Abstract
Background
Activated vitamin D has anti-inflammatory properties. 25-Hydroxyvitamin D [25(OH)D] deficiency might contribute to subclinical interstitial lung disease (ILD).
Objective
We examined associations between serum 25(OH)D concentrations and subclinical ILD among middle-aged to older adults who were free of cardiovascular disease at baseline.
Methods
We studied 6302 Multi-Ethnic Study of Atherosclerosis (MESA) participants who had baseline serum 25(OH)D concentrations and computed tomography (CT) imaging spanning ≤ 10 y. Baseline cardiac CT scans (2000–2002) included partial lung fields. Some participants had follow-up cardiac CT scans at exams 2–5 and a full-lung CT scan at exam 5 (2010–2012), with a mean ± SD of 2.1 ± 1.0 scans. Subclinical ILD was defined quantitatively as high-attenuation areas (HAAs) between –600 and –250 Hounsfield units. We assessed associations of 25(OH)D with adjusted HAA volumes and HAA progression. We also examined associations between baseline 25(OH)D and the presence of interstitial lung abnormalities (ILAs) assessed qualitatively (yes or no) from full-lung CT scans at exam 5. Models were adjusted for sociodemographic characteristics, lifestyle factors (including smoking), and lung volumes.
Results
The cohort's mean ± SD characteristics were 62.2 ± 10 y for age, 25.8 ± 10.9 ng/mL for 25(OH)D concentrations, and 28.3 ± 5.4 for body mass index (kg/m2); 53% were women, with 39% white, 27% black, 22% Hispanic, and 12% Chinese race/ethnicities. Thirty-three percent had replete (≥30 ng/mL), 35% intermediate (20 to <30 ng/mL), and 32% deficient (<20 ng/mL) 25(OH)D concentrations. Compared with those with replete concentrations, participants with 25(OH)D deficiency had greater adjusted HAA volume at baseline (2.7 cm3; 95% CI: 0.9, 4.5 cm3) and increased progression over a median of 4.3 y of follow-up (2.7 cm3; 95% CI: 0.9, 4.4 cm3) (P < 0.05). 25(OH)D deficiency was also associated with increased prevalence of ILAs 10 y later (OR: 1.5; 95% CI: 1.1, 2.2).
Conclusions
Vitamin D deficiency is independently associated with subclinical ILD and its progression, based on both increased HAAs and ILAs, in a community-based population. Further studies are needed to examine whether vitamin D repletion can prevent ILD or slow its progression. The MESA cohort design is registered at www.clinicaltrials.gov as NCT00005487.
Keywords: vitamin D, interstitial lung disease, CT scan, epidemiology, risk factors
Introduction
Interstitial lung disease (ILD) is a heterogeneous set of disorders characterized by alveolar injury, inflammation, and fibrosis. A number of risk factors for ILD have been identified, including age, smoking, genetic risk, occupational and environmental exposures, infection, and radiation (1–3), in addition to cases that are idiopathic. By the time patients are diagnosed with a fibrotic ILD such as idiopathic pulmonary fibrosis, the majority die within ∼5 y, and few treatment options are available. Therefore, the identification of additional modifiable risk factors is of utmost importance for disease prevention.
To understand the pathogenesis of ILD and its progression, there is a need to study subclinical or “early stage” ILD, defined as individuals having features of ILD based on imaging but who are either asymptomatic or who have symptoms that have not been attributed to ILD (4). Two phenotypes of subclinical ILD based on computed tomography (CT) have been used in previous studies as follows: First, high-attenuation areas (HAAs; i.e., increased areas of lung attenuation seen on CT scans) are a quantitatively measured phenotype of subclinical ILD. HAAs have been shown to be associated with biomarker evidence of lung injury, reduced lung function, and increased risk of visibly identifiable subclinical ILD on CT scans at 10-y follow-up (5, 6). HAAs are also associated with clinically meaningful outcomes, such as increased all-cause mortality and risk of hospitalizations from respiratory diseases (5, 7). Second, interstitial lung abnormalities (ILAs) seen on chest CT scans are a qualitative, visually assessed phenotype of subclinical ILD, associated with genetic markers of idiopathic pulmonary fibrosis (8), increased pulmonary symptoms (8), reduced pulmonary function (9, 10), and greater rates of all-cause mortality (11).
Previous mechanistic studies found that 1,25-dihydroxyvitamin D, the active form of vitamin D, has anti-inflammatory and antifibrotic properties (12–14). Previous studies have suggested that vitamin D deficiency, measured by serum or plasma 25-hydroxyvitamin D [25(OH)D] concentrations, may be a potential risk factor for diseases of the respiratory system (15). For example, patients with chronic obstructive pulmonary disease and asthma have a higher prevalence of 25(OH)D deficiency (15–18). Concentrations of serum or plasma 25(OH)D have also been associated with pulmonary function testing in a large consortium of studies conducted in general community populations (19). However, the association between vitamin D status with ILD specifically remains less certain. One previous cross-sectional analysis of 118 patients with ILD found that lower 25(OH)D concentrations were associated with worse lung function (17). However, to our knowledge, there are currently no large cross-sectional or longitudinal studies that have examined the relation between 25(OH)D concentrations and subclinical ILD.
Therefore, we examined the associations between vitamin D status and subclinical ILD, based on CT imaging, over 10 y in the Multi-Ethnic Study of Atherosclerosis (MESA), a community-based cohort of middle-aged and older adults. We hypothesized that 25(OH)D deficiency would be associated with subclinical ILD and its progression, independent of other ILD risk factors such as smoking.
Methods
Study population
The MESA is a multicenter prospective cohort study established to investigate the prevalence, correlates, and progression of subclinical cardiovascular disease (CVD) in individuals without clinical CVD at baseline (20). In 2000–2002, MESA recruited 6814 men and women aged 45–84 y from 6 US communities: Forsyth County, North Carolina; New York City, New York; Baltimore, Maryland; St. Paul, Minnesota; Chicago, Illinois; and Los Angeles, California. Participants were white (38%), black (28%), Hispanic (22%), or Chinese (12%). Exclusion criteria for MESA included a previous history of CVD, weight >136 kg, any impediment to long-term participation, and chest CT within the past year (20). After the baseline examination (exam 1), participants took part in ≤4 follow-up visits at exam 2 (2002–2004), exam 3 (2004–2005), exam 4 (2005–2007), and exam 5 (2010–2012).
