Trauma professionals rely heavily on diagnostic tests these days, particularly CT after blunt trauma. Apparently, the neurosurgeons at the Massachusetts General Hospital are asking for CT angiography of the brain on occasion in patients with TBI. Ostensibly, this is to rule out cases when a brain aneurysm causes a car crash or other blunt trauma.
WTF? Now, I know that we occasionally agonize over older victims of blunt trauma who come in pre-terminal or in arrest. Did they have an MI which caused the event, or did the trauma stop their heart? I had no idea that a ruptured/rupturing aneurysm was such a problem in blunt trauma.
So the surgeons at the MGH decided to critically look at this issue to see if the extra head scan was warranted.
Here are the factoids:
- 600 patients with blunt TBI over a one year study period were reviewed
- 22% underwent CT angio in addition to the normal head CT
- 66% had the CT angio immediately, 27% within 24 hours, and 7% beyond 24 hours after arrival
- Specialists who requested the study were neurosurgeons (23), radiologists (15!), neurologists (7!!)
- Reasons for getting the study: look for cause of subarachnoid hemorrhage (aneurysm) in 43, look for vascular injuries near a skull fracture in 14, rule out stroke in 4, and no particular reason in 71
- Head CTA changed management in only 1 patient, prompting a formal angiogram which was negative
- 33 patients (25%) had incidental findings on CTA, but none required any intervention in the hospital or on later followup
Bottom line: There is no value to adding CT angio of the head to the usual scan protocols. Having said that, if the patient was witnessed to lose consciousness prior to the event, and the CT shows subarachnoid hemorrhage in a more typical middle cerebral artery distribution, you might consider it to look for an aneurysm. That’s a lot of if’s. Just move the zebra off the CT scanner first. And as you can see from the last factoid above, if we scan it, we will find stuff. Fortunately, most of that stuff doesn’t need further workup or treatment.
Reference: Is CT angiography of the head useful in management of traumatic brain injury? J Am Col Surg 220(6):1027-1031, 2015.
Head CT after blunt head trauma is routine. And in many hospitalized patients, repeat head CT scan is also routine. Sometimes the routine includes many repeat CT scans. But when is the last time you’ve gotten that repeat scan on a neurologically normal patient and found “actionable” pathology? By that, I mean a finding that needs some type of intervention, not just “serial monitoring?”
An interesting paper published by neurosurgeons at McMaster University in Canada looked at the value of repeat head scans in patients with mild TBI, defined as a GCS of 13-15. I wrote about this one several months ago when it was just an abstract. Now, the full paper has been published so we can scrutinize it more closely.
The authors looked at their own experience, but also did a meta-analysis of 15 other studies in the literature. They grouped the patients into those who underwent intervention (hyperosmolar infusions, ICP monitor insertion, surgery) based on clinical findings vs findings on repeat head CT. Papers included in the meta-analysis were limited to larger studies (>30 subjects), and ones in which repeat head CT was performed and the reason for intervention was clear.
In their own series, they identified 445 patients who underwent repeat head CT. This generally occurred within 24 hours, but was done more urgently if neurologic changes occurred. Interesting findings included:
- Intracranial hemorrhage was unchanged in 80% of patients and increased in 20%
- 25 patients (6%) had a change in management after the repeat head CT
- Of these, 23 had the change based on deterioration of the neurologic exam, not the CT
- Only 2 had an intervention based on the repeat head CT ( mannitol administration due to increased edema, despite no change in exam)
- The meta-analysis showed similar findings
Bottom line: This is one of several recent studies questioning the usefulness of the “routine” repeat head CT. It’s time to work with our neurosurgeons and agree that a repeat CT is not needed in low-risk, hospitalized patients who can have regular neurologic exams. I would suggest that we limit this course of management to patients with a GCS of 15 only. Repeat head CT should still be used in patients who are on any type of medication that interferes with clotting, as these can lead to insidious bleeds. But to really make this work, we need to figure out how long the patient needs to be monitored, and the cost/benefit analysis of a longer hospital stay vs repeat scan and early discharge.
Reference: The Value of Scheduled Repeat Cranial Computed Tomography After Mild Head Injury: Single-Center Series and Meta-analysis. Neurosurgery 72(1):56-64, 2013.
Repeating the head CT in patients with head injury has almost become routine. This practice varies greatly and depends on the neurosurgeon’s preference in many centers. It occurs most commonly when there is blood inside the skull, any type of blood. But does this practice make sense? Sure, if you’ve got a small epidural it seems reasonable. But what about that wisp of subarachnoid blood?
Another paper being presented at the Congress of Neurologic Surgeons this fall describes a review of 445 cases at their hospital and a meta-analysis of 15 studies in the literature. The authors looked at the practice of repeat CT scanning with respect to the good old clinical exam. They stratified all patients who underwent an intervention after repeat CT into two groups, based on changing clinical exam or CT findings.
They found that a significant number of patients required some management change based on deteriorating neurologic exam, whereas very few required it based on the repeat CT. The authors concluded that it is not necessary to rescan a head trauma patient if their neurologic status is stable or improving.
Bottom line: This is preliminary data, and it is only available in abstract form, so don’t change your practice yet. We need more information on how many patients were reviewed and how good the meta-analysis was. However, you should begin to question whether rescanning everyone is necessary or prudent. Not all head injuries are alike, and some of the trivial ones, like subarachnoid blood in a young adult probably don’t need a repeat scan. More to come when this is presented and/or published.
Reference: The Value of Scheduled Repeat Cranial Computed Tomography Following Mild Head Injury: Single Center Experience and Meta-Analysis. Paper 152, presented at the Congress of Neurologic Surgeons, October 2012.
Emergency physicians and trauma surgeons routinely assess patients with potential neurotrauma and decide whether to obtain CT scans and/or neurosurgical consultations. The criteria they use to make these decisions are not always clear.
The neurosurgery department at the University of California – Davis performed a prospective study that looked at the appropriateness of consults they received and of CTs of the head ordered by other physicians in trauma and non-trauma patients. A total of 99 patients entered the study (32 head trauma, 29 spine trauma, 34 other disease, 4 not documented).
After reviewing the consultations, they found that 69 were appropriate, 32 were not appropriate, and 7 could not be classified. Additionally, they felt that 10 of the head CTs in injured patients (31%) were not indicated.
“Appropriateness” was difficult to define well in this study, and there is certainly a great deal of subjectivity involved. The authors recommend using the Canadian CT Head Rule to fine-tune use of head CT in trauma patients.
The bottom line: 1 in 4 consults were not appropriate, and 1 in 3 head CTs were not indicated. Despite its flaws, this study shows that we need to be better at evaluating our patients to reduce unnecessary consults and radiation!
Reference: (In)appropriate neurosurgical consultation. van Essen et al. Clinical Neurology and Neurosurgery. In press, for publication 10/2010.