Abstract 15057: Vitamin D Deficiency is Independently Associated With Subclinical Interstitial Lung Disease: The Multi-Ethnic Study of Atherosclerosis (MESA)
Background: Vitamin D deficiency and interstitial lung disease (ILD) have both been associated with cardiovascular disease (CVD). Yet, the relationship between vitamin D and ILD is not well known. Vitamin D deficiency may be a risk factor for ILD since vitamin D has anti-inflammatory properties. We hypothesized that 25-hydroxyvitamin D [25(OH)D] deficiency will be independently associated with subclinical ILD in a large population-based cohort study.
Methods: We studied 6302 MESA participants without clinical CVD who had baseline serum 25[OH]D levels and cardiac CTs that included partial lung fields in 2000-2002. Some participants had follow up cardiac CT scans at Exams 2-4 and a full lung CT scan at Exam 5 (2010-2012), with an average number of scans of 2.1 (1.0). Subclinical ILD was defined quantitatively as log-transformed volume (continuous) of high attenuation areas (HAA) between -600 and -250 Hounsfield units adjusted for lung volume. We used mixed model effect methods to assess the longitudinal association between baseline 25(OH)D stratified into three groups (<20, 20-29, ≥30 ng/ml) with HAA change. We also examined non-concurrent cross sectional relationship between baseline 25(OH)D and presence of interstitial lung abnormalities (ILA) (yes/no) determined qualitatively by radiologists from full lung CTs at Exam 5. We adjusted for demographics, smoking, and lifestyle factors.
Results: The mean (SD) age of our cohort was 62.2(10) years, 53% were women. Over a median follow up of 3.6 years, participants with lower 25(OH)D levels had increased progression of HAA [Figure]. Compared to those with replete levels (≥30 ng/ml), 25(OH)D deficiency (<20 ng/ml) was associated with increased HAA volume [exponentiated HAA ratio 1.02 (1.01, 1.03)] and with presence of ILAs [RR 1.43 (1.01, 2.02)].
Conclusions: Vitamin D deficiency was independently and modestly associated with subclinical ILD based on increased HAA volume and ILAs on high resolution CT scans.
Author Disclosures: S.M. Kim: None. D. Zhao: None. P.L. Lutsey: None. E. Guallar: None. A. Podolanczuk: None. S.M. Kawut: None. G. Raghu: None. I. de Boer: None. B. Kesternbaum: None. D.J. Lederer: None. E.D. Michos: Consultant/Advisory Board; Modest; Siemens Diagnostic.
- © 2016 by American Heart Association, Inc.