Prognostic Value of Ambulatory and Home Blood Pressures Compared With Office Blood Pressure in the General Population
http://www.100md.com
循环学杂志 2005年第4期
Clinica Medica (R.S., R.F., M.B., G.G., G.M.) and Medicina del Lavoro (R.F., G.C.)
Dipartimento di Medicina Clinica, Prevenzione e Biotecnologie Sanitarie, Università Milano-Bicocca, Ospedale San Gerardo, Monza
Dipartimento di Statistica (G.C.), Università Milano-Bicocca, Centro Interuniversitario di Fisiologia Clinica e Ipertensione (G.G., G.M.)
IRCCS Ospedale Maggiore
IRCCS Istituto Auxologico Italiano (G.G., G.M.), Milano, Italy.
Abstract
Background— Studies in hypertensive patients suggest that ambulatory blood pressure (BP) is prognostically superior to office BP. Much less information is available in the general population, however. Obtaining this information was the purpose of the Pressioni Arteriose Monitorate e Loro Associazioni (PAMELA) study.
Methods and Results— Office, home, and 24-hour ambulatory BP values were obtained in 2051 subjects between 25 and 74 years of age who were representative of the general population of Monza (Milan, Italy). Subjects were followed up for an average of 131 months, during which time cardiovascular and noncardiovascular fatal events were recorded (n=186). Office, home, and ambulatory BP values showed a significant exponential direct relationship with risk of cardiovascular or all-cause death. The goodness of fit of the relationship was greater for systolic than for diastolic BP and for night than for day BP, but its overall value was not better for home or ambulatory than for office BP. The slope of the relationship, however, was progressively greater from office to home and ambulatory BP. Home and night BP modestly improved the goodness of fit of the risk model when added to office BP.
Conclusions— In the PAMELA population, risk of death increased more with a given increase in home or ambulatory than in office BP. The overall ability to predict death, however, was not greater for home and ambulatory than for office BP, although it was somewhat increased by the combination of office and outside-of-office values. Systolic BP was almost invariably superior to diastolic BP, and night BP was superior to day BP.
Key Words: blood pressure monitoring ; heart rate ; mortality
Introduction
Two sets of evidence suggest that in hypertension and diabetes mellitus, ambulatory blood pressure (BP) may be clinically more important than office BP. Cross-sectional studies have shown that 24-hour, day, or night average BP values correlate with the organ damage accompanying hypertension and diabetes mellitus more closely than office values.1–4 Longitudinal studies of patients with untreated or treated hypertension or diabetes have shown that BP values averaged over the whole or part of 24 hours predict progression of organ damage5 or the risk of cardiovascular disease better than or in addition to BP values obtained in the office environment.6–18
Information on the prognostic value of ambulatory compared with office BP in the general population has been obtained only in a study of residents of Ohasama, Japan, based on the fatal cardiovascular events that occurred during a follow-up period of several years.19–22 The Italian Pressioni Arteriose Monitorate e Loro Associazioni (PAMELA Study) examined cross-sectionally a large sample representative of the population of Monza, a town in the northeast outskirts of Milan, from 1990 to 1993 to establish the normality range of ambulatory and home BP values,23 as well as the relationship between these pressures and echocardiographically detected abnormalities of the heart vis-à-vis office BP.24 From recruitment to December 31, 2003, contact with participants in the study was maintained by mail and phone. This method allowed the prognostic value of ambulatory compared with office BP to be assessed over a long observation period. It further allowed us to assess the prognostic value of home BP, for which prospective evidence also is limited largely to data obtained in the Ohasama population.25,26
Methods
The methodology used in the PAMELA Study has been reported in detail elsewhere.23 Briefly, 3200 individuals were randomly selected from the Monza residents to represent the town population for gender and age decades (25 to 74 years) according to the criteria used in the WHO MONICA study conducted in the same geographic area.27 The overall participation rate was 64% consistently in each age-sex stratum The demographic characteristics of nonparticipants were similar to those of participants, as were cardiovascular risk factors based on information collected via telephone interviews.
Entry Data
Participants were invited to come to the outpatient sector of the local hospital (San Gerardo) in the morning of a working day (Monday through Friday) where several data were collected. Data relevant to the present study include 3 sphygmomanometric BP measurements with the subject in the sitting position starting 10 minutes after the beginning of the medical visit and 24-hour ambulatory BP monitoring through an oscillometric device (Spacelabs 90207, Spacelabs) with BP readings set at 20-minute intervals. The subjects were sent home after the device was checked for accuracy with instructions to participate in their usual activities and to come back the next morning for device removal. Additional data included 2 home BP measurements (approximately at 7 AM and 7 PM) made with a semiautomatic device (Model HP 5331, Philips) on the arm contralateral to that used for ambulatory monitoring, 3 more sphygmomanometric sitting BP measurements taken after removal of the ambulatory BP device, and information (history and physical examination) on cardiovascular risk factors, including overweight, smoking habits, serum cholesterol, blood glucose, diabetes mellitus, and history of previous cardiovascular morbid events. Total serum cholesterol and blood glucose levels were measured in all subjects by a radioenzymatic method.
Follow-Up
the time of the medical visit to December 31, 2003, information on subjects’ vital status was requested by letter. A copy of the death certificate was obtained for all subjects who died. The causes of death reported in the certificate were coded according to the 10th revision of the International Classification of Diseases.28
Data Analysis
The 3 office BP measurements obtained at the initial visit and following ambulatory BP monitoring were separately averaged and pooled. The 2 home BP measurements were averaged. As reported in detail elsewhere,23 ambulatory BP values were edited for artifacts according to preselected criteria29 and averaged for the 24 hours, the day (7 AM to 11 PM), and the night (11 PM to 7 AM). Valid ambulatory blood pressure readings were obtained from 95.9% of the readings planned (ie, 72 readings over 24 hours) with a homogeneous distribution (2.9 per hour) throughout the entire recording period and a similar percentage of valid readings compared with the expected reading over the day (95.7%) and night (96.5%). Cox proportional-hazard models were fitted to explore the relationship between each BP or heart rate (HR) and the natural logarithm of the relative risk of cardiovascular or all-cause death. ; Coefficients (the average change in the natural logarithm of the relative risk of death per each 1–mm Hg or 1-bpm increase in BP or HR) were estimated by maximizing the logarithm of the partial likelihood function. For all models, the likelihood ratio test (difference between the likelihood of the model with and without the BP measurement as a covariate) was computed to evaluate whether BP (or HR) explained a significant portion of the risk of death.30 The likelihood ratio test was also computed for models that added home and ambulatory BP to the data relating the risk of mortality to office BP to see whether this improved the goodness of fit of the model and thus the ability to predict risk. The likelihood ratio test has an asymptotic 2 distribution with degrees of freedom equal to the number of estimated parameters. Kaplan-Meier survival curves were constructed by stratifying the cohort according to office, home, and ambulatory BP and HR above and below the corresponding median value. No adjustment for age, sex, and other cardiovascular risk factors was made because comparisons between the predictive value of various BP values involved the same sample. For each BP, the log-rank test was used to compare the curves. Throughout the text, values are given as mean±SD, except for the ; coefficients, which are given as mean±SEM. For all hypotheses tested, 2-tailed values of P<0.05 were considered statistically significant.
