by Nicholas Otts, MD
Edited by Gal Altberg, MD and Abbas Husain, MD
A Common Problem:
9 month female presents to an ED with vomiting, diarrhea, and decreased PO intake at home.
Is this child dehydrated? What PE signs am I relying on? What urine or lab values matter?
A Small Dose of EBM:
A good systematic review suggested that the most useful individual physical exam findings for predicting 5% dehydration are abnormal capillary refill time, abnormal skin turgor, and abnormal respiratory pattern (1).
However, a combination of PE findings is better than any of the above individual signs; also, this combination can include other common PE findings such as dry mucous membranes, sunken fontanelle, etc.
As for labs (BUN, BUN/Cr ratio, etc.), serum bicarbonate is best, but only as a tool to help you rule out dehydration (if greater than 15 or 17) (1,2,3).
Further, all urine values (specific gravity, ketones, output, etc.), which some use as POC testing, have been found to be not helpful (1,4).
A Brighter Future:
Since multiple studies have suggested a normal bicarb as a method of ruling out moderate dehydration (and thus, ruling out the need to establish an IV in a dehydrated patient), a brighter future may be found in POC testing in the EM. Yale did a recent study establishing significant cost savings of point of care (i-Stat) testing of serum electrolytes (including bicarb) vs. traditional lab work in pediatric patients with gastroenteritis (5). If this kind of technology becomes more widespread, PEM attendings may be able to avoid many IV placements in future patients with gastroenteritis.
An explanation. Let’s tackle some research on the subject.
There is a wealth of research on pediatric dehydration, but few studies use a valid gold standard. The gold standard for dehydration is weight loss, but it is impossible to have an accurate weight change in a patient when we aren’t measuring their well weight just before they get sick.
Thus, good studies use a comparison of the rehydration weight to the acute weight as an estimation. (rehydration weight - acute weight) / rehydration weight.
There was a systematic review from JAMA in ‘04 of pediatric dehydration that used the above standard in its analysis. It identified studies from 1966-2003 on this topic (1603) and eliminated those that did not meet the above gold standard for analysis, among others. It created tables of the LR (likelihood ratios) all the common PE findings and lab values that could be used to assess dehydration (1).
Some bottom line numbers: abnormal cap refill time LR 4.1, 95% CI 1.7-9.8, abnormal skin turgor LR 2.5, 95% CI 1.5-4.2, and abnormal respiratory pattern LR 2.0, 95% CI 1.5-2.7) (1).
Labwork appears to have some value if grossly abnormal ( such as BUN/Cr > 45) but the bicarb was the only value that had good data (LR .18 to .22 for ruling out dehydation) (1,2,3)
Further, for those that like using a U/A to rule in or out the need for an IV placement, a prospective cohort study in ‘07 also used the above gold standard and looked specifically at urine studies--specific gravity, urine ketones, and output--and did not find predictive value for dehydration (4)
As always, there are limitations in every study, and even the gold standard used for analysis of dehydration (with the rehydration weight), is not ideal. But it is helpful to create discussion from the current literature in those that will assess this kind of patient on every shift.
1. Steiner M, DeWalt D, Byerley, J. Is this Child Dehydrated? JAMA, June 9, 2004 - Vol 291, No.22
2. Vega RM, Avner JR. A prospective study of the usefulness of clinical and laboratory parameters for predicting percentage of dehydration in children. Pediatric Emergency Care. 1997; 13:179-182.
3. Yilmaz K, Krabocuoglu M, Citak A, Uzel N. Evaluation of laboratory tests in deydrated children with acute gastroenteritis. J Paediatr Child Health. 2002; 38: 226-228.
4. Steiner, M, Nager A, Wang V. Urine Specific Gravity and Other Urinary Indices: Inaccurate Test for Dehydration. Pediatric Emergency Care, Vol 23, Number 5, May 2007
5. Whitney R, Santucci K, Hsiao A, Chen L. Cost effectiveness of point-of-care testing for dehydration in the pediatric ED. American Journal of Emergency Medicine 34 (2016) 1573-1575.