Rotavirus is the most common cause of childhood morbidity worldwide, accounting for 453,000 deaths children <5 years of age.1 Approximately half of these deaths occur in Africa. In 2009, the World Health Organization (WHO) recommended 2 new rotavirus vaccines (Rotarix; GSK Biologicals, Rixensart, Belgium and RotaTeq; Merck & Co, West Point, PA) for use in all children worldwide.2 These vaccines have been tested in various regions of the world and demonstrated heterogeneity in efficacy that correlates with the socioeconomic condition of the population, ranging from 76% to 98% in high- and middle-income countries to ~39–48% in low-income settings.3–7 However, even with the lower efficacy in the poorer settings, benefits of vaccine were substantial in these settings in terms of absolute reduction in severe rotavirus disease because of the very high baseline burden.6
Rotavirus vaccines have been widely adopted in many middle- and high-income countries.8 In these countries, in addition to reductions in rotavirus disease in vaccinated children,9,10 alterations in the timing and seasonal pattern of the annual rotavirus epidemic11 and indirect benefits to unvaccinated members of the community12,13 are some of the notable findings that have been documented after the introduction of vaccine. While changes in strain patterns have also been seen in some countries after vaccine introduction,14–16 similar secular changes have also been seen in countries without vaccination in Expanded Programme on Immunization and thus these changes might represent natural variation in rotavirus strain prevalence. Having established baseline surveillance for rotavirus disease and strains before the introduction of vaccine was crucial for the successful monitoring of vaccine impact in these early adopter countries and gaining a better understanding of the epidemiology of rotavirus.8
In 2011, Sudan became the first low-income country in Africa to introduce a rotavirus vaccine in its national immunization program. The decision to introduce rotavirus vaccine in Sudan was based in part on the high burden of rotavirus disease in the country, which was demonstrated through the establishment of surveillance for rotavirus diarrhea at sentinel hospitals nationwide and is described in this article. Against this baseline pattern, the impact of rotavirus vaccine can be robustly measured and changes in disease ecology after the introduction of vaccine soundly interpreted.
MATERIALS AND METHODS
Active Rotavirus Surveillance
Sudan is a low-income country in eastern North Africa, with an annual birth cohort of 1.4 million children. In 2007, the Ministry of Health of Sudan joined a regional rotavirus surveillance network for the Eastern Mediterranean Region of the WHO. After pilot surveillance, a stable system consisting of 8 pediatric hospitals in Sudan has been sustained since June 1, 2009. All hospitals in the network followed a standard WHO surveillance protocol for enrolling all children <5 years of age who are hospitalized for acute gastroenteritis, which is defined as acute onset of 3 or more loose stools or 2 or more episodes of vomiting in a 24-hour period, which were not explained by another diagnosis.17 Acute onset was defined as symptom(s) onset of <14 days before presentation.
Regional epidemiologists were assigned as focal points for the surveillance program and were responsible for training hospital staff, collating surveillance data, providing feedback to hospital staff and monitoring surveillance indicators. Physicians at each hospital were routinely reminded to enroll all children meeting the surveillance case definition. Surveillance coordinators at each hospital ensured enrollment of all potential participants by reviewing daily admission logs and by frequent discussions with physicians. Surveillance coordinators collected demographic and clinical information on a case report form and obtained a stool sample (>5 mL) as soon as feasible during the hospitalization. The stool samples were refrigerated immediately and shipped weekly to a national laboratory for rotavirus testing using a commercially available enzyme-linked immunosorbent assay (IDEIA, Rotavirus Test, Dako Diagnostic). At the national laboratory, rotavirus-positive samples were stored at −70°C for genotyping at a future date.
