Sepsis and meningitis are rare but feared diagnoses in children, especially when they are unable to provide any meaningful history. Based upon available evidence, it can be assumed that delayed diagnosis in a dangerous condition would lead to worse outcomes. Children under the age of one month are almost universally treated as if they are immunosuppressed for this reason, and routinely are admitted and managed as an inpatient with antibiotics empirically. Children who are outside this age range, yet remain non-verbal or have limited abilities to express themselves pose a clinical challenge, and evidence for management in this age group is limited, as these children are susceptible to many common, limited, self-resolving illnesses that may mimic these diagnoses. The question that this article hopes to address is how often children are found to have something benign, but are later found to suffer from a more dangerous, potentially deadly illness, and what the consequences of this assumption may be.

Clinical Question

How often do children who are diagnosed with septicemia or meningitis have prior ED visits within 5 days to their ultimate diagnosis and hospitalization?


Children 30 days to 5 years diagnosed with meningitis or septicemia who were hospitalized or died.


Primary:Frequency of having a related ED visit prior to diagnosis with septicemia or meningitis.
Secondary:Length of stay, critical care use, mortality


Retrospective cohort study looking at administrative data collected in Ontario, Canada.


• Age less than 30 days or older than 5 years.
• Admitted for less than 4 days.
• Inpatients within the past 14 days for any reason.
• Admission diagnoses reviewed, and cases excluded if deemed unrelated (11 cases).

Primary Results

Critical Findings:

  • 1 in 5 patients (114/521) had a prior related ED visit within 5 days of their ultimate diagnosis and admission.
  • There was no difference in mortality, hospital length of stay or critical care usage between those with immediate admission and those with admission on second presentation.

Other Results

  • >96% of those discharged came back to the ED within 72 hours of discharge, with an average time of 24.5 hours
  • 1/3 of patients went to a different emergency department on their repeat visit.
  • No child discharged from an ED died in the hospital
  • 6 children who were admitted on their first ED visit died while in the hospital.
  • The 30-day mortality was similar, but because of confidentiality, further information was withheld to protect patient confidentiality.
  • Those discharged were
    • Older (17 months vs. 11 months old) and had lower initial triage severity scores.
    • More likely to be seen in community Emergency Departments without pediatric consultations available.


  • 5 years of retrospective data encompassing nearly 2.4 million ED visits.
  • Hospitals included small community EDs staffed only by family medicine doctors and those with EM fellowships, as well as large academic centers with emergency medicine and pediatric emergency medicine specialists.


  • Low mortality – only 6 of 521
  • Not linked to death registry
    • Theoretically, children who died and never presented to an emergency department were not included in the study.
  • Data was anonymized administrative data, so no clinical information available.
  • Visits to non-ED providers were not captured (ie clinics, urgent care)

Author's Conclusions

“In this cohort, repeated ED visits among children with meningitis or septicemia were common, yet they had health outcomes similar to those of children admitted on initial visit.”

Our Conclusions

Meningitis and sepsis are rare, and while misses may occur, it is more likely that patients who are treated and discharged, then later “bounce back” and are diagnosed with these diseases, did not have these diseases upon initial presentation. Instead the initial presentation may merely represent something less severe that allows septicemia or meningitis to develop.

In a large Western population sample, no difference in patient-centered outcomes was found amongst patients discharged with a similar appearing initial presentation. It appears that current practice does a good job at identifying those children who have a serious bacterial illness. While it is possible that in some cases, the diagnosis was missed initially, it is more likely that these children did not have septicemia or meningitis, but instead developed it and subsequently presented to the Emergency Department.

Potential Impact To Current Practice

This article is more reassuring than practice changing, as it appears that providers do a good job of identifying critically ill children. Indeed, it does not appear that any specific screening test would add anything to diagnosis, as these diseases are rare, and it appears that careful examination may be the best tool available.

Bottom Line

In this large, retrospective cohort study, 1 in 5 kids who were diagnosed with a serious bacterial infection (sepsis or meningitis) presented to an ED and were discharged within 5 days of their ultimate diagnosis. However, there were no differences in outcomes between the group that was recognized on 1st presentation and those that were initially discharged.

Read More

Green SM et al. Sick Kids Look Sick. Ann Emer Med 2015 Jun;65(6);633-5. PMID: 25536869

Don’t Forget the Bubbles: Consequences of Missing Meningitis or Septiceamia on First Presentation to ED