All MESA participants had cardiac CT scans that included partial lung fields at exam 1. Some participants additionally had follow-up cardiac CT scans at exams 2–5, a full-lung CT scan at exam 5, or both (Figure 1). For our analyses, we excluded 341 participants with missing 25(OH)D concentrations, 2 with 25(OH)D concentrations >150 ng/mL (consistent with high-dose supplementation), and 167 who were missing relevant covariates for our primary analytical model, for a total sample size of 6304 participants at baseline (Figure 1). Among these included participants, the mean ± SD number of cardiac CT scans was 2.1 ± 1.0 and the numbers (percentages) of participants with a total of 1, 2, 3, and 4 CT scans were as follows: 781 (12.4%), 2723 (43.2%), 2218 (35.2%), and 581 (9.2%). Among the 6304 participants included in this sample, there were 2668 participants enrolled in the MESA Lung Study who underwent a full-lung CT scan at exam 5.
FIGURE 1.
Participant flow in the study. CT, computed tomography; MESA, Multi-Ethnic Study of Atherosclerosis; 25(OH)D, 25-hydroxyvitamin D.
The institutional review boards at all MESA study sites and the coordinating center approved the study protocols. All of the participants provided written informed consent at each study visit.
Laboratory assays
Serum was collected at the baseline exam after an 8- to 12-h overnight fast and stored at −80°C until being thawed in 2011 for 25(OH)D analysis. Serum 25-hydroxyergocalciferol [25(OH)-ergocalciferol] and 25-hydroxycholecalciferol [25(OH)-cholecalciferol] concentrations were measured by using LC-MS on a Quattro Micro mass spectrometer (Waters) (21). Interassay CVs were 8.5% for 25(OH)-cholecalciferol at a concentration of 24.8 ng/mL and 11.8% for 25(OH)-ergocalciferol at a concentration of 7.0 ng/mL (22). Calibration of serum 25(OH)D concentrations was verified according to the National Institutes of Standards and Technology (23). The total 25(OH)D concentration was determined by summing 25(OH)-ergocalciferol and 25(OH)-cholecalciferol concentrations. Serum 25(OH)D was adjusted for seasonal variation by using a cosinor approach, as has previously been described (22). To convert 25(OH)D to nanomoles per liter from nanograms per milliliter, multiply by 2.496. Total and HDL cholesterol were measured from ethylenediaminetetraacetic acid (EDTA) plasma by using the cholesterol oxidase method (Roche Diagnostics); HDL cholesterol was measured after precipitation of non-HDL cholesterol with magnesium/dextran. Serum high-sensitivity C-reactive protein (hs-CRP) concentrations were measured using the BN-II nephelometer (Dade Behring Inc.).
Covariates
Trained study staff collected information on demographic characteristics, comorbidities, medication use, behavioral variables, and physical examination measurements with the use of standardized protocols at each study examination (20). Covariates considered in this analysis included age, sex, race/ethnicity (self-classified as white, Chinese, black, or Hispanic), BMI (kg/m2), highest educational attainment (less than high school, high school, or college or higher), smoking status (never, former, or current, as well pack-years smoking), current alcohol usage, intentional physical activity (metabolic equivalent task minutes per week), diabetes mellitus (yes or no; defined as fasting glucose ≥126 mg/dL, self-reported physician's diagnosis, or diabetic medication use), total and HDL cholesterol (expressed as mg/dL; continuous), use of lipid-lowering medications (yes or no), systolic blood pressure (expressed as mm Hg), use of antihypertensive medications (yes or no), hs-CRP (expressed as mg/L), estimated glomerular filtration rate (mL × min−1 × 1.73 m−2) with the use of the Chronic Kidney Disease Epidemiology Collaboration equation (24), the CT scanner model, total lung volume imaged on CT scan (expressed as cm3), and percentage of emphysema [defined as the percentage of lung with less than –950 Hounsfield units (HU)].
CT definition of subclinical ILD
Cardiac CT scans were performed on multidetector CT or electron beam tomography scanners with the use of standardized protocols on all participants at baseline (25). By design, a randomly assigned half of participants received a follow-up cardiac CT scan at exam 2 and the other randomly assigned half at exam 3 (21). In addition, a random sample of participants underwent another cardiac CT scan at exam 4 (26). In addition, participants in the MESA Lung Study underwent a full-lung scan following the MESA Lung/Spiromics protocol in exam 5 using 64-slice scanners (GE Healthcare and Siemens Heathcare) (27). The percentage of participants who underwent a CT scan at each visit is shown in Figure 1.
As part of an ancillary study (the MESA Lung Study), quantitative lung-density measures were assessed from the lung fields of these cardiac CT scans, which imaged ∼65% of the lung volume from the carina to the lung bases (28). Briefly, 2 sequential scans on separate breath-holds were performed in succession at full inspiration on each participant. The scan with the higher air volume was used for analyses, except in cases of discordant quality, in which case the higher-quality scan was used. Image attenuation was assessed by using a modified version of the Pulmonary Analysis Software Suite at a single reading center by trained readers without knowledge of the participants’ 25(OH)D status. Lung-density measures obtained from cardiac CT scans have been shown to provide reproducible and valid results compared with full-lung CT scans (28). The methods for the assessment of HAAs from the cardiac CTs and assessment ILAs from the full-lung CT scans have been previously reported (5, 7, 29, 30).
Previous work in MESA considered subclinical parenchymal lung disease by HAAs defined by the lung volume (expressed as cm3) with a CT attenuation value between −600 and −250 HU, as previously described (5, 6). This range of CT lung attenuation includes ground-glass and reticular abnormalities and is low enough to clearly exclude more dense areas, such as complete atelectasis, consolidations, medium and large blood vessels, and pulmonary nodules, which are all more dense than water (HU of 0) (5).
ILAs were scored qualitatively (as present or absent) from the full-lung CT scans at exam 5 by trained readers if the following abnormalities were present in ≥5% of nondependent lung fields: ground-glass or reticular abnormalities, honey-combing, traction bronchiectasis, diffuse centrilobular nodularity, nonemphysematous cysts, or architectural distortion, as previously described (5, 11, 31).
Statistical analysis
Serum 25(OH)D concentrations were categorized into 3 groups [deficient (<20 ng/mL), intermediate (20 to <30 ng/mL), and replete (≥30 ng/mL)], as per the Endocrine Society guidelines (32). Mixed-effects models were used to assess the cross-sectional and longitudinal associations between baseline 25(OH)D concentrations and change in volume (expressed as cm3) of HAAs from exams 1 to 5 over a 10-y period, allowing for random variations in baseline values and annual change in HAAs among participants. Continuous HAA values were log-transformed and used as a dependent variable. To estimate the marginal mean values of HAAs over the follow-up across the 25(OH)D categories, we calculated the adjusted mean of log-transformed HAAs from fitted mixed model while averaging out the other covariates and fixing 25(OH)D concentrations. For estimating baseline HAA values, we additionally forced the time of follow-up to be 0. The estimated marginal mean log(HAA) was then exponentiated to reflect the geometric mean of the HAA.
