PEDIATRICS Vol. 122 No. 3 September 2008, pp. 545-549 (doi:10.1542/peds.2007-3141)
ARTICLE |
Validation of the Clinical Dehydration Scale for Children With Acute Gastroenteritis
a Pediatric Research in Emergency Therapeutics Program, Division of Pediatric Emergency Medicine, BC Children's Hospital, Vancouver, Canada
b Department of Pediatrics, Child and Family Research Institute, University of British Columbia, Vancouver, Canada
c Pediatric Outcomes Research Team, Division of Pediatric Medicine, Hospital for Sick Children, Toronto, Canada
d Department of Pediatrics, Faculty of Medicine, University of Toronto, Toronto, Canada
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
OBJECTIVE. We previously created a clinical dehydration scale. Our objective was to validate the clinical dehydration scale with a new cohort of patients with acute gastroenteritis who were assessed in a tertiary emergency department in a developed country.
METHODS. A prospective observational study was performed in an emergency department at a large pediatric tertiary center in Canada. Children 1 month to 5 years of age with symptoms of acute gastroenteritis who were assessed in the emergency department were enrolled consecutively during a 4-month period. The main outcome measures were length of stay, proportion of children receiving intravenous fluid rehydration, and proportions of children with abnormal serum pH values or bicarbonate levels.
RESULTS. A total of 205 children were enrolled, with a mean age of 22.4 ± 14.9 months; 103 (50%) were male. The distribution of severity categories was as follows: no dehydration (score of 0), n = 117 (57%); some dehydration (score of 1–4), n = 83 (41%); moderate/severe dehydration (score of 5–8), n = 5 (2%). The 3 dehydration categories were significantly different with respect to the validation hypotheses (length of stay, mean ± SD: none, 245 ± 181 minutes; some, 397 ± 302 minutes; moderate/severe, 501 ± 389 minutes; treatment with intravenous fluids: none, n =17, 15%; some, n = 41, 49%; moderate/severe, n = 4, 80%; number of vomiting episodes in the 7 days before the emergency department visit: none, 8.4 ± 7.7 episodes; some, 13 ± 10.7 episodes; moderate/severe, 30.2 ± 14.8 episodes).
CONCLUSION. The clinical dehydration scale and the 3 severity categories were valid for a prospectively enrolled cohort of patients who were assessed in our tertiary emergency department. The scoring system was valuable in predicting a longer length of stay and the need for intravenous fluid rehydration for children with symptoms of acute gastroenteritis.
Key Words: scale • validation • dehydration • gastroenteritis
Abbreviations: CDS—clinical dehydration scale • ED—emergency department • AGE—acute gastroenteritis • LOS—length of stay
Acute gastroenteritis (AGE) is one of the most common reasons for hospitalization of young children. Among US children <5 years of age, gastroenteritis causes an estimated 220000 admissions per year, with a total hospital length of stay (LOS) of 925000 days,1 and results in direct costs for outpatient and hospital visits of more than $2 billion per year.2
It was suggested previously that the clinical ability of physicians to estimate degrees of dehydration is lacking and laboratory studies are needed to enhance accurate assessment.3 Until recently, the level of dehydration was frequently determined without using a unified validated scale.4–6 Several authors have attempted to determine a reliable measure of dehydration by using combinations of clinical findings.3,4,6–8 A systematic review found that the best individual examination signs for assessment of dehydration were prolonged capillary refill time, abnormal skin turgor, and abnormal respiratory pattern.9 Grouping signs can improve diagnostic accuracy.
We reported previously on the development of a clinical dehydration scale (CDS) for children 1 to 36 months of age with AGE in the emergency department (ED).10 With the use of formal measurement methods (item selection and reduction, reliability, discriminatory power, validity, and responsiveness), the CDS consists of 4 clinical items, which may be summed for a total score ranging from 0 to 8 (Table 1). We assessed the measurement properties (sensitivity, specificity, and likelihood ratios) of the CDS to predict 3 dehydration categories. The final 3 categories were no dehydration (CDS score of 0), some dehydration (CDS score of 1–4), and moderate/severe dehydration (CDS score of 5–8). The aim of the current study was to validate prospectively the CDS dehydration categories with a new cohort of children.
|
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
This was a prospective observational study of children 1 month to 5 years of age with suspected AGE in the ED. For the purpose of the study, we defined AGE according to the American Academy of Pediatrics practice parameter, that is, "diarrheal disease of rapid onset, with or without accompanying symptoms and signs such as nausea, vomiting, fever, or abdominal pain."2
Our tertiary care pediatric ED has 
50000 visits annually and provides specialized pediatric emergency care for a population of 3 to 4 million children in southern Ontario. Between January 1, 2005, and May 6, 2005, parents of children with vomiting or diarrhea during the 24 hours before arrival at the ED were approached by a research assistant and asked to sign an informed consent form for participation in the study. We excluded children with diarrhea for >10 days (which suggested a chronic diarrheal disease), any suspected cause of dehydration other than presumed gastroenteritis, or a chronic disease (renal disease, gastrointestinal disease, cystic fibrosis, including coexisting malnutrition, or a failure to thrive). We also excluded children who had received treatment with intravenous fluids within the previous 24 hours or had visited our ED for the same illness in the 7 days before arrival (because this might have altered the treatment of the child during the second visit).
The triage nurse in the ED completed the CDS assessment (Table 1) after the history was recorded and a physical examination was performed. The patient then continued to be treated by the attending ED physician without any active study-related intervention.
We collected information on age, gender, length of illness, and numbers of episodes of vomiting and diarrhea, as reported by the parents. In addition, we collected information on the chief complaint, final diagnosis at discharge or admission, whether a trial of oral rehydration was implemented, whether intravenous fluids were administered, and serum pH and bicarbonate results if plasma samples were obtained. The normal value for serum pH was considered 
7.32 and that for serum bicarbonate levels 18 mEq/L, according to the reference ranges for the laboratory at the Hospital for Sick Children (P.C.P., C. Macarthur, MD, A. Khambalia, PhD, R.D.G., and J.N.F., unpublished data, 2007). We then calculated the LOS in the ED, defined as the time of arrival in triage to the time of discharge or the time of the decision to admit the patient to the observation unit (in the ED) or the general pediatrics inpatient unit.
Three primary validation hypotheses were constructed a priori. The first hypothesis was that the 3 dehydration categories (none, some, or moderate/severe) would be positively associated with the LOS. The second hypothesis was that the 3 dehydration categories would be positively associated with the proportion of children receiving intravenous fluid rehydration. The third hypothesis was that the 3 dehydration categories would be positively associated with the proportion of children with abnormal serum pH values or bicarbonate levels. A secondary validation hypothesis was that the 3 dehydration categories (none, some, or moderate/severe) would be positively associated with the number of vomiting and diarrhea episodes before presentation to the ED. The study was approved by the Hospital for Sick Children Research Ethics Board, and informed parental consent was obtained.
As recommended previously by the World Health Organization, using the CDS we classified dehydration as none, some, or moderate/severe.11,12 In accordance with the definitions we used previously (P.C.P., C. Macarthur, MD, A. Khambalia, PhD, R.D.G., and J.N.F., unpublished data, 2007) and for the purpose of the analysis, we categorized the CDS results into no dehydration (defined as a CDS score of 0), some dehydration (defined as a CDS score of 1–4), and moderate/severe dehydration (defined as a CDS score of 5–8).
Data were entered into Microsoft Excel (Microsoft, Redmond, Washington) with a double-data entry technique. Statistical analyses were conducted with SPSS for Windows 13 (SPSS, Chicago, IL). Baseline characteristics were reported by using descriptive statistics. The 3 validation hypotheses were analyzed by using analysis of variance (for continuous data) and the 
2 test (for proportional data). A P value of <.05 was considered significant.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Baseline Characteristics
During the study period, we recruited a total of 211 patients. We excluded 2 patients because scores were not determined by the triage nurse and 3 patients because they had incomplete data. Of the 206 completed records, 103 (50%) were for boys; the mean age was 22.4 ± 14.9 months.
The dehydration scores were 0 for 117 children (57%), 1 to 4 for 84 children (41%), and 5 for 5 children (2%). A total of 196 children (95%) were discharged from the hospital, and 10 (5%) were admitted to the hospital.
Children with increased severity of dehydration tended to be older (P = .06). Fifty-nine children (29%) had venous blood samples obtained. Seventeen children (29%) had pH values of 
7.31 (range: 7.14–7.31), and the rest had values of 7.32 (range: 7.32–7.43). Twenty-three children (39%) had bicarbonate levels of 17 mEq/L (range: 11–17 mEq/L), and the rest had levels of 18 mEq/L (range: 18–25 mEq/L). Fifty-five (93%) of the children who had blood samples taken were treated at least in part with intravenous fluid rehydration.
The median LOS for the cohort was 231 minutes (mean ± SD: 313 ± 252 minutes). Twenty-one children (10%) were discharged with no rehydration attempted in the ED. One hundred seventy-seven children (86%) were treated with oral rehydration, and 54 (31%) of the 177 needed intravenous fluid administration, as determined by the ED staff physician. Eight children (4%) were given intravenous fluids without an oral rehydration trial.
Validation Hypotheses
The results for the validation hypotheses are presented in Table 2. The 3 dehydration categories (none, some, or moderate/severe) were positively associated with the LOS (none, 245 ± 181 minutes; some, 397 ± 302 minutes; moderate/severe, 501 ± 389 minutes; P < .001). The 3 dehydration categories were positively associated with the proportions of children who received intravenous fluid rehydration (none, 15%; some, 49%; moderate/severe, 80%; P = .001). The 3 dehydration categories demonstrated a trend toward positive associations with the proportions of children with abnormal serum pH values or bicarbonate levels, but this was not statistically significant. A box plot comparing the LOS in the ED for children in the 3 groups is presented in Fig 1.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
We described previously the development of the CDS for use in young children with AGE in the ED.10 The 4-item scale (which includes general appearance, eyes, mucous membranes, and tears) takes a triage nurse
2 minutes to complete and is simple to use. Furthermore, by using the 8-point CDS, we assessed the diagnostic test properties of 3 dehydration severity categories, that is, no dehydration (CDS score of 0), some dehydration (CDS score of 1–4), and moderate/severe dehydration (CDS score of 5–8) (P.C.P., C. Macarthur, MD, A. Khambalia, MD, R.D.G., and J.N.F., unpublished data, 2007). We now demonstrate, with this new, prospectively enrolled cohort of patients, that the dehydration categories correlate well with LOS and treatment with intravenous fluid rehydration. The validation in this study supports the use of the CDS and the 3 dehydration severity categories for diagnostic and therapeutic purposes in children 1 month to 5 years of age with AGE presenting to an urban tertiary care ED in a developed country. The CDS can be used to help guide clinicians in treatment decisions (oral or intravenous fluid rehydration) and disposition (discharge, observation unit, or hospital admission). Although it is true that most definitions of dehydration are based on changes (losses) in weight,13 baseline weights rarely are available to clinicians, especially for children >1 year of age, who do not visit their primary care providers frequently. A validated scale such as the one we have developed can assist in treatment decisions in the absence of a recent accurate weight.
Several previous efforts to devise a simple dehydration scale have been reported. Gorelick et al4 and Duggan et al6 reported clinical predictors of dehydration similar to the 4 items in our scale. Mackenzie et al5 found decreased peripheral perfusion, deep breathing, and decreased skin turgor to be associated with 4% dehydration. However, some of those studies had limitations, because they had different outcome measures6 regarding the level of dehydration, included physicians in training,5 or incorporated older children6 or only admitted patients.5 More recently, Steiner et al9 systematically reviewed 11 clinical signs and results of 7 laboratory tests from a total of 7 studies, before the publication of our dehydration scale. They found that, although the most useful individual signs for predicting 5% dehydration in children were abnormal capillary refill time (likelihood ratio: 4.1; 95% confidence interval: 1.7–9.8), abnormal skin turgor (likelihood ratio: 2.5; 95% confidence interval: 1.5–4.2), and abnormal respiratory pattern (likelihood ratio: 2.0; 95% confidence interval: 1.5–2.7), combinations of examination signs were better than any individual sign in predicting dehydration.
There are several potential limitations to our study. First, the study was conducted in only 1 center, a tertiary pediatric center. Future studies should be multicentered, including developing countries and small centers, to increase the generalizability of our findings. Second, we had a relatively small number of children in the moderate/severe dehydration category, which limited our ability to analyze data separately for that group. However, this is similar to other studies conducted in North America, where children present early during the course of their illness. Third, our validation efforts used LOS, and it is important to note that this indicator is influenced by multiple factors, such as staffing and the busyness of the ED. Practice variations among staff physicians may account for some of the differences in children's LOS and the decision to administer fluids intravenously, but we suggest that such practice variations were minimized by the fact that all faculty members in this ED were trained in pediatric emergency medicine. Furthermore, minimizing practice variations is not a requirement for developing and testing a validation hypothesis. If practice variations were eliminated through implementation of a set of criteria (such as clinical practice guidelines), then it would not be possible to assess validation of the scale; for example, assessment of the relationship between CDS scores and intravenous fluid use would be confounded by the established criteria for intravenous fluid use. Fourth, it is possible that, while we were conducting this validation study, some staff members started using the scale and changed their practices because of the study (Hawthorne effect). However, we suggest that the relatively short period of the study and the novel use of the CDS in our department accounted for a negligible Hawthorne effect. Triage nurses were not sharing the results of the scoring with physicians, and the forms containing the scores were not part of the patients' medical records. Furthermore, the CDS was not part of the recommended guidelines, and its use was not presented to the staff as a recommended guideline. Fifth, because of the relatively small number of children for whom blood tests were performed, we were unable to determine the validity of our third hypothesis, that is, that the 3 dehydration categories would be positively associated with the proportions of children with abnormal serum pH values or bicarbonate levels. Future research is needed to provide information on this hypothesis.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
The CDS and the 3 dehydration severity categories were valid for a prospectively enrolled cohort of patients assessed in our tertiary ED. The CDS was valuable in predicting a longer LOS and the need for intravenous fluid rehydration for children with symptoms of AGE. The CDS may be added to a decision-making algorithm to guide the treatment of children with AGE.
|
ACKNOWLEDGMENTS |
|---|
Dr Goldman is supported by the Child and Family Research Institute (Vancouver, Canada). The Pediatric Outcomes Research Team is supported by a grant from the Hospital for Sick Children Foundation (Toronto, Canada).
|
FOOTNOTES |
|---|
Accepted Dec 18, 2007.
Address correspondence to Ran D. Goldman, MD, Division of Pediatric Emergency Medicine, BC Children's Hospital, 4480 Oak St, Vancouver, BC V6H 3V4, Canada. E-mail: rgoldman{at}cw.bc.ca
The authors have indicated they have no financial relationships relevant to this article to disclose.
| What's Known on This Subject Acute gastroenteritis is one of the most common reasons for hospitalization of young children. We previously created a clinical dehydration scale but had not yet validated it.
|
| What This Study Adds The clinical dehydration scale and the 3 severity categories were valid for a prospectively enrolled cohort of patients assessed in a tertiary emergency department.
|
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
1. Cicerello HG, Glass RI. Current concepts of the epidemiology of diarrheal diseases. Semin Pediatr Infect Dis. 1994;5 (3):163 –167
2. American Academy of Pediatrics, Provisional Committee on Quality Improvement, Subcommittee on Acute Gastroenteritis. Practice parameter: the management of acute gastroenteritis in young children. Pediatrics. 1996;97 (3):424–435
3. Vega RM, Avner JR. A prospective study of the usefulness of clinical and laboratory parameters for predicting percentage of dehydration in children. Pediatr Emerg Care. 1997;13 (3):179–182
4. Gorelick MH, Shaw KN, Murphy KO. Validity and reliability of clinical signs in the diagnosis of dehydration. Pediatrics. 1997;99 (5). Available at: www.pediatrics.org/cgi/content/full/99/5/e6
5. Mackenzie A, Barnes G, Shann F. Clinical signs of dehydration in children. Lancet. 1989;2 (8663):605 –607[Web of Science][Medline]
6. Duggan C, Refat M, Hashem M, Wolff M, Fayad I, Santosham M. How valid are clinical signs of dehydration in infants? J Pediatr Gastroenterol Nutr. 1996;22 (1):56 –61[CrossRef][Web of Science][Medline]
7. Ozuah PO, Avner JR, Stein RE. Oral rehydration, emergency physicians, and practice parameters: a national survey.
Pediatrics. 2002;109
(2):259
–261
8. Plata Rueda E, Diaz Cruz G. Clinical and biochemical evaluation of the degree of dehydration in children with acute diarrhea [in Spanish]. Bol Med Hosp Infant Mex. 1974;31 (3):561 –576[Medline]
9. Steiner MJ, DeWalt DA, Byerley JS. Is this child dehydrated?
JAMA. 2004;291
(22):2746
–2754
10. Friedman JN, Goldman RD, Srivastava R, Parkin PC. Development of a clinical dehydration scale for use in children between 1 and 36 months of age. J Pediatr. 2004;145 (2):201 –207[CrossRef][Web of Science][Medline]
11. World Health Organization, Program for Control of Diarrhoeal Diseases. A Manual for the Treatment of Diarrhoea. Geneva, Switzerland: World Health Organization;1990
12. King CK, Glass R, Bresee JS, Duggan C. Managing acute gastroenteritis among children: oral rehydration, maintenance, and nutritional therapy. MMWR Recomm Rep. 2003;52 (RR-16):1 –16[Medline]
13. Liebelt EL. Clinical and laboratory evaluation and management of children with vomiting, diarrhea and dehydration. Curr Opin Pediatr. 1998;10 (5):461 –469[CrossRef][Medline]
PEDIATRICS (ISSN 1098-4275). ©2008 by the American Academy of Pediatrics
CiteULike 
Connotea 
Del.icio.us 
Digg 
Facebook 
Reddit 
Technorati 
Twitter What's this?
This article has been cited by other articles:
|
|
 |
|
  U. Uhlig, N. Pfeil, G. Gelbrich, C. Spranger, S. Syrbe, B. Huegle, B. Teichmann, T. Kapellen, P. Houben, W. Kiess, et al. Dimenhydrinate in Children With Infectious Gastroenteritis: A Prospective, RCT Pediatrics, October 1, 2009; 124(4): e622 - e632. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
