AJH 2000;13: by the American Journal of Hypertension, Ltd /00/$20.00

Transcription

1 AJH 2000;13: Comparison of Three Blood Pressure Measurement Methods for the Evaluation of Two Antihypertensive Drugs: Feasibility, Agreement, and Reproducibility of Blood Pressure Response Stéphanie Ragot, Nathale Genès, Laurent Vaur, and Daniel Herpin Our objective was to compare three different methods of blood pressure measurement through the results of a controlled study aimed at comparing the antihypertensive effects of trandolapril and losartan. Two hundred and twenty-nine hypertensive patients were randomized in a double-blind parallel group study. After a 3-week placebo period, they received either 2 mg trandolapril or 50 mg losartan once daily for 6 weeks. At the end of both placebo and active treatment periods, three methods of blood pressure measurement were used: a) office blood pressure (three consecutive measurements); b) home self blood pressure measurements (SBPM), consisting of three consecutive measurements performed at home in the morning and in the evening for 7 consecutive days; and c) ambulatory blood pressure measurements (ABPM), 24-h BP recordings with three measurements per hour. Of the 229 patients, 199 (87%) performed at least 12 valid SBPM measurements during both placebo and treatment periods, whereas only 160 (70%) performed good quality 24-h ABPM recordings during both periods (P <.0001). One hundredforty patients performed the three methods of measurement well. At baseline and with treatment, agreement between office measurements and ABPM or SBPM was weak. Conversely, there was a good agreement between ABPM and SBPM. The mean difference (SBP/DBP) between ABPM and SBPM was / at baseline and / at the end of the treatment period. The correlation between SBPM and ABPM expressed by the r coefficient and the P values were the following: at baseline 0.79/0.70 (< 0.001/<.0001), with active treatment 0.74/0.69 (0.0001/.0001). Hourly and 24-h reproducibility of blood pressure response was quantified by the standard deviation of BP response. Compared with office blood pressure, both global and hourly SBPM responses exhibited a lower standard deviation. Hourly reproducibility of SBPM response (10.8 mm Hg/6.9 mm Hg) was lower than hourly reproducibility of ABPM response (15.6 mm Hg/11.9 mm Hg). In conclusion, SBPM was easier to perform than ABPM. There was a good agreement between these two methods whereas concordance between SBPM or ABPM and office measurements was weak. As hourly reproducibility of SBPM response is better than reproducibility of both hourly ABPM and office BP response, SBPM seems to be the most appropriate method for evaluating residual antihypertensive effect. Am J Hypertens 2000; 13: American Journal of Hypertension, Ltd. KEY WORDS: Blood pressure measurement, reproducibility, trandolapril, losartan. Received May 7, Accepted October 25, From CHU la Milétrie, Service de Cardiologie, Poitiers; and HMR France, Medical Department, Paris la Défense, France. Address correspondence and reprint requests to Dr. Nathalie Genès, HMR France, Tour HMR, 1 Terrasse Bellini Paris la Défense Cedex, France by the American Journal of Hypertension, Ltd /00/$20.00 Published by Elsevier Science, Inc. PII S (99)

2 AJH JUNE 2000 VOL. 13, NO. 6, PART 1 BP TO ASSESS ANTIHYPERTENSIVE EFFICACY 633 The optimal technique to measure blood pressure (BP) remains controversial. Officebased blood pressure recorded by physicians using mercury sphygmomanometers has long been the clinical standard. Indeed, it has been used to describe the relationship between blood pressure and cardiovascular morbidity and mortality 1 and to establish the efficacy of antihypertensive drug therapy. 2 Nevertheless, office blood pressure has several limitations including inaccurate BP level estimates, a white coat effect leading to misclassification of hypertensive status, and poor prediction of target organ damage. 3,4 These concerns have prompted efforts to validate alternative BP measurement methods outside the medical setting. Twenty-four hour ambulatory blood pressure monitoring (ABPM) and home self blood pressure measurement (SBPM) have been proposed. 5 Previous reports have pointed out that both techniques are more closely correlated than office BP to left ventricular mass and urinary albumin excretion. 6 9 In addition, it has been reported that the high number of measures accounted for higher reproducibility in determining individual BP levels or individual BP changes. 