We also performed a nonconcurrent cross-sectional analysis assessing the associations between 25(OH)D (measured once at exam 1) and the presence of ILAs (yes or no; measured once at exam 5). For this analysis, logistic regression was used to estimate ORs of ILAs associated with baseline 25(OH)D concentrations.
For all analyses, we analyzed 3 progressively adjusted models as shown in Supplemental Figure 1. Model 1 (a limited model of demographic characteristics and CT variables) was adjusted for time since baseline, age, sex, race/ethnicity, study site, scanner mode, percentage of emphysematous area, and total volume of imaged lung. Model 2 (our primary analytical model) was additionally adjusted for baseline education, alcohol usage, baseline and change in BMI, smoking status, pack-years of smoking, and physical activity. Model 3 [an exploratory model that included cardiovascular risk factors that may potentially mediate associations between 25(OH)D and vascular/lung injury] was included baseline log-transformed hs-CRP, baseline and change in systolic blood pressure, use of antihypertensive medications, total and HDL cholesterol, use of lipid-lowering medications, diabetes, and estimated glomerular filtration rate.
To assess for nonlinear relations between 25(OH)D concentration and trajectories of HAAs, we also used restricted cubic splines with the use of follow-up time with knots at the 5th, 45th, 75th, and 95th percentiles (values of 0, 1.3, 2.7, and 9.8 y) of its distribution adjusted for model 2 covariates (our primary model) and stratified by 25(OH)D concentrations at baseline. Effect modifications by subgroups including age, sex, race/ethnicity, smoking status, BMI, and diabetes were also explored by including an interaction of subgroups and 25(OH)D concentrations.
Two-sided P values <0.05 were considered significant. All of the statistical analyses were conducted with the use of Stata version 14 (StataCorp).
Results
Baseline characteristics of the 6304 participants by 25(OH)D categories are shown in Table 1. The mean ± SD characteristics of the cohort were 62.2 ± 10 y for age, 25.8 ± 10.9 ng/mL for 25(OH)D concentration, and 28.3 ± 5.4 for BMI. Our study sample included 53% women and 39% white, 27% black, 22% Hispanic, and 12% Chinese race/ethnicities. Among our sample, 33% had replete (≥30 ng/mL), 35% intermediate (20 to <30 ng/mL), and 32% deficient (<20 ng/mL) 25(OH)D concentrations. On average, participants with 25(OH)D deficiency were more likely to be younger, of black race/ethnicity, and current smokers. On average, 25(OH)D-deficient individuals had lower amounts of total intentional exercise, greater use of antihypertensive medications, had higher BMIs, higher prevalence of diabetes, higher hs-CRP, and a slightly higher volume of HAAs (all P < 0.05).
TABLE 1.
Baseline characteristics of participants by 25(OH)D categories: the MESA, 2000–20021
| 25(OH)D concentration2 | ||||
|---|---|---|---|---|
| Replete (≥30 ng/mL; range: 30–149.9 ng/mL) | Intermediate (20 to <30 ng/mL) | Deficient (<20 ng/mL; range: 1.4–19.9 ng/mL) | P-trend | |
| Participants, n | 2064 | 2189 | 2051 | |
| Age, y | 63.4 ± 10 | 62.4 ± 10 | 60.7 ± 10 | <0.001 |
| Men, n (%) | 926 (45) | 1098 (50) | 918 (45) | <0.001 |
| Race/ethnicity, n (%) | <0.001 | |||
| White | 1193 (58) | 842 (39) | 413 (20) | |
| Chinese-American | 267 (13) | 353 (16) | 159 (8) | |
| Black | 215 (10) | 454 (21) | 1027 (50) | |
| Hispanic | 389 (19) | 540 (25) | 452 (22) | |
| Education, n (%) | <0.001 | |||
| Less than high school | 332 (16) | 423 (19) | 385 (19) | |
| High school, technical school, or associate degree | 918 (45) | 981 (45) | 1010 (49) | |
| College, graduate, or professional school | 814 (39) | 785 (36) | 656 (32) | |
| Smoking, n (%) | <0.001 | |||
| Never | 1035 (50) | 1179 (54) | 1012 (49) | |
| Former | 823 (40) | 766 (35) | 686 (33) | |
| Current | 206 (10) | 244 (11) | 353 (17) | |
| Pack-years of smoking | 11.8 ± 22 | 10.2 ± 20 | 11.5 ± 21 | 0.02 |
| Intentional exercise,3 MET-min/wk | 1138 (330–2438) | 788 (105–1875) | 630 (0–1680) | <0.001 |
| BMI, kg/m2 | 27 ± 5 | 28 ± 5 | 30 ± 6 | <0.001 |
| Systolic blood pressure, mm Hg | 125 ± 21 | 126 ± 22 | 129 ± 22 | <0.001 |
| Diastolic blood pressure, mm Hg | 71 ± 10 | 72 ± 10 | 73 ± 11 | <0.001 |
| Plasma total cholesterol, mg/dL | 196 ± 34 | 195 ± 36 | 192 ± 37 | 0.01 |
| Plasma HDL cholesterol, mg/dL | 54 ± 16 | 50 ± 14 | 50 ± 14 | <0.001 |
| eGFR, mL ⋅ min−1 ⋅ 1.73 m−2 | 74 ± 15 | 78 ± 16 | 82 ± 17 | <0.001 |
| Serum hs-CRP,3 mg/L | 1.6 (0.7–3.9) | 1.8 (0.8–3.8) | 2.4 (1.0–5.0) | <0.001 |
| Use of antihypertension medication, n (%) | 705 (34) | 809 (37) | 803 (39) | 0.003 |
| Use of lipid-lowering medication, n (%) | 363 (18) | 376 (17) | 297 (15) | 0.01 |
| Diabetes, n (%) | 162 (8) | 310 (14) | 308 (15) | <0.001 |
| Volume of HAAs,3 cm3 | 116 (96–139) | 120 (101–143) | 121 (103 –146) | <0.001 |
| Presence of ILA,4n (%) | 79 (9.1) | 109 (11.7) | 84 (9.7) | 0.17 |
1Values are means ± SDs or n (%) unless otherwise noted, n = 6304. eGFR, estimated glomerular filtration rate; HAA, high-attenuation area; hs-CRP, high-sensitivity C-reactive protein; ILA, interstitial lung abnormality; MESA, Multi-Ethnic Study of Atherosclerosis; MET, metabolic equivalent task; 25(OH)D, 25-hydroxyvitamin D.
