Age-Related Changes in Blood Pressure STEN LANDAHL, CALLE BENGTSSON, JOHAN A. SIGURDSSON, ALVAR SVANBORG, AND KURT SVARDSUDD SUMMARY This report is based on three different representative population samples of a total of 1304 men (50-79 years old) and 1246 women (38-79 years old) observed for up to 12 years. Subjects' consumption of antihypertensive drugs and blood pressure levels in subjects with and without such treatment are presented. The prevalence of treatment with antihypertensive drugs (including /3-blockers and diuretics for other Indications) increased from 2% at age 50 years to 37% at 79 years of age among the men and from 1% at 38 years to 61 % at 79 years of age among the women. The mean systolic/diastolic blood pressure in untreated subjects increased from 138/91 mm Hg at age 50 years to 159/91 mm Hg at age 70 years in the men and from 123/79 mm Hg at age 38 years to 168/93 mm Hg at age 70 years in the women. At age 79 years the mean systolic/diastolic blood pressure was 155/83 mm Hg in the men and 161785 mm Hg in the women. In a longitudinal follow-up of reexamined subjects, there was an increase in systolic blood pressure levels up to age 75 years and a reduction in diastolic blood pressure after age 75 years in both sexes. (Hypertension 8: 1044-1049, 1986) KEY WORDS • blood pressure levels • antihypertensive treatment longitudinal study • epidemiological study Downloaded from http://ahajournals.org by on December 8, 2021 S SBP or DBP at higher ages. 4 6 Generally, reference values for arterial blood pressures are lacking for elderly people. At the same time, however, the prescription and consumption of antihypertensive drugs increase sharply at ages above 60 years. 13 ' 14 This report is based on results from three different longitudinal population studies of men and women living in Goteborg and representing the age interval 38 to 79 years in women and 50 to 79 years in men. Goteborg is the second largest city in Sweden, with 450,000 inhabitants within the city borders and about 1 million in the metropolitan area. It has a considerable amount of industry, the biggest harbor in the country, and an advanced school and university system. Goteborg is located on the west coast and has a milder climate than the middle and northern parts of Sweden. All three studies were designed to obtain samples representative of the general population,13-17 and reports of comparisons between responders and nonresponders are available.15"19 The aim of this report was to describe blood pressure levels studied both longitudinally over 9- to 12-year periods and cross-sectionally in the total sample population as well as in those without drugs influencing the blood pressure. EVERAL previous studies from industrialized countries have reported age-associated changes in both systolic (SBP) and diastolic (DBP) blood pressures. These changes seem to be different for SBP and DBP and have also been reported to be different in male and female subjects.1"12 The majority of these studies have been performed on age groups below 60 to 65 years of age. Blood pressure measurements on representative groups of populations above these ages are lacking, although certain previous reports indicate that there is no further age-related rise in From the Department of Geriatric and Long-term Care Medicine (S. Landahl, A. Svanborg), Vasa Hospital, and the Department of Primary Health Care (C. Bengtsson, J.A. Sigurdsson), and the Department of Medicine I (K. Svfirdsudd), 5stra Hospital, University of Goteborg, Gdteborg, Sweden. The study of 70-year-old persons in Goteborg was made possible through grants from the Swedish Ministry of Health and Social Affairs, the Commission for Social Research, the Swedish Medical Research Council, the Gdteborg Medical Services Administration, the Goteborg Administration of Social Services, and the Wilhelm and Martina Lundgren Foundation. Address for reprints: Sten Landahl, Department of Geriatric and Long-term Care Medicine, Vasa Hospital, S-411 33 Go'teborg, Sweden. Received June 10, 1985; accepted June 3, 1986. 