Epidemiological drivers of mother to child HIV transmission in West Bengal,India: A retrospective cohort study

Introduction

Prevention of Parent to Child Transmission (PPTCT) aims to lower the transmission rate of HIV acquired vertically from a positive mother to her child. Vertical transmission of HIV occurs during the antenatal, intranatal, and postnatal lactating periods. This can be prevented with appropriate interventions, such as administering antiretroviral therapy (ART) to mothers and antiretroviral (ARV) prophylaxis to exposed infants. The risk of vertical transmission of HIV is 30–45% in the absence of any intervention, but can be minimized to less than 2% with intervention. ¹ There are several demographic features and practices which are supposed to influence the transmission dynamics. In India, an estimated 79,800 children aged 0–14 years are living with HIV infection. An estimated 20,520 pregnancies annually are expected to be associated with maternal HIV. Prevention of Parent to Child Transmission services provide pregnant women access to cascade services like diagnosis, treatment, and prevention. The study ultimately aimed to ensure the integration of PPTCT services within existing Reproductive and Child Health programs. ²

Mother to child transmission (MTCT) of HIV is preventable to a large extent by administering ART to HIV-positive women and ARV prophylaxis to HIV-exposed babies. But first, certain epidemiological factors that play an important role in deciding HIV-related outcomes for exposed babies must be understood. Some international studies have pointed out epidemiological attributes like mode of delivery, infant feeding options, adherence to PPTCT cascades, etc. can impact HIV-acquisition outcomes in children of HIV-positive women. ³

In the Indian context, there are a few studies which deal with MTCT of HIV but the studies related to correlation with socio demographic attributes with prevention of MTCT are numbered. Even in the international arena, most of the studies conducted in relation to MTCT reported mode of delivery, treatment adherence, antiretroviral literacy, infant feeding options, nature of antiretroviral prophylaxis etc.

The outcome of PPTCT program in India varies from state to state in spite of effective and uniform program strategy. This may be attributed to region-specific variations in demographic features and practices. There are a limited number of studies which have tried to find correlations of vertical transmission of HIV with related demographic attributes. This study is unique as it dealt with more than ten attributes which may play crucial role in determining mother to child HIV transmission rate.

The objective of this study was to identify the factors that contribute towards modifying the vertically-acquired HIV transmission rate.

Materials and method

Operational case definition

(1) HIV-positive child: a child with HIV infection detected through Nucleic Acid PCR (Figure 1)

OPEN IN VIEWER

Figure 1.

Inclusion and exclusion criteria matrix.

Exclusion criteria

(1) Lost to follow-up cases with unknown pregnancy outcomes.

Results

Out of the 518 HIV-exposed babies, 41 (7.9%) were found to be HIV-reactive on dry blood spot Nucleic Acid PCR, and, hence, considered to be HIV-positive. The remaining 477 HIV-exposed babies (92.1%) were non-reactive and, hence, considered to be HIV-negative. The distribution of demographic and HIV-related characteristics are presented in Table 1. The biological sex of 48% (20) and 46% (222) of all HIV-positive and HIV-negative babies, respectively, was female. The mean birth weight of HIV-positive babies was 2.5 (+/−0.338) kg while that of HIV-negative babies was 2.7 (+/−0.523) kg. The median birth order for HIV-positive and HIV-negative babies was one and two, respectively. 31% of mothers of HIV-positive babies were illiterate as compared to only 16% of mothers of HIV-negative babies. Maternal detection in most of the mothers (79%) of HIV-negative babies happened during the antenatal period, whereas the same for a substantial proportion of mothers (43%) of HIV-positive babies happened during the postnatal period. 8% of mothers of HIV-positive babies underwent normal delivery as compared to 67% mothers with HIV-negative babies who delivered by Cesarean surgery (C-section). 39% of HIV-positive babies were administered ARV prophylaxis at birth as compared to 87% of babies in the HIV-negative cohort. While 49% of babies from the HIV-positive group received nevirapine, the remaining 51% did not receive any prophylaxis. 99.6% babies in the HIV-negative group received any form of prophylaxis, with 92% being given nevirapine, followed by zidovudine (6%), and lopinavir/ritonavir (1%). [Table 1]

OPEN IN VIEWER

Table 1.

Distribution of different maternal and child attribute between HIV positive and negative babies.

