Background:

Fatty acids, including n-6 series, modulate immune function, but their effect on CD4 cell counts, death, or hospitalization in HIV-infected patients on antiretroviral therapy is unknown.

Methods:

In a randomized trial for effects of multivitamins in HIV-infected patients in Uganda, we used gas chromatography to measure plasma n-6 fatty acids at baseline; determined CD4 counts at baseline, 3, 6, 12, and 18 months; and recorded hospitalization or death events. The associations of fatty acids with CD4 counts and events were analyzed using repeated-measures analysis of variance and Cox regression, respectively.

Results:

Among 297 patients with fatty acids measurements, 16 patients died and 69 were hospitalized within 18 months. Except for linoleic acid, n-6 fatty acids levels were positively associated with CD4 counts at baseline but not during follow-up. In models that included all 5 major n-6 fatty acids, age; sex; body mass index; anemia status; use of antiretroviral therapy, multivitamin supplements, and alcohol; and the risk of death or hospitalization decreased significantly with an increase in linoleic acid and gamma-linolenic acid levels, whereas associations for dihomo-gamma-linolenic acid, arachidonic acid, and aolrenic acid were null. The hazard ratios (95% confidence intervals) per 1 SD increase in linoleic acid and gamma-linolenic acid were 0.73 (0.56–0.94) and 0.51 (0.36–0.72), respectively. Gamma-linolenic acid remained significant (hazard ratio= 0.51; 95% confidence interval: 0.35 to 0.68) after further adjustment for other plasma fatty acids.

Conclusions:

Lower levels of gamma-linolenic acid are associated with lower CD4 counts and an increased risk of death or hospitalization. These results suggest a potential for using n-6 fatty acids to improve outcomes from antiretroviral therapy.