We assessed the determinants of onset of hypertension in a large, prospective population-based study of perimenopausal women from the Kuopio Osteoporosis Risk Factor and Prevention (OSTPRE) study. The data collection started in 1989, when a baseline postal inquiry was sent to all women aged 47–56 years (n=14 220) residing in the Kuopio Province in Eastern Finland. Names, social security numbers and addresses were obtained from the Population Register Centre of Finland. A total of 11 798 women responded at baseline and at 5-year follow-up in 1994. After the exclusion of 1777 women with prevalent hypertension at baseline and women with missing height or weight information, the study population consisted of 9485 without established hypertension at baseline. New cases of established hypertension during the follow-up (n=908) were ascertained with the Registry of Specially Refunded Drugs of the Finnish Social Insurance Institution (SII). According to the National Health Insurance, the SII granted 90% reimbursement for drug costs in defined chronic illnesses necessitating continuous medication, like arterial hypertension. Weight and weight gain both raised the risk by 5% per kg (P<0.001). Weight gain of 4–6 kg increased the risk of hypertension 1.25 times and a gain of more than 7 kg 1.65 times compared with the control (zero) group. To conclude, the onset of hypertension in peri- and early postmenopausal women was related to an increase in body weight despite controlling for initial body weight, reported physical activity and use of HRT. Therefore, preventing weight gain by dietary means and exercise is of great importance at menopausal age.
Men have higher blood pressure (BP) levels than women and the difference can be seen already in adolescence through adulthood. On average, systolic (SBP) and diastolic (DBP) blood pressure levels are 6–7 and 3–5 mmHg, respectively, higher in men than women between comparable age groups.1,2 However, after the age of 60 years, this sexual dimorphism in BP levels gradually attenuates and becomes even reversed especially due to marked increase in SBP levels in women.1,2 In this regard, the role of menopause and loss of ovarian function has gained interest. Indeed, cross-sectional studies show that there occurs a significant increase in SBP and DBP levels after the onset of menopause.3,4 In a study by Staessen et al,4 the increase in SBP per decade was 5 mmHg greater in perimenopausal and postmenopausal women as compared with premenopausal women. Further, the prevalence of hypertension was increased four-fold in postmenopausal as compared to premenopausal women.4 However, in the few longitudinal studies available5,6,7,8 the menopausal transition of hypertension has not been convincingly demonstrated. The current view is that BP does not rise during the transition phase from perimenopause to menopause, but rather that the increase in BP takes an average of 5–20 years to develop.7,9,10 However, environmental determinants of the postmenopausal hypertension may be more critical than the loss of ovarian function per se. As the longitudinal studies on this topic are scarce, we analysed the determinants of the onset of persistent hypertension in 9485 peri–postmenopausal women in a population-based study followed-up to 5 years.
Subjects and methods
This study was performed on the study population from the Kuopio Osteoporosis Risk Factor and Prevention (OSTPRE) study. The data collection started in May 1989, when a baseline postal inquiry was sent to all women aged 47–56 years (n=14 220) residing in the Kuopio Province in Eastern Finland.11 Names, social security numbers and addresses were obtained from the Population Register Centre of Finland. After 5 years, in May 1994, a follow-up inquiry was sent to all the living members (n=13 989) of the original cohort. In all, 11 798 (84%) women responded both in 1989 and 1994. The enquiries included detailed questions about body weight and height, gynaecological history including parity and use of hormonal replacement therapy (HRT), physical activity, smoking habits, work disability, use of dairy products and alcohol, chronic heath disorders and physical loading at work. Reproducibility and validation of the baseline enquiry was checked with repeated inquires on a subsample.12
The data from records of the baseline and 5-year follow-up inquiry were pooled together with the records of the Registry of Specially Refunded drugs of the Finnish Social Insurance Institution (later SII). The final study population (n=9485) was formed by including cases with information on height at the baseline and weight at baseline and in the 5-year follow-up (n=11 155), but excluding cases (n=1777) with hypertension already diagnosed, that is, medication reimbursed by the SII before the baseline in 1989.
New cases of established hypertension during follow-up (N=908) were ascertained with the Registry of Specially Refunded Drugs of the SII. According to the National Health Insurance, the SII granted, in 1989, 90% reimbursement for drug costs in defined chronic illnesses necessitating continuous medication, like arterial hypertension. This National Health Insurance is statutory and comprehensive and covers equitably the whole Finnish population. SII is supervised by the Finnish Parliament and the costs of reimbursement are financed by tax revenues. To acquire the reimbursement, one has to complete an application to the SII accompanied by the statement from his/her physician. This medical certificate must include the history of hypertension, repeated BP measurements, physical examination, relevant laboratory investigations and radiological findings targeting especially to show possible signs of secondary complications of hypertension. According to the National Health Insurance the criteria for reimbursement of medication in chronic hypertension were, (in 1989) as follows:
DBP 105 mmHg or over continuously and no signs of secondary complications.
