| 31) Leosdottir M (2007) | The Malmö Diet and Cancer Study | 28,098 middle-aged men and women without CVD.
(Sweden) | 8.4
(1991-96 to 2002) | 1,556?
(583 women, and
973 men) | Cardiovascular events | Total fat (not defined, but including trans fatty acids) |
| Women | Men |
HR = 0.98 (0.77-1.25; P = 0.8) for the highest vs lowest quartile of consumption.
Amount specific data (% energy intake):
30.8: HR = 1.
36.5: HR = 0.88 (0.69-1.12).
40.3: HR = 0.87 (0.68-1.11).
46.2: HR = 0.98 (0.77-1.25).
|
HR = 1.02 (0.84-1.23; P = 0.8) for the highest vs lowest quartile of consumption.
Amount specific data (% energy intake):
31.8: HR = 1.
37.8: HR = 1.03 (0.86-1.24).
41.7: HR = 1.06 (0.88-1.28).
47.7: HR = 1.02 (0.84-1.23).
|
Age, smoking habits, alcohol consumption, socioeconomic status, marital status, physical activity, BMI, fibre intake, and blood pressure. |
| 31) Leosdottir M (2005) | The Malmö Diet and Cancer Study | 28,098 Swedish middle-aged individuals (11,063 men, and 17,035 women). | 6.6
(1991-96 to 2000) | See variables | Cardiovascular mortality | Total fat (not defined) |
Women
(97? cases) | Men
(242? cases) |
RR = 0.74 (0.40-1.36; P = 0.25) for the highest vs lowest quartile of consumption.
Amount specific data (% of daily energy):
30.8: RR = 1.
36.5: RR = 0.99 (0.57-1.72).
40.3: RR = 0.80 (0.45-1.43).
46.1: RR = 0.74 (0.40-1.36).
|
RR = 0.65 (0.45-0.94; P = 0.03) for the highest vs lowest quartile of consumption.
Amount specific data (% of daily energy):
31.7: RR = 1.
37.8: RR = 0.76 (0.53-1.09).
41.7: RR = 0.74 (0.52-1.07).
47.7: RR = 0.65 (0.45-0.94; P = 0.023).
|
Age, alcohol consumption, smoking, social class, marital status, physical activity, BMI and fibre intake. |
| 28) Erkkilä AT (2003) | The Finnish cohort of the EUROASPIRE Study | 415 subjects (285 men and 130 women) aged 33-74 with CAD admitted to Kuopio University Hospital. | 5 | 44 | CVD risk (CVD death, AMI, or stroke | Fat (not defined) | RR = 1.22 (0.89-1.67; P = 0.216) per 1-SD increment in intake. | Age, sex, diagnostic category, energy intake, serum cholesterol, serum triaglycerol, diabetes, BMI, and education. |
| 13) Tanasescu M (2004) | The Nurses' Health Study | 5,672 women with type 2 diabetes, but without CVD or cancer. | 1980 to 1994-98 | 619? | CVD events (nonfatal myocardial infarction, fatal coronary heart disease, and stroke) | Total fat (not defined) |
RR = 1.09 (0.81-1.47; P = 0.56) for the highest vs lowest quintile of consumption.
Amount specific data (% of energy):
29.3: RR = 1.
34.3: RR = 1.11 (0.86-1.44).
37.8: RR = 1.11 (0.85-1.45).
41.3: RR = 1.17 (0.88-1.54).
47.0: RR = 1.09 (0.81-1.47). | Age, smoking, postmenopausal hormone use, parental history of MI before age 60, alcohol intake, moderate vigorous activities, BMI, total caloric intake, protein intake, fiber intake, multivitamin use, vitamin E supplement use, medication use, and cholesterol. |
Prospective studies of dietary fat and coronary heart disease:
| Author | Cohort name | Subjects | Years of follow-up | Cases | End point | Consumption of | Relative Risk (RR) | Adjustments |
| 44) Corella D (2010) | The Spanish Cohort of EPIC-Heart
(The Spanish cardiovascular part of The EPIC Study) | 41,440 healthy subjects (15,632 men, and 25,806 women) aged 30-69, and without CHD.
(Nested case-control) | 10
(1992-96 to 2004) | 534 | CHD incidence (fatal or nonfatal myocardial infarction or angina requiring a revascularization procedure) | Total fat (not defined) |
| g/day: | % energy: | consumption > 30% energy: |
| Cases consumed less fat (97.8 g/day) than controls (101.8 g/day; P = 0.034).
|
Cases consumed nonsignificantly less fat (35.7% energy) than controls (36.3% energy; P = 0.066).
|
The same amount of cases consumed > 30% energy from fat (82.8%) as controls (85.5; P = 0.15).
|
Unadjusted.
Controls were matched to cases by center, sex, age, and time of enrollment. |
| 37) Nettleton JA (2008) | The Atherosclerosis Risk In Communities (ARIC) Study. | 14,153 African-American and white adults age 45-64.
(USA) | 13.3
(1987-89 to 2003) | 1,140? (639 men, 501 women) | Incident heart failure | Total fat (not defined) | Cases consumed more total fat (33.2% of energy) than controls (32.9% of energy; P = 0.086). | Total energy. |
| 35) Xu J (2006) | The Strong Heart Study | 2,938 American Indians aged 47-79, and free of CHD, dialysis treatment, liver cirrhosis, or having had a kidney transplant.
(USA) | 7.2
(1993-95 to 2002) | See variables | First CHD event (nonfatal CHD [definite MI, definite CHD, and electrocardiogram-evident definite MI], and fatal CHD [death from definite MI, definite sudden death due to CHD, definite fatal CHD, and possible fatal CHD]) | Total fat (not defined) |
CHD event
(403 cases) | Nonfatal CHD
(298 cases) |
HR = 1.03 (0.77-1.40; P = 0.97) for the highest vs lowest quartile of consumption.
Amount specific data (% of energy):
24.0: HR = 1.
32.3: HR = 1.04 (0.78-1.40).
38.4: HR = 0.91 (0.67-1.23).
45.9: HR = 1.03 (0.77-1.40).
|
HR = 1.12 (0.79-1.59; P = 0.71) for the highest vs lowest quartile of consumption.
Amount specific data (% of energy):
24.0: HR = 1.
32.3: HR = 1.09 (0.77-1.54).
38.4: HR = 0.92 (0.64-1.31).
45.9: HR = 1.12 (0.79-1.59).
|
CHD death.
Stratified by age:
47-59
(46 cases) | 60-79
(92 cases) |
HR = 3.57 (1.21-10.49; P = 0.01) for the highest vs lowest quartile of consumption.
Amount specific data (% of energy):
24.8: HR = 1.
33.0: HR = 1.44 (0.45-4.58).
39.1: HR = 2.42 (0.83-7.06).
46.6: HR = 3.57 (1.21-10.49).