1044 1045 AGE-RELATED CHANGES IN BLOOD PYiESSUREJLandahl et al. Downloaded from http://ahajournals.org by on December 8, 2021 Study Populations and Methods The number of participants at different ages is shown in Table 1. The study of men born in 1913 and 192313 started in 1963 and included a sample of 973 men, all born in 1913. Theresponserate was 88%, and comparisons between responders and nonresponders have been described.18- " Certain differences in mean income, marital status, and alcoholism were observed. A total of 703 of the men have been observed for 10 years. Another 50-year-old cohort of 226 men (bom in 1923) was included in the study 10 years later. The study of women in Goteborg started in 1968—1969 with a sample of 1622 women bom in 1930, 1922, 1918, 1914, and 1908. The response rate was 90%, and comparisons between responders and nonresponders have been reported.16 The nonresponders were studied through telephone interviews and records from outpatient and inpatient departments. No differences were found with respect to place of birth, presence of heart disease, and blood pressure. This report deals with the women bom in 1918 and examined at ages 38, 44, and 50 years and those bom in 1930 and examined at 50, 56, and 62 years of age. Seventy-five percent of these women had been observed for 12 years. The "70-year-old people in Goteborg" study started in 1971-1972 with 973 women and men examined out of a sample of 1148 persons bom in 1901-1902 (a response rate of 85%). The representativity of this sample has been demonstrated previously.17 The nonresponders (15%) did not differ with respect to such variables as income, marital status, or previous hospital care. Recent observations showed a similar mortality at age 79 years in responders and nonresponders (A. Svanborg, unpublished data, 1986). We were able to follow up 46% of the originally examined men (208) and 63% of the women (328) at age 79 years. Another 70-year-old cohort bom in 1906-1907 (473 men and 561 women) was examined in 1976-1977. TABLE 1. Number of Participants at Different Ages Women Age (yr) Men 38 372* — 44 336* — 50 308* 855t — 54 792t — 56 35Ot — 60 718t — 325* 62 70 449§ 524|| 75 329§ 410|| 208§ 79 328|| •Women born in 1930. tMen born in 1913. tWomen born in 1918. §Men born in 1901-1902. I Women born in 1901-1902. Blood pressure was measured manually with the cuff method after the subjects had spent 5 minutes at rest in a sitting position and wasregisteredto the nearest 5 mm Hg in every subject except for the women bom in 1918 and 1930 and for the 60-year-old men, when the nearest 2 mm Hg was used. As far as DBP is concerned, Korotkoff s Phase IV and V were measured, but only Phase V data are used in this report, with the exception of the 50-year-old men examined in 1963, for whom only Phase IV was used. The majority of antihypertensive drugs used in these populations included thiazide and thiazidelike diuretics, £-adrenergic receptor blocking agents, hydralazine, prazosin, calcium influx inhibitors, spironolactone, and a-methyldopa. Results The use of antihypertensive drugs at the time of the analyses in subjects examined between 1963 and 1981 is shown in Table 2. The percentage treated was higher in women than in men, and the prevalence of such treatment was approximately doubled in both sexes for every decade above 50 years. Table 2 also shows the proportion treated because of hypertension, which was known for the age interval 70 to 79 years. The mean SBP at various ages among the different cohorts is given in Table 3 and Figure 1. All persons examined on each occasion, except for those treated with antihypertensive drugs are included. There was an increase in the mean SBP up to age 70 years and a moderate decrease from age 70 to 75 years. The agerelated increase in SBP was less pronounced in men, resulting in a higher (about 10 mm Hg) mean SBP in women than in men after age 60 years. In men, there was also a moderate decrease in the mean SBP after age 70 to 75 years. In the total study population (treated patients included), the mean SBP in women was 1 to 7 mm Hg higher and in men 1 to 3 mm Hg higher TABLE 2. Treatment with Blood Pressure-Lowering Drugs* Percentage Age (yr) 38 44 50 54 56 60 62 70 75 79 Men — — 2 6 — 10 — Women 3 4 8 — 15 — 20 39 (30) 49 (39) 61 (39) 19 (13) 29 (17) 37 (19) Numbers in parentheses indicate the percentage of subjects with a history of antihypertensive treatment. */9-blockers, diuretics, and/or other antihypertensive drugs. HYPERTENSION 1046 TABLE 3. Systolic Blood Pressure at Various Ages in Subjects Without Antihypertensive Treatment Men