Abstract
Background
Severe dehydration due to acute infectious diarrhea remains a leading cause of death among young children worldwide. Diarrhea with severe dehydration is a clinical syndrome with distinct management per the World Health Organization (WHO) Integrated Management of Childhood Illness (IMCI) and the WHO Global Task Force on Cholera Control (GTFCC) guidelines. We sought to characterize the pathogens causing severe dehydration using data from the Global Enteric Multicenter Study.
Methods
We used the IMCI and GTFCC guidelines to define severe dehydration and quantitative polymerase chain reaction–based attribution models to assign the etiology of diarrhea associated with severe dehydration.
Results
The IMCI or GTFCC guidelines classified 2284 of the 5304 (43%) cases with moderate-to-severe diarrhea as having severe dehydration. In one-third of the cases with severe dehydration, no pathogens were attributed. The top pathogens attributed to children with guidelines-classified severe dehydration varied by age and were similar among those requiring intravenous hydration and hospitalization. Rotavirus (30.9%), Cryptosporidium (12.0%), and heat-stable (ST) enterotoxigenic Escherichia coli (ETEC) (10.3%) were the most common pathogens for ages 0–11 months, while Shigella/enteroinvasive E coli (EIEC) (25.8%), rotavirus (19.3%), and ST-ETEC (10.9%) were the most common for ages 12–23 months. Shigella/EIEC (25.9%), Vibrio cholerae (10.4%), and rotavirus (9.2%) were the most common among ages 24–59 months.
Conclusions
The findings inform prioritization of pathogens, in addition to V cholerae, that cause severe dehydration for future preventive and treatment efforts. The schema for prioritization is driven primarily by age stratifications.
Keywords: diarrhea, dehydration, pediatric
Diarrhea with severe dehydration causes significant morbidity and mortality in low- and middle-income countries. Attributable causes of severe dehydration varied by age and by definition of severe dehydration, informing pathogen prioritization for prevention and treatment strategies.
Diarrheal illness, primarily acute infectious diarrhea, remains a leading cause of death among young children globally, causing 500 000 deaths per year primarily in low- and middle-income countries (LMICs) [1, 2]. In addition to morbidity and mortality from an acute episode, diarrheal illness can also result in additional sequelae including impaired growth and cognitive development [3]. Deaths due to diarrheal illness in young children occur primarily due to severe dehydration, associated with hypovolemic shock and electrolyte disturbances. Thus, the prevention and optimized management of severe dehydration can save lives.
In LMICs, the management of diarrheal illness, including of severe dehydration, is syndromic [4]. The World Health Organization (WHO) Integrated Management of Childhood Illness (IMCI) guidelines provide guidance on treatment of pediatric diarrhea using the dehydration level, chronicity, and presence of blood in the stool [5, 6]. Notably, the level of dehydration determines the rehydration plan (ie, oral vs intravenous [IV] fluid resuscitation) and indication to use antibiotics. If children >2 years of age are diagnosed with severe dehydration in a cholera-endemic area, IMCI guidelines recommend antibiotics for treatment of presumed cholera [5]. Notably, in those with severe dehydration, cholera endemicity is the only pathogen-specific factor in the consideration for empiric antibiotic use. The Global Task Force on Cholera Control (GTFCC) also provides a definition for severe dehydration associated with diarrhea that aids in the identification and treatment of potential cholera cases [7]. If children >2 years of age with suspected cholera and severe dehydration are hospitalized, GTFCC guidelines recommend antibiotic treatment [8]. With cholera, antibiotics have the potential to shorten the duration of symptoms and shedding as well as reduce the volume of stool loss in the setting of a cholera infection [9, 10]. However, global overuse of antibiotics is associated with increasing population-level antibiotic resistance in addition to individual-level adverse effects [11].
Asymptomatic and convalescent shedding of diarrheal pathogens has complicated the interpretation of epidemiologic studies. The emergence of molecular diagnostic techniques, paired with case-control study design, has allowed for the attribution of infectious etiologies associated with pediatric diarrhea. Although prior studies have described pathogens associated with varying levels of diarrheal illness severity, data on the etiologies attributed specifically to severe dehydrating diarrhea are limited [12–14]. While Vibrio cholerae is well known to cause acute watery diarrhea leading to severe dehydration [15], other pathogens have the potential to cause severe diarrhea and dehydration [16, 17], and use of antibiotics in viral and protozoal infections may result in more harm than benefit. Thus, our primary objective was to use a large dataset from a multicountry study of pediatric diarrhea to examine the pathogens associated with severe dehydration [18]. A secondary objective was to assess the test performance of the IMCI case definition in identifying V cholerae.
METHODS
Study Population
Our study is a secondary analysis of data from the Global Enteric Multicenter Study (GEMS). GEMS was a prospective, case-control study of children aged 0–59 months with moderate-to-severe diarrhea (MSD) at 7 sites in sub-Saharan Africa (Kenya, Mali, Mozambique, and The Gambia), and South Asia (Bangladesh, India, and Pakistan). Data collection occurred during December 2007–March 2011, prior to the widespread use of the rotavirus vaccine [19]. The methods have been described in detail previously [12, 18, 20]. In brief, cases were identified when presenting for care with the new onset (after 7 diarrhea-free days) of acute (<7 days’ duration) and moderate-to-severe (defined as at least 1 of the following: sunken eyes, loss of skin turgor, IV rehydration administered or prescribed, visible blood in loose stool, hospitalized with diarrhea or bloody stools) diarrhea. A total of 9439 cases with MSD were enrolled along with 1 to 3 matched nondiarrheal controls; for this analysis, we used only diarrhea cases. Demographic and clinical information was collected from caregiver interviews at enrollment and approximately 60-day follow-up. Clinicians also examined cases to assess signs of dehydration and to determine treatment. Stool specimens were collected from cases and controls. Our analysis was restricted to cases with MSD.