Omitting HDL cholesterol and LDL cholesterol from the model did not change the results (data not shown).
|
HR = 0.77 (0.41-1.45; P = 0.24) for the highest vs lowest quartile of consumption.
Amount specific data (% of energy):
23.0: HR = 1.
31.2: HR = 1.20 (0.67-2.17).
37.3: HR = 0.73 (0.38-1.40).
44.7: HR = 0.77 (0.41-1.45).
|
Effect modification:
No significant interaction was found for CHD death and sex or diabetes. Or for CHD risk (either nonfatal or fatal) between age groups and all other confounders.Variable of interest as a % of energy, sex, age, study center, diabetes status, BMI, HDL, LDL, triaglycerol, smoking, alcohol, hypertension, protein, and total energy. |
| 34) Ness AR (2005) | The Boyd Orr Cohort | 4,028 people (1,995 men, and 2,033 women) from England and Scotland. | average 37
(1948 to 2000) | 298 | CHD mortality | Fat (not defined) |
RR = 0.56 (0.30-1.08; P = 0.07) for the highest vs lowest quartile of consumption.
Amount specific data (g):
30.1-63.3: RR = 1.
63.3-77.6: RR = 0.83 (0.54-1.26).
77.7-96.4: RR = 0.67 (0.41-1.09).
93.5-229.0: RR = 0.56 (0.30-1.08).
Cut-off points for quintile 4-5 were incorrectly defined in the original article. | Age, energy, sex, childhood family food expenditure, father's social class, district of residence as a child, period of birth, season when studied as a child, and Townsend score for current address or place of death. |
| 31) Leosdottir M (2007) | The Malmö Diet and Cancer Study | 28,098 middle-aged men and women without CVD.
(Sweden) | 8.4
(1991-96 to 2002) | 908?
(no data about men vs women) | Acute coronary events | Total fat (not defined, but including trans fatty acids) |
| Women | Men |
HR = 0.77 (0.55-1.08; P = 0.1) for the highest vs lowest quartile of consumption.
Amount specific data (% energy intake):
30.8: HR = 1.
36.5: HR = 0.81 (0.59-1.11).
40.3: HR = 0.72 (0.52-1.01).
46.2: HR = 0.77 (0.55-1.08).
|
HR = 0.99 (0.78-1.25; P = 0.8) for the highest vs lowest quartile of consumption.
Amount specific data (% energy intake):
31.8: HR = 1.
37.8: HR = 1.04 (0.83-1.30).
41.7: HR = 0.96 (0.76-1.21).
47.7: HR = 0.99 (0.78-1.25).
|
Age, smoking habits, alcohol consumption, socioeconomic status, marital status, physical activity, BMI, fibre intake, and blood pressure. |
| 29) Jakobsen MU (2004) | 4 different cohorts
(The 1914 cohort; The 1936 cohort, The MONICA I, and the MONICA III) | 3,686 subjects (1,837 men and 1,849 women) aged 30-71, and free of diabetes or CHD.
(Denmark) | 16
(1964-93 to 1998) | 228? men, and
98? women | CHD events (fatal and nonfatal heart disease) | Total fat (not defined) |
Stratified by gender:
| Women | Men |
HR = 1.10 (0.95-1.28) for an intake of a 5% higher level of energy.
HR = 1.12 (0.93-1.36) for an intake of a 5% higher level of energy instead of carbohydrates.
|
HR = 1.00 (0.91-1.10) for an intake of a 5% higher level of energy.
HR = 0.98 (0.87-1.10) for an intake of a 5% higher level of energy instead of carbohydrates.
|
Stratified by gender and age:
| Women < 60 years | Women ≥ 60 years | Men < 60 years | Men ≥ 60 years |
HR = 1.66 (1.12-2.47) for an intake of a 5% higher level of energy.
HR = 1.74 (1.15-2.64) for an intake of a 5% higher level of energy instead of carbohydrates.
|
HR = 1.03 (0.88-1.21) for an intake of a 5% higher level of energy.
HR = 1.05 (0.86-1.28) for an intake of a 5% higher level of energy instead of carbohydrates.
|
HR = 1.17 (0.96-1.42) for an intake of a 5% higher level of energy.
HR = 1.15 (0.93-1.41) for an intake of a 5% higher level of energy instead of carbohydrates.
|
HR = 0.96 (0.86-1.07) for an intake of a 5% higher level of energy.
HR = 0.93 (0.81-1.06) for an intake of a 5% higher level of energy instead of carbohydrates.
|
Variable of interest, total energy intake, and cohort identification.
Fat for carbohydrates was also adjusted for % energy from protein, % of energy from the other major types of fatty acids, familial history of MI, smoking, leisure time physical activity, educational attainment, alcohol, dietary fiber, and dietary cholesterol. |
| 28) Erkkilä AT (2003) | The Finnish cohort of the EUROASPIRE Study | 415 subjects (285 men and 130 women) aged 33-74 with CAD admitted to Kuopio University Hospital. | 5 | See variables | CAD | Fat (not defined) |
CAD death
(16 cases) | CAD death or AMI
(34 cases) |
| RR = 1.03 (0.63-1.70; P = 0.902) per 1-SD increment in intake.
|
RR = 1.05 (0.73-1.52; P = 0.799) per 1-SD increment in intake.
|
Age, sex, diagnostic category, energy intake, serum cholesterol, serum triaglycerol, diabetes, BMI, and education. |
| 28) Erkkilä AT (2003) | The Finnish cohort of the EUROASPIRE Study | 415 subjects (285 men and 130 women) aged 33-74 with CAD admitted to Kuopio University Hospital. | 5 | 38 | Revascularization | Fat (not defined) | RR = 1.31 (0.94-1.82; P = 0.113) per 1-SD increment in intake. | Age, sex, diagnostic category, energy intake, serum cholesterol, serum triaglycerol, diabetes, BMI, and education. |
| 26) Soinio M (2003) | No cohort name defined | 366 men and 295 women with diabetes type 2, but free from CHD, aged 45-64 from Kuopio.
(Finland) | 7
(1982-84 to 1989) | See variables | CHD (nonfatal MI or CHD death) | Total fat (not defined) |
CHD death:
Men
(41 cases) | Women
(24 cases) |
| No significant difference in consumption in % of energy was found between cases (38.9) and controls (39.6).
|
Cases consumed a smaller % of energy from fat (36.3) than controls (39.3; P = 0.016).
|
CHD: Women with CHD had a lower reported intake of total fat (P = < 0.001; 43 cases. No data shown).Unadjusted. |
| 24) Boniface DR (2002) | The Health and Lifestyle Survey | 1,225 men and 1,451 women aged 40-75, and without heart disease, diabetes, anti-hypertensive treatment, or being on a special diet.
(Great Britain) | 16
(1984-85 to 2000) | See variables | CHD death | Total fat (not defined) |
Men
(98 cases) | Women
(57 cases) |
16 y death rate = 8.2% (5.1-12.3; P = 0.19) for the highest vs lowest quintile of consumption.
