Postural hypotension and diuretic therapy in the elderly MARTIN G. MYERS,* MD, FRCP[C]; PATRICIA M. KEARNS, M SC N. DAVID S. KENNEDY, B 5C RORY H. FISHER, MB, FRCP[C]

Blood pressures were recorded in 319 ambulatory subjects, largely men, aged 50 to 99 years. The mean systolic pressures were maximal in the seventh and eighth decades (136.0 and 132.1 mm Hg with the subjects supine and erect, respectively), whereas the mean diastolic pressures fell progressively after age 69. The distribution of postural changes in mean blood pressure was similar in each decade; a decrease of 20 mm Hg or more was noted in 3A0/o of the subjects aged 80 to 99 years and in 4.10/0 of those aged 50 to 79 years. The frequency of postural hypotension was 4.60/o in subjects treated with diuretics and 3A0/o in those not so treated. Blood pressures and the frequency of postural hypotension did not progressively increase with age in this elderly population. Les tensions arterielles ont et6 enregis. trees chez 319 sujets ambulatoires, en majorite des hommes. &g6s de 50 a 99 ans. Les tensions systoliques moyennes ont 6t6 maximales au cours des septieme et huitieme decennies (136.0 et 132.1 mm Hg, respectivement, pour les sujets en positions couchee et debout) alors que les tensions diastoliques s'abaissaient progressivement apres l'Age de 69 ans. La distribution des changements de tension arterielle avec les changements de posture .tait Ia mAine pour chaque d6cade: une diminution de 20 mm Hg ou plus a ete not6e chez 3A0/o des sujets Ages de 80 a 99 ans et chez 4.10/o de ceux qui etaient Ages de 50 a 79 ans. La frAquence de l'hypotension orthostatique etait de 4.60/o chez les sujets recevant des diuretiques et de 3.40/o chez ceux qui ne recevaient pas un tel traitement. Les tensions arterielles et Ia frequence de l'hypotension orthostatique n'augmentaient pas progressivement avec l'ige parmi cette population ig6e.

this condition may be partly responsible for the frequent falls, episodic dizziness, unsteadiness and confusion often seen in older individuals.1-3 One report noted an increased frequency of postural hypotension in persons who had had a stroke,4 while other investigators speculated about inappropriate antihypertensive therapy in the elderly resulting in transient neurologic upsets or syncope.3 In view of the possible relation between neurologic disturbances and postural hypotension, we thought it would be of interest to know what proportion of the elderly population experiences excessive postural falls in blood pressure. The study described below was undertaken to determine the possible effects of increasing age on the frequency of postural hypotension in those aged between 50 and 99 years. Since diuretics may be among the drugs most commonly prescribed for the elderly,6 special attention was directed towards the possible influence of these compounds on the blood pressures of this population. Methods

Screening for postural hypotension was conducted among 319 residents of a veterans' geriatric unit and domiciliary institution. All but 16 of the subjects were male and the mean age of the group was 75.5 years. Each of the participants was able to walk without assistance and could stand unsupported. Blood pressures were recorded after the subject had been recumbent for a minimum of 5 minutes and erect for 2 minutes. All readings at this time were taken by the same examiner (D.S.K.), who used a Postural hypotension has been im- Hawksley random-zero sphygmomaplicated as a possible cause of neuro- nometer,7 an instrument designed to logic disturbances in the elderly. minimize observer bias. Blood presVarious reports have suggested that sure was recorded in duplicate, and the average of two readings taken several minutes apart was used to From the department of medicine, obtain the final measurement. The Sunnybrook Medical Centre, Toronto diastolic pressure was taken as the Reprint requests to: Dr. Martin 0. point at which the Korotkoff sounds Myers, Sunnybrook Medical Centre, disappeared. Postural hypotension 2075 Bayview Ave., Toronto, Ont. was evaluated by examining the M4N 3M5

