- Email: [email protected]
Spontaneous Subarachnoid Hemorrhage with Negative Angiography Managed in a Stroke Unit: Clinical and Prognostic Characteristics
Spontaneous Subarachnoid Hemorrhage with Negative Angiography Managed in a Stroke Unit: Clinical and Prognostic Characteristics s Mosqueira, MD, Florentino Nombela, MD, PhD, Beatrice Canneti, MD, Antonio Jesu Francisco Gilo, MD, and Jose Vivancos, MD, PhD
Background: Perimesencephalic subarachnoid hemorrhage (p-SAH) is linked to a benign prognosis compared with aneurysmal SAH. However, the outcome in nonperimesencephalic angiographically negative SAH (np-SAH) is not well established. We reviewed our experience and evaluated the clinical and prognostic differences between patients with p-SAH and np-SAH. Methods: Retrospective observational study based on data collected prospectively from all patients admitted to our hospital with SAH during the past 11 years. After selecting patients with normal angiography, we categorized them as either p-SAH or np-SAH according to the Rinkel criteria. Demographic, clinical, radiologic, and prognostic features were recorded. Results: We collected a total of 41 (12.53%) angiographically negative SAH: 17 p-SAH (41.46%) and 24 np-SAH (58%-53%). The np-SAH group included the 6 patients with Glasgow Coma Scale (GCS) less than 15 (P 5 .083), and all 5 patients with Hunt & Hess (H&H) scores more than II (P 5 .045), being the GCS 5 15 and H&H less than II in the rest of np-SAH and in all of the p-SAH patients. The average hospital stay in days was longer in the np-SAH group (24 6 7.08) than in the p-SAH group (17 6 5.11; P 5 .55). Hydrocephalus requiring external ventricular drainage (EVD) was only recorded in the np-SAH group (29.16%, P 5 .029). On discharge, all patients had H&H grade less than II and modified Rankin Scale measured 3 months later was less than 2 in both groups. Conclusions: Our results agree with other studies showing a low complication rate for SAH patients with a normal angiography, especially in the p-SAH group. The prognosis appears to be less favorable in terms of a more frequent need for EVD, so a more cautious approach is recommended in this subgroup. Key Words: Nonaneurysmal subarachnoid hemorrhage—perimesencephalic hemorrhage—angiographically negative subarachnoid hemorrhage—CT scan—angiography—neuroradiology. Ó 2015 by National Stroke Association
From the Stroke Unit, Neurology Department, Hospital Universitario de La Princesa, Instituto de Investigaci on Sanitaria Princesa, Madrid, Spain. Received March 23, 2015; revision received May 18, 2015; accepted June 14, 2015. None of the authors have any relevant interests to declare. Address correspondence to Beatrice Canneti Hospital Universitario La Princesa. Diego de Le on, 62. 28006. Madrid, Spain. E-mail: canneti. [email protected]. 1052-3057/$ - see front matter Ó 2015 by National Stroke Association http://dx.doi.org/10.1016/j.jstrokecerebrovasdis.2015.06.011
Introduction Spontaneous subarachnoid hemorrhage (SAH) is an event usually caused by the presence of an underlying vascular disease and most frequently caused by the rupture of a saccular aneurysm, although other causes have been described (arterial dissection, tumors, and so forth). In about 15% of all patients, angiographic studies show no vascular pathology that would cause bleeding. Such cases are referred to as idiopathic SAH, angiographically
Journal of Stroke and Cerebrovascular Diseases, Vol. -, No. - (---), 2015: pp 1-7
1
B. CANNETI ET AL.
2
negative, or SAH with normal or nondiagnostic angiography.1-3 In some of these patients, the extravasated blood is typically confined to the cisterns surrounding the midbrain, it does not extend to the ventricular system or exit the posterior fossa, and the parenchyma is not affected. This pattern corresponds to the perimesencephalic subarachnoid hemorrhage (p-SAH).4 A p-SAH pattern observed on initial computed tomography (CT) has a characteristically high predictive value for a normal angiographic study,5 although aneurysm in the posterior circulation is responsible for the bleed in about 10% of cases.5-7 The p-SAH pattern is defined according to the Rinkel criteria for findings in simple cranial CT in the acute phase.4 This pattern has been linked to an excellent long-term clinical prognosis, with a much lower complication rate than that of aneurysmal SAH.5,8-15 Because of differences in behavior in the acute phase, and in medium- and long-term prognosis for these SAH patterns, skipping the second angiography has become a common practice in the management of these patients,16,17 without new reports so far of any increases in long-term complications in p-SAH patients.18 However, there are also angiographically negative SAH cases in which the CT blood pattern is not confined to the basal cisterns (BCs) but extends to the ventricular system or peripheral locations such as the Sylvian fissure or the convexity. These cases do not therefore fulfill diagnostic criteria for p-SAH, and they have been referred to as nonperimesencephalic, aneurysmlike, or diffuse angiographically negative SAH. Although most of these cases are idiopathic, researchers have identified different etiologies, including hidden thrombosed or microscopic aneurysms that are not detectable by conventional angiographic techniques, as well as other less common causes such as arteriovenous malformations masked by vasospasm or a deep venous anomaly.6,19-23 In these patients, the diagnostic and therapeutic approach is less well defined. At present, it has not been conclusively shown that a negative angiographic study indicates that outcomes will be as favorable as in p-SAH. The studies published so far suggest that prognosis is poorer in these patients, but because the group is etiologically heterogeneous, treating patients on a case-by-case basis is recommended. The aim of this study was to describe the demographic, clinical, radiologic, and prognostic features of angiographically negative cases of subarachnoid hemorrhage attended in a tertiary care hospital. By reviewing cranial CT studies performed in the acute phase, we can compare patients meeting the criteria for p-SAH to those who do not (np-SAH). We attempt to establish whether the latter is a distinct subtype of angiographically negative SAH.
Patients and Methods We conducted a retrospective observational study based on data collected prospectively from all patients admitted to our hospital and diagnosed with SAH during the past 11 years. From among these patients, we selected those whose initial angiographic study was normal and categorized them as p-SAH or np-SAH using the criteria described by Rinkel et al4 for the initial CT. According to those criteria, we excluded patients with acute CT performed after the first 72 hours. The Rinkel criteria specify that: (1) the center of the hemorrhage is located immediately anterior to the midbrain or brainstem with possible extension of blood to the anterior part of the ambient cistern or to the basal part of the Sylvian fissure; (2) no complete filling of the anterior interhemispheric fissure and no extension to the lateral Sylvian fissure, except for minute amounts of blood; (3) absence of frank intraventricular hemorrhage, except for minute amounts of blood; and (4) absence of intracerebral hematoma. Patients in whom CT was performed more than 72 hours after symptom onset were placed in the np-SAH group, and we indicated whether this was the reason for not meeting the Rinkel criteria. We also recorded the predominant location of bleeding in the np-SAH group and established 3 subgroups: bleeding mostly limited to BCs, convexity/cortical location, and intraventricular bleeding. Cases of SAH with negative CT findings (SAH diagnosed by cerebrospinal fluid analysis) was not included in this study. We collected the baseline characteristics from both groups: sex, age, vascular risk factors, and previous antithrombotic treatment. We also recorded data describing clinical presentation: location and type of headache, associated symptoms, presence of meningeal signs, altered consciousness according to the Glasgow Coma Scale (GCS) and the Hunt & Hess (H&H) scale in the acute phase. Our team also analyzed the type and number of angiographic studies and administrative variables such as the admitting department and length of hospital stay. To assess patient progress and prognosis, we registered the incidence of seizures, presence of decreased level of consciousness at time of admission, incidence of rebleeding, presence of vasospasm shown by ultrasound and angiography, presence of acute hydrocephalus (in patients requiring an external ventricular drainage [EVD]), cardiac electric conduction complications, electrolyte disturbances, and number of deaths. Clinical outcome at time of discharge was assessed on the GCS and H&H grades at discharge and the modified Rankin Scale (mRS) at 3 months. The study is guided by the basic ethical principles in the Declaration of Helsinki. The highest standards of professional conduct have always been maintained, and patient confidentiality has been ensured at all times. It has
ANGIOGRAPHICALLY NEGATIVE SAH IN A STROKE UNIT
been applied the national legislation on data protection (Organic Law 15/1999 on Protection of Personal Data). In the statistical analysis, the chi-square test was used to compare proportions. The Fisher exact test was also applied when necessary. P values less than .05 were considered statistically significant, and P values between .05 and .10 showed a trend toward significance. Statistical analysis was completed with SPSS version 17 for Windows.
