Multidrug-Resistant and Extensively Drug-Resistant Tuberculosis: Implications for the HIV Epidemic and Antiretroviral Therapy Rollout in South Africa

SUPPLEMENT ARTICLE CSE Global Theme Issue on Poverty and Human Development Multidrug-Resistant and Extensively Drug-Resistant Tuberculosis: Implications for the HIV Epidemic and Antiretroviral Therapy Rollout in South Africa Jason R. Andrews, 1 N. Sarita Shah, 2 Neel Gandhi, 2 Tony Moll, 3 and Gerald Friedland, 1 on behalf of the Tugela Ferry Care and Research (TF CARES) Collaboration 1 Yale University School of Medicine, New Haven, Connecticut; 2 Albert Einstein College of Medicine, Bronx, New York; and 3 Philanjalo and Church of Scotland Hospital, Tugela Ferry, KwaZulu-Natal, South Africa Drug-resistant tuberculosis (TB) is emerging as a major clinical and public health challenge in areas of sub- Saharan Africa where there is a high prevalence of human immunodeficiency virus (HIV) infection. TB drugresistance surveillance in this region has been limited by laboratory capacity and the public health infrastructure; however, with the maturation of the HIV epidemic, the burden of drug-resistant TB is increasing rapidly. The recent discovery of large numbers of cases of multidrug-resistant (MDR) TB and extensively drug-resistant (XDR) TB in South Africa likely represents an unrecognized and evolving epidemic rather than sporadic, localized outbreaks. The combination of a large population of HIV-infected susceptible hosts with poor TB treatment success rates, a lack of airborne infection control, limited drug-resistance testing, and an overburdened MDR-TB treatment program provides ideal conditions for an MDR-TB and XDR-TB epidemic of unparalleled magnitude. In the present article, we review the history of drug-resistant TB in South Africa, describe its interaction with the HIV epidemic and the resultant consequences, and suggest measures necessary for controlling MDR-TB and XDR-TB in this context. A successful response to the emergence of MDR-TB and XDR-TB will necessitate increased resources for and collaboration between TB and HIV programs. Multidrug-resistant (MDR) tuberculosis (TB), defined as disease caused by Mycobacterium tuberculosis strains with resistance to, at least, isoniazid and rifampicin, is a growing public health and clinical problem worldwide. It is estimated that 424,000 cases of MDR-TB occur annually, representing 14% of the global burden of TB [1]. TB drug resistance is caused by inadequate therapy enabling selection and growth of resistant populations (i.e., acquired resistance) or by infection with a drug- Potential conflicts of interest: none reported. Financial support: Doris Duke Charitable Foundation; Irene Diamond Fund; Yale University; Center for AIDS Research at the Albert Einstein College of Medicine and Montefiore Medical Center, funded by the National Institutes of Health (NIH grant AI-51519). Supplement sponsorship is detailed in the Acknowledgments. Reprints or correspondence: Dr. Jason R. Andrews, Yale University School of Medicine AIDS Program, 135 College St., Ste. 323, New Haven, CT 06510-2483 (Jasonandr@gmail.com). The Journal of Infectious Diseases 2007; 196:S482 90 2007 by the Infectious Diseases Society of America. All rights reserved. 0022-1899/2007/19611S3-0007$15.00 DOI: 10.1086/521121 resistant strain (i.e., primary resistance). Diagnosis is established by laboratory methods requiring advanced laboratory capacity; however, these methods are not available in most resource-limited settings. Compared with treatment of drug-susceptible TB, treatment of MDR-TB is longer, more complicated, more expensive, and less successful [2]. Although efforts to expand drug-susceptibility testing (DST) and the availability of second-line drug (SLD) therapy have been emphasized over the past decade, the majority of MDR-TB cases that occur globally are still undiagnosed and untreated. The confluence of the MDR-TB and HIV epidemics could facilitate a precipitous rise in the global burden of MDR-TB [3]. To date, drug-resistance surveillance in the region of sub-saharan Africa, where there is a high prevalence of HIV infection, has been limited by poor health infrastructure and a paucity of laboratories capable of performing DST. Fewer than one-half of African countries have been surveyed since 1994, and trend data are available from only 1 country, Botswana [4]. From the available data, it has been estimated that S482 JID 2007:196 (Suppl 3) Andrews et al.