changes in mean blood pressure (diastolic plus one third of the difference between the systolic and diastolic pressures) between the supine and the erect positions. The medical records for each individual were examined, and a note was made of any medications, particularly antihypertensive agents and diuretics, being administered at the time of blood pressure screening. The significance of changes in mean blood pressure was assessed with the use of Student's paired t-test. Results Blood pressures in each decade The blood pressures of the 319 elderly individuals when supine, grouped by decade of age (between 50 and 99 years), are shown in Fig. 1. The mean systolic blood pressure increased between the sixth (123.7 mm Hg) and the seventh (136.0 mm Hg) decades and remained relatively constant thereafter. In contrast, the mean diastolic pressure fell progressively after age 69, decreasing to 64.4 mm Hg in the 13 subjects in their 90s. With the subjects erect the mean systolic pressure reached a plateau in the eighth decade (132.1 mm Hg) and the mean diastolic pressure gradually decreased until age 90, when the mean was 62.3 mm Hg (Fig. 2). Postural changes in blood pressure The postural changes in mean blood pressure within each decade are shown in Fig. 3. With the exception of the 10th decade, in which the number of subjects was relatively small, there was no appreciable difference in the postural changes in mean blood pressure. Of the 319 individuals 191 exhibited an increase in mean blood pressure on arising. In only 12 of the remaining 128 subjects did the mean blood pressurc fall more than 20 mm Hg. The individuals with postural hypotension were evenly distributed in the decades between 50 and 99 years. Of the 319 subjects 34 (10.7%) had a postural fall in systolic blood

CMA JOURNAL/SEPTEMBER 23, 1978/VOL. 119

581

140

-IT

56

z

1201-

E w

iooF

a.

801-

a 2

140156

z E

E w

Co C,,

w 0.

a 0 0 .8

0

-i

-

i

-

I.

pressure greater than 20 mm Hg. The distribution of postural changes in systolic and mean blood pressures were similar in each decade. Diuretic therapy and blood pressure The percentage of subjects receiving diuretic therapy increased progressively between the sixth and the ninth decades (Table I). However, there was no corresponding increase in the percentage with postural hypotension: a postural decrease in blood pressure greater than 20 mm Hg was noted in 7 (4.1%) of the 170 subjects between the ages of 50 and 79 years, compared with 5 (3.4%) of the 149 subjects between the ages of 80 and 99 years (Fig. 3). The postural changes in mean blood pressure in the 87 subjects treated with diuretics were similar to those in the 232 subjects not so treated. Four (4.6%) of the 87 individuals receiving diuretic therapy showed a decrease in mean blood pressure greater than 20 mm Hg. Similarly, 8 (3.4%) of the 232 subjects not receiving diuretic therapy exhibited postural hypotension (Fig. 4). Of the 87 subjects treated with diuretics 42 were receiving furosemide (mean daily dose 36.8 mg), 23 hydrochlorothiazide (35.8 mg), 4 spironolactone (50.0 mg) and 19 spironolactone (25 mg) and hydrochiorothiazide (25 mg). Three patients were receiving more than one diuretic. Three patients in the group not treated with diuretics were receiving methyldopa (500 mg or less daily), and one patient taking hydrochlorothiazide was also receiving minoxidil and propranolol. Effect of institutionalization on blood pressure To examine the possible effects of institutionalization on the absolute blood pressure measurements in this population, serial readings were taken

in 22 subjects during the first 2 weeks after admission. None of these individuals received hypotensive medication during this period. There was a significant decrease (P < 0.01) in systolic blood pressure (mean ± standard error of the mean), from 133.6 ± 4.3 mm Hg on day 1 to 126.7 ± 3.8 mm Hg on day 7 and 123.7 ± 5.0 mm Hg on day 14 (Fig. 5). The diastolic pressure fell slightly during this period but the difference between the initial level (78.0 ± 3.2 mm Hg) and the level at 24 days (73.2 ± 2.8 mm Hg) was not significant. The decrease in blood pressure at-

50 F

40

tributed to institutionalization may have occurred, at least in part, as a result of accommodation to having the pressure taken by a new observer with new apparatus. To assess the importance of such accommodation, serial blood pressure measurements were performed for 14 days in 20 subjects who had resided in the institution for at least 2 months. The mean systolic pressure rose slightly from day 1 (135.0 ± 6.7 mm Hg) to day 3 (141.1 ± 7.8 mm Hg) and thereafter fell to 128.9 ± 6.8 and 127.8 ± 8.3 mm Hg on days 7 and 14, respectively. The mean diastolic pressure showed

