Aneurysm Treatment <24 Versus h After Subarachnoid Hemorrhage

Transcription

1 Neurocrit Care (2014) 21:4 13 DOI /s ORIGINAL ARTICLE Aneurysm Treatment <24 Versus h After Subarachnoid Hemorrhage Simone C. Oudshoorn Gabriel J. E. Rinkel Andrew J. Molyneux Richard S. Kerr Sanne M. Dorhout Mees Daan Backes Ale Algra Mervyn D. I. Vergouwen Published online: 18 March 2014 Ó Springer Science+Business Media New York 2014 Abstract Introduction In patients with aneurysmal subarachnoid hemorrhage (asah), it is unclear whether aneurysm treatment <24 h after ictus results in better outcomes than treatment h after asah. We studied whether aneurysm occlusion <24 h is associated with better outcomes than occlusion h after asah. Methods We used two cohorts of patients with asah: (1) the UMC Utrecht cohort with patients admitted between 2008 and 2012 and (2) the International Subarachnoid Aneurysm Trial cohort. Aneurysm treatment was categorized into <24 h and h after ictus. We calculated adjusted risk ratios (arrs) with 95 % confidence intervals (CIs) using Poisson regression analyses for poor functional outcome (death or dependency) for both cohorts separately, and performed a pooled analysis based on individual patient data. We also performed a worst-case scenario analysis wherein all patients with rebleeding >3 h after admission were re-categorized into the group with aneurysm treatment h after asah. Results We included 1,238 patients (UMC Utrecht cohort: n = 330; ISAT: n = 908). The arr for poor outcome after treatment <24 h was in the UMC Utrecht cohort 1.84 (95 % CI: ), in ISAT 1.14 (95 % CI ), in the pooled analysis 1.37 (95 % CI ), and in the worst-case scenario pooled analysis 1.24 (95 % CI ). Conclusion Our results suggest that aneurysm occlusion can be performed in day time within 72 h after ictus, instead of on an emergency basis. However, due to the retrospective, non-randomized design of our study, our results cannot be considered as definitive evidence. Keywords Subarachnoid hemorrhage Aneurysm Treatment Clipping Coil embolization Timing Introduction S. C. Oudshoorn G. J. E. Rinkel S. M. Dorhout Mees D. Backes A. Algra M. D. I. Vergouwen (&) Department of Neurology and Neurosurgery, Brain Center Rudolf Magnus, University Medical Center Utrecht, Room G03-228, Heidelberglaan 100, 3584 CX Utrecht, The Netherlands m.d.i.vergouwen@umcutrecht.nl A. J. Molyneux R. S. Kerr Neurovascular Research Unit, Nuffield Department of Surgery, University of Oxford, Oxford, UK A. J. Molyneux R. S. Kerr Oxford Radcliffe Hospitals NHS Trust, John Radcliffe Hospital, Oxford, UK A. Algra Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Utrecht, The Netherlands Since a pivotal trial in the 1980s found no overall difference in outcome after aneurysmal subarachnoid hemorrhage (asah) between aneurysm treatment within 3 days of rupture versus postponed aneurysm treatment more than 7 days after rupture, timing of aneurysm treatment has shifted from a delayed phase to an early phase within 72 h after ictus [1]. The ESO guidelines recommend aneurysm treatment as early as logistically and technically possible within 72 h after asah to reduce the risk of rebleeding [2]. The guidelines of the American Heart Association recommend aneurysm treatment as early as feasible to reduce the rate of rebleeding (Class I, level of Evidence B) [3]. As a clinical implication, many centers nowadays try to occlude aneurysms on an emergency basis

2 Neurocrit Care (2014) 21: within a couple of hours, also during night time. However, evidence that aneurysm occlusion within 24 h after ictus improves outcome is lacking. Treatment within 24 h may even be detrimental, because processes induced by the initial ischemia from aneurysmal rupture may provoke delayed cerebral ischemia (DCI) [4]. Aneurysm treatment in itself also induces cerebral ischemia [5 9], and may therefore add to the risk of DCI if performed within the initial aneurysm rupture. Moreover, treatment within 24 h is a logistic and economic burden and increases burden on neurosurgeons and interventional neuroradiologists. Our objective was to investigate if aneurysm treatment within 24 h is indeed associated with better functional outcomes than treatment h after ictus. Methods For this observational study, we used two cohorts of patients with aneurysmal asah: patients retrieved from (1) the University Medical Center (UMC) Utrecht SAH database and (2) the International Subarachnoid Aneurysm Trial (ISAT) database. From both cohorts, we extracted the following variables: age, sex, clinical condition according