Amount specific data (Quintiles not defined):
Q1: DR = 6.6% (3.8-10.5).
Q2: DR = 6.2% (3.5-10.0).
Q3: DR = 8.8% (5.6-13.0).
Q4: DR = 10.2% (6.7-14.8).
Q5: DR = 8.2% (5.1-12.3).
RR = 1.01 (0.93-1.10; P = 0.85) for a 100 g/week increase, after adjustment for age, alcohol, smoking, exercise, and social class.
|
16 y death rate = 5.2% (2.9-8.4; P = 0.0025) for the highest vs lowest quintile of consumption.
Amount specific data (Quintiles not defined):
Q1: DR = 1.4% (0.4-3.5).
Q2: DR = 2.7% (1.2-5.3).
Q3: DR = 4.1% (2.2-7.1).
Q4: DR = 6.2% (3.7-9.6).
Q5: DR = 5.2% (2.9-8.4).
RR = 1.19 (1.03-1.37; P = 0.018) for a 100 g/week increase, after adjustment for age, alcohol, smoking, exercise, and social class (56 cases).
|
Unadjusted. |
| 22) Bostick RM (1999) | The Iowa Women's Health Study | 34,486 women aged 55-69 and without a history of IHD. | 8
(1986-1994) | 387? | Ischemic heart disease mortality | Total fat (not defined) | Cases and noncases did not differ at P ≤ 0.05 in intake of total fat (no data shown). | Unadjusted. |
| 18) Esrey KL (1996) | The Lipid Research Clinics Prevalence Follow-Up Study | 4,546 men and women aged 30-79 and free of CVD.
(USA) | 12.4
(1972-76 to 1987) | See variables | CHD mortality | Total fat (not defined) |
Stratified by age
30-59
(52? cases) | 60-79
(40? cases) |
| RR = 1.04 (1.01-1.08; P = < 0.05) for fat as 1% of total energy.
|
RR = 0.99 (0.95-1.03) for fat as 1% of total energy.
|
Age, sex, energy intake, serum lipids, systolic blood pressure, cigarette smoking status, BMI, and glucose intolerance. |
| 16) Kromhout D (1995) | Cohort name not defined | 272 subjects (137 men and 135 women) aged 64-87 from a general practice in Rotterdam.
(The Netherlands) | 17
(1971-1987) | 58? | Primary & secondary CHD mortality | Total fat (not defined) | No association was found with CHD mortality (no data shown). | Age, gender, and prevalence of myocardial infarction and angina pectoris. |
| 15) Knekt P (1994) | The Finnish Mobile Clinic Health Cohort | 5,133 men and women aged 30-69, and free of heart disease. | 14
(1968-72 to 1984) | 186? men, and 58? women. | CHD mortality | Total fat (not defined) |
| Men: | Women: |
| Cases consumed nonsignificantly more (136 g) than controls (129 g; P = 0.06).
|
Cases did not consume significantly more (92 g) than controls (88 g; P = 0.40).
|
Effect modification: No significant interaction was found with smoking, hypertension, and serum cholesterol (data not shown).Age. |
| 14) Fehily AM (1993) | The Caerphilly Study | 2,423 men aged 45-59 from South Wales. | 5
(1983-?) | 137 | Incident ischaemic heart disease (IHD death, myocardial infarction) | Total fat (not defined, but including fat from meats, butter, lard, milk, cream, cheese, and eggs) |
All men:
(137 cases) | Men with no evidence of IHD at baseline:
(70 cases) | Men with IHD at baseline
(including angina, MI, ischaemia):
(67 cases) |
RO = 1.3 (P = NS) for the highest vs lowest quintile of consumption.
Amount specific data (% of energy from fat):
≤ 34.2: RO = 1.0
34.3-38.1: RO = 1.3
38.2-41.6: RO = 1.1
41.7-45.8: RO = 2.2
> 45.8: RO = 1.3
% of energy: Cases did not consume significantly more (40.9% of energy) than controls (40.1% of energy).
|
RO = 1.3 for the highest vs lowest quintile of consumption.
Amount specific data (% of energy from fat):
≤ 34.2: RO = 1.0
34.3-38.1: RO = 1.5
38.2-41.6: RO = 1.0
41.7-45.8: RO = 1.3
> 45.8: RO = 1.3
|
RO = 1.4 for the highest vs lowest quintile of consumption.
Amount specific data (% of energy from fat):
≤ 34.2: RO = 1.0
34.3-38.1: RO = 1.1
38.2-41.6: RO = 1.3
41.7-45.8: RO = 2.7
> 45.8: RO = 1.4
|
Age, BMI, smoking, and evidence of IHD at baseline. |
| 13) Halton TL (2006) | The Nurses' Health Study | 82,802 women aged 34-59, and without diabetes, cancer, or cardiovascular disease. | 20
(1980-2000) | 1,994? | Coronary heart disease incidence (nonfatal myocardial infarctions or fatal coronary events) | Total fat (not defined) |
RR = 0.99 (0.79-1.23; P = 0.86) for the highest vs lowest decile of consumption.
Amount specific data (Deciles not defined):
D1: RR = 1.
D2: RR = 1.18 (0.99-1.42).
D3: RR = 1.02 (0.84-1.23).
D4: RR = 1.04 (0.86-1.26).
D5: RR = 0.99 (0.81-1.20).
D6: RR = 1.07 (0.88-1.30).
D7: RR = 1.10 (0.88-1.30).
D8: RR = 1.03 (0.84-1.26).
D9: RR = 1.11 (0.91-1.36).
D10: RR = 0.99 (0.79-1.23). | Age, BMI, smoking status, postmenopausal hormone use, hours of physical activity, alcohol intake, aspirine use, use of multivitamins, use of vitamin E supplement use, history of hypertension, history of hypercholesterolemia, parental history of myocardial infarction, protein, and total calories. |
| 13) Oh K (2005) | The Nurses' Health Study | 78,778 women free of CVD, diabetes, hypercholesterolemia, or cancer. | 20
(1980-2000) | 1,766? | CHD incidence (nonfatal myocardial infarction and CHD death) | Total fat (not defined) |
RR = 0.92 (0.77-1.09; P = 0.49) for the highest vs lowest quintile of consumption.
Amount specific data (% of energy):
28.3: RR = 1.
32.6: RR = 0.94 (0.81-1.08).
35.6: RR = 0.91 (0.79-1.06).
38.7: RR = 0.98 (0.84-1.15).
44.0: RR = 0.92 (0.77-1.09).