SUBJECTS NOT ON DIURETICS (n=232) DIURETIC-TREATED SUBJECTS (N=87)

30

nT]

20 F

10 F

0I.III

-40 -30 -20 -10 0

+10 +20 +30 -40 -30 -20 -10 0 +10 +20 +30 +40

POSTURAL CHANGE IN MEAN B P (mm Hg)

iges in mean blood pressure of sul

smaller changes, rising from day 1 (75.8 ± 3.6 mm Hg) to day 3 (76.6 -+- 3.9 mm Hg), then falling to 73.5 -.- 3.2 and 7.40 ± 4.3 mm Hg on days 7 and 14 respectively. None of these changes were significant. Discussion

There is no doubt that postural hypotension may be a serious problem in some elderly individuals. However, the above findings suggest that the frequency of this condition is somewhat less than has been suspected. If one considers mean blood pressure as an indicator of perfusion pressure in the systemic circulation, then it appears that about 4% of persons over the age of 50 years will exhibit clinically important postural hypotension (mean decrease in pressure greater than 20 mm Hg). In previous studies postural hypotension has been found in 17% and 24% of the population more than 65 to 70 years of age.8'9 In these studies postural changes in blood pressure were defined in terms of systolic blood pressures. From our findings it is apparent that this results in an overestimation of the frequency of severe postural decreases in blood pressure. Among the individuals in our study group with a decrease in systolic pressure greater than 20 mm Hg (10.7% of the group), many showed little change in the diastolic readings, with the mean pressure while they were erect unchanged from that while they were supine. This suggests that the mean pressure may be the best noninvasive measurement of perfusion pressure when one is evaluating the presence of postural hypotension. To compare our results with those of earlier series,8'9 we included only ambulatory subjects who were able to stand unsupported. Eighty-nine less mobile individuals whose blood pressures were measured while they were supine and sitting exhibited postural changes similar to those recorded in the more ambulatory group. Thus, the exclusion of persons unable to stand unsupported did not appear to alter the observed frequency of postural hypotension in our elderly group. Our study also found similar changes in blood pressure in each of the decades between 50 and 99 years. Regardless of whether one considers changes in systolic pressure

or in mean pressure, there was no progressive increase in the frequency of postural hypotension with age. This suggests that the decrease in baroreceptor sensitivity with . may not be clinically significant in the ambulatory elderly population. Cardiovascular reflexes may be impaired in older persons, but not enough to produce postural hypotension in a high proportion of them. Since there is no standard time after which one records blood pressure with the subject erect to assess postural hypotension, it is difficult to make exact comparisons of results between studies. Most authors have measured the blood pressure in erect individuals after 1 to 2 minutes of standing.7'8 It is not unreasonable to expect an older individual to rest at the side of the bed for this short interval before arising. Hence, the practice of measuring postural changes in blood pressure after 2 minutes may be more applicable to the clinical setting and may better reflect the chances of encountering symptomatic postural hypotension than data derived from studies of baroreceptor function. Diuretic therapy does not seem to affect the frequency of postural hypotension in the elderly. The 87 individuals who were treated with various diuretic agents because of hypertension, congestive heart failure and other unknown disorders did not show an increased tendency to become hypotensive while standing. Also, although the percentage of persons receiving diuretic therapy increased progressively with each decade, the frequency of postural hypotension remained relatively unchanged between the ages of 50 and 99 years. The percentage of subjects in our study treated with diuretics is similar to that reported by others,8 and suggests that the participants in our study were not overtreated when compared with older residents in the community. Nevertheless, in some individuals currently not receiving diuretic therapy symptomatic postural hypotension may have developed in the past when they were treated with these agents. It is generally believed that antihypertensive agents are more likely to induce postural hypotension in the elderly than in younger people.12 This may be true of the sympatholytic drugs, but it appears less likely with diuretics when they are used alone,

