| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
ABSTRACT |
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Since 1992, we have performed several longitudinal, village-based studies using active case detection to examine the epidemiology of pediatric diarrhea in Egyptian children.4–9 As in other developing countries, these projects demonstrate that diarrheal diseases continue to be a major disease burden. All-cause diarrhea was 2.9 and 5.5 episodes per child-year of follow-up among children less than 3 years of age in Abees and Abu Homos, respectively.4,6,8 Enterotoxigenic Escherichia coli (ETEC) was commonly isolated with 0.6 episodes and 1.5 episodes per child-year for the same two areas, respectively.4,8 Besides ETEC, Campylobacter, rotavirus, and Shigella have been isolated in 0.6, 0.24, and 0.2 episodes per child-year.5,8,9 For these enteropathogens, dehydration varied from a low of 12% for ETEC-associated to a high of 56% for rotavirus-associated diarrhea.
To complement these studies, starting in May 2000 we developed and implemented passive surveillance at two Egyptian hospitals located in the Nile River Delta. The study was completed in May 2002. The primary objective of this study was to assess the epidemiology and clinical characteristics of severe diarrhea, diarrhea severe enough to bring a child to medical attention. We expect that these results will be useful for decision analysis models for prioritizing vaccine requirements.
|
MATERIALS AND METHODS |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Enrollment. It was common during the summer months for more than 20 children a day to be evaluated for diarrhea. To maintain quality data collection, a prestudy decision was made to sequentially register and enroll only every fifth child. If a guardian did not want to participate, the caretaker of the next registered child was asked to enroll their child.
Data collection. After informed consent, a clinical history and physical examination was recorded on a structured, pre-tested questionnaire.
Specimen collection, handling, and storage. Two rectal swabs and a stool sample were obtained from each study child. One swab was placed into a tube of Cary-Blair (CB) transport medium and the other into a tube containing buffered-glycerol-saline (BGS) enrichment medium. After collection, the inoculated tubes were refrigerated at 2°C to 8°C while the stool sample was split into seven aliquots and then frozen at –20°C. Within 3 days of collection, specimens were sent to U.S. Naval Medical Research Unit No. 3 (NAMRU3) located approximately 3 hours by car from the study clinics. During transportation, rectal swabs were placed in a cold box with ice packs while frozen specimens were placed in a container with dry ice. Upon arrival at NAMRU3, all specimens were accessioned and rectal swabs were sent to the microbiology laboratory for culture, and the frozen samples were transferred to a –70°C freezer.
Microbiology. Using conventional microbiologic techniques, both rectal swabs were streaked onto standard laboratory agar media for the recovery and identification of common enteric bacterial pathogens: Shigella, Salmonella, Campylobacter, and Vibrio.10–13 Campylobacter isolates were further speciated using hippurate hydrolysis,14 and Shigella isolates were serogrouped and serotyped by slide agglutination using commercial antisera (Difco Laboratories, Livonia, MI).15 In addition, five E. coli–like colonies were picked and tested for the expression of heat-labile enterotoxin (LT) and/or heat-stable enterotoxin (ST) using a GM1 ELISA.16,17 Colonies testing positive for LT or ST were further tested for expression of colonization factor (CF) antigens CFA/I, CFA/II (coli surface antigen [CS] 1, CS2, CS3), CFA/III, CFA/IV (CS4, CS5, CS6), CS7, CS17, PCFO159, and PCFO166 by monoclonal antibody dot blot assay.18 Following the manufacturer’s instructions, a commercial enzyme immunoassay (Premier Rotaclone, Meridian Bioscience, Inc., Cincinnati, OH) was used to identify rotavirus antigen from the submitted stool samples.