Results Between 1 January, 2002, and 31 March, 2013, a total of 327 patients with spontaneous SAH were attended in our hospital. The source of the bleed was located in 286 cases (87.46%), whereas 41 (12.53%) were angiographically negative. According to the pattern of blood distribution on the baseline CT in the latter group, a total of 17 patients with p-SAH (41.46%) and 24 with np-SAH (58.53%) were identified; they represented 5.19% and 7.33% of the SAH cases, respectively. A total of 16 patients (66.66%) were placed in the np-SAH group based on blood distribution, whereas the remaining 8 (33.33%) were included based on time elapsed between symptom onset and completion of the initial CT. According to the initial CT, most of the npSAH patients (21, 87.5%) had blood in the BCs, 14 patients (58.3%) showed some cortical bleeding, and 6 patients (25%) had intraventricular blood.
General Characteristics and Clinical Presentation The mean age at presentation was 51.52 6 9.43 years in the p-SAH group and 57.43 6 10.82 years in np-SAH group. Sex ratio was close to 50% in both groups. Except for dyslipidemia, which was more frequent in the npSAH group (P 5 .028), no significant differences were found for any vascular risk factors or for history of antithrombotic treatment (Table 1). Table 1. Demographic variables Demographic variables
p-SAH, np-SAH, n 5 17, n (%) n 5 24, n (%)
P
Sex (M) 9 (52.94) 12 (50) ns Age, y 51.52 6 9.43 57.43 6 10.82 ns Arterial 2 (11.76) 7 (29.16) ns hypertension Active smoking 4 (23.52) 5 (20.83) ns Diabetes mellitus 1 (5.88) 2 (8.33) ns Dyslipidemia 1 (5.88) 9 (37.5) .028 Antithrombotic 1 (5.88) 4 (16.66) ns treatment Abbreviations: M, male; np-SAH, nonperimesencephalic angiographically negative SAH; ns, not significant; p-SAH, perimesencephalic SAH; SAH, subarachnoid hemorrhage.
3
Table 2. Clinical presentation
Clinical presentation Thunderclap headache Occipital headache Valsalva/exercise Nausea/vomiting Syncope Meningeal signs GCS GCS 15 Stuporous (GCS 14-9) Coma (8-3) Hunt & Hess I-II III-IV V
p-SAH, n 5 17, n (%)
np-SAH, n 5 24, n (%)
15 (88.23) 16 (66.66) 9 (52.94) 8 (33.33) 8 (47.05) 8 (33.33) 13 (76.47) 21 (87.5) 0 2 (8.33) 11 (64.7) 11 (45.83) 17 (100) 0 0 17 0 0
P ns ns ns ns ns ns
18 (75) ns 5 (20.83) .083 1 (4.16) 19 (79.16) .045 5 (20.83) 0
Abbreviations: GCS, Glasgow Coma Scale; np-SAH, nonperimesencephalic angiographically negative SAH; ns, not significant; p-SAH, perimesencephalic SAH; SAH, subarachnoid hemorrhage.
Regarding clinical presentation (Table 2), there were no significant differences between the 2 groups except for GCS. Headache was more frequent in the p-SAH group (88%) than in the np-SAH group (66%), whereas the percentage of patients with nausea or vomiting was slightly higher in the np-SAH group (87% versus 76%), although none of these differences showed statistical significance. Only 2 patients lost consciousness, and both were in the np-SAH group (8.3%). Clinical condition on admission was poorer in the np-SAH group. Level of consciousness was normal (GCS, 15) in all patients with p-SAH, whereas the np-SAH group contained 5 stuporous patients (20.83%) and 1 comatose patient (4.16%); these differences showed a trend toward statistical significance (P 5 .083). The entire p-SAH group and most np-SAH patients (79.16%) had an H&H grade between I and II. The 5 patients with grades higher than II (III or IV) all belonged to the np-SAH group; the difference was statistically significant (P 5 .045). No grades of V on the H&H scale were recorded.
Diagnostic Process All p-SAH patients and most np-SAH patients (79.16%) were admitted to the stroke unit (Table 3). The rest of the np-SAH group was distributed between the Neurosurgery department (4 patients, 16.6%) and the ICU (1 patient, 4.16%). As indicated by current guidelines, all patients underwent arteriography at some point during hospitalization. However, the most common initial angiographic study in both groups was CT angiography (88.23% and 75%,
B. CANNETI ET AL.
4
Table 3. Administrative variables Admitting department Stroke unit Neurosurgery ICU Hospital stay (d)
p-SAH, n (%) np-SAH, n (%) 17 (100) 0 0 17 6 5.11
19 (79.16) 4 (16.66) 1 (4.16) 24 6 7.08
P ns ns ns .055
Abbreviations: ICU, intensive care unit; np-SAH, nonperimesencephalic angiographically negative SAH; ns, not significant; p-SAH, perimesencephalic SAH; SAH, subarachnoid hemorrhage.
respectively). The second study performed was an arteriography in all patients with p-SAH and most of those with np-SAH (70.83%). MR angiography was performed for the remaining 12.5% of the np-SAH group. Median time between symptom onset and the initial angiographic study was less than 24 hours p-SAH group and 3 days in the np-SAH group. Approximately half of the patients in each group underwent 3 or more angiographic studies, none of which showed vascular abnormalities that would have caused the bleed (Table 4).
Figure 1. Main SAH complications. *P , .05. Abbreviations: np-SAH, nonperimesencephalic angiographically negative SAHp-SAH, perimesencephalic SAH; SAH, subarachnoid hemorrhage.
the patients with np-SAH showed CT evidence of hydrocephalus (not significant [ns]). Seven patients in this group required EVD placement (17.07%), whereas there were no cases in the p-SAH group (P 5 .029). Our series showed a significant association between this complication and presence of intraventricular blood. Rebleeding and ischemia were events that only occurred in the np-SAH group. Angiographic vasospasm was detected in 1 patient in the p-SAH group versus 3 (12.5%) in
Outcomes The average hospital stay in days was longer in the np-SAH group (24 6 7.08) than in the p-SAH group (17 6 5.11; P 5 .55; Table 4). Complications associated with SAH are summarized in Figure 1 and Table 5. All were more frequent in the np-SAH group. Epileptic seizures, decreased level of consciousness, and acute hydrocephalus requiring EVD were only present in the np-SAH group. A higher percentage of Table 4. Angiographic studies
Angiographic studies Type of first angiography Arteriography CT MRI Type of second angiography Total of second arteriographies Arteriography CT MRI $3 Angiographies
p-SAH, n 5 17, n (%)
np-SAH, n 5 24, n (%)
P ns
2 (11.76) 15 (88.23) 0
8 (33.33) 18 (75) 0 ns
17 (100)
20 (83.33)
17 (100) 0 0 10 (58.82)
17 (85) 0 3 (15) 16 (66.66)
Table 5. Complications and outcomes p-SAH, n 5 17, n (%) Rebleeding 0 Ventricular dilatation 4 (23.52) in CT Ischemia 0 Angiographic vasospasm 1 (6.88) Hydrocephalus (EVD) 0 Seizures 0 Decreased level of 0 consciousness H&H at discharge I-II 16 (94.11) III-IV 0 V 0 mRS at 3 mo mRS 0-2 7 (41.17) mRS 3-4 0 mRS 5-6 0 TCU showing vasospasm Yes 3 (17.64) No 7 (41.17) Not performed 7 (41.17)
np-SAH, n 5 24, n (%)
P
1 (4.16) 8 (33.33)
ns ns
1 (4.16) ns 3 (12.5) ns 7 (29.16) .029 2 (8.33) ns 4 (16.66) ns ns 20 (83.33) 0 0 ns 13 (54.16) 0 0 ns 2 (8.33) 11 (45.83) 11 (45.83)
ns
Abbreviations: CT, computed tomography; MRI, magnetic resonance imaging; np-SAH, nonperimesencephalic angiographically negative SAH; ns, not significant; p-SAH, perimesencephalic SAH; SAH, subarachnoid hemorrhage.