as in our subjects. Judicious use of diuretics may therefore lower blood pressure in older persons without invariably producing postural effects. A number of reports, including that of the Framingham study, have indicated that blood pressure tends to rise with age during the early and middle years of life.'3 However, it is not clear whether this relation also exists after the age of 60 years. The general belief that hypertension is more common in the elderly is not wholly supported by the results of well designed population surveys. Cross-sectional data from the Framingham study for persons between the ages of 36 and 74 years have suggested that in men the systolic blood does not rise beyond the seventh decade and the diastolic pressure falls precipitously after age 56." Similar findings were made in a large population survey of elderly persons in a California retirement community;'4 among the male participants the levelling off of systolic blood pressure occurred in the 70s. Since most of our subjects were also male, we were not surprised to find similar tendencies in our study group. The absolute mean blood pressure readings in our group were lower than those reported from the Framingham study." The differences are small, however, and may, at least in part, be accounted for by the effects of institutionalization or possibly of diuretic therapy. The importance of diuretic therapy is difficult to assess since the reports of the Framingham study and the California survey did not indicate the proportion of subjects receiving diuretics. This information would also be helpful in determining the influence of diuretic therapy on the tendency of blood pressure to rise with age. In our series the proportion of individuals receiving diuretic therapy was highest in the upper decades, a factor that could be responsible for the mean pressure readings' being lower than expected. That the blood pressure has not been noted to rise progressively after age 60 years in cross-sectional population surveys could be the result of a "survivor effect". Persons in whom a condition such as hypertension develops in the early or middle years of life may suffer a fatal complication before the seventh decade. Thus, any elderly population studied at one time may be composed predominant-

584 CMA JOURNAL/SEPTEMBER 23, 1978/VOL. 119

ly of survivors. This survivor effect is a potential source of error in most geriatric research; longitudinal studies over many years are necessary to eliminate this factor. However, the cross-sectional approach does provide considerable information on the status of the survivors who make up the current geriatric population. Since many of us treat these individuals, it is important that we understand various aspects of their physical state when we actually see them (i.e., in old age) as well as appreciate what has happened to their less fortunate cohorts. In conclusion, it appears that blood pressure and the frequency of postural hypotension do not progressively' increase in the upper decades of life. Furthermore, diuretic therapy does not necessarily result in postural hypotension in the geriatric age group, and the judicious use of these agents would be unlikely to result in serious neurologic complications. It seems preferable to use diuretics as the initial antihypertensive agents in the elderly whenever possible, although small amounts of several drugs may be necessary to treat nonresponders or those with diuretic-induced side effects. We thank Mrs. G. McMillan and Mrs. K. Adamson for technical and secretarial assistance. Dr. Myers is the recipient of a senior research fellowship from the Ontario Heart Foundation. References 1. Postural hypotension in the elderly (E). Br Med J 4: 246, 1973 2. CAIRD Fl, DALL JLC: The significance of heart disease in old age, in Textbook of Geriatric Medicine and Gerontology,

BROCKLEHURST

JC

(ed),

Churchill Livingstone, London, 1973, p 137 3. WOLLNER L, SPALDING JMK: The autonomic nervous system, ibid, p 246 4. Gi.oss M: The effect of posture on subjects with cerebrovascular disease. Q J Med 39: 485, 1970 5. JACKSON G, PIERsCIANowsKI JA, MAHON W, et al: Inappropriate antihypertensive therapy in the elderly. Lancet 2: 1317, 1976

6. LAW R, CHALMERS C: Medicine and elderly people: a general practice survey. Br Med J 1: 565, 1976 7. WRIGHT BM, DORE CF: A randomzero sphygmomanometer. Lancet 1: 337, 1970 8. JOHNSON RH, SMITH AC, SPALDING JMK, et al: Effect of posture on blood-pressure in elderly patients. Lan-

cet 1: 731, 1965

9. CAIRD Fl, ANDREWS GR, KENNEDY RD: Effect of posture on blood pressure in the elderly. Br Heart J 35: 527, 1973 10. GRIBBIN B, PICKERING TG, SLEIGHT P, et al: Effect of age and high blood pressure on baroreflex sensitivity in man. Circ Res 29: 424, 1971 11. DUKE PC, WADE JG, HICKEY RF, et al: The effect of age on baroreceptor reflex function in man. Can