to the World Federation of Neurological Surgeons (WFNS) scale [10], aneurysm location, maximum aneurysm size, timing of aneurysm treatment, modality of aneurysm treatment, occurrence of rebleeding and DCI, and functional outcome. Timing of aneurysm treatment was categorized into <24 h after ictus and h after asah. In the UMC Utrecht cohort, poor functional outcome was defined as a Glasgow Outcome Scale (GOS) score of 1 3 three months after ictus [11]. In ISAT, poor functional outcome was defined as a modified Rankin Scale score of 3 6 two months after asah [12]. UMC Utrecht Cohort The UMC Utrecht SAH database is a prospectively collected database that includes >3,500 consecutive patients with confirmed SAH. For the purpose of this study, we included all patients with asah admitted between January 1, 2008, and January 1, Approval was obtained from the Institutional Research Ethics Board. In the UMC Utrecht cohort, the time interval between ictus and aneurysm treatment was measured in hours. Patients were included if the aneurysm was treated within 72 h after ictus, or if aneurysm treatment was postponed to >72 h after ictus because of rebleeding C3 h after admission (these episodes of rebleeding could have been prevented by prompt aneurysm occlusion), or if patients were admitted in good clinical condition but received no aneurysm treatment because of rebleeding C3 h after admission. Aneurysm treatment included clipping and endovascular treatment. Exclusion criteria were (1) death imminent; (2) untreatable aneurysm; and (3) age <16 years. In the UMC Utrecht, all aneurysms are treated during daytime by experienced vascular neurosurgeons or interventional neuroradiologists. In the UMC Utrecht cohort, the occurrence of rebleeding was recorded between admission and aneurysm treatment, and defined as a sudden clinical deterioration with signs of increased hemorrhage on CT scan compared with previous CT imaging or found at autopsy, or a sudden clinical deterioration suspect for rebleeding with fresh blood in the ventricular drain in which no CT scan or autopsy was obtained. For clinical deterioration due to DCI we used the definition that was recently proposed by an international multidisciplinary research group [13]. Functional outcome 3 months after SAH was typically assessed by a stroke nurse, either during a visit to the outpatient clinic or by telephone interview. ISAT Cohort ISAT is a randomized trial that compared the efficacy of neurosurgical clipping with endovascular coiling in 2,143 patients with asah [12]. Patients were eligible for inclusion in ISAT if there was a confirmed SAH within the previous 28 days, and an aneurysm that was suitable for both neurosurgical clipping and endovascular coiling, but with uncertainty on which treatment was most suitable. Only patients 18 years or older were included. In ISAT, the WFNS score was recorded at time of randomization, the time interval between ictus and aneurysm treatment was measured in days, and the occurrence of rebleeding was recorded between randomization and aneurysm treatment. Patients were included if the aneurysm was treated at day 0 (maximum 24 h after ictus), day 1 or 2 (maximum 72 h after ictus) after ictus, if rebleeding occurred between randomization and aneurysm treatment with treatment beyond day 2, or if rebleeding occurred after randomization and no aneurysm treatment was performed. For the purpose of this study, aneurysm treatment at day 0 was considered <24 h after ictus, and treatment at days 1 or 2 as h after asah. Patients were excluded if no follow-up data were available on either functional outcome or case fatality 2 months after ictus. Modality of aneurysm treatment was defined as the type of aneurysm treatment the patient actually received, also if treatment crossover occurred. In the ISAT manual, clinical deterioration due to DCI was defined as clinical grounds of delayed ischemic neurological deficit, not due to operative factors or other factors such as procedural vessel occlusion, hydrocephalus or aneurysmal rebleeding, and by the clinicians judged to be due to vasospasm [14]. Clinical outcome was assessed by self-reported questionnaires

3 6 Neurocrit Care (2014) 21:4 13 with the modified Rankin Scale score two months after ictus [15]. Outcome Measures Primary outcome was poor functional outcome. Secondary outcomes in the UMC Utrecht cohort were case fatality three months after asah and clinical deterioration due to DCI. Secondary outcomes in the ISAT cohort were case fatality 2 months after asah and clinical deterioration due to DCI. Statistical Analyses First, we performed an on-treatment analysis and compared the primary and secondary outcome measures between the groups of patients with aneurysm treatment <24 h and h after asah. Risk ratios (RR) with 95 % confidence intervals (CI) were calculated with the group of patients treated h after ictus as the reference group. Adjusted risk ratios (arrs) were calculated using Poisson regression analyses with adjustments for the following predefined variables: age, sex, WFNS score, location of aneurysm, maximum aneurysm size, and type of aneurysm treatment. To compare the proportion of patients with DCI in patients with aneurysm treatment <24 h and h after ictus, we calculated arrs with 95 % CIs. For the pooled analyses, adjustments were made for all six predefined variables plus asah cohort. Second, we performed a worst-case scenario analysis for the primary outcome measure. In the UMC Utrecht cohort, we re-categorized all patients with intention of aneurysm treatment who had rebleeding C3 h after admission to the group of patients treated h after admission. This analysis was performed to assure that all patients who had rebleeding within the initial 24 h were analyzed as if the intention was to perform aneurysm treatment in the h interval. For the same reason, we included in the h interval all patients with postponement of aneurysm treatment to >72 h because of rebleeding and patients with no aneurysm treatment because of rebleeding. For ISAT, we re-categorized all patients with intention of aneurysm treatment and rebleeding to the group of patients treated h after asah. For this worst-case scenario analysis, risk ratios were calculated with adjustments for all predefined variables as described previously. Third, we performed a sensitivity analysis for the primary outcome measure in which timing of aneurysm treatment in both cohorts was defined similar to the original categorization in ISAT (treatment at day 0 vs. treatment at days 1 2). Fourth, we performed subgroup analyses for the primary outcome measure for type of aneurysm treatment (clipped or coiled) and for WFNS score. For these subgroup analyses, risk ratios were calculated with adjustments for all predefined variables except for the variable under investigation. For the on-treatment, worst-case scenario, and sensitivity analyses, we first performed the analyses for both cohorts separately, and subsequently a pooled analysis was performed based on individual patient data. For the pooled analyses, we made adjustments not only for the predefined variables, but also for the asah cohort since it was a confounder. Subgroup analyses were performed in the pooled dataset only to avoid small numbers of patients in certain subgroups. Finally, we performed a literature search and metaanalysis of the studies found. We searched Pubmed with the terms subarachnoid hemorrhage AND aneurysm treatment AND ( timing OR 24 h ) AND outcome for publications comparing poor functional outcome (death or dependency) in patients with asah and aneurysm treatment <24 h versus aneurysm treatment >24 h after ictus, regardless of study quality, and language restricted to English, French, German, and Spanish. We calculated crude RR with 95 % CI for individual studies and pooled RRs. Results During the study period, 569 patients with asah were admitted to the UMC Utrecht. For the analyses, we included 330 patients (<24 h: n = 134; h: n = 180; treatment postponed to >72 h because of rebleeding C3 h after admission: n = 4; no aneurysm treatment despite intention to treatment but rebleeding C3 h after admission: n = 12) (Fig. 1). Baseline characteristics of patients treated <24 h and h after ictus are listed in Table 1. Rebleeding before aneurysm occlusion occurred in 14 patients (10 %) with aneurysm treatment <24 h and in 5 patients (3 %) with aneurysm treatment h after ictus. From the ISAT cohort, we included 908 patients (day 0: n = 83; days 1 2: n = 804; beyond day 2: n = 9; no aneurysm treatment because of rebleeding after randomization: n = 12) (Fig. 2). Reasons for exclusion were aneurysm treatment beyond day 2 after asah without rebleeding (n = 1,231) and missing data on outcome (n = 4). Baseline characteristics are listed in Table 1. Rebleeding occurred in 1 patient (1 %) treated at day 0 and in 17 patients (2 %) treated at days 1 2. Data on functional outcome 2 months after ictus were available in 81 patients treated at day 0 and 800 patients treated at days 1 2 after asah. Median number of days from admission to randomization in the included patients was 0 (IQR 0-0).

4 Neurocrit Care (2014) 21: Fig. 1 Flowchart of UMC Utrecht cohort Patients presenting with asah during study period, n=569 Excluded: Death imminent or short life-expectancy, n = 99 Untreatable aneurysm, n = 6 Age <16 years, n = 1 intention to aneurysm treatment, n = 463 Excluded: Good clinical condition on admission, but deterioration <3 admission, n = 8 GOS after three months unknown, n = 2 <24 asah, n = asah, n = 180 >72 asah, n = 127 No aneurysm treatment despite good clinical condition on admission, but admission, n = 12 admission, n = 5 admission, n = 1 admission, n = 4 admission, n = 12 Included in group <24 hours, n = 134 Included in group hours, n = 180 Included in worstcase scenario analysis, n = 4 Included in worstcase scenario analysis, n = 12 On-treatment Analyses Results of the on-treatment analyses are shown in Table 2. Treatment <24 h was not associated with better functional outcome (arr in pooled analysis: 1.37; 95 % CI ) or a reduction in case fatality (arr in pooled analysis: 1.51; 95 % CI ). The arr for clinical deterioration due to DCI after treatment <24 h was in the UMC Utrecht cohort 1.13 (95 % CI ), in ISAT 1.02 (95 % CI ), and in the pooled analysis 1.01 (95 % CI ). Worst-Case Scenario Analysis Accounting for Potentially Preventable Rebleeding Results of the worst-case scenario analyses are shown in Table 2. In the UMC Utrecht cohort, 12 patients with intention to aneurysm treatment had rebleeding C3 h after

5 8 Neurocrit Care (2014) 21:4 13 Table 1 Baseline characteristics, in-hospital complications, and outcomes according to time to treatment UMC Utrecht cohort ISAT cohort Pooled databases 0 24 h h 0 24 h h 0 24 h h N = 134 N = 180 N = 83 N = 804 N = 217 N = 984 Median age (IQR) 55 (47 62) 56 (47 66) 53 (45 60) 51 (42 58) 54 (46 61) 51 (43 59) Females 88 (66) 131 (73) 59 (71) 500 (62) 147 (68) 631 (64) WFNS score on admission 1 (GSC 15) 42 (31) 87 (48) 32 (39) 483 (60) 74 (34) 570 (58) 2 (GCS 13 14) 39 (29) 43 (24) 32 (39) 218 (27) 71 (33) 261 (27) 3 (GCS 13 14) 6 (5) 9 (5) 3 (4) 43 (5) 9 (4) 52 (5) 4 (GCS 7 12) 25 (19) 26 (14) 7 (8) 36 (5) 32 (15) 62 (6) 5 (GCS < 7) 21 (16) 11 (6) 3 (4) 12 (2) 24 (11) 23 (2) Unclear 1 (1) 4 (2) 6 (7) 12 (2) 7 (3) 16 (2) Posterior circulation aneurysm 7 (5) 28 (16) 3 (4) 22 (3 %) 10 (5) 50 (5) Median aneurysm size in mm (IQR) 6 (4 8) 6 (4 8) 6 (4 8) 5 (4 8) 6 (4 8) 6 (4 8) Type of treatment Endovascular 65 (49) 94 (52) 51 (61) 431 (54) 116 (54) 525 (53) Clipping 69 (52) 86 (48) 32 (39) 373 (46) 101 (47) 459 (47) Complications Rebleeding 14 (10) a 5 (3) 1 (1) 17 (2) 15 (7) 22 (2) DCI 37 (28) 36 (20) 22 (27) 195 (24) 59 (27) 231 (24) Poor functional outcome 51 (38) 31 (17) 32/81 (40) 215/800 (27) 83/215 (39) 246/980 (25) Case-fatality 20 (15) 13 (7) 11 (13) 58 (7) 31 (14) 71 (7) Data are number (%) unless indicated otherwise IQR interquartile range, WFNS World Federation of Neurological Surgeons, DCI delayed cerebral ischemia, CT computed tomography a In 5 of these 14 patients rebleeding occurred > 3 h after admission. These 5 patients were in the worst case scenario analysis grouped into the h treatment group admission, and therefore did not receive aneurysm treatment. In addition, five patients with aneurysm treatment <24 h after ictus and four patients with aneurysm treatment >72 h after asah had rebleeding C3 h after admission. If these 21 patients were added to the group of patients with aneurysm treatment h after ictus, the arr for poor outcome after treatment <24 h was 1.50 (95 % CI ). In ISAT, 12 patients had rebleeding after randomization, and therefore no aneurysm treatment was performed. In addition, 9 patients with aneurysm treatment beyond day 2 had rebleeding and 1 patient with aneurysm treatment at day 0. If these 22 patients were added to the group of patients with aneurysm treatment h after ictus, the arr for