Abbreviations: CT, computed tomography; EVD, external ventricular drainage; H&H, Hunt & Hess; mRS, modified Rankin Scale; np-SAH, nonperimesencephalic angiographically negative SAH; ns, not significant; p-SAH, perimesencephalic SAH; SAH, subarachnoid hemorrhage; TCU, transcranial ultrasound.
ANGIOGRAPHICALLY NEGATIVE SAH IN A STROKE UNIT
Figure 2. Hospital stay, Hunt&Hess (H&H) at discharge, and modified Rankin Scale (mRS) at 3 months. P , .05. Abbreviations: np-SAH, nonperimesencephalic angiographically negative SAH; p-SAH, perimesencephalic SAH; SAH, subarachnoid hemorrhage.
the np-SAH group (ns). Vasospasm was detected by transcranial ultrasound in 3 np-SAH patients and 2 p-SAH patients, but these data should be interpreted with caution because they were missing in about half the patients in each group. No deaths were recorded in either group. Neither of the groups contained cases of cardiac abnormalities or electrolyte imbalances. At time of discharge, all patients displayed grades below II on the H&H scale; mRS at 3 months was less than 2 in both groups (Fig. 2).
Discussion Our results support those from other studies describing a more favorable overall prognosis in SAH with negative angiography; these cases show a very low complication rate and good long-term outcomes compared with aneurysmal SAH. During the past 10 years, we have not encountered any patients in whom a second angiographic study revealed an aneurysm that was not detected by the first one. This contrasts with experiences in other series, which point to rates of positives in second studies ranging from 4% to 30%.13,24-26 This being the case, we must highlight that several of these series predate many of the recently introduced technical advances in neuroimaging and specifically in angiography (digital subtraction techniques and so forth). This may result in overestimation of the number of normal second angiograms leading to false negative results, difficult to ensure in our case because of the small size of our sample. Because none of the second arteriographies carried out in our study let us identify aneurysms or other vascular malformations, this test might be regarded as superfluous. However, the high mortality associated with rebleeding (up to 4% during the first 24 hours and up to 15%-25% during the first 14 days, depending on the series),26,27 together with the low morbidity of arteriography, tilts the risk–benefit balance toward repeating this test. This is especially important if vasospasm is present because
5
detecting saccular aneurysms is especially difficult in this context.28 An analysis of only those SAH cases that do not meet p-SAH criteria reveals poorer clinical status at baseline, a longer mean hospital stay, and significantly higher rates of hydrocephalus requiring EVD. Seizures and decreased level of consciousness occurred only in the np-SAH group. Regarding the longer hospitalization stay in the p-SAH group, it should be taken into consideration that 4 patients of this group (16.6%) were admitted in the ICU, which could lead to a longer length of stay as admission to a critical care unit results in a longer stay even with the same disease severity. Poorer prognosis in np-SAH patients with regard to complications is consistent with reports in other published series (Table 6). The meta-analysis by Boswell et al31 reports a significantly lower probability of rebleeding in the p-SAH group (odds ratio, 2.78; P 5 .013). Kang et al38 determined a 27.6% rate of radiologic vasospasm in their np-SAH group versus no cases in the p-SAH group (P 5 .006). They also observed a 37.9% rate of hydrocephalus in the imaging study among np-SAH patients compared with 8.7% of the p-SAH group (odds ratio 6.42, P 5 .026). Regarding the latter complication, however, 4 articles did not report any statistically significant differences.5,37,39,42 Ildan et al19 compared seizure frequency between these 2 groups and found a seizure rate of 3.6% in p-SAH patients versus 6.3% in np-SAH patients (ns). The same study reports statistically significant higher rates of permanent ischemic damage and subsequent death among patients with np-SAH (P 5 .05). Long-term functional prognosis seems to be good in both groups. Nevertheless, we were unable to obtain complete data on the 3-month mRS in a significant number of patients because the medium- and long-term follow-up was not performed in our center. However, any neurosurgical/neurointerventional complications or deaths during the outpatient follow-up if they occurred, would have been notified and treated in our unit. In our study, the only baseline feature to appear more frequently in the np-SAH group is dyslipidemia, which is consistent with results from the Zhong et al series and others. Zhong et al30 also found a higher frequency of hypertension, diabetes, and smoking in the np-SAH subgroup and in aneurysmal SAH. Exposure to these factors could explain the poorer response of the parenchyma and vessels to stress, thereby suggesting a poorer prognosis after bleeding. We have not identified any other baseline clinical data pointing to a poorer outcome, despite the differences in the clinical presentation of np-SAH and p-SAH. The main limitation of this study is because of its retrospective deign and sample size, so conclusions should be taken with caution.
B. CANNETI ET AL.
6
Table 6. Previous idiopathic SAH articles comparing p-SAH and np-SAH
Author
Year
Journal
Patel NB29
2014
Zhong W30 Boswell S31 Khan AA32
2014 2013 2013
Moscovici S33 Lin N34 Canovas D35 Sarabia R1
2013 2012 2012 2010
J Neurol Surg; A Cent Eur Neurosurg Turk; Neurosurg J Stroke Cerebrovasc Dis Clin Neurology and Neurosurgery Neurol Res Neurocrit Care Eur J Neurol. Neurocrit Care
Arauz A36 Jung JY37 Kang DH38 Hui FK2 Andaluz N39 Topcuoglu MA28 Lagares A40 Ildan F19 Ruigrok YM41 Canhao P42 Van Calenbergh F6 Rinkel GJ43
2007 2006 2009 2009 2008 2003 2002 2002 2002 1995 1993 1991
Neurologıa J Clin Neurosci J Clin Neurosci Neurocrit Care Neurosurgery J Neurosurg. Neurocirugia (Astur) Neurosurgery Cerebrovasc Dis Acta Neurochir (Wien) Surg Neurol Lancet
Type of analysis
Time, y
Retrospective
N (p-SAH/ np-SAH)
Better prognosis p-SAH
88
Yes
Retrospective Retrospective Retrospective
3 8 5
83 31 50
Yes Yes Yes
Retrospective Retrospective Retrospective Retrospective, multicenter Retrospective Retrospective Retrospective Retrospective Retrospective Retrospective Retrospective Retrospective Retrospective Retrospective Retrospective Retrospective
6 5 13 3
56 66 108 220
Yes Yes Yes Yes
5 18
50 143 52 94 92 86 97 84 88 71 62 113
No Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
6 5 6.5 10 5.4 7 7
Abbreviations: np-SAH, nonperimesencephalic angiographically negative SAH; p-SAH, perimesencephalic SAH; SAH, subarachnoid hemorrhage.
Although p-SAH and np-SAH share a prognosis that is good overall, we must pay particular attention to the distribution of blood on the initial CT in nonaneurysmal SAH cases because the pattern appears to be associated with an increased risk of complications in np-SAH.
Conclusions SAH managed in a Stroke Unit with initial normal arteriography findings has a generally good prognosis compared with that of aneurysmal SAH. However, one type of nonaneurysmal SAH, np-SAH, may be less benign than p-SAH, and it presents a higher risk of complications in the acute phase. Extreme caution is therefore needed with these cases, and we recommend applying a diagnostic and therapeutic approach similar to that used in aneurysmal SAH. Acknowledgments: To the Spanish Neurological Society (SEN) for the language support.
References 1. Sarabia R, Lagares A, Fern andez Alen JA, et al. Idiopathic subarachnoid hemorrhage: a multicentre series of 220 patients. Neurocirugia (Astur) 2010;21:441-451.