nearly 60,000 cases, or 14% of the global burden, occur annually in Africa. Of the World Health Organization (WHO) regions, Africa has the lowest percentage of national TB programs providing MDR-TB management [5]. These circumstances, combined with the maturing HIV epidemic and an overburdened health care infrastructure, may portend high rates of mortality and a rapidly expanding epidemic of MDR-TB in sub-saharan Africa [3]. The recent emergence of extensively drug-resistant (XDR) TB worldwide and, specifically, the large numbers of cases occurring in South Africa represent a serious threat to both TB and HIV programs throughout South Africa and the region [6, 7]. South Africa is home to one of the largest populations of HIV-infected individuals in the world and has more patients receiving antiretroviral therapy (ART) than does any other country [8]. The increase in TB drug resistance in this context may undermine the strides that the national ART rollout has made and may potentially limit its successful and continued expansion. In the present article, we review the recent history of drug-resistant TB in South Africa, describe its interaction with the HIV epidemic and the resultant consequences, and discuss measures for its control in the context of populations with a high prevalence of HIV infection. DRUG-RESISTANT TB IN SOUTH AFRICA With nearly 20% of its adult population infected with HIV [9], South Africa is home to one of the world s worst HIV epidemics and not unrelated one of the highest burdens of TB. In 2005, it was estimated that 285,000 incident cases of TB developed in the country, representing the seventh highest total number of incident cases in the world and the second highest total in Africa [5]. Fifty-eight percent of patients with TB were coinfected with HIV. Of significant concern, the TB cure rate (54%) was the lowest TB cure rate among the 10 countries with the highest TB burden. Compared with most sub-saharan African countries, South Africa has an advanced public health infrastructure and far greater capacity for TB drug-resistance surveillance. Between 1965 and 1991, a total of 25 annual surveys of drug resistance were conducted by the Tuberculosis Research Institute of the South African Medical Research Council (MRC) [10]. The results of these studies indicated a dramatic decrease in the prevalence of both primary and acquired drug-resistant TB. In 1995, the MRC s TB research program was suspended because of budgetary constraints. Several smaller drug-resistance surveys, including 2 provincial surveys, were undertaken, but national data were not collected again until 2001. The data from these interim surveys, described below, suggested that the prevalence of MDR-TB remained low in the mid-1990s but began to increase in the latter half of the decade [11 14]. Smaller, localized studies of TB drug resistance in the mid- 1990s found generally low rates of MDR-TB among patients with newly treated TB (!1%) and among patients with previously treated TB (!3%) [11, 14, 15]. TB drug resistance was not associated with HIV infection, and primary drug resistance appeared to be limited, with acquired resistance playing a dominant role. Between 1993 and 1997, an outbreak of TB resistant to 4 first-line drugs isoniazid, rifampicin, ethambutol, and streptomycin affected at least 21 individuals in Cape Town, none of whom were infected with HIV. More recently, in Western Cape, the emergence of outbreak-associated MDR-TB strains with patterns of resistance to pyrazinamide, streptomycin, and ethambutol has been reported [16, 17]. The most recent and systematic national estimates of drugresistant TB were based on a survey of 9 provinces performed between 2001 and 2002 by the Tuberculosis Lead Programme of the MRC in South Africa, which is part of the global network of supranational reference laboratories assembled by the Global Project on Anti-tuberculosis Drug Resistance Surveillance of the WHO [18]. Nationwide, MDR-TB was detected in 1.6% of isolates recovered from patients with newly diagnosed TB and in 6.6% of isolates recovered from patients with previously treated TB. One-fourth of all MDR-TB isolates were found to be resistant to all 4 first-line drugs against which they were tested, but SLD-susceptibility testing was not performed, so estimates of the percentage of XDR-TB isolates are not available (table 1). Among patients with newly diagnosed TB, the prevalence of MDR-TB ranged from 0.9% in Western Cape to 2.6% in Mpumalanga. The prevalence of MDR-TB among previously treated patients with TB ranged from 3.9% in Western Cape to 13.7% in Mpumalanga. In KwaZulu-Natal, the prevalence of MDR- TB among patients with newly diagnosed TB (1.7%), as well as that among patients with previously treated TB (7.7%), was similar to the national average. However the sampling strategy was not fully successful; some sites were oversampled, and many were undersampled, possibly hiding hot spots within the province. Nevertheless, because of population size and TB prevalence, the largest number of cases was estimated to have arisen in KwaZulu-Natal. In the 5 years since the last formal survey was conducted, it is likely that the number of cases of drugresistant TB has increased substantially in South Africa, in parallel with the increase in the number of TB cases, fueled by the dramatic increase in the HIV epidemic. XDR-TB The emergence of XDR-TB defined as TB that is resistant to isoniazid, rifampicin, quinolones, and at least 1 of 3 injectable SLDs (i.e., kanamycin, capreomycin, or amikacin) in every region of the world has raised additional alarms about the MDR-TB, XDR-TB, and HIV in South Africa JID 2007:196 (Suppl 3) S483

Table 1. Any drug resistance and multidrug resistance (MDR) among patients with culture-positive tuberculosis (TB), by South African province, 2001 2002. Patients with any drug resistance, % Patients with MDR-TB, % Province With no history of TB treatment a With a history of TB treatment a With no history of TB treatment a With a history of TB treatment a Eastern Cape 11.3 17.7 1.0 7.4 Free State b 8.6 9.2 1.8 1.7 Gauteng 6.6 12.7 1.4 5.5 KwaZulu-Natal 6.6 18.4 1.7 7.7 Limpopo 7.1 17.0 2.4 6.8 Mpumalanga 9.4 23.4 2.6 13.7 North West 8.1 19.1 2.2 6.9 Western Cape 5.6 7.9 0.9 3.9 South Africa c 7.7 15.5 1.6 6.6 NOTE. Current levels of resistance are not known and are likely higher (see the Drug-Resistant TB in South Africa section). Adapted from [18] with permission from K. Weyer. a According to patient and/or records. b Rates suspect because of misclassification of previous TB treatment. c National rate weighted by province. future of TB control [7]. A review of global DST data found 347 isolates of XDR-TB (then defined as MDR-TB with additional resistance to any 3 classes of SLDs) worldwide, accounting for 2% of all TB isolates surveyed and 15% of MDR- TB isolates; importantly, data from African and Asian countries (other than South Korea) and information on HIV status were not available for inclusion [7]. In 2005, large numbers of patients with MDR-TB and XDR- TB were identified at a rural hospital in Tugela Ferry, KwaZulu- Natal. Systematic surveillance undertaken at the hospital between January 2005 and March 2006 revealed that, of 542 patients with positive sputum TB culture results, 221 (41%) had MDR-TB and 53 (10%) had TB caused by M. tuberculosis strains with resistance to all 