Metabolic Syndrome and Neurocognitive Deficits in HIV Infection

Beverly Yu, BS; Elizabeth Pasipanodya, PhD; Jessica L. Montoy; PhD; Raeanne C. Moore, PhD; Sara Gianella, MD; Allen McCutchan, MD; Ron Ellis, MD; Robert K. Heaton, PhD; Dilip V. Jeste, MD; David J. Moore, PhD; María J. Marquine, PhD


J Acquir Immune Defic Syndr. 2019;81(1):95-101. 

In This Article



Two hundred one adults (109 PLHIV and 92 HIV−) from the Multi-dimensional Successful Aging study at the University of California San Diego participated in the current study. PLHIV were recruited from a variety of sources that serve adults with HIV in the San Diego area, including community clinics and health care providers. HIV-uninfected participants were recruited from our existing participant pool (ie, participants who were enrolled in previous studies at our research center and who agreed to be contacted for future studies) via flyers posted throughout the community (ie, colleges, coffee shops, gyms) and via presentations by study staff at community organizations. Further details on this study have been published previously.[29,30] Briefly, the inclusion criteria were being between the ages of 35 and 65 years, English-speaking, and capable of providing informed consent. We attempted to recruit an equivalent number of persons in each of three 10-year age cohorts (35–45, 46–54, and 56–65). Exclusionary criteria were a history of non-HIV neurologic disorder (eg, head injury with loss of consciousness greater than 30 minutes or neurologic complications, seizure disorder, stroke with neurologic or neuropsychiatric consequences), current psychotic disorder, and a history of a severe learning disability [eg, Wide Range Achievement-Fourth Edition (WRAT-4)[31] score of <70]. Participants who had positive urine toxicology or Breathalyzer test on the day of testing were rescheduled. Additional inclusion criteria for the present analyses were having data available on neurocognition and MetS. Only data from baseline visits were included, which took place between May 2013 and January 2016. All participants successfully completed an assessment of capacity to consent[32] to participate in clinical research and subsequently provided written informed consent.

Materials and Procedures

Study procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1975, as revised in 2000.

Neuromedical Evaluation

HIV serostatus was determined using enzyme-linked immunoabsorbent assays and confirmatory Western blot analyses in all participants. Metabolic risk factors [ie, elevated triglycerides, reduced high-density lipoprotein (HDL-C), elevated waist circumference, diabetes mellitus, and elevated blood pressure] were determined via laboratory assessments (phlebotomy and anthropomorphic) and current medication use (via self-report and/or record review). These assessments were performed by a clinician or a trained staff under clinical supervision. Per NCEP ATP III criteria, measurements for any of these conditions at or above specific thresholds or current treatment for these conditions was considered as meeting criteria for the relevant metabolic risk factor.[8,33] These thresholds were:[8] (1) waist circumference > 102 cm in men and >88 cm in women, (2) triglycerides ≥ 150 mg/dL, (3) HLD-C < 40 mg/dL in men and <50 mg/dL in women, (4) blood pressure ≥ 130/85 mm Hg, and (5) fasting glucose ≥ 100 mg/dL. MetS was defined as the presence of 3 or more metabolic risk factors. BMI was also calculated using measurements of participants' height and weight.

Among PLHIV, a number of disease and treatment characteristics were collected, including self-reported estimated duration of HIV infection, use of ART and other medications, lowest measured (nadir) CD4 T-cell count, and historical AIDS diagnosis. Current CD4 cell count and HIV RNA plasma levels were measured at a Clinical Laboratory Improvement Amendments-certified medical center laboratory using reverse transcriptase–polymerase chain reaction with a lower limit of quantitation at 50 viral copies per milliliter. HIV viral load was considered undetectable if values were below the lower limit of quantitation. Following established criteria, the Veteran's Aging Cohort Study Index (a composite measure of disease status and risk for all-cause mortality) was also calculated.[34,35]

Neurocognitive Evaluation

Participants completed a comprehensive neurocognitive test battery, which assessed neurocognitive domains in 7 ability areas: verbal fluency, abstraction/executive functioning, speed of information processing, verbal and visual learning, delayed recall, attention/working memory, and complex motor skills.[36] This battery was designed in agreement with the international consensus conference recommendations.[37] Following previously established and published procedures, domain deficit scores (DDSs) in each of the 7 domains, ranging from 0 (no impairment) to 5 (severe impairment), were computed from T-scores that were demographically corrected for the effects of age, education, sex, and race.[38,39] These DDSs were then averaged to obtain a summary measure of global NCI [Global Deficit Score (GDS)]. Participants also completed the WRAT-4[31] Reading Test, an estimate of the premorbid cognitive functioning/quality of education.

Psychiatric Evaluation

The presence/absence of current and lifetime major depressive disorder and substance use disorders (ie, of alcohol, tobacco, methamphetamine, cocaine, and 7 other illicit substances) was evaluated using the Composite International Diagnostic Interview[40,41] (CIDI v.2.1), which follows the diagnostic criteria based on the Diagnostic and Statistical Manual of Mental Disorders—Version IV.[42]

Statistical Analyses

HIV serostatus group differences on demographic factors, neurocognitive deficits, MetS (and its components), and psychiatric comorbidities were assessed via independent sample t tests (or nonparametric equivalent) and χ 2 tests. To examine whether HIV status modified the association between MetS and neurocognitive deficits, we ran a multivariable linear regression model on GDS with terms for HIV, MetS, their interaction, and significant covariates. Covariates considered included estimated premorbid neurocognitive function (WRAT-4 Reading Scaled Score [SS]) and psychiatric characteristics. Demographic characteristics were not considered as covariates given that GDS scores adjust for age, sex, education, and race. Covariates that differed by HIV status and were associated with GDS (at P < 0.10), based on independent sample t tests (or nonparametric equivalent) and χ 2 tests, were considered candidate covariates and included in the multivariable model.

Next, we investigated whether MetS was associated with GDS among PLHIV after considering the impact of significant covariates. To do so, we first examined the univariable association of WRAT-4 Reading SS and psychiatric and HIV disease characteristics to GDS among PLHIV via a series of Pearson product moment correlation coefficients and independent sample t tests. Variables associated with GDS at P > 0.10 were entered, along with MetS, as predictors in a multivariable model on GDS within the PLHIV group. To investigate the association between MetS and specific neurocognitive domains, we first ran separate independent sample t tests on each of the domains by MetS group (yes/no) among PLHIV. For domains significantly associated with MetS, we conducted separate multivariable linear regression models adjusting for significant covariates identified in the analyses described earlier. We followed a similar approach in our investigation of the association between specific components of the MetS and GDS.