Prevalence of rotavirus infection among children with acute diarrhoea after rotavirus vaccine introduction in Kenya, a hospital cross-sectional study

Diarrhoeal diseases remain a leading cause of morbidity and mortality among children in the world, more so in developing countries with rotavirus infection being the most common cause of severe, acute diarrhoea [1]. Globally, it was estimated to cause 527,000 deaths in the year 2008 among children below 5 years of age, [2] this has since reduced to 215,000 in the year 2015 in the same age group [1]. More than 80% of these deaths continue to occur in South Asia and Sub-Saharan Africa [3]. Early complimentary feeding, nutritional status, dehydration and age less than 2 years are important risk factors associated with rotavirus diarrhoea. [4‐6]. The peak infection age range with rotavirus is 3–24 months, the highest rate being between the ages of 6–11 months [7].The reported prevalence of rotavirus diarrhoea among children below 5 years hospitalized with diarrhoea from global surveillance networks and hospital based studies varies greatly ranging from 6 to 56% [8, 9]. In Kenya, the rotavirus prevalence was 40% among children below 5 years of age hospitalized for treatment of acute gastroenteritis [9] in the period 2006 to 2008. The clinical presentation of rotavirus illness ranges from mild, watery diarrhoea to severe diarrhoea with vomiting and fever that can result in dehydration with shock, electrolyte imbalance, and even death [10]. The Vesikari clinical severity scoring system (VCSSS) has been used in clinical trials in assessing rotavirus vaccine efficacy and effectiveness as a tool for defining the primary end point, which is severe rotavirus gastroenteritis [11]. It has also been used in clinical studies as a measure of acute gastroenteritis severity [12‐14]. The parameters and categories of this severity scale are shown in Table 1.

Vaccination has been shown to be the best way to prevent severe rotavirus disease [15, 16]. Currently available rotavirus vaccines have been shown to be effective in reducing the rotavirus disease burden with observed efficacy rates of about 65% in developing countries in Africa [15]. Rotavirus vaccines have been included in most national immunization programs in the world to date. However, it was not until July 2014, that the rotavirus vaccine was incorporated into the Kenya Expanded Program of Immunization (KEPI) [17].

The rotavirus vaccines available in Kenya are Rotarix®, manufactured by GlaxoSmithKline, administered orally in a 2-dose schedule, currently issued countrywide under the KEPI and RotaTeq®, manufactured by Merck & Co. Inc. and administered orally in a 3-dose schedule, mostly in private facilities.

In South Africa, a decline was reported in rotavirus prevalence and hospitalisations among children below 5 years after introduction of rotavirus vaccine in 2009 [18]. Similarly, a Rotavirus Sentinel Surveillance performance feedback report by WHO in 2016 reported a significant decline of rotavirus infection among countries in East and Southern Africa from 44% in 2010 to 25% in 2015, after introduction of rotavirus vaccine from the year 2014 [19]. Though much is known about the morbidity and mortality of rotavirus diarrhoea before introduction of rotavirus vaccine, there has not been a study to determine the change in the clinical profile of children being treated with acute diarrhoea after the introduction of the rotavirus vaccine in Kenyatta National Hospital. Therefore, this study aimed to determine the prevalence of rotavirus diarrhoea and severity of acute diarrhoea among children aged 3–24 months at Kenyatta National Hospital after rotavirus vaccine introduction in Kenya and also to determine the rotavirus vaccination status among the children.

This study used data from survey conducted from August 2016 to April 2017 during the paediatrics residency period of the lead author. The study was conducted in the paediatric emergency unit and wards of Kenyatta National Hospital, Kenya’s largest public teaching and referral hospital; situated in the capital city, Nairobi. The hospital serves the low and middle-income population from Nairobi and its environs as well as referrals from other hospitals in the country and the greater Eastern Africa region.

In this study, the observed prevalence of rotavirus diarrhoea was 14.5% which was lower than what was observed by Osano in 2008 at 38.2% [10] and Karanja in 2009 at 39.5% [21]. These studies were conducted before rotavirus vaccine was introduced into the Kenya National Immunization Program (KEPI). A Rotavirus Sentinel Surveillance performance feedback report by World Health Organization reported a decline of rotavirus associated diarrhoea among countries in East and Southern Africa from 40% in 2014 to 25% in 2015, after introduction of rotavirus vaccine from the year 2014 [19] showing a reduction in the burden of rotavirus with the introduction of the rotavirus vaccine. Our result however, may not be a true reflection of rotavirus burden in the community because of lack of data on the prevalence of rotavirus diarrhoea before introduction of vaccine in similar setting. Nonetheless, in a community study in Nicaragua they observed a 40% lower incidence rate of watery diarrhoea episodes suggestive of a reduction in rotavirus infection in the vaccine period as compared with the pre-vaccine period. This reduction may be attributable to herd immunity that results from an overall protective effect of the immunization program on both immunized and non-immunized children [22]. Similarly, this study noted that the children who presented with acute diarrhoea and were not vaccinated did not have rotavirus positive stools; it is postulated that it may be as a result of the overall protective effect of the rotavirus vaccine on both the immunized and non-immunized children [22].

The Vesikari Clinical Severity scoring system for gastroenteritis used in this study, elicited the distribution of severe diarrhoea as similar among those children who were rotavirus positive compared to those who were negative. Comparability of severity of rotavirus diarrhoea with other studies is difficult due to varied classification systems, some studies have described severity of rotavirus associated diarrhoea using the vesikari clinical scoring system, while others have described severity using the hydration status [21, 23] or the need for hospitalization as a marker for severity of illness [10]. Gatinu’s study in 2007 reported a 47.9% prevalence of severe dehydration as a marker of severe rotavirus disease [23]. The hospitalization rate in our study was observed to be above 60%. This could be attributable to associated comorbidities as a majority of the children admitted had associated comorbidities that necessitated hospitalization, the commonest being pneumonia. Severe dehydration commonly presents as fast and deep acidotic breathing due to electrolyte imbalances and metabolic acidosis as a result of fluid loss and may be misdiagnosed as pneumonia due to similar presentation [24]. However, other studies demonstrate concurrent pneumonia infection in children presenting with diarrhoea [24]. The results indicate that the risk of rotavirus infection increased with age of the child which is quite contrary to most studies which report increased risk of rotavirus infection in infants [5, 10, 25]. According to the World Health Organization scientific working group, most cases of rotavirus infections occur in children between 6 and 24 months with a peak incidence at 9 to 12 months [7]. It is postulated that younger children tend to be at an increased risk of developing severe dehydration due to their small body size, as they lose a greater portion of their total fluid volume during the illness [26].

We found that the exclusively breastfed children were about one and a half more times likely to have rotavirus diarrhoea, though it was not statistically significant. It is thought that breastfeeding reduces gastrointestinal infections as breast milk contains secretory antibodies such as secretory IgA, immune cells and other defense factors such as lactoferrin, oligosaccharides and human milk glycans that protect the intestinal epithelium against pathogens [27]. However, the specific role of breastfeeding in the prevention of rotavirus diarrhoea has not been well established but it is generally considered to at least reduce the severity of the disease [28]. There have been conflicting results as to whether breastfeeding is protective or not. Naficy et al. found a lower incidence of rotavirus diarrhoea in infants that received breast milk [29] and others have shown evidence that breastfeeding offers protection against only severe rotavirus infections [28]. On the contrary, Gurwith and Totterdell found no evidence of protection against clinical rotavirus disease by maternal milk [30, 31]. The role of exclusive breastfeeding needs to be explored further in a study designed to establish whether exclusive breastfeeding protects against rotavirus diarrhoea. Malnourished children were found to have less risk of developing severe diarrhoea in our study. Interestingly, it is postulated that malnutrition is associated with protection from rotavirus diarrhoea for various reasons, among these, the possibility of shortening of villi in malnourished infants that may inhibit rotavirus entry and replication [32]. However, some studies show that nutritional status has no significant correlation with severity of rotavirus diarrhoea [33].

The rotavirus vaccination status among the children in this study was found to be at 96.7% against rotavirus. However, vaccination of the child was verified from the maternal and child booklet and/or word of mouth from the parent. For the parents who did not recall the names of the vaccines received, they described the vaccine by route of administration and the age of the child when they received the particular vaccine. The latter method of vaccine verification is unlikely to be as accurate as the written and dated records with a high likelihood of overreporting [34, 35]. The Kenya Demographic Health Survey 2014/2015 reported that 79% of children had received all the basic vaccinations [36]. There was no specific rotavirus vaccine coverage report in the Kenya Demographic Health Survey by 2014 as the vaccine had just been rolled out for use in the country in mid-2014 however, according to WHO/UNICEF the rotavirus vaccine achieved only 38% coverage in 50% of the national target Kenyan population in 2014 and 66% in 2015 [37]. Contributing factors associated with the high rotavirus vaccination status were not explored in this study.

Interpretation of results from this study should be done with caution because most results were not statistically significant which could be due to the sample size of the study.

The burden of rotavirus associated diarrhoea among children aged 3–24 months at Kenyatta National Hospital in 2017 was observed to be 14.5%. The results provide the level of burden of rotavirus infection but are not able to conclude on the attributable effect of introduction of the rotavirus vaccine. Though not statistically significant, there seems to be some interesting pattern for both rotavirus infection and severity of diarrhea with child/parent characteristics. The rotavirus vaccination status was 96.7% among the children. Community based surveillance studies are needed to establish the prevalence of rotavirus at a population level and identify associated risk factors for infection.