Preoperative white matter lesions are independent predictors of long-term survival after internal carotid endarterectomy

Multivariable Analysis of Pre-operative Brain Atrophy as a Predictor of Long Term Mortality After Carotid Endarterectomy

OBJECTIVE

Brain atrophy is associated with an increased mortality rate in elderly trauma patients and in patients treated with mechanical thrombectomy for acute ischaemic stroke. In the setting of ischaemic stroke, the association between brain atrophy and death is stronger than that of sarcopenia. It has previously been shown that lower masseter area, as a marker of sarcopenia, is linked to lower survival after carotid endarterectomy (CEA). The aim of this study was to investigate whether brain atrophy is also associated with long term mortality in patients undergoing CEA.

METHODS

A cohort of patients treated with CEA between 2004 and 2010 was retrieved from the Tampere University Hospital vascular registry and those with available pre-operative computed tomography (CT) imaging were analysed retrospectively. CT images were evaluated for brain atrophy index (BAI) and masseter muscle surface area and density. The association between BAI and mortality was investigated with Cox regression.

RESULTS

Two hundred and thirty-three patients with a median (interquartile range [IQR]) age of 71 years (64.0, 77.0) were included. Most patients were operated on for symptomatic stenosis (n = 203; 87.1%). The median (IQR) duration of follow up was 115.0 months (66.0, 153.0), and 155 patients (66.5%) died during follow up. BAI was statistically significantly correlated with age (r = .489), average masseter density (r = -.202), and smoking (r = -.186; all p <.005). Increased BAI was statistically significantly associated with overall mortality (hazard ratio [HR] 1.45, 95% confidence interval [CI] 1.25 - 1.68, per one standard deviation [SD] increase) in the univariable analysis, and the association remained (HR 1.23, 95% CI 1.04 - 1.46, per one SD increase) in the multivariable models. Age, peripheral artery disease, and chronic obstructive pulmonary disease were also independently associated with mortality. The optimal cutoff value for BAI was 0.133.

CONCLUSION

Brain atrophy independently predicts the long term post-operative mortality rate after CEA in a cohort containing mainly symptomatic patients. Future studies are needed to validate the results in prospective settings and in asymptomatic patients.

Evaluating severity of white matter lesions from computed tomography images with convolutional neural network

PURPOSE

Severity of white matter lesion (WML) is typically evaluated on magnetic resonance images (MRI), yet the more accessible, faster, and less expensive method is computed tomography (CT). Our objective was to study whether WML can be automatically segmented from CT images using a convolutional neural network (CNN). The second aim was to compare CT segmentation with MRI segmentation.

METHODS

The brain images from the Helsinki University Hospital clinical image archive were systematically screened to make CT-MRI image pairs. Selection criteria for the study were that both CT and MRI images were acquired within 6 weeks. In total, 147 image pairs were included. We used CNN to segment WML from CT images. Training and testing of CNN for CT was performed using 10-fold cross-validation, and the segmentation results were compared with the corresponding segmentations from MRI.

RESULTS

A Pearson correlation of 0.94 was obtained between the automatic WML volumes of MRI and CT segmentations. The average Dice similarity index validating the overlap between CT and FLAIR segmentations was 0.68 for the Fazekas 3 group.

CONCLUSION

CNN-based segmentation of CT images may provide a means to evaluate the severity of WML and establish a link between CT WML patterns and the current standard MRI-based visual rating scale.

Cerebral Small Vessel Disease in Standard Pre-operative Imaging Reports Is Independently Associated with Increased Risk of Cardiovascular Death Following Carotid Endarterectomy

OBJECTIVE

Cerebral white matter lesions (WMLs) and lacunar infarcts are surrogates of cerebral small vessel disease (SVD). WML severity as determined by trained radiologists predicts post-operative stroke or death in patients undergoing carotid endarterectomy (CEA). It is unknown whether routine pre-operative brain imaging reports as part of standard clinical practice also predict short and long term risk of stroke and death after CEA.

METHODS

Consecutive patients from the Athero-Express biobank study that underwent CEA for symptomatic high degree stenosis between March 2002 and November 2014 were included. Pre-operative brain imaging (computed tomography [CT] or magnetic resonance imaging [MRI]) reports were reviewed for reporting of SVD, defined as WMLs or any lacunar infarcts. The primary outcome was defined as any stroke or any cardiovascular death over three year follow up. The secondary outcome was defined as the 30 day peri-operative risk of stroke or cardiovascular death.

RESULTS

A total of 1038 patients were included (34% women), of whom 659 (63.5%) had CT images and 379 (36.5%) MRI images available. Of all patients, 697 (67%) had SVD reported by radiologists. Patients with SVD had a higher three year risk of cardiovascular death than those without (6.5% vs. 2.1%, adjusted HR 2.52 [95% CI 1.12-5.67]; p = .026) but no association was observed for the three year risk of stroke (9.0% vs. 6.7%, for patients with SVD vs. those without, adjusted HR 1.24 [95% CI 0.76-2.02]; p = .395). No differences in 30 day peri-operative risk were observed for stroke (4.4% vs. 2.9%, for patients with vs. those without SVD; adjusted HR 1.49 [95% CI 0.73-3.05]; p = .28), and for the combined stroke/cardiovascular death risk (4.4% vs. 3.5%, adjusted HR 1.20 [95% CI 0.61-2.35]; p = .59).

CONCLUSION

Presence of SVD in pre-operative brain imaging reports can serve as a predictor for the three year risk of cardiovascular death in symptomatic patients undergoing CEA but does not predict peri-operative or long term risk of stroke.

Long-term effects of white matter changes on the risk of stroke recurrence after carotid artery stenting in patients with symptomatic carotid artery stenosis

INTRODUCTION

Cerebral white matter changes (WMC) are associated with increased 30-day perioperative risk of stroke in patients undergoing carotid artery stenting (CAS). However, there is no data showing their impact on postoperative long-term stroke recurrence or survival. It remains unknown whether this effect is independent of classic cardiovascular risk factors or not. We tried to assess the effects of WMC on long-term stroke recurrence after CAS in patients with symptomatic carotid stenosis.

METHODS

A database of patients with symptomatic carotid stenosis who had undergone CAS was sampled in a single Chinese medical center from 2007 to 2014. Copies of baseline brain imaging were analyzed by two investigators to evaluate the severity of WMC. We analyzed the association between WMC and stroke recurrence after CAS by reviewing case histories and conducting telephone interviews, with a mean follow-up time of 28months.

RESULTS

107 patients with an average age of 66years fulfilled the inclusion criteria, of which 70 were examined with CT and the remainder by MRI. In our cohort, 91.6% of the participants were male. There were 29 patients with diffuse WMC. In univariate analysis, patients with diffuse WMC had more contralateral occlusion than those with non-diffuse WMC. In patients with diffuse WMC, more stroke recurrences were observed compared with those with non-diffuse WMC (hazard ratio [HR] 3.516; 95% CI 1.176, 10.510, P=0.024).

CONCLUSIONS

In patients with symptomatic carotid artery stenosis, diffuse WMC were a risk factor for stroke recurrences after CAS. Larger studies are warranted to confirm this finding and explore the potential clinical impact of WMC so as to better determine treatment strategies for patients with symptomatic carotid artery stenosis.