Anaesth Soc J 23: 111, 1976 12. Hypertension in the elderly (E). Lancet 1: 684, 1977 13. KANNEL WB: Blood pressure and the development of cardiovascular disease

in the aged, in Cardiology in Old Age, CAIRn Fl, DALL JLC, KENNEDY RD (eds), Plenum Pub, New York, 1976, p 143 14. COLANDREA MA, FRIEDMAN GD, NICHAMAN MZ, et al: Systolic hypertension in the elderly: an epidemiologic assessment. Circulation 41: 239, 1970

BOOKS continued from page 578 HUMAN BLOOD GROUPS. Proceedings of the Fifth International Convocation on Immunology. Buffalo, N.Y., June 7-10, 1976. Edited by J.F. Mohn, R.W. Plunkett, R.K. Cunningham and others. 462 pp. IIlust. S. Karger AG, Basel, 1977. $35.75. ISBN 3-8055-2422-6 INTENSIVE CARE RADIOLOGY. Imaging of the Critically III. Edited by Lawrence R. Goodman and Charles E. Putman. 363 pp. Illust. The C.V. Mosby Company, Saint Louis, 1978. $39.25. ISBN 0-80161894-0 INTERNATIONAL COLLABORATION: Problems and Opportunities. Proceedings of a Symposium Held in Toronto on the Occasion of the Designation by the World Health Organization of the Addiction Research Foundation as a Collaborating Centre for Research and Training on Drug Dependence. Edited by Barbara Rutledge and E. Kaye Fulton. 206 pp. Illust. Addiction Research Foundation, Toronto, 1977. $14.95, paperbound MAN AND MOVEMENT: Principles of Physical Education. 2nd ed. Harold M. Barrow. 396 pp. Lea & Febiger, Philadelphia; the Macmillan Company of Canada Limited, Toronto, 1977. $17.95. ISBN 08121-0599-0 MANUAL OF CLINICAL PROBLEMS IN INTERNAL MEDICINE. Annotated with Key References. 2nd ed. Jerry L. Spivak and H. Verdain Barnes. 513 pp. Little, Brown and Company (Inc.), Boston, 1978. $10.95, spiralbound. ISBN 0-316-80714-1 MANUAL OF THE INTERNATIONAL STATISTICAL CLASSIFICATION OF DISEASES, INJURIES, AND CAUSES OF DEATH. Vol. 2. Alphabetical Index. Based on the Recommendations of the Ninth Revision Conference, 1975 and Adopted by the Twenty-ninth World Health Assembly. World Health Organization. 659 pp. World Health Organization, Geneva, 1978. $15. ISBN 92-4-154005-2

