Background: Highly active antiretroviral therapy (HAART) and drugs that are used to treat multidrug-resistant tuberculosis have potentially overlapping adverse effects. Few South African studies have documented adverse effects in the multidrug-resistant tuberculosis population. This study examined the adverse effects profile in a sample of the outpatient population at the King George V Hospital Multidrug-Resistant Tuberculosis Clinic in Durban, KwaZulu-Natal.

Method: The method was an anonymous, retrospective record review of 350 patients with multidrug-resistant tuberculosis, who were attending the King George V Hospital Multidrug-Resistant Tuberculosis Clinic (2010–2011). Adverse effect profiles in patients with multidrug-resistant tuberculosis only, and those who were co-infected with the human immunodeficiency virus (HIV) who were on and not on HAART, were documented and analysed.

Results: Adverse events were recorded for 80.6% of patients. These included hearing loss (28.7%); peripheral neuropathy (23.2%); diarrhoea, nausea and vomiting (20.5%); arthralgia (15.9%); rashes and dermatological effects (excluding Stevens-Johnson syndrome) (14%); abdominal pain and dyspepsia (10.3%); and psychoses and confusion (8.3%). In this study population, 72.6% of patients were HIV positive, and 85% were concomitantly on HAART and multidrug-resistant tuberculosis treatment. Adverse events were significantly more common in patients who were HIV positive than in patients who were HIV negative with regard to peripheral neuropathy (p-value < 0.001), psychosis and confusion (p-value = 0.04), hearing loss (p-value = 0.047), and thyroid disease (p-value < 0.001). The use of HAART in patients who were HIV positive and on multidrug-resistant tuberculosis treatment was not significantly associated with the overall incidence of adverse events (p-value = 0.432). However, the calculated likelihood ratios of several individual adverse events occurring in these patients was greater. Patients who were HIV negative experienced the least adverse events.

Conclusion: The high percentage of patients in the sample population (45%) who was found to be multidrug-resistant tuberculosis positive de novo or while on standard tuberculosis treatment suggests that drug sensitivity testing for all patients with tuberculosis should be considered. The findings of this study support the current national policy that all patients with tuberculosis should be tested for HIV, and that all patients who are HIV positive and with multidrug-resistant tuberculosis should be on HAART. Clinicians should be supported in their function of examining, managing and recording adverse events. Reporting adverse events to the Department of Health should be encouraged. The development of a standardised recording instrument may mitigate the under-reporting of adverse events. The adverse effects profile in this study population differs from that reported in other studies.