SEARCH

SEARCH BY CITATION

Keywords:

  • age;
  • mortality;
  • myocardial infarction;
  • serum lipids;
  • smoking;
  • women

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Subjects and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Abstract. Lindqvist P, Bengtsson C, Lissner L, Björkelund C (Department of Primary Health Care, Göteborg University, Gothenburg, Sweden). Cholesterol and triglyceride concentration as risk factors for myocardial infarction and death in women, with special reference to influence of age. J Intern Med 2002; 251: 484–489.

Objective. To evaluate the importance of serum cholesterol and triglyceride concentrations as predictors of myocardial infarction and death in women of different ages.

Design. Prospective observational study, initiated in 1968–69.

Setting. Gothenburg, Sweden, with about 430 000 inhabitants.

Subjects. A population-based sample of 1462 women aged 38, 46, 50, 54 and 60 years at start of the study, followed up for 24 years.

Main outcome measures. Within each age group, myocardial infarction and death were predicted by serum cholesterol and triglyceride concentrations and smoking in a multivariate model.

Results. In the total population only serum triglyceride concentration was a strong independent risk factor for both end-points studied. Serum triglyceride concentration measured in 38- and 46-year-old women had no predictive value with respect to 24-year incidence of myocardial infarction or death. In 50-, 54- and 60-year-old women, high serum triglyceride concentration consistently predicted myocardial infarction and total mortality. Serum cholesterol concentration, on the other hand, showed evidence of direct association for 24-year all-cause mortality in the younger premenopausal group. Serum cholesterol had no predictive value for myocardial infarction or mortality in the peri- and postmenopausal ages.

Conclusions. There appears to be age-specificity in association between serum lipids and these end-points in women, serum cholesterol concentration being more important for younger women and serum triglyceride concentration more important for postmenopausal women as risk factors, observations which need further attention.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Subjects and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Hyperlipidaemia is considered to be an important risk factor of death and incidence of cardiovascular disease. Of the serum lipids, cholesterol and triglyceride concentrations have most often been measured in clinical routine work. The question of whether cholesterol or triglyceride concentration is more important as a predictor in this respect has been debated [1-3]. The influence of age has not been studied in detail in previous studies in women.

In the present population study of women in Gothenburg, Sweden, the baseline examination took place in 1968–69, and the participants have then been followed up for 24 years. End-points in relation to different concentrations of serum lipids were studied. The purpose of this paper was to evaluate the importance of cholesterol and triglyceride concentration as predictive risk factors for myocardial infarction and early death for women in different age strata, and in the population sample as a whole. As smoking per se is a risk factor for coronary heart disease and death [4,5], and also a stronger risk factor in women than in men [6], smoking was taken into consideration as a background factor in the statistical analyses. Additional variables, such as hypertension, physical activity at leisure and work, waist-to-hip ratio, body mass index, mental stress, educational level and hormone-replacement therapy (HRT) were also investigated in this respect.

Subjects and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Subjects and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Participation, data registration and end-points during the follow-up period

In 1968–69, 1462 women in Gothenburg, Sweden, who were aged 38, 46, 50, 54 or 60 years, were recruited to the study [7], shown in Table 1. The systematic sampling based on the date of birth and the high percentage of those women sampled that participated (90.1%) ensured that the participants were a representative cross-section of women in the age groups studied. The population sample was re-examined in 1974–75 [8], 1980–81 [9] and 1992–93 [10]. In 1992–93, 830 of the 1462 women sampled in 1968–69 participated, 70.2% of the women who were alive.

Table 1.   Participants in the baseline study in 1968–69, and number and percentage of end-points during the 24-year follow-up period Thumbnail image of
Mortality.

The women sampled were checked with the Revenue Office Register of Gothenburg and the Swedish National Death Registry for vital status and with the Swedish Person and Address Registry for place of residence. Death certificates were obtained from the Central Bureau of Statistics. Cause of death as registered on death certificates was recorded. There were no missing data with respect to vital status after 24 years.

Myocardial infarction.