Estimates of National Rotavirus Disease Burden
We applied the method used by the WHO to estimate nationwide burden of mortality of deaths, hospitalizations and outpatient visits related to rotavirus in Sudan.18,19 To calculate the number of rotavirus deaths occurring annually in Sudan among children <5 years of age, we multiplied the prevalence of rotavirus hospitalizations among children <5 years of age with gastroenteritis (36%) from the current surveillance by the number of diarrhea deaths in that age group as estimated by the WHO.20 For hospitalizations, we multiplied the prevalence of rotavirus hospitalizations by the number of annual hospitalizations related to diarrhea and dehydration among children <5 of age in Sudan during 2010, as recorded by the Ministry of Health in the Annual Health Statistical report. Prevalence of rotavirus is lower for milder disease compared with severe disease (ie, hospitalizations). Thus, for outpatient rotavirus events, we multiplied the number of outpatient events related to diarrhea from this report by the estimated proportion of rotavirus prevalence among milder cases of diarrhea in low-income countries (19%).18 Subsequently, we calculated the cumulative risk that a child would experience these events (death, hospitalization or outpatient visit) before reaching the age of 5. Here, we assumed that the number of these events in each year among a group of children <5 years of age would approximate the number of events occurring in a single birth cohort that was followed to age 5. Cumulative risks were expressed as the ratio of 1 to the quotient of the annual birth cohort (n = ~1,400,000) and the number of respective events among children <5 years of age.
All data were collected and stored in an Access (Microsoft, Redmond, WA) database and analyzed using Excel (Microsoft, Redmond, WA) and SAS 9.1 (Cary, NC). To assess seasonal trends, we plotted the number of monthly visits for all-cause and rotavirus-associated diarrhea among children <5 years of age during the 2-year surveillance period. The severity of acute gastroenteritis was categorized based on the 20 point Vesikari scale and differences compared between children testing positive versus negative for rotavirus. Proportions were compared using the χ2 test for unequal odds, mean values were compared using the unpaired 2-sided t-test and median values were compared using the Wilcoxon rank-sum test.
Active Rotavirus Surveillance
From June 2009 to May 2011, we enrolled a total of 10,910 children with acute gastroenteritis in the rotavirus surveillance system at the 8 hospitals in Sudan. Of these children, 3957 (36%) tested positive for rotavirus, with prevalence being equal during both surveillance years (Table 1). Prevalence of rotavirus detection ranged from 25% to 48% between the 8 sites.
Nearly all of the rotavirus hospitalizations (91%) were among children <2 years of age, with 61% of the events occurring during the first year of life (Fig. 1). Only 1% (26/3985) of the rotavirus hospitalizations occurred among children <2 months of age and another 8% (300/3985) occurred between 2 and 4 months of life, before the age when rotavirus vaccination is completed.
Rotavirus hospitalizations occurred year-round with detection rates being >10% during all months of the year. Two peaks in rotavirus prevalence were observed in Sudan during March to May and November to December (Fig. 2). No difference in seasonal pattern was observed between the 8 surveillance sites.
Of all children hospitalized with gastroenteritis, 59% and 60% of the rotavirus-positive and rotavirus-negative cases were male, respectively. Fever was reported among 80% of the children and over half had 2–3 or more episodes of vomiting and diarrhea per day (Table 2). Among children with rotavirus gastroenteritis, moderate to severe dehydration was reported in ~98% of the children, with 96% of them scoring >11 on the Vesikari severity scoring scale; 88% required intravenous hydration. Duration of illness was >3 days in 32% of the children, with hospitalization lasting for a median duration of 1 day. More children with rotavirus-negative gastroenteritis (38%) had illness lasting longer than 3 days compared with those with rotavirus-positive gastroenteritis (32%; P < 0.001). Similarly, the proportion of children with Vesikari score ≥15 was higher among those with rotavirus-negative gastroenteritis (34%) versus those with rotavirus-positive gastroenteritis (27%; P < 0.001).
