Colum P. Nolan and R. Loch Macdonald
✓ The authors tested the null hypothesis that published literature with a high level of evidence does not support the assertion that subarachnoid hemorrhage (SAH) causes cerebral vasospasm, which in turn causes cerebral infarction and poor outcome after aneurysmal SAH. The medical literature on SAH was searched in MEDLINE. The author's personal files of all published literature on SAH were reviewed. References cited in Cochrane reviews as well as the published papers that were reviewed were also retrieved.
There is no question that SAH causes what the authors have chosen to call “angiographic vasospasm.” However, the incidence and severity of vasospasm in recent series of patients is not well defined. There is reasonable evidence that vasospasm causes infarction, but again, accurate data on how severe and how diffuse vasospasm has to be to cause infarction and how often vasospasm is the primary cause of infarction are not available. There are good data on the incidence of cerebral infarction after SAH, and these data indicate that it is highly associated with poor outcome. The link between angiographic vasospasm and poor outcome is particularly poorly described in terms of what would be considered data of a high level of evidence.
The question as to whether there is a clear pathway from SAH to vasospasm to cerebral infarction to poor outcome seems so obvious to neurosurgeons as to make it one not worth asking. Nevertheless, the obvious is not always true or accurate, so it is important to note that published literature only weakly supports the causative association of vasospasm with infarction and poor outcome after SAH. It behooves neurosurgeons to document this seemingly straightforward pathway with high-quality evidence acceptable to the proponents of evidence-based medicine.
R. Loch Macdonald
R. Loch Macdonald
George M. Ibrahim, Aria Fallah and R. Loch Macdonald
At present, the administration of prophylactic antiepileptic medication following aneurysmal subarachnoid hemorrhage (SAH) is controversial, and the practice is heterogeneous. Here, the authors sought to inform clinical decision making by identifying factors associated with the occurrence of seizures following aneurysm rupture.
Exploratory analysis was performed on 413 patients enrolled in CONSCIOUS-1 (Clazosentan to Overcome Neurological Ischemia and Infarction Occurring after Subarachnoid Hemorrhage), a prospective randomized trial of clazosentan for the prevention of angiographic vasospasm. The association among clinical, laboratory, and radiographic covariates and the occurrence of seizures following SAH were determined. Covariates with a significance level of p < 0.20 on univariate analysis were entered into a multivariate logistic regression model. Receiver operating characteristic (ROC) curve analysis was used to define optimal predictive thresholds.
Of the 413 patients enrolled in the study, 57 (13.8%) had at least 1 seizure following SAH. On univariate analysis, a World Federation of Neurosurgical Societies grade of IV–V, a greater subarachnoid clot burden, and the presence of midline shift and subdural hematomas were associated with seizure activity. On multivariate analysis, only a subarachnoid clot burden (OR 2.76, 95% CI 1.39–5.49) and subdural hematoma (OR 5.67, 95% CI 1.56–20.57) were associated with seizures following SAH. Using ROC curve analysis, the optimal predictive cutoff for subarachnoid clot burden was determined to be 21 (of a possible 30) on the Hijdra scale (area under the curve 0.63).
A greater subarachnoid clot burden and subdural hematoma are associated with the occurrence of seizures after aneurysm rupture. These findings may help to identify patients at greatest risk for seizures and guide informed decisions regarding the prescription of prophylactic anticonvulsive therapy. Clinical trial registration no.: NCT00111085 (ClinicalTrials.gov).