Skip to main content

24-09-2013 | Stroke | Article

Clinical SAH decision rule refined


Free abstract

medwireNews: Researchers have consolidated their previously published decision rules for subarachnoid hemorrhage (SAH) into a single rule with 100% sensitivity.

The team previously developed three decision rules for investigating SAH in the emergency department. In the current study, which appears in JAMA, they tested the rules in a new cohort of 2131 patients with possible SAH and found that they were highly sensitive, at 95.5% to 98.5%.

But Jeffrey Perry (University of Ottawa, Ontario, Canada) and team aimed to achieve a rule with perfect sensitivity, reasoning that such a rule “might have greater clinical acceptance than a rule with lower sensitivity.”

Adding two items (thunderclap headache and limited neck flexion) to one of the original rules resulted in 100% sensitivity, but at the expense of very limited specificity, at 15.3% compared with 27.6% to 35.6% for the original three rules.

The resulting new rule, called the Ottawa SAH Rule, requires physicians to further investigate for SAH if a patient has one or more of the following variables: age of at least 40 years; neck pain or stiffness; witnessed loss of consciousness; onset during exertion; thunderclap headache; and limited neck flexion when examined.

Of note, 54 patients proved to have serious conditions other than SAH, and the Ottawa SAH Rule classified all but four of these patients (with ischemic stroke, meningitis, and tumors) in the high-risk group.

In an accompanying editorial, David Newman-Toker (Johns Hopkins University School of Medicine, Baltimore, Maryland, USA) and Jonathan Edlow (Harvard Medical School, Boston, Massachusetts, USA) agree that a rule with 100% sensitivity has “an intrinsic appeal,” but caution that it still requires further validation. At a conservative estimate, using the 95% confidence interval, the rule might only be 97.2% sensitive with a negative likelihood ratio of 0.39.

They say that the “high-stakes scenarios” that prompt development of decision rules for uncommon but potentially lethal disorders “require both a highly accurate rule and an ineffective, risky, or cumbersome diagnostic alternative.”

However, the current means of assessing the likelihood for SAH – computed tomography followed by lumber puncture for unclear cases – is itself a reasonable approach, although the editorialists note that lumber puncture is not always done where indicated.

Finally, they point out that value of a rule in general practice is dependent on its accurate application by physicians and the avoidance of “rules creep,” in this case most likely a too-liberal interpretation of sudden-onset (thunderclap) headache.

medwireNews ( is an independent clinical news service provided by Springer Healthcare Limited. © Springer Healthcare Ltd; 2013

By Eleanor McDermid, Senior medwireNews Reporter


Related topics