Many interventional trials have demonstrated

AJH 1998;11:165 173 Short-Term Effects of Withdrawing Angiotensin Converting Enzyme Inhibitor Therapy on Home Self-Measured Blood Pressure in Hypertensive Patients Laurent Vaur, Guillaume Bobrie, Claire Dutrey-Dupagne, Isabelle Dubroca, Bernard Vaisse, Michel Bouvier d Yvoire, François Elkik, Gilles Chatellier, and Joël Menard The aim of this study was to compare blood pressure rise after interruption of two angiotensin converting enzyme (ACE) inhibitors in hypertensive patients. After a 2-week placebo run-in period, hypertensive patients were treated with either trandolapril 2 mg once daily or perindopril 4 mg once daily for 4 weeks in a double-blind design. A placebo was then administered for 1 week. Three periods of 1-week home self-measured blood pressure (SMBP) were programmed: end of placebo run-in period, end of treatment period, and final withdrawal placebo period. Every day, three consecutive measurements were requested both in the evening and in the morning. Individual reversion to baseline BP level was studied in the subgroup of patients responding to therapy (evening diastolic SMBP decrease >6 mm Hg). The ratio (R) of mean post-drug DBP lowering (residual effect) over evening on-drug DBP lowering (full effect) was used to study reversion to baseline. Patients exhibiting a lower value than the median of this ratio were called Reverters, whereas others were called Nonreverters. One hundred-nineteen patients entered the analysis. During the treatment period, mean SMBP decreased significantly, from 150 14/ 97 7mmHgto139 15/91 9 mm Hg (all P <.001). The on-drug BP level was similar in the evening in the two treatment groups. However, both systolic and diastolic morning SMBP levels were significantly lower in the trandolapril group. After drug discontinuation, the mean BP level significantly rose to 144 14/94 9 mm Hg (all P.01) but remained lower than the baseline BP values (P.003 for SBP and P.002 for DBP). The post-drug BP level was significantly lower in the trandolapril group than in the perindopril group. Seventy-four patients were responders to therapy. In this subgroup, the median of the R ratio used to analyze reversion to baseline after drug discontinuation was 44%. Nonreverters were characterized by a sustained on-drug BP decrease, compared to Reverters. We therefore conclude that ACE inhibitor treatment withdrawal is accompanied by a rapid rise in BP (within 48 h), followed by a 5-day BP plateau that is lower than the initial level. Reverters to baseline after drug discontinuation were more likely to be insufficiently controlled during therapy, particularly in the morning. The longer duration of action of trandolapril was associated with a lower BP level during both the morning during the active treatment phase and the 1-week posttreatment phase. Am J Hypertens 1998;11:165 173 1998 American Journal of Hypertension, Ltd. KEY WORDS: Home self-measured blood pressure, treatment withdrawal, angiotensin converting enzyme inhibitor therapy. From the Laboratoires Roussel, Direction Médicale, Paris; Service de Néphrologie et Hypertension Artérielle, Centre Hospitalier, Vannes; HMR France, Paris; Centre Hospitalier Régional Universitaire La Timone, Marseille; and Centre d Investigations Cliniques, Hôpital Broussais, Paris, France. Address correspondence and reprint requests to Doctor L. Vaur, Laboratoires Roussel, Direction Médicale, Tour Roussel-Hoechst, 1 Terrasse Bellini, 92910 Paris-La Défense, France. Many interventional trials have demonstrated that antihypertensive agents significantly reduce long-term morbidity and mortality in mild to severe hypertensive patients. 1 3 The reduction in the incidence of stroke has been shown to be the most striking and 1998 by the American Journal of Hypertension, Ltd. 0895-7061/98/$19.00 Published by Elsevier Science, Inc. PII S0895-7061(97)00420-2

166 VAUR ET AL AJH FEBRUARY 1998 VOL. 11, NO. 2 constant benefit. The reduction in the incidence of coronary events was lower than expected, though highly significant. These benefits were proportional to the extent of blood pressure lowering and related to the duration of the treatment period. They were obtained in large-scale trials testing diuretics or -blockers v placebo. Although not demonstrated, it is suggested that the use of other drug classes may lead to at least the same benefits. 4,5 This is the main reason most hypertensive patients who begin receiving medication should expect to be treated on a life-long basis. However, discontinuation of therapy may occur under two circumstances. First, the practitioner may decide to interrupt the medication. This is supported by the hypothesis of resetting the BP control mechanism. Some authors assumed that the carotid sinus barostat mechanism, having been reset at a higher level during periods of hypertension, might be reset at a lower level if BP was consistently reduced. 6 If treatment led to resetting of baroreceptors or any other BP control mechanism, it should be possible to interrupt medication in the expectation of a remission period. Discontinuous rather than continuous treatment may then be possible for at least some patients. Treatment interruption might also identify people in whom BP will remain low for other reasons and for whom further treatment is unnecessary. 6 8 Discontinuation of therapy may also occur because of patient s behavior. 9,10 It has long been recognized that patients often do not wholly comply with the instructions for use of their medication. Protracted intervals between doses may occur, particularly when dosing frequency is greater than once daily. This commonly occurs in the form of drug holidays, where doses are omitted for 1 or more days. 10,11 Withdrawal of antihypertensive therapy and drug holidays raise the question of how and when BP rises after interruption of therapy. 12 The present study was conducted to answer this question by analyzing the course of BP during the week after discontinuation of two ACE inhibitors in hypertensive patients. Home self-measured blood pressure (SMBP) was chosen as the BP evaluation method because standardized measurements repeated during several consecutive days were needed. METHODS Study Design After a 2-week single-blind placebo run-in period, patients entered a 4-week double-blind active treatment period, where they were randomized to receive either 2 mg trandolapril once daily or 4 mg perindopril once daily. This phase was followed by a 1-week single-blind placebo period. Patients were instructed to take their medication at 8:00 am. Compliance and exact time of drug intake were checked by electronic pill-boxes (MEMS; Aprex Ltd, Fremont, CA). The protocol was approved by the Comité Consultatif de Protection des Personnes se Prêtant à la Recherche Biomédicale of Rennes, France. Blood Pressure Measurements Office BP was measured at the end of the placebo run-in phase. Three consecutive measurements were performed by the physician using a mercury sphygmomanometer with the patient in the sitting position after a 5-min rest. The mean of these three measurements was taken for consideration for inclusion of the patient in the study. Home self-measurement of blood pressure (SMBP) was performed during the last week of the placebo run-in period, the last week of the active treatment period, and the week of the placebo post-treatment period. The weeks of SMBP of periods 2 and 3 were consecutive. Every day, a series of three consecutive measurements was requested in the morning before drug intake (8:00 am) and repeated in the evening (8:00 pm). Measurements were performed in the sitting position, after a 5-min rest, using the A&D UA 751 device (A&D Enginnering, Milpitas, CA), which is a printer-equipped, semiautomatic, digitized device based on the oscillometric method. It has been validated previously by comparison to a random-zero mercury sphygmomanometer. 