Diastolic blood pressure between 100 and 104 mmHg and male under 50 years old/female under 40 years old or several cases of cardiovascular disorders or deaths under the age of 55 years in close relatives, diabetes or familial hypercholesterolemia or SBP continuously over 180 mmHg (under 50 years) or 200 mmHg (over 50 years), signs of secondary complications (heart failure, left ventricular hypertrophy, renal failure, coronary hearth disease, arteriosclerosis of large vessels, retinopathy, etc).
DBP 95 mmHg or over and diabetic nephropathy or other severe renal insufficiency.
In borderline cases with fluctuating BP levels follow-up of at least 6 months was required.
Standard package of SPSS software was used for data processing. The final statistical model included 12 variables. The following variables were treated as categorical variables: smoking (no/yes), leisure time physical activity (I/no activity, II/1–4 h per week, III/5 or more hours per week), use of hormone-replacement therapy (HRT) before baseline (no/yes) and during follow-up ((I/no, II/occasional, III/continuous), menopausal status (pre/post), work status (at work/retired) and physical loading at work (I/sedentary or slight, II/intermediate, III/heavy). Other variables (age, weight, weight change, height, parity) were processed as continuous. A woman was regarded as postmenopausal, when 6 months had elapsed since the last menstruation. With those subjects who had started HRT before menopause, the time of menopause was located at the start of usage of HRT. Association between risk factors and incidence of hypertension was analysed with logistic regression. Hypertension was treated as a dependent variable. In multivariate analyses, all 12 covariates were put in model at the same time. Additional subanalysis was performed with weight gain as a four-category variable: 0 kg or less, 1–3 kg, 4–6 kg, 7 kg or more.
The final study population consisted of 9485 participants. There was minimal variation in the number of subjects in the statistical analysis due to question-specific missing information. The baseline characteristics of the women are presented in Table 1. The mean weight (±s.d.) at the baseline was 67.3±11.1 kg and it increased on average by 2.4±4.8 kg during the follow-up. Mean BMI at baseline was 25.8±4.0 kg/m2.
In all, 67.5% of the subjects were postmenopausal at the baseline and 2796 (27.5%) of the women turned to postmenopausal phase during the study period. At the end of study period, 5% were still premenopausal. By the baseline, 33.3% of the women had used HRT and the mean duration of HRT was 1.1+2.3 years. In the follow-up inquiry, 58.7% of the women announced that they had not used HRT during follow-up and correspondingly 25.6 and 15.7% of the women informed occasional or continuous use of HRT during the follow-up, respectively. Of the participants, 5.4% had bilateral ophorectomy performed before baseline.
At the baseline, 12% (n=1192) of the participants were retired and 16% (n=1530) retired during the study period. The most frequent diagnoses for retirement during the study period were degenerative lumbar and joint disorders (n=715), depression (n=130), neurotic anxiety (n=74), coronary heart disease (n=98) and asthma (n=66). A total of 842 (8.9%) acknowledged smoking habits in both baseline and follow-up inquiries. Starting and cessation of smoking was quite rare, because only 3.5% women changed their smoking habits during the study period. Of the study population, 47% did not perform any leisure time physical activity at the baseline, 3309 (36.7%) had leisure time physical activity at baseline and in follow-up studies, while 32.8% were inactive both at the baseline and at the end of follow-up (Table 1).
At the baseline, 9.0% of the postmenopausal and 10.7% of the premenopausal women had reimbursed antihypertensive drugs. Altogether 908 new cases of established hypertension were registered during follow-up. Incidence of hypertension was associated with baseline BMI: 6.9% for BMI⩽18.9 kg/m2, 6.4% for 19–24.9 kg/m2, 10.8% for 25–29.9 kg/m2 and 16.7% for BMI⩾30 kg/m2, respectively (P<0.011).
In univariate logistic regression analysis (Table 2) weight at the baseline, increase of weight, premenopausal status at baseline, retirement during follow-up and physical inactivity were associated with the new onset of hypertension. Body weight at baseline and its increase during follow-up raised the risk of hypertension by 3.2% (P<0.001) and 3.8% (P<0.001) per kg, respectively. Postmenopausal women had about 17% greater risk of hypertension than premenopausal women. Retirement during follow-up was associated with an increased risk of hypertension of nearly 50% (P<0.001). Age, body height, smoking, use of HRT, physical load at work and parity were not associated with the risk of hypertension.
