We continue on with the series with Knowledge Bomb #10. The purpose and motivation for this series is outlined in the first entry. This entry we discuss the use of a CT only approach for patients with symptoms concerning for subarachnoid hemorrhage.
Differentiating benign, though uncomfortable, headaches from those that may represent more dangerous pathologies is a daily practice in most EDs. The motivation for this edition of the Knowledge Bomb series stemmed from a string of shifts during which this became a recurring theme
At the conclusion of these shifts I came away with increased lumbar puncture proficiency, mostly out of concern for possible subarachnoid hemorrhage. But was this necessary? With the CT scanners used today and the right patient do we still need CSF?
Sensitivity of Early Brain Computed Tomography To Exclude Aneurysmal Subarachnoid Hemorrhage: A Systematic Review and Meta-Analysis. Nicole M. Dubosh, M. Fernanda Bellolio, Alejandro A. Rabinstein, and Jonathon A. Edlow. Stroke. 2016;47:750-755. PMID:26797666
This was a systematic review and meta-analysis seeking to calculate pooled sensitivities, specificities and likelihood ratios for CT scanning done within 6 hours of patient presentation. It only included studies in which CT scans were read by attending radiologists and performed with modern-day CT scanners. It excluded patients with traumatic insult, those patients less than 15 years of age and patients with headache for longer than 6 hours.
The authors calculated the following values based on the above studies included in their meta-analysis:
- Pooled sensitivity: 0.987 (95% CI, 0.971-0.994)
- Pooled specificity: 0.999 (95% CI, 0.993-1.0)
- Pooled negative LR: 0.010 (95% CI, 0.003-0.034)
- Negative CT within 6 hours changes post-test probability to 0.2%
- Miss rate of 1-2 per 1000 cases
The authors of this meta-analysis argue that given the correct patient population the use of CT scan alone is sufficient for excluding subarachnoid hemorrhage.
As emergency department providers we must be confident that we have gone far enough to rule out such a dangerous condition. I think this comes down to a matter of defining levels of acceptable risk within our clinical practice and joint decision-making with our patients.
For me, given a patient who seems to understand their situation after an honest conversation of the risks and benefits of each course of action, I am more likely to put away my needle.
Dr. Bradley Davis is a current PGY2 at UIC’s EM Program.
Click to Read Attending Response - Dr. Adeeb Zaer
Although a systematic review is tempting to follow, and avoiding an LP with a paper to back you up is even more enticing, it is not the end of the discussion. As of right now, ACEP has a grade B recommendation that an LP be done after CT for evaluation of sudden onset severe headache. This is grounds to say that if an LP is not recommended to the patient after the CT scan, even if within 6 hours, you are not complying to the standard of care for ED physicians throughout the country. If a bad outcome were to result, you would have no defense.
This article does not change my practice of recommending an LP after the CT, but does better help me to inform the patient of the risks, benefits, and likely statistics of a pathologic outcome found from the procedure.
Dr. Adeeb Zaer is an Attending physician and newly appointed Education Director at Mercy Hospital – one of the hospital sites where UIC residents rotate.
- Sensitivity of Early Brain Computed Tomography To Exclude Aneurysmal Subarachnoid Hemorrhage: A Systematic Review and Meta-Analysis. Nicole M. Dubosh, M. Fernanda Bellolio, Alejandro A. Rabinstein, and Jonathon A. Edlow. Stroke. 2016;47:750-755.