Decision-making and neurosurgeons' agreement in the management of aneurysmal subarachnoid haemorrhage based on computed tomography angiography

Development of a Complication- and Treatment-Aware Prediction Model for Favorable Functional Outcome in Aneurysmal Subarachnoid Hemorrhage Based on Machine Learning

BACKGROUND

Current prognostic tools in aneurysmal subarachnoid hemorrhage (aSAH) are constrained by being primarily based on patient and disease characteristics on admission.

OBJECTIVE

To develop and validate a complication- and treatment-aware outcome prediction tool in aSAH.

METHODS

This cohort study included data from an ongoing prospective nationwide multicenter registry on all aSAH patients in Switzerland (Swiss SOS [Swiss Study on aSAH]; 2009-2015). We trained supervised machine learning algorithms to predict a binary outcome at discharge (modified Rankin scale [mRS] ≤ 3: favorable; mRS 4-6: unfavorable). Clinical and radiological variables on admission ("Early" Model) as well as additional variables regarding secondary complications and disease management ("Late" Model) were used. Performance of both models was assessed by classification performance metrics on an out-of-sample test dataset.

RESULTS

Favorable functional outcome at discharge was observed in 1156 (62.0%) of 1866 patients. Both models scored a high accuracy of 75% to 76% on the test set. The "Late" outcome model outperformed the "Early" model with an area under the receiver operator characteristics curve (AUC) of 0.85 vs 0.79, corresponding to a specificity of 0.81 vs 0.70 and a sensitivity of 0.71 vs 0.79, respectively.

CONCLUSION

Both machine learning models show good discrimination and calibration confirmed on application to an internal test dataset of patients with a wide range of disease severity treated in different institutions within a nationwide registry. Our study indicates that the inclusion of variables reflecting the clinical course of the patient may lead to outcome predictions with superior predictive power compared to a model based on admission data only.

Surgical Interventions for Cervical Radiculopathy without Myelopathy: A Systematic Review and Meta-Analysis

BACKGROUND

The effectiveness of surgical interventions for cervical degenerative disorders has been investigated in multiple systematic reviews. Differences in study population (e.g., patients with myelopathy and/or radiculopathy) were often neglected. Therefore, the objective of this study was to investigate the effectiveness of surgical interventions for patients with symptoms of cervical radiculopathy without myelopathy by conducting a systematic review and meta-analysis based on randomized controlled trials (RCTs).

METHODS

A comprehensive systematic search was conducted in MEDLINE, Embase, and CENTRAL (Cochrane Central Register of Controlled Trials) to identify RCTs that investigated the effectiveness of surgical interventions using an anterior or posterior approach compared with other interventions for patients with pure cervical radiculopathy. Outcomes were success rates (Odom criteria, similar rating scales, or percentage of patients who improved), complication and reoperation rates, work status, disability (Neck Disability Index), and pain (arm and neck). The Cochrane risk-of-bias tool was used to assess the likelihood of the risk of bias. A random-effects model was used. Heterogeneity among study results (I2 ≥ 50% or p < 0.05) was explored by conducting subgroup analyses. Funnel plots were used to assess the likelihood of publication bias.

RESULTS

A total of 21 RCTs were included, comprising 1,567 patients. For all outcomes, among all surgical techniques, only 1 pooled estimate showed a significant effect on success rate, which was in favor of anterior cervical discectomy with fusion compared with anterior cervical discectomy without an intervertebral spacer (p = 0.02; risk ratio [RR] = 0.87; 95% confidence interval [CI] = 0.77 to 0.98). Complication rates were higher when autologous bone graft from the iliac crest was used as an intervertebral spacer (p < 0.01; RR = 3.40; 95% CI = 1.56 to 7.43), related to donor-site morbidity.

CONCLUSIONS

This meta-analysis demonstrated consistent results regarding clinical outcome for pure cervical radiculopathy among all studied interventions. Complication and reoperation rates were also similar, with the exception of higher complication rates in patients in whom autologous bone grafts were used. On the basis of clinical outcome and safety, there is no superior surgical intervention for pure cervical radiculopathy.

LEVEL OF EVIDENCE

Therapeutic Level I. See Instructions for Authors for a complete description of levels of evidence.

Patterns of care for ruptured aneurysms of the middle cerebral artery: analysis of a Swiss national database (Swiss SOS)

OBJECTIVE

The objective of this study was to determine patterns of care and outcomes in ruptured intracranial aneurysms (IAs) of the middle cerebral artery (MCA) in a contemporary national cohort.

METHODS

The authors conducted a retrospective analysis of prospective data from a nationwide multicenter registry of all aneurysmal subarachnoid hemorrhage (aSAH) cases admitted to a tertiary care neurosurgical department in Switzerland in the years 2009-2015 (Swiss Study on Aneurysmal Subarachnoid Hemorrhage [Swiss SOS]). Patterns of care and outcomes at discharge and the 1-year follow-up in MCA aneurysm (MCAA) patients were analyzed and compared with those in a control group of patients with IAs in locations other than the MCA (non-MCAA patients). Independent predictors of a favorable outcome (modified Rankin Scale score ≤ 3) were identified, and their effect size was determined.

RESULTS

Among 1866 consecutive aSAH patients, 413 (22.1%) harbored an MCAA. These MCAA patients presented with higher World Federation of Neurosurgical Societies grades (p = 0.007), showed a higher rate of concomitant intracerebral hemorrhage (ICH; 41.9% vs 16.7%, p < 0.001), and experienced delayed cerebral ischemia (DCI) more frequently (38.9% vs 29.4%, p = 0.001) than non-MCAA patients. After adjustment for confounders, patients with MCAA were as likely as non-MCAA patients to experience DCI (aOR 1.04, 95% CI 0.74-1.45, p = 0.830). Surgical treatment was the dominant treatment modality in MCAA patients and at a significantly higher rate than in non-MCAA patients (81.7% vs 36.7%, p < 0.001). An MCAA location was a strong independent predictor of surgical treatment (aOR 8.49, 95% CI 5.89-12.25, p < 0.001), despite statistical adjustment for variables traditionally associated with surgical treatment, such as (space-occupying) ICH (aOR 1.73, 95% CI 1.23-2.45, p = 0.002). Even though MCAA patients were less likely to die during the acute hospitalization (aOR 0.52, 0.30-0.91, p = 0.022), their rate of a favorable outcome was lower at discharge than that in non-MCAA patients (55.7% vs 63.7%, p = 0.003). At the 1-year follow-up, 68.5% and 69.6% of MCAA and non-MCAA patients, respectively, had a favorable outcome (p = 0.676).

CONCLUSIONS

Microsurgical occlusion remains the predominant treatment choice for about 80% of ruptured MCAAs in a European industrialized country. Although patients with MCAAs presented with worse admission grades and greater rates of concomitant ICH, in-hospital mortality was lower and long-term disability was comparable to those in patients with non-MCAA.