Etiology, Presentation, and Risk Factors for Diarrheal Syndromes in 3 Sub-Saharan African Countries After the Introduction of Rotavirus Vaccines From the Vaccine Impact on Diarrhea in Africa (VIDA) Study

Abstract

Background: Diarrheal disease is heterogeneous, including watery diarrhea (WD) and dysentery, some cases of which become persistent diarrhea (PD). Changes in risk over time necessitate updated knowledge of these syndromes in sub-Saharan Africa.

Methods: The Vaccine Impact on Diarrhea in Africa (VIDA) study was an age-stratified, case-control study of moderate-to-severe diarrhea among children <5 years old in The Gambia, Mali, and Kenya (2015-2018). We analyzed cases with follow-up of about 60 days after enrollment to detect PD (lasting ≥14 days), examined the features of WD and dysentery, and examined determinants for progression to and sequelae from PD. Data were compared with those from the Global Enteric Multicenter Study (GEMS) to detect temporal changes. Etiology was assessed from stool samples using pathogen attributable fractions (AFs), and predictors were assessed using χ2 tests or multivariate regression, where appropriate.

Results: Among 4606 children with moderate-to-severe diarrhea, 3895 (84.6%) had WD and 711 (15.4%) had dysentery. PD was more frequent among infants (11.3%) than in children 12-23 months (9.9%) or 24-59 months (7.3%), P = .001 and higher in Kenya (15.5%) than in The Gambia (9.3%) or Mali (4.3%), P < .001; the frequencies were similar among children with WD (9.7%) and those with dysentery (9.4%). Compared to children not treated with antibiotics, those who received antibiotics had a lower frequency of PD overall (7.4% vs 10.1%, P = .01), and particularly among those with WD (6.3% vs 10.0%; P = .01) but not among children with dysentery (8.5% vs 11.0%; P = .27). For those with watery PD, Cryptosporidium and norovirus had the highest AFs among infants (0.16 and 0.12, respectively), while Shigella had the highest AF (0.25) in older children. The odds of PD decreased significantly over time in Mali and Kenya while increasing significantly in The Gambia.

Conclusions: The burden of PD endures in sub-Saharan Africa, with nearly 10% of episodes of WD and dysentery becoming persistent.

Keywords: diarrhea; dysentery; global; infection; persistent.

Figure 1.

Clinical features among children with bloody or watery syndromes of moderate-to-severe diarrhea (MSD), based on findings at enrollment from medical history and physical examination. The level of severity was assessed for each syndrome at physical examination using 2 severity scores: the World Health Organization (WHO) dehydration assessment and the modified Vesikari score. ***P < .001 (χ² test).

Figure 2.

Attributable fractions and 95% confidence intervals among pathogens significantly associated with moderate-to-severe diarrhea among children residing at 3 sites in sub-Saharan Africa, by diarrhea syndrome, in the Vaccine Impact on Diarrhea in Africa (VIDA) study and the Global Enteric Multicenter Study (GEMS). Pathogen detection was performed by means of quantitative polymerase chain reaction using a customized TaqMan Array Card. A, B, and C display results for the 0–11-, 12–23-, and 24–59-month age strata, respectively. Abbreviations: aEPEC, atypical enteropathogenic Escherichia coli; B. fragilis, Bacteroides fragilis; C. difficile, Clostridioides difficile; E. bieneusi, Enterocytozoon bieneusi; EAEC, enteroaggregative E. coli; EIEC, enteroinvasive E. coli; H. pylori, Helicobacter pylori; LT-ETEC, heat-labile enterotoxigenic E. coli; ST-ETEC, heat-stable enterotoxigenic E. coli; tEPEC, typical enteropathogenic E. coli.

Figure 3.

Pathogen-specific odds that an episode of moderate-to-severe diarrhea will become persistent (duration ≥14 days) among children from 3 sites in sub-Saharan Africa participating in the Vaccine Impact on Diarrhea in Africa (VIDA) study. A multivariate logistic regression model was used, combining watery and bloody diarrheal episodes, with adjustment for child age, height-for-age z score (HAZ) at enrollment, and study site. Results of pathogen × site interactions are shown in Supplementary Figure 2. Abbreviations: B. fragilis, Bacteroides fragilis; C. difficile, Clostridioides difficile; EIEC, enteroinvasive Escherichia coli; H. pylori, Helicobacter pylori; LT-ETEC, heat-labile enterotoxigenic E. coli; ST-ETEC, heat-stable enterotoxigenic E. coli; tEPEC, typical enteropathogenic E. coli.