poor outcome after treatment <24 h was 1.12 (95 % CI ). In the worstcase scenario pooled analysis, the arr for poor outcome after treatment <24 h was 1.24 (95 % CI ). Sensitivity Analysis Results of the sensitivity analyses are shown in Table 2. If time of aneurysm treatment in the UMC Utrecht cohort was categorized in days instead of hours similar to the original categorization in ISAT, 286 patients were included in the UMC Utrecht cohort. The arr for poor outcome after treatment at day 0 in the UMC Utrecht cohort was 1.57 (95 % CI ), and in the pooled analysis was 1.26 ( ). Subgroup Analyses Results of subgroup analyses of pooled cohorts are shown in Fig. 3. Treatment <24 h was not associated with better functional outcome in any subgroup. Meta-analysis of Previous Studies Four recent studies compared functional outcomes in patients treated <24 and >24 h after asah (Table 3) [16 19]. One of these studies performed a subgroup analysis in which patients treated <24 h were compared with those treated 1 3 days after asah [16]. The pooled RR for aneurysm treatment <24 h versus >24 h was 0.76 (95 % CI ) and for aneurysm treatment <24 h versus h was 1.34 ( ) (Fig. 4a, b).

6 Neurocrit Care (2014) 21: Fig. 2 Flowchart of ISAT cohort asah with intention to aneurysm treatment, n = 2143 Excluded: Missing data on outcome, n = 4 at day 0, n = 83 at days 1-2, n = 804 beyond day 2, n = 1240 No aneurysm treatment despite good clinical condition on admission, but rebleeding after randomization, n = 12 rebleeding after randomization, n=1 rebleeding after randomization, n=17 rebleeding after randomization, n=9 admission, n = 12 Included in group <24 hours, n = 83 Included in group hours, n = 804 Included in worstcase scenario analysis, n=9 Included in worstcase scenario analysis, n=12 Table 2 Poor functional outcome and case fatality according to time to treatment UMC Utrecht cohort ISAT cohort Pooled cohorts 0 24 h h a 0 24 h h a 0 24 h h a Poor functional outcome (n/n (%)) 51/134 (38) 31/180 (17) 32/81 (40) 215/800 (27) 83/215 (39) 246/980 (25) Unadjusted risk ratio 2.21 ( ) ( ) ( ) b 1 Adjusted risk ratios On-treatment analysis 1.84 ( ) ( ) ( ) b 1 Worst-case scenario analysis 1.50 ( ) ( ) ( ) b 1 Sensitivity analysis 1.57 ( ) ( ) ( ) b 1 Case-fatality (n/n (%)) 20/134 (15) 13/180 (7) 11/83 (13) 58/804 (7) 31/217 (14) 71/984 (7) Unadjusted risk ratio 2.07 ( ) ( ) ( ) b 1 Adjusted risk ratios On-treatment analysis 1.88 ( ) ( ) ( ) b 1 Data are number (%) or risk ratio (95 % CI) WFNS World Federation of Neurological Surgeons a Reference group b All pooled analyses were (also) adjusted for asah cohort Discussion In two large independent cohorts of asah patients, aneurysm treatment <24 h did not result in better outcomes than treatment h after ictus. This result was robust even if we performed a worst-case scenario analysis accounting for cases of rebleeding that could have been prevented if the aneurysm was treated immediately after

7 10 Neurocrit Care (2014) 21:4 13 Fig. 3 Results of subgroup analyses of pooled cohorts. Adjusted risk ratios with 95 % CI for poor functional outcome. <24 h are compared with those treated h after ictus admission. Also in none of the subgroup analyses was treatment <24 h associated with better functional outcome than treatment h after ictus. The meta-analysis including previous studies shows that aneurysm treatment <24 h SAH is better than treatment >24 h, but not h after ictus. One study investigated the relationship between timing of aneurysm treatment and rebleeding and did not find such a relationship [17]. In contrast with two previous studies suggesting that aneurysm treatment <24 h is associated with better outcomes [16, 19], we did not include patients treated >72 h after ictus, since delayed treatment beyond 72 h mostly results from late admission, logistic problems, or