Incidence, Clinical Characteristics and Risk Factors of Late Infection in Solid Organ Transplant Recipients: Data from the RESITRA Study Group

Authors


  • This paper was partially presented at the 45th Interscience Congress on Antimicrobial Agents and Chemotherapy, Washington (16–19 December 2005), oral presentation K-1811.

*Corresponding author: Rafael San Juan Garrido, rafasjg@yahoo.es

Abstract

Information describing the incidence and clinical characteristics of late infection (LI) in solid organ transplantation (SOT) is scarce. The aim of this study was to define the incidence, clinical characteristics and risk factors for LI (>6 months) as compared with infection in the early period (<6 months) after SOT. By the online database of the Spanish Network of Infection in Transplantation (RESITRA) we prospectively analyzed 2702 SOT recipients from September 2003 to February 2005. Univariate and multivariate analysis using logistic regression were performed to calculate the risk factors associated with the development of LI. A total of 131 patients developed 176 LI episodes (8%). Global incidence of LI was 0.4 per 1000 transplant-days, ranging from 0.3/1000 in kidney transplants to 1.4 in lung transplants. Independent risk factors for LI in were: acute rejection in the early period (OR 1.5; CI 95%: 1.1–2.3), chronic graft malfunction (OR 2; CI 95%: 1.4–3), re-operation (OR 1.9; CI 95%: 1.3–2.8) relapsing viral infection apart from CMV (OR 1.9; CI 95%: 1.1–3.5), previous bacterial infection (OR 1.8; CI 95%: 1.2–2.6) and lung transplantation (OR 4.5; CI 95%: 2.6–7.8). Severe LI occurs in a subgroup of high-risk SOT recipients who deserve a more careful follow-up and could benefit from prolonged prophylactic measures similar to that performed in the early period after transplantation.

Introduction

According to the classical timetable of infection in solid organ transplantation (SOT) the risk for nosocomial and opportunistic infections has been considered to be maximum during the first 6 months (1). Indeed, prophylaxis and preemptive treatment strategies are mostly directed to that period of time. Late infection (LI) occurring after the sixth month of transplantation is supposed to be caused by the same pathogens that in other patients with chronic conditions (1), but there is scarce information about which is the real incidence, etiology and risk factors for LI in SOT recipients (2–9).

The aims of the present study were to calculate the incidence of LI in a broad prospective cohort of SOT recipients, to describe clinical characteristics of such infectious episodes, and to analyze the risk factors for developing LI.

Methods

RESITRA network

RESITRA is a Research Network financed by the Spanish Ministry of Health, formed by 16 Transplant Centers and 3 Reference Laboratories that prospectively includes all SOT and Haematopoietic Stem Cell transplant (HSCT) recipients from these hospitals. It was created in April 2003. Each hospital has a research team (MD, nurse and technician) who manage the research tools. An online database and a serum, plasma, microorganism and DNA bank are used as research tools. Pre-, peri-transplant and follow-up (days 0, 7, 14, 30, 60, 90, 180, 270, 360 and 720 after transplantation) data are prospectively included in the online database, as well as all infections (diagnostic work-up, clinical presentation, therapy and outcome) and rejection episodes, always according to the predefined definitions published in the RESITRA website (http://www.resitra.retics.net/web/index.asp) (10) (based on universally accepted criteria). Data-collection is made through PDF e-forms, which are sent to an SQL-server database located in the website. Data-exploitation is made through managerial and statistical databases generated from the SQL-server database, after a validating process managed by each Hospital coordinator. From April to July 2003, research tools were created and implemented, and at the end of July 2003, the first patients were recruited.

For the present study, we included all SOT recipients included in the RESITRA prospective cohort from September 2003 to the end of February 2005 with a minimum follow-up of 3 months for the calculation of infection incidence and a minimum of 6 months of follow-up for the analysis of risk factors of LI.

Infection episode definition:  We reported all bacterial, fungal, viral and parasitic infectious episodes in accordance to the clinical and microbiological criteria included in the predefined RESITRA definitions (10) occurring in SOT recipients included in the RESITRA cohort that required at least 24 h of hospital management. We also included episodes of asymptomatic CMV viraemia that required antiviral treatment (preemptive therapy). Asymptomatic bacteriuria and non-complicated cystitis episodes were not included. Recrudescence of B or C hepatitis in patients was not considered as a new infectious episode in the present study. Those episodes occurring in the fist 6 months after transplantation were considered as early infections (EI) and those occurring after the sixth month were considered LIs (LI).

