Treatment of drug‐resistant tuberculosis in patients with HIV‐1 infection
Abstract
This is a protocol for a Cochrane Review (Intervention). The objectives are as follows:
To determine the optimum drug regimen for the treatment of drug resistant TB, including MDR and XDR TB, in patients with HIV and TB co‐infection.
Background
Description of the condition
Tuberculosis is one of the most significant infectious causes of death worldwide. Despite that effective drugs for tuberculosis have been available for over 50 years, over 9 million new cases of active tuberculosis occur annually WHO 2009. HIV‐1 sero‐positive individuals carry a disproportionate burden of tuberculosis, as much of the burden of TB occurs in areas of the world hardest hit by the HIV‐1 pandemic WHO 2009. In addition, HIV‐1 and TB co‐infected individuals are more likely to progress to active disease, have more severe disease, and have higher mortality Rajeswaran 2006; Cain 2007.
Description of the intervention
The development of resistance to first line drugs has complicated tuberculosis treatment considerably and has become a major threat to global tuberculosis control Dye 2002. Drug resistance is associated with higher rates of treatment failure, higher rates of relapse, and increased mortality Gandhi 2006. MDR‐TB (multi drug resistant TB ‐ defined as resistance to rifampin and isoniazid) and XDR‐TB (extensively drug resistant TB ‐ defined as MDR TB with additional resistance to the flouroquinolones and one of the three second line injectable drugs) have become more prevalent worldwide. The WHO estimates that 510,545 new case of MDR‐TB occurred in 2007 (4.9% of all cases) and 25% of all countries have reported at least 1 case of XDR‐TB. This number may underestimate the global burden given the limited available to drug sensitivity testing in TB high burden countries WHO 2009. Drug resistant TB has made treatment more difficulty, with lengthened treatments and use of second line mycobacterial drugs. These second line drugs are often more toxic and have may have more interactions with other medications, including drugs used to treat HIV Caminero 2006; Nathanson 2004.
How the intervention might work
A better understanding of outcomes in drug resistant tuberculosis in HIV positive individuals could have important treatment implications. First, assessing which regiments have an association with treatment failure and relapse could allow one to tailor regiments further based on drug resistant and HIV status. Second, in HIV positive individuals on ART, understanding medication interactions that affect outcomes could be an important factor in co‐administration of ART and drug resistant TB. Third, a better understanding of treatment of drug resistant TB could lead to prolonging duration of therapy for current regiments associated with poor outcomes in HIV.
Why it is important to do this review
There are limited data available to guide therapy of drug resistant TB in HIV‐1 infected individuals. It is important to determine the optimal management of different patterns of TB drug resistance in order to improve outcomes in HIV positive patients. We propose to evaluate the available evidence from published and unpublished studies to determine what treatment regimens result in maximum clinical benefit, as measured by microbiological cure and mortality, among HIV infected individuals with evidence of drug resistant TB, including MDR and XDR TB.
Objectives
To determine the optimum drug regimen for the treatment of drug resistant TB, including MDR and XDR TB, in patients with HIV and TB co‐infection.
Methods
Criteria for considering studies for this review
Types of studies
Randomized or quasi‐randomized controlled trials of tuberculosis treatment among HIV‐1 infected individuals with evidence of resistance to at least one antimycobacterial drug will be included. Should data from clinical trials be insufficient, data from observational studies (e.g. cohort, case control, cross‐sectional studies) will be considered for this review. Studies performed in general and specific populations, and in hospitals or clinics, or both, will be included. Studies performed in any country and published in any language will be included. Studies that relied on historical or ecological controls will be excluded because they provide unreliable data for association.
Types of participants
HIV‐1 infected individuals in studies assessing the relationship between TB drug resistance and treatment outcomes will be included.
Types of interventions
All pharmacologic interventions used to treat drug resistant tuberculosis will be included in this review. We will compare available combinations of pharmacologic interventions as well as different durations of therapy in this review.
Types of outcome measures
Primary outcomes
-
Mortality
-
Time to Microbiologic Cure
-
Time to Smear Negativity
-
Time to Relapse
Secondary outcomes
-
Treatment default
-
Treatment completion
-
Development of adverse events
-
Development of IRIS (Immune Reconstitution Inflammatory Syndrome)
-
Change in antiretroviral drug regimen
-
Change in markers of HIV‐1 disease progression (CD4 count and/or HIV‐1 RNA)
Search methods for identification of studies
See search methods used in reviews by the Cochrane Collaborative Review Group on HIV Infection and AIDS.
Electronic searches
We will formulate a comprehensive and exhaustive search strategy in an attempt to identify all relevant studies regardless of language or publication status (published, unpublished, in press and in progress). Full details of the Cochrane HIV/AIDS Review Group methods and the journals hand‐searched are published in the section on Collaborative Review Groups in The Cochrane Library.
We will search the following databases, from 1980 to the search date:
-
Cochrane Central Register of Controlled Trials (CENTRAL)
-
MEDLINE
-
EMBASE
-
LILACS
We will also search the metaRegister of Controlled Trials (mRCT) to identify trials in progress Reference lists of all studies that are included in the pool of retrieved studies will be examined to identify further studies. We will make an effort to contact authors of included studies for information about other studies. If the outcomes of interest are not clearly stated in the article, the authors will be contacted to provide further information.
