Arjen Lindenholz

114 CHAPTER 5 Hypertension was defined as systolic blood pressure greater than 140 mmHg or diastolic blood pressure greater than 90 mmHg or use of antihypertensive drugs; 23 diabetes mellitus as fasting serum glucose levels greater than 6.9 mmol/L or use of antidiabetic drugs; 24 and hyperlipidemia as either total serum cholesterol level greater than 6.2 mmol/L, LDL cholesterol level greater than 4.1 mmol/L, HDL cholesterol level less than 1.0 mmol/L, triglyceride level greater than 2.2 mmol/L, or use of lipid-lowering drugs. 25 From the collected data, the SecondManifestations of ARTerial disease (SMART) vascular risk score was calculated to combine risk factors (age, sex, current smoking status, systolic blood pressure, diabetes mellitus, coronary artery disease, cerebrovascular disease, abdominal aortic aneurysm, peripheral artery disease, time since first cardiovascular diagnose, HDL cholesterol, total cholesterol, eGFR, and high-sensitivity CRP) into one score. 26 This score predicts the 10-year risk of recurrent vascular events in participants with clinically manifest atherosclerosis. Data evaluation and MRI analysis All MRI scans were assessed with an offline workstation (Philips DICOM Viewer 3.0, Philips Healthcare, Best, Netherlands). Two readers (AK, 8 years of experience and AL, 3 years of experience) independently assessed the MRI series of all participants ( n = 90). Readers were blinded for any patient characteristics. The degree of reliabilityandagreementwas calculatedandcomparedwith the levelsof agreement published in recent literature. 27,28 In addition, after 15 image assessments, the two readers held an interim consensus meeting as quality assurance for lesion- definition adherence and to improve the accuracy and consistency of vessel wall assessment for the whole dataset. Any discrepancies were resolved, and, in the case of disagreement, a third reader (JH, neuroradiologist with > 10 years of experience) was consulted. The count results of the most trained reader (AK, 8 years of experience) were used for the descriptive results ( Supplemental Table 1 ). Arterial vessel segments that were assessed bilaterally for vessel wall lesions included the distal internal carotid artery (ICA, clinoid, supra-clinoid and terminal segments), middle cerebral artery (MCA; M1 and M2 segments), proximal anterior cerebral artery (ACA; A1 and A2 segments), posterior cerebral artery (PCA; P1 and P2 segments), intracranial segment of the vertebral artery (VA), and the basilar artery (BA). In general, the process of lesion detection, with the comparison with the normal-appearing vessel segments either on the ipsilateral and contralateral side and possible lesion enhancement is in agreement with the radiology practice of interpreting intracranial vessel wall lesions in our radiology department, as has been described in more detail in a recent publication in Radiology . 18 A vessel wall lesion was defined as a visually defined focal or diffuse thickening of the vessel wall greater than 50% compared with the adjacent or contralateral vessel wall thickness on both pre- and postcontrast images. The vessel wall images could be