Perimesencephalic nonaneurysmal subarachnoid hemorrhage

INTRODUCTION AND DEFINITION — Subarachnoid hemorrhage (SAH) refers to bleeding within the subarachnoid space, which lies between the arachnoid and pia mater and is normally filled with cerebrospinal fluid. Most spontaneous (nontraumatic) cases of SAH are caused by rupture of an intracranial aneurysm, but approximately 20 percent of patients do not have an established vascular lesion on initial four-vessel cerebral angiography [1,2]. The causes of nonaneurysmal SAH (NASAH) are potentially diverse, and often the mechanism of bleeding in these cases is not identified.

In 1985, a subtype of NASAH, so-called perimesencephalic NASAH (PM-NASAH), was identified in a case series of 13 patients who had a characteristic pattern of localized blood on computed tomography (CT), normal cerebral angiography, and a benign course that distinguished these patients from both aneurysmal SAH and other patients with NASAH [3]. This observation has subsequently been confirmed by other reports [2,4-7]. It is important to recognize the perimesencephalic pattern of bleeding, because only a small percent will be due to aneurysm rupture; thus, invasive diagnostic testing can be limited in these patients and morbidity thereby reduced.

This topic will review the pathogenesis, clinical features, diagnosis, and management of PM-NASAH. Other causes of NASAH are reviewed separately. (See "Nonaneurysmal subarachnoid hemorrhage".)

Aneurysmal SAH is reviewed elsewhere.

●(See "Aneurysmal subarachnoid hemorrhage: Epidemiology, risk factors, and pathogenesis".)

●(See "Aneurysmal subarachnoid hemorrhage: Clinical manifestations and diagnosis".)

●(See "Aneurysmal subarachnoid hemorrhage: Treatment and prognosis".)

EPIDEMIOLOGY — The reported proportion of cases of NASAH that are perimesencephalic NASAH varies between 21 and 68 percent [1,2,6,8-11]. In a 2018 systematic review, the annual rate of PM-NASAH was estimated to be 0.5 per 100,000 persons over 18 years of age [12]. The pooled mean age of onset was 53 years, with reported cases ranging in age from 3 to 90 years.

In contrast with aneurysmal SAH, there is no clear female predisposition for PM-NASAH; in the systematic review, females accounted for 42 percent of PM-NASAH cases [12].

Case-control studies suggest that hypertension and cigarette smoking are risk factors for PM-NASAH, but these appear to be somewhat less prevalent among patients with PM-NASAH compared with patients with aneurysmal SAH [8,13-15]. In the 2018 systematic review, the pooled odds ratios (compared with controls) for hypertension (1.6) and smoking (1.1) did not reach statistical significance, but there was substantial heterogeneity in the analyses [12].

PM-NASAH is not known to have a genetic predisposition; however, rare cases in first-degree relatives have been described [16].

ETIOLOGY AND PATHOGENESIS

Aneurysmal PM-SAH — A ruptured saccular aneurysm is the cause of SAH in patients with a perimesencephalic pattern of bleeding in only 2 to 9 percent of patients [1,3,5,17-20]. The aneurysm in these cases arises from the posterior circulation: the basilar tip, the vertebrobasilar junction, or the posterior inferior cerebellar, superior cerebellar, or posterior cerebral artery [5,21-23]. Posterior circulation aneurysms are less common than those arising from the anterior circulation. When posterior circulation aneurysms rupture, a perimesencephalic pattern of SAH results less than 17 percent of the time [19,20,22].

Nonaneurysmal PM-SAH — In the majority of cases, PM-SAH is not due to aneurysm rupture (ie, PM-NASAH), and the etiology is not defined, even after extensive evaluation. Theories regarding its origin in some or all cases include [24]:

●Rupture of a perforating artery – The perimesencephalic location of the blood clots is consistent with bleeding from a perforating artery arising from the posterior circulation [25]. In one case, this pathogenesis has been substantiated pathologically [26]. In addition, hypertension, a known risk factor for perforating artery disease, is also a risk factor for PM-NASAH. (See 'Epidemiology' above.)

In one case series of 25 patients with PM-NASAH, an acute lacunar infarction (usually attributed to occlusive disease of the perforating arteries) was found in four patients [27]. However, the acute lacunar infarctions in this and other reports have been in the distribution of the anterior circulation, and their relationship to the PM-NASAH is not clear [1,28].

●Venous origin – A venous source for PM-NASAH is suggested by the limited extension of blood and low rate of subsequent rebleeding, both suggesting a low-pressure bleeding source. PM-NASAH often occurs in the setting of physical exertion, which in this paradigm sequentially produces increased intrathoracic pressure, impaired internal jugular venous return, elevated intracranial venous pressure, and leakage of venous blood from susceptible blood vessels [29]. Physical exertion is also believed to increase rupture from arterial pathology, including saccular aneurysms.

In multiple case series, a higher incidence of variant forms of venous drainage on conventional or computed tomography (CT) venography was found in patients with PM-NASAH when compared with aneurysmal SAH [30-32]. In these patients, the basal vein of Rosenthal and/or the perimesencephalic veins drain directly into the dural sinuses rather than the vein of Galen, potentially making them more susceptible to venous congestion. Other reported venous anomalies in patients with PM-NASAH include vein of Galen stenosis, venous infarction, and cerebral venous thrombosis [33-37].

●Basilar artery wall hematoma – In some cases of PM-NASAH, an abnormal contour of the basilar artery has been observed, either a small bulge or luminal narrowing [24,38]. This was interpreted as possible vasospasm in some reports, but others have speculated that an intramural hematoma could account for the abnormal blood vessel appearance and also be the source of bleeding. It is postulated that rupture of the vasa vasorum is the source of the relatively limited bleeding that occurs in PM-NASAH. This is in contrast with an arterial dissection that results from an intimal tear and produces massive SAH. (See "Nonaneurysmal subarachnoid hemorrhage", section on 'Intracranial arterial dissection'.)

The role of a putative basilar artery wall hematoma in PM-NASAH remains unproven. In one case series, systematic examination of basilar artery contours in 27 patients with PM-NASAH found no variations that suggested this origin of bleeding [9]. This hypothesis awaits further study with high-resolution neuroimaging studies.

●Other causes – Other potential causes of PM-NASAH include rupture of a cryptic vascular malformation, high cervical spinal dural arteriovenous fistula, cavernous angioma, or capillary telangiectasia [39-41]. Some patients with PM-NASAH may have an occult aneurysm; however, the low incidence of rebleeding in these patients suggests that this is rare. (See "Nonaneurysmal subarachnoid hemorrhage".)

CLINICAL PRESENTATION — The clinical manifestations of PM-NASAH are generally less severe than those of aneurysmal SAH, but the presentation is otherwise similar, with sudden onset of headache, nausea, and vomiting being the most common symptoms [12]. Meningismus and photophobia are also reported [1,7]. Headache onset is instantaneous in approximately 85 percent, and more gradual (beginning in minutes rather than seconds) in the remainder [12]. Transient focal deficits, including as sensory symptoms, weakness, gait impairment, and speech arrest, are reported by approximately 10 percent of patients.

Individual patients with PM-NASAH and other forms of SAH cannot be distinguished based upon their clinical symptoms [20]. However, group comparisons of patients with PM-NASAH and aneurysmal SAH find differences in the prevalence of certain clinical features and measures of disease severity. Some examples of these differences, as well as similarities, are:

●Severity at presentation – On initial evaluation in the hospital, over 90 percent of patients with PM-NASAH are categorized in lower Hunt and Hess grades, a measure of the clinical severity of the SAH (table 1) [1,6,7,42-44]. By contrast, patients with aneurysmal SAH present are likelier to present with more severe symptoms [6,43].

●Frequency of cognitive impairment – Loss of consciousness at the onset of SAH is reported less frequently in patients with PM-NASAH compared with those with aneurysmal SAH [12]. Amnesia at the onset of bleeding has been reported in up to one-third of patients, possibly due to transient hydrocephalus or a seizure [45].

●Headache – Sentinel headache, a prodromal feature of up to 40 percent of patients with aneurysmal SAH, is not reported in PM-NASAH [1]. However, patients with PM-NASAH may report severe neck pain prior to acute headache. Headache onset in PM-NASAH may be rapid but typically not instantaneous like in aneurysmal SAH.

●Exertional triggers – The clinical setting of physical exertion appears to be as common in PM-NASAH as in aneurysmal SAH [3,7,24,29]. The onset of both PM-NASAH and aneurysmal SAH are more common during daytime and evening hours compared with nighttime [46]. One analysis of the circadian fluctuation found an increased risk of PM-NASAH during the day and late evening hours compared with the hours between 12:00 and 6:00 AM [47].

DIAGNOSTIC EVALUATION — SAH should be considered in any patient complaining of severe headache of sudden onset. Urgent CT of the head should immediately follow consideration of the diagnosis. If the suspicion is high and the CT scan fails to show blood in subarachnoid space, a lumbar puncture must be obtained. (See "Aneurysmal subarachnoid hemorrhage: Clinical manifestations and diagnosis", section on 'Evaluation and diagnosis'.)

The diagnosis of PM-NASAH is made in a patient who meets specific computed tomography (CT) criteria described below and in whom at least one angiographic study has failed to demonstrate an aneurysm. (See 'Radiologic criteria' below and 'Initial cerebral angiography' below.)

Radiologic criteria — The pattern of blood on CT scan identifies a patient with PM-SAH and a low likelihood of an aneurysm (image 1 and figure 1) [1,5,48-50]. Specific radiographic criteria include the following (all must be true) [12,50]:

●The diagnostic head CT scan without contrast scan is done within 72 hours after the ictus.

●The center of the hemorrhage is located immediately anterior to and in contact with brainstem in prepontine, interpeduncular, or suprasellar cistern.

●Blood is limited to the prepontine, interpeduncular, suprasellar, crural, ambient, and/or quadrigeminal cistern and/or cisterna magna.

●No extension of blood into sylvian or interhemispheric fissures.

●Interventricular blood is limited to incomplete filling of the fourth ventricle and occipital horns of the lateral ventricle.

●No intraparenchymal blood.

In some patients, the visualized blood is limited to the quadrigeminal or pretruncal cisterns [48,49,51]. There is high interobserver agreement among radiologists in the identification of PM-SAH (kappa 0.87 to 0.96) [5,50].

The timing of the CT scan is critical. A diagnosis of PM-NASAH should not be considered reliable on a CT performed more than 48 to 72 hours after the clinical ictus, as washout and resorption of blood may have taken place and obscured the initial extent of hemorrhage [1,7,52].

Occasionally, an aneurysmal source of bleeding is found to be the cause of a PM-SAH that fulfills all of the above criteria, but an aneurysm is more likely when one or more of the criteria are not fulfilled (image 2). Thus, angiographic evaluation is essential for all patients with PM-SAH. (See 'Initial cerebral angiography' below.)

Evaluation — Imaging evaluation is essential for all patients with PM-SAH to exclude an aneurysmal source of bleeding (algorithm 1). (See 'Initial cerebral angiography' below.)

Cerebral venous thrombosis can rarely present with PM-NASAH as its primary manifestation [34-37]. The thrombosis may be visualized on venous phase of digital subtraction angiography and/or on magnetic resonance imaging venography (MRV) or computed tomography venography (CTV). It is important to exclude the possibility of acute cerebral venous sinus thrombosis when PM-SAH is associated with venous hemorrhagic infarction, and it is reasonable to evaluate for cerebral venous thrombosis if there is a history of prior venous thrombosis or a clinical susceptibility for hypercoagulability. (See "Cerebral venous thrombosis: Etiology, clinical features, and diagnosis".)

If discovered, the treatment paradigm for cerebral venous thrombosis would differ significantly than the typical course of management for PM-NASAH. (See "Cerebral venous thrombosis: Treatment and prognosis".)

Initial cerebral angiography — The likelihood of finding an aneurysm on angiogram in a patient with a PM-SAH has varied from 2 to 9 percent in different reports. (See 'Etiology and pathogenesis' above.)

Because of the very high and immediate morbidity and mortality associated with aneurysmal bleeding, every patient with PM-SAH must undergo angiographic evaluation. The commonly used angiographic techniques include computed tomography angiography (CTA), magnetic resonance angiography (MRA), and conventional digital subtraction angiography (DSA). The best modality for evaluation of cerebral vasculature in PM-SAH is controversial.

Preference for DSA — Digital subtraction angiography (DSA) has the highest resolution for the detection of intracranial aneurysms and remains the gold-standard imaging test. Three-dimensional rotational techniques can further improve the diagnostic yield of DSA by identifying very small aneurysms that may be difficult to visualize with standard two-dimensional angiography [53]. While invasive, DSA has a relatively low morbidity in this setting. In one meta-analysis, the combined risk of permanent and transient neurologic complications following DSA was significantly lower in patients with SAH compared with those with a transient ischemic attack (TIA) or stroke (1.8 versus 3.7 percent) [54].