Age (yr) VOL 8, No 11, NOVEMBER 1986 TABLE 4. Diastolic Blood Pressure at Various Ages in Subjects Without Antihypertensive Treatment Men Women Women SBP (mm Hg) No. SBP (mm Hg) No. Age (yr) DBP (mm Hg) No. DBP (mm Hg) No. — — 123±14.4 369* 38 — — 79±9.1 369* 38 44 — — 124±16.4 322* 44 — — 81 ±8.6 322* 50 138 ±20.5 841t 133±19.4 284* 50 91 + 12.8 8411 84±9.8 284* 54 142 ±20.1 747t — — 54 9O±11.8 747t — 56 — 298* 56 — — 84±9.1 60 144±22.2 62 — 70 159 + 25.0 75 79 — 134±18.8 — 60 88 ±12.7 643t — — 143+19.8 259* 62 — — 86 ±10.0 259t 362§ 168±25.5 321|| 70 91 ±12.6 362§ 93±13.2 321|| 90±12.4 234§ 89+12.9 209|| 83±11.3 130§ 85±9.9 127|| — 643t 158±24.5 234§ 166±22.5 209|| 75 155 + 24.6 13O§ 161 ±2 2 . 1 127|| 79 Systolic blood pressure (SBP) expressed as mean ± SD. •Women bom in 1930. tMen born in 1913. tWomen bom in 1918. §Men born in 1901-1902. ||Women bom in 1901-1902. than in those without treatment. However, the pattern of blood pressure change with age in the total population was identical to that among untreated subjects. The mean DBP in women without antihypertensive treatment also increased with age up to 70 years (Table 4; see Figure 1). In men, however, no change in DBP Downloaded from http://ahajournals.org by on December 8, 2021 SBP mmHg /^>9 160 • 140 120 - -Ago 40 — 298t 50 60 70 80 DBP mmHg — Diastolic blood pressure (DBP) expressed as mean ± SD. •Women bom in 1930. tMen bom in 1913. tWomen bom in 1918. §Men bom in 1901-1902. [Women bom in 1901-1902. was observed between 50 and 70 years of age. After age 70 to 75 years, the DBP decreased in both men and women. In the total study population (treated patients included), the mean DBP was 1 to 3 mm Hg higher in both sexes than among those with no antihypertensive treatment. Table 5 shows the longitudinally observed changes in blood pressure observed in subjects without hypotensive drugs: women were examined at the age intervals 38 to 50, 50 to 62, and 70 to 79 years, and men were examined at 50 to 60 and 70 to 79 years. The SBP increased significantly in both sexes at least up to age 75 years. The lower value at the age interval 70 to 79 years compared with that at 70 to 75 years is due to a decrease in SBP between ages 75 and 79 years. The DBP in women still showed an increase at the age interval 50 to 62 years but showed a decline after age 75 years, and no change was noted in DBP in men until after age 75 years, when a significant decline was observed. Table 6 shows the blood pressure levels in those probands born in 1901-1902 who werereexaminedat both age 75 and 79 years. The DBP decreased after age 75 years in both sexes. The blood pressures were also lower in this group of survivors at ages 70 and 75 years than in the total study population. The longitudinal changes showed a unimodal distribution. 100- 80- -Age 40 50 60 70 80 FIGURE 1. Systolic (SBP) and diastolic (DBP) blood pressures in subjects without antihypertensive treatment. Crosssectional data. Discussion Previous epidemiological studies of possible agerelated changes in blood pressure are numerous (for a review, see References 1-12) and have yielded somewhat controversial results. To a great extent, the discrepancies are due to variations in population sampling procedures and the samples investigated, to differences in age-standardized morbidity and longevity be- AGE-RELATED CHANGES IN BLOOD PRESSURE/Landahl et al. 1047 TABLE 5. Longitudinal Changes in Systolic and Diastolic Blood Pressure Among Reexamined Subjects Without Antihypertensive Treatment ASBP (mm Hg) Age interval Whole interval No. ADBP (mm Hg) Per year Whole interval Per year 0.5 Women 38-50 284 12.4± 16.12* 1.0 6.2±8.64* 50-62 251 12.5 ±17.63* 1.0 3.2 ±8.76* 0.3 70-75 127 1.2 -1.5+11.24 -0.3 70-79 127 6.2 ±20.09* 4.3±23.38t — - 4 . 5 ±1 1 . 