Comparison of Angiotensin II ReceptorBlockers: Impact of Missed Doses of

Candesartan Cilexetil and Losartan inSystemic Hypertension

Giuseppe Mancia, MD, Raffaella Dell’Oro, MD, Carlo Turri, MD, and Guido Grassi, MD

Blood pressure remains poorly controlled in the hyper-tensive population due in large part to low or unsatis-factory patient compliance. Clinical studies that incorpo-rate an intentionally missed dose have been designed toevaluate the impact of poor patient compliance on theeffectiveness of antihypertensive medications. In thesestudies, ambulatory blood pressure monitoring is con-tinued throughout the dosing interval and beyond inorder to determine when systolic and diastolic bloodpressure increase into the hypertensive range. In an8-week, randomized, double-blind, placebo-controlledtrial in patients with mild-to-moderate hypertension, theantihypertensive effects of candesartan cilexetil 16 mgwere maintained after a missed dose, whereas systolicand diastolic blood pressure increased toward baseline

levels after a missed dose of losartan 100 mg. Cande-sartan cilexetil provided a significantly greater reductionin sitting systolic (p 5 0.004) and diastolic blood pres-sure (p 5 0.008) than losartan when measured 48 hoursafter the last dose. Moreover, the homogeneity of anti-hypertensive effects was greater after candesartan cilex-etil than losartan based on calculation of the smoothnessindex from ambulatory systolic and diastolic measure-ments during the first 24-hour period after dosing andduring the 12-hour period after the missed dose. Theseresults demonstrate that missing or delaying a dose ofcandesartan cilexetil has less impact on antihyperten-sive efficacy than missing or delaying a dose oflosartan. Q1999 by Excerpta Medica, Inc.

Am J Cardiol 1999;84:28S–34S

Despite the availability of safe and effective anti-hypertensive medications, blood pressure contin-

ues to be poorly controlled in the hypertensive popu-lation.1 Only a minority of patients achieve blood-pressure levels,140/90 mm Hg as recommended innational and international guidelines for the manage-ment of patients with hypertension.2,3 The Third Na-tional Health and Nutrition Examination Survey(NHANES III) found only about one fourth of nonin-stitutionalized adults with hypertension in the UnitedStates had achieved blood-pressure control.4 Evenamong hypertensive patients being treated with anti-hypertensive medications, less than half had achievedblood pressure control. Similarly, a survey of 73 gen-eral practitioners working at 14 local health officesthroughout Italy found only about one third of hyper-tensive subjects had achieved blood-pressure levels,140/90 mm Hg.5 Notably, results from the secondphase of NHANES III suggest that the percentage ofpatients with controlled blood pressure has remainedstatic in recent years.3

Several factors are believed to contribute to poorblood-pressure control. It is often difficult to predictwhich patients will respond to a given antihyperten-sive drug class, because the underlying pathophysiol-

ogy of hypertension is complex.6 However, currentguidelines define an individualized stepwise approachto hypertension management, in which drugs areadded or substituted based on a patient’s response totreatment.2,3 Nevertheless, many physicians appear re-luctant to aggressively treat hypertension to achieveblood-pressure control.5,7 Some physicians expressconcern about the J-curve hypothesis, which althoughunproved, suggests that excessive blood-pressure low-ering may increase cardiovascular risk. Others fail toaggressively treat hypertension because of concernabout potential drug side effects, and still others ap-pear uncertain about the benefits of antihypertensivetreatment, particularly in patients with mild hyperten-sion. Even in cases where an appropriate treatmentplan has been defined, patients may still not achieveblood-pressure control because they do not follow thephysician’s instructions. Many experts believe thatpoor blood-pressure control is largely due to low orunsatisfactory patient compliance.

PATIENT COMPLIANCEPatient compliance is a complex phenomenon that

is impacted by numerous factors, including ethnicbackground, educational level, and economic status ofthe patient, and the complexity and side effects asso-ciated with treatment. The educational and economiclevels of patients are related inversely to patient com-pliance and blood-pressure control. In NHANES III,the percentage of hypertensive patients with con-trolled blood pressure was low in both men andwomen and in all ethnic groups, but it was lowest

From the Clinica Medica, Universita di Milano–Bicocca, Ospedale S.Gerardo, Monza, Milan, Italy.

Address for reprints: Giuseppe Mancia, MD, University of Milan–Bicocca, S. Gerardo Hospital, Department of Internal Medicine,Monza, Milan, Italy.