4,10 The objective of the present study was to compare the three methods of BP measurements, ie, office BP, ABPM, and SBPM, through the results of a controlled study aimed at comparing the effects of the angiotensin-converting enzyme (ACE) inhibitor trandolapril with those of the angiotensin II antagonist losartan. MATERIALS AND METHODS Patients Men and women older than age 18 years with mild to moderate hypertension were eligible to enter the study. At the end of a 3-week placebo run-in period, office diastolic blood pressure (DBP) had to be confirmed within the 95 to 114 mm Hg range. The main exclusion criteria were any serious chronic disease, known hypersensitivity to ACE inhibitors or angiotensin II antagonists, and severe renal failure. No antihypertensive treatment other than the studied medications (ie, trandolapril or losartan) was allowed during the study. The protocol was approved by the Comité Consultatif de Protection des Personnes dans la Recherche Biomédicale of Poitou-Charentes, France and all patients gave their written informed consent. Study Design This was a double-blind, randomized, parallel group study. After the 3-week single-blind placebo run-in period, patients were randomly allocated to either 2 mg trandolapril or 50 mg losartan once daily in the morning over a 6-week period. Patients were instructed to take the medication between 8 am and 10 am. Office BP Measurements Office BP was measured both at the end of the placebo period, to check that the inclusion criteria had been met, and at the end of the treatment period. Physicians were recommended to perform, in the morning between 8 am and 10 am, before drug intake, three consecutive measurements in the sitting position after a 5-min rest using a mercury sphygmomanometer. Systolic blood pressure (SBP) was defined by phase I and DBP by phase V of the Korotkoff s sounds. The mean of the measurements was considered for office BP analysis. Home Self Blood Pressure Measurements (SBPM) Home SBPM were performed during the last week of both the placebo period and the treatment period. For each period, SBPM were planned for 7 consecutive days. Each day a series of three consecutive measurements was requested in the morning (8 am) before drug intake and repeated in the evening (8 pm). Recordings were performed at home in the sitting position after a 5-min rest using a printer-equipped semiautomatic digitized device based on the oscillometric method (A & D, UA 751, Milpitas, CA). It had previously been validated by carrying out a comparison with a random mercury sphygmomanometer. 11 The first day of each period was considered as training and the correspondent values were not analyzed. The following values were considered as irrelevant and discarded: DBP 40 or 150 mm Hg, SBP 70 or 250 mm Hg, (SBP DBP) 10 mm Hg, measurements performed outside predefined morning (4 am 12am) or evening times (4 pm 12pm). Data for patients who performed at least six measurements in the morning and six measurements in the evening during each period were analyzed. Individual home blood pressure was defined for each period as the mean of all available values. The mean of morning measurements was considered for morning SBPM, the mean of evening measurements for evening SBPM, and the mean of all measurements for global SBPM. Morning SBPM occurred 24 h after the previous drug intake and measured the trough effect. Evening SBPM occurred 12 h after the drug intake when the full effect of the drug was expected. The ratio of morning SBPM decrease over evening SBPM decrease has been calculated for each treatment. This ratio can be considered as a substitute for the trough-to-peak ratio for the evaluation of the time-effect profile of a drug. 12 Ambulatory BP Measurements (ABPM) Ambulatory 24-h blood pressure recordings were performed at the end of both the placebo period and the treatment period. Ambulatory recordings were made using the SpaceLabs (90207; Redmond, WA) ambulatory blood pressure monitor with blood pressure being measured every 20 min. All recordings had to begin between 8 am and 10 am. Recordings beginning after

3 634 RAGOT ET AL AJH JUNE 2000 VOL. 13, NO. 6, PART 1 TABLE 1. FEASIBILITY OF THE THREE METHODS OF BP MEASUREMENT: NUMBER OF PATIENTS WITH VALID MEASUREMENTS AT BOTH D0 AND D42 ACCORDING TO EACH METHOD Losartan Trandolapril Total Overall population Office BP population (98%) SBPM population (87%) ABPM population (70%) Office BP SBPM ABPM population (61%) BP blood pressure; SBPM home self blood pressure measurement; ABPM ambulatory blood pressure measurement. 12 am were discarded, as were recordings with a number of measurements less than 50% of the maximum requested measurements. The same values as those defined for SBPM were considered invalid and were discarded from analysis. Mean hourly BP was calculated as the average of the measurements taken within the time range. Recordings with two consecutive hourly mean values missing were discarded. Statistical Analysis Data are shown as mean standard deviation. The between-group comparisons were done using the Student s t test. A comparison of BP recorded by physicians in the office, by patients at home, and by ABPM was done according to the method of Bland and Altman. 