|Home | About | Journals | Submit | Contact Us | Français|
Nicaragua was the first developing nation to implement universal infant rotavirus immunization with the pentavalent rotavirus vaccine (RV5). Initial studies of vaccine effectiveness in Nicaragua and other developing nations have focused on the prevention of hospitalizations and severe rotavirus diarrhea. However, rotavirus diarrhea is more commonly treated in the primary care setting, with only 1–3% of rotavirus cases receiving hospital care. We measured the prevalence of rotavirus infection in primary care clinics in León, Nicaragua, after introduction of the immunization program. In the post-vaccine period, 3.5% (95% confidence interval = 1.9–5.8) of children seeking care for diarrhea tested positive for rotavirus. A high diversity of rotavirus genotypes was encountered among the few positive samples. In conclusion, rotavirus was an uncommon cause of childhood diarrhea in this primary care setting after implementation of a rotavirus immunization program.
Rotavirus causes an estimated 114 million cases of diarrhea in the world annually.1 Approximately 1–3% of rotavirus cases receive care in the hospital, whereas 15–20% of cases receive care in the primary care setting.1 Less severe cases of diarrhea, which are more likely to be treated in primary care clinics, are responsible for substantial morbidity.2–4
To reduce the burden of rotavirus diarrhea, a growing number of nations are adding the rotavirus vaccine to their national immunization schedules. In clinical trials conducted primarily in the United States and Europe, the pentavalent rotavirus vaccine (RV5) reduced clinic visits for rotavirus diarrhea by 86%.5 Subsequent clinical trials and effectiveness studies in developing nations show less robust protection; these studies have focused on the prevention of severe disease and hospital visits.6–9 Less is known about the vaccine's impact on the primary care setting in the developing world.
In October of 2006, Nicaragua became the first developing nation to initiate universal infant rotavirus immunization with RV5. Infants are offered the vaccine at 2, 4, and 6 months of age as part of Nicaragua's Expanded Program on Immunization (EPI). Before the rotavirus immunization program in Nicaragua, rotavirus was isolated among 28.4% of children seeking care for diarrhea in a combination of primary and secondary healthcare facilities.10 After the immunization program, a study based in Nicaraguan hospitals found a vaccine effectiveness of 46% against hospitalization for rotavirus diarrhea.6 Also, Ministry of Health (MOH) data from health facilities show a 15% decrease in diarrhea visits among infants during the rotavirus season.11
The aim of this study was to measure the prevalence of rotavirus infection in the primary care setting in León, Nicaragua, after RV5 introduction. Secondary aims included examining risk factors for rotavirus infection and also characterizing rotavirus genotypes circulating in the post-vaccine era.
We conducted surveillance for rotavirus infection among children receiving care for diarrhea at six government-administered primary care clinics distributed throughout León, Nicaragua. The study was conducted over a 12-month period from April of 2008 to March of 2009.
León is Nicaragua's second largest city, with an estimated population of 192,000 people in 2009. During the time of the study, the MOH reported that RV5 coverage in León was 98%, 93%, and 77% for receipt of the first, second, and third doses, respectively. The government-administered primary care clinics in León are free of charge and provide preventive and acute care. The clinics have capacity for intravenous hydration, but children requiring inpatient management are transferred to León's University Hospital. The MOH estimates that 90% of the population receives care in these clinics, whereas 10% receives care from private physicians.
A rotavirus prevalence study was conducted in seven government-administered primary care clinics in León during a 12-month period (August of 2002 to July of 2003) before vaccine introduction. Using the same recruitment criteria as the present study, the pre-vaccine study found a rotavirus prevalence among children with diarrhea of 14.0% (69/493; 95% confidence interval [CI] = 10.9–17.1).12 For all health facilities in León during this pre-vaccine year studied, the MOH reported 2,745 diarrhea visits among children under the age of 5 years. During the present year studied, the MOH reported 2,494 diarrhea visits among children under the age of 5 years.