Results
Entry and Follow-Up Data
Entry data were obtained for 2051 subjects. During the follow-up period (average, 131 months), there were 186 deaths, ie, 9.1% of the sample. Of these deaths, 56 (2.7%) were cardiovascular. Cardiovascular death was ascribed to myocardial infarction or coronary heart disease (n=27), stroke resulting from vascular occlusion or hemorrhage (n=15), heart failure (n=6), sudden death (n=2), rupture of an aortic aneurysm (n=3), hypertension (n=1), and systemic embolism (n=2). Table 1 shows that compared with subjects alive at the end of the follow-up, those who died of cardiovascular diseases were older and more likely to be male. Prevalence of smoking was similar in the 2 groups, but hypertension (office BP 140 mm Hg systolic or 90 mm Hg diastolic or antihypertensive treatment), hypercholesterolemia (serum total cholesterol 240 mg/dL), overweight (body mass index 26 kg/m2), diabetes mellitus (fasting serum glucose 140 mg/dL or antidiabetic treatment), total serum cholesterol, blood glucose, and a history of cardiovascular disease were much more frequent in those who had a fatal cardiovascular event. The all-cause death group resembled the cardiovascular death group, although the differences with alive subjects were less pronounced. BP values were progressively less from office to home, day, 24 hours, and nighttime, with all values being greater in the cardiovascular and all-cause death groups than in the group of alive subjects. Except for a slightly greater office value in those who died of all causes, HR did not differ between groups.
BP and Mortality
All entry BP values showed a direct exponential relationship with the risk of cardiovascular or all-cause death. As shown in Table 2, the goodness of fit of the relationship of BP to the risk of death was almost invariably much less for diastolic than for systolic BP, for which the values were usually higher than those required for minimal statistical significance and not less for office than for home or ambulatory BP values. The ; coefficients (ie, the increase in risk per 1–mm Hg increase in BP) were progressively greater from the first day, the second day, or the pooled office BP values to home and 24-hour BP values, with lower values for the day than for the night. Thus, as shown in Figure 1, the risk of cardiovascular and all-cause mortality increased more steeply from office to home, day, 24-hour, and night BP. This is further illustrated in Figure 2, which shows that at different initial systolic BPs, a 10–mm Hg BP increase raised the risk of cardiovascular death in a progressively greater fashion from office to home, day, 24-hour, and night values, as was also the case (except for daytime) for diastolic BP. Figure 1 also shows that the same increase in risk occurred at progressively lower values from office to nighttime. This finding is further illustrated in Figure 3, which shows at which office, home, and ambulatory BP values the 10%, 20%, and 30% risk of death was reached.
The risk of cardiovascular and all-cause death showed no correlation with HR values measured in the office, at home, or over 24 hours (Table 2). Kaplan-Meier survival curves showed similar survival rates in groups with HR value above and below the median (log-rank test always not significant), in contrast with the widely different survival rates in groups below and above median values of office, home, and ambulatory BP values (Figure 4); this log-rank test was always statistically significant (P<0.0001).
Multivariate Analysis
Discussion
The 11-year follow-up of subjects from the PAMELA population shows that office BP measured in 1 or 2 visits, home BP, and ambulatory BP were all higher in individuals who died of cardiovascular or other causes compared with those who remained alive. It also shows that all pressures were predictive of the risk of death with no prognostic superiority of self-measured or ambulatory values over the clinically collected values both for the incidence of fatal events and for their timing during the follow-up period. This finding does not confirm the conclusion drawn by previous studies that ambulatory or home BP values are prognostically superior to BP values obtained in the clinic environment by the conventional approach.7,10–12,18,20,21 Rather, it suggests that in the general population and for long observation periods, BP values measured under different conditions have a similar predictive ability.
The reasons for the difference between the results of the present study and those that have previously concluded a prognostic superiority of ambulatory or home BP need to be addressed. Three possibilities can be advanced. First, the longer follow-up and greater number of diagnostically safe events available in our study may have made our data less prone to chance than those obtained in several previous studies. Second, ambulatory BP may be prognostically superior to office BP in hypertensive patients (to whom most previous studies refer), whereas little prognostic difference between these BP values may exist in normotensive individuals who represent most of the population. Third, in previous studies, the finding of a prognostic superiority of ambulatory compared with office BP may have been favored by the methods used to analyze the data. In a recent study of hypertensive patients, for example, the subgroup with an office systolic BP between 140 and 159 mm Hg was reported to have a different cardiovascular morbidity rate in relation to a 24-hour BP above or below 135 mm Hg.12 However, ambulatory and office BP values are significantly related to each other4; the correlation coefficient was high as 0.70 in this population.23 Thus, it is thus that these 2 subgroups also had higher and lower office values, respectively (eg, in the 150 and 140 mm Hg range), making the interpretation of morbidity data resulting from differences in ambulatory BP only somewhat speculative.
Although similarly predictive of the risk of death, office, home, and ambulatory BP values showed some important differences. First, the same degree of risk was reached at BP values that were progressively and markedly lower from office to home, day, 24-hour, and night values, which is in line with the differences between these pressures seen in the population.23,31–34 Second, the increment in risk per 1–mm Hg increase in BP was progressively and markedly greater from office to ambulatory values, with a 10–mm Hg increase in systolic BP leading to a several-times-greater increase in risk if based on 24-hour or night values than on office values. This is in line with the steeper changes in cardiovascular risk with an increase in ambulatory compared with office BP reported in previous studies.7,10–12,18,20,21 However, a steeper relationship between ambulatory BP and risk can be accounted for by the fact that as means of a large number of values, 24-hour BP values are distributed over a narrower range than individual BP values or means obtained by a few values only, without necessarily implying a greater predictive ability. A narrower distribution range characterized ambulatory BP values in the PAMELA population.35 Third, adding home or ambulatory (night) BP values to conventional BP significantly improved the goodness of fit of the risk model. This finding suggests that although not progressively superior to one another, these pressures possess different types of information, and their combination may be clinically useful. It would seem from our data that the combination that is useful is that between clinic and out-of-clinic BP, regardless of whether it is obtained at home or over 24 hours, because the 3 BP values together were not more predictive than 2 values.