Definitions. Time from disease onset until evaluation was termed "illness length." Stooling patterns were described by the "mean number of stools" in the 24 hours before evaluation and "mean maximal stools" on any day since the illness began. Dehydration was defined according to definitions of the World Health Organization.19 An episode lasting less than 14 days without visible blood was termed "acute watery diarrhea," while "dysentery" was defined as an illness lasting less than 14 days with visible blood or mucus in the stool. "Persistent diarrhea" was defined as an episode with or without blood lasting longer than 14 days. "Fever" was defined as a rectal temperature > 38.5°C and "severe fever" as rectal temperature > 40.0°C. When parents reported a history of fever, the fevers were classified as present or absent. Seizures were defined as a sudden, violent, involuntary contraction of a muscle group. Summer months were from May to October and cool months were from November to April. If an organism was identified from the stool sample, the child was classified as "infected" with that organism. If a single pathogen was isolated from the stool, then the episode was referred to as "pathogen"-associated diarrhea (e.g., ETEC-associated diarrhea). "Undiagnosed diarrhea" included any episode in which a studied bacterial pathogen or rotavirus was undetected.
Analysis.
We used a 
2 test of independence to determine if there was a statistically significant difference in the proportion of children with a clinical characteristic such as vomiting among enteropathogens (i.e., Campylobacter, ETEC, Shigella, rotavirus, or undiagnosed agents). If a statistically significant difference was detected among the enteropathogens, then a series of 2 tests were used to determine which pair (e.g., ETEC versus rotavirus) of enteropathogens differed by that clinical characteristic. When the symptom was measured on a continuous scale (e.g., mean number of stools in 24 hours), the normality of the distribution was tested using Shapiro-Wilk test,20 and analysis of variance (ANOVA) was used to determine if there were statistically significant differences in the mean of that variable when categorized by enteropathogen. When a statistically significant difference was detected, Student’s t tests were used to identify which pair of means was statistically different. To determine if a pathogen showed seasonality, we tested if the proportion of summer cases (May–October) was statistically greater than 50% of all cases using a single sample binomial test of significance.21
All data was double entered into Epi-info version 6. Stat-exact 4.0 (Cytel Software Corporation, Cambridge, MA); SAS 6.12 and SAS 9.0 (SAS, Cary, NC) were used for all analyses. Statistical significance was two-tailed and set at P < 0.05 for all analyses.
This study was approved by the Institutional Review Board of the U.S. Naval Medical Research Unit No. 3, Arab Republic of Egypt and the Ministry of Health and Population, Arab Republic of Egypt. The work was performed under work unit 6000.RAD1.D.0301. Written informed consent was obtained from each parent during enrollment.
|
RESULTS |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Rectal swabs were collected from 99% of children in Benha and 100% of children in Abu Homos. A stool sample was obtained from 90% (N = 505) of the children from Benha and 73% (N = 521) of the children from Abu Homos. The greater frequency of stool collection in Benha appeared to be due to a greater willingness on the part of Benha parents to wait until there children produced a stool.
Disease burden. Because one in five children were enrolled, we extrapolated that 2,805 clinic visits to Benha Clinic and 3,570 clinic visits to Abu Homos for treatment of diarrhea were made during the study. Based on hospital records, this represented 12% and 19% of all clinic visits for children under 6 years of age at Benha and Abu Homos Clinics, respectively.
Diarrhea cases.
The number of children with diarrhea at these clinics was inversely correlated with age (Table 1
). Children less than 12 months were 56% and those from birth to 24 months were 87% of the population. This distribution of cases was similar to that of Benha (P = 0.5). Female children were somewhat less likely to be brought for treatment at both clinics.
|
). The distribution of these enteropathogens differed by study site (P < 0.0001) due to differences in the isolation rate of Campylobacter (P = 0.03), rotavirus (P < 0.0001), mixed infections (P = 0.02), and undiagnosed diarrhea cases (P = 0.0001). The distribution of diarrhea episodes stratified by age and enteropathogen showed a distinct pattern. Except for ETEC cases in Abu Homos, the number of episodes seeking treatment increased from birth to 5 months old, peaked at 6 to 11 months old, and decreased over the next two successive age categories (i.e., 12 to 23 and 24 to 59 months). Two-hundred sixty diarrhea cases associated with rotavirus (161 Benha, 99 Abu Homos) were detected. Of these 260 cases, rotavirus was the sole pathogen in 204 (78%) cases. Of the 258 ETEC-associated episodes (111 Benha, 148 Abu Homos), ETEC was the sole pathogen isolated from the stool in 201 cases. ETEC more frequently produced LT (N = 115, 45%) followed by ST (N = 100, 39%). The remaining ETEC isolates produced LTST (case = 21, 8%) or mixed enterotoxin types such as LT and LTST (N = 22, 8%).
