Abstract: Most patients with multidrug-resistant tuberculosis (MDR-TB) in South Africa are HIV-infected, but the safety and tolerability of cotreatment are unknown. The authors reviewed all adverse events (AEs) for patients with MDR-TB in a home-based treatment program in rural KwaZulu-Natal. Of 91 MDR-TB patients, 74 (81%) were HIV-positive and receiving antiretroviral therapy. AEs were common, but most were mild and did not require therapy modification. The most common severe AEs were hypothyroidism (36%) and psychosis (5%). Patients receiving concurrent antiretroviral therapy did not experience AEs more frequently than those on MDR-TB therapy alone. Concurrent treatment for MDR-TB/HIV can be safely administered in a home-based care setting.
*Department of Medicine, Montefiore Medical Center and Albert Einstein College of Medicine, Bronx, NY
†Philanjalo and Church of Scotland Hospital, Tugela Ferry, South Africa
‡Centre for the AIDS Programme of Research in South Africa (CAPRISA), Durban, South Africa
§Department of Epidemiology and Population Health, Montefiore Medical Center and Albert Einstein College of Medicine, Bronx, NY
‖Health Systems Research Unit, Medical Research Council, Cape Town, South Africa
¶Department of Medicine, Nelson R. Mandela School of Medicine, University of KwaZulu-Natal, Durban, South Africa; Departments of
**Epidemiology, Yale University School of Medicine, New Haven, CT
††Department of Epidemiology, Emory University Rollins School of Public Health, Atlanta, GA.
Correspondence to: James C. M. Brust, MD, Division of General Internal Medicine, Montefiore Medical Center, 111 E. 210th Street, Bronx, NY 10467 (e-mail: firstname.lastname@example.org).
Supported by the US President’s Emergency Plan for AIDS Relief. Statistical support was provided by the Center for AIDS Research at Albert Einstein College of Medicine/Montefiore Medical Center (P30 AI051519).
J. C. M. Brust is supported by the National Institutes of Health (K23 AI083088). N. R. Gandhi and N. S. Shah are recipients of the Doris Duke Charitable Foundation Clinical Scientist Development Award (NRG 2007070, NSS 2007071). G. H. Friedland is also supported by the Doris Duke Charitable Foundation (2007018), the Gilead Foundation, and the Irene Diamond Fund (R05130). All remaining authors have no funding or conflicts of interest to disclose.
Presented in part at the XIX International AIDS Conference, July 22–27, 2012, Washington, DC (Abstract THPDB0106).
Received August 26, 2012
Accepted December 04, 2012
The emergence of multidrug-resistant tuberculosis (MDR-TB; resistance to at least isoniazid and rifampin) has posed numerous challenges for TB-endemic countries worldwide. MDR-TB is associated with high mortality rates, particularly in the setting of HIV coinfection.1 Second-line TB medications are more toxic and less potent than treatment for drug-susceptible TB, and most countries have adopted centralized inpatient models of care for MDR-TB, in part because of concerns about monitoring and treating adverse drug events. However, with limited numbers of hospital beds and specialists, some countries have adopted community-based or ambulatory treatment for MDR-TB.2–4
Adverse events (AEs) associated with MDR-TB treatment have been well described2,5–11 and play an important role in treatment because they impact regimen choice, medication adherence, and retention in care. Among patients coinfected with MDR-TB and HIV, there is an additional concern of additive or synergistic toxicities when second-line TB medications are given concurrently with antiretroviral therapy (ART). However, there have been few data addressing this issue—the majority of AE reports have been in HIV-negative populations.5–11 As the HIV and drug-resistant TB epidemics converge, establishing the safety of coadministration of these treatments is essential. Moreover, as a greater proportion of patients with MDR-TB are treated in outpatient or community-based settings, it is important to establish that ambulatory treatment can be carried out safely and effectively.
The objective of this study was to examine the frequency and severity of AEs in patients with MDR-TB and HIV coinfection treated at an integrated MDR-TB/HIV home-based treatment program in KwaZulu-Natal, South Africa.
Tugela Ferry is a resource-limited rural area in KwaZulu-Natal province, South Africa, with a high incidence of MDR-TB.12 More than 80% of patients with MDR-TB are coinfected with HIV.1 The details of the ambulatory MDR-TB/HIV treatment program in Tugela Ferry have been previously described.13 Patients with suspected or proven MDR-TB are referred to the local specialized MDR-TB treatment hospital and clinic. All patients are tested for HIV and, if positive, are initiated on ART as soon as possible. After a brief admission, patients are treated at home by a visiting injection team (intensive phase) or lay treatment-supporter (continuation phase) with family support. Patients undergo clinical evaluation and laboratory testing monthly with full blood count and chemistries. Thyroid-stimulating hormone (TSH) is tested 6 monthly. Sputum smear, culture, and drug-susceptibility testing are performed monthly.
Patients are treated with a standardized MDR-TB regimen consisting of kanamycin (15 mg/kg, max 1000 mg daily), ofloxacin (800 mg daily), cycloserine (10–20 mg/kg, max 750 mg daily, divided twice a day), ethionamide (10–20 mg/kg, max 750 mg daily, divided twice a day), pyrazinamide (20–30 mg/kg, max 1600 mg daily), and ethambutol (15–20 mg/kg, max 1200 mg daily). No patients received capreomycin or para-aminosalicylic acid (PAS). The intensive phase, when patients receive kanamycin, lasts 4 months after culture conversion (6 months minimum). Total duration of treatment is a minimum of 24 months. The South African first-line ART regimen consists of efavirenz, lamivudine, and either stavudine (40 mg twice a day) or tenofovir (beginning April 2010).
AE Monitoring and Management
Patients are screened at baseline for any preexisting symptoms. After hospital discharge, they are seen monthly at the clinic by a clinician. At each visit, they are screened for any AEs using a standardized screening instrument, which includes 15 common complaints, such as peripheral neuropathy (numbness, burning, or pain), nausea, and rash. The clinician evaluates positive findings and grades them in severity using a modified Division of AIDS toxicity table.14 Modifications in therapy are at the discretion of the treating clinician. All patients undergo formal audiometry at baseline and every 1–2 months while receiving kanamycin.
We reviewed the medical records of patients with culture-confirmed MDR-TB who initiated treatment at the MDR-TB center between November 1, 2008, and April 15, 2011. Patients were excluded if they had resistance to kanamycin, amikacin, capreomycin, or fluoroquinolone.
We identified any reported AEs as of November 15, 2011. Hospital/clinic records were manually reviewed to corroborate self-reports of confusion, psychosis, depression, and insomnia and to identify any interventions. We defined psychosis as severe if cycloserine or terizidone was stopped, even if a grade was not provided. Laboratory results for hemoglobin, potassium, creatinine, and alanine aminotransferase (ALT) were graded using the Division of AIDS toxicity table. TSH was graded as: normal (0.34–5.6 mIU/L), elevated (5.6–8.0 mIU/L), and severely elevated (>8.0 mIU/L).
AEs were pooled to determine the proportion of patients who experienced at least 1 AE. We then calculated the proportion experiencing each specific AE at least once and those experiencing severe AEs (grade ≥3). Finally, we analyzed the proportion of patients experiencing each AE during 6-month time blocks. Intrapatient trends over time were analyzed using generalized estimating equations. AE proportions and trends were compared among those who did and did not receive concurrent ART.
The study was approved by the ethics committees at the University of KwaZulu-Natal, Albert Einstein College of Medicine, Yale University, and KwaZulu-Natal Department of Health.
During the study period, 101 patients with culture-confirmed MDR-TB initiated treatment at the decentralized clinic. Of these, 10 either died or were transferred during their initial hospitalization and had no AE data available. The remaining 91 patients were included in this analysis.
