophylaxis versus placebo/no treatment
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Allocation concealment17 Risk Ratio (M-H, Random, 95% CI)Subtotals only
1.1 Adequate3142Risk Ratio (M-H, Random, 95% CI)0.26 [0.06, 1.20]
1.2 Inadequate/unclear141695Risk Ratio (M-H, Random, 95% CI)0.67 [0.45, 0.99]
2 Blinding of participants and investigators17 Risk Ratio (M-H, Random, 95% CI)Subtotals only
2.1 Blinding51135Risk Ratio (M-H, Random, 95% CI)0.62 [0.39, 0.98]
2.2 No blinding12702Risk Ratio (M-H, Random, 95% CI)0.65 [0.33, 1.27]
3 Intention-to-treat analysis (ITT)17 Risk Ratio (M-H, Random, 95% CI)Subtotals only
3.1 ITT undertaken91448Risk Ratio (M-H, Random, 95% CI)0.62 [0.40, 0.98]
3.2 ITT not undertaken8389Risk Ratio (M-H, Random, 95% CI)0.65 [0.32, 1.29]
4 All-cause mortality and time of outcome assessment or trial publication date17 Risk Ratio (M-H, Random, 95% CI)Subtotals only
4.1 Outcome at 9-12 months101370Risk Ratio (M-H, Random, 95% CI)0.63 [0.40, 0.97]
4.2 Outcome at 4-6 months7468Risk Ratio (M-H, Random, 95% CI)0.64 [0.31, 1.33]
4.3 Outcome in trials published before 199710678Risk Ratio (M-H, Random, 95% CI)0.71 [0.25, 2.08]
4.4 Outcome in trials published in 1997 or later71160Risk Ratio (M-H, Random, 95% CI)0.62 [0.41, 0.94]
Analysis 3.1.

Comparison 3 Effect of methodological quality on all-cause mortality in studies of prophylaxis versus placebo/no treatment, Outcome 1 Allocation concealment.

Analysis 3.2.

Comparison 3 Effect of methodological quality on all-cause mortality in studies of prophylaxis versus placebo/no treatment, Outcome 2 Blinding of participants and investigators.

Analysis 3.3.

Comparison 3 Effect of methodological quality on all-cause mortality in studies of prophylaxis versus placebo/no treatment, Outcome 3 Intention-to-treat analysis (ITT).

Analysis 3.4.

Comparison 3 Effect of methodological quality on all-cause mortality in studies of prophylaxis versus placebo/no treatment, Outcome 4 All-cause mortality and time of outcome assessment or trial publication date.

Comparison 4. Ganciclovir versus aciclovir
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 CMV disease and CMV infection in all treated patients8 Risk Ratio (M-H, Random, 95% CI)Subtotals only
1.1 CMV disease in all patients71113Risk Ratio (M-H, Random, 95% CI)0.37 [0.23, 0.60]
1.2 CMV organ involvement71034Risk Ratio (M-H, Random, 95% CI)0.28 [0.15, 0.49]
1.3 CMV syndrome61009Risk Ratio (M-H, Random, 95% CI)0.40 [0.16, 1.02]
1.4 CMV infection6815Risk Ratio (M-H, Random, 95% CI)0.44 [0.28, 0.67]
1.5 CMV disease in patients treated with ganciclovir for 3 months4703Risk Ratio (M-H, Random, 95% CI)0.28 [0.09, 0.82]
1.6 CMV disease in patients treated with ganciclovir for 2-4 weeks then aciclovir3410Risk Ratio (M-H, Random, 95% CI)0.38 [0.22, 0.64]
2 CMV antibody +ve recipients6 Risk Ratio (M-H, Random, 95% CI)Subtotals only
2.1 All symptomatic CMV disease5722Risk Ratio (M-H, Random, 95% CI)0.27 [0.13, 0.55]
2.2 CMV infection5522Risk Ratio (M-H, Random, 95% CI)0.30 [0.16, 0.58]
3 CMV +ve donors / CMV -ve recipients5 Risk Ratio (M-H, Random, 95% CI)Subtotals only
3.1 All symptomatic CMV disease5246Risk Ratio (M-H, Random, 95% CI)0.64 [0.41, 0.99]
3.2 CMV infection4228Risk Ratio (M-H, Random, 95% CI)0.63 [0.36, 1.09]
4 CMV -ve donor / CMV -ve recipient3 Risk Ratio (M-H, Random, 95% CI)Subtotals only
4.1 CMV disease341Risk Ratio (M-H, Random, 95% CI)0.45 [0.07, 3.07]
5 Effect of prophylaxis for different transplanted organs7 Risk Ratio (M-H, Random, 95% CI)Subtotals only
5.1 CMV disease in kidney transplant patients2168Risk Ratio (M-H, Random, 95% CI)0.30 [0.07, 1.35]
5.2 CMV disease in liver transplant patients5791Risk Ratio (M-H, Random, 95% CI)0.37 [0.23, 0.59]
5.3 CMV disease in heart or lung transplant patients275Risk Ratio (M-H, Random, 95% CI)0.55 [0.10, 3.00]
5.4 CMV infection in kidney transplant patients2168Risk Ratio (M-H, Random, 95% CI)0.20 [0.04, 0.95]
5.5 CMV infection in liver transplant patients4572Risk Ratio (M-H, Random, 95% CI)0.42 [0.25, 0.73]
5.6 CMV infection in heart or lung transplant patients275Risk Ratio (M-H, Random, 95% CI)0.88 [0.50, 1.55]
6 Death8 Risk Ratio (M-H, Random, 95% CI)Subtotals only
6.1 Death associated with CMV disease6832Risk Ratio (M-H, Random, 95% CI)0.33 [0.07, 1.58]
6.2 All-cause mortality81138Risk Ratio (M-H, Random, 95% CI)1.13 [0.82, 1.58]
7 Additional outcomes8 Risk Ratio (M-H, Random, 95% CI)Subtotals only
7.1 Acute rejection61009Risk Ratio (M-H, Random, 95% CI)0.98 [0.87, 1.10]
7.2 Graft loss3268Risk Ratio (M-H, Random, 95% CI)0.55 [0.27, 1.13]
7.3 Other viral infections4740Risk Ratio (M-H, Random, 95% CI)0.81 [0.32, 2.01]
7.4 Invasive fungal infections3401Risk Ratio (M-H, Random, 95% CI)0.67 [0.40, 1.10]
7.5 Bacterial infections1167Risk Ratio (M-H, Random, 95% CI)1.10 [0.78, 1.53]
7.6 Protozoal infections1167Risk Ratio (M-H, Random, 95% CI)0.34 [0.01, 8.16]
7.7 Obliterative bronchiolitis in lung transplant recipients125Risk Ratio (M-H, Random, 95% CI)0.81 [0.42, 1.54]
7.8 Leucopenia6955Risk Ratio (M-H, Random, 95% CI)3.28 [1.48, 7.25]
7.9 Kidney dysfunction4661Risk Ratio (M-H, Random, 95% CI)0.96 [0.83, 1.10]
7.10 Neurological dysfunction2306Risk Ratio (M-H, Random, 95% CI)1.01 [0.24, 4.15]
Analysis 4.1.

Comparison 4 Ganciclovir versus aciclovir, Outcome 1 CMV disease and CMV infection in all treated patients.

Analysis 4.2.

Comparison 4 Ganciclovir versus aciclovir, Outcome 2 CMV antibody +ve recipients.

Analysis 4.3.

Comparison 4 Ganciclovir versus aciclovir, Outcome 3 CMV +ve donors / CMV -ve recipients.

Analysis 4.4.

Comparison 4 Ganciclovir versus aciclovir, Outcome 4 CMV -ve donor / CMV -ve recipient.

Analysis 4.5.

Comparison 4 Ganciclovir versus aciclovir, Outcome 5 Effect of prophylaxis for different transplanted organs.

Analysis 4.6.

Comparison 4 Ganciclovir versus aciclovir, Outcome 6 Death.

Analysis 4.7.

Comparison 4 Ganciclovir versus aciclovir, Outcome 7 Additional outcomes.

Comparison 5. Ganciclovir / aciclovir versus ganciclovir
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 CMV disease and CMV infection in all treated patients1 Risk Ratio (M-H, Random, 95% CI)Totals not selected
1.1 CMV disease1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
1.2 CMV infection1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
2 Death1 Risk Ratio (M-H, Random, 95% CI)Totals not selected
2.1 All-cause mortality1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
3 Additional outcomes1 Risk Ratio (M-H, Random, 95% CI)Totals not selected
3.1 EBV infection1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
Analysis 5.1.

Comparison 5 Ganciclovir / aciclovir versus ganciclovir, Outcome 1 CMV disease and CMV infection in all treated patients.

Analysis 5.2.

Comparison 5 Ganciclovir / aciclovir versus ganciclovir, Outcome 2 Death.

Analysis 5.3.

Comparison 5 Ganciclovir / aciclovir versus ganciclovir, Outcome 3 Additional outcomes.

Comparison 6. Valganciclovir versus ganciclovir
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 CMV disease or infection in CMV donor +ve / recipient -ve1 Risk Ratio (M-H, Random, 95% CI)Totals not selected
1.1 CMV disease by 6 months1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
1.2 CMV disease by 1 year1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
1.3 CMV syndrome by 6 months1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
1.4 CMV syndrome by 1 year1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
1.5 Tissue invasive CMV disease by 6 months1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
1.6 Tissue invasive CMV disease by 1 year1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
1.7 CMV disease in liver transplant recipients by 6 months1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
1.8 CMV disease in renal transplant recipients by 6 months1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
1.9 CMV disease in heart transplant recipients by 6 months1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
1.10 CMV disease in renal-pancreas transplant recipients by 6 months1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
1.11 CMV infection by 6 months1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
1.12 CMV infection by 1 year1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
2 Death1 Risk Ratio (M-H, Random, 95% CI)Totals not selected
2.1 Death due to CMV disease1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
2.2 All-cause mortality1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
3 Additional outcomes1 Risk Ratio (M-H, Random, 95% CI)Totals not selected
3.1 Acute rejection in all recipients1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
3.2 Graft loss1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
3.3 Opportunistic infections1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
3.4 Neutrophil count < 1000/mm³1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
3.5 Medications ceased because of neutropenia1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
3.6 Anaemia (< 80 g/L)1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
3.7 Thrombocytopenia1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
3.8 Tremor1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
Analysis 6.1.

Comparison 6 Valganciclovir versus ganciclovir, Outcome 1 CMV disease or infection in CMV donor +ve / recipient -ve.

Analysis 6.2.

Comparison 6 Valganciclovir versus ganciclovir, Outcome 2 Death.

Analysis 6.3.

Comparison 6 Valganciclovir versus ganciclovir, Outcome 3 Additional outcomes.

Comparison 7. Valaciclovir versus ganciclovir or valganciclovir
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 CMV disease and CMV infection in all treated patients3 Risk Ratio (M-H, Random, 95% CI)Subtotals only
1.1 CMV disease3188Risk Ratio (M-H, Random, 95% CI)0.74 [0.15, 3.75]
1.2 CMV infection3188Risk Ratio (M-H, Random, 95% CI)1.37 [0.78, 2.39]
1.3 CMV disease in donor +ve or -ve/recipient +ve163Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
1.4 CMV infection in donor +ve or -ve/recipient +ve163Risk Ratio (M-H, Random, 95% CI)0.45 [0.09, 2.31]
1.5 CMV disease in donor +ve/recipient -ve112Risk Ratio (M-H, Random, 95% CI)0.33 [0.02, 6.86]
1.6 CMV infection in donor +ve/recipient -ve112Risk Ratio (M-H, Random, 95% CI)1.86 [0.86, 4.01]
2 Death2 Risk Ratio (M-H, Random, 95% CI)Subtotals only
2.1 All-cause mortality2154Risk Ratio (M-H, Random, 95% CI)1.03 [0.15, 6.90]
3 Additional outcomes3 Risk Ratio (M-H, Random, 95% CI)Subtotals only
3.1 Acute rejection3188Risk Ratio (M-H, Random, 95% CI)0.91 [0.22, 3.73]
3.2 Graft loss2107Risk Ratio (M-H, Random, 95% CI)1.34 [0.23, 7.86]
3.3 Leucopenia169Risk Ratio (M-H, Random, 95% CI)1.03 [0.40, 2.62]
3.4 Thrombocytopenia169Risk Ratio (M-H, Random, 95% CI)0.63 [0.30, 1.33]
3.5 Anaemia168Risk Ratio (M-H, Random, 95% CI)0.5 [0.19, 1.31]
3.6 Neurological dysfunction169Risk Ratio (M-H, Random, 95% CI)1.54 [0.62, 3.87]
3.7 Dose reduction or cessation for adverse effects169Risk Ratio (M-H, Random, 95% CI)0.62 [0.25, 1.51]
3.8 Other herpes virus infections183Risk Ratio (M-H, Random, 95% CI)1.86 [0.18, 19.73]
3.9 Non-viral infections183Risk Ratio (M-H, Random, 95% CI)0.59 [0.44, 0.80]
4 Renal function at end of study3 Std. Mean Difference (IV, Random, 95% CI)Subtotals only
4.1 Serum creatinine3188Std. Mean Difference (IV, Random, 95% CI)-0.23 [-0.51, 0.06]
4.2 Calculated GFR169Std. Mean Difference (IV, Random, 95% CI)0.41 [-0.06, 0.89]
Analysis 7.1.

