 |
|
|
| |
No Adverse Pregnancy Outcome Patterns With Common Antiretrovirals
|
| |
| |
International Workshop on Clinical Pharmacology of HIV, Hepatitis, and Other Antiviral Drugs Virtual Meeting, September 28-30, 2020
By Mark Mascolini for NATAP and Virology Education
Systematic review of 924 pregnancies during 2004-2019 in France discerned no pattern of adverse pregnancy outcomes after mothers took common antiretroviral combinations [1]. One possible exception was low birth weight after exposure to nucleoside-only combinations, but that analysis involved only 16 pregnancies.
Researchers working in Toulouse noted that a million or more pregnant women with HIV take antiretrovirals every year and so risk adverse pregnancy outcomes or adverse outcomes in infants exposed to antiretrovirals during pregnancy. In 2018, for example, the World Health Organization warned of neural tube defects linked to use of the integrase inhibitor dolutegravir [2].
To get a countrywide sense of potential adverse effects of current antiretrovirals during pregnancy, this Toulouse team analyzed data submitted to the ANRS pharmacovigilance system from 2004 to 2019. This systematic review included all HIV cohorts and trials in which pregnant women or women who might become pregnant took antiretrovirals. All women were older than 18.
As adverse pregnancy outcomes, the researchers considered abortion, ectopic pregnancy, stillbirth, prematurity (birth before 37 weeks gestation), low birth weight (below 2500 grams), and congenital abnormalities. They defined a favorable pregnancy outcome as a live birth without congenital abnormalities, prematurity, or low birth weight. Logistic regression (adjusted for maternal age, study location, design, and type, antiretroviral initiation period, and antiretroviral switching during pregnancy) estimated the odds ratio for each outcome with each antiretroviral therapy group compared with nonnucleoside (NNRTI)-based therapy.
The analysis included 924 pregnancies reported in 34 studies. Mothers had a median age of 31 years. The most used combinations were based on protease inhibitors (PIs) (49%), NNRTIs (42%), integrase inhibitors (5%), and nucleosides/nucleotides (2%). More than half of women, 57%, had started antiretrovirals before conception, 9% started in the first trimester, 21% in the second or third trimester, and 13% did not take antiretrovirals.
The researchers counted 624 live births for a live birth rate of 77.2%. Among all pregnancies, 75.8% were live births with no adverse pregnancy outcomes, 14.4% had low birth weight, 13.1% were premature, and 1.4% had congenital abnormalities. Among 184 nonlive births (23.4%), 9.9% were spontaneous abortions, 8.7% induced abortions, 3.2% stillbirths, and 1.1% ectopic pregnancies.
Logistic regression saw no differences in pregnancy outcomes between NNRTI regimens (the comparative reference) and nucleoside-only regimens, PI-based regimens, integrase inhibitor-based regimens, or other regimens for spontaneous abortions or prematurity. Odds of low birth weight were 7.5-fold higher (95% confidence interval 1.49 to 37.83) with nucleoside-only regimens than with NNRTI combinations. But no other regimens differed from NNRTI regimens in low birth weight. The authors suggest interpreting the nucleoside finding with caution because of the small size of that group (16 women) and the potential for indication bias.
Overall, the ANRS team believes their analysis "provides reassuring pharmacovigilance data."
References
1. Saint-Lary L, Benevent J, Damase-Michel C, Vayssiere C, Sommet A, Leroy V. Systematic review and meta-analysis of adverse perinatal outcomes associated with prenatal exposure to protease inhibitor combination in pregnant women with HIV infection. International Workshop on Clinical Pharmacology of HIV, Hepatitis, and Other Antiviral Drugs Virtual Meeting, September 28-30, 2020. Abstract 30.
2. Chouchana L, Pariente A, Pannier E, Tsatsaris V, Treluyer JM. Dolutegravir and neural tube defects: a new insight. Lancet Infect Dis. 2020;20:405-406. https://www.thelancet.com/journals/laninf/article/PIIS1473-3099(20)30117-1/fulltext
|
| |
|
|
|
|
|
