Postural Changes in Blood Pressure and the Prevalence of Orthostatic Hypotension Among Home-Dwelling Elderly Aged 75 Years Or Older

Journal of Human Hypertension (2009) 23, 33–39 & 2009 Macmillan Publishers Limited All rights reserved 0950-9240/09 $32.00 www.nature.com/jhh ORIGINAL ARTICLE Postural changes in blood pressure and the prevalence of orthostatic hypotension among home-dwelling elderly aged 75 years or older

P Hiitola1,2, H Enlund3, R Kettunen4, R Sulkava5,6 and S Hartikainen7,8,9 1Department of Social Pharmacy, University of Kuopio, Kuopio, Finland; 2Kuopio Research Centre of Geriatric Care, Kuopio, Finland; 3Department of Pharmacy Practice, Kuwait University, Kuwait; 4Department of Medicine, Paijat-Hame Central Hospital, Lahti, Finland; 5Division of Geriatrics, School of Public Health and Clinical Nutrition, University of Kuopio, Finland, Finland; 6Rheumatism Foundation Hospital, Heinola, Finland; 7Faculty of Pharmacy, University of Kuopio, Kuopio, Finland; 8Kuopio Research Centre of Geriatric Care, Kuopio, Finland and 9Leppa¨virta Health Centre, Leppa¨virta, Finland

This cross-sectional analysis of a population-based half of the home-dwelling elderly when they stood cohort investigates the postural changes in blood up from a supine to a standing position. The pressure (BP) and heart rate and assesses the pre- total prevalence of OH was 34% (n ¼ 220). No significant valence of orthostatic hypotension (OH) and its associa- gender or age differences were seen. The prevalence tions with the medicines used by an elderly population. of OH was related to the total number of medicines The study population (n ¼ 1000) was a random sample of in regular use (Po0.05). OH and postural changes persons aged 75 years or older in the City of Kuopio, in BP are more common among the home-dwelling Finland. In 2004, altogether, 781 persons participated in elderly than reported in previous studies. The preva- the study. After the exclusion of persons living in lence of OH is related to the number of medicines institutional care (n ¼ 82) and those without orthostatic in regular use. There is an obvious need to measure test (n ¼ 46), the final study population comprised 653 orthostatic BP of elderly persons, as low BP and OH are home-dwelling elderly persons. OH was defined as a important risk factors especially among the frail elderly X20 mm Hg drop of systolic BP or a X10 mm Hg drop persons. of diastolic BP or both 1 or 3 min after standing up Journal of Human Hypertension (2009) 23, 33–39; from supine position. Systolic BP dropped for more than doi:10.1038/jhh.2008.81; published online 24 July 2008

Keywords: blood pressure; orthostatic hypotension; pulse pressure; elderly; population-based study