6 drugs tested (isoniazid, rifampicin, ethambutol, streptomycin, ciprofloxacin, and kanamycin) [6]. The mortality rate among patients with XDR-TB was exceptionally high (98%), with a median duration of survival of just 16 days from the time of collection of diagnostic sputum samples. All patients for whom HIV status was known were HIV positive. Most patients with XDR-TB had never been treated for TB, and high levels of genetic clustering were observed among isolates. Together, these findings were considered to be evidence of primary drug resistance associated with recent transmission, and the high proportion of patients who were hospitalized in the past year was suggestive of nosocomial transmission. The XDR-TB related deaths noted among several health care workers at this hospital further supported this view. The high prevalence of MDR-TB (present among 41% of all culture-positive patients) was of further concern, and preliminary data from a case-control study revealed an alarming mortality rate of 68% among patients with MDR- TB at this site [19]. From January 2005 to March 2007, a total of 433 patients with MDR-TB, including 239 patients with XDR-TB, were identified at the Church of Scotland Hospital in KwaZulu-Natal [20]. XDR-TB isolates have been recovered in 140 health care facilities across the province and in every province in the country. XDR-TB has now been reported in HIV-uninfected individuals in South Africa [21]. Mortality remains high, and mathematical projections suggest that, even with the implementation of effective infection-control measures, the incidence of XDR- TB will continue to increase in the coming years [22]. HIV INFECTION AND MDR-TB AND XDR-TB The association between HIV infection and the development of MDR-TB has not yet been fully clarified. Among the most important unresolved questions is whether HIV infection is an independent risk factor for the development of MDR-TB. Several studies have shown increased rates of drug-resistant TB among HIV-infected individuals [23 25], whereas other studies, including the global network of supranational reference laboratories assembled by the Global Project on Anti-tuberculosis Drug Resistance Surveillance of the WHO, have failed to confirm this finding [14, 15, 26, 27]. The South African national TB survey conducted in 2001 found no significant difference in the prevalence of HIV infection between patients with drug-susceptible TB and those with drug-resistant TB [18]. The higher rates of drug-resistant TB found in the smaller studies could result from the fact that recently circulating strains are more likely to be drug resistant, and HIV-infected individuals are more likely to manifest TB disease more rapidly (as primary disease). Consequently, HIV-infected individuals are disproportionately represented in the early stages of outbreaks S484 JID 2007:196 (Suppl 3) Andrews et al.

of drug-resistant TB. The well-known institution-based outbreaks of MDR-TB in the 1990s, as well as the XDR-TB cases that developed in KwaZulu-Natal, for example, occurred disproportionately among HIV-infected individuals [6, 28 30]. Although the epidemiological impact of HIV infection on TB drug resistance remains to be established, several clinical observations have been made concerning the association between HIV disease and the development of drug resistance. Patients who are coinfected with HIV have varying degrees of intestinal absorption of TB drugs [31, 32] and of treatment failure with standard regimens [33, 34], both of which potentially increase the risk of acquiring or amplifying TB drug resistance [35]. Regardless of whether HIV infection is an independent risk factor for the development of MDR-TB at the individual level, the increase in the pool of immunocompromised patients who serve as both hosts and vectors for all forms of TB, including MDR-TB and XDR-TB, is certain to increase the absolute burden of drug-resistant TB at the population level. Moreover, at the programmatic level, the HIV epidemic has overwhelmed and disrupted established TB-control programs, causing increases in the treatment failure rates and increasing the opportunity for drug-resistant TB to emerge and spread among both HIV-infected and -uninfected persons. RESPONDING TO DRUG-RESISTANT TB IN POPULATIONS OF SOUTH AFRICA WITH A HIGH PREVALENCE OF HIV INFECTION HIV infection adversely affects the prevalence and outcome of MDR-TB and XDR-TB. In turn, MDR-TB and XDR-TB create growing threats to the response to HIV infection, including the ART rollout (Appendix A). Although an established infrastructure for TB involving directly observed treatment, short course (DOTS), predated that of ART clinics in the country, newly established HIV treatment programs have been developed independently of TB programs, and both have continued to function largely separately in most areas of the country. This represents a missed opportunity for the programs to strengthen one another and potentially improve the outcome for both diseases [36, 37]. With the increased attention and resources in South Africa in response to the XDR-TB crisis, this is a critical time in which to identify ways that the responses to both TB and HIV infection can be coordinated and integrated. Below, we discuss the control and management of MDR-TB and XDR-TB in populations of South Africa with a high prevalence of HIV infection (Appendix B). Infection control. MDR-TB and XDR-TB can cause explosive nosocomial outbreaks in HIV-infected populations, as is evidenced by the MDR-TB outbreaks that occurred in New York and elsewhere in the mid-1990s [30]. Moreover, compared with drug-susceptible TB, MDR-TB is associated with a longer period of infectiousness, resulting from delayed recognition of drug resistance and inappropriate therapy, and may generate more secondary cases [38]. In South Africa, large common patient wards without basic measures for airborne infection control, combined with the high HIV infection prevalence among inpatients, provide a prime setting for the rapid spread of drug-resistant organisms. Indeed, nosocomial transmission is suspected to have been responsible for many of the initial 53 XDR-TB cases reported from rural KwaZulu-Natal, and further investigation has revealed previous hospitalization to be a strong risk factor for MDR-TB and XDR-TB [6, 19]. In naturally ventilated hospitals, such as most hospitals in South Africa, a recent study combining a carbon dioxide tracer gas technique with the Wells-Riley model of airborne infection revealed that as many as one-third of patients can be infected by exposure to a single infectious patient for 24 h [39]. Very few hospitals have the capacity to isolate or even create cohorts of patients with MDR-TB and XDR-TB, and nosocomial superinfection may be rife. HIV clinics are sites of clustering of TBsusceptible individuals, and, in the presence of unrecognized TB [40], they provide another opportune setting for the rapid dissemination of drug-resistant TB. Improvements in infection control, active screening for TB symptoms, heightened clinical suspicion for drug-resistant TB, and regular DST surveillance in areas of patient congregation are essential to prevent or to allow early identification of outbreaks of drug-resistant TB in South African hospitals and clinics. Although challenging, simple and relatively low-cost measures could have a cumulative effect