Abstract
Background: Muscle mass, intermuscular adipose tissue, and strength are important indicators of physical function. Dietary fatty acids (FAs) have been associated with muscle parameters such as larger size and higher strength, but large, population-based longitudinal data in older adults who are at risk of functional decline are lacking.
Objective: The objective of this study was to investigate associations between plasma phospholipid polyunsaturated fatty acids (PUFAs) and measures of muscle size, intermuscular adipose tissue, and muscle strength cross-sectionally and after 5 y of follow-up.
Methods: Data are from the Age, Gene/Environment Susceptibility–Reykjavik Study, a prospective cohort aged 66–96 y at baseline. The analytic sample included 836 participants with cross-sectional measures of muscle parameters and 459 participants with data on change in muscle parameters. PUFAs were assessed at study baseline through use of GC. Muscle parameters were assessed at baseline and after a median of 5.2 y. Muscle area and intermuscular adipose tissue were assessed with computed tomography. Maximal grip strength and knee extension strength were assessed with dynometers. Relative changes in muscle parameters (%) were calculated. Multivariate linear regression was performed to calculate unstandardized regression coefficients and P values for trends across tertiles of FAs are reported.
Results: Higher concentrations of total PUFAs were cross-sectionally associated with larger muscle size (P-trend: 0.002) and with greater knee extension strength (P-trend: 0.038). Higher concentrations of arachidonic acid were associated with smaller muscle size (P-trend: 0.015). Greater linoleic acid was associated with less intermuscular adipose tissue (P-trend: 0.004), whereas eicosapentaenoic acid (20:5n–3) was positively associated (P-trend: 0.047). Longitudinal analyses showed positive associations for α-linolenic acid with increased knee extension strength (P-trend: 0.014). No other associations were observed.
Conclusions: These data illustrate the complex relation between plasma phospholipid PUFAs and muscle parameters; inconsistent cross-sectional relations with muscle size, intermuscular adipose tissue, and strength, and little evidence of a role in changes in muscle parameters.
Keywords: epidemiology, muscle loss, muscle parameters, older adults, polyunsaturated fatty acids
Introduction
The loss of muscle mass and/or muscle strength that occurs with aging (1–3) is indicative of declining health, functional impairment (4), disability (4, 5), and mortality (6, 7). Aging is also accompanied by adipose infiltration of muscle tissue [greater intermuscular adipose tissue (IMAT)10], which is associated with low muscle strength (8) and insulin resistance (9). There have been numerous efforts to prevent or reverse loss of muscle mass and muscle strength (10, 11), but effective treatments are lacking (12), underscoring the importance of identifying modifiable factors related to muscle in old age.
Several studies suggest a potential role of FAs on muscle, particularly n–3 PUFAs, which predominately consist of EPA and DHA. A study of cancer-related muscle loss, which shares features with age-related muscle loss (13), reported that individuals with low plasma n–3 PUFAs had lower muscle mass and greater muscle mass loss than individuals with higher n–3 PUFAs (14). A small study of older men and women reported a positive correlation between n–3 PUFA consumption and leg strength and an inverse correlation with chair rise time, although not independent of protein consumption (15). In a study of nearly 3000 community-dwelling older men and women, a positive association between fatty fish consumption and grip strength was reported (16). Clinical trials provide further evidence of a relation with n–3 PUFAs; supplemental n–3 PUFAs enhance the rate of muscle protein synthesis in older adults (17), and improvements in muscle strength and functional capacity were observed when strength training was combined with fish oil supplementation in older women (18).
However, previous studies are limited by their study size, cross-sectional study design, estimates of PUFA consumption from FFQs, or a nonrepresentative sample, e.g., only women (18) or nonobese subjects and those free from chronic disease (17). Computed tomography (CT)–derived measures of muscle size, which provide a more precise estimate of skeletal muscle than other imaging modalities (e.g., dual-energy X-ray absorptiometry) (19), and captures IMAT are also limited (20).
The objective of this study was to define relations between 2 domains of PUFAs: plasma phospholipid PUFAs and lifetime fish oil consumption with thigh muscle size and IMAT from CT as well as muscle strength from knee extension and grip strength in a substudy of older men and women from the Age, Gene/Environment Susceptibility (AGES-Reykjavik) Study. We further aimed to determine if PUFAs are associated with changes in these muscle parameters after 5 y of follow-up.
Methods
Study population.
The AGES-Reykjavik Study is a random sample of 5764 survivors from the Reykjavik Study, a single-center, population-based cardiovascular cohort begun in 1967 to study heart disease. At study baseline (2002–2006), participants were aged 66–96 y. A follow-up examination was carried out between 2007 and 2011 for 3316 participants. Details of the study design are provided by Harris et al. (21). All participants provided written informed consent, and the study was approved by the institutional review board.
The analytic sample was drawn from 2 substudies in AGES-Reykjavik with plasma FA data: ICELAND-MI, a study of cardiac MRI (22, 23) and a case cohort study of fracture (unpublished data). Participants in ICELAND-MI and the fracture cohort were randomly selected AGES-Reykjavik Study participants who met eligibility for MRI (i.e., no implanted devices or severe kidney disease) (n = 1012). Participants who were fracture cases were excluded because of nonrandom sampling. Individuals missing data on fish oil consumption, thigh muscle size or IMAT, grip strength, knee extension strength, or covariates were excluded resulting in an analytic sample of 836 with baseline measures (referred to as the cross-sectional sample) and 459 with follow-up measures (longitudinal sample). Compared with the overall AGES-Reykjavik Study population, our sample had fewer women (cross-sectional only, P = 0.009), were younger (longitudinal only, P < 0.001), more educated (longitudinal only, P = 0.010), and less likely to be current smokers (cross-sectional only, P = 0.011). Participants in our samples were also more likely to report more moderate-to-vigorous physical activity, less likely to have diabetes, and had greater muscle size and strength and less IMAT than the overall AGES-Reykjavik Study population (P < 0.05 for all).
FAs.
Plasma samples were collected after an overnight fast and stored at −80°C until analysis. FAs were measured in the phospholipid fraction, which provides a measure of short-term dietary intake and FAs available to peripheral tissues. Analyses were carried out at the Biomarker Lab, Fred Hutchinson Cancer Research Center (Seattle, WA). Plasma lipids were extracted with use of the Folch method (24). Phospholipids were isolated from other lipids with use of thin layer chromatography (25). FA methyl esters were prepared by direct transesterification (26) and separated by GC (Agilent 7890 gas chromatograph flame ionization detector; Supelco-fused silica 100-m capillary column SP-2560; initial 160°C for 16 min, ramp 3.0°c/min to 240°C, hold for 15 min). Identification, precision, and accuracy were continuously evaluated with use of both model mixtures of known FA methyl esters and established in-house control pools. FAs were expressed as a relative weight proportion (%). We focused on total PUFAs, total n–3 and n–6 PUFAs, and individual PUFAs previously associated with muscle size and/or strength: linoleic acid (LA) (18:2n–6), arachidonic acid (AA) (20:4n–6), α-linolenic acid (ALA) (18:3n–3), EPA (20:5n–3), and DHA (22:6n–3). The coefficient of variation from pooled quality control samples for LA, AA, EPA, and DHA were 0.43%, 0.62%, 2.05%, and 1.44%, respectively.