No effect modification was found by age or BMI. | Age, BMI, smoking, alcohol intake, parental history of MI, history of hypertension, menopausal status and hormone use, aspirin use, multivitamin use, vitamin E supplement use, physical activity, and energy, protein, and cholesterol intake. |
| 13) Hu FB (1999) | The Nurses' Health Study | 80,082 women aged 34-59, and without CHD, stroke, cancer, hypercholesterolemia, or diabetes. | 14
(1980-1994) | 939? | Coronary heart disease (nonfatal myocardial infarction and fatal CHD) | Total fat (not defined) | RR from the standard multivariate model:
RR = 1.03 (0.98-1.08) for an increase of 10 g. | Age, time period, BMI, cigarette smoking, menopausal status, parental history of myocardial infarction, multiple vitamin use, vitamin E supplement use, alcohol consumption, history of hypertension, aspirin use, vigorous exercise, cholesterol intake, and total energy intake. |
| 13) Hu FB (1997) | The Nurses' Health Study | 80,082 women aged 34-59 and without CVD, diabetes, hypercholesterolemia, or cancer. | 14
(1980-1994) | 939? | Coronary heart disease incidence (nonfatal myocardial infarction or fatal coronary disease) | Total fat (not defined) |
RR = 1.04 (0.83-1.28; P = 0.50) for the highest vs lowest quintile of consumption.
Amount specific data (% of energy):
29.1: RR = 1.
33.9: RR = 0.91 (0.74-1.13).
37.1: RR = 1.01 (0.82-1.25).
40.6: RR = 1.03 (0.83-1.27).
46.1: RR = 1.04 (0.83-1.28). | Age, time period, BMI, cigarette smoking, menopausal status, parental history of MI before age 65, multivitamin use, vitamin E supplement use, alcohol consumption, history of hypertension, aspirin use, vigorous exercise, % energy from protein, and dietary cholesterol. |
| 13) Ascherio A (1996) | The Health Professionals Follow Up Study | 43,757 men aged 40-75.
(USA) | 6
(1986-1992) | See variables | Coronary heart disease | Total fat (not defined) |
Total MI (including non-fatal MI and fatal cHD):
(734 cases) | Fatal CHD:
(229 cases) |
RR = 1.02 (0.78-1.34; P = 0.42) for the highest vs lowest quintile of consumption.
Amount specific data (g/day):
118: RR = 1.
136: RR = 1.00 (0.77-1.29).
149: RR = 1.05 (0.82-1.36).
160: RR = 1.07 (0.82-1.39).
171: RR = 1.02 (0.78-1.34).
RR for a proportion of energy of 5%:
RR = 1.01 (0.94-1.08).
|
RR = 1.22 (0.75-2.00; P = 0.31) for the highest vs lowest quintile of consumption.
Amount specific data (g/day):
118: RR = 1.
136: RR = 0.93 (0.57-1.51).
149: RR = 1.34 (0.84-2.12).
160: RR = 1.26 (0.79-2.03).
171: RR = 1.22 (0.75-2.00).
RR for a proportion of energy of 5%:
RR = 1.08 (0.95-1.22).
|
Age, BMI, smoking habits, alcohol consumption, physical activity, history of hypertension or high blood cholesterol, family history of before age 60, profession, and fibre intake. |
| 11) Ness AR (2002) | The Diet and Reinfarction Trial (DART) | 2,033 men aged < 70 who had survived a MI, and were rectruited from hospitals in England. | 1983-87 to 2000 | 738 | CHD mortality | Fat advice | Background: Men were randomized to fat advice. The advice aimed to achieve a reduction in total fat intake and increase in the (polyunsaturated fat=saturated fat) P=S ratio.
HR = 0.91 (0.79-1.05).
Hazard by follow-up period in years:
0-2: HR = 0.98 (0.74-1.30).
2-5: HR = 1.02 (0.75-1.40).
5-10: HR = 0.89 (0.68-1.16).
10+: HR = 0.76 (0.57-1.03).
At 2 y the percentage of energy from fat was around 35% in those not given fat advice and just over 32% in those given fat advice. The P=S ratio was around 0.4 in those not given fat advice compared to just under 0.8 in the fat advice group. | Unadjusted.
Adjustment for history of MI, angina, hypertension at baseline; X-ray evidence of cardiomegaly, pulmonary congestion or pulmonary oedema at baseline; and treatment (at entry) with b-blockers, other anti-hypertensives, digoxin/anti-arrhythmics, or anticoagulants did not change the HRs. |
| 11) Burr ML (1989) | The Diet And Reinfarction Trial (DART) | 2,033 men aged < 70, admitted to 21 hospitals, and who had recovered from acute MI (excluding diabetic patients).
(Randomised controlled trial). | 2 | See variables | CHD | Experimental group: fat advice (not defined, but designed to reduce fat intake to 30% of total energy, and to increase the polyunsaturated/saturated ratio to 1-0).
Control group: No fat advice. |
IHD events
(276 cases) | IHD death
(194 cases) | Non-fatal MI
(82 cases) |
RR = 0.91 (0.71-1.15) for fat advice vs no fat advice.
This RR remained 0.91 (0.72-1.16) after adjustment for history of MI, angina, or hypertension; X-ray evidence of cardiomegaly, pulmonary congestion, or pulmonary oedema; and treatment (at entry) with beta-blockers, other antihypertensives, digoxin/antiarrhytmics, or anticoagulants.
|
There were 97 IHD deaths (9.5 %) in the experimental group, and 97 (9.6 %) in the control group.
|
There were 35 nonfatal MI's (3.4 %) in the experimental group, and 47 (4.6 %) in the control group.
|
-Men were allocated to receive or not to receive advice on each of 3 dietary factors: (1) fat advice, (2) fish advice, and (3) fibre advice, thus creating 8 possible combinations.
-At 6 months and 2 years, the subjects was administered a detailed dietary questionnaire.
-The % of fat energy at 6 months and 2 years were 32.1/32.3 for the fat advice group, and 35.3/35.0 for no fat advice.
-The P/S ratios at 6 months and 2 years were 0.78/0.78 for the fat advice group, and 0.40/0.44 for no fat advice.
-Fat advice dropped cholesterol levels, whereas no fat advice showed a rise of cholesterol levels. The difference between these 2 groups was significant (3.6 % at 6 months; P = < 0.001). This difference remained after 2 years (3.5 % difference; P = < 0.01).
-HDL cholesterol rose equally in both groups during the first 6 months and thereafter remained constant.
-In 111 men given fat advice the mean percentage of linoleic acid was 28-1% (95% CI = 26.9-29.2%) compared with 25-0% (24.0-26.0%) in 92 men not given fat advice (p = < 0.001).
-On average the weights of the subjects did not change much during the trial: the overall mean weight rose by 0-6 kg.Unadjusted. |
| 8) Kushi LH (1985) | The Ireland-Boston Diet-Heart Study | 1,001 middle-aged men of Irish descent. | 20
(1959-65 to 1982) | 110? | CHD mortality (death from coronary or ischemic disease) | Fat (not defined) | Cases consumed the same % of calories from fat (39.4) as noncases (38.5; P = 0.12). | Age and cohort. |
| 6) Marr JW (1981) | Cohort name not defined | 337 UK bank and busmen. | 10-20 | 50? | CHD death + nonfatal disease | Total fat (not defined) | Cases consumed significantly less fat (118 g/d) than noncases (129 g/d; P = < 0.01). | Age and occupation. |
| 5) Gillman MW (1997) | The Framingham Heart Study | 832 men aged 45-65, and free of cardiovascular disease.
(USA) | 18-22
(1966-69 to ?) | 83? | Coronary heart disease death | Total fat (not defined) | RR = 1.03 (0.95-1.12) for each increment of 3% of total energy. | Systolic blood pressure, cigarette smoking, glucose intolerance, BMI, physical activity, left ventricular hypertrophy, and intake of energy, alcohol, and fruits and vegetables. |
| 5) Posner BM (1991) | The Framingham Study | 793 men aged 45-65 without preexisting CVD or cancer. | 16
(1966-69 to ?) | See variables | CHD incidence (angina pectoris, coronary insufficiency, MI, sudden death, or nonsudden death from CHD) | Total fat (not defined) |
Consumption in g:
Men aged 45-55
(99? cases) | Men aged 56-65
(114? cases) |
| RR = 0.74 (0.58-0.94) for consumption 90 g (NCEP recommendation) vs 118.0 g.
|
RR = 0.99 (0.85-1.15) for consumption 90 g (NCEP recommendation) vs 109.3 g.
|
Consumption in % of energy intake:
Men aged 45-55
(99? cases) | Men aged 56-65
(114? cases) |
| RR = 0.71 (0.56-0.90) for intake 30% (NCEP recommendation) vs 39.7%.
|
RR = 1.00 (0.83-1.19) for intake 30% (NCEP recommendation) vs 38.3%.
|
Energy intake, physical activity, serum cholesterol level, systolic blood pressure, left ventricular hypertrophy, cigarette smoking, glucose intolerance, and Metropolitan relative weight. |
| 5) Gordon T (1981) | The Framingham Study | 859 men aged 45-64 and free of CHD. | 4
(1966-69 to ?) | 14? | CHD death | Total fat (not defined) | Cases consumed the same amount (112 g) as noncases (114 g). | Age. |
| 5) Gordon T (1981) | The Framingham Study | 859 men aged 45-64 and free of CHD. | 4
(1966-69 to ?) | See variables | CHD incidence | Total fat (not defined) |
Total CHD
(MI, CHD death and angina pectoris)
(79? cases) | MI or CHD death
(51? cases) | Other CHD
(angina pectoris)
(28? cases) |
| Cases consumed the same amount (112 g) as noncases (114 g).
|
Cases consumed the same amount (106 g) as noncases (114 g).
|
Cases consumed the same amount (119 g) as noncases (114 g).
|
Consumption in % of calories:
Total CHD
(MI, CHD death and angina pectoris)
(79? cases) | MI or CHD death
(51? cases) | Other CHD
(angina pectoris)
(28? cases) |
| Cases consumed fat in the same % of calories (40.2) as noncases (38.8).
|
Cases consumed fat in the same % of calories (40.0) as noncases (38.8).
|
Cases consumed fat in the same % of calories (40.0) as noncases (38.8).
|
Age. |
| 5) Gordon T (1981) | The Puerto Rico Heart Health Program | 8,218 men aged 45-64 and free of CHD. | 6
(1965-68 to ?) | 71? | CHD death | Total fat (not defined) | Cases consumed the same amount (94 g) as noncases (96 g). | Age. |
| 5) Gordon T (1981) | The Puerto Rico Heart Health Program | 8,218 men aged 45-64 and free of CHD. | 6
(1965-68 to ?) | See variables | CHD incidence | Total fat (not defined) |
Total CHD
(MI, CHD death and angina pectoris)
(286? cases) | MI or CHD death
(163? cases) | Other CHD
(angina pectoris)
(123? cases) |
| Cases consumed the same amount (94 g) as noncases (95 g).
|
Cases consumed the same amount (92 g) as noncases (95 g).
|
Cases consumed the same amount (96 g) as noncases (95 g).
|
Consumption in % of calories:
Total CHD
(MI, CHD death and angina pectoris)
(286? cases) | MI or CHD death
(163? cases) | Other CHD
(angina pectoris)
(123? cases) |
| Cases consumed fat in the a higher % of calories (36.6) than noncases (35.3; P = < 0.01).
|
Cases consumed fat in a higher % of calories (36.7) than noncases (35.3; P = < 0.05).
|
Cases consumed fat in the same % of calories (36.4) as noncases (35.3).
|
Age. |
| 4) Garcia-Palmieri MR (1980) | The Puerto Rico Heart Health Program | 8,218 men aged 45-64 who were free of CHD at entry. | 6
(1965-?) | See variables | CHD incidence | Total fat (not defined) |
Total CHD (MI, CHD death, coronary insufficiency, and angina pectoris):
Urban men
(213 cases) | Rural men
(73 cases) |
Consumption in g: No significant difference was found between cases (98 g), and noncases (99 g).
Consumption in % of calories: No significant difference was found between cases (38.0%), and noncases (36.6%).
|
Consumption in g: No significant difference was found between cases (81 g), and noncases (86 g).
Consumption in % of calories: No significant difference was found between cases (32.3%), and noncases (32.2%).
|
MI and CHD death:
Urban men
(129 cases) | Rural men
(34 cases) |
Consumption in g: No significant difference was found between cases (95 g), and noncases (99 g).
Consumption in % of calories: No significant difference was found between cases (37.7%), and noncases (36.6%).
|
Consumption in g: No significant difference was found between cases (75 g), and noncases (86 g).
Consumption in % of calories: No significant difference was found between cases (32.0%), and noncases (32.2%).
|
CHD death:
Urban men
(57 cases) | Rural men
(14 cases) |
| Consumption in g: No significant difference was found between cases (98 g), and noncases (100 g).
|
Consumption in g: No significant difference was found between cases (75 g), and noncases (86 g).
|
Coronary insufficiency:
Urban men
(31 cases) | Rural men
(14 cases) |
| Consumption in g: No significant difference was found between cases (95 g), and noncases (99 g).
|
Consumption in g: No significant difference was found between cases (90 g), and noncases (86 g).
|
Angina pectoris:
Urban men
(53 cases) | Rural men
(25 cases) |
| Consumption in g: No significant difference was found between cases (106 g), and noncases (99 g).
|
Consumption in g: No significant difference was found between cases (79 g), and noncases (86 g).
|
Age. |
| 3) Reed DM (1990) | The Honolulu Heart Program | 7,591 men of Japanese ancestry living in Hawaii, and free of CHD and stroke. | 19
(1965-68 to 1985) | See variables | Coronary heart disease incidence | Total fat (not defined) |
Coronary heart disease
(666? cases) | Atherosclerosis of coronary arteries at autopsy
(258? cases) | MI determined at autopsy
(258? cases) |
| RR = 1.3 (1.0-1.6; P = < 0.05) for the highest vs lowest quartile of intake.
|
RR = 3.3 (0.4-28.4) for the highest vs lowest quartile of intake.
|
RR = 9.2 (1.3-63.1; P = < 0.05) for the highest vs lowest quartile of intake.
|
Difference between the highest vs lowest quartile = 24% of calories.Age, systolic blood pressure, cigarette smoking, serum cholesterol, serum glucose, alcohol intake, usual dietary pattern [western vs eastern], and animal protein. |
| 3) McGee (1985) | The Honolulu Heart Program | 7,088 men of Japanese ancestry without CVD or cancer, and living on the island of Oahu. | 10
(1965-68 to ?) | 99? | CHD death | Total dietary fat (not defined) |
| Consumption in g: | Consumption in % of calories: |
| No significant association was found (risk was > 1; no data shown).
|
No significant association was found (risk was > 1; no data shown).
|
Age, systolic blood pressure, BMI, physical activity index, and cigarettes smoked per day. |
| 3) McGee DL (1984) | The Honolulu Heart Program | 7,088 men of Japanese ancestry aged 45-68, and without coronary heart disease, stroke, or cancer.
(Oahu island) | 10
(1965-68 to ?) | See variables | Coronary heart disease incidence (myocardial infarction, CHD death, angina pectoris, or coronary insufficiency) | Fat (not defined) |
Grams.
Age adjusted:
Total CHD
(456? cases) | MI or CHD death
(309? cases) | Angina pectoris or coronary insufficiency
(147? cases) |
| Cases did not consume significantly more (86.4 g) than controls (86.3 g).
|
Cases did not consume significantly more (86.9 g) than controls (86.3 g).
|
Cases did not consume significantly less (85.2 g) than controls (86.3 g).
|
% of calories:
Total CHD
(456? cases) | MI or CHD death
(309? cases) | Angina pectoris or coronary insufficiency
(147? cases) |
| Cases consumed significantly more than controls (SLC = 0.13; P = < 0.01).
|
Cases consumed significantly more than controls (SLC = 0.19; P = < 0.01).
|
Cases did not consume significantly less than controls (SLC = -0.01).
|
SLC = Standardized logistic coefficient.Age, systolic blood pressure, serum cholesterol, cigarettes, body weight, physical activity, and intakes of calories, carbohydrates, starch, other carbohydrates, alcohol, vegetable protein, protein, and cholesterol. |
| 5) Gordon T (1981) | The Honolulu Heart Study | 7,272 men aged 45-64 of Japanese ancestry and free of CHD. | 6
(1965-68 to ?) | 78? | CHD death | Total fat (not defined) | Cases consumed the same amount (86 g) as noncases (87 g). | Age. |
| 5) Gordon T (1981) | The Honolulu Heart Study | 7,272 men aged 45-64 of Japanese ancestry and free of CHD. | 6
(1965-68 to ?) | See variables | CHD incidence | Total fat (not defined) |
Total CHD
(MI, CHD death and angina pectoris)
(264? cases) | MI or CHD death
(164? cases) | Other CHD
(angina pectoris)
(100? cases) |
| Cases consumed the same amount (86 g) as noncases (87 g).
|
Cases consumed the same amount (86 g) as noncases (87 g).
|
Cases consumed the same amount (87 g) as noncases (87 g).
|
Consumption in % of calories:
Total CHD
(MI, CHD death and angina pectoris)
(264? cases) | MI or CHD death
(164? cases) | Other CHD
(angina pectoris)
(100? cases) |
| Cases consumed fat in a higher % of calories (34.9) than noncases (33.4; P = < 0.01).
|
Cases consumed fat in a higher % of calories (35.6) than noncases (33.4; P = < 0.05).
|
Cases consumed fat in the same % of calories (33.6) as noncases (33.4).
|
Age. |
| 3) Yano K (1978) | The Honolulu Heart Study | 7,705 men aged 45-68 of Japanese ancestry and living in Hawaii. | 6
(1965-68 to ) | See variables | CHD incidence (death attributable to CHD including sudden death; nonfatal myocardial infarction (MI); acute coronary insufficiency (CI) ascertained by severe chest pain lasting more than 30 minutes, with documented transient ST-T wave changes on ECG and without elevation of enzyme levels; angina pectoris (AP) ascertained by episodic substernal pain brought on by exertion and relieved by rest.) | Total fat (not defined) |
Consumption in amounts:
CHD death + MI
(179? cases) | Acute CI
(27? cases) | Angina
(88? cases) |
| Cases consumed the same amount of fat (85 g/d) as noncases (86 g/d; P = not significant).
|
Cases consumed the same amount of fat (80 g/d) as noncases (86 g/d; P = not significant).
|
Cases consumed the same amount of fat (87 g/d) as noncases (86 g/d; P = not significant).
|
Consumption as proportion of calories:
CHD death + MI
(179? cases) | Acute CI
(27? cases) | Angina
(88? cases) |
| Cases consumed a larger proportion of calories of fat (35 %) than noncases (33 %; P = < 0.01).
|
Cases consumed the same proportion of calories of fat (31 %) as noncases (33 %; P = not significant).
|
Cases consumed the same proportion of calories of fat (34 %) as noncases (33 %; P = not significant).
|
Age. |
1) Paul O/
Oglesby P (1963) | The Western Electric Study | 1,885 American men of Polish and Bohemian ancestry aged 40-55 without a history of MI or angina pectoris. | 4.5
(1957-?) | 88? | CHD risk (angina pectoris, MI, or CHD death. excluding sudden death) | Total fat (not defined) | The top 15% of calories from fat was contrasted with the bottom 15% (both 296 men). The amount of CHD cases was similar in both groups (14 for high fat, and 16 for low fat, respectively). | Unadjusted. |
Prospective studies of dietary fat and stroke:
| Author | Cohort name | Subjects | Years of follow-up | Cases | End point | Consumption of | Relative Risk (RR) | Adjustments |
| 41) Boden-Albala B (2009) | The Northern Manhattan Study (NOMAS) | 3,183 Hispanic, non-Hispanic white, and non-Hispanic black men and women aged > 40, and without stroke.
(USA) | 5.5
(1993-2001 to ?) | 142 | Ischemic stroke risk | Total fat (not defined) |
HR = 1.6 (0.6-3.9; No P-value) for the highest vs lowest quintile of consumption.
Amount specific data (grams/day):
25: HR = 1.
41: HR = 0.9
54: HR = 0.7
72: HR = 1.3
115: HR = 1.6
No 95% CI's were defined, but the figure showed that all quintiles included the 1-value.
HR = 1.4 (0.8-2.4; P = 0.08) for consumption ≥ 45% of total daily calories from fat.
HR = 1.7 (1.0-2.9) for ≥ 65 vs < 65 g/day.
Stratified by gender:
| Men | Women |
| HR = 2.2 (No 95% CI defined, but the 1 was excluded) for ≥ 65 vs < 65 g.
|
HR = 1.4 for ≥ 65 vs < 65 g.
|
Stratified by race/ethnicity:
| White | Black | Hispanic |
| HR = 1.7 for ≥ 65 vs < 65 g.
|
HR = 1.6 for ≥ 65 vs < 65 g.
|
HR = 1.7 (No 95% CI defined, but the 1 was excluded) for ≥ 65 vs < 65 g.
|
Stratified by age:
| < 70 | > 70 |
| HR = 1.6 for ≥ 65 vs < 65 g.
|
HR = 1.6 (No 95% CI defined, but the 1 was excluded) for ≥ 65 vs < 65 g.
|
Stratified by BMI:
| < 30 | > 30 |
| HR = 1.7 (No 95% CI defined, but the 1 was excluded) for ≥ 65 vs < 65 g.
|
HR = 1.5 for ≥ 65 vs < 65 g.
|
Age, race/ethnicity, sex, education, hypertension, diabetes, coronary artery disease, moderate alcohol, current and previous smoking, BMI, leisure time physical activity, sodium, potassium, fruit/vegetable, calcium, fiber, vitamin E, and calories. |
| 34) Ness AR (2005) | The Boyd Orr Cohort | 4,028 people (1,995 men, and 2,033 women) from England and Scotland. | average 37
(1948 to 2000) | 83 | Stroke mortality | Fat (not defined) |
RR = 0.76 (0.26-2.16; P = 0.2) for the highest vs lowest quartile of consumption.
Amount specific data (g):
30.1-63.3: RR = 1.
63.3-77.6: RR = 2.05 (1.09-3.85).
77.7-96.4: RR = 0.69 (0.29-1.63).
93.5-229.0: RR = 0.76 (0.26-2.16).
Cut-off points for quintile 4-5 were incorrectly defined in the original article. | Age, energy, sex, childhood family food expenditure, father's social class, district of residence as a child, period of birth, season when studied as a child, and Townsend score for current address or place of death. |
| 31) Leosdottir M (2007) | The Malmö Diet and Cancer Study | 28,098 middle-aged men and women without CVD.
(Sweden) | 8.4
(1991-96 to 2002) | 648?
(no data about men vs women) | Ischemic stroke | Total fat (not defined, but including trans fatty acids) |
| Women | Men |
HR = 1.12 (0.79-1.59; P = 0.4) for the highest vs lowest quartile of consumption.
Amount specific data (% energy intake):
30.8: HR = 1.
36.5: HR = 0.91 (0.65-1.30).
40.3: HR = 1.06 (0.75-1.49).
46.2: HR = 1.12 (0.79-1.59).
|
HR = 0.97 (0.71-1.33; P = 0.9) for the highest vs lowest quartile of consumption.
Amount specific data (% energy intake):
31.8: HR = 1.
37.8: HR = 1.02 (0.75-1.40).
41.7: HR = 1.26 (0.94-1.70).
47.7: HR = 0.97 (0.71-1.33).
|
Age, smoking habits, alcohol consumption, socioeconomic status, marital status, physical activity, BMI, fibre intake, and blood pressure. |
| 30) Sauvaget C (2004) | The Adult Health Study
(A subcohort of the Hiroshima/Nagasaki Life Span Study) | 3,731 male and female atomic bomb survivors aged 35-89.
(Japan) | Median 14
(1983-2001) | 60 | Cerebral infarction death | Fat from animal and vegetable products (not defined) |
RH = 0.75 (0.37-1.54; P = 0.4609) for the highest vs lowest tertile of consumption.
Amount specific data (g/day):
27: RH = 1.
47: RH = 0.98 (0.53-1.80).
76: RH = 0.75 (0.37-1.54). | Stratified by sex and age.
Adjusted for radiation dose, city, BMI, smoking status, alcohol habits, and medical history of hypertension and diabetes. |
| 25) Iso H (2003) | No cohort name defined | 4,775 individuals (2,269 men and 2,506 women) aged 40-69, and without stroke.
(Japan) | 14.3
(1973-88 to 1997) | 68 | Intraparenchymal hemorrhage incidence (lasting ≥ 24 hours or until death) | Total fat (not defined) |
RR = 0.46 (0.20-1.04; P = 0.05) for the highest vs lowest quartile of consumption.
Amount specific data (g/day):
23.1: RR = 1.
34.8: RR = 1.10 (0.60-2.02).
45.6: RR = 0.79 (0.40-1.56).
61.2: RR = 0.46 (0.20-1.04).
Other strokes: No correlations were found with risk of other stroke subtypes (subarachnoid hemmorhage [41 cases], and ischemic stroke [166 cases]; Data not shown). | Age, sex, total energy, BMI, hypertension, diabetes, serum total cholesterol, smoking status, ethanol intake, and menopausal status. |
| 20) Ross RK (1997) | No cohort name defined | 18,244 men aged 45-64 from Shanghai, and without cancer.
(China) | 1986-89 to 1994 | 245 | Stroke/cerebrovascular accident death | Fat (not defined) |
RR = 1.3 (0.9-1.8) for the highest vs lowest tertile of consumption.
Amount specific data (Tertiles not defined):
T1: RR = 1.
T2: RR = 1.2 (0.8-1.7).
T3: RR = 1.3 (0.9-1.8). | Education, marital status, BMI, lifetime cigarette smoking, lifetime ethanol intake, and history of hypertension. |
| 13) He K (2003) | The Health Professionals Follow Up Study | 43,732 men aged 40-75, and without a history of CVD or diabetes mellitus.
(USA) | 14
(1986-2000) | See variables | Stroke risk | Total fat (not defined) |
Ischaemic stroke
(embolism or thrombosis)
(455 cases) | Haemorrhagic stroke
(subarachnoid and intracerebral)
(125 cases) |
RR = 0.91 (0.65-1.28; P = 0.77) for the highest vs lowest quintile of consumption.
Amount specific data (g/d):
54: RR = 1.
64: RR = 0.93 (0.68-1.26).
70: RR = 1.02 (0.74-1.39).
77: RR = 1.02 (0.74-1.42).
86: RR = 0.91 (0.65-1.28).
|
RR = 1.16 (0.58-2.32; P = 0.83) for the highest vs lowest quintile of consumption.
Amount specific data (g/d):
54: RR = 1.
64: RR = 1.76 (0.95-3.25).
70: RR = 1.49 (0.79-2.83).
77: RR = 1.18 (0.60-2.35).
86: RR = 1.16 (0.58-2.32).
|
Age, smoking, BMI, physical activity, history of hypertension, smoking status, aspirin use, multivitamin use, and consumption of alcohol, potassium, fibre, vitamin E, total servings of fruit and vegetables, total energy, and hypercholesterolaemia. |
| 13) Iso H (2001) | The Nurses' Health Study | 85,764 women aged 34-59, and without cancer, angina, MI, coronary revascularization, stroke, or other CVD.
(USA) | 14
(1980-1994) | 74 | Intraparenchymal hemorrhage risk | Total fat (not defined) |
RR = 0.92 (0.45-1.88; P = 0.77) for the highest vs lowest quintile of consumption.
Amount specific data (g/d):
52: RR = 1.
63: RR = 0.85 (0.42-1.72).
70: RR = 0.69 (0.33-1.45).
77: RR = 0.81 (0.40-1.67).
87: RR = 0.92 (0.45-1.88). | Age, smoking status, time interval, BMI, alcohol intake, menopausal status and postmenopausal hormone use, vigorous exercise, usual aspirin use, multivitamin use, vitamin E use, n3 fatty acid intake, calcium intake, and histories of hypertension, diabetes, and high cholesterol levels, and total energy. |
| 12) Simon JA (1995) | The Multiple Risk Factor Intervention Trial | 12,866 US men, aged 35-57.
(Nested case-control) | 6.9
(1973-76 to ?) | 96 | Stroke incidence (fatal or nonfatal) | Total fat (not defined) | Cases consumed significantly more (40.2% kcal) than controls (37.6% kcal; P = < 0.05). | Unadjusted.
Cases matched to controls by age, clinical center, treatment group, and date of randomization. |
| 10) Khaw KT (1987) | The Rancho Bernardo Cohort | 859 men and women aged 50-79 without CVD, of which 50% had hyperlipidemia.
(Southern California) | 12 | 24
(9 men, and 15 women) | Stroke-associated mortality | Total fat (not defined) | RR = 1.08 (0.59-1.97; P = 0.80) per 20 g. | Age, sex, calories, and potassium. |
| 9) Seino F (1997) | The Shibata Study | 954 men and 1,329 women, age ≥ 40, and free of stroke.
(Japan) | 15.5
(1977-1992) | 141? | Stroke incidence (cerebral hemorrhage, cerebral infarction and subarachnoid hemorrhage. Excluding transient cerebral ischemic attack) | Total fat (not defined) |
RR = 0.94 (0.48-1.85; P = 0.44) for the highest vs lowest quartile of consumption.
Amount specific data (g/day):
33: RR = 1.
41: RR = 0.60 (0.31-1.17).
48: RR = 0.82 (0.44-1.54).
59: RR = 0.94 (0.48-1.85). | Sex, age, diastolic blood pressure, atrial fibrillation, and energy. |
| 7) Keli SO (1994) | The Zutphen Study
(The Dutch contribution to The Seven Countries Study) | 552 men aged 50-69.
(The Netherlands) | 1970-1985 | 42 | First stroke incidence (a sudden onset of neurological paralysis of > 24-hour duration or leading to death) | Total fat (not defined) | Stroke cases consumed the same amount (42.8% of energy) as noncases (41.5% of energy). | Unadjusted. |
| 5) Gillman MW (1997) | The Framingham Heart Study | 832 men aged 45-65, and free of cardiovascular disease.
(USA) | 18-22
(1966-69 to ?) | 61? | Ischemic stroke incidence (atherothrombotic brain infarction and embolus) | Total fat (not defined) |
RR = 0.84 (0.75-0.92) for each increment of 3% of total energy.
Further adjustment for the total serum and HDL cholesterol levels and the physical activity index produced comparable results (no data shown).
Stratified by occurence of CHD during the follow-up period:
(age, and energy adjusted)
No CHD
(43? cases) | CHD
(18? cases) |
| RR = 0.85 (0.77-0.94) for each increment of 3% of total energy.
|
RR = 1.05 (0.88-1.23) for each increment of 3% of total energy.
|
Systolic blood pressure, cigarette smoking, glucose intolerance, BMI, physical activity, left ventricular hypertrophy, and intake of energy, alcohol, and fruits and vegetables. |
| 3) Reed DM (1990) | The Honolulu Heart Program | 7,591 men of Japanese ancestry living in Hawaii, and free of CHD and stroke. | 19
(1965-68 to 1985) | See variables | Stroke incidence | Total fat (not defined) |
Thromboembolic stroke
(250? cases) | Hemorrhagic stroke
(111? cases) |
| RR = 0.8 (0.6-1.2) for the highest vs lowest quartile of intake.
|
RR = 1.2 (0.7-1.9) for the highest vs lowest quartile of intake.
|
Atherosclerosis at autopsy:
Atherosclerosis of large cerebral arteries
(198? cases) | Atherosclerosis of small cerebral arteries
(198? cases) |
| RR = 0.6 (0.2-1.3) for the highest vs lowest quartile of intake.
|
RR = 0.7 (0.2-1.8) for the highest vs lowest quartile of intake.
|
Autopsy determined stroke:
Cerebral infarction
(198? cases) | Cerebral hemorrhage
(198? cases) |
| RR = 0.6 (0.2-1.7) for the highest vs lowest quartile of intake.
|
RR = 0.8 (0.2-4.6) for the highest vs lowest quartile of intake.
|
Difference between the highest vs lowest quartile = 24% of calories.Age, systolic blood pressure, cigarette smoking, serum cholesterol, serum glucose, alcohol intake, usual dietary pattern [western vs eastern], and animal protein. |
| 3) McGee (1985) | The Honolulu Heart Program | 7,088 men of Japanese ancestry without CVD or cancer, and living on the island of Oahu. | 10
(1965-68 to ?) | 61? | Stroke death | Total dietary fat (not defined) |
| Consumption in g: | Consumption in % of calories: |
| No significant association was found (risk was < 1; no data shown).
|
An inverse association was found (No data shown; P = < 0.05).
|
Age, systolic blood pressure, BMI, physical activity index, and cigarettes smoked per day. |
| 3) Kagan A (1985) | The Honolulu Heart Program | 7,088 men aged 45-68 of Japanese ancestry on the island of Oahu. | 10
(1965-68 to ?) | See variables | Stroke incidence | Fat (not defined) |
Total stroke:
(208? cases) | Thrombo-embolic stroke:
(134? cases) | Intracranial hemorrhage:
(58? cases) |
Incidence per thousand according to level consumption in g/day:
≤ 54.8: 38.1
54.9-72.8: 28.9
72.9-90.5: 24.3
90.6-115.3: 26.4
115.4+: 29.4 (P = 0.002)
|
Incidence per thousand according to level consumption in g/day:
≤ 54.8: 25.6
54.9-72.8: 18.6
72.9-90.5: 15.6
90.6-115.3: 17.0
115.4+: 20.8 (P = 0.020)
|
Incidence per thousand according to level consumption in g/day:
≤ 54.8: 9.0
54.9-72.8: 6.5
72.9-90.5: 8.0
90.6-115.3: 7.6
115.4+: 8.6 (P = 0.459).
No difference in consumption was found with subarachnoid hemorrhage, or cerebral hemorrhage (85.1, 84.8, and 85.5 g for SAH, CH, and control, respectively).
|
The inverse relationship to fat intake became non-significant when blood pressure, left ventricular hypertrophy, serum glucose, cigarette smoking, and proteinuria were included in the model.Age. |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | |