Sex differences in HIV outcomes in the highly active antiretroviral therapy era: a systematic review

Abstract

To assess sex disparities in AIDS clinical and laboratory outcomes in the highly active antiretroviral therapy (HAART) era we conducted a systematic review of the published literature on mortality, disease progression, and laboratory outcomes among persons living with HIV and starting HAART. We performed systematic PubMed and targeted bibliographic searches of observational studies published between January, 1998, and November, 2013, that included persons starting HAART and reported analyses of mortality, progression to AIDS, or virologic or immunologic treatment outcomes by sex. Risk ratios (relative risks, odd ratios, and hazard ratios) and 95% confidence intervals were obtained. Sixty-five articles were included in this review. Thirty-nine studies were from North America and Europe and 26 were from Latin America, Asia, and Africa. Forty-four studies (68%) showed no statistically significant difference in risk of mortality, progression to AIDS, or virologic or immunologic treatment outcomes by sex. Decreased risk of death among females compared to males was observed in 24 of the 25 articles that included mortality analyses [pooled risk ratio 0.72 (95% confidence interval=0.69-0.75)], and decreased risk of death or AIDS was observed in 9 of the 13 articles that examined the composite outcome [pooled risk ratio=0.91 (0.84-0.98)]. There was no significant effect of sex on the risk of progression to AIDS [pooled risk ratio=1.15 (0.99-1.31)]. In this systematic review, females starting HAART appeared to have improved survival compared to males. However, this benefit was not associated with decreased progression to either AIDS or to differences in virologic or immunologic treatment outcomes.

FIG. 1.

Search results. PubMed Search Query: (HIV[mh] OR Acquired Immunodeficiency Syndrome[mh]) AND (Prognosis[mh] OR Mortality[mh] OR Disease Progression[mh]) AND (gender[tw] OR sex[mh] OR sex[tw] OR (male[mh] AND female[mh])) AND English[la] AND adult[mh] AND human[mh] AND 1998:2013[dp].

FIG. 2.

(a) Risk of death. (a) “Early” mortality following HAART initiation, as defined by the study. (b) “Late” mortality following HAART initiation, as defined by the study. (c) Mortality among patients with CD4⁺ lymphocyte count 351–500 cells/μl at HAART initiation. (d) Mortality among patients with CD4⁺ lymphocyte count >500 cells/μl at HAART initiation. n, number of individuals in the study cohort; Hgb, hemoglobin (as a marker for anemia), indicated as whether or not included in multivariable analyses; uHR, univariable hazard ratio; aHR, adjusted hazard ratio from multivariable analyses; ES, effect size of risk ratio, reported as comparing females to males; 95% CI, 95% confidence interval; Favors Females, a risk ratio less than one indicates a decreased risk of death for females compared to males; Favors Males, a risk ratio greater than one indicates an increased risk of death for females compared to males. (b) Risk of AIDS. n, number of individuals in the study cohort; Hgb, hemoglobin (as a marker for anemia), indicated as whether or not included in multivariable analyses; aIRR, adjusted incidence rate ratio from multivariable analyses; aHR, adjusted hazard ratio from multivariable analyses; aOR, adjusted odds ratio from multivariable analyses; ES, effect size of risk ratio, reported as comparing females to males; 95% CI, 95% confidence interval; Favors Females, a risk ratio less than one indicates a decreased risk of AIDS for females compared to males; Favors Males, a risk ratio greater than one indicates an increased risk of AIDS for females compared to males. (c) Risk of AIDS or death. n, number of individuals in the study cohort; Hgb, hemoglobin (as a marker for anemia), indicated as whether or not included in multivariable analyses; TLC, total lymphocyte count; uHR, univariable hazard ratio; aHR, adjusted hazard ratio from multivariable analyses; ES, effect size of risk ratio, reported as comparing females to males; 95% CI, 95% confidence interval; Favors Females, a risk ratio less than one indicates a decreased risk of AIDS or death for females compared to males; Favors Males, a risk ratio greater than one indicates an increased risk of AIDS or death for females compared to males.

FIG. 2.

(a) Risk of death. (a) “Early” mortality following HAART initiation, as defined by the study. (b) “Late” mortality following HAART initiation, as defined by the study. (c) Mortality among patients with CD4⁺ lymphocyte count 351–500 cells/μl at HAART initiation. (d) Mortality among patients with CD4⁺ lymphocyte count >500 cells/μl at HAART initiation. n, number of individuals in the study cohort; Hgb, hemoglobin (as a marker for anemia), indicated as whether or not included in multivariable analyses; uHR, univariable hazard ratio; aHR, adjusted hazard ratio from multivariable analyses; ES, effect size of risk ratio, reported as comparing females to males; 95% CI, 95% confidence interval; Favors Females, a risk ratio less than one indicates a decreased risk of death for females compared to males; Favors Males, a risk ratio greater than one indicates an increased risk of death for females compared to males. (b) Risk of AIDS. n, number of individuals in the study cohort; Hgb, hemoglobin (as a marker for anemia), indicated as whether or not included in multivariable analyses; aIRR, adjusted incidence rate ratio from multivariable analyses; aHR, adjusted hazard ratio from multivariable analyses; aOR, adjusted odds ratio from multivariable analyses; ES, effect size of risk ratio, reported as comparing females to males; 95% CI, 95% confidence interval; Favors Females, a risk ratio less than one indicates a decreased risk of AIDS for females compared to males; Favors Males, a risk ratio greater than one indicates an increased risk of AIDS for females compared to males. (c) Risk of AIDS or death. n, number of individuals in the study cohort; Hgb, hemoglobin (as a marker for anemia), indicated as whether or not included in multivariable analyses; TLC, total lymphocyte count; uHR, univariable hazard ratio; aHR, adjusted hazard ratio from multivariable analyses; ES, effect size of risk ratio, reported as comparing females to males; 95% CI, 95% confidence interval; Favors Females, a risk ratio less than one indicates a decreased risk of AIDS or death for females compared to males; Favors Males, a risk ratio greater than one indicates an increased risk of AIDS or death for females compared to males.

FIG. 3.

(a) Risk of virologic suppression. n, number of individuals in the study cohort; uHR, univariable odds ratio; uOR, univariable odds ratio; aHR, adjusted hazard ratio from multivariable analyses; aOR, adjusted odds ratio from multivariable analyses; ES, effect size of risk ratio, reported as comparing females to males; 95% CI, 95% confidence interval; Favors Males, a risk ratio less than one indicates a decreased likelihood of virologic suppression for females compared to males; Favors Females, a risk ratio greater than one indicates an increased likelihood of virologic suppression for females compared to males; (b) Risk of virologic failure. n, number of individuals in the study cohort; uHR, univariable odds ratio; uOR, univariable odds ratio; aHR, adjusted hazard ratio from multivariable analyses; aOR, adjusted odds ratio from multivariable analyses; ES, effect size of risk ratio, reported as comparing females to males; 95% CI, 95% confidence interval; Favors Females, a risk ratio less than one indicates a decreased risk of virologic failure for females compared to males; Favors Males, a risk ratio greater than one indicates an increased risk of virologic failure for females compared to males.

FIG. 4.

(a) Risk of immunologic response. (a) Odds of achieving CD4⁺ lymphocyte count >200 cells/μl after 12 months for patients with CD4⁺ lymphocyte count <200 cells/μl at HAART initiation. (b) Odds of achieving CD4⁺ lymphocyte count >350 cells/μl after 12 months for patients with CD4⁺ lymphocyte count <200 cells/μl at HAART initiation. n, number of individuals in the study cohort; aHR, adjusted hazard ratio from multivariable analyses; aOR, adjusted odds ratio from multivariable analyses; aRR, adjusted relative risk from multivariable analyses; ES, effect size of risk ratio, reported as comparing females to males; 95% CI, 95% confidence interval; Favors Males, a risk ratio less than one indicates a decreased likelihood of immunologic response for females compared to males; Favors Females, a risk ratio greater than one indicates an increased likelihood of immunologic response for females compared to males. (b) Risk of immunologic failure. (a) Odds of CD4⁺ lymphocyte count <200 cells/μl after 12 months of HAART. (b) Odds of CD4⁺ lymphocyte count <200 cells/μl after 24 months of HAART. n, number of individuals in the study cohort; uHR, univariable hazard ratio; aHR, adjusted hazard ratio from multivariable analyses; aOR, adjusted odds ratio from multivariable analyses; ES, effect size of risk ratio, reported as comparing females to males; 95% CI, 95% confidence interval; Favors Females, a risk ratio less than one indicates a decreased risk of immunologic failure for females compared to males; Favors Males, a risk ratio greater than one indicates an increased risk of immunologic failure for females compared to males.