Incidence of infection:  We used as denominator the number of SOT recipients with a minimum of 3 months of follow-up for the calculation of the incidence of EI and all the SOT recipients of the cohort with a follow-up over 6 months for the calculation of the incidence of LI (expressed in cases of infection per 100 SOT recipients). We also calculated the incidence per 1000 transplantation-days using as denominator the number of days at risk. The number of transplantation days consisted in the sum of all the days at risk of each SOT recipient: from the transplantation day to the day 180 or the end of follow-up (31 March 2004, death or graft loss) in the early period and from the day 180 to the end of follow-up in the late period (31 March 2004, death or graft loss).

Analyzed variables as possible risk factors for developing LI were divided into the following groups: patient characteristics, and pretransplantation, intraoperative and postoperative variables. The preoperative period referred to the month immediately before transplantation. Patient characteristics encompassed the following variables: age, sex, diabetes, chronic obstructive lung disease (COPD) and previous transplant. Pretransplantation variables (1 month before transplantation) included type and severity of underlying disease, renal failure (serum creatinine >2.5 mg/dL for more than 7 days) and hemodialysis, leukopenia (<3000 leukocytes/mm3), ICU admission with or without mechanical ventilation, CMV, HBV and HCV serological condition, donor/recipient matching, corticosteroid, antibiotic or antifungal therapy for more than 7 days, previous latent or active tuberculosis infection of the recipient and donor age and presence of donor infection around transplantation time. Intraoperative factors included length of transplant surgery, cold ischemia time of the graft, the total number of blood product units given during surgery, specific technical procedures (single or double renal transplant, type of biliar derivation in liver or pancreas transplant) and type of antibiotic prophylaxis. The following postoperative factors were analyzed: length of ICU stay, mechanical ventilation and vascular amine requirements, reoperation (new surgical procedure), urgent retransplantation (in the first 7 days after transplant), repeat ICU stay (excluding patients with retransplantation or invasive aspergillosis) with or without mechanical ventilation, renal failure, hemodialysis, recrudescence of the underlying disease (including hepatitis virus recrudescence in liver transplant), early bacterial infections, recrudescence of the underlying disease (including hepatitis virus recrudescence in liver transplant), early bacterial infections, CMV prophylaxis (CMV treatment in high-risk patients without any evidence of CMV infection), early CMV infection or disease, fungal prophylaxis and early fungal infection and Pneumocystis jirovecii prophylaxis. The immune-related factors included: crossmatching for HLA compatibility, leukopenia, acute and chronic rejection, type of immunosuppressive treatment (including administered for acute rejection episodes) and chronic graft malfunction.

Data analysis

Continuous variables were expressed as the mean and standard deviation (SD) for those with a normal distribution, and as the median (mostly for those with a skewed distribution). Discrete variables were expressed as percentages.

Student's unpaired t-test was used to compare continuous variables, the Mann–Whitney U-test to compare continuous variables with non-normal distribution, and the chi-squared or Fisher exact test to compare proportions. All statistical tests were two-tailed and the threshold of statistical significance was p < 0.05.

Odds ratios (ORs) were calculated for variables with statistically significant differences between patients with or without LI. Binary logistic regression was applied individually to each variable to obtain the OR in the univariate analysis. Quantitative variables were previously converted into qualitative variables for that task.

Statistically significant variables (p < 0.05) in the univariate analysis were introduced in a multivariate model by use of forward stepwise logistic regression (SPSS 12.0) to identify the independent risk factors for LI. In addition, clinically relevant factors with p-values <0.1 that were considered to be potential confounders (on the basis of experience and the literature) were forced into the multivariate model to investigate their effect. We analyzed the global risk factors of LI and the specific risk factors for the transplantation groups with most included patients (kidney-transplant, liver and heart transplant).

Results

A total of 2702 SOT recipients were included in the RESITRA cohort from September 2003 to February 2005. A total of 2169 patients reached the late period (above 6 months of follow-up), with a mean follow-up of 400 days (SD 128 days) and a median of 402 days.

Incidence and type of infections in the early and late period

We compared the incidence of infection and the distribution of the infectious syndromes between the early and late period posttransplantation (Tables 1–3). In the early posttransplant period (<6 months) we also analyzed the infections occurring during the first month (more linked to the surgical procedure) and compared them with those that were reported between the second and the sixth month after transplant (more related to immunosuppression). As is shown in Tables 1 and 2, the incidence of early infection (3.5 episodes per 1000 transplant-days) was significantly higher than in the late period (0.4 episodes per 1000 transplantation days) (p < 0.0001). The incidence of LI ranged from less than 0.3/1000 transplant-days in kidney transplant to 1.4 in lung transplant.

Table 1.  Incidence of infection per transplant type
Transplant typeEarly periodLate period
First month after Tx Incidence1Second to sixth month after Tx Incidence1Over six months after Tx Incidence1
  1. Tx = transplant.

  2. 1Episodes/1000 transplant-days.

Heart8.782.330.34
Liver11.521.900.31
Kidney-pancreas14.474.270.76
Lung13.263.291.4
Kidney4.912.020.28
Total8.272.130.37
Table 2.  Incidence of infectious syndromes in the early and late periods
Infectious syndromesEarly periodLate period
First month after Tx Total: 755 No episodes (%)Second to sixth month after Tx Total: 835 No episodes (%)Over six months after Tx Total: 176 No episodes (%)
  1. Tx = transplant; CMV = cytomegalovirus. Episodes/1000 transplant-days are shown.

  2. 1Includes mostly CMV viral syndrome without organ affectation.

  3. 2Non-complicated cystitis or asymptomatic bacteriuria excluded.

Viral symptomatic disease10.1 0.260.01
Lower respiratory tract infection1.960.10.09
Skin and soft tissue infection1.620.210.04
Asymptomatic CMV viraemia requiring treatment0.230.230.03
Genitourinary infection20.860.20.04
Intraabdominal infection0.990.160.04
Endovascular infection (also catheter-related)0.840.10.01
Upper respiratory tract infection (including tracheobronchitis)0.830.10.03
Bacteremia and/or disseminated infection0.680.10.02
Colitis and/or gastroenteritis0.220.110.02
Stomatitis and/or oesophagitis0.280.080.008
CNS infection0.01
Bone and/or joint infection0.0010.002
Table 3.  Infectious syndromes distribution and isolated microorganisms in the early and late periods
Infectious syndromesEarly periodLate periodp-value1
First month after Tx Total: 757 No episodes (%)Second to sixth month after Tx Total: 838 No episodes (%)Over 6 months after Tx Total: 176 No episodes (%)
  1. CMV = cytomegalovirus; NFGNB = non-fermenting Gram-negative bacilli; Tx = transplant. Episodes/1000 transplant-days are shown. 1Comparison between rates of infection distribution before and after the sixth month. Only p-values below 0.1 are shown.

  2. 2Includes mostly CMV viral syndrome without organ affectation.

  3. 3Non-complicated cystitis or asymptomatic bacteriuria excluded.

  4. 4Mostly Herpes-simplex and Herpes-zoster viruses.

  5. 5Includes filamentous fungi, BK virus, mycobacteria, Listeria, Nocardia, Legionella, P. jirovecii, Leishmania.

Viral symptomatic disease220(2.6%)119 (14.2%)7 (4%)p < 0.0001
Lower respiratory tract infection165 (21.8%)96 (11.5%)45 (25.6%)p < 0.0001
Skin and soft tissue infection136 (18%)96 (11.5%)19 (10.8%) 
Asymptomatic CMV viraemia requiring treatment20 (2.6%)106 (12,6%)16 (9.1%) 
Genitourinary infection373 (9.6%)93(11,1%)19 (10.8%) 
Intraabdominal infection83 (11%)71 (8.5%)19 (10.8%) 
Endovascular infection (also catheter-related)71 (9.3%)47 (5.6%)5 (2.8%)0.06
Upper respiratory tract infection (including tracheobronchitis)69 (9.2%)46 (5.4%)15(8.5%) 
Bacteremia and/or disseminated infection57 (7.5%)48 (5.7%)10 (5.7%) 
Colitis and/or gastroenteritis18 (2.5%)52 (6.2%)11 (6.3%) 
Stomatitis and/or oesophagitis24 (3.2%)38 (4.6%)4 (2.3%) 
CNS infection03 (0.4%)0 
Bone and/or joint infection01 (0.1%)1 (0.6%) 
Others21(2.7%)22 (2.6%)5 (2.8%) 
Total: 711Total: 640Total: 126 
No isolates (%)No isolates (%)No isolates (%) 
Isolated microorganisms
 Gram-positive bacteria272 (38.2%)144 (22.5%)23 (18.3%)0.02
 CMV39 (5.4%)150 (23.4%)26 (20.6%) 
 Enteric bacteria167 (23.5%)140 (21.9%)25 (19.8%) 
 NFGNB and other Gram-negative bacteria135 (19%)96 (15%)15(11.9%) 
 Other virus apart from CMV456 (7.9%)61(9.5%)20 (15.9%)0.002
 Yeast31 (4.4%)32 (5.1%)3 (2.4%) 
 Other immunosuppression related microorganisms511 (1.6%)17 (2.6%)14 (11.1%)<0.0001

