Validation of cerebrospinal fluid findings in aneurysmal subarachnoid hemorrhage
a b s t r a c t
Background: Recently proposed cutoff criteria for cerebrospinal fluid (CSF) analyses might safely exclude a diag- nosis of Aneurysmal subarachnoid hemorrhage (aSAH).
Objective: The objective of this study was to examine the sensitivity of a CSF red blood cell (RBC) count greater than 2000 x 106/L (ie, 2000 RBCs per microliter) or the presence of visible CSF xanthochromia in identifying pa- tients with aSAH. Methods: We identified a retrospective case series of patients diagnosed with aSAH after lumbar puncture in an integrated health delivery system between January 2000 and June 2013 by chart review. All identified patients had at least 1 cerebral aneurysm that was treated with a neurosurgical or endovascular intervention during the index hospitalization. The lowest CSF RBC count was used for validation analysis. Cerebrospinal fluid color was determined by visual inspection. Xanthochromia was defined as pink, orange, or yellow pigmentation of CSF supernatant.
Results: Sixty-four patients met study inclusion criteria. Of these, 17 (33%) of 52 underwent LP within 12 hours of headache onset, and 49 (84%) of 58 exhibited CSF xanthochromia. The median CSF RBC count was 63250 x 106/L. The sensitivity of a CSF RBC count of greater than 2000 x 106/L in identifying aSAH was 96.9% (95% confidence interval, 89.3%-99.1%). Additional consideration of CSF xanthochromia resulted in a sensitivity of 100% (95% con- fidence interval, 94.3%-100%).
Conclusions: All patients in this case series of patients with aSAH had either a CSF RBC count greater than 2000 x 106/L or visible CSF xanthochromia, increasing the likelihood that this proposed cutoff strategy may safely iden- tify patients who warrant further investigation for an aneurysmal cause of subarachnoid hemorrhage.
(C) 2015
Introduction
Aneurysmal subarachnoid hemorrhage (aSAH) occurs with an annu- al incidence of approximately 1 in 10000 among North American popu- lations [1]. Although most aSAH cases are identified by cranial computed tomography , the imperfect sensitivity of CT combined with the increased morbidity associated with Delays in diagnosis makes lumbar puncture the definitive test to exclude a diagnosis
?? Funding source: Kaiser Permanente Northern California Community Benefits Grant.
? Conflicts of interest: None.
* Corresponding author at: Department of Emergency Medicine, 275 W. MacArthur Blvd, Oakland, CA 94611. Tel.: +1 510 918 1976; fax: +1 510 752 7667.
of aSAH [2]. Cerebrospinal fluid (CSF) is examined for the presence of red blood cells and xanthochromia, the latter being determined either by visual inspection or by spectrophotometry depending on local practices. Although some advocate that xanthochromia is a sine qua non factor for a diagnosis of aSAH, this assumption is based on LP perfor- mance at 12 or more hours from ictus combined with the use of spectro- photometry to detect CSF bilirubin, neither of which are usual practices within North America [3-5].
As a result, when abnormal numbers of RBCs are present in the CSF of a patient with possible “CT-negative” SAH (subarachnoid hemor- rhage), angiography is often recommended to exclude a vascular etiol- ogy of hemorrhage [6-8]. Unfortunately false-positive LPs because of elevated RBC counts are common, most often attributed to traumatic technique (“traumatic taps,” with reported rates up to 30%) [9-11]. Re- sultant workups can result in unnecessary and potentially harmful diag- nostic testing; rates of neurologic complications after cerebral digital
http://dx.doi.org/10.1016/j.ajem.2015.05.012
0735-6757/(C) 2015
1250 D.G. Mark et al. / American Journal of Emergency Medicine 33 (2015) 1249–1252
subtraction angiography (DSA) have been reported as high as 1.3% [12]. In addition, when considering that there is a 1% to 2% background prev- alence of cerebral aneurysms in the general population, unplanned sur- gical treatment of incidental aneurysms becomes a real concern [13,14]. Researchers have proposed various methods to identify traumatic taps. Although often suggestive, analyses using the percentage decrease in RBCs between sequential tubes have failed to produce a clearly ac- cepted cutoff value [9-11,15]. This is largely because CSF collection vol- umes are not standardized, and also because traumatic taps and SAH are not mutually exclusive conditions. Two prior retrospective studies have examined potential CSF RBC count cutoffs below which SAH can be ex- cluded; one reported a negative predictive value of 100% for a cutoff of 500 x 106/L or less (ie, 500 cells per microliter or per cubic millimeter) in detecting 11 cases of SAH with vascular etiologies among a cohort of 594 patents, and the other reported 100% sensitivity for a cutoff of 100 x 106/L among 26 cases of SAH of varying etiologies [9,10]. However, nei- ther of these studies specifically examined patients with aSAH, the diag- nosis that carries the most potential morbidity from delayed recognition
and treatment [16].
Recently, data from the largest prospective study examining diag- nostic testing for SAH revealed that, among 1739 patients undergoing LP, the presence of either a CSF RBC count 2000 x 106/L or more or vis- ible xanthochromia identified 15 patients with aSAH with 100% sensi- tivity (95% confidence interval [CI], 74.7-100.0) [11]. Given this potentially promising approach, we sought to provide a narrower CI for sensitivity by applying these CSF analysis criteria to a 13-year case series of patients diagnosed with aSAH after LP.
Methods
Case cohort selection and setting
Chart review data from 2 independent studies with overlapping study periods spanning January 2000 through June 2013 were used to assemble the case series. Both studies were approved by the Kaiser Foundation Research Institute Institutional Review Board with a waiver of the requirement for informed consent. In both cases, electronic health records of patients treated within the Kaiser Permanente Northern Cal- ifornia (KPNC) integrated health delivery system were screened for case inclusion if they had an emergency department (ED) or hospital en- counter with an associated International Classification of Diseases, Ninth Revision (ICD–9), diagnosis code of subarachnoid hemorrhage (430). During the study period, emergency care within KPNC was provided at between 17 and 21 community EDs serving between 2 and 3.4 mil- lion Kaiser Foundation Health Plan members. Patients were electroni- cally excluded from further screening if they were younger than 18 years or had an ICD–9-coded diagnosis of head or Neck trauma within 24 hours of the index encounter.
Inclusion criteria for this case series included initial presentation to a KPNC ED, performance of an LP in the ED, and subsequent cerebral an- giography demonstrating an aneurysm that was treated with neurosur- gical or endovascular intervention during the index hospitalization. We chose urgent aneurysm repair as part of our case inclusion criteria to avoid inclusion of incidentally discovered aneurysms. However, we also identified patients who had cerebral aneurysms identified but did not undergo urgent surgical or Endovascular repair for use in a sensitiv- ity analysis.
Methods and measurements
Three emergency physician abstractors (D.G.M., M.K.V., and S.R.O.) blinded to the current study hypothesis conducted structured explicit chart reviews. Data on age, sex, and race were electronically collected. Abstractors manually documented the study inclusion criteria (age, per- formance of LP in the ED, aneurysm on cerebral angiography, and subse- quent treatment) along with CSF RBC counts, the presence of CSF
xanthochromia, the official written radiology interpretation of the initial cranial CT, the location and size of the culprit aneurysm, and whether the time to LP was 12 hours or greater from headache onset, when avail- able. It is common practice in our system for patients with negative ini- tial cerebral angiography to undergo a second Imaging study within the following 2 weeks, most often cerebral DSA, and these studies were also reviewed for delayed recognition of a cerebral aneurysm.
Computed tomography examinations were performed without con- trast using multislice cine technology. General radiologists or neurora- diologists made the final interpretation of CT images. A blinded review of 40% of the final sample was conducted by 2 emergency physicians (D.R.V. and D.W.B.) to establish the inter-rater agreement on the follow- ing variables: time to LP of 12 hours or greater from headache onset, presence of CSF xanthochromia, and lowest CSF RBC cell count.
Cerebrospinal fluid was assessed for xanthochromia by visual in- spection of the supernatant against a white background, as per usual practice in North American laboratories [17]. Although xanthochromia in its literal interpretation means “yellow color,” spectrophotometric evidence of bilirubin, the primary colorimetric breakdown product of RBCs responsible for yellow CSF supernatant discoloration, has been ob- served in CSF ranging from red to green [18]. As such, it is now common practice to refer any discoloration of CSF supernatant as xanthochromia, particularly pink, orange, and yellow [19-21]. Thus, we explicitly con- sidered any of these reported CSF colors (pink, orange, or yellow) as representative of xanthochromia, in addition to explicitly reported xanthochromia. We did not include red discoloration in our definition of xanthochromia because significant amounts of oxyhemoglobin (causative of red discoloration of CSF supernatant) are detectable by spectrophotometry within 1 hour in ex vivo CSF samples containing RBCs greater than 20000 x 106/L [21].
We used the lowest available CSF cell count for our primary analysis
for 2 principal reasons: (1) cell count and color analysis were only per- formed on a single tube in multiple instances and (2) it is not uncom- mon for tubes to be either reported or collected out of order, resulting in inaccurate apparent trends in RBC counts. Thus, for practical pur- poses, we assumed that the lowest RBC count represented the last tube collected in all cases, consistent with the definition used in the der- ivation study of interest [11].
Outcomes and analysis
The outcomes of interest were the independent and combined sen- sitivities of a CSF RBC count greater than 2000 x 106/L and/or the pres- ence of visible CSF xanthochromia in identifying patients with aSAH. Binomial CIs were calculated using the Wilson method [22]. All statisti- cal analyses were performed using STATA v 13.0 (College Station, TX).
Results
A total of 2236 charts were reviewed for inclusion criteria. Sixty-four patients with LP results who underwent Urgent treatment of aSAH were identified, comprising the final case series. The median age at diagnosis was 51.5 years, and 69% were female. Twenty-four (38%) had evidence of SAH noted on the final CT report. Nineteen patients (30%) had only a single CSF tube analyzed for cell count and/or color. One patient had only CSF color results reported because of a clotted sample, and 6 pa- tients did not have CSF color results reported. The median RBC count was 63250 x 106/L (range, 408-721000 x 106/L), and 49 (84%) of 58 ex- hibited CSF xanthochromia (47% reported as “xanthochromia” and 38% reported as “pink”- the remaining were 16% reported as “colorless”). The median percentage decrease in cell counts between tubes was 17% (range, 0%-83%). Of charts with available data, 17 (33%) of 52 underwent LP within 12 hours of headache onset. The most common aneurysm location was the anterior communicating artery (38%) followed by the posterior communicating artery (33%). Cohort charac- teristics are summarized in Table 1. Inter-rater agreement was 100%
D.G. Mark et al. / American Journal of Emergency Medicine 33 (2015) 1249–1252 1251
Table 1
Patient characteristics and outcomes in the entire cohort (n = 64) and subcohort of patients with LP performed within 12 hours of headache onset (n = 17)
Variable |
Entire cohort (n = 64) |
Time to LP b 12 hours (n = 17) |
Age, y (median, IQR) |
51.5 (45-65) |
54 (50-69) |
Female (%) |
44/64 (69) |
15/17 (88) |
Evidence of SAH on cranial CT (%) |
24/64 (38) |
6/17 (35) |
LP within 12 hours of headache onset (%) |
17/52 (33) |
n/a |
CSF RBC count x106/L (median, IQR) |
63250 (24100-205000) |
175000 (46735-270000) |
CSF xanthochromia (%) |
49/58 (84) |
10/14 (71) |
Percentage decrease in RBC counts between CSF sample tubes (median, IQR) |
17 (9-31) |
15 (5-34) |
Aneurysm location (%) |
||
ACOM |
38 |
|
PCOM |
33 |
|
MCA |
8 |
|
Vertebral |
6 |
|
Other? |
15 |
Abbreviations: IQR, interquartile range; ACOM, anterior communicating artery; PCOM, posterior communicating artery; MCA, Middle cerebral artery.
* Other: basilar (2), posterior inferior cerebellar (2), internal carotid (2), multiple (2), anterior choroid, paraclinoid.
for time to LP of 12 hours or greater from headache onset, presence of CSF xanthochromia, and lowest CSF RBC cell count.
The lowest CSF RBC count was greater than 2000 x 106/L in 62 pa- tients, resulting in a sensitivity of 96.9% (95% CI, 89.3%-99.1%) for RBC count alone. The 2 remaining patients both had visible CSF xanthochromia (xanthochromia in the single patient with a clotted sample and pink in a patient with an RBC count of 408 x 106/L who underwent LP within 12 hours of headache onset) resulting in a com- bined sensitivity of 100% (95% CI, 94.3%-100%) for either a CSF RBC count greater than 2000 x 106/L or xanthochromia (Table 2).
We performed a sensitivity analysis by including 2 patients who had negative CT but positive LP results and evidence of cerebral aneurysms that were not emergently treated (aneurysms presumed to be inciden- tal findings). Both of those patients had lowest CSF RBC counts greater than 2000 x 106/L (14300 and 31040 x 106/L, respectively); thus, the sensitivity of this discriminatory threshold remained unchanged at 100%.
Discussion
This case series adds further evidence in support of a proposed cutoff of greater than 2000 x 106/L CSF RBCs, or the presence of visible xanthochromia, in warranting further investigation for aneurysmal causes of SAH, with a sensitivity of 96.9% for RBC count alone and 100% sensitivity with the additional consideration of visible xanthochromia. These findings were unchanged when including pa- tients with presumed unruptured cerebral aneurysms, on the chance that those clinical determinations were erroneous. Although we were unable to provide an estimate of specificity, the prospective data pub- lished by Perry et al [11] suggest that this approach may significantly re- duce extraneous testing for aSAH, given a reported specificity of 91.2% when considering the presence of more than 1 x 106/L CSF RBCs as ab- normal (although the calculated specificity does decrease slightly to ap- proximately 88% if one subtracts the approximately 160 patients in that study with between 2 and 5 x 106/L CSF RBCs from the number of patients without aSAH who have a low risk CSF RBC count, given that N 5 x 106/L is a more commonly used threshold for defining an
Sensitivity of CSF findings for the diagnosis of aSAH
Sensitivity 95% CI
abnormal CSF RBC count). Regardless, considering the considerable lack of enthusiasm among emergency physicians for any testing strategy for aSAH that detects less than 99% to 100% of cases, sensitivity is of paramount importance in the adoption of any such rule-out criteria [23].
A key criticism of this LP interpretation strategy is that it relies on a lack of visible xanthochromia as a rule-out criterion. Given both the im- perfect sensitivity of visible xanthochromia for SAH as well as concerns over poor interobserver and intraobserver agreement in its determina- tion, this is a valid criticism [5,19,24,25]. However, although spectro- photometry has a better evidence base for being a more objective, reproducible assay with high sensitivity [26,27], its general lack of avail- ability in North America [17] leaves a significant knowledge-to-practice translation gap that can only be filled by investigations that incorporate relevant clinical practices, including the performance of LP within 12 hours of headache onset [11]. That being said, we fully acknowledge that, with only 1 patient in this study and 1 patient in the reference study by Perry et al with both aSAH and a CSF RBC count less than 2000 x 106/L, the reliability of visible xanthochromia in this subset of patients remains uncertain and requires further validation [11].
Another important factor in this discussion is the method of angio- graphic investigation used in response to LP analyses suggestive of SAH. Increasingly, multislice CT angiography has been promoted as a stand-alone test for excluding cerebral aneurysms, specifically in cases of CT-negative SAH [28-33]. In contrast, other reports have emphasized the suboptimal sensitivity of CT angiography when compared with ce- rebral DSA, particularly for aneurysms less than 5 mm in size [34,35]. This is of particular concern given that a large prospective registry of more than 2000 patients with aSAH (the International Subarachnoid Aneurysm Trial) found that a slight majority of cerebral aneurysms im- plicated in SAH are 5 mm or smaller in size [36]. Accordingly, DSA still remains the preferred method of investigation for cerebral aneurysms in the setting of SAH [37]. However, because DSA has been associated with rates of neurologic complications up to 1.3%, with a 0.5% rate of permanent disability [12], there is a clear need to improve the specificity of CSF analyses for aSAH as performed in North America. The prospec- tive study by Perry et al [11], in concert with the results presented here, takes an important step in that direction.
Limitations of this study include a retrospective design, relatively small case series, and the lack of a control group to offer an estimate of specificity, as discussed earlier. Given that case identification relied upon ICD–9 diagnosis, this study may suffer from ascertainment bias if
* Xanthochromia is defined as a noted visible presence of yellow, orange, or pink dis- coloration of CSF supernatant.
CSF RBC count N 2000 x 106/L CSF xanthochromia? |
96.9% 84.5% |
89.3-99.1 73.1-91.6 |
there were cases of aSAH that underwent LP but had the results misinterpreted such that SAH was not appropriately coded. Such misdi- |
Either CSF RBC count N 2000 x 106/L or xanthochromia |
100% |
94.3-100 |
agnoses in the setting of LP, however, are unusual [16]. In addition, we |
reviewed 50 instances of probable misdiagnosis among 452 patients with aSAH; none of the misdiagnoses involved the performance of a
1252 D.G. Mark et al. / American Journal of Emergency Medicine 33 (2015) 1249–1252
LP (unpublished data). Overall, given the relative rarity of aSAH cases where LP is performed at the diagnostic stage, we feel that a case series approach is a valuable option for exploring possible refinements to CSF analysis in clinical practice, with the understanding that any such find- ings must be prospectively validated.
Conclusions
When performing LP to investigate possible SAH, nearly all patients with ruptured cerebral aneurysms will have a CSF RBC count of greater than 2000 x 106/L. Additional consideration of visible CSF xanthochromia may detect the remaining few patients with CSF RBC counts less than 2000 x 106/L. These findings require further prospective validation.
Acknowledgments
The authors thank Yun-Yi Hung, PhD, and Natalia Udaltsova, PhD, for their assistance with electronic data queries and management. This work was funded by a Kaiser Permanente Northern California Commu- nity Benefits Grant.
References
- de Rooij NK, Linn FH, van der Plas JA, Algra A, Rinkel GJ. Incidence of subarachnoid haemorrhage: a systematic review with emphasis on region, age, gender and time trends. J Neurol Neurosurg Psychiatry 2007;78:1365-72.
- Edlow JA, Panagos PD, Godwin SA, Thomas TL, Decker WW. Clinical policy: Critical issues in the evaluation and management of adult patients presenting to the emer- gency department with acute headache. J Emerg Nurs 2009;35:e43-71.
- Vermeulen M, Hasan D, Blijenberg BG, Hijdra A, van Gijn J. Xanthochromia after sub- arachnoid haemorrhage needs no revisitation. J Neurol Neurosurg Psychiatry 1989; 52:826-8.
- Beetham R. Recommendations for CSF analysis in subarachnoid haemorrhage. J Neurol Neurosurg Psychiatry 2004;75:528.
- Chu K, Hann A, Greenslade J, Williams J, Brown A. Spectrophotometry or visual in- spection to most reliably detect xanthochromia in subarachnoid hemorrhage: sys- tematic review. Ann Emerg Med 2014;64:256-64.
- Wallace AN, Dines JN, Zipfel GJ, Derdeyn CP. Yield of catheter angiography after com- puted tomography negative, lumbar puncture positive subarachnoid hemorrhage [corrected]. Stroke 2013;44:1729-31.
- Chalouhi N, Witte S, Penn DL, Soni P, Starke RM, Jabbour P, et al. Diagnostic yield of cerebral angiography in patients with computed tomography-negative, lum- bar puncture-positive subarachnoid hemorrhage. Neurosurgery 2013;73: 282-7 [discussion 287-8].
- Bakker NA, Groen RJ, Foumani M, Uyttenboogaart M, Eshghi OS, Metzemaekers JD, et al. Appreciation of CT-negative, lumbar puncture-positive subarachnoid haemor- rhage: risk factors for presence of aneurysms and diagnostic yield of imaging. J Neurol Neurosurg Psychiatry 2014;85:885-8.
- Czuczman AD, Thomas LE, Boulanger AB, Peak DA, Senecal EL, Brown DF, et al. Interpreting red blood cells in lumbar puncture: distinguishing true subarachnoid hemorrhage from traumatic tap. Acad Emerg Med 2013;20:247-56.
- Gorchynski J, Oman J, Newton T. Interpretation of traumatic Lumbar punctures in the setting of possible subarachnoid hemorrhage: who can be safely discharged? Cal J Emerg Med 2007;8:3-7.
- Perry JJ, Alyahya B, Sivilotti ML, Bullard MJ, Emond M, Sutherland J, et al. Differenti- ation between traumatic tap and aneurysmal subarachnoid hemorrhage: prospec- tive cohort study. BMJ 2015;350:h568.
- Willinsky RA, Taylor SM, TerBrugge K, Farb RI, Tomlinson G, Montanera W. Neuro- logic complications of cerebral angiography: Prospective analysis of 2,899 proce- dures and review of the literature. Radiology 2003;227:522-8.
- Schievink WI. Intracranial aneurysms. N Engl J Med 1997;336:28-40.
- Winn HR, Jane Sr JA, Taylor J, Kaiser D, Britz GW. Prevalence of asymptomatic inci- dental aneurysms: review of 4568 arteriograms. J Neurosurg 2002;96:43-9.
- Heasley DC, Mohamed MA, Yousem DM. Clearing of red blood cells in lumbar punc- ture does not rule out ruptured aneurysm in patients with suspected subarachnoid hemorrhage but negative head CT findings. AJNR Am J Neuroradiol 2005;26:820-4.
- Kowalski RG, Claassen J, Kreiter KT, Bates JE, Ostapkovich ND, Connolly ES, et al. Ini- tial misdiagnosis and outcome after subarachnoid hemorrhage. JAMA 2004;291: 866-9.
- Edlow JA, Bruner KS, Horowitz GL. Xanthochromia. Arch Pathol Lab Med 2002;126: 413-5.
- Petzold A, Keir G, Sharpe LT. Spectrophotometry for xanthochromia. N Engl J Med 2004;351:1695-6.
- Marshman LA, Duell R, Rudd D, Johnston R, Faris C. Intraobserver and interobserver agreement in visual inspection for xanthochromia: implications for subarachnoid hemorrhage diagnosis, computed tomography validation studies, and the Walton rule. Neurosurgery 2014;74:395-9 [discussion 399-400].
- Seehusen DA, Reeves MM, Fomin DA. Cerebrospinal fluid analysis. Am Fam Physician
Graves P, Sidman R. Xanthochromia is not pathognomonic for subarachnoid hemor- rhage. Acad Emerg Med 2004;11:131-5.
- Zhou XH, Li CM, Yang Z. Improving interval estimation of binomial proportions. Philos Transact A Math Phys Eng Sci 2008;366:2405-18.
- Perry JJ, Eagles D, Clement CM, Brehaut J, Kelly AM, Mason S, et al. An international study of emergency physicians’ practice for acute headache management and the need for a clinical decision rule. CJEM 2009;11:516-22.
- Arora S, Swadron SP, Dissanayake V. Evaluating the sensitivity of visual xanthochromia in patients with subarachnoid hemorrhage. J Emerg Med 2010;39: 13-6.
- Linn FH, Voorbij HA, Rinkel GJ, Algra A, van Gijn J. Visual inspection versus spectro- photometry in detecting bilirubin in cerebrospinal fluid. J Neurol Neurosurg Psychi- atry 2005;76:1452-4.
- Petzold A, Keir G, Sharpe TL. Why human color vision cannot reliably detect cerebro- spinal fluid xanthochromia. Stroke 2005;36:1295-7.
- Cruickshank A, Auld P, Beetham R, Burrows G, Egner W, Holbrook I, et al. Revised na-
tional guidelines for analysis of cerebrospinal fluid for bilirubin in suspected sub- arachnoid haemorrhage. Ann Clin Biochem 2008;45:238-44.
McDonald JS, Kallmes DF, Lanzino G, Cloft HJ. Use of CT angiography and digital sub- traction angiography in patients with ruptured cerebral aneurysm: evaluation of a large multihospital data base. AJNR Am J Neuroradiol 2013;34:1774-7.
- Rana AK, Turner HE, Deans KA. Likelihood of aneurysmal subarachnoid haemor- rhage in patients with normal unenhanced CT, CSF xanthochromia on spectropho- tometry and negative CT angiography. J R Coll Physicians Edinb 2013;43:200-6.
- McCormack RF, Hutson A. Can computed tomography angiography of the brain re- place lumbar puncture in the evaluation of acute-onset headache after a negative noncontrast cranial computed tomography scan? Acad Emerg Med 2010;17: 444-51.
- Lim LK, Dowling RJ, Yan B, Mitchell PJ. Can CT angiography rule out aneurysmal sub- arachnoid haemorrhage in CT scan-negative subarachnoid haemorrhage patients? J Clin Neurosci 2014;21:191-3.
- Jethwa PR, Punia V, Patel TD, Duffis EJ, Gandhi CD, Prestigiacomo CJ. Cost- effectiveness of Digital subtraction angiography in the setting of computed tomo- graphic angiography negative subarachnoid hemorrhage. Neurosurgery 2013;72: 511-9 [discussion 519].
- Agid R, Andersson T, Almqvist H, Willinsky RA, Lee SK, terBrugge KG, et al. Negative CT angiography findings in patients with spontaneous subarachnoid hemorrhage: When is digital subtraction angiography still needed? AJNR Am J Neuroradiol 2010;31:696-705.
- Lu L, Zhang LJ, Poon CS, Wu SY, Zhou CS, Luo S, et al. Digital subtraction CT angiog- raphy for detection of intracranial aneurysms: comparison with three-dimensional digital subtraction angiography. Radiology 2012;262:605-12.
- McKinney AM, Palmer CS, Truwit CL, Karagulle A, Teksam M. Detection of aneurysms by 64-section multidetector CT angiography in patients acutely suspected of having an intracranial aneurysm and comparison with digital subtraction and 3D rotational angiography. AJNR Am J Neuroradiol 2008;29:594-602.
- Molyneux A, Kerr R, Stratton I, Sandercock P, Clarke M, Shrimpton J, et al. Interna- tional Subarachnoid Aneurysm Trial (ISAT) of neurosurgical clipping versus endovascular coiling in 2143 patients with ruptured intracranial aneurysms: a randomised trial. Lancet 2002;360:1267-74.
- Connolly Jr ES, Rabinstein AA, Carhuapoma JR, Derdeyn CP, Dion J, Higashida RT, et al. Guidelines for the management of aneurysmal subarachnoid hemorrhage: a guideline for healthcare professionals from the American Heart Association/ American Stroke Association. Stroke 2012;43:1711-37.