The Diagnostic Evaluation of Secondary Headache Disordershead_1841 Vincent T. Martin, MD

Transcription

1 CURRENT LITERATURE: CLINICAL SCIENCE The Diagnostic Evaluation of Secondary Headache Disordershead_1841 Vincent T. Martin, MD One of the greatest challenges within the field of headache is the identification of secondary headache disorders. Physicians are particularly fearful of missing serious causes of secondary headache disorders such as brain tumors, subarachnoid hemorrhage, temporal arteritis, meningitis, and stroke. Missing one of these disorders could result in significant morbidity to the patient and leave the physician vulnerable to costly litigation. Unfortunately, the signs and symptoms of secondary headache disorders can sometimes be subtle. Physicians must ask specific questions to identify red flags of secondary headache disorders, conduct a detailed neurological examination and perform appropriate diagnostic testing in order to establish a diagnosis. In following commentary, we will review articles published in the last 2 years related to the diagnostic evaluation of secondary headache disorders. Pearls: Headache Dodick D. Semin Neurol. 2010;30: Distinguishing primary headache from secondary headache is the first objective of every new clinical encounter with a patient complaining of headache. The history is king in headache medicine 90% of patients presenting with headache have a primary headache disorder and the examination is normal. The history must be elicited because patients will not always volunteer seminal information. A standard series of questions must be asked of each patient to guide an appropriate diagnostic evaluation and ensure that secondary causes are not overlooked. The second objective, of course, is making the correct diagnosis of the primary headache disorder. Although at first glance this appears obvious and almost patronizing, making the correct diagnosis is often not a priority, nor is it a process that is emphasized in undergraduate and postgraduate training programs. Knowing some simple rules and standard questions will make the process almost fail proof. Dodick 1 has recently updated the SNOOP mnemonic to help physicians to remember the red flags for secondary headache From the Division of General Internal Medicine, Department of Internal Medicine, Cincinnati, OH, USA. Address all correspondence to V.T. Martin, Division of General Internal Medicine, Department of Internal Medicine, 231 Albert Sabin Way RM 6603, Cincinnati, OH , USA. Accepted for publication October 7, Headache 2011 American Headache Society disorders. The new mnemonic has been called SNOOP4. (Table 1) The primary difference between the current and the past SNOOP mnemonic is the addition of the 4 P s, which stand for progressive headache, precipitation by valsalva, postural aggravation, and papilledema. Progressive refers to the evolution of a headache disorder into a daily, continuous pattern. Precipitation by valsalva indicates a worsening with activities that cause patients to valsalva (eg, bearing down, lifting weights, etc). Postural aggravation is the accentuation of headache when going from a lying to a standing position or vice versa. It may also occur with neck movements, which could represent a sign of cervicogenic headache. Papilledema occurs with increased intracranial pressure that can result from the various causes of intracranial hypertension. Certain red flags can increase the likelihood of specific types of secondary headache disorders. For example, the most likely diagnosis with a sudden onset headache would be a subarachnoid hemorrhage. However, other disorders such as dural venous thrombosis, pituitary apoplexy, and carotid dissection can occur also in patients with sudden onset headache, but are much less frequent. 2 Orthostatic headaches (eg, headaches reproducibly worse upon standing) are most commonly associated with intracranial hypotension, but can also occur less frequently with postural orthostatic tachycardia syndrome. 