complex aneurysm characteristics, and therefore should not be combined with that of patients treated within 72 h. In the subgroup analysis that found a statistically non-significant reduction in poor outcome in patients treated <24 h versus those treated h after asah, no adjustments were made for baseline characteristics [16]. Neither previous study performed a worst-case scenario analysis as we did, to account for the rebleeding that might have resulted in postponement or cancellation of aneurysm treatment [16, 17, 19]. Despite the theoretical advantage of preventing episodes of rebleeding by emergency aneurysm occlusion, we found no improved outcome from aneurysm treatment <24 h, but a worse outcome instead. A first explanation for lack of effectiveness of aneurysm treatment <24 h is that many instances of rebleeding occur soon after admission. Indeed, in the UMC Utrecht cohort, two-thirds of episodes of rebleeding occurred within 3 h after admission. It will be almost impossible to perform aneurysm occlusion within this time frame. Moreover, the beneficial effect of aneurysm treatment <24 h on prevention of rebleeding may be offset by impeded recovery from early brain injury after asah. In recent years, there is increased interest in early brain injury after asah resulting from increased intracranial pressure, microvascular alterations, platelet aggregation, acute vasospasm, and reperfusion injury [20 24]. These factors may contribute to the finding of a worse outcome in the <24 h aneurysm treatment group. The strength of our study is that we investigated two cohorts of asah patients, in which functional outcome was collected prospectively and blinded for the present research question. In the UMC Utrecht cohort, loss to follow-up was less than 1 %. We performed regression analyses and a worst-case scenario analysis to account for rebleeding which did not change our conclusions. We were also able to perform pooled analyses based on individual patient data. A limitation of our study is that it was a retrospective, non-randomized comparison. As a result, no proper intention-to-treat analysis could be performed. However, the

8 Neurocrit Care (2014) 21: Table 3 Overview of studies comparing outcomes in patients with aneurysm treatment <24 h versus >24 h after ictus Study N a Comparison of timing of aneurysm treatment in hours Number of patients with poor outcome (n/n (%)) b Crude risk ratio (95 % CI) Phillips et al. [16] 408 <24 versus >24 h <24 h: 16/199 (8 %) 0.56 ( ) >24 h: 30/209 (14 %) 382 <24 versus h <24 h: 16/199 (8 %) 0.64 ( ) h: 23/183 (13 %) Wong et al. [17] 276 <24 versus >24 h <24 h: 54/148 (36 %) 0.90 ( ) >24 h: 52/128 (41 %) Sandström et al. [18] 122 <24 versus >24 h <24 h: 33/80 (41 %) 0.87 ( ) >24 h: 20/42 (48 %) Gu et al. [19] 96 <24 versus >24 h <24 h: 7/56 (13 %) 0.42 ( ) >24 h: 12/40 (30 %) Oudshoorn et al. 314 <24 versus h <24 h: 51/134 (38 %) 2.21 ( ) (UMCU) h: 31/180 (17 %) 881 <24 versus h <24 h: 32/81 (40 %) 1.47 ( ) (ISAT) h: 215/800 (27 %) mrs modified Rankin scale score, CI confidence interval a Total number of patients with available outcomes b Defined as modified Rankin scale score of 3 6 or GOS score of 1 3 A Study or Subgroup Treatment <24 h Events Total Treatment >24 h Events Total Weight Risk Ratio M-H, Fixed, 95% CI Year Risk Ratio M-H, Fixed, 95% CI Phillips 2011 Wong 2012 Gu 2012 Sandström % 44.5% 11.2% 20.9% 0.56 [0.32, 1.00] 0.90 [0.67, 1.21] 0.42 [0.18, 0.96] 0.87 [0.57, 1.31] Total (95% CI) % 0.76 [0.61, 0.94] Total events Heterogeneity: Chi² = 4.66, df = 3 (P = 0.20); I² = 36% Test for overall effect: Z = 2.48 (P = 0.01) Favours treatment <24 h Favours treatment >24 h B Study or Subgroup Treatment <24 h Events Total Treatment h Events Total Weight Risk Ratio M-H, Random, 95% CI Year Risk Ratio M-H, Random, 95% CI Phillips 2011 Oudshoorn 2014 