Backgrounds:

Severe acute malnutrition (SAM) is the most extreme form of malnutrition, necessitating hospitalization if the child has severe oedema, infection, or lack of appetite. Mortality continues to remain unacceptably high for hospitalized children, with deaths frequently being attributed to infectious causes.

Methods

Children with SAM aged <60 months were enrolled after admission at two tertiary referral hospitals in Harare, Zimbabwe, and one in Lusaka, Zambia between August 2016 and March 2018. Children had baseline data and anthropometry collected and were assessed by a doctor daily to document diagnoses. Independent risk factors associated with mortality were determined using backwards stepwise Cox regression.

Results:

70 of 745 (9.4%) children died in hospital, mean time to death 9.9 days. Age between 6-23 months [adjusted hazard ratio 6.53, 95%CI 2.24 – 19.02], presence of shock [aHR 8.18, 95%CI 3.79 – 17.65], sepsis [aHR 3.13, 95%CI 1.44 – 6.80], persistent diarrhoea [aHR 2.27, 95%CI 1.18 – 4.37], lack of a household toilet [aHR 4.35, 95%CI 1.65 – 11.47], mid-upper arm circumference [aHR 0.726, 95%CI 0.591 – 0.892] and oedema [aHR 2.22, 95%CI 1.23 – 4.05] were all independent risk factors for in-patient mortality [Figure 1A-E]. HIV was not independently associated with mortality, although there was a significant interaction with shock: children with HIV were 6.21 times more likely to be shocked (95%CI 1.41 – 27.2) than HIV-ve children.

Conclusions/Learning Points:

In this lagest prospective study of children with SAM since WHO guidelies were introduced, sepsis, shock, persistent diarrhoea, and household sanitation are all independent predictors of hospital mortality, showing importance of infection in inpatient deaths, despite universal antibiotic administration. HIV, associated with mortality in several previous studies of SAM, was not independently associated with mortality.

(Ethical approval/informed consent obtained)

Backgrounds:

Despite falling vertical transmission rates globally, children born to mothers living with HIV have higher mortality than children born to mothers without HIV. We tested the hypothesis that antenatal cytomegalovirus (CMV) viraemia is associated with this excess mortality.

Methods

Women from the SHINE trial in rural Zimbabwe (ClinicalTrials.gov registration number NCT01824940) were recruited during pregnancy and infants were followed for 18 months. Antenatal CMV DNA was quantified by real-time PCR at a median gestation of 16-weeks. Cox regression models were used to estimate hazard ratios (HR) for mortality.

Results:

Among 563 women with HIV, 69% had no detectable CMV (group 1), 24% had a CMV viral load below 45 copies/mL (group 2), 7% had a CMV viral load 45 copies/mL or greater (group 3). For each log rise in antenatal CMV viral load, risk of child death by 18 months of age increased by 14%, after adjusting for maternal HIV viral load and CD4 count, gestational age at blood sampling and randomised trial arm (adjusted HR (aHR) 1.14; 95%CI 1.04, 1.25; P=0.007). Risk of mortality of children in group 1 was similar to 3989 children born to mothers without HIV (5.1% vs. 5.0%; aHR 0.99; 95%CI 0.60, 1.61; P=0.96). Risk of mortality of children in group 2 was 1.75-fold higher than children born to mothers without HIV (8.8% vs. 5.0%; aHR 1.75; 95%CI 1.00, 3.07; P=0.05) and in group 3 was 4-fold higher (21.1% vs. 5.0%; aHR 4.14; 95%CI 2.10, 8.16; P<0.001) than children born to mothers without HIV.

Conclusions/Learning Points:

Antenatal CMV viraemia is associated with mortality of HIV-exposed children. Identifying and treating CMV viraemia during pregnancy may provide a new target for improving clinical outcomes amongst children born to mothers with HIV.

Backgrounds:

Low birthweight (LBW) infants, encompassing those born small-for-gestational age (SGA) and/or preterm, have poor health outcomes in developing countries, including increased risk of vaccine failure, infection and death. The role of immune development and inflammation remain poorly understood in LBW infants. This study aimed to ascertain biomarkers of enteropathy, T-cell activation and systemic inflammation associated with LBW.

Methods

Using stored samples from a large cluster-randomised trial in rural Zimbabwe, we measured biomarkers of enteropathy and systemic inflammation using ELISA, and T-cell activation by flow cytometry, comparing: (i) term, appropriate for gestational age (AGA); (ii) preterm, AGA; (iii) term, SGA; and (iv) preterm SGA. We used generalised estimating equations and generalised linear models to explore differences between groups.

Results:

We include data from 1574 infants with visits at 1-3 months of age. Faecal biomarkers of intestinal inflammation were associated with LBW: compared to term AGA infants, neopterin was raised in premature SGA infants (mean 7.54nmol/L vs. 6.67nmol/L; P<0.01); myeloperoxidase was raised in premature AGA infants (8.11 ng/ml vs. 7.95; P=0.01); and alpha-1 antitrypsin was raised in premature AGA infants (13.60ng/ml vs.13.04ng/ml; p=0.03). Compared to term AGA infants, term SGA infants had higher proportions of activated (HLA-DR+) and proliferating (Ki67+) CD4+ T-cells (13.4% vs. 7.9% (P<0.01) and 8.0% vs. 3.2% (P=0.02), respectively). These effects were consistent after adjusting for plausible confounders. There were no differences in C-reactive protein between groups.

Conclusions/Learning Points:

We identified associations between LBW, intestinal inflammation and T-cell activation, with prematurity having greater association with gut inflammation and SGA with lymphocyte activation. Mechanistic studies are needed to determine whether these pathways might be amenable to intervention to improve clinical outcomes among LBW infants.

Backgrounds:

It is unclear whether there are persistent school-age health, growth and developmental disparities between children who are HIV-exposed but uninfected (CHEU) versus HIV-unexposed (CHU). Long-term follow up of a cohort of children recruited to the SHINE trial, who are now aged 7-8 years, offers the opportunity to compare school-age outcomes in a setting where 15% of children are HIV-exposed.

Methods

We measured physical function using handgrip strength, broad jump and the shuttle run test to provide a standardized score. Cognitive function was evaluated using the Kaufman Assessment Battery for Children (KABC-II), with additional tools measuring executive function, literacy, numeracy, fine motor skills and socioemotional function. Growth was assessed by anthropometry, body composition (using bioimpedance analysis) and skinfold thicknesses. A detailed caregiver questionnaire measured demographics, socioeconomic status, nurturing, child discipline, food and water insecurity. Results for CHEU and CHU were compared using generalized estimating equations with an exchangeable working correlation structure to account for clustering.

Results:

328 rural Zimbabwean children (89 CHEU, 239 CHU) were assessed at age 7-8 years. CHEU were 2 months older than CHU (95%CI, 1mo to 3mo, p<0.001), and had marginally reduced height-for-age Z-score, HAZ (-0.22, 95% CI -1.89, 0.01 p=0.06). CHEU had significantly reduced total KABC-II scores (-4.6 marks, 95%CI -6.8, -2.3; p<0.001) and weak evidence for reduced literacy and numeracy scores (-5 marks, 95%CI -11, 1; p=0.10). CHEU had slightly higher socioemotional issues as measured by the strength and difficulties questionnaire (1.4 marks, 95% CI -0.04, 2.86, p=0.06).

Conclusions/Learning Points:

Our results indicate that antenatal HIV exposure has a persistent effect on CHEU growth and neurodevelopment at age 7 years, particularly for cognitive function.