3 Other red flags are less specific and can occur with a wide variety of headache disorders. See Table 1 for the secondary headache disorders associated with specific red flags. The probability that a given red flag will identify a patient with a secondary headache disorder depends upon the red flag and the clinical setting (eg, primary care, emergency room settings, and specialty practices) in which the patients are seen. The incidence of secondary headache disorders will generally be higher in emergency room settings then in primary care or specialty practices secondary to the higher acuity of medical illnesses encountered in that setting. The red flags that have the most data to support their inclusion as red flags for secondary headache disorders include the following: (1) headache of sudden onset; (2) headache associated with neurologic signs and symptoms; and (3) headache onset after 50 years of age. The predictive value of these red flags for secondary headache disorders will be discussed below. There have been several studies detailing the probability of secondary headache disorders in patients with sudden onset headache. Landtblom and colleagues 2 reported an incidence of secondary headache disorders of 21.4% in those patients... Conflict of Interest: Dr. Martin is a consultant for Allergan, Merck, and Nautilus; a speaker for Allergan; and has received grants from GSK and Endo. 346

2 347 Headache February 2011 Table 1. SNOOP4 Mnemonic for Secondary Headache Disorders Mnemonic Clinical Presentation Common Secondary Headache Disorders Systemic Neurologic Unexplained fever, chills, weight loss New onset headache in patient with malignancy, immunosuppression or HIV Complaints of motor weakness, sensory loss, diplopia or ataxia Primary or metastatic tumors, meningitis, brain abscess, temporal arteritis Malignant, inflammatory, and vascular disorders of the brain Abnormal neurological examination Onset sudden Headache reaches peak intensity in <1 minute Vascular events such as subarachnoid hemorrhage (most common), CVA, carotid dissection, cerebral vasoconstriction syndromes, dural venous thrombosis Onset after age 50 New onset headache after age 50 Neoplastic, inflammatory disorders, and temporal arteritis Pattern change Progressive headache (evolution to daily headache) Malignant, inflammatory, and vascular disorders of the brain Precipitated by valsalva Postural aggravation Papilledema Chiari malformation, primary and metastatic lesions of brain, hydrocephalus Low pressure headache syndromes, cervicogenic headaches, intracranial hypertension, POTS Malignant and inflammatory disorders of brain, idiopathic intracranial hypertension, dural venous thrombosis Adapted from Dodick D. Semin Neurol. 2010;30: CVA = cerebrovascular accident; HIV = human immunodeficiency virus; POTS = postural orthostatic tachycardia syndrome. presenting to the emergency room with sudden onset headaches (eg, onset <10 seconds). Subarachnoid hemorrhage (SAH), cerebral infarction, intracerebral hemorrhage, aseptic meningitis, cerebral edema, and dural venous thrombosis occurred in 11.3%, 3.6%, 2.2%, 2.9%, 0.7%, and 0.7%, respectively. Linn and colleagues 4 found an incidence of secondary headaches disorders of 46% (eg, subarachnoid hemorrhage in 30%, intracerebral hemorrhage in 8%, and others in 8%) in 110 primary care patients with sudden severe headaches peaking in 1 minute. Another study 5 of 434 emergency room patients with sudden onset headache demonstrated an incidence of 43% for secondary headache disorders with subarachnoid hemorrhage occurring in 16.3%. Perry and colleagues 6 enrolled 592 emergency room patients with headaches reaching maximal severity in <1 hour and discovered subarachnoid hemorrhage in 10.3% of patients. Therefore, the probability of secondary headache disorders ranges from 21% to 46% in patients presenting with sudden onset headache while SAH occurs in 10-30%. Neurologic signs and symptoms are among the most predictive for secondary headache disorders. Ramirez and colleagues 7 found secondary headache disorders in 44% of emergency room patients with a complaint of headache and an abnormal neurologic examination as compared with 2% in those with a normal exam. The positive likelihood ratio for secondary headache disorders was 16.2 for an abnormal neurologic exam in this study. Locker and colleagues 8 reported that an up-going plantar reflex increased the likelihood of a secondary headache to 55% in 353 emergency room patients with a complaint of non-traumatic headache. Another study 9 conducted within an emergency room setting demonstrated a positive likelihood ratio of 3.56 for an abnormal neurological examination in the prediction of secondary headache disorders. Thus, an abnormal neurological examination confers a probability of secondary headache disorders of 44-55% within emergency room settings. Older age has also been associated with the diagnosis of secondary headache disorders. Goldstein and colleagues 10 collected data from emergency room visits of patients presenting with a complaint of headache as part of the National Ambulatory Medical Care Survey. They reported that patients 50 years of age were 3.3 times more likely to receive a serious secondary headache diagnosis (eg, stroke, meningitis, encephalitis, intracerebral and subarachnoid hemorrhage, temporal arteritis, hypertensive encephalopathy, and benign intracranial hypertension) than those <50 years of age. Overall, 6% of patients 50 years of age had secondary headache disorders compared with 1% of those <50 years of age. This study however did not report which of these headaches were new onset. Kernig and colleagues 11 found that serious secondary headache disorders were 2-3 times more common in patients 50 years of age that presented to the primary care physician with a new complaint of headache. The incidence of secondary headaches (eg, malignant brain tumors, temporal arteritis, cerebrovascular events, and benign spaceoccupying lesions) was 4% in patients that were 50 years of age and 0.25% in those <50 years of age. Ramirez-Lassepas and colleagues 7 demonstrated intracranial pathology in 9.8% of emergency room patients presenting with headache that were

3 348 Headache February years of age as compared with 2% in those <55 years of age. Locker et al 9 reported a likelihood ratio of 2.34 for secondary headache disorders in emergency room patients greater than 50 years of age presenting with a complaint of headache. These data suggest that physicians should be particularly vigilant for secondary headache disorders in patients with headaches that are 50 years of age or older. The above studies confirm that the red flags of sudden onset headache and abnormal neurologic exam are more predictive for secondary headache disorders than older age. These results are most applicable to emergency room patients as most of the above studies were conducted within emergency room settings. What Happens to New-Onset Headache Presented to Primary Care? A Case Cohort Study Using Electronic Primary Care Records Kernick D, Stapley S, Goadsby P, Hamilton W. Cephalalgia. 2008;28: In the UK, 4% of general practitioner consultations are for headache, yet the natural history of these presentations is unknown. The objective of this study was to describe the outcome of new headache presentations to the general practitioner. This was a prospective case control study in adults over a period of 1 year using data from the General Practitioner Research Database, UK. Records of patients who presented with primary headache (migraine, tension-type headache, cluster headache) or undifferentiated headache (no further descriptor) were examined for the subsequent year for subarachnoid hemorrhage, primary brain tumor, benign spaceoccupying lesion, temporal arteritis, stroke, and transient ischemic attack (TIA). We identified 21,758 primary headaches and 63,921 undifferentiated headaches. The likelihood ratio was 29 (9.9, 92) for a subarachnoid hemorrhage after an undifferentiated headache and increased with age. The 1-year risk of a malignant brain tumor with a new undifferentiated headache was 0.15%, rising to 0.28% above the age of 50 years. For primary headache the risk was 0.045%. The risk for a benign space-occupying lesion was 0.05% for an undifferentiated and 0.009% for a primary headache. The risk of temporal arteritis was the highest of the conditions studied, 0.66% in the undifferentiated and 0.18% in the primary headache group. Accepting the limitations of this approach, our data can inform management guidelines for new presentations of headache in primary care and confirm the need for follow-up, even if a primary headache diagnosis is made. Secondary headache disorders can be caused by benign as well as serious life threatening causes. However, one could argue it is the serious and potentially life threatening disorders that are most important to recognize. These would include malignant brain tumors, subarachnoid hemorrhage, dural venous thrombosis, meningitis/encephalitis, brain abscess, temporal arteritis, arterial dissections, and cerebrovascular accidents. The 2 studies mentioned below provide the best estimate of the incidence of serious secondary headache disorders within primary care patients. The authors from the above-mentioned study queried the General Practitioners Research Database for diagnoses of secondary headache disorders (eg, subarachnoid hemorrhage, brain tumors, temporal arteritis, stroke/tia, and benign spaceoccupying lesions) in the year following an office visit for a complaint of new onset headache. 11 They found that the incidence of secondary headache disorders is highly dependent on the diagnosis of headache given to the patient by their primary care physician. They reported an overall incidence of 2.49% in patients given an undifferentiated headache diagnosis and 0.70% in those with a diagnosis of a primary headache disorders (eg, migraine, tension-type and cluster headaches). It is possible however that the incidence of serous secondary headache disorders may have been overrepresented in this manuscript as a complaint of headache may have been unrelated to the future development of stroke and TIA. In addition, benign space-occupying lesions such as meningiomas do not always produce headache and are often found incidentally on imaging studies. If one removes these 2 diagnoses then the incidence reduces to 1.00% in the undifferentiated group and 0.25% in the primary headache group. These data indicate that serious causes of secondary headache disorders are rare in primary care and primary care physicians should be particularly vigilant for secondary headache disorders in patients without a straightforward diagnosis of a primary headache disorder. A Brazilian study 12 determined the incidence of secondary headache disorders in 561 primary care patients presenting with a chief complaint of headache. Overall, 44% had a secondary cause for their headaches, but most suffered from benign medical conditions leading to headache (eg, 15.7% with benign febrile illnesses, 11.1% with hypertension, 7.8% with sinusitis) Similar to the above study, they found an incidence of serious secondary headache disorders of 1.41%. See Table 2 for the specific types of secondary headache disorders found within these 2 primary care studies. In contrast, the incidence of serious secondary headache disorders is slightly higher in emergency room patients with a complaint of headache. Friedman and colleagues 13 reported that serious secondary headache disorders (cerebrovascular accident, meningitis/encephalitis, hydrocephalus, idiopathic intracranial hypertension, hypertensive crisis) were diagnosed in 6.0% of 480 emergency room patients with headache. Goldstein and colleagues 10 found that 2.0% of 5198 emergency room visits for a complaint of headache received a diagnosis of a serious secondary

4 349 Headache February 2011 Table 2. Incidence of Serious Secondary Headache Disorders in Primary Care Type of Headache Disorder Kernick et al 2008 Bigal et al 2000 (%) Undifferentiated Headache Diagnosis (%) Primary Headache Diagnosis (%) Subarachnoid hemorrhage Malignant brain tumor Temporal arteritis Stroke/TIA Benign space-occupying lesions Meningitis N/A N/A 0.35 Total (including stroke/tia) Total (excluding stroke/tia) Adapted from Kernick D et al. Cephalalgia. 2008;28: and Bigal et al. Headache. 2000;40: Includes one patient with an intracerebral granuloma and another with a brain abscess. N/A = non-available from study; TIA = transient ischemic attack. headache disorder (stroke, meningitis, encephalitis, intracerebral and subarachnoid hemorrhage, temporal arteritis, hypertensive encephalopathy, and benign intracranial hypertension). Two other studies 7,8 reported an incidence of 3.4% and 3.8%, respectively, of serious secondary headache. The above studies confirm that the probability of serious secondary headache disorders ranges from 0.7% to 2.49% in primary care patients presenting to their physician with a complaint of headache, but increases from 2% to 6% in emergency room patients with headache. The higher incidence of serious causes of secondary headache disorders in emergency room setting is likely secondary to the greater acuity of medical illnesses encountered in this patient population. Non-Traumatic Headaches in the Emergency Department: Evaluation of a Clinical Pathway Dutto L, Meineri P, Melchio R, Bracco C, Lauria G, Sciolla A, Pomero F, Sturlese U, Grasso E, Tartaglino B. Headache. 