Quantitative MRI in Multiple Sclerosis Patients with Hypertension
Justin Abbatemarco1, Kunio Nakamura3, Scott Husak2, Daniel Ontaneda1, Robert Bermel1, Devon Conway1
1Mellen Center, 2Quantitative Health Sciences, Cleveland Clinic, 3Biomedical Engineering, Cleveland Clinic Lerner Research Institute
Objective:
To examine the impact of hypertension on quantitative magnetic resonance imaging (MRI) measures in multiple sclerosis (MS).
Background:
Cardiovascular comorbidities, including hypertension, are common in MS and are associated with worse outcomes.
Design/Methods:
We conducted a cross-sectional analysis of all patients who underwent quantitative MRI at our center from June 2015 - August 2019. Brain MRIs were quantitatively analyzed via a fully-automated method to calculate T2 lesion volume (T2LV), brain parenchymal fraction (BPF), and spinal cord area (SCA). Our electronic medical record (Epic) was queried to determine the number of hypertension diagnosis codes entered for each patient. Those with ≥2 entries were considered to have hypertension. Univariable and multivariable linear regression models were used to determine the relationship between hypertension and quantitative MRI metrics.
Results:
We identified 918 patients who underwent quantitative MRI, of whom 126 had hypertension. The average age among hypertensives was 52.2 years with average disease duration of 17.2 years. Those without hypertension had an average age and disease duration of 45.1 years and 13.3 years respectively. On univariable analysis, patients with hypertension had BPF that was 1.4% lower (p<0.0001), T2LV that was 7.5 mL greater (p<0.0001), and SCA that was 1.8 mm2 less (p=0.05) than those without hypertension. After adjusting for age, hypertension was still significantly associated with T2LV (p=0.0005), but not BPF (p=0.15) or SCA (p=0.4). When adjusting for age, disease duration, MS phenotype, sex, and race, there were no significant associations with hypertension.
Conclusions:
Hypertension was associated with greater T2LV, even after controlling for age, suggesting this comorbidity is an important mediator of disease activity. When adjusting for additional clinical factors, hypertension was not associated with our MRI measures suggesting its effects may be mediated through one of our covariates. Additional analysis is needed to determine if chronic small vessel disease contributes to higher T2LV in hypertensive patients.