Our study provides several other results. First, home BP adequately predicted the risk of death despite the fact that only 2 home BP values were available, which did not allow us to fully explore the potential of this approach, ie, to collect values over days and weeks. This should further support the use of self-measured BP in clinical practice. Second, office, home, and ambulatory systolic BP values all showed a closer relationship with risk of death than the corresponding diastolic BP values. This may be due to the fact that indirect methods for measuring blood pressure are more accurate for systolic than for diastolic values. This is particularly the case when, as in our study, the BP measuring device is based on an oscillometric method that does not directly measure diastolic BP.36 It is more likely, however, that this reflects, at least in part, the prognostic superiority of systolic over diastolic BP shown by classic epidemiological studies,37,38 which our findings extend to home and ambulatory systolic values. Third, BP values within 24 hours were prognostically different; night BP showed a steeper relationship with the risk of death than day BP, and more importantly, the goodness of fit to the risk model was greater for night than for day BP values. The reasons for this prognostic superiority of night over day BP are not clear, although 2 hypotheses can be advanced. First, the prognostic value of day BP may be reduced by the contribution to the risk of BP variability,1,4,39 which is much more pronounced during the day than during the night.40 Second, during the night, reduction of arteriolar tone may allow greater transmission of BP values existing in the large artery compartment to microcirculation, enhancing their damaging effects. Regardless of the possible explanations, however, this further supports the recommendation to extend ambulatory BP monitoring to a 24-hour time interval. Finally, there was in our subjects no relationship between the risk or time of death and HR, regardless of whether the measurements were made at different office visits or at home and over 24 hours. This is not in line with the report of a prognostic value of HR for cardiovascular and all-cause death.41 A limitation of our study is that the number of cardiovascular deaths was small (n=56), so the study power was limited. However, the data were similar when the much greater number of all-cause deaths (n=186) was considered. This strengthens the study conclusion, although with the limitation that it reflects what happens in an overall low-cardiovascular-risk population.
In conclusion, the long follow-up of the PAMELA sample provides evidence that office, home, and ambulatory BPs are similarly predictive of the risk of cardiovascular and all-cause death, with a superiority of systolic over diastolic BP and of night over daytime values. It further shows that the risk already increases at lower values and more steeply for ambulatory and home than for office BP. Finally, it shows that the different information provided by clinic and out-of-clinic BP may be combined to improve prediction of risk.
References
Mancia G, Parati G. Ambulatory blood pressure monitoring and organ damage. Hypertension. 2000; 36: 894–900.
Mancia G, Zanchetti A, Agabiti-Rosei E, Benemio G, De Cesaris R, Fogari R, Pessina A, Porcellati C, Rappelli A, Salvetti A, Trimarco B. Ambulatory blood pressure is superior to clinic blood pressure in predicting treatment-reduced regression of left ventricular hypertrophy. Circulation. 1997; 95: 1464–1470.
Jula A, Puukka P, Karanko H. Multiple clinic and home blood pressure measurements versus ambulatory blood pressure monitoring. Hypertension. 1999; 34: 261–266.
Mancia G, Parati G, Hennig M, Flatau B, Omboni S, Glavina F, Costa B, Scherz R, Bond G, Zanchetti A, for the ELSA Investigators. Relationship between blood pressure variability and carotid artery damage in hypertension: baseline data from the European Lacidipine Study on Atherosclerosis (ELSA). J Hypertens. 2001; 19: 1981–1989.
Lurbe E, Redon J, Kesani A, Pascual JM, Tacons J, Alvarez V, Batlle D. Increase in nocturnal blood pressure and progression to microalbuminuria in type I diabetes. N Engl J Med. 2002; 347: 797–805.
Perloff D, Sokolow M, Cowan R. The prognostic value of ambulatory blood pressures. JAMA. 1983; 249: 2793–2798.
Redon J, Campos C, Narciso ML, Rodicio JL, Pascual JM, Ruilope LM. Prognostic value of ambulatory blood pressure monitoring in refractory hypertension: a prospective study. Hypertension. 1998; 31: 712–718.
Khattar RS, Swales JD, Banfield A, Dore C, Senior R, Lahiri A. Prediction of coronary and cerebrovascular morbidity blood pressure monitoring in essential hypertension. Circulation. 1999; 100: 1071–1076.
Robinson TG, Dawson SL, Ahmed U, Manktelow B, Fotherby MD, Potter JF. Twenty-four hour systolic blood pressure predicts long-term mortality following acute stroke. J Hypertens. 2001; 19: 2127–2134.
Staessen JA, Thijs L, Fagard R, O’Brien ET, Clement D, de Leeuw PW, Mancia G, Nachev C, Palatini P, Parati G, Tuomilehto J, Webster J. Predicting cardiovascular risk using conventional vs ambulatory blood pressure in older patients with systolic hypertension: Systolic Hypertension in Europe Trial Investigators. JAMA. 1999; 282: 539–546.
Verdecchia P, Reboldi G, Porcellati C, Schillaci G, Pede S, Bentivoglio M, Angeli F, Norgiolini S, Ambrosio G. Risk of cardiovascular disease in relation to achieved office and ambulatory blood pressure control in treated hypertensive subjects. J Am Coll Cardiol. 2002; 39: 878–885.
Clement DL, De Buyzere ML, De Bacquer DA, de Leeuw PW, Duprez DA, Fagard RH, Gheeraert PJ, Missault LH, Braun JJ, Six RO, Van Der Niepen P, O’Brien E, for the Office Versus Ambulatory Pressure Study Investigators. Prognostic value of ambulatory blood pressure in patients with treated hypertension. N Engl J Med. 2003; 348: 2407–2415.
Zweiker R, Eber B, Schumacher M, Toplak H, Klein W. "Non-dipping" related to cardiovascular events in essential hypertensive patients. Acta Med Austriaca. 1994; 21: 86–89.
Bjorklund K, Lind L, Zethelius B, Andren B, Lithell H. Isolated ambulatory hypertension predicts cardiovascular morbidity in elderly men. Circulation. 2003; 107: 1297–1302.
Yamamoto Y, Akiguchi I, Oiwa K, Hayashi M, Kimura J. Adverse effect of nighttime blood pressure on the outcome of lacunar infarct patients. Stroke. 1998; 29: 570–576.
Kario K, Pickering TG, Matsuo T, Hoshide S, Schwartz JE, Shimada K. Stroke prognosis and abnormal nocturnal blood pressure falls in older hypertensives. Hypertension. 2001; 38: 852–857.
