figure_sick_by_toilet_12153A 6-year-old male is brought to the emergency department (ED) after falling from the monkey bars at a local playground. Physical examination reveals no scalp hematoma, and the child appears alert and well oriented. You decide to observe him over the next 30 minutes hours to determine if he develops any disconcerting symptoms. After 15 minutes of observation within the ED the patient has an episode of vomiting witnessed by the nurses. The patient’s mother wants to know if this means he has failed his observation period and needs to receive a head CT. Your answer?


EDs frequently evaluate children with blunt head trauma accounting for more than 450,000 annual ED visits. Due to increased awareness of radiation risks extrapolated from population based studies, particularly to the pediatric cohort, a great deal of attention has been given to decision instruments designed to reduce the rates of CT scanning in children. The most widely received and utilized rule, the PECARN rule,1 allows one to clinically classify risk based upon distinct and measurable clinical parameters. The instrument functions as a whole, and each branch point is reliant upon the interconnected network of variables that form the rule as derived. Since its derivation, however, questions have arisen regarding the individual variables most commonly encountered in clinical practice, such as vomiting. So Dayan et al from the PECARN research group looked to determine if the isolated variable of vomiting independently predicted traumatic brain injury (TBI) in children.2

Study Objectives

Determine the association between children presenting with vomiting and traumatic brain injury with blunt head trauma

Study Methods

  • Secondary analysis of prospective observational cohort
  • Patients
  • Exclusion criteria
    • Ground-level falls
    • Running into stationary objects
    • Penetrating head trauma
    • Preexisting neurological disease
    • Syncope/seizure
    • Bleeding disorders
    • VP shunts
  • Isolated vomiting was defined using the “extensive definition” (Figure 1) rather than the PECARN definition
  • Outcome measures
    • Any TBI on CT
    • Clinically important TBI (ciTBI)
      • Death, neurosurgical procedure, intubation for 24 hours, hospitalization for 2 or more nights


  • 5,392 patients included in the final analysis
  • 4,577 (84.9%) with non-isolated vomiting
    • Had at least 1 other symptom of head trauma based on the extensive definition
  • 815 (15.1%) with isolated vomiting, as defined by the “extensive” definition (Figure 1)
Isolated vomitingNon-isolated vomiting
Clinically important TBI2/815 (0.2%; 95% CI 0-0.9%)114/4,577 (2.5%; 95% CI 2.1-3.0%)
TBI on CT5/298 (2.5%; 95% CI 0.5-3.9%)211/3,284 (6.4%; 95% CI 5.6-7.3%)
  • Note:  The different denominators for “TBI on CT” were due to not all included patients in the final analysis actually receiving a CT.
  • Note: There was no association found between vomiting and specific clinical variables (i.e. LOC) to aid prediction.


Though this study is severely limited by its post-hoc methodology and rather disparate definition of ciTBI, it confirms what is standard practice across EDs — that is obtaining a head CT for any evidence of vomiting in blunt head trauma. Isolated vomiting as a symptom is instinctually more disconcerting than many other symptoms as providers colloquially associate its presence with an artificially high risk assessment of poor outcomes. However, this study finds that isolated vomiting (extensively defined) had an extraordinarily low risk of association with clinically relevant brain injury.

In the case of isolated vomiting, the benefit is unclear as the cohort had a total 0.2% risk of ciTBI, which is well below a reasonable testing threshold to recommend routing CT screening. Balance this with the most recent retrospective cohort study3 assessing leukemia and brain tumor risk from pediatric head CT, which estimates that one case of leukemia and one brain tumor will be caused by every 10,000 children CT-scanned. While this is an extraordinarily low malignancy risk, it needs to be counterbalanced by the purported benefit of CT in the appropriate clinical presentation should such a benefit exist.

Even with non-isolated vomiting, the aggregate risk is low enough to recommend against the routine and wonton scanning of children presenting in this fashion without alternative considerations. What seems clear from this and many previous studies is that observation remains key for true risk assessment and clinical stratification of potential disease. An observation period in the ED can serve to reduce one’s pre-test probability for TBI well below the already low disease thresholds established within this study. This would save a great many patients from an unnecessary CT and help identify those who will most benefit from the diagnostic capabilities of CT.

Take Home Points

  1. Isolated vomiting has an extraordinarily low risk of ciTBI.
  2. Non-isolated vomiting confers a higher risk, but the risk should be counterbalanced by the risk of CT in the pediatric cohort.
  3. ED observation periods may aid in further risk stratification and safely reduce CTs in children.
Kuppermann N, Holmes J, Dayan P, et al. Identification of children at very low risk of clinically-important brain injuries after head trauma: a prospective cohort study. Lancet. 2009;374(9696):1160-1170. [PubMed]
Dayan P, Holmes J, Atabaki S, et al. Association of traumatic brain injuries with vomiting in children with blunt head trauma. Ann Emerg Med. 2014;63(6):657-665. [PubMed]
Pearce M, Salotti J, Little M, et al. Radiation exposure from CT scans in childhood and subsequent risk of leukaemia and brain tumours: a retrospective cohort study. Lancet. 2012;380(9840):499-505. [PubMed]
William Paolo, MD

William Paolo, MD

Residency Program Director
Assistant Professor of Emergency Medicine
SUNY-Upstate Medical Center