Recognition of infants at high risk for vertical HIV transmission at delivery in rural Western Cape Province, South Africa.

Background: Despite South Africa's substantial reduction in vertical HIV transmission (VHT), national paediatric HIV elimination is not yet attained. National and Western Cape Province (WC) HIV guidelines recommend enhanced postnatal prophylaxis for infants at high risk for VHT, identified in the WC 2015/2016 guidelines by any single high-risk criterion (maternal antiretroviral therapy (ART) <12 weeks, absent/ unsuppressed maternal HIV viral load (HIV-VL) <12 weeks before/including delivery, spontaneous preterm labour, prolonged rupture of membranes, chorioamnionitis). Accuracy of high-risk infant identification is unknown. Objectives: Primarily, to determine the proportion of infants at high risk for VHT, the accuracy of labour-ward risk classification, the criteria determining high-risk statuses and the criteria missed among unrecognised high-risk infants; secondarily, to determine maternal factors associated with high-risk infants. Methods: Infants born to women living with HIV at a rural regional hospital (May 2016 - April 2017) were retrospectively evaluated using data from the labour ward VHT register, standardised maternity case records, National Health Laboratory Service database and WC Provincial Health Data Centre. The study-derived risk status for each infant was determined using documented presence/absence of risk criteria and compared with labour ward assigned risk to determine accuracy. Proportions of high-risk and unrecognised high-risk infants with each high-risk criterion were determined. Maternal characteristics associated with having a high-risk infant were evaluated using multivariable logistic regression. Results: For liveborn infants, labour ward assigned risk classifications were 40% (n=75/188) high risk, 50% (n=94/188) low risk and 10% (n=19/188) unclassified. Study-derived risk was high risk for 69% (n=129/188) and low risk for 31% (n=59/188), yielding a high-risk classification sensitivity of 51% (95% confidence interval (CI) 42 - 60) and specificity of 69% (95% CI 56 - 80). Absent/unsuppressed HIVVL <12 weeks before delivery accounted for 83% (n=119/143) of study-derived high-risk exposures and 81% (n=60/74) of missed high-risk exposures. Fewer mothers of high-risk infants had >4 antenatal visits (38% v. 81%, p<0.01) and first antenatal visit <20 weeks' gestation (57% v. 77%, p=0.01). Only the number of antenatal visits remained associated with having a high-risk infant after adjusting for gestation at first visit and timing of HIV diagnosis and ART initiation: each additional antenatal visit conferred a 39% (95% CI 25 - 50) reduction in the odds of having a high-risk infant. Conclusion: Labour ward risk classification failed to recognise half of high-risk infants. Infant high-risk status as well as non-detection thereof were driven by suboptimal maternal HIV-VL monitoring. Reinforcing visit frequency later in pregnancy may improve antenatal HIV-VL monitoring, and point-of-care HIV-VL monitoring at delivery could improve recognition of virally unsuppressed mothers and their high-risk infants.

Recognition of infants at high risk for vertical HIV transmission at delivery in rural Western Cape Province, South Africa

Despite South Africa's substantial reduction in vertical HIV transmission (VHT), national paediatric HIV elimination is not yet attained. National and Western Cape Province (WC) HIV guidelines recommend enhanced postnatal prophylaxis for infants at high risk for VHT, identified in the WC 2015/2016 guidelines by any single high-risk criterion (maternal antiretroviral therapy (ART) <12 weeks, absent/ unsuppressed maternal HIV viral load (HIV-VL) <12 weeks before/including delivery, spontaneous preterm labour, prolonged rupture of membranes, chorioamnionitis). Accuracy of high-risk infant identification is unknown. Objectives. Primarily, to determine the proportion of infants at high risk for VHT, the accuracy of labour-ward risk classification, the criteria determining high-risk statuses and the criteria missed among unrecognised high-risk infants; secondarily, to determine maternal factors associated with high-risk infants. Methods. Infants born to women living with HIV at a rural regional hospital (May 2016 - April 2017) were retrospectively evaluated using data from the labour ward VHT register, standardised maternity case records, National Health Laboratory Service database and WC Provincial Health Data Centre. The study-derived risk status for each infant was determined using documented presence/absence of risk criteria and compared with labour ward assigned risk to determine accuracy. Proportions of high-risk and unrecognised high-risk infants with each high-risk criterion were determined. Maternal characteristics associated with having a high-risk infant were evaluated using multivariable logistic regression. Results. For liveborn infants, labour ward assigned risk classifications were 40% (n=75/188) high risk, 50% (n=94/188) low risk and 10% (n=19/188) unclassified. Study-derived risk was high risk for 69% (n=129/188) and low risk for 31% (n=59/188), yielding a high-risk classification sensitivity of 51% (95% confidence interval (CI) 42 - 60) and specificity of 69% (95% CI 56 - 80). Absent/unsuppressed HIVVL <12 weeks before delivery accounted for 83% (n=119/143) of study-derived high-risk exposures and 81% (n=60/74) of missed high-risk exposures. Fewer mothers of high-risk infants had >4 antenatal visits (38% v. 81%, p<0.01) and first antenatal visit <20 weeks' gestation (57% v. 77%, p=0.01). Only the number of antenatal visits remained associated with having a high-risk infant after adjusting for gestation at first visit and timing of HIV diagnosis and ART initiation: each additional antenatal visit conferred a 39% (95% CI 25 - 50) reduction in the odds of having a high-risk infant. Conclusion. Labour ward risk classification failed to recognise half of high-risk infants. Infant high-risk status as well as non-detection thereof were driven by suboptimal maternal HIV-VL monitoring. Reinforcing visit frequency later in pregnancy may improve antenatal HIV-VL monitoring, and point-of-care HIV-VL monitoring at delivery could improve recognition of virally unsuppressed mothers and their high-risk infants

Horizontal HIV transmission to children of HIV-uninfected mothers: A case series and review of the global literature.

BACKGROUND: Vertical transmission is the predominant route for acquisition of HIV infection in children, either in utero, intrapartum or postnatally through breast feeding. Less frequently, children may acquire HIV by horizontal transmission. Horizontal transmission may be healthcare-associated (infusion of HIV-contaminated blood products, use of contaminated needles, syringes and medical equipment, or through ingestion of HIV in expressed breastmilk). Community-acquired HIV transmission to children may occur following surrogate breastfeeding, pre-mastication of food, and sexual abuse. METHODS: Children with suspected horizontally acquired HIV infection were identified by retrospective folder review of existing patients (2004-2014) and by prospective interview and examination of new patients (from 2009 onwards), at a hospital-based paediatric antiretroviral clinic in Cape Town, South Africa. The global literature on horizontal HIV transmission to children (1 January 1986-1 November 2019) was reviewed, to contextualize the local findings. RESULTS: Among the 32 children with horizontal HIV transmission (15 identified retrospectively and 17 prospectively), the median age at first diagnosis was 79 months (interquartile range 28.5-91.5); most children (90.6%) had advanced HIV disease at presentation. HIV transmission was considered healthcare-associated in 15 (46.9%), community-associated in ten (31.3%), possibly healthcare or community-associated in five (15.6 %); and unknown in two children (6.3%). CONCLUSION: Horizontal HIV transmission to children is an important public health issue, with prevention efforts requiring intervention at healthcare facility- and community-level. Greater effort should be made to promptly identify and comprehensively investigate each horizontally HIV-infected child to establish possible routes of transmission and inform future prevention strategies.

Neurodevelopmental and behavioural outcomes of HIV-exposed uninfected and HIV-unexposed children at 2-3 years of age in Cape Town, South Africa.

Successful vertical HIV transmission prevention programmes (VTP) have resulted in an expanding population of HIV-exposed uninfected (HEU) infants whose growth, health and neurodevelopmental outcomes could have consequences for future resource allocation. We compared neurodevelopmental and behavioural outcomes in a prospective cohort of 2-3 year old HEU and HIV-unexposed uninfected (HU) children.Women living with and without HIV and their infants were enrolled within three days of birth from a low-risk midwife obstetric unit in Cape Town, South Africa during 2012 and 2013, under WHO Option A VTP guidelines. HIV-uninfected children aged 30-42 months were assessed using the Bayley scales of Infant Development-Third edition (BSID) and Strengths and Difficulties questionnaire (SDQ).Thirty-two HEU and 27 HU children (mean birth weight 3048g vs 3096g) were assessed. HEU children performed as well as HU children on BSID cognitive, language and motor domains. Mean scores fell within the low average range. Mothers of HEU children reported fewer conduct problems but stunting was associated with increased total difficulties on the SDQ.HEU and HU children's performance on the BSID was similar. In this low-risk cohort, HIV exposure did not confer additional risk. Stunting was associated with increased behavioural problems irrespective of HIV exposure.

Maternal HIV and Paediatric Lung Health.

With improved prevention of mother to child transmission of HIV, paediatric HIV disease is less common. However, the number of HIV exposed but uninfected infants is growing. Exposure to maternal HIV impacts infant respiratory health through an increase in known risk factors such as increased preterm birth and low birth weight, suboptimal breastfeeding, increased psychosocial stressors and increased exposure to infective pathogens. Exposure to the HIV virus and altered maternal immune environment result in immunologic changes in the infant that may contribute to respiratory disease risk. HIV exposed infants are at increased risk for severe pneumonia with poorer outcomes compared to unexposed infants. Maternal ART and optimal nutrition, including breastfeeding in high infective disease burden settings, reduce morbidity and mortality in HIV exposed infants and should be prioritized. The impact of exposure to maternal HIV on normal lung growth and risk for chronic respiratory disease is unknown and warrants further investigation.

Determinants of tuberculosis treatment completion among newborns in a high-burden setting.

BACKGROUND: Newborns exposed to Mycobacterium tuberculosis are at high risk of progression to tuberculosis (TB) disease. DESIGN AND SETTING: A prospective cohort study conducted in Cape Town, South Africa, from January 2011 to June 2012. TB-exposed newborns requiring isoniazid preventive therapy (IPT) or anti-tuberculosis treatment were followed to 6 months of age. Appropriate tuberculosis treatment referral, maternal and socio-economic determinants were evaluated. The primary outcome, completion of treatment (6 months IPT, 3 months IPT with a negative tuberculin skin test, or 6 months' treatment for disease) was measured at 6 months. Data were collected from folders and care giver interviews. Cox regression was used to determine hazard ratios (HR) for non-completion of treatment. RESULTS: Fifty-six (63% human immunodeficiency virus [HIV] exposed) TB-exposed newborns were included; median gestational age and mean birth weight were respectively 36 weeks and 2242 g. Of the 56 newborns, 44 (79%) were followed to 6 months; 29/44 (66%) completed anti-tuberculosis treatment without study team intervention. Appropriate treatment referral was associated with a lower hazard of non-completion of treatment (unadjusted HR 0.34, 95%CI 0.12-0.93). This relationship was maintained in multivariable adjustment for maternal HIV status and type of care giver (adjusted HR 0.26, 95%CI 0.09-0.77). CONCLUSIONS: Appropriate anti-tuberculosis treatment referral improves completion of treatment in infants.

Home environmental tobacco smoke exposure in Canadian children.

BACKGROUND: Children living with a smoker experience increased environmental tobacco smoke (ETS) exposure, even when the smoker refrains from smoking inside the house, compared to children not living with a smoker. Given the risks of ETS in children, it was hypothesized that households with children are less likely than those without children to experience home ETS exposure. DESIGN: Cross-sectional analysis of the Canadian Community Health Survey 2009-2010 for the association between children aged <12 years in the household and home ETS exposure, using logistic regression and considering household education and income as confounders or effect modifiers. A subgroup analysis was conducted comparing younger child households (at least one child aged <6 years) to older child households (only children aged 6-11 years). RESULTS: Of 66 631 households included, home ETS exposure occurred in 25% of households without children and 22% of households with children. Households with children were less likely than those without children to experience ETS exposure (OR 0.83, 95%CI 0.80-0.87). Effect modification by education and income was observed. No difference was observed in ETS exposure between older child and younger child households (OR 0.98, 95%CI 0.91-1.05). CONCLUSION: Households with children are marginally less likely than households without children to experience home ETS exposure.

Severe infections in HIV-exposed uninfected infants: clinical evidence of immunodeficiency.

We describe the clinical and basic immunological findings of eight HIV-exposed uninfected infants hospitalized with serious infectious morbidity and referred for immunological evaluation. The median age at presentation was 5.5 (1.5-15) months. Infections included Pneumocystis jiroveci pneumonia (three), cytomegalovirus colitis with perforation (one), Pseudomonas sepsis (two), hemorrhagic varicella (one) and Group A streptococcal meningitis and endocarditis (one). Five required intensive care, four for assisted ventilation and one for post-surgical care. Follow-up to 36 months suggested resolution of a transient immunodeficiency in two infants, one of whom had CD4 and the other B-cell depletion. Further studies are indicated in HIV-exposed uninfected infants.