5 3 * -0.5 1.0 0.7+ 12.83 0.8 1.1+9.81 — -5.6+11.48* Men 70-75 130 9.5± 18.91* 4.1±19.80t 70-79 130 3.7±23.O8 50-^0 639 0.1 0.2 -0.6 Changes in systolic (SBP) and diastolic (DBP) blood pressure expressed as mean ± SD. *p < 0.001, tp < 0.05 (paired t test). Downloaded from http://ahajournals.org by on December 8, 2021 tween, for example, populations in developing and developed countries, and to cultural and environmental differences.12 The Swedish population is presently the oldest in the world, with no less than 17% aged 65 years or more. Compared with most other countries, numeric and cultural differences between different age groups are rather small. The majority of epidemiological studies in countries with high longevity indicate that the SBP rises with age. This is obviously the case in Goteborg, as well as in Framingham, Massachusetts, USA, 7 and Glostrup, Denmark,11 up to age 70 to 75 years. The age-related rise indicated by both cross-sectional and longitudinal comparisons was found to be faster in women than in men. Cross-sectional comparison showed that the regression lines representing SBP versus age in the two sexes cross each other at age 45 years in the Framingham study but not until age 60 to 70 years in the present investigation. Available data indicate that in the Japanese population the female line does not approach the male line until ages above 70 years.20 Longitudinal blood pressure measurement results from the Framingham study7 indicate that, although the agerelated rise was faster in women than in men and the TABLE 6. Systolic and Diastolic Blood Pressures in Subjects 70-79 Years of Age Who Attended All Three Examinations and Received No Antihypertensive Treatment Men (/» = 130) Women (n= 127) (mm Hg) SBP (mm Hg) DBP (mm Hg) 152±19.8 88±9.5 156±17.9 89+10.4 156±22.4* 89+10.4 163±19.8t 88+11.1 155 ±24.6 8 3 ±1 1 . 3 j 161 ±22.1 84 + 9.9J Age (yr) SBP (mm Hg) 70 75 79 DBP Systolic (SBP) and diastolic (DBP) blood pressure expressed as mean + SD. *p < 0.05, tp < 0.001 (paired t test), compared with same-sex SBP in 70-year-old age group; tp < 0.001 (paired / test), compared with same-sex DBP in 75-year-old age group. female lines approach the male ones at about age 70 years, these lines never cross. The present results showed a rate of change of about 1 mm Hg per year in both men and women up to age 75 years longitudinally, while the cross-sectional comparisons showed a faster increase in women than in men. These findings might either imply a higher selective mortality of women with higher blood pressure than of men, or result from the fact that the proportion of women treated with blood pressure-lowering drugs was much higher (see Table 2), resulting in different sex distributions of the untreated groups. There are no data available in the Swedish mortality statistics indicating that women below the age of 70 to 75 years have a higher risk than men of having cardiovascular complications due to hypertension. We therefore conclude from the present results that available epidemiological data indicate that the rate of age-related increase in SBP is faster in women than in men and that at age 70 years, a higher clinical reference value for SBP has to be accepted in women than in men. How great such a physiological difference should be accepted to be — cross-sectionally women had blood pressure levels that were 10 mm Hg higher — remains to be studied in populations with a similar rate of treatment with antihypertensive drugs. The present report is based on three different population studies in samples that obviously allow generalizations to be made about the total population of Goteborg. Indirect evidence such as available information on the consumption of antihypertensive drugs in different areas of Sweden10 indicates that the present observations on age-related changes in SBP and DBP also are relevant for the rest of the population in Sweden. A comparison of the age-related change in DBP is more difficult to make because of the proportionally higher measurement error in relation to the age differences. The fact that KorotkofFs Phase IV was used in the longitudinal analyses for men born in 1913 was shown to be of minor importance, since the change 1048 HYPERTENSION between 54 and 60 years of age showed no difference compared with Phase V data. Also, the change in DBP between 70 to 75 and 79 years of age was similar for both phases. Most results were based onregistrationof blood pressure to the nearest 5 mm Hg, except for the two youngest female cohorts and the 60-year-old male cohort. This methodological difference is of minor importance as far as analyses of longitudinal changes within the different cohorts are concerned but may affect the comparison between the absolute blood pressure levels for different cohorts. In men, previous studies1"12 have shown an age-related increase, at least up to age 50 to 60 years and thereafter a leveling off. The present study of men older than 50 years showed a further decline starting above age 70 to 75 years. There was a sex difference with a further rise also in DBP up to age 70 years in women, while both cross-sectional and longitudinal data showed a decline in the DBP of men and women after age 75 years, after which no sex differences in these age-related trends in blood pressure levels can be seen. Downloaded from http://ahajournals.org by on December 8, 2021 For ethical reasons all the participants in the population studies in Goteborg have had some medical intervention (e.g., the institution of antihypertensive treatment). From this point of view, the best cross-sectional comparison can be made between the "starting" ages (i.e., 38, 50, and 70 years) in the total study population. However, our main conclusions have been based on both the cross-sectional analysis and the results from longitudinal follow-ups in subjects without any drug treatment influencing the blood pressure. The decline in blood pressures during the age interval 75 to 79 years thus could not be due to any intervention given during this age interval. It is also important to bear in mind that the distribution of the blood pressure changes was close to normal with a slightly more pronounced decrease. The reason for the blood pressure decline has not been addressed in this report. Other studies of the same sample21 have shown a marked change of body mass and body composition at these ages. Possible relationships between body mass, body cell mass, and blood pressure can be anticipated. However, it should be emphasized that similar changes in body mass and body composition also took place at the age interval 60 to 70 years, when there still was an increase in SBP in both sexes and in DBP in women. A marked decline in body height could also affect blood pressure, for instance, by changing the demand of blood pressure to ensure adequate cerebral blood flow. Selective mortality and observer bias may also be factors of some importance for this decline in blood pressure. One reason for the increase of SBP up to age 70 to 75 years is a gradual deterioration of elastic filaments in the aorta and large arteries. This exchange of elastic filaments with stiffer collagen elements increases the "Windkessel" phenomenon in these vessels, leading to increased pulse pressure. A decreasing ability with age to handle exogenous blood pressure-affecting factors such as salt intake, extracellular volume, and serum proteins may also be of importance.22 Several previous VOL 8, No 11, NOVEMBER 1986 studies23"29 have investigated possible age-related changes in the production of catecholamines. The results are conflicting, but the majority of these studies conclude that increased norepinephrine levels are balanced by a decrease in receptor sensitivity. Plasma renin activity and renin concentrations have been reported to be lowered by increasing age, but similarly in hypertensive and normotensive elderly.2630 Also, plasma and urine aldosterone concentrations decrease with age.31 The reason why the age-related SBP changes in women are more pronounced than in men is unknown. Neither do we know any reasons for the earlier leveling off of the DBP in men than in women. Possible relationships to hormonal changes causing, for example, differences in salt and water content of the body might be considered. Recent studies have shown a less pronounced blood pressure increase after age 50 years in women who are postmenopausal than in women who still menstruate.32 References 1. B0e J, Humcrfelt S, Wedcrvang F. The blood pressure in a population: blood pressure readings and height and weight determinations in the adult population of the city of Bergen. Acta Med Scand [suppl] 1957;321:77-111 2. McDonough JR, Garrison GE, Hamer CG. Blood pressure and hypertensive disease among negroes and whites: a study in Evans County, Georgia. Ann Intern Med 1964;61:208-228 3. Johnson BC, Epstein FH, Kjelsberg MD. Distributions and familial studies of blood pressure and serum cholesterol levels in a total community: Tecumseh, Michigan. J Chronic Dis 1965;18:147-161 4. Eilertsen E, Humerfelt S. The blood pressure in a representative population sample. Acta Med Scand 1968;183:293-305 5. Miall WE, Chinn S. Blood pressure and ageing: results of a 15-17 year