Introduction (BP) drops lower than those fulfilling the OH criteria.2,16 Orthostatic hypotension (OH) is a major health Orthostatic hypotension can be caused by several problem in the elderly. It affects 6–30% of home- 9 1–5 medicines, such as antihypertensives, antidepres- dwelling elderly persons. OH is even more 9,17 9,14,18 6–11 sants, a-adrenergic blocking agents and prevalent in residential care. Postural hypoten- medication for Parkinson’s disease,6 haemodynamic sion and OH are risk factors for dizziness, syncope conditions such as hypovolaemia and cardiac and falls, which can lead to functional impair- 1,7,10,12,13 insufficiency, and neurogenic causes such as multi- ment. Hospitalization, prolonged bed rest ple system atrophy and, for example, diabetic and diseases such as Parkinson’s disease are also 19 6,14,15 neuropathy. associated with OH. OH is a risk factor for As far as we know, there are no wide population- stroke, and it is associated with cardiovascular 2,3,16 based studies concerning the changes in blood and and all-cause mortality. An association with pulse pressures following postural change and the mortality has been reported even for blood pressure prevalence of OH and its associations with medication in elderly populations. Therefore, the aim of this study was to examine the changes in blood and Correspondence: P Hiitola, Department of Social Pharmacy, pulse pressures following postural change to deter- University of Kuopio, Box 1627, Kuopio 70211, Finland. E-mail: [email protected] mine the prevalence of OH and to find its associa- Received 22 February 2008; revised 15 June 2008; accepted 18 tions with the use of medicines in home-dwelling June 2008; published online 24 July 2008 elderly persons aged 75 years or older. Postural changes in blood pressure and orthostatic hypotension P Hiitola et al 34 Materials and methods (DBP) decrease of at least 10 mm Hg (diastolic OH) within 3 min of standing up.22 The BP recordings Study subjects and setting were made after 10 min rest. BP was measured in This study is part of the population-based GeMS lying, sitting and standing positions (at 1 and 3 min) study (Geriatric Multidisciplinary Strategy for the by a trained nurse using a calibrated mercury Good Care of the Elderly), a multidisciplinary health column sphygmomanometer or a calibrated air- intervention survey focusing on the clinical epide- pressure sphygmomanometer. Pulse pressure was miology of diseases, functional capacity, medication calculated as the difference between SBP and DBP. and use of services in a population of elderly The largest changes from baseline in SBP and DBP persons aged 75 years or older. The target popula- were also examined. The clinical criteria for de- tion comprised all the inhabitants of the City of mentia were those of Diagnostic and Statistical Kuopio in Eastern Finland, aged 75 years or older on Manual of Mental Disorders-IV.23 1 November 2003. From this population, a random sample of 1000 persons was drawn. In this cohort, 55 died before the examination, Statistical analysis two moved away and 162 persons refused to The data management and analysis were performed take part in the survey. The remaining 781 partici- using SPSS 14.0 for Windows (Statistical Package pants attended a structured clinical examination for Social Sciences 14.0). Chi-square test was used and an interview conducted by a trained nurse. for analysing the statistical significance of the Physiotherapists tested their functional capacity, differences in categorical variables. Non-parametric strength and balance. Mann–Whitney U-test was used in analysing the Of all the examined elderly persons (n ¼ 781), we statistical significance of the differences in contin- excluded those in institutional care (n ¼ 82) and the uous variables like age and the number of medicines home-dwelling elderly without an orthostatic test used. The T-test was used for analysing the (n ¼ 46). This study is based on information from the differences in pulse pressures between the OH 653 home-dwelling elderly with an orthostatic groups. test. Of the participants, 70% (n ¼ 454) were females and 30% (n ¼ 199) were males. The mean age of our participants in 2004 was 81 years (range of 75–99 years). Ethical issues Written informed consent was obtained from the study participants or their relatives. The study was approved by the Research Ethics Committee of Data collection Northern Savo Hospital District and Kuopio Uni- The basic demographic and clinical data were versity Hospital. collected by means of interviews. Trained nurses interviewed the participants about their use of medicines and recorded the medicines they were Results currently taking. The participants were also asked to bring their prescription forms and medicine con- There were no statistically significant differences in tainers with them for the interview. If the person basic demographic characteristics, body mass index, him/herself could not answer the questions, a diseases or BPs between participants having OH and relative or a caregiver gave the required information. those not having OH (Table 1). If the participant was unable to visit the clinic, a trained nurse made a home visit. Medical records from the municipal health centre, home nursing Postural changes in BP among all participants service, local hospitals and Kuopio University In the orthostatic test, SBP did not change in 13% Hospital were also available. (1 min) and 18% (3 min) of the participants. About Both regularly and irregularly taken prescribed 20% of the elderly persons had a rise and 68% and non-prescribed medicines were recorded. If a (1 min) and 60% (3 min) had a drop in SBP medicine was taken daily or at regular intervals, it (Figure 1). SBP dropped by 30 mm Hg or more after was recorded as being in regular use. If it was taken 1 min in 10% and by 10–19 mm Hg in 27% of those only when needed, it was recorded as being in tested. DBP dropped in 39% (1 min) and in 32% irregular use. The medicines were classified accord- (3 min). ing to the Anatomic Therapeutic Chemical classifi- The lowest SBP value was 95 mm Hg (1 min) and cation system, version 2004, recommended by the 100 mm Hg (3 min) and the lowest DBP value World Health Organization for drug utilization 48 mm Hg (1 min) and 50 mm Hg (3 min). The biggest studies.20,21 drop in SBP from a supine to a standing position The criteria for OH were those defined by the after 1 min of standing was 55 mm Hg and that after American Autonomic Society and the American 3 min of standing was 50 mm Hg. In DBP, the biggest Academy of Neurology: a systolic BP (SBP) decrease drop was 20 mm Hg after both 1 and 3 min of of at least 20 mm Hg (systolic OH) or a diastolic BP standing.