on reducing nosocomial transmission and should be emphasized and monitored [41, 42]. Administrative and personal infection-control measures, such as encouraging the use of respirator masks, practicing proper cough hygiene, isolating patients with suspected or diagnosed drug-resistant TB, reducing hospitalization rates, reducing the length of inpatient stay, and introducing rapid DST may individually and collectively reduce transmission. Furthermore, environmental measures in particular, improving natural ventilation, which is low cost and highly effective [39], as well as installing high-efficiency particulate air filters and implementing UV germicidal irradiation may also reduce transmission in settings with a high prevalence of HIV infection [41]. Protecting health care staff from infection with drug-resistant M. tuberculosis is critical; failure to provide a safe environment for health care workers will jeopardize both TB and HIV programs. In South Africa, there have been reports of health care workers refusing to care for patients with XDR-TB, requesting reassignment, and, in some cases, leaving their posts. In most South African health care settings, including MDR-TB referral hospitals, N-95 particulate respirator masks are not available for staff. Even if such masks are available, the culture of not MDR-TB, XDR-TB, and HIV in South Africa JID 2007:196 (Suppl 3) S485

wearing masks is long-standing. Given the burden of HIV infection among health care providers, it is essential to make particulate respirator masks universally available and to develop policies and foster practices for their use. All staff should also be encouraged to undergo HIV testing, with the recommendation for discreet reassignment of any HIV-infected staff from TB wards, to reduce the risk of drug-resistant TB among health care providers. Clinical management. Diagnosis of any form of TB, including MDR-TB and XDR-TB, is more challenging in the presence of HIV disease [43 45]. South Africa has one of the most advanced TB laboratory systems in sub-saharan Africa; its 18 laboratories for culture and DST performed 207,319 cultures and 17,418 DSTs in 2005 [46]. Nevertheless, culture and DST for all patients with newly diagnosed TB are not currently recommended by the guidelines of the South African TB-control program [47]. Thus, patients with unrecognized MDR-TB and XDR-TB are likely to receive inadequate therapy and spread resistant organisms to others before a culture and DST are performed. Culture and DST targeted at patients with relapse, treatment default, or treatment failure are currently supported by the guidelines; as the numbers of cases of MDR-TB and XDR-TB continue to increase, however, the extension of culture and DST to all patients with suspected TB will be necessary to minimize mortality and continued transmission. To achieve this goal, an increase in laboratory capacity for DST is critical. Moreover, resistance testing for SLD, in particular, is not routinely performed in most provinces. XDR-TB may continue to be undiagnosed unless the capacity for SLD DST is more widely established. Compared with therapy for drug-susceptible TB, treatment of MDR-TB and XDR-TB requires a longer duration; is considerably more complicated, expensive, and toxic; and results in lower treatment success rates [2, 48, 49]. HIV-infected individuals undergoing treatment for MDR-TB have lower rates of treatment success and higher mortality rates than do HIV-uninfected patients [50, 51]. One study of a case series of patients with MDR- TB in South Africa found that the treatment success rate for HIV-uninfected patients was 53%, compared with 38% for HIVinfected patients [52]. These treatment success rates, even among HIV-uninfected individuals, are significantly lower than the international norms for a well-functioning MDR-TB program [49]. The South African Tuberculosis Control Programme s policy of employing standardized SLD regimens for all patients requires reevaluation and consideration of selection of individual regimens on the basis of DST findings. Although the benefit of ART on outcomes in patients with HIV infection and drug-resistant TB has not been rigorously assessed, its well-established mortality benefit, in conjunction with treatment for drug-susceptible TB, will likely extend to MDR-TB and XDR-TB [53 55]. Encouraging HIV testing of all patients with TB and encouraging timely initiation of ART are essential components in the management of MDR-TB and XDR-TB. The separation of sites for MDR-TB and XDR-TB treatment from sites for ART may lead to delays in initiation of TB treatment or ART and to logistical difficulties for coinfected patients, potentially resulting in treatment default [56]. The use of SLDs concomitantly with antiretroviral drugs may result in problematic drug-drug interactions as well as additive adverse effects and toxicities. One of the greatest challenges to successful MDR-TB and XDR-TB treatment in South Africa is rapid death associated with HIV coinfection. In an observational cohort study conducted in Tugela Ferry, the median duration of survival from the time of collection of sputum samples for culture and DST was just 22 days for patients with MDR-TB and 14 days for patients with XDR-TB; both periods are much shorter than the period required for conventional culture and DST results to become available [19]. Most patients with MDR-TB and XDR- TB in this setting are dying before drug-resistant TB is diagnosed. The current system requires a waiting period of 4 6 weeks for DST results, if available, before action can be taken. Early identification, isolation, and treatment of patients with suspected cases of MDR-TB and XDR-TB are critical to stem ongoing transmission in the general TB ward, where those with drug-susceptible TB are exposed to patients with MDR-TB or XDR-TB. Rapid death and a high risk of nosocomial transmission of MDR-TB and XDR-TB have stimulated intensive interest in the development and use of rapid DSTs. Several promising techniques for rapid, low-cost diagnosis have recently emerged and are gaining increasingly widespread assessment and use. Notable among these techniques are the microscopic-observation drug-susceptibility (MODS) technique [57], which will soon debut in trials in South Africa, and 2 molecular assays (a line probe assay [Genotype MTBDRplus assay; Hain Lifesciences and FIND Diagnostics] and a phage-based assay [FastPlaque; Biotec Laboratories and FIND Diagnostics]), both of which will be launched in large trials in the country this year. MODS has been validated for 4 first-line drugs, and its culture and DST results are typically available after a median duration of 7 days. However, data from settings where the prevalence of HIV infection is high are limited. The molecular tests can be performed directly on sputum samples and can provide results within 2 days, but they determine only whether the organisms are resistant to rifampicin, and they perform poorly in patients with smear-negative results [58, 59]. Because, in many settings, rifampicin resistance is almost always accompanied by isoniazid resistance, some experts have suggested that empirical SLDs and third-line drugs be considered for HIVinfected patients who have a positive rapid rifampicin test result while awaiting confirmation of results by culture and DST [60]. S486 JID 2007:196 (Suppl 3) Andrews et al.

None of the proposed rapid technologies presently are configured to identify resistance to SLDs. In light of the emergence of XDR-TB, extension of this technology to SLDs is urgently needed. Programmatic strengthening, integration, and decentralization. HIV testing among patients with TB is one of the most essential components of TB/HIV integration and may serve as a critical entry point into HIV treatment and care in settings with high prevalences of HIV and TB [34, 61]. In South Africa, only 22% of patients with TB underwent HIV testing in 2005, despite a national policy of providing counseling and testing for all patients with TB [5]. This low percentage reflects the lack of strong TB/HIV collaborative activities and indicates an urgent need for integration of the responses of the HIV and TB programs. Similarly, aggressive screening of HIV-infected patients for TB will enable early diagnosis of all forms of TB, including drug-resistant TB, thereby potentially reducing mortality, the duration of infectiousness, and transmission of TB to others. Treatment of HIV/AIDS is a critical component in the control of MDR-TB and XDR-TB. ART has been shown to reduce TB incidence, and this will likely extend to MDR- TB and XDR-TB [40]. The control and management of MDR-TB in South Africa are ostensibly based on the WHO DOTS-Plus model, with MDR-TB treatment, including SLD, available to all patients with documented MDR-TB. However, the current capacity for MDR-TB control and management in South Africa does not approach international standards of care. South Africa does not have an MDR-TB treatment program that is supported by the Green Light Committee, which provides access to low-cost SLDs; the choice of SLDs has been limited, and capreomycin and para-aminosalicylic acid have become available to MDR- TB treatment facilities only very recently. Treatment success rates for MDR-TB are!50%, and the mortality rate is 120% [46, 52]. Treatment default has been a major problem [46, 62]. Because there are few facilities that provide SLDs, there is massive overcrowding and long waiting lists for inpatient therapy in some areas. When treated in inpatient settings, patients are at risk for superinfection with more-resistant organisms, including XDR-TB, because of poor infection control, inadequate facilities for isolation, and prolonged inpatient stays. Patients receiving outpatient therapy often have to travel long distances on a monthly basis to obtain treatment, which puts them at high risk for treatment default. As the numbers of cases of MDR-TB and XDR-TB continue to increase in South Africa, this MDR-TB treatment system is becoming overwhelmed, and immediate solutions are needed. Decentralization of treatment is necessary to address some of these problems and requires substantial increases in material and human resources. Just as, in some settings, the ART rollout has adopted lessons learned from the success of the TB DOTS model, the management of MDR-TB and XDR-TB in the outpatient setting may benefit from models of health care delivery brought about by HIV care. The duration of therapy is typically 3 4 times as long as that of short-course chemotherapy for TB and requires more medical oversight and management. The ART rollout in South Africa and elsewhere faced these same challenges of providing chronic care with a need for practical, careful, and longterm oversight. To overcome these challenges, HIV care programs have developed strong medication adherence programs, which include treatment literacy training; community-based therapy, including the use of community health workers or treatment supporters for observing therapy and monitoring for illnesses and drug toxicities; nutritional support with food parcels and micronutrient supplements; clinic transportation support or stipends; and other techniques [53]. Such a model of delivery of care for MDR-TB has been utilized in Peru with great success [63]. Developing such community-based treatment for MDR-TB and XDR-TB and integrating it with HIV care and treatment in South Africa is a daunting challenge that must now be faced and met. CONCLUSION The emergence of MDR-TB and XDR-TB in settings where the prevalence of HIV infection is high is no longer a worrisome possibility it has already occurred in South Africa. It has been 5 years since systematic surveys have been performed in the country, and the high numbers of cases of MDR-TB and XDR- TB being reported in such settings across the country more likely represent a growing problem of endemicity rather than sporadic outbreaks. The high prevalence of both TB and HIV, with weakening of TB programs and deepening of populationbased immunosuppression, has created this disastrous scenario. The difficulties associated with the diagnosis, control, and management of drug-resistant TB urgently require plans and action. An approach that coordinates and integrates the responses to HIV and MDR-TB/XDR-TB is necessary to avert the faltering or the collapse of the ART rollout and TB DOTS programs. Acknowledgments Supplement sponsorship. This article was published as part of a supplement entitled The Realities of Antiretroviral Therapy Rollout: Challenges to Successful Programmatic Implementation, sponsored by the Harvard Medical School Division of AIDS, the Harvard University Center for AIDS Research, and the Harvard Initiative for Global Health. MDR-TB, XDR-TB, and HIV in South Africa JID 2007:196 (Suppl 3) S487