Even during the early period the incidence of infection declined dramatically after the first month (from 8.27 episodes per 1000 transplant-days to 2.12 episodes per 1000 transplant-days in the 2–6 month period; p < 0.0001) except for CMV disease, that was significantly more frequent in the second to sixth month after transplant period.

Distribution of infectious syndromes (Table 3) was similar in the early and late period except for viral symptomatic disease (mostly CMV disease) and endovascular infection that were proportionally more frequent in the early period and lower respiratory tract infection, which was proportionally more prevalent in the late period. The related mortality of the infectious episodes was similar in both periods (5.5% in early period vs. 8.5% in the late period; p = 0.6).

Isolated microorganisms in both periods and their proportional distribution are also resumed in Table 3. Infection by Gram-positive bacteria predominated in the early period and infection by Herpes-simplex, Herpes-zoster and other microorganisms related to underlying immunosuppression predominated in the late period.

Risk factors for LI in SOT

A total of 131/2169 SOT recipients (6%) suffered from LI. Per type of transplanted organ, LI occurred in 14/206 heart recipients (6.8%), 40/707 liver recipients (5.7%), 55/1092 kidney recipients (5%), 4/55 kidney-pancreas recipients (9.1%) and 18/120 lung recipients (15%). The incidence of infectious syndromes and the distribution of isolated microorganisms per transplant type are shown in Tables 4 and 5.

Table 4.  Incidence1 of infections in the late period per transplant type
 KidneyKidney-pancreasLiverHeartLung
  1. CMV = cytomegalovirus.

  2. 1Episodes/1000 transplant-days.

  3. 2Non-complicated cystitis or asymptomatic bacteriuria excluded.

  4. 3Includes mostly CMV viral syndrome without organ affectation.

Infectious syndromesTotal: 70Total: 7Total: 48Total: 16Total: 35
Lower respiratory tract infection0.060.110.070.110.53
Intraabdominal infection0.010.110.1100
Skin and soft tissue infection0.0300.030.060.16
Genitourinary infection20.050.110.0300.04
Asymptomatic CMV viraemia requiring treatment0.0300.010.040.24
Upper respiratory tract infection (including tracheobronchitis)0.02000.040.28
Viral symptomatic disease30.020.2210.040
Bacteremia and/or disseminated infection0.010.110.0300
Colitis and/or gastroenteritis0.020.110.020.020.04
Endovascular infection (also catheter-related)0.0100.0100
Stomatitis and/or oesophagitis0.010000.08
Table 5.  Isolated microorganisms distribution in the late period per transplant type
 Kidney (%)Kidney-pancreas (%)Liver (%)Heart (%)Lung (%)
  1. CMV = cytomegalovirus.

  2. 1Includes mostly CMV viral syndrome without organ affectation.

  3. 2Mostly Herpes-simplex and Herpes-zoster viruses.

  4. 3Includes filamentous fungi, BK virus, mycobacteria, Listeria, Nocardia, Legionella, P. jirovecii, Leishmania.

Microorganisms isolatedTotal: 54Total: 7Total: 27Total: 11Total: 27
Gram-positive bacteria9 (16.7)3(43)4 (14.8)1 (9.1)4 (14.8)
CMV111 (20.4)1 (11)06 (54.5)8 (29.6)
Enteric bacteria12 (22.2)1 (11)10 (37)1 (9.1)1 (3.7)
NFGNB and other Gram-negative bacteria4 (7.4)  1 (11)4 (14.8)06 (22.2)
Other virus apart from CMV219 (18.5)1 (11)4 (14.8)2 (18.2)3 (11.1)
Yeast1 (1.9)  02 (7.4) 01 (3.7)
Other immunosuppression related microorganisms37 (13)  03 (11.1)1 (9.1)4 (14.8)

In Table 6 the principal characteristics of the 2169 SOT recipients followed-up for more than 6 months are shown. The data from 131 patients with LI were compared with those from the remaining 2038 SOT recipients who did not suffer any significant infectious episode in the late period. The mean time to developing LI was 301 days (SD 103) with a median of 267 days. On the other hand the mean time of follow-up of the patients in the control group was of 397 days (SD 128 days) with a median of 398 days.

Table 6.  Characteristics of SOT recipients included in the study cohort (follow-up over 6 months)
  1. SOT = solid organ transplant; Pre-Tx = previous to transplantation; CMV = cytomegalovirus; HCV = hepatitis C virus; ATG = anti-tymocyte globulin; OKT3 = anti-CD3 monoclonal antibody; GCV = gancyclovir; VGCV = valgancyclovir; Tx = Transplant.

  2. 1Mainly for Herpes-simplex infection prophylaxis.

  3. 2Intravenous gancyclovir during two weeks followed or not by oral valgancyclovir during 3 months in high-risk patients without any evidence of CMV infection.

Sex (M/F) (%)59/41
Pre-Tx diabetes16.8%
Donor age (years)47 ± 18
Recipient age (years)50 ± 14
Transplantation type
 Kidney1092 (50.3%)
 Liver707 (32.6%)
 Heart206 (9.5%)
 Lung120 (5.5%)
 Kidney-pancreas44 (2%)
 CMV serological mismatch (seropositive donor/seronegative recipient)179 (8.3%)
 HCV infection365(17%)
Immunosuppression including
 Cyclosporine613 (28.3%)
 Tacrolimus1029 (47.5%)
 Mycophenolate1329 (61.4%)
 Rapamycin122 (5.6%)
 Azatioprin84 (3.9%)
 ATG/OKT3108 (5%)
 Basiliximab/daclizumab577 (26.6%)
 Investigational immunosuppressive drugs26 (1.2%)
 Anti-calcineurinic free regimen498 (23%)
 Maintenance corticosteroids free regimen124 (5.7%)
 Major changes imunosuppression during the follow-up48 (2.2%)
Antimicrobial prophylaxis including:
 Prophylaxis against P. jirovecii1714 (79%)
 Fluconazole prophylaxis (first 3 months after Tx)477 (22%)
 Prophylaxis against filamentous fungi (first 3 months after Tx)297 (13.7%)
 Prophylaxis with acyclovir (first 3 months after Tx)1187 (8.6%)
 GCV/VGCV prophylaxis in the first 3 months2557 (25.7%)
 Allograft chronic dysfunction539 (24.9%)
 At least one episode of acute rejection475 (21.9%)
 Global mortality27 (1.2%)

Following a multivariable analysis model the risk factors related to LI after SOT (Table 7) were: lung transplantation (OR 4.5; 2.6–7.8), chronic graft malfunction (OR 2; 1.4–3), reoperation (OR 1.8; 1.1–3.2), positivity of anti-HBc serology in the donor (OR 2.21; 1–4.1), and bacterial infection (OR 1.8; 1.2–2.6), viral infection other than CMV (OR 1.9; 1.1–3.5) or 1 or more episodes of acute rejection (OR 1.4; 0.99–2.2) during the early period. Antimicrobial prophylaxis had no influence in the development of LI. Although CMV prophylaxis appeared to be a risk factor of LI by univariate analysis, this factor was not retained in the final multivariate model. P. jirovecii prophylaxis was performed in more than 80% of patients in all transplant types. Antimicrobial prophylaxis differed regarding the type of SOT, as is shown in Table 8. We did not find significant differences in the spectrum of infections between patients with or without antimicrobial prophylaxis.

Table 7.  Global risk factors for late infection in SOT
 Univariate analysisMultivariate analysis
ORCI 95%p-valueORCI 95%p-value
  1. 1Mostly in the immediate post-Tx period. CMV = cytomegalovirus; anti-HBc Ab = anti-core HBV antibodies.

  2. 2Mostly Herpes-simplex and Herpes-zoster virus infection. HBV = hepatitis B virus; HCV = hepatitis C virus.

Fluconazole prophylaxis11.50.9–2.30.08 
Prophylaxis against CMV1.51.1–2.20.036 
Anti-HBc Ab positive donor1.91.1–3.50.022.21.1–4.10.02
Lung transplantation3.82.3–6.3<0.00014.52.6–7.8<0.0001
Use of tacrolimus0.670.5–0.90.03 
Reoperation for dehiscence or stenosis2.31.4–40.0021.91.3–2.80.04
Bacterial infection (early period)2.21.6–3.2<0.00011.81.2–2.60.002
Viral infection apart from CMV (early period)22.11.2–3.80.011.91.1–3.50.04
Acute rejection episode (early period)21.4–3<0.00011.51.1–2.30.02
Chronic graft malfunction1.91.3–2.70.0121.4–3<0.0001
Table 8.  Percentage of SOT recipients receiving antimicrobial prophylaxis per transplant type
 Kidney (%)Kidney-pancreas (%)Liver (%)Heart (%)Lung (%)
  1. GCV = gancyclovir; VGCV = valgancyclovir; SOT = solid organ transplant.

  2. 1Mainly for Herpes-simplex infection prophylaxis.

  3. 2Intravenous gancyclovir during two weeks followed or not by oral valgancyclovir during 3 months in high-risk patients without any evidence of CMV infection.

Prophylaxis against P. jirovecii7895709495
Fluconazole prophylaxis (first 3 months after Tx)2.575331551
Prophylaxis against filamentous fungi(first 3 months after Tx)1026193140
Prophylaxis with acyclovir(first 3 months after Tx)11.69 7 3721
GCV/VGCV prophylaxis in the first 3 months22384133290

A more detailed analysis of risk factors for LI in each type of transplanted organ (Table 9) confirmed that some factors (as chronic graft malfunction and previous infectious episodes in the early period) were common for many SOT recipients, but there were specific risk factors for each transplant group. Thus, CMV mismatch and the use of sirolimus as immune-suppressor drug were independent risk factors in kidney and pancreas transplant recipients. Twelve out of 118 (10%) single kidney transplant recipients who received sirolimus as initial immunosuppressive drug (not as a salvage treatment) developed episodes of LI (pneumonia: 4, acute pyelonephritis: 3, surgical site infection: 1, catheter-related bacteremia: 1, herpetic esophagitis: 1, localized Herpes-zoster: 1 and acute parotiditis: 1).

Table 9.  Risk factors for late infection in each type of SOT
 Univariate analysisMultivariate analysis
ORCI 95%p-valueORCI 95%p-value
  1. 1Renal failure episodes after the second posttransplant week are considered. CMV = cytomegalovirus; anti-HBc Ab = anti-core HBV antibodies.

Kidney and pancreas
 CMV mismatch2.71.4–5.50.0042.41.3–4.60.006
 Sirolimus use1.90.99–3.90.062.31.1–4.80.024
 At least two episodes of bacterial infection (early period)3.62–6.5<0.00013.62–6.6<0.0001
 Renal failure episode in the early period12.40.9–5.90.053 
 Chronic graft malfunction2.61.5–4.5<0.00012.71.5–4.7<0.0001
Liver
 Anti-HBc Ab positive donor3.81.7–8.40.0013.31.3–8.30.01
 Jejune bile derivation4.61.6–130.0044.81.6–14.30.02
 Bacterial infection (early period)2.51.3–4.70.007 
 Viral infection apart from CMV (early period)3.41.4–8.10.00631.2–7.60.03
 Acute rejection episode (early period)2.21.2–4.30.02 
 Chronic graft malfunction2.41.2–4.50.0092.31.2–4.5<0.0001
Heart
 Previous cardiac valves disease4.91.3–170.014.361.1–17.20.035
 Chronic pulmonary obstructive disease4.51.1–180.036.81.47–310.001

In liver recipients, some distinctive risk factors for LI apart from chronic graft malfunction were detected. Thus, positivity of anti-HBc serology in the donor was confirmed as a risk factor for LI in this subgroup of patients (OR 3.3; CI 1.3–8.3). The incidence of LI was 12.5% (7/56) in liver transplant recipients with anti-HBc positive donors in comparison with 5% (32/650) in those with anti-HBc seronegative donors (p = 0.4). Other identified risk factor was biliar derivation to jejune, which was made on 38 patients in the entire cohort of liver transplant patients; 5 of them (13.2%) developed LI compared with 3% in the other 668 patients from the cohort with choledoccal derivation (p = 0.0046). As was to be expected these 5 patients developed infectious episodes related to biliary tract (colangitis 4, and secondary peritonitis 1) and 1 patient developed bacterial pneumonia.

