Diagnostic errors lead to death or disability for an estimated 150 000 patients in the United States each year.1 The emergency department is a known high-risk location for misdiagnosis.2 Missed ischemic stroke and brain hemorrhage are recognized sources of diagnostic error, with approximately 9% of cerebrovascular events missed at first emergency department contact,3 including an estimated 20% of subarachnoid hemorrhages in patients presenting with normal mental status.4 Because effective treatments are available, diagnostic delays increase morbidity and mortality 3to 8-fold,4,5 so accurate early diagnosis is important. Ratesofmissedsubarachnoidhemorrhage in the1980sand 1990s were estimated at 32%,6 although more recent estimates suggest the rate is approximately 12%, with about half occurring in the emergency department.7 Some improvement isprobably fromnewergenerationcomputed tomography(CT),7,8butmost isnot.9 The current recommended standard of care is to obtain cranial CT for patients with new, rapid-onset, severe headaches and, for those with nondiagnosticCTswhoare still suspectedofhavingpossible subarachnoid hemorrhages, to follow with diagnostic lumbar puncture.10 This “CT-LP” rule is a proven method, with sensitivity for subarachnoidhemorrhage close to 100%whenperformed correctly.11 However, because the real-world emergency department miss rate for subarachnoid hemorrhage is approximately 6%,7 the CT-LP rule either is not applied to all at-risk patients or is used incorrectly (eg, lumbar puncture is obtained too early or too late, when spinal fluid findingsmay be misleading). In this issue of JAMA, Perry and colleagues12 seek to enhance the clinical capabilities for diagnosing subarachnoid hemorrhage through validation and refinement of the Ottawa SAH Rule. The authors present the results of a prospective, cross-sectional study involving 2131 patients with acute headacheanddemonstrate that theirbestbedsidedecisionrule identified all cases of subarachnoid hemorrhage (n = 132) among emergency department patients presentingwith new, isolatedheadaches. The final rule relies on thepresenceof any 1 of 6 findings (age ≥40 years; neck pain or stiffness; witnessed loss of consciousness; onset during exertion; thunderclap headache [instantly peaking pain]; limited neck flexion on examination) and has an estimated sensitivity of 100% for detecting atraumatic subarachnoid hemorrhage. This rule offers the potential to reduce missed subarachnoid hemorrhage and decrease unnecessary, invasive diagnostic testing for patients with low-risk headaches. Is the Ottawa SAH Rule clinically useful? Any test with near-perfect sensitivity has an intrinsic appeal because a negative result effectively rules out the target disorder. The rule proposed by Perry et al also has a “rule-out power” or negative likelihood ratio (ie, extent to which the odds of having a diagnosis will change following a negative test result) of 0.024, translating to a 42-fold reduction (ie, 1.0/ 0.024) in the likelihood of subarachnoid hemorrhage. For instance, a patient with acute headache and a pretest probability of 10% who has a negative result with the Ottawa SAH rule would have a posttest probability of 0.3% (ie, convert pretest probability [10%] to pretest odds [1:9]; multiply by 0.024 to obtain posttest odds [0.024:9]; then convert back to probability: 0.024/[0.024 + 9] = 0.00266 = 0.3%. For clinicians uncomfortable converting probability to odds and back, pocket-card nomograms provide a simple graphical interpretation of pretest and posttest probabilities using likelihood ratios). This is a clinically useful result because “very low” residual risks (<1%) of dangerous disorders may be considered acceptable in the emergency department,13 particularly when shared decision-making approaches that consider patient preferences (eg, whether a patient prefers the small residual risk of missed subarachnoid hemorrhage or the risks of a false-positive lumbar puncture following a traumatic tap, including follow-up angiography) are used effectively.14 However, there are several important caveats for applicationof this decision rule. Effective use of anydecision rule requires careful attention to clinical details affecting its generalizability.Does thepatientmeet all original inclusion criteria, such as having a headache that peaked in less than an hour? Hasanexaminationbeenperformedcarefullyenough toverify that neurologic status is truly normal, including no papilledema? Is subarachnoid hemorrhage the only target diagnosis being considered, or are unstudied, rare, yet important causes of sudden-onset headache (eg, cerebral venous sinus thrombosis, pituitary apoplexy, arterial dissection) still part of the differential diagnosis? Are other unstudied variables (eg, family history of brain aneurysms) present that might complicate interpretation of the rule? In clinical practice, “rules creep” can lead to overly broad applicationofadecision rule. Suchcreep in thesettingofheadache could be toward patientswho presentwith severe headaches that are more gradual in onset. This misuse could present a problem for patients, especially if the rulewere used to Related article page 1248 Opinion
Read full abstract
Jan 28, 2024
Marit V Forslund + 10
Open Access
