10 Natural History and Management Options of Angionegative Subarachnoid Hemorrhage
Jorn Van Der Veken, Aye Aye Gyi, and Amal Abou-Hamden
Abstract
Keywords: angionegative subarachnoid hemorrhage perimesencephalic bleed diffuse subarachnoid hemorrhage nontraumatic subarachnoid hemorrhage
10.1 Introduction
In 15 to 20% of patients presenting with a nontraumatic subarachnoid hemorrhage (SAH), a cause for the hemorrhage cannot be detected on the initial vascular imaging. This group is referred to as the angionegative SAH group. There is no consensus on the timing, number, and type of investigations needed to define an angionegative SAH. Digital subtraction angiography (DSA) has been the imaging modality of choice, but with the advent of high-quality 3D computed tomography angiography (CTA) and magnetic resonance angiography (MRA), routine use of this more conventional and invasive investigation has been challenged. Based on radiological findings, angionegative SAHs are further classified into the following:
●Perimesencephalic SAH (PMSAH): It was initially described by van Gijn et al in 19851 and has a characteristic radiographic pattern of hemorrhage restricted to the perimesencephalic cisterns on CTB.1, 2
●Non-PMSAH or diffuse SAH (dSAH): This is characterized by a more widespread distribution of SAH with a visible extension into the anterior part of the ambient cistern or to the basal part of the Sylvian fissures, mimicking an aneurysmal SAH.1, 2
This classification refers to the initial computed tomography of the brain (CTB) ideally performed within 24 hours after the ictus, as delayed imaging might mimic a PMSAH due to blood resorption.1, 2 Although often considered a single pathological entity, distinguishing non-PMSAH from PMSAH is particularly important as they have a different natural course and management.
In a recent systematic review and meta-analysis, demographic characteristics of patients (n = 3,853) were described in 49 of 58 included studies.3 The mean age of included patients was 53.8 years. Forty-five studies including 3,530 patients reported diagnoses stratified by bleeding pattern and found that 1,763/3,530 (49.9%) had a PMSAH and 1,577/3,530 (44.7%) had a non-PMSAH.3 Unlike aneurysmal SAH, there is a male predominance in angionegative SAH (54.3%). A summary of demographic data from the systematic review is presented in Table 10.1.3
Table 10.1 General demographics of the study patients adapted from Mohan et al3
Studies | Type of study | No. of patients | Mean age (y) | Male (%) |
Rinkel et al4 | Retro | 65 | 53.0 | 61.5% |
Goergen et al5 | Retro | 18 | NR | NR |
Van Calenbergh et al6 | Retro | 62 | 47.0 | 40.0% |
Hütter et al7 | Retro | 20 | 49.0 | NR |
Canhão et al8 | Pros and retro | 71 | 49.9 | 56.0% |
Tatter et al9 | Retro | 40 | NR | NR |
Duong et al10 | Pros | 92 | 49.0 | 54.0% |
Berdoz et al11 | Retro | 52 | 52.8 | 65.0% |
Linn et al12 | Pros | 23 | 56.0 | 74.0% |
Madureira et al13 | Pros | 18 | NR | NR |
Marquardt et al14 | Pros | 21 | 55.0 | 52.0% |
Franz et al15 | Retro | 34 | 50.1 | 52.0% |
Ildan et al16 | Retro | 84 | 49.5 | NR |
Alén et al17 | Pros | 44 | 51.9 | 63.6% |
Lang et al18 | Pros | 57 | 54.7 | 56.0% |
Topcuoglu et al19 | Retro | 86 | 54.3 | 63.0% |
Caeiro et al20 | Pros | 33 | NR | NR |
Jung et al21 | Pros | 143 | 52.3 | 42.0% |
Matsuyama et al22 | Pros | 9 | 50.0 | 57.0% |
Andaluz and Zuccarello23 | Retro | 92 | 49.4 | 34.0% |
Kang et al24 | Retro | 52 | 55.4 | 53.8% |
Whiting et al25 | Retro | 89 | 56.0 | 51.0% |
Beseoglu et al26 | Retro | 21 | 57.2 | 60.0% |
Caeiro et al27 | Pros | 37 | NR | 43.0% |
Nayak et al28 | Retro | 190 | 57.0 | 61.6% |
Alfieri et al29 | Pros | 38 | 44.3 | NR |
Fontanella et al30 | Retro | 102 | 53.0 | 62.7% |
Kong et al31 | Pros | 31 | 49.7 | NR |
Oda et al32 | Retro | 15 | NR | NR |
Pyysalo et al33 | Retro | 97 | 52.0 | 36.0% |
Cánovas et al34 | Retro | 108 | 52.4 | 44.9% |
Delgado Almandoz et al35 | Retro | 72 | 53.1 | 36.1% |
Gross et al36 | Retro | 77 | 59.8 | 48.0% |
Kostić et al37 | Pros | 36 | 48.3 | 55.6% |
Lin et al38 | Pros | 68 | 59.5 | 51.5% |
Maslehaty et al39 | Pros and retro | 179 | NR | 60.0% |
Yu et al40 | Retro | 28 | 60.3 | 51.1% |
Zhong et al41 | Retro | 49 | 54.0 | 49.0% |
Boswell et al42 | Retro | 31 | 56.3 | 55.0% |
Dalyai et al43 | Retro | 254 | NR | NR |
Khan et al44 | Retro | 50 | 52.5 | 60.0% |
Muehlschlegel et al45 | Retro | 93 | 54.0 | 59.0% |
Prat et al46 | Retro | 63 | 52.4 | 54.0% |
Tsermoulas et al47 | Retro | 62 | 51.0 | 51.2% |
Woodfield et al48 | Retro | 240 | 51.0 | 62.0% |
Ellis et al49 | Pros | 173 | 55.0 | 55.0% |
Konczalla et al50 | Pros | 125 | 56.0 | 70.0% |
Kumar et al51 | Pros | 39 | 50.5 | 69.0% |
Mensing et al52 | Pros | 79 | 53.0 | 54.0% |
Qureshi et al53 | Retro | 5 | 59.8 | NR |
Canneti et al54 | Retro | 41 | 55.0 | 51.0% |
Dalbjerg et al55 | Retro | 95 | 53.0 | 48.4% |
Konczalla et al56 | Retro | 152 | 58.0 | 59.0% |
Konczalla et al57 | Pros | 173 | 56.0 | NR |
Sprenker et al58 | Retro | 26 | NR | NR |
Walcott et al59 | Retro | 138 | 55.6 | 53.6% |
Konczalla et al60 | Retro | 225 | 57.0 | NR |
Moscovici et al61 | Retro | 56 | 53.4 | 54.0% |
Abbreviations: NR, not reported; Pros, prospective; Retro, retrospective; WFNS, World Federation of Neurosurgical Societies. |
A pooled analysis of data from 9 studies with 646 patients showed that a significantly greater number of patients in the non-PMSAH group had a poor-grade SAH (World Federation of Neurosurgical Societies [WFNS] grade > 2) than PMSAH group (24.6 vs. 7.2%; Table 10.2).3
Table 10.2 Proportion of study participants with poor grade SAH (WFNS grade greater than 2)3
Studies | Total patients | Poor WFNS grade, PMSAH | Poor WFNS grade, non-PMSAH |
Van Calenbergh et al6 | 44 | 3/20 | 9/24 |
Berdoz et al11 | 52 | 0/22 | 9/30 |
Beseoglu et al26 | 26 | 1/17 | 4/9 |
Kostić et al37 | 34 | 1/18 | 4/16 |
Khan et al44 | 40 | 1/17 | 3/23 |
Moscovici et al61 | 56 | 0/25 | 22/31 |
Konczalla et al50 | 125 | 7/73 | 12/52 |
Konczalla et al57 | 173 | 8/87 | 15/86 |
Sprenker et al58 | 96 | 1/25 | 6/71 |
Total | 646 | 22/304 (7.2%) | 84/342 (24.6%) |
Abbreviations: PMSAH, perimesencephalic subarachnoid hemorrhage; WFNS, World Federation of Neurosurgical Societies. |
10.2 Selected Papers on the Natural History of Angionegative SAH
●Hui FK, Tumialán LM, Tanaka T, Cawley CM, Zhang YJ. Clinical differences between angiographically negative, diffuse subarachnoid hemorrhage and perimesencephalic subarachnoid hemorrhage. Neurocrit Care 2009;11(1):64–70
10.3 Natural History
The natural history of angionegative SAH is generally more favorable compared to their angiopositive counterparts. However, complications are well recognized and these include vasospasm, hydrocephalus, permanent ischemic complications, chronic headaches, and rehemorrhage.3, 62 A pooled analysis of published data (Table 10.3) suggests that the incidence of rebleeding in the non-PMSAH group was approximately 3.5% compared to 1.1% in PMSAH group. Development of hydrocephalus was found to be considerably worse in the non-PMSAH group (25.5 vs. 9.0%) with requirement for external ventricular drainage (41 vs. 9.6%), and eventual permanent cerebrospinal fluid diversion (20.6 vs. 0%).3 Furthermore, non-PMSAH patients had higher rates of radiological vasospasm (14.0 vs. 8.3%) and seizures (5.2 vs. 2.5%) when compared to those with PMSAH patients.3 Another publication reported a higher rate of hydrocephalus and radiological vasospasm for the non-PMSAH group.62 A forest plot meta-analysis of available complications data from both studies3, 62 showed a higher risk of overall hydrocephalus (odds ratio [OR], 2.69; 95% confidence interval [CI], 2.10–3.45) and radiological vasospasm (OR, 2.57; 95% CI, 1.82–3.62) for the non-PMSAH group (Fig. 10.1).
Table 10.3 Pooled analysis of published data on the complications stratified by bleeding pattern
Complications | Studies | Events/no. of patients (%) | PMSAH (%) | non-PMSAH (%) |
Rebleed | Hui et al62 | NR | NR | NR |
Mohan et al3 | 56/1,675 (3.3%) | 4/506 (1.1%) | 16/530 (3.5%) | |
Hydrocephalus | Hui et al62 | 36/94 (38.3%) | 3/36 (9.6%) | 33/36 (50.8%) |
Mohan et al3 | 469/2,399 (19.5%) | 104/915 (9.0%) | 241/924 (25.5%) | |
Radiological vasospasm | Hui et al62 | 22/94 (23.4%) | 3/22 (9.6%) | 19/22 (28.6%) |
Mohan et al3 | 180/1,435 (12.5%) | 49/530 (8.3%) | 116/630 (14.0%) | |
Seizures | Hui et al62 | NR | NR | NR |
Mohan et al3 | 12/379 (3.2%) | 3 /164 (2.5%) | 8 /179 (5.2%) | |
Abbreviations: NR, not reported; PMSAH, perimesencephalic subarachnoid hemorrhage. |
A pooled analysis of functional outcome data stratified by bleeding patterns reported in the systematic review suggested that the non-PMSAH patients were more likely to experience poor functional outcomes (modified Rankin Scale [mRS] ≥ 3) than those with PSAH at 3 to 6 months (12.8 vs. 6.1%) and ≥ 1 year (14.4 vs. 7.3%).3 Ultimately, only 76% of the non-PMSAH patients achieved complete recovery and independent living, compared to 96.7% of the PMSAH patients.3
The length of hospital stay was reported in 9 studies (608 patients) with median and mean values of 10 and 8.3 days in PMSAH patients versus 8.3 and 4 days in non-PMSAH patients (Table 10.4).3
Table 10.4 Length of stay (days) for patients with angionegative SAH stratified by SAH types
Study | Total | Subgroups | |||||||
n | LOS | PMSAH (n) | LOS | Non-PMSAH (n) | LOS | Radiologically negative (n) | LOS | ||
Goergen et al5 | 18 | NR | 9 | 12a | NR | NA | NR | NA | |
Andaluz and Zuccarello23 | 92 | 6.3b | 45 | 4.3b | 47 | 8.3b | 0 | NA | |
Beseoglu et al26 | 21 | 15.3b | 12 | 11.2b | 9 | 20.7b | 0 | NA | |
Nayak et al28 | 190 | 14 | NR | NA | NR | NA | NR | NA | |
Delgado Almandoz et al35 | 25 | 2–7a | 15 | 2–7a | 8 | 2–7a | 2 | 2–7 | |
34 | 8–14a | 12 | 8–14a | 20 | 8–14a | 2 | 8–14a | ||
13 | ≥ 15a | 2 | ≥ 15a | 11 | ≥ 15a | 0 | ≥ 15 | ||
Boswell et al42 | 31 | NR | 14 | 12.5b | 16 | 13.2b | 1 | NR | |
Khan et al44 | 50 | NR | 17 | 8.3 | 23 | 10 | 10 | 5 | |
Muehlschlegel et al45 | 93 | 11 | 36 | NR | 48 | NR | 9 | NR | |
Canneti et al54 | 41 | 21.1b | 17 | 17b | 24 | 24b | 0 | ||
Abbreviations: LOS, length of stay; NA, not applicable; NR, not reported; PMSAH, perimesencephalic subarachnoid hemorrhage. Source: adapted from Mohan et al.3 aDischarge within days of admission. bMean/median. |
Fig. 10.1 Forest plots of complications data in perimesencephalic subarachnoid hemorrhage (PMSAH) versus non-PMSAH comparing (a) hydrocephalus and (b) radiological vasospasm.

Stay updated, free articles. Join our Telegram channel

Full access? Get Clinical Tree