2To convert 25(OH)D concentrations to nanomoles per liter from nanograms per milliliter, multiply by 2.496.
3Values are medians (IQRs).
4ILA was measured at exam 5 (n = 2667).
Table 2 shows the cross-sectional associations between 25(OH)D concentrations and volume of HAAs (expressed as cm3) in lung fields on cardiac CTs at exam 1. Compared with those with replete vitamin D (≥30 ng/mL), those with intermediate (20 to <30 ng/mL) and deficient (<20 ng/mL) 25(OH)D concentrations had greater average HAA volumes (expressed as cm3) of 2.2 (95% CI: 0.6, 3.7; P = 0.007) and 2.7 (95% CI: 0.9, 4.5; P = 0.003), respectively, after adjustment for demographic characteristics and lifestyle factors in our primary model (model 2). This was attenuated after further adjustment for CVD and inflammatory factors that might mediate associations between vitamin D deficiency and lung injury (model 3).
TABLE 2.
Cross-sectional analysis: adjusted average volume of HAAs in lung fields of cardiac CTs by serum 25(OH)D categories at MESA exam 1 (2000–2002)1
| 25(OH)D concentrations2 | Model 1 | Model 2 | Model 3 |
|---|---|---|---|
| Adjusted mean HAA volume, cm3 | |||
| Replete (≥30 ng/mL) | 123.2 (121.9, 124.5) | 125.3 (124.0, 126.6) | 125.7 (124.4, 127.1) |
| Intermediate (20 to <30 ng/mL) | 127.2 (125.9, 128.5) | 127.4 (126.2, 128.7) | 126.9 (125.7, 128.2) |
| Deficient (<20 ng/mL) | 130.4 (129.0, 131.8) | 128.0 (126.6, 129.4) | 127.3 (125.9, 128.7) |
| Difference in adjusted mean HAA volume compared with reference, cm3 | |||
| Replete (≥30 ng/mL) | Reference | Reference | Reference |
| Change between intermediate and replete | 4.0 (2.4, 5.6) | 2.2 (0.6, 3.7) | 1.2 (–0.4, 2.8) |
| P value for change between intermediate and replete | <0.001 | 0.007 | 0.13 |
| Change between deficient and replete | 7.2 (5.4, 9.0) | 2.7 (0.9, 4.5) | 1.6 (–0.2, 3.3) |
| P value for change between deficient and replete | <0.001 | 0.003 | 0.08 |
1Values are means (95% CIs) unless otherwise indicated. Results were derived from a mixed-effects model constraining time since baseline and change in covariates at 0 and other baseline covariates at their means. Model 1 (demographics and CT variables) was adjusted for baseline age, sex, race/ethnicity, study site and scanner mode, total volume of imaged lung, percentage of emphysema, and time since baseline. Model 2 (plus confounding lifestyle variables) was adjusted as in model 1 plus baseline BMI, educational attainment, smoking status, pack-years of smoking, alcohol usage, and physical activity. Model 3 (plus cardiovascular risk factors) was adjusted as in model 2 plus baseline systolic blood pressure, use of antihypertensive medications, total and HDL cholesterol, use of lipid-lowering medications, diabetes, CRP, and eGFR categories. CRP, C-reactive protein; CT, computed tomography; eGFR, estimated glomerular filtration rate; HAA, high-attenuation area; MESA, Multi-Ethnic Study of Atherosclerosis; 25(OH)D, 25-hydroxyvitamin D.
2To convert 25(OH)D concentrations to nanomoles per liter from nanograms per milliliter, multiply by 2.496.
Table 3 shows the longitudinal associations between baseline 25(OH)D concentrations and change in volume of HAA (expressed as cm3) in lung fields on cardiac CT in ≤10 y of follow-up (median follow-up of 4.3 y). Compared with those with replete 25(OH)D concentrations (≥30 ng/mL), those with intermediate (20 to <30 ng/mL) and deficient (<20 ng/mL) 25(OH)D concentrations had greater average increase in HAA volumes of 2.1 (95% CI: 0.6, 3.7; P = 0.007) and 2.7 (95% CI: 0.9, 4.4; P = 0.003), respectively, during the follow-up period after adjustment for demographic characteristics and lifestyle factors in our primary model (model 2). This was again attenuated after further adjustment for CVD and inflammatory factors that might mediate associations between vitamin D deficiency and lung injury (model 3).
TABLE 3.
Longitudinal analysis: adjusted average volume of HAAs in lung fields across cardiac CT scans obtained from MESA exams 1–5 (2000–2013), by serum 25(OH)D categories at exam 11
| 25(OH)D concentrations 2 | Participants/visits, n/n | Model 1 | Model 2 | Model 3 |
|---|---|---|---|---|
| Adjusted mean HAA volume, cm3 | ||||
| Replete (≥30 ng/mL) | 2064/5037 | 122.1 (121.0, 123.3) | 124.0 (122.9, 125.2) | 124.7 (123.5, 125.8) |
| Intermediate (20 to <30 ng/mL) | 2189/5313 | 126.1 (124.9, 127.2) | 126.1 (125.1, 127.2) | 125.9 (124.8, 127.0) |
| Deficient (<20 ng/mL) | 2051/4856 | 129.3 (128.0, 130.6) | 126.7 (125.5, 127.9) | 126.2 (125.1, 127.4) |
| Difference in adjusted mean HAA volume compared with | ||||
| reference, cm3 | ||||
| Replete (≥30 ng/mL) | — | Reference | Reference | Reference |
| Change between intermediate and replete | — | 4.0 (2.3, 5.6) | 2.1 (0.6, 3.7) | 1.2 (-0.4, 2.8) |
| P value for change between intermediate and replete | — | <0.001 | 0.007 | 0.13 |
| Change between deficient and replete | — | 7.1 (5.3, 9.0) | 2.7 (0.9, 4.4) | 1.6 (-0.2, 3.3) |
| P value for change between deficient and replete | — | <0.001 | 0.003 | 0.08 |
1Values are means (95% CIs) unless otherwise indicated. Results were derived from a mixed-effects model constraining other covariates at their means. The numbers (%) of participants with a total of 1, 2, 3, and 4 CT scans are as follows: 781 (12.4%), 2723 (43.2%), 2218 (35.2%), and 581 (9.2%). Model 1 (demographics and CT variables) adjusted for baseline age, sex, race/ethnicity, study site and scanner mode, total volume of imaged lung, percentage of emphysema, and time since baseline. Model 2 (plus confounding lifestyle variables) adjusted as in model 1 plus baseline BMI, educational attainment, smoking status, pack-years of smoking, alcohol usage, and physical activity. Model 3 (plus cardiovascular risk factors) adjusted as in model 2 plus baseline systolic blood pressure, use of antihypertensive medications, total and HDL cholesterol, use of lipid-lowering medications, diabetes, CRP, and eGFR categories. CRP, C-reactive protein; CT, computed tomography; eGFR, estimated glomerular filtration rate; HAA, high-attenuation area; MESA, Multi-Ethnic Study of Atherosclerosis; 25(OH)D, 25-hydroxyvitamin D.
