Abstract. Subarachnoid hemorrhage (SAH) is a diagnosis often considered in patients presenting to the ED with acute sudden headaches, but with normal physical examinations. Standard of care today is for these patients to be investigated by noncontrast CT scan followed by lumbar puncture (LP) for negative CTs. However, given that most investigated patients have benign headaches, most of the CT and LP results are normal. The authors studied, by means of a theoretical analysis, the impact of an alternative diagnostic model, in which LP would be the first (and, in most cases, only) diagnostic test for patients suspected of SAH who meet lone acute sudden headache (LASH) criteria. Given reasonable assumptions, for every 100 patients investigated, the ''LP-first'' model would result in 79 to 83 fewer CT scans and only seven to 11 additional LPs, as compared with traditional strategies. Among ED headache patients meeting LASH criteria, the authors believe use of this model could result in more efficient use of resources, minimal additional morbidity, and equal diagnostic accuracy for SAH. Key words: lumbar puncture; headache; subarachnoid hemorrhage; lone acute sudden headache. ACADEMIC EMERGENCY MEDI-CINE 1999; 6:131 -136 N ONTRAUMATIC headache is a common source of ED visits, representing up to 4.5% of presenting complaints.1 Subarachnoid hemorrhage (SAH) is the cause of approximately 1% of all headaches seen in the ED, 1 and up to a third of acute sudden headaches.
2,3Subarachnoid hemorrhage occurs in young people, with a median age of 50 years, and has a 50% mortality at six months.2,4 -6 Among survivors, just over half regain their premorbid neurologic state six months after SAH. 4,6,7 Clearly emergency physicians must maintain a high index of suspicion for SAH and must, quite appropriately, be liberal in initiating investigations for this devastating, but potentially curable, disease.Among Computed tomography scanning alone cannot be relied on to rule out SAH, since its sensitivity at 24 hours varies from 81% to 100%, and decreases rapidly thereafter, 9 -15 even with newergeneration CT scanners.12,15,16 Also, CT sensitivity varies with the neurologic status of the patient; CT is only 86% sensitive in alert patients with SAH, as compared with 99.4% sensitivity in stuporous or comatose patients with SAH. 6,11 Normal CTs, therefore, must be followed by a lumbar puncture (LP). Cerebrospinal fluid (CSF) xanthochromia as detected by spectrophotometry is virtually 100% sensitive for SAH, so long as it is done no earlier than 12 hours following the bleeding, in order to allow enough time for xanthochromia to develop. 13,17 -19 This very high sensitivity of LP decays very slowly, remaining at virtually 100% for two weeks following a hemorrhage. 10,13,17 Given that LASH patients should be investigated for SAH, but that most have benign headaches, a great many negative CT scans and LPs are carried out. Is there a more efficient way to investigate patients meeting LASH criteria? Specifically, could we forgo the CT and pro...