Associations of ω-3 Fatty Acids With Interstitial Lung Disease and Lung Imaging Abnormalities Among Adults

Abstract

Docosahexaenoic acid (DHA), an ω-3 polyunsaturated fatty acid, attenuates interstitial lung disease (ILD) in experimental models, but human studies are lacking. We examined associations of circulating levels of DHA and other polyunsaturated fatty acids with hospitalization and death due to ILD over 12 years in the Multi-Ethnic Study of Atherosclerosis (MESA; n = 6,573). We examined cross-sectional associations with CT lung abnormalities in MESA (2000-2012; n = 6,541), the Framingham Heart Study (2005-2011; n = 3,917), and the Age, Gene/Environment Susceptibility-Reykjavik Study (AGES-Reykjavik) (2002-2006; n = 1,106). Polyunsaturated fatty acid levels were determined from fasting blood samples and extracted from plasma phospholipids (MESA and AGES-Reykjavik) or red blood cell membranes (Framingham Heart Study). Higher DHA levels were associated with a lower risk of hospitalization due to ILD (per standard-deviation increment, adjusted rate ratio = 0.69, 95% confidence interval (CI): 0.48, 0.99) and a lower rate of death due to ILD (per standard-deviation increment, adjusted hazard ratio = 0.68, 95% CI: 0.47, 0.98). Higher DHA was associated with fewer interstitial lung abnormalities on computed tomography (per natural log increment, pooled adjusted odds ratio = 0.65, 95% CI: 0.46, 0.91). Higher DHA levels were associated with a lower risk of hospitalization and death due to ILD and fewer lung abnormalities on computed tomography in a meta-analysis of data from population-based cohort studies.

Keywords: computed tomography; fatty acids; interstitial lung disease.

Figure 1

Selection of participants from the Multi-Ethnic Study of Atherosclerosis (MESA) with polyunsaturated fatty acid (PUFA) measurements and interstitial lung disease (ILD)-related hospitalizations and mortality, 2000–2012.

Figure 2

Selection of participants with polyunsaturated fatty acid (PUFA) measurements and interstitial lung abnormalities (ILA) from the Multi-Ethnic Study of Atherosclerosis (MESA) (2000–2012) (A), Framingham Heart Study (2005–2011) (B), and Age, Gene/Environment Susceptibility-Reykjavik Study (AGES) (2002–2006) (C) populations. CT, computed tomography.

Figure 3

Selection of participants with polyunsaturated fatty acid (PUFA) measurements and high-attenuation areas (HAA) from the Multi-Ethnic Study of Atherosclerosis (MESA) (2000–2012) (A) and Framingham Heart Study (2005–2011) (B) populations. CT, computed tomography; eGFR, estimated glomerular filtration rate. Four MESA participants with valid PUFA and HAA measurements were missing data on more than 1 variable.

Figure 4

Association of plasma phospholipid docosahexaenoic acid level with interstitial lung disease mortality in data from 3 cohort studies, 2000–2012. The model adjusted for generalized propensity score 2, which included age, sex, race/ethnicity, study site, smoking status, pack-years of cigarette smoking, waist circumference, height, body mass index, estimated glomerular filtration rate, educational attainment, alcohol use, coronary artery calcium level, use of diabetes medication, use of insulin, fasting glucose level, total amount of intentional exercise (MET-minutes/week), systolic and diastolic blood pressures, use of antihypertensive medication, total and high-density lipoprotein cholesterol levels, use of cholesterol medication, C-reactive protein level, D-dimer level, percent emphysema, and cancer history. The solid curve (smoothed regression line) shows the overall effect estimate, and the thin dashed lines show the 95% confidence interval. Each vertical hash mark in the rug plot along the x-axis represents 1 participant. MET, metabolic equivalent of task.

Figure 5

Results from a meta-analysis of associations between ω-3 polyunsaturated fatty acid (PUFA) levels and interstitial lung abnormalities in data from 3 cohort studies, 2000–2012. A) Docosahexaenoic acid; B) eicosapentaenoic acid; C) docosapentaenoic acid; D) total ω-3 polyunsaturated fatty acids (docosahexaenoic acid + eicosapentaenoic acid + docosapentaenoic acid). Odds ratios (ORs) were calculated per natural log unit increment of PUFA level. Models for each cohort adjusted for age, sex, smoking status, pack-years of cigarette smoking, and body mass index. The MESA model also adjusted for race/ethnicity. For all 4 random-effects analyses, I² = 0. Bars, 95% confidence intervals (CIs). AGES, Age, Gene/Environment Susceptibility-Reykjavik Study; FHS, Framingham Heart Study; MESA, Multi-Ethnic Study of Atherosclerosis.