Fish oil.
Fish oil consumption was assessed as a measure of long-term n–3 PUFA consumption by a FFQ. The questionnaire was administered at study baseline (late life) and participants recalled intake during early life (ages 14–19) and midlife (ages 40–50). The questionnaire was validated for midlife and late-life consumption (27, 28). The most commonly consumed fish in Iceland are cod and haddock (29), both of which contain low concentrations of n–3 PUFAs. Therefore, we focused on fish liver oil consumption (referred to as fish oil hereafter), which is rich in n–3 PUFAs and has been common in the Icelandic diet for many decades. Consumption was classified as never, <daily, or daily.
Muscle parameters.
CT measurements were performed in the mid-thigh with use of a 4-detector system (Sensation; Siemens Medical Systems) as described previously (30). Thigh cross-sectional muscle area (cm2) was determined from a single 10-mm-thick trans-axial section (31). The fascial plane was used as an outline to segment muscle from subcutaneous fat. Adipose tissue between and within muscle (IMAT; cm2) was determined as pixels with radiographic density between −150 and −30 Hounsfield units and multiplied by pixel area.
The maximal isometric muscle strength, handgrip, and knee extension were measured on the dominant side with use of an adjustable dynamometer chair. Knee extension (N) was measured with the knee angle at 60° (31), and handgrip strength (N) was measured with use of a hand-held dynamometer with the elbow flexed at 90° (32).
CT, knee extension, and grip strength were repeated at the follow-up exam (median: 5.16 y, IQR: 5.06–5.21 y) with use of the same protocols as at baseline. Relative changes (%) in muscle size, IMAT, and knee and grip strength were calculated.
Covariates.
Covariates associated with muscle size and/or strength were assessed at baseline. Values in the text are means ± SDs. Age, education, smoking status, and physical activity were assessed by questionnaire. BMI was determined from measured height and weight with use of standard protocols (21). Type 2 diabetes, chronic obstructive pulmonary disease, and coronary heart disease were determined from self-report, medication use, and clinical assessment. Microalbuminuria was defined as urinary albumin-to-creatinine ratio ≥300 mg/g. Plasma C-reactive protein was analyzed with use of reagents from Roche Diagnostics on a Hitachi 912 analyzer according to manufacturer instructions.
Statistical analysis.
Interaction terms for FAs by sex in relation to muscle parameters were nonsignificant, and thus, data are presented for men and women together. The distribution of FAs varied; e.g., DHA met normal assumptions if log transformed, whereas PUFAs would require cubic transformation and EPA 1/square root. To facilitate interpretation of results, FAs were examined in tertiles rather than continuous measures. To achieve an equal distribution of men and women across tertiles of PUFAs, and adequately adjust for potential confounding due to sex, we computed sex-specific tertiles. Multivariate linear regression was used to test associations of FAs and fish oil with cross-sectional measures of muscle size, IMAT, and muscle strength. For the longitudinal analyses we performed similar regression analyses with relative changes in muscle size, IMAT, and muscle strength as outcome measures. Tertile 1 and never consuming fish oil were the referent groups. Results are reported as regression coefficients with 95% CIs. Model 1 was adjusted for age, sex, and physical activity. Model 2 was adjusted for model 1 plus education, smoking status, BMI, diabetes, chronic obstructive pulmonary disease, coronary heart disease, microalbuminuria, C-reactive protein, and time between assessments for the longitudinal sample. A P value for trend was calculated to examine linear trends across the tertiles with use of the categorical variable in adjusted regression models. Analyses were performed with STATA version 12.1 (StataCorp). A 2-sided P value < 0.05 was considered statistically significant.
Results
Baseline characteristics of participants in the cross-sectional and longitudinal samples are shown in Table 1. The age of the cross-sectional sample was 76.7 ± 5.60 y of which the majority were women. On average participants were overweight (BMI: 27.1 ± 4.11 kg/m2) with 1.35 ± 2.33 h/wk of moderate-to-vigorous physical activity. The distribution of plasma phospholipid FAs at the AGES-Reykjavik Study baseline for the cross-sectional and longitudinal sample are shown in Table 2.
TABLE 1.
Baseline characteristics of older men and women within the AGES-Reykjavik observational cohort study1
| Cross-sectional cohort (n = 836) | Longitudinal cohort (n = 459) | |
| Women, n (%) | 448 (53.6) | 249 (54.3) |
| Age at baseline, y | 76.7 ± 5.60 | 74.9 ± 4.98 |
| Education, n (%) | ||
| Primary school | 191 (22.9) | 89 (19.4) |
| Secondary | 426 (51.0) | 241 (52.5) |
| College | 128 (15.3) | 73 (15.9) |
| University | 91 (10.9) | 56 (12.2) |
| Smoking status, n (%) | ||
| Never | 334 (40.0) | 195 (42.5) |
| Former | 401 (48.0) | 216 (47.1) |
| Current | 101 (12.1) | 48 (10.5) |
| BMI, kg/m2 | 27.1 ± 4.11 | 27.2 ± 3.93 |
| Moderate-to-vigorous physical activity, h/wk | 1.35 ± 2.33 | 1.73 ± 2.59 |
| Diabetes, n (%) | 87 (10.4) | 40 (8.7) |
| Chronic obstructive pulmonary disease, n (%) | 23 (2.8) | 11 (2.4) |
| Coronary heart disease, n (%) | 177 (21.2) | 89 (19.4) |
| Microalbuminuria, n (%) | 64 (7.7) | 27 (5.9) |
| Plasma C-reactive protein, mg/L | 4.05 ± 8.04 | 3.64 ± 8.11 |
| Thigh muscle area, cm2 | 112 ± 25.6 | 116 ± 25.8 |
| Thigh intermuscular fat area, cm2 | 17.3 ± 7.78 | 17.1 ± 7.99 |
| Knee extension strength, N | 329 ± 117 | 353 ± 117 |
| Grip strength, N | 294 ± 110 | 313 ± 114 |
Continuous variables are means ± SDs and categorical variables are n (%). AGES-Reykjavik, Age, Gene/Environment Susceptibility–Reykjavik.
TABLE 2.
