HIV and obstetric complications and fetal outcomes in Vellore,India

Introduction

Worldwide, 18.5 million women are infected with HIV. Over 90% of newly detected HIV infections occur in developing countries and over four million people (38% women) are infected with HIV in India. Antenatal clinics in India have reported a greater than 1% seropositivity. ¹

HIV-infected women are more likely than uninfected women to have adverse pregnancy outcomes. ² They experience higher rates of spontaneous abortion, stillbirth, fetal abnormality, maternal mortality, perinatal mortality, preterm birth, low birth weight, intrauterine growth restriction, postpartum haemorrhage and anaemia. ³ Infant mortality rates in HIV-infected women are especially high in developing countries. ² Some link these adverse outcomes to maternal CD4 and infant HIV infection along with complicating conditions such as vaginal bleeding, hypertension, anaemia, pre-eclampsia, eclampsia and diabetes or confounding sociodemographic factors. ² Conflicting results on the effects of antiretroviral therapy (ART) on pregnancy outcomes have further complicated the issue. Although many of the long-term risks and benefits of ART on obstetric outcomes are still unknown, some studies suggest that ART during pregnancy may be associated with premature delivery, maternal hyperglycaemia, anaemia, exacerbation of diabetes mellitus and mitochondrial dysfunction. ³ However, a large meta analysis found no differences between women who received combination therapy, monotherapy or no therapy and rates of preterm labour, low birth weight, low Apgar scores or stillbirth. ⁴ Most of these studies have been conducted in western or sub-Saharan African countries, with sparse information available from India.

Methods

A retrospective chart and perinatal audit review, approved by the institutional review board at the Christian Medical College Hospital (CMCH), was performed on 23,386 women who delivered at CMCH 2000–2002. Details of obstetric and fetal complications were collected. All subjects were screened for HIV by enzyme-linked immunosorbent assay (Genedia and Genscreen). Financial capability and time of presentation determined whether women received mother-to-child transmission (MTCT) prophylaxis and what type of prophylaxis was administered. Those who had scheduled appointments and were on zidovudine (AZT) prophylaxis had their routine dose before elective caesarean sections. Women who received AZT prophylaxis but delivered vaginally did not receive any AZT during labour because intravenous AZT was not available in India during the study period. Those who did not have appointments and were diagnosed in labour and women who opted for nevirapine (NVP) due to the cost received NVP at the onset of labour. All infants were given HIV prophylaxis, primarily with AZT syrup (2 mg/kg 4 times a day for 6 weeks).

Demographics and HIV history were collected for all HIV-infected women. Statistical analyses were performed with SPSS software (release 10.0, SPSS, Chicago, IL, USA). Comparisons of obstetric complication and fetal outcome rates were performed, including mode of delivery, maternal and fetal deaths, pregnancy-induced hypertension, anaemia, eclampsia, diabetes, abruptio placenta, previa placenta, haemorrhage, infections and anomalies. Anaemia is defined as haemoglobin < 8.4 g% and the test was done using a coulter machine. Relative risk and 95% confidence intervals were calculated. Chi-square statistic, or Yates correction for small-expected cell sizes, was used to determine potential correlate variables between HIV status, obstetric complications and fetal outcomes.

Results

Of the 23,386 women, 0.5% (n = 109) of the women were HIV-infected. Table 1 lists demographic, delivery and MTCT prophylaxis information for the HIV-infected women: 11% were illiterate; 95% were housewives and 96% of their husband's were HIV positive. All those who were registered with us before 34 weeks of gestation were offered MTCT prophylaxis with AZT 300 mg twice daily and newborn syrup AZT 2 mg/kg four times a day for six weeks (Thai short, long regimen) All women diagnosed as HIV positive in labour or late registration were offered single dose NVP in labour. Registered women who could not afford AZT for four weeks were given single dose NVP in labour. All HIV-positive women who were registered in the antenatal clinic were given the option of delivering by elective caesarean section as one of the methods for prevention of MTCT. Those who did not want a caesarean were allowed to deliver vaginally. All HIV-positive women who came as direct admissions into the labour ward were allowed to complete labour and deliver vaginally. Emergency caesareans were only done if there were obstetric indications.

The correlations between HIV status, obstetrical complications and fetal outcomes are shown in Table 2. HIV-infected women were more likely than uninfected women to have pregnancy-induced hypertension, anaemia, breech presentations and stillborn babies. To reduce MTCT, more HIV-infected women had caesarean sections (P = 0.001). Of the 12 fetal deaths in HIV-infected women, nine (75%) were macerated stillbirths, one (8%) was due to an obstructed labour and ruptured uterus, one (8%) to multiple congenital anomalies and one (8%) to unknown causes. Of the nine macerated stillbirths: five (56%) had no detectible obstetrical, medical or surgical cause; two (22%) were secondary to severe pre-eclampsia and preterm delivery; one (11%) to prematurity and one (11%) from maternal malarial infection. Among uninfected women: 476 (47%) of fetal deaths were from asphyxiation; 272 (27%) from macerated stillbirths; 163 (16%) from anomalies; 49 (5%) from prematurity; 22 (2%) from other causes and (21) 2% from unknown causes. In both univariate and multivariate analysis, fetal deaths were more likely to occur in HIV-infected women who did not receive MTCT prophylaxis (P = 0.001) or who had breech fetal presentation (P = 0.001).

Discussion

As in previous studies, there were a greater number of adverse pregnancy outcomes among HIV-infected women in our study. ² The association between maternal HIV infection and fetal mortality is particularly strong in developing countries. ² The lack of maternal MTCT prophylaxis was associated with fetal mortality; however, unknown variations in sociodemographics, medical history, time of presentation, prenatal care and other details of delivery may account for some of these differences. As in previous studies, breech presentations were also associated with fetal mortality. ⁵ In our study, the five macerated stillbirths of unknown cause, the one due to prematurity as well as the one fetal death of unknown cause may have been linked to HIV infection.

This study was limited by the inability to obtain information on prenatal care, medical history, financial status and time of presentation. It is difficult to determine whether adverse outcomes associated with a lack of maternal prophylaxis were related to differences in sociodemographics and prenatal care. Although the women were counselled and referred for tests on their infants, the HIV status of the infants is unknown. Further investigations should be made to determine the additional obstetric risks of HIV-infected women and their infants in order to provide proper monitoring, care and preventive measures. HIV prophylaxis and optimal prenatal care should be available to HIV-infected women.

The conclusion from this study is that the HIV-infected women were significantly more likely to have pregnancy-induced hypertension, anaemia, breech presentations and stillborn babies. Therefore, medical personnel need to be aware of these issues when providing care for HIV-infected pregnant women.