Multidetector CTA as an alternative — Of the two noninvasive angiographic techniques, computed tomography angiography (CTA) and magnetic resonance angiography (MRA), CTA has been best studied in the setting of acute SAH. Different meta-analyses have documented sensitivities and specificities of CTA for aneurysm detection to be 83 to 97 and 88 to 97 percent, respectively [55-59]. There is likely to be appreciable interinstitutional variability, stemming from expertise in interpretation and also in the specific technology used. A 2011 meta-analysis of CTA diagnosis of intracranial aneurysms found that, compared with single-detector CTA, use of multidetector CTA was associated with an overall improved sensitivity and specificity for aneurysm detection (both >97 percent) as well as improved detection of smaller aneurysms ≤4 mm in diameter [57]. Another systematic review and meta-analysis restricted to patients with SAH had similar findings [58]. Multidetector CTA also allows for superior depiction of aneurysm morphology [60].

A decision analysis reported that CTA alone had higher utilities in the setting of PM-SAH compared with other diagnostic strategies including no investigation, DSA alone, and CTA plus DSA [17]. However, certain assumptions in the analysis, including a low prevalence of aneurysm (4 percent), a high complication rate of angiography (2.7 percent morbidity and mortality), and a 97 percent sensitivity and specificity of CTA, favored this outcome and can be disputed.

Despite the improvement in CTA technology in accurately identifying small ruptured aneurysms, the specific CTA technology, along with expertise of interpretation, varies at different institutions. As a result, we recommend DSA over CTA as a first test for most patients with PM-SAH. Centers with a high volume of cases of SAH and demonstrated, documented experience with multidetector CTA for this purpose may choose CTA over DSA.

The role of repeat angiography — For patients with PM-NASAH with initial angiography that was technically difficult, complicated by vasospasm, or associated with severe clinical grade, we repeat imaging with DSA with three-dimensional rotational angiography or multidetector CTA one to two weeks following the onset of PM-SAH. In addition, rebleeding suggests the presence of an occult aneurysm and is an indication for repeat DSA. For other patients with PM-NASAH and negative initial angiography, it is reasonable to avoid repeat testing.

When DSA is repeated after an initial first study in patients with SAH, an aneurysm may be revealed in 2 to 24 percent of patients [1,6,10,61]. Reasons for an initial false-negative angiogram include technical or reading errors, small aneurysm size, and obscuration of the aneurysm because of vasospasm, hematoma, or thrombosis within the aneurysm [6,23].

A false-negative angiogram in the setting of PM-SAH is a rare but reported event [6,10,23,42,61-63]. A study of 273 patients with angiogram-negative SAH found that patients with PM-NASAH had a higher rate of variant venous drainage and that the presence of typical venous drainage in such patients was associated with a higher rate of positive findings among those who underwent repeat angiography as well as an increased risk of rebleeding from an arterial source [32]. Given the low yield of repeat angiography, many clinicians do not pursue this unless the initial angiogram is felt to be technically compromised because of vasospasm or other factors [1-4,64-69]. Others disagree and advocate repeating DSA in all patients [52,59]. A conservative, middle-road approach advocates follow-up imaging with noninvasive CTA [6,63].

Magnetic resonance imaging — Magnetic resonance imaging (MRI) has no additional benefit for the detection of a bleeding source compared with CT, CTA, and DSA and is not cost-effective in the evaluation of patients with PM-SAH [12,61,70-72]. However, some centers may continue to perform MRI in selected patients with suspected PM-NASAH if the pattern of bleeding is unusual.

Other testing — Patients with SAH should undergo basic laboratory testing including complete blood count, serum chemistries, and coagulation studies. A baseline electrocardiogram (ECG) should also be obtained.

CLINICAL COURSE AND COMPLICATIONS — Most clinical case series report that patients with PM-NASAH have a lower incidence and severity of complications compared with patients with aneurysmal SAH and patients with non-PM-NASAH [2,11,43,73-75].

Rebleeding — Rebleeding is rare among patients with PM-SAH after initial negative angiography [76]. In a 2018 systematic review, in-hospital rebleeding was reported in only 3 of over 1220 patients with PM-NASAH [12]. In those patients in whom rebleeding occurs, an occult cerebral aneurysm seems likely because the pattern of recurrent hemorrhage is often not perimesencephalic [6,77].

Rebleeding represents an indication for repeat digital subtraction angiography (DSA).

Vasospasm — Vasospasm with cerebral ischemia is a leading cause of death and disability after aneurysm rupture, occurring in 20 to 30 percent of patients with aneurysmal SAH [78]. In PM-NASAH, vasospasm is less common [79] and typically associated with milder clinical symptoms. In a 2018 systematic review of patients with PM-NASAH, angiographic vasospasm was detected in 9 percent (95% CI 6 to 14 percent) [12]. However, there are reported cases of symptomatic, diffuse, and severe vasospasm affecting the anterior and posterior circulations in patients with PM-NASAH [2,80-85]. Symptomatic vasospasm has been precipitated by DSA in some cases [1,7,73]. The lower incidence of vasospasm in PM-NASAH compared with aneurysmal SAH may be due to the lower volume or less oxyhemoglobin from venous source of bleeding in these patients, although this specific association has not been demonstrated.

Transcranial Doppler (TCD) sonography is useful for detecting and monitoring vasospasm in spontaneous SAH [86,87]. (See "Aneurysmal subarachnoid hemorrhage: Treatment and prognosis", section on 'Vasospasm and delayed cerebral ischemia' and "Aneurysmal subarachnoid hemorrhage: Treatment and prognosis", section on 'Monitoring'.)

Hydrocephalus — Hydrocephalus can be an early or delayed complication of SAH. A 2018 systematic review of PM-NASAH found that acute hydrocephalus occurred in 14 percent (95% CI 10 to 18 percent) [12], a rate similar to aneurysmal SAH; symptomatic hydrocephalus, defined as radiographic hydrocephalus with a diminished level of consciousness leading to drainage, was less frequent, occurring in 3 percent (95% CI 2 to 6 percent). One case series found that among their patients with PM-NASAH, filling of all perimesencephalic cisterns was a prerequisite for developing hydrocephalus [88]. In more than half of the cases, hydrocephalus resolved spontaneously without intervention (ventriculostomy and/or shunting), perhaps because of rapid washout of blood limited to the cisterns.

Seizures — Acute seizures have been described in PM-NASAH, but this is uncommon [12]. PM-NASAH does not appear to be associated with the later development of epilepsy.

Hyponatremia — Hyponatremia following SAH is due to increased secretion of antidiuretic hormone. In one series, 10 of 35 patients with PM-NASAH developed hyponatremia, only five of whom had sodium levels less than 130 mmol/L [73]. (See "Cerebral salt wasting".)

Cardiac and ECG abnormalities — Cardiac abnormalities and electrocardiographic (ECG) changes are commonly seen after SAH, ranging in severity from transient ST segment depression to life-threatening cardiac arrhythmias (see "Aneurysmal subarachnoid hemorrhage: Treatment and prognosis", section on 'Cardiopulmonary complications'). In one case series of PM-NASAH, baseline ECG changes were found in 22 percent on admission and transient ECG changes occurred in most [1,73]. No clinically significant cardiac complications have been reported in the setting of PM-NASAH.

TREATMENT

Initial management while aneurysm is being excluded — Patients with PM-SAH should be treated as though there is an underlying aneurysm until this has been satisfactorily excluded. Patients are admitted to an intensive care setting for constant hemodynamic and cardiac monitoring, given stool softeners, kept at bed rest, and given analgesia to diminish hemodynamic fluctuations and lower the risk of rebleeding. Anticoagulation and antiplatelet agents should be discontinued. Pneumatic compression stockings to limit risk of deep vein thrombosis should be utilized while patients are immobile. Patients in whom PM-SAH is discovered in the outpatient setting should be transferred to an inpatient setting for further management. (See "Aneurysmal subarachnoid hemorrhage: Treatment and prognosis".)

Management after exclusion of aneurysm — After exclusion of cerebral aneurysm, patients with PM-NASAH should continue to receive symptomatic care with analgesics as needed, cardiac monitoring, serum chemistry monitoring, and monitoring of neurologic status [1]. Blood pressure should be monitored and controlled, but strict bed rest is not required. The risk of epileptic seizures is low and prophylactic antiseizure medications are not recommended [7].

The calcium channel blocker nimodipine has been shown to improve outcomes in patients with aneurysmal SAH [89] (see "Aneurysmal subarachnoid hemorrhage: Treatment and prognosis", section on 'Nimodipine'). The mechanism of its action is uncertain but may involve some type of neuroprotection against vasospasm-induced cerebral ischemia. It is less certain that patients with other mechanisms of SAH also benefit from nimodipine. There are no data as to whether nimodipine provides any benefit for patients with PM-NASAH, who as a group incur little, if any, serious neurologic mortality and morbidity. At the same time, nimodipine is unlikely to be associated with significant adverse effects in this group. We suggest that patients with PM-SAH be treated and monitored as if there is a cerebral aneurysm with nimodipine and transcranial Doppler (TCD) monitoring until imaging evaluation has satisfactorily excluded aneurysmal SAH as the etiology. (See 'The role of repeat angiography' above.)

In patients who required anticoagulation prior to the onset of PM-NASAH, there are limited data regarding the optimal time of anticoagulation reinitiation. Early rebleeding has been described with restarting anticoagulation in the immediate posthemorrhagic period [90]. We suggest avoiding therapeutic anticoagulation for the first three to five days following onset of PM-NASAH when possible.

While complications following PM-NASAH are less likely than after aneurysmal SAH, they should be managed similarly. (See "Aneurysmal subarachnoid hemorrhage: Treatment and prognosis", section on 'Early complications'.)

LONG-TERM PROGNOSIS — Compared with aneurysmal SAH, patients with PM-NASAH have a good outcome [2,6,73]. Long-term follow-up studies reveal that rebleeding is exceptional and life expectancy is not altered [43,91-93]. With rare exception, patients do not have significant neurologic deficits after PM-NASAH. Some uncontrolled studies suggest that many patients report residual headaches, depression, minor cognitive deficits, and fatigue several years after the PM-NASAH, in some cases affecting employment status [4,91,93,94]. Persistent anosmia is an infrequent (5 percent) complication of PM-NASAH [95]. However, one study found that quality of life two years after PM-NASAH was not significantly different from controls [96].

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Stroke in adults".)

SUMMARY AND RECOMMENDATIONS

●Definition – Perimesencephalic nonaneurysmal subarachnoid hemorrhage (PM-NASAH) refers to a subset of patients with subarachnoid hemorrhage (SAH) who have a characteristic pattern of localized blood on computed tomography (CT), normal cerebral angiography, and a benign course. The annual rate of PM-NASAH was estimated to be 0.5 per 100,000 persons over 18 years of age. (See 'Introduction and definition' above and 'Epidemiology' above.)

●Clinical features – The clinical presentation of PM-NASAH overlaps that of aneurysmal SAH with abrupt onset of headache, meningismus, photophobia, nausea, and vomiting. As a group, patients with PM-NASAH have milder clinical features than those with aneurysmal SAH. (See 'Clinical presentation' above.)

●Diagnosis – The diagnosis of PM-NASAH is made in a patient who meets specific computed tomography (CT) criteria in whom at least one angiographic study has failed to demonstrate an aneurysm.

The CT findings that define PM-SAH (image 3) include blood limited to the prepontine, interpeduncular, suprasellar, crural, ambient, and/or quadrigeminal cistern and/or cisterna magna. There may be extension into the basal parts of the sylvian fissures and small amount of spillover in the ventricles but not into the lateral sylvian fissure or anterior interhemispheric fissure. (See 'Radiologic criteria' above.)

●Evaluation to exclude aneurysm – An intracranial aneurysm should be excluded in all patients with PM-SAH (algorithm 1). We suggest using digital subtraction angiography (DSA). In centers that have a large experience with reliable CT angiography (CTA), this may replace DSA.

Repeat imaging with three-dimensional DSA or multidetector CTA should be done in patients with an initial negative study in whom concerns for underlying aneurysm remain because of underlying vasospasm or the technical quality of the initial study. Rare patients who have recurrent hemorrhage should also have a repeat study. (See 'Initial cerebral angiography' above.)

●Management

•Patients with PM-SAH who are found to have a cerebral aneurysm should be managed as are other patients with aneurysmal SAH. (See "Aneurysmal subarachnoid hemorrhage: Treatment and prognosis".)

•Patients with PM-NASAH have a lower incidence of complications than patients with aneurysmal SAH. Patients should be clinically monitored for complications including rebleeding, seizures, clinical vasospasm, hydrocephalus, hyponatremia, and cardiac complications. (See 'Clinical course and complications' above.)

•Aside from monitoring, specific treatments are not required for patients with PM-NASAH. However, until a cerebral aneurysm has been excluded on follow-up imaging, we suggest using prophylactic nimodipine to ameliorate complications of potential vasospasm (Grade 2B). (See 'Treatment' above.)

●Prognosis – The long-term prognosis for patients with PM-NASAH is, in general, excellent. (See 'Long-term prognosis' above.)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges David Brock, MD, CIP, who contributed to earlier versions of this topic review.