Should all labs be using spectrophotometry?
A: Ensuring that patients with a sudden severe headache do not have a subarachnoid hemorrhage—a type of hemorrhagic stroke—is a priority for emergency physicians. While relatively few patients turn out to have a subarachnoid hemorrhage, the consequences of a missed diagnosis for those who do often include severe morbidity or death. Fortunately, computed tomography is increasingly able to diagnose most patients with subarachnoid hemorrhage with a sensitivity of 92.9% overall and 100% when the imaging is conducted within 6 hours of headache onset (BMJ 2011;343:d4277). However, the sensitivity for subarachnoid hemorrhage is much less—85.7%—when imaging is delayed. In this setting, further analysis is indicated, including lumbar puncture to assess cerebrospinal fluid for xanthochromia, a yellow discoloration.
Xanthochromia is assessed visually, or by spectrophotometry. Rarely used in North America, spectrophotometry is the standard of care in the United Kingdom and many European hospitals. Most North American hospitals assess both the red blood cell count and visual xanthochromia. While there are few false positives for visual xanthochromia, red blood cell counts could be elevated due to a traumatic spinal tap, in which the blood is secondary to the procedure.
There are at least four different definitions of positive spectrophotometry, including: Traditional, an optical density >0.023 at a wavelength of 415 nm; and Chalmers and Kiley, a net bilirubin absorption >0.015 positive, 0.010–0.015 borderline using absorbances at 415 nm and 440 nm relative to a baseline joining absorbances at 530 nm and 360 nm. The Chalmers revised definition uses an optical density >0.014 at 476 nm, while the United Kingdom National External Quality Assurance Service (UK NEQAS) is based on net bilirubin and oxyhemoglobin absorbances at 476 nm and 415 nm, respectively, relative to a baseline joining the 530 nm and 360 nm absorbances.
We previously conducted a study assessing these four definitions of spectrophotometry (Stroke 2006;37:2467–72). The UK NEQAS definition performed best in terms of sensitivity and specificity, 100% and 83%, respectively. However, this still would have resulted in an increase in cerebral angiography of 254% versus our study’s baseline angiography rate of 5.9%. A recent systematic review and meta-analysis compared visual xanthochromia to spectrophotometry xanthochromia as defined by UK NEQAS (Ann Emerg Med 2014;64:256–64). These authors found the sensitivity and specificity of visual xanthochromia—without considering red blood cell counts—to be 83% and 96%, respectively. Spectrophotometry had a sensitivity of 87% and a specificity of 86%.
Because the prevalence of aneurysmal hemorrhage in these patients approaches 1%, which is similar to the rate of detection of incidental aneurysms in the population at large, spectrophotometry would be expected to have a real-world positive-predictive value of about 5%. In other words, the test would be falsely positive in 19 of 20 patients. One can add this to the list of reasons why many clinicians are reluctant to embrace spectrophotometry.
Prior to incorporating a technically difficult diagnostic test involving an expensive machine which reports a result that is both labile and difficult to interpret, we need better evidence that this will translate into improved care and outcomes. Spectrophotometry advocates need to demonstrate clearly the benefits this test offers for the CT-negative headache patient with visually clear and colorless CSF with low number of red blood cells. Supporters of this technology also need to show how its benefits offset the unnecessary angiography and inevitable identification of asymptomatic aneurysms in patients with otherwise benign headaches.
Jeffrey J. Perry, MD, MSc (Epi), CCFP-EM, is an associate professor of emergency medicine and research chair in emergency neurological research at the University of Ottawa, Ontario, Canada. He is also an emergency physician at The Ottawa Hospital and a senior scientist at The Ottawa Hospital Research Institute.
+Email: [email protected]