Verdecchia P, Porcellati C, Schillaci G, Borgioni C, Ciucci A, Battistelli M, Guerrieri M, Gatteschi C, Zampi I, Santucci A. Ambulatory blood pressure: an independent predictor of prognosis in essential hypertension. Hypertension. 1994; 24: 793–801.
Bobrie G, Chatellier G, Genes N, Clerson P, Vaur L, Vaisse B, Menard J, Mallion JM. Cardiovascular prognosis of "masked hypertension" detected by blood pressure self-measurement in elderly treated hypertensive patients. JAMA. 2004; 291: 1342–1349.
Kikuya M, Hozawa A, Ohokubo T, Tsuji I, Michimata M, Matsubara M, Ota M, Nagai K, Araki T, Satoh H, Ito S, Hisamichi S, Imai Y. Prognostic significance of blood pressure and heart rate variabilities: the Ohasama Study. Hypertension. 2000; 36: 901–906.
Ohkubo T, Hozawa A, Nagai K, Kikuya M, Tsuji I, Ito S, Satoh H, Hisamichi S, Imai Y. Prediction of stroke by ambulatory blood pressure monitoring versus screening blood pressure measurement in a general population: the Ohasama Study. J Hypertens. 2000; 18: 847–854.
Imai Y, Ohokubo T, Sakuma M, Tsuji II, Satoh H, Nagai K, Hisamichi S, Abe K. Predictive power of screening blood pressure, ambulatory blood pressure and blood pressure measured at home for overall and cardiovascular mortality: a prospective observation in a cohort from Ohasama, northern Japan. Blood Press Monit. 1996; 1: 251–254.
Imai Y, Hozawa A, Ohkubo T, Tsuji I, Yamaguchi J, Matsubara M, Michimata M, Hashimoto J, Fujiwara T, Nagai K, Kitaoka H, Satoh H, Hisamichi S. Predictive value of automated blood pressure measurement: what can we learn from the Japanese population: the Ohasama Study. Blood Press Monit. 2001; 6: 335–339.
Mancia G, Sega R, Bravi C, De Vito G, Valagussa F, Cesana G, Zanchetti A. Ambulatory blood pressure normality: results from the PAMELA Study. J Hypertens. 1995; 13: 1377–1390.
Sega R, Trocino G, Lanzarotti A, Carugo S, Cesana G, Schiavina R, Valagussa F, Bombelli M, Giannattasio C, Zanchetti A, Mancia G. Alterations of cardiac structure in patients with isolated office, ambulatory, or home hypertension: data from the general population [Pressione Arteriose Monitorate e Loro Associazioni (PAMELA) Study]. Circulation. 2001; 104: 1385–1392.
Hozawa A, Ohkubo T, Nagai K, Kikuya M, Matsubara M, Tsuji I, Ito S, Satoh H, Hisamichi S, Imai Y. Prognosis of isolated systolic and isolated diastolic hypertension as assessed by self-measurement of blood pressure at home: the Ohasama Study. Arch Intern Med. 2000; 160: 3301–3306.
Imay I, Poncelet P, DeBuyzere M, Padfield PL, Van Monfrans GA. Prognostic significance of self-measurements of blood pressure. Blood Press Monit. 2000; 5: 137–143.
WHO MONICA Project Principal Investigators. The World Health Organization MONICA project: a major international collaboration. J Clin Epidemiol. 1988; 41: 105–114.
Anderson RN, Rosenberg HM. Disease classification: measuring the effect of the tenth revision of the International Classification of Diseases on cause-of-death in the United States. Stat Med. 2003; 15: 1551–1570.
Groppelli A, Omboni S, Parati G, Mancia G. Evaluation of non-invasive blood pressure monitoring devices Spacelabs 90202 and 90207 versus resting and ambulatory 24 hour intraarterial blood pressure. Hypertension. 1992; 20: 227–232.
Marubini E, Valsecchi MG. Analyzing Survival Data From Clinical Trials and Observational Studies. Chichester, UK: J. Wiley & Sons; 1995.
Scettini C, Bianchi M, Nieto F, Sandoya E, Senra H. Ambulatory blood pressure: normality and comparison with other measurements: Hypertension Working Group. Hypertension. 1999; 34: 818–825.
Staessen JA, O’Brien ET, Amery AK, Atkins N, Baumgart P, De Cort P, Degaute JP, Dolenc P, De Gaudemaris R, Enstrom I. Ambulatory blood pressure in normotensive and hypertensive subjects: results from an international database. J Hypertens. 1994; 12 (suppl 7): S1–S12.
Wiinberg N, Hoegholm A, Christensen HR, Bang LE, Mikkelsen KL, Nielsen PE, Svendsen TL, Kampmann JP, Madsen NH, Bentzon MW. 24 hour ambulatory blood pressure in 352 normal Danish subjects, related to age and gender. Am J Hypertens. 1995; 8: 978–986.
Mancia G, O’Brien E, Imai Y, Redon J. Task Force II: Ambulatory blood pressure monitoring in population studies. Blood Press Monit. 1999; 4: 295–301.
Mancia G, Sega R, Grassi G, Cesana G, Zanchetti A. Defining ambulatory and home blood pressure normality: further considerations based on data from the PAMELA Study. J Hypertens. 2001; 19: 995–999.
Guidelines Committee. 2003 European Society of Hypertension–European Society of Cardiology guidelines for the management of arterial hypertension. J Hypertens. 2003; 21: 1011–1053.
Franklin SS, Khan SA, Wong ND, Larson MG, Levy D. Is pulse pressure useful in predicting risk for coronary heart disease; The Framingham Heart Study. Circulation. 1999; 100: 354–360.
Domanski M, Mitchell G, Pfeffer M, Neaton JD, Norman J, Svendsen K, Grimm R, Cohen J, Stamler J, for the MRFIT Research Group. Pulse pressure and cardiovascular disease-related mortality: follow-up of the Multiple Risk Factor Intervention Trial (MRFIT). JAMA. 2002; 287: 2677–2683.
Sander D, Kukla C, Klingelhofer J, Winbeck K, Conrad B. Relationship between circadian blood pressure patterns and progression of early carotid atherosclerosis: a 3-year follow-up study. Circulation. 2000; 102: 1536–1541.
Mancia G, Ferrari A, Gregorini L, Parati G, Pomidossi G, Bertinieri G, Grassi G, di Rienzo M, Pedotti A, Zanchetti A. Blood pressure and heart rate variabilities in normotensive and hypertensive human beings. Circ Res. 1983; 53: 96–104.
Palatini P, Julius S. Heart rate and the cardiovascular risk. J Hypertens. 1997; 15: 3–17.(Roberto Sega, MD; Rita Fa)
Dipartimento di Medicina Clinica, Prevenzione e Biotecnologie Sanitarie, Università Milano-Bicocca, Ospedale San Gerardo, Monza
Dipartimento di Statistica (G.C.), Università Milano-Bicocca, Centro Interuniversitario di Fisiologia Clinica e Ipertensione (G.G., G.M.)
IRCCS Ospedale Maggiore
IRCCS Istituto Auxologico Italiano (G.G., G.M.), Milano, Italy.
Abstract
Background— Studies in hypertensive patients suggest that ambulatory blood pressure (BP) is prognostically superior to office BP. Much less information is available in the general population, however. Obtaining this information was the purpose of the Pressioni Arteriose Monitorate e Loro Associazioni (PAMELA) study.
Methods and Results— Office, home, and 24-hour ambulatory BP values were obtained in 2051 subjects between 25 and 74 years of age who were representative of the general population of Monza (Milan, Italy). Subjects were followed up for an average of 131 months, during which time cardiovascular and noncardiovascular fatal events were recorded (n=186). Office, home, and ambulatory BP values showed a significant exponential direct relationship with risk of cardiovascular or all-cause death. The goodness of fit of the relationship was greater for systolic than for diastolic BP and for night than for day BP, but its overall value was not better for home or ambulatory than for office BP. The slope of the relationship, however, was progressively greater from office to home and ambulatory BP. Home and night BP modestly improved the goodness of fit of the risk model when added to office BP.
Conclusions— In the PAMELA population, risk of death increased more with a given increase in home or ambulatory than in office BP. The overall ability to predict death, however, was not greater for home and ambulatory than for office BP, although it was somewhat increased by the combination of office and outside-of-office values. Systolic BP was almost invariably superior to diastolic BP, and night BP was superior to day BP.
Key Words: blood pressure monitoring ; heart rate ; mortality
Introduction
Two sets of evidence suggest that in hypertension and diabetes mellitus, ambulatory blood pressure (BP) may be clinically more important than office BP. Cross-sectional studies have shown that 24-hour, day, or night average BP values correlate with the organ damage accompanying hypertension and diabetes mellitus more closely than office values.1–4 Longitudinal studies of patients with untreated or treated hypertension or diabetes have shown that BP values averaged over the whole or part of 24 hours predict progression of organ damage5 or the risk of cardiovascular disease better than or in addition to BP values obtained in the office environment.6–18
Information on the prognostic value of ambulatory compared with office BP in the general population has been obtained only in a study of residents of Ohasama, Japan, based on the fatal cardiovascular events that occurred during a follow-up period of several years.19–22 The Italian Pressioni Arteriose Monitorate e Loro Associazioni (PAMELA Study) examined cross-sectionally a large sample representative of the population of Monza, a town in the northeast outskirts of Milan, from 1990 to 1993 to establish the normality range of ambulatory and home BP values,23 as well as the relationship between these pressures and echocardiographically detected abnormalities of the heart vis-à-vis office BP.24 From recruitment to December 31, 2003, contact with participants in the study was maintained by mail and phone. This method allowed the prognostic value of ambulatory compared with office BP to be assessed over a long observation period. It further allowed us to assess the prognostic value of home BP, for which prospective evidence also is limited largely to data obtained in the Ohasama population.25,26
Methods
The methodology used in the PAMELA Study has been reported in detail elsewhere.23 Briefly, 3200 individuals were randomly selected from the Monza residents to represent the town population for gender and age decades (25 to 74 years) according to the criteria used in the WHO MONICA study conducted in the same geographic area.27 The overall participation rate was 64% consistently in each age-sex stratum The demographic characteristics of nonparticipants were similar to those of participants, as were cardiovascular risk factors based on information collected via telephone interviews.
Entry Data
Participants were invited to come to the outpatient sector of the local hospital (San Gerardo) in the morning of a working day (Monday through Friday) where several data were collected. Data relevant to the present study include 3 sphygmomanometric BP measurements with the subject in the sitting position starting 10 minutes after the beginning of the medical visit and 24-hour ambulatory BP monitoring through an oscillometric device (Spacelabs 90207, Spacelabs) with BP readings set at 20-minute intervals. The subjects were sent home after the device was checked for accuracy with instructions to participate in their usual activities and to come back the next morning for device removal. Additional data included 2 home BP measurements (approximately at 7 AM and 7 PM) made with a semiautomatic device (Model HP 5331, Philips) on the arm contralateral to that used for ambulatory monitoring, 3 more sphygmomanometric sitting BP measurements taken after removal of the ambulatory BP device, and information (history and physical examination) on cardiovascular risk factors, including overweight, smoking habits, serum cholesterol, blood glucose, diabetes mellitus, and history of previous cardiovascular morbid events. Total serum cholesterol and blood glucose levels were measured in all subjects by a radioenzymatic method.
Follow-Up
the time of the medical visit to December 31, 2003, information on subjects’ vital status was requested by letter. A copy of the death certificate was obtained for all subjects who died. The causes of death reported in the certificate were coded according to the 10th revision of the International Classification of Diseases.28
Data Analysis
The 3 office BP measurements obtained at the initial visit and following ambulatory BP monitoring were separately averaged and pooled. The 2 home BP measurements were averaged. As reported in detail elsewhere,23 ambulatory BP values were edited for artifacts according to preselected criteria29 and averaged for the 24 hours, the day (7 AM to 11 PM), and the night (11 PM to 7 AM). Valid ambulatory blood pressure readings were obtained from 95.9% of the readings planned (ie, 72 readings over 24 hours) with a homogeneous distribution (2.9 per hour) throughout the entire recording period and a similar percentage of valid readings compared with the expected reading over the day (95.7%) and night (96.5%). Cox proportional-hazard models were fitted to explore the relationship between each BP or heart rate (HR) and the natural logarithm of the relative risk of cardiovascular or all-cause death. ; Coefficients (the average change in the natural logarithm of the relative risk of death per each 1–mm Hg or 1-bpm increase in BP or HR) were estimated by maximizing the logarithm of the partial likelihood function. For all models, the likelihood ratio test (difference between the likelihood of the model with and without the BP measurement as a covariate) was computed to evaluate whether BP (or HR) explained a significant portion of the risk of death.30 The likelihood ratio test was also computed for models that added home and ambulatory BP to the data relating the risk of mortality to office BP to see whether this improved the goodness of fit of the model and thus the ability to predict risk. The likelihood ratio test has an asymptotic 2 distribution with degrees of freedom equal to the number of estimated parameters. Kaplan-Meier survival curves were constructed by stratifying the cohort according to office, home, and ambulatory BP and HR above and below the corresponding median value. No adjustment for age, sex, and other cardiovascular risk factors was made because comparisons between the predictive value of various BP values involved the same sample. For each BP, the log-rank test was used to compare the curves. Throughout the text, values are given as mean±SD, except for the ; coefficients, which are given as mean±SEM. For all hypotheses tested, 2-tailed values of P<0.05 were considered statistically significant.
Results
Entry and Follow-Up Data
Entry data were obtained for 2051 subjects. During the follow-up period (average, 131 months), there were 186 deaths, ie, 9.1% of the sample. Of these deaths, 56 (2.7%) were cardiovascular. Cardiovascular death was ascribed to myocardial infarction or coronary heart disease (n=27), stroke resulting from vascular occlusion or hemorrhage (n=15), heart failure (n=6), sudden death (n=2), rupture of an aortic aneurysm (n=3), hypertension (n=1), and systemic embolism (n=2). Table 1 shows that compared with subjects alive at the end of the follow-up, those who died of cardiovascular diseases were older and more likely to be male. Prevalence of smoking was similar in the 2 groups, but hypertension (office BP 140 mm Hg systolic or 90 mm Hg diastolic or antihypertensive treatment), hypercholesterolemia (serum total cholesterol 240 mg/dL), overweight (body mass index 26 kg/m2), diabetes mellitus (fasting serum glucose 140 mg/dL or antidiabetic treatment), total serum cholesterol, blood glucose, and a history of cardiovascular disease were much more frequent in those who had a fatal cardiovascular event. The all-cause death group resembled the cardiovascular death group, although the differences with alive subjects were less pronounced. BP values were progressively less from office to home, day, 24 hours, and nighttime, with all values being greater in the cardiovascular and all-cause death groups than in the group of alive subjects. Except for a slightly greater office value in those who died of all causes, HR did not differ between groups.
BP and Mortality
All entry BP values showed a direct exponential relationship with the risk of cardiovascular or all-cause death. As shown in Table 2, the goodness of fit of the relationship of BP to the risk of death was almost invariably much less for diastolic than for systolic BP, for which the values were usually higher than those required for minimal statistical significance and not less for office than for home or ambulatory BP values. The ; coefficients (ie, the increase in risk per 1–mm Hg increase in BP) were progressively greater from the first day, the second day, or the pooled office BP values to home and 24-hour BP values, with lower values for the day than for the night. Thus, as shown in Figure 1, the risk of cardiovascular and all-cause mortality increased more steeply from office to home, day, 24-hour, and night BP. This is further illustrated in Figure 2, which shows that at different initial systolic BPs, a 10–mm Hg BP increase raised the risk of cardiovascular death in a progressively greater fashion from office to home, day, 24-hour, and night values, as was also the case (except for daytime) for diastolic BP. Figure 1 also shows that the same increase in risk occurred at progressively lower values from office to nighttime. This finding is further illustrated in Figure 3, which shows at which office, home, and ambulatory BP values the 10%, 20%, and 30% risk of death was reached.
The risk of cardiovascular and all-cause death showed no correlation with HR values measured in the office, at home, or over 24 hours (Table 2). Kaplan-Meier survival curves showed similar survival rates in groups with HR value above and below the median (log-rank test always not significant), in contrast with the widely different survival rates in groups below and above median values of office, home, and ambulatory BP values (Figure 4); this log-rank test was always statistically significant (P<0.0001).
Multivariate Analysis
Discussion
The 11-year follow-up of subjects from the PAMELA population shows that office BP measured in 1 or 2 visits, home BP, and ambulatory BP were all higher in individuals who died of cardiovascular or other causes compared with those who remained alive. It also shows that all pressures were predictive of the risk of death with no prognostic superiority of self-measured or ambulatory values over the clinically collected values both for the incidence of fatal events and for their timing during the follow-up period. This finding does not confirm the conclusion drawn by previous studies that ambulatory or home BP values are prognostically superior to BP values obtained in the clinic environment by the conventional approach.7,10–12,18,20,21 Rather, it suggests that in the general population and for long observation periods, BP values measured under different conditions have a similar predictive ability.
The reasons for the difference between the results of the present study and those that have previously concluded a prognostic superiority of ambulatory or home BP need to be addressed. Three possibilities can be advanced. First, the longer follow-up and greater number of diagnostically safe events available in our study may have made our data less prone to chance than those obtained in several previous studies. Second, ambulatory BP may be prognostically superior to office BP in hypertensive patients (to whom most previous studies refer), whereas little prognostic difference between these BP values may exist in normotensive individuals who represent most of the population. Third, in previous studies, the finding of a prognostic superiority of ambulatory compared with office BP may have been favored by the methods used to analyze the data. In a recent study of hypertensive patients, for example, the subgroup with an office systolic BP between 140 and 159 mm Hg was reported to have a different cardiovascular morbidity rate in relation to a 24-hour BP above or below 135 mm Hg.12 However, ambulatory and office BP values are significantly related to each other4; the correlation coefficient was high as 0.70 in this population.23 Thus, it is thus that these 2 subgroups also had higher and lower office values, respectively (eg, in the 150 and 140 mm Hg range), making the interpretation of morbidity data resulting from differences in ambulatory BP only somewhat speculative.
Although similarly predictive of the risk of death, office, home, and ambulatory BP values showed some important differences. First, the same degree of risk was reached at BP values that were progressively and markedly lower from office to home, day, 24-hour, and night values, which is in line with the differences between these pressures seen in the population.23,31–34 Second, the increment in risk per 1–mm Hg increase in BP was progressively and markedly greater from office to ambulatory values, with a 10–mm Hg increase in systolic BP leading to a several-times-greater increase in risk if based on 24-hour or night values than on office values. This is in line with the steeper changes in cardiovascular risk with an increase in ambulatory compared with office BP reported in previous studies.7,10–12,18,20,21 However, a steeper relationship between ambulatory BP and risk can be accounted for by the fact that as means of a large number of values, 24-hour BP values are distributed over a narrower range than individual BP values or means obtained by a few values only, without necessarily implying a greater predictive ability. A narrower distribution range characterized ambulatory BP values in the PAMELA population.35 Third, adding home or ambulatory (night) BP values to conventional BP significantly improved the goodness of fit of the risk model. This finding suggests that although not progressively superior to one another, these pressures possess different types of information, and their combination may be clinically useful. It would seem from our data that the combination that is useful is that between clinic and out-of-clinic BP, regardless of whether it is obtained at home or over 24 hours, because the 3 BP values together were not more predictive than 2 values.