At least two out of the following three criteria had to be fulfilled in order to accept the diagnosis of myocardial infarction: central chest pain, typical electrocardiographic changes and serum transaminase elevation. For a fatal myocardial infarction, the diagnosis had to be stated in the death certificate. Two of the subjects had a myocardial infarction before the first examination and these prevalent cases were excluded.

Ethical consideration.

The study has been approved by the local ethics committee.

Clinical examination

At baseline blood samples were drawn from the women after an overnight fast for measurement of total serum cholesterol and serum triglyceride concentration. Serum cholesterol and triglyceride concentrations were determined according to the methods of Levine and Zak [11] and Lofland [12]. Body mass index and waist-to-hip ratio were calculated as described earlier [13]. Information about smoking was obtained via a standardized interview and subjects who smoked one cigarette or more per day were defined as smokers. Information about the additional variables was also obtained from the interview and the definition of these variables have previously been described as for hypertension [14], physical activity at leisure and work [15], mental stress [16], educational level [17] and HRT [18].

Statistical methods

A Cox regression model was used for the statistical analysis; the analysis was carried out by including cholesterol concentration and triglyceride concentration as independent variables, and taking the other variable together with smoking into consideration as background factors when one of these was related to myocardial infarction and death. Age was also included as a background variable, when the statistical analyses referred to the total population sample. Cholesterol and triglyceride concentrations were analysed as continuous variables. The relative risk (RR) corresponds to 1 mM unit change in cholesterol or triglyceride concentration. The impact of other variables such as hypertension, waist-to-hip ratio, body mass index, physical activity at leisure and work, mental stress, educational level and hormonal replacement therapy on the RR for change in cholesterol and triglyceride concentration was controlled separately.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Subjects and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

End-points

The end-points studied were mortality and myocardial infarction and are shown in Table 1.

Cholesterol and triglyceride concentrations as predictors of myocardial infarction and death in the total population sample

According to the Cox regression model, no difference of statistical significance from those expected was observed for incidence of myocardial infarction and total mortality in relation to baseline values of serum cholesterol concentration. However, with increasing serum triglyceride concentration there was an increased incidence of myocardial infarction (RR 2.08, CI 1.59–2.72, P < 0.001) and death (RR 1.58, CI 1.30–1.92, P < 0.001) (Table 2).

Table 2.   Serum cholesterol and serum triglycerides as risk factors for myocardial infarction and death in women divided into different age groups. Relative risks (RR) reflect 1 unit increase in lipid concentration, adjusted for the other lipid factor and smoking Thumbnail image of

Triglyceride and cholesterol concentrations as predictors of myocardial infarction and death in relation to age at baseline

Based on a Cox regression model (Table 2), the incidence of myocardial infarction and mortality increased with increasing serum triglyceride values. The risk ratio was statistically significantly increased in women aged 50, 54 and 60 years at baseline. For serum cholesterol concentration the incidence of total mortality was statistically significantly increased only in women aged 38 years at baseline. In the other age cohorts there was no statistically significant association between cholesterol and either myocardial infarction or death.

Smoking

In the total population sample smoking was a significant risk factor for both myocardial infarction (RR 1.85, CI 1.21–2.82, P < 0.01) and death (RR 1.56, CI 1.22–2.00, P < 0.001) analysed with cholesterol and triglyceride concentrations and age as background factors. When examined after age-stratification at baseline, smoking was a significant or close to significant risk factor only in some of the younger age cohorts, age 38–50 years.

Effect of other variables besides lipids and smoking

After individual adjustment for additional covariates (waist-to-hip ratio, hypertension, mental stress, physical activity at leisure and work, HRT, body mass index and educational level), the initially observed associations were compared with further adjusted result. On the whole, the RRs were not much affected. However, for waist-to-hip ratio and body mass index some attenuation of the RR for triglycerides was seen particularly for myocardial infarction. None of these tested variables had a major impact on the RR associated with cholesterol. We concluded that the associations between triglycerides and both end-points could be partly, but not entirely, explained by anthropometric measures.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Subjects and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The prospective population study of women in Gothenburg, Sweden, has now been going on for more than 30 years, and the present report refers to the 24-year period between 1968 and 1969 and 1992–1993, 1992–93 being the time for the fourth follow-up study of the population sample, at which time different end-points were also confirmed by adding information from different registers to the information given by the participating women. In this way the accuracy of the end-points is to be considered as very high. This is further strengthened by the facts that the participation rates initially and throughout the period have been very high, and that we can follow all participants, including those who have not participated in the follow-up studies, by means of different registers.

On the other hand, the numbers of participants in the baseline study were limited, particularly when studying different age cohorts separately. This is the reason why we waited 24 years to study risk factors for different end-points with respect to different age cohorts. Even after 24 years, the numbers of end-points are limited, especially in the youngest age cohort because of rather low incidence rates, and in the oldest age cohort because of low initial number of participants in that age cohort, which means that conclusions still must be drawn with caution.

When studying the incidence rates of the total population sample, our previous results from our 12-year follow-up [1] and the Stockholm Prospective Study [19] are confirmed, serum triglyceride concentration being a much stronger risk factor than serum cholesterol concentration for the end-points studied. When studying the different age cohorts separately in the Cox regression model, one of the observations worth noting is the consistent association between serum triglyceride concentration and incidence of myocardial infarction and total mortality in the older age cohorts. The lack of statistical significance between triglyceride concentration and the end-points in the youngest age cohort may partly be explained by a low number of end-points in this age, but there is probably also an additional effect of age indicating that serum triglyceride concentration is more important as a risk factor with increasing age. The other observation worth noting is the association between serum cholesterol concentration and all-cause mortality in the youngest age cohort, women 38 years at baseline. A similar trend, albeit not statistically significant, was observed between serum cholesterol and incidence of myocardial infarction in this age cohort.

The statistical analyses have been based on the baseline values of serum cholesterol and serum triglyceride concentrations, although the follow-up period has been as long as 24 years. It is, however, reasonable to believe that the differences observed would have been even more obvious, had it been possible to relate the end-points studied to serum lipid values and smoking habits followed continuously during the study period.

The role of triglycerides as an independent risk factor for cardiovascular disease has been debated during a long time. However, our results are consistent with the results from the Framingham Heart Study in 1972 [20], in which women under the age of 50 and who had high serum cholesterol values were associated with an increased risk of coronary heart disease. In women over the age of 50 high levels of the triglyceride-rich pre-β lipoprotein (Sf 20–400) gave a better prediction of the risk of coronary heart disease. The findings in the Framingham Heart Study was virtually replicated in the Australian population study by Simons [21], in women over the age of 60, for which an independent gradient of coronary artery disease risk was found with increasing triglyceride levels and a similar gradient with decreasing high-density lipoprotein cholesterol. Renewed analyses from the Framingham Heart Study [22] showed men and women who had high triglyceride levels and low high-density lipoprotein levels had a significantly higher rate of coronary artery disease.

In studies on low-density lipoprotein particle size and atherogenicity, a strong relation to triglyceride and high-density lipoprotein levels has been found. Small, dense low-density lipoprotein particles were significantly associated with a threefold increased risk of myocardial infarction, increased levels of triglycerides, very low-density lipoproteins and intermediate-density lipoproteins and decreased plasma levels of high-density lipoprotein cholesterol [23]. Further, the statistical relationship between elevated triglycerides and small, dense low-density lipoprotein is valid for about 50% of subjects with small dense low-density lipoprotein [24].

In a study by Mulcahy et al. [25], women with coronary heart disease compared with controls differed in serum cholesterol concentration, having higher cholesterol concentrations especially in the youngest age group, women aged 35–44 years. A high low-density lipoprotein cholesterol level is a strong predictor of coronary artery disease risk in younger women and somewhat weaker predictor in older women [26]. In a retrospective review of medical records, Arnold and Moodie [27] concluded that hypercholesterolaemia is a risk factor for coronary artery disease in young women with an average age of 28 ± 2.4 years. In a population-based cohort study in the Netherlands by Verschuren and Kromhout [28], primarily aimed at the age group around 40 years of age, a strong positive relationship between total cholesterol concentration and mortality from coronary artery disease and cardiovascular disease was observed in both men and women.

In recent years triglyceride levels have been shown to be an independent risk factor for coronary heart disease in men [29] irrespective of serum levels of high-density lipoprotein cholesterol and low-density lipoprotein cholesterol. Also in a meta-analysis of combined data from 17 population-based prospective studies by Hokanson and Austin [30], triglyceride levels were shown to be a significant risk factor for cardiovascular disease, with a risk increase of 32 and 76% for men and women, respectively. Also in a most recent study amongst Japanese men and women with low mean values of total cholesterol the incidence of coronary heart disease was predicted by serum triglycerides [31].

It is likely that our results, where elevated serum triglyceride levels are seen as a predominant risk factor for myocardial infarction and death in women of older age, is a reflection of increase in triglyceride-rich lipoproteins, particularly very low-density lipoproteins, which may be directly atherogenic. In some respect it could reflect other underlying derangements of lipid metabolism.

We have emphasized earlier that men and women differ in respect of what is the predominant lipid risk factors, hypercholesterolaemia being the most dangerous risk factor in men and hypertriglyceridaemia in women. It now seems that there is a difference between younger and older women, elevated serum cholesterol being the most dangerous risk factor in younger women and serum triglycerides in older women.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Subjects and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The study was supported by the Swedish Medical Research Council (27X-04578-27C) and the Swedish Council for Planning and Coordination of Research (0950:5 A19-5/67).

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Subjects and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • 1
    Lapidus L, Bengtsson C, Lindquist O, Sigurdsson JA, Rybo E. Triglycerides – main lipid risk factor for cardiovascular disease in women? Acta Med Scand 1985; 217: 4819.
  • 2
    Carlson LA, Böttiger LE. Ischaemic heart-disease in relation to fasting values of plasma triglycerides and cholesterol. Stockholm Prospective Study. Lancet 1972; 1: 8658.
  • 3
    Kannel WB, Castelli WP, Gordon T. Cholesterol in the prediction of atherosclerotic disease: new perspectives based on the Framingham Study. Ann Intern Med 1979; 90: 8591.
  • 4
    Kiyohara Y, Ueda K, Fujishima M. Smoking and cardiovascular disease in the general population in Japan. J Hypertens Suppl 1990; 8: S915.
  • 5
    Nyboe J, Jensen G, Appleyard M, Schnohr P. Smoking and the risk of first acute myocardial infarction. Am Heart J 1991; 122: 43847.
  • 6
    Njolstad I, Arnesen E, Lund-Larsen PG. Smoking, serum lipids, blood pressure and sex differences in myocardial infarction. Circulation 1996; 93: 4506.
  • 7
    Bengtsson C, Blohmé G, Hallberg L et al. The study of women in Gothenburg 1968–1969 – a population study. General design, purpose and sampling results. Acta Med Scand 1973; 193: 3118.
  • 8
    Bengtsson C, Hallberg L, Hällström T et al. The population study of women in Göteborg 1974–75 – the second phase of a longitudinal study. General design, purpose and sampling results. Scand J Soc Med 1978; 6: 4954.
  • 9
    Bengtsson C, Gredmark T, Hallberg L et al. The population study of women in Gothenburg 1980–81 – the third phase of a longitudinal study. Comparison between participants and non-participants. Scand J Soc Med 1989; 17: 1415.
  • 10
    Bengtsson C, Ahlqwist M, Andersson K, Björkelund C, Lissner L, Söderström M. The prospective population study of women in Gothenburg, Sweden, 1968–69 to 1992–93. Scand J Prim Health Care, 1997; 15: 2149.
  • 11
    Levine JB, Zak B. Automated determination of serum total cholesterol. Clin Chim Acta 1964; 10: 3814.
  • 12
    Lofland HB. A semiautomated procedure for the determination of triglycerides in serum. Anal Biochem 1964; 9: 393400.
  • 13
    Bengtsson C, Björkelund C, Lapidus L, Lissner L. Associations of lipid concentrations and obesity with mortality in women: 20 year follow up of participants in prospective population study in Gothenburg, Sweden. BMJ 1993; 307: 13858.
  • 14
    Sigurdsson JA, Bengtsson C. Symptoms and signs in relation to blood pressure and antihypertensive treatment. Acta Med Scand 1983; 213: 18390.
  • 15
    Lissner L, Bengtsson C, Björkelund C, Wedel H. Physical activity levels and changes in relation to longevity. A prospective study of Swedish women. Am J Epidemiol 1996; 143: 5462.
  • 16
    Bengtsson C, Hällström T, Tibblin G. Social factors, stress experience and personality traits in women with ischaemic heart disease, compared to a population sample of women. Acta Med Scand 1973; Suppl 549: 8292.
  • 17
    Cabrera C, Helgesson Ö, Wedel H, Björkelund C, Bengtsson C, Lissner L. Socioeconomic status and mortality in Swedish women: opposing trends for cardiovascular disease and cancer. Epidemiology 2001; 12: 5326.DOI: 10.1097/00001648-200109000-00012
  • 18
    Rödström K, Bengtsson C, Lissner L, Björkelund C. Pre-existing risk factor profiles in users and non-users of hormone replacement therapy: prospective cohort study in Gothenburg, Sweden. BMJ 1999; 319: 8903.
  • 19
    Carlson LA, Böttiger LE. Risk factors for ischemic heart disease in men and women. Acta Med Scand 1985; 218: 20711.
  • 20
    Kannel WB, Castelli WP. The Framingham Study of coronary disease in women. Med Tms (NY) 1972; 100: 17395.
  • 21
    Simons L. Triglyceride levels and the risk of coronary artery disease: a view from Australia. Am J Cardiol 1992; 70 (Suppl.):14H18H.
  • 22
    Castelli WP. Epidemiology of triglycerides: a view from Framingham. Am J Cardiol 1992; 70: 3H9H.
  • 23
    Austin MA, Breslow JL, Hennekens CH, Buring JE, Willett WC, Krauss RM. Low-density lipoprotein subclass patterns and risk of myocardial infarction. JAMA, 1988; 260: 191721.
  • 24
    Superko HR. Small, dense, low-density lipoproteins and atherosclerosis. Curr Ather Rep 2000; 2: 22631.
  • 25
    Mulcahy R, Hickey N, Maurer B. Coronary heart disease in women. Study of the risk factors in 100 patients less than 60 years of age. Circulation 1967; 36: 57786.
  • 26
    Manolio TA, Pearson TA, Wenger NK, Barret-Connor E, Payne GH, Harlan WR. Cholesterol and heart disease in older persons and women: review of an NHLBI workshop. Ann Epidemiol 1992; 2: 16176.
  • 27
    Arnold AZ, Moodie DS. Coronary artery disease in young women: risk factor analysis and long-term follow up. Cleve Clin J Med 1993; 60: 3938.
  • 28
    Verschuren WMM, Kromhout D. Total cholesterol concentration and mortality at a relatively young age: do men and women differ? BMJ 1995; 311: 77983.
  • 29
    Assman G, Cullen P, Schulte H. The Munster Heart Study (PROCAM). Results of follow-up at 8 years. Eur Heart J 1998; 19 (Suppl. A): A2A11.
  • 30
    Hokanson JE, Austin MA. Plasma triglyceride level is a risk factor for cardiovascular disease independent of high-density lipoprotein cholesterol level: a meta-analysis of population-based prospective studies. J Cardiovasc Risk 1996; 3: 2139.
  • 31
    Iso H, Naito Y, Sato S et al. Serum triglycerides and risk of coronary heart disease among Japanese men and women. Am J Epidemiol 2001; 153: 4909.