APPENDIX A THREATS POSED BY MULTIDRUG-RESISTANT AND EXTENSIVELY DRUG-RESISTANT TUBERCULOSIS (TB) TO ANTIRETROVIRAL THERAPY ROLLOUT AND TB PROGRAMS Increased morbidity and mortality among patients with TB/HIV coinfection Transmission of drug-resistant TB to HIV-infected patients in hospital inpatient settings Transmission of drug-resistant TB in congregate HIV outpatient care settings Risk of and concerns associated with transmission of drug-resistant TB to health care staff Additive complexities and complications of treatment with antiretroviral agents and second-line TB drugs Further overburdening of TB diagnostic and treatment facilities Potential reversal of positive gains in TB and HIV program collaboration and integration Competition for human and financial resources APPENDIX B ELEMENTS OF SUCCESSFUL RESPONSE TO MULTIDRUG-RESISTANT (MDR) AND EXTENSIVELY DRUG-RESISTANT (XDR) TUBERCULOSIS (TB) IN SOUTH AFRICA IN THE CONTEXT OF HIGH LEVELS OF HIV INFECTION PREVALENCE Improving diagnosis Enhancing surveillance for drug-resistant TB Expanding laboratory capacity for first-line and secondline drug-susceptibility testing Testing and deploying rapid diagnostic procedures and strategies to identify MDR-TB and XDR-TB Improving rates of HIV testing among patients with MDR-TB or XDR-TB Enhancing screening for TB among HIV-infected patients Enhancing MDR-TB and XDR-TB contact tracing Reducing transmission Instituting and monitoring infection-control measures in congregate care settings Encouraging HIV testing for health care workers to reduce the risk of TB transmission Improving treatment Accelerating universal access to antiretroviral therapy (ART) Making ART accessible in a timely manner to patients with MDR-TB and XDR-TB Increasing expertise in and capacity for MDR-TB and XDR-TB treatment Establishing decentralized and/or community-based MDR-TB and XDR-TB treatment Reducing the MDR-TB and XDR-TB treatment default rate by addressing barriers to care Improving procurement and availability of second-line drugs Evaluating potential treatment regimens for XDR-TB among HIV-infected patients Developing and more rapidly testing new and novel anti- TB drugs, as well as reducing the time to their approval References 1. Zignol M, Hosseini MS, Wright A, et al. Global incidence of multidrugresistant tuberculosis. J Infect Dis 2006; 194:479 85. 2. Iseman M. Treatment of multidrug-resistant tuberculosis. N Engl J Med 1993; 329:784 91. 3. Wells CD, Cegielski JP, Nelson LJ, et al. HIV infection and multidrugresistant tuberculosis the perfect storm. J Infect Dis 2007; 196(Suppl 1):S86 107. 4. Aziz MA, Wright A, Laszlo A, et al. Epidemiology of antituberculosis drug resistance (the Global Project on Anti-tuberculosis Drug Resistance Surveillance): an updated analysis. Lancet 2006; 368:2142 54. 5. World Health Organization (WHO). Global tuberculosis control: surveillance, planning, financing: WHO report 2007. WHO/HTM/TB/ 2007.376. Geneva: WHO, 2007. 6. Gandhi NR, Moll A, Sturm AW, et al. Extensively drug-resistant tuberculosis as a cause of death in patients co-infected with tuberculosis and HIV in a rural area of South Africa. Lancet 2006; 368:1575 80. 7. Shah N, Wright A, Bai GH, et al. Worldwide emergence of extensively drug-resistant tuberculosis (XDR TB): global survey of second-line drug resistance among Mycobacterium tuberculosis isolates. Emerg Infect Dis 2007; 13:380 7. 8. World Health Organization (WHO). Towards universal access: scaling up priority HIV/AIDS interventions in the health sector; progress report, April, 2007. Geneva: WHO, 2007. 9. Joint United Nations Programme on HIV/AIDS (UNAIDS). 2006 Report on the global AIDS epidemic. Geneva: UNAIDS, 2006. 10. Weyer K, Kleeberg HH. Primary and acquired drug resistance in adult black patients with tuberculosis in South Africa: results of a continuous national drug resistance surveillance programme involvement. Tuber Lung Dis 1992; 73:106 12. 11. Davies GR, Pillay M, Sturm AW, Wilkinson D. Emergence of multidrug-resistant tuberculosis in a community-based directly observed treatment programme in rural South Africa. Int J Tuberc Lung Dis 1999; 3:799 804. 12. Weyer K, Groenewald P, Zwarenstein M, Lombard CJ. Tuberculosis drug resistance in the Western Cape. S Afr Med J 1995; 85:499 504. 13. Weyer K, Lancaster J, Balt E, Durrheim D. Tuberculosis drug resistance in Mpumalanga province, South Africa (abstract 669-PD). In: Proceedings of the Global Congress on Lung Health, 29th World Conference of the International Union against Tuberculosis and Lung Disease (Bangkok). Int J Tuberc Lung Dis 1998; 2:S165. 14. Wilkinson D, Pillay M, Davies GR, Sturm AW. Resistance to antituberculosis drugs in rural South Africa: rates, patterns, risks, and transmission dynamics. Trans R Soc Trop Med Hyg 1996; 90:692 5. 15. Churchyard GJ, Corbett EL, Kleinschmidt I, Mulder D, De Cock KM. S488 JID 2007:196 (Suppl 3) Andrews et al.

Drug-resistant tuberculosis in South African gold miners: incidence and associated factors. Int J Tuberc Lung Dis 2000; 4:433 40. 16. Louw GE, Warren RM, Donald PR, et al. Frequency and implications of pyrazinamide resistance in managing previously treated tuberculosis patients. Int J Tuberc Lung Dis 2006; 10:802 7. 17. Victor TC, Streicher EM, Kewley C, et al. Spread of an emerging Mycobacterium tuberculosis drug-resistant strain in the Western Cape of South Africa. Int J Tuberc Lung Dis 2007; 11:195 201. 18. Weyer K, Lancaster J, Brand J, van der Walt M, Levin J. Survey of tuberculosis drug resistance in South Africa: 2001 2002. Pretoria, South Africa: Medical Research Council of South Africa, 2004. 19. Andrews J. Clinical predictors of drug resistance and mortality among tuberculosis patients in a rural South African hospital: a case-control study. New Haven, CT: Yale AIDS Program, Department of Internal Medicine, Yale University School of Medicine, 2007:79. 20. Moll A, Gandhi N, Andrews J, et al. Extensively drug-resistant (XDR) TB now more common than MDR TB in Tugela Ferry, KwaZulu Natal, South Africa. In: Program and abstracts of the 38th Union World Conference on Lung Health (Cape Town). Paris: International Union Against Tuberculosis and Lung Disease, 2007. 21. Moodley P, Moll A, Shah NS, et al. Multi- and extensively drug-resistant Mycobacterium tuberculosis in KwaZulu Natal. In: Program and abstracts of the 47th Interscience Conference on Antimicrobial Agents and Chemotherapy (Chicago). Washington, DC: American Society for Microbiology, 2007. 22. Basu S, Andrews J, Poolman EM, et al. The epidemic-level impact of preventing nosocomial transmission of extensively drug-resistant (XDR) tuberculosis in rural South African district hospitals. The Lancet 2007 (in press). 23. Campos PE, Suarez PG, Sanchez J, et al. Multidrug-resistant Mycobacterium tuberculosis in HIV-infected persons, Peru. Emerg Infect Dis 2003; 9:1571 8. 24. Gordin FM, Nelson ET, Matts JP, et al. The impact of human immunodeficiency virus infection on drug-resistant tuberculosis. Am J Respir Crit Care Med 1996; 154:1478 83. 25. Punnotok J, Shaffer N, Naiwatanakul T, et al. Human immunodeficiency virus-related tuberculosis and primary drug resistance in Bangkok, Thailand. Int J Tuberc Lung Dis 2000; 4:537 43. 26. Kenyon TA, Mwasekaga MJ, Huebner R, Rumisha D, Binkin N, Maganu E. Low levels of drug resistance amidst rapidly increasing tuberculosis and human immunodeficiency virus co-epidemics in Botswana. Int J Tuberc Lung Dis 1999; 3:4 11. 27. World Health Organization (WHO). Anti-tuberculosis drug resistance in the world: third global report. WHO/International Union against Tuberculosis and Lung Disease Global Project on Anti-tuberculosis Drug Resistance Surveillance. WHO/HTM/TB/2004.343. Geneva: WHO, 2004. 28. Centers for Disease Control and Prevention. Multidrug-resistant tuberculosis outbreak on an HIV ward Madrid, Spain, 1991 1995. MMWR Morb Mortal Wkly Rep 1996; 45:330 3. 29. Moro ML, Gori A, Errante I, et al. An outbreak of multidrug-resistant tuberculosis involving HIV-infected patients of two hospitals in Milan, Italy. Italian Multidrug-Resistant Tuberculosis Outbreak Study Group. AIDS 1998; 12:1095 102. 30. Frieden TR, Sherman LF, Maw KL, et al. A multi-institutional outbreak of highly drug-resistant tuberculosis: epidemiology and clinical outcomes. JAMA 1996; 276:1229 35. 31. Tappero JW, Bradford WZ, Agerton TB, et al. Serum concentrations of antimycobacterial drugs in patients with pulmonary tuberculosis in Botswana. Clin Infect Dis 2005; 41:461 9. 32. Patel KB, Belmonte R, Crowe HM. Drug malabsorption and resistant tuberculosis in HIV-infected patients. N Engl J Med 1995; 332:336 7. 33. Chaisson RE, Clermont HC, Holt EA, et al. Six-month supervised intermittent tuberculosis therapy in Haitian patients with and without HIV infection. Am J Respir Crit Care Med 1996; 154:1034 8. 34. Murray J, Sonnenberg P, Shearer SC, Godfrey-Faussett P. Human immunodeficiency virus and the outcome of treatment for new and recurrent pulmonary tuberculosis in African patients. Am J Respir Crit Care Med 1999; 159:733 40. 35. Ridzon R, Whitney CG, McKenna MT, et al. Risk factors for rifampin mono-resistant tuberculosis. Am J Respir Crit Care Med 1998; 157: 1881 4. 36. Friedland G, Abdool Karim S, Abdool Karim Q, et al. Utility of tuberculosis directly observed therapy programs as sites for access to and provision of antiretroviral therapy in resource-limited countries. Clin Infect Dis 2004; 38(Suppl 5):S421 8. 37. Friedland G, Churchyard GJ, Nardell E, eds. Tuberculosis and HIV coinfection: current state of knowledge and research priorities. J Infect Dis 2007; 196(Suppl 1):S1 125. 38. Blower SM, Chou T. Modeling the emergence of the hot zones : tuberculosis and the amplification dynamics of drug resistance. Nat Med 2004; 10:1111 6. 39. Escombe AR, Oeser CC, Gilman RH, et al. Natural ventilation for the prevention of airborne contagion. PLoS Med 2007; 4:e68. 40. Lawn SD, Myer L, Bekker L-G, Wood R. Burden of tuberculosis in an antiretroviral treatment programme in sub-saharan Africa: impact on treatment outcomes and implications for tuberculosis control. AIDS 2006; 20:1605 12. 41. Bock NN, Jensen PA, Miller B, Nardell E. Tuberculosis infection control in resource-limited settings in the era of expanding HIV care and treatment. J Infect Dis 2007; 196(Suppl 1):S108 13. 42. World Health Organization (WHO). Guidelines for the prevention of tuberculosis in health care facilities in resource-limited settings. Geneva: WHO, 1999. 43. Johnson JL, Vjecha MJ, Okwera A, et al. Impact of human immunodeficiency virus type-1 infection on the initial bacteriologic and radiographic manifestations of pulmonary tuberculosis in Uganda. Makerere University Case Western Reserve University Research Collaboration. Int J Tuberc Lung Dis 1998; 2:397 404. 44. Long R, Scalcini M, Manfreda J, Jean-Baptiste M, Hershfield E. The impact of HIV on the usefulness of sputum smears for the diagnosis of tuberculosis. Am J Public Health 1991; 81:1326 8. 45. Perlman DC, el-sadr WM, Nelson ET, et al. Variation of chest radiographic patterns in pulmonary tuberculosis by degree of human immunodeficiency virus-related immunosuppression. The Terry Beirn Community Programs for Clinical Research on AIDS (CPCRA). The AIDS Clinical Trials Group (ACTG). Clin Infect Dis 1997; 25:242 6. 46. Green-Thompson R. South Africa. In: Program and abstracts of the Meeting of the Global XDR-TB Task Force, 9 10 October 2006 (Geneva). Geneva: World Health Organization, 2006. 47. South African Department of Health. The South African Tuberculosis Control Programme practical guidelines. Pretoria: South African Department of Health, 2000. 48. Rajbhandary SS, Marks SM, Bock NN. Costs of patients hospitalized for multidrug-resistant tuberculosis. Int J Tuberc Lung Dis 2004;8: 1012 6. 49. Nathanson E, Lambregts-van Weezenbeek C, Rich ML, et al. Multidrug-resistant tuberculosis management in resource-limited settings. Emerg Infect Dis 2006; 12:1389 97. 50. Kawai V, Soto G, Gilman RH, et al. Tuberculosis mortality, drug resistance, and infectiousness in patients with and without HIV infection in Peru. Am J Trop Med Hyg 2006; 75:1027 33. 51. Munsiff SS, Ahuja SD, Li J, Driver CR. Public-private collaboration for multidrug-resistant tuberculosis control in New York City. Int J Tuberc Lung Dis 2006; 10:639 48. 52. van der Wal M, Weyer K, Lancaster J. A standardized approach to management of drug-resistant TB in South Africa [PowerPoint presentation]. South Africa Medical Research Council Expert Consultation on Drug-Resistant Tuberculosis. Johannesburg: Medical Research Council of South Africa, 2006. 53. Gandhi N, Moll A, Pawinski R, Zeller K, Lalloo U, Friedland G. Favorable outcomes of the integration of TB and HIV treatment in rural South Africa: the Sizonq oba study (abstract MOPE0181). In: Program MDR-TB, XDR-TB, and HIV in South Africa JID 2007:196 (Suppl 3) S489

and abstracts of the XVIth International AIDS Conference (Toronto). Geneva: International AIDS Society, 2006. 54. Dean GL, Edwards SG, Ives NJ, et al. Treatment of tuberculosis in HIV-infected persons in the era of highly active antiretroviral therapy. AIDS 2002; 16:75 83. 55. Manosuthi W, Chottanapand S, Thongyen S, Chaovavanich A, Sungkanuparph S. Survival rate and risk factors of mortality among HIV/ tuberculosis-coinfected patients with and without antiretroviral therapy. J Acquir Immune Defic Syndr 2006; 43:42 6. 56. Friedland G, Abdool Karim S, Abdool Karim Q, et al. Utility of tuberculosis directly observed therapy programs as sites for access to and provision of antiretroviral therapy in resource-limited countries. Clin Infect Dis 2004; 38(Suppl 5):S421 8 57. Moore DAJ, Evans CAW, Gilman RH, et al. Microscopic-observation drug-susceptibility assay for the diagnosis of TB. N Engl J Med 2006; 355:1539 50. 58. Morgan M, Kalantri S, Flores L, Pai M. A commercial line probe assay for the rapid detection of rifampicin resistance in Mycobacterium tuberculosis: a systematic review and meta-analysis. BMC Infect Dis 2005; 5:62. 59. Pai M, Kalantri S, Pascopella L, Riley LW, Reingold AL. Bacteriophagebased assays for the rapid detection of rifampicin resistance in Mycobacterium tuberculosis: a meta-analysis. J Infect 2005; 51:175 87. 60. Global XDR-TB Task Force working group. Management of XDR TB suspects in high and low HIV prevalence settings [PowerPoint presentation]. Meeting of the Global XDR-TB Task Force, 9 10 October 2006 (Geneva). Geneva, World Health Organization, 2006. 61. World Health Organization (WHO). Interim policy on collaborative TB/HIV activities. Geneva: WHO, 2004. 62. Holtz TH, Lancaster J, Laserson KF, Wells CD, Thorpe L, Weyer K. Risk factors associated with default from multidrug-resistant tuberculosis treatment, South Africa, 1999 2001. Int J Tuberc Lung Dis 2006; 10:649 55. 63. Mitnick C, Bayona J, Palacios E, et al. Community-based therapy for multidrug-resistant tuberculosis in Lima, Peru. N Engl J Med 2003; 348:119 28. S490 JID 2007:196 (Suppl 3) Andrews et al.