Children ages 10 weeks to 36 months with an onset of diarrhea within 14 days before the clinic visit were recruited. Diarrhea was defined as more than or equal to three loose or liquid stools over a 24-hour period. Two study nurses were present at each site and recruited eligible children during clinic work hours. Each clinic is required to report numbers of diarrhea visits to the MOH. We used these reports to estimate our recruitment rate.
Study nurses administered a questionnaire to the parent(s) that assessed participant characteristics and obtained anthropometric measurements and temperature. A detailed vaccine history was obtained from the medical record. The study received Institutional Review Board approvals from the University of Nicaragua, León, Nicaragua (UNAN-León), and the University of North Carolina, Chapel Hill, NC. Informed consent was obtained from a parent or legal guardian.
Stool samples were collected in sterile containers or from soiled diapers during the clinic visit or within 3 days after the visit, and they were transported to the UNAN-León microbiology laboratory at 4°C. A commercial enzyme-linked immunosorbent assay (ELISA; Oxoid, Cambridge, United Kingdom) was used for identification of rotavirus. G (G1, G2, G3, G4, G8, G9, G10, and G12) and P (P4, P6, P8, P9, P10, and P11) genotypes were investigated by reverse transcription–polymerase chain reaction (RT-PCR) using primers and methods previously described.13
The prevalence of rotavirus was estimated by the proportion of samples that tested positive for rotavirus. The corresponding exact 95% CI was calculated assuming that samples from the same child were independent. Characteristics of children and diarrhea episodes with and without rotavirus were compared using the Wilcoxon rank sum test or Fisher's exact test. In the analysis of child characteristics, a child who provided two samples was classified by the result of the first sample. Each child's weight for age percentile was calculated using World Health Organization standards. Children were considered up to date if they received all recommended vaccine dosages for their age without being more than 1 month behind. STATA 9 software (College Station, TX) was used for all analyses.
Overall, 392 children presenting with diarrhea were enrolled in the study. Eighteen children presented with two diarrhea episodes for a total of 410 episodes. These 410 episodes among children under age 3 years represented 59% of the diarrhea visits reported to the MOH in the under age 5 years age group from these six clinics; therefore, our recruitment rate was at least 59%. Stool samples were obtained for 403 of 410 episodes (98.3%). Five children who were not able to provide a stool sample were excluded from the analysis.
Fourteen of four hundred three (3.5%; 95% CI = 1.9–5.8) stool samples tested positive for rotavirus, of which four samples were PG1, two samples were PG2, one sample was PG3, one sample was PG4, and three samples were mixed infections. One sample was negative for the genotypes investigated, and two samples were not investigated for genotypes because of insufficient stool quantity. The three mixed infections were P:G2/G9, P/P:G1/G9, and P:G1/G2/G4. Among those children who tested positive for rotavirus, 10 children had received all three doses of the vaccine, 3 children were partially immunized, and 1 child had not received vaccine. All of the G and P types reported above were observed among fully immunized children. The child with the P:G1/G2/G4 infection was unimmunized. Two children presented with rotavirus diarrhea within 2 days after receiving the vaccine.
Characteristics of the children are shown in Table 1. Children had an average age of 12.4 months, and 71.9% were breastfed. All participants who had received RV5 had also received the oral polio vaccine on the same day for at least one RV5 dose. The children who tested positive for rotavirus were less likely to have an indoor toilet. There was no difference in immunization status between those children who tested positive for rotavirus and those children with diarrhea of another cause.
Clinical characteristics of the diarrhea episodes experienced among the children are shown in Table 2. There were no differences in these clinical characteristics among episodes caused by rotavirus and episodes caused by other etiologies. There were also no differences in clinical characteristics of rotavirus episodes between those children immunized and those children unimmunized with RV5.
Rotavirus infection was uncommon among children seeking care for diarrhea in the primary care setting in León, Nicaragua. The low prevalence of rotavirus infection (3.5%; 95% CI = 1.9–5.8) occurred despite high rates of breastfeeding and co-administration of the oral polio vaccine, factors that are hypothesized to contribute to the lower effectiveness of the rotavirus vaccine in developing world settings.14 The rotavirus prevalence that we encountered was lower than reported in pre-vaccine studies conducted in the primary care setting in the same city12 and other Latin American countries (prevalence range = 14–35%).15–18 We also confirmed broad coverage of the rotavirus vaccine among the children in the study. By distributing the vaccine through Nicaragua's EPI, 90.7% of the children were up to date on the vaccine.
We observed high genetic diversity among the few rotavirus-positive samples, which was also observed before vaccine introduction in Central America.19–21 In addition, we observed that, among the fully immunized children who developed rotavirus infection, one-half were infected with P types not included in RV5. Finally, we observed a high proportion of mixed infections or infections with more than one rotavirus genotype. Mixed infections may allow for exchange of genetic material, resulting in unusual strains of unknown pathogenic potential.22 Future surveillance is warranted to investigate whether any genotypes are increasing in incidence in the population.
Although the vaccine's effectiveness in the primary care setting is supported by the low rotavirus prevalence compared with pre-vaccine studies, we did not find a statistical difference in immunization status between children with and without rotavirus diarrhea. Our study may have been limited in power to detect such a difference. An additional study limitation includes the inherent annual variation in rotavirus prevalence. Lastly, 10.9% of the participants were ineligible to receive the vaccine because of age greater than 2 months when the immunization program was introduced. In this group, any changes in rotavirus prevalence may be because of indirect effects of the vaccine.
In conclusion, rotavirus was not a common cause of childhood diarrhea in this Central American primary care setting after the introduction of a rotavirus immunization program. Future studies are warranted to investigate diarrhea etiologies in the post-rotavirus vaccine era, to both help direct clinicians in the appropriate treatment of children with diarrhea and guide the choice of future vaccine candidates and prevention efforts.
We would like to thank Mercedes Delgadillo, Dr. Xiomara Toruño, and the staff of the Centro de Investigación en Demografía y Salud (CIDS) for their contributions to this study. We would also like to thank Dr. Gilberto Moreno, Sistemas Locales de Atención Integral a la Salud (SILAIS), and the primary care nurses who assisted with the study.
Disclaimer: Some of the authors were awarded investigator-initiated research grants from Merck or served as consultants for Merck for unrelated studies. This statement is made in the interest of full disclosure and not because the authors consider this relationship to be a conflict of interest.
Financial support: This research was supported by a grant from the Gorgas Memorial Institute. Sylvia Becker-Dreps is supported by Grant 4K01TW008401-03 from the National Institutes of Health Fogarty International Center.
Authors' addresses: Sylvia Becker-Dreps, Department of Family Medicine, University of North Carolina School of Medicine, Chapel Hill, NC, E-mail: ude.cnu@dbs. Margarita Paniagua and Filemon Bucardo, Departamento de Microbiología, Campus Médico Universidad Nacional Autónoma de Nicaragua, León, Nicaragua, E-mails: xm.moc.oohay@natiagaugainap and moc.liamtoh@odracub_ilif. Luis Enrique Zambrana, Centro de Investigación en Demografía y Salud (CIDS), Universidad Nacional Autónoma de Nicaragua, León, Nicaragua, E-mail: moc.liamg@atelocramal. Michael G. Hudgens, Department of Biostatistics, University of North Carolina Gillings School of Global Public Health, Chapel Hill, NC, E-mail: ude.cnu.liame@snegduhm. David J. Weber, Division of Infectious Diseases, University of North Carolina School of Medicine, Chapel Hill, NC, E-mail: ude.cnu.hcnu@rebewD. Douglas R. Morgan, Division of Gastroenterology, University of North Carolina School of Medicine, Chapel Hill, NC, E-mail: ude.cnu.dem@nagrom_salguod. Félix Espinoza, Centro de Enfermedades Infecciosas (CEI), Departamento de Microbiología, Campus Médico Universidad Nacional Autónoma de Nicaragua, León, Nicaragua, E-mail: moc.liamg@80xilefazonipse.