Young women (15–24 years) are particularly vulnerable to sexually transmitted infections and HIV . In this particular population, pregnancy is commonly unplanned and undesired, and quite frequently occurs in women who had never been tested for HIV [1, Reference Koenig2]. Age, lower therapeutic adherence, and a combination of socio-economic factors pose young pregnant women with HIV at risk of adverse pregnancy outcomes, lower treatment coverage, less stringent suppression of HIV, and vertical transmission [Reference Lima3]. It is therefore of importance to better define the prevalence of undiagnosed HIV infection, unplanned pregnancy, adverse pregnancy outcomes and suboptimal HIV suppression in this fragile population. To address this issue, we used data from a large national cohort of pregnant women with HIV and explored temporal trends in the proportion of women with HIV with pregnancy at young age (<25 years), and assessed in this specific group the rates of planned pregnancy, HIV diagnosis in pregnancy, antiretroviral (ARV) treatment at conception and during pregnancy, HIV viral suppression at third trimester, HIV vertical transmission, and main pregnancy outcomes.
Data from the Italian National Program on Surveillance on Antiretroviral Treatment in Pregnancy, an ongoing observational study established in Italy in 2001, were used [Reference Floridia4]. Only HIV-positive pregnant women are eligible for the study, and treatment of HIV infection is decided by the treating physician, usually according to national guidelines. Women provide consent based on a patient information sheet approved by the competent Ethics Committee (deliberation 578, 28 September 2001, I.N.M.I. Lazzaro Spallanzani Ethics Committee, Rome). For the current analysis, all pregnancies with known maternal age and available date of HIV diagnosis were considered eligible. The study period (2001–2016) was divided in three intervals of 5 years each (2001–2005, 2006–2010, 2011–2016), and temporal trends were analyzed using the χ 2 test for trend. Other categorical variables were compared using the χ 2 test, with odds ratios (ORs) and 95% confidence intervals (CIs) calculated. In order to adjust for potential confounders, also defined in previous analyses [Reference Floridia4, Reference Floridia5], the role of young age as predictor of unplanned pregnancy and undiagnosed HIV infection was further evaluated in multivariable logistic regression analyses. Gender- and gestational age-adjusted Z-scores for birthweight were calculated according to recent national references [Reference Bertino6]. Major birth defects were defined according to the Antiretroviral Pregnancy Registry definition [Reference Scheuerle and Tilson7]. For all analyses, P values <0·05 were considered significant. All statistical analyses were performed with the SPSS software, version 22 (IBM Corp, Released 2013, Armonk, New York, USA).
As of 24 February 2017, 2979 pregnancies had available information and were included in the analysis. The median age at entry in pregnancy was 33 years (interquartile range 29–36), with 9·0% (269/2979) of the women <25 of age. Younger (<25 years) and older (⩾25 years) had identical mean CD4 count levels at entry (493/mm3 in younger women vs. 494/mm3 in older women, P = 0·966), but younger women were more frequently of foreign origin (68·2% vs. 45·6%, OR 2·56, 95% CI 1·94–3·37, P < 0·001). Only a small proportion of pregnancies were planned (925/2540, 36·4%), with a significantly lower probability of planned pregnancy among younger women (23·2% vs. 37·7%, OR 0·50, 95% CI 0·36–0·69, P < 0·001). The main temporal trends in pregnancy planning, diagnosis of HIV in pregnancy, and ARV treatment coverage in mothers and infants are summarized in Table 1. The proportion of young women increased significantly over time, from 7·0% in 2001–2005 to 12·2% in 2011–2016. The proportion of planned pregnancy also increased over time, but this increase was statistically significant only in the entire group and in the subgroup of older women. The rate of HIV diagnosis in current pregnancy did not change over time, but young women were significantly more likely (46·5% compared with 20·9% in older women) to have HIV diagnosed in pregnancy (OR 3·29, 95% CI 2·54–4·25, P < 0·001). Among the women with HIV already known, overall treatment coverage at conception was 68·2%, with a significant increase in recent years, from 61·9% in 2001–2005 to 80·1% in 2011–2016 (P < 0·001). This significant temporal trend was present in both age groups (Table 1), but younger women were significantly less likely to be on treatment at conception (overall: 56·3% vs. 69·0%, OR 0·58, 95% CI 0·45–0·81, P = 0·002). Subsequent treatment coverage in pregnancy was almost universal, with ARV treatment administered in 96·5% of all pregnancies (details in Table 1), and in 99·2% (2177/2195) of the pregnancies that did not end in miscarriage, stillbirth, or voluntary termination (98·5% in women <25 years vs. 99·3% in older women, P = 0·240). ARV prophylaxis in infants also showed no difference by maternal age (97·2% and 97·3% in infants from younger and older women, respectively; OR 0·94, 95% CI 0·37–2·39, P = 0·895). The main pregnancy outcomes, reported in Table 2, showed no major differences by age group. Among the pregnancies that ended in a live birth, the two age groups had similar rate of undetectable HIV viral load at third trimester (58·2% vs. 65·2%, Table 2). Given the possible role of some confounders in the significant differences observed, we performed two logistic regression analyses that analyzed the role of other potential relevant cofactors as predictors of pregnancy planned and HIV diagnosis in pregnancy. Such analyses confirmed the independent role of young age as a predictor of unplanned pregnancy and undiagnosed HIV infection before pregnancy. The details of univariate and multivariate analyses for these two outcomes, with adjusted ORs for age and the other covariates are reported in Table 3.
* χ 2 for trend.
† Women with HIV infection diagnosed before pregnancy only.
OR, odds ratio; CI, confidence interval.
* χ 2.
† Birthweight <10th percentile by age and gender, singletons only.
‡ Livebirths only.
Univariate analyses, contingency tables with χ 2 test; UOR, unadjusted odds ratio; CI, confidence interval; multivariate analysis, multivariable logistic regression models; AOR, adjusted odds ratio.
This longitudinal study provided new findings on HIV and pregnancy in young women. First, we showed that within the population of pregnant women with HIV in Italy, the proportion of younger women is increasing. This finding is consistent with the high number of new HIV infections reported by WHO in young people , and with other local studies that reported a recent increase in the proportion of young pregnant women with HIV [Reference Fatti8]. We also showed that, despite a positive temporal trend indicating increasing rates of planned pregnancy, a very high proportion of pregnancies (63%) remain unplanned among women with HIV. National data are scarce, but the rate of planned pregnancy in the general population is likely to be higher [Reference Krause9]. In terms of predictors, younger age and foreign origin were two strong and independent determinants of both unplanned pregnancy and diagnosis of HIV in pregnancy. This significant role of younger age and foreign origin in pregnancy planning is consistent with the marked differences reported in fertility and abortion rates between Italian and foreign women [Reference Marchesi10], and with the higher rate of induced abortion observed in younger women with HIV [Reference Ammassari11]. Overall, these findings underline the urgent need to promote more efficiently HIV testing and reproductive counseling among young [Reference Kenny12, Reference Echenique13] and migrant women [Reference Lauria14].
It has already been shown that women with HIV are less likely to receive adequate and timely prenatal care compared with women without HIV [Reference Ng15]. The later entry in HIV care observed for younger women indicates an even more vulnerable subgroup within a population who already receives suboptimal care. Although the analysis of maternal treatment coverage and pregnancy outcomes showed no significant differences by maternal age in all the outcomes considered, younger women, being quite commonly unaware of their HIV infection status at booking in pregnancy, were less frequently on ARV treatment at conception, as already described, with a later entry in HIV care and treatment [Reference Fatti8]. Given the strong consequential associations that link late maternal HIV diagnosis and later start of treatment to unsuppressed HIV viral load and mother-to-child transmission, a later entry in HIV care may potentially translate in some infants infected with HIV [Reference Fatti8, Reference Momplaisir16], and should not be therefore underestimated. In this study, the number of infants vertically infected with HIV was very low, not allowing any conclusion, but younger women had slightly lower rates of viral load suppression at third trimester, and slightly higher rates of vertical transmission.
The present findings therefore confirm the urgency of implementing as much as possible the WHO recommendations, with particular attention to promoting and enhancing HIV testing before pregnancy among young women and among foreign women [1, Reference Musumari17]. Women's knowledge of their own HIV-positive status is a significant predictor of family planning [Reference Habte and Namasasu18], and there is evidence that women already aware of being infected before pregnancy have an earlier attendance of antenatal care, with potential significant benefits [Reference Gill19]. Although HIV testing in pregnancy is extremely relevant and effective, earlier interventions may provide additional benefits to young and foreign women in terms of family planning and engagement in care. Potentially relevant strategies include promotion of testing and antenatal counseling by cross-cultural phone counseling [Reference Taglieri20], wider and better implementation of contraception counseling in pre- and post-natal maternal care [Reference Lauria14], immediate testing in the presence of HIV indicator diseases [Reference Scognamiglio21], and rapid HIV testing in outpatients belonging to vulnerable/at-risk populations [Reference Uccella22].
The authors thank Cosimo Polizzi and Alessandra Mattei of the Istituto Superiore di Sanità in Rome, Italy, for providing technical secretarial for this study. No compensation was received for this contribution.
This work (currently not funded) was supported in the past by public research grants (ref.: H85E08000200005) from the Italian Medicines Agency (AIFA). The funder had no role in study design, data collection, data analysis, manuscript preparation, and/or publication decision. No funding was received for this work from any of the following organizations: National Institutes of Health (NIH); Wellcome Trust; and the Howard Hughes Medical Institute (HHMI). The corresponding author had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
DECLARATION OF INTEREST
Ethics approval was obtained from the Ethics Committee of the I.N.M.I. Lazzaro Spallanzani in Rome (ref. deliberation 578/2001, 28 September 2001).