An ETEC colonization factor was identified in 127 (49%) of the ETEC isolates. However, CF expression varied by enterotoxin expression. CFs were less frequently identified among ETEC strains expressing LT (21%, N = 24) than in strains expressing ST (70%, N = 70), LTST (71%, N = 15), or mixed strains (82%, N = 18). CFs were heterogeneous. CFA/I (N = 45), CFA/IV (N = 33), and CS7 (N = 15) were the three most common adhesins, followed by CFA/II (N = 12), PCFO166 (N = 10), CFA/III (N = 4), PCFO159 (N = 2), CS17 (N = 3), and two mixed CFs (CS6 and CFA/II; CS7 and CFA/I).
Campylobacter was isolated from 38 stools (9 Benha, 29 Abu Homos) and was the sole pathogen in 26 (68%) cases. Ninety-five percent of isolates were C. jejuni. Shigella was isolated from 23 children (9 Benha, 14 Abu Homos) and was the sole pathogen in 19 (73%) cases. Of the 23 isolates, 19 (83%) were S. flexneri, three (13%) were S. dysenteriae, and one (4%) was S. sonnei.
Diarrhea seasonality.
Children were more likely to be brought for treatment during the warm through early cool season in Abu Homos (Figure 1). ETEC was the prominent organism during the warm season. In Benha, diarrhea was more common during the warm season of the first year but showed no increase during the warm season of the second year. For both years in both clinics, but especially, the first year in Benha, there was a noticeable increase in cases during the late summer, early cool season primarily due to rotavirus-associated diarrhea. The increase in rotavirus isolation during the first year contributed to the higher proportion of rotavirus-associated cases in Benha (23%) than Abu Homos (10%). Although occurring mainly during the warmer months, Campylobacter and Shigella occurred sporadically at other times of the year.
|
Clinical characteristic.
Children were brought for treatment within 3 to 5 days of disease onset, but children who had rotavirus-associated diarrhea were taken for health care 1 day earlier than children with undiagnosed diarrhea (P = 0.003) (Table 2). Loose motions in the 24 hours before treatment differed by enteropathogen with the frequency of stooling for Shigella-infected children being statistically greater for them than for children with ETEC (P = 0.03), Campylobacter (P = 0.04), and undiagnosed diarrhea (P = 0.02), but statistically similar to that of children with rotavirus (P = 0.8).
|
Mucoid stools were more frequent when patients were infected with Campylobacter than with rotavirus (P = 0.0006), ETEC (P = 0.03), and Shigella (P = 0.04) but not against undiagnosed diarrhea cases (P = 0.7). At the time of medical evaluation, 17% (N = 209) children had a documented fever with 1% having a temperature 
40°C. Neither the historical (P = 0.1) nor objective (P = 0.7) cases of fever were more frequently associated with any specific enteropathogen. Vomiting was statistically more common among children with rotavirus-associated diarrhea than for children infected with ETEC (P < 0.0001), Campylobacter (P < 0.002), Shigella (P < 0.04), and undiagnosed diarrhea (P < 0.0001).
Children infected with Shigella reported four (22%) seizures, and the frequency of seizures was greater for Shigella than ETEC (P = < 0.0001), rotavirus (P = < 0.0001), and undiagnosed diarrhea (P = < 0.0001). Only children with Campylobacter-associated diarrhea had rates of seizures similar to that of Shigella-associated diarrhea (P = 0.2).
Rotavirus-associated diarrhea was linked to a significant number of dehydrated children at the time of evaluation. Rotavirus cases presented with 39% "some" and 3% "severe" dehydration. Rotavirus-associated dehydration was statistically greater than for children infected with ETEC (P = 0.0003), Campylobacter (P = 0.03), and undiagnosed diarrhea cases (P = 0.0001) but not statistically for Shigella (P = 0.72).