28S ©1999 by Excerpta Medica, Inc. 0002-9149/99/$20.00All rights reserved. PII S0002-9149(99)00731-6

among Mexican Americans.4 The impact of drug-induced side effects on patient compliance is oftenunderestimated by physicians, because many patientsfail to inform their physicians that they have discon-tinued treatment due to side effects, some of whichimpact the patient’s quality of life, such as sexualdysfunction withb blockers. Finally, the complexityof treatment also impacts patient compliance, withpatients being more likely to remain compliant on asimple treatment plan. For example, patient compli-ance is higher with a once-daily antihypertensive med-ication than with an agent that must be taken severaltimes a day.

Patient compliance is difficult to assess quantita-tively. However, electronic methods, such as the med-ication event monitoring system, have been developedthat record when a drug container is opened andclosed. Although such studies do not prove whetherthe patient has actually ingested the medication, theyprovide an indication of when the patient is likely tohave received treatment. Such studies indicate thatpatient compliance is actually lower than suggested bypill counts or patient self-reports.8,9 Moreover, the useof electronic monitoring suggests that poor patientcompliance is expressed in several different ways.Some patients do not take any pills after a physicianvisit, and only start taking their medication just beforetheir next visit. More often, however, patients takemedications at irregular times and not at the end of thedosing interval as prescribed by the physician. Theymay take the medication earlier than prescribed, butfrequently they take it later than at the prescribedinterval.

The erratic pattern of patient compliance is illus-trated by results of several studies.10 Moreover, theresults of other studies indicate that a significant pro-portion of patients delay taking medication beyond theprescribed dosing interval.9,11 As a result, antihyper-tensive medications that “forgive” patients for thisdelay are receiving increasing attention. Such forgiv-ing medications not only control blood pressure dur-ing the dosing interval, but their efficacy extendsbeyond a 24-hour period if the agent is given oncedaily or beyond 12 hours if administered twice daily.

The effect of delaying treatment beyond the dosinginterval on blood-pressure control has been addressedin so-called “missed-dose” studies. Blood pressure isrecorded continuously before and after a dose is in-tentionally skipped to determine when blood pressureincreases into the hypertensive range after the misseddose. In one study, ambulatory blood-pressure moni-toring was conducted before and after 6 months oftreatment with the long-acting calcium antagonist,amlodipine.12 Monitoring was conducted for 24 hoursafter the amlodipine dose and then continued for anadditional 24 hours after the next scheduled dose wasskipped. The antihypertensive effect of amlodipinewas maintained throughout the 24-hour period follow-ing the missed dose, which indicated that this agentbelongs to the category of drugs that forgive patients

for missing or delaying a dose. Other investigatorshave reported similar results. Amlodipine13 and alsonifedipine GITS (Gastrointestinal Therapeutic Sys-tem)14 remained effective after a missed dose, whereasblood-pressure lowering with diltiazem or perindoprilwas diminished.13,15 Missed-dose studies have alsobeen conducted with other antihypertensive drugclasses, and show that some, but not all, members ofeach drug class may forgive patients for missing ordelaying a dose.16,17

ANGIOTENSIN II RECEPTORBLOCKERS: THE CHAMP STUDY

The angiotensin II receptor blockers (ARBs) arethe newest drug class to become available for thetreatment of hypertension. These agents include can-desartan cilexetil, irbesartan, losartan, telmisartan, andvalsartan, all of which are approved for once-dailyadministration. The impact of a missed dose on theantihypertensive effects of candesartan cilexetil andlosartan was evaluated in the CHAMP Study.17 In thisrandomized, double-blind, placebo-controlled trial,patients 20–80 years of age with mild-to-moderatehypertension entered a run-in period during whichthey received placebo, and then they were randomizedin a 3 : 3 : 1 ratio to once-daily treatment withcandesartan cilexetil, losartan, or placebo. Candesar-tan cilexetil was administered at a daily dosage of 8mg for the first 4 weeks, and then the dosage was forcetitrated to 16 mg daily for the final 4 weeks of thestudy. Losartan was administered at a daily dosage of50 mg during the first 4-week period and 100 mg dailyduring the final 4-week period. Ambulatory blood-pressure monitoring was conducted for 36 hours atbaseline, and after 4 and 8 weeks of treatment.

During the first 24-hour period after dosing, bothcandesartan cilexetil and losartan lowered ambulatorysystolic and diastolic blood pressure relative to base-line levels (Figure 1). At the week 8 assessment,reduction in systolic blood pressure was significantlygreater with candesartan cilexetil 16 mg than withlosartan 100 mg (p5 0.008). Candesartan cilexetilalso exhibited a trend for producing greater reductionsin ambulatory diastolic blood pressure during the 24-hour period following administration (p5 0.073).However, during the subsequent 12-hour period,which followed a missed dose, ambulatory systolicand diastolic blood pressure remained lowered in thecandesartan cilexetil group, whereas they increasedtoward baseline levels in the losartan group. Duringthis period after the missed dose, candesartan cilexetilprovided a significantly greater reduction in both am-bulatory systolic (p5 0.004) and diastolic blood pres-sure (p 5 0.022, Figures 1A and B). In addition,sitting blood-pressure measurements were performedin the clinic at 48 hours after the last drug dose.Candesartan cilexetil 16 mg provided a significantlygreater reduction in sitting systolic (p5 0.004) anddiastolic blood pressure (p5 0.008) than losartan 100

A SYMPOSIUM: ANGIOTENSIN II RECEPTOR BLOCKADE IN HYPERTENSION 29S

FIGURE 1. Candesartan cilexetil vs losartan: mean change from baseline to week 8 in am-bulatory systolic blood pressure (A) and diastolic blood pressure (B) during the 36-hour pe-riod after the last drug dose. The arrow indicates the time of the missed dose. (Adapted withpermission from Am J Hypertens.17)

FIGURE 2. Reduction in sitting systolic and diastolic blood pressure at 48 hours after the lastdose of candesartan cilexetil 16 mg or losartan 100 mg: mean (and 95% confidence interval)change from baseline to week 8. The changes from baseline in sitting systolic and diastolicblood pressure were significantly greater at 48 hours after candesartan cilexetil than afterlosartan. (Adapted with permission from Am J Hypertens.17)

30S THE AMERICAN JOURNAL OF CARDIOLOGYT VOL. 84 (10A) NOVEMBER 18, 1999

mg when measured 48 hours after the last dose (Figure2). Thus, the results of the CHAMP study demonstratethat missing a dose of candesartan cilexetil had muchless effect on blood-pressure control than skipping adose of losartan.17 These results were presented at theAmerican Society of Hypertension 14th ScientificMeeting by Lacourcie`re and Asmar.17

HOMOGENEITY OFANTIHYPERTENSIVE TREATMENT

The homogeneity of antihypertensive treatmentduring the entire dosing interval is another importantconsideration because the degree of blood-pressurevariability is significantly and independently associ-ated with increased risk for end-organ damage seen inhypertension.18,19 Calculation of the trough-to-peakratio has been a popular approach for assessing thehomogeneity of antihypertensive treatment and forjustifying administration of drugs in a once-daily reg-

imen. It is generally believed that a low trough-to-peak ratio has 3 disadvantages. First, it may lead toexcessive decrease in blood pressure at peak response,resulting in underperfusion of vital organs. Second, itmay result in inability to control blood pressure duringthe latter part of the dosing interval. Third, it maycause a general increase in blood-pressure variabilitydue to the effect of pharmacologic variability on thenormal spontaneous variability seen in hypertensiveindividuals.20

Recent evidence suggests that homogeneity of an-tihypertensive treatment may be better assessed bycalculating the smoothness index.21 Whereas thetrough-to-peak ratio is determined from only 2 blood-pressure measurements—at peak and at trough—dur-ing a 24-hour dosing interval, the smoothness index iscalculated from the standard deviation of all hourlyblood-pressure measurements during the 24-hour pe-riod and normalized for the mean blood-pressure de-crease during this period. As a result, homogeneity of

FIGURE 3. Reproducibility of trough : peak (TP) ratio and smoothness index (SI) after 3 and12 months of treatment. DBP5 diastolic blood pressure; SBP 5 systolic blood pressure.(Adapted with permission from J Hypertens.21)

A SYMPOSIUM: ANGIOTENSIN II RECEPTOR BLOCKADE IN HYPERTENSION 31S

treatment is assessed throughout the 24-hour intervalrather than at 2 arbitrary time points.

The advantage of using the smoothness index in-stead of the trough-to-peak ratio for assessing thehomogeneity of antihypertensive treatment is illus-trated by results from the Study on Ambulatory Mon-itoring of Blood Pressure and Lisinopril Evaluation(SAMPLE).21,22 In this trial, patients with hyperten-sion and left ventricular hypertrophy were treated withlisinopril 20 mg plus hydrochlorothiazide 12.5–25 mgas needed to achieve blood-pressure control. Ambu-latory blood-pressure monitoring was conducted after3 and 12 months of treatment, and the homogeneity ofantihypertensive treatment assessed by calculating thetrough-to-peak ratio and the smoothness index. Thesmoothness index, which was calculated from theambulatory systolic and diastolic blood pressure,showed greater reproducibility between measurements

at 3 and 12 months than did the trough-to-peak ratio,which was calculated from average changes in systolicand diastolic blood pressure over 2-hour periods se-lected at peak and trough times (Figure 3).21 More-over, after 12 months of treatment, the regression inleft ventricular hypertrophy was significantly corre-lated with the smoothness index calculated from bothsystolic and diastolic blood pressure (both p,0.001),but not with the trough-to-peak ratio (Figure 4).21,22Inthis study, erratic values of trough-to-peak ratio wereoften encountered, whereas the smoothness index hada normal distribution, which offered a statistical ad-vantage when differences between treatments werecalculated.

The smoothness index and trough-to-peak ratiohave been calculated for hypertensive patients whowere treated with placebo and lower (4–8 mg oncedaily) or higher (12–16 mg) dosages of candesartan

FIGURE 4. Correlation coefficient values between changes in left ventricular mass index(LVMI) induced by antihypertensive treatment and trough : peak (TP) ratio (upper panels)and smoothness index (SI). The relation was significant only for SI. (Adapted with permissionfrom Hypertension.22)

32S THE AMERICAN JOURNAL OF CARDIOLOGYT VOL. 84 (10A) NOVEMBER 18, 1999

cilexetil. The smoothness index was zero in the pla-cebo group, but had a positive value in both cande-sartan cilexetil groups, with a significantly greatervalue in the higher-dosage group (Figure 5). However,when the trough-to-peak ratio was used to assess thehomogeneity of antihypertensive treatment, a signifi-cant difference between the 2 dosage levels was notevident, because the trough-to-peak ratio exhibitedlower reproducibility than the smoothness index.

SUMMARYClinical studies in which a dose is missed inten-

tionally provide important evidence on which antihy-pertensive medications are likely to be most forgivingwhen a patient misses or delays taking the next dose.Ambulatory blood-pressure monitoring during thetime period after the missed dose identifies whenblood pressure increases into the hypertensive range.The results of the CHAMP study demonstrate thatcandesartan cilexetil, unlike losartan, maintains blood-pressure control after a missed dose. Similarly, othertrials suggest that specific members of each drug class,such as amlodipine among the calcium antagonistsand lisinopril and trandolapril among the ACE inhib-itors, also provide blood-pressure control beyond theonce-daily dosing interval. A new measure, thesmoothness index, which indicates the homogeneityof the antihypertensive effects during the dosing in-terval, is inversely related to blood-pressure variabil-ity, an independent factor for end-organ damage inhypertension. The smoothness index is more repro-ducible than the trough-to-peak ratio, and it correlatedsignificantly with regression of left ventricular hyper-trophy in the SAMPLE study with lisinopril. Based onresults from a comparative clinical trial, candesartan

cilexetil has a significantly greater smoothness indexfor both systolic and diastolic blood pressure thanlosartan during the first 24 hours after administrationas well as during the 12-hour period following amissed dose. The results of these clinical studies dem-onstrate that missing or delaying a dose of candesartancilexetil has less impact on antihypertensive efficacythan missing or delaying a dose of losartan.

1. Mancia G, Sega R, Milesi C, Cesana G, Zanchetti A. Blood-pressure controlin the hypertensive population.Lancet1997;349:454–457.2. WHO-ISH Hypertension Guidelines Committee. 1999 World Health Organi-zation—International Society of Hypertension Guidelines for the Management ofHypertension.J Hypertens1999;17:151–185.3. Joint National Committee on Prevention, Detection, Evaluation, and Treatmentof High Blood Pressure. The Sixth report of the Joint National Committee onPrevention, Detection, Evaluation, and Treatment of High Blood Pressure. ArchIntern Med1997;157:2413–2446.4. Burt VL, Whelton P, Roccella EJ, Brown C, Cutler JA, Higgins M, Horan MJ,Labarthe D. Prevalence of hypertension in the US adult population: results fromthe Third National Health and Nutrition Examination Survey, 1988-1991.Hy-pertens1995;25:305–313.5. Avanzini F, Alli C, Colombo P, Corsetti A, Colombo F, Tognoni G. Controlof hypertension in Italy: results of the Study on Antihypertensive Treatment inGeneral Practice (STAP).G Ital Cardiol 1998;28:760–766.6. Menard J, Chatellier G. Limiting factors in the control of BP: why is there agap between theory and practice?J Human Hypertens1995;9(suppl 2):S19–S23.7. Berlowitz DR, Ash AS, Hickey EC, Friedman RH, Glickman M, Kader B,Moskowitz MA. Inadequate management of blood pressure in a hypertensivepopulation.N Engl J Med1998;339:1957–1963.8. Straka RJ, Fish JT, Benson SR, Suh JT. Patient self-reporting of compliancedoes not correspond with electronic monitoring: an evaluation using isosorbidedinitrate as a model drug.Pharmacother1997;17:126–132.9. Lee JY, Kusek JW, Greene PG, Bernhard S, Norris K, Smith D, Wilkening B,Wright JT Jr. Assessing medication adherence by pill count and electronicmonitoring in the African American Study of Kidney Disease and Hypertension(AASK) pilot study.Am J Hypertens1996;9:719–725.10. Urquhart J. Patient non-compliance with drug regimes: measurement, clinicalcorrelates, economic impact.Eur Heart J1996;17(suppl A):8–15.11. Leenen FH, Wilson TW, Bolli P, Larochelle P, Myers M, Handa SP, BoileauG, Tanner J. Patterns of compliance with once versus twice daily antihypertensivedrug therapy in primary care: a randomized clinical trial using electronic moni-toring. Can J Cardiol1997;13:914–920.

FIGURE 5. Values of smoothness index for systolic (SBP) and diastolic blood pressure (DBP)during placebo and candesartan cilexetil (C) treatment at low and high doses.

A SYMPOSIUM: ANGIOTENSIN II RECEPTOR BLOCKADE IN HYPERTENSION 33S

12. Mancia G, Cattaneo BM, Omboni S, Grassi G. Clinical benefits of aconsistent blood pressure reduction.J Hypertension1998;16(suppl 6):S35–S39.13. Leenen FH, Fourney A, Notman G, Tanner J. Persistence of anti-hypertensiveeffect after ‘missed doses’ of calcium antagonist with long (amlodipine) vs short(diltiazem) elimination half-life.Br J Clin Pharmacol1996;41:83–88.14. Zanchetti A. Antihypertensive effects of nifedipine GITS on clinic andambulatory blood pressure in essential hypertensives.High Blood Press1994;3:45–56.15. Zannad F, Bernaud CM, Fay R. Double-blind, randomized, multicentrecomparison of the effects of amlodipine and perindopril on 24 h therapeuticcoverage and beyond in patients with mild to moderate hypertension.J Hypertens1999;17:137–146.16. Vaur L, Dutrey-Dupagne C, Boussac J, Genes N, Bouvier d’Yvoire M, ElkikF, Meredith PA. Differential effects of a missed dose of trandolapril and enalaprilon blood pressure control in hypertensive patients.J Cardiovasc Pharmacol1995;26:127–131.17. Lacourciere Y, Asmar R. Comparison of the impact of missed doses of

candesartan cilexetil and losartan in ambulatory hypertensive patients. (Abstr.)Am J Hypertens1999;12:143A.18. Mancia G, Parati G, Omboni S, Ulian L, Zanchetti A. Ambulatory bloodpressure monitoring.Clin Exp Hypertens1999;21:703–715.19. Frattola A, Parati G, Cuspidi C, Albini F, Mancia G. Prognostic value of 24hour blood pressure variability.J Hypertens1993;11:1133–1137.20. Mancia G, Ferrari A, Gregorini L, Parati G, Pomidossi G, Bertinieri G, GrassiG, Di Rienzo M, Pedotti A, Zanchetti A. Blood pressure and heart rate variabil-ities in normotensive and hypertensive human beings.Circ Res1983;53:96–104.21. Parati G, Omboni S, Rizzoni D, Agabiti-Rosei E, Mancia G. The smoothnessindex: a new, reproducible and clinically relevant measure of the homogeneity ofthe blood pressure reduction with treatment for hypertension.J Hypertens1998;16:1685–1691.22. Omboni S, Fogari R, Palatini P, Rappelli A, Mancia G. Reproducibility andclinical value of the trough-to-peak ratio of the antihypertensive effect: evidencefrom the sample study.Hypertension1998;32:424–429.

34S THE AMERICAN JOURNAL OF CARDIOLOGYT VOL. 84 (10A) NOVEMBER 18, 1999