Definition of Diarrhea With Severe Dehydration
We used 2 definitions of diarrhea with severe dehydration in this analysis: (1) WHO IMCI criteria and (2) GTFCC criteria. GTFCC guidance aims to strengthen surveillance for cholera. The guideline includes criteria for severe dehydration that could apply to diarrhea caused by any pathogen. Severe dehydration per the WHO IMCI criteria is defined as 2 of the following signs: lethargic or unconscious, sunken eyes, not able to drink or drinking poorly, skin pinch goes back very slowly [5]. All variables were available using the data collected in GEMS. Severe dehydration per the GTFCC guidelines is defined as 1 or more of the following: lethargic or loss of consciousness, absent or weak pulse, respiratory distress; or at least 2 of the following: sunken eyes, not able to drink or drinks poorly, skin pinch goes back very slowly [7]. The GEMS data did not include pulse, and this was excluded from the definition in our analysis. We also compared the IMCI and GTFCC clinical criteria to clinical outcomes that would suggest severe dehydration: need for IV hydration and need for hospitalization.
Etiology
We used the quantitative real-time polymerase chain reaction–based (qPCR) majority attribution models developed by Liu et al to ascribe the etiology of diarrhea [21]. qPCR data were available for 5304 of the cases and thus our analysis was restricted to these cases. The subset of cases with qPCR data was previously randomly selected from each age strata and study site [21]. Using previously derived episode-specific attributable fraction (AFe) of etiology, we used a cut-off of >0.5 to designate attribution of a pathogen to a particular episode [22]. Of note, the invasion plasmid antigen H (ipaH) qPCR target was used for identification of Shigella and is also described in enteroinvasive Escherichia coli (EIEC); thus, we were unable to distinguish between Shigella and EIEC [21].
Statistical Analysis
We performed all analyses using R version 4.2.3 software [23]. We stratified the cases of MSD with qPCR data by dehydration severity: (1) severe as defined by the IMCI definition, (2) severe as defined by a modified (excluding pulse) GTFCC definition, (3) severe as defined by IMCI and GTFCC definitions, (4) severe as defined by either IMCI or GTFCC definition, and (5) not severe by either definition. We also stratified using other clinical proxies for severe dehydration: cases receiving IV hydration and cases requiring hospitalization. We examined the etiologies in each severity class. Data were further stratified by age (<12 months, 12–23 months, and 24–59 months). We also calculated descriptive statistics for demographic and clinical characteristics. Unadjusted and age-adjusted odds ratios (ORs) were calculated for severe dehydration by etiology using logistic regression.
Additionally, we assessed the test performance of the IMCI clinical case definition in identifying diarrhea of cholera etiology (Supplemental Table 1). The IMCI case definition outlines when antibiotics would be given for presumed cholera and was as follows: diarrhea with severe dehydration in a child aged ≥2 years in a cholera-endemic area [5]. An AFe cut-off of >0.5 (as defined above) for V cholerae was used as the gold standard to define true test positivity for sensitivity and specificity calculations to remain consistent with the other analyses. We also used V cholerae culture positivity as a gold standard for comparison. Areas with known cholera at the time of the study were defined as the following 4 sites: India, Pakistan, Bangladesh, and Mozambique.
Patient Consent Statement
The GEMS study protocol was approved by ethics committees and applicable scientific review boards at each field site and the University of Maryland, Baltimore. Individual, informed consent was obtained [18].
RESULTS
Of the 9439 children with MSD in GEMS, qPCR was available from 5304 cases, and these were included in this analysis. Among the 5304 cases, 2284 (43%) were classified as having severe dehydration by either the IMCI or GTFCC guidelines. The GTFCC guidelines had a broader definition of severe dehydration, and all cases that were labeled as severe with the IMCI guidelines were also labeled as severe with the GTFCC guidelines (n = 1893); an additional 391 were only categorized as severe with the GTFCC guidelines (SupplementalTable 2). Approximately 40% of children with severe dehydration were aged <1 year, compared to 32% in those without severe dehydration (Table 1).
Table 1.
