Criteria abstracted from The Users' Guide to Medical Literature, from the Health Information Research Unit and Clinical Epidemiology and Biostatistics, McMaster University

Highlighted lines and questions below provide links to the pertinent description of criteria in The EBM User's Guide, now available at the Canadian Centres for Health Evidence

Article Reviewed:

Please visit the new Evidence Based Journal Club Reviews

Does an isolated history of loss of consciousness or amnesia predict brain injuries in children after blunt head trauma?

Palchak MJ, Holmes JF, Vance CW, et al.

Pediatrics. 2004 Jun;113(6):e507-13 [abstract]

Reviewed by Katri Typpo MD, Pediatric Critical Care Section, Department of Pediatrics, Baylor College of Medicine, Houston TX

Review posted January 13, 2005

I. What is being studied?:

The study objective:

To determine whether an isolated loss of consciousness (LOC) and/or amnesia is predictive of traumatic brain injury (TBI) in children with blunt head trauma.

The study design:

A prospective observational cohort design.

Prospective enrollment of children < 18 years old presenting to a level I trauma center ED during a 38 month time period, with blunt head trauma who had history or physical examination findings consistent with head trauma. History of LOC, amnesia, seizures, vomiting, current headache, dizziness, nausea, vision change, abnormal mental status, focal neurologic deficits, clinical signs of skull fracture, and scalp trauma were all felt to be consistent with head trauma.

The outcomes assessed:

They primary outcome assessed was the presence or absence of TBI. TBI was defined as either: TBI identified on head CT, or TBI requiring intervention. All head CT's were read by a faculty pediatric radiologist. TBI requiring intervention was defined by several features. The presence of one or more of the following was adequate: need for neurosurgical procedure, antiepileptic medication (AED) beyond 7 days, the presence of a persistent neurological deficit, or more than 2 nights of hospitalization. The charts were then reviewed and patients followed up by phone to evaluate the association between isolated LOC/amnesia with the presence of TBI.

II. Are the results in the study valid?

Primary questions:

1. Was there a representative and well-defined sample of patients at a similar point in the course of the disease?

Yes. Of 2043 eligible children, 77% were enrolled. They excluded children with head trauma resulting from ground level falls or from walking or running into stationary objects if the only finding was a scalp laceration or abrasion. While this seems reasonable, they may have missed the small population of children who do have TBI with relatively insignificant trauma. These children often have a previously unrecognized medical problem. I do think this population should be included as they could present with a history of LOC and therefore would be pertinent to the aims of this study.

2. Was follow-up sufficiently long and complete?

Yes. The follow up was sufficiently long, 7 days after ED evaluation is sufficiently long to evaluate for clinical deterioration and the need for further intervention with TBI. It is also sufficiently short to minimize recall bias. Patients were either followed up by telephone or with a written questionnaire; 88% were able to be contacted. At study completion, the county morgue records and hospital trauma registry were reviewed for the names of patients who were not reached by telephone or did not respond to the questionnaire Ð no cases of missed TBI were identified in this remaining 12%.

Secondary questions:

3. Were objective and unbiased outcome criteria used?

Yes and no. The presence of TBI on CT scan is an objective and unbiased outcome measure. CT scans with equivocal readings were given to a pediatric radiologist blinded to clinical information for a definitive reading. However, not all children with LOC and/or amnesia underwent CT scan, so one of the outcomes could have been missed in some patients.

Overall, I felt that the clinical criteria were as unbiased and objective as they could be. They used a convenience sample to assess interobserver differences in clinical variables. Interobserver agreement for both history of LOC and amnesia were high with kappa scores of 0.86 and 0.63, respectively.

It is difficult to define clinical criteria that are objective and unbiased given that they depend on an individual physicians' judgment. The need for neurosurgical intervention is fairly unbiased and objective. Surgeons are unlikely to operate without clearly defined pathology. On the other hand, hospitalization for 2 or more days and the use of AED's for greater than one week may at times be the result of variations in practice. Some physicians may not feel comfortable stopping AED's and will wait for subspecialty follow up to do so. This could result in a child being on an AED for more than 7 days. Some physicians may hospitalize children with isolated skull fractures despite the fact that the literature does not support the need for routine hospitalization. Others might observe a child in the hospital for more than 2 days out of concern for an occult TBI. Even within a single institution, there is often a large variation in practice.

4. Was there adjustment for important prognostic factors?

No.

III. What are the results?

1. How large is the likelihood of the outcome event(s) in a specified period of time?

For patients with isolated LOC and/or amnesia, 0 of 142 had TBI on CT scan, and 0 of 164 had TBI in need of acute intervention.

However, TBI was clearly more common in patients with LOC and/or amnesia (not as isolated findings). 9.4% of patients with LOC and/or amnesia had TBI on CT while only 2.7% of those without this history had TBI on CT (difference of 6.7%, 95% CI 4.1, 9.3). From these data we can calculate sensitivity, specificity and likelihood ratios (LR). The sensitivity is 0.86 (95% CI 0.79, 0.94), specificity 0.37 (95% CI 0.34, 0.40) and the LR (of a positive CT) is 1.3 (95% CI 1.25, 1.52). Although significant, this is only weakly suggestive. The post-test likelihood of TBI on CT given no LOC and/or amnesia is ~2%.

2. How precise are the estimates of likelihood?

The confidence intervals they provide are very precise. For patients with isolated LOC, the confidence interval for TBI on CT scan was 0%-2.4%, and the confidence interval for TBI based on the need for acute intervention was 0% to 2.2%. For patients with an isolated LOC and/or amnesia, the confidence interval for TBI on CT scan was 0% to 2.1%, and the confidence interval for TBI based on the need for acute intervention was 0% to 1.8%.

IV. Will the results help me in caring for my patients?

1. Were the study patients and their management similar to my own?

Yes. The study population is similar to my own in that it is a pediatric ED and hospital. Our institution has similar guidelines for the evaluation of children who present with a head injury. Often a history of LOC is enough to warrant pursuit of a head CT or observation at home vs. hospital. The American Academy of Pediatrics has guidelines for the evaluation of children > 2 years of age with head trauma. These guidelines include a provision to use head CT vs. observation in the population of kids with LOC and/or amnesia. (1)

2. Will the results lead directly to selecting or avoiding therapy?

Yes. I will be far less likely to order a head CT in a child who presents with an isolated history of LOC or amnesia. I will also pay closer attention to those children who have LOC or amnesia with other signs or symptoms as they are more likely to have a TBI. This may have implications for changing the current AAP guidelines to avoid head CT in children with isolated amnesia or LOC.

3. Are the results useful for reassuring or counseling patients?

Yes. You can counsel patients that for children who have isolated LOC or amnesia there is really minimal risk for TBI.

I think that this is reassuring to parents. It is still necessary to have parents observe their child at home according to AAP guidelines because the risk for TBI will rise if their child begins to have other signs or symptoms of TBI.

References:

1. The management of minor closed head injury in children. Committee on Quality Improvement, American Academy of Pediatrics. Commission on Clinical Policies and Research, American Academy of Family Physicians. Pediatrics. 1999 Dec;104(6):1407-15.[abstract]

 

Comments

Back to the EB Journal Club Index