2. Hui F, Schuette A, Moskowitz S, et al. Antithrombotic states and outcomes in patients with angiographically negative subarachnoid hemorrhage. Neurosurgery 2011; 68:125-130. discussion 30. 3. Whiting J, Reavey Cantwell J, Velat G, et al. Clinical course of nontraumatic, nonaneurysmal subarachnoid hemorrhage: a single-institution experience. Neurosurg Focus 2009;26:E21. 4. van Gijn J, van Dongen KJ, Vermeulen M, et al. Perimesencephalic hemorrhage: a nonaneurysmal and benign form of subarachnoid hemorrhage. Neurology 1985; 35:493-497. 5. Rinkel GJ, Wijdicks EF, Vermeulen M, et al. Nonaneurysmal perimesencephalic subarachnoid hemorrhage: CT and MR patterns that differ from aneurysmal rupture. AJNR Am J Neuroradiol 1991;12:829-834. 6. Calenbergh FV, Plets C, Goffin J, et al. Nonaneurysmal subarachnoid hemorrhage: prevalence of perimesencephalic hemorrhage in a consecutive series. Surg Neurol 1993;39:320-323. 7. Pinto AN, Ferro JM, Canh~ ao P, et al. How often is a perimesencephalic subarachnoid haemorrhage CT pattern caused by ruptured aneurysms? Acta Neurochir (Wien) 1993;124:79-81. 8. Alexander MS, Dias PS, Uttley D. Spontaneous subarachnoid hemorrhage and negative cerebral panangiography. Review of 140 cases. J Neurosurg 1986;64:537-542. 9. Cioffi F, Pasqualin A, Cavazzani P, et al. Subarachnoid haemorrhage of unknown origin: clinical and tomographical aspects. Acta Neurochir (Wien) 1989;97:31-39.
ANGIOGRAPHICALLY NEGATIVE SAH IN A STROKE UNIT 10. Giombini S, Bruzzone MG, Pluchino F. Subarachnoid hemorrhage of unexplained cause. Neurosurgery 1988; 22:313-316. 11. Eskesen V, Sørensen EB, Rosenørn J, et al. The prognosis in subarachnoid hemorrhage of unknown etiology. J Neurosurg 1984;61:1029-1031. 12. Barlow P. Incidence of delayed cerebral ischaemia following subarachnoid haemorrhage of unknown cause. J Neurol Neurosurg Psychiatry 1985;48:132-136. 13. Nishioka H, Torner JC, Graf CJ, et al. Cooperative study of intracranial aneurysms and subarachnoid hemorrhage: a long-term prognostic study. III. Subarachnoid hemorrhage of undetermined etiology. Arch Neurol 1984;41:1147-1151. 14. Brismar J, Sundb€ arg G. Subarachnoid hemorrhage of unknown origin: prognosis and prognostic factors. J Neurosurg 1985;63:349-354. 15. Spallone A, Ferrante L, Palatinsky E, et al. Subarachnoid haemorrhage of unknown origin. Acta Neurochir (Wien) 1986;80:12-17. 16. Kaim A, Proske M, Kirsch E, et al. Value of repeatangiography in cases of unexplained subarachnoid hemorrhage (SAH). Acta Neurol Scand 1996;93:366-373. 17. Rinkel GJ, van Gijn J, Wijdicks EF. Subarachnoid hemorrhage without detectable aneurysm. A review of the causes. Stroke 1993;24:1403-1409. 18. Greebe P, Rinkel GJ. Life expectancy after perimesencephalic subarachnoid hemorrhage. Stroke 2007;38: 1222-1224. 19. Ildan F, Tuna M, Erman T, et al. Prognosis and prognostic factors for unexplained subarachnoid hemorrhage: review of 84 cases. Neurosurgery 2002;50:1015-1024. discussion 24–5. 20. Forster DM, Steiner L, Hakanson S, et al. The value of repeat pan-angiography in cases of unexplained subarachnoid hemorrhage. J Neurosurg 1978;48:712-716. 21. Alen JF, Lagares A, Campollo J, et al. Idiopathic subarachnoid hemorrhage and venous drainage: are they related? Neurosurgery 2008;63:1106-1111. discussion 11–2. 22. Yuji M, Kiyoyuki Y, Ai M, et al. Significance of a small bulge on the basilar artery in patients with perimesencephalic nonaneurysmal subarachnoid hemorrhage. Report of two cases. J Neurosurg 2003;98:426-429. 23. Tatter SB, Crowell RM, Ogilvy CS. Aneurysmal and microaneurysmal ‘‘angiogram-negative’’ subarachnoid hemorrhage. Neurosurgery 1995;37:48-55. 24. Juul R, Fredriksen TA, Ringkjøb R. Prognosis in subarachnoid hemorrhage of unknown etiology. J Neurosurg 1986; 64:359-362. 25. Pathirana N, Refsum SE, McKinstry CS, et al. The value of repeat cerebral angiography in subarachnoid haemorrhage. Br J Neurosurg 1994;8:141-146. 26. Iwanaga H, Wakai S, Ochiai C, et al. Ruptured cerebral aneurysms missed by initial angiographic study. Neurosurgery 1990;27:45-51. 27. Inamasu J, Nakamura Y, Saito R, et al. ‘‘Occult’’ ruptured cerebral aneurysms revealed by repeat angiography: result from a large retrospective study. Clin Neurol Neurosurg 2003;106:33-37. 28. Topcuoglu MA, Ogilvy CS, Carter BS, et al. Subarachnoid hemorrhage without evident cause on initial angiog-
7
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
raphy studies: diagnostic yield of subsequent angiography and other neuroimaging tests. J Neurosurg 2003; 98:1235-1240. Patel NB, Patel AD, Wilkinson J, et al. An analysis of 88 patients with diffuse and ‘‘benign’’ perimesencephalic subarachnoid hemorrhage. J Neurol Surg A Cent Eur Neurosurg 2014;75:299-304. Zhong W, Zhao P, Wang D, et al. Different clinical characteristics between perimesencephalic subarachnoid hemorrhage and diffuse subarachnoid hemorrhage with negative initial angiography. Turk Neurosurg 2014; 24:327-332. Boswell S, Thorell W, Gogela S, et al. Angiogram-negative subarachnoid hemorrhage: outcomes data and review of the literature. J Stroke Cerebrovasc Dis 2013;22:750-757. Khan AA, Smith JD, Kirkman MA, et al. Angiogram negative subarachnoid haemorrhage: outcomes and the role of repeat angiography. Clin Neurol Neurosurg 2013;115:1470-1475. Moscovici S, Fraifeld S, Ramirez-de-Noriega F, et al. Clinical relevance of negative initial angiogram in spontaneous subarachnoid hemorrhage. Neurol Res 2013; 35:117-122. Lin N, Zenonos G, Kim AH, et al. Angiogram-negative subarachnoid hemorrhage: relationship between bleeding pattern and clinical outcome. Neurocrit Care 2012; 16:389-398. Canovas D, Gil A, Jato M, et al. Clinical outcome of spontaneous non- aneurysmal subarachnoid hemorrhage in 108 patients. Eur J Neurol 2012;19:457-461. Arauz A, Lopez M, Cantu C, et al. [Nonaneurysmal subarachnoid hemorrhage. Study of long-term follow-up]. Neurologia 2007;22:502-506. Jung JY, Kim YB, Lee JW, et al. Spontaneous subarachnoid haemorrhage with negative initial angiography: a review of 143 cases. J Clin Neurosci 2006;13:1011-1017. Kang DH, Park J, Lee SH, et al. Does nonperimesencephalic type non-aneurysmal subarachnoid hemorrhage have a benign prognosis? J Clin Neurosci 2009;16:904-908. Andaluz N, Zuccarello M. Yield of further diagnostic work-up of cryptogenic subarachnoid hemorrhage based on bleeding patterns on computed tomographic scans. Neurosurgery 2008;62:1040-1046. discussion 7. Lagares A, Gomez PA, Lobato RD, et al. [Idiopathic subarachnoid hemorrhage; comparison of different bleeding patterns and long-term outcome]. Neurocirugia 2002; 13:110-119. Ruigrok YM, Rinkel GJ, Van Gijn J. CT patterns and long-term outcome in patients with an aneurysmal type of subarachnoid hemorrhage and repeatedly negative angiograms. Cerebrovasc Dis 2002;14:221-227. Canh~ ao P, Ferro JM, Pinto AN, et al. Perimesencephalic and nonperimesencephalic subarachnoid haemorrhages with negative angiograms. Acta Neurochir (Wien) 1995; 132:14-19. Rinkel GJ, Wijdicks EF, Hasan D, Kienstra GE, Franke CL, Hageman LM, et al. Outcome in patients with subarachnoid haemorrhage and negative angiography according to pattern of haemorrhage on computed tomography. Lancet 1991;338:964-968.
Background: Perimesencephalic subarachnoid hemorrhage (p-SAH) is linked to a benign prognosis compared with aneurysmal SAH. However, the outcome in nonperimesencephalic angiographically negative SAH (np-SAH) is not well established. We reviewed our experience and evaluated the clinical and prognostic differences between patients with p-SAH and np-SAH. Methods: Retrospective observational study based on data collected prospectively from all patients admitted to our hospital with SAH during the past 11 years. After selecting patients with normal angiography, we categorized them as either p-SAH or np-SAH according to the Rinkel criteria. Demographic, clinical, radiologic, and prognostic features were recorded. Results: We collected a total of 41 (12.53%) angiographically negative SAH: 17 p-SAH (41.46%) and 24 np-SAH (58%-53%). The np-SAH group included the 6 patients with Glasgow Coma Scale (GCS) less than 15 (P 5 .083), and all 5 patients with Hunt & Hess (H&H) scores more than II (P 5 .045), being the GCS 5 15 and H&H less than II in the rest of np-SAH and in all of the p-SAH patients. The average hospital stay in days was longer in the np-SAH group (24 6 7.08) than in the p-SAH group (17 6 5.11; P 5 .55). Hydrocephalus requiring external ventricular drainage (EVD) was only recorded in the np-SAH group (29.16%, P 5 .029). On discharge, all patients had H&H grade less than II and modified Rankin Scale measured 3 months later was less than 2 in both groups. Conclusions: Our results agree with other studies showing a low complication rate for SAH patients with a normal angiography, especially in the p-SAH group. The prognosis appears to be less favorable in terms of a more frequent need for EVD, so a more cautious approach is recommended in this subgroup. Key Words: Nonaneurysmal subarachnoid hemorrhage—perimesencephalic hemorrhage—angiographically negative subarachnoid hemorrhage—CT scan—angiography—neuroradiology. Ó 2015 by National Stroke Association
From the Stroke Unit, Neurology Department, Hospital Universitario de La Princesa, Instituto de Investigaci on Sanitaria Princesa, Madrid, Spain. Received March 23, 2015; revision received May 18, 2015; accepted June 14, 2015. None of the authors have any relevant interests to declare. Address correspondence to Beatrice Canneti Hospital Universitario La Princesa. Diego de Le on, 62. 28006. Madrid, Spain. E-mail: canneti. [email protected]. 1052-3057/$ - see front matter Ó 2015 by National Stroke Association http://dx.doi.org/10.1016/j.jstrokecerebrovasdis.2015.06.011
Introduction Spontaneous subarachnoid hemorrhage (SAH) is an event usually caused by the presence of an underlying vascular disease and most frequently caused by the rupture of a saccular aneurysm, although other causes have been described (arterial dissection, tumors, and so forth). In about 15% of all patients, angiographic studies show no vascular pathology that would cause bleeding. Such cases are referred to as idiopathic SAH, angiographically
Journal of Stroke and Cerebrovascular Diseases, Vol. -, No. - (---), 2015: pp 1-7
1
B. CANNETI ET AL.
2
negative, or SAH with normal or nondiagnostic angiography.1-3 In some of these patients, the extravasated blood is typically confined to the cisterns surrounding the midbrain, it does not extend to the ventricular system or exit the posterior fossa, and the parenchyma is not affected. This pattern corresponds to the perimesencephalic subarachnoid hemorrhage (p-SAH).4 A p-SAH pattern observed on initial computed tomography (CT) has a characteristically high predictive value for a normal angiographic study,5 although aneurysm in the posterior circulation is responsible for the bleed in about 10% of cases.5-7 The p-SAH pattern is defined according to the Rinkel criteria for findings in simple cranial CT in the acute phase.4 This pattern has been linked to an excellent long-term clinical prognosis, with a much lower complication rate than that of aneurysmal SAH.5,8-15 Because of differences in behavior in the acute phase, and in medium- and long-term prognosis for these SAH patterns, skipping the second angiography has become a common practice in the management of these patients,16,17 without new reports so far of any increases in long-term complications in p-SAH patients.18 However, there are also angiographically negative SAH cases in which the CT blood pattern is not confined to the basal cisterns (BCs) but extends to the ventricular system or peripheral locations such as the Sylvian fissure or the convexity. These cases do not therefore fulfill diagnostic criteria for p-SAH, and they have been referred to as nonperimesencephalic, aneurysmlike, or diffuse angiographically negative SAH. Although most of these cases are idiopathic, researchers have identified different etiologies, including hidden thrombosed or microscopic aneurysms that are not detectable by conventional angiographic techniques, as well as other less common causes such as arteriovenous malformations masked by vasospasm or a deep venous anomaly.6,19-23 In these patients, the diagnostic and therapeutic approach is less well defined. At present, it has not been conclusively shown that a negative angiographic study indicates that outcomes will be as favorable as in p-SAH. The studies published so far suggest that prognosis is poorer in these patients, but because the group is etiologically heterogeneous, treating patients on a case-by-case basis is recommended. The aim of this study was to describe the demographic, clinical, radiologic, and prognostic features of angiographically negative cases of subarachnoid hemorrhage attended in a tertiary care hospital. By reviewing cranial CT studies performed in the acute phase, we can compare patients meeting the criteria for p-SAH to those who do not (np-SAH). We attempt to establish whether the latter is a distinct subtype of angiographically negative SAH.
Patients and Methods We conducted a retrospective observational study based on data collected prospectively from all patients admitted to our hospital and diagnosed with SAH during the past 11 years. From among these patients, we selected those whose initial angiographic study was normal and categorized them as p-SAH or np-SAH using the criteria described by Rinkel et al4 for the initial CT. According to those criteria, we excluded patients with acute CT performed after the first 72 hours. The Rinkel criteria specify that: (1) the center of the hemorrhage is located immediately anterior to the midbrain or brainstem with possible extension of blood to the anterior part of the ambient cistern or to the basal part of the Sylvian fissure; (2) no complete filling of the anterior interhemispheric fissure and no extension to the lateral Sylvian fissure, except for minute amounts of blood; (3) absence of frank intraventricular hemorrhage, except for minute amounts of blood; and (4) absence of intracerebral hematoma. Patients in whom CT was performed more than 72 hours after symptom onset were placed in the np-SAH group, and we indicated whether this was the reason for not meeting the Rinkel criteria. We also recorded the predominant location of bleeding in the np-SAH group and established 3 subgroups: bleeding mostly limited to BCs, convexity/cortical location, and intraventricular bleeding. Cases of SAH with negative CT findings (SAH diagnosed by cerebrospinal fluid analysis) was not included in this study. We collected the baseline characteristics from both groups: sex, age, vascular risk factors, and previous antithrombotic treatment. We also recorded data describing clinical presentation: location and type of headache, associated symptoms, presence of meningeal signs, altered consciousness according to the Glasgow Coma Scale (GCS) and the Hunt & Hess (H&H) scale in the acute phase. Our team also analyzed the type and number of angiographic studies and administrative variables such as the admitting department and length of hospital stay. To assess patient progress and prognosis, we registered the incidence of seizures, presence of decreased level of consciousness at time of admission, incidence of rebleeding, presence of vasospasm shown by ultrasound and angiography, presence of acute hydrocephalus (in patients requiring an external ventricular drainage [EVD]), cardiac electric conduction complications, electrolyte disturbances, and number of deaths. Clinical outcome at time of discharge was assessed on the GCS and H&H grades at discharge and the modified Rankin Scale (mRS) at 3 months. The study is guided by the basic ethical principles in the Declaration of Helsinki. The highest standards of professional conduct have always been maintained, and patient confidentiality has been ensured at all times. It has
ANGIOGRAPHICALLY NEGATIVE SAH IN A STROKE UNIT
been applied the national legislation on data protection (Organic Law 15/1999 on Protection of Personal Data). In the statistical analysis, the chi-square test was used to compare proportions. The Fisher exact test was also applied when necessary. P values less than .05 were considered statistically significant, and P values between .05 and .10 showed a trend toward significance. Statistical analysis was completed with SPSS version 17 for Windows.
Results Between 1 January, 2002, and 31 March, 2013, a total of 327 patients with spontaneous SAH were attended in our hospital. The source of the bleed was located in 286 cases (87.46%), whereas 41 (12.53%) were angiographically negative. According to the pattern of blood distribution on the baseline CT in the latter group, a total of 17 patients with p-SAH (41.46%) and 24 with np-SAH (58.53%) were identified; they represented 5.19% and 7.33% of the SAH cases, respectively. A total of 16 patients (66.66%) were placed in the np-SAH group based on blood distribution, whereas the remaining 8 (33.33%) were included based on time elapsed between symptom onset and completion of the initial CT. According to the initial CT, most of the npSAH patients (21, 87.5%) had blood in the BCs, 14 patients (58.3%) showed some cortical bleeding, and 6 patients (25%) had intraventricular blood.
General Characteristics and Clinical Presentation The mean age at presentation was 51.52 6 9.43 years in the p-SAH group and 57.43 6 10.82 years in np-SAH group. Sex ratio was close to 50% in both groups. Except for dyslipidemia, which was more frequent in the npSAH group (P 5 .028), no significant differences were found for any vascular risk factors or for history of antithrombotic treatment (Table 1). Table 1. Demographic variables Demographic variables
p-SAH, np-SAH, n 5 17, n (%) n 5 24, n (%)
P
Sex (M) 9 (52.94) 12 (50) ns Age, y 51.52 6 9.43 57.43 6 10.82 ns Arterial 2 (11.76) 7 (29.16) ns hypertension Active smoking 4 (23.52) 5 (20.83) ns Diabetes mellitus 1 (5.88) 2 (8.33) ns Dyslipidemia 1 (5.88) 9 (37.5) .028 Antithrombotic 1 (5.88) 4 (16.66) ns treatment Abbreviations: M, male; np-SAH, nonperimesencephalic angiographically negative SAH; ns, not significant; p-SAH, perimesencephalic SAH; SAH, subarachnoid hemorrhage.
3
Table 2. Clinical presentation
Clinical presentation Thunderclap headache Occipital headache Valsalva/exercise Nausea/vomiting Syncope Meningeal signs GCS GCS 15 Stuporous (GCS 14-9) Coma (8-3) Hunt & Hess I-II III-IV V
p-SAH, n 5 17, n (%)
np-SAH, n 5 24, n (%)
15 (88.23) 16 (66.66) 9 (52.94) 8 (33.33) 8 (47.05) 8 (33.33) 13 (76.47) 21 (87.5) 0 2 (8.33) 11 (64.7) 11 (45.83) 17 (100) 0 0 17 0 0
P ns ns ns ns ns ns
18 (75) ns 5 (20.83) .083 1 (4.16) 19 (79.16) .045 5 (20.83) 0
Abbreviations: GCS, Glasgow Coma Scale; np-SAH, nonperimesencephalic angiographically negative SAH; ns, not significant; p-SAH, perimesencephalic SAH; SAH, subarachnoid hemorrhage.
Regarding clinical presentation (Table 2), there were no significant differences between the 2 groups except for GCS. Headache was more frequent in the p-SAH group (88%) than in the np-SAH group (66%), whereas the percentage of patients with nausea or vomiting was slightly higher in the np-SAH group (87% versus 76%), although none of these differences showed statistical significance. Only 2 patients lost consciousness, and both were in the np-SAH group (8.3%). Clinical condition on admission was poorer in the np-SAH group. Level of consciousness was normal (GCS, 15) in all patients with p-SAH, whereas the np-SAH group contained 5 stuporous patients (20.83%) and 1 comatose patient (4.16%); these differences showed a trend toward statistical significance (P 5 .083). The entire p-SAH group and most np-SAH patients (79.16%) had an H&H grade between I and II. The 5 patients with grades higher than II (III or IV) all belonged to the np-SAH group; the difference was statistically significant (P 5 .045). No grades of V on the H&H scale were recorded.
Diagnostic Process All p-SAH patients and most np-SAH patients (79.16%) were admitted to the stroke unit (Table 3). The rest of the np-SAH group was distributed between the Neurosurgery department (4 patients, 16.6%) and the ICU (1 patient, 4.16%). As indicated by current guidelines, all patients underwent arteriography at some point during hospitalization. However, the most common initial angiographic study in both groups was CT angiography (88.23% and 75%,
B. CANNETI ET AL.
4
Table 3. Administrative variables Admitting department Stroke unit Neurosurgery ICU Hospital stay (d)
p-SAH, n (%) np-SAH, n (%) 17 (100) 0 0 17 6 5.11
19 (79.16) 4 (16.66) 1 (4.16) 24 6 7.08
P ns ns ns .055
Abbreviations: ICU, intensive care unit; np-SAH, nonperimesencephalic angiographically negative SAH; ns, not significant; p-SAH, perimesencephalic SAH; SAH, subarachnoid hemorrhage.
respectively). The second study performed was an arteriography in all patients with p-SAH and most of those with np-SAH (70.83%). MR angiography was performed for the remaining 12.5% of the np-SAH group. Median time between symptom onset and the initial angiographic study was less than 24 hours p-SAH group and 3 days in the np-SAH group. Approximately half of the patients in each group underwent 3 or more angiographic studies, none of which showed vascular abnormalities that would have caused the bleed (Table 4).
Figure 1. Main SAH complications. *P , .05. Abbreviations: np-SAH, nonperimesencephalic angiographically negative SAHp-SAH, perimesencephalic SAH; SAH, subarachnoid hemorrhage.
the patients with np-SAH showed CT evidence of hydrocephalus (not significant [ns]). Seven patients in this group required EVD placement (17.07%), whereas there were no cases in the p-SAH group (P 5 .029). Our series showed a significant association between this complication and presence of intraventricular blood. Rebleeding and ischemia were events that only occurred in the np-SAH group. Angiographic vasospasm was detected in 1 patient in the p-SAH group versus 3 (12.5%) in
Outcomes The average hospital stay in days was longer in the np-SAH group (24 6 7.08) than in the p-SAH group (17 6 5.11; P 5 .55; Table 4). Complications associated with SAH are summarized in Figure 1 and Table 5. All were more frequent in the np-SAH group. Epileptic seizures, decreased level of consciousness, and acute hydrocephalus requiring EVD were only present in the np-SAH group. A higher percentage of Table 4. Angiographic studies
Angiographic studies Type of first angiography Arteriography CT MRI Type of second angiography Total of second arteriographies Arteriography CT MRI $3 Angiographies
p-SAH, n 5 17, n (%)
np-SAH, n 5 24, n (%)
P ns
2 (11.76) 15 (88.23) 0
8 (33.33) 18 (75) 0 ns
17 (100)
20 (83.33)
17 (100) 0 0 10 (58.82)
17 (85) 0 3 (15) 16 (66.66)
Table 5. Complications and outcomes p-SAH, n 5 17, n (%) Rebleeding 0 Ventricular dilatation 4 (23.52) in CT Ischemia 0 Angiographic vasospasm 1 (6.88) Hydrocephalus (EVD) 0 Seizures 0 Decreased level of 0 consciousness H&H at discharge I-II 16 (94.11) III-IV 0 V 0 mRS at 3 mo mRS 0-2 7 (41.17) mRS 3-4 0 mRS 5-6 0 TCU showing vasospasm Yes 3 (17.64) No 7 (41.17) Not performed 7 (41.17)
np-SAH, n 5 24, n (%)
P
1 (4.16) 8 (33.33)
ns ns
1 (4.16) ns 3 (12.5) ns 7 (29.16) .029 2 (8.33) ns 4 (16.66) ns ns 20 (83.33) 0 0 ns 13 (54.16) 0 0 ns 2 (8.33) 11 (45.83) 11 (45.83)
ns
Abbreviations: CT, computed tomography; MRI, magnetic resonance imaging; np-SAH, nonperimesencephalic angiographically negative SAH; ns, not significant; p-SAH, perimesencephalic SAH; SAH, subarachnoid hemorrhage.
