icon-    folder.gif   Conference Reports for NATAP  
  XIX International AIDS Conference
July 22-27, 2012
Washington, DC
Back grey_arrow_rt.gif
Mitochondrial Upsets May Underlie Metabolic Disorders in HIV-Positive Adolescents
  XIX International AIDS Conference, July 22-27, 2012, Washington, DC

Mark Mascolini

A US study comparing HIV-infected youngsters with HIV-exposed but uninfected (HEU) youth yielded data suggesting that mitochondrial dysfunction lies behind metabolic abnormalities in HIV-positive adolescents [1]. This Pediatric HIV/AIDS Cohort Study (PHACS) linked every 1 mg/dL higher venous lactate to an 18-mg/dL higher triglyceride reading.

Metabolic abnormalities often affect HIV-positive people, including perinatally infected children and adolescents taking antiretrovirals for a decade or more. These metabolic derangements may be caused by mitochondrial dysfunction fostered by antiretroviral therapy or chronic HIV infection, PHACS researchers suggested. To test those hypotheses, they gauged mitochondrial function as oxidative phosphorylation (OSPHOS) enzyme activities and lactate levels, then compared those measures in HIV-infected and HEU adolescents. In the HIV-positive group, the researchers also determined associations between mitochondrial function and fasting glucose, insulin, and HOMA-defined insulin resistance.

Researchers recruited youngsters without known non-HIV mitochondrial disorders from the PHACS Adolescent Master Protocol. The investigators gathered demographic and body mass index data in both HIV-positive and HEU youth, and they recorded CD4 counts, viral loads, antiretroviral exposure, and fasting insulin and glucose in HIV-positive youngsters. Principal outcomes were venous and point-of-care (fingerprick) lactate, venous pyruvate, and PBMC NADH dehydrogenase (CI) and cytochrome c oxidase (CIV) enzyme activities.

The PHACS team enrolled 191 HIV-positive adolescents and 117 HEU youngsters. The HIV-positive adolescents were older than HEU youth (average 15.8 versus 12.7 years, P < 0.001), and a higher proportion of HIV-positive youngsters were non-Hispanic blacks (70% versus 56%, P = 0.05). But the gender distribution was similar in the two groups (54% and 51% boys). Body mass index Z scores were significantly lower in adolescents with HIV (average 0.46 versus 0.91, P = 0.005).

In the HIV group, 50% had CDC stage B or C HIV infection, 58% were taking a protease inhibitor (PI)-based antiretroviral combination, only 11% had a viral load below 400 copies, and median CD4 count stood at 624. HIV-positive youngsters had a median fasting glucose of 86 mg/dL (interquartile range [IQR] 81 to 91), median fasting insulin of 12.1 uu/mL (IQR 8.0 to 20.9), and HOMA insulin resistance of 2.5 (IQR 1.7 to 4.5). Median total cholesterol stood at 159 mg/dL (IQR 137 to 186) and median triglycerides at 86.5 mg/dL (IQR 63 to 116).

Median point-of-care lactate levels (measured by fingerprick) were marginally higher in adolescents with HIV (1.45 mg/dL, IQR 1.0 to 1.9) than in HEU youth (1.4 mmol/L, IQR 1.1 to 1.9), and that difference was not significant (P = 0.98). But median venous lactates were significantly lower in the HIV group (1.0 mg/dL, IQR 0.79 to 1.40) than in HEU youngsters (1.26 mg/dL, IQR 0.89 to 1.70) (P < 0.001).

Median venous pyruvate was also significantly lower in the HIV group (0.09 mg/dL, IQR 0.05 to 0.11) than in the HEU group (0.10 mmol/L, IQR 0.07 to 0.13) (P = 0.005). Pyruvate may be metabolized to lactate or to acetyl CoA.

Median CI OXPHOS enzyme activity was similar in HIV-positive and HEU adolescents (37.9 and 36.9 OD/min/ug e-6, P = 0.71), but median CIV OXPHOS enzyme activity was higher in the HIV group (69.4 versus 60.8 OD/min/ug e-6, P = 0.048).

In children with HIV, insulin resistance was associated with higher venous lactate (P = 0.046) and pyruvate (P = 0.028), while high triglycerides were associated with higher point-of-care lactate (P = 0.024) and venous lactate (P < 0.001). Venous lactate correlated positively with total cholesterol (r = 0.16, P = 0.04) and with triglycerides (r = 0.37, P < 0.0001) in HIV-positive children. Low "good" high-density lipoprotein (HDL) cholesterol was associated with lower PBMC OXPHOS CI enzyme activity (P = 0.024) and lower OXPHOS CIV enzyme activity (P = 0.085).

Multivariate analysis identified associations between longer PI duration and higher triglycerides (+2.63 mg/dL per year of PIs, P = 0.03) and longer nonnucleoside duration and higher triglycerides (+4.19 mg/dL per year of nonnucleosides, P = 0.004). Higher venous lactate was also associated with higher triglycerides (+17.7 mg/dL per 1 mg/dL lactate, P = 0.0008).

The PHACS investigators concluded that (1) insulin resistance is associated with higher lactates and pyruvate in HIV-positive children, (2) high triglycerides are associated with higher lactates, (3) low HDL cholesterol is associated with lower OXPHOS CI and CIV enzyme activities, and (4) venous lactate is independently associated with higher triglycerides. They proposed the overall conclusion that "mitochondrial dysfunction induced by either HIV or antiretrovirals may be responsible for the observed metabolic changes" in HIV-positive youngsters.

Previous studies yielded additional findings on mitochondrial function in HIV-positive, HIV-exposed, and HIV-negative youth. A Spanish cross-sectional comparison of 47 asymptomatic antiretroviral-treated youngsters and 27 healthy HIV-negative controls found significantly lower mitochondrial DNA (mtDNA) in PBMCs from the HIV group, but similar levels of mitochondrial RNA in the two groups [2]. CIV protein subunit content and enzymatic activity were also similar in the two groups.

A study of 2931 children in Pediatric AIDS Clinical Trials Group protocols 219 and 219C used two definitions of mitochondrial dysfunction, the Enquete Perinatale Francaise criteria and the Mitochondrial Disease Classification criteria [3]. One third of children (33.5%) met one or both criteria, and mortality was highest in the 96 children who met both criteria. This study did not have an HIV-negative comparison group.

A comparison of HIV-uninfected infants born to HIV-positive or HIV-negative mothers found lower geometric mean PBMC mtDNA levels at birth in the HIV-exposed group [4]. In HIV-exposed infants, mtDNA levels were lowest in those not exposed to antiretrovirals, higher in those exposed to zidovudine alone, and highest in those exposed to combination nucleoside analogs.


1. Miller TL, Wang J, Jacobson DL, et al. Mitochondrial function and metabolic abnormalities in children with perinatally-acquired HIV infection in the Pediatric HIV/AIDS Cohort Study (PHACS). XIX International AIDS Conference. July 22-27, 2012. Abstract MOAB0203. Slides online at http://pag.aids2012.org/session.aspx?s=239#3.

2. Moren C, Noguera-Julian A, Rovira N, et al. Mitochondrial assessment in asymptomatic HIV-infected paediatric patients on HAART. Antivir Ther. 2011;16:719-724.

3. Crain MJ, Chernoff MC, Oleske JM, et al. Possible mitochondrial dysfunction and its association with antiretroviral therapy use in children perinatally infected with HIV. J Infect Dis. 2010;202:291-301.

4. Aldrovandi GM, Chu C, Shearer WT, et al. Antiretroviral exposure and lymphocyte mtDNA content among uninfected infants of HIV-1-infected women. Pediatrics. 2009;124:e1189-e1197.