Background: Non-traumatic subarachnoid hemorrhage (SAH) is a devastating disease that accounts for less than 10% of all strokes but is one of the most common cause of stroke-related deaths. However, for a small subset of these patients, specifically those in which a bleed source cannot be identified (angiogram-negative), outcomes tend to be more favorable. This group can be further subdivided into patients with a perimesencephalic bleed (PMH) and non-perimesencephalic (nPMH) bleeding patterns, with some studies suggesting differences in outcomes between the two groups. In this study, the authors aimed to characterize the incidence, complications, and clinical outcomes of both nPMH and PMH angiogram-negative SAH.
Methods: This was a retrospective study of patients with angiogram-negative SAH admitted to the NeuroICU at Helsinki University Hospital from September 2004 to January 2018. Patients were considered eligible if they met the following inclusion criteria: adults 18 years of age and older with at least one angiographic study (MRA, CTA, DSA) that was negative for the source of the bleed. Exclusion criteria included: history of antecedent trauma, presence of an intracranial vascular lesion (e.g. aneurysm, arteriovenous shunt lesion such as arteriovenous malformation or fistula, etc.), vasculitis, and reversible cerebral vasoconstriction syndrome (RCVS). All patients were treated according to the institutional SAH pathways, which included administration of Nimodipine and Tranexamic acid.
Individual chart review of each patient was performed to extract a number of clinical, demographic, and radiographic characteristics, including presenting clinical syndrome and severity, presence of clinical and radiographic hydrocephalus, vasospasm, delayed cerebral ischemia (DCI), and need for active vasospasm treatment.
The primary outcome was a dichotomized Glasgow Outcome Score (GOS) at 90 days, with scores of 4-5 representing favorable outcomes and scores 3 or less representing unfavorable outcomes. Additional outcomes included rates of re-bleeding as well as mortality at 1 year. Univariate and multivariate analyses were performed to identify factors associated with outcome.
Results: One hundred and thirty-nine patients were identified between 2004 and 2018 with angiogram-negative SAH, of which 108 patients comprised the final cohort, after excluding those who either did not meet inclusion criteria or lacked follow-up data.
The median age in this group was 58 and roughly 50% were women. The majority of patients were low-grade bleeds (69% had a GCS score of 15 on admission and nearly 75% had a modified Fisher score of 0-2) and 39% had a PMH pattern of bleeding. Acute hydrocephalus was present in 30% of patients, of which more than half (56%) required emergent CSF diversions, with median duration of drain placement of approximately 1 week. Radiographic vasospasm was noted in 66% of patients, and 6% developed DCI.
Vascular imaging was performed in all patients, and the median number performed on each patient was 2. The majority were CTAs (78%), followed by DSA (17%) and MRA (5%). Of note, 28% underwent at least one DSA.
For the entire cohort, 84% had a favorable GOS score (4-5) at 90 days. Among those with a favorable outcome, 42% had a PMH bleed pattern on admission, and among those with an unfavorable outcome, 13% had a PMH bleeding pattern. Overall 1-year mortality was 5% and re-bleeding occurred in 2% of the population with a median follow-up time of 7.6 years. While a number of variables including age, severity of bleed and clinical syndrome on admission, and nPMH bleeding pattern were associated with poor outcome in univariate analysis, multivariate analysis only identified acute hydrocephalus as an independent predictor of unfavorable outcome (OR 4.93; 95% CI 1.14-21.41, p = 0.033).
When separated by PMH versus nPMH bleeding, 95% of those with PMH had a favorable GOS score of 4-5 at 90 days, while in the nPMH group, 79% of patients had a favorable 90-day functional outcome. Multivariate regression was only performed in the nPMH group as only two patients in the PMH group had poor outcomes at 90 days, which again only identified acute hydrocephalus as an independent predictor of poor outcome (OR 8.56, 95% CI 1.59-46.23).
Commentary: The primary finding of this study is that in keeping with prior literature, the majority (84%) of patients with angiogram-negative SAH have favorable functional outcomes at 90 days with low overall mortality (5%) and rate of re-bleeding (2%). In univariate analysis, age, history of smoking, anti-thrombotic medication use, poor SAH grade on admission, hydrocephalus, and vasospasm were all associated with poor functional outcomes at 90 days. However, only presence of acute hydrocephalus on admission was an independent predictor of poor outcome in the multivariate analysis. Roughly one-third of patients in this cohort developed hydrocephalus, which is lower than what is seen in aneurysmal SAH (approximately 40%). Additionally, 16% of patients in this study developed vasospasm, which is again lower than what is seen in patients with aneurysmal SAH (33%). The relatively benign clinical course and favorable functional outcomes seen in angiogram-negative SAHs perhaps alludes to the etiology of these bleeds, which are thought to either reflect a venous, low pressure hemorrhage or leakage from small perforating vessels in the lenticulostriate system. However definitive etiology in many of these cases remain unknown.
The primary strength of this study is that it is a large cohort of patients treated in the same institution over a 14 year period, minimizing confounding effects of different treatment practices. It also reinforces that the distinction that is made between PMH and nPMH is also clinically relevant. However, a major limitation of the study is that not all patients underwent DSA to evaluate for the presence of a vascular lesion. Although non-invasive vascular imaging modalities such as MRA and CTA are becoming increasingly sophisticated in their ability to detect small lesions, DSA remains the gold standard for diagnosis. Therefore, it is conceivable that patients characterized as “angiogram-negative” in fact had a small aneurysm that went undetected, potentially skewing the outcomes in a negative direction, as we know that aneurysmal SAH tends to have higher rates of inpatient complications, death, and long term disability. However, the study is nonetheless helpful in establishing outcomes and predictors of outcome after angiogram-negative subarachnoid hemorrhages, which is important for the neurointensivist to understand in order to best care for these patients.
Preethi Ramchand, MD
Clinical Assistant Professor of Neurosurgery and Neurology
Divisions of Neurocritical Care and Interventional Neuroradiology
Medical Director of Neurocritical Care: Main Line Health
Main Line Health Jefferson Neurosurgery