Demographic and Clinical Characteristics of Included Cases
| Characteristic | Severe Dehydration: GTFCC or IMCI | Severe Dehydration: IMCI and GTFCC Criteria Met | Severe Dehydration: GTFCC Criteria Only | Not Severe Dehydration: Neither IMCI Nor GTFCC Criteria Met | Total |
|---|---|---|---|---|---|
| No. of cases | 2284 | 1893 | 391 | 3020 | 5304 |
| Female sex | 998 (43.7) | 823 (43.5) | 175 (44.8) | 1288 (42.6) | 2286 (43.1) |
| Male sex | 1286 (56.3) | 1070 (56.5) | 216 (55.2) | 1732 (57.4) | 3018 (56.9) |
| Age, mo | |||||
| 0–11 | 939 (41.1) | 764 (40.4) | 175 (44.8) | 970 (32.1) | 1909 (36.0) |
| 12–23 | 778 (34.1) | 642 (33.9) | 136 (34.8) | 1057 (35.0) | 1835 (34.6) |
| 24–59 | 567 (24.8) | 487 (25.7) | 80 (20.5) | 993 (32.9) | 1560 (29.4) |
| Study site | |||||
| Bangladesh | 106 (4.6) | 56 (3.0) | 50 (12.8) | 771 (22.5) | 877 (16.5) |
| India | 383 (16.8) | 356 (18.8) | 27 (6.9) | 466 (15.4) | 849 (16.0) |
| Kenya | 448 (19.6) | 356 (18.8) | 27 (6.9) | 466 (15.4) | 849 (16.0) |
| Mali | 308 (13.5) | 279 (14.7) | 29 (7.4) | 526 (17.4) | 834 (15.7) |
| Mozambique | 284 (12.4) | 173 (9.1) | 111 (28.4) | 200 (6.6) | 484 (9.1) |
| Pakistan | 446 (19.5) | 392 (20.7) | 54 (13.8) | 342 (11.3) | 788 (14.9) |
| The Gambia | 309 (13.5) | 264 (13.9) | 45 (11.5) | 376 (12.5) | 685 (12.9) |
| Diarrheal episode duration prior to presentation, d, mean (SD) | 2.77 (1.35) | 2.79 (1.36) | 2.64 (1.29) | 2.82 (1.34)a | 2.80 (1.34) |
| Lethargy/loss of consciousness | 576 (25.2) | 558 (29.5) | 18 (4.6) | 0 (0) | 576 (10.9) |
| Respiratory distress | 806 (35.3) | 429 (22.7) | 377 (96.4) | 0 (0) | 806 (15.2) |
| Sunken eyes | 2139 (93.7) | 1875 (99.0) | 264 (67.5) | 2151 (71.2) | 4290 (80.9) |
| Poor drinking | 902 (39.5) | 883 (46.6) | 19 (4.9) | 89 (2.9) | 991 (18.7) |
| Slow skin recoil | 1002 (43.9) | 992 (52.4) | 10 (2.6) | 38 (1.3) | 1040 (19.6) |
| Predominant stool type | |||||
| Bloody | 123 (5.4) | 92 (4.9) | 31 (7.9) | 373 (12.4) | 496 (9.4) |
| Rice watery | 171 (7.5) | 159 (8.4) | 12 (3.1) | 132 (4.4) | 303 (5.7) |
| Simple watery | 1466 (64.2) | 1239 (65.5) | 227 (58.1) | 1609 (53.3) | 3075 (58.0) |
| Sticky/mucoid | 524 (22.9) | 403 (21.3) | 121 (30.9) | 906 (30.0) | 1430 (27.0) |
| Rehydration | |||||
| None | 119 (5.2) | 44 (2.3) | 75 (19.2) | 242 (8.0) | 361 (6.8) |
| Oral | 1523 (66.7) | 1281 (67.7) | 242 (61.9) | 2551 (84.5) | 4074 (76.8) |
| IV | 642 (28.1) | 568 (30.0) | 74 (18.9) | 227 (7.5) | 869 (16.4) |
| Antibiotics given | 1688 (73.9) | 1380 (72.9) | 308 (78.8) | 2443 (80.9) | 4131 (77.9) |
| Hospitalized | 617 (27.0) | 451 (23.8) | 166 (42.5) | 546 (18.1) | 1163 (21.9) |
| Death by ∼60-d follow-upb | 85 (4.1) | 68 (4.0) | 17 (4.8) | 28 (1.0) | 113 (2.3) |
| Median time from enrollment to death, d | 10.0 | 10.0 | 12.0 | 21.0 | 13.0 |
Data are presented as No. (%) unless otherwise indicated.
Abbreviations: GTFCC, World Health Organization Global Task Force on Cholera Control; IMCI, World Health Organization Integrated Management of Childhood Illness; IV, intravenous; SD, standard deviation.
aOne case (not severe dehydration) missing data for diarrheal episode duration.
bMissing status at 60-day follow-up for 483 cases. Percentages were calculated using observed sample sizes as follows: IMCI and GTFCC criteria, n = 1698; GTFCC only, n = 351; GTFCC or IMCI, n = 2049; neither IMCI nor GTFCC, n = 2772; Total, n = 48.
Clinical Characteristics
We did not identify any difference in diarrheal episode duration at time of presentation between those with and without severe dehydration. Most children with severe dehydration had sunken eyes and/or slow skin recoil (99.0% and 93.7% had sunken eyes and 52.4% and 43.9% had slow skin recoil per IMCI and GTFCC classifications, respectively, Supplemental Table 3). Twenty-eight percent of children with severe dehydration received IV rehydration and 27.0% were hospitalized. Death by 60 days of follow-up was more common in the cases with severe dehydration; 4.1% died compared to 1% of those without severe dehydration (Table 1).
Etiologies
Among the children with severe dehydration, approximately one-third did not have any attributable pathogens (33% for IMCI definition, 35% GTFCC). Most children had 1 attributable pathogen (53% for IMCI definition, 52% GTFCC) and some had ≥2 attributable pathogens (14% for IMCI definition, 13% GTFCC). The most common attributed pathogens varied by age group. In children 0–11 months of age, the most common pathogens in those with severe dehydration meeting either criteria were rotavirus (30.9%), Cryptosporidium (12.0%), heat-stable (ST) enterotoxigenic E coli (ETEC) (10.3%), Shigella/EIEC (7.3%), and adenovirus 40/41 (7.0%). In those 12–23 months of age, they were Shigella/EIEC (25.8%), rotavirus (19.3%), ST-ETEC (10.9%), Cryptosporidium (7.6%), and adenovirus 40/41 (4.6%). In those 24–59 months of age, Shigella/EIEC (25.9%), V cholerae (10.4%), rotavirus (9.2%), ST-ETEC (7.8%), and Helicobacter pylori (6.7%) were the most common (Table 2). Among those without severe dehydration, the top 5 etiologies were the same for each age group as for those with severe dehydration, though the proportion varied. In those 0–11 months of age, rotavirus was attributed in one-third (30.9%) of the cases compared to one-fourth (23.7%) in those without severe dehydration; Cryptosporidium and ST-ETEC were also attributed more frequently in those with severe dehydration. For the cases in children aged 24–59 months, V cholerae was attributed in 10.4% of cases with severe dehydration compared to 2.6% of the cases without severe dehydration (Table 2).
Table 2.