National Estimates of Rotavirus Deaths, Hospitalizations and Outpatient Visits in Sudan
We applied the prevalence of rotavirus among children hospitalized for diarrhea (36%) to the national estimates of diarrhea deaths (n = 25,606) and diarrhea and dehydration hospitalizations (n = 63,220) to estimate that 9200 deaths and 22,800 hospitalizations related to rotavirus occur each year among children <5 years of age in Sudan. Applying the 19% prevalence of rotavirus among children with outpatient diarrhea in developing countries, we estimate that 55,400 of the 291,318 annual visits for diarrhea in Sudan are related to rotavirus. Thus, we estimate that before the introduction of rotavirus vaccine, 1 in 152 Sudanese children died, 1 in 61 was hospitalized and 1 in 25 received outpatient care for rotavirus infection before reaching 5 years of age.
The establishment of a rotavirus surveillance network in Sudan led to the confirmation of rotavirus infection as a common cause of severe childhood diarrhea and allowed the National Immunization Technical Advisory Group to make an evidence-based recommendation to Ministry officials for introducing rotavirus vaccine. Before the implementation of a rotavirus vaccine program, rotavirus infections in Sudan represented some 36% of the severe diarrhea cases among children <5 years of age. Applying these surveillance data to the number of diarrhea deaths and hospitalizations in Sudan, we estimate that the risk of death for a child <5 years of age in Sudan was 1 in 151 and the risk of hospitalization was 1 in 61, amounting to nearly 9200 childhood deaths and 22,800 childhood hospitalizations per year. This first description of the burden of rotavirus disease in Sudan strongly supports the country’s decision to introduce the vaccine for control of rotavirus disease and lays the framework for future monitoring of vaccine impact.
Rotavirus seasonality in Sudan is similar to other low-income countries,21 in that symptomatic infections are prevalent year-round, with some increase during November to December and March to May. Most of the surveillance sites in Sudan are in a desert climate, where temperatures are lower during November to March compared with April to September, and periods of heavy rainstorms and higher relative humidity may occur from July to September. The increase in prevalence of rotavirus during the drier, cooler months with lower relative humidity is consistent with some previous observations from other settings providing some support to the hypotheses that these climate conditions might favor the increase in aerial transport of dried, contaminated fecal material22–26; however, this observation has not been consistently noted in all locations, thus raising skepticism of a single unifying explanation for the seasonality of rotavirus disease.
Any observed changes in seasonal pattern of rotavirus disease after the introduction of rotavirus vaccine after the recent rollout of vaccine in Sudan might provide further insight on this perplexing issue. Modelers have suspected that birth rates might influence the seasonal patterns of rotavirus disease, in that countries with lower birth rates have more seasonal disease and those with higher birth rates have year-round disease, in part because of the more rapid accumulation of susceptible infants through birth in the latter setting sustain year-round transmission.27 In settings with higher birth rates, vaccination, by virtue of reducing the effective rate of accumulation of susceptible infants through, could lead to the appearance of strongly seasonal disease during the first few years of vaccination if higher birth rates were indeed the primary driver of sustained year-round transmission in prevaccine years.
Sudan is the first low-income country in Africa to introduce a rotavirus vaccine. The heterogeneity in the efficacy of rotavirus vaccines in the poorer settings and the programmatic and logistical challenges that could emerge under unpredictable real-world conditions (eg, cold-chain problems, delays in timeliness vaccination) necessitate close monitoring of rotavirus disease after the introduction of vaccine.28 Sudan’s well-established and robust rotavirus surveillance system bodes well for evaluation of the performance and impact of rotavirus vaccine in this low-income setting. Through this surveillance, a reduction in the absolute burden of rotavirus disease would first be noticed among infants during the first year or 2 after attaining high coverage and then in older children during subsequent years.28,29 However, given that vaccine efficacy is expected to be ~50% in Sudan on the basis of clinical trial efficacy in similar setting, even with high coverage of 80–90% and the prevaccine rotavirus prevalence of 36%, a substantial amount of cases would still be detected at these surveillance sites. To allay concerns of vaccine failure as a cause for ongoing detection of rotavirus cases, it may be prudent to leverage these surveillance sites and apply robust epidemiologic methods such as the case-control approach to directly assess vaccine effectiveness against rotavirus hospitalizations.28,29 The establishment of a functional rotavirus surveillance network in Sudan was the necessary first step to conduct such postlicensure assessments that would ultimately ensure optimal effectiveness in target populations with high mortality.
Some limitations of this surveillance should be considered. First, rotavirus disease is known to have secular variation—prevalence of rotavirus disease changes over time depending on clinical management and healthcare seeking patterns as well as prevalence of other enteric pathogens causing diarrhea. Although the consistency of the burden estimates with other similar settings is reassuring, our burden estimates only reflect 2 years of surveillance. Second, the estimates of diarrhea deaths are not based on direct rotavirus testing which is impractical for establishing etiology in diarrhea deaths, but is based on the assumption that the proportion of severe, dehydrating diarrhea is similar to the proportion of children dying from diarrhea. Lastly, data on children enrolled through the sentinel surveillance system may not be generalizable to the entire Sudanese population, although the fact that we had 8 sentinel sites scattered throughout the country is reassuring.
In summary, the high burden of rotavirus disease in Sudan indicates that recent introduction of rotavirus vaccine in Sudan should prevent a substantial proportion of diarrhea fatalities and hospitalizations in the country. Continued surveillance of rotavirus diarrhea using the same systematic surveillance methodology that was applied before the introduction of vaccine will allow for assessment of the benefits of vaccination within a few years of vaccination and identify potential areas of concern that warrant future attention. Ultimately, these data would allow policy makers to assess the value of vaccination for controlling severe and fatal rotavirus disease.
1. Tate JE, Burton AH, Boschi-Pinto C, et al. 2008 estimate of worldwide rotavirus-associated mortality in children younger than 5 years before the introduction of universal rotavirus vaccination programmes: a systematic review and meta-analysis. Lancet Infect Dis. 2012;12:136–141
2. WHO. . Rotavirus vaccines: an update. Wkly Epidemiol Rec. 2009;84:533–540
3. Zaman K, Dang DA, Victor JC, et al. Efficacy of pentavalent rotavirus vaccine against severe rotavirus gastroenteritis in infants in developing countries in Asia: a randomised, double-blind, placebo-controlled trial. Lancet. 2010;376:615–623
4. Vesikari T, Matson DO, Dennehy P, et al.Rotavirus Efficacy and Safety Trial (REST) Study Team. Safety and efficacy of a pentavalent human-bovine (WC3) reassortant rotavirus vaccine. N Engl J Med. 2006;354:23–33
5. Ruiz-Palacios GM, Pérez-Schael I, Velázquez FR, et al.Human Rotavirus Vaccine Study Group. Safety and efficacy of an attenuated vaccine against severe rotavirus gastroenteritis. N Engl J Med. 2006;354:11–22
6. Madhi SA, Cunliffe NA, Steele D, et al. Effect of human rotavirus vaccine on severe diarrhea in African infants. N Engl J Med. 2010;362:289–298
7. Armah GE, Sow SO, Breiman RF, et al. Efficacy of pentavalent rotavirus vaccine against severe rotavirus gastroenteritis in infants in developing countries in sub-Saharan Africa: a randomised, double-blind, placebo-controlled trial. Lancet. 2010;376:606–614
8. Patel MM, Steele D, Gentsch JR, et al. Real-world impact of rotavirus vaccination. Pediatr Infect Dis J. 2011;30(1 suppl):S1–S5
9. Tate J, Mutuc JD, Panozzo CA, et al. Sustained decline in rotavirus detections in the United States following introduction of rotavirus vaccine in 2006. Pediatr Infect Dis. 2011;30:S30–S34