Our study provides several other results. First, home BP adequately predicted the risk of death despite the fact that only 2 home BP values were available, which did not allow us to fully explore the potential of this approach, ie, to collect values over days and weeks. This should further support the use of self-measured BP in clinical practice. Second, office, home, and ambulatory systolic BP values all showed a closer relationship with risk of death than the corresponding diastolic BP values. This may be due to the fact that indirect methods for measuring blood pressure are more accurate for systolic than for diastolic values. This is particularly the case when, as in our study, the BP measuring device is based on an oscillometric method that does not directly measure diastolic BP.36 It is more likely, however, that this reflects, at least in part, the prognostic superiority of systolic over diastolic BP shown by classic epidemiological studies,37,38 which our findings extend to home and ambulatory systolic values. Third, BP values within 24 hours were prognostically different; night BP showed a steeper relationship with the risk of death than day BP, and more importantly, the goodness of fit to the risk model was greater for night than for day BP values. The reasons for this prognostic superiority of night over day BP are not clear, although 2 hypotheses can be advanced. First, the prognostic value of day BP may be reduced by the contribution to the risk of BP variability,1,4,39 which is much more pronounced during the day than during the night.40 Second, during the night, reduction of arteriolar tone may allow greater transmission of BP values existing in the large artery compartment to microcirculation, enhancing their damaging effects. Regardless of the possible explanations, however, this further supports the recommendation to extend ambulatory BP monitoring to a 24-hour time interval. Finally, there was in our subjects no relationship between the risk or time of death and HR, regardless of whether the measurements were made at different office visits or at home and over 24 hours. This is not in line with the report of a prognostic value of HR for cardiovascular and all-cause death.41 A limitation of our study is that the number of cardiovascular deaths was small (n=56), so the study power was limited. However, the data were similar when the much greater number of all-cause deaths (n=186) was considered. This strengthens the study conclusion, although with the limitation that it reflects what happens in an overall low-cardiovascular-risk population.
In conclusion, the long follow-up of the PAMELA sample provides evidence that office, home, and ambulatory BPs are similarly predictive of the risk of cardiovascular and all-cause death, with a superiority of systolic over diastolic BP and of night over daytime values. It further shows that the risk already increases at lower values and more steeply for ambulatory and home than for office BP. Finally, it shows that the different information provided by clinic and out-of-clinic BP may be combined to improve prediction of risk.
References
Mancia G, Parati G. Ambulatory blood pressure monitoring and organ damage. Hypertension. 2000; 36: 894–900.
Mancia G, Zanchetti A, Agabiti-Rosei E, Benemio G, De Cesaris R, Fogari R, Pessina A, Porcellati C, Rappelli A, Salvetti A, Trimarco B. Ambulatory blood pressure is superior to clinic blood pressure in predicting treatment-reduced regression of left ventricular hypertrophy. Circulation. 1997; 95: 1464–1470.
Jula A, Puukka P, Karanko H. Multiple clinic and home blood pressure measurements versus ambulatory blood pressure monitoring. Hypertension. 1999; 34: 261–266.
Mancia G, Parati G, Hennig M, Flatau B, Omboni S, Glavina F, Costa B, Scherz R, Bond G, Zanchetti A, for the ELSA Investigators. Relationship between blood pressure variability and carotid artery damage in hypertension: baseline data from the European Lacidipine Study on Atherosclerosis (ELSA). J Hypertens. 2001; 19: 1981–1989.
Lurbe E, Redon J, Kesani A, Pascual JM, Tacons J, Alvarez V, Batlle D. Increase in nocturnal blood pressure and progression to microalbuminuria in type I diabetes. N Engl J Med. 2002; 347: 797–805.
Perloff D, Sokolow M, Cowan R. The prognostic value of ambulatory blood pressures. JAMA. 1983; 249: 2793–2798.
Redon J, Campos C, Narciso ML, Rodicio JL, Pascual JM, Ruilope LM. Prognostic value of ambulatory blood pressure monitoring in refractory hypertension: a prospective study. Hypertension. 1998; 31: 712–718.
Khattar RS, Swales JD, Banfield A, Dore C, Senior R, Lahiri A. Prediction of coronary and cerebrovascular morbidity blood pressure monitoring in essential hypertension. Circulation. 1999; 100: 1071–1076.
Robinson TG, Dawson SL, Ahmed U, Manktelow B, Fotherby MD, Potter JF. Twenty-four hour systolic blood pressure predicts long-term mortality following acute stroke. J Hypertens. 2001; 19: 2127–2134.
Staessen JA, Thijs L, Fagard R, O’Brien ET, Clement D, de Leeuw PW, Mancia G, Nachev C, Palatini P, Parati G, Tuomilehto J, Webster J. Predicting cardiovascular risk using conventional vs ambulatory blood pressure in older patients with systolic hypertension: Systolic Hypertension in Europe Trial Investigators. JAMA. 1999; 282: 539–546.
Verdecchia P, Reboldi G, Porcellati C, Schillaci G, Pede S, Bentivoglio M, Angeli F, Norgiolini S, Ambrosio G. Risk of cardiovascular disease in relation to achieved office and ambulatory blood pressure control in treated hypertensive subjects. J Am Coll Cardiol. 2002; 39: 878–885.
Clement DL, De Buyzere ML, De Bacquer DA, de Leeuw PW, Duprez DA, Fagard RH, Gheeraert PJ, Missault LH, Braun JJ, Six RO, Van Der Niepen P, O’Brien E, for the Office Versus Ambulatory Pressure Study Investigators. Prognostic value of ambulatory blood pressure in patients with treated hypertension. N Engl J Med. 2003; 348: 2407–2415.
Zweiker R, Eber B, Schumacher M, Toplak H, Klein W. "Non-dipping" related to cardiovascular events in essential hypertensive patients. Acta Med Austriaca. 1994; 21: 86–89.
Bjorklund K, Lind L, Zethelius B, Andren B, Lithell H. Isolated ambulatory hypertension predicts cardiovascular morbidity in elderly men. Circulation. 2003; 107: 1297–1302.
Yamamoto Y, Akiguchi I, Oiwa K, Hayashi M, Kimura J. Adverse effect of nighttime blood pressure on the outcome of lacunar infarct patients. Stroke. 1998; 29: 570–576.
Kario K, Pickering TG, Matsuo T, Hoshide S, Schwartz JE, Shimada K. Stroke prognosis and abnormal nocturnal blood pressure falls in older hypertensives. Hypertension. 2001; 38: 852–857.