13 In this method the average of measurements evaluated by two different methods (ie, office and SBPM) is plotted against their difference for both SBP and DBP. Correlation coefficients were also calculated according to the leastsquares method. The reproducibility of BP response obtained by the three methods of BP measurement was analyzed. Hourly and 24-h reproducibility of BP response were quantified by the standard deviation of the mean difference between baseline and on-drug evaluation. As office BP was measured at a single time point, hourly reproducibility and 24-h reproducibility were confounded for this method. For SBPM, hourly reproducibility was drawn from the mean of evening and morning SBPM responses whereas 24-h reproducibility was drawn from the global SBPM response. For ABPM, hourly reproducibility was drawn from the mean of each hourly ABPM response and 24-h reproducibility was drawn from the global 24-h ABPM response. Statistical significance was achieved when P.05. RESULTS Patients and Feasibility of the Three Methods of BP Measurement (Table 1) Two hundred and twentynine patients took part in the study. There were 128 men (56%) and 101 women (44%); mean age was years; 45% had never been previously treated for hypertension. The mean duration of hypertension was 5 7 years. Their baseline office BP level was /100 4mmHg. Four patients were withdrawn from office BP analysis due to major protocol violations. Office BP analysis was then performed in 225 patients (98%). One hundred and ninety-nine patients (87%) performed at least 12 valid home blood pressure measurements during the placebo period and the treatment period. The remaining 30 patients provided fewer than the requested measurements. These 199 patients were analyzed for SBPM efficacy. The average number of valid measurements was 33 3 during the placebo period and 32 4 during the treatment period. Time of morning BP measurement was 7:38 am ( 0.48) during the placebo period and 7:40 am ( 0.52) during the treatment period. The evening BP measurement was performed at 7:25 pm ( 0.48) during the placebo phase and at 7:37 pm ( 1.05) during the treatment phase. One hundred and sixty patients (70%) performed a good-quality 24-h ABPM both during the placebo and the treatment period. The average number of valid measurements was 67 6 during the placebo period and 65 6 during the treatment period. The remaining 69 patients were withdrawn from ABPM analysis either for poor quality of at least one of the two recordings or for beginning the recording later than 12 am. However, as the devices were often put on early in the morning, many 24-h recordings finished around 7 am the following day. Therefore, we were not in a position to satisfactorily analyze the trough effect of the drugs with this technique. The number of patients performing SBPM well was significantly higher than that of patients performing ABPM well (P.0001). Only 140 patients (61%) performed all of the three methods of BP measurements well. Comparative Effects of Trandolapril and Losartan The antihypertensive effects of trandolapril and losartan assessed by the three methods of BP measurement are depicted in Table 2. Whatever the method used, blood pressure significantly decreased during the 6-week treatment. No significant difference between the two treatment groups was observed. Global SBPM decrease, which was the main efficacy criterion, was /5.2 6 mm Hg in the trandolapril group and 8.8 9/4.3 6 mm Hg in the losartan group (0.58/ 0.30). Evening SBPM decrease was / mm Hg in the trandolapril group and / mm Hg in the losartan group (0.92/ 0.91). Morning SBPM decrease was / mm Hg in the trandolapril group and / mm Hg in the losartan group (0.35/0.07). The

4 AJH JUNE 2000 VOL. 13, NO. 6, PART 1 BP TO ASSESS ANTIHYPERTENSIVE EFFICACY 635 TABLE 2. ANTIHYPERTENSIVE EFFECTS OF LOSARTAN AND TRANDOLAPRIL ACCORDING TO THE THREE METHODS OF BLOOD PRESSURE (BP) MEASUREMENT Losartan Trandolapril P Value Office BP (n 225) SBP/DBP baseline (mm Hg) / / /.93 SBP/DBP treatment (mm Hg) / / /.19 SBP/DBP decrease (mm Hg) / / /.20 SBPM (n 199) SBP/DBP baseline (mm Hg) / / /.20 SBP/DBP treatment (mm Hg) / / /.47 SBP/DBP decrease (mm Hg) 8.8 9/ / /.30 ABPM (n 160) SBP/DBP baseline (mm Hg) / / /.49 SBP/DBP treatment (mm Hg) / / /.52 SBP/DBP decrease (mm Hg) 7.4 9/ / /.96 SBP systolic blood pressure; DBP diastolic blood pressure; SBPM home self blood pressure measurement; ABPM ambulatory blood pressure measurement. ratios of morning BP decrease to evening BP decrease were 72%/79% for trandolapril and 61%/51% for losartan. Twenty-one patients exhibited an adverse event in the trandolapril group (including four cases of coughing) and 10 patients did so in the losartan group (including two cases