2009;49: Objective. To determine the impact and efficacy of a clinical pathway in the management of patients with non-traumatic and afebrile headache (NTAH) in the emergency department (ED). Background. Non-traumatic and afebrile headache is one of the most common neurological symptoms in the ED. However, data about the application of an evidence-based operative protocol are lacking. Methods. A before-after intervention study comparing adult patients presenting to the ED with atraumatic headache was conducted during a 6-month period from April to September 2005 and with the same type of patients with the same period in 2006 after a clinical pathway had been implemented. According to their clinical presentations, patients of the 2006 group were divided into 3 subgroups and managed following the established protocol. Results. A total of 686 patients were enrolled in the study, of which 374 were those presenting to our ED with NTAH in 2006 and managed with the aid of the study protocol; the other 312 patients were those who presented in 2005, before the intervention. The study protocol was strictly applied to 247 patients (66%) of the 2006 group. There were fewer neurological consultations after the intervention (41.2% vs 52.5%, difference: -11.3%, 95% confidence intervals [CI]: -18.7% to -3.9%; P =.003); likewise, admissions were significantly reduced after the intervention (9.0% vs 14.7%, difference -5.7%, 95% CI: to -0.08%; P =.02). No significant differences were found between the 2 groups for number of computed tomography (CT) head scans (42.2% vs 38.4%, difference 3.7%, 95% CI: -3.5% to 11%; P =.3). Mean length of stay in the ED was lower after the intervention, although not significantly ( minutes vs minutes, difference: -9.8 minutes, 95% CI: to 5.7; P =.09). A 6-month follow-up was completed involving 302 (96.7%) patients in the first group and 370 (98.9%) in the second group. There was only 1 misdiagnosis after the intervention while 2 incorrect diagnoses were made before the intervention (0.27% vs 0.60%, difference: -0.33, 95% CI: -2.1% to 0.9%; P =.50). Conclusions. Our diagnostic protocol for NTAH appears to be safe and sensitive in diagnosing malignant headaches. In addition, it may improve use of resources by reducing the need for neurological consultations and admissions without increasing the number of CT scans or prolonging length of stay in the ED. Furthermore, when using the protocol ED physicians seem

5 350 Headache February 2011 more confident in their evaluations of headache resulting in fewer requests for specialist input. This study created a diagnostic algorithm in which emergency room patients with a complaint of headache were placed into 1 of 3 different groups based of the characteristics of their headaches. Group 1 included patients with the following characteristics: (1) first or worst headache of their life; (2) thunderclap headache; (3) headache accompanied with neurologic findings on exam; (4) headache with vomiting or syncope; and (5) headache following exertion or sexual intercourse. Group 2 identified patients with recent onset of headaches (weeks to months), patients over 40 years of age or worsening/persistent headaches. Group 3 included patients with headaches that were similar to past headaches with the exception that they were of longer duration as well as more severe and refractory to their standard abortive therapies. Those in group 1 were encouraged to have CT scans of the head as well as a lumbar puncture (LP) if the CT scan was negative when subarachnoid hemorrhage was suspected. Group 2 patients were recommended to have CT scans and neurological consultation if deemed to be medically necessary. Group 3 patients did not receive a CT scan and were treated with parenteral ketorolac and metoclopramide. This study is novel because it tries to place red flags into groups that identify specific types of secondary headache disorders. Group 1 was specifically designed to identify patients with acute presentations of secondary headache disorders. It would primarily identify vascular disorders such as subarachnoid and intracranial hemorrhage, carotid dissection, cerebrovascular accident, intracranial aneurysms, and dural venous thrombosis. Group 2 recognized more indolent cause of headaches such as malignant and benign brain tumors, temporal arteritis, idiopathic intracranial hypertension, and brain abscess. Group 3 was meant to find patients with worsening of their underlying primary headache disorder. Groups 1 and 2 identified all patients with serious secondary headache disorder with 1 exception (eg, one cerebral neoplasm that was placed into group 3). These groupings were very similar to those first proposed in a study by Cortelli and colleagues. 