Comparison 7 Valaciclovir versus ganciclovir or valganciclovir, Outcome 1 CMV disease and CMV infection in all treated patients.

Analysis 7.2.

Comparison 7 Valaciclovir versus ganciclovir or valganciclovir, Outcome 2 Death.

Analysis 7.3.

Comparison 7 Valaciclovir versus ganciclovir or valganciclovir, Outcome 3 Additional outcomes.

Analysis 7.4.

Comparison 7 Valaciclovir versus ganciclovir or valganciclovir, Outcome 4 Renal function at end of study.

Comparison 8. Different ganciclovir regimens
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 IV doses given at different frequencies1 Risk Ratio (M-H, Random, 95% CI)Totals not selected
1.1 CMV disease1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
1.2 CMV syndrome1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
1.3 Invasive CMV disease1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
1.4 CMV infection1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
1.5 All-cause mortality1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
1.6 Death due to CMV disease1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
1.7 Bacteraemia1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
1.8 Bronchiolitis obliterans syndrome1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
1.9 Leucopenia1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
2 Oral versus IV ganciclovir2 Risk Ratio (M-H, Random, 95% CI)Subtotals only
2.1 CMV disease294Risk Ratio (M-H, Random, 95% CI)0.57 [0.16, 2.05]
2.2 CMV syndrome294Risk Ratio (M-H, Random, 95% CI)0.48 [0.11, 2.11]
2.3 CMV invasive organ disease164Risk Ratio (M-H, Random, 95% CI)1.0 [0.07, 15.30]
2.4 CMV infection130Risk Ratio (M-H, Random, 95% CI)1.05 [0.41, 2.70]
2.5 All-cause mortality164Risk Ratio (M-H, Random, 95% CI)5.00 [0.62, 40.44]
2.6 Acute rejection294Risk Ratio (M-H, Random, 95% CI)0.85 [0.45, 1.59]
2.7 Graft loss134Risk Ratio (M-H, Random, 95% CI)1.0 [0.07, 14.72]
2.8 Leucopenia due to ganciclovir164Risk Ratio (M-H, Random, 95% CI)0.69 [0.35, 1.39]
2.9 Medications ceased due to leucopenia164Risk Ratio (M-H, Random, 95% CI)1.0 [0.27, 3.66]
Analysis 8.1.

Comparison 8 Different ganciclovir regimens, Outcome 1 IV doses given at different frequencies.

Analysis 8.2.

Comparison 8 Different ganciclovir regimens, Outcome 2 Oral versus IV ganciclovir.

Comparison 9. Extended duration compared with three months of valganciclovir
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 CMV disease2 Risk Ratio (M-H, Random, 95% CI)Subtotals only
1.1 CMV disease at end of treatment2454Risk Ratio (M-H, Random, 95% CI)0.20 [0.12, 0.35]
1.2 CMV disease at 9 months1310Risk Ratio (M-H, Random, 95% CI)0.39 [0.25, 0.60]
1.3 CMV disease at 12 months1318Risk Ratio (M-H, Random, 95% CI)0.44 [0.29, 0.66]
1.4 CMV disease at 24 months1318Risk Ratio (M-H, Random, 95% CI)0.55 [0.38, 0.79]
2 CMV syndrome2454Risk Ratio (M-H, Random, 95% CI)0.39 [0.24, 0.64]
3 CMV invasive disease2 Risk Ratio (M-H, Random, 95% CI)Subtotals only
3.1 Number at 12 months2454Risk Ratio (M-H, Random, 95% CI)0.17 [0.02, 1.34]
3.2 Number at 24 months1318Risk Ratio (M-H, Random, 95% CI)0.70 [0.12, 4.14]
4 CMV infection2 Risk Ratio (M-H, Random, 95% CI)Subtotals only
4.1 CMV infection at end of treatment2454Risk Ratio (M-H, Random, 95% CI)0.27 [0.10, 0.71]
4.2 CMV infection at 9 months1318Risk Ratio (M-H, Random, 95% CI)0.72 [0.56, 0.94]
4.3 CMV infection at 12 months1318Risk Ratio (M-H, Random, 95% CI)0.73 [0.57, 0.95]
5 All-cause mortality1 Risk Ratio (M-H, Random, 95% CI)Totals not selected
5.1 Number at 12 months1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
5.2 Number at 2 years1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
6 Graft loss1 Risk Ratio (M-H, Random, 95% CI)Totals not selected
6.1 Number at 12 months1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
6.2 Number at 24 months1 Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]
7 Acute rejection2 Risk Ratio (M-H, Random, 95% CI)Subtotals only
7.1 Biopsy proved acute rejection < 100 days1318Risk Ratio (M-H, Random, 95% CI)0.70 [0.32, 1.51]
7.2 Biopsy proven acute rejection at 12 months2454Risk Ratio (M-H, Random, 95% CI)0.64 [0.43, 0.95]
7.3 Biopsy proven acute rejection at 24 months1318Risk Ratio (M-H, Random, 95% CI)0.62 [0.35, 1.08]
8 Other outcomes2 Risk Ratio (M-H, Random, 95% CI)Subtotals only
8.1 Opportunistic infections2456Risk Ratio (M-H, Random, 95% CI)0.71 [0.33, 1.57]
8.2 Post-transplant diabetes mellitus1244Risk Ratio (M-H, Random, 95% CI)1.17 [0.58, 2.36]
9 Adverse effects2 Risk Difference (M-H, Random, 95% CI)Subtotals only
9.1 Total treatment related adverse effects2456Risk Difference (M-H, Random, 95% CI)0.08 [-0.01, 0.16]
9.2 Treatment related serious adverse effects2456Risk Difference (M-H, Random, 95% CI)0.02 [-0.02, 0.07]
9.3 Leukopenia1320Risk Difference (M-H, Random, 95% CI)0.12 [0.01, 0.22]
9.4 Leucopenia leading to VGCV cessation1320Risk Difference (M-H, Random, 95% CI)0.04 [0.00, 0.07]
9.5 Termination due to treatment related adverse effects1136Risk Difference (M-H, Random, 95% CI)0.07 [-0.04, 0.18]
9.6 Hospitalisations due to CMV disease1418Risk Difference (M-H, Random, 95% CI)-0.10 [-0.17, -0.04]
9.7 Hospitalisations due to adverse effects1418Risk Difference (M-H, Random, 95% CI)0.04 [-0.05, 0.13]
9.8 CMV mutations known to confer ganciclovir resistance2208Risk Difference (M-H, Random, 95% CI)0.02 [-0.08, 0.11]
Analysis 9.1.

Comparison 9 Extended duration compared with three months of valganciclovir, Outcome 1 CMV disease.

Analysis 9.2.

Comparison 9 Extended duration compared with three months of valganciclovir, Outcome 2 CMV syndrome.

Analysis 9.3.

Comparison 9 Extended duration compared with three months of valganciclovir, Outcome 3 CMV invasive disease.

Analysis 9.4.

Comparison 9 Extended duration compared with three months of valganciclovir, Outcome 4 CMV infection.

Analysis 9.5.

Comparison 9 Extended duration compared with three months of valganciclovir, Outcome 5 All-cause mortality.

Analysis 9.6.

Comparison 9 Extended duration compared with three months of valganciclovir, Outcome 6 Graft loss.

Analysis 9.7.

Comparison 9 Extended duration compared with three months of valganciclovir, Outcome 7 Acute rejection.

Analysis 9.8.

Comparison 9 Extended duration compared with three months of valganciclovir, Outcome 8 Other outcomes.

Analysis 9.9.

Comparison 9 Extended duration compared with three months of valganciclovir, Outcome 9 Adverse effects.

Appendices

Appendix 1. Electronic search strategies

DatabaseSearch terms
CENTRAL
  1. MeSH descriptor Cytomegalovirus, this term only in MeSH products

  2. MeSH descriptor Cytomegalovirus Infections explode all trees in MeSH products

  3. MeSH descriptor Cytomegalovirus Vaccines explode all trees

  4. cytomegalovirus* in All Fields in CENTRAL

  5. cmv* in All Fields in CENTRAL

  6. (#1 OR #2 OR #3 OR #4 OR #5)

  7. (organ or renal or kidney or heart or lung or liver or pancreas) adj transplant in All Fields in all products

  8. MeSH descriptor Organ Transplantation, this term only

  9. MeSH descriptor Heart Transplantation explode all trees

  10. MeSH descriptor Lung Transplantation explode all trees

  11. MeSH descriptor Kidney Transplantation, this term only

  12. MeSH descriptor Liver Transplantation, this term only

  13. MeSH descriptor Pancreas Transplantation, this term only

  14. (#7 OR #8 OR #9 OR #10 OR #11 OR #12 OR #13)

  15. (#6 AND #14)

MEDLINE (OVID SP)
  1. Cytomegalovirus/

  2. exp Cytomegalovirus Infections/

  3. Cytomegalovirus Vaccines/

  4. cytomegalovirus.tw.

  5. cmv.tw.

  6. or/1-5

  7. Organ Transplantation/

  8. exp Heart Transplantation/

  9. exp Lung Transplantation/

  10. Kidney Transplantation/

  11. Liver Transplantation/

  12. Pancreas Transplantation/

  13. ((organ or renal or kidney or heart or lung or liver or pancreas) adj transplant$).tw.

  14. or/8-13

  15. 6 and 14

EMBASE (OVID SP)
  1. exp CYTOMEGALOVIRUS/

  2. Cytomegalovirus Infection/

  3. Cytomegalovirus Antibody/

  4. Cytomegalovirus Vaccine/

  5. cytomegalovirus.tw.

  6. CMV.tw.

  7. or/1-6

  8. exp organ transplantation/

  9. ((organ or renal or kidney or heart or lung or liver or pancreas) adj transplant$).tw.

  10. or/8-9

  11. 7 and 10

Appendix 2. Risk of bias assessment tool

Potential source of biasAssessment criteria

Random sequence generation

Selection bias (biased allocation to interventions) due to inadequate generation of a randomised sequence

Low risk of bias: Random number table; computer random number generator; coin tossing; shuffling cards or envelopes; throwing dice; drawing of lots; minimization (minimization may be implemented without a random element, and this is considered to be equivalent to being random).
High risk of bias: Sequence generated by odd or even date of birth; date (or day) of admission; sequence generated by hospital or clinic record number; allocation by judgement of the clinician; by preference of the participant; based on the results of a laboratory test or a series of tests; by availability of the intervention.
Unclear: Insufficient information about the sequence generation process to permit judgement.

Allocation concealment

Selection bias (biased allocation to interventions) due to inadequate concealment of allocations prior to assignment

Low risk of bias: Randomisation method described that would not allow investigator/participant to know or influence intervention group before eligible participant entered in the study (e.g. central allocation, including telephone, web-based, and pharmacy-controlled, randomisation; sequentially numbered drug containers of identical appearance; sequentially numbered, opaque, sealed envelopes).
High risk of bias: Using an open random allocation schedule (e.g. a list of random numbers); assignment envelopes were used without appropriate safeguards (e.g. if envelopes were unsealed or non-opaque or not sequentially numbered); alternation or rotation; date of birth; case record number; any other explicitly unconcealed procedure.
Unclear: Randomisation stated but no information on method used is available.

Blinding of participants and personnel

Performance bias due to knowledge of the allocated interventions by participants and personnel during the study

Low risk of bias: No blinding or incomplete blinding, but the review authors judge that the outcome is not likely to be influenced by lack of blinding; blinding of participants and key study personnel ensured, and unlikely that the blinding could have been broken.
High risk of bias: No blinding or incomplete blinding, and the outcome is likely to be influenced by lack of blinding; blinding of key study participants and personnel attempted, but likely that the blinding could have been broken, and the outcome is likely to be influenced by lack of blinding.
Unclear: Insufficient information to permit judgement

Blinding of outcome assessment

Detection bias due to knowledge of the allocated interventions by outcome assessors.

Low risk of bias: No blinding of outcome assessment, but the review authors judge that the outcome measurement is not likely to be influenced by lack of blinding; blinding of outcome assessment ensured, and unlikely that the blinding could have been broken.
High risk of bias: No blinding of outcome assessment, and the outcome measurement is likely to be influenced by lack of blinding; blinding of outcome assessment, but likely that the blinding could have been broken, and the outcome measurement is likely to be influenced by lack of blinding.
Unclear: Insufficient information to permit judgement

Incomplete outcome data

Attrition bias due to amount, nature or handling of incomplete outcome data.