of coughing). Agreement Between the Three Methods of BP Measurement Table 3 summarizes the agreement between the three methods of measurement according to the Bland and Altman method. At baseline there was a lack of agreement between office blood pressure measurements and SBPM. For SBP the mean difference was 17.5 mm Hg and the 95% confidence interval of the difference for the population was 49.1/ 14.1 mm Hg. The same trend was observed between office BP measurements and ABPM (mean difference: 22.1 mm Hg and 95% confidence interval: 48.5/4.3 mm Hg). Conversely the agreement was better between SBPM and ABPM, with a mean difference of only 4.6 mm Hg and a 95% confidence interval of the difference for the population of 15.7/25.0 mm Hg (Figures 1, 2, 3). Similar results were observed for DBP at baseline and for SBP and DBP at the end of the treatment period. Using the least-squares method, there was, at baseline, a weak but nonetheless statistically significant linear correlation between office blood pressure measurements and SBPM (r 0.46, P.001 for SBP and r 0.31, P.002 for DBP). A similar pattern was observed between office blood pressure measurements and ABPM (r 0.48, P.0001/r 0.23, P.006), whereas the correlation between SBPM and ABPM was stronger (r 0.79, P.0001/r 0.70, P.0001). At the end of the treatment period (D42), the same trend was observed, ie, a weak correlation between office blood pressure measurements and either SBPM (r 0.42, P.0001/r 0.44, P.0001) or ABPM TABLE 3. AGREEMENT BETWEEN THE THREE METHODS OF BP MEASUREMENT ACCORDING TO THE BLAND AND ALTMAN METHOD SBP (D0) DBP (D0) SBP (D42) DBP (D42) SBPM v office Mean difference Limits of agreement (mean 2 SD) 49.1/ / / /18.2 ABPM v office Mean difference Limits of agreement (mean SD) 48.5/ / / /12.5 SBPM v ABPM Mean difference Limits of agreement (mean 2 SD) 15.7/ / / /17/9 Abbreviations as in Table 2.

5 636 RAGOT ET AL AJH JUNE 2000 VOL. 13, NO. 6, PART 1 FIGURE 1. Comparison of systolic home self blood pressure measurements (SBPM) and systolic office blood pressure measurements at baseline according to the method of Bland and Altman. 13 Mean blood pressure was obtained by averaging pressure measured at home and by physician and the difference was calculated as pressure measured at home minus pressure measured by physician. measurements (r 0.43, P.0001/r 0.48, P.0001) and a stronger correlation between SBPM and ABPM (r 0.74, P.0001/r 0.69, P.0001). Correlations between SBP and DBP responses according to each technique are shown in Table 4. There were linear correlations between the three evaluation methods. The stronger correlation was observed between SBPM and ABPM responses (r 0.40, P.0001 for SBP and r 0.35, P for DBP). FIGURE 3. Comparison of systolic home self blood pressure measurements (SBPM) and systolic ambulatory blood pressure measurements (ABPM) at baseline according to the method of Bland and Altman. 13 Mean blood pressure was obtained by averaging pressure measured at home and by ambulatory device and the difference was calculated as pressure measured at home minus pressure measured by ambulatory device. Reproducibility of BP Response Evaluated by the Three Methods of BP Measurement Table 5 summarizes the hourly reproducibility and the 24-h reproducibility of the blood pressure response evaluated by the three methods of BP measurement. Compared with office blood pressure, both global and hourly SBPM responses exhibited a lower standard deviation. The division by two of the number of measurements between global SBPM and hourly SBPM (ie, morning or evening SBPM) did not affect the standard deviation of this method. As far as ABPM is concerned, results were quite different: 24-h ABPM response reproducibility was as good as global SBPM response reproducibility, whereas hourly ABPM response reproducibility was worse than office BP response reproducibility. DISCUSSION The initial objective of the present study was to compare the antihypertensive effects of the angiotensin II antagonist losartan with those of the ACE inhibitor trandolapril in hypertensive patients. Although acting on the same physiopathologic pathway, ie, the renin FIGURE 2. Comparison of systolic ambulatory blood pressure measurements (ABPM) and systolic office blood pressure measurements at baseline according to the method of Bland and Altman. 13 Mean blood pressure was obtained by averaging the pressure measured by the ambulatory device and by physician and the difference was calculated as pressure measured by ambulatory device minus pressure measured by physician. TABLE 4. CORRELATIONS BETWEEN OFFICE BP DECREASE, SBPM DECREASE, AND AMBULATORY BP DECREASE AFTER A 6-WEEK TREATMENT BY EITHER TRANDOLAPRIL OR LOSARTAN (n 140) DBP decrease SBP decrease r P r P Office v SBPM Office v ABPM 0.13 NS SBPM v ABPM Abbreviations as in Table 2.