The infecting enteropathogen also affected the rates of hospitalization in children with diarrhea. Likely related to the increased frequency of dehydration, children with rotavirus-associated diarrhea were more likely to be hospitalized and showed a statistically significant difference compared with ETEC (P = 0.0005), Campylobacter (P = 0.01), and undiagnosed diarrheal cases (P = 0.0001), but statistically similar to that of Shigella (P = 0.27).
|
DISCUSSION |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Similar to our community-based studies, ETEC was the predominant bacterial pathogen recovered. As expected, ETEC episodes were more severe in the current study than ETEC episodes identified in community studies.4,8,23 Like the community-based studies, no clinical feature allowed ETEC to be differentiated from other diarrheal etiologies.4
Consistent with our community-based studies, ST-ETEC was more commonly isolated in the warmer months, while LT-ETEC showed no seasonal variation.4,8 ETEC expressing a CF were also more common in the warmer months likely due to the higher proportion of ST-ETEC strains that express a CF.8
In agreement with other studies of pediatric diarrhea, we found rotavirus to be common with rates equal to those of ETEC. This is in contrast to our field-based study where rotavirus was the fourth leading cause of diarrhea after ETEC-, Campylobacter-, and Shigella-associated diarrhea.5 This highlights that while rotavirus may induce fewer episodes than other enteropathogens, it appears to induce a greater proportion of severe cases. Also in the current study, rotavirus occurred more often during the cooler months, a pattern similar to that in other temperate climates24,25 and similar to a report made from Bilbeis, Egypt.23 These data contrast with results from our community-based study that identified rotavirus more often in the warmer months.5 Taken together, these studies may suggest that rotavirus is a disease of late warm and early cool season but seasonality may vary year-to-year.
For rotavirus, certain clinical features were suggestive of the infection. A rapid onset of symptoms, frequent watery stools, recurrent vomiting, dehydration, and hospitalization occurred more commonly with rotavirus than the other enteric pathogens.
Campylobacter- and Shigella-associated diarrhea were identified less frequently than ETEC- or rotavirus-associated cases. Campylobacter- and Shigella-associated cases were occasionally associated with blood and/or mucus in stools as reported elsewhere in Egypt,8,26 while Campylobacter infections were also found to be associated with convulsions, possible febrile seizures, that have also been reported elsewhere.27,28
A potential limitation to the current study is the generalizability of the data from two clinics of the Nile River Delta to the pediatric population of Egypt. An additional concern is that differences between the two study sites may have been due to dissimilarity in interviewing, clinical methods, or sample collection. In an attempt to minimize this bias, we structured and standardized all our data collection procedures and then provided the same training and supervision to investigators and staff at both facilities. As rotavirus testing can only be performed on stool sample, the lower percentage of stool samples available from Abu Homos may partially explain why rotavirus was less commonly identified at that site. To check for this possibility, an analysis was performed that was limited to children from whom a stool sample was available for rotavirus testing and the difference in rotavirus isolation rates remained.
Study implications. Observations from this investigation may be important for inclusion in data analysis models for prioritization of control strategies, including the development and deployment of new enteric vaccines. These results are similar to those of other developing countries where watery diarrhea is common and dysentery infrequent. The data suggest that interventions should concentrate on children less than 2 years of age, where disease is most common. Among these children, rotavirus-associated diarrhea was the most prevalent and severe. Although laboratory confirmation of rotavirus would be preferable, even in the absence of laboratory confirmation, it would be desirable for medical staff to initiate surveillance for rotavirus-like symptoms (i.e., vomiting and severe diarrhea).
Received April 8, 2005. Accepted for publication June 20, 2005.
Acknowledgments: The authors thank Dr. Anne Marie Svennerholm for providing the reagents used in testing for enterotoxigenic E. coli and Ms. Manal Mostafa for assisting with the data management and analysis of the study. This study was approved by institutional review boards at the Naval Medical Research Unit No. 3 in compliance with all Federal regulations governing the protection of human subjects. All subjects or their guardians gave voluntary, informed consent for participation prior to enrollment in this study.