Abbreviations: CT, computed tomography; EVD, external ventricular drainage; H&H, Hunt & Hess; mRS, modified Rankin Scale; np-SAH, nonperimesencephalic angiographically negative SAH; ns, not significant; p-SAH, perimesencephalic SAH; SAH, subarachnoid hemorrhage; TCU, transcranial ultrasound.
ANGIOGRAPHICALLY NEGATIVE SAH IN A STROKE UNIT
Figure 2. Hospital stay, Hunt&Hess (H&H) at discharge, and modified Rankin Scale (mRS) at 3 months. P , .05. Abbreviations: np-SAH, nonperimesencephalic angiographically negative SAH; p-SAH, perimesencephalic SAH; SAH, subarachnoid hemorrhage.
the np-SAH group (ns). Vasospasm was detected by transcranial ultrasound in 3 np-SAH patients and 2 p-SAH patients, but these data should be interpreted with caution because they were missing in about half the patients in each group. No deaths were recorded in either group. Neither of the groups contained cases of cardiac abnormalities or electrolyte imbalances. At time of discharge, all patients displayed grades below II on the H&H scale; mRS at 3 months was less than 2 in both groups (Fig. 2).
Discussion Our results support those from other studies describing a more favorable overall prognosis in SAH with negative angiography; these cases show a very low complication rate and good long-term outcomes compared with aneurysmal SAH. During the past 10 years, we have not encountered any patients in whom a second angiographic study revealed an aneurysm that was not detected by the first one. This contrasts with experiences in other series, which point to rates of positives in second studies ranging from 4% to 30%.13,24-26 This being the case, we must highlight that several of these series predate many of the recently introduced technical advances in neuroimaging and specifically in angiography (digital subtraction techniques and so forth). This may result in overestimation of the number of normal second angiograms leading to false negative results, difficult to ensure in our case because of the small size of our sample. Because none of the second arteriographies carried out in our study let us identify aneurysms or other vascular malformations, this test might be regarded as superfluous. However, the high mortality associated with rebleeding (up to 4% during the first 24 hours and up to 15%-25% during the first 14 days, depending on the series),26,27 together with the low morbidity of arteriography, tilts the risk–benefit balance toward repeating this test. This is especially important if vasospasm is present because
5
detecting saccular aneurysms is especially difficult in this context.28 An analysis of only those SAH cases that do not meet p-SAH criteria reveals poorer clinical status at baseline, a longer mean hospital stay, and significantly higher rates of hydrocephalus requiring EVD. Seizures and decreased level of consciousness occurred only in the np-SAH group. Regarding the longer hospitalization stay in the p-SAH group, it should be taken into consideration that 4 patients of this group (16.6%) were admitted in the ICU, which could lead to a longer length of stay as admission to a critical care unit results in a longer stay even with the same disease severity. Poorer prognosis in np-SAH patients with regard to complications is consistent with reports in other published series (Table 6). The meta-analysis by Boswell et al31 reports a significantly lower probability of rebleeding in the p-SAH group (odds ratio, 2.78; P 5 .013). Kang et al38 determined a 27.6% rate of radiologic vasospasm in their np-SAH group versus no cases in the p-SAH group (P 5 .006). They also observed a 37.9% rate of hydrocephalus in the imaging study among np-SAH patients compared with 8.7% of the p-SAH group (odds ratio 6.42, P 5 .026). Regarding the latter complication, however, 4 articles did not report any statistically significant differences.5,37,39,42 Ildan et al19 compared seizure frequency between these 2 groups and found a seizure rate of 3.6% in p-SAH patients versus 6.3% in np-SAH patients (ns). The same study reports statistically significant higher rates of permanent ischemic damage and subsequent death among patients with np-SAH (P 5 .05). Long-term functional prognosis seems to be good in both groups. Nevertheless, we were unable to obtain complete data on the 3-month mRS in a significant number of patients because the medium- and long-term follow-up was not performed in our center. However, any neurosurgical/neurointerventional complications or deaths during the outpatient follow-up if they occurred, would have been notified and treated in our unit. In our study, the only baseline feature to appear more frequently in the np-SAH group is dyslipidemia, which is consistent with results from the Zhong et al series and others. Zhong et al30 also found a higher frequency of hypertension, diabetes, and smoking in the np-SAH subgroup and in aneurysmal SAH. Exposure to these factors could explain the poorer response of the parenchyma and vessels to stress, thereby suggesting a poorer prognosis after bleeding. We have not identified any other baseline clinical data pointing to a poorer outcome, despite the differences in the clinical presentation of np-SAH and p-SAH. The main limitation of this study is because of its retrospective deign and sample size, so conclusions should be taken with caution.
B. CANNETI ET AL.
6
Table 6. Previous idiopathic SAH articles comparing p-SAH and np-SAH
Author
Year
Journal
Patel NB29
2014
Zhong W30 Boswell S31 Khan AA32
2014 2013 2013
Moscovici S33 Lin N34 Canovas D35 Sarabia R1
2013 2012 2012 2010
J Neurol Surg; A Cent Eur Neurosurg Turk; Neurosurg J Stroke Cerebrovasc Dis Clin Neurology and Neurosurgery Neurol Res Neurocrit Care Eur J Neurol. Neurocrit Care
Arauz A36 Jung JY37 Kang DH38 Hui FK2 Andaluz N39 Topcuoglu MA28 Lagares A40 Ildan F19 Ruigrok YM41 Canhao P42 Van Calenbergh F6 Rinkel GJ43
2007 2006 2009 2009 2008 2003 2002 2002 2002 1995 1993 1991
Neurologıa J Clin Neurosci J Clin Neurosci Neurocrit Care Neurosurgery J Neurosurg. Neurocirugia (Astur) Neurosurgery Cerebrovasc Dis Acta Neurochir (Wien) Surg Neurol Lancet
Type of analysis
Time, y
Retrospective
N (p-SAH/ np-SAH)
Better prognosis p-SAH
88
Yes
Retrospective Retrospective Retrospective
3 8 5
83 31 50
Yes Yes Yes
Retrospective Retrospective Retrospective Retrospective, multicenter Retrospective Retrospective Retrospective Retrospective Retrospective Retrospective Retrospective Retrospective Retrospective Retrospective Retrospective Retrospective
6 5 13 3
56 66 108 220
Yes Yes Yes Yes
5 18
50 143 52 94 92 86 97 84 88 71 62 113
No Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
6 5 6.5 10 5.4 7 7
Abbreviations: np-SAH, nonperimesencephalic angiographically negative SAH; p-SAH, perimesencephalic SAH; SAH, subarachnoid hemorrhage.
Although p-SAH and np-SAH share a prognosis that is good overall, we must pay particular attention to the distribution of blood on the initial CT in nonaneurysmal SAH cases because the pattern appears to be associated with an increased risk of complications in np-SAH.
Conclusions SAH managed in a Stroke Unit with initial normal arteriography findings has a generally good prognosis compared with that of aneurysmal SAH. However, one type of nonaneurysmal SAH, np-SAH, may be less benign than p-SAH, and it presents a higher risk of complications in the acute phase. Extreme caution is therefore needed with these cases, and we recommend applying a diagnostic and therapeutic approach similar to that used in aneurysmal SAH. Acknowledgments: To the Spanish Neurological Society (SEN) for the language support.
References 1. Sarabia R, Lagares A, Fern andez Alen JA, et al. Idiopathic subarachnoid hemorrhage: a multicentre series of 220 patients. Neurocirugia (Astur) 2010;21:441-451.
