The Scenario

It is gastro season. There is vomiting and diarrhea everywhere. Just after seeing your fifth patient with the aforementioned symptoms a colleague asks you about a patient that they recently saw with, you guessed it, vomiting and diarrhea. They noted that the patient only had mild dehydration and did a great job taking adequate oral fluids after some coaching on not allowing the child to chug. However, the parents were still concerned that their child would continue to have symptoms and thus become severely dehydrated. Therefore, the physician ordered a normal saline bolus and then discharged the child home. They admitted that they felt that hemodynamically the bolus was not warranted but felt inclined to do so because of parental concerns. On the following day the family returned with their child, still with symptoms of gastroenteritis, but also still able to take oral fluids. Again, the patient was well hydrated on exam. Your colleague bus wondered whether or not the bolus was, in a way, responsible for the parents returning. They hypothesized that the parents saw the treatment as definitive therapy, and when symptoms continued they felt that therapy had failed and therefore packed up the car and headed back to the ED.

After a dramatic pause……

You produce the following article.

Emergency Department Revisits in Children With Gastroenteritis: a Retrospective Observational Cohort Study
Freedman et al.
J Pediatr Gastroenterol Nutr. 2013 Jun 30. 

The authors of this study endeavored to determine whether or not administration of IV fluids in the ED for kids with gastro were independently associated with a reduction in unscheduled emergency department (ED) revisits within 7 days. The designed a retrospective cohort, and found 3,346 patients under 18 years of age diagnosed with gastroenteritis. 497 received IV fluids (15%) and 543 returned to the ED (16%). Linear regression analysis was only done on 2,874 children (as the remainder had missing data). The authors identified the following 5 independent predictors of a return ED visit:

    1. intravenous rehydration (OR = 1.76; 95%CI: 1.36, 2.26)
    2. number of vomiting episodes (1.20; 95% CI: 1.04, 1.28/5 episode increase)
    3. days of diarrhea (OR = 0.92; 95%CI: 0.88, 0.97/day increase)
    4. frequency of diarrhea (1.19; 95% CI: 1.03, 1.38/5 episode increase)
    5. age (OR = 0.94; 95%CI: 0.91, 0.98/year)

 

Will this change my practice in any way?

I won’t be any more or less likely to order fluids because of this article. What I think it highlights for me, and what I’ll take back to my interaction with residents, nurses and families is that we need to clearly explain why we do what we do. Do a good physical exam and look for signs of dehydration. Give fluids if the child is hemodynamically compromised, or if they fail PO challenge. Use ondansetron as warranted. Don’t promise that fluids will make it all better – or at least imply as such by omitting the all important discharge education discussion topics:

  • The symptoms will continue
  • You still need to encourage PO intake
  • Your child is dehydrated if (provide specific signs for the parents to watch for) and you should return if…

 

I do feel that this study was limited by the fact that almost 15% of its patients were not included in the linear regression analysis. This certainly could have changed things were the missing patients included. An interesting companion article by Sturm et al. in the Annals of Emergency Medicine in 2010 noted that patients who receive ondansetron were more likely to return to the pediatric ED within 72 hours (OR 1.45; 95% CI 1.27 to 1.65) and be admitted on that return visit (OR 1.74; 95% CI 1.39 to 2.19). Interestingly their initial aims of their study were to see whether or not ondansetron was masking an alternative diagnosis in children with vomiting. It too, is well worth a read.

Here’s a little bonus video for those of you with pop culture addled minds…