follow-up study in South Wales. Clin Sci Mol Med 1973;45:23-33 6. Blood pressure of persons 18-74 years: United States 1971-72. Washington, DC: U.S. Department of Health, Education and Welfare, 1975 (Vital and health statistics; series 11; no 150) 7. Kannel WB, Gordon T. Evaluation of cardiovascular risk in the elderly: the Framingham study. Bull NY Acad Med 1978^4:573-591 8. Stamler R, Stamler J, Riedlinger WF, Algera G, Roberts RH. Weight and blood pressure:findingsin hypertension screening of 1 million Americans. JAMA 1978;240:1607-1610 9. Svardsudd K, Tibblin G. A longitudinal blood-pressure study: change of blood-pressure during ten years in relation to age and initial level: the study of men born in 1913. J Chronic Dis 1980;33:627-636 10. Amery A, Hansson L, Andreri L, Gudbrandsson T, Sivertsson R, Svensson A. Hypertension in the elderly. Acta Med Scand 1981;210:221-229 11. The Glostrup Population Studies. In: Hagerup L, Eriksen M, Schroll M, et al., eds. Collection of epidemiologic tables. The Danish Heart Foundation (ISBN 87-981OO7-1-8) 12. Waldron I, Nowotarski M, Freimer M, et al. Cross-cultural variation in blood pressure. Soc Sci Med 1982;16:419-430 13. BoethiusG. The treatment of hypertension: an analysis of drug prescription data. Acta Med Scand [suppl] 1976;602:12O-123 14. Landahl S, Steen B. LSkemedelskonsumtion hos 70- och 75aringar. Lakartidningen I981;78:2458-246O 15. Tibblin G. High blood pressure in men aged 50: a population study of men born in 1913. Acta Med Scand [suppl] 1967; 470:1-84 AGE-RELATED CHANGES IN BLOOD PRESSIME/Landahl et al. 16. Bengtsson C, Blohme' G, Hailberg L, et al. The study of women in Gothenburg 1968-69, a population study: general design, purpose and sampling result. Ada Med Scand 1973; 193: 311-318 17. Rinder L, Roupe S, Steen B, Svanborg A. 70-Year-old people in Gflteborg: a population study in an industrialized Swedish city. I. General presentation of the study. Acta Med Scand 1975;198:397-407 18. Tibblin G. A population study of 50-year-old men: an analysis of the non-participation group. Acta Med Scand 1965;178: 453-^54 19. Silverbage-Carlsson G, Svardsudd K. Comparison between participants and non-participants in a population study: the study of men born in 1913 and 1923. Scand J Sec Med 1985;13:15-22 20. Svanborg A, Shibata H, Hatano S, Matsuzaki T. Comparisons of ecology, age and state of health in Japan and Sweden, present and previous leaders in longevity. Acta Med Scand 1985;218:5-17 21. Steen B, Isaksson B, Svanborg A. Body composition at 70 and 75 years of age. Eur J Clin Exp Gerontol 1979;1:185-200 22. Freis ED. Salt, volume and the prevention of hypertension. Circulation 1976;53:589-595 23. Vestal RE, Wood AJJ, Shand DG. Reduced /3-adrenoceptor sensitivity in the elderly. Clin Pharmacol 1979;26:181-186 24. Bertel O, BQhler FR, Kiowski W, Lutold BE. Decreased betaadrenoreceptor responsiveness as related to age, blood pres- 25. 26. 27. 28. 29. 30. 31. 32. 1049 sure, and plasma catecholamines in patients with essential hypertension. Hypertension 1980;2:130-138 Doyle V, O'Malley K, Kelly JG. Human lymphocyte betaadrenoceptor density in relation to age and hypertension. J Cardiovasc Pharmacol 1982;4:738-74O Agabiti-Rosei E, Alicandri C, Beschi M, et al. Relationships between plasma catecholamines, renin, age and blood pressure in essential hypertension. Cardiology 1983;7O:3O8-316 Cutler NR, Hodes JE. Assessing the noradrenergic system in normal aging: a review of methodology. Exp Aging Res 1983;9:123-127 Sigurdsson JA, Bengtsson C, Tibblin E, Wojciechowski J. Prevalence of secondary hypertension in a population sample of Swedish women. Eur Heart J 1983;4:424-433 Masuo K, Ogihara T, Kumahara Y, Yamatodani A, Wada H. Increased plasma norepinephrine in young patients with essential hypertension under three sodium intakes. Hypertension 1984;6:315-321 Forette F, Henry JF, Hervy MP. Hypertension in the elderly. In: Amery A, ed. Hypertensive cardiovascular disease: pathophysiology and treatment. Boston: Martinus Nijhoff, 1982: 347-364 Crane MG, Harris JJ, Johns VJ. Hyporeninemic hypertension. Am J Med 1972^2:457^166 Lindquist O. Influence of the menopause on ischaemic heart disease and its risk factors and on bone mineral content. Acta Obstet Gynecol Scand [suppl] 1982;! 10:1-36 Downloaded from http://ahajournals.org by on December 8, 2021