14 However, their study recommended use of a fourth group of patients with severe headache, fever, and/or neck stiffness. This would identify patients with meningitis and encephalitis as well as those with sepsis and sinus infections. The present study did not include this fourth group as they excluded patients with fever. I believe that this fourth group is important as infectious causes for secondary headache could otherwise be missed. Locker et al 9 conducted a prospective study to determine the multivariate predictors of serious secondary headache disorders (eg, any secondary headache disorder other than those related to medication side effects, diagnostic/therapeutic procedures, musculoskeletal disorders, dental disorders, herpes zoster, or systemic illness). They found 3 independent predictors of secondary headache disorders that included age >50 (positive likelihood ratio [LR] = 2.34), sudden onset (LR = 1.74), and abnormal neurologic examination (LR = 3.56). Any 1 of the 3 predictors yielded a sensitivity of 98.6, specificity of 34.4, positive LR of 1.5, and negative LR of These data suggest an extremely low probability of a secondary headache disorder in patients <50 years of age when their headaches are not of sudden onset and their neurologic examinations are normal. The above 2 studies have attempted to define specific groupings of patients with high or low risk of secondary headache disorders. The strength of the first study is that specific groupings were created that place patients into 3 different diagnostic categories. By placing patients into different categories it guides the appropriate choice of diagnostic tests by emergency room physicians. Because group 1 primarily identifies those with vascular causes of secondary headache, these patients should receive a CT scan followed by an LP if the initial CT is negative. Depending on the clinical suspicion for other headache disorders, one might also consider performing a magnetic resonance imaging (MRI) (with and without contrast), CT angiogram/magnetic resonance angiograms (MRA) and magnetic resonance venograms (MRV) to rule out carotid dissection, intracranial aneurysm, dural venous thrombosis, intracranial hypotension, etc. Group 2 would indentify patients with more indolent causes of secondary headache disorders. These patients would receive a CT of the head in the emergency room (ER) and then later as an outpatient would get a MRI (with/without contrast) of the brain if the initial CT was negative. If greater than 50 years of age, then an erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) might be performed to exclude temporal arteritis. Patients in group 3 might not need any neuroimaging whatsoever in the ED. The second study defines a group of patients at low risk for secondary headache disorders in the ED namely, those that lack the 3 most potent predictors of serious secondary headache disorders (eg, age 50 years, abnormal neurologic exam and sudden onset headache). These studies provide valuable insight into the risk stratification of headache patients in the ED, but need to be replicated before application within ED settings. Is the Combination of Negative Computed Tomography Result and Negative Lumbar Puncture Result Sufficient to Rule Out Subarachnoid Hemorrhage? Perry J, Spacek A, Forbes M, Wells G, Mortensen M, Syminton C, Fortin C, Fortin N, Stiell I. Ann Emerg Medical. 2008;51: Study Objective. Current clinical practice assumes a negative CT head scan result and a negative LP result together are adequate to rule out subarachnoid hemorrhage in patients with acute headache. Our objective is to determine the sensitivity of a

6 351 Headache February 2011 negative CT result combined with a negative LP result to exclude subarachnoid hemorrhage. Methods. This prospective cohort study was conducted at 2 tertiary care EDs during 3 years. We enrolled all patients who were older than 15 years, had a non-traumatic acute headache and normal neurologic examination result, and who had a CT head scan and an LP if the CT result was negative (ie, no blood in the subarachnoid space). Patients were followed-up with a structured telephone questionnaire 6 to 36 months after their ED visit and electronic hospital records review to ensure no missed subarachnoid hemorrhage. We calculated sensitivity, specificity, and likelihood ratios of the strategy of CT and then LP for subarachnoid hemorrhage. Results. Five hundred and ninety-two patients were enrolled, including 61 with subarachnoid hemorrhage. The mean patient age was 43.6 years, with 59.1% female patients. All cases of subarachnoid hemorrhage were identified on initial CT or LP. One patient without subarachnoid hemorrhage was subsequently diagnosed with cerebral aneurysm, requiring surgery. The strategy classified patients with subarachnoid hemorrhage with sensitivity, specificity and positive and negative likelihood ratios (with 95% CIs of 100% (95% CI 94% to 100%), 67% (95% CI 63% to 71%), 3.03 (95% CI 2.69 to 3.53), and 0. For diagnosis of subarachnoid hemorrhage or aneurysm, these were 98% (95% CI 91% to 100%), 67% (95% CI 63% to 71%), 2.98 (95% CI 2.63 to 3.38), and 0.02 (95% CI 0.00 to 0.17), respectively. Conclusions. To our knowledge, this is the largest prospective study evaluating the accuracy of a strategy of CT and LP to rule out subarachnoid hemorrhage in alert ED patients with acute headache. This study validates clinical practice that a negative CT with a negative LP is sufficient to rule out subarachnoid hemorrhage. A common diagnostic strategy for a sudden onset headache is to order a non-contract CT scan and then perform an LP if the CT was negative. However, is it reasonable to stop at that point or must patients undergo additional tests to exclude intracranial aneurysms? This is one of the most controversial questions in the field of headache. The answer to the question lies in whether you are trying to identify the future occurrence of subarachnoid hemorrhage or intracranial aneurysms or both? The above study addresses the probability of future subarachnoid hemorrhages in patients with acute headache after a negative head CT and LP. This prospective multicenter study identified 61 subarachnoid hemorrhages in 592 patients that presented to the ER with acute headache (headache reaching peak intensity within 1 hour) through use of head CT followed by an LP if clinically necessary. The other 561 were followed prospectively to determine if they developed a subarachnoid hemorrhage. They were contacted by telephone to determine if they had subarachnoid hemorrhage at least 6 months after the initial emergency room visit. In addition, hospital records were searched to ascertain admissions for subarachnoid hemorrhoid or intracranial aneurysm. Follow-up was obtained in 89.6% of the cohort. No new cases of subarachnoid hemorrhage were identified during the follow-up period, but 1 case of intracranial aneurysm was found. Savitz and colleagues 15 found similar results to those obtained by Perry. They pooled the results of 7 past studies (including the one above) to determine the risk of subarachnoid hemorrhage after a negative CT and LP. There were 813 patients in these studies and no patients developed subarachnoid hemorrhage during the follow-up period which ranged from 6 months to 3 years. Follow-up was achieved in 89.6% to 100% of cases. The results of the above studies suggest that clinically recognizable subarachnoid hemorrhages are quite unlikely after a negative CT and LP within a 6-month to 3-year time period after a sudden onset headache. Is it possible that intracranial aneurysms could have been missed in the above studies? The obvious answer to this question is yes. Intracranial aneurysms occur in 2-6% of the population, but the vast majority are asymptomatic and will not lead to subarachnoid hemorrhage. 16 We will invariably find intracranial aneurysms if we look for them, but their mere presence does not indicate that they were the cause of the sudden onset headache. If the newly discovered aneurysms are 1 cm or show signs of rupture then this would be more convincing evidence that they are clinically significant. However, it is difficult to know if nonruptured aneurysms <1 cm in size are responsible for the headaches or simply found incidentally. To further complicate the issue, Rinkel and colleagues 17 estimated the risk ratio of rupture for a symptomatic aneurysm was 8.3 (95% CIs, 4 to 17) as compared with an asymptomatic aneurysm. If these data are correct then aneurysms found in the context of a sudden onset headache may be more likely to rupture than those found without symptoms. However, there are no studies to document the risk of rupture of aneurysms found on neuroimaging after a negative CT and LP. This could represent a subgroup of patients with a vastly different risk for rupture than those that have not had an evaluation. Therefore, it is difficult to ascertain the clinical significance of smaller aneurysms found on neuroimaging for sudden onset headache. To my knowledge, there is only 1 study 18 that has performed CT angiograms (CTA) in patients presenting to the ER with a complaint of sudden onset headache. Study patients underwent a non-contrast head CT and a CTA, but also received an LP if the head CT was negative. They found aneurysms in 5.1% of the study population. Three of the 5 patients with aneurysms on CTA had negative LPs and head CTs. One of the 3 was found to be a false positive CTA after receiving cerebral angiography. Both of the confirmed aneurysms were 7 mm in size and each was later surgically clipped. This study confirmed that CTA will indentify

7 352 Headache February 2011 aneurysms even in those with negative CTs and LPs, but it is impossible to tell if these aneurysms were responsible for the sudden onset headaches. Should we perform CTAs or MRAs in all patients with sudden onset headaches? There simply is not enough evidence to answer this question. We can tell patients that a negative head CT and LP confer a low risk of subarachnoid hemorrhage in the 6-18 months after a sudden onset headache. Neuroimaging will identify aneurysms even when the head CT and LP are negative, but that the clinical significance of these aneurysms is unknown if they are small (eg, <10 mm in size). One might consider performing a CTA or MRA in those with a high pretest likelihood of subarachnoid bleed and/or aneuryms. This might include those patients with sudden onset headache and any of the following characteristics: (1) syncope/altered mental status; (2) neurological signs or symptoms; and (3) personal or family history of intracranial aneurysms, polycystic kidney disease or connective tissue diseases (eg, Marfans, Ehlers Danlos syndrome). Whether to perform these studies on patients with low to intermediate pretest probabilities might depend on the patient s preference after an informed discussion of the risks and benefits of such a strategy. References 1. Dodick DW. Pearls: Headache. Semin Neurol. 2010;30: Landtblom AM, Fridriksson S, Boivie J, Hillman J, Johansson G, Johansson I. Sudden onset headache: A prospective study of features, incidence and causes. Cephalalgia. 2002;22: Mokri B, Low PA. Orthostatic headaches without CSF leak in postural tachycardia syndrome. Neurology. 2003;61: Linn FH, Wijdicks EF, van der Graaf Y, Weerdesteyn-van Vliet FA, Bartelds AI, van Gijn J. Prospective study of sentinel headache in aneurysmal subarachnoid haemorrhage. Lancet. 1994;344: Bo SH, Davidsen EM, Gulbrandsen P, Dietrichs E. Acute headache: A prospective diagnostic work-up of patients admitted to a general hospital. Eur J Neurol. 2008;15: Perry JJ, Spacek A, Forbes M, et al. Is the combination of negative computed tomography result and negative lumbar puncture result sufficient to rule out subarachnoid hemorrhage? Ann Emerg Med. 2008;51: Ramirez-Lassepas M, Espinosa CE, Cicero JJ, Johnston KL, Cipolle RJ, Barber DL. Predictors of intracranial pathologic findings in patients who seek emergency care because of headache. Arch Neurol. 1997;54: Locker T, Mason S, Rigby A. Headache management are we doing enough? An observational study of patients presenting with headache to the emergency department. Emerg Med J. 2004;21: Locker TE, Thompson C, Rylance J, Mason SM. The utility of clinical features in patients presenting with nontraumatic headache: An investigation of adult patients attending an emergency department. Headache. 2006;46: Goldstein JN, Camargo CA Jr, Pelletier AJ, Edlow JA. Headache in United States emergency departments: Demographics, work-up and frequency of pathological diagnoses. Cephalalgia. 2006;26: Kernick D, Stapley S, Goadsby PJ, Hamilton W. What happens to new-onset headache presented to primary care? A case-cohort study using electronic primary care records. Cephalalgia. 2008;28: Bigal ME, Bordini CA, Speciali JG. Etiology and distribution of headaches in two Brazilian primary care units. Headache. 2000;40: Friedman BW, Hochberg ML, Esses D, et al. Applying the International Classification of Headache Disorders to the emergency department: An assessment of reproducibility and the frequency with which a unique diagnosis can be assigned to every acute headache presentation. Ann Emerg Med. 2007;49: , 419 e Cortelli P, Cevoli S, Nonino F, et al. Evidence-based diagnosis of nontraumatic headache in the emergency department: A consensus statement on four clinical scenarios. Headache. 2004;44: Savitz SI, Levitan EB, Wears R, Edlow JA. Pooled analysis of patients with thunderclap headache evaluated by CT and LP: Is angiography necessary in patients with negative evaluations? J Neurol Sci. 2009;276: Schievink WI. Intracranial aneurysms. N Engl J Med. 1997;336: Rinkel GJ, Djibuti M, Algra A, van Gijn J. Prevalence and risk of rupture of intracranial aneurysms: A systematic review. Stroke. 1998;29: Carstairs SD, Tanen DA, Duncan TD, et al. Computed tomographic angiography for the evaluation of aneurysmal subarachnoid hemorrhage. Acad Emerg Med. 2006;13: