Predictors of 30-day mortality using machine learning approach following carotid endarterectomy.

BACKGROUND: Preoperative prognostication of 30-day mortality in patients with carotid endarterectomy (CEA) can optimize surgical risk stratification and guide the decision-making process to improve survival. This study aims to develop and validate a set of predictive variables of 30-day mortality following CEA. METHODS: The patient cohort was identified from the American College of Surgeons National Surgical Quality Improvement Program (2005-2016). We performed logistic regression (enter, stepwise, and forward) and least absolute shrinkage and selection operator (LASSO) method for the selection of variables, which resulted in 28-candidate models. The final model was selected based upon clinical knowledge and numerical results. RESULTS: Statistical analysis included 65,807 patients with 30-day mortality in 0.7% (n = 466) patients. The median age of our cohort was 71.0 years (range, 16-89 years). The model with 9 predictive factors which included age, body mass index, functional health status, American Society of Anesthesiologist grade, chronic obstructive pulmonary disorder, preoperative serum albumin, preoperative hematocrit, preoperative serum creatinine, and preoperative platelet count-performed best on discrimination, calibration, Brier score, and decision analysis to develop a machine learning algorithm. Logistic regression showed higher AUCs than LASSO across these different models. The predictive probability derived from the best model was converted into an open-accessible scoring system. CONCLUSION: Machine learning algorithms show promising results for predicting 30-day mortality following CEA. These algorithms can be useful aids for counseling patients, assessing preoperative medical risks, and predicting survival after surgery.

Risk Factors for Cerebral Vasospasm in Patients with Aneurysmal Subarachnoid Hemorrhage: A Tertiary Care Center Experience.

Objectives Cerebral vasospasm in subarachnoid hemorrhage (SAH) is associated with high morbidity and mortality. There is a lack of consensus on the risk factors leading to cerebral vasospasm in patients with aneurysmal subarachnoid hemorrhage (aSAH). In this retrospective study, our objective was to determine the association of risk factors for cerebral vasospasm aSAH. Methods A total of 259 charts of aSAH patients consecutively admitted to the surgical intensive care unit of Hamad General Hospital from January 2007 to December 2016 were reviewed and included. The patient's demographic data, including comorbidities like hypertension (HTN), was recorded. Variables of interest included measurements of the neurological deficit on admission, the severity of SAH, treatment modality, and the initial computerized tomography scan of the head for intraventricular hemorrhage, intracerebral hemorrhage, or hydrocephalus. Multivariate analysis and multiple logistic regression analyzed the relationship to identify the association of independent variables. Results Out of the 259 patients, 34% ( n = 87) suffered from cerebral vasospasm. The severity of SAH was associated with the development of cerebral vasospasm ( p < 0.05). The presence of HTN and neurological deficits on admission were associated with an increased risk of cerebral vasospasm ( p < 0.05, p < 0.01, respectively). Hydrocephalus requiring treatment using external ventricular drains decreased the risk of cerebral vasospasm ( p < 0.05). Intraventricular and intracerebral hemorrhage were not associated with cerebral vasospasm ( p = 0.25, p = 0.16). The endovascular treatment of cerebral aneurysms was associated with an increased risk of cerebral vasospasm ( p < 0.05). Conclusion Cerebral vasospasm is common among patients admitted with aSAH. It is significantly associated with the history of HTN, the neurological deficit on admission that corelates more strongly to the motor deficit on admission, the severity of hemorrhage (modified Fischer score), and endovascular treatment. External ventricular drainage was associated with a decrease in cerebral vasospasm. The present study's findings shed light on cerebral vasospasm's risk factors in the country and the region.

Patient-Reported Outcome for Endovascular Treatment versus Microsurgical Clipping in Aneurysmal Subarachnoid Hemorrhage.

OBJECTIVE: Aneurysmal subarachnoid hemorrhage has a high mortality with significant impact on quality of life despite effective management strategies including endovascular treatment and/or microsurgical clipping. Although the modalities have undergone clinical comparison, they have not been evaluated on patient-reported outcomes (PROs). This study compared endovascular versus microsurgical treatment using a PRO measure. METHODS: We conducted a cross-sectional telephonic survey of adult patients conducted at Hamad General Hospital, Doha, Qatar between 2017 and 2019. Candidate study participants were identified from procedure logs and hospital electronic health records for endovascular treatment (N = 32) versus microsurgical clipping (N = 32) of cerebral aneurysm. The primary outcome measure was the short version of the Stroke-Specific Quality of Life (SS-QoL) measure. The secondary outcome measure was the screened clinician-reported modified Rankin Scale (mRS) for all screened patients (n = 137). Mean scores were compared for the 2 treatment groups. RESULTS: The SS-QoL mean score was 4.23 (standard deviation ± 0.77) in endovascular treatment and 4.19 ± 0.19 in surgical clipping (P = 0.90). In exploratory analysis, mean physical domain score was 3.17 ± 0.60 versus 2.98 ± 0.66 in endovascular treatment and surgical clipping groups, respectively. Mean psychosocial domain scores were 4.43 ± 0.85 versus 4.18 ± 0.0.92, respectively. In multivariable analysis, none of the clinical variables were significantly related to SS-QoL except vasospasm irrespective of intervention received. In secondary outcome analysis, modified Rankin Scale score was higher for endovascular treatment (P = 0.04). CONCLUSIONS: Published evidence has supported clinical benefits of endovascular treatment for cerebral aneurysm treatment, but this study did not find any difference in PROs. Future studies of treatments should include PRO to identify potential differences from the patient's perspective.

Baseline Occlusion Angiographic Appearance on Mechanical Thrombectomy Suggests Underlying Etiology and Outcome.

Background: Etiology of a large vessel occlusion is relevant in the management of acute ischemic stroke patients and often difficult to determine in the acute phase. Aims: We aim to investigate whether the angiographic appearance of the occlusion is related to its etiology and outcome. Materials and Methods: Patients without cervical carotid occlusions who underwent mechanical thrombectomy in our center from April 2015 to September 2018 were studied. Demographics, clinical and radiological variables and outcome measures, including etiological classification of stroke, were collected. Underlying intracranial atherosclerosis was estimated according to the presence of stenosis after recanalization. Patients were assigned to groups based on the appearance of the occlusion observed in the first angiogram as "tapered" or "non-tapered." Differences were searched amongst them. Results: 131 patients met inclusion criteria. 31 (23.6%) were "tapered" and 100 (76.3%) non-tapered. Tapered presented lower mean baseline NIHSS (10.3 ± 6.2 vs. 16.1 ± 7.2; p < 0.001), smaller acute infarct cores as CTP CBV ASPECTS (8.6 ± 1.6 vs. 7.2 ± 2.4; p = 0.003), higher proportion of instant re-occlusions (26.7 vs. 8.2%; p = 0.025), fewer complete recanalization (45.2 vs. 71.0%; p = 0.028), and more persistent occlusions (37.5 vs. 10.6%; p = 0.011) on follow up MRA. There were no differences in reperfusion rates (83.9 vs. 84.0%; p = 0.986) nor in good long term functional outcome (50.0 vs. 51.1%; p = 0.921). Intracranial atherosclerosis etiology was more common in tapered than in non-tapered occlusions (54.8 vs. 18.0%; p < 0.001). Conclusion: The angiographic appearance of an occlusion in mechanical thrombectomy patients may determine its etiology, predict likelihood of successful recanalization, and risk of reocclusion.