Journal of Human Hypertension Postural changes in blood pressure and orthostatic hypotension P Hiitola et al 35 Table 1 Characteristics of the study population according to the OH status (n ¼ 653)

OH-positive OH-negative Total P-valuea n ¼ 220 (%) n ¼ 433 (%) n ¼ 653 (%)

Age 75–79 100 (46) 228 (53) 328 (50) 0.147 80–84 69 (31) 128 (29) 197 (30) 85+ 51 (23) 77 (18) 128 (20)

Sex Male 64 (29) 135 (31) 199 (31) 0.653

BMI, mean 26.0 (95% CI 25.3–26.4) 26.7 (95% CI 26.3–27.2) 26.5 (95% CI 26.1–26.8) 0.062 Systolic blood pressure, sitting 145 (95% CI 142–149) 145 (95% CI 143–147) 145 (95% CI 144–147) 0.991 position, mean (mm Hg) Diastolic blood pressure, sitting 79 (95% CI 77–80) 80 (95% CI 79–81) 80 (95% CI 79–81) 0.103 position, mean (mm Hg) Dementia 37 (17) 52 (12) 89 (14) 0.093 Diabetes mellitus 67 (16) 32 (15) 99 (15) 0.838 Hypertonia 127 (58) 252 (58) 379 (58) 0.866 Parkinson’s disease 6 (3) 5 (1) 11 (2) 0.122

Abbreviations: BMI, body mass index; CI, confidence interval; OH, orthostatic hypotension. aFrom the w2 test (stratified variables) or Mann–Whitney U-test (continuous variables).

prevalence of a combination of systolic and diastolic OH increased significantly with age (Po0.05). The prevalence of any orthostatic reaction was similar regardless of SBP in a sitting position (Table 2). Systolic OH (3 min) was most common among the participants with SBPX160 mm Hg (20%), and its prevalence decreased in the lower BP groups (7%). The prevalence of diastolic OH (1 min) was more prevalent in the participants whose SBP was o120 mm Hg (18%) in a sitting position, and it decreased in the higher BP groups (6%).

Orthostatic hypotension and the use of medicines The association between the prevalence of OH and the number of regularly used medicines was statistically significant (Po0.05, Table 3). The more medicines there were in regular use, the more common was OH. OH was not associated with the number of irregularly used medicines or total medication. Further analysis showed that OH was Figure 1 Changes in BP after postural changes in all elderly not associated with the number of BP-lowering participants (n ¼ 653). BP, blood pressure. medicines in use or the number of drugs that traditionally have been associated with OH. Neither any particular antihypertensive nor causative med- Prevalence of orthostatic hypotension icine was associated with OH. Orthostatic hypotension was found in 34% (14% after 1 min only, 4% after 3 min only and 16% both at 1 and 3 min) of this elderly population. The Orthostatic hypotension and heart rate prevalence of only systolic OH was 23% and only In the orthostatic test, after 1 min of standing, 64% diastolic OH 4% and both systolic and diastolic OH of the participants had no change in their heart rate, 7%. OH was equally common in men and women. 27% showed an increase and 9% showed a decrease The OH criteria were fulfilled by 31% of the persons in heart rate. After 3 min of standing, the respective aged 75–79 years, 35% of those aged 80–84 years proportions were almost identical 66, 25 and 9%. and 40% of those aged 85 years or older (P ¼ 0.15, After 1 min of standing, the increases in age as a continuous variable, P ¼ 0.09). Only the heart rates between OH-positive and OH-negative

Journal of Human Hypertension Postural changes in blood pressure and orthostatic hypotension P Hiitola et al 36 Table 2 Prevalence of different types of OH and mean blood pressure drops (mm Hg) by systolic blood pressure in sitting position (n ¼ 653)

Type of OH reaction Systolic blood pressure in sitting position (mm Hg)

o120 120–139 140–159 X160 Total n ¼ 57 n ¼ 178 n ¼ 239 n ¼ 179 n ¼ 653

Any OH reaction, n (%) 18 (32) 66 (37) 75 (31) 61 (34) 220 (34) Systolic OH 1 min, n (%) 14 (25) 52 (29) 62 (26) 43 (24) 171 (26) Mean drop (range) À9.5 (15, À35) À9.5 (35, À55) À9.3 (25, À50) À8.1 (33, À50) À9.0 (35, À55) Systolic OH 3 min, n (%) 4 (7) 27 (15) 46 (19) 35 (20) 112 (17) Mean drop (range) À4.4 (20, À25) À5.9 (20, À40) À6.6 (33, À50) À6.6 (30, À45) À6.2 (33, À50) Diastolic OH 1 min, n (%) 10 (18) 23 (13) 14 (6) 11 (6) 58 (9) Mean drop (range) À2.7 (12, À15) À1.4 (36, À15) À0.5 (16, À20) 0.1 (16, À20) 0.8 (36, À20) Diastolic OH 3 min, n (%) 2 (4) 17 (10) 11 (5) 9 (5) 39 (6) Mean drop (range) À0.4 (15, À10) À0.7(18, À16) 0.2 (13, À20) 0.6 (18, À18) 0 (18, À20)

Abbreviation: OH, orthostatic hypotension.

Table 3 Use of medications and mean numbers (95% CI) of used medicines according to the presence of OH (n ¼ 653)

Medical groups (ATC code) OH-positive OH-negative Total P-valuea n ¼ 220 (%) n ¼ 433 (%) n ¼ 653 (%)

Diuretics (C03) 69 (31) 107 (25) 176 (27) 0.077 b-blocking agents (C07) 114 (52) 218 (50) 332 (51) 0.741 Calcium channel blockers (C08) 57 (26) 112 (26) 169 (26) 1.000 Agents acting on renin–angiotensin 69 (31) 159 (37) 228 (35) 0.193 system (C09) Organic nitrates (C01DA) 59 (27) 116 (27) 175 (27) 1.000 a-blockers (G04CA) 13 (6) 16 (4) 29 (4) 0.228 Drugs for Parkinson’s disease (N04B) 7 (3) 5 (1) 12 (2) 0.118 Antipsychotics (N05A) 11 (5) 23 (5) 34 (5) 1.000 Tricyclic antidepressants (N06AA) 6 (3) 3 (1) 9 (1) 0.068

Causative medicationb 1 70 (44) 123 (44) 193 (44) 2 62 (39) 104 (37) 166 (37) 3 or more medicines 29 (18) 56 (20) 85 (19) Mean (95% CI) 1.31 (1.2–1.5) 1.17 (1.1–1.3) 1.22 (1.1–1.3) 0.106

Antihypertensivesc 1 71 (32) 143 (33) 214 (33) 2 62 (28) 121 (28) 183 (28) 3 or more medicines 36 (17) 70 (16) 106 (16) Mean (95% CI) 1.43 (1.3–1.6) 1.39 (1.3–1.5) 1.41 (1.3–1.5) 0.900

Regularly used medicines Mean (95% CI) 5.03 (4.6–5.4) 4.59 (4.3–4.9) 4.74 (4.5–5.0) 0.049

Irregularly used medication Mean (95%) 1.27 (1.1–1.5) 1.27 (1.2–1.4) 1.27 (1.2–1.4) 0.624

Total medication Mean (95%) 6.30 (5.8–6.7) 5.86 (5.5–6.2) 6.01 (5.7–6.3) 0.074

Abbreviations: ATC, Anatomic Therapeutic Chemical classification system; CI, confidence interval; OH, orthostatic hypotension. aFrom the w2 test (stratified variables) or Mann–Whitney U-test (continuous variables). bCausative medication (regular use): antihypertensive, organic nitrates, a-blockers, drugs for Parkinson’s disease, antipsychotics and tricyclic antidepressants. cAntihypertensives: diuretic, b-blocking agents, calcium channel blockers and agents acting on renin–angiotensin system.