10. Cortes JE, Curns AT, Tate JE, et al. Rotavirus vaccine and health care utilization for diarrhea in U.S. children. N Engl J Med. 2011;365:1108–1117
11. Curns AT, Panozzo CA, Tate JE, et al. Remarkable postvaccination spatiotemporal changes in United States rotavirus activity. Pediatr Infect Dis J. 2011;30(1 suppl):S54–S55
12. Lopman BA, Curns AT, Yen C, et al. Infant rotavirus vaccination may provide indirect protection to older children and adults in the United States. J Infect Dis. 2011;204:980–986
13. Curns AT, Steiner CA, Barrett M, et al. Reduction in acute gastroenteritis hospitalizations among US children after introduction of rotavirus vaccine: analysis of hospital discharge data from 18 US states. J Infect Dis. 2010;201:1617–1624
14. Kirkwood CD, Boniface K, Barnes GL, et al. Distribution of rotavirus genotypes after introduction of rotavirus vaccines, Rotarix® and RotaTeq®, into the National Immunization Program of Australia. Pediatr Infect Dis J. 2011;30(1 suppl):S48–S53
15. Hull JJ, Teel EN, Kerin TK, et al. United States rotavirus strain surveillance from 2005 to 2008: genotype prevalence before and after vaccine introduction. Pediatr Infect Dis J. 2011;30(1 suppl):S42–S47
16. Carvalho-Costa FA, Volotao E, Santos de Assis RM, et al. Laboratory-based rotavirus surveillance during the introduction of a vaccination program, Brazil, 2005–2009 Pediatr Infect Dis. 2011;30:S35–S41
17. WHO. . Generic protocols for (i) hospital-based surveillance to estimate the burden of rotavirus gastroenteritis in children and (ii) a community-based survey on utilization of health care services for gastroenteritis in children. 2002 Document WHO/V & B/0215 Geneva;:1–67
18. Parashar UD, Hummelman EG, Bresee JS, et al. Global illness and deaths caused by rotavirus disease in children. Emerging Infect Dis. 2003;9:565–572
19. Parashar UD, Burton A, Lanata C, et al. Global mortality associated with rotavirus disease among children in 2004. J Infect Dis. 2009;200(suppl 1):S9–S15
21. Cunliffe NA, Kilgore PE, Bresee JS, et al. Epidemiology of rotavirus diarrhoea in Africa: a review to assess the need for rotavirus immunization. Bull World Health Organ. 1998;76:525–537
22. Levy K, Hubbard AE, Eisenberg JN. Seasonality of rotavirus disease in the tropics: a systematic review and meta-analysis. Int J Epidemiol. 2009;38:1487–1496
23. Haffejee IE. The epidemiology of rotavirus infections: a global perspective. J Pediatr Gastroenterol Nutr. 1995;20:275–286
24. Brandt CD, Kim HW, Rodriguez WJ, et al. Rotavirus gastroenteritis and weather. J Clin Microbiol. 1982;16:478–482
25. Moe K, Shirley JA. The effects of relative humidity and temperature on the survival of human rotavirus in faeces. Arch Virol. 1982;72:179–186
26. Ansari SA, Springthorpe VS, Sattar SA. Survival and vehicular spread of human rotaviruses: possible relation to seasonality of outbreaks. Rev Infect Dis. 1991;13:448–461
27. Pitzer VE, Viboud C, Lopman BA, et al. Influence of birth rates and transmission rates on the global seasonality of rotavirus incidence. J R Soc Interface. 2011;8:1584–1593
28. Patel MM, Parashar UD. Assessing the effectiveness and public health impact of rotavirus vaccines after introduction in immunization programs. J Infect Dis. 2009;200(suppl 1):S291–S299
29. WHO. . Generic protocol for monitoring impact of rotavirus vaccination on rotavirus disease burden and viral strains. 2009 Document WHO/IVB/0816 Geneva;:1–73