Cavernous Malformation Hemorrhagic Presentation at Diagnosis Associated with Low 25-Hydroxy-Vitamin D Level.

BACKGROUND: Cavernous malformations (CM) are angiographically occult vascular malformations that may be incidental or present with intracerebral or spinal hemorrhage, seizures, or nonhemorrhagic focal neurologic deficit (FND). Recently in vitro data have suggested vitamin D may play a role in stabilizing CCM2 endothelial cells. Little is known about the effect of vitamin D in human CM disease. METHODS: Beginning in 2015, consecutive patients at our institution with radiologically confirmed CM were recruited to participate in a prospective clinical registry as well as 25-hydroxy-vitamin D study. A structured interview, survey, and examination were performed at baseline. Medical records and magnetic resonance imaging studies were reviewed and data collected included comorbid conditions, medication use, and location of CM. Standard definition of clinical hemorrhage, FND, and seizures was used. Univariate and multivariate logistic regression models were used, and OR, 95% CIs, and likelihood-ratio p values were calculated to determine the influence of the 25-hydroxy-vitamin D level on clinical presentation with hemorrhage. RESULTS: Of 213 patients enrolled in the clinical registry between January 2015 and October 2018, 70 participated in the vitamin D study (median age: 38.3 years; 51.4% female). Of the 70 participants, 30 (42.9%) presented with hemorrhage. 25-Hydroxy-vitamin D levels were performed within 1 year of symptoms in 64.1% of patients. Patients presenting with hemorrhage had a lower 25-hydroxy-vitamin D level compared to those presenting with seizure without hemorrhage, FND, or as an incidental finding (median 25.5 ng/mL; range 11-59 hemorrhage vs. median 31.0; range 14-60, no hemorrhage; p = 0.04). After adjusting for age, month of blood draw, and body mass index, 25-hydroxy-vitamin D remained a significant predictor of hemorrhagic presentation. Brainstem location also predicted hemorrhage at presentation. CONCLUSION: Low 25-hydroxy-vitamin D level was more common in patients with CM presenting with hemorrhage. This study supports the potential role of modifiable factor in the initial clinical presentation of CM. Further study is needed to determine the role of vitamin D on prospective hemorrhage risk and whether supplementation may be beneficial.

Effect of Continuous-Flow Left Ventricular Assist Device Support on Coronary Artery Endothelial Function in Ischemic and Nonischemic Cardiomyopathy.

BACKGROUND: The coronary vasculature encounters a reduction in pulsatility after implementing durable continuous-flow left ventricular assist device (CF-LVAD) circulatory support. Evidence exists that appropriate pulsatility is required to maintain endothelial cell homeostasis. We hypothesized that coronary artery endothelial function would be impaired after CF-LVAD intervention. METHODS AND RESULTS: Coronary arteries from patients with end-stage heart failure caused by ischemic cardiomyopathy (ICM; n=16) or non-ICM (n=22) cardiomyopathy were isolated from the left ventricular apical core, which was removed for the CF-LVAD implantation. In 11 of these patients, paired coronary arteries were obtained from an adjacent region of myocardium after the CF-LVAD intervention (n=6 ICM, 5 non-ICM). Vascular function was assessed ex vivo using isometric tension procedures in these patients and in 7 nonfailing donor controls. Maximal endothelium-dependent vasorelaxation to BK (bradykinin; 10-6-10-10 M) was blunted (P<0.05) in arteries from patients with ICM compared with non-ICM and donor controls, whereas responses to sodium nitroprusside (10-4-10-9 M) were similar among the groups. Contrary to our hypothesis, vasorelaxation responses to BK and sodium nitroprusside were similar before and 219±37 days after CF-LVAD support. Of these patients, an exploratory subgroup analysis revealed that BK-induced coronary artery vasorelaxation was greater (P<0.05) after (87±6%) versus before (54±14%) CF-LVAD intervention in ICM patients, whereas sodium nitroprusside-evoked responses were similar. CONCLUSIONS: Coronary artery endothelial function is not impaired by durable CF-LVAD support and in ICM patients appears to be improved. Investigating coronary endothelial function using in vivo approaches in a larger patient population is warranted.