Increased incidence of peripheral neuropathy with co-administration of stavudine and isoniazid in HIV-infected individuals.

Abstract

The treatment of HIV-infected individuals often requires the co-administration of many drugs. Potential drug interactions should be carefully considered in order to minimize adverse reactions. Distal sensory neuropathy (DSN) is the commonest neurological complication found in HIV-infected individuals [1]. Estimates of its incidence vary, but it has been documented in up to a third of AIDS patients. It presents with paraesthesia, numbness and tingling in a symmetrical pattern in the distal extremities, and signs of diminished sensation to pinprick and vibration, loss of ankle jerks, but preservation of joint position sense and motor function. Infectious, metabolic, inflammatory, nutritional and drug-related aetiologies have all been implicated. DSN can be very distressing and if drug-related may lead to a switch of therapy. Isoniazid (INH) used in the treatment and prophylaxis of tuberculosis can produce DSN, although this is rare with the concurrent use of pyridoxine. The antiretroviral agent stavudine (d4T) is a nucleoside analogue, which produces DSN in up to a fifth of patients at standard doses [2]. We sought to determine the incidence of DSN in patients receiving both INH and d4T for HIV-related tuberculosis. A retrospective case-note review was performed for all patients using INH and antiretroviral agents since 1996, when d4T became widely available in the UK. DSN was diagnosed clinically (symptoms and signs as above). Resolution was deemed present after the sustained cessation of both symptoms and signs. A total of 30 individuals were identified. Two had pre-existing DSN and were excluded from the analysis. Twenty-two took d4T in combination with other drugs. Six used only non-d4T containing regimes. All started isoniazid before d4T. Twelve out of 22 patients (55%) developed DSN with a median time to onset of 5 months (range 2–15 months). Their ages ranged from 28 to 64 years (median 37) and CD4 cell counts on starting d4T from 24 to 234 cells/mm3 (median 129). Neither of these variables was significantly different in the 10 patients without DSN (two-tailed Student's t-test). None of the 22 had other identified risk factors for DSN. In nine of the 12, DSN resolved on changing antiretroviral treatments. Eight switched to zidovudine and one to ritonavir/saquinavir. Three remain on d4T with on-going symptoms. Of the six who received INH but no d4T, one developed DSN caused by zalcitabine (nucleoside analogue) and this resolved on switching therapy. In the subjects in this study the co-administration of INH with d4T led to a greatly increased incidence of DSN when compared with the use of d4T alone (11% in our centre;P < 0.001 using the chi-squared test). This effect was independent of other risk factors including age and CD4 cell count. The central role of d4T in the development of DSN is supported by the observed resolution on stopping this drug. We suggest that when treating HIV and tuberculosis co-infection with isoniazid, the concomitant use of d4T-containing regimes should be avoided if possible. This may also have implications for tuberculosis prophylaxis in this population [3]. Ronan A. M. Breen Marc C. I. Lipman Margaret A. Johnson