UMCU Oudshoorn 2014 ISAT % 34.7% 37.2% 0.64 [0.35, 1.17] [1.50, 3.25] [1.10, 1.97] 2014 Total (95% CI) % 1.34 [0.76, 2.37] Total events Heterogeneity: Tau² = 0.20; Chi² = 11.54, df = 2 (P = 0.003); I² = 83% Test for overall effect: Z = 1.01 (P = 0.31) Favours treatment 24 h Favours treatment h Fig. 4 a Pooled (crude) relative risk estimates for poor functional outcome for patients with aneurysm treatment <24 h versus >24 h after ictus. b Pooled (crude) relative risk estimates for poor functional outcome for patients with aneurysm treatment <24 h versus h after ictus worst-case scenario analysis, where we regrouped all patients with rebleeding >3 h after admission into the h treatment group to simulate a worst-case intention-to-treat analysis, did not change the results essentially. The non-randomized design may have resulted in baseline differences between the groups, because the decision to treat the aneurysm <24 h or h after ictus may have been influenced by the clinical condition of the patient. Although we performed regression analyses with adjustments for variables such as WFNS score and aneurysm size and location, and performed subgroup analyses for WFNS score which did not change our findings, it could be that

9 12 Neurocrit Care (2014) 21:4 13 there were other factors that we did not account for. Another limitation is that definitions in ISAT were somewhat different than in the UMC Utrecht cohort. In ISAT, day of aneurysm treatment was expressed in days after asah, with day of ictus defined as day 0. As a result, treatment at day 1 also included patients who were treated within 24 h after asah. To account for differences in definition, we performed a sensitivity analysis using the ISAT definition in both cohorts, which did not change the results essentially. In addition, in ISAT, rebleeding was defined as recurrent bleeding between randomization and treatment. Therefore, cases with rebleeding between admission and randomization were not included in this analysis, which explains the lower incidence of rebleeding in ISAT compared with the UMC Utrecht cohort. Our results suggest that in clinical practice, aneurysm occlusion can be performed in day time within 72 h after ictus, instead of on an emergency basis. However, due to the retrospective, non-randomized design, our results are no sound evidence for either strategy. Such evidence should preferably come from a randomized trial comparing aneurysm treatment <24 h versus treatment h after ictus, but the feasibility of such a trial is questionable. Acknowledgments Funding None. None. Conflict of interest Andrew Molyneux reports a consulting agreement with Sequent Medical Inc. All other authors have no conflicts of interest. References 1. Ohman J, Heiskanen O. Timing of operation for ruptured supratentorial aneurysms: a prospective randomized study. J Neurosurg. 1989;70: Steiner T, Juvela S, Unterberg A, Jung C, Forsting M, Rinkel G, European Stroke Organization. European Stroke Organization guidelines for the management of intracranial aneurysms and subarachnoid haemorrhage. Cerebrovasc Dis. 2013;35: Connolly ES Jr, Rabinstein AA, Carhuapoma JR, Derdeyn CP, Dion J, Higashida RT, American Heart Association Stroke Council, Council on Cardiovascular Radiology and Intervention, Council on Cardiovascular Nursing, Council on Cardiovascular Surgery and Anesthesia, Council on Clinical Cardiology, 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: Brilstra EH, Rinkel GJ, Algra A, van Gijn J. Rebleeding, secondary ischemia, and timing of operation in patients with subarachnoid hemorrhage. Neurology. 2000;55: Cronqvist M, Wirestam R, Ramgren B, Brandt L, Nilsson O, Saveland H, et al. Diffusion and perfusion MRI in patients with ruptured and unruptured intracranial aneurysms treated by endovascular coiling: complications, procedural results, MR findings and clinical outcome. J Neuroradiol. 2005;47: Grunwald IQ, Papanagiotou P, Politi M, Struffert T, Roth C, Reith W. Endovascular treatment of unruptured intracranial aneurysms: occurrence of thromboembolic events. J Neurosurg. 2006;58: Hahnemann ML, Ringelstein A, Sandalcioglu IE, Goericke S, Moenninghoff C, Wanke I, et al. Silent embolism after stentassisted coiling of cerebral aneurysms: diffusion-weighted MRI study of 75 cases. J Neurointerv Surg. 2013;. doi: / neurintsurg Kang DH, Kim BM, Kim DJ, Suh SH, Kim DI, Kim YS, et al. MR-DWI-positive lesions and symptomatic ischemic complications after coiling of unruptured intracranial aneurysms. Stroke. 2013;44: Soeda A, Sakai N, Murao K, Sakai H, Ihara K, Yamada N, et al. Thromboembolic events associated with Guglielmi detachable coil embolization with use of diffusion-weighted MR imaging. Part II. Detection of the microemboli proximal to cerebral aneurysm. AJNR Am J Neuroradiol. 2003;24: Teasdale GM, Drake CG, Hunt W, Kassell N, Sano K, Pertuiset B, et al. A universal subarachnoid hemorrhage scale: report of a committee of the World Federation of Neurosurgical Societies. J Neurol Neurosurg Psychiatry. 1988;51: Jennett B, Teasdale G, Braakman R, Minderhoud J, Knill-Jones R. Predicting outcome in individual patients after severe head injury. Lancet. 1976;1: Molyneux A, Kerr R, Stratton I, Sandercock P, Clarke M, Shrimpton J, International Subarachnoid Aneurysm Trial (ISAT) Collaborative Group, et al. International Subarachnoid Aneurysm Trial (ISAT) of neurosurgical clipping versus endovascular coiling in 2143 patients with ruptured intracranial aneurysms: a randomised trial. Lancet. 2002;360: Vergouwen MD, Vermeulen M, van Gijn J, Rinkel GJ, Wijdicks EF, Muizelaar JP, et al. Definition of delayed cerebral ischemia after aneurysmal subarachnoid hemorrhage as an outcome event in clinical trials and observational studies: proposal of a multidisciplinary research group. Stroke. 2010;41: Dorhout Mees SM, Kerr RS, Rinkel GJ, Algra A, Molyneux AJ. Occurrence and impact of delayed cerebral ischemia after coiling and after clipping in the International Subarachnoid Aneurysm Trial (ISAT). J Neurol. 2012;259: van Swieten JC, Koudstaal PJ, Visser MC, Schouten HJ, van Gijn J. Interobserver agreement for the assessment of handicap in stroke patients. Stroke. 1988;19: Phillips TJ, Dowling RJ, Yan B, Laidlaw JD, Mitchell PJ. Does treatment of ruptured intracranial aneurysms within 24 hours improve clinical outcome? Stroke. 2011;42: Wong GK, Boet R, Ng SC, Chan MT, Gin T, Zee B, et al. Ultraearly (within 24 hours) aneurysm treatment after subarachnoid hemorrhage. World Neurosurg. 2012;77: Sandström N, Yan B, Dowling R, Laidlaw J, Mitchell P. Comparison of microsurgery and endovascular treatment on clinical outcome following poor-grade subarachnoid hemorrhage. J Clin Neurosci. 2013;20: Gu DQ, Zhang X, Luo B, Long XA, Duan CZ. Impact of ultraearly coiling on clinical outcome after aneurysmal subarachnoid hemorrhage in elderly patients. Acad Radiol. 2012;19: Sehba FA, Hou J, Pluta RM, Zhang JH. The importance of early brain injury after subarachnoid hemorrhage. Prog Neurobiol. 2012;97: Sehba FA, Mostafa G, Knopman J, Friedrich V Jr, Bederson JB. Acute alterations in microvascular basal lamina after subarachnoid hemorrhage. J Neurosurg. 2004;101: Sehba FA, Mostafa G, Friedrich V Jr, Bederson JB. Acute microvascular platelet aggregation after subarachnoid hemorrhage. J Neurosurg. 2005;102:

10 Neurocrit Care (2014) 21: Bederson JB, Levy AL, Ding WH, Kahn R, DiPerna CA, Jenkins AL III, et al. Acute vasoconstriction after subarachnoid hemorrhage. J Neurosurg. 1998;42: Claassen J, Carhuapoma JR, Kreiter KT, Du EY, Connolly ES, Mayer SA. Global cerebral edema after subarachnoid hemorrhage: frequency, predictors, and impact on outcome. Stroke. 2002;33: