Antiretroviral Therapy for Mitochondrial Toxicity in HIV-Infected Pregnant Women

Background: The use of antiretroviral therapy (ART) in HIV-infected women is crucial to restore and maintain the immune system and prevent HIV transmission during pregnancy, labor, delivery and breastfeeding. Furthermore, ART reduce the risk of Mother-to-child transmission (MTCT). Therefore, ART has been associated with mitochondrial defections that could induce preeclampsia, preterm birth, low birthweight, intrauterine growth restriction (IUGR), stillbirth and sudden infant death. Objective: To evaluate the effect of antiretroviral therapy on mitochondrial defections in HIV-infected pregnant women. Methods: We searched eligible studies in MEDLINE, Scorpus and WHO Global Index Medicus. Included studies were assessing the effects of antiretroviral therapy on genetic mitochondrial diseases in HIV infected pregnant women and HIV exposed infants. JTL searched eligible studies in different databases and both JLT and JLT critically appraised included studies. Results: We found five observational studies with low risk of bias. All studies illustrated that ART increased the mean of mitochondrial defections. The results were statistically significant in all studies with P<0.05. Conclusion: Mitochondrial lesions were very common in HIV infected pregnant women and HIV exposed infants. However, further investigations are needful to strengthen this evidence. In fact, mitochondrial defections have been associated with increased rates of preterm delivery, stillbirth, IUGR, and sudden infant death [5-8]. Reviewing the literature, ART can induce subclinical transplacental mitochondrial lesions. Therefore, it is unknown whether HIV in pregnancy could injure mitochondria. Mitochondrial defections in a given organism could lead to the development of mitochondriopathies [9]. Mitochondriopathies are classified as inherited or acquired (derived from toxic substances), with both sharing similar clinical consequences [10]. This scientific paper is focused on acquired mitochondriopathies due to ART affecting both mitochondrial DNA and proteins. The mutations responsible for genetic mitochondrial diseases can be present in both the nuclear or mitochondrial genome [10]. Genetic mitochondrial diseases are already present in 1 in 5000 newborns and 1 in 200 women may carry one of these deleterious mutations [11]. Studies have shown that ART highly increase mitochondrial mutations.

In fact, mitochondrial defections have been associated with increased rates of preterm delivery, stillbirth, IUGR, and sudden infant death [5][6][7][8]. Reviewing the literature, ART can induce subclinical transplacental mitochondrial lesions. Therefore, it is unknown whether HIV in pregnancy could injure mitochondria. Mitochondrial defections in a given organism could lead to the development of mitochondriopathies [9].
Mitochondriopathies are classified as inherited or acquired (derived from toxic substances), with both sharing similar clinical consequences [10]. This scientific paper is focused on acquired mitochondriopathies due to ART affecting both mitochondrial DNA and proteins. The mutations responsible for genetic mitochondrial diseases can be present in both the nuclear or mitochondrial genome [10]. Genetic mitochondrial diseases are already present in 1 in 5000 newborns and 1 in 200 women may carry one of these deleterious mutations [11]. Studies have shown that ART highly increase mitochondrial mutations.

Objectives
To evaluate the effects of antiretroviral therapy on mitochondrial defections in HIV-infected pregnant women and HIV exposed infants.

Methods
We systematically searched in MEDLINE, Scorpus and WHO Global Index Medicus. We included observational studies assessing the effects of antiretroviral therapy on genetic mitochondrial diseases in HIV infected pregnant women and HIV exposed infants.
The criteria for considering included studies for this mini review were HIV pregnant women on ART and HIV exposed newborn. We assessed maternal mtDNA depletion and newborn mtprotein. We included observational studies comparing cases and controls. We considered all types of ART regimens in HIV pregnant women. No other restriction criteria were imposed in terms of inclusion or exclusion studies. JTL searched eligible studies in different databases. JLT and JLT worked independently to review full-text versions articles. The reviewers independently extracted data for study characteristics, patient characteristics, and outcomes (unadjusted and adjusted associations). Observational studies were assessed with Newcastle-Ottawa Scale. The following domains were used for bias assessment:

Results and Discussion
We screened a total number of 217 studies, and included 5 studies; four studies have been conducted in Spain and one in Canada. All the included studies were observational studies among which four crosssectional and one prospective cohort study. All of studies analyzed the occurrence of maternal mtDNA and mtprotein in infants.
We included 5 observational studies among which four cross sectional studies and one prospective cohort study [5,7,8,12,13] (Table  1). The results were reported narratively. All mitochondrial studies conducting in HIV-pregnancies have described an increased frequency of mtDNA depletions.
The first study [8] reported the mean (SE) depletion of 44.45 ± 3.77% with P=0.001 in HIV-pregnant women and the mean (SE) mt proteins were reduced in HIV exposed infants compared to the control group 48.79 ± 3.41% with P<0.001. The second study [9] found mt proteins were reduced in HIV-infected pregnant women, infants and foetus with respectively: 25.8%, 38.6% and 13.6% (P<0.05). The third study [10] reported the mean (SE) maternal mtDNA of 42.66 ± 5.94% and mt proteins of 12.82 ± 5.73%. The mtDNA mean (SD) in the fourth study [11] was reduced by 39.20% ± 2.78%. The last study [12] revealed that the mean (SE) mtDNA/nDNA ratio was reduced by -18.0 ± 6.1. All the results were statistically significant (P<0.05). All included studies were low risk of bias from 6 to 7 on Newcastle-Ottawa Scale [13].

Conclusion
Summary considerations on study quality in general, the studies did not show major problems of selection bias, the majority use the same population (HIV pregnant women and HIV exposed infants) compared to pregnant women HIV negative and HIV non-exposed infants respectively, limiting then problems of comparability. Multiple linear regressions were used in all included studies, and then adjustment for confounding factors was adequate. Ascertainment of outcome has not major biases in relation to the assessment of mtDNA and mtprotein because those outcomes are ascertained through laboratory record. Losses to follow-up would not possibly represent a problem. In most studies, losses to follow-up were not reported and the possible risk of bias is not predictable. All included studies included double-NRTI based regimens. Among them, three studies included AZT contained regimens [5,7,8]. As explained above, AZT based regimens may cause severe mitochondrial deflections implying highly statistically significant results in AZT based regimens studies.
A potential limitation of our search strategy is that only few databases were considered. It is likely to associated publication bias. However, this is a brief research has confirmed the results of previous studies conducted on the effects of ART on mitochondrial diseases. Besides, we did not find any study conducted in Sub-Saharan Africa where HIV prevalence is the highest in the world.
In fact, ARTs are indispensable in the treatment and prevention of HIV infection. Until now, the use of ART during pregnancy is considered safe; therefore, there are still many concerns about maternal and foetal outcomes that are not clearly highlighted in clinical practice. Besides, certain ART regimens have more mitochondrial toxicities than others. By the way, specific ART regimens should be used in pregnancy. In addition, the timing of prescribing ART in pregnancy could lower mitochondrial toxicity. Then, mitochondrial toxicity associated to ART inducing newborn and maternal morbidity need further studies, particularly in Sub-Saharan Africa. This could establish whether HIVpregnancy may be an additional risk for the onset of mitochondrial toxicity, because it is currently unknown.