continued on page 598

Brief Prescribing Information

Caution should be observed in prescribing Anafranil in hyperthyroid patients or in patients receiving thyroid medication conjointly. Transient cardiac arrhythmias have occurred in rare instances in patients who have been Antidepressant receiving other tricyclic compounds conIndications and Clinical uses comitantly with thyroid medication. Anafranil (clomipramine hydrochloride) is Obstructive jaundice and bone marrow deindicated in the drug treatment of depressive pression with agranulocytosis have been reported. Periodic blood cell counts and liver illness, including manic depressive psychosis, function tests are recommended in patients depressed phase, and involutional melancholia receiving treatment with Anafranil over Anafranil appears to have a mild sedative prolonged periods. effect which may be helpful in alleviating the anxiety component often accompanying Adverse Reactions depression. Anafranil also appears to be of The following adverse reactions have been some value as an adjunct in the management reported with Anafranil or other tricyclic of manifestations of agitated depression which antidepressants: sometimes exacerbate obsessive compulsive Central Nervous System Effects: neurosis. drowsiness, fatigue, insomnia, extra-pyramidal Contraindications effects such as tremor and ataxia, headache, Anafranil should not be given in conjunction anorexia and convulsions. Peripheral neurowith or within fourteen days of treatment with a pathy has also been reported with tricyclic monoamine oxidase inhibitor. Combined therapy compounds. of this type could lead to the appearance of Behavioural Effects: serious hypertensive crises and death may occur. agitation, excitement, hypomania or manic episodes, activation of psychosis, confusion, Anafranil is contraindicated in patients with existing liver damage and should not be disturbed concentration, visual hallucinations. Autonomic Nervous System Effects: administered to patients with a history of blood dry mouth, blurred vision, difficulty with accomdyscrasias. modation, constipation, paralytic ileus, Anafranil is contraindicated in patients who disturbances of micturition, excessive sweating, have shown hypersensitivity to the drug. nausea and vomiting. Anafranil is contraindicated in patients with Cardiovascular Effects: glaucoma, as the condition may be aggravated hypotension, particularly orthostatic hypodue to the atropine-like effect of the drug. tension with associated vertigo, tachycardia, Use in Pregnancy: syncope, arrhythmia, asystole, EKG changes The safety of use in pregnant women has not (including flattening or inversion of Twave) been established. Therefore, Anafranil should and disturbances in cardiac conduction. not be administered to women of childbearing Haematological and Other Toxic Effects: potential, particularly during the first trimester of agranulocytosis has been reported; it reprepregnancy, unless, in the opinion of the physisents hypersensitivity a reaction. Eosinophilia cian, the expected benefit to the patient may also occur. Obstructive jaundice, allergic outweighs the potential risk to the fetus. skin reactions, photosensitization, occasional Warnings disturbances of appetite, abdominal pain, The following warnings apply to Anafranil and changes in libido, and weight gain. other tricyclic antidepressant agents: Dosage and Administration Tricyclic agents may lower the convulsive Except in Elderly Patients and Adolescents: threshold and should, therefore, be used with 25mg 3 times daily initially, increase up to caution in patients with convulsive disorders. 150mg daily, or more, as required. Electrocardiographic studies suggest that Anafranil should not be used in the presence of Dosage in excess of 200mg daily is not usually recommended for office patients. Occasionally pronounced cardiac or circulatory failure, recent in more severe hospitalized patients, dosages myocardial infarction or ischaemic heart disease. Anafranil also has a hypotensive up to 300 mg may be required. In Elderly Patients and Adolescents: action which may be detrimental in these cir20 to 30mg daily, increased by 10mg daily, if cumstances. The drug should, therefore, be used with caution in patients who are suscepnecessary, depending on tolerance and response. tible to hypotensive episodes. Tricyclic agents may produce urinary retention Availability and should be used with caution in patients Each pale yellow, sugar-coated, lenticular with urinary pathology, particularly in the tablet with Geigy imprinted, contains 25mg presence of prostatic hypertrophy. clomipramine hydrochloride. Particularly in the elderly and in hospitalized Also available in pale yellow, triangular patients the tricyclic antidepressants may give sugar-coated tablets imprinted Geigy, containrise to paralytic ileus and therefore appropriate ing 10mg clomipramine hydrochloride. measures should be taken if constipation occurs. In bottles of 50 and 500. Anafranil should be kept in a safe place, well Product monograph supplied on request. out of the reach of children. References: 1. Rompel, H.: The Treatment of Depression, Med. Proc. 13, 631, (1967) Precautions 2. Clarke, F.C.: The Treatment of Depression in General Practice. S. At r. Med. In seriously depressed patients the possibility of J. 43, 23, (1969) suicide should be bome in mind and may 3. Lasich, A.J.: Clinical Evaluation of a NewAnti-Depressant(Anafranil), Med. Proc. 14, 312(1968) persist until significant remission occurs. Therefore, these patients should be carefully supervised during treatment with Anafranil, and hospitalization or concomitant electroconvulsive therapy may be required. Activation of latent schizophrenia or aggravation of existing psychotic manifestations in schizophrenic patients may occur; patients with see outside back cover manic-depressive tendencies may experience hypomanic or manic shifts; and hyperactive or agitated patients may become over-stimulated. A reduction in dose or discontinuation of Anafranil should be considered under these circumstances. Since Anafranil may produce sedation, particularly during the initial phase of therapy, patients should be cautioned about the danger of engaging in activities requiring mental alertness, judgement and physical coordination. It should be home in mind that Anafranil may block the pharmacological effects of hypoPAAB ccPp tensive drugs, such as guanethidine and similar agents. Dorval, Quo. H9S iBi G-7004-R-1

1&nafranil

Geigy

CMA JOURNAL/SEPTEMBER 23, 1978/VOL. 119 585