Low risk of bias: No missing outcome data; reasons for missing outcome data unlikely to be related to true outcome (for survival data, censoring unlikely to be introducing bias); missing outcome data balanced in numbers across intervention groups, with similar reasons for missing data across groups; for dichotomous outcome data, the proportion of missing outcomes compared with observed event risk not enough to have a clinically relevant impact on the intervention effect estimate; for continuous outcome data, plausible effect size (difference in means or standardized difference in means) among missing outcomes not enough to have a clinically relevant impact on observed effect size; missing data have been imputed using appropriate methods.
High risk of bias: Reason for missing outcome data likely to be related to true outcome, with either imbalance in numbers or reasons for missing data across intervention groups; for dichotomous outcome data, the proportion of missing outcomes compared with observed event risk enough to induce clinically relevant bias in intervention effect estimate; for continuous outcome data, plausible effect size (difference in means or standardized difference in means) among missing outcomes enough to induce clinically relevant bias in observed effect size; ‘as-treated’ analysis done with substantial departure of the intervention received from that assigned at randomisation; potentially inappropriate application of simple imputation.
Unclear: Insufficient information to permit judgement

Selective reporting

Reporting bias due to selective outcome reporting

Low risk of bias: The study protocol is available and all of the study’s pre-specified (primary and secondary) outcomes that are of interest in the review have been reported in the pre-specified way; the study protocol is not available but it is clear that the published reports include all expected outcomes, including those that were pre-specified (convincing text of this nature may be uncommon).
High risk of bias: Not all of the study’s pre-specified primary outcomes have been reported; one or more primary outcomes is reported using measurements, analysis methods or subsets of the data (e.g. subscales) that were not pre-specified; one or more reported primary outcomes were not pre-specified (unless clear justification for their reporting is provided, such as an unexpected adverse effect); one or more outcomes of interest in the review are reported incompletely so that they cannot be entered in a meta-analysis; the study report fails to include results for a key outcome that would be expected to have been reported for such a study.
Unclear: Insufficient information to permit judgement

Other bias

Bias due to problems not covered elsewhere in the table

Low risk of bias: The study appears to be free of other sources of bias.
High risk of bias: Had a potential source of bias related to the specific study design used; stopped early due to some data-dependent process (including a formal-stopping rule); had extreme baseline imbalance; has been claimed to have been fraudulent; had some other problem.
Unclear: Insufficient information to assess whether an important risk of bias exists; insufficient rationale or evidence that an identified problem will introduce bias.

What's new

DateEventDescription
3 January 2013New search has been performedNew studies included
3 January 2013New citation required and conclusions have changedRisk of bias assessment incorporated

History

Protocol first published: Issue 3, 2002
Review first published: Issue 4, 2005

DateEventDescription
18 March 2010AmendedContact details updated.
13 May 2009AmendedContact details updated.
13 August 2008AmendedConverted to new review format.
7 January 2008New citation required and conclusions have changedSubstantive amendment, 6 additional publications identified, 2 new studies included
16 October 2004AmendedTitle changed. Background, methods edited to reflect limitation of review to prophylaxis with antiviral medication. Quality assessment criteria added.

Contributions of authors

Review update in 2013

  • EMH, ML, ACW and JCC contributed to the data extraction, quality assessment, data analysis and rewriting of the review update.

Review update in 2008

  • EMH, ACW, JCC, GFMS contributed to the data extraction, quality assessment, data analysis and rewriting of the review update.

Original review 2005

  • EMH identified and extracted data from included studies, contacted authors, analysed and interpreted the results and wrote the manuscript.

  • CAJ conceived, designed and developed the protocol and search strategy for the review, identified and extracted data from included studies and participated in revision of the manuscript.

  • ACW analysed and interpreted the results and participated in the revision of the manuscript.

  • GFMS checked the analysis and interpretation of the results and participated in the revision of the manuscript.

  • PGB and KK identified and extracted data from included studies and participated in revision of the manuscript.

  • DV developed the protocol and search strategy for the review.

  • JCC conceived, designed and developed the protocol, analysed and interpreted the results and edited the drafting and revision of the manuscript.

Declarations of interest

None known.

Characteristics of studies

Characteristics of included studies [ordered by study ID]

2VAL Study 2010 Kidney

Methods
  • Study design: parallel RCT

  • Time frame: November 2007 and ongoing

  • Follow-up period: 4 months (preliminary data); planned for 36 months

  • Loss to follow-up: 0%

Participants
  • Country: Czech Republic

  • Setting: tertiary single centre

  • Kidney transplant recipients aged ≥ 18 years; D/R+, D+/R-, D-/R-

Treatment group 1

  • Number: 19

  • Mean age ± SD: 46 ± 14 years

  • Sex (M/F): 13/6

Treatment group 2

  • Number: 17

  • Mean age ± SD: 47 ± 10 years

  • Sex (M/F): 10/7

Exclusion criteria

  • Unknown or D-/R- serology; systemic antiviral drug intake within 2 weeks; active viral infection; significant leukopenia or thrombocytopenia; participation in another study; allergy to study medications

Interventions

Treatment group 1

  • VGCV: 900 mg orally/d for 12 weeks

Treatment group 2

  • VACV: 2000 mg 4 times/d for 12 weeks

Co-interventions

  • CSA, TAC, MMF, prednisone, ALG 1/19 valganciclovir, 5/17 valaciclovir

Outcomes
  1. CMV disease

  2. CMV infection: CMV DNA by PCR

  3. Graft loss

  4. Acute rejection

  5. Adverse effects

NotesPreliminary results at 4 months only. Full data to be analysed when all patients have completed 12 months. Information on results and randomisation sequence obtained from authors
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskRandom numbers table, block randomisation (1:1 ratio, blocks of 4)
Allocation concealment (selection bias)Low riskSealed envelopes opened after patient enrolled
Blinding of participants and personnel (performance bias)
All outcomes
High riskOpen label. Lack of blinding could influence clinical assessment of symptoms of possible CMV disease
Blinding of outcome assessment (detection bias)
All outcomes
High riskOpen label. Lack of blinding could influence clinical assessment of symptoms of possible CMV disease
Incomplete outcome data (attrition bias)
All outcomes
Unclear riskFull data on follow-up not yet reported
Selective reporting (reporting bias)Unclear riskFull data on outcomes not yet reported
Other biasLow riskGrants from Ministry of Health

Ahsan 1997 Kidney

Methods
  • Study design: parallel RCT

  • Time frame: March 1995 to December 1995

  • Follow-up period: 9 months

  • Loss to follow-up: 0%

Participants
  • Country: USA

  • Setting: tertiary single centre

  • Kidney transplant recipients; D/R+, D+/R-, D-/R-; if diabetic or receiving OKT-3

Treatment group

  • Number: 22

  • Mean age ± SEM: 50.4 ± 2.3 years

  • Sex (M/F): 10/11

  • CD/LD: 18/3

Control group

  • Number: 22

  • Mean age ± SEM: 47.6 ± 2.1 years

  • Sex (M/F): 12/11

  • CD/LD: 7/15

Exclusion criteria: NS

Interventions

Treatment group

  • GCV: 750 mg orally twice/d for 12 weeks starting day 1

Control group

  • No treatment

Co-interventions

  • CSA, AZA, prednisone, OKT-3 (CD recipients)

Outcomes
  1. CMV disease

  2. CMV infection: CMV culture, IgM3

  3. All-cause mortality

  4. Death due to CMV disease

  5. Acute rejection

  6. Graft loss

  7. Opportunistic infections

Notes
  1. Exclusions post randomisation but pre-intervention: none

  2. Stop or end point: NS

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskComputerised generated code with 4 patients in each block
Allocation concealment (selection bias)Unclear riskRandomisation stated but no information provided
Blinding of participants and personnel (performance bias)
All outcomes
High riskOpen label study. Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias)
All outcomes
High riskOpen label study. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias)
All outcomes
Low riskOne patient excluded but reason unlikely to be related to true outcome
Selective reporting (reporting bias)High riskIncomplete reporting of adverse effects
Other biasUnclear riskNo information about pharmaceutical sponsorship

Badley 1997 Liver

Methods
  • Study design: parallel RCT

  • Time frame: January 1991 to June 1994

  • Follow-up period: 1 year

  • Loss to follow-up: 0%

Participants
  • Country: USA

  • Setting: tertiary multicentre

  • First liver transplant

Treatment group

  • Number: 83

  • Age range: 16 to 68 years

  • Sex (M/F): 50/33

Control group

  • Number: 84

  • Age range: 16 to 68 years

  • Sex (M/F): 46/38

Exclusion criteria

  • Allergy to GCV/ACV; creatinine > 3 mg/dL or GFR < 10; stage 3/4 coma post-transplant; existing CMV infection

Interventions

Treatment group

  • GCV: 5 mg/kg IV twice/d for 14 days starting first day post-transplant

  • ACV: 800 mg orally 4 times/d to 120 days

Control group

  • ACV: 800 mg orally 4 times/d to 120 days

Co-interventions

  • CSA, AZA (one centre), prednisone

Outcomes
  1. CMV disease

  2. CMV syndrome

  3. CMV invasive organ disease

  4. CMV infection: CMV culture

  5. All-cause mortality

  6. Acute rejection

  7. Opportunistic infections

  8. Adverse effects

Notes
  1. Exclusions post randomisation but pre-intervention: 3 excluded

  2. Stop or end point: NS

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low risk"Block randomisation scheme was used to generate a series of 150 randomly selected treatment assignments for each transplant centre"
Allocation concealment (selection bias)Low riskPatient randomisation and all statistical analyses were performed at coordinating centre
Blinding of participants and personnel (performance bias)
All outcomes
High riskMedications schedules differ between intervention groups. Assessment of primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias)
All outcomes
High riskMedications schedules differ between intervention groups. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias)
All outcomes
Low riskThree patients excluded but exclusions unlikely to be related to outcomes
Selective reporting (reporting bias)High riskNo graft loss reported
Other biasLow riskStudy carried out under NIH contracts

Balfour 1989 Kidney

Methods
  • Study design: parallel RCT

  • Time frame: August 1985 to May 1988

  • Follow-up period: 1 year

  • Loss to follow-up: 6% at 1 year, 0% at 6 months

Participants
  • Country: USA

  • Setting: tertiary single centre

  • Cadaveric kidney transplant recipients > 10 years

Treatment group

  • Number: 53

  • Median age (range): 43 years (15 to 67)

  • Sex (M/F): 36/17

Control group

  • Number: 51

  • Median age (range): 42 years (17 to 68)

  • Sex (M/F): 34/17

Exclusion criteria

  • Intolerance of ACV

Interventions

Treatment group

  • ACV: 800 mg orally 4 times/d for 12 weeks starting day of transplant

Control group

  • Placebo: 1 tablet 4 times/d for 12 weeks starting day of transplant

Co-interventions

  • CSA, AZA, prednisone

Outcomes
  1. CMV disease

  2. CMV syndrome

  3. CMV invasive organ disease

  4. CMV infection: CMV culture, rising CMV antibody

  5. All-cause mortality

  6. Death due to CMV disease

  7. Acute rejection

  8. Graft loss

  9. Opportunistic infections

  10. Adverse events

Notes
  1. Exclusions post randomisation but pre-intervention: none reported

  2. Stop or end point: NS

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskRandomisation scheme generated by computer program
Allocation concealment (selection bias)Unclear riskNo information provided
Blinding of participants and personnel (performance bias)
All outcomes
Low riskPlacebo controlled. Placebo tablets identical in appearance to acyclovir
Blinding of outcome assessment (detection bias)
All outcomes
Low riskPlacebo controlled. Placebo tablets identical in appearance to acyclovir
Incomplete outcome data (attrition bias)
All outcomes
Low risk14 patients (6 intervention, 8 placebo) excluded but reasons unlikely to be related to true outcome
Selective reporting (reporting bias)Low riskAll expected outcomes reported
Other biasLow riskReport partial support from NIH, Minnesota Medical Foundation and Burroughs Wellcome

Barkholt 1999 Liver

Methods
  • Study design: parallel RCT

  • Time frame: May 1993 to December 1994

  • Follow-up period: 3 months

  • Loss to follow-up: 0%

Participants
  • Country: Sweden

  • Setting: tertiary single centre

  • Liver transplant recipients; all CMV serostatus

Treatment group

  • Number: 28

  • Mean age ± SD: 41 ± 17 years

  • Sex (M/F): 16/12

Control group

  • Number: 27

  • Mean age± SD: 47 ± 15 years

  • Sex (M/F): 12/15

Exclusion criteria

  • Age < 6 years; HIV infection; CMV therapy in previous 4 weeks

Interventions

Treatment group

  • ACV: 800 mg (1 tablet) orally 4 times/d for 12 weeks starting 6 hours pre-transplant

Control group

  • Placebo: 1 tablet orally 4 times/d for 12 weeks starting 6 hours pre-transplant

Co-interventions

  • CSA, AZA, prednisone

Outcomes
  1. CMV disease

  2. CMV infection: CMV culture, CMV DNA, IgM

  3. All-cause mortality

  4. Death due to CMV disease

  5. Acute rejection

  6. Graft loss

  7. Opportunistic infections

  8. Adverse reactions

Notes
  1. Exclusions post randomisation but pre-intervention: 5

  2. Stop or end point: NS

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNo information provided
Allocation concealment (selection bias)Unclear riskNo information provided
Blinding of participants and personnel (performance bias)
All outcomes
Low riskPlacebo controlled
Blinding of outcome assessment (detection bias)
All outcomes
Low riskPlacebo controlled. Patients with verified CMV infection were withdrawn from study drug without breaking the code
Incomplete outcome data (attrition bias)
All outcomes
Low risk5 excluded (3 given acyclovir outside study; 2 under 6 years) but reasons unlikely to be related to true outcome
Selective reporting (reporting bias)Low riskAll expected outcomes reported
Other biasHigh riskSupported by Wellcome Research Laboratories

Brennan 1997 Kidney

Methods
  • Study design: parallel RCT

  • Time frame: NS

  • Follow-up period: 6 months

  • Loss to follow-up: 0%

Participants
  • Country: USA

  • Setting: tertiary single centre

  • Kidney transplant recipients; D/R+, D+/R- recipients

Treatment group

  • Number: 19

  • Mean age ± SEM: 50.6 ± 2.8 years

  • Sex (M/F): 13/6

Control group

  • Number: 23

  • Mean age ± SEM: 44.2 ± 3.0 years

  • Sex (M/F): 5/18

Exclusion criteria

  • D-/R- recipients

Interventions

Treatment group

  • GCV: 1000 mg orally 3 times/d for 12 weeks starting at transplant

Control group

  • No treatment except ACV low dose to prevent Herpes simplex

Co-interventions

  • CSA, AZA, prednisone, ATG

Outcomes
  1. CMV disease

  2. CMV syndrome

  3. CMV invasive organ disease

  4. CMV infection: CMV DNA

  5. All-cause mortality

  6. Acute rejection

  7. Opportunistic infections

  8. Adverse effects

Notes
  1. Exclusions post randomisation but pre-intervention: None

  2. Stop or end point: NS

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)High riskOdd and even numbers according to last digit of medical record number. Information obtained from authors
Allocation concealment (selection bias)High riskOdd and even numbers according to last digit of medical record number. Information obtained from authors
Blinding of participants and personnel (performance bias)
All outcomes
High riskMedications differ between intervention groups. Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias)
All outcomes
High riskMedications differ between intervention groups. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo missing data on primary outcome
Selective reporting (reporting bias)High riskIncomplete outcome reporting. No report of graft loss
Other biasHigh riskHoffman-La Roche Laboratory pharmaceutical sponsorship

Cohen 1993 Liver

Methods
  • Study design: parallel RCT

  • Time frame: NS

  • Follow-up period: 18 months

  • Loss to follow-up: 0%

Participants
  • Country: UK

  • Setting: tertiary single centre

  • Liver transplant recipients; D/R+, D+/R-

Treatment group

  • Number: 33

  • Mean age: 42.4 years

  • Sex (M/F): 15/18

Control group

  • Number: 32

  • Mean age: 46.3 years

  • Sex (M/F): 16/16

Exclusion criteria

  • Acute kidney injury; multiple organ system failure; D-/R- recipients

Interventions

Treatment group

  • GCV: 5 mg/kg IV twice/d for 14 days starting on day 14

Control group

  • No treatment

Co-interventions

  • CSA, AZA, prednisone

Outcomes
  1. CMV disease

  2. CMV syndrome

  3. CMV invasive organ disease

  4. CMV infection: CMV culture, IgM

  5. All-cause mortality

  6. Death due to CMV disease

  7. Acute rejection

  8. Graft loss

  9. Adverse effects

Notes
  1. Exclusions post randomisation but pre-intervention: None

  2. Stop or end point: NS

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low risk"65 patients were randomised in a distribution determined by random numbers"
Allocation concealment (selection bias)Low riskInformation obtained from authors that method used would not allow investigator/participant to know allocation before participant entered study
Blinding of participants and personnel (performance bias)
All outcomes
High riskOpen label study. Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias)
All outcomes
High riskOpen label study. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias)
All outcomes
Low riskAll patients completed follow-up
Selective reporting (reporting bias)High riskIncomplete reporting of outcomes. No or limited report on other infections or adverse effects
Other biasUnclear riskNo report on pharmaceutical sponsorship

Conti 1995 Kidney

Methods
  • Study design: parallel RCT

  • Time frame: January 1992 to January 1994

  • Follow-up period: 12 months

  • Loss to follow-up: 0%

Participants
  • Country: USA

  • Setting: tertiary single centre

  • Kidney transplant recipients; D/R+; receiving ALG for induction or rejection

Treatment group

  • Number: 22

  • Mean age: 43 years

  • Sex (M/F): 11/11

Control group

  • Number: 18

  • Mean age: 45 years

  • Sex (M/F): 12/6

Exclusion criteria: NS

Interventions

Treatment group

  • GCV: 5 mg/kg/d IV during ALG therapy (median 10 days) starting on first day of ALG

Control group

  • No treatment

Co-interventions

  • CSA, AZA, prednisone, ALG

Outcomes
  1. CMV disease

  2. CMV syndrome

  3. CMV invasive organ disease

  4. All-cause mortality

  5. Acute rejection

  6. Graft loss

  7. Opportunistic infections

  8. Adverse effects

Notes
  1. Exclusions post randomisation but pre-intervention: None

  2. Stop or end point: NS

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear risk"Patients were randomly assigned" but method of sequence generation not stated
Allocation concealment (selection bias)Unclear risk"Patients were randomly assigned" but no information provided on method used
Blinding of participants and personnel (performance bias)
All outcomes
High riskParticipants in control group received no specific intervention. Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias)
All outcomes
High riskParticipants in control group received no specific intervention. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias)
All outcomes
Low riskAll patients evaluated
Selective reporting (reporting bias)High riskIncomplete reporting of outcomes. No report or limited reporting of CMV infection/adverse effects
Other biasUnclear riskSupported in part by grant from National Kidney Foundation. No report on pharmaceutical sponsorship

Duncan 1993 Lung

Methods
  • Study design: parallel RCT

  • Time frame: NS

  • Follow-up period: 1 year

  • Loss to follow-up: 0%

Participants
  • Country: USA

  • Setting: tertiary single centre

  • Lung transplant recipients; D/R+, D+/R-; neutrophils > 1000/mm³, creatinine > 2.5 mg/dL

Treatment group

  • Number: 13

  • Age: 41.8 ± 9.6 years (mean ± SD)

  • Sex (M/F): 9/4

Control group

  • Number: 12

  • Age: 45.6 ± 8.4 years

  • Sex (M/F): 7/5

Exclusion criteria

  • D-/R-

Interventions

Treatment group

  • GCV: 5 mg/kg 4 times/d IV x 14 days starting day 7; 5 mg/kg/d IV for days 21 to 28; 5 mg/kg IV 5 times/wk to day 90

Control group

  • GCV: 5 mg/kg 4 times/d IV x 14 days starting day 7; 5 mg/kg/d IV for days 21 to 28

  • ACV: 800 mg orally 4 times/d to day 90

Co-interventions

  • CSA, AZA

Outcomes
  1. CMV tissue invasive disease

  2. CMV infection: CMV culture of bronchial lavage

  3. All-cause mortality

  4. Death due to CMV disease

  5. Obliterative bronchiolitis

  6. Graft loss

  7. Adverse effects

Notes
  1. Exclusions post randomisation but pre-intervention: None

  2. Stop or end point: NS

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNo information provided other than that patients were stratified according to CMV serostatus and type of transplant
Allocation concealment (selection bias)Unclear riskSaid to be "randomly assigned" but no other information provided
Blinding of participants and personnel (performance bias)
All outcomes
High riskMedications differ between intervention groups. Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias)
All outcomes
High riskMedications differ between intervention groups. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias)
All outcomes
Low riskConsecutive lung transplant recipients randomised. Results from all reported.
Selective reporting (reporting bias)High riskIncomplete outcome reporting. No or limited reporting of CMV disease, acute rejection, opportunistic infections
Other biasUnclear riskNo report of pharmaceutical sponsorship

Egan 2002 Heart

Methods
  • Study design: parallel RCT

  • Time frame: September 1994 to February 1998

  • Follow-up period: 6 months

  • Loss to follow-up: 0%

Participants
  • Country: UK

  • Setting: tertiary single centre

  • Heart transplant recipients; D/R+

Treatment group

  • Number: 14

  • Mean age (range): 51.6 years (39 to 63)

  • Sex (M/F): 11/1

Control group

  • Number: 13

  • Mean age (range): 50.4 years (31 to 62)

  • Sex (M/F): 10/3

Exclusion criteria

  • Active herpes infection; required other antiviral agents

Interventions

Treatment group

  • VACV: 2000 mg orally 4 times/d for 90 days starting within 72 hours of transplant

Control group

  • ACV: 200 mg orally 4 times/d for 90 days starting within 72 hours of transplant for herpes simplex

Co-interventions

  • CSA, AZA, prednisone, ATG

Outcomes
  1. CMV disease

  2. CMV syndrome

  3. CMV invasive organ disease

  4. CMV infection: CMV antigenaemia, culture

  5. All-cause mortality

  6. Death due to CMV disease

  7. Acute rejection

  8. Graft loss

  9. Opportunistic infections

  10. Adverse effects

Notes
  1. Exclusions post randomisation but pre-intervention: none

  2. Stop or end point: NS

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low risk"Computer generated randomization schedule (block size 4)"
Allocation concealment (selection bias)Low risk"Allocation by opening sealed envelopes corresponding to patient number in sequence"
Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskControl group given low dose acyclovir to "maintain double blind by effective prophylaxis of herpes simplex outbreaks" but no information that acyclovir and valacyclovir tablets were indistinguishable
Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskControl group given low dose acyclovir to "maintain double blind by effective prophylaxis of herpes simplex outbreaks" but no information that acyclovir and valacyclovir tablets were indistinguishable
Incomplete outcome data (attrition bias)
All outcomes
Low riskAll enrolled patients were included in the analysis including 2 patients randomised in error
Selective reporting (reporting bias)Low riskAll expected outcomes reported
Other biasHigh riskFunding provided by Glaxo Wellcome Research and Development

Flechner 1998 Kidney

Methods
  • Study design: parallel RCT

  • Time frame: April 1996 to December 1997

  • Follow-up period: 6 to 27 months

  • Loss to follow-up: 0%

Participants
  • Country: USA

  • Setting: tertiary single centre

  • Kidney transplant recipients > 18 years and < 101 kg; D/R+, D+/R-

Treatment group

  • Number: 40

  • Mean age: 47.9 years

  • Sex (M/F): 30/10

Control group

  • Number: 39

  • Mean age: 50.2 years

  • Sex (M/F): 31/8

Exclusion criteria

  • D-/R-; Allergy to GCV/ACV; AIDS; WBC < 3000; platelets < 100,000; previous viral hepatitis

Interventions

Treatment group

  • GCV: 1000 mg orally 3 times/d for 84 days starting on day 1

Control group

  • ACV: 800 mg orally 4 times/d for 84 days starting on day 1

Co-interventions

  • CMV IgG given to D+/R- recipients in each group; CSA, AZA (⅓), MMF (⅔), OKT-3

Outcomes
  1. CMV disease

  2. CMV syndrome

  3. CMV invasive organ disease

  4. CMV infection: CMV culture

  5. All-cause mortality

  6. Death due to CMV disease

  7. Acute rejection

  8. Opportunistic infections

Notes
  1. Exclusions post-randomisation but pre-intervention: None

  2. Stop or end point: NS

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskComputer generated list. Information provided by authors
Allocation concealment (selection bias)Low riskCentral research coordinator
Blinding of participants and personnel (performance bias)
All outcomes
High riskMedications differ between intervention groups. Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias)
All outcomes
High riskMedications differ between intervention groups. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias)
All outcomes
Low riskAll participants were followed to death/graft loss or June 1998
Selective reporting (reporting bias)High riskIncomplete outcome reporting. No report of graft loss
Other biasUnclear riskNo information provided about pharmaceutical sponsorship

Gane 1997 Liver

Methods
  • Study design: parallel RCT

  • Time frame: December 1993 to April 1995

  • Follow-up period: 1 year

  • Loss to follow-up: 0%

Participants
  • Country: USA, Europe

  • Setting: tertiary multicentre

  • Primary liver transplant recipients aged > 18 years; D/R+, D+/R-

Treatment group

  • Number: 150

  • Mean age ± SD: 46.8 ± 11.6 years

  • Sex (M/F): 92/58

Control group

  • Number: 154

  • Mean age ± SD: 48.1 ± 10.9 years

  • Sex (M/F): 82/72

Exclusion criteria

  • Multiple organ transplant; D-/R- (2 patients inadvertently randomised and included in analysis); unable to take oral medications; neutrophils < 1000; platelets < 25,000; creatinine > 300

Interventions

Treatment group

  • GCV: 1000 mg (4 tablets) orally 3 times/d until day 98 starting within 10 days of transplant

Control group

  • Matching placebo: 4 tablets orally 3 times/d until day 98 starting within 10 days of transplant

Co-interventions

  • CSA, TAC (52 patients), ALG (61 patients)

Outcomes
  1. CMV disease

  2. CMV syndrome

  3. CMV invasive organ disease

  4. CMV infection: CMV antigenaemia, IgM, CMV culture

  5. All-cause mortality

  6. Death due to CMV disease

  7. Acute rejection

  8. Graft loss

  9. Opportunistic infection

  10. Adverse effects

Notes
  1. Exclusions post randomisation but pre-intervention: None

  2. Stop or end point: NS

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNo information provided
Allocation concealment (selection bias)Unclear risk"Randomised trial" but no further information provided
Blinding of participants and personnel (performance bias)
All outcomes
Low riskMatching placebo capsules
Blinding of outcome assessment (detection bias)
All outcomes
Low riskMatching placebo capsules
Incomplete outcome data (attrition bias)
All outcomes
Low riskComplete 12 month data available on all participants
Selective reporting (reporting bias)Low riskAll expected outcomes reported
Other biasHigh riskGrant support from Roche Global Development

Gavalda 1997 Liver

Methods
  • Study design: parallel RCT

  • Time frame: June 1991 to November 1993

  • Follow-up period: 12 months

  • Loss to follow-up: 0%

Participants
  • Country: Spain

  • Setting: tertiary single centre

  • Primary liver transplant recipient; D/R+

Treatment group

  • Number: 37

  • Median age (range): 57 years (34 to 66)

  • Sex (M/F): 25/12

Control group

  • Number: 36

  • Median age (range): 54 years (20 to 65)

  • Sex (M/F): 23/13

Exclusion criteria

  • Second transplant recipients

Interventions

Treatment group

  • ACV: 400 mg orally 5 times/d for 16 weeks starting 3 to 30 days (median 7 days) post-transplant

Control group

  • No treatment

Co-interventions
CSA, prednisone

Outcomes
  1. CMV disease

  2. CMV syndrome

  3. CMV invasive organ disease

  4. CMV infection: CMV culture

  5. All-cause mortality

  6. Opportunistic infections

  7. Adverse effects

Notes
  1. Exclusions post randomisation but pre-intervention: None

  2. Stop or end point: NS

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNo information provided
Allocation concealment (selection bias)Unclear risk"Randomized study" but no other information provided
Blinding of participants and personnel (performance bias)
All outcomes
High riskControl group received no medication. Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias)
All outcomes
High riskControl group received no medication. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias)
All outcomes
Low riskConsecutive adult recipients enrolled. 7 did not complete study. All included in analysis
Selective reporting (reporting bias)High riskIncomplete outcome reporting. No or limited reporting of acute rejection, adverse effects
Other biasUnclear riskNo information provided on pharmaceutical sponsorship

Green 1997 Liver

Methods
  • Study design: parallel RCT

  • Time frame: July 1992 to March 1994

  • Follow-up period: 1 year

  • Loss to follow-up: 0%

Participants
  • Country: USA

  • Setting: tertiary single centre

  • First liver transplant recipients aged < 18 years

Treatment group

  • Number: 24

  • Mean age: 4.9 years

  • Sex (M/F): NS

Control group

  • Number: 24

  • Mean age: 4.3 years

  • Sex (M/F): NS

Exclusion criteria

  • Multi-organ recipients

Interventions

Treatment group

  • GCV: 5 mg/kg twice/d IV for 14 days starting day 1

  • ACV: 800 mg/m² orally 4 times/d to 1 year

Control group

  • GCV: 5 mg/kg twice/d IV for 14 days starting day 1

Co-interventions

  • TAC, prednisone

Outcomes
  1. CMV disease

  2. CMV syndrome

  3. CMV invasive tissue disease

  4. CMV infection: CMV culture

  5. All-cause mortality

  6. Opportunistic infections

Notes
  1. Exclusions post randomisation but pre-intervention: None

  2. Stop or end point: NS

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskStratified according to donor/recipient serostatus. Method not reported.
Allocation concealment (selection bias)Unclear risk"A randomized trial" but no further information provided
Blinding of participants and personnel (performance bias)
All outcomes
High riskControl group received no medication after initial two weeks of ganciclovir therapy. Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias)
All outcomes
High riskControl group received no medication after initial two weeks of ganciclovir therapy. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias)
All outcomes
Low riskAll patients enrolled in study were included in analysis
Selective reporting (reporting bias)High riskIncomplete reporting of outcomes. No or limited reporting of acute rejection, graft loss, adverse effects
Other biasUnclear riskStudy ended following interim analysis which showed no benefit of prolonged course of acyclovir and families requesting that their children receive acyclovir rather than enter trial. No information provided on pharmaceutical sponsorship

Hertz 1998 Heart/lung

Methods
  • Study design: parallel RCT

  • Time frame: January 1993 to January 1996

  • Follow-up period: 1 year

  • Loss to follow-up: 0%

Participants
  • Country: USA

  • Setting: tertiary single centre

  • Lung or heart/lung transplant recipients; D/R+; D+/R-

Treatment group

  • Number: 35

  • Mean age ± SD: 46.4 ± 11.4 years

  • Sex (M/F): 15/20

Control group

  • Number: 37

  • Mean age ± SD: 49.1 ± 8.7 years

  • Sex (M/F): 14/23

Exclusion criteria

  • D-/R-

Interventions

Treatment group

  • GCV: 5 mg/kg twice/d IV on days 8 to 21; 5 mg/kg IV 3 times/wk to 90 days

Control group

  • GCV: 5 mg/kg twice/d IV on days 8 to 21; 5 mg/kg IV daily to 90 days

Co-interventions

  • CSA, AZA, prednisone

Outcomes
  1. CMV disease

  2. CMV syndrome

  3. CMV tissue invasive disease

  4. CMV infection: CMV culture of bronchial lavage

  5. All-cause mortality

  6. Death due to CMV disease

  7. Opportunistic infections

  8. Adverse effects

Notes
  1. Exclusions post randomisation but pre-intervention: None

  2. Stop or end point: NS

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNo information provided
Allocation concealment (selection bias)Unclear risk"Randomized trial" in title but no information provided
Blinding of participants and personnel (performance bias)
All outcomes
High riskDifferent interventions given to groups. Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias)
All outcomes
High riskDifferent interventions given to groups. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias)
All outcomes
Low riskOne patient unable to complete therapy but included in analyses
Selective reporting (reporting bias)High riskIncomplete outcome reporting. No or limited reporting of graft loss, adverse effects
Other biasUnclear riskNo information provided about pharmaceutical sponsorship

Hibberd 1995 Kidney

Methods
  • Study design: parallel RCT

  • Time frame: November 1990 to September 1992

  • Follow-up period: 6 months

  • Loss to follow-up: 1.8% (2 lost at 32 days and 78 days)

Participants
  • Country: USA

  • Setting: tertiary multicentre

  • Kidney transplant recipients; receiving ALG preparations for induction or treatment of rejection; D/R+

Treatment group

  • Number: 64

  • Mean age ± SEM: 44.2 ± 1.62 years

  • Sex (M/F): 36/28

Control group

  • Number: 49

  • Mean age ± SEM: 42.8 ± 1.99 years

  • Sex (M/F): 33/16

Exclusion criteria

  • Aged < 20 years; pregnant; multi-organ recipient; treatment with other antiviral agent

Interventions

Treatment group

  • GCV: 2.5 mg/kg/d IV during ALG therapy (median duration 9 days) starting within 24 hours of first dose of ALG

Control group

  • No treatment

Co-interventions

  • CSA, AZA, prednisone, ALG or OKT-3

Outcomes
  1. CMV disease

  2. CMV syndrome

  3. CMV invasive organ disease

  4. CMV infection: CMV culture

  5. All-cause mortality

  6. Death due to CMV disease

  7. Graft loss

  8. Adverse effects

Notes
  1. Exclusions post randomisation but pre-intervention: None

  2. Stop or end point: NS

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear risk"Separate randomization lists for each center" but no other information available
Allocation concealment (selection bias)Unclear risk"Patients were randomly assigned" but no other information available
Blinding of participants and personnel (performance bias)
All outcomes
High risk"Investigators at each site knew which patients received the study drug". Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias)
All outcomes
High risk"Investigators at each site knew which patients received the study drug". Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias)
All outcomes
Low riskAll participants included in the analyses
Selective reporting (reporting bias)High riskIncomplete outcome reporting. No or limited reporting of acute rejection, adverse effects
Other biasHigh riskSupported in part by a grant from Ortho Pharmaceutical Corporation. Ganciclovir provided by Syntex Laboratories Inc

IMPACT 2010 Kidney

Methods
  • Study design: parallel RCT

  • Time frame: March 2006 to August 2008 (final data collection date for primary outcome measure)

  • Follow-up period: 24 months

  • Loss to follow-up: 6/326 did not receive experimental therapy. 103 subsequently withdrew from treatment but all who received at least one dose of medication and underwent post randomisation safety assessment were included in ITT analysis for safety. All who received at least one dose of therapy and were D+/R- were included in efficacy study

Participants
  • Countries: 65 transplant centres in 13 countries

  • Setting: tertiary multicentre

  • Kidney transplant recipients

Treatment group 1

  • Number: 156

  • Mean age ± SD: 47 ± 13.5 years

  • Sex (M/F): 116/40

Treatment group 2

  • Number: 164

  • Mean age ± SD: 48.5 ± 13.8 years

  • Sex (M/F): 119/45

Exclusion criteria

  • CMV disease; HIV; hepatitis B; hepatitis C at enrolment; received CMV IgG in previous 1 month; multi-organ transplant

Interventions

Treatment group 1

  • 200 days group

    • VGCV: 900 mg/d orally for 200 days started as soon as able to tolerate oral medications and by 10 days post-transplant

Treatment group 2

  • 100 days group

  • VGCV: 900 mg/d orally for 100 days started as soon as able to tolerate oral medications and by 10 days post-transplant followed by placebo orally for 100 days

Co-interventions

  • Induction therapy with ATG (52, 52) or IL2Ra (79, 72)

Outcomes
  1. CMV disease

  2. CMV infection: CMV DNA by PCR, CMV antigenaemia

  3. All-cause mortality

  4. Acute rejection

  5. Graft loss

  6. Opportunistic infections

  7. Adverse effects

  8. Death due to CMV disease

  9. Ganciclovir resistant mutations

NotesFurther Information sought from the authors on sequence generation and allocation concealment but no response obtained
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear risk"Patients randomized sequentially in a 1:1 ratio at each study centre in the order in which they were enrolled". No other information provided
Allocation concealment (selection bias)Low riskCentral randomisation
Blinding of participants and personnel (performance bias)
All outcomes
Low riskDouble blind. Placebo and active drug "were indistinguishable"
Blinding of outcome assessment (detection bias)
All outcomes
Low risk"Study investigators, site staff and sponsors were fully blinded to treatment allocation until after analysis of the primary endpoint"
Incomplete outcome data (attrition bias)
All outcomes
Low riskITT analysis. Patients excluded who did not receive at least one dose of medication but only 8 patients excluded and numbers unlikely to influence true outcome
Selective reporting (reporting bias)Low riskAll expected outcomes reported
Other biasHigh riskFunded by F Hoffman-La-Roche. Medical writers funded by sponsors. "There is an agreement between the Principal Investigators and the Sponsor that restricts the principal investigators' rights to discuss or publish trial results after the trial is completed"

Kletzmayr 1996 Kidney

Methods
  • Study design: parallel RCT

  • Time frame: NS

  • Follow-up period: 1 year

  • Loss to follow-up: 5.6%

Participants

Country: Austria

  • Setting: tertiary single centre

  • Kidney transplant recipients; D+/R-

Treatment group

  • Number: 22

  • Mean age ± SD: 46 ± 14 years

  • Sex (M/F): 17/5

Control group

  • Number: 10

  • Mean age ± SD: 44 ± 13 years

  • Sex (M/F): 7/3

Exclusion criteria: NS

Interventions

Treatment group

  • ACV: 800 mg 3 times/d orally for 3 months starting first post-op day

Control group

  • No treatment

Co-interventions

  • CSA, AZA, prednisone

Outcomes
  1. CMV disease

  2. CMV infection: CMV antigenaemia, CMV culture, IgM

  3. All-cause mortality

  4. Acute rejection

  5. Graft loss

Notes
  1. Exclusions post randomisation but pre-intervention: None

  2. Stop or end point: NS

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear risk"Patients randomized... in a 2:1 ratio". No information on sequence generation provided
Allocation concealment (selection bias)Unclear risk"Patients were randomly assigned". No information provided on method
Blinding of participants and personnel (performance bias)
All outcomes
High riskControl group received no specific treatment. Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias)
All outcomes
High riskControl group received no specific treatment. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias)
All outcomes
Low risk4/36 excluded from analysis
Selective reporting (reporting bias)High riskIncomplete outcome reporting. No or limited reporting of opportunistic infections/adverse effects
Other biasUnclear riskNo information provided on pharmaceutical sponsorship

Leray 1995 Kidney

Methods
  • Study design: parallel RCT

  • Time frame: January 1991 to July 1994

  • Follow-up period: Unclear

  • Loss to follow-up: 0%

Participants
  • Country: France

  • Setting: tertiary single centre

  • Kidney transplant recipients; D+/R-

Treatment group

  • Number: 13

  • Age: NS

  • Sex (M/F): NS

Control group

  • Number: 10

  • Age: NS

  • Sex (M/F): NS

Exclusion criteria: NS

Interventions

Treatment group

  • GCV: 5 mg/kg IV twice/d for 14 days starting 14 days post-transplant

Control group

  • No treatment

Co-interventions

  • CSA, AZA, prednisone, ALG

Outcomes
  1. CMV disease

  2. CMV infection: CMV antigenaemia, CMV culture, IgM

  3. Acute rejection

  4. Adverse effects

Notes
  1. Exclusions post randomisation but pre-intervention: none reported

  2. Stop or end point: NS

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNo information provided
Allocation concealment (selection bias)Unclear risk"On day 14 patients were randomized". No other information provided
Blinding of participants and personnel (performance bias)
All outcomes
High riskControl group received no specific therapy. Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias)
All outcomes
High riskControl group received no specific therapy. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias)
All outcomes
Unclear riskUnclear if any patients were excluded from analysis
Selective reporting (reporting bias)Unclear riskAbstract only available
Other biasUnclear riskNo information provided on sponsorship

Lowance 1999 Kidney

Methods
  • Study design: parallel RCT

  • Time frame: July 1992 to December 1996

  • Follow-up period: 12 months

  • Loss to follow-up: 0%

Participants
  • Country: USA/Europe

  • Setting: tertiary multicentre

  • Kidney transplant recipients; D/R+, D+/R-

Treatment group

  • Number: 306; D/R+ (204); D+/R- (102)

  • Mean age ± SD: D/R+ (43.6 ± 13.1 years); D+/R- (40.3 ± 14.2 years)

  • Sex (M/F): D/R+ 153/51; D+/R- 60/42

Control group

  • Number: 310; D/R+ (204); D+/R- (106)

  • Mean age ± SD: D/R+ (45.1 ± 13 years); D+/R- (45.6 ± 13.5 years)

  • Sex (M/F): D/R+ 124/80; D+/R- 65/41

Exclusion criteria

  • D-/R-; active herpes infection; antiviral therapy in previous 2 months

Interventions

Treatment group

  • VACV: 2000 mg orally 4 times/d for 90 days starting within 3 days of transplant

Control group

  • Placebo: orally 4 times/d for 90 days starting within 3 days of transplant

Co-interventions

  • CSA, AZA, TAC (6), MMF (7), ATG or ALG (251), OKT-3 (102)

Outcomes
  1. CMV disease

  2. CMV syndrome

  3. CMV invasive organ disease

  4. CMV infection: CMV culture

  5. All-cause mortality

  6. Death due to CMV disease

  7. Acute rejection

  8. Opportunistic infections

  9. Adverse effects

Notes
  1. Exclusions post randomisation but pre-intervention: None

  2. Stop or end point: NS

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear risk"Randomly assigned in 1:1 ratio according to study site". No other information provided
Allocation concealment (selection bias)Unclear risk"Randomly assigned" but method of allocation unstated
Blinding of participants and personnel (performance bias)
All outcomes
Low riskMatching placebo tablets
Blinding of outcome assessment (detection bias)
All outcomes
Low riskMatching placebo tablets
Incomplete outcome data (attrition bias)
All outcomes
Low riskAll patients included in intention to treat analysis
Selective reporting (reporting bias)High riskNot all expected outcomes reported. No graft loss data reported
Other biasHigh riskSupported by Glaxo Wellcome. Employees included as authors

Macdonald 1995 Heart

Methods
  • Study design: parallel RCT

  • Time frame: NS

  • Follow-up period: 12 months

  • Loss to follow-up: 0%

Participants
  • Country: Australia

  • Setting: tertiary single centre

  • Heart transplant recipients; D/R+, D+/R-

Treatment group

  • Number: 28

  • Mean age ± SD: 48 ± 15 years

  • Sex (M/F): 24/4

Control group

  • Number: 28

  • Mean age ± SD: 45 ± 15 years

  • Sex (M/F): 25/3

Exclusion criteria

  • D-/R-

Interventions

Treatment group

  • GCV: 5 mg/kg IV 3 times/wk for 6 weeks starting pre-transplant

Control group

  • Placebo: IV 3 times/wk for 6 weeks starting pre-transplant

Co-interventions

  • CSA, AZA, prednisone, ATG

Outcomes
  1. CMV disease

  2. CMV syndrome

  3. CMV invasive organ disease

  4. CMV infection: CMV culture

  5. All-cause mortality

  6. Opportunistic infections

  7. Adverse effects

Notes
  1. Exclusions post randomisation but pre-intervention: None

  2. Stop or end point: NS

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskTable of random numbers. Separate randomisation sequences were used according to serostatus
Allocation concealment (selection bias)Unclear riskMethod of allocation not stated
Blinding of participants and personnel (performance bias)
All outcomes
Low riskMatching placebo administered to control group
Blinding of outcome assessment (detection bias)
All outcomes
Low riskMatching placebo administered to control group
Incomplete outcome data (attrition bias)
All outcomes
Low riskConsecutive patients enrolled and all included in analysis
Selective reporting (reporting bias)High riskIncomplete outcome reporting. No report of graft loss
Other biasUnclear riskNo report on pharmaceutical sponsorship

Martin 1994 Liver

Methods
  • Study design: parallel RCT

  • Time frame: February 1991 to August 1991

  • Follow-up period: 24 weeks

  • Loss to follow-up: 0%

Participants
  • Country: USA

  • Setting: tertiary single centre

  • Liver transplant recipients aged > 18 years

Treatment group

  • Number: 68

  • Mean age ± SD: 48.1 ± 13.2 years

  • Sex (M/F): 43/25

Control group

  • Number: 71

  • Mean age ± SD: 47 ± 12.9 years

Sex (M/F): 35/36

Exclusion criteria

  • Fulminant hepatic failure; stage 3/4 hepatic coma; hepatic malignancies with pre-operative chemotherapy

Interventions

Treatment group

  • GCV: 5 mg/kg twice/d IV for 14 days starting 2 days post-transplant

  • ACV: 800 mg orally 4 times/d to 10 weeks

Control group

  • ACV: 800 mg orally 4 times/d for 10 weeks starting 2 days post-transplant

Co-interventions

  • TAC

Outcomes
  1. CMV disease

  2. CMV syndrome

  3. CMV invasive tissue disease

  4. CMV infection: CMV culture, IgM

  5. All-cause mortality

  6. Acute rejection

  7. Graft loss

  8. Adverse effects

Notes
  1. Exclusions post randomisation but pre-intervention: none

  2. Stop or end point: NS

  3. Four excluded after randomisation (active CMV (1), death from sepsis (2), unable to take medication (1)) and one randomised to ganciclovir given acyclovir and analysed in acyclovir group

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low risk"Fixed block randomization scheme (block size = 4)"
Allocation concealment (selection bias)Unclear riskNo information provided on allocation
Blinding of participants and personnel (performance bias)
All outcomes
High riskGroups received different medications by different routes. Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias)
All outcomes
High riskGroups received different medications by different routes. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias)
All outcomes
Low risk4/143. Missing outcome data unlikely to be related to true outcome
Selective reporting (reporting bias)High riskDid not report opportunistic infections
Other biasUnclear riskNo information on pharmaceutical sponsorship provided

Merigan 1992 Heart

Methods
  • Study design: parallel RCT

  • Time frame: NS

  • Follow-up period: 120 days

  • Loss to follow-up: 0%

Participants
  • Country: USA

  • Setting: tertiary multicentre

  • Heart transplant recipients; D/R+, D+/R-.

Treatment group

  • Number: 76

  • Mean age ± SEM: 47.1 ± 1.55 years

  • Sex (M/F): 68/8

Control group

  • Number: 73

  • Mean age ± SEM: 47.6 ± 1.4 years

  • Sex (M/F): 63/10

Exclusion criteria

  • D-/R-; combined heart-lung transplant recipients; antiviral agents in previous 7 days; WBC < 1500; platelets < 50,000; GFR < 10 or > 400

Interventions

Treatment group

  • GCV: 5 mg/kg IV twice/d for 14 days starting on day 1 post-transplant but delay for 2 to 7 days in 21%

Control group

  • Placebo: IV twice/d for 14 days starting on day 1 post-transplant but delay for 2 to 7 days in 23%

Co-interventions

  • CSA, AZA, prednisone, OKT-3

Outcomes
  1. CMV disease

  2. CMV syndrome

  3. CMV invasive organ disease

  4. CMV infection: CMV culture

  5. All-cause mortality

  6. Opportunistic infections

  7. Adverse effects

Notes
  1. Exclusions post randomisation but pre-intervention: None

  2. Stop or end point: Study stopped after interim assessment after 80 patients enrolled when difference between treatment groups evident

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear risk"Stratified at randomization according to their CMV serostatus". Otherwise no information provided
Allocation concealment (selection bias)Unclear riskPatients were... randomly assigned". No information provided on allocation
Blinding of participants and personnel (performance bias)
All outcomes
Low riskPatients in control group received infusions of placebo medication
Blinding of outcome assessment (detection bias)
All outcomes
Low riskPatients in control group received infusions of placebo medication
Incomplete outcome data (attrition bias)
All outcomes
Low riskAll patients included in analysis
Selective reporting (reporting bias)High riskIncomplete outcome reporting. No report of graft loss
Other biasHigh riskSupported by Public Health Service grant and by grant from Syntex Corporation (employees included as authors)

Nafar 2005 Kidney

Methods
  • Study design: parallel RCT

  • Time frame: September 2001 to November 2001

  • Follow-up period: 12 months

  • Loss to follow-up: 0%

Participants
  • Country: Iran

  • Setting: tertiary single centre

  • Kidney transplant recipients; D+/R+; ATG required for rejection; second transplant; deceased donor transplant

  • Mean age ± SD: 37.8 ± 9.8 years

Treatment group

  • Number: 16 (17 entered the study)

  • Age: NS

  • Sex (M/F): 11/5

Control group

  • Number: 14 (17 entered study)

  • Age: NS

  • Sex (M/F): 9/5

Exclusion criteria: NS

Interventions

Treatment group

  • GCV: 1000 mg oral 3 times/d for 3 months

Control group

  • GCV: 5 mg/kg/d IV for 2 weeks

Co-interventions

  • ATG for induction or rejection; other immunosuppression NS

Outcomes
  1. CMV disease

  2. CMV viraemia: CMV antigenaemia

  3. Acute rejection

  4. Adverse effects

  5. Kidney function at 12 months

Notes
  1. Exclusions post randomisation but pre-intervention: None

  2. One patient from treatment group excluded following graft loss; 3 excluded from control group (graft loss 1, pre-existing CMV antigenaemia, refusal to be followed).

  3. Stop or end point: NS

  4. Additional data requested from authors: none

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNo information available
Allocation concealment (selection bias)Unclear risk"Randomized prospective trial" in title but no other information provided
Blinding of participants and personnel (performance bias)
All outcomes
High riskDifferent interventions given to groups. Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias)
All outcomes
High riskDifferent interventions given to groups. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias)
All outcomes
High risk4/34 excluded. 3 excluded from !V ganciclovir arm
Selective reporting (reporting bias)High riskDrug toxicity and side effects not reported
Other biasUnclear riskNo information provided on pharmaceutical sponsorship

Nakazato 1993 Liver

Methods
  • Study design: parallel RCT

  • Time frame: August 1990 to November 1991

  • Follow-up period: 1 year

  • Loss to follow-up: 0%

Participants
  • Country: USA

  • Setting: tertiary single centre

  • Liver transplant recipients

Treatment group

  • Number: 52

  • Mean age ± SD: 38.7 ± 21.5 years

  • Sex (M/F): NS

Control group

  • Number: 52

  • Mean age ± SD: 34.9 ± 22.8 years

  • Sex (M/F): NS

Exclusion criteria: NS

Interventions

Treatment group

  • GCV: 5 mg/kg/d IV during inpatient periods in first 3 months post-transplant

  • ACV: 5 mg/kg/d oral to 3 months

Control group

  • ACV: 5 mg/kg/d IV during inpatient periods in first 3 months post-transplant

  • ACV: 5 mg/kg/d oral to 3 months

Co-interventions

  • IgG IV 200 mg/kg/d during inpatient periods in first 3 months post-transplant; CSA (81), TAC (23), prednisone

Outcomes
  1. CMV disease: CMV culture/histopathology and symptoms

  2. All-cause mortality

  3. Acute rejection

  4. Graft loss

  5. Opportunistic infections

  6. Adverse effects

Notes
  1. Exclusions post randomisation but pre-intervention: None

  2. Stop or end point: NS

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNo information provided
Allocation concealment (selection bias)Unclear risk"Preliminary report of a randomized trial..." in title. Otherwise no information provided
Blinding of participants and personnel (performance bias)
All outcomes
High riskDifferent interventions given to groups. Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias)
All outcomes
High riskDifferent interventions given to groups. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias)
All outcomes
Low riskAll patients included in analyses
Selective reporting (reporting bias)High riskNo CMV infection or adverse effects reported
Other biasHigh riskSupported in part by Sandoz Pharmaceuticals

Palmer 2010 Lung

Methods
  • Study design: parallel RCT

  • Time frame: July 2003 to January 2007

  • Follow-up period: 13 months

  • Loss to follow-up: 45/136 withdrawn but all included in analysis

Participants
  • Country: USA

  • Setting/Design: tertiary multicentre (11 centres)

  • Single or double first lung transplant recipient; aged ≥ 18 years; adequate haematological, kidney and liver function; D/R+, D+/R-; received IV GCV for 2 weeks post-transplant; able to tolerate oral medications; negative PCR/bronchoscopy for CMV at baseline and at day 75 when randomisation occurred

Treatment group 1

  • Number: 70

  • Age (IQR): 56 (45 to 62) years

  • Sex (M/F): 29/41

Treatment group 2

  • Number: 66

  • Age (IQR): 55 (42 to 61) years

  • Sex (M/F): 38/28

Exclusion criteria

  • Re-transplant, on ventilator; current/previous GCV outside study; invasive fungal disease; using disallowed medications; previous severe reaction to GCV; diarrhoea; malabsorption; liver/kidney/haematological dysfunction

Interventions

Treatment group 1

  • 12 months group

    • IV GCV: for 2 weeks starting within 24 hours of transplant

    • Oral VGCV: 900 mg/d for 3 months

    • Oral VGCV: 900 mg/d for 9 months

Treatment group 2

  • 3 months group

    • IV GCV: for 2 weeks starting within 24 hours of transplant

    • Oral VGCV: 900 mg/d for 3 months

    • Placebo: for 9 months

Co-interventions

  • TAC 50/70 and 46/66. ALG 23/70 and 21/66

Outcomes
  1. CMV disease

  2. CMV infection: CMV-DNA by PCR on blood and/or broncholavage

  3. All-cause mortality (data from 1 centre)

  4. Acute rejection

  5. Opportunistic infections

  6. Adverse reactions

Notes
  • Information on absolute numbers with outcomes requested from investigators. Response received but information not available

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskRandomised 1.1 stratified by site at 3 months. Computer-generated randomised list managed centrally
Allocation concealment (selection bias)Low riskRandomised at 3 months. Independent pharmacist dispensed medically centrally
Blinding of participants and personnel (performance bias)
All outcomes
Low riskPlacebo controlled
Blinding of outcome assessment (detection bias)
All outcomes
Low riskBronchoscopies performed by investigators blinded to treatment group
Incomplete outcome data (attrition bias)
All outcomes
Low riskAll patients included in analysis
Selective reporting (reporting bias)High riskIncomplete reporting of outcomes. Reports of deaths only available for one institution
Other biasHigh riskFunded by Roche Pharmaceuticals. All data analyses performed at Duke Clinical Research Institute

Pavlopoulou 2005 Kidney

Methods
  • Study design: parallel RCT

  • Time frame: April 1999 to September 2000

  • Follow-up period: 6 months

  • Loss to follow-up: 0%

Participants
  • Country: Greece

  • Setting: tertiary single centre

  • Kidney transplant recipient; D/R+, D+/R-

Treatment group

  • Number: 43

  • Mean age ± SD: 40.7 ± 12 years

  • Sex (M/F): 34/9

Control group

  • Number: 40

  • Mean age ± SD: 43.1 ± 15 years

  • Sex (M/F): 29/11

Exclusion criteria

  • Active herpes viral infection; antiviral therapy in previous 14 days

Interventions

Treatment group

  • VACV: 2000 mg oral 4 times/d starting within 72 hours of transplant for 3 months

Control group

  • GCV: 1000 mg oral 3 times/d starting within 72 hours of transplant for 3 months

Co-interventions

  • CSA or TAC, sirolimus (11), IL2R antagonists 23 (treatment) and 25 (control), ATG 4 (treatment) and 2 (control)

Outcomes
  1. CMV disease

  2. CMV infection: CMV-DNA

  3. All-cause mortality

  4. Acute rejection

  5. Opportunistic infections

  6. Adverse reactions

  7. Kidney function at 6 months

Notes
  1. Exclusions post randomisation but pre-intervention: NS

  2. Stop or end point: NS

  3. Additional data requested from authors: None

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskAssigned randomly in 1:1 ratio but no other information provided
Allocation concealment (selection bias)Unclear riskNo information provided
Blinding of participants and personnel (performance bias)
All outcomes
High riskOpen label. Different interventions given to groups. Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias)
All outcomes
High riskOpen label. Different interventions given to groups. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias)
All outcomes
Low riskAll patients included in analyses
Selective reporting (reporting bias)High riskIncomplete outcome reporting. Limited reporting of adverse effects
Other biasUnclear riskNo information provided on pharmaceutical sponsorship

Paya 2004 All

Methods
  • Study design: parallel RCT

  • Time frame: April 2000 to August 2001

  • Follow-up period: 12 months

  • Loss to follow-up: 0%

Participants
  • Country: USA/Europe/Canada/Australia

  • Setting: tertiary multicentre

  • Solid organ transplant recipient aged >12 years (liver, kidney, heart, kidney-pancreas); D+/R-; first transplant; adequate liver and kidney function

Treatment group

  • Number: 245

  • Mean age: 45.7 years

  • Sex (M/F): 179/66

Control group

  • Number: 127

  • Mean age: 45.3 years

  • Sex (M/F): 95/32

Exclusion criteria

  • Retransplant; history of CMV infection/disease; CMV therapy in previous 30 days; severe uncontrolled diarrhoea; malabsorption

Interventions

Treatment group

  • VGCV: 900 mg oral daily starting within 10 days of transplant for 100 days

Control group

  • GCV: 1000 mg oral 3 times/d starting within 10 days of transplant for 100 days

Co-interventions

  • Immunosuppression according to protocol of centre

Outcomes
  1. CMV disease

  2. CMV syndrome

  3. CMV tissue invasive disease

  4. CMV infection: CMV-DNA; infection confirmed in central lab

  5. All-cause mortality

  6. Death due to CMV disease

  7. Acute rejection

  8. Graft loss

  9. Opportunistic infections

  10. Adverse reactions

NotesExclusions post-randomisation but pre-intervention: 2 excluded from safety analysis as did not receive medication, 8 excluded from primary outcome analysis as not D+/R-
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskStratified according to organ transplanted and assigned in 2:1 ratio at each centre
Allocation concealment (selection bias)Low risk"Treatment randomization numbers were assigned by telephone via a central randomization center"
Blinding of participants and personnel (performance bias)
All outcomes
Low riskDouble-dummy. Placebo tablets given to both groups
Blinding of outcome assessment (detection bias)
All outcomes
Low riskEnd points adjudicated by independent (of sponsor and study) blinded Endpoint Committee
Incomplete outcome data (attrition bias)
All outcomes
Low riskITT population included 364/372 patients. Safety 370/372. Reasons for missing outcomes data unlikely to be related to true outcome
Selective reporting (reporting bias)Low riskExpected outcomes all reported
Other biasHigh riskStudy funded by Hoffman-La Roche

Pouteil-Noble 1996 Kidney

Methods
  • Study design: parallel RCT

  • Time frame: NS

  • Follow-up period: 6 months

  • Loss to follow-up: 0%

Participants
  • Country: France

  • Setting: tertiary single centre

  • Kidney transplant recipients; all CMV serostatus

Treatment group

  • Number: 24

  • Age: NS

  • Sex (M/F): NS

Control group

  • Number: 26

  • Age: NS

  • Sex (M/F): NS

Exclusion criteria: NS

Interventions

Treatment group

  • GCV: 5 mg/kg/d IV for 14 days starting on day of transplant

  • ACV: 800 mg oral 3 times/d from day 14 to 3 months

Control group

  • Placebo: given as for treatment arm

Co-interventions: NS

Outcomes
  1. CMV disease

  2. CMV infection: CMV culture, IgM

  3. All-cause mortality

Notes
  1. Exclusions post randomisation but pre-intervention: None

  2. Stop or end point: NS

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNo information provided except stratification for CMV serostatus
Allocation concealment (selection bias)Low riskAdequate allocation (information received from authors)
Blinding of participants and personnel (performance bias)
All outcomes
Low riskControl group received placebo
Blinding of outcome assessment (detection bias)
All outcomes
Low riskControl group received placebo
Incomplete outcome data (attrition bias)
All outcomes
Low riskAll patients included in analyses
Selective reporting (reporting bias)Unclear riskAbstract only
Other biasUnclear riskWork supported by Wellcome Laboratories and Hospices Civils de Lyon

Reischig 2005 Kidney

Methods
  • Study design: parallel RCT

  • Time frame: April1999 to December 2000; January 2001 to January 2003

  • Follow-up period: 12 months

  • Loss to follow-up: 0%

Participants
  • Country: Czech Republic

  • Setting: tertiary single centre

  • Kidney transplant recipients; D/R+, D+/R-

Treatment group

  • Number: 35

  • Mean age ± SD: 45 ± 12 years

  • Sex (M/F): 26/9

Control group

  • Number: 36

  • Mean age ± SD: 48 ± 11 years

  • Sex (M/F): 25/11

Exclusion criteria

  • D-/R-; unknown CMV status; active CMV infection; treatment with antiviral agents; WBC < 4000; platelets < 150,000; allergy to study drugs

Interventions

Treatment group

  • VACV: 2000 mg oral 4 times/d starting within 3 days of transplant for 3 months

Control group

  • GCV: 1000 mg oral 3 times/d starting within 3 days of transplant for 3 months

Co-interventions

  • ACV low dose to prevent herpes simplex; CSA, MMF, prednisone, ATG or OKT-3 (9), anti-IL2R monoclonal antibody/sirolimus (6)

Outcomes
  1. CMV disease

  2. CMV infection: CMV-DNA, CMV antigenaemia, CMV culture

  3. All-cause mortality

  4. Acute rejection

  5. Graft loss

  6. Adverse reactions

Notes
  1. Exclusions post randomisation but pre-intervention: None

  2. Stop or end point: NS

  3. Additional data requested from authors: data on quality assessment and results obtained

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskRandom number generator used. (Information from authors)
Allocation concealment (selection bias)Low riskAdequate allocation based on information from authors
Blinding of participants and personnel (performance bias)
All outcomes
High riskDifferent medication schedules in each group. Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias)
All outcomes
High riskDifferent medication schedules in each group. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias)
All outcomes
Low riskRandomised consecutive patients. All patients included in analyses
Selective reporting (reporting bias)High riskIncomplete outcome reporting. No data of opportunistic infections
Other biasLow risk"The study was independent and not funded by any commercial sources"

Rondeau 1993 Kidney

Methods
  • Study design: parallel RCT

  • Time frame: January 1990 to July 1992

  • Follow-up period: 3 months

  • Loss to follow-up: 0%

Participants
  • Country: France

  • Setting: tertiary multicentre

  • Kidney transplant recipients; D+/R-

Treatment group

  • Number: 17

  • Mean age ± SEM: 43.8 ± 2.9 years

  • Sex (M/F): 13/4

Control group

  • Number: 15

  • Mean age ± SEM: 43.5 ± 3.3 years

  • Sex (M/F): 6/9

Exclusion criteria

  • Living related donor transplant recipients; WBC < 1500; platelets < 50,000; treatment with another antiviral agent

Interventions

Treatment group

  • GCV: 5 mg/kg IV twice/d for 14 days starting day 14 post-transplant

Control group

  • No treatment

Co-interventions: NS

Outcomes
  1. CMV disease

  2. CMV syndrome

  3. CMV invasive organ disease

  4. CMV infection: CMV culture, IgM

  5. All-cause mortality

  6. Acute rejection

  7. Graft loss

Notes
  1. Exclusions post randomisation but pre-intervention: None reported

  2. Stop or end point: NS

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNo information available
Allocation concealment (selection bias)Unclear risk"On day 14 after transplantation, patients were randomized...". No further information available
Blinding of participants and personnel (performance bias)
All outcomes
High riskControl group received no specific therapy. Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias)
All outcomes
High riskControl group received no specific therapy. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias)
All outcomes
Low riskAll patients included in analyses
Selective reporting (reporting bias)High riskIncomplete outcome reporting. No or limited reporting of opportunistic infections/adverse effects
Other biasLow riskWork supported in part by grants from non-pharmaceutical sources

Rostaing 1994 Kidney

Methods
  • Study design: parallel RCT

  • Time frame: April 1992 to February 1993

  • Follow-up period: mean 12 months

  • Loss to follow-up: 0%

Participants
  • Country: France

  • Setting: tertiary single centre

  • Kidney transplant recipients; D/R+

Treatment group

  • Number: 19

  • Mean age ± SD: 50.4 ± 11.3 years

  • Sex (M/F): 13/6

Control group

  • Number: 18

  • Mean age ± SD: 45.1 ± 11.1 years

  • Sex (M/F): 14/4

Exclusion criteria

  • D+/R-; D-/R- recipients

Interventions

Treatment group

  • ACV: 6 mg/kg/d IV for 3 days starting day 1 then ACV 800 mg oral 4 times/d for 3 months

Control group

  • No treatment

Co-interventions

  • CSA, AZA, prednisone, ATG

Outcomes
  1. CMV disease

  2. CMV syndrome

  3. CMV invasive organ disease

  4. CMV infection: CMV culture

  5. All-cause mortality

  6. Acute rejection

  7. Graft loss

  8. Adverse effects

Notes
  1. Exclusions post randomisation but pre-intervention: None reported

  2. Stop or end point: NS

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNo information available
Allocation concealment (selection bias)Unclear risk"The patients were randomized to receive either acyclovir or nothing..". No other information available
Blinding of participants and personnel (performance bias)
All outcomes
High riskControl group received no specific therapy. Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias)
All outcomes
High riskControl group received no specific therapy. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias)
All outcomes
Low riskAll patients included in analysis
Selective reporting (reporting bias)High riskNo data on opportunistic infections or adverse reactions
Other biasUnclear riskNo information provided about pharmaceutical sponsorship

Rubin 2002 All

Methods
  • Study design: parallel RCT

  • Time frame: November 1996 to January 1999

  • Follow-up period: 12 months

  • Loss to follow-up: 0% of evaluated patients

Participants
  • Country: USA

  • Setting: tertiary multicentre

  • First kidney, liver or heart transplant recipients aged >12 years; D+/R-

Treatment group

  • Number: 77

  • Mean age ± SD: 46 ± 13 years

  • Sex (M/F): 60/17

Control group

  • Number: 78

  • Mean age ± SD: 45 ± 12 years

  • Sex (M/F): 61/17

Exclusion criteria

  • D/R+; D-/R-

Interventions

Treatment group

  • GCV: 5 mg/kg/d IV for 5 to 10 days starting within 72 hours of transplant, then GCV 1000 mg oral 3 times/d to 12 weeks

Control group

  • GCV: 5 mg/kg/d IV for 5 to 10 days starting within 72 hours of transplant then ACV 400 mg oral 3 times/d to 12 weeks

Co-interventions

  • CSA (141), TAC (27), AZA (57), MMF (101), antibody therapy (56)

Outcomes
  1. CMV disease

  2. CMV syndrome

  3. CMV invasive organ disease

  4. CMV infection: CMV antigenaemia, CMV culture

  5. All-cause mortality

  6. Acute rejection

  7. Opportunistic infections

  8. Adverse effects

  9. Time to CMV disease

Notes
  1. Exclusions post randomisation but pre-intervention: None

  2. Stop or end point: NS

  3. 11 (5 acyclovir, 6 ganciclovir) were deemed unable to be evaluated: 7 did not qualify for protocol, 1 died, 3 lost to follow-up

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskStratification for organ transplanted. Central randomisation. Otherwise no information available
Allocation concealment (selection bias)Low riskCentral randomisation
Blinding of participants and personnel (performance bias)
All outcomes
High riskPatients received different oral medications. Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias)
All outcomes
High riskPatients received different oral medications. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias)
All outcomes
Low risk11/166 excluded from analyses. Reasons for missing data unlikely to be related to true outcome
Selective reporting (reporting bias)High riskIncomplete reporting of outcomes. No report of graft loss
Other biasHigh riskFunded in part by a grant from F. Hoffman-LaRoche

Saliba 1993 Liver

Methods
  • Study design: parallel RCT

  • Time frame: February 1990 to February 1991

  • Follow-up period: 3 months

  • Loss to follow-up: 0%

Participants
  • Country: France

  • Setting/Design: tertiary single centre

  • Liver transplant recipients; D/R+

Treatment group

  • Number: 60

  • Mean age ± SD: 45.3 ± 12 years

  • Sex (M/F): 36/24

Control group

  • Number: 60

  • Mean age ± SD: 44.5 ± 13 years

  • Sex (M/F): 35/35

Exclusion criteria

  • D+/R-; D-/R- recipients

Interventions

Treatment group

  • ACV: 500 mg/m²/d IV for 10 days, then 800 mg oral 4 times/d to 3 months

Control group

  • No treatment

Co-interventions

  • CSA, AZA, prednisone

Outcomes
  1. CMV disease

  2. CMV infection: CMV culture

  3. Adverse effects

Notes
  1. Exclusions post randomisation but pre-intervention: None

  2. Stop or end point: NS

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNo information available
Allocation concealment (selection bias)Low riskAdequate allocation concealment (information from authors)
Blinding of participants and personnel (performance bias)
All outcomes
High riskControl group received no specific therapy. Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias)
All outcomes
High riskControl group received no specific therapy. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias)
All outcomes
Low riskConsecutive recruitment. All patients included in analyses
Selective reporting (reporting bias)Unclear riskAbstract only
Other biasUnclear riskNo information provided on pharmaceutical sponsorship

Winston 1995 Liver

Methods
  • Study design: parallel RCT

  • Time frame: NS

  • Follow-up period: 4 months

  • Loss to follow-up: 0%

Participants
  • Country: USA

  • Setting: tertiary single centre

  • First liver transplant recipients aged > 12 years; all serologies

Treatment group

  • Number: 124

  • Mean age (range): 52 years (20 to 72)

  • Sex (M/F): 67/57

Control group

  • Number: 126

  • Mean age (range): 47 years (20 to 74)

  • Sex (M/F): 67/59

Exclusion criteria

  • Second transplants

Interventions

Treatment group

  • GCV: 6 mg/kg/d IV to day 30; GCV 6 mg/kg/d IV Monday to Friday to day 100

Control group

  • ACV: 10 mg/kg IV 8 hourly until discharge; ACV 800 mg oral 4 times/d to day 100

Co-interventions

  • CSA, TAC (38), AZA, prednisone

Outcomes
  1. CMV disease

  2. CMV syndrome

  3. CMV invasive organ disease

  4. CMV infection: CMV culture, isolation from any site

  5. All-cause mortality

  6. Death due to CMV disease

  7. Acute rejection

  8. Opportunistic infections

  9. Adverse effects

Notes
  1. Exclusions post randomisation but pre-intervention: None

  2. Stop or end point: NS

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskRandomisation stratified according to CMV status but no other information provided
Allocation concealment (selection bias)Unclear riskNo information provided
Blinding of participants and personnel (performance bias)
All outcomes
High riskPatients given different medications by different routes. Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias)
All outcomes
High riskPatients given different medications by different routes. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias)
All outcomes
Low riskAll patients included in analyses
Selective reporting (reporting bias)Low riskAll expected outcomes reported
Other biasHigh riskSupported in part by non-pharmaceutical grants. Ganciclovir from Syntex Research

Winston 2003 Liver

Methods
  • Study design: parallel RCT

  • Time frame: NS

  • Follow-up period: 12 months

  • Loss to follow-up: 0%

Participants
  • Country: USA

  • Setting: tertiary single centre

  • Liver transplant recipients; D/R+

Treatment group

  • Number: 110

  • Mean age (range): 51 years (7 to 78)

  • Sex (M/F): 58/52

Control group

  • Number: 109

  • Mean age (range): 51 years (7 to 71)

  • Sex (M/F): 58/51

Exclusion criteria

  • D+/R-; D-/R- recipients

Interventions

Treatment group

  • GCV: 6 mg/kg/d IV to day 14 starting day of transplant; GCV 1000 mg oral 3 times/d to day 100

Control group

  • GCV: 6 mg/kg/d IV to day 14 starting day of transplant; ACV 800 mg oral 4 times/d to day 100

Co-interventions

  • CSA (58), TAC (164), AZA (128), MMF (85), prednisone

Outcomes
  1. CMV disease: CMV DNA, CMV culture

  2. CMV syndrome

  3. CMV tissue invasive disease

  4. All-cause mortality

  5. Death due to CMV disease

  6. Acute rejection

  7. Opportunistic infections

  8. Adverse effects

Notes
  1. Exclusions post randomisation but pre-intervention: Unclear

  2. Stop or end point: NS

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNo information provided
Allocation concealment (selection bias)Unclear risk"Patients were assigned randomly" but no other information available
Blinding of participants and personnel (performance bias)
All outcomes
High riskDifferent interventions given to groups with different dose frequency. Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias)
All outcomes
High riskDifferent interventions given to groups with different dose frequency. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias)
All outcomes
Low riskAll patients included in analyses
Selective reporting (reporting bias)High riskIncomplete reporting of outcomes. No report of CMV infection and graft loss
Other biasHigh riskSupported in part by a research grant from Roche Laboratories

Winston 2004 Liver

  1. a

    ACV - aciclovir; AIDS - acquired immunodeficiency syndrome; ALG - antilymphocyte globulin; AT - antithymocyte globulin; AZA - azathioprine; CD/LD - cadaveric donor/living donor; CMV, cytomegalovirus; CMVIgG - cytomegalovirus gamma G immunoglobulin; CSA - cyclosporin; D/R+ - donor CMV positive or negative/recipient CMV positive; D+/R- - donor CMV positive/recipient CMV negative; D-/R- - donor CMV negative/recipient CMV negative; DNA - deoxyribonucleic acid; GCV - ganciclovir; GFR - glomerular filtration rate; HI -, human immunovirus; IgG - immunoglobulin G; IgM - immunoglobulin M; IgM 3 - immunoglobulin M 3; IL2Ra - interleukin 2 receptor alpha; IQR - interquartile range; ITT - intention-to-treat; IV - intravenous; MMF - mycophenolate mofetil; NS - not stated; OKT-3 - monoclonal anti CD3 antibody; PCR - polymerase chain reaction; TAC - tacrolimus; VACV - valaciclovir; VGCV - valganciclovir; WBC - white blood cell

Methods
  • Study design: parallel RCT

  • Time frame: June 1997 to April 2000

  • Follow-up period: 1 year

  • Loss to follow-up: 0%

Participants
  • Country: USA

  • Setting: tertiary single centre

  • Liver transplant recipients; D+/R-

Treatment group

  • Number: 32

  • Mean age (range): 49 years (13 to 67)

  • Sex (M/F): 24/8

Control group

  • Number: 32

  • Mean age (range): 46 years (6 to 73)

  • Sex (M/F): 23/9

Exclusion criteria

  • D/R+; D-/R-

Interventions

Treatment group

  • GCV: 6 mg/kg IV daily days 1 to 14; GCV 1000 mg oral 3 times/d on days 15 to 86

Control group

  • GCV: 6 mg/kg IV daily days 1 to 14; GCV 6 mg/kg IV Monday to Friday from days 15 to 86

Co-interventions

  • CSA (10), TAC (54), MMF (29), AZA (3), prednisone

Outcomes
  1. CMV disease

  2. CMV syndrome

  3. CMV tissue invasive disease

  4. All-cause mortality

  5. Opportunistic infections

  6. Adverse effects

Notes
  1. Exclusions post randomisation but pre-intervention: None

  2. Stop or end point: NS

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNo information available
Allocation concealment (selection bias)Unclear risk"Randomized controlled trial" in title but no other information provided
Blinding of participants and personnel (performance bias)
All outcomes
High riskDifferent interventions given to groups. Primary outcome of CMV disease could be influenced by lack of blinding
Blinding of outcome assessment (detection bias)
All outcomes
High riskDifferent interventions given to groups. Primary outcome of CMV disease could be influenced by lack of blinding
Incomplete outcome data (attrition bias)
All outcomes
Low riskAll patients followed for 1 year or until death
Selective reporting (reporting bias)High riskIncomplete outcome reporting. No report of CMV infection, graft loss
Other biasUnclear riskSupported in part by research grant from Roche Laboratories

Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion
  1. a

    ATG - antithymocyte globulin; CMV - cytomegalovirus; GCV - ganciclovir; HLA - human leukocyte antigen; IgG - Immunoglobulin G; RCT - randomised controlled trial

Ahsan 1998Not RCT (sequential)
Arbo 2000Economic evaluation of previous study
Brennan 1997Pre-emptive study
Brennan 2001Review article
Devolder 2010Ineligible intervention. Compares different methods to encourage compliance
Dickinson 1996IgG to prevent CMV
Falagas 1997Included both non-randomised patients and patients from a previous study
Fehir 1989Nonrandomised patients included
Ferreira 2004Prospective study of different immunosuppressive regimens. Not RCT
Fishman 2000Retrospective study
Gerna 2003Diagnostic test systematic review
Gerna 2008Pre-emptive therapy compared with prophylaxis
Greger 1988Ineligible intervention
Griffiths 1997Review article
Griffiths 2010Study of pre-emptive therapy vs. monitoring
Grundmann 1986Ineligible intervention. CMV IgG
Hecht 1988Not an RCT
Huurman 2006RCT of ATG versus daclizumab, not antiviral medication
Jung 2001Pre-emptive study
Jurim 1996Subgroup of previous study; outcome hepatitis B
Khoury 2006Pre-emptive study
Kim 2000Economic evaluation of previous study
Kletzmayr 2000Not RCT. Historical controls
Kliem 2008Pre-emptive study
Koetz 2001Pre-emptive study
Kuypers 1999Review article
Laske 1991Review article
Laske 1992Review article
Luan 2009Retrospective study
Lumbreras 1993Not RCT. Historical controls
MacDonald 1991Ineligible intervention. CMV IgG
Marker 1980Treatment not prophylaxis of CMV disease
Martin 1993Review article
Martin 1994Review article
Martin 1995Review article
Mattes 2004Ineligible intervention. Comparing 2 pre-emptive regimens. Results cannot be separated for bone marrow and solid organ transplant recipients
McGavin 2001GCV review
Moreno 1999Not RCT
Mullen 1998Retrospective study
Murray 1997Pre-emptive study
Paya 2002Pre-emptive study
Pescovitz 2009Pharmacokinetic study
Pouteil 1991Study of influence of HLA on CMV infection within RCT of different immunosuppressive regimens
PROTECT Study 2010Comparing pre-emptive therapy with prophylaxis
Queiroga 2003Pre-emptive study
Rayes 2001Pre-emptive study
Reischig 2008Pre-emptive study
Sagedal 2003Pre-emptive study
Said 2007Appears to be sequential study not RCT
Schafers 1988Not RCT (sequential)
Schnitzler 2000Re-analysis of previous study (1992)
Singh 1994Pre-emptive study
Singh 1995Not RCT
Singh 2000Pre-emptive study
Snydman 1991aReview article
Snydman 1991bCompares results to previous study
Snydman 1994Compares results to previous study
Snydman 2001Historical controls
Speich 1999Not RCT (sequential)
Stratta 1992Non-randomised patients included
Tong 2002Not an RCT
Turgeon 1998Not RCT (sequential)
Valantine 1995IgG study
VICTOR Study 2007Treatment of CMV disease not prophylaxis
Yang 1998Pre-emptive study
Yang 1999Unable to determine if patients randomised

Characteristics of ongoing studies [ordered by study ID]

Villano 2010

Trial name or titleA randomized, double-blind study to assess the efficacy and safety of prophylactic use of maribavir versus oral ganciclovir for the prevention of cytomegalovirus disease in recipients of orthotopic liver transplants
Methods
  • Allocation: randomized

  • Endpoint classification: safety/efficacy study

  • Intervention model: parallel assignment

  • Masking: double blind (subject, caregiver, investigator)

  • Primary purpose: prevention

Participants

Inclusion criteria

  • Male and female, ≥ 18 years

  • Orthotopic liver transplant recipient

  • Donor CMV seropositive / Recipient CMV seronegative

  • Enrolled within 10 days after liver transplant

  • Able to swallow tablets

Exclusion criteria

  • Multiple organ transplant

  • HIV infection

  • CMV disease

  • Use of other anti-CMV therapy at time of enrolment

Interventions
  • Maribavir: 100 mg twice a day for 14 weeks

  • Ganciclovir: 1000 mg 3 times/d for 14 weeks

Outcomes
  • CMV disease 6 months post-transplant

  • CMV disease 100 days and 12 months post-transplant

  • Incidence of CMV infection 100 days and 12 months post-transplant

  • Incidence of graft rejection 100 days and 12 months post-transplant

  • Incidence of retransplantation 100 days and 12 months post-transplant

  • Mortality 100 days and 12 months post-transplant

Starting dateJuly 2007
Contact informationStephen Villano, MD, Viropharma, Inc.
NotesStudy completed 2009