Materials and Methods

The Big Heart Disease Lie

Heart Disease Causes and Treatments

Get Instant Access

We limited our review to prospective cohort studies that measured vitamin C status before ascertainment of morbidity and mortality and that used appropriate statistical techniques to estimate relative risk (RR). We also limited studies to those that excluded prevalent cases at study entry to avoid potential biases resulting from their inclusion; prevalent cases may be more likely to change their dietary habits as a result of their disease or their disease may affect blood ascorbate levels. Restriction of the cohort to those free of disease at baseline also allows examination of the benefits of vitamin C in primary prevention of CVD. We included studies assessing vitamin C status measured by both blood assay or dietary intake (including or excluding supplement use). We reviewed studies examining various end points, and studies are reviewed in descending order of inclusiveness of CVD end points. We also address whether adjustment was made for potentially confounding variables, particularly smoking, and how these may affect interpretation of findings.

Studies Using Blood Ascorbate Concentrations

Three studies have examined the relation of blood ascorbate levels to CVD, with two finding significant associations (Table 7.1). Using the broadest definition of CVD [9 th revision of the International Classification of Disease (ICD-9), codes 390-459], Simon et al. (10) found that 30- to 75-y-old participants in the second National Health and Nutrition Examination Survey (NHANES II) with saturated (>1.1 mg/dL) or normal (0.5-1.0 mg/dL) compared with marginal (<0.4 mg/dL) serum ascorbate levels had a 34 and 33% reduced risk, respectively, of dying from CVD over 12-16 y. In a cohort of roughly the same age but more than twice as large, Khaw et al. (11) found a significant association between plasma ascorbate quintile and CVD mortality (excluding rheumatic heart disease and diseases of arterioles and veins from previous definition) after adjustment for age during 4 y of follow-up. Levels identified as most protective were consistent with saturation but multivariate-adjusted RR for quintiles were not presented. However, the multivariate-adjusted RR was 0.64 (0.51-0.78) in men and 0.81 (0.62-1.06) in women per increase of 0.35 mg/dL serum ascorbate. In the same NHANES II cohort used in the study by Simon et al. (10), Loria et al. (12) did not find an association between serum ascorbate quartile and a narrower definition of

CVD, including end points likely to be affected by antioxidant mechanisms (e.g., ischemic heart disease, cerebrovascular disease, sudden death, and diseases of arterioles). Adjustment for potential confounders in multivariate analyses from all three studies were generally comparable; however, the first two studies used smoking status (i.e., current, past, never) whereas the last study used average number of cigarettes smoked per day.

Two of four studies found a significant relation between blood ascorbate and CHD mortality defined using ICD-9 codes 410-414 (Table 7.1). Plasma ascorbate quintiles were associated with CHD mortality during 4 y of follow-up; however, these data were adjusted only for age; multivariate-adjusted RR in men was 0.63 (0.42-0.94) and women 0.56 (0.36-0.87) per increase of 0.35 mg/dL serum ascorbate (11). In a smaller study of free-living men and women >60 y old, Sahyoun et al. (13) found that participants with plasma ascorbate 1.0-1.5 mg/dL vs. 0.9 mg/dL had a decreased risk of heart disease mortality; participants with levels >1.6 mg/dL also had an increased risk although it was not significant. Although this study adjusted for a number of confounders, adjustment was not made for smoking. Two other studies examining CHD mortality in middle-aged men (14) and elderly men and women (15) found no significant association with plasma ascorbate. Another study examining the relation of plasma ascorbate with nonfatal and fatal CHD in middle-aged men failed to find a significant association (16). Only one study examined the relation between plasma ascorbate and stroke; elderly men and women in the highest (>0.49 mg/dL) compared with the lowest (<0.21 mg/dL) tertile had a 30% decreased risk of dying from stroke although the RR risk was borderline significant and was adjusted only for age and sex (15).

Studies Using Vitamin C Intake

Two studies examined the relation between vitamin C intake and broad definitions of CVD in representative samples of U.S. adults (Table 7.2). The first study, using data from NHANES I with 13-16 y of follow-up, found that vitamin C intakes >50 mg/d in combination with vitamin C supplement use had a standardized mortality ratio (SMR) of 0.7 (17). However, SMR, unlike RR estimates, are estimated using an external comparison group and do not permit adjustment for potential confounding factors. In NHANES II, vitamin C intakes of 32-73 mg/d among men were associated with an increased (60%) risk of CVD mortality after adjustment for CVD risk factors, supplement use, and total energy intake, but there was no increased risk in the middle two dietary intake quartiles (Loria, unpublished data). There was no association between vitamin C intake and CVD mortality among women in NHANES II. This study also used a narrower definition of CVD than the NHANES I study, excluding rheumatic, hypertensive and pulmonary heart disease, and diseases of veins.

Four studies examined the relation between vitamin C intake and fatal and nonfatal CHD; however, none yielded significant associations. In a study of U.S. male health professionals who were 40-75 y of age and followed for 4 y, total vitamin C intakes (diet and supplements) were not associated with the risk of CHD after adjust-

TABLE 7.1

Prospective Cohort Studies Examining Serum or Plasma Ascorbate Concentrations and Cardiovascular Disease (CVD)a

Reference Cohort Gender Age range (y) Cohort size CVD

Women 10,636

Women 3724

Endpoints Exposure

390-459

<0.4

1.0

0.5-1

0.67

0.5-0.9

1.1-2.7

0.66

0.4-0.99

400-438

0.4

1.0

0.7

0.9

0.6-1.5

0.8

0.67

0.4-1.2

1.0

0.29

0.2-0.6

1.3

0.29

0.2-0.6

0.5

1.0

0.9

0.41

0.2-0.9

1.0

0.36

0.2-0.9

1.2

0.6

0.3-1.3

1.5

0.41

0.2-1.0

410-414,427.5,

<0.5

1.4

0.9-2.3

428.9,429.2,

0.5-0.9

1.00

0.6-1.5

430-438,

0.9-1.3

1.30

0.9-2.0

440-444

>1.3

1.0

(same)

<0.7

0.9

0.6-1.5

0.7-1.2

0.9

0.6-1.4

1.2-1.5

0.9

0.6-1.4

>1.5

1

CHD

(11) UK

Men

45-79

8860

2-6

410-414

0.4

1

0.7

1.18

0.7-2.1

0.8

0.92

0.5-1.7

1.0

0.35

0.2-0.8

1.3

0.32

0.2-0.8

Women

10,636

(same)

0.5

1

0.9

0.23

0.7-0.9

1.0

0.39

0.2-1.2

1.2

0.32

0.1-1.0

1.5

0.07

0.1-0.7

(14) Basel

Men

Middle-age

2974

12

410-414

<0.4

1.25

NS

(15) Britain

Both

65+

730

20

410-414

<0.21

1

0.21-0.49

0.9

0.6-1.3

>0.49

0.9

0.6-1.3

(13) Boston

Both

60+

725

12

Heart disease

0.9

1

mortality

1.0-1.5

0.51

0.3-0.9

1.6

0.53

0.3-1.1

Fatal or nonfatal MI

(16) Eastern

Men

42-60

1605

3-8

Fatal or

<0.20

2.08

0.8-5.3

Finland

nonfatal MI

0.2-0.58

0.87

0.4-1.9

0.59-0.88

0.62

0.3-1.5

0.89-1.14

0.92

0.4-2.1

>1.14

1.0

Cerebrovascular diseases

(15) Britain

Both

65+

730

20

433-438

<0.21

1

0.21-0.49

1.1

0.7-1.7

>0.49

0.7

0.4-1.1

a Abbreviations: ICD-9, International Classification of Diseases, 9th Revision; RR, relative risk; CI, confidence interval; CHD, coronary heart disease; NS, nonsignificant; MI, myocardial infarction.

a Abbreviations: ICD-9, International Classification of Diseases, 9th Revision; RR, relative risk; CI, confidence interval; CHD, coronary heart disease; NS, nonsignificant; MI, myocardial infarction.

TABLE 7.2

Prospective Cohort Studies Examining Vitamin C Intake and Cardiovascular Disease (CVD)a

TABLE 7.2

Prospective Cohort Studies Examining Vitamin C Intake and Cardiovascular Disease (CVD)a

Age

Cohort

Follow-

Endpoints

Exposure

Reference

Cohort

Gender

range (y)

size

up (y)

(ICD-9)

(mg/d)

RR

95% CI

CVD mortality

(17)

U.S.

Men

25-74

4479

13-16

390-459

<50 mg/d 50 + nosup 50 + sup

0.5-0.8

Women

6869

(same)

<50 mg/d 50 + nosup 50 + sup

0.6-0.96

Unpublished^

U.S.

Men

30-75

3347

12-16

410-414,427.5, 428.9,429.2, 430-438, 440-444

<32 32-73 74-139 140+

1.30 1.60 1.20 1.00

0.8-2.2 1.0-2.6 0.7-2.11

Women

3724

(same)

<29 29-73 74-134 135+

0.80 0.80 0.90 1.00

0.5-1.4 0.5-1.2 0.6-1.4

Fatal and nonfatal CHD

(18)

U.S.

Men

40-75

39,910

CABG, angioplasty, stroke

92 149 218 392 1162

0.9

0.8-1.3 0.9-1.5 0.8-1.3 0.7-1.2

(19)

U.S.

Women

30-55

87,245

8

CHD

Q5

0.8

NS

(20)

Wales

Men

45-59

2423

5

410-414, MI, angina

<34.7 34.8-43.5 43.6-52.3 52.4-66.4 >66.5

1.50 1.30 1.40 1.30 1.00

NS

(21)

Rotterdam

Both

55-95

4802

4

I21-I24 (ICD-10)

<87 87-126 >126

1.00 0.94 0.88

0.58-1.37 0.56-1.38

CHD mortality

(22) Finland

Men

30-69

2748

12-16

410-414

<60

1.00

(ICD-8)

61-85

0.90

0.6-1.2

>85

1.00

0.7-1.4

Women

2385

<61

1.00

62-91

0.50

0.3-0.99

>91

0.50

0.2-0.98

(23) Chicago

Men

40-55

5397

24

410-412

21-82

1.00

83-112

1.03

113-393

0.75

100

0.63

0.45-1.02

(15) Britain

Both

65+

730

20

410-414

<28

1.00

28-44.9

0.90

0.7-1.4

>45

0.80

0.6-1.2

(13) Boston

Both

60+

725

12

Heart

90

1.00

disease

91-387

0.50

0.3-0.9

mortality

388

0.50

0.3-1.1

(24) Iowa

Women

55-69

34,486

8

410,412,

<112.3

1.00

429.2

112.4-161.3

1.08

0.69-1.69

161.4-226.7

0.85

0.53-1.37

226.8-391.2

0.99

0.61-1.59

>391.3

1.49

0.96-2.30

Cerebrovascular disease

(15) Britain

Both

65+

730

20

433-438

<28

1.00

28-44.9

0.80

0.5-1.2

>45

0.40

0.2-0.6

(25) Chicago

Men

40-55

1843

30

430-434

22-74

436-438

75-97

0.87

0.60-1.26

98-122

0.87

0.60-12.6

123-393

0.71

0.47-1.05

(26) Iowa

Women

55-69

34,492

11

430-438

82.4

1.00

138.3

0.69

0.43-1.10

190.6

0.77

0.48-1.21

280.9

0.70

0.42-1.13

678.7

1.24

0.81-1.89

a Abbreviations: ICD-9, International Classification of Diseases, 9th revision; RR, relative risk; CI, confidence interval; sup, supplement use; CHD, coronary heart disease; MI, myocardial infarction; CABG, coronary artery bypass grafting; Q, quintile; NS, nonsignificant.

CopyrigBat€fr20(te|rjjAO(npu Presd a Abbreviations: ICD-9, International Classification of Diseases, 9th revision; RR, relative risk; CI, confidence interval; sup, supplement use; CHD, coronary heart disease; MI, myocardial infarction; CABG, coronary artery bypass grafting; Q, quintile; NS, nonsignificant.

CopyrigBat€fr20(te|rjjAO(npu Presd ment for CHD risk factors and intakes of vitamin E and carotene (18). Among 30- to 55-y-old female nurses in the United States followed for 8 y, the risk of CHD was lower for women in the highest compared with the lowest quintile of total intake although this association was not significant (19). A study of middle-aged men living in Wales failed to find significant associations (20). Another study examining morbidity and mortality from acute MI did not yield significant associations in Dutch adults followed for 4 y (21).

Five studies examined the relation between vitamin C intake and fatal CHD; only one found a significant association although another tended to be significant. Dietary vitamin C intake was significantly associated with CHD mortality (ICD-8, codes 410-414) among Finnish women but not Finnish men (22). Women in the middle (<61 mg/d) and highest (61-85 mg/d) tertile had a 50% lower risk of CHD mortality than women in the lowest tertile (<60-61 mg/d) (22). Middle-aged men in Chicago followed for 24 y who had intakes between 113 and 393 mg/d compared with 21-82 mg/d had a decreased risk of dying from acute MI (ICD-9, codes 410-412) although it only was borderline significant (23). The RR was significant, however, in nonsmokers when analyses were stratified (23). Another study examining CHD mortality failed to find a significant association in elderly adults (15). In a smaller study of free-living men and women (60 y old, Sahyoun et al. (13) did not find a significant relation between total vitamin C intake and heart disease mortality, presumably CHD. Another study examining mortality from acute MI failed to find a significant association in Iowa women followed for 6 y (24).

One of three studies that examined the relation between vitamin C intake and stroke found a significant association, whereas another reported only a borderline significant finding. Adults >65 y old followed for 20 y who consumed >45 mg vitamin C/d had a 60% reduced risk of dying from stroke compared with those consuming <28 mg/d (15). However, RR estimates were adjusted for age and sex only and did not account for other potential confounders. Middle-aged men in Chicago followed for 30 y who had intakes between 123 and 393 mg/d compared with 22-74 mg/d had a decreased risk of stroke although it only was borderline significant (25). In a study of Iowa women 55-69 y of age, there was no increased risk of dying from stroke after

11 y (26). Discussion

The evidence regarding vitamin C status and CVD from prospective cohort studies was reviewed, taking into account the variability in study design. Findings were more likely to be positive when serum ascorbate concentration vs. intake was used to assess vitamin C status. This observation is not surprising given that these indicators may measure quite different aspects of vitamin C status, i.e., estimated body ascorbic acid pool vs. vitamin C intake from foods and supplements. The relationship between vitamin C intakes and serum ascorbate level is complex because serum levels are saturable (1,27), absorption is dose dependent (1), and may be modified by other factors, such as cigarette smoking, disease state, and medications (1,4,28-30). On the basis of this review of prospective studies, available stores may be the more important aspect of vitamin C status with respect to CVD than intake level, although the potential for misclassification associated with dietary assessment methods may also account for greater inconsistency in studies examining intake levels.

Findings also tended to be positive when a wider definition of CVD was used; this pattern was observed in studies using both serum concentrations and intake data. Studies that examined CHD, either fatal or nonfatal, as an end point yielded inconsistent findings, although those using serum ascorbate concentrations tended to be positive compared with those using vitamin C intake. The evidence for a relationship between stroke and vitamin C status was also weak. No studies examined the association of vitamin C on morbidity alone; it is possible that vitamin C may affect morbidity differently than mortality. A tendency to positive associations with wider definitions including end points both likely and less likely to be affected by antioxidant mechanisms suggests that vitamin C may be protective of CVD through multiple mechanisms, including but not limited to oxidation.

Most of the studies reviewed controlled for important potential confounders; however, some analyses were adjusted only minimally for age and sex and these results should be interpreted with caution, especially those that did not adjust for cigarette smoking. In one of the two most recent studies examining serum ascorbate in a large cohort, Khaw et al. (11) adjusted only for age and sex in analyses by ascorbate quintile but presented multivariate-adjusted models for continuous ascorbate level. In two different analyses of the other recently published large cohort, NHANES II, one study controlled for smoking status (i.e., current, past, never) and found a significant association (10), whereas the other adjusted for number of cigarettes smoked and failed to find a significant association (12). This discrepancy raises the question of possible residual confounding by smoking although these studies also used differing end points, vitamin C status classification, inclusion criteria, and classification of other adjustment variables, all of which may have contributed to the discrepant findings. More research is required on appropriate adjustment for cigarette smoking in statistical models, given the important association of smoking with serum ascorbate levels. Cigarette smoking decreases serum ascorbate levels (28,31) and increases CVD risk possibly by generating oxygen-derived free radicals (1,32). Further consideration of whether smoking should be treated as a potential confounder, effect modifier, part of the causal mechanism, or some combination of these is warranted.

Was this article helpful?

0 0
A Disquistion On The Evils Of Using Tobacco

A Disquistion On The Evils Of Using Tobacco

Among the evils which a vitiated appetite has fastened upon mankind, those that arise from the use of Tobacco hold a prominent place, and call loudly for reform. We pity the poor Chinese, who stupifies body and mind with opium, and the wretched Hindoo, who is under a similar slavery to his favorite plant, the Betel but we present the humiliating spectacle of an enlightened and christian nation, wasting annually more than twenty-five millions of dollars, and destroying the health and the lives of thousands, by a practice not at all less degrading than that of the Chinese or Hindoo.

Get My Free Ebook


Post a comment