Distribution of plasma phospholipid FAs at AGES-Reykjavik Study baseline1
| Tertile 1 | Tertile 2 | Tertile 3 | |
| Cross-sectional cohort, n | 278 | 278 | 280 |
| Total PUFAs | 36.5 (30.0–38.0)2 | 38.4 (37.6–39.2) | 39.8 (38.8–42.8) |
| n–6 PUFAs | 23.3 (14.9–26.4) | 28.0 (26.3–29.6) | 31.0 (29.2–37.4) |
| n–3 PUFAs | 7.69 (4.36–9.16) | 10.3 (8.70–12.3) | 14.4 (11.3–22.6) |
| LA | 14.4 (7.38–16.5) | 17.6 (16.4–18.9) | 20.6 (18.8–27.3) |
| AA | 5.27 (3.81–6.05) | 6.70 (5.96–7.46) | 8.71 (7.33–16.5) |
| ALA | 0.16 (0.09–0.19) | 0.21 (0.18–0.24) | 0.31 (0.23–0.66) |
| EPA | 1.49 (0.69–2.05) | 2.53 (1.87–3.49) | 4.88 (3.13–10.9) |
| DHA | 4.80 (2.50–5.73) | 6.31 (5.49–7.21) | 8.17 (6.82–12.1) |
| Longitudinal cohort, n | 153 | 153 | 153 |
| Total PUFAs | 36.9 (32.0–38.2) | 38.6 (37.9–39.3) | 40.0 (39.0–42.8) |
| n–6 PUFAs | 24.0 (14.9–27.2) | 28.4 (26.6–29.8) | 31.2 (29.3–37.4) |
| n–3 PUFAs | 7.72 (4.36–9.16) | 10.1 (8.77–12.0) | 14.1 (10.8–22.6) |
| LA | 14.8 (7.38–16.8) | 17.9 (16.4–19.1) | 20.8 (19.0–27.3) |
| AA | 5.35 (3.82–6.18) | 6.80 (6.08–7.47) | 8.82 (7.38–16.5) |
| ALA | 0.15 (0.09–0.19) | 0.21 (0.19–0.24) | 0.30 (0.24–0.63) |
| EPA | 1.51 (0.69–2.06) | 2.53 (1.87–3.41) | 4.79 (2.94–10.9) |
| DHA | 4.79 (2.50–5.69) | 6.23 (5.48–7.10) | 7.99 (6.63–12.1) |
Tertiles are sex-specific and therefore values may overlap. AA, arachidonic acid; AGES-Reykjavik, Age, Gene/Environment Susceptibility–Reykjavik; ALA, α-linolenic acid; LA, linoleic acid.
Mean relative weight; range in parentheses (all such values).
PUFAs with cross-sectional muscle parameters.
Table 3 shows the cross-sectional associations between PUFAs with muscle parameters. Total PUFAs were positively associated with thigh muscle size with adjustment for age, sex, and physical activity (model 1). Associations remained significant in model 2 after additional adjustment for education, smoking status, BMI, diabetes, chronic obstructive pulmonary disease, coronary heart disease, microalbuminuria, and C-reactive protein [tertile 2: 2.86 cm2 (95% CI: 0.49, 5.22) and tertile 3: 3.83 cm2 (95% CI: 1.41, 6.25)]. Higher concentrations of AA were associated with lower thigh muscle size in fully adjusted models in tertile 2: −2.77 cm2 (95% CI: −5.14, −0.40) and tertile 3: −3.02 cm2 (95% CI: −5.45, −0.58). Tertiles 2 and 3 of LA were inversely associated with IMAT in both models [tertile 2: −1.51 cm2 (95% CI: −2.67, −0.35) and tertile 3: −1.74 cm2 (95% CI: −2.91, −0.56) (model 2)]. Tertile 3 of EPA was positively associated with IMAT after full adjustment for covariates [tertile 3: 1.20 cm2 (95% CI: 0.01, 2.38)].
TABLE 3.
Plasma phospholipid FAs in relation to baseline thigh muscle parameters and muscle strength among 836 participants1
| Muscle size, cm2 |
IMAT, cm2 |
Knee extension strength, N |
Grip strength, N |
|||||
| Model 1 | Model 2 | Model 1 | Model 2 | Model 1 | Model 2 | Model 1 | Model 2 | |
| Total PUFAs | ||||||||
| T2 | 2.93 (0.05, 5.81) | 2.86 (0.49, 5.22) | −0.83 (−2.12, 0.47) | −0.92 (−2.10, 0.25) | 11.1 (−3.13, 25.3) | 11.9 (−2.10, 25.9) | −1.93 (−14.1, 10.2) | −1.58 (−13.8, 10.6) |
| T3 | 4.31 (1.42, 7.20) | 3.83 (1.41, 6.25) | −1.11 (−2.41, 0.19) | −1.16 (−2.36, 0.05) | 16.5 (2.26, 30.8) | 15.2 (0.89, 29.6) | 11.0 (−1.21, 23.2) | 9.72 (−2.75, 22.2) |
| P-trend | 0.04 | 0.002 | 0.10 | 0.06 | 0.024 | 0.038 | 0.08 | 0.13 |
| n–6 PUFAs | ||||||||
| T2 | −0.46 (−3.34, 2.41) | −2.29 (−4.65, 0.08) | 0.80 (−0.49, 2.09) | 0.16 (−1.02, 1.33) | 2.99 (−11.2, 17.2) | 0.39 (−13.6, 14.4) | 8.68 (−3.41, 20.8) | 8.82 (−3.33, 21.0) |
| T3 | 2.53 (−0.39, 5.45) | 0.86 (−1.57, 3.28) | −0.29 (−1.60, 1.03) | −1.00 (−2.21, 0.21) | 3.32 (−11.1, 17.8) | 3.61 (−10.8, 18.0) | −6.40 (−18.7, 5.90) | −5.86 (−18.3, 6.63) |
| P-trend | 0.09 | 0.49 | 0.67 | 0.10 | 0.65 | 0.62 | 0.31 | 0.36 |
| n–3 PUFAs | ||||||||
| T2 | 0.01 (−2.86, 2.88) | −0.70 (−3.07, 1.67) | 0.67 (−0.61, 1.96) | 0.57 (−0.61, 1.74) | 6.59 (−7.55, 20.7) | 2.71 (−11.3, 16.7) | 18.1 (6.06, 30.1) | 16.6 (4.51, 28.7) |
| T3 | −1.38 (−4.27, 1.50) | 0.03 (−2.38, 2.45) | −0.09 (−1.38, 1.20) | 0.66 (−0.53, 1.86) | 7.11 (−7.12, 21.3) | 5.18 (−9.06, 19.4) | 12.3 (0.20, 24.4) | 10.1 (−2.21, 22.5) |
| P-trend | 0.35 | 0.98 | 0.89 | 0.28 | 0.33 | 0.48 | 0.047 | 0.11 |
| LA | ||||||||
| T2 | −1.20 (−4.07, 1.67) | −0.70 (−3.05, 1.65) | −1.75 (−3.03, −0.47) | −1.51 (−2.67, −0.35) | 2.71 (−11.4, 16.9) | 2.49 (−11.4, 16.4) | 8.07 (−4.00, 20.1) | 6.80 (−5.27, 18.9) |
| T3 | 0.65 (−2.23, 3.54) | 1.41 (−0.97, 3.79) | −2.07 (−3.36, −0.79) | −1.74 (−2.91, −0.56) | 1.67 (−12.6, 15.9) | 0.90 (−13.2, 15.0) | −2.73 (−14.9, 9.40) | −3.88 (−16.1, 8.33) |
| P-trend | 0.66 | 0.25 | 0.002 | 0.004 | 0.82 | 0.90 | 0.66 | 0.54 |
| AA | ||||||||
| T2 | −0.77 (−3.65, 2.12) | −2.77 (−5.14, −0.40) | −0.39 (−1.69, 0.91) | −1.12 (−2.30, 0.06) | −5.78 (−20.0, 8.47) | −7.69 (−21.7, 6.36) | −5.04 (−17.2, 7.12) | −4.30 (−16.5, 7.92) |
| T3 | 0.64 (−2.25, 3.52) | −3.02 (−5.45, −0.58) | 0.81(−0.48, 2.10) | −0.63 (−1.84, 0.58) | −8.57 (−22.8, 5.66) | −9.22 (−23.6, 5.21) | −7.32 (−19.5, 4.83) | −5.38 (−17.9, 7.17) |
| P-trend | 0.66 | 0.015 | 0.21 | 0.31 | 0.24 | 0.21 | 0.24 | 0.40 |
| ALA | ||||||||
| T2 | 0.03 (−2.84, 2.89) | 0.54 (−1.81, 2.90) | −1.02 (−2.30, 0.26) | −0.88 (−2.05, 0.28) | 0.99 (−13.1, 15.1) | 2.13 (−11.8, 16.0) | 1.42 (−10.6, 13.5) | 1.34 (−10.8, 13.4) |
| T3 | −0.16 (−3.02, 2.71) | 0.87 (−1.49, 3.24) | −0.93 (−2.21, 0.35) | −0.51 (−1.68, 0.66) | −5.50 (−19.6, 8.63) | −4.36 (−18.3, 9.59) | 5.70 (−6.36, 17.8) | 5.28 (−6.85, 17.4) |
| P-trend | 0.92 | 0.47 | 0.16 | 0.39 | 0.45 | 0.54 | 0.35 | 0.39 |
| EPA | ||||||||
| T2 | 2.01 (−0.86, 4.87) | 0.70 (−1.66, 3.07) | 0.58 (−0.71, 1.86) | 0.25 (−0.93, 1.42) | 12.0 (−2.14, 26.1) | 8.12 (−5.83, 22.1) | 18.3 (6.25, 30.3) | 17.1 (5.01, 29.2) |
| T3 | 0.24 (−2.63, 3.11) | 0.81 (−1.58, 3.20) | 0.76 (−0.53, 2.04) | 1.20 (0.01, 2.38) | 15.5 (1.32, 29.6) | 13.3 (−0.81, 27.4) | 14.0 (1.98, 26.0) | 11.8 (−0.38, 24.1) |
| P-trend | 0.87 | 0.50 | 0.25 | 0.047 | 0.032 | 0.06 | 0.023 | 0.06 |
| DHA | ||||||||
| T2 | 1.40 (−1.46, 4.27) | 0.58 (−1.78, 2.95) | −0.01 (−1.30, 1.28) | −0.17 (−1.34, 1.00) | 12.5 (−1.63, 26.6) | 9.79 (−4.17, 23.7) | 13.4 (1.39, 25.5) | 12.1 (−0.04, 24.2) |
| T3 | −0.62 (−3.52, 2.27) | 0.30 (−2.12, 2.72) | −0.19 (−1.49, 1.11) | 0.42 (−0.78, 1.62) | 7.32 (−6.93, 21.6) | 4.91 (−9.37, 19.2) | 12.8 (0.69, 25.0) | 10.5 (−1.91, 22.9) |
| P-trend | 0.67 | 0.81 | 0.78 | 0.49 | 0.31 | 0.50 | 0.038 | 0.10 |
Multivariable linear regression analyses were performed to compute regression coefficients (95% CIs). Tertile 1 is the reference group (coefficient = 0.00), n = 278; n = 278 for tertile 2; and n = 280 for tertile 3. Model 1 was adjusted for age, sex, and physical activity. Model 2 was adjusted for model 1 plus education, smoking status, BMI, diabetes, chronic obstructive pulmonary disease, coronary heart disease, microalbuminuria, and C-reactive protein. AA, arachidonic acid; ALA, α-linolenic acid; IMAT, intermuscular adipose tissue; LA, linoleic acid; T, tertile.
Greater knee extension strength was observed for participants in tertile 3 of total PUFAs [15.2 N (95% CI: 0.89, 29.6) (model 2)]. Total n–3 PUFAs and DHA were positively associated with grip strength in model 1, however, associations were attenuated in model 2. EPA was positively associated with knee extension and grip strength in model 1 but was no longer significant with adjustment for additional covariates (model 2).
PUFAs with relative changes in muscle parameters.
Loss of muscle size and strength and increased fatty infiltration of muscle was evident over follow-up. Change (%) in muscle size, IMAT, knee extension strength, and grip strength were the following: −4.87 ± 7.87, 8.28 ± 25.5, −17.8 ± 21.2, and −3.42 ± 25.3, respectively. PUFAs in relation to changes in muscle size, IMAT, knee extension, and grip strength are shown in Table 4. There were no associations with any of the FAs with change in muscle size or grip strength. Relative to tertile 1, tertile 3 of AA was associated with greater losses in knee extension strength relative to tertile 1 [−5.31% (95% CI: −10.1, −0.54) (model 1)]. This association became nonsignificant in model 2. Higher concentrations of ALA were associated with greater IMAT relative to tertile 1 in tertile 2 [8.07% (95% CI: 2.47, 13.7)] and in tertile 3 [6.01% (95% CI: 0.40, 11.6) (model 1)]. However, further adjustments attenuated the associations. Tertile 3 of ALA was associated with less loss in knee extension strength relative to tertile 1 [5.97% (95% CI: 1.17, 10.8) (model 2)]. No other associations between FAs and change in muscle parameters or strength were observed.
TABLE 4.
Plasma phospholipid FAs in relation to relative change in muscle parameters and muscle strength among 459 participants over 5 y of follow-up1
| Muscle size, % |
IMAT, % |
Knee extension strength, % |
Grip strength, % |
|||||
| Model 1 | Model 2 | Model 1 | Model 2 | Model 1 | Model 2 | Model 1 | Model 2 | |
| Total PUFAs | ||||||||
| T2 | 0.51 (−1.27, 2.29) | 0.88 (−0.88, 2.65) | −1.98 (−7.63, 3.68) | −2.18 (−7.87, 3.52) | 2.21 (−2.58, 6.99) | 2.57 (−2.27, 7.41) | 1.34 (−4.37, 7.05) | 1.58 (−4.21, 7.38) |
| T3 | 0.04 (−1.77, 1.84) | 0.35 (−1.46, 2.17) | −2.78 (−8.51, 2.95) | −1.61 (−7.45, 4.23) | 2.54 (−2.31, 7.38) | 2.14 (−2.82, 7.10) | 0.74 (−5.04, 6.52) | 0.60 (−5.35, 6.54) |
| P-trend | 0.97 | 0.71 | 0.34 | 0.59 | 0.31 | 0.40 | 0.80 | 0.85 |
| n–6 PUFAs | ||||||||
| T2 | −0.81 (−2.58, 0.97) | −0.29 (−2.05, 1.48) | 1.49 (−4.17, 7.14) | 0.73 (−4.95, 6.41) | −1.34 (−6.12, 3.45) | −0.33 (−5.16, 4.51) | −0.16 (−5.86, 5.53) | 0.42 (−5.35, 6.19) |
| T3 | −0.97 (−2.79, 0.84) | −0.50 (−2.33, 1.34) | −0.05 (−5.83, 5.73) | −1.09 (−6.98, 4.79) | −1.22 (−6.11, 3.67) | −0.21 (−5.22, 4.80) | 3.64 (−2.18, 9.45) | 3.88 (−2.10, 9.86) |
| P-trend | 0.29 | 0.60 | 0.99 | 0.72 | 0.62 | 0.93 | 0.22 | 0.20 |
| n–3 PUFAs | ||||||||
| T2 | 0.39 (−1.38, 2.16) | 0.39 (−1.36, 2.15) | −2.52 (−8.14, 3.11) | −2.43 (−8.07, 3.22) | −2.69 (−7.45, 2.06) | −2.47 (−7.28, 2.33) | −3.12 (−8.79, 2.54) | −2.76 (−8.50, 2.98) |
| T3 | 0.10 (−1.69, 1.89) | −0.33 (−2.14, 1.48) | −3.18 (−8.86, 2.50) | −1.84 (−7.66, 3.97) | −0.21 (−4.59, 5.01) | −1.10 (−6.04, 3.85) | −3.62 (−9.34, 2.10) | −3.93 (−9.84, 1.99) |
| P-trend | 0.91 | 0.73 | 0.27 | 0.53 | 0.94 | 0.65 | 0.21 | 0.19 |
| LA | ||||||||
| T2 | 0.11 (−1.66, 1.89) | 0.46 (−1.29, 2.20) | −3.54 (−9.17, 2.10) | −3.60 (−9.20, 2.01) | −1.67 (−6.43, 3.09) | −1.37 (−6.14, 3.39) | −1.48 (−7.14, 4.17) | −1.10 (−6.78, 4.58) |
| T3 | −0.30 (−2.08, 1.48) | −0.33 (−2.10, 1.44) | −1.92 (−7.58, 3.73) | −2.63 (−8.33, 3.08) | 2.54 (−2.23, 7.32) | 2.19 (−2.66, 7.04) | 5.03 (−0.65, 10.7) | 5.19 (−0.59, 11.0) |
| P-trend | 0.74 | 0.72 | 0.51 | 0.36 | 0.29 | 0.39 | 0.08 | 0.08 |
| AA | ||||||||
| T2 | 0.47 (−1.30, 2.25) | 0.53 (−1.23, 2.29) | 1.26 (−4.41, 6.92) | 1.25 (−4.42, 6.91) | −0.14 (−4.90, 4.62) | 0.52 (−4.28, 5.32) | −4.61 (−10.3, 1.09) | −4.38 (−10.1, 1.37) |
| T3 | −0.76 (−2.55, 1.02) | −0.37 (−2.19, 1.45) | −0.59 (−6.27, 5.09) | −0.50 (−6.37, 5.36) | −5.31 (−10.1, −0.54) | −4.02 (−8.99, 0.95) | −1.29 (−7.00, 4.42) | −0.97 (−6.93, 4.98) |
| P-trend | 0.40 | 0.70 | 0.84 | 0.87 | 0.029 | 0.12 | 0.66 | 0.74 |
| ALA | ||||||||
| T2 | −1.00 (−2.77, 0.79) | −1.15 (−2.91, 0.61) | 8.07 (2.47, 13.7) | 7.32 (1.70, 12.9) | −0.82 (−5.55, 3.90) | −1.11 (−5.88, 3.66) | −2.11 (−7.80, 3.57) | −2.27 (−8.03, 3.48) |
| T3 | −0.37 (−2.15, 1.40) | −0.49 (−2.26, 1.28) | 6.01 (0.40, 11.6) | 5.50 (−0.15, 11.2) | 6.51 (1.77, 11.2) | 5.97 (1.17, 10.8) | 2.29 (−3.41, 7.981) | 1.88 (−3.91, 7.68) |
| P-trend | 0.68 | 0.60 | 0.037 | 0.06 | 0.007 | 0.014 | 0.43 | 0.51 |
| EPA | ||||||||
| T2 | −1.31 (−3.08, 0.46) | −0.90 (−2.66, 0.86) | 5.06 (−0.56, 10.7) | 4.68 (−0.97, 10.3) | −0.18 (−4.95, 4.59) | 0.68 (−4.14, 5.50) | −3.81 (−9.45, 1.86) | −3.27 (−9.03, 2.49) |
| T3 | −0.24 (−2.01, 1.53) | −0.66 (−2.44, 1.11) | 1.18 (−4.45, 6.81) | 1.89 (−3.81, 7.59) | 0.01 (−4.77, 4.79) | −0.86 (−5.72, 4.00) | −2.96 (−8.64, 2.73) | −3.08 (−8.89, 2.72) |
| P-trend | 0.79 | 0.46 | 0.68 | 0.50 | 0.99 | 0.74 | 0.31 | 0.29 |
| DHA | ||||||||
| T2 | 0.21 (−1.56, 1.98) | 0.16 (−1.60, 1.93) | −5.23 (−10.8, 0.39) | −4.51 (−10.2, 1.15) | −1.98 (−6.74, 2.78) | −2.59 (−7.41, 2.24) | −2.69 (−8.44, 2.97) | −2.75 (−8.51, 3.02) |
| T3 | −0.02 (−1.81, 1.77) | −0.23 (−2.05, 1.60) | −3.05 (−8.71, 2.62) | −2.10 (−7.95, 3.76) | −0.61 (−5.41, 4.20) | −2.09 (−7.08, 2.90) | −4.55 (−10.3, 1.16) | −4.89 (−10.8, 1.07) |
| P-trend | 0.98 | 0.81 | 0.29 | 0.47 | 0.80 | 0.41 | 0.12 | 0.11 |
Multivariable linear regression analyses were performed to compute regression coefficients (95% CIs). Tertile 1 is the reference group (coefficient = 0.00), n = 153; n = 153 for tertile 2; and n = 153 for tertile 3. Model 1 was adjusted for age, sex, physical activity, and time between assessments. Model 2 was adjusted for model 1 plus education, smoking status, BMI, diabetes, chronic obstructive pulmonary disease, coronary heart disease, microalbuminuria, and C-reactive protein. AA, arachidonic acid; ALA, α-linolenic acid; IMAT, intermuscular adipose tissue; LA, linoleic acid; T, tertile.
Fish oil consumption and cross-sectional and relative changes in muscle parameters.
Fish oil consumption across the lifetime in relation to cross-sectional muscle parameters and changes in muscle parameters is shown in Supplemental Tables 1 and 2. Daily consumption of fish oil was common in our sample (old age: 64.0%, midlife: 45.2%, early life: 33.5%), which is reflective of the Icelandic population (33).
There were no associations between fish oil consumption in old age and any cross-sectional or prospective muscle parameters, confirming the largely null associations with plasma phospholipid n–3 PUFAs. Furthermore, there were no associations between fish oil consumption in midlife with any muscle parameters after full adjustment for covariates. There was a trend toward daily fish oil consumption in early life and lower baseline grip strength (P = 0.019), but no association with change in grip strength (P = 0.91).
Discussion
This is one of the first population-based studies, to our knowledge, examining the association between circulating concentrations of FAs and comprehensive cross-sectional and longitudinal muscle parameters that are important indicators of physical function in older adults. Our results illustrate a complex relation between FAs and muscle parameters. Total PUFAs were positively associated with cross-sectional muscle size and knee extension strength, but there were inconsistent associations with individual FAs, including the n–3 PUFAs, which are widely hypothesized to play a role in muscle mass and/or strength (17, 20). Of note, none of the cross-sectional associations were confirmed in our analysis of changes in muscle parameters.
Based on previous findings, we hypothesized that higher PUFA concentrations would be associated with favorable muscle measures and less diminution of muscle strength (15–18). For example, based on clinical trial evidence, it would have been reasonable to expect positive associations between PUFA concentrations and muscle size because PUFA supplementation was associated with an enhanced rate of muscle protein synthesis (17) or higher muscle strength (18). In addition, results of the Hertfordshire Study, a large retrospective cohort study of nearly 3000 participants aged 59–73 y, showed an increase in grip strength for each additional portion of fatty fish consumed per week (16). We did not find associations between PUFAs and grip strength. Differences between observed associations might be explained by different study design and/or age group.
The lack of associations between FAs and changes in muscle parameters casts doubt on the relevance of the cross-sectional associations we observed and those observed in prior cross-sectional studies (16). Furthermore, the lack of consistent associations between FAs and muscle parameters that are interrelated (e.g., total PUFAs were associated with knee extension strength but not grip strength or IMAT) does not provide convincing support for a role of PUFAs. Despite evidence from older adults of a stimulatory effect of fish oil supplementation on muscle protein synthesis (17), we did not find relations between fish oil, total n–3, or individual long-chain n–3 PUFAs with change in muscle size over a 5-y time period. It is important to note that the trial by Smith et al. (17) measured acute effects of high doses of fish oil (4 g/d) in a small sample (7 in control group, 8 in fish oil group) of individuals with different demographics than our population (BMI: <30 kg/m2 and free from disease). It is possible that our divergent results are due to a threshold effect of n–3 PUFAs on muscle, but a previous study of cancer patients with a mean age of 62 y reported a linear association between CT-derived muscle size and circulating (nonsupplemented) concentrations of phospholipid EPA (14).
n–3 PUFAs have been investigated in a wide range of chronic conditions because of their anti-inflammatory potential whereby n–3-derived lipid mediators (eicosanoids) inhibit formation of proinflammatory eicosanoids from n–6 PUFAs (34). Reports of the benefits of n–3 PUFAs in chronic diseases with inflammatory processes, such as cardiovascular disease, cancer, and bone loss, were promising (35–38), but subsequent null studies have dampened enthusiasm (39, 40). We add to the debate over FAs and health outcomes by suggesting that phospholipid n–3 PUFAs that reflect supplementation and dietary sources do not play a role in processes related to (change of) muscle size, IMAT, or strength.
The strengths of the study include the measurement of circulating FAs in a relatively large sample of older adults, CT assessment of muscle size and IMAT, and assessments of both upper and lower body strength. The availability of self-reported data on fish oil supplementation at 3 key time points and longitudinal measures of muscle size, IMAT, and muscle strength, although in a smaller number of participants, uniquely enabled characterization of PUFAs in relation to change in muscle parameters. There are also study limitations to consider. PUFAs were only determined at baseline and therefore changes in exposure over time were limited to self-reported data on fish oil consumption. The self-reported data did not indicate any relations between chronic PUFA exposure and outcome. However, multiple measurements of phospholipid plasma PUFAs may be of additional value to investigate the role of dynamic circulating PUFAs in relation to muscle parameters. The Icelandic population is characterized by high fish oil consumption relative to the United States, although n–3 PUFA concentrations are lower than in Asian countries (41). This may affect the generalizability of our results. It is also possible that the likelihood of detecting associations in a population with relatively high n–3 PUFA concentrations is reduced because the reference group includes individuals who have a quite favorable FA profile. We did not have information on the composition of fish oil supplements and were limited to assessing muscle parameters in relation to the frequency rather than dose of n–3 PUFAs. However, because phospholipid EPA and DHA were generally not related to muscle parameters, this does not limit the interpretation of our findings. Finally, participants in the longitudinal cohort had to be well enough to complete follow-up measurements. The limited associations we observed between baseline PUFAs and muscle parameters may have been driven by less-healthy participants who did not survive long enough for follow-up muscle measures and inclusion in the longitudinal cohort. However, the longitudinal cohort was not typified by healthy aging; loss of muscle size and strength, and gain of IMAT, were the predominant features. Future studies should investigate whether changes in PUFA profiles are associated with muscle parameters. In addition, because this is the first longitudinal population-based study, to our knowledge, our results need to be confirmed by studies investigating the association between PUFAs and changes in muscle parameters in different study populations.
In conclusion, our results suggest inconsistent cross-sectional relations between plasma phospholipid PUFAs and muscle size, IMAT, and strength.
Supplementary Material
Acknowledgments
We thank Pho Diep for technical assistance with the FA analyses. GE, VG, TBH, and RAM designed and conducted the research; XS, SS, KS, IAB, and RAM provided essential data; IR and RAM performed statistical analyses and wrote the manuscript; XS, MV, SS, TA, KS, IAB, and TBH critically revised the manuscript; and IR takes responsibility for the integrity of the data and had primary responsibility for final content. All authors read and approved the final manuscript.
Footnotes
Abbreviations used: AA, arachidonic acid; AGES-Reykjavik, Age, Gene/Environment Susceptibility–Reykjavik; ALA, α-linolenic acid; CT, computed tomography; IMAT, intermuscular adipose tissue; LA, linoleic acid.
References
- 1.Newman AB, Kupelian V, Visser M, Simonsick E, Goodpaster B, Nevitt M, Kritchevsky SB, Tylavsky FA, Rubin SM, Harris TB. Sarcopenia: alternative definitions and associations with lower extremity function. J Am Geriatr Soc 2003;51:1602–9. [DOI] [PubMed] [Google Scholar]
- 2.Visser M, Pahor M, Tylavsky F, Kritchevsky SB, Cauley JA, Newman AB, Blunt BA, Harris TB. One- and two-year change in body composition as measured by DXA in a population-based cohort of older men and women. J Appl Physiol 2003;94:2368–74. [DOI] [PubMed] [Google Scholar]
- 3.Goodpaster BH, Park SW, Harris TB, Kritchevsky SB, Nevitt M, Schwartz AV, Simonsick EM, Tylavsky FA, Visser M, Newman AB. The loss of skeletal muscle strength, mass, and quality in older adults: the health, aging and body composition study. J Gerontol A Biol Sci Med Sci 2006;61:1059–64. [DOI] [PubMed] [Google Scholar]
- 4.Janssen I, Heymsfield SB, Ross R. Low relative skeletal muscle mass (sarcopenia) in older persons is associated with functional impairment and physical disability. J Am Geriatr Soc 2002;50:889–96. [DOI] [PubMed] [Google Scholar]
- 5.Marsh AP, Rejeski WJ, Espeland MA, Miller ME, Church TS, Fielding RA, Gill TM, Guralnik JM, Newman AB, Pahor M. Muscle strength and BMI as predictors of major mobility disability in the Lifestyle Interventions and Independence for Elders pilot (LIFE-P). J Gerontol A Biol Sci Med Sci 2011;66:1376–83. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Murphy RA, Ip EH, Zhang Q, Boudreau RM, Cawthon PM, Newman AB, Tylavsky FA, Visser M, Goodpaster BH, Harris TB, et al. Transition to sarcopenia and determinants of transitions in older adults: a population-based study. J Gerontol A Biol Sci Med Sci 2014;69:751–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Moisey LL, Mourtzakis M, Cotton BA, Premji T, Heyland DK, Wade CE, Bulger E, Kozar RA; Nutrition and Rehabilitation Investigators Consortium (NUTRIC). Skeletal muscle predicts ventilator-free days, ICU-free days, and mortality in elderly ICU patients. Crit Care 2013;17:R206. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Goodpaster BH, Carlson CL, Visser M, Kelley DE, Scherzinger A, Harris TB, Stamm E, Newman AB. Attenuation of skeletal muscle and strength in the elderly: the Health ABC Study. J Appl Physiol 2001;90:2157–65. [DOI] [PubMed] [Google Scholar]
- 9.Goodpaster BH, Thaete FL, Kelley DE. Thigh adipose tissue distribution is associated with insulin resistance in obesity and in type 2 diabetes mellitus. Am J Clin Nutr 2000;71:885–92. [DOI] [PubMed] [Google Scholar]
- 10.Morley JE. Sarcopenia: diagnosis and treatment. J Nutr Health Aging 2008;12:452–6. [DOI] [PubMed] [Google Scholar]
- 11.Sayer AA, Robinson SM, Patel HP, Shavlakadze T, Cooper C, Grounds MD. New horizons in the pathogenesis, diagnosis and management of sarcopenia. Age Ageing 2013;42:145–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Cooper C, Dere W, Evans W, Kanis JA, Rizzoli R, Sayer AA, Sieber CC, Kaufman JM, Abellan van Kan G, Boonen S, et al. Frailty and sarcopenia: definitions and outcome parameters. Osteoporos Int 2012;23:1839–48. [DOI] [PubMed] [Google Scholar]
- 13.Rolland Y, Abellan van Kan G, Gillette-Guyonnet S, Vellas B. Cachexia versus sarcopenia. Curr Opin Clin Nutr Metab Care 2011;14:15–21. [DOI] [PubMed] [Google Scholar]
- 14.Murphy RA, Mourtzakis M, Chu QS, Reiman T, Mazurak VC. Skeletal muscle depletion is associated with reduced plasma (n-3) fatty acids in non-small cell lung cancer patients. J Nutr 2010;140:1602–6. [DOI] [PubMed] [Google Scholar]
- 15.Rousseau JH, Kleppinger A, Kenny AM. Self-reported dietary intake of omega-3 fatty acids and association with bone and lower extremity function. J Am Geriatr Soc 2009;57:1781–8. [DOI] [PubMed] [Google Scholar]
- 16.Robinson SM, Jameson KA, Batelaan SF, Martin HJ, Syddall HE, Dennison EM, Cooper C, Sayer AA; Hertfordshire Cohort Study Group. Diet and its relationship with grip strength in community-dwelling older men and women: the Hertfordshire Cohort Study. J Am Geriatr Soc 2008;56:84–90. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Smith GI, Atherton P, Reeds DN, Mohammed BS, Rankin D, Rennie MJ, Mittendorfer B. Dietary omega-3 fatty acid supplementation increases the rate of muscle protein synthesis in older adults: a randomized controlled trial. Am J Clin Nutr 2011;93:402–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Rodacki CL, Rodacki AL, Pereira G, Naliwaiko K, Coelho I, Pequito D, Fernandes LC. Fish-oil supplementation enhances the effects of strength training in elderly women. Am J Clin Nutr 2012;95:428–36. [DOI] [PubMed] [Google Scholar]
- 19.Mourtzakis M, Prado CM, Lieffers JR, Reiman T, McCargar LJ, Baracos VE. A practical and precise approach to quantification of body composition in cancer patients using computed tomography images acquired during routine care. Appl Physiol Nutr Metab 2008;33:997–1006. [DOI] [PubMed] [Google Scholar]
- 20.Murphy RA, Mourtzakis M, Chu QS, Baracos VE, Reiman T, Mazurak VC. Nutritional intervention with fish oil provides a benefit over standard of care for weight and skeletal muscle mass in patients with nonsmall cell lung cancer receiving chemotherapy. Cancer 2011;117:1775–82. [DOI] [PubMed] [Google Scholar]
- 21.Harris TB, Launer LJ, Eiriksdottir G, Kjartansson O, Jonsson PV, Sigurdsson G, Thorgeirsson G, Aspelund T, Garcia ME, Cotch MF, et al. Age, Gene/Environment Susceptibility-Reykjavik Study: multidisciplinary applied phenomics. Am J Epidemiol 2007;165:1076–87. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Schelbert EB, Cao JJ, Sigurdsson S, Aspelund T, Kellman P, Aletras AH, Dyke CK, Thorgeirsson G, Eiriksdottir G, Launer LJ, et al. Prevalence and prognosis of unrecognized myocardial infarction determined by cardiac magnetic resonance in older adults. JAMA 2012;308:890–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.van Ballegooijen AJ, Visser M, Cotch MF, Arai AE, Garcia M, Harris TB, Launer LJ, Eiriksdottir G, Gudnason V, Brouwer IA. Serum vitamin D and parathyroid hormone in relation to cardiac structure and function: the ICELAND-MI Substudy of AGES-Reykjavik. J Clin Endocrinol Metab 2013;98:2544–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Folch J, Lees M, Sloane Stanley GH. A simple method for the isolation and purification of total lipides from animal tissues. J Biol Chem 1957;226:497–509. [PubMed] [Google Scholar]
- 25.Schlierf G, Wood P. Quantitative determination of plasma free fatty acids and triglycerides by thin-layer chromatography. J Lipid Res 1965;6:317–9. [PubMed] [Google Scholar]
- 26.Lepage G, Roy CC. Direct transesterification of all classes of lipids in a one-step reaction. J Lipid Res 1986;27:114–20. [PubMed] [Google Scholar]
- 27.Eysteinsdottir T, Gunnarsdottir I, Thorsdottir I, Harris T, Launer LJ, Gudnason V, Steingrimsdottir L. Validity of retrospective diet history: assessing recall of midlife diet using food frequency questionnaire in later life. J Nutr Health Aging 2011;15:809–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Eysteinsdottir T, Thorsdottir I, Gunnarsdottir I, Steingrimsdottir L. Assessing validity of a short food frequency questionnaire on present dietary intake of elderly Icelanders. Nutr J 2012;11:12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Gunnarsdottir I, Gunnarsdottir BE, Steingrimsdottir L, Maage A, Johannesson AJ, Thorsdottir I. Iodine status of adolescent girls in a population changing from high to lower fish consumption. Eur J Clin Nutr 2010;64:958–64. [DOI] [PubMed] [Google Scholar]
- 30.Sigurdsson G, Aspelund T, Chang M, Jonsdottir B, Sigurdsson S, Eiriksdottir G, Gudmundsson A, Harris TB, Gudnason V, Lang TF. Increasing sex difference in bone strength in old age: the Age, Gene/Environment Susceptibility-Reykjavik study (AGES-REYKJAVIK). Bone 2006;39:644–51. [DOI] [PubMed] [Google Scholar]
- 31.Johannesdottir F, Aspelund T, Siggeirsdottir K, Jonsson BY, Mogensen B, Sigurdsson S, Harris TB, Gudnason VG, Lang TF, Sigurdsson G. Mid-thigh cortical bone structural parameters, muscle mass and strength, and association with lower limb fractures in older men and women (AGES-Reykjavik Study). Calcif Tissue Int 2012;90:354–64. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Era P, Parssinen O, Gausenilsson I, Heikkinen E, Steen B. Intraocular-pressure in samples of elderly Finnish and Swedish men and women. Acta Ophthalmol (Copenh) 1994;72:581–7. [DOI] [PubMed] [Google Scholar]
- 33.Steingrimsdottir L, Thorgeirsdottir H, Olafsdottir AS. The diet of Icelanders. Dietary Survey of the Icelandic Nutrition Council 2002. Main findings. Rannsóknir Manneldisráds Íslands V [cited 2014 Jul 17]. Available from: http://www.landlaeknir.is/servlet/file/store93/item11603/skyrsla.pdf.
- 34.Wall R, Ross RP, Fitzgerald GF, Stanton C. Fatty acids from fish: the anti-inflammatory potential of long-chain omega-3 fatty acids. Nutr Rev 2010;68:280–9. [DOI] [PubMed] [Google Scholar]
- 35.Yokoyama M, Origasa H, Matsuzaki M, Matsuzawa Y, Saito Y, Ishikawa Y, Oikawa S, Sasaki J, Hishida H, Itakura H, et al. Effects of eicosapentaenoic acid on major coronary events in hypercholesterolaemic patients (JELIS): a randomised open-label, blinded endpoint analysis. Lancet 2007;369:1090–8. [DOI] [PubMed] [Google Scholar]
- 36.Gissi- HF Investigators, Tavazzi L, Maggioni AP, Marchioli R, Barlera S, Franzosi MG, Latini R, Lucci D, Nicolosi GL, Porcu M, et al. Effect of n-3 polyunsaturated fatty acids in patients with chronic heart failure (the GISSI-HF trial): a randomised, double-blind, placebo-controlled trial. Lancet 2008;372:1223–30. [DOI] [PubMed] [Google Scholar]
- 37.Fearon KC, Von Meyenfeldt MF, Moses AG, Van Geenen R, Roy A, Gouma DJ, Giacosa A, Van Gossum A, Bauer J, Barber MD, et al. Effect of a protein and energy dense N-3 fatty acid enriched oral supplement on loss of weight and lean tissue in cancer cachexia: a randomised double blind trial. Gut 2003;52:1479–86. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Rahman MM, Bhattacharya A, Fernandes G. Docosahexaenoic acid is more potent inhibitor of osteoclast differentiation in RAW 264.7 cells than eicosapentaenoic acid. J Cell Physiol 2008;214:201–9. [DOI] [PubMed] [Google Scholar]
- 39.Fearon KC, Barber MD, Moses AG, Ahmedzai SH, Taylor GS, Tisdale MJ, Murray GD. Double-blind, placebo-controlled, randomized study of eicosapentaenoic acid diester in patients with cancer cachexia. J Clin Oncol 2006;24:3401–7. [DOI] [PubMed] [Google Scholar]
- 40.Risk and Prevention Study Collaborative Group, Roncaglioni MC, Tombesi M, Avanzini F, Barlera S, Caimi V, Longoni P, Marzona I, Milani V, et al. n-3 Fatty acids in patients with multiple cardiovascular risk factors. N Engl J Med 2013;368:1800–8. [DOI] [PubMed]
- 41.Sekikawa A, Steingrimsdottir L, Ueshima H, Shin C, Curb JD, Evans RW, Hauksdottir AM, Kadota A, Choo J, Masaki K, et al. Serum levels of marine-derived n-3 fatty acids in Icelanders, Japanese, Koreans, and Americans–a descriptive epidemiologic study. Prostaglandins Leukot Essent Fatty Acids 2012;87:11–6. [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.