Changes in Ultrasound-Assessed Carotid Intima-Media Thickness and Plaque With a Mediterranean Diet A Substudy of the PREDIMED Trial Aleix Sala-Vila, Edwin-Saúl Romero-Mamani, Rosa Gilabert, Isabel Núñez, Rafael de la Torre, Dolores Corella, Valentina Ruiz-Gutiérrez, María-Carmen López-Sabater, Xavier Pintó, Javier Rekondo, Miguel-Ángel Martínez-González, Ramon Estruch, Emilio Ros Objective—The Prevención con Dieta Mediterránea (PREDIMED) trial showed that a Mediterranean diet (MedDiet) supplemented with either extra virgin olive oil or 30 g/d of mixed nuts reduced incident cardiovascular events compared with a control (low fat) diet. The mechanisms of cardiovascular protection afforded by MedDiets remain to be uncovered. We assessed the effect of both supplemented MedDiets on internal carotid intima-media thickness (ICA-IMT) and plaque height, the ultrasound features that best predict future cardiovascular events, in subjects at high cardiovascular risk. Approach and Results—In a PREDIMED subcohort (n=175), plaque height and carotid IMT of 3 prespecified segments (ICA, bifurcation, and common) were sonographically assessed at baseline and after intervention for a mean of 2.4 years. We evaluated 164 subjects with complete data. In a multivariate model, mean ICA-IMT progressed in the control diet group (mean [95% confidence interval], 0.052 mm [−0.014 to 0.118 mm]), whereas it regressed in the MedDiet+nuts group (−0.084 mm [−0.158 to −0.010 mm]; P=0.024 versus control). Similar results were observed for maximum ICAIMT (control, 0.188 mm [0.077 to 0.299 mm]; MedDiet+nuts, −0.030 mm [−0.153 to 0.093 mm]; P=0.034) and maximum plaque height (control, 0.106 mm [0.001 to 0.210 mm]; MedDiet+nuts, −0.091 mm [−0.206 to 0.023 mm]; P=0.047). There were no changes in ICA-IMT or plaque after the MedDiet+extra virgin olive oil. Conclusions—Compared with a control diet, consumption of a MedDiet supplemented with nuts is associated with delayed progression of ICA-IMT and plaque. The results contribute mechanistic evidence for the reduction of cardiovascular events observed in the PREDIMED trial. Clinical Trial Registration—URL: http://www.controlled-trials.com. Unique identifier: ISRCTN35739639. (Arterioscler Thromb Vasc Biol. 2014;34:439-445.) Key Words: carotid intima-media thickness ◼ carotid stenosis ◼ diet, Mediterranean ◼ nuts ◼ olive oil
C
ardiovascular disease (CVD) mortality rates in Spain and other Mediterranean areas are low compared with those of Northern Europe1,2 and the United States.2 This North-to-South gradient in CVD risk could be explained in part by local dietary habits, such as adherence to the Mediterranean diet (MedDiet). This hypothesis is supported by observational studies,3,4 a secondary prevention trial (the Lyon Diet Heart Study),5 and the recent results of the randomized controlled clinical trial Prevención con
Dieta Mediterránea (PREDIMED; www.predimed.org). In the PREDIMED study, conducted in 7447 participants at high cardiovascular risk but no CVD at enrollment, advice on the MedDiet and its supplementation with either extra virgin olive oil (EVOO) or mixed nuts reduced by 30% the incidence of a composite end point of cardiovascular death, myocardial infarction, and stroke in comparison with a control diet (advice on a low-fat diet) after intervention for a mean of 4.8 years.6
Received on: July 22, 2013; final version accepted on: November 14, 2013. From the Lipid Clinic, Endocrinology and Nutrition Service, Institut d’Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS), Hospital Clínic, Barcelona, Spain (A.S.-V., E.R.); Ciber Fisiopatología de la Obesidad y Nutrición (CIBERobn), Instituto de Salud Carlos III (ISCIII), Madrid, Spain (A.S.-V., E.-S.R.-M., I.N., R.d.l.T., D.C., V.R.-G., M.-C.L.-S., X.P., J.R., M.-Á.M.-G., R.E., E.R.); Department of Internal Medicine, IDIBAPS, Hospital Clinic, Barcelona, Spain (E.-S.R.-M., R.E.); Diagnostic Imaging Centre, IDIBAPS, Hospital Clínic, Barcelona, Spain (R.G., I.N.); Human Pharmacology and Clinical Neurosciences Research Group, Institut de Investigacions Mèdiques Hospital del Mar, Barcelona, Spain (R.d.l.T.); Department of Preventive Medicine, University of Valencia, Valencia, Spain (D.C.); Instituto de la Grasa, Consejo Superior de Investigaciones Cientificas, Sevilla, Spain (V.R.-G.); Department of Nutrition and Food Science, School of Pharmacy, XaRTA, INSA, University of Barcelona, Barcelona, Spain (M.-C.L.-S.); Lipids and Vascular Risk Unit, Internal Medicine, Hospital Universitario de Bellvitge, Hospitalet de Llobregat, Barcelona, Spain (X.P.); Department of Cardiology, University Hospital of Alava, Vitoria, Spain (J.R.); and Department of Preventive Medicine and Public Health, School of Medicine, University of Navarra– Clínica Universidad de Navarra, Pamplona, Spain (M.-Á.M.-G.). The online-only Data Supplement is available with this article at http://atvb.ahajournals.org/lookup/suppl/doi:10.1161/ATVBAHA.113.302327/-/DC1. Correspondence to Aleix Sala-Vila, PhD, Clínica de Lípids, Hospital Clínic, Villarroel 170, Edifici Helios, despatx 8, 08036 Barcelona, Spain. E-mail [email protected] © 2013 American Heart Association, Inc. Arterioscler Thromb Vasc Biol is available at http://atvb.ahajournals.org
DOI: 10.1161/ATVBAHA.113.302327
Downloaded from http://atvb.ahajournals.org/ at439 University of Pittsburgh--HSLS on March 14, 2015
440 Arterioscler Thromb Vasc Biol February 2014
Nonstandard Abbreviations and Acronyms CCA CVD EVOO ICA IMT MedDiet PREDIMED
common carotid artery cardiovascular disease extra virgin olive oil internal carotid artery intima-media thickness Mediterranean diet Prevención con Dieta Mediterránea
The biological mechanisms by which EVOO and nuts, main components of the MedDiet intervention in PREDIMED,6 protect against CVD remain to be uncovered. Both foods
have a favorable fatty acid profile.7,8 Their richness in bioactive phytochemicals with antioxidant and anti-inflammatory properties probably explains why their consumption decreases circulating inflammatory biomarkers related to atherogenesis.9 Thus, it is plausible that EVOO and nuts might reduce CVD via an antiatherogenic effect, thus delaying the development of atherosclerosis, the pathological basis of most CVD events. Artery vessel wall enlargement, an early feature of atherosclerosis, can be easily evaluated by ultrasound determination of carotid artery intima-media thickness (IMT), a noninvasive, well-standardized, and validated imaging technique.10 IMT has been usually measured in the common carotid artery (CCA) because this carotid segment is easily visualized and
Table 1. Baseline Clinical Characteristics, Lipid Profiles, Treatment Regimes, and Sonographic Variables of the Study Population by Group Allocation Control Diet (n=61)
MedDiet+EVOO (n=57)
MedDiet+Nuts (n=46)
P Value*
Male, n (%)
23 (38)
28 (49)
25 (54)
0.203
Age, y
66 (64–67)
67 (65–68)
66 (64–67)
0.481
Family history of early-onset CVD, n (%)
27 (44)
27 (47)
21 (46)
0.876
BMI, kg/m2
29.7 (28.8–30.6)
Waist circumference, cm Leisure-time physical activity, MET-min/d
29.2 (28.2–30.2)
29.9 (28.9–30.9)
0.596
99 (97–101)
100 (97–102)
101 (99–103)
0.548
274 (214–334)
282 (224–341)
300 (227–373)
0.847
Current smoker, n (%)
10 (16)
10 (18)
7 (15)
0.951
Former smoker, n (%)
19 (31)
13 (23)
18 (39)
0.200
Dyslipidemia, n (%)
50 (82)
42 (74)
33 (72)
0.402
Total cholesterol, mg/dL
206 (197–216)
216 (206–226)
212 (201–222)
0.386
HDL cholesterol, mg/dL
51 (48–54)
53 (49–57)
50 (47–53)
0.433
LDL cholesterol, mg/dL
133 (124–141)
133 (124–142)
134 (125–143)
0.974
Triacylglycerols, mg/dL
128 (101–155)
156 (119–194)
139 (112–167)
0.419
Use of statins, n (%)
34 (56)
25 (44)
20 (43)
0.328
Hypertension, n (%)
51 (84)
41 (72)
35 (76)
0.306
Systolic blood pressure, mm Hg
152 (148–157)
147 (142–152)
152 (146–158)
0.284
Diastolic blood pressure, mm Hg
83 (81–85)
79 (77–82)
80 (78–83)
0.066
32 (70)
0.661
Use of antihypertensive drugs, n (%) Fasting glucose, mg/dL
43 (71) 118 (107–130)
36 (63) 120 (109–131)
120 (109–131)
0.752
Diabetes mellitus, n (%)
20 (33)
22 (39)
16 (35)
0.801
Use of oral antidiabetic agents, n (%)
15 (25)
12 (21)
7 (15)
0.495
4 (7)
3 (5)
2 (4)
0.880
Use of insulin, n (%) Carotid ultrasound† CCA-IMTmean, mm
0.79 (0.75–0.83)
0.76 (0.72–0.80)
0.81 (0.77–0.86)
0.247 0.543
CCA-IMTmax, mm
0.95 (0.88–1.02)
0.97 (0.90–1.04)
1.01 (0.93–1.09)
BIF-IMTmean, mm
0.97 (0.91–1.04)
0.92 (0.85–0.99)
0.98 (0.90–1.06)
0.431
BIF-IMTmax, mm
1.38 (1.25–1.52)
1.34 (1.19–1.48)
1.44 (1.28–1.59)
0.680
ICA-IMTmean, mm‡
0.74 (0.66–0.82)
0.73 (0.65–0.82)
0.82 (0.73–0.92)
0.297
ICA-IMTmax, mm‡
1.01 (0.86–1.16)
1.05 (0.89–1.20)
1.12 (0.95–1.30)
0.608
Plaquemax, mm§
2.10 (1.88–2.32)
1.89 (1.67–2.12)
2.13 (1.88–2.39)
0.308
Data are expressed as mean (95% confidence interval) except for quantitative variables, expressed as n (%). BIF-IMT indicates intima-media thickness of bifurcation; BMI, body mass index; CCA-IMT, intima-media thickness of common carotid artery; CVD, cardiovascular disease; EVOO, extra virgin olive oil; HDL, high-density lipoprotein; ICA-IMT, intima-media thickness of internal carotid artery; LDL, low-density lipoprotein; MedDiet, Mediterranean diet; and MET-min, minutes at a given metabolic equivalent level (units of energy expenditure in physical activity, 1 MET-min is roughly equivalent to 1 kcal). *χ2 test and ANOVA, as appropriate. †Adjusted for sex, age, ever smoking, BMI, energy intake, use of statins, use of antidiabetic drugs, and use of antihypertensive drugs. ‡Measured in 155 participants (n=58, n=55, and n=42, respectively). §Measured in subjects with focal intrusions into the lumen ≥1.2-mm thick (n=46, n=46, and n=34, respectively).
Downloaded from http://atvb.ahajournals.org/ at University of Pittsburgh--HSLS on March 14, 2015
Sala-Vila et al MedDiet and Carotid Atherosclerosis 441 Table 2. Baseline Intake of Energy and Nutrients and Changes by Intervention Group Control Diet (n=61)
MedDiet+EVOO (n=57)
MedDiet+Nuts (n=46)
P Value*
Energy, kcal/d Baseline
2387 (2273 to 2501)a
2173 (2054 to 2292)b
2407 (2276 to 2538)a
0.013
Change
−444 (−549 to −339)
−187 (−297 to −77)
−288 (−409 to −166)
0.005
a
b
Carbohydrates, g/d Baseline
237.8 (229.6 to 245.9)
234.1 (225.5 to 242.6)
232.9 (223.5 to 242.3)
0.711
Change
−34.1 (−42.0 to −26.2)a
−42.0 (−50.4 to −33.7)
−49.5 (−58.7 to −40.3)b
0.045
Protein, g/d Baseline
94.3 (90.8 to 97.9)
98.5 (94.8 to 102.2)a
31.5 (87.4 to 95.5)b
0.043
Change
−11.1 (−14.2 to −8.0)
−8.0 (−11.2 to −4.7)
−8.8 (−12.3 to −5.2)
0.364
Fat, g/d Baseline
101.3 (98.0 to 104.5)
104.6 (101.2 to 108.0)
104.7 (101.0 to 108.5)
0.265
Change
−14.1 (−17.5 to −10.6)a
−10.7 (−14.3 to −7.0)
−7.2 (−11.3 to −3.2)b
0.040
Saturated fatty acids, g/d Baseline
26.5 (25.1 to 27.8)
27.8 (26.4 to 29.2)
27.4 (25.8 to 28.9)
0.377
Change
−5.5 (−6.9 to −4.2)
−5.6 (−7.0 to −4.2)
−5.6 (−7.2 to −4.1)
0.992
Monounsaturated fatty acids, g/d Baseline
52.4 (50.4 to 54.5)
54.5 (52.4 to 56.7)
55.6 (53.3 to 58.0)
0.113
Change
−6.1 (−8.3 to −3.9)
−3.9 (−6.3 to −1.6)
−4.6 (−7.2 to −2.0)
0.403
15.4 (14.4 to 16.4)
0.559
Polyunsaturated fatty acids, g/d Baseline
15.2 (14.3 to 16.1)
Change
−2.9 (−3.9 to −2.0)
14.7 (13.7 to 15.6) a
−1.7 (−2.7 to −0.7)
a
1.7 (0.6 to 2.8)
b
C
ardiovascular disease (CVD) mortality rates in Spain and other Mediterranean areas are low compared with those of Northern Europe1,2 and the United States.2 This North-to-South gradient in CVD risk could be explained in part by local dietary habits, such as adherence to the Mediterranean diet (MedDiet). This hypothesis is supported by observational studies,3,4 a secondary prevention trial (the Lyon Diet Heart Study),5 and the recent results of the randomized controlled clinical trial Prevención con
Dieta Mediterránea (PREDIMED; www.predimed.org). In the PREDIMED study, conducted in 7447 participants at high cardiovascular risk but no CVD at enrollment, advice on the MedDiet and its supplementation with either extra virgin olive oil (EVOO) or mixed nuts reduced by 30% the incidence of a composite end point of cardiovascular death, myocardial infarction, and stroke in comparison with a control diet (advice on a low-fat diet) after intervention for a mean of 4.8 years.6
Received on: July 22, 2013; final version accepted on: November 14, 2013. From the Lipid Clinic, Endocrinology and Nutrition Service, Institut d’Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS), Hospital Clínic, Barcelona, Spain (A.S.-V., E.R.); Ciber Fisiopatología de la Obesidad y Nutrición (CIBERobn), Instituto de Salud Carlos III (ISCIII), Madrid, Spain (A.S.-V., E.-S.R.-M., I.N., R.d.l.T., D.C., V.R.-G., M.-C.L.-S., X.P., J.R., M.-Á.M.-G., R.E., E.R.); Department of Internal Medicine, IDIBAPS, Hospital Clinic, Barcelona, Spain (E.-S.R.-M., R.E.); Diagnostic Imaging Centre, IDIBAPS, Hospital Clínic, Barcelona, Spain (R.G., I.N.); Human Pharmacology and Clinical Neurosciences Research Group, Institut de Investigacions Mèdiques Hospital del Mar, Barcelona, Spain (R.d.l.T.); Department of Preventive Medicine, University of Valencia, Valencia, Spain (D.C.); Instituto de la Grasa, Consejo Superior de Investigaciones Cientificas, Sevilla, Spain (V.R.-G.); Department of Nutrition and Food Science, School of Pharmacy, XaRTA, INSA, University of Barcelona, Barcelona, Spain (M.-C.L.-S.); Lipids and Vascular Risk Unit, Internal Medicine, Hospital Universitario de Bellvitge, Hospitalet de Llobregat, Barcelona, Spain (X.P.); Department of Cardiology, University Hospital of Alava, Vitoria, Spain (J.R.); and Department of Preventive Medicine and Public Health, School of Medicine, University of Navarra– Clínica Universidad de Navarra, Pamplona, Spain (M.-Á.M.-G.). The online-only Data Supplement is available with this article at http://atvb.ahajournals.org/lookup/suppl/doi:10.1161/ATVBAHA.113.302327/-/DC1. Correspondence to Aleix Sala-Vila, PhD, Clínica de Lípids, Hospital Clínic, Villarroel 170, Edifici Helios, despatx 8, 08036 Barcelona, Spain. E-mail [email protected] © 2013 American Heart Association, Inc. Arterioscler Thromb Vasc Biol is available at http://atvb.ahajournals.org
DOI: 10.1161/ATVBAHA.113.302327
Downloaded from http://atvb.ahajournals.org/ at439 University of Pittsburgh--HSLS on March 14, 2015
440 Arterioscler Thromb Vasc Biol February 2014
Nonstandard Abbreviations and Acronyms CCA CVD EVOO ICA IMT MedDiet PREDIMED
common carotid artery cardiovascular disease extra virgin olive oil internal carotid artery intima-media thickness Mediterranean diet Prevención con Dieta Mediterránea
The biological mechanisms by which EVOO and nuts, main components of the MedDiet intervention in PREDIMED,6 protect against CVD remain to be uncovered. Both foods
have a favorable fatty acid profile.7,8 Their richness in bioactive phytochemicals with antioxidant and anti-inflammatory properties probably explains why their consumption decreases circulating inflammatory biomarkers related to atherogenesis.9 Thus, it is plausible that EVOO and nuts might reduce CVD via an antiatherogenic effect, thus delaying the development of atherosclerosis, the pathological basis of most CVD events. Artery vessel wall enlargement, an early feature of atherosclerosis, can be easily evaluated by ultrasound determination of carotid artery intima-media thickness (IMT), a noninvasive, well-standardized, and validated imaging technique.10 IMT has been usually measured in the common carotid artery (CCA) because this carotid segment is easily visualized and
Table 1. Baseline Clinical Characteristics, Lipid Profiles, Treatment Regimes, and Sonographic Variables of the Study Population by Group Allocation Control Diet (n=61)
MedDiet+EVOO (n=57)
MedDiet+Nuts (n=46)
P Value*
Male, n (%)
23 (38)
28 (49)
25 (54)
0.203
Age, y
66 (64–67)
67 (65–68)
66 (64–67)
0.481
Family history of early-onset CVD, n (%)
27 (44)
27 (47)
21 (46)
0.876
BMI, kg/m2
29.7 (28.8–30.6)
Waist circumference, cm Leisure-time physical activity, MET-min/d
29.2 (28.2–30.2)
29.9 (28.9–30.9)
0.596
99 (97–101)
100 (97–102)
101 (99–103)
0.548
274 (214–334)
282 (224–341)
300 (227–373)
0.847
Current smoker, n (%)
10 (16)
10 (18)
7 (15)
0.951
Former smoker, n (%)
19 (31)
13 (23)
18 (39)
0.200
Dyslipidemia, n (%)
50 (82)
42 (74)
33 (72)
0.402
Total cholesterol, mg/dL
206 (197–216)
216 (206–226)
212 (201–222)
0.386
HDL cholesterol, mg/dL
51 (48–54)
53 (49–57)
50 (47–53)
0.433
LDL cholesterol, mg/dL
133 (124–141)
133 (124–142)
134 (125–143)
0.974
Triacylglycerols, mg/dL
128 (101–155)
156 (119–194)
139 (112–167)
0.419
Use of statins, n (%)
34 (56)
25 (44)
20 (43)
0.328
Hypertension, n (%)
51 (84)
41 (72)
35 (76)
0.306
Systolic blood pressure, mm Hg
152 (148–157)
147 (142–152)
152 (146–158)
0.284
Diastolic blood pressure, mm Hg
83 (81–85)
79 (77–82)
80 (78–83)
0.066
32 (70)
0.661
Use of antihypertensive drugs, n (%) Fasting glucose, mg/dL
43 (71) 118 (107–130)
36 (63) 120 (109–131)
120 (109–131)
0.752
Diabetes mellitus, n (%)
20 (33)
22 (39)
16 (35)
0.801
Use of oral antidiabetic agents, n (%)
15 (25)
12 (21)
7 (15)
0.495
4 (7)
3 (5)
2 (4)
0.880
Use of insulin, n (%) Carotid ultrasound† CCA-IMTmean, mm
0.79 (0.75–0.83)
0.76 (0.72–0.80)
0.81 (0.77–0.86)
0.247 0.543
CCA-IMTmax, mm
0.95 (0.88–1.02)
0.97 (0.90–1.04)
1.01 (0.93–1.09)
BIF-IMTmean, mm
0.97 (0.91–1.04)
0.92 (0.85–0.99)
0.98 (0.90–1.06)
0.431
BIF-IMTmax, mm
1.38 (1.25–1.52)
1.34 (1.19–1.48)
1.44 (1.28–1.59)
0.680
ICA-IMTmean, mm‡
0.74 (0.66–0.82)
0.73 (0.65–0.82)
0.82 (0.73–0.92)
0.297
ICA-IMTmax, mm‡
1.01 (0.86–1.16)
1.05 (0.89–1.20)
1.12 (0.95–1.30)
0.608
Plaquemax, mm§
2.10 (1.88–2.32)
1.89 (1.67–2.12)
2.13 (1.88–2.39)
0.308
Data are expressed as mean (95% confidence interval) except for quantitative variables, expressed as n (%). BIF-IMT indicates intima-media thickness of bifurcation; BMI, body mass index; CCA-IMT, intima-media thickness of common carotid artery; CVD, cardiovascular disease; EVOO, extra virgin olive oil; HDL, high-density lipoprotein; ICA-IMT, intima-media thickness of internal carotid artery; LDL, low-density lipoprotein; MedDiet, Mediterranean diet; and MET-min, minutes at a given metabolic equivalent level (units of energy expenditure in physical activity, 1 MET-min is roughly equivalent to 1 kcal). *χ2 test and ANOVA, as appropriate. †Adjusted for sex, age, ever smoking, BMI, energy intake, use of statins, use of antidiabetic drugs, and use of antihypertensive drugs. ‡Measured in 155 participants (n=58, n=55, and n=42, respectively). §Measured in subjects with focal intrusions into the lumen ≥1.2-mm thick (n=46, n=46, and n=34, respectively).
Downloaded from http://atvb.ahajournals.org/ at University of Pittsburgh--HSLS on March 14, 2015
Sala-Vila et al MedDiet and Carotid Atherosclerosis 441 Table 2. Baseline Intake of Energy and Nutrients and Changes by Intervention Group Control Diet (n=61)
MedDiet+EVOO (n=57)
MedDiet+Nuts (n=46)
P Value*
Energy, kcal/d Baseline
2387 (2273 to 2501)a
2173 (2054 to 2292)b
2407 (2276 to 2538)a
0.013
Change
−444 (−549 to −339)
−187 (−297 to −77)
−288 (−409 to −166)
0.005
a
b
Carbohydrates, g/d Baseline
237.8 (229.6 to 245.9)
234.1 (225.5 to 242.6)
232.9 (223.5 to 242.3)
0.711
Change
−34.1 (−42.0 to −26.2)a
−42.0 (−50.4 to −33.7)
−49.5 (−58.7 to −40.3)b
0.045
Protein, g/d Baseline
94.3 (90.8 to 97.9)
98.5 (94.8 to 102.2)a
31.5 (87.4 to 95.5)b
0.043
Change
−11.1 (−14.2 to −8.0)
−8.0 (−11.2 to −4.7)
−8.8 (−12.3 to −5.2)
0.364
Fat, g/d Baseline
101.3 (98.0 to 104.5)
104.6 (101.2 to 108.0)
104.7 (101.0 to 108.5)
0.265
Change
−14.1 (−17.5 to −10.6)a
−10.7 (−14.3 to −7.0)
−7.2 (−11.3 to −3.2)b
0.040
Saturated fatty acids, g/d Baseline
26.5 (25.1 to 27.8)
27.8 (26.4 to 29.2)
27.4 (25.8 to 28.9)
0.377
Change
−5.5 (−6.9 to −4.2)
−5.6 (−7.0 to −4.2)
−5.6 (−7.2 to −4.1)
0.992
Monounsaturated fatty acids, g/d Baseline
52.4 (50.4 to 54.5)
54.5 (52.4 to 56.7)
55.6 (53.3 to 58.0)
0.113
Change
−6.1 (−8.3 to −3.9)
−3.9 (−6.3 to −1.6)
−4.6 (−7.2 to −2.0)
0.403
15.4 (14.4 to 16.4)
0.559
Polyunsaturated fatty acids, g/d Baseline
15.2 (14.3 to 16.1)
Change
−2.9 (−3.9 to −2.0)
14.7 (13.7 to 15.6) a
−1.7 (−2.7 to −0.7)
a
1.7 (0.6 to 2.8)
b