participants were not statistically significant (31 vs Pulse pressure was significantly higher in the 24%, P ¼ 0.121). The increase in heart rate was o10 OH-positive participants than in the OH-negative beats per minute in 71% of OH-positive and in participants in a supine position (Po0.001), after 80% of OH-negative participants. In the rest of the 1 min of standing (Po0.001) and after 3 min of participants, the increase in heart rate varied standing (P ¼ 0.011). The decrease in pulse pressure between 10 and 20 beats per minute. Similar results was more pronounced in the OH-positive than in the were obtained at 3 min. OH-negative persons after rising from a supine to a

Journal of Human Hypertension Postural changes in blood pressure and orthostatic hypotension P Hiitola et al 37 Table 4 Mean pulse pressures in OH-positive and OH-negative participants and changes in pulse pressure during the upright postural changes (mm Hg)

OH-positive (n ¼ 218) OH-negative (n ¼ 433) P-valuea

Supine position 77 68 o0.001 Sitting position 67 65 0.339 Standing position 1 min 59 65 o0.001 Standing position 3 min 63 66 0.011 Decrease in pulse pressure from supine 18 3 o0.001 to standing position after 1 min Decrease in pulse pressure from supine 14 2 o0.001 to standing position after 3 min

Abbreviation: OH, orthostatic hypotension. aFrom a T-test. standing position after 1 (Po0.001) and 3 min of cular complications and mortality.25 However, the standing (Po0.001) (Table 4). mean pressure is not the only factor resulting in a high cardiovascular risk in old patients. Another important finding was that OH in elderly Discussion persons is independent of BP in sitting position and that diastolic OH after 1 min of standing was found Systolic BP dropped in two-thirds of the home- to be more prevalent in the persons with low BP in a dwelling elderly, and every third had OH. This sitting position. This finding is notable because figure was higher than that in many previous diastolic OH after 1 min predicted cardiovascular studies.1,2,5 mortality in old persons.3 In addition, a diastolic BP Orthostatic hypotension and low BP have been drop, even when it is small enough not to fulfil the found to be associated with dizziness and an criteria of OH, after 1 min of standing up identifies increased risk of falling.1,10,13 Moreover, only scant the elderly persons at a high risk for myocardial attention has been paid to a drop of BP too small to infarction.16 This might be due to the load the heart fulfill the criteria of orthostatic reaction. It might be is exposed to upon rising up, and it may provoke that, in frail elderly persons with low BP, even a coronary insufficiency and a decrease of stroke small drop in BP can be a risk factor for falling. The volume in frail elderly person. haemodynamic response to an orthostatic reaction Previous studies have reported an increase in the involves many organs. Orthostatic challenges are prevalence of OH with advancing age.1,2 This trend responsible for blood redistribution in the lower was not very obvious in our study population aged parts of the body and for decreases in BP and cardiac 75 years or older. Only the prevalence of a output.24 In response to that, carotid and aortic wall combination of systolic and diastolic orthostatic receptors activate the baroreflex, which reduces reactions increased significantly with advancing parasympathetic and increases sympathetic activity. age. However, our population was older (mean age These autonomic changes cause vasoconstriction, 81 years) than many of those previously reported, tachycardia and positive inotropy and lead to partial and they had multiple comorbidities and medica- restoration of arterial pressure. In the present study, tions. There was no difference in the prevalence of OH-positive participants showed an increase of OH between men and women, which is consistent heart rate slightly more often than the OH-negative with the previous studies. persons, whereas a minority of OH-negative persons We found that a high number of regularly used maintained their BP by increasing the heart rate. medicines are associated with OH. Poon and Braun9 This might indicate that reflex tachycardia is found a relationship between OH and potentially secondary in preventing an orthostatic reaction even causative medication. In our study, we did not find in elderly persons. In addition, the heart rate such an association. However, in our study popula- changes in both the groups were minor, less than tion, the use of medicines was high; the range of 10 beats per minute, which might suggest baroreflex medicines in the regular use was from 0 to 19 and impairment.15 the range of total medicines was from 0 to 23. In old In the present study, pulse pressure in a supine individuals, with numerous medicines in use, it is position was higher in OH-positive than in OH- often impossible to determine whether OH is caused negative subjects. This can be explained by the fact by certain medicines or the underlying diseases that an increased resting pulse pressure indicates or both. stiffness of the major arteries. Still, after rising to a To get a proper diagnosis of OH, the methods and standing position, the decrease of pulse pressure protocol of measuring BP should be correct. In was also more pronounced in OH-positive than in clinical practice, the skills and knowledge of nurses OH-negative persons. The increased pulse pressure are often inadequate for diagnosing OH in elderly has been associated with a risk of major cardiovas- patients.26 In our study, the study nurses were

