PEDIATRICS Vol. 115 No. 4 April 2005, pp. 1108-1109 (doi:10.1542/peds.2004-2656)
Use of Serum Electrolyte Panels in Gastroenteritis
Kyung E. Rhee, MDMichael Silverstein, MD, MPH
Department of Pediatrics
Boston University School of Medicine
Boston, MA 02118
To the Editor.—
Although we appreciate the recent publication by Wathen et al1 on the role of serum electrolytes among children requiring intravenous (IV) rehydration, we disagree with the authors' conclusions. The Wathen et al study sought to determine the utility of routine serum electrolytes on children 2 months to 9 years old who were receiving IV fluids (IVFs) for dehydration. To do so, the investigators undertook a prospective study, without a comparison group, in which a serum electrolyte panel (SEP) was routinely drawn for each study subject. Treating physicians were asked whether they would have drawn an SEP under ordinary circumstances. Investigators then determined, by physician report, how the "study-imposed" electrolyte panel changed clinical management. Based on the findings that physicians could not accurately predict which children would have a clinically significant electrolyte abnormality and the fact that electrolyte results "changed management" in 
10% of subjects, the authors concluded that "routine ordering of a SEP for dehydrated children receiving IVFs not only would detect some apparently occult CSEAs [clinically significant electrolyte abnormalities] but also would affect clinical management."
However, the important unanswered question is: Would routine ordering of serum electrolytes affect clinical management adversely?
In the Wathen et al series of 182 dehydrated patients, 48% had at least 1 abnormal electrolyte value, and 10.4% had a "clinically significant" abnormality, defined by investigators as one that changed a hypothetical clinical management plan expressed before knowing the SEP results. The fundamental limitation of this study is that the investigators failed to determine if such changes in management actually benefited the patients. In each case of altered management, the change represented more aggressive treatment, longer observation, or admission. Whether these changes led to better clinical outcomes remains uncertain.
An analogy might be made between the results of this study and recent trends in the inpatient management of infants and young children with bronchiolitis. Recent studies have suggested that routine testing for oxygen saturation have resulted in prolonged hospital stays among infants with bronchiolitis.2 Longer hospital stays permit more opportunity for medical errors, unnecessary cascade testing, and potentially worse outcomes. Based on the results of Wathen et al, we are concerned that routine ordering of serum electrolytes on all children undergoing IV rehydration might have similar adverse effects among dehydrated children.
The authors also argue that their data indicate that physicians cannot predict accurately which children will have clinically significant electrolyte abnormalities. In only one third of the cases reported in the study, physicians reported that they would have ordered serum electrolytes. This hypothetical ordering carried a mere 58% sensitivity for detecting a clinically significant abnormality (again, defined as a laboratory-value abnormality that changed clinical management). We believe that this sensitivity analysis is misleading. First, performance characteristics of any test (be it a laboratory test or, as in this case, clinical acumen) must be judged in light of a "gold standard." The authors report no acceptable gold standard, which in this case would have been clinical outcomes. Furthermore, analyzing the performance characteristics of such a subjective measure as whether a physician would have ordered serum electrolytes, although potentially useful, strays from more conventional uses of sensitivity and specificity. Outcomes of a physician's recent cases, other personal experiences and education, and institutional norms likely influence a physician's decision to order serum electrolytes. A better presentation of the results might have been the predictive value of laboratory ordering, in this case the predictive value of not ordering an SEP and the subsequent lack of change in clinical management. Among all cases in which the physician reported that he or she would not have ordered an SEP, 93% of the clinical decisions were unchanged after seeing the results of the tests. Although we acknowledge that this impressive negative predictive value is driven largely by the low prevalence of clinical management changes, it suggests that physicians indeed are capable of deciding how to treat patients based on clinical acumen.
Wathen et al suggest that obtaining a routine SEP is useful among dehydrated children undergoing IV rehydration. Although this may be true, their study fails to provide the evidence to support such a conclusion. We fear that in the absence of better evidence, widespread adherence to this recommendation might lead to overly aggressive treatment, longer hospitalizations, and exposure to iatrogenic complications, as has been the case with bronchiolitis. We do not argue against the utility of serum electrolytes among some children requiring IV rehydration. However, in light of the fact that only 34% of treating physicians in this study would have ordered electrolytes, we do suggest that the question of whether such ordering should be left to the physician's discretion or instituted routinely be addressed scientifically. Until a randomized, controlled trial is done to compare physician-driven versus blanket ordering of serum electrolytes (and considers clinical outcomes), we should admit to a lack of clinical evidence for this question. Until that time, to claim that routine serum electrolyte ordering among patients requiring IV rehydration is useful may cause more harm than good.
REFERENCES
- Wathen JE, MacKenzie T, Bothner JP. Usefulness of the serum electrolyte panel in the management of pediatric dehydration treated with intravenously administered fluids.
Pediatrics. 2004;114
:1227
–1234
[Abstract/Free Full Text] - Schroeder AR, Marmor AK, Pantell RH, Newman TB. Impact of pulse oximetry and oxygen therapy on length of stay in bronchiolitis hospitalizations.
Arch Pediatr Adolesc Med. 2004;158
:527
–530
[Abstract/Free Full Text]
PEDIATRICS (ISSN 1098-4275). ©2005 by the American Academy of Pediatrics
CiteULike 
Connotea 
Del.icio.us 
Digg 
Facebook 
Reddit 
Technorati 
Twitter What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