In heart transplant recipients only two risk factors for LI could be detected in the multivariable analysis: COPD and previous valve disease. These factors probably reflexes a worse functional cardiorespiratory state in these patients which conditioned that respiratory tract infection was the most frequent diagnosed syndrome (8/17 episodes; 47%). More than the half of the episodes (9/17; 57%) had viral etiology including six episodes due to CMV infection (three episodes of CMV viral syndrome, one episode of CMV pneumonitis, one episode of CMV retinitis and one episode of asymptomatic CMV viraemia requiring preemptive treatment), two episodes of single-segmental Herpes-zoster infection and one episode of Herpes-simplex virus stomatitis.

Although lung transplant had the highest rate of infection we could not find any significant difference in the analyzed variables between patients with or without late infection probably due to the limited number of patients included.

Discussion

Currently the rate of LI in SOT come from reports of different series in the 1990s (1) that lack of uniform definition criteria of infection and provide figures expressed in percentages of patients with infection and therefore not adjusted by the time at risk.

To the best of our knowledge, the present study—performed with data obtained from a large prospective cohort provided by the RESITRA research network—is the first one that has prospectively evaluated the incidence of LI in SOT adjusted by the time at risk. The global incidence was 0.4 episodes per 1000 transplantation-days. Heart, kidney and liver transplant recipients presented the lowest rates (around 0.3 per 1000 transplant-days) whereas pancreas and lung transplant recipients had the highest incidence rates (0.76 and 1.4 per 1000 transplant-days, respectively).

One of the most relevant findings of this study was that, against what has been classically reported (1), there were only slight differences in the etiology of infectious complications occurring early or late after transplantation, and even most important, the infection-related mortality was similar, as was previously reported in a study analyzing the long-term outcome of liver transplant recipients (8). Therefore, although infection if much less common after the sixth month of transplantation it is as severe as in the early period. Fortunately these late infectious episodes occur only in a subgroup of high-risk patients. Patients with chronic graft malfunction, previous infectious episodes in the first 6 months of transplantation and with graft-related reoperations are at higher risk of LI. On the other hand, some organ-specific risk factors were identified in this study. Thus, in single kidney transplant recipients primary immunosuppressive treatment with sirolimus was independently related with a higher risk of LI. Sirolimus-containing regimens have shown higher surgical wound complication rates than other regimens because of a poorer wound healing (mostly lymphocele formation) in renal transplant recipients (11–13) and have recently been related with a higher risk of global infection in liver transplant (14). Anyway, the accurate relationship of this new immune-suppressant drug and LI deserves specific consideration. In liver transplant recipients, jejune bile derivation and a liver donor with anti-HBc antibodies defined a population of higher risk for developing LI. Although jejune bile derivation technique has been related with a higher risk of bacterial infection (15,16), liver grafts from donors with anti-HBc antibodies have not been related with a higher risk of infection or a worse graft survival (17) and its relationship with LI is difficult to explain. Finally LI was associated in heart transplant recipients with a worse previous functional cardiorespiratory estate (patients with COPD or valvular disease).

Some drawbacks of the present study deserve specific consideration. Given that the median follow-up of the patients included in the study cohort is 13 months the results of the present study regarding LI should be applied only to the 6th to 24th month after transplantation. On the other hand, the present study is applied to global LI, so specific infections should deserve an individual approach in further studies. Finally, with the design of the study it is difficult to demonstrate causality in some of the detected risk factors for LI and, therefore, such variables should be used exclusively to identify a high-risk population of SOT recipients.

In conclusion, severe infection occurs over the sixth month after SOT in a subgroup of patients fulfilling the risk factors formerly discussed and has similar clinical characteristics than infectious episodes occurring in the early period. These high-risk patients deserve a more careful follow-up and could benefit from more prolonged directed prophylactic measures similar to those performed in the early period after transplantation.

Ancillary