Fifty (55%) patients were women, and the median age was 34 years [interquartile ratio (IQR), 29–41]. Seventy-six (84%) patients were HIV coinfected, with a median CD4 count of 207 cells per cubic millimeters (IQR, 89–411 cells per cubic millimeter) at MDR-TB treatment initiation. Sixty-six (87%) HIV-infected patients were already receiving ART at MDR-TB treatment initiation, and 8 of the remaining 10 patients started ART a median of 45 days (IQR, 19–90 days) later. One patient defaulted treatment before initiating ART and the other was an elite controller. At the time of analysis, patients had been followed for a median of 652 days (IQR, 476–728 days).
Clinical AEs were extremely common, with 90 (99%) patients reporting at least 1 AE. The most common were peripheral neuropathy (73%), injection site pain (66%), rash (53%), and nausea/vomiting (42%) (Table 1). Clinician grading was inconsistent, with only 38% of AEs receiving a grade. The most common severe AEs were psychosis (5%), hearing loss (8%), and peripheral neuropathy (5%). Cycloserine or terizidone was stopped in all cases of psychosis. Four patients became severely depressed (2 were suicidal) and required discontinuation of cycloserine or terizidone. Nine (10%) patients required dose reductions of kanamycin for hearing loss, and 6 (7%) changed from stavudine to zidovudine or tenofovir because of peripheral neuropathy. Formal audiometry results were available for 35 patients, of whom 24 (69%) had some degree of hearing loss and 4 (11%) had grade 3 or higher (defined as >55 dB threshold shift in 2 or more contiguous frequencies).
Eighty-nine (98%) patients had at least 1 laboratory result available during the treatment. Laboratory AEs were common, but rarely clinically significant (Table 1). Forty-six (52%) patients had a grade ≥1 elevation in ALT, of whom 6 (7%) had a grade 3 or 4 elevation. All 6 of these elevations resolved on repeat testing without any change in regimen. Similarly, 32 (36%) patients experienced grade ≥1 elevation in potassium and 11 (12%) experienced grade 3 or 4 elevation, but all were normal on repeat testing. Mild hypokalemia was also common [37 (42%) grade ≥1], but only 3 (3%) patients had severe hypokalemia (K+ ≤2.4 mEq/L); all improved with supplemental oral potassium. Hypothyroidism was very common. Among 74 (81%) patients with at least 1 TSH result, 38 (51%) had an elevated TSH and 26 (36%) had a level >8.0 mIU/L, requiring initiation of levothyroxine. For each AE, there was no significant difference in frequency or severity between patients who were and were not receiving concurrent ART.
Changes in AEs Over Time
Examining 6-month within-treatment time blocks, we found that all clinical AEs decreased in frequency over the course of treatment (Fig. 1). These temporal decreases were statistically significant (P < 0.05), except for depression. Anemia improved over the course of treatment, as did hypokalemia and elevations in creatinine; the latter 2 may be related to discontinuation of kanamycin in the continuation phase. Hypothyroidism also improved over time, possibly because of levothyroxine therapy. The frequency of ALT elevations or hyperkalemia did not change over the course of treatment. There was no significant difference in temporal trends between patients receiving and not receiving concurrent ART. At the end of the study period, 78 (86%) patients were either cured (n = 37) or still on treatment (n = 41).
This study demonstrates the safety of concurrent treatment of MDR-TB and HIV in an ambulatory setting with home-based care. Overall, we found very high rates of AEs, but the majority were mild and did not require discontinuation of either MDR-TB treatment or ART. Interestingly, we found no significant differences in AE frequency between subjects who were and were not treated with concomitant ART, even among AEs with a plausible mechanism of additive or synergistic toxicity (eg, peripheral neuropathy and neuropsychiatric effects).
Eighteen (20%) subjects in our study either reported or were found to have confusion/psychosis, but on chart review, only 5 (5%) were considered sufficiently severe by the treating clinician to discontinue cycloserine or terizidone. Neuropsychiatric toxicity from cycloserine is well known,15 and there has been concern about coadministration with efavirenz. Four of the 5 patients with severe psychosis in our cohort received concurrent efavirenz, but this was not stopped and psychosis resolved in all patients. In Lesotho, 16% of patients with MDR-TB experienced psychosis from cycloserine, but the authors did not note how many discontinued cycloserine or how many were receiving concurrent efavirenz.2 In Istanbul, 21% of patients experienced psychiatric symptoms, but this was a combined category that included psychosis, depression, and anxiety, making it difficult to compare with our results.
Hypothyroidism was very common in our study, and 36% of patients required levothyroxine therapy. Although hypothyroidism was previously believed to be rare in MDR-TB treatment, recent reports suggest that it is more common.16–18 In Lesotho, 69% of patients with MDR-TB treated with a combination regimen including both ethionamide and PAS had a TSH level >10 mIU/L. Although hypothyroidism was less common in our cohort, none of our patients were treated with PAS, suggesting that PAS and ethionamide have an additive effect in causing hypothyroidism. Some studies reporting lower rates of hypothyroidism only measured TSH in symptomatic patients, thus missing subclinical cases.5,8
Irreversible hearing loss from kanamycin or amikacin is perhaps the most devastating AE associated with MDR-TB treatment. In our study, audiometry results were not available for most patients, and consequently, our results of self-report are a minimal estimate. The available audiometry results, however, confirm high rates of subclinical hearing loss. Further study is needed to determine if changes in dose or frequency prevent ongoing hearing loss without affecting TB outcomes. Reported rates of ototoxicity vary widely in the literature, depending largely on the availability of routine audiometry testing rather than symptom-based testing or clinical self-report. In the cohort from Istanbul, 50% of patients receiving amikacin discontinued injections prematurely because of ototoxicity.8
Nearly all AEs were reported less frequently over time, even though patients were actively asked about specific AEs at each visit throughout treatment. Although certain AEs are related to specific therapies (eg, ototoxicity), the improvement of most over time despite continuation of therapy suggests either that they represent the general discomfort of being ill (improving as the underlying illness is treated) or that patients become accustomed to medication side effects and learn to tolerate them until the end of treatment. In either case, our results are encouraging and demonstrate that patients continue treatment with favorable outcomes.13
This study has several important limitations. First, patients were not routinely screened for AE-related symptoms while they were admitted to the hospital. This may have missed AEs shortly after initiating therapy when they are likely to be more common. Second, with the exception of psychosis and confusion, AEs were primarily identified by self-report and, given the inconsistent severity grading by the clinicians, it is difficult to determine their clinical significance. Third, although we did not find a difference in the incidence of AEs between HIV-coinfected subjects receiving concomitant ART and HIV-uninfected subjects, our study only included 15 HIV-negative subjects and may not have had adequate power to detect a small difference.
With a growing case burden and limited hospital beds, South Africa has moved to a community-based treatment model for MDR-TB nationwide.19 We have found that, although AEs are frequent, patients with MDR-TB—with or without HIV coinfection—can be safely treated at home with supportive care. Our data support the recent World Health Organization treatment guidelines for MDR-TB calling for ambulatory models of care.20 With intensive patient/family education, close clinical follow-up, and rapid assessment and management of AEs, patients remain in treatment and achieve successful outcomes even in rural resource-limited settings with high HIV prevalence.21
The authors thank Dr Alois Mngadi and Sister (Prof Nurse) Lee-Megan Larkan for their outstanding clinical management of the study subjects.
1. Gandhi NR, Shah NS, Andrews JR, et al.. HIV coinfection in multidrug- and extensively drug-resistant tuberculosis results in high early mortality. Am J Respir Crit Care Med. 2010;181:80–86.
2. Seung KJ, Omatayo DB, Keshavjee S, et al.. Early outcomes of MDR-TB treatment in a high HIV-prevalence setting in Southern Africa. PLoS One. 2009;4:e7186.
3. Mitnick C, Bayona J, Palacios E, et al.. Community-based therapy for multidrug-resistant tuberculosis in Lima, Peru. N Engl J Med. 2003;348:119–128.
4. Heller T, Lessells RJ, Wallrauch CG, et al.. Community-based treatment for multidrug-resistant tuberculosis in rural KwaZulu-Natal, South Africa. Int J Tuberc Lung Dis. 2010;14:420–426.
5. Furin JJ, Mitnick CD, Shin SS, et al.. Occurrence of serious adverse effects in patients receiving community-based therapy for multidrug-resistant tuberculosis. Int J Tuberc Lung Dis. 2001;5:648–655.
6. Shin SS, Pasechnikov AD, Gelmanova IY, et al.. Adverse reactions among patients being treated for MDR-TB in Tomsk, Russia. Int J Tuberc Lung Dis. 2007;11:1314–1320.
7. Bloss E, Kuksa L, Holtz TH, et al.. Adverse events related to multidrug-resistant tuberculosis treatment, Latvia, 2000-2004. Int J Tuberc Lung Dis. 2010;14:275–281.
8. Torun T, Gungor G, Ozmen I, et al.. Side effects associated with the treatment of multidrug-resistant tuberculosis. Int J Tuberc Lung Dis. 2005;9:1373–1377.
9. Nathanson E, Gupta R, Huamani P, et al.. Adverse events in the treatment of multidrug-resistant tuberculosis: results from the DOTS-Plus initiative. Int J Tuberc Lung Dis. 2004;8:1382–1384.
10. Baghaei P, Tabarsi P, Dorriz D, et al.. Adverse effects of multidrug-resistant tuberculosis treatment with a standardized regimen: a report from Iran. Am J Ther. 2011;18:e29–e34.
11. Vega P, Sweetland A, Acha J, et al.. Psychiatric issues in the management of patients with multidrug-resistant tuberculosis. Int J Tuberc Lung Dis. 2004;8:749–759.
12. Zager EM, McNerney R. Multidrug-resistant tuberculosis. BMC Infect Dis. 2008;8:10.
13. Brust JC, Shah NS, Scott M, et al.. Integrated, home-based treatment for MDR-TB and HIV in rural South Africa: an alternate model of care. Int J Tuberc Lung Dis. 2012;16:998–1004.
15. Lewis WC, Calden G, Thurston JR, et al.. Psychiatric and neurological reactions to cycloserine in the treatment of tuberculosis. Dis Chest. 1957;32:172–182.
16. Dutta BS, Hassan G, Waseem Q, et al.. Ethionamide-induced hypothyroidism. Int J Tuberc Lung Dis. 2012;16:141.
17. Satti H, Mafukidze A, Jooste PL, et al.. High rate of hypothyroidism among patients treated for multidrug-resistant tuberculosis in Lesotho. Int J Tuberc Lung Dis. 2012;16:468–472.
18. Isaakidis P, Varghese B, Mansoor H, et al.. Adverse events among HIV/MDR-TB co-infected patients receiving antiretroviral and second line anti-TB treatment in Mumbai, India. PloS One. 2012;7:e40781.
19. Department of Health, South Africa. Multi-Drug Resistant Tuberculosis: A Policy Framework on Decentralized and Deinstitutionalized Management for South Africa. 2011 Available at: http://www.doh.gov.za/docs/policy/2011/policy_TB.pdf
. Accessed December 26, 2012.
20. Falzon D, Jaramillo E, Schunemann HJ, et al.. WHO guidelines for the programmatic management of drug-resistant tuberculosis: 2011 update. Eur Respir J. 2011;38:516–528.
21. Brust JC, Lygizos M, Chaiyachati K, et al.. Culture conversion among HIV co-infected multidrug-resistant tuberculosis patients in Tugela Ferry, South Africa. PLoS One. 2011;6:e15841.