2. Hui F, Schuette A, Moskowitz S, et al. Antithrombotic states and outcomes in patients with angiographically negative subarachnoid hemorrhage. Neurosurgery 2011; 68:125-130. discussion 30. 3. Whiting J, Reavey Cantwell J, Velat G, et al. Clinical course of nontraumatic, nonaneurysmal subarachnoid hemorrhage: a single-institution experience. Neurosurg Focus 2009;26:E21. 4. van Gijn J, van Dongen KJ, Vermeulen M, et al. Perimesencephalic hemorrhage: a nonaneurysmal and benign form of subarachnoid hemorrhage. Neurology 1985; 35:493-497. 5. Rinkel GJ, Wijdicks EF, Vermeulen M, et al. Nonaneurysmal perimesencephalic subarachnoid hemorrhage: CT and MR patterns that differ from aneurysmal rupture. AJNR Am J Neuroradiol 1991;12:829-834. 6. Calenbergh FV, Plets C, Goffin J, et al. Nonaneurysmal subarachnoid hemorrhage: prevalence of perimesencephalic hemorrhage in a consecutive series. Surg Neurol 1993;39:320-323. 7. Pinto AN, Ferro JM, Canh~ ao P, et al. How often is a perimesencephalic subarachnoid haemorrhage CT pattern caused by ruptured aneurysms? Acta Neurochir (Wien) 1993;124:79-81. 8. Alexander MS, Dias PS, Uttley D. Spontaneous subarachnoid hemorrhage and negative cerebral panangiography. Review of 140 cases. J Neurosurg 1986;64:537-542. 9. Cioffi F, Pasqualin A, Cavazzani P, et al. Subarachnoid haemorrhage of unknown origin: clinical and tomographical aspects. Acta Neurochir (Wien) 1989;97:31-39.
ANGIOGRAPHICALLY NEGATIVE SAH IN A STROKE UNIT 10. Giombini S, Bruzzone MG, Pluchino F. Subarachnoid hemorrhage of unexplained cause. Neurosurgery 1988; 22:313-316. 11. Eskesen V, Sørensen EB, Rosenørn J, et al. The prognosis in subarachnoid hemorrhage of unknown etiology. J Neurosurg 1984;61:1029-1031. 12. Barlow P. Incidence of delayed cerebral ischaemia following subarachnoid haemorrhage of unknown cause. J Neurol Neurosurg Psychiatry 1985;48:132-136. 13. Nishioka H, Torner JC, Graf CJ, et al. Cooperative study of intracranial aneurysms and subarachnoid hemorrhage: a long-term prognostic study. III. Subarachnoid hemorrhage of undetermined etiology. Arch Neurol 1984;41:1147-1151. 14. Brismar J, Sundb€ arg G. Subarachnoid hemorrhage of unknown origin: prognosis and prognostic factors. J Neurosurg 1985;63:349-354. 15. Spallone A, Ferrante L, Palatinsky E, et al. Subarachnoid haemorrhage of unknown origin. Acta Neurochir (Wien) 1986;80:12-17. 16. Kaim A, Proske M, Kirsch E, et al. Value of repeatangiography in cases of unexplained subarachnoid hemorrhage (SAH). Acta Neurol Scand 1996;93:366-373. 17. Rinkel GJ, van Gijn J, Wijdicks EF. Subarachnoid hemorrhage without detectable aneurysm. A review of the causes. Stroke 1993;24:1403-1409. 18. Greebe P, Rinkel GJ. Life expectancy after perimesencephalic subarachnoid hemorrhage. Stroke 2007;38: 1222-1224. 19. Ildan F, Tuna M, Erman T, et al. Prognosis and prognostic factors for unexplained subarachnoid hemorrhage: review of 84 cases. Neurosurgery 2002;50:1015-1024. discussion 24–5. 20. Forster DM, Steiner L, Hakanson S, et al. The value of repeat pan-angiography in cases of unexplained subarachnoid hemorrhage. J Neurosurg 1978;48:712-716. 21. Alen JF, Lagares A, Campollo J, et al. Idiopathic subarachnoid hemorrhage and venous drainage: are they related? Neurosurgery 2008;63:1106-1111. discussion 11–2. 22. Yuji M, Kiyoyuki Y, Ai M, et al. Significance of a small bulge on the basilar artery in patients with perimesencephalic nonaneurysmal subarachnoid hemorrhage. Report of two cases. J Neurosurg 2003;98:426-429. 23. Tatter SB, Crowell RM, Ogilvy CS. Aneurysmal and microaneurysmal ‘‘angiogram-negative’’ subarachnoid hemorrhage. Neurosurgery 1995;37:48-55. 24. Juul R, Fredriksen TA, Ringkjøb R. Prognosis in subarachnoid hemorrhage of unknown etiology. J Neurosurg 1986; 64:359-362. 25. Pathirana N, Refsum SE, McKinstry CS, et al. The value of repeat cerebral angiography in subarachnoid haemorrhage. Br J Neurosurg 1994;8:141-146. 26. Iwanaga H, Wakai S, Ochiai C, et al. Ruptured cerebral aneurysms missed by initial angiographic study. Neurosurgery 1990;27:45-51. 27. Inamasu J, Nakamura Y, Saito R, et al. ‘‘Occult’’ ruptured cerebral aneurysms revealed by repeat angiography: result from a large retrospective study. Clin Neurol Neurosurg 2003;106:33-37. 28. Topcuoglu MA, Ogilvy CS, Carter BS, et al. Subarachnoid hemorrhage without evident cause on initial angiog-
7
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
raphy studies: diagnostic yield of subsequent angiography and other neuroimaging tests. J Neurosurg 2003; 98:1235-1240. Patel NB, Patel AD, Wilkinson J, et al. An analysis of 88 patients with diffuse and ‘‘benign’’ perimesencephalic subarachnoid hemorrhage. J Neurol Surg A Cent Eur Neurosurg 2014;75:299-304. Zhong W, Zhao P, Wang D, et al. Different clinical characteristics between perimesencephalic subarachnoid hemorrhage and diffuse subarachnoid hemorrhage with negative initial angiography. Turk Neurosurg 2014; 24:327-332. Boswell S, Thorell W, Gogela S, et al. Angiogram-negative subarachnoid hemorrhage: outcomes data and review of the literature. J Stroke Cerebrovasc Dis 2013;22:750-757. Khan AA, Smith JD, Kirkman MA, et al. Angiogram negative subarachnoid haemorrhage: outcomes and the role of repeat angiography. Clin Neurol Neurosurg 2013;115:1470-1475. Moscovici S, Fraifeld S, Ramirez-de-Noriega F, et al. Clinical relevance of negative initial angiogram in spontaneous subarachnoid hemorrhage. Neurol Res 2013; 35:117-122. Lin N, Zenonos G, Kim AH, et al. Angiogram-negative subarachnoid hemorrhage: relationship between bleeding pattern and clinical outcome. Neurocrit Care 2012; 16:389-398. Canovas D, Gil A, Jato M, et al. Clinical outcome of spontaneous non- aneurysmal subarachnoid hemorrhage in 108 patients. Eur J Neurol 2012;19:457-461. Arauz A, Lopez M, Cantu C, et al. [Nonaneurysmal subarachnoid hemorrhage. Study of long-term follow-up]. Neurologia 2007;22:502-506. Jung JY, Kim YB, Lee JW, et al. Spontaneous subarachnoid haemorrhage with negative initial angiography: a review of 143 cases. J Clin Neurosci 2006;13:1011-1017. Kang DH, Park J, Lee SH, et al. Does nonperimesencephalic type non-aneurysmal subarachnoid hemorrhage have a benign prognosis? J Clin Neurosci 2009;16:904-908. Andaluz N, Zuccarello M. Yield of further diagnostic work-up of cryptogenic subarachnoid hemorrhage based on bleeding patterns on computed tomographic scans. Neurosurgery 2008;62:1040-1046. discussion 7. Lagares A, Gomez PA, Lobato RD, et al. [Idiopathic subarachnoid hemorrhage; comparison of different bleeding patterns and long-term outcome]. Neurocirugia 2002; 13:110-119. Ruigrok YM, Rinkel GJ, Van Gijn J. CT patterns and long-term outcome in patients with an aneurysmal type of subarachnoid hemorrhage and repeatedly negative angiograms. Cerebrovasc Dis 2002;14:221-227. Canh~ ao P, Ferro JM, Pinto AN, et al. Perimesencephalic and nonperimesencephalic subarachnoid haemorrhages with negative angiograms. Acta Neurochir (Wien) 1995; 132:14-19. Rinkel GJ, Wijdicks EF, Hasan D, Kienstra GE, Franke CL, Hageman LM, et al. Outcome in patients with subarachnoid haemorrhage and negative angiography according to pattern of haemorrhage on computed tomography. Lancet 1991;338:964-968.