Early infant diagnosis of HIV infection at the John F. Kennedy Medical Center, Monrovia, Liberia

Abstract

Background: Vertical transmission accounts for majority of new HIV infections among children worldwide. Ninety percent of HIV-positive children reside in Sub- Saharan Africa with their infection predominantly acquired via vertical transmission. In 2004, the vertical transmission rate of HIV in Africa was estimated at 25 − 40% but, remarkably, the rate has significantly decreased to less than 5% in most African countries following implementation and expansion of prevention of MTCT (PMTCT) programs.

Objective: To determine the rate of and factors associated with vertical transmission of HIV among attendees of early infant diagnosis (EID) program of an academic and community-based tertiary facility in Liberia.

Design: A retrospective cross-sectional analysis.

Methods: A retrospective review of medical records of babies seen at Pediatric Unit of Infectious Disease Clinic of John F Kennedy Medical Center (JFKMC) in Monrovia, Liberia between January 1, 2016 and December 31, 2020. All subjects were children born to HIV-positive mothers and who had HIV DNA PCR testing performed between the ages of 6 weeks and 6 months. Children who suffered early neonatal death and those who did not undergo PCR testing were excluded. Demographics of mother to child pairs as well as factors known to influence vertical transmission of HIV such as partial (15.8%) or full (84.2%) participation in prevention of MTCT (PMTCT) programs, mode of delivery, breastfeeding and utilization of post-exposure prophylaxis were collected and assessed. Binomial logistic regression analyses were used to assess factors associated with vertical transmission.

Results: During the study timeframe, 284 children had a HIV DNA PCR test with a male:female ratio − 1.3:1. Sixteen tested positive (conducted at a mean of 155 days post birth) giving a vertical transmission rate of 5.6%. For 239 mothers (84.2%) who had full PMTCT, 1.3% of their children tested positive, while for 45 mothers (15.8%) who had partial PMTCT, 28.8% of their children being positive. Two hundred and seventy six children (97%) had exclusive breastfeeding, 13 of whom tested positive while 2 children who were mixed fed tested positive. Children who had Nevirapine vs no prophylaxis (OR = 1.89[95% CI 1.16 − 2.96]), were delivered via caesarian section vs vaginal delivery (OR= 2.26[95% CI 1.92 − 4.12].) and full versus partial participation in PMTCT programs (OR = 4.02[95% CI 2.06 − 4.13] were more likely to have negative HIV test.

Conclusion: Vertical transmission rate was found to be high in Liberia and may be driven by suboptimal PMTCT program participation including post-exposure prophylaxis for infants. Therefore, strategies to scale up and improve uptake of PMTCT services are needed to mitigate the burden of HIV among children.

Keywords:

• HIV testing
• early diagnosis
• HIV infections
• infant
• Immunologic tests

Introduction

Globally, Vertical transmission of HIV accounts for the majority of new HIV infections among children.¹ Sub-Saharan Africa bears the highest burden, harboring ninety percent of children living with HIV most of whom acquired HIV through vertical transmission (95%).¹ The high prevalence of HIV among women of reproductive age and the high fertility rate of African women are some of the factors that contribute to the comparatively high prevalence of transmission of HIV to infants.²

It is critically important to diagnose HIV in children as early as possible, and commence treatment, as the disease progression is more rapid than in adults and associated with high rates of early mortality in infants with perinatal HIV infection.³ Early Infant Diagnosis (EID) is the critical step for the diagnosis and subsequent implementation of early treatment with combination antiretroviral therapy (cART). Early Infant diagnosis is based on PCR tests which detect viral nucleic acids that is particularly ideal for HIV-exposed infants within 6 weeks of birth and up to 18 months of age who may still have circulating maternal antibodies that can bias antibody-based tests.^4–6

Furthermore, EID provides the basis for entry into treatment programs for infected babies. Prompt and timely initiation of Highly Active Anti-Retroviral Therapy (HAART) in babies who acquired improves treatment outcomes and contributes to epidemic control.^3,7 In addition, EID provides an opportunity for communication between health care providers and parents or guardians around resources for follow-up of uninfected HIV- exposed infants(HEIs) to ensure they remain negative, and ways to prevent infections for future children.⁸

Several countries have reported declining rates of HIV infection among infants following implementation of routine EID ⁹ and implementation and expansion of other PMTCT programs. For example, Uganda reported a decline in infant diagnoses rates from 10% to 4% from 2011 to 2015 while South Africa reported a 40% decline in HIV incidence from 2011 to 2015, and Nigeria reported a decline from 9.7% in 2008 to 4.9% in 2014.¹⁰ However, these rates are still relatively high compared to those reported from developed countries like USA.¹¹

There is dearth of information on EID in Liberia; therefore, this study was conducted to assess HIV positivity rates and associated factors among attendees of the EID program of the largest tertiary facility in Liberia, John F. Kennedy Medical Center (JFKMC) in Monrovia. This study adds to the body of literature available on EID in West-African children.

Materials and methods

The study was a retrospective review of medical records of children seen at the Pediatric Unit of the Infectious Disease Clinic of JFKMC between January 1, 2016 and December 31, 2020. Included in the study, were children born to HIV-positive mothers who had HIV DNA PCR testing between the ages of 6 weeks to 6 months. Children who suffered early neonatal death and those who did not undergo PCR testing were excluded.

Samples were collected as dried blood spots on filter paper by trained nurse counselors from infants of mothers living with HIV using heel or big toe prick. The filter paper was dried overnight at room temperature and sealed in humidity free bags and, after a period of five to seven days, sent to the molecular laboratory. One spot from each filter paper was tested by COBAS Ampliprep/Taqman HIV-1 Version 2.0 real time PCR assay (ROCHE Diagnostics, Indianapolis, IN) according to manufacturer instructions. Results are transported from the laboratory back to the health facility and then the provider informs the children’s caregivers on their post-test appointment day. The date of collection of the dried blood spots card and subsequent result as well as prophylaxis medication offered were recorded in the child’s medical record. All positive tests were confirmed by collecting a second sample from the child and performing another DNA-PCR test. However, HIV treatment was typically commenced while the confirmatory test results were awaited.

Medical records of the patients were manually retrieved, and a structured database was used to extract relevant data which included age and gender of children, maternal use of anti-retrovirals (ARVs) during pregnancy, mode of delivery and delivery complications if they occurred, use of Nevirapine as post-exposure prophylaxis (PEP) and mode of infant feeding. In addition, results of DNA PCR were extracted from the file of each baby. However, information on maternal viremia was not available.

Prevention of Mother to Child Transmission (PMTCT) participation was categorized as full, partial or none as follows:

• FULL –maternal receipt of ARVs, elective Caesarian section or PMTCT protocol for vaginal delivery, infant ARVs for PEP, and appropriate infant feeding option (exclusive breastfeeding or use of breast milk substitutes for six months followed by complementary feeding) Any patient with incomplete of these pieces of information or not clear in the case note was excluded from the study.

• PARTIAL – when one or more of the components were not implemented

• NONE – when no component implemented

The following definitions were also utilized during data extraction:

• Exclusive breastfeeding was defined as feeding an infant with breast milk only.¹² This excluded the use of formula feed, or any other liquids or solids.¹² The use of the prescribed medications and oral rehydration salt (ORS) for diarrhea was allowed, as per WHO definitions.¹²

• Mixed feeding was defined as feeding an infant with both breast milk and formula feed, or any other liquid, or solids.¹²

• Replacement feeding/not breastfed at all, refers to avoiding all breast milk and feeding an infant with an appropriate replacement milk.¹²

• Mother to child transmission rate was defined as the proportion of HIV-exposed infants tested who had positive results.

Approval for the study was obtained from the Ethics Committee of JFKMC, before commencement of the study.

Analysis

Study data from study database were entered into SPSS soft-ware version 20.0. IBM Corp., Armonk, NY, USA for descriptive statistical analysis. Furthermore, demographics of mother to child pairs as well as factors known to influence MTCT of HIV such as degree of participation in PMTCT programs (no, partial or full), mode of delivery (vaginal birth versus caesarian section), breastfeeding (exclusive or mixed) and utilization of post-exposure prophylaxis were assessed for association with HIV test positivity using Binomial logistic regression analyses. P-values <0.05 were considered statistically significant.

Results

A total of 284 children had a PCR test performed with a male: female ratio of 1.3:1. Sixteen of the children tested positive giving an MTCT rate of 5.6% with no predominant gender affected. Test turn-around time after sample collection ranged from 121 to 183 days, with a mean of 155 days. All mothers participated in one or more components of PMTCT with 239 (84.2%) having full PMTCT, and 45 (15.8%) having only partial PMTCT. Only 1.3% (3/239) of the children from mothers with full PMTCT had positive PCR test but 28.8% (13/45) of babies were positive from mothers with partial PMTCT (Table 1).All but eight children (276) had exclusive breastfeeding among whom 13 tested positive. Two children that were mixed fed in the first six months of life were both positive and 1 out of 6 children who received replacement feeding tested positive (Table 2).

Table 1. HIV PCR Test result by PMTCT participation.

PMTCT	POSITIVE (%)	NEGATIVE (%)	TOTAL (%)
FULL	3 (1.1)	236 (83.1)	239 (84.2)
PARTIAL	13 (4.5)	32 (11.3)	45 (15.8)
TOTAL	16 (5.6)	268 (94.4)	284 (100)

Table 2. HIV PCR Test result by infant feeding option.

FEEDING OPTION	POSITIVE (%)	NEGATIVE (%)	TOTAL (%)
BREASTFEEDING	13 (4.5)	263 (92.6)	276 (97.2)
REPLACEMENT FEEDING	1 (0.4)	5 (1.8)	6 (2.1)
MIXED FEEDING	2 (0.7)	0 (0)	2 (0.7)
TOTAL	16 (5.6)	268 (94.4)	284 (100)

Regression analyses showed that after controlling for other factors in the model, children who were exclusively breastfed were 2 times (OR= 2.12 [95% CI 2.51 − 4.13]) more likely to have negative PCR results when compared to those who had mixed feeding (p < 0,056), while children who received only replacement feeding were 3 times (0 R= 3 [95% CI 1.83 − 3.13]) more likely to have negative PCR results compared to those that received exclusive breastfeeding or mixed feeding (p < 0.004). Children who had Nevirapine prophylaxis were more likely to have negative PCR results compared to children who had no prophylaxis (OR = 1.89 [95% CI 1.16 − 2.96]), and children delivered via elective cesarean section were also more likely to have negative PCR results compared to children who were delivered vaginally (OR = 2.26 [95% CI 1.92 − 4.12]). Similarly, odds of having a negative PCR result was 4 (4.02, 95% CI 2.06 − 4.13) for full versus partial PMTCT participation (Table 3). All differences were statistically significant.

Table 3. Logistic regression predicting likelihood of negative DNA-PCR based on nevirapine prophylaxis, exclusive breastfeeding, elective cesarean section and PMTCT.

Characteristics	OR	95% CI	p value
Nevirapine Prophylaxis^a	1.89	1.16 − 2.96	0.001
Exclusive Breastfeeding^b	5.12	3.41 − 6.13	0.056
Replacement feeding^c	3.19	1.83 − 3.13	0.004
Elective Caesarean Section^d	2.26	1.92 − 4.12	0.003
PMTCT^e	4.02	2.06 − 4.13	0.002

^aProphylaxis compared to no prophylaxis.

^bExclusive breastfeeding to mixed feeding.

^cRelacement feeding to exclusive breastfeeding & mixed feeding.

^dCesarean section to vaginal delivery.

^eFull PMTCT to partial.

Discussion

The Mother to-Child Transmission rate found among the study population was 5.6%, which is similar to the rate reported in other developing countries within the sub-Saharan Africa region.⁹ For example, a contemporary study by Ademola et al in Nigeria reported a rate of 4.9% in 2019.¹⁰ While at par with MTCT rates of other African countries, the rates are still relatively high compared to those in developed countries like USA with MTCT rate of <1%.¹¹ This difference is likely due to the challenges of full implementation and uptake of elimination of vertical transmission components in our setting. Furthermore, because of limited capacity for and implementation of EID, the positivity rates may be an underestimation of the true burden of vertical transmission in Liberia. Impact criteria by WHO for validation of elimination of vertical transmission include: for HIV, ≤50 new pediatric infections per 100,000 live births and a transmission rate of either <5% in breastfeeding populations or <2% in non-breastfeeding populations.¹³ Required process criteria for validation of vertical transmission include: 95% of pregnant women to receive antenatal care (ANC); 95% of pregnant women to receive HIV and syphilis testing in pregnancy; and 95% of pregnant women diagnosed with HIV or syphilis to receive treatment.¹³ These are yet to be achieved in Liberia.

In our EID program, we primarily utilized dried blood spot testing,¹⁴ a method that has been shown to remarkably improve rates of EID in low-resource settings. However, the mean test turn-around time in our study was 155 days. This is very high when compared to turn-around times that have been reported in settings similar to ours: Kenya (25 days), Lesotho (63 days), Nigeria (47 days), and Tanzania (35 days).^15–18 Our test turn-around time is very long in Liberia as dried blood spot (DBS) testing is not available at all levels of health facilities, therefore all DBS samples collected from all facilities must be sent to a central testing reference laboratory. Thus, there is a significant delay after a specimen is obtained, including lag time with transportation of the specimen to the testing laboratory, and transmission of the results from the laboratory back to the provider and then the children’s caregivers. Unfortunately, majority of infants found to be HIV-positive in our program initiated ART later than age of 12 weeks, past the period of peak mortality. Therefore, this occurrence likely limits the survival gains of commencing early infant antiretroviral therapy when diagnosis is delayed and children may experience morbidity (rapid disease progression) and mortality. Furthermore, from a programmatic and clinical standpoint, prolonged delays in HIV diagnosis in setting of use of single antiretroviral agent- nevirapine- for postexposure prophylaxis, may fuel drug resistance.

Breastfeeding is important for child survival^19,20 and it is recommended that mothers living with HIV, especially in low and middle income countries (LMIC), should breastfeed for 12 months and may continue breastfeeding for up to 24 months or longer, while being fully supported for ART adherence.²¹ This is to protect against malnutrition and death resulting from diseases associated with poor food and water hygiene. Breastfeeding may be stopped once a nutritionally adequate and safe diet without breast milk can be provided, a luxury for many women in resource limited settings.²¹However, while exclusive breastfeeding by HIV positive mothers is a risk factor for HIV infection, in some studies,²² mixed feeding as compared to exclusive breastfeeding, is a more significant predictor of HIV infection as it is linked to continuing risk of postnatal infection as breastfeeding occurs for even longer durations. This finding is similar with other related studies done in Addis Ababa, Ethiopia²³ and in Zimbabwe.²⁴A postulated mechanism is that irritation of infant’s immature gastrointestinal tract caused by additional foods which might facilitate entry of HIV viral particles from the mother’s breast milk into the child’s blood stream.²⁵

For social and cultural reasons in Africa, elective caesarean section is not a popular method of preventing perinatal transmission of HIV though it is a well-established method of preventing vertical transmission from mothers who are viremic at time of delivery. All the babies delivered through elective caesarean section (C-section) in our study were found to be HIV negative. This is in agreement with studies from Nigeria² and these data could be utilized to enhance communication around the role of C-sectioning mitigating vertical transmission with caregivers and HIV-infected mothers.

In this study, partial PMTCT intervention which included missing components such as non-utilization of cART during pregnancy and infant Nevirapine prophylaxis or inappropriate infant feeding options were significantly associated with vertical transmission of HIV. This is similar to related studies in Nigeria and Ethiopia.^26,27 This finding highlights the need to ensure that the comprehensive package or entire bundle of PMTCT components are implemented together and delivered as appropriate to optimize and maximize the benefit of having HIV-free children.

Our study has significant limitations: information collected from medical records may have been incomplete for many reasons including failure to chart events or treatments given, erroneous recall of data by providers or caregivers and or missing documentation as the records are paper-based. We did not measure adherence to ART by mothers or adherence to and duration of post-exposure prophylaxis use in infants in our program, therefore mother and child pairs may have experienced suboptimal “protection” from HIV transmission where adherence was poor. The various components of PMTCT have varying effectiveness with decreasing vertical transmission, so for those with partial uptake, it is difficult to determine which missing component drove negative outcomes. Our results are reflective of an EID program that is severely under-resourced and accessed by women and children from urban and semi-urban areas in Liberia, thus may not be generalizable to setting with dissimilar characteristics and resources. Lastly, sample size was small such that secondary analysis of factors associated with having a negative HIV test, may have been underpowered for significance.

Conclusion

Vertical transmission rate is still relatively high in a tertiary facility in Liberia especially among mothers with partial PMTCT participation. Specifically, mixed feeding, non-utilization Nevirapine prophylaxis and vaginal delivery were associated with higher rates of vertical transmission of HIV. Furthermore, strategies to scale up and bundle PMTCT services are needed to mitigate the burden of HIV among children.

Funding

The author(s) reported there is no funding associated with the work featured in this article.