Searching other resources
In addition to the above searches, we will also search abstracts from the following conferences: the Union World Conference on Lung Health from 1990 to present, the Keystone Symposium on Tuberculosis from 1990 to present, and the American Thoracic Society Conference from 1990 to present.
We will search the abstracts of the Conference on Retroviruses and Opportunistic Infections (CROI), 1994‐2011; International AIDS Society, International AIDS Conference (IAC), 1985‐2010; and International AIDS Society, Conference on HIV Pathogenesis, Treatment and Prevention (IAS), 2001‐2011
Data collection and analysis
The development of the search strategy was performed with the assistance of the Cochrane Infectious Diseases Group. The titles, abstracts, and descriptor terms of all downloaded material from the electronic searches will be read by MA, DH and JW and irrelevant reports will be discarded. All citations identified will be independently evaluated by MA, DH and JW for relevance according to the pre‐specified criteria. When there is uncertainty as to the relevance of the study, the full article will be obtained.
Selection of studies
MA, DH and JW will independently apply the inclusion criteria. Studies will be reviewed for relevance based on study design and outcome measures. We will seek further information from authors if studies do not have enough information to determine eligibility.
Data extraction and management
Data will be independently extracted by MA, DH and JW using standardized data extraction forms. We will use separate forms for randomized trials, case‐control studies, and prospective cohort studies. The following characteristics will be extracted from each study:
-
Administrative details: Author(s), published or unpublished, year of publication year in which study was conducted
-
Details of study: Study design, type, duration, completeness of follow‐up for cohort studies, country/location of the study, setting (i.e., urban or rural, hospital or clinic), method of recruitment and number of participants
-
Characteristics of participants: Age, gender, socioeconomic status, and HIV stage, history of treated TB, mycobacterial culture data, if available
-
Details of HIV disease ‐WHO stage, CD4 count, HIV‐1 RNA (where available), ART regimen, duration of HIV‐1 disease, duration of ART treatment, HIV‐1 genotype resistance profile and opportunistic infection history
-
Details of intervention: tuberculosis treatment regimen and length of treatment
-
Details of outcomes: Time to microbiologic cure, time to smear negativity, time to relapse, mortality, treatment default, and treatment completion.
Assessment of risk of bias in included studies
Bias will be assessed as follows:
Performance bias:
-
Were study staff blinded to the subjects treatment group?
Detection bias:
-
Was HIV status ascertained using an ELISA, Western blot, or particle agglutination test?
-
Was HIV status confirmed using a second test from this list?
-
Was initial diagnosis of tuberculosis made by clinical symptoms, smear microscopy or culture results?
Attrition bias:
-
Were individuals in both the control and intervention group followed for the same period of time?
-
Were at least 80% of participants in all groups included in the final analysis or was the description of those not included suggestive of bias?
Selection bias:
-
Were cases selected from the general population?
-
Were controls selected from the same population as the cases?
Measures of treatment effect
Treatment effect will be assessed based on the outcomes of interest for the Cochrane review. The outcomes include time to microbiologic cure, time to smear negativity, time to relapse, mortality, treatment completion and treatment default.
Unit of analysis issues
To avoid difficulties, studies with similar units of analysis will be grouped together for the purpose of analysis. Studies with multiple treatment groups, for instance, will be grouped together and cluster randomized studies will be grouped together. Studies with different units of analysis will not be pooled for analysis.
Dealing with missing data
If a study meets our inclusion criteria but has missing data, attempts will be made to contact the authors for clarification of relevant information.
Assessment of heterogeneity
Different interventions will be grouped together and analyzed separately. A chi square test will be used to assess the presence of heterogeneity within each intervention group. If statistically significant heterogeneity is observed, a meta‐analysis will not be performed.
Assessment of reporting biases
Funnel plots will be generated to assess for the presence of reporting bias. If asymmetry is present suggesting reporting bias, we will conduct a thorough evaluation to determine if there are other reasons for asymmetry. If no other reasons for asymmetry are found, we will document the possible presence of reporting bias.
Data synthesis
Randomized clinical trials data will not be combined with data from observational studies. Observational studies will be grouped and assessed by type (i.e., cohort study, case‐control study). Studies also will be grouped by type of intervention and analyzed as a group. A meta analysis will be performed only if there is no significant heterogeneity detected. If significant heterogeneity is detected, a narrative analysis will be performed.
Subgroup analysis and investigation of heterogeneity
Studies will be grouped according to type of intervention. Sub‐group analysis will be performed on separate cohorts grouped by drug resistance profile, intervention provided, and endemic case rates of tuberculosis.
Sensitivity analysis
As a result of the different approaches to performing a systematic review, we will conduct sensitivity analysis to determine if minor changes in methodology affect the overall findings of our review. Sensitivity analysis will be performed by using one or all of the following strategies:
-
Inclusion and exclusion criteria may be modified to assess the effect on our results.
-
Studies with missing data may be re‐analyzed using a reasonable range of missing values.
-
Data may be re‐analyzed using different statistical approaches.
