Blunt Head Trauma Imaging Clinical Decision Rules
Search Strategy: You log on to your library proxy from the PC at your work station. First you check the “Conditions” headings in Clinical Evidence, hoping for a nice review article on the current standard of care. “Head Injury” (or a similar subject) is not included, although “Wrinkles”, “Halitosis”, and “Ear Wax” are listed. You then search MD Consult looking for “Head Injury” with a focus on diagnosis. This yields 206 journal matches. Limiting this search to adult studies after the year 2000 reduces the number to 163 matches. Further attempts to limit the search all produce an insufficient number of articles, so you move on to Pub Med. A diagnostic search for “Head Injury” under Pub Med’s Clinical Queries yields 389 references and combining this search with one for “Computed Tomography” produces 93 references, a number through which you are happy to search and find several relevant articles:
After volunteering for a disaster medical team you are dispatched without delay to serve in a field hospital in downtown New Orleans near Charity Hospital, which has not yet reopened to serve the public near the recently hurricane ravaged French Quarter.*
You are on night duty in the field hospital when several patients are brought to you from a bar fight, all with head injuries. The patients are all attendees of a medical conference and include: a Canadian, who smells of alcohol but does not appear intoxicated; an American, who appears intoxicated; and a Mormon, who hasn’t been drinking. The details of each patient’s history and exam are listed separately. In order to obtain a CT scan you must transfer each patient to Baton Rouge, a 1½ hour ambulance drive away. (Helicopters are not flying because of fear of being shot at by disturbed citizens.)
Hoping to reduce the utilization of scarce medical resources, yet protect the health of your distinguished colleagues, you look to the recent medical literature to help you decide who to send for CT scanning.
First years: Indications for computed tomography in patients with minor head injury. N Engl J Med. 2000 Jul 13;343(2):100-5. (New Orleans Head CT Rules)
Second years: The Canadian CT Head Rule for patients with minor head injury. Lancet. 2001 May 5;357(9266):1391-6. (Canadian Head CT Rules)
Third years: Comparison of the Canadian CT Head Rule and the New Orleans Criteria in patients with minor head injury. JAMA. 2005 Sep 28;294(12):1511-8.
Fourth years: Developing a Decision Instrument to Guide Computed Tomographic Imaging of Blunt Head Injury Patients. J Trauma. 2005 Oct;59(4):954-9. (Nexus II)
Bonus: External validation of the Canadian CT Head Rule and the New Orleans Criteria for CT scanning in patients with minor head injury. JAMA. 2005 Sep 28;294(12):1519-25. (Optional for all, including attendings)
Article 1: Indications for Computed Tomography in Patients with Minor Head Injury, NEJM 2000; 343: 100-105
Article 2: Clinical Prediction or Decision Rule The Canadian CT Head Rule for Patients with Minor Head Injury, Lancet 2001; 357: 1391-1396
Article 3: Clinical Prediction or Decision Rule Comparison of the Canadian CT Head Rule and the New Orleans Criteria in Patients with Minor Head Injury, JAMA 2005; 294: 1511-1518
Article 4: Developing a Decision Instrument to Guide Computed Tomographic Imaging of Blunt Head Injury Patients (NEXUS II), J Trauma 2005; 59: 954-959
Bonus Article: External validation of the Canadian CT Head Rule and the New Orleans Criteria for CT scanning in patients with minor head injury. JAMA. 2005 Sep 28;294(12):1519-25. (Optional for all, including attendings)
We assessed the three available Clinical Decision Rules: New Orleans, Canadian, and NEXUS 2. Here is the bottom line for each of them:
derived and validated at single hospital among patients who suffered a loss of consciousness with an initial GCS of 15, thus limiting one’s ability to generalize findings to other populations. The authors failed to utilize accepted methods for the development of a CDR and utilized a surrogate outcome measure of positive head CT rather than clinically significant intracranial injury resulting in a rate of operative intervention of only 0.4% (6 patients) among the entire cohort. The very subjective risk factor of “trauma above the clavicles” would likely increase the number of head CT’s performed at many institutions if liberally applied to all blunt head injury patients. The seven-item rule appears simple to remember and apply with an excellent negative Likelihood Ratio (zero) in this derivation study, but requires multi-center validation before widespread use.
derivation study in Canadian hospitals among patients who suffered a loss of consciousness or definite amnesia with an initial GCS above 13, so unable to generalize findings to other populations, or even to the same group of Canadians, without subsequent multi-center prospective validation. A logical rule consisting of five high risk factors and two additional intermediate risk factors which require at least two hours of observation in the ED to fully assess. The rule has an excellent negative Likelihood Ratio, but a useless positive Likelihood Ratio, so the rule is helpful to lower the post-test probability of a clinically significant intracranial injury in the absence of all seven risk factors with little effect on raising the post-test probability if one of the risk factors is present.
very large, multicenter derivation study utilizing established CDR methods, although the selection of candidate variables is obscure and seemingly incomplete. NEXUS II provides a simple eight-variable rule of items immediately available in the ED to assess blunt head trauma patients of all ages for whom head CT scanning is being contemplated by emergency physicians. Similar to the New Orleans Head CT Rule, the seemingly nebulous scalp hematoma would likely increase the number of head CT’s performed at many institutions if applied independent of baseline clinical intuition. NEXUS II lacks the ability of the Canadian rule to lower post-test probability of clinically significant intracranial injury. For example, if one’s pre-test probability was 10%, utilizing NEXUS II would lower the post-test probability to 1.4% while the Canadian Rule would lower it to 0.3%.