Interventions for the prevention of mycobacterium avium complex in adults and children with HIV

  • Review
  • Intervention

Authors


Abstract

Background

Mycobacterium avium complex (MAC) infection is a common complication of advanced acquired immunodeficiency syndrome (AIDS) disease and is an independent predictor of mortality and shortened survival.

Objectives

To determine the effectiveness and safety of interventions aimed at preventing MAC infection in adults and children with HIV infection.

Search methods

We searched MEDLINE, EMBASE, and The Cochrane Library (search date December 2012).

Selection criteria

Randomised controlled trials comparing different strategies for preventing MAC infection in HIV-infected individuals.

Data collection and analysis

Two reviewers independently assessed trial eligibility and quality, and extracted data. Where data were incomplete or unclear, a third reviewer resolved conflicts and/or trial authors were contacted for further details. Development of MAC infection and survival were compared using risk ratios (RR) and 95% confidence intervals (CI). The quality of evidence has been assessed using the GRADE methodology.

Main results

Eight studies met the inclusion criteria.

Placebo-controlled trials

There was no statistically significant difference between clofazimine and no treatment groups in the number of patients that developed MAC infection (RR 1.01; 95% CI 0.37 to 2.80). Rifabutin (one study; RR 0.48; 95% CI 0.35 to 0.67), azithromycin (three studies; RR 0.37; 95% CI 0.19 to 0.74) and clarithromycin (one study; RR 0.35; 95% CI 0.21 to 0.58) were more effective than placebo in preventing the development of MAC infection. There was no statistically significant difference between those treated with clofazimine (one study; RR 0.98; 95% CI 0.41 to 2.32), rifabutin (one study RR 0.91; 95% CI 0.78 to 1.05), azithromycin (three studies, pooled RR 0.96; 95% CI 0.69 to 1.32) and placebo in number of reported deaths. One study found that the risk of death was reduced by 22% in patients treated with clarithromycin compared to those treated with placebo (RR 0.78; 95% CI 0.64 to 0.96).  

Monotherapy vs. monotherapy

Patients treated with clarithromycin (RR 0.60; 95% CI 0.41 to 0.89) and azithromycin (RR 0.60; 95% CI 0.40 to 0.89) were 40% less likely to develop MAC infection than those treated with rifabutin. There was no statistically significant difference between those treated with clarithromycin (RR 0.98; 95% CI 0.83 to 1.15), azithromycin (RR 0.98; 95% CI 0.77 to 1.24) and rifabutin in the number of reported deaths

Combination therapy versus monotherapy

There was no statistically significant difference between patients treated with a combination of rifabutin and clarithromycin and those treated with clarithromycin alone (RR 0.74; 95% CI 0.46 to 1.20); and those treated with combination of rifabutin and azithromycin and those treated with azithromycin alone (RR 0.59; 95% CI 1.03). Patients treated with a combination of rifabutin plus clarithromycin were 56% less likely to develop MAC infection than those treated with rifabutin alone (RR 0.44; 95% CI 0.29 to 0.69). Patients treated with a combination of rifabutin plus azithromycin were 65% less likely to develop MAC infection than those treated with rifabutin alone (RR 0.35; 95% CI 0.21 to 0.59). There was no statistically significant difference in the number of reported deaths in all the four different comparisons of prophylactic agents.

Authors' conclusions

Based on limited data, azithromycin or clarithromycin appeared to be a prophylactic agent of choice for MAC infection. Further studies are needed, especially direct comparison of clarithromycin and azithromycin. In additions, studies that will compare different doses and regimens are needed.

Plain language summary

Interventions for the prevention of mycobacterium avium complex in adults and children with HIV

Mycobacterium avium complex (MAC) infection is a common complication of advanced acquired immunodeficiency syndrome (AIDS) disease and can shorten the survival of these patients. We sought to examine effectiveness of all drugs for preventing MAC infection in adults and children with HIV infection. This review included eight trials conducted in the USA and Europe, published between 1993 and 2003.

We found evidence (very-low to low grade) that azithromycin or clarithromycin appeared to be a drug of choice for preventing MAC infection. There is a need for further studies to compare direct evidence between clarithromycin and azithromycin and studies to determine the optimal doses required for effective prevention of MAC infection.

Background

Mycobacterium avium complex (MAC) infection is a common complication of advanced acquired immunodeficiency syndrome (AIDS) disease and is an independent predictor of mortality (Shafran 1998), and shortened survival (Horsburgh 1991; Inderlied 1993; Benson 1993). MAC organisms have been isolated from a variety of environmental sources, including water, soil, birds, pigs, and cattle (Inderlied 1993). The exact route of acquisition from environmental sources to humans is uncertain, but indirect evidence favours the acquisition through either the gastrointestinal or respiratory tracts (Shafran 1998). A critical requirement of the pathogenesis of disseminated MAC infection is severe CD4 lymphocyte depletion (Shafran 1998).

In the era before the introduction of highly active antiretroviral therapy (HAART), MAC had an annual incidence of approximately 20% after the occurrence of a first event defining the presence of AIDS (Nightingale 1992). In a North American study, over 50% of patients with AIDS had extensive tissue involvement with MAC at postmortem examination (Hawkins 1986). The widespread use of chemoprophylaxis and the availability of potent antiretroviral treatment (ART) have resulted in an overall decrease in the incidence of serious opportunistic infections (Heifets 1993). MAC, however, still remains a cause of serious disease, either with classical systemic manifestations or as a focal inflammatory lymphadenitis occurring shortly after the initiation of protease inhibitor-containing regimens (Phillips 1999). The prevalence of HIV-associated MAC infection in Africa ranges between 1% and 10% (Gilks 1995; Clive 2001; Maartens 2002). The lower prevalence reported from some African countries could be due to a lack of diagnostic facility for MAC, delayed diagnosis of MAC and shortened survival of AIDS patients (Gilks 1995; Clive 2001; WHO 2004).

On a clinical basis, the diagnosis of disseminated MAC is difficult, and several aspects of the natural history of the infection still are not clear (MacGregor 2005). Prevention of MAC extends survival (Horsburgh 1991; Cohn 1997), but the optimal regimen is unknown. The most common drugs used for prophylaxis against MAC are azithromycin, clarithromycin, ethambutol, rifabutin, rifampicin, ciprofloxacin and amikacin. Two studies reported that clarithromycin has significant antimicrobial activity against MAC and is efficacious in preventing MAC disease (Naik 1989; Dautzenberg 1997). Two studies reported that monotherapy with clarithromycin prevents MAC, reduces the level of mycobacteremia and results in clinical improvement. However, drug resistance emerges within 12-16 weeks of therapy and is associated with clinical relapse (Heifets 1993; Chaisson 1994).

Accordingly, multidrug regimens have been proposed to lower the rate of relapse. A United States Public Health Service task force has listed a number of agents that can be combined with an antibacterial agent which includes the macrolides (clarithromycin or azithromycin) for the treatment of MAC bacteremia (MMWR 1993). For primary prophylaxis in adult AIDS patients before the first episode of MAC infection, a single drug, clarithromycin or azithromycin, are recommended; and for secondary prophylaxis in adult AIDS patients who have had a previous MAC infection, macrolides (clarithromycin or azithromycin) and an anti-tuberculous drug, ethambutol, are recommended (WHO 2004). Also, lifelong secondary prophylactic regimens for MAC is recommended with concomitant administration of HAART (WHO 2004; Benson 2004). The relative contribution of these agents to a combination regimen is unknown, particularly following HAART treatment (Chaisson 1997; Benson 2004).

A controversy that has developed following the emergence of HAART is whether it is necessary to continue MAC prophylaxis in individuals who had CD4 counts below 50 cells/mm3 when commencing antiretroviral therapy but who experienced a favourable response to ART leading to a marked increased in their CD4 T lymphocyte counts (Shafran 1998). In light of these uncertainties, we aim to evaluate the current evidence about interventions for the prevention of MAC infection associated with HIV infection in adults and children.

Objectives

To determine the effectiveness and safety of interventions aimed at preventing MAC infection in adults and children with HIV infection.

Methods

Criteria for considering studies for this review

Types of studies

For the primary purpose of determining the effects of any intervention, only randomised controlled clinical trials (RCTs) of preventive treatments will be considered, irrespective of whether the control group received a placebo. We will exclude quasi-randomised studies.

Types of participants

HIV-infected adults and children.

Types of interventions

Any intervention aimed at preventing HIV-related MAC infection, including:

  • Macrolides (clarithromycin, azithromycin)

  • Ethambutol

  • Rifabutin

  • Fluoroquinolones

  • Amikacin

  • Clofazimine

The intervention may be a single drug or a combination. Also, the Interventions may be combined with ART as long as the intervention and control group both receive the same antiretroviral therapy.

Types of outcome measures

Primary outcome:

  • Detection of disseminated MAC infection as evidenced by a positive culture from blood or other normally sterile site

Secondary outcome:

  • All-cause mortality

  • Any adverse events - any drug-related adverse events will be reported.

Search methods for identification of studies

Electronic searches

See: HIV/AIDS Cochrane Review Group search strategy.

Identification of studies was done with the assistance of the HIV/AIDS Review Group Trials Search coordinator. We formulated a comprehensive and exhaustive search strategy in an attempt to identify all relevant studies regardless of language or publication status (i.e. published, unpublished, in press, and in progress).

We searched the following electronic databases in December 2012

  1. PubMed, 1 January 1980 to 20 December 2012

  2. EMBASE, 1 January 1980 to 20 December 2012

  3. The Cochrane Central Register of Controlled Trials (Issue 12 2012)

The detailed search strategy for each of the databases are documented in Table 1; Table 2; and Table 3 respectively.

Table 1. Search strategy for MEDLINE
NumberSearch terms
#1HIV Infections[MeSH] OR HIV[MeSH] OR hiv[tw] OR hiv-1*[tw] OR hiv-2*[tw] OR hiv1[tw] OR hiv2[tw] OR hiv infect*[tw] OR human immunodeficiency virus[tw] OR human immunedeficiency virus[tw] OR human immuno-deficiency virus[tw] OR human immune-deficiency virus[tw] OR ((human immun*) AND (deficiency virus[tw])) OR acquired immunodeficiency syndrome[tw] OR acquired immunedeficiency syndrome[tw] OR acquired immuno-deficiency syndrome[tw] OR acquired immune-deficiency syndrome[tw] OR ((acquired immun*) AND (deficiency syndrome[tw]))
#2"mycobacterium avium complex" [Mesh] OR mycobacterium avium complex [tw] OR mycobacterium avium OR MAC
#3#1 OR #2
#4randomized controlled trial [pt] OR controlled clinical trial [pt] OR randomized controlled trials [mh] OR random allocation [mh] OR double-blind method [mh] OR single-blind method [mh] OR clinical trial [pt] OR clinical trials [mh] OR ("clinical trial" [tw]) OR ((singl* [tw] OR doubl* [tw] OR trebl* [tw] OR tripl* [tw]) AND (mask* [tw] OR blind* [tw])) OR ( placebos [mh] OR placebo* [tw] OR random* [tw] OR research design [mh:noexp] OR comparative study [mh] OR evaluation studies [mh] OR follow-up studies [mh] OR prospective studies [mh] OR control* [tw] OR prospectiv* [tw] OR volunteer* [tw]) NOT (animals [mh] NOT human [mh])
#5#3 AND #4
#6#5 Field: All Fields, Limits: from 1980 to 2008, HUMAN
Table 2. Search strategy for EMBASE
NumberSearch terms
#1'human immunodeficiency virus infection'/exp
#2'human immunodeficiency virus'/exp
#3hiv:ti OR hiv:ab
#4'hiv-1':ti OR 'hiv-1':ab
#5'hiv-2':ti OR 'hiv-2':ab
#6'human immunodeficiency virus':ti OR 'human immuno deficiency':ab
#7'human immuno-deficiency virus':ti OR 'human immuno-deficiency virus':ab
#8'human immunedeficiency virus':ti OR 'human immune deficiency virus':ab
#9'human immune-deficiency virus':ti OR 'human immune-deficiency virus':ab
#10'acquired immune-deficiency syndrome':ti OR 'acquired immune-deficiency syndrome':ab
#11'acquired immunedeficiency syndrome':ti OR 'acquired immunedeficiency syndrome':ab
#12'acquired immunodeficiency syndrome':ti OR 'acquired immunodeficiency syndrome':ab
#13'acquired immuno-deficiency syndrome':ti OR 'acquired immuno-deficiency syndrome':ab
#14#1 OR #2 OR #3 OR #4 OR #5 OR #6 OR #7 OR #8 OR #9 OR #10 OR #11 OR #12 OR #13
#15'mycobacterium avium complex'
#16Mac
#17' mycobacterium avium'
#18#15 OR #16 OR #17
#19random*:ti OR random*:ab
#20factorial*:ti OR factorial*:ab
#21cross ?over*:ti OR cross?over:ab OR crossover*:ti OR crossover *:ab
#22placebo*:ti OR placebo*:ab
#23((doubl*:ti AND blind*:ti) OR (doubl*:ab AND blind*:ab))
#24((singl*:ti AND blind*:ti) OR (singl*:ab AND blind*:ab))
#25assign*:ti OR assign*:ab
#26volunteer*:ti OR volunteer*:ab
#27'crossover procedure'/de
#28'double-blind procedure'/de
#29'single-blind procedure'/de
#30'randomized controlled trial'/de
#31allocat*:ti OR allocat*:ab
#32#19 OR #20 OR #21 OR #22 OR #23 OR #24 OR #25 OR #26 OR #27 OR #28 OR #29 OR #30 OR #31
#33#14 AND #18 AND #32
Table 3. Search strategy for CENTRAL
  
NumberSearch terms
#1hiv OR hiv-1* OR hiv-2* OR hiv1 OR hiv2 OR (HIV INFECT*) OR (HUMAN IMMUNODEFICIENCY VIRUS) OR (HUMAN IMMUNEDEFICIENCY
VIRUS) OR (HUMAN IMMUNE-DEFICIENCY VIRUS) OR (HUMAN IMMUNO-DEFICIENCY VIRUS) OR (HUMAN IMMUN* DEFICIENCY VIRUS) OR
(ACQUIRED IMMUNODEFICIENCY SYNDROME) OR (ACQUIRED IMMUNEDEFICIENCY SYNDROME) OR (ACQUIRED IMMUNO-DEFICIENCY
SYNDROME) OR (ACQUIRED IMMUNE-DEFICIENCY SYNDROME) OR (ACQUIRED IMMUN* DEFICIENCY SYNDROME) in All Fields in all products
#2MeSH descriptor HIV Infections explode all trees in MeSH products
#3MeSH descriptor HIV explode all trees in MeSH products
#4(#1 OR #2 OR #3)
#5(MYCOBACTERIUM AVIUM COMPLEX) OR (MYCOBACTERIUM AVIUM) in All Fields in all products
#6(#4 AND #5)

Searching other resources

We also searched the WHO International Clinical Trials Registry Platform. In addition, we checked the reference lists of all studies identified by the above methods and examined any systematic reviews, meta-analyses or treatment guidelines we identified.

Data collection and analysis

Selection of studies

Two authors (MU and OU) independently read the titles, abstracts, and descriptor terms of the search output from the different databases to identify potentially eligible studies. Full text articles were obtained for all citations identified as potentially eligible. Both authors (MU and OU) independently inspected these to establish the relevance of the articles according to the pre-specified criteria. The studies were reviewed for relevance based on study design, types of participants, interventions, and outcome measures. We gave reasons for excluding potentially relevant studies in an excluded studies table.

Data extraction and management

We extracted data independently using the form we designed and agreed upon. Both authors verified the extracted data. Extracted information included the study, participant, intervention and outcome details. We summarized the eligible study using the RevMan software.The authors independently extracted the data, entered them into RevMan; and all disagreements were resolved by discussion.

Assessment of risk of bias in included studies

MU and OU independently examined the components of each included trial for risk of bias using a standard form. This included information on the sequence generation, allocation concealment, blinding (participants, personnel, and outcome assessor), incomplete outcome data, selective outcome reporting, and other sources of bias. The methodological components of the studies were assessed and classified as adequate, inadequate, or unclear as per the Cochrane Handbook of Systematic Reviews of Interventions (Higgins 2011). Where differences arose, they were resolved by discussions with the third reviewer (IY).

Sequence generation

  • Low risk of bias: investigators described a random component in the sequence generation process such as the use of random number table, coin tossing, cards or envelops shuffling, etc.

  • High risk of bias: investigators described a non-random component in the sequence generation process such as the use of odd or even date of birth, algorithm based on the day/date of birth, hospital or clinic record number.

  • Unclear: insufficient information to permit judgment of the sequence generation process.

Allocation concealment

  • Low risk of bias: participants and the investigators enrolling participants cannot foresee assignment (e.g. central allocation; or sequentially numbered, opaque, sealed envelopes).

  • High risk of bias: participants and investigators enrolling participants can foresee upcoming assignment (e.g. an open random allocation schedule (e.g. a list of random numbers); or envelopes were unsealed or non­opaque or not sequentially numbered).

  • Unclear: insufficient information to permit judgment of the allocation concealment or the method not described.

Blinding

  • Low risk of bias: blinding of the participants, key study personnel and outcome assessor, and unlikely that the blinding could have been broken. Or lack of blinding unlikely to introduce bias. No blinding in the situation where non-blinding is not likely to introduce bias.

  • High risk of bias: no blinding, incomplete blinding and the outcome is likely to be influenced by the lack of blinding.

  • Unclear: insufficient information to permit judgment of adequacy or otherwise of the blinding.

Incomplete outcome data

  • Low risk of bias: no missing outcome data, reasons for missing outcome data unlikely to be related to true outcome, or missing outcome data balanced in number across groups.

  • High risk of bias: reason for missing outcome data likely to be related to true outcome, with either imbalance in number across groups or reasons for missing data.

  • Unclear: insufficient reporting of attrition or exclusions.

Selective Reporting

  • Low risk of bias: a protocol is available which clearly states the primary outcome as the same as in the final trial report.

  • High risk of bias: the primary outcome differs between the protocol and final trial report.

  • Unclear: no trial protocol is available or there is insufficient reporting to determine if selective reporting is present.

Other forms of bias

  • Low risk of bias: there is no evidence of bias from other sources.

  • High risk of bias: there is potential bias present from other sources (e.g. early stopping of trial, fraudulent activity, extreme baseline imbalance or bias related to specific study design).

  • Unclear: insufficient information to permit judgment of adequacy or otherwise of other forms of bias.

Measures of treatment effect

Outcome measures for dichotomous data (e.g. development of MAC infection and death) were calculated as a relative risk with 95% confidence intervals.

Assessment of heterogeneity

We did not have a sufficient number of studies to perform a meta-analysis. Therefore we could not assess statistical heterogeneity. However, the included studies differed in terms of the prophylactic agents used.

Data synthesis

MU and OU independently extracted data from the included studies. The data were summarised in RevMan 5.1. All of the entries were checked by both authors.

Results

Description of studies

Results of the search

We prepared a PRISMA statement flowchart to describe how we processed the references identified through the search results (see Figure 1). The literature searches yielded 1769 citations after duplicates were removed. After scanning titles and abstracts, a total of 21 potentially relevant articles were identified and full-text copies were assessed independently against the inclusion criteria by the two authors.

Figure 1.

Study flow diagram.

Included studies

Eight studies (Abrams 1993; Nightingale 1993; Havlir 1996; Pierce 1996; Oldfield 1998; Benson 2000; Currier 2000; El-Sadr 2000) met the inclusion criteria (see 'Characteristics of included studies'). All trials were identified from published literature. The studies were published between 1993 and 2000.

Types of participants and settings.

Most of the studies were multicenter studies conducted in the USA (Abrams 1993; Nightingale 1993; Havlir 1996; Oldfield 1998; Benson 2000; Currier 2000; El-Sadr 2000). One study was a multicenter study conducted in the USA and Europe (Pierce 1996). All studies included HIV-infected patients with advanced stage (as measured by low CD4+ count). The definition of low CD4+ count varied across the studies. This review was based on data from 4972 participants. The sample size ranged from 110 (Abrams 1993) to 1178 (Benson 2000).

Types of intervention

One study compared efficacy and safety of clofazimine with no treatment (Abrams 1993). One study each compared efficacy and safety of placebo with clarithromycin (Pierce 1996) and rifabutin (Nightingale 1993). Three studies compared efficacy and safety of azithromycin with placebo (Oldfield 1998; Currier 2000; El-Sadr 2000). Two studies had three-arms (Havlir 1996; Benson 2000). Benson 2000 compared efficacy and safety of clarithromycin and rifabutin alone and in combination. Havlir 1996 compared prophylactic regimens: rifabutin, azithromycin or both drugs.

The dosage of the prophylactic agents is as follows:

Types of outcome

The primary end point of all studies was development of confirmed MAC infection. The definition of confirmed cases of MAC infection varied across studies. Most studies diagnosed MAC infection by a positive blood culture. Some studies defined confirmed MAC infection by growth from a usually sterile site, such as blood, lymph node, liver or bone marrow. All studies reported all-cause mortality.

Lost to follow-up

Six of the eight studies reported percentage lost to follow-up in each treatment arm (see Table 4). The percentage lost to follow-up in placebo (or no treatment) arm ranged from 1.2% in El-Sadr 2000 to 15.2% in Abrams 1993. While the percentage lost to follow-up in the active treatment arm ranged from 1.6% in El-Sadr 2000 to 9.3% in Currier 2000. There is no evidence of differential lost to follow-up between treatment arms in all of the six studies that reported lost to follow-up.

Table 4. Loss to follow-up
Study IDInterventionLoss to follow-up (%)P-value
Abrams 1993   
 Clofazamine3.70.048
 No treatment15.2 
Nightingale 1993   
 Rifabutin6.10.518
 Placebo7.1 
Havlir 1996   
 Rifabutin5.50.316
 Azithromycin4.3 
 Combination2.7 
Pierce 1996   
 Clarithromycin6.60.921
 Placebo6.9 
Currier 2000   
 Azithromycin9.30.681
 Placebo10.3 
El-Sadr 2000   
 Azithromycin1.60.688
 Placebo1.2 

Excluded studies

Eleven studies were excluded (see 'Characteristics of excluded studies'). Most (n=6) studies were excluded because they were reviews of other studies (Ostroff 1995; Siegal 1996; Cohn 1997; Currier 1997; Ong 1999; Chu 2011). Two studies were excluded because they were non-RCT (Dube 1999; Green 2004). Dube 1997 was not a prophylactic study, the study examined the role of ethambutol for prevention of relapse of MAC infection). McNabb 2000 reported no relevant outcomes for this review, the study examined adherence to azithromycin and clarithromycin. Moore 1995 was an open-label follow-up and secondary analysis of Nightingale 1993.

Risk of bias in included studies

See Risk of Bias graph and Risk of Bias summary tables for the included studies Figure 2 and Figure 3. In general, the overall methodological quality of the included studies was acceptable.

Figure 2.

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

Figure 3.

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

Allocation

Three studies did not report how the random sequence was generated (Abrams 1993; Havlir 1996; El-Sadr 2000). The risk of selection bias (in terms of random sequence generation) was low in four studies (Nightingale 1993; Pierce 1996; Benson 2000; Currier 2000). Only one study (Benson 2000) used proper allocation concealment (i.e. central allocation method), the other seven studies did not describe how the allocation was concealed.

Blinding

With the exception of Abrams 1993, most studies were double blind. The participants and outcome assessors were masked to the treatment allocations in two studies (Benson 2000; Currier 2000). Five studies did not report who was masked (El-Sadr 2000; Havlir 1996; Nightingale 1993; Oldfield 1998; Pierce 1996)There was a high risk of performance bias and detection bias in Abrams 1993 , because the study was an open-label RCT.

Incomplete outcome data

No information about attrition was found in two studies (Benson 2000; Havlir 1996). The attrition rates were similar in treatment arms in other six studies (Nightingale 1993; Abrams 1993; Pierce 1996; Currier 2000; El-Sadr 2000).

Selective reporting

Apparently, there was a low risk of selecting reporting bias in all the studies.

Other potential sources of bias

There was no sufficient information to determine whether there is a risk of other potential sources of bias in three studies (Abrams 1993; Nightingale 1993; Havlir 1996). There was a low risk of other potential sources of bias in three studies (Benson 2000; Currier 2000; El-Sadr 2000). There was a high risk of other potential sources of bias in one (Pierce 1996), because a large proportion of patients withdrew from the study.

Effects of interventions

  1. Monotherapy versus placebo

    1. Detection of disseminated MAC infection(Analysis 1.1): There was no statistically significant difference between clofazimine and no treatment groups in the number of patients developed MAC infection (RR 1.01; 95% CI 0.37 to 2.80). Rifabutin (one study; RR 0.48; 95% CI 0.35 to 0.67), azithromycin (three studies; RR 0.37; 95% CI 0.19 to 0.74) and clarithromycin (one study; RR 0.35; 95% CI 0.21 to 0.58) were more effective than placebo in preventing the development of MAC infection.

    2. All-cause mortality(Analysis 1.2): There was no statistically significant difference between those treated with clofazimine and no treatment in the number of reported deaths (RR 0.98; 95% CI 0.41 to 2.32). Similarly, there were no statistically significant differences between those treated with rifabutin (one study RR 0.91; 95% CI 0.78 to 1.05), azithromycin (three studies, pooled RR 0.96; 95% CI 0.69 to 1.32) and placebo in the number of reported deaths. Pierce 1996 found that the risk of death was reduced by 22% in patients treated with clarithromycin compared to those treated with placebo (RR 0.78; 95% CI 0.64 to 0.96).  

  2. Monotherapy versus another monotherapy

    1. Detection of disseminated MAC infection (Analysis 2.1): Benson 2000 and Havlir 1996 found that clarithromycin and azithromycin were more effective in preventing MAC infection than rifabutin respectively. Patients treated with clarithromycin (RR 0.60; 95% CI 0.41 to 0.89) and azithromycin (RR 0.60; 95% CI 0.40 to 0.89) were 40% less likely to develop MAC infection than those treated with rifabutin.

    2. All-cause mortality (Analysis 2.2): There was no statistically significant difference between those treated with clarithromycin (RR 0.98; 95% CI 0.83 to 1.15), azithromycin (RR 0.98; 95% CI 0.77 to 1.24) and rifabutin in the number of reported deaths.

  3. Combination therapy versus monotherapy

    1. Detection of disseminated MAC infection (Analysis 3.1): There was no statistically significant difference between patients treated with a combination of rifabutin and clarithromycin and those treated with clarithromycin alone (RR 0.74; 95% CI 0.46 to 1.20) in preventing MAC infection. The combination of rifabutin and azithromycin showed trends toward effective in preventing MAC infection than those treated with azithromycin alone (RR 0.59; 95% CI 1.03), however, this does not reach statistically significant level (p=0.06). Patients treated with a combination of rifabutin plus clarithromycin were 56% less likely to develop MAC infection than those treated with rifabutin alone (RR 0.44; 95% CI 0.29 to 0.69). Patients treated with a combination of rifabutin plus azithromycin were 65% less likely to develop MAC infection than those treated with rifabutin alone (RR 0.35; 95% CI 0.21 to 0.59).  

    2. All-cause mortality (Analysis 3.2): There was no statistically significant difference in the number of reported deaths in all the four different comparisons of prophylactic agents.

Discussion

Summary of main results

This review evaluated the effects of different strategies for preventing MAC infection. We found evidence from placebo-controlled studies that patients treated with rifabutin, clarithromycin and azithromycin were less likely to develop a MAC infection. Both clarithromycin and azithromycin when administered as a single agent were more effective than rifabutin monotherapy. Similarly, the combination of rifabutin with clarithromycin or azithromycin were more effective than rifabutin alone. The combination of rifabutin and azithromycin was no more effective than azithromycin alone. In addition, the combination of clarithromycin and rifabutin was no more effective than clarithromycin alone. In terms of survival benefits, only clarithromycin demonstrated benefit over placebo. However, there is no evidence of survival differences between any monotherapy versus monotherapy and combination therapy versus any monotherapy. The combination therapy has been reported to be associated with higher adverse events than monotherapy (Cohn 1997).

There is evidence from both randomised controlled trials (Currier 1997; El-Sadr 2000) and cohort studies (Dworkin 2000; Green 2004) that patients receiving effective highly active antiretroviral therapy with CD4 cell counts above 100 cells/mm3 have a low risk of developing disseminated MAC infection. Consequently, guide for HIV/AIDS Clinical Care (US DHHS 2011) recently published in January 2011 by the U.S. Department of Health and Human Services Health Resources and Services Administration recommended that primary prophylaxis for MAC can be discontinued in persons who have responded to effective ART with sustained increases in CD4 counts to >100 cells/µL for at least 3 months; Careful observation and monitoring are required, and prophylaxis should be restarted if the patient's CD4 count decreases to <50 cells/µL; secondary prophylaxis can be discontinued in patients who received at least 12 months of treatment for MAC, are asymptomatic, and have sustained (for at least 6 months) CD4 counts of >100 cells/µL on ART; and secondary prophylaxis should be reintroduced if the CD4 count decreases to <100 cells/µL.

Overall completeness and applicability of evidence

The bulk of the evidence is placebo-controlled trials. The number of efficacy studies of single agent therapy with another agent or combination therapy is limited. All studies were conducted in high-income countries, as such it is not clear if these findings are directly applicable in low- and middle-income countries due to differences in the risk profile of developing MAC infection. Therefore caution is needed in extrapolating the findings to other regions.

Potential biases in the review process

We performed a comprehensive search of several databases and sources to identify studies. There was no language restriction. We reduced potential bias in the conduct of this review by having two of the authors independently scan through the search output, extract data, and assess the methodological quality of each study.

Agreements and disagreements with other studies or reviews

We identified two previous systematic reviews (Chu 2011; Cohn 1997). Chu and colleagues (Chu 2011) conducted a methodological study on adjusted indirect comparison using MAC prophylaxis in HIV as an illustration. Using adjusted indirect calculations, Chu et al, reported that the efficacy of rifabutin compared to placebo ranged from 41% to 44%; and efficacies of clarithromycin and azithromycin were 73% and 72% respectively. Cohn 1997 reviewed studies that investigated different prevention strategies for MAC. We included two additional studies published after Cohn 1997 was published. The results of this review are in agreement with Cohn's systematic review (Cohn 1997).

Authors' conclusions

Implications for practice

Based on limited data, azithromycin or clarithromycin appeared to be a prophylactic agent of choice for MAC infection.

Implications for research

Further studies are needed, especially direct comparison of clarithromycin and azithromycin. In addition, studies that will compare different doses and regimens are needed.

Acknowledgements

Muhammed Mubashir Babatunde Uthman was awarded a Reviews for Africa Programme Fellowship (www.mrc.ac.za/cochrane/rap.htm), funded from the Nuffield Commonwealth Programme, through The Nuffield Foundation.

Data and analyses

Download statistical data

Comparison 1. Monotherapy versus no treatment or placebo
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Development of MAC disease6 Risk Ratio (M-H, Fixed, 95% CI)Subtotals only
1.1 Clofazimine vs. no treatment199Risk Ratio (M-H, Fixed, 95% CI)1.01 [0.37, 2.80]
1.2 Rifabutin vs. placebo11146Risk Ratio (M-H, Fixed, 95% CI)0.48 [0.35, 0.67]
1.3 Azithromycin vs. placebo31337Risk Ratio (M-H, Fixed, 95% CI)0.37 [0.19, 0.74]
1.4 Clarithromycin vs. placebo1667Risk Ratio (M-H, Fixed, 95% CI)0.35 [0.21, 0.58]
2 All-cause mortality6 Risk Ratio (M-H, Fixed, 95% CI)Subtotals only
2.1 Clofazimine vs. no treatment199Risk Ratio (M-H, Fixed, 95% CI)0.98 [0.41, 2.32]
2.2 Rifabutin vs. placebo11146Risk Ratio (M-H, Fixed, 95% CI)0.91 [0.78, 1.05]
2.3 Azithromycin vs. placebo31337Risk Ratio (M-H, Fixed, 95% CI)0.96 [0.69, 1.32]
2.4 Clarithromycin vs. placebo1667Risk Ratio (M-H, Fixed, 95% CI)0.78 [0.64, 0.96]
Analysis 1.1.

Comparison 1 Monotherapy versus no treatment or placebo, Outcome 1 Development of MAC disease.

Analysis 1.2.

Comparison 1 Monotherapy versus no treatment or placebo, Outcome 2 All-cause mortality.

Comparison 2. Monotherapy versus monotherapy
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Development of MAC disease2 Risk Ratio (M-H, Fixed, 95% CI)Subtotals only
1.1 Clarithromycin vs.rifabutin1789Risk Ratio (M-H, Fixed, 95% CI)0.60 [0.41, 0.89]
1.2 Azithromycin vs. rifabutin1446Risk Ratio (M-H, Fixed, 95% CI)0.60 [0.40, 0.89]
2 All-cause mortality2 Risk Ratio (M-H, Fixed, 95% CI)Subtotals only
2.1 Clarithromycin vs.rifabutin1789Risk Ratio (M-H, Fixed, 95% CI)0.98 [0.83, 1.15]
2.2 Azithromycin vs. rifabutin1446Risk Ratio (M-H, Fixed, 95% CI)0.98 [0.77, 1.24]
Analysis 2.1.

Comparison 2 Monotherapy versus monotherapy, Outcome 1 Development of MAC disease.

Analysis 2.2.

Comparison 2 Monotherapy versus monotherapy, Outcome 2 All-cause mortality.

Comparison 3. Combination therapy versus monotherapy
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Development of MAC disease2 Risk Ratio (M-H, Fixed, 95% CI)Subtotals only
1.1 Rifabutin plus Clarithromycin vs. clarithromycin1787Risk Ratio (M-H, Fixed, 95% CI)0.74 [0.46, 1.20]
1.2 Rifabutin plus Clarithromycin vs. rifabutin1780Risk Ratio (M-H, Fixed, 95% CI)0.44 [0.29, 0.69]
1.3 Rifabutin plus azithromycin vs. azithromycin1441Risk Ratio (M-H, Fixed, 95% CI)0.59 [0.34, 1.03]
1.4 Rifabutin plus azithromycin vs. rifabutin1441Risk Ratio (M-H, Fixed, 95% CI)0.35 [0.21, 0.59]
2 All-cause mortality2 Risk Ratio (M-H, Fixed, 95% CI)Subtotals only
2.1 Rifabutin plus Clarithromycin vs. clarithromycin1787Risk Ratio (M-H, Fixed, 95% CI)1.10 [0.94, 1.28]
2.2 Rifabutin plus Clarithromycin vs. rifabutin1780Risk Ratio (M-H, Fixed, 95% CI)1.07 [0.92, 1.25]
2.3 Rifabutin plus azithromycin vs. azithromycin1441Risk Ratio (M-H, Fixed, 95% CI)1.00 [0.78, 1.27]
2.4 Rifabutin plus azithromycin vs. rifabutin1441Risk Ratio (M-H, Fixed, 95% CI)0.97 [0.77, 1.24]
Analysis 3.1.

Comparison 3 Combination therapy versus monotherapy, Outcome 1 Development of MAC disease.

Analysis 3.2.

Comparison 3 Combination therapy versus monotherapy, Outcome 2 All-cause mortality.

What's new

DateEventDescription
8 April 2013AmendedFixed affiliation

History

Protocol first published: Issue 2, 2008
Review first published: Issue 4, 2013

DateEventDescription
17 February 2013AmendedWe planned to include the following secondary outcomes: frequency of episodes, global impression of change, completion of prophylaxis, and time-to-treatment failure; however, they were not reported in the primary trials and as such were no longer considered in this review
28 October 2008AmendedConverted to new review format.

Contributions of authors

Dr Uthman Muhammed Mubashir Babatunde is the first author and guarantor for this review. Dr Uthman AbdulRahman Olalekan is the second author. Dr Yahaya Ismail is the third author (to resolve disputes).

Declarations of interest

None known.

Sources of support

Internal sources

  • Department of Epidemiology and Community Health, University of Ilorin Teaching Hospital, Nigeria.

External sources

  • Review for African Programme Fellowship, South Africa.

  • South African Cochrane Centre (SACC), HIV/AIDS mentoring programme, South Africa.

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Abrams 1993

MethodsRCT
ParticipantsPatients with advanced HIV
Interventions50mg/day clofazimine vs. no treatment
Outcomes

Primary outcomes

  1. Disseminated MAC infection (diagnosed by blood culture positive for acid-fast bacilli or positive bone marrow biopsy)

  2. Death

NotesCommunity-based clinical trial conducted in USA
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot described in the study
Allocation concealment (selection bias)Unclear riskNot described in the study
Blinding of participants and personnel (performance bias)
All outcomes
High riskOpen-label study
Blinding of outcome assessment (detection bias)
All outcomes
High riskOpen-label study
Incomplete outcome data (attrition bias)
All outcomes
Low riskMissing outcome data balanced in number across groups.
Selective reporting (reporting bias)Low riskCompares favourably with protocol on www.ClinicalTrials.gov (Identifier: NCT00002058)
Other biasUnclear riskThere was no sufficient information to determine whether there is a risk of other potential sources of bias

Benson 2000

MethodsRCT
ParticipantsHIV-infected persons with ≤ 100 CD4 T lymphocytes/ʯL
InterventionsClarithromycin (500mg 2 times a day) vs. rifabutin (450mg once per day) vs. clarithromycin plus rifabutin
Outcomes

Primary outcome

  1. Development of MAC infection as defined by a single blood culture positive for MAC after randomisation or the isolation of MAC from another normally sterile site plus at least 1 sign or symptom of MAC infection, as previously defined

Secondary outcome

  1. Death

  2. treatment-limiting adverse effects

NotesMulticenter study conducted in USA
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low risk"... use permuted blocks of six"
Allocation concealment (selection bias)Low risk"randomisation was done centrally by ..."
Blinding of participants and personnel (performance bias)
All outcomes
Low risk"... patients .. remained blinded to treatment assignment and dose..."
Blinding of outcome assessment (detection bias)
All outcomes
Low risk".. participating clinicians remained blinded to treatment assignment..."
Incomplete outcome data (attrition bias)
All outcomes
Unclear riskNo published information on the study was found.
Selective reporting (reporting bias)Low riskCompares favourably with protocol on www.ClinicalTrials.gov (Identifier: NCT00001030)
Other biasLow riskNo other potential sources of bias identified

Currier 2000

MethodsRCT
ParticipantsDocumented HIV infection and CD4+ cell counts of 0.05 X 109 cells/L or less followed by a documented increase in CD4+ cell count to 0.10X109 cells/L or more on two separate occasions at least 4 weeks apart, with no intervening values less than 0.10 X109 cells/L.
InterventionsAzithromycin (1200mg once weekly) vs. matching placebo
Outcomes

Primary outcome

  1. development of MAC infection, defined by growth from a usually sterile site, such as blood, lymph node, liver, or bone marrow.

Secondary outcome

  1. Death

NotesMulticenter study conducted in USA
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskRandomisation process was appropriate.
Allocation concealment (selection bias)Unclear riskNo published information on the study was found.
Blinding of participants and personnel (performance bias)
All outcomes
Low risk"Both participants and staff were blinded to treatment assignment throughout the study"
Blinding of outcome assessment (detection bias)
All outcomes
Low risk"Both participants and staff were blinded to treatment assignment throughout the study"
Incomplete outcome data (attrition bias)
All outcomes
Low riskMissing outcome data balanced in number across groups.
Selective reporting (reporting bias)Low riskNo evidence of selective outcome reporting
Other biasLow riskNo other potential sources of bias identified

El-Sadr 2000

MethodsRCT
ParticipantsHIV-infected patients with CD4+ cell count of less than 50 per cubic millimeter (at any time in the past), followed by a count of more than 100 cells per cubic millimeter on two consecutive occasions in response to antiretroviral therapy
InterventionsAzithromycin (1200mg once weekly) vs. matching placebo
Outcomes

Primary outcome

  1. development of confirmed MAC infection

Secondary outcome

  1. Death

NotesMulticenter study conducted in USA
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNo published information on the study was found.
Allocation concealment (selection bias)Unclear riskNo published information on the study was found.
Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskNo information was provided on who was masked, though the study was described as double-blind
Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskNo information was provided on who was masked, though the study was described as double-blind
Incomplete outcome data (attrition bias)
All outcomes
Low riskMissing outcome data balanced in number across groups.
Selective reporting (reporting bias)Low riskCompares favourably with protocol on www.ClinicalTrials.gov (Identifier: NCT00000947)
Other biasLow riskNo other potential sources of bias identified.

Havlir 1996

MethodsRCT
ParticipantsHIV-infected patients with fewer than 100 CD4 cell per cubic millimeter
InterventionsRifabutin (300mg daily) vs. azithromycin (1200mg weekly) vs. rifabutin plus azithromycin
Outcomes

Primary outcome

  1. Development of MAC infection, which was diagnosed on the basis of a positive culture of MAC from blood or another sterile body site

NotesMulticenter study conducted in USA
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNo published information on the study was found.
Allocation concealment (selection bias)Unclear riskNo published information on the study was found.
Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskNo information was provided on who was masked, though the study was described as double-blind
Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskNo information was provided on who was masked, though the study was described as double-blind
Incomplete outcome data (attrition bias)
All outcomes
Unclear riskNo published information on the study was found.
Selective reporting (reporting bias)Low riskNo evidence of selective outcome reporting
Other biasUnclear riskThere was no sufficient information to determine whether there is a risk of other potential sources of bias

Nightingale 1993

MethodsRCT
ParticipantsHIV-infected patients with ≤ 200 CD4 cell per cubic millimeter
InterventionsRifabutin (300mg daily) vs. placebo
Outcomes

Primary outcome

  1. Development of MAC infection

Secondary outcome

  1. Death

NotesMulticenter study conducted in USA
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskPermuted block
Allocation concealment (selection bias)Unclear riskNo published information on the study was found.
Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskNo information was provided on who was masked, though the study was described as double-blind
Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskNo information was provided on who was masked, though the study was described as double-blind
Incomplete outcome data (attrition bias)
All outcomes
Low riskMissing outcome data balanced in number across groups.
Selective reporting (reporting bias)Low riskNo evidence of selective outcome reporting
Other biasUnclear riskThere was no sufficient information to determine whether there is a risk of other potential sources of bias

Oldfield 1998

MethodsRCT
ParticipantsHIV-infected with one documented CD4 cell count <100/mm3 within the preceding 12 months
InterventionsAzithromycin (300-mg weekly) vs. matching placebo
Outcomes

Primary outcome

Development of MAC infection documented by culture of a specimen

NotesMulticenter study conducted in USA
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskPatients were randomly assigned in one-to-one ratio by a computer-generated random code
Allocation concealment (selection bias)Unclear riskNo published information on the study was found.
Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskNo published information on the study was found.
Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskNo published information on the study was found.
Incomplete outcome data (attrition bias)
All outcomes
Low riskMissing outcome data balanced in number across groups.
Selective reporting (reporting bias)Low riskCompares favourably with protocol on www.ClinicalTrials.gov (Identifier: NCT00002309)
Other biasUnclear riskNo published information on the study was found.

Pierce 1996

MethodsRCT
ParticipantsHIV-infected patients with CD4 count 100 or less per cubic millimeter
InterventionsClarithromycin (500mg) vs. matching placebo
Outcomes

Primary outcome

  1. development of disseminated MAC infection as confirmed by a positive culture from blood or another sterile site

Secondary outcome

  1. All-cause mortality

NotesMulticenter study conducted in USA and Europe
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low risk... use a computer-generated randomisation sequence
Allocation concealment (selection bias)Unclear riskNo published information on the study was found.
Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskNo information was provided on who was masked, though the study was described as double-blind
Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskNo information was provided on who was masked, though the study was described as double-blind
Incomplete outcome data (attrition bias)
All outcomes
Low riskMissing outcome data balanced in number across groups.
Selective reporting (reporting bias)Low riskNo evidence of selective outcome reporting
Other biasHigh riskLarge proportion of patients withdrew from the study

Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion
Chu 2011Systematic review
Cohn 1997Review article
Currier 1997Review article
Dube 1997Not a prophylaxis study, prevention of lapse of MAC
Dube 1999Non-RCT
Green 2004Non-RCT
McNabb 2000No relevant outcome, not efficacy study
Moore 1995Re-analysis of Nightingale 1993
Ong 1999Review article
Ostroff 1995Review article
Siegal 1996Review article