Pathogens and Dehydration Severity, Stratified by Age Group
| Pathogen | Dehydration Measure and Age Group | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Age 0–11 mo (n = 1909) | Age 12–23 mo (n = 1835) | Age 24–59 mo (n = 1560) | ||||||||||
| GTFCC or IMCI Guidelines | Received IV Rehydration | Hosp.a Required | Not Severely Dehydratedb | GTFCC or IMCI Guidelines | Received IV Rehydration | Hosp. Requireda | Not Severely Dehydratedb | GTFCC or IMCI Guidelines | Received IV Rehydration | Hosp. Requireda | Not Severely Dehydratedb | |
| No. of cases | 939 | 380 | 460 | 970 | 778 | 284 | 397 | 1057 | 567 | 205 | 306 | 993 |
| Rotavirus | 290 (30.9) | 149 (39.2) | 181 (39.3) | 230 (23.7) | 150 (19.3) | 77 (27.1) | 136 (34.3) | 196 (18.5) | 52 (9.2) | 21 (10.2) | 34 (11.1) | 62 (6.2) |
| Shigella/EIEC | 69 (7.3) | 21 (5.5) | 34 (7.4) | 96 (9.9) | 201 (25.8) | 67 (23.6) | 125 (31.5) | 444 (42.0) | 147 (25.9) | 61 (29.8) | 139 (45.4) | 437 (44.0) |
| ST-ETEC | 97 (10.3) | 48 (12.6) | 35 (7.6) | 49 (5.1) | 85 (10.9) | 36 (12.7) | 39 (9.8) | 84 (7.9) | 44 (7.8) | 18 (8.8) | 19 (6.2) | 54 (5.4) |
| Cryptosporidium spp | 113 (12.0) | 49 (12.9) | 48 (10.4) | 76 (7.8) | 59 (7.6) | 29 (10.2) | 31 (7.8) | 47 (4.4) | 7 (1.2) | 1 (0.5) | 4 (1.3) | 4 (0.4) |
| Adenovirus 40/41 | 66 (7.0) | 37 (9.7) | 46 (10.0) | 57 (5.9) | 36 (4.6) | 11 (3.9) | 13 (3.3) | 42 (4.0) | 10 (1.8) | 1 (0.5) | 4 (1.3) | 16 (1.6) |
| Vibrio cholerae | 16 (1.7) | 13 (3.4) | 4 (0.9) | 8 (0.8) | 26 (3.3) | 21 (7.4) | 10 (2.5) | 18 (1.7) | 59 (10.4) | 42 (20.5) | 30 (9.8) | 26 (2.6) |
| Helicobacter pylori | 6 (0.6) | 5 (1.3) | 5 (1.1) | 5 (0.5) | 13 (1.7) | 3 (1.1) | 4 (1.0) | 14 (1.3) | 38 (6.7) | 12 (5.9) | 18 (5.9) | 57 (5.7) |
| Astrovirus | 20 (2.1) | 10 (2.6) | 6 (1.3) | 22 (2.3) | 23 (3.0) | 3 (1.1) | 6 (1.5) | 31 (2.9) | 7 (1.2) | 0 (0) | 2 (0.7) | 9 (0.9) |
| Salmonella spp | 7 (0.7) | 2 (0.5) | 3 (0.7) | 0 (0) | 17 (2.2) | 8 (2.8) | 10 (2.5) | 16 (1.5) | 14 (2.5) | 7 (3.4) | 7 (2.3) | 13 (1.3) |
| Norovirus GII | 19 (2.0) | 7 (1.8) | 9 (2.0) | 19 (2.0) | 4 (0.5) | 1 (0.4) | 4 (1.0) | 6 (0.6) | 9 (1.6) | 0 (0) | 4 (1.3) | 15 (1.5) |
| Sapovirus | 3 (0.3) | 0 (0) | 0 (0) | 5 (0.5) | 7 (0.9) | 1 (0.4) | 2 (0.5) | 19 (1.8) | 20 (3.5) | 4 (2.0) | 6 (2.0) | 21 (2.1) |
| Campylobacter jejuni/C coli | 28 (3.0) | 17 (4.5) | 22 (4.8) | 47 (4.8) | 0 (0) | 0 (0) | 0 (0) | 6 (0.6) | 0 (0) | 0 (0) | 2 (0.7) | 5 (0.5) |
| tEPEC | 16 (1.7) | 11 (2.9) | 13 (2.8) | 10 (1.0) | 10 (1.3) | 6 (2.1) | 5 (1.3) | 6 (0.6) | 1 (0.2) | 0 (0) | 0 (0) | 0 (0) |
| Aeromonas spp | 8 (0.9) | 1 (0.3) | 6 (1.3) | 5 (0.5) | 1 (0.1) | 0 (0) | 2 (0.5) | 5 (0.5) | 16 (2.8) | 9 (4.4) | 6 (2.0) | 25 (2.5) |
| Entamoeba histolytica | 0 (0) | 0 (0) | 0 (0) | 1 (0.1) | 1 (0.1) | 0 (0) | 0 (0) | 2 (0.2) | 7 (1.2) | 0 (0) | 0 (0) | 18 (1.8) |
| Cyclospora cayetanensis | 2 (0.2) | 1 (0.3) | 1 (0.2) | 2 (0.2) | 2 (0.3) | 0 (0) | 2 (0.5) | 5 (0.5) | 2 (0.4) | 0 (0) | 1 (0.3) | 3 (0.3) |
| Cystoisospora belli | 2 (0.2) | 1 (0.3) | 1 (0.2) | 1 (0.1) | 0 (0) | 0 (0) | 0 (0) | 0 (0) | 0 (0) | 0 (0) | 0 (0) | 0 (0) |
| EAEC | 0 (0) | 0 | 0 (0) | 0 (0) | 1 (0.1) | 0 | 1 (0.3) | 0 (0) | 0 (0) | 0 | 0 (0) | 0 (0) |
| LT-ETEC | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| STEC | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
Estimates are shown as No. (%).
Abbreviations: EAEC, enteroaggregative Escherichia coli; EIEC, enteroinvasive Escherichia coli; ETEC, enterotoxigenic Escherichia coli; GTFCC, World Health Organization Global Task Force on Cholera Control; Hosp., hospitalization; IMCI, World Health Organization Integrated Management of Childhood Illness; IV, intravenous; LT, heat-labile; ST, heat-stable; STEC, Shiga toxin–producing Escherichia coli; tEPEC, typical enteropathogenic Escherichia coli.
aHospitalization refers to admission to the hospital; some children received treatment without requiring admission.
bNot meeting either GTFCC or IMCI criteria.
In children with MSD, the highest odds of having severe dehydration was in those episodes attributed to V cholerae (OR, 3.05 [95% confidence interval {CI}, 2.17–4.33]) when adjusted for age. Episodes attributed to Shigella/EIEC had among the lowest odds of having severe dehydration (OR, 0.50 [95% CI, .44–.57]) (Table 3).
Table 3.
Odds Ratios for Severe Dehydration (as Defined by World Health Organization Global Task Force on Cholera Control or Integrated Management of Childhood Illness Guidelines) in Children With Moderate-to-Severe Diarrhea, by Pathogen
| Pathogen | OR (95% CI) | Adjusteda OR (95% CI) |
|---|---|---|
| Rotavirus | 1.42 (1.24–1.64) | 1.30 (1.13–1.50) |
| Shigella/EIEC | 0.46 (.41–.53) | 0.50 (.44–.57) |
| ST-ETEC | 1.66 (1.36–2.04) | 1.62 (1.33–1.99) |
| Cryptosporidium spp | 1.94 (1.53–2.45) | 1.74 (1.37–2.21) |
| Adenovirus 40/41 | 1.30 (1.00–1.70) | 1.19 (.91–1.56) |
| Vibrio cholerae | 2.64 (1.89–3.73) | 3.05 (2.17–4.33) |
| Helicobacter pylori | 0.99 (.70–1.40) | 1.21 (.85–1.73) |
| Astrovirus | 1.07 (.73–1.55) | 0.99 (.68–1.44) |
| Salmonella spp | 1.74 (1.08–2.86) | 1.91 (1.18–3.15) |
| Norovirus GII | 1.06 (.66–1.69) | 1.04 (.64–1.65) |
| Sapovirus | 0.88 (.55–1.39) | 0.98 (.61–1.56) |
| Campylobacter jejuni/C coli | 0.63 (.40–.99) | 0.55 (.34–.86) |
| tEPEC | 2.25 (1.22–4.27) | 2.01 (1.09–3.82) |
| Aeromonas spp | 0.94 (.56–1.57) | 1.08 (.63–1.80) |
| Entamoeba histolytica | 0.50 (.21–1.09) | 0.69 (.28–1.47) |
| Cyclospora cayetanensis | 0.79 (.27–2.14) | 0.84 (.28–2.28) |
| Cystoisospora belli | 2.65 (.25–56.96) | 2.27 (.22–48.83) |
| EAEC | NA | NA |
| LT-ETEC | NA | NA |
| STEC | NA | NA |
Odds ratios were unable to be calculated for EAEC, LT-ETEC, and STEC due to the low number of cases (1, 0, and 0, respectively).
Abbreviations: CI, confidence interval; EAEC, enteroaggregative Escherichia coli; EIEC, enteroinvasive Escherichia coli; ETEC, enterotoxigenic Escherichia coli; LT, heat-labile; NA, not applicable; OR, odds ratio; ST, heat-stable; STEC, Shiga toxin–producing Escherichia coli; tEPEC, typical enteropathogenic Escherichia coli.
aAdjusted for age at enrollment.
Measures of Dehydration
Overall, the pathogen hierarchy did not change when severe dehydration was restricted to IMCI-only guidelines compared to GTFCC (Supplemental Table 4). The IMCI and GTFCC guidelines classified many more children as severely dehydrated than using the clinical outcomes of IV rehydration (16.4% of children) and hospitalization (21.9%) as proxies for severe dehydration (Table 2). Although the total number of children classified as severely dehydrated varied if using a clinical measure such as IV hydration or hospitalization, the top pathogenremained similar in each age group (Table 2 and Supplemental Table 5). Notable exceptions were that rotavirus, rather than Shigella/EIEC, represented the most common pathogen identified in children aged 12–23 months requiring IV rehydration or hospitalization and that V cholerae was attributed to 20.5% of the children requiring IV hydration compared to 10.4% of the children classified as severely dehydrated by GTFCC or IMCI guidelines (Table 2).
Test Performance
Given that IMCI case definition serves to identify potential cholera cases, such that they could be provided appropriate rehydration and antibiotics, we examined the performance of this case definition in identifying V cholerae–attributed episodes. The test performance of the IMCI case definition at identifying V cholerae in countries with known cholera, using AFe >0.5 as the gold standard, showed sensitivity of 67% and specificity of 72%. The positive predictive value (PPV) was 19% and the negative predictive value (NPV) was 96% (Supplemental Table 6). Test performance was similar using V cholerae culture positivity as a gold standard (Supplemental Table 7).
DISCUSSION
In this analysis of children seeking medical care for MSD with severe dehydration, we showed that the most common pathogens associated with severe dehydration as classified by WHO IMCI or GTFCC criteria varied by age group. Rotavirus was the most common in those aged <12 months, and Shigella/EIEC was the most common attributable etiology in those 12–59 months of age. Rotavirus was also among the top 5 pathogens in 12–23 months of age as well as 24–59 months of age. Cryptosporidium was also common among those <24 months of age (attributed to approximately 10% of cases), while V cholerae was common among those 24–59 months of age (attributed to approximately 10% of cases). The prevalence of rotavirus is less surprising given that the data collection occurred prior to the widespread use of the rotavirus vaccine. However, the prevalence of other etiologies aside from V cholerae has implications for both public health, such as prioritization of vaccine development efforts, and the clinical management of individual patients, through informing syndromic guidance of diarrheal treatment.
The leading 6 identified etiologies of children with diarrhea presenting with severe dehydration as classified by WHO IMCI or GTFCC criteria were rotavirus, Shigella/EIEC, ST-ETEC, Cryptosporidium, adenovirus, and V cholerae. These also represented the top 6 identified etiologies of children who received IV fluids and required hospitalization in the study. While Shigella/EIEC is a leading cause of severe dehydration, episodes of MSD attributed to Shigella/EIEC had among the lowest odds of having severe dehydration. This finding could be related to the high burden of diarrhea attributable to Shigella/EIEC in the GEMS population. The other most common pathogens had higher odds of having severe dehydration. Salmonella and enteropathogenic E coli both had higher odds of severe dehydration but were not frequently attributed as etiologies of diarrhea; further evaluation of their potential to cause severe dehydration is warranted.
Our findings are similar to other multisite studies of pediatric diarrheal illnesses using molecular detection that occurred after the implementation of the rotavirus vaccine. In the Global Pediatric Diarrhea Surveillance network (GPDS), which examined children aged <5 years with diarrhea requiring hospitalization after the introduction of the rotavirus vaccine, rotavirus was the leading cause of diarrhea requiring hospitalization, with Shigella/EIEC as the second most prevalent cause [13]. Interestingly, norovirus was the third most common cause in GPDS, while norovirus was only attributed in 1.4% of children with severe dehydration in our study [13]. In the Malnutrition and Enteric Disease Study (MAL-ED), which was a longitudinal study of diarrheal episodes, a modified Vesikari score was used to define severity of diarrheal episode, and rotavirus, Shigella/EIEC, adenovirus, and Cryptosporidium were associated with high severity [14]. Additionally, previous analyses of the GEMS dataset have described etiologies associated with MSD (which includes children with and without severe dehydration) and showed similar findings with Shigella/EIEC, adenovirus, ST-ETEC, V cholerae, rotavirus, and Cryptosporidium representing the top 6 pathogens [21]. Overall, these findings highlight the importance of considering pathogens other than V cholerae in children presenting with severe dehydration. Shigella/EIEC is of particular importance as antibiotic treatment is also recommended but may differ from that for V cholerae.
Overall, the IMCI and GTFCC clinical criteria for severe dehydration identified similar cases. The GTFCC criteria categorized slightly more patients as severely dehydrated compared to the IMCI criteria. However, the distribution of pathogens was very similar between the 2. Although previous studies have demonstrated that the IMCI guidelines are not accurate for predicting severe dehydration compared to a gold standard of weight loss [24], the pathogens that are associated with severe dehydration in our analysis are similar to pathogens in studies that used different metrics for dehydration, such as the Vesikari scale [14, 25, 26].
We found that severe dehydration alone in a child aged >2 years is poorly predictive of diarrhea due to V cholerae. Using the IMCI clinical criteria, in GEMS countries that were cholera endemic, only 19% of children >2 years of age with severe dehydration were attributed to V cholerae. Additionally, the PPV of the IMCI criteria, 19%, is not clinically useful. The IMCI guidelines recommend that all children >2 years of age meeting criteria for severe dehydration in cholera-endemic countries should be treated with antibiotics. The low PPV suggests that a significant number would be inappropriately deemed in need of antibiotics for treatment of V cholerae. Conversely, the NPV is 96% and therefore 96% of patients without severe dehydration do not have V cholerae, suggesting that few V cholerae–positive patients would be missed using these clinical criteria. Those with MSD and V cholerae had the highest odds of severe dehydration, emphasizing the already known risk of severe dehydration with cholera [27]. The inadequacy of the clinical criteria highlights the importance of the development of better diagnostic tools to help guide appropriate treatment [28, 29].
While IMCI guidelines recommend that children presenting with severe dehydration should receive IV fluids, we found that only 30% of children who met IMCI criteria for severe dehydration received IV fluids (Table 1). Previous research has shown that adherence to the WHO guidelines for management of diarrheal diseases may be near impossible due to a combination of factors including inadequate staffing, limited resources, and differences between patient and clinician expectations, among others [30–32]. Additionally, a small number of children classified as severely dehydrated did not require any hydration (oral or IV) at the healthcare center. The IMCI and GTFCC definitions of severe dehydration incorporate both provider and caregiver assessment and it is possible that provider gestalt, which determined need for rehydration, differed from the clinical criteria classification that has been observed in prior studies [24]. Furthermore, structural barriers such as accessibility and availability of supplies as well as training of staff affect ability to adhere to treatment guidelines [32]. Technologies to provide clinical decision support for this challenge may improve guideline adherence [33, 34].
Our study has several limitations. First, GEMS only included children presenting with MSD, and thus we are unable to determine the proportion of each etiology that presents with severe dehydration. Second, the GTFCC guidelines also include pulse as a factor to determine dehydration severity, and our analysis was limited by the absence of pulse in the GEMS dataset. The inclusion of pulse data would have potentially added more cases to the severe category that were not captured with the other criteria, and it is unclear if these missing severe dehydrated cases would have similar or different etiologies. Additionally, our analysis represents a subset of all the cases in GEMS as qPCR data were only available for 5304 of 9439 total cases, which limits the generalizability of the results. Finally, GEMS data represent an older dataset and were collected prior to the availability of the rotavirus vaccine and may not represent the current landscape.