6 AJH JUNE 2000 VOL. 13, NO. 6, PART 1 BP TO ASSESS ANTIHYPERTENSIVE EFFICACY 637 TABLE 5. REPRODUCIBILITY OF BP RESPONSE EVALUATED BY THE THREE METHODS OF MEASUREMENT AFTER A 6-WEEK TREATMENT BY EITHER TRANDOLAPRIL OR LOSARTAN (n 140) Hourly BP Response 24-h BP Response Number of BP Measures (Median) BP Response (mm Hg) Mean SD (mm Hg) Number of BP Measures (Median) BP Response (mm Hg) Mean SD (mm Hg) DBP Office SBPM ABPM SBP Office SBPM ABPM Abbreviations as in Table 2. angiotensin aldosterone system (RAAS), these two drug categories have potential differences because of different action mechanisms. On the one hand, ACE inhibitors may act by the prevention of the breakdown of bradykinin. 14 On the other hand, a more complete blockade of the RAAS may be achieved by angiotensin II antagonists; it has been suggested that significant amounts of angiotensin II are also generated by non angiotensin-converting enzymes, such as chymases. 15 In this study, as in previous studies, we found no differences in terms of overall efficacy between the angiotensin II antagonist and the ACE inhibitor A trend towards a more sustained trough effect was seen with the SBPM technique in the trandolapril group, compatible with the well-established long duration of action of this drug. 19,20 Unfortunately, due to a large number of missing values during the last2hof the recordings, we were not in a position to confirm or to deny this trend with ABPM. This trial provides interesting methodologic findings. Indeed, studies aimed at comparing office sphygmomanometry, ABPM, and home SBPM are infrequent. 21 The feasibility of both SBPM and ABPM has long been questioned. On the one hand, SBPM protocol requires the patient to perform three measurements in the morning and in the evening over several consecutive days. On the other hand, ABPM has the disadvantage of imposing a BP measurement every min day and night during 24 h. It has been said that SBPM technique needs a greater physician involvement and a selection of appropriate patients to comply with the SBPM protocol. In a large-scale, 1700-patient study performed in general practice, 65% of the patients correctly performed baseline SBPM whereas only 55% of the patients correctly performed both baseline and on-drug SBPM. 22 However, in carefully monitored studies involving trained investigators, the feasibility was better, reaching 80% of patients. 20,23 In the present study, 87% of patients satisfactorily performed two series of SBPM. This success rate was higher than the 70% obtained with the ABPM technique. Therefore, the SBPM protocol seems easier to achieve than 24-h ABPM recordings. Obviously, office BP, SBPM, and ABPM do not measure the same blood pressure. In this study, betweenmethod correlations, although statistically significant, were far from being perfect. These kinds of results have been previously reported. 5,24,25 The weakest correlations were those involving baseline office diastolic blood pressure. This is probably due to the truncated nature of this variable, which had to be within the 95 to 114 mm Hg range. The strongest correlations were found between SBPM levels and 24-h ABPM levels. In addition, average SBPM levels and ABPM levels were very close, whereas office BP levels were significantly higher. As far as BP responses to therapy are concerned, between-method correlations were much weaker: the summation of baseline and on-drug variabilities accounted for these poor results. Here again, the strongest correlations were found between SBPM and ABPM evaluations This confirms the results of a previous study, 21 where the agreement between SBPM and ABPM responses (r 0.6, P.005) was much greater, compared with the relationship between casual and ABPM responses (r 0.4, P.05). As is discussed later, the superiority of SBPM and 24-h ABPM over office BP might be related both to the higher precision of the former two methods and to the assessment of the measurements outside the medical setting, which avoids any white coat effect. One useful mathematical tool to compare the reproducibility of different methods is the standard deviation of a considered parameter. The smaller the standard deviation, the narrower the distribution of the individual responses and then the greater the repro-

7 638 RAGOT ET AL AJH JUNE 2000 VOL. 13, NO. 6, PART 1 ducibility of the method used. In this study, the reproducibility of the BP responses measured by office sphygmomanometry, global SBPM, and 24-h ABPM were compared. As shown in Table 5, the standard deviation of either global SBPM or 24-h ABPM was 30% smaller than that of office sphygmomanometry. Thus, these two former techniques should be preferred when assessment of overall drug efficacy is needed. We also compared the reproducibility of the blood pressure response evaluated by the three methods at a given time point, ie, within 1 prespecified hour. Compared with office BP, SBPM still exhibited a favorable low standard deviation whereas the 1-h standard deviation of ABPM increased dramatically. This limited hourly reproducibility of ABPM, compared with 24-h reproducibility, had been underlined by some authors. 26,27 It can be partly explained by the small number of hourly readings. However, the uncontrolled conditions of ABPM recordings may also play a role too; differences in behavioral conditions between baseline and on-drug evaluation may explain some of the intraindividual variability. Indeed, in the present study, the 1-h standard deviation of three ABPM measurements was even worse than that of three office readings. This is of great importance in that antihypertensive therapies are often evaluated at trough to assess their residual efficacy. In this particular context, morning SBPM should be preferred to ABPM. In conclusion, the higher agreement between SBPM and ABPM and the better reproducibility of either SBPM and 24-h ABPM responses suggest that both techniques are superior to office readings for evaluating the overall effect of antihypertensive therapy. However, ABPM reproducibility of BP response, unlike SBPM reproducibility, is deeply compromised when the efficacy is analyzed over hourly periods. Thanks to repeated measurements in standardized conditions at fixed times, SBPM seems the most appropriate method for evaluating the trough effect of an antihypertensive drug. REFERENCES 1. MacMahon S, Peto R, Cutler J, Collins R, Sorlie P, Neaton J, Abott R, Godwin J, Dyer A, Stamler J: Blood pressure, stroke, and coronary heart disease. Part 1. Prolonged differences in blood pressure: prospective observational studies corrected for the regression dilution bias. Lancet 1990;335: Collins R, Peto R, MacMahon S, Hebert P, Fiebach NH, Erberlein KA, Godwin J, Qizilbash N, Taylor JO, Hennekens C: Blood pressure, stroke, and coronary heart disease. Part 2. Short term reduction in blood pressure: overview of randomized drug trial in their epidemiological context. Lancet 1990;335: Pickering TG: The ninth Sir George Pickering memorial lecture. Ambulatory monitoring and the definition of hypertension. J Hypertens 1992;10: Coats AJS, Radaelli A, Clark SJ, Conway J, Sleight P: The influence of ambulatory blood pressure monitoring on the design and interpretation of trials in hypertension. J Hypertens 1992;10: Appel LJ, Stason WB: Ambulatory blood pressure monitoring and blood pressure self-measurement of blood pressure in the diagnosis and management of hypertension. Ann Intern Med 1993;118: Abe H, Yokouchi M, Nagata S, Ashida T, Yoshimi H, Kawano Y, Kojima S, Matsuoka S, Kuramochi M, Omae T: Relation of office and home blood pressure to left ventricular hypertrophy and performance in patients with hypertension. High Blood Press 