Characteristics		Positive babies (N = 41)		Negative babies (N = 477
		n	%	n	%
Gender of the baby	M	21	51.2	255	53.5
	F	20	48.8	222	46.5
Mean birth wt. of the babies (Kg)	—	2.5 (+/−0.338)		2.7 (+/− 0.523)
Median birth order	—	1(IQR, 1–2)		2(IQR, 1–2)
Educational status of the mother	Illiterate	13	31.7	80	16.8
	Just literate	9	22	29	6.1
	Primary	10	24.4	137	28.7
	Secondary	6	14.6	173	36.3
	Higher secondary and above	3	7.3	58	12.2
Time of detection of HIV of mother	ANC	15	36.6	379	79.5
	DIL	0	0	19	4
	PNC	18	43.9	15	3.1
	Before pregnancy	6	14.6	64	13.4
	mother’s status not known	2	4.9	0	0
Mode of delivery	Normal delivery	36	87.8	322	67.5
	CS	5	12.2	155	32.5
Mean number of weeks ART taken during pregnancy before delivery	—	19.4 (+/− 15.3)		20.6 (+/−13.1)
Prophylaxis initiated at birth	—	16	39	407	87.3
Prophylaxis offered	Nevirapine	20	48.8	442	92.7
	zidovudine	0	0	27	5.7
	lopinavir/ritonavir	0	0	6	1.3
	No prophylaxis	21	51.2	2	0.4

Bivariate analysis, through unpaired t-test of quantitative variables, has been demonstrated in Table 2. Significantly lower mean birth weight of baby and lesser duration (weeks) of ART for mother were observed in the case of HIV-positive babies compared to HIV-negative babies (p < 0.05). However, no significant difference in the age of the mother and the birth order of the baby was observed between the HIV-positive and HIV-negative groups. [Table 2]

OPEN IN VIEWER

Table 2.

Bivariate analysis (unpaired t-test) of quantitative maternal and child attributes between HIV positive and negative babies.

Indicators	HIV status of child	Mean	SD	t	p
Birth Weight	Positive	2.54	0.33	−2.79	0.007
	Negative	2.69	0.52
Age of the mother	Positive	24.29	3.65	−1.21	0.23
	Negative	25.04	4.91
Weeks on ART of mother	Positive	10.41	14.77	−4.74	<0.001
	Negative	20.6	13.06

Bivariate analysis of categorical variables revealed a significant association between MTCT of HIV and several factors such as low birth weight (<2.7 kg), maternal detection during the postnatal period, vaginal delivery, and maternal age (>25 years). However, in multivariate adjusted analysis using binary logistic regression, the association of the first three attributes with HIV infection in the children was found to be significant (p < 0.05). No significant association was observed between HIV-transmission and birth order and sex of the babies. [Table 3]

OPEN IN VIEWER

Table 3.

Bivariate and Multivariate analysis of categorical maternal and child attributes between HIV positive and negative babies.

Attributes		Negative Babies (N = 477)		Positive babies (N = 41)		Crude OR	p value	95% CI	Adjusted OR	p value	95% CI
		n	%	n	%
Birth weight	≥2.5 Kg	256	53.7	10	24.4	3.591	0.001	1.7–7.4	2.667	0.016	1.2–5.9
	<2.5 Kg	221	46.3	31	75.6	—	—	—	—	—	—
HIV Detection time of mother	Before pregnancy and ANC period	443	92.9	21	51.2	12.409	<0.001	6.1–25.1	11.165	<0.001	5.2–23.8
	During delivery, PNC period and status remained unknown	34	7.1	20	48.8	—	—	—	—	—	—
Mode of delivery	CS	155	32.5	5	12.8	3.273	0.015	1.3–8.5	2.822	0.045	1.01–7.73
	Vaginal delivery	322	67.5	34	87.2	—	—	—	—	—	—
Birth order	1	370	77.6	35	87.5	0.494	0.151	0.24–1.5	0.661	0.443	0.22–1.9
	>1	107	22.4	6	12.5	—	—	—	—	—	—
weeks ART taken during pregnancy	<20	249	52.2	19	46.3	1.265	0.472	0.7–2.4	1.496	0.277	0.7–3.1
	≥20	228	47.8	22	53.7	—	—	—	—	—	—
Maternal age	≤24.9	249	52.2	15	35	2.082	0.040	0.9–3.7	1.652	0.220	0.7–3.6
	>25	228	47.8	26	65	—	—	—	—	—	—
Gender of the baby	M	255	53.5	21	51.2	1.094	0.783	0.6–2.1	0.943	0.876	0.5–1.9
	F	222	46.5	20	48.8	—	—	—	—	—	—

Discussion

It was observed in the current study that the birth weight of HIV-exposed babies, time of detection of HIV status of women, mode of delivery, maternal age, and duration of ART during pregnancy contributed significantly in perinatal transmission of HIV among the representative study population. While studying the attributes of the HIV-positive babies in detail, the authors observed that the circumstances that led to or influenced MTCT of HIV was different for every baby. Most of the babies whose mothers had been detected during the postnatal period were HIV-positive. Other information like the type of sexual relationship the mother had engaged in (monogamy/polygamy), the place of maternal detection, mode of delivery and maternal ART initiation, which vary geographically and with the question of whether the mother had disclosed her HIV status at hospital before delivery (as HIV infection is stigmatized) was also found to be relevant to determining and reducing MTCT risk. This study is unique and highly relevant as only a limited number of similar studies to identify epidemiological drivers for MTCT of HIV have so far been conducted in India, and there is dearth of such studies in the eastern part of India as well as in low HIV prevalence settings. One Karnataka-based study demonstrated a significant reduction in vertical transmission risk through the administration of nevirapine prophylaxis to HIV-exposed infants at birth. ⁴ The present study not only corroborated the abovementioned finding but also showed that nevirapine prophylaxis was found to be superior to zidovudin or protease inhibitor prophylaxis in reducing MTCT risk for the HIV-exposed babies. This finding may be attributed to the fact that any prophylactic regimen other than nevirapine could only be given in exceptional circumstances—prior exposure to nevirapine or HIV-2, mixed infection of the mother, mother on second or third line of ART, or other conditions which might lead to archived resistance to non-nucleoside reverse transcriptase inhibitors for the mother.

In order to reduce MTCT of HIV, the basic PMTCT services play the most important role like antenatal services, institutional delivery, ARV prophylaxis for the exposed babies, maternal antiretroviral treatment. Absence of these services may result in increased risk of HIV transmission and an Ethiopia based study also highlighted the same. ⁵ Maternal ART is an independent factor predicting chance of transmission from infected mother to her child. This was echoed in several studies as it was also demonstrated in a study in Mexico. ⁶ Most of the studies from the different parts of the world demonstrated the association of MTCT with maternal ART, ARV prophylaxis, infant feeding options. One Egyptian study demonstrated that exposure of pregnant women to ART for less than 4 weeks, absence of ARV prophylaxis, mixed feeding, and absence of cotrimoxazole prophylaxis increased the risk of MTCT. ⁷ Several studies established the important role of ARV prophylaxis in reduction of MTCT.^8,9 Our study dealt with all the factors to establish correlation of MTCT with maternal ART, ARV prophylaxis to exposed infant, duration of ART, place of delivery etc.

Regarding other demographic attributes, a Nigeria-based study showed that parental education and social class can have an impact on the vertical transmission of HIV. ¹⁰ Parental education is considered to be a key factor in understanding the need for good obstetric care and the importance of ART in pregnancy for better obstetric outcomes. It is also expected that, with an increase in educational attainment, HIV-positive pregnant women tend to show good compliance with ART as well as PPTCT service cascades, which is crucial to favorable obstetric outcomes. Though the present study did not have a socio-economic component to it, it concluded that maternal education has some protective effect in vertical transmission of HIV. Another India-based study conducted in West Bengal also arrived at this conclusion though the findings were not statistically significant. ¹¹

Some international studies have demonstrated that vertical transmission is gender-sensitive. A Kenya-based study on various maternal-and-child-related determinants showed that HIV-exposed female children are more prone to perinatal HIV infection than their male counterparts. But birth weight and gestational age of maternal HIV detection were not found to be related to vertical transmission of HIV-1. ¹² As regards the present study, the gender-sensitivity of vertical transmission of HIV was not statistically significant and birth weight was not found to be a determinant for HIV transmission in both bivariate and multivariate analyses. Regarding gestational age of HIV detection in the mother, the present study, unlike the Kenya-based study, demonstrated the protective efficacy of early detection of maternal HIV infection in the prevention of MTCT. Early detection allows for early administration of ART, and with an increase in maternal exposure to ART during the antenatal period, the chances of securing viral load suppression will be higher at the time of delivery, resulting in a favorable PPTCT outcome. This proposition was corroborated by another Kenya-based study which showed that poor PPTCT uptake, like late infant enrollment, poor adherence to infant prophylaxis, undetected maternal HIV infection, poor access to ART, lack of clinic-based HIV education, and poor maternal adherence of ART are associated with MTCT. ¹³

Another study with French perinatal cohort also corroborated the fact that the rate of MTCT increased with viral load, very premature delivery, and short duration of ART. ¹⁴ The authors of the present study also investigated the extent to which factors like duration of ART during pregnancy and of ARV prophylaxis contribute to HIV infection in exposed babies. While no significant difference in MTCT outcome due to the duration of ARV prophylaxis being administered to the babies was observed, the duration of maternal exposure to ART during pregnancy was found to have protective efficacy in preventing MTCT of HIV.

Since during the study period, mandatory viral load testing during pregnancy was not introduced in the country our study failed to elicit any information on this issue directly and our study did not have any component whether it was a pre-term delivery or term delivery.

In another study conducted in Zimbabawe, mixed feeding was found to be an independent risk factor associated with HIV infection among children while being exclusively breast-fed for less than 6 months was found to be protective. ¹⁵ The study population of the present study hardly had HIV-exposed babies who had been mixed-fed, and feeding options like exclusive breast-feeding and exclusive replacement feeding were not found to be associated with MTCT risk. It is most likely that this effect was due to the saturation of ART coverage.

Regarding the mode of delivery of HIV-positive babies, a Western European study demonstrated that elective Caesarian Section can prevent MTCT even at low maternal viral loads. Diverging mode of delivery patterns in Europe reflected uncertainties regarding the risk-benefit balance of elective Caesarian Section for women on successful ART. ¹⁶ In the present study, the same protective efficacy of Caesarian Section was demonstrated in both bivariate and multivariate analyses. During normal delivery, the fetus has to undergo much wear and tear while traveling through the birth canal, which may lead to injuries resulting in increased chance of exposure to maternal HIV.

Another study in Cameroon suggested that maternal age, type of pregnancy and twin sets were new risk factors for MTCT. As per the study result, younger mothers were more likely to transmit HIV to their children, and more children from single pregnancies were HIV-positive than children from multiple pregnancies. There were more HIV infections in male-female twin sets. ¹⁷ The present study had only two cases of multiple pregnancies with favorable PPTCT outcome, which is why it multiple pregnancies were not included in the spectrum of statistical analysis. However, the study also found that while bivariate analysis showed that older women have significantly favorable PPTCT outcomes, multivariate analysis failed to demonstrate this significant relationship. There may be several reasons for younger women having poor PPTCT outcomes. Younger women tend to acquire HIV and get pregnant almost at the same time, and after the acquisition of HIV during periconceptional period, the chance of antenatal transmission becomes very high due to an initial hike in viral load. Younger women have pelvis during the maturing stage which may cause more wear and tear of the baby during delivery.

While there are hardly any studies that demonstrate a correlation between place of delivery and PPTCT outcomes, one Nigeria-based study showed that HIV-positive pregnant women with higher levels of educational attainment and the child’s higher birth order were associated with more facility level delivery. ¹⁸ The present study tried to determine whether there is any relationship between place of delivery (delivery at public or private facility/home birth) and PPTCT outcomes but no significant correlation was observed.

Increased duration of maternal ART has got favorable impact in MTCT. One Namibia based study demonstrated that preconceptional ART initiation had the lowest risk of MTCT. ¹⁹ Another Canada based study also demonstrated that lower duration of maternal ART before delivery increased the chance of transmission. ²⁰ All these results are in tune with our study outcome.

As has been demonstrated in the above discussion, there are few studies which have dealt with the effect of so many demographic attributes on PPTCT within a single study. Even fewer studies of this kind have been conducted in India. This makes the present study both unique and important for securing a reduction in MTCT through improved PPTCT interventions.

Conclusion

Mother to child transmission of HIV is not solely related to effective maternal treatment and infant prophylaxis, but involves a range of demographic factors. Therefore, interventions like HIV education and counseling, adapted to address the unique needs and circumstances of each HIV-positive pregnant woman, may be effective in managing this vertically transmitted infection. Molecular-level studies are needed to further corroborate these facts.

Limitation of the study

Our study did not have the component of socio-economic status of the family of the positive pregnant women. Moreover, in our cohort we had only two cases of twin pregnancies, so outcome in the multiple pregnancies could not be studied. Similarly, presence of HIV 2 or HIV 1 and 2 mixed infections were not considered as we got no such cases in our study cohort.