Overall, our study highlights important pathogens aside from V cholerae that cause diarrhea with severe dehydration, a leading cause of death in young children worldwide. Rotavirus was a leading pathogen in our study and remained so in the GPDS study after introduction of the rotavirus vaccine, highlighting the importance of improving rotavirus vaccine coverage. Shigella/EIEC, ST-ETEC, Cryptosporidium, and V cholerae also are important causes of severe dehydration. Vaccine and treatment advances should be targeted at these pathogens to reduce the mortality associated with severe dehydration. Shigella/EIEC was prevalent across all age groups and was also an important cause of hospitalization in the GPDS study and severe diarrhea in MAL-ED, suggesting that Shigella/EIEC represent an important high-priority target for vaccine development. Besides rotavirus, V cholerae is the only other pathogen with an approved vaccine, although unfortunately with supply issues, brief length of protection, and suboptimal effectiveness in young children, its use as a preventive strategy is limited [35]. Additionally, the IMCI and GTFCC clinical criteria are poor predictors of a V cholerae infection, which emphasizes the importance of continued work on development of readily available diagnostics to guide appropriate treatment.
Supplementary Data
Supplementary materials are available at Open Forum Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.
Supplementary Material
Contributor Information
Anna Jones, Department of Pediatrics, University of Utah School of Medicine, Salt Lake City, Utah, USA.
Sharia M Ahmed, Division of Epidemiology, Department of Internal Medicine, University of Utah School of Medicine, Salt Lake City, Utah, USA.
James A Platts-Mills, Division of Infectious Diseases and International Health, University of Virginia, Charlottesville, Virginia, USA.
Karen L Kotloff, Department of Pediatrics, Center for Vaccine Development and Global Health, University of Maryland School of Medicine, Baltimore, Maryland, USA.
Adam C Levine, Department of Emergency Medicine, Warren Alpert Medical School of Brown University, Providence, Rhode Island, USA.
Eric J Nelson, Department of Pediatrics and Environmental and Global Health, Emerging Pathogens Institute, University of Florida, Gainesville, Florida, USA.
Andrew T Pavia, Division of Pediatric Infectious Diseases, Department of Pediatrics, University of Utah School of Medicine, Salt Lake City, Utah, USA.
Ashraful I Khan, Infectious Diseases Division, International Center for Diarrhoeal Disease Research, Bangladesh (icddr,b), Dhaka, Bangladesh.
Daniel T Leung, Division of Infectious Diseases, Department of Internal Medicine, University of Utah School of Medicine, Salt Lake City, Utah, USA.
Notes
Author contributions. Concept and design: S. M. A. and D. T. L. Data analysis: A. J. Drafting of the primary manuscript: A. J. Critical revision of the manuscript for important intellectual content: All authors. Administrative, technical, or material support: All authors.
Data availability. This submitted manuscript is a secondary analysis. GEMS data are publicly and freely available by request through ClinEpiDB.org. The previously derived attributable fractions (AFe) for etiologies will be made available upon request.
Financial support . This work was supported by the National Institutes of Health (grant numbers R01AI135114 and K24166087 to D. T. L.; Ruth L. Kirschstein National Research Service Award T32AI055434 to S. M. A.; and grant number R38 HL143605 to A. J. through Utah Stimulating Access to Research in Residency [StARR]).
References
- 1. Perin J, Mulick A, Yeung D, et al. Global, regional, and national causes of under-5 mortality in 2000–19: an updated systematic analysis with implications for the Sustainable Development Goals. Lancet Child Adolesc Health 2022; 6:106–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Roth GA, Abate D, Abate KH, et al. Global, regional, and national age-sex-specific mortality for 282 causes of death in 195 countries and territories, 1980–2017: a systematic analysis for the Global Burden of Disease Study 2017. Lancet 2018; 392:1736–88. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. World Gastroenterology Organisation . Acute diarrhea in adults and children: a global perspective. 2012. Available at: https://www.worldgastroenterology.org/guidelines/acute-diarrhea/acute-diarrhea-english. Accessed 14 August 2024.
- 4. Nadri J, Sauvageot D, Njanpop-Lafourcade BM, et al. Sensitivity, specificity, and public-health utility of clinical case definitions based on the signs and symptoms of cholera in Africa. Am J Trop Med Hyg 2018; 98:1021–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. World Health Organization . Integrated management of childhood illness: distance learning course. 2014. Available at: https://apps.who.int/iris/handle/10665/104772. Accessed 18 July 2023.
- 6. World Health Organization . The treatment of diarrhoea. 2005. Available at: https://www.who.int/publications/i/item/9241593180. Accessed 3 September 2024.
- 7. Global Task Force on Cholera Control Surveillance Working Group . Interim guidance document on cholera surveillance. Global task force on cholera control. 2017. Available at: https://www.gtfcc.org/wp-content/uploads/2023/02/gtfcc-public-health-surveillance-for-cholera-interim-guidance.pdf. Accessed 21 August 2024.
- 8. Global Task Force on Cholera Control . Interim technical note: use of antibiotics for the treatment and control of cholera. 2022. Available at: https://www.gtfcc.org/wp-content/uploads/2019/10/gtfcc-technical-note-on-use-of-antibiotics-for-the-treatment-of-cholera-1.pdf. Accessed 30 May 2024.
- 9. Harris JB, LaRocque RC, Qadri F, Ryan ET, Calderwood SB. Cholera. Lancet 2012; 379:2466–76. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Nelson EJ, Nelson DS, Salam MA, Sack DA. Antibiotics for both moderate and severe cholera. N Engl J Med 2011; 364:5–7. [DOI] [PubMed] [Google Scholar]
- 11. Rostami A, Zadeh FA, Ebrahimzadeh F, Jafari-Sales A, Gholami S. Globally Vibrio cholera antibiotics resistance to RNA and DNA effective antibiotics: a systematic review and meta-analysis. Microb Pathog 2022; 172:105514. [DOI] [PubMed] [Google Scholar]
- 12. Kotloff KL, Nataro JP, Blackwelder WC, et al. Burden and aetiology of diarrhoeal disease in infants and young children in developing countries (the Global Enteric Multicenter Study, GEMS): a prospective, case-control study. Lancet 2013; 382:209–22. [DOI] [PubMed] [Google Scholar]
- 13. Cohen AL, Platts-Mills JA, Nakamura T, et al. Aetiology and incidence of diarrhoea requiring hospitalisation in children under 5 years of age in 28 low-income and middle-income countries: findings from the Global Pediatric Diarrhea Surveillance network. BMJ Glob Health 2022; 7:e009548. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Platts-Mills JA, Babji S, Bodhidatta L, et al. Pathogen-specific burdens of community diarrhoea in developing countries (MAL-ED): a multisite birth cohort study. Lancet Glob Health 2015; 3:e564–75. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Clemens JD, Nair GB, Ahmed T, Qadri F, Holmgren J. Cholera. Lancet 2017; 390:1539–49. [DOI] [PubMed] [Google Scholar]
- 16. Keita AM, Doh S, Sow SO, et al. Prevalence, clinical severity, and seasonality of adenovirus 40/41, astrovirus, sapovirus, and rotavirus among young children with moderate-to-severe diarrhea: results from the vaccine impact on diarrhea in Africa (VIDA) study. Clin Infect Dis 2023; 76(Suppl 1):S123–31. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Yeasmin S, Hasan SMT, Chisti MJ, Khan MA, Faruque ASG, Ahmed T. Factors associated with dehydrating rotavirus diarrhea in children under five in Bangladesh: an urban-rural comparison. PLoS One 2022; 17:e0273862. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Kotloff KL, Blackwelder WC, Nasrin D, et al. The Global Enteric Multicenter Study (GEMS) of diarrheal disease in infants and young children in developing countries: epidemiologic and clinical methods of the case/control study. Clin Infect Dis 2012; 55(Suppl 4):S232–45. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. World Health Organization . Global Health Observatory repository. Rotavirus—immunization coverage estimates by country. 2023. Available at: https://apps.who.int/gho/data/view.main.ROTACv?lang=en. Accessed 30 May 2024.
- 20. Panchalingam S, Antonio M, Hossain A, et al. Diagnostic microbiologic methods in the GEMS-1 case/control study. Clin Infect Dis 2012; 55(Suppl 4):S294–302. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Liu J, Platts-Mills JA, Juma J, et al. Use of quantitative molecular diagnostic methods to identify causes of diarrhoea in children: a reanalysis of the GEMS case-control study. Lancet 2016; 388:1291–301. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Brintz BJ, Howard JI, Haaland B, et al. Clinical predictors for etiology of acute diarrhea in children in resource-limited settings. PLoS Negl Trop Dis 2020; 14:e0008677. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. R Core Team . R: a language and environment for statistical computing. 2023. Available at: https://www.R-project.org/. Accessed 1 July 2024.
- 24. Jauregui J, Nelson D, Choo E, et al. External validation and comparison of three pediatric clinical dehydration scales. PLoS One 2014; 9:e95739. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Lee GO, Richard SA, Kang G, et al. A comparison of diarrheal severity scores in the MAL-ED multisite community-based cohort study. J Pediatr Gastroenterol Nutr 2016; 63:466. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Platts-Mills JA, Houpt ER, Liu J, et al. Etiology and incidence of moderate-to-severe diarrhea in young children in Niger. J Pediatr Infect Dis Soc 2021; 10:1062–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Andrews JR, Leung DT, Ahmed S, et al. Determinants of severe dehydration from diarrheal disease at hospital presentation: evidence from 22 years of admissions in Bangladesh. PLoS Negl Trop Dis 2017; 11:e0005512. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Ahmed SM, Brintz BJ, Pavlinac PB, et al. Clinical prediction rule to guide diagnostic testing for shigellosis and improve antibiotic stewardship for pediatric diarrhea. Open Forum Infect Dis 2023; 10:ofad119. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Pavlinac PB, Platts-Mills JA, Liu J, et al. Azithromycin for bacterial watery diarrhea: a reanalysis of the AntiBiotics for Children With Severe Diarrhea (ABCD) trial incorporating molecular diagnostics. J Infect Dis 2024; 229:988–98. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Biswas D, Hossin R, Rahman M, et al. An ethnographic exploration of diarrheal disease management in public hospitals in Bangladesh: from problems to solutions. Soc Sci Med 2020; 260:113185. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Elshabassi N, Garbern SC, Rosen RK, et al. Understanding variations in diarrhea management across healthcare facilities in Bangladesh: a formative qualitative study. J Infect Dev Ctries 2023; 17:665–76. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Deichsel EL, Keita AM, Verani JR, et al. Management of diarrhea in young children in sub-Saharan Africa: adherence to World Health Organization recommendations during the Global Enteric Multisite Study (2007–2011) and the Vaccine Impact of Diarrhea in Africa (VIDA) study (2015–2018). Clin Infect Dis 2023; 76(Suppl 1):S23–31. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Levine AC, Gainey M, Qu K, et al. A comparison of the NIRUDAK models and WHO algorithm for dehydration assessment in older children and adults with acute diarrhoea: a prospective, observational study. Lancet Glob Health 2023; 11:e1725–33. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Khan AI, Mack JA, Salimuzzaman M, et al. Electronic decision support and diarrhoeal disease guideline adherence (mHDM): a cluster randomised controlled trial. Lancet Digit Health 2020; 2:e250–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Malembaka EB, Bugeme PM, Hutchins C, et al. Effectiveness of one dose of killed oral cholera vaccine in an endemic community in the Democratic Republic of the Congo: a matched case-control study. Lancet Infect Dis 2024; 24:514–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.