Journal of Human Hypertension Postural changes in blood pressure and orthostatic hypotension P Hiitola et al 38 instructed and trained to measure BPs accurately. Cardiovascular Health Study. For the CHS collabora- Our study had the limitation that OH testing was tive research group. Hypertension 1992; 19: 508–519. performed only once. Furthermore, the time of the 2 Masaki KH, Schatz IJ, Burchfiel CM, Sharp DS, Chiu D, day for measuring BP, meals and daily activities Foley D et al. Orthostatic hypotension predicts differed between persons, which might have influ- mortality in elderly men: The Honolulu Heart Program. Circulation 1998; 98: 2290–2295. enced our results. The reliability of the collection of 3 Luukinen H, Koski K, Laippala P, Kivela SL. Prognosis medication data was improved by asking the of diastolic and systolic orthostatic hypotension in participants to bring their medicine containers and older persons. Arch Intern Med 1999; 159: 273–280. prescription forms with them for the interview, and 4Ra¨iha¨ I, Luutonen S, Piha J, Seppa¨nen A, Toikka T, medication information was also checked from the Sourander L. Prevalence, predisposing factors, and medical records and caregivers or family members, prognostic importance of postural hypotension. Arch if needed. Intern Med 1995; 155: 930–935. Orthostatic hypotension and drops in BP seem to 5 Atli T, Keven K. Orthostatic hypotension in the healthy be more common among home-dwelling elderly elderly. Arch Gerontol Geriatr 2006; 43: 313–317. persons than has previously been reported. Our 6 Ooi WL, Barrett S, Hossain M, Kelley-Gagnon M, Lipsitz LA. Patterns of orthostatic blood pressure results suggest the need for testing OH in patients change and their clinical correlates in a frail, elderly aged 75 years or older regardless of the BP levels in a population. JAMA 1997; 277: 1299–1304. sitting position. Clinicians should assess medication 7 Ooi WL, Hossain M, Lipsitz LA. The association in persons with OH to prevent risks for falling and between orthostatic hypotension and recurrent falls in cardiovascular hazards. nursing home residents. Am J Med 2000; 108: 106–111. 8 Weiss A, Grossman E, Beloosesky Y, Grinblat J. Orthostatic hypotension in acute geriatric ward: Is it a consistent finding? Arch Intern Med 2002; 162: What is known about the topic 2369–2374. K Orthostatic hypotension is a major health problem in the 9 Poon IO, Braun U. High prevalence of orthostatic elderly and it affects 6–30% of home-dwelling elderly hypotension and its correlation with potentially 1–5 persons. causative medications among elderly veterans. J Clin K Orthostatic hypotension is a risk factor for dizziness and Pharmcol Ther 2005; 30: 173–178. falls as well for stroke, and it is associated with cardiovascular and all-cause mortality.2,3,16 10 Vloet LC, Pel-Little RE, Jansen PA, Jansen RW. High prevalence of postprandial and orthostatic hypoten- What this study adds sion among geriatric patients admitted to Dutch K This study reports that systolic blood pressure dropped in hospitals. J Gerontol A Biol Sci Med Sci 2005; 60: two-thirds of the home-dwelling elderly when they stood 1271–1277. up from a supine to a standing position and every third had 11 Weiss A, Beloosesky Y, Kornowski R, Yalov A, Grinblat orthostatic hypotension. This figure was higher than that in J, Grossman E. Influence of orthostatic hypotension on many previous studies. mortality among patients discharged from an acute K Pulse pressure in a supine position was higher in OH- geriatric ward. J Gen Intern Med 2006; 21: 602–606. positive than in OH-negative subjects, and after rising to a 12 Heitterachi E, Lord SR, Meyerkort P, McCloskey I, standing position, the decrease of pulse pressure was also more pronounced in OH-positive than in OH-negative Fitzpatrick R. Blood pressure changes on upright persons. tilting predict falls in older people. Age Ageing 2002; K High number of regularly used medicines is associated with 31: 181–186. orthostatic hypotension. 13 Kario K, Tobin JN, Wolfson LI, Whipple R, Derby CA, Singh D et al. Lower standing systolic blood pressure as a predictor of falls in the elderly: a community- based prospective study. J Am Coll Cardiol 2001; 38: 246–252. Acknowledgements 14 Mets TF. Drug-induced orthostatic hypotension in We thank statistician Piia Lavikainen for her help older patients. Drugs Aging 1995; 6: 219–228. 15 Gupta M, Lipsitz LA. Orthostatic hypotension in the with the statistics and research secretary Pa¨ivi elderly: diagnosis and treatment. Am J Med 2007; 120: Heikura for updating and managing the database. 841–847. This study was supported by Social Insurance 16 Luukinen H, Koski K, Laippala P, Airaksinen KE. Institution of Finland and City of Kuopio. Orthostatic hypotension and the risk of myocardial infarction in the home-dwelling elderly. J Intern Med 2004; 255: 486–493. 17 Liu BA, Topper AK, Reeves RA, Gryfe C, Maki BE. Conflict of interest Falls among older people: Relationship to medication None. use and orthostatic hypotension. J Am Geriatr Soc 1995; 43: 1141–1145. 18 Souverein PC, Van Staa TP, Egberts AC, De la Rosette JJ, Cooper C, Leufkens HG. Use of alpha-blockers and the References risk of hip/femur fractures. J Intern Med 2003; 254: 548–554. 1 Rutan GH, Hermanson B, Bild DE, Kittner SJ, LaBaw F, 19 Bradley JG, Davis KA. Orthostatic Hypotension. Am Tell GS. Orthostatic hypotension in older adults. The Fam Physician 2003; 68: 2393–2398.