Verdecchia P, Porcellati C, Schillaci G, Borgioni C, Ciucci A, Battistelli M, Guerrieri M, Gatteschi C, Zampi I, Santucci A. Ambulatory blood pressure: an independent predictor of prognosis in essential hypertension. Hypertension. 1994; 24: 793–801.
Bobrie G, Chatellier G, Genes N, Clerson P, Vaur L, Vaisse B, Menard J, Mallion JM. Cardiovascular prognosis of "masked hypertension" detected by blood pressure self-measurement in elderly treated hypertensive patients. JAMA. 2004; 291: 1342–1349.
Kikuya M, Hozawa A, Ohokubo T, Tsuji I, Michimata M, Matsubara M, Ota M, Nagai K, Araki T, Satoh H, Ito S, Hisamichi S, Imai Y. Prognostic significance of blood pressure and heart rate variabilities: the Ohasama Study. Hypertension. 2000; 36: 901–906.
Ohkubo T, Hozawa A, Nagai K, Kikuya M, Tsuji I, Ito S, Satoh H, Hisamichi S, Imai Y. Prediction of stroke by ambulatory blood pressure monitoring versus screening blood pressure measurement in a general population: the Ohasama Study. J Hypertens. 2000; 18: 847–854.
Imai Y, Ohokubo T, Sakuma M, Tsuji II, Satoh H, Nagai K, Hisamichi S, Abe K. Predictive power of screening blood pressure, ambulatory blood pressure and blood pressure measured at home for overall and cardiovascular mortality: a prospective observation in a cohort from Ohasama, northern Japan. Blood Press Monit. 1996; 1: 251–254.
Imai Y, Hozawa A, Ohkubo T, Tsuji I, Yamaguchi J, Matsubara M, Michimata M, Hashimoto J, Fujiwara T, Nagai K, Kitaoka H, Satoh H, Hisamichi S. Predictive value of automated blood pressure measurement: what can we learn from the Japanese population: the Ohasama Study. Blood Press Monit. 2001; 6: 335–339.
Mancia G, Sega R, Bravi C, De Vito G, Valagussa F, Cesana G, Zanchetti A. Ambulatory blood pressure normality: results from the PAMELA Study. J Hypertens. 1995; 13: 1377–1390.
Sega R, Trocino G, Lanzarotti A, Carugo S, Cesana G, Schiavina R, Valagussa F, Bombelli M, Giannattasio C, Zanchetti A, Mancia G. Alterations of cardiac structure in patients with isolated office, ambulatory, or home hypertension: data from the general population [Pressione Arteriose Monitorate e Loro Associazioni (PAMELA) Study]. Circulation. 2001; 104: 1385–1392.
Hozawa A, Ohkubo T, Nagai K, Kikuya M, Matsubara M, Tsuji I, Ito S, Satoh H, Hisamichi S, Imai Y. Prognosis of isolated systolic and isolated diastolic hypertension as assessed by self-measurement of blood pressure at home: the Ohasama Study. Arch Intern Med. 2000; 160: 3301–3306.
Imay I, Poncelet P, DeBuyzere M, Padfield PL, Van Monfrans GA. Prognostic significance of self-measurements of blood pressure. Blood Press Monit. 2000; 5: 137–143.
WHO MONICA Project Principal Investigators. The World Health Organization MONICA project: a major international collaboration. J Clin Epidemiol. 1988; 41: 105–114.
Anderson RN, Rosenberg HM. Disease classification: measuring the effect of the tenth revision of the International Classification of Diseases on cause-of-death in the United States. Stat Med. 2003; 15: 1551–1570.
Groppelli A, Omboni S, Parati G, Mancia G. Evaluation of non-invasive blood pressure monitoring devices Spacelabs 90202 and 90207 versus resting and ambulatory 24 hour intraarterial blood pressure. Hypertension. 1992; 20: 227–232.
Marubini E, Valsecchi MG. Analyzing Survival Data From Clinical Trials and Observational Studies. Chichester, UK: J. Wiley & Sons; 1995.
Scettini C, Bianchi M, Nieto F, Sandoya E, Senra H. Ambulatory blood pressure: normality and comparison with other measurements: Hypertension Working Group. Hypertension. 1999; 34: 818–825.
Staessen JA, O’Brien ET, Amery AK, Atkins N, Baumgart P, De Cort P, Degaute JP, Dolenc P, De Gaudemaris R, Enstrom I. Ambulatory blood pressure in normotensive and hypertensive subjects: results from an international database. J Hypertens. 1994; 12 (suppl 7): S1–S12.
Wiinberg N, Hoegholm A, Christensen HR, Bang LE, Mikkelsen KL, Nielsen PE, Svendsen TL, Kampmann JP, Madsen NH, Bentzon MW. 24 hour ambulatory blood pressure in 352 normal Danish subjects, related to age and gender. Am J Hypertens. 1995; 8: 978–986.
Mancia G, O’Brien E, Imai Y, Redon J. Task Force II: Ambulatory blood pressure monitoring in population studies. Blood Press Monit. 1999; 4: 295–301.
Mancia G, Sega R, Grassi G, Cesana G, Zanchetti A. Defining ambulatory and home blood pressure normality: further considerations based on data from the PAMELA Study. J Hypertens. 2001; 19: 995–999.
Guidelines Committee. 2003 European Society of Hypertension–European Society of Cardiology guidelines for the management of arterial hypertension. J Hypertens. 2003; 21: 1011–1053.
Franklin SS, Khan SA, Wong ND, Larson MG, Levy D. Is pulse pressure useful in predicting risk for coronary heart disease; The Framingham Heart Study. Circulation. 1999; 100: 354–360.
Domanski M, Mitchell G, Pfeffer M, Neaton JD, Norman J, Svendsen K, Grimm R, Cohen J, Stamler J, for the MRFIT Research Group. Pulse pressure and cardiovascular disease-related mortality: follow-up of the Multiple Risk Factor Intervention Trial (MRFIT). JAMA. 2002; 287: 2677–2683.
Sander D, Kukla C, Klingelhofer J, Winbeck K, Conrad B. Relationship between circadian blood pressure patterns and progression of early carotid atherosclerosis: a 3-year follow-up study. Circulation. 2000; 102: 1536–1541.
Mancia G, Ferrari A, Gregorini L, Parati G, Pomidossi G, Bertinieri G, Grassi G, di Rienzo M, Pedotti A, Zanchetti A. Blood pressure and heart rate variabilities in normotensive and hypertensive human beings. Circ Res. 1983; 53: 96–104.
Palatini P, Julius S. Heart rate and the cardiovascular risk. J Hypertens. 1997; 15: 3–17.(Roberto Sega, MD; Rita Fa)