2To convert 25(OH)D concentrations to nanomoles per liter from nanograms per milliliter, multiply by 2.496.
Figure 2 shows the multivariable-adjusted trajectory in HAAs from exam 1 to 5 associated with baseline 25(OH)D concentrations (deficient and replete groups are shown). The spline model generally shows greater HAA volume seen with deficient 25(OH)D concentrations compared with replete concentrations throughout the follow-up time period.
FIGURE 2.
Trajectories of HAA volume in lung fields in MESA cardiac CT scans over follow-up from exam 1 to 5 associated with serum 25(OH)D concentrations at baseline. Curves represent the adjusted trajectories of log-transformed HAAs based on restricted cubic splines for follow-up time from exam 1 (time 0), with knots at the 5th, 45th, 75th, and 95th percentiles (values of 0, 1.3, 2.7, and 9.8 y, respectively) of their sample distributions. Models were stratified by 25(OH)D concentrations, with only the categories of <20 and ≥30 ng/mL shown for space considerations. The solid lines and light-gray shading represent the mean (95% CI) HAA trajectory for deficient (<20 ng/mL) 25(OH)D and the dashed lines and dark-grey shading represent the trajectory for the replete (≥30 ng/mL) 25(OH)D concentrations. Models were adjusted for age, sex, race/ethnicity, study site and scanner mode, total volume of imaged lung, emphysema, time since baseline, BMI, educational attainment, smoking status, pack-years of smoking, alcohol usage, and physical activity. To convert 25(OH)D concentrations to nanomoles per liter from nanograms per milliliter, multiply by 2.496. CT, computed tomography; HAA, high-attenuation area; MESA, Multi-Ethnic Study of Atherosclerosis; 25(OH)D, 25-hydroxyvitamin D.
Longitudinal associations between 25(OH)D concentrations (deficient compared with replete) and HAA volume change by subgroups are presented in Supplemental Figure 2. There were no significant interactions by age, sex, race/ethnicity, smoking status, BMI, or diabetes (P-interaction > 0.05 for all).
Table 4 shows the adjusted association between 25(OH)D concentrations at baseline and the presence of ILAs among those participants who underwent a full-lung CT scan at exam 5 (n = 2668). Compared with those with replete 25(OH)D concentrations (≥30 ng/mL), those with intermediate (20 to <30 ng/mL) and deficient (<20 ng/mL) 25(OH)D concentrations had greater odds of ILAs [OR (95% CI): 1.6 (1.1, 2.2) (P = 0.006) and 1.5 (1.1, 2.2) (P = 0.03)], respectively, after adjustment for demographic characteristics and lifestyle factors in our primary model (model 2). These associations remained similar and significant after further adjustment for potentially mediating CVD and inflammatory risk factors (model 3).
TABLE 4.
Cross-sectional associations between serum 25(OH)D categories at exam 1 and presence of ILAs on MESA full-lung CTs at exam 51
| ORs (95% CIs) | ||||
|---|---|---|---|---|
| 25(OH)D concentrations2 | Participants (n = 2668) | Model 1 | Model 2 | Model 3 |
| Replete (≥30 ng/mL) | 870 | Reference (1) | Reference (1) | Reference (1) |
| Intermediate (20 to <30 ng/mL) | 935 | 1.6 (1.2, 2.2) | 1.6 (1.1, 2.2) | 1.6 (1.1, 2.2) |
| P | 0.004 | 0.006 | 0.007 | |
| Deficient (<20 ng/mL) | 863 | 1.6 (1.1, 2.3) | 1.5 (1.1, 2.2) | 1.5 (1.1, 2.2) |
| P | 0.02 | 0.03 | 0.03 | |
1Results were derived from logistic regression models. Model 1 (demographics and CT variables) was adjusted for baseline age, sex, race/ethnicity, study site and scanner mode, total volume of imaged lung, percentage of emphysema, and time since baseline. Model 2 (plus confounding lifestyle variables) was adjusted as in model 1 plus for baseline BMI, educational attainment, smoking status, pack-years of smoking, alcohol usage, and physical activity. Model 3 (plus cardiovascular risk factors) was adjusted as in model 2 plus for baseline systolic blood pressure, use of antihypertensive medications, total and HDL cholesterol, use of lipid-lowering medications, diabetes, CRP, and eGFR categories. CRP, C-reactive protein; CT, computed tomography; eGFR, estimated glomerular filtration rate; ILA, interstitial lung abnormality; MESA, Multi-Ethnic Study of Atherosclerosis; 25(OH)D, 25-hydroxyvitamin D.
2To convert 25(OH)D concentrations to nanomoles per liter from nanograms per milliliter, multiply by 2.496.
Discussion
In a large multiracial, community-based cohort of middle-aged and older adults followed for ≤10 y, we found that deficient and intermediate 25(OH)D concentrations, compared with replete concentrations, were cross-sectionally and longitudinally associated with evidence of subclinical ILD on the basis of increased areas of high lung attenuation and the presence of ILAs. This is important considering that a recent study showed that increased HAAs found incidentally by cardiac CT scans were associated with clinically relevant events such as future ILD-related hospitalization and mortality (7). Although our study is observational and can only describe associations, our findings lend support to the potential role of adequate vitamin D concentrations for the prevention of early ILD, which needs to be confirmed in future interventional studies.
Our study results are consistent with previously published data that showed adverse associations between 25(OH)D deficiency and various other respiratory disorders, particularly chronic obstructive pulmonary disease and asthma. Notably, data from NHANES-III showed that higher 25(OH)D concentrations were associated with greater lung function (forced expiratory volume in first second) in a cross-sectional study in 14,000 participants (18). Previous work in MESA, published as part of a large consortium of population-based studies, also found that concentrations of serum 25(OH)D were associated with pulmonary function testing (19). In addition, a large Cochrane meta-analysis found that vitamin D supplementation reduced the risk of severe asthma exacerbation in those with mild-to-moderate asthma and in those with low baseline 25(OH)D status (33). Another meta-analysis found that vitamin D supplementation might reduce the risk of developing acute respiratory tract infections (34).
However, little was previously known about the association of 25(OH)D with ILD specifically. Patients with idiopathic pulmonary fibrosis have higher rates of mortality during the winter season, which coincides with the time of year with the lowest 25(OH)D concentrations (35). One previous analysis that evaluated the relation between low 25(OH)D concentrations and lung function among patients with ILD was cross-sectional (17). In this study in patients with ILD (67 with connective tissue disease–associated ILD and 51 with other ILD forms) there was a high prevalence of 25(OH)D deficiency, with 38% having 25(OH)D concentrations <20 ng/mL (17). We now show that lower concentrations of 25(OH)D were associated with CT markers of subclinical ILD, including greater HAA volume at baseline, greater progression over follow-up (≤10 y), and increased prevalence of ILAs. These associations were independent of possible lifestyle confounders such as smoking and exercise, which are associated with both lung disease and low 25(OH)D concentrations (36).
ILD encompasses a spectrum of multiple diagnoses. Currently, the possible mechanisms linking vitamin D deficiency to ILD are not well understood. There have been a number of animal studies that suggest that activated vitamin D may be important for anti-inflammatory effects such as reducing the proliferation of hematopoietic stem cells (37, 38). In addition, vitamin D may be important in downregulating metalloproteinases (39) and reducing proliferation of fibroblasts (14), which are both processes involved in the pathogenesis of idiopathic pulmonary fibrosis. Further experimental work is needed to establish mechanisms.
Limitations and strengths
Our study has many strengths in that, to our knowledge, this is the largest community-based study to date that has analyzed the relation between 25(OH)D concentrations and subclinical ILD progression over long-term follow-up with adjustments for many potentially confounding variables. It is also important to note that we found consistent associations between 25(OH)D deficiency and higher cross-sectional HAA, HAA progression over time, and the presence of ILAs. The longitudinal design reduces concerns for reverse causation.
Nonetheless, our findings should be considered in the context of several limitations. First, we had only a single measure of 25(OH)D available at baseline, which may not reflect long-term vitamin D status. Second, the cardiac CT scans included only partial lung fields (∼65% of lung volume), which may have missed additional HAAs given that ILD often tends to be present at the bases. However, all of the results were adjusted for lung volume imaged to account for differences in the imaged lung volumes. Third, not all of the participants had the same follow-up periods between the CT scans, but we accounted for time between scans in our analyses. Fourth, longitudinal analyses are prone to survival bias due to differential follow-up. Fifth, the study did not adjust for environmental exposures, which can contribute to ILD development. Sixth, participants with clinical ILD were not specifically excluded from MESA; however, these were community-dwelling adults selected without regard to smoking or respiratory diseases. In addition, MESA's exclusion of any adults with a chest CT scan within the past year also eliminates many who might have chronic lung diseases. Therefore, in this context, we feel that the presence of HAAs and ILAs represents subclinical interstitial abnormalities (7). Finally, although we adjusted for numerous potentially confounding covariates, residual confounding might still explain the associations seen.
Conclusions
In conclusion, we found an independent, but modest, association between lower 25(OH)D concentrations and increased HAA volume in the lung fields of cardiac CTs, which is a marker for subclinical ILD, in ≤10 y of follow-up. We also found that lower 25(OH)D concentrations were associated with prevalent ILAs on full-lung CT scans 10 y later. Our study is consistent with previous published studies that showed a relation between 25(OH)D deficiency and other types of respiratory disorders. Because 25(OH)D concentrations are relatively easy to monitor and easy to treat with supplementation modest sunlight exposure, or both, our work calls attention to this potentially modifiable risk factor for the early stages of ILD. However, further interventional studies are needed. The Lung VITamin D and OmegA-3 TriaL (VITAL) is a randomized clinical trial that currently is underway to assess the effect of vitamin D supplementation on chronic respiratory disease exacerbations (40). However, further studies are needed to examine whether vitamin D repletion can prevent or slow the progression of ILD specifically.
Supplementary Material
Acknowledgments
We thank the staff of the MESA study for their important contributions. A full list of participating MESA investigators and institutions can be found at http://www.mesa-nhlbi.org. The authors’ responsibilities were as follows—SMK and EDM: designed the research, wrote the manuscript, and take final responsibility for all content; DZ: performed statistical analysis; IHdB and BRK: provided essential materials (vitamin D measurement); RGB and DJL: provided essential materials (CT data); DZ, AJP, PLL, EG, SMK, RGB, IHdB, BRK, and DJL: edited the manuscript for critical scientific input; and all authors: read and approved the final manuscript.
Notes
An abstract from this work was presented at the 2016 American Heart Association Scientific Sessions in New Orleans, Louisiana. The full manuscript has not been previously published and is not under consideration elsewhere.
EDM was supported by NIH/National Institute of Neurological Disorders and Stroke grant R01NS072243. EDM and DZ were also supported by the Blumenthal Scholars Fund for Preventive Cardiology. Funding for 25-hydroxyvitamin D measurements was provided by R01HL096875 from the NIH (to IHdB and BRK). The Multi-Ethnic Study of Atherosclerosis (MESA) study was funded by contracts HHSN268201500003I, N01-HC-95159, N01-HC-95160, N01-HC-95161, N01-HC-95162, N01-HC-95163, N01-HC-95164, N01-HC-95165, N01-HC-95166, N01-HC-95167, N01-HC-95168, and N01-HC-95169 from the National Heart, Lung, and Blood Institute, and by grants UL1-TR-000040, UL1-TR-001079, and UL1-TR-001420 from the National Center for Advancing Translational Sciences. The MESA Lung and Lung Fibrosis funding was funded by R01-HL077612, R01-HL093081, RC1-HL100543, R01-HL-103676, T32-HL-105323, and K24-HL-131937. The MESA Air ancillary, which supported the computed tomography scans at exam 5, was developed under a Science to Achieve Results (STAR) research assistance agreement, no. RD831697, awarded by the US Environmental Protection Agency (EPA). It has not been formally reviewed by the EPA.