Financial support: Global Emerging Infections System (GEIS), Work Unit No.: 847705.82000.25GB.E0018
* Address correspondence to Thomas F. Wierzba, Ph.D., M.P.H., 6190 Kathmandu Place, Dulles, VA 20189. E-mail: wierzbat{at}who.org.np 
Authors’ addresses: Thomas F. Wierzba, 6190 Kathmandu Place, Dulles, VA, 20189, Telephone: +977-1-444-4950, Fax: +977-1-553-0150, E-mail: wierzbat{at}who.org.np. Ibrahim Adib Abdel-Messih, U.S. Naval Medical Research Unit No. 3, PSC 452, Box 5000, FPO AE 09835, Telephone: +20-2-342-1375, Fax: +20-2-342-9625, E-mail: adibi{at}namru3.med.navy.mil. Remon Abu-Elyazeed, 150 Beach Road #22-00, Gateway west, Singapore 189720, Telephone: 65-232 8338, Fax: 65-6732 8678, E-mail: remon.abu-elyazeed{at}gsk.com. Shannon D. Putnam, U.S. Naval Medical Research Unit No. 2, FPO AP 96520, Telephone: +62-21-421-4457, Fax: +62-21-424-4507, E-mail: putnam{at}namru2.org. Karim A. Kamal, U.S. Naval Medical Research Unit No. 3, PSC 452, Box 5000, FPO AE 09835, Telephone: +20-2-342-1375, Fax: +20-2-342-9625, E-mail: kamalk{at}menanet.net. Patrick Rozmajzl, Naval Medical Research Center, 503 Robert Grant Avenue, Silver Spring, MD 20910-7500, Telephone: 1-301-319-9667, Fax: 301-319-7460, E-mail: rozmajzlp{at}nmrc.navy.mil. Salwa F. Ahmed, U.S. Naval Medical Research Unit No. 3, PSC 452, Box 5000, FPO AE 09835, Telephone: +20-2-342-1375, Fax: +20-2-342-9625, E-mail: fouads{at}namru3.med.navy.mil. Hind I. Shaheen, U.S. Naval Medical Research Unit No. 3, PSC 452, Box 5000, FPO AE 09835, Telephone: +20-2-342-1375, Fax: +20-2-342-9625, E-mail: shaheenh{at}namru3.med.navy.mil. Abdel Fatah, Benha Fever Hospital, Benha, Qalubya Governorate, Egypt, Telephone: +20-2-421764, Fax: +20-2-421764. Khaled Zabedy, Abu Homos District Hospital, Beheira Governorate, Egypt, Telephone: +20-45-256 0044, Fax: +20-45-256 0130. John Sanders, U.S. Naval Medical Research Unit No. 3, PSC 452, Box 5000, FPO AE 09835, Telephone: +20-2-342-1375, Fax: +20-2-342-9625, E-mail: sandersj{at}namru3.navy.med.mil. Robert Frenck, U.S. Naval Medical Research Unit No. 3, PSC 452, Box 5000, FPO AE 09835, Telephone: +20-2-342-1375, Fax: +20-2-342-9625, E-mail: rfrenck{at}uclacvr.labiomed.org.
Reprint requests: Thomas F. Wierzba, Ph.D., M.P.H., 6190 Kath-mandu Place, Dulles, VA 20189, Telephone: +977-1-444-4950, Fax: +977-1-553-0150, E-mail: wierzbat{at}who.org.np.
|
REFERENCES |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  H. I. Shaheen, I. A. Abdel Messih, J. D. Klena, A. Mansour, Z. El-Wakkeel, T. F. Wierzba, J. W. Sanders, S. B. Khalil, D. M. Rockabrand, M. R. Monteville, et al. Phenotypic and Genotypic Analysis of Enterotoxigenic Escherichia coli in Samples Obtained from Egyptian Children Presenting to Referral Hospitals J. Clin. Microbiol., January 1, 2009; 47(1): 189 - 197. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