Journal of Human Hypertension Postural changes in blood pressure and orthostatic hypotension P Hiitola et al 39 20 WHO. Guidelines for ATC Classification and DDD edn. American Psychiatric Association: Washington, Assignment. WHO Collaborating Centre for Drug DC, 1994. Statistics Methodology: Oslo, 2003. 24 Boddaert J, Tamim H, Verny M, Belmin J. Arterial 21 National Agency for Medicines. Classification of stiffness is associated with orthostatic hypotension in Medicines (ATC) and Defined Daily Doses (DDD) elderly subjects with history of falls. J Am Geriatr Soc 2004. National Agency for Medicines: Helsinki, 2004. 2004; 52: 568–572. 22 The Consensus Committee of the American Auto- 25 Blacher J, Staessen JA, Girerd X, Gasowski J, Thijs L, nomic Society and the American Academy of Wang JG et al. Pulse pressure not mean pressure Neurology. Consensus statement on the definition determines cardiovascular risk in older hypertensive of orthostatic hypotension, pure autonomic failure, patients. Arch Intern Med 2000; 160: 1085–1089. and multiple system atrophy. Neurology 1996; 26 Vloet LC, Smits R, Frederiks CM, Hoefnagels WH, 46: 1470. Jansen RW. Evaluation of skills and knowledge on 23 American Psychiatric Association. Diagnostic and orthostatic blood pressure measurements in elderly Statistical Manual of Mental Disorders: DSM-IV. 4th patients. Age Ageing 2002; 31: 211–216.

Journal of Human Hypertension