Author disclosures: SMK, DZ, AJP, PLL, EG, SMK, and BRK, no conflicts of interest. EDM received an honorarium from Siemens Healthcare Diagnostics in 2016, unrelated to this work. IHdB has the following disclosures: Ironwood Pharma and Boehringer-Ingelheim (consulting) and Medtronic and Abbott (equipment and supplies donated to institution for research), all of which are unrelated to this work. DJL has been a consultant for Roche, Veracyte, Philips Respironics, Fibrogen, Global Blood Therapeutics, Sanofi Genzyme, and Immuneworks. RGB has funding from the COPD Foundation.
The views expressed in this document are solely those of the authors, and the EPA does not endorse any products or commercial services mentioned in this publication.
Supplemental Figures 1 and 2 are available from the “Supplementary data” link in the online posting of the article and from the same link in the online table of contents at https://academic.oup.com/jn/.
Abbreviations used:
- CT
computed tomography
- CVD
cardiovascular disease
- HAA
high-attenuation area
- hs-CRP
high-sensitivity C-reactive Protein
- HU
Hounsfield units
- ILA
interstitial lung abnormality
- ILD
interstitial lung disease
- MESA
Multi-Ethnic Study of Atherosclerosis
- 25(OH)D
25-hydroxyvitamin D
- 25(OH)-cholecalciferol
25-hydroxycholecalciferol
- 25(OH)-ergocalciferol
25-hydroxyergocalciferol
References
- 1. Wallis A, Spinks K. The diagnosis and management of interstitial lung diseases. BMJ 2015;350:h2072. [DOI] [PubMed] [Google Scholar]
- 2. Rosas IO, Dellaripa PF, Lederer DJ, Khanna D, Young LR, Martinez FJ. Interstitial lung disease: NHLBI Workshop on the Primary Prevention of Chronic Lung Diseases. Ann Am Thorac Soc 2014;11(Suppl 3):S169–77. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Azadeh N, Limper AH, Carmona EM, Ryu JH. The role of infection in interstitial lung diseases: a review. Chest 2017;152(4):842–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Doyle TJ, Hunninghake GM, Rosas IO. Subclinical interstitial lung disease: why you should care. Am J Respir Crit Care Med 2012;185(11):1147–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Podolanczuk AJ, Oelsner EC, Barr RG, Hoffman EA, Armstrong HF, Austin JH, Basner RC, Bartels MN, Christie JD, Enright PL et al. High attenuation areas on chest computed tomography in community-dwelling adults: the MESA study. Eur Respir J 2016;48(5):1442–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Lederer DJ, Enright PL, Kawut SM, Hoffman EA, Hunninghake G, van Beek EJ, Austin JH, Jiang R, Lovasi GS, Barr RG. Cigarette smoking is associated with subclinical parenchymal lung disease: the Multi-Ethnic Study of Atherosclerosis (MESA)–Lung Study. Am J Respir Crit Care Med 2009;180(5):407–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Podolanczuk AJ, Oelsner EC, Barr RG, Bernstein EJ, Hoffman EA, Easthausen IJ, Hinckley Stukovsky K, RoyChoudhury A, Michos ED, Raghu G et al. High attenuation areas on chest CT and clinical respiratory outcomes in community-dwelling adults. Am J Respir Crit Care Med 2017;196(11):1434–42. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Hunninghake GM, Hatabu H, Okajima Y, Gao W, Dupuis J, Latourelle JC, Nishino M, Araki T, Zazueta OE, Kurugol S et al. MUC5B promoter polymorphism and interstitial lung abnormalities. N Engl J Med 2013;368(23):2192–200. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Araki T, Putman RK, Hatabu H, Gao W, Dupuis J, Latourelle JC, Nishino M, Zazueta OE, Kurugol S, Ross JC et al. Development and progression of interstitial lung abnormalities in the Framingham Heart Study. Am J Respir Crit Care Med 2016;194(12):1514–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Washko GR, Hunninghake GM, Fernandez IE, Nishino M, Okajima Y, Yamashiro T, Ross JC, Estepar RS, Lynch DA, Brehm JM et al. Lung volumes and emphysema in smokers with interstitial lung abnormalities. N Engl J Med 2011;364(10):897–906. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Putman RK, Hatabu H, Araki T, Gudmundsson G, Gao W, Nishino M, Okajima Y, Dupuis J, Latourelle JC, Cho MH et al. Association between interstitial lung abnormalities and all-cause mortality. JAMA 2016;315(7):672–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Calton EK, Keane KN, Newsholme P, Soares MJ. The impact of vitamin D levels on inflammatory status: a systematic review of immune cell studies. PLoS One 2015;10(11):e0141770. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Finklea JD, Grossmann RE, Tangpricha V. Vitamin D and chronic lung disease: a review of molecular mechanisms and clinical studies. Adv Nutr 2011;2(3):244–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Ramirez AM, Wongtrakool C, Welch T, Steinmeyer A, Zugel U, Roman J. Vitamin D inhibition of pro-fibrotic effects of transforming growth factor beta1 in lung fibroblasts and epithelial cells. J Steroid Biochem Mol Biol 2010;118(3):142–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Gilbert CR, Arum SM, Smith CM. Vitamin D deficiency and chronic lung disease. Can Respir J 2009;16(3):75–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Foong RE, Zosky GR. Vitamin D deficiency and the lung: disease initiator or disease modifier? Nutrients 2013;5(8):2880–900. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Hagaman JT, Panos RJ, McCormack FX, Thakar CV, Wikenheiser-Brokamp KA, Shipley RT, Kinder BW. Vitamin D deficiency and reduced lung function in connective tissue-associated interstitial lung diseases. Chest 2011;139(2):353–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Black PN, Scragg R. Relationship between serum 25-hydroxyvitamin d and pulmonary function in the third national health and nutrition examination survey. Chest 2005;128(6):3792–8. [DOI] [PubMed] [Google Scholar]
- 19. Xu J, Bartz TM, Chittoor G, Eiriksdottir G, Manichaikul AW, Sun F, Terzikhan N, Zhou X, Booth SL, Brusselle GG et al. Large meta-analysis provides eviden ce for an association of serum vitamin D with pulmonary function. Am J Respir Crit Care Med 2016;193[abstract A2776]. [Google Scholar]
- 20. Bild DE, Bluemke DA, Burke GL, Detrano R, Diez Roux AV, Folsom AR, Greenland P, Jacob DR Jr., Kronmal R, Liu K et al. Multi-Ethnic Study of Atherosclerosis: objectives and design. Am J Epidemiol 2002;156(9):871–81. [DOI] [PubMed] [Google Scholar]
- 21. Tibuakuu M, Zhao D, de Boer IH, Guallar E, Bortnick AE, Lutsey PL, Budoff MJ, Kizer JR, Kestenbaum BR, Michos ED. Relation of serum vitamin D to risk of mitral annular and aortic valve calcium (from the Multi-Ethnic Study of Atherosclerosis). Am J Cardiol 2017;120(3):473–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Sachs MC, Shoben A, Levin GP, Robinson-Cohen C, Hoofnagle AN, Swords-Jenny N, Ix JH, Budoff M, Lutsey PL, Siscovick DS et al. Estimating mean annual 25-hydroxyvitamin D concentrations from single measurements: the Multi-Ethnic Study of Atherosclerosis. Am J Clin Nutr 2013;97(6):1243–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Phinney KW, Bedner M, Tai SS, Vamathevan VV, Sander LC, Sharpless KE, Wise SA, Yen JH, Schleicher RL, Chaudhary-Webb M et al. Development and certification of a standard reference material for vitamin D metabolites in human serum. Anal Chem 2012;84(2):956–62. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D; Modification of Diet in Renal Disease Study Group. A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Ann Intern Med 1999;130(6):461–70. [DOI] [PubMed] [Google Scholar]
- 25. Carr JJ, Nelson JC, Wong ND, McNitt-Gray M, Arad Y, Jacobs DR Jr., Sidney S, Bild DE, Williams OD, Detrano RC. Calcified coronary artery plaque measurement with cardiac CT in population-based studies: standardized protocol of Multi-Ethnic Study of Atherosclerosis (MESA) and Coronary Artery Risk Development in Young Adults (CARDIA) study. Radiology 2005;234(1):35–43. [DOI] [PubMed] [Google Scholar]
- 26. Bild DE, McClelland R, Kaufman JD, Blumenthal R, Burke GL, Carr JJ, Post WS, Register TC, Shea S, Szklo M. Ten-year trends in coronary calcification in individuals without clinical cardiovascular disease in the Multi-Ethnic Study of Atherosclerosis. PLoS One 2014;9(4):e94916. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Sieren JP, Newell JD Jr., Barr RG, Bleecker ER, Burnette N, Carretta EE, Couper D, Goldin J, Guo J, Han MK et al. Spiromics protocol for multicenter quantitative computed tomography to phenotype the lungs. Am J Respir Crit Care Med 2016;194(7):794–806. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Hoffman EA, Jiang R, Baumhauer H, Brooks MA, Carr JJ, Detrano R, Reinhardt J, Rodriguez J, Stukovsky K, Wong ND et al. Reproducibility and validity of lung density measures from cardiac CT scans—the Multi-Ethnic Study of Atherosclerosis (MESA) Lung Study. Acad Radiol 2009;16(6):689–99. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Podolanczuk AJ, Raghu G, Tsai MY, Kawut SM, Peterson E, Sonti R, Rabinowitz D, Johnson C, Barr RG, Hinckley Stukovsky K et al. Cholesterol, lipoproteins and subclinical interstitial lung disease: the MESA study. Thorax 2017;72(5):472–4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Sack CS, Doney BC, Podolanczuk AJ, Hooper LG, Seixas NS, Hoffman EA, Kawut SM, Vedal S, Raghu G, Barr RG et al. Occupational exposures and subclinical interstitial lung disease: the MESA (Multi-Ethnic Study of Atherosclerosis) Air and Lung Studies. Am J Respir Crit Care Med 2017;196(8):1031–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Bernstein EJ, Barr RG, Austin JHM, Kawut SM, Raghu G, Sell JL, Hoffman EA, Newell JD Jr., Watts JR Jr., Nath PH et al. Rheumatoid arthritis-associated autoantibodies and subclinical interstitial lung disease: the Multi-Ethnic Study of Atherosclerosis. Thorax 2016;71(12):1082–90. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Holick MF, Binkley NC, Bischoff-Ferrari HA, Gordon CM, Hanley DA, Heaney RP, Murad MH, Weaver CM, Endocrine S. Evaluation, treatment, and prevention of vitamin D deficiency: an Endocrine Society clinical practice guideline. J Clin Endocrinol Metab 2011;96(7):1911–30. [DOI] [PubMed] [Google Scholar]
- 33. Martineau AR, Cates CJ, Urashima M, Jensen M, Griffiths AP, Nurmatov U, Sheikh A, Griffiths CJ. Vitamin D for the management of asthma. Cochrane Database Syst Rev 2016;9:CD011511. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Martineau AR, Jolliffe DA, Hooper RL, Greenberg L, Aloia JF, Bergman P, Dubnov-Raz G, Esposito S, Ganmaa D, Ginde AA et al. Vitamin D supplementation to prevent acute respiratory tract infections: systematic review and meta-analysis of individual participant data. BMJ 2017;356:i6583. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Olson AL, Swigris JJ, Raghu G, Brown KK. Seasonal variation: mortality from pulmonary fibrosis is greatest in the winter. Chest 2009;136(1):16–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Chin K, Zhao D, Tibuakuu M, Martin SS, Ndumele CE, Florido R, Windham BG, Guallar E, Lutsey PL, Michos ED. Physical activity, vitamin D, and incident atherosclerotic cardiovascular disease in whites and blacks: the ARIC study. J Clin Endocrinol Metab 2017;102(4):1227–36. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Luong QT, Koeffler HP. Vitamin D compounds in leukemia. J Steroid Biochem Mol Biol 2005;97(1-2):195–202. [DOI] [PubMed] [Google Scholar]
- 38. O'Kelly J, Hisatake J, Hisatake Y, Bishop J, Norman A, Koeffler HP. Normal myelopoiesis but abnormal T lymphocyte responses in vitamin D receptor knockout mice. J Clin Invest 2002;109(8):1091–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Kim SH, Baek MS, Yoon DS, Park JS, Yoon BW, Oh BS, Park J, Kim HJ. Vitamin D inhibits expression and activity of matrix metalloproteinase in human lung fibroblasts (HFL-1) cells. Tuberc Respir Dis (Seoul) 2014;77(2):73–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Gold DR, Litonjua AA, Carey VJ, Manson JE, Buring JE, Lee IM, Gordon D, Walter J, Friedenberg G, Hankinson JL et al. Lung VITAL: rationale, design, and baseline characteristics of an ancillary study evaluating the effects of vitamin D and/or marine omega-3 fatty acid supplements on acute exacerbations of chronic respiratory disease, asthma control, pneumonia and lung function in adults. Contemp Clin Trials 2016;47:185–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.