Incidence of and risk factors for HIV-associated distal sensory polyneuropathy.

Abstract

To assess the incidence of and risk factors for distal sensory polyneuropathy (DSP) in a cohort of HIV-infected subjects. We followed 272 subjects semiannually for up to 30 months. DSP was diagnosed if subjects had decreased or absent ankle jerks, decreased or absent vibratory perception at the toes, or decreased pinprick or temperature in a stocking distribution. Subjects were further classified at each visit as having asymptomatic DSP (ADSP) (signs only) or symptomatic DSP (SDSP) if, in addition to the neurologic signs, paresthesias or pain was reported. At baseline, 45% of the subjects did not meet criteria for DSP, 20% met criteria for ADSP, and 35% met criteria for SDSP. Dideoxynucleoside therapy was used by 23% of the patients, and this treatment was independent of their neuropathy status. In longitudinal univariate analyses, history of AIDS diagnoses (hazard ratio [HR] = 1.89; p = 0.02) and lower CD4 cell count (HR = 0.69; p = 0.0006) were risk factors for incident DSP (ADSP or SDSP). However, for incident SDSP only, in addition to history of AIDS diagnoses, mood and neurologic (other than DSP) and functional abnormalities were significant risk factors. Functional abnormalities remained a significant risk factor in a multiple regression analysis. The presence of ADSP and the use of dideoxynucleosides at baseline were not significant risk factors for incident SDSP. The Kaplan-Meier estimate of the 1-year incidence of SDSP was 36%. Subjects with moderate-to-severe immunosuppression from HIV infection commonly have SDSP. However, sex, use of dideoxynucleosides, and presence of ADSP were not significant risk factors for SDSP.