1992;1: James GD, Pickering TG, Yee LS, Harshfield GA, Riva S, Laragh JH: The reproducibility of average ambulatory, home and clinic pressures. Hypertension 1988;11: Clement DL, De Buyzere M, Duprez D: Prognostic value of ambulatory blood pressure monitoring. J Hypertens 1994;12: Bauduceau B, Genès N, Chamontin B, Vaur L, Renault M, Etienne S, Marre M: Ambulatory blood pressure and urinary albumine excretion in diabetic (non-insulin-dependent and insulin-dependent) hypertensive patients. Am J Hypertens 1998;11: Mengden T, Bättig B, Vetter W: Self-measurement of blood pressure improves the accuracy and reduces the number of subjects in clinical trials. J Hypertens 1991; 9(suppl 6):S336 S Jamieson MJ, Webster J, Witte K, Higgins MM, Mac Donald TM, de Beaux A, Petrie JC: An evaluation of the A & D UA-751 semi-automated cuff-oscillometric sphygmomanometer. J Hypertens 1990;8: Menard J, Chatellier G, Day M, Vaur L: Self measurement of blood pressure at home to evaluate drug effects by the trough:peak ratio. J Hypertens 1994;12:S21 S Bland JM, Altman DG: Statistical methods for assessing agreement between two methods of measurement. Lancet 1986;i: Linz W, Wiemer G, Gohlke P, Unger T, Shölkens BA: Contribution of kinins to the cardiovascular actions of angiotensin-converting enzyme inhibitor. Pharmacol Rev 1995;47: Wood AJJ: Angiotensin receptors and their antagonists. N Engl J Med 1996;334: Tikkanen I, Omvik P, Jensen HAE: Comparison of the angiotensin II antagonist losartan with the angiotensin converting enzyme inhibitor enalapril in patients with essential hypertension. J Hypertens 1995;13: Gradman AH, Arcuri KE, Goldberg AI, Ikeda LS, Nelson EB, Snavely DB, Sweet CS: A randomized placebocontrolled, double-blind parallel study of various doses of losartan potassium compared with enalapril maleate in patients with essential hypertension. Hypertension 1995;25: Black HR, Graff A, Shute D, Stoltz R, Ruff D, Levine J, Shi Y, Mallows S: Valsartan, a new angiotensin II antagonist for the treatment of essential hypertension: efficacy, tolerability and safety compared to an angio-

8 AJH JUNE 2000 VOL. 13, NO. 6, PART 1 BP TO ASSESS ANTIHYPERTENSIVE EFFICACY 639 tensin-converting enzyme inhibitor, lisinopril. J Hum Hypertens 1997;11: Zannad F, Matzinger A, Larché J: Trough/peak ratios of once daily angiotensin converting enzyme inhibitors and calcium antagonists. Am J Hypertens 1996;9: Vaur L, Bobrie G, Dutrey-Dupagne C, Dubroca I, Vaisse B, Bouvier d Yvoire M, Elkik F, Chatellier G, Ménard J: Short-term effects of withdrawing angiotensin converting enzyme inhibitor therapy on home selfmeasured blood pressure in hypertensive patients. Am J Hypertens 1998;11: Mengden T, Weisser B, Vetter B: Ambulatory 24-h blood pressure versus self-measured blood pressure in pharmacologic trials. J Cardiovasc Pharmacol 1994; 24(suppl 2):S20 S Chatellier G, Dutrey-Dupagne C, Vaur L, Zannad F, Genès N, Elkik F, Ménard J: Home self-blood pressure measurement in general practice: the SMART study. Am J Hypertens 1996;9: Bobrie G, Day M, Tugayé A, Ménard J: Self blood pressure measurement at home. Clin Exp Hypertens 1993;15: Zannad F, Vaur L, Dutrey-Dupagne C, Genès N, Chatellier G, Elkik F, Ménard J: Assessment of drug efficacy using home self-blood pressure measurement: the SMART study. J Hum Hypertens 1996;10: Thjis L, Staessen JA, Celis H, de Gaudemaris R, Imai Y, Julius S, Fagard R: Reference values for self-recorded blood presssure. Arch Intern Med 1998;158: Mancia G, Omboni S, Parati G, Trazzi S, Mutti E: Limited reproducibility of hourly blood pressure values obtained by ambulatory blood pressure monitoring: implications for studies on antihypertensive drugs. J Hypertens 1992;10: Staessen J, Thijs L, Mancia G, Parati G, O Brien ET: Clinical trials with ambulatory blood pressure monitoring: fewer patients needed? Lancet 1994;344: