A retrospective comparative study of adverse drug reactions among HIV (+) and HIV (-) adult patients taking antitubercular drugs, in 2006-2007
Tuberculosis (TB) is an infectious bacterial infection caused by Mycobacterium tuberculosis. It is estimated that a third of the world population is infected, with eight million people progressing to active TB disease each year, two million of whom die of the disease. About one third of people infected with HIV are also infected with TB and 70% of these people live in sub-Saharan Africa. In Kenya, the data for 2006 indicated that the national average HlV prevalence in TB patients was 52%. Although standard TB treatment consisting of isoniazid (H), rifampicin (R), pyrazinamide (Z) and ethambutol (E) is effective in treating active TB, it has been associated with many adverse drug reactions (ADRs) more so in HIV (+) patients and poses a significant challenge to completion of treatment, therefore; they need to be carefully monitored. Main objective: To determine the prevalence of ADRs among HIV (+) and HIV(-) adult patients taking anti-TB drugs at Kenyatta National Hospital (KNH). Study Design: A retrospective cohort study carried out at the medical records department of KNH. Three hundred and fourteen patient files meeting inclusion criteria during the study period January 2006 to December 2007 were selected randomly for inclusion in the study, of which 157 were for adult HIV(+) patients taking anti-TB drugs while 157 were for their HIV(-) counterparts. Data extracted from patient files were analyzed using the Statistical Package for social sciences (SPSS) version 13.0. Results: A total of 83 ADRs were recorded among HIV (+) patients compared to 28 ADRs among HIV(-) patients. ADRs were more common in HIV (+) patients 70 (44.6%) compared to HIV (-) patients 26 (16.6%), (OR = 2.692 [-1.819- 3.985], p<O.OOI). More HIV(+) patients than HIV (-) patients taking anti- TB drugs had more than one ADR; 11 (7%) versus 2 (1.3%), (OR = 5.50 [1.239 - 24.412], p=0.02). Seventy two (45.86%) HIV (+) patients were also taking ARVs while on anti-TB drugs. Overall, the most frequent ADR was gastrointestinal disturbance which was recorded in 34 (21.7%) HIV (+) patients compared to 16 (10.2%) HIV(-) patients, p=0.006. Peripheral neuropathy was recorded in 26 (16.6%) HIV (+) patients compared to 7 (4.5%) HIV (-) patients, p=0.0005. Cutaneous reactions were more common in HIV(+) patients 10 (6.4%) than HIV (-) patients 2 (1.3%), p=0.02. Nine (5.7%) HIV (+) patients compared to 2 (1.3%) HIV (-) patients had hepatotoxicity, p=0.03. There were no statistically significant differences between HIV (+) patients and HIV (-) patients with regard to the prevalence of ocular toxicity (1.9% vs 0.6%, p=O.3)and ototoxicity (0.6% vs 0%,p=0.3). Interruption of anti-TB treatment occurred in 8 (5.1%) HIV (+) and 4 (2.55%) HIV(-) patients. Although, gastrointestinal disturbances and peripheral neuropathy were the most commonly recorded ADRs in both the HIV (+) and HIV (-) patients, hepatotoxicity was the main cause of TB treatment interruption; 4.46% inHIV (+) and in 1.27%HIV (-) patients. Conclusion: ADRs were more likely to occur in adult HIV (+) patients taking anti-TB drugs and on ARVs than their HIV (-)counterparts taking anti-TB drugs alone. Gastrointestinal disturbances and peripheral neuropathy were the most common ADRs in both HIV(+) and HIV(-) adult patients taking anti-TB drugs. However, hepatotoxicity was the main cause of TB treatment interruption in both the HIV(+) and HIV(-) adult taking anti-TB drugs. Recommendations: ADRs surveillance systems should be established in hospitals as they will have a major impact on ADRs monitoring and control. Clinical monitoring should be done regularly as hepatotoxicity is the main cause of TB treatment interruption. More studies on the incidence and risk factors for ADRs in Africans are required as they will help to prevent serious adverse effects during TB and combined HIV/TB treatment. Any future studies should preferably be prospective to overcome the numerous limitations of retrospective studies. Grading of ADRs will be necessary in future studies to identify serious ADRs.