In multivariate analysis (Table 3), weight at baseline, increase of weight, postmenopausal status at baseline, height and retirement during follow-up were all predictive of hypertension (P<0.01). Weight and increase of weight both raised the risk by 5% per kg (P<0.001), whereas height decreased the risk by 4% per 1 cm (P<0.001). Retirement during the follow-up increased the risk by 29% (P<0.001). Postmenopausal phase at baseline decreased the risk by 17% (P<0.05). Other covariates (age, smoking, use of HRT before or after baseline, physical load at work and parity) included in this model did not have a significant association to onset of hypertension.
In further analysis the risk of hypertension increased with the weight increase (Figure 1), so that the threshold was at 4 kg. This amount of weight increase was associated with an increased risk (P<0.05) as compared to the zero group (=weight losers or no change). Weight gain of 4–6 kg increased the risk of hypertension 1.25 times and gain more than 7 kg 1.65 times compared with the control (no weight change) group.
The main finding of our study was that the incidence of hypertension in peri- and postmenopausal women was related to increase in body weight regardless of initial body weight, reported physical activity and use of HRT. Further, the findings imply that the onset of hypertension coupled with weight gain may become manifested within a fairly short period.
The public health implication of this study is that menopause seems to be quite a critical time with respect to the onset of hypertension and even mild weight gain increases the risk for a chronic disorder such as hypertension. It is important to note that the results do not reflect menopause per se; rather they reflect the age-associated increase in body weight currently occurring at alarming rate throughout the developed and even developing countries;13 as a matter of fact, the mean increase in body weight during the 5 years is comparable to what has previously occurred in the Finnish population.14 Thus, our findings are in agreement with earlier ones suggesting that even modest weight increase may be an important determinant of hypertension in peri–postmenopausal women.15,16,17 It is known for decades that weight gain increases BP, although the mechanism(s) by which increased fat mass actually increases blood pressures remains to be clarified. On the other hand, there are plenty of data showing that weight loss decreases BP. However, studies assessing the effect of weight loss on BP have frequently failed to take into account the effect of continued weight loss or negative energy balance on BP.18,19,20 Actually there are quite a few studies that should have assessed the effects of weight maintenance after marked weight loss on BP. The Swedish Obesity Study showed that after marked surgically induced weight loss there was no improvement in the long-term frequency of hypertension,21 which are in accordance with our study22 in which diet-induced weight loss in abdominally obese subjects decreased ambulatory BP levels after weight loss, but during weight maintenance the BP returned to pretreatment levels. As the achievement of permanent weight loss is notoriously difficult, it seems that prevention of age-related weight gain is of enormous importance in preventing the onset of hypertension and its adverse consequences.
The strength of this study is the large population-based sample and therefore it is most likely representative of the catchment area. Further, the study design was prospective and controlled for the major confounders. On the other hand, our study has obvious limitations. First, the body weight was self-reported. People's ability to report their weight varies and the variation is associated with certain personal characteristics.23,24,25,26 In particular, obese subjects often underestimate their body weight whereas lean ones do over-report their weight. However, in the subanalysis of this study, the self-reported weight was quite reproducible (r=0.93),12 and given the large number of subjects studied it is unlikely that consistent underestimation could have markedly biased the results. Further, our criteria for hypertension was the limit for drug reimbursement by the Social Insurance Institution, which are remarkably higher than what current guidelines suggest. Thus, the criteria were not very sensitive; but on the other hand, specificity was likely to have been quite high.
To conclude, the onset of hypertension in peri- and postmenopausal women was related to an increase in body weight even when the effects of other determinants, like initial body weight, reported physical activity and use of HRT, were taken into account. Therefore preventing weight gain by dietary means and exercise is of great importance at a perimenopausal age.
Staessen J et al. Reference values for ambulatory blood pressure: a meta-analysis. J Hypertens 1990; 8 (Suppl 6): S57–S64.
Burt VL et al. Prevalence of hypertension in the US adult population. Results from the Third National Health and Nutrition Examination Survey. 1988–1991. Hypertension 1995; 25: 305–313.
Weiss NS . Relationship of menopause to serum cholesterol and arterial blood pressure: the United States Health Examination Survey of adults. Am J Epidemiol 1972; 96: 237–241.
Staessen J et al. The influence of menopause on blood pressure. J Hum Hypertens 1989; 6: 427–433.
Matthews KA, Meilahn, Kuller LH . Menopause and risk factors for coronary heart disease. N Engl J Med 1989; 321: 641–645.
Hjotrland MC, McNamara PM, Kannel WB . Some atherogenic concomitants of menopause: the Framingham Study. Am J Epidemiol 1976; 103: 304–311.
Luoto R et al. Blood pressure and menopausal transition: the Atheroscerosis Risk in Communities Study (1987–95). J Hypertens 200; 18: 27–33.
Staessen JA, Ginocchio G, Thijs L, Fagard R . Conventional and ambulatory blood pressure and menopause in a prospective population study. J Hum Hypertens 1997; 11: 507–514.
Reckelhoff JF . Gender differences in the regulation of blood pressure. Hypertension 2001; 37: 1199–1208.
Dubey RK, Oparil S, Imthurn B, Jackson EK . Sex hormones and hypertension. Cardiovasc Res 2002; 53: 688–708.
Honkanen R, Alhava E, Saarikoski S, Tuppurainen M . Osteoporosis risk factors in perimenopausal women. Calcif Tissue Int 1991; 49 (Suppl): 74–75.
Tuppurainen M et al. Osteoporosis risk factors, Gynaecological history and fractures in perimenopausal women — results of the baseline enquiry of the Kuopio Osteoporosis Risk factor and Prevention Study. Maturitas 1993; 17: 89–100.
James PT, Leach R, Kalamara E, Shayeghu M . The worldwide obesity epidemic. Obes Res 2001; 9: 228S–233S.
Rissanen A, Heliovaara M, Aromaa A . Overweight and anthropometric changes in adulthood: a prospective study of 17,000 Finns. Int J Obes 1988; 12: 391–401.
Grobbee DE et al. Importance of body weight in determining rise and level of blood pressure in postmenopausal women. Hypertens 1988; 4 (Suppl): S614–S616.
Portaluppi F, Pansini F, Manfredini R, Mollica G . Relative influence of menopausal status, age, and body mass index on blood pressure. Hypertension 1997; 29: 976–979.
Huang Z et al. Body weight, weight change, and risk of hypertension in women. Ann Intern Med 1998; 128: 81–88.
MacMahon SW et al. Comparison of weight reduction with metoprolol in treatment of hypertension in young overweight patients. Lancet 1985; 1: 1233–1236.
De Simone G et al. Weight reduction lowers blood pressure independently of salt restriction. J Endocrinol Invest 1992; 15: 339–343.
Blumenthal JA et al. Exercise and weight loss reduce blood pressure in men and women with mild hypertension: effects on cardiovascular, metabolic, and hemodynamic functioning. Arch Intern Med 2000; 160: 1947–1958.
Sjöström CD, Peltonen M, Wedel H, Sjöström L . Differentiated long-term effects of intentional weight loss on diabetes and hypertension. Hypertension 2000; 36: 20–25.
Laaksonen DE et al. Weight loss and weight maintenance, ambulatory blood pressure and cardiac autonomic tone in obese persons with the metabolic syndrome. J Hypertens 2003; 21: 371–378.
Jalkanen L, Tuomilehto J, Tanskanen A, Puska P . Accuracy of self-reported body weight compared to measured body weight, A population study. Scand J Soc Med 1987; 15: 191–198.
Kuskowska-Wolk A, Karlsson P, Stolt M, Rössner S . The predictive validity of body mass index based on self-reported weight and height. Int J Obes 1989; 13: 441–453.
Kuskowska-Wolk A, Bergström R, Boström G . Relationship between questionnaire data and medical records of height, weight and body mass index. Int J Obes 1992; 16: 1–9.
Palta M, Prineas RJ, Berman R, Hannan P . Comparison of self-reported and measured height and weight. Am J Epidemiol 1982; 2: 223–230.
This study was supported by the Academy of Finland, the grants Nos, 45052 and 39955.
About this article
Cite this article
Juntunen, M., Niskanen, L., Saarelainen, J. et al. Changes in body weight and onset of hypertension in perimenopausal women. J Hum Hypertens 17, 775–779 (2003). https://doi.org/10.1038/sj.jhh.1001611
- blood pressure
- body weight
Long-Distance Skiing and Incidence of Hypertension: A Cohort Study of 206,889 Participants in a Long-Distance Cross-Country Skiing Event
Combined effect of weight gain within normal weight range and parental hypertension on the prevalence of hypertension; from the J-MICC Study
Journal of Human Hypertension (2019)
Effect of estrodiol on leptin receptors expression in regulating fat distribution and adipocyte genesis
Gynecological Endocrinology (2016)
Inverse relationship between a genetic risk score of 31 BMI loci and weight change before and after reaching middle age
International Journal of Obesity (2016)