Clinical profile of children under 5 years of age with rotavirus diarrhoea in a hospital setting in Kisangani, DRC, after the introduction of the rotavirus vaccine, a cross-sectional study

For several decades, acute diarrhoea has been a major cause of morbidity and mortality in children under 5 years of age worldwide, especially in resource-limited settings such as sub-Saharan Africa (SSA) [1, 2]. Rotavirus is the main causative agent of acute diarrhoea in children under 5 years of age. It is estimated to be responsible for more than 258 million annual episodes of diarrhoea worldwide, causing up to 128,500 deaths per year in children under five, the majority of whom are in SSA [2, 3]. Regardless of where they live, almost all children are infected with rotavirus at least once before the age of 5 years, despite efforts to improve immunisation, hygiene and access to safe water in many parts of the world [4, 5]. In the Democratic Republic of the Congo, rotavirus infection is a real public health problem and before the introduction of the rotavirus vaccine, the rotavirus mortality rate was estimated to be around 7% of the global under-five mortality rate in DRC [3, 6]. The clinical manifestations of rotavirus infection vary greatly with age and previous exposure to the virus, ranging from asymptomatic infection to severe manifestations of diarrhoea, vomiting and dehydration [5, 7]. The assessment of the severity of rotavirus infection by the Vesikari score has been proposed by several authors and to date it is one of the means of assessing the impact of vaccination on severe disease [8‐10]. Several studies have demonstrated the change in the epidemiological and clinical profile of rotavirus infection in some parts of the world since the introduction of different rotavirus vaccines [9‐11]. Presented as the best way to protect children against rotavirus in several countries [11, 12]; it is since 2019, the DRC, with the support of the World Health Organisation (WHO), has introduced the Rotasiil vaccine into the routine vaccination of children, following the example of other countries in Africa and this vaccine was recommended for all children in the DRC within the framework of the National Expanded Programme on Immunisation [13, 14]. This vaccine, manufactured by the Serum Institute of India, includes five genotypes G1, G2, G3, G4 and G9 that are frequently involved in acute diarrhoea in children in Africa [6, 15]. Although data on the impact of vaccination on the epidemiological and clinical changes of rotavirus infection in children under 5 years of age are available globally and in some SSA countries [7, 10], there are still few data available for the DRC, especially the city of Kisangani. The objective of this study was to describe the clinical characteristics of children with acute rotavirus diarrhoea after the introduction of Rotasiil in the DRC.

A total of 178 children under 5 years of age with acute diarrhoea were admitted to the four hospitals during the study period; 4 (2.2%) children were excluded because no consent was provided and 9 (5%) had no stool sample. Therefore, a total of 165 (92.6%) children under 5 years of age with acute diarrhoea were included in the study over a 6-month period at the research sites. Analysis of stool samples for rotavirus and adenovirus antigenic testing revealed 59 cases of rotavirus, 36% CI₉₅ [27, 45]; 10 cases of adenovirus, 6.1%; 2 cases of rotavirus and adenovirus co-infection, 1.2%; and 94 negative cases, 57%.

The aim of this study was to describe the clinical characteristics of children under 5 years of age with acute diarrhoea after the introduction of rotavirus vaccination of children in Kisangani city hospitals. The hospital prevalence of rotavirus acute diarrhoea in children under 5 years was 36% CI₉₅ [27, 45]. Prior to the introduction of vaccination, Heylen E and Bibi found a prevalence ranging from 18.8 to 37.1% with an average of 29% in children under 5 years of age [17]. However, in Lubumbashi, Sangaji found a high prevalence of 53.9% in children under 5 years of age suffering from acute diarrhoea [18], while in sentinel sites, the prevalence of acute rotavirus diarrhoea in children under 5 years of age was 60% [6]. We believe that the high proportion of unvaccinated children could explain this prevalence. Looking at the socio-demographic characteristics of children with acute rotavirus diarrhoea, we noted that children in the 1–11 month age group were the most affected in 36 out of 59 cases. In Kenya Moendo, observed that the prevalence of rotavirus diarrhoea was high in children aged 17–24 months [10], while in the Sangaji study it was children under 12 months who were most affected [18], although in Rwanda Claudine showed that the risk of acute diarrhoea in general was highest in children aged 12–23 months [19]. The female/male sex ratio was 0.4 in contrast to other studies in Lubumbashi where the sex ratio was about 1. We noted that the proportion of children vaccinated against rotavirus was lower in children with rotavirus than in children without rotavirus (73.9% negative vs 26.1% rotavirus), Zakaret also found a low proportion of 23.9% in his study [20]. This is in contrast to the Moendo study in Kenya where almost all children were vaccinated [10]. This situation could be explained by the low promotion of the rotavirus vaccine in Kisangani and context of the Covid 19 pandemic, despite the fact that this vaccine is recommended for all children according to the National Expanded Programme on Immunisation in the DRC and that this vaccine would contribute to the protection of vaccinated children, as already demonstrated elsewhere by other authors [9, 10, 20, 21]. We noted that the mothers of children with rotavirus diarrhoea had in most cases a secondary level of education (55 cases, i.e. 59.1% negative vs. 38 cases, i.e. 40.9% rotavirus) and were mostly unemployed (64 cases, i.e. 63.4% negative vs. 37 cases, i.e. 36.6% rotavirus); they used tap water in their households (83 cases, i.e. 62.9% negative vs. 49 cases, i.e. 37.1% rotavirus), although the difference was not statistically significant.

Clinically, in a bivariate analysis between the two groups, we observed that there was a statistically significant difference in the mean frequency per day/per admission of diarrhoea (5.9 vs 9.6) and the mean duration in days of rotavirus diarrhoea (3.6 vs 6.1). The same was true of the proportions of vomiting (54.7% vs. 45.3%) and dehydration in their various degrees. In this study, it was noted that the mean clinical severity score of Vesikari was higher in children with rotavirus with a statistically significant difference (9.4 vs 13.1) and by comparing the different categories, we observed that 69% of the rotavirus infection cases were severe contrary to Moendo who reported 28.3% of the severe cases and about 39.6% of the mild cases [10]. Comparing the clinical characteristics of rotavirus infection between vaccinated and unvaccinated children, we observed a higher score in unvaccinated children than in vaccinated children with a statistically significant difference (12.7 vs 10.7 p-value 0.024). Studies have shown that vaccination protects children against severe forms of rotavirus infection [15, 22] and we believe this would explain the difference in Vesikari score between vaccinated and unvaccinated children. In Lebanon, Zakaret reported a mean score of 10.9 ± 1.7, closer to Shim Dong who found a mean score of 10 ± 3.3 in a study evaluating the predictive value of this sore between viral and bacterial causes of paediatric gastroenteritis [20, 22, 23]. Our observation could be explained by the fact that these are severe cases that frequent hospital facilities more often, although some authors think that the performance of this score should be improved [24]. Cough as a respiratory symptom was observed between the two groups with no statistically significant difference (64.3% vs 35.7% p value 0.4). Apart from clinical intestinal manifestations, cough or convulsions have been described during rotavirus infection in children by other authors [25], whereas in this study cough was observed without significant difference from children without rotavirus infection. In a multiple linear regression analysis, we looked for factors that could influence the Vesikari score in children with rotavirus infection (see Table 5); the results of this analysis showed that the 24–59 month age range of children with rotavirus infection was associated with a clinical severity score of approximately minus 4 while it was associated with minus 3 for mothers unemployed with a statistically significant difference. The high incidence of rotavirus infection in children under 2 years of age has been reported in several studies and this is the period when primary rotavirus infection is usually observed [18, 26, 27], this would explain the frequency of severe clinical forms compared to children over 2 years of age and we don’t have an explanation for the relationship with unemployed mothers. Finally, we believe that the interpretation of the results of this study must be made taking into account the limitations of the sample size and the rotavirus diagnostic method we used. However, this study provides important and reliable information in these four hospitals where rotavirus screening is not part of routine hospital care.

This study showed that rotavirus infection in the hospital setting after the introduction of the rotavirus vaccine in Kisangani is characterised by severe clinical manifestations of high frequency watery diarrhoea accompanied by severe dehydration and affects in the majority of cases unvaccinated children in the age group of 1 to 11 months. These results highlight the burden of rotavirus-associated diarrhoea in children under 5 years of age in Kisangani, Democratic Republic of the Congo. Surveillance studies are needed to identify risk factors associated with rotavirus infection.