13 Each BP value was printed on a ticket with the date and exact time of measurement. Patients Male and female patients aged 18 70 years with mild to moderate essential hypertension were eligible to enter the study. At the end of the 2-week placebo run-in period, office diastolic blood pressure (DBP) had to be confirmed in the range of 95 119 mm Hg and the median of home self-measured DBP had to be greater than 85 mm Hg. The main exclusion criteria were any serious chronic disease, known hypersensibility or contraindication to ACE inhibitors, and obese arm, impairing the quality of SMBP. Any antihypertensive treatments other than the medications studied were not allowed during the study. Self-Measurement Data Management The first day of the placebo run-in period and the treatment period were considered as training and the corresponding values were not analyzed. The theoretical number of available measurements was therefore 36 during the baseline period (18 in the morning and 18 in the evening), 39 during the active treatment period (21 in the morning and 18 in the evening), and 33 during the post-treatment period (15 in the morning and 18 in the evening). Aberrant values were deleted according to the following rules: DBP 40 or DBP 150 mm Hg, SBP 60 or 250 mm Hg, pulse pressure 10 mm Hg when SBP 110 mm Hg; measurements performed

AJH FEBRUARY 1998 VOL. 11, NO. 2 SMBP AND ACE INHIBITOR WITHDRAWAL 167 outside the predefined morning and evening times (4:00 am to noon for morning SMBP or 4:00 pm to midnight for evening SMBP); or measurements performed outside the predefined intervals (12 4 h after drug intake for evening SMBP or 24 4 h for morning SMBP). The data of patients having at least six morning and six evening validated measurements for each of the three periods were analyzed. Individual home BP was defined for each period as the mean of all available values. The mean of morning measurements was considered for morning SMBP, the mean of evening measurements for evening SMBP, and the mean of all measurements for global SMBP. Analysis of BP Rise After Drug Discontinuation Individual reversion to baseline BP level was studied in the subgroup of patients responding to therapy. As previously described, patients were defined as responders when evening diastolic SMBP decreased by at least 6 mm Hg during the treatment phase, compared to baseline. 14 The ratio (R) of mean post-drug DBP lowering over evening on-drug DBP lowering was used to study the degree of reversion of each patient. Patients exhibiting a lower value than the median of the R ratio were called Reverters to baseline BP levels. Patients with higher value were called Non- Reverters. Characteristics of Reverters and Nonreverters were compared. Statistical Analysis Data are summarized as means standard deviation (SD). The difference between two qualitative variables was tested by the 2 test. The difference between two quantitative variables was tested by the Student s paired or unpaired t test, as appropriate. Analyses of covariance were performed to adjust either on-drug or post-treatment SMBP to baseline SMBP. A P.05 was considered significant. Calculations were performed using the SAS statistical package (SAS institute Inc, Cary, NC). RESULTS Baseline Characteristics of Patients Of the 152 patients entering the study, 33 were excluded from the analysis because of either major protocol deviations (16 patients) or failure to achieve the required minimum number of valid BP self measurements (17 patients). The remaining 119 patients were included in the statistical evaluation. There were 58 men and 61 women, aged 53 10 years. At the end of the 2-week placebo run-in period, mean casual blood pressure was 163 13 mm Hg for SBP and 103 6mmHgfor DBP. During the same period, mean home self-measured blood pressure was 150 14 mm Hg for SBP and 97 7 mm Hg for DBP. The average number of home BP measurements was 34 3 during the placebo run-in period, 35 5 during the treatment period, and 31 4 during the post-treatment period. Morning SMBP occured at 7:53 0:59 am and evening SMBP at 7:50 1:07 pm. There were 62 patients in the perindopril group and 57 patients in the trandolapril group. The groups were well matched with regard to age, gender, weight, casual blood pressure, and home blood pressure upon entry to the study; indeed, no statistical differences between the groups were observed. Compliance assessed by electronic pill-boxes was over 80% in all the 119 patients entering the analysis. Mean compliance was 99.4 4% in the perindopril group and 99.0 4% in the trandolapril group. Drug intake occured at 8:20 1:37 am in the perindopril group and at 8:39 1:39 am in the trandolapril group. Effects on BP Course in the Overall Population Figure 1 and Table 1 refer to home BP levels during the three SMBP periods of the study in the overall population. During the treatment period, the average BP level was 139 15 mm Hg for SBP and 91 9mmHgfor DBP. The BP level adjusted for baseline was lower in the trandolapril group (137 9 mm Hg/90 6mm Hg) than in the perindopril group (141 9mmHg/ 92 6 mm Hg) (Table 2). The difference was statistically significant only for SBP (P.04). As shown in Table 2, the difference in BP level between the two treatment groups reached statistical significance in the morning but not in the evening. After drug discontinuation the mean BP rose consistently for 2 days and then was stable during the following 4 days. The average BP level during the post-treatment phase was 144 14 mm Hg for SBP and 94 9 mm Hg for DBP. These values were significantly higher than the mean on-drug BP levels (P.01 for both SBP and DBP) but lower than the baseline BP levels (P.003 for SBP and.002 for DBP). The adjusted BP level during the post-treatment phase was lower in the trandolapril group (142 9/93 5 mm Hg) than in the perindopril group (146 9mm Hg/95 5 mm Hg). There was a statistical difference between the two groups for both SBP (P.007) and DBP (P.05). Blood Pressure Rise After Drug Discontinuation in Responder Patients: Reverters v Nonreverters Seventy-four of the 119 patients were responders to therapy. The distribution of the ratio (R) of mean post-drug DBP lowering over evening on-drug DBP lowering is shown in Figure 2. Its median was 44%. Residual BP lowering of Reverters was close to zero; their median R ratio was 11% for SBP and 15% for DBP. In the Nonreverter group, BP lowering was consistently sustained despite withdrawing therapy; their median R ratio was 66% for SBP and 75% for DBP (Table 3). The BP rise after drug discontinuation in Reverters

168 VAUR ET AL AJH FEBRUARY 1998 VOL. 11, NO. 2 FIGURE 1. Home self-measured blood pressure (SMBP) levels during the three SMBP periods of the study (n 119). D0, first intake of active drug; D28, last intake of active drug. The first plot of each day corresponds to the mean morning SMBP; the second plot of each day corresponds to the mean evening SMBP. and Nonreverters is shown in Figure 3. The zero line corresponds to the baseline SMBP level. The plots represent the difference between this level and each post-treatment SMBP evaluation over time. They fit well with an exponential function whose general formula is: y y 0 (y y 0 )(1 e t ) where y 0 is the difference between baseline SMBP and last evening on-drug SMBP, y the plateau level, and the time constant. The resulting equations for Reverters and Nonreverters are given in Figure 3. The plateau level (y ) is close to mean baseline SMBP level in Reverters (1.4 mm Hg for SBP and 0.5 mm Hg for DBP), whereas it is consistently lower in Nonreverters (12.8 mm Hg for SBP and 8.4 mm Hg for DBP). Six days after withdrawal, the proportion of therapeutic effect (y /y 0 ) still present was 9% (SBP) and 5% (DBP) in Reverters v 55% (SBP) and 59% (DBP) in Nonreverters. Characteristics of Reverters and Nonreverters are depicted in Table 2. The differences in residual effects between Reverters and Nonreverters are summarized. As shown, these differences seem to be linked to the extent of on-drug BP lowering, particularly in the morning. TABLE 1. HOME BP COURSE OVER THE THREE PERIODS Baseline Period (SMBP [mm Hg]) SMBP (mm Hg) Drug Period PvBaseline Placebo PostTreatment Period SMBP (mm Hg) PvBaseline SBP Morning 148 16 140 16.0002 143 15.02 Evening 151 15 138 15.0001 145 15.001 Global 150 14 139 15.0001 144 14.003 DBP Morning 97 8 92 10.0001 94 9.01 Evening 97 8 89 9.0001 93 10.002 Global 97 7 91 9.0001 94 9.002 Abbreviations: SMBP, home self-measured blood pressure; SBP, systolic blood pressure; DBP, diastolic blood pressure.

AJH FEBRUARY 1998 VOL. 11, NO. 2 SMBP AND ACE INHIBITOR WITHDRAWAL 169 TABLE 2. BASELINE AND ON-DRUG HOME SMBP IN THE TWO TREATMENT GROUPS Baseline SMBP Unadjusted On-Drug SMBP On-Drug SMBP Adjusted for Baseline SMBP P Group T Group P P Group T Group P P Group T Group P Morning SBP 149 17 147 15 NS 143 16 137 16.05 142 10 138 10.02 DBP 98 9 96 8 NS 94 10 90 10.03 93 7 91 7.04 Evening SBP 151 15 151 15 NS 139 14 137 16 NS 139 11 137 11 NS DBP 97 9 97 7 NS 90 9 89 9 NS 90 7 89 7 NS Mean SBP 150 15 149 14 NS 141 14 137 15 NS 141 9 137 9.04 DBP 97 8 96 7 NS 92 9 89 9 NS 92 6 90 6 NS Abbreviations: SMBP, home self-measured blood pressure; T, trandolapril; P, perindopril; SBP, systolic blood pressure; DBP, diastolic blood pressure. Blood Pressure Rise After Drug Discontinuation in Responder Patients: Trandolapril v Perindopril After drug discontinuation in responder patients, the residual SBP/DBP lowering was greater in the trandolapril group (10.5 9.5/6.2 5.6 mm Hg) than in the perindopril group (5.4 8.2/4.1 3.8 mm Hg). These differences were statistically significant for both SBP (P.01) and DBP (P.04). The BP rise after discontinuation of either drug has also been modeled, as shown in Figure 4. The plateau level (y ) is closer to the mean baseline SMBP level in the perindopril group (3.0 mm Hg for SBP and 1.9 mm Hg for DBP) than in the trandolapril group (7.5 mm Hg for SBP and 4.0 mm Hg for DBP). The proportion of therapeutic effect (y /y 0 ) still present 6 days after drug discontinuation was 23% (for both SBP and DBP) in the perindopril group v 49% (SBP) and 45% (DBP) in the trandolapril group. Characteristics of the two groups are depicted in Table 4. As shown, a significant difference in BP lowering was also present during therapy. In the evening, 12 h after dosing, the difference between the two groups was mild and only significant for DBP lowering (P.05). In the morning, 24 h after dosing, the difference was pronounced and significant for both SBP (P.03) and DBP (P.01). DISCUSSION Many studies have been designed to examine the long-term outcome among patients who stopped tak- FIGURE 2. Classification of Reverters and Nonreverters according to the individual degree of reversion to baseline BP levels. The R ratio is defined by the ratio of residual BP lowering (mean baseline SMBP level mean posttreatment BP level) over evening on-drug BP lowering.

170 VAUR ET AL AJH FEBRUARY 1998 VOL. 11, NO. 2 TABLE 3. CHARACTERISTICS OF REVERTERS AND NONREVERTERS AMONG RESPONDER PATIENTS Reverters (n 37) Nonreverters (n 37) P Residual effect after drug discontinuation Residual SBP lowering* (mm Hg) 2.5 7.2 13.1 7.7 NA R ratio for SBP (median in %) 15 66 y (mm Hg) 1.4 12.8 Residual DBP lowering* (mm Hg) 1.5 3.2 8.6 3.2 NA R ratio for DBP (median in %) 11 75 y (mm Hg) 0.5 8.4 Explanatory variables Age (years) 50 11 52 10 NS Male gender (%) 51 41 NS Patients receiving perindopril (%) 59 46 NS Baseline SBP/DBP (mm Hg) 148 15/99 10 151 15/97 6 NS On-drug evening SBP/DBP lowering (mm Hg) 17 9/12 5 20 11/12 4 NS On-drug morning SBP/DBP lowering (mm Hg) 7 9/5 6 15 10/9 5.001/.002 On-drug global SBP/DBP lowering (mm Hg) 12 8/8 5 17 10/11 4.01/.02 Abbreviations: SBP, systolic blood pressure; DBP, diastolic blood pressure; NA, not applicable (the statistical comparison would be incorrect as the two groups are issued from the same truncated sample); NS, not significant. * Residual BP lowering, difference between mean baseline BP level and mean post-treatment BP level. R ratio, ratio of residual BP lowering over evening on-drug BP lowering. y, plateau level assessed by exponential modeling. ing chronic antihypertensive medication and were normotensive at the time of such cessation. 6,7,8,15 In these studies, the observation period lasted from 1 to 20 years (in the Framingham study 8 ) and casual BP measurements were taken every 1 to 6 months. They showed that the proportion of subjects remaining in initial remission ranged from 20% to 80% at 1 year. This proportion was then approximately halved every 2 years and no remission that exceeded 14 years was observed. One might consider that long-term remissions appear surprisingly frequent. In fact, these figures are FIGURE 3. Evolution of BP in responders after treatment discontinuation, in Reverters v Nonreverters. Formulas of exponential functions are Reverters: y 10.8 10.3 (1 exp ( 0.044 t)); Nonreverters: y 14.2 5.7 (1 exp ( 0.082 t)).

AJH FEBRUARY 1998 VOL. 11, NO. 2 SMBP AND ACE INHIBITOR WITHDRAWAL 171 FIGURE 4. Evaluation of BP in responders after treatment discontinuation, in trandolapril v perindopril. Formulas of exponential functions are trandolapril: y 13.4 7.6 (1 exp ( 0.051 t)); perindopril: y 11.4 7.8 (1 exp ( 0.054 t)). partly explained both by a regression to the mean phenomenon and by errors in the initial diagnosis of definite hypertension. Alternatively, Korner et al 16 have shown that BP took longer to recover hypertensive levels after withdrawal from prolonged treatment. The normalization of hypertension-induced vascular remodeling obtained after prolonged treatment with some drugs such as ACE inhibitors may explain some cases of sustained remission. Indeed, redevelopment of hypertension was much slower after obtaining regression of both vascular hypertrophy and left ventricular hypertrophy. 16 There appeared to be no dependable way to distinguish in advance Reverters to previous hypertensive BP levels from TABLE 4. EFFECTS OF TRANDOLAPRIL AND PERINDOPRIL ON BLOOD PRESSURE RISE AFTER WITHDRAWAL AMONG RESPONDER PATIENTS Trandolapril (n 35) Perindopril (n 39) P Residual effect after drug discontinuation Residual SBP lowering* (mm Hg) 10.5 9.5 5.4 8.2.01 R ratio for SBP (median in %) 55 33 y (mm Hg) 9.7 5.1 Residual DBP lowering* (mm Hg) 6.2 5.6 4.1 3.8.04 R ratio for DBP (median in %) 59 42 y (mm Hg) 5.8 3.6 Explanatory variables Age (years) 50 9 52 11 NS Male gender (%) 59 31.02 Baseline SBP/DBP (mm Hg) 148 14/97 6 151 16/99 9 NS On-drug evening SBP/DBP lowering (mm Hg) 20 12/13 5 16 7/11 4 NS/.05 On-drug morning SBP/DBP lowering (mm Hg) 14 12/9 7 8 8/6 4.03/.01 On-drug global SBP/DBP lowering (mm Hg) 17 11/11 6 12 7/8 3.03/.007 Abbreviations: SBP, systolic blood pressure; DBP, diastolic blood pressure. * Residual BP lowering, difference between mean baseline BP level and mean post-treatment BP level. R ratio, ratio of residual BP lowering over evening on-drug BP lowering. y, plateau level assessed by exponential modeling.

172 VAUR ET AL AJH FEBRUARY 1998 VOL. 11, NO. 2 Nonreverters. However, the likelihood of reversion seemed to increase with the level of the pretreatment BP. 8,17 These studies do not allow us to define the precise time of reversion because BP levels are checked at best monthly. On the contrary, our study was focused on short-term follow-up. To our knowledge, this is the first study addressing the BP course during the few days following withdrawal of chronic antihypertensive therapy. Indeed, the use of home self-measurement of BP allowed us to evaluate individual BP levels twice daily during several consecutive days. Results showed that mean BP rose rapidly to reach a plateau. The level of this plateau was lower than the pretreatment BP level. This could be explained by a regression to the mean phenomenon, placebo effect over time, or persistance of true therapeutic activity. A similar phenomenon has been described previously in a study exploring SMBP during the first 5 days after shortterm administration of enalapril. 18 We decided to focus the individual analysis of reversion on the subgroup of responder patients. Indeed, BP cannot be expected to rise after withdrawing therapy in nonresponders. It appears to be difficult to individualize Reverters from Nonreverters, in that the distribution of the variable (R ratio) exploring reversion is unimodal and continuous. Fifteen percent of patients reverted totally to baseline, whereas BP did not rise at all in 11% of patients. The BP rise in the majority of patients was insufficient to reach baseline BP level. We decided arbitrarily to halve the responder population in Reverters and Nonreverters according to the median of the R ratio. Nonreverters were significantly better controlled during the treatment phase. No other difference between the two groups was found. It means that BP control under therapy is fundamental as far as reversion is considered. The more effective the BP control, the lower the reversion rate. This result may have two medical explanations. First, effective treatment may lead to the resetting of baroreceptors or any other BP control mechanism in well-controlled patients. 6 Second, structural changes may have occurred in previously altered small arteries of such patients, leading to a decrease in peripheral resistances. 19 In the present study, the BP level was consistently higher in the perindopril group than in the trandolapril group during the 6 days after drug cessation. It reverted close to baseline BP level in the perindopril group, whereas about half of the therapeutic effect was still present in the trandolapril group. Two explanations can be put forward. The duration of action of trandolapril is so long that a pharmacologic effect is still present 6 days after withdrawal. This hypothesis is unlikely as far as plasma ACE inhibition is concerned. However, the persistence of some tissue ACE inhibition has been demonstrated in animal models 1 week after cessation. 20 Furthermore, the duration of inhibition of local microvascular ACE in SHR has been shown to be longer with trandolapril than with perindopril. 21 Alternatively, trandolapril administration may have consistently modified BP control mechanisms or structural abnormalities. The BP control was better sustained during therapy in the trandolapril group, particularly in the morning. The well-documented long clinical duration of action of trandolapril supports this hypothesis. 22 24 All drugs will cease to have a pharmacologic action at some time following cessation of therapy. However, it has been suggested that the clinical consequences of drug cessation may be partly drug specific. 17 Withdrawal syndromes have been described in hypertensive patients who abruptly stop taking -blockers. 17,25 The severity of symptoms often depends on whether the patients have ischemic heart disease. Symptoms related to sympathetic overactivity have also been described when -blockers are suddenly stopped. 17 Some authors have studied ambulatory BP monitoring profiles during the 48-h period following cessation of various antihypertensive agents. Reversion occurs within 48 h with some drugs, whereas BP control appears well maintained during this period with others. For example, the magnitude and duration of the BP response over 24 to 48 h was statistically larger after a dose of betaxolol, compared to atenolol 26 ; amlodipine maintained a stable antihypertensive efficacy for 48 h after the dose, whereas BP reverted to baseline BP level 24 h after the last dose of diltiazem. 27 About half of the peak effect was maintained 48 h after withdrawal of trandolapril, whereas the residual effect was only 10% to 20% 48 h after the last dose of enalapril. 22 These discrepancies seem to be related to the duration of action of the drugs as measured by, for instance, their trough-to-peak ratios. The closer to 100% the trough-to-peak ratio, the greater the 48-h post-dose residual effect will be. 28 It would be of great interest to compare the short-term effects of withdrawing agents from various classes exhibiting different durations of action. In responder patients, ACE inhibitor treatment withdrawal is accompanied by a rapid rise in BP (within 48 h), followed by a 4-day BP plateau that is lower than the initial level. The final BP level is variable among patients and may be at least partially determined by the BP-lowering efficacy of the treatment. The better the BP control under therapy, the lower the reversion rate to pretreatment BP level will be. In this respect, the intrinsic properties of the drug may be of some importance. These findings are based on short-term treatment; results could be somewhat different after longer treatment periods.

AJH FEBRUARY 1998 VOL. 11, NO. 2 SMBP AND ACE INHIBITOR WITHDRAWAL 173 REFERENCES 1. Collins R, Peto R, MacMahon S, et al: Blood pressure, stroke, and coronary heart disease. Part 2. Short-term reductions in blood pressure: overview of randomised drug trials in their epidemiological context. Lancet 1990;335:827 838. 2. Dahlöf B, Lindholm LH, Hansson L, et al: Morbidity and mortality in the Swedish Trial in Old Patients with Hypertension (STOP-Hypertension). Lancet 1991;338: 1281 1285. 3. SHEP Cooperative Research Group: Prevention of stroke by antihypertensive drug treatment in older persons with isolated systolic hypertension. Final results of the Systolic Hypertension in the Elderly Program (SHEP). JAMA 1991;265:3255 3264. 4. Gong L, Zhang W, Zhu Y, et al: Shangai trial of nifedipine in the elderly (STONE). J Hypertens 1996;14:1237 1245. 5. Hansson L: Cardioprotective potential of antihypertensives. Am J Hypertens 1995;8(suppl):49S 52S. 6. Medical Research Council Working Party on Mild Hypertension: Course of blood pressure in mild hypertensives after withdrawal of long term antihypertensive treatment. Br Med J 1996;293:988 992. 7. Burton R: Withdrawing antihypertensive treatment. Hypertension may settle with time. Br Med J 1991;303: 324 325. 8. Dannenberg AL, Kannel WB: Remission of hypertension. The natural history of blood pressure treatment in the Framingham Study. JAMA 1987;257:1477 1483. 9. Girvin B, Johnston GD: The implication of noncompliance with antihypertensive medication. Drugs 1996;52: 186 195. 10. Meredith PA: Therapeutic implications of drug holidays. Eur Heart J 1996;17 (suppl A):21 24. 11. Mallion JM, Dutrey-Dupagne C, Vaur L, et al: Benefits of electronic pill boxes in evaluating treatment compliance of patients with mild to moderate hypertension. J Hypertens 1996;14:137 144. 12. Beltman FW, Heesen WF, Kok RHJ, et al: Predictive value of ambulatory blood pressure shortly after withdrawal of antihypertensive drugs in primary care patients. Br Med J 1996;313:404 406. 13. Jamieson MJ, Webster J, Witte K, et al: An evaluation of the A&D UA-751 semi-automated cuff-oscillometric sphygmomanometer. J Hypertens 1990;8:377 381. 14. Zannad F, Vaur L, Dutrey-Dupagne C, et al: Assessment of drug efficacy using home self-blood pressure measurement: the SMART study. J Hum Hypertens 1996;10:341 347. 15. Maland LJ, Lutz LJ, Castle CH: Effects of withdrawing diuretic therapy on blood pressure in mild hypertension. Hypertension 1983;5:539 544. 16. Korner PI, Jennings GL, Esler MD: Pathogenesis of human primary hypertension: a new approach to the identification of causal factors and to therapy. J Hypertens 1986;4 (suppl):s149 S153. 17. Schobel HP, Schmieder RE, Messerli FH: Risks versus benefits of withdrawing antihypertensive therapy. Drug Safety 1992;7:395 403. 18. Chatellier G, Day M, Bobrie G, et al: Feasibility study of N-of-1 trials with blood pressure self-monitoring in hypertension. Hypertension 1995;25:294 301. 19. Christensen KL, Jespersen LT, Mulvany MJ: Development of blood pressure in spontaneously hypertensive rats after withdrawal of long-term treatment related to vascular structure. J Hypertens 1989;7:83 90. 20. Chevillard C, Brown NL, Jouquey S: Cardiovascular actions and tissue converting enzyme inhibitory effects of chronic enalapril and trandolapril treatment of spontaneously hypertensive rats. J Cardiovasc Pharmacol 1989;14:297 301. 21. Vicaut E, Hou X: Duration of inhibition of local microvascular ACE in spontaneously hypertensive rats. J Hypertens 1995;13:1758 1763. 22. Vaur L, Dutrey-Dupagne C, Genès N, et al: Differential effects of a missed dose of trandolapril and enalapril on blood pressure control in hypertensive patients. J Cardiovasc Pharmacol 1995;26:127 131. 23. Zannad F, Vaur L, Dutrey-Dupagne C, et al: Antihypertensive effects of trandolapril and nitrendipine in the elderly: a controlled trial with special emphasis on time effect profiles. J Hum Hypertens 1996;10:51 55. 24. Cesarone MR, De Sanctis MT, Laurora G, et al: Effects of trandolapril on 24-hour ambulatory blood pressure in patients with mild-to-moderate essential hypertension. J Cardiovasc Pharmacol 1994;23 (suppl 4):S65 S72. 25. Psaty BM, Koepsell TD, Wagner EH, et al: The relative risk of incident coronary heart disease associated with recently stopping the use of -blockers. JAMA 1990; 263:1653 1657. 26. Johnson BF, Whelton A, McMahon FG: Betaxolol versus atenolol in hypertension: a comparison of efficacy, duration of response and effects of withdrawal (abstr). Am J Hypertens 1990;3(5)part 2:121a. 27. Leenen FHH, Fourney A, Notman G, et al: Persistence of anti-hypertensive effect after missed doses of calcium antagonist with long (amlodipine) vs short (diltiazem) elimination half-life. Br J Clin Pharmacol 1986; 41:83 88. 28. Meredith P, Elliot HL: Antihypertensive treatment and trough:peak ratio:general considerations. J Hypertens 1994;12(suppl 8):S79 S83.