clinical investigation effect ofdiltiazem on symptomatic...

9
15 Clinical Investigation Effect of Diltiazem on Symptomatic and Asymptomatic Episodes of ST Segment Depression Occurring During Daily Life and During Exercise P. Theroux, MD; M. Baird, MD; M. Juneau, MD; W. Warnica, MD; P. Klinke, MD; W. Kostuk, MD; P. Pflugfelder, MD; E. Lavallee, BSc; C. Chin, BSc; E. Dempsey, MSc; M. Grace, PhD; Y. Lalonde, MSc; and D. Waters, MD Background. Silent myocardial ischemia is an adverse prognostic marker in patients with coronary disease; however, controlled data on the effect of treatment are sparse and contra- dictory, and the relations among the occurrence of ST segment depression, drug efficacy, and heart rate are unclear. Methods and Results. Sixty patients with stable coronary artery disease, a positive treadmill exercise test and asymptomatic ST segment depression on ambulatory electrocardiographic recording were assessed in a multicenter, double-blind, placebo-controlled, cross-over trial. Treadmill exercise tests and 72-hour electrocardiographic recordings were obtained at the end of two 2-week treatment periods with sustained-release diltiazem 180 mg b.i.d. or equivalent placebo. Episodes of asymptomatic ST depression decreased by 50% or more in 70% of the patients from a median number of 4.5 (range, 0-19) to 1.5 (range, 0-13) (p=0.0001); their cumulative duration also decreased from 78.5 (range, 0-60) to 24.5 (range, 0-411) minutes (p=0.001). No circadian variation was found in the efficacy of diltiazem. The occurrence of ischemic type ST segment depression was modulated by changes in heart rate rather than by absolute heart rate. Diltiazem also improved exercise test end points but to a lesser extent. Time to ST segment depression increased to 341±+148 from 296±f154 seconds (p=0.005). Although less frequent with diltiazem administration (45 versus 54 patients, p<0.03), exercise-induced ST depression Was more often asymptomatic (98% versus 72% of patients, p<O.0001). Conclusions. Diltiazem reduces the frequency and severity of ischemic type ST depression in patients with stable coronary artery disease. (Circulation 1991;84:15-22) Silent myocardial ischemia has emerged as a controversial problem in modern cardiology. Although, both its pathophysiological basis and its importance in routine practice remain to be more fully explored, the concept could potentially modify our clinical approach to patients with coro- nary artery disease. Clearly, silent ischemia occurs frequently,1-3 can be diagnosed accurately,3-8 and carries prognostic implications in patients with un- stable9,10 and stablell-'3 coronary artery disease. These considerations justify a more in-depth investi- From the Canadian Multicenter Diltiazem Study Group, Montreal. Supported by Nordic Laboratories, Inc., Montreal. Address for reprints: Pierre Theroux, MD, Montreal Heart Institute, 5000 East Belanger Street, Montreal, Quebec H1T 1C8, Canada. Received September 6, 1990; revision accepted February 12, 1991. gation of the pathophysiology, clinical features, and management of silent ischemia. Ambulatory electro- cardiographic (ECG) monitoring provides an easy and accessible method to study silent ischemia and the effect of various interventions on it.6-8 Small or un- controlled studies have suggested that silent ischemia can be partly prevented by antianginal drugs14-25 or revascularization procedures.26 However, larger and controlled studies have contradicted some of these results.27 This placebo-controlled, randomized, double-blind study was designed to determine the effect on silent ischemia of one antianginal drug, sustained-release diltiazem. In addition to this primary goal, the trial was expected to provide useful data to define the features of symptomatic and asymptomatic ST seg- ment depression in a population of patients with stable coronary artery disease, to provide an insight by guest on July 14, 2018 http://circ.ahajournals.org/ Downloaded from

Upload: lamcong

Post on 01-Jul-2018

215 views

Category:

Documents


0 download

TRANSCRIPT

15

Clinical Investigation

Effect of Diltiazem on Symptomatic andAsymptomatic Episodes of ST SegmentDepression Occurring During Daily Life

and During ExerciseP. Theroux, MD; M. Baird, MD; M. Juneau, MD; W. Warnica, MD; P. Klinke, MD;

W. Kostuk, MD; P. Pflugfelder, MD; E. Lavallee, BSc; C. Chin, BSc; E. Dempsey, MSc;

M. Grace, PhD; Y. Lalonde, MSc; and D. Waters, MD

Background. Silent myocardial ischemia is an adverse prognostic marker in patients withcoronary disease; however, controlled data on the effect of treatment are sparse and contra-dictory, and the relations among the occurrence of ST segment depression, drug efficacy, andheart rate are unclear.Methods and Results. Sixty patients with stable coronary artery disease, a positive treadmill

exercise test and asymptomatic ST segment depression on ambulatory electrocardiographicrecording were assessed in a multicenter, double-blind, placebo-controlled, cross-over trial.Treadmill exercise tests and 72-hour electrocardiographic recordings were obtained at the endof two 2-week treatment periods with sustained-release diltiazem 180 mg b.i.d. or equivalentplacebo. Episodes of asymptomatic ST depression decreased by 50% or more in 70% of thepatients from a median number of 4.5 (range, 0-19) to 1.5 (range, 0-13) (p=0.0001); theircumulative duration also decreased from 78.5 (range, 0-60) to 24.5 (range, 0-411) minutes(p=0.001). No circadian variation was found in the efficacy of diltiazem. The occurrence ofischemic type ST segment depression was modulated by changes in heart rate rather than byabsolute heart rate. Diltiazem also improved exercise test end points but to a lesser extent. Timeto ST segment depression increased to 341±+148 from 296±f154 seconds (p=0.005). Althoughless frequent with diltiazem administration (45 versus 54 patients, p<0.03), exercise-inducedST depression Was more often asymptomatic (98% versus 72% of patients, p<O.0001).

Conclusions. Diltiazem reduces the frequency and severity of ischemic type ST depression inpatients with stable coronary artery disease. (Circulation 1991;84:15-22)

Silent myocardial ischemia has emerged as acontroversial problem in modern cardiology.Although, both its pathophysiological basis

and its importance in routine practice remain to bemore fully explored, the concept could potentiallymodify our clinical approach to patients with coro-nary artery disease. Clearly, silent ischemia occursfrequently,1-3 can be diagnosed accurately,3-8 andcarries prognostic implications in patients with un-stable9,10 and stablell-'3 coronary artery disease.These considerations justify a more in-depth investi-

From the Canadian Multicenter Diltiazem Study Group,Montreal.

Supported by Nordic Laboratories, Inc., Montreal.Address for reprints: Pierre Theroux, MD, Montreal Heart

Institute, 5000 East Belanger Street, Montreal, Quebec H1T 1C8,Canada.

Received September 6, 1990; revision accepted February 12, 1991.

gation of the pathophysiology, clinical features, andmanagement of silent ischemia. Ambulatory electro-cardiographic (ECG) monitoring provides an easy andaccessible method to study silent ischemia and theeffect of various interventions on it.6-8 Small or un-controlled studies have suggested that silent ischemiacan be partly prevented by antianginal drugs14-25 orrevascularization procedures.26 However, larger andcontrolled studies have contradicted some of theseresults.27

This placebo-controlled, randomized, double-blindstudy was designed to determine the effect on silentischemia of one antianginal drug, sustained-releasediltiazem. In addition to this primary goal, the trialwas expected to provide useful data to define thefeatures of symptomatic and asymptomatic ST seg-ment depression in a population of patients withstable coronary artery disease, to provide an insight

by guest on July 14, 2018http://circ.ahajournals.org/

Dow

nloaded from

16 Circulation Vol 84, No 1 July 1991

into the pathophysiology of silent ischemia bothduring daily life and during exercise, and to yieldclues to the mechanism of its suppression by dil-tiazem. For this purpose, the relation between heartrate and episodes of ST segment depression wasclosely monitored, and the effects of the drug on STdepression during ambulatory ECG monitoring andduring exercise testing were compared.

MethodsPatient Selection

Sixty patients who meet all of the following criteriawere included in the study: 1) documented coronaryartery disease, 2) stable symptoms in Canadian Car-diovascular Society classes 1-3 without rest, noctur-nal, progressive, or unstable angina and without arecent (.3 months) myocardial infarction, 3) a pos-itive treadmill exercise test according to the Bruceprotocol, with 1.0 mm or more horizontal ordownsloping ST depression measured 0.08 msec afterthe J point, with or without angina, and 4) thepresence of 10 minutes or more of silent ST depres-sion 1 mm or more in amplitude or four or moreasymptomatic episodes, each lasting for at least 1minute during 24 hours of a continuous 72-hourscreening ambulatory ECG. The criteria used for thediagnosis of coronary artery disease were 1) presenceof at least one coronary artery stenosis greater than50% lumen diameter reduction, 2) a well-docu-mented previous myocardial infarction, or 3) a posi-tive thallium exercise test manifested by a significantexercise-induced reversible defect. Holter ECG eli-gibility could be determined at the clinical center, butconfirmation within 72 hours by the central ECGlaboratory was required or the patient would bedropped from the study.

Exclusion criteria included 1) cardiac surgerywithin the previous 3 months, 2) stroke within theprevious 6 months; 3) congestive heart failure, 4)known valvular heart disease, 5) the presence ofatrial arrhythmias, sinus bradycardia of 50 beats/minor less, ventricular preexcitation, pacemaker, con-duction abnormalities, left ventricular hypertrophy,or ST depression on the resting 12-lead ECG; 6)hypertension of 160 mm Hg or more systolic or 90mm Hg or more diastolic and hypotension of 100mm Hg or less systolic. The study was approved bythe respective Hospital Ethics Committees, and allpatients gave written, informed consent for partici-pation in the study.One hundred ninety-six patients were screened in

the five clinical centers, and 118 were excludedbefore study entry; the most-common exclusion fac-tor was failure to exhibit ST depression during the 72hours of ambulatory ECG monitoring. Eighteen pa-tients began the trial and could not be included in theanalysis because they had no ambulatory ECG re-cording or treadmill exercise test during the double-blind phase of the study: 10 were ineligible based onthe central laboratory Holter ECG reading, two

refused further participation, and six experiencedadverse reactions.At least 7 days before the screening visit, patients

discontinued all (3-blocking agents, digoxin, vasodila-tors including long-acting nitrates and calcium chan-nel blockers, and any other drug that can interferewith interpretation of ST segment changes. Use ofthese medications was not permitted during thestudy. Sublingual nitroglycerin was allowed to treatangina episodes.

Study DesignThe study was randomized, double-blind, and

placebo-controlled. After a 7-day single-blind pla-cebo run-in screening phase, patients received sus-tained-release diltiazem capsules 180 mg b.i.d., atotal daily dose of 360 mg, or matching placebocapsules for two consecutive 14-day double-blindtreatment phases. Each patient received each of thetwo treatments in a random sequence according toa 2x 2 Latin square design. Compliance as evalu-ated by pill counts was 98+4% during diltiazem and97±+-5% during placebo administration.

Ambulatory ECG MonitoringECGs were recorded on 24-hour cassette tapes

with a calibrated Marquette Series 8500 recorder(Marquette Electronics Inc., Milwaukee, Wis.). A72-hour continuous ambulatory ECG was obtainedat the start of the placebo run-in phase to identifyqualified patients for the study and at the end of eachtreatment phase, with diltiazem and with placebo.Bipolar electrodes were attached with the exploringelectrodes applied to record modified V2 and V5leads. Patients were instructed to maintain theirnormal daily activities and to record their activities,symptoms, and medications in a diary.

All tapes were analyzed by the Ambulatory ECGAnalysis Central Laboratory at the Montreal HeartInstitute using the Cardiodata MK4 Playback System(Cardiodata, Northborough, Mass.). Tapes were in-dependently read by a technician and one physician;both were unaware of patient identity and treatmentassignment. An ischemic episode was defined by thestandard criteria of transient ST depression of 1 mmor more lasting for at least 1 minute. In the presentstudy, the duration of the ischemic episode wascalculated from the onset of ST depression to returnto baseline. An isoelectric segment lasting for at least1 minute was required between episodes. All epi-sodes of ST depression that were identified on thetrend analysis were validated by visual inspection ofthe ECG tracings, recorded at a paper speed of 25mm/sec before, at the beginning, peak, and end of theST depression. Artifacts, apparent positionalchanges, and upsloping ST depression of 1 mm orless at 0.06 msec after the J point were disregarded.Borderline ST depression was also not counted un-less the configuration was clearly ischemic and simi-lar to other episodes of ST depression of 1 mm ormore in the same patient The total number of

by guest on July 14, 2018http://circ.ahajournals.org/

Dow

nloaded from

ThUroux et al Diltiazem and Silent Ischemia 17

ischemic episodes and the cumulative duration wereused as study end points. Heart rate was calculated 5minutes before and at the peak of ST segment shiftand as the mean per hour for the 72-hour recordings.

Intraobserver variability was studied by rereadingin a blind fashion 50 tapes selected at random amongthe 941 study tapes. The coefficient of correlationbetween the two readings from the same patient was0.93 for the number of episodes and 0.91 for the totalduration. The interobserver reliability was studied byhaving 50 randomly selected tapes interpreted by anindependent laboratory that also used the Cardio-data MK4 Playback System. The coefficients of cor-relation between the two independent readings ofthe same patient were 0.87 for number of episodesand 0.90 for total duration of ischemic episodes.

Exercise Tolerance TestingA maximal symptom-limited treadmill exercise test

was performed at the baseline screening visit to identifyeligible patients. The test was repeated at the end ofeach double-blind treatment phase, just before the endof the 72-hour continuous ECG recording period. Alltests were performed in the morning, at the same timeof day for each patient, and 2-3 hours after ingestion ofa dose of the study drug. The Bruce protocol was used,and the following were recorded: blood pressure andheart rate at each stage and at each end point, reasonsfor stopping the test, time of onset of angina and of1-mm ST depression, peak ST depression, maximumST segment shift, and total exercise duration.' Theexercise tests were interpreted by the principal investi-gator at each clinical center.

Statistical AnalysisData were analyzed using the SAS (Statistical Analy-

sis System, Cary, N.C.), version 5.18 with Virtual Mem-ory System. Because of the nonnormal distribution ofangina frequency, nitroglycerin consumption, and fre-quency and duration of ischemia on Holter monitoring,statistical tests were performed by use of a parametricprocedure (ANOVA) with the ranked data (Iman-Connover method28). Results for these data are pre-sented as medians and ranges to comply with thenonnormal distribution and as means+SD as justifiedby the rank transformation of the data. Data on clinicalcharacteristics and on the exercise test parametersfollowed a normal distribution and were analyzed by x2and t statistics. They are presented as mean±SD.Comparisons between the two treatments during thedouble-blind phase of the study were performed fol-lowing the Grizzle model for a two-period Latin squaredesign for repeated measures. Any possible residual(carry-over) effect and period effect can also be deter-mined by this model. A center-by-treatment interactionwas ruled out using a two-way analysis of variance forrepeated measures. Interobserver and intraobservervariability in ambulatory ECG interpretation and therelation between episodes of ST segment depressionand changes in heart rate were studied with regressioncorrelation coefficents.

TABLE 1. Characteristics of the Study Population at EntryInto Study

ClinicalAge (yr)Sex (n)MaleFemale

Previous myocardial infarction (n)Previous hypertension (n)Diabetes mellitus (n)Previous smokers (n)Nonsmokers (n)Angina episodes(Number per wk)

Functional class (n)123

Electrocardiographic (n)Q wave present1st degree AV blockSignificant arrhythmias

AngiographicWith coronary angiography (n)

One-vessel diseaseTwo-vessel diseaseThree-vessel disease

59.9+8 (ranges, 42-79)

564

18140

3751

(93)( 7)(32)(23)( 0)(64)(85)

1.49±2.6

44 (73)13 (22)3 ( 5)

2441

44151712

(31)

( 5)( 1)

(73)

Values are mean±SD where applicable. Values in parenthesesare percentages.AV, atrioventricular.

The significance level for inferential procedureswas fixed at 5%.

ResultsStudy Population

All 60 patients in this study had documented coro-nary artery disease and reversible ischemia, expressedas transient ST depression, both during daily life andduring a treadmill exercise test. Coronary artery diseasewas documented by coronary angiography in 44 pa-tients (73%), by a previous myocardial infarction in 18(32%), and by a; thallium exercise test showing areversible perfusion defect in 29 (48%). Five patientshad all three diagnostic criteria; 25 had two, and 30 hadone criterion. Among the 30 patients with one diagnos-tic criterion, the disease was documented by angiogra-phy in 20 and by exercise thallium scintigraphy in 10.Angina during the treadmill test was present in 17patients (28%); exercise was terminated during Brucestage 2 in 21, during stage 3 in 32, during stage 4 in six,and during stage 5 in one. The clinical, ECG, andangiographic features of the patients at the time ofentry into the study are listed in Table 1. The mediannumber of chest pain episodes per week was 0 (range,0-7) with a mean of 1.49+2.6 per patient. By entrycriteria,'no patient had chest pain at rest or during thenight or had a clinical history suggestive of variable

by guest on July 14, 2018http://circ.ahajournals.org/

Dow

nloaded from

18 Circulation Vol 84, No 1 July 1991

No. episodes10 -

8-

6 -

4 - (0.

2-

0--

Total duration (mmn)250

- 200

- 150

- 100

- 50

r0-400) 0

Placebo Diltiazem Placebo Diltiazem

FIGURE 1. Bar graph of total episodes and cumulative duration of ST depression during the double-blind crossoverphase of thestudy. Diltiazem significantly reduced both the number of episodes and their total duration. Dotted lines within the bars show themedian and the numbers the ranges.

threshold or vasospastic angina. Seventy-three percentof the patients were in functional class 1, 22% in class2, and 5% in class 3.During the 72-hour ambulatory ECG screening

period, the median number of episodes of silentischemia was 6 (range, 1-21) with a mean of7.25 + 5.05 per patient, with a total duration of 133.5(range, 19-820) minutes and a mean of 175±162minutes. In contrast, the median number of episodesof angina was 0 (range, 0-5) with a mean of0.34±0.84 per patient, with a median duration of 0(range, 0-76) and a mean of 7.73 ± 16.6 minutesduring this period.

Ambulatory ECG MonitoringAmong the 60 patients studied, 10 became com-

pletely free of silent ischemic episodes with diltia-zem, 31 improved by 50% or more, and nine im-proved by less than 50%. The number of episodesremained unchanged in two patients and increased inthe remaining eight. The cumulative duration of STdepression exhibited a similar pattern.The median number of episodes was 4.5 (range,

0-19) with placebo and 1.5 (range, 0-13) with dil-tiazem administration. Means were, respectively,5.59±3.78 and 2.83+2.45, for a 50% reduction(p=O.OOOl) (Figure 1). The cumulative duration ofST depression was similarly reduced by 44%, from amedian of 78.5 (range, 0-60) minutes to 24.5 (range,0-411) minutes, with a mean of 119±107 to 67±92minutes (p=0.001). The beneficial effect of diltiazemwas maintained throughout the circadian cycle: thepercent reduction in episodes was 54% from mid-night to 6:00 AM, 48% from 6:00 AM to noon, 57%from noon to 6:00 PM, and 47% from 6:00 PM tomidnight.Average heart rate during the 72-hour placebo

monitoring period was 74.3+14.3 beats/min com-pared with 68.5± 12.9 during the diltiazem treatmentperiod (p<0.001). This reduction in heart rate wasmaintained around the clock, being only slightly less

during the night compared with that during the day.The distribution of episodes of ST depression duringboth placebo and diltiazem treatment phases did notparallel the absolute heart rate distribution but wasrelated to heart rate variation (Figure 2). Thus, theheart rate acceleration observed between 6:00 and9:00 AM was accompanied by an increasing incidenceof ischemic episodes and the slowing between 8:00 PMand midnight by a decreased incidence. With themore constant heart rate observed at other times,episodes of ST segment depression were stable, beingslightly less with slower heart rates.

Episodes of ST depression occurred at a slowerheart rate with diltiazem than with placebo treat-ment, 66.1 +-9.8 beats/min versus 74.3-+9.8 (p=0.001). Heart rate was, however, similar at the pointof maximal ST depression, 114.1±f 17.6 and116.9+16.4 beats/min. Blood pressure was recordedat the beginning and end of the monitoring period;systolic blood pressure was similar with placebo anddiltiazem treatment. 129+15 and 128±15 mm Hg,but diastolic blood pressure was slightly lower withdiltiazem, 77+9 versus 81±9 (p<0.002).The mean number of ventricular premature beats

per hour was 26.4±97.7 during screening, 31.3±111during placebo treatment, and 33.4±135 during dil-tiazem treatment. Ventricular tachycardia was ob-served in 14 patients during the screening phase, in12 patients during the placebo phase, and in onlyfour patients during the diltiazem phase (p<0.05).

Exercise Treadmill TestTable 2 lists the hemodynamic, clinical, and ECG

results of the exercise tests performed duringscreening and at the end of each treatment period.Heart rate and rate-pressure product both at restand at peak exercise were significantly lower withdiltiazem than with placebo treatment. ST depres-sion occurred in 54 patients during the placebo testand in 45 during the diltiazem test (p=0.03). Exer-cise was stopped because of angina or ST depres-

by guest on July 14, 2018http://circ.ahajournals.org/

Dow

nloaded from

Theroux et al Diltiazem and Silent Ischemia 19

Heart rate (bpm)90 -

80 -

70 -

60 -

50

-_ Diltiazem-a- Placebo

0 12

Time of day

18 246

No. episodes ST segment depression

0,81

0,6

0,4

0,2

_ Diltiazem-a- Placebo

0 6 12 18

Time of dayFIGURE 2. Plots ofheart rate (top panel) and distribution ofepisodes of STdepression (bottom panel) during 24 hours with placeboand diltiazem treatment. Diltiazem reduced both variables evenly during 24 hours, except for a slightly greater effect on episodes of STsegment depression when the heart rate was faster and when episodes of ST depression were more frequent. The occurrence of STdepression correlated much better with heart rate changes than with absolute heart rates. Accelerating heart rate observed between 6:00and 9:00AM is accompanied by an increasing incidence ofepisodes and the slowing;from 8:00 PM to midnight by a decreasing incidence.At other times, when heart rate is more constant, episodes ofST segment depression are fewer with less variation in frequency.

sion in 35% of the placebo tests and in 20% of thediltiazem tests (p =0.05). Time to 1-mm ST depres-sion improved by 15% with diltiazem treatment,from 296+154 to 341±148 seconds (p=0.005). To-tal exercise duration improved by 6% with diltiazemtreatment, from 467± 156 to 494± 143 seconds(p=0.0004), and furthermore, the amount of ST

segment depression was reduced during the dil-tiazem tests, from 2.25+1.03 to 1.78+1.09 mm

(p=0.0002). No correlations existed between theimprovement in exercise tolerance and either meanheart rate reduction on the ambulatory ECG orreduction in the number of episodes of asymptom-atic ST depression.

24

A 9 9 5

by guest on July 14, 2018http://circ.ahajournals.org/

Dow

nloaded from

20 Circulation Vol 84, No 1 July 1991

TABLE 2. Hemodynamic, Clinical, and Electrocardiographic Results of Exercise Testing

Parameter Screening Placebo Diltiazem p

HemodynamicRestHeart rate (beats/min) 74.5+11.2 77.0+12.2 68.8+12.5 0.0001

Systolic BP (mm Hg) 134.3±13.7 128.7±14.6 128.4+15.4 0.9

Rate-pressure product 10,032+1,981 9,945+2,157 8,910+2,218 0.0001Peak exercise

Heart rate (beats/min) 140.6+18.6 143.7+17.3 134.7±17.5 0.0001

Systolic BP (mm Hg) 175.3+24.3 172.3±21.3 174.7±24.5 0.2Rate-pressure product 24,861±5,849 24,894+4,998 23,733+5,401 0.01

ClinicalST depression 60 (100) 54 (90) 45 (79) 0.03With angina 17 (28) 15 (28) 1(2) <0.0001

Without angina 43 (72) 39 (72) 44 (98) <0.0001Reason for stopping

Pain or ST depression 28 (46) 21 (35) 12 (20) 0.05Fatigue or dyspnea 31 (51) 39 (64) 48 (80) 0.05

Total exercise duration (sec) 420±121 467±156 494+143 0.0004

ElectrocardiographicTime to 1 mm ST depression (sec) 257±132 296±154 341±148 0.005Time to maximum ST depression 403+ 113 447+156 479+141 0.005Maximum ST depression (mm) 2.38±0.83 2.25±1.03 1.78±1.09 0.0002

Values are mean+SD. Values in parentheses are percentages.BP, blood pressure.p values compare placebo and diltiazem tests.

Clinical ObservationsAnginal pain and nitroglycerin use were infrequent

during the study. The median number of reportedchest pain episodes was 0 (range, 0-45) duringplacebo treatment with 0 (range, 0-26) nitroglycerintablets consumed. Respective means were 2.8+7.2and 1.66±5.2. Patients reported 50% fewer episodeswith a median of 0 (range, 0-21) and a mean of1.43±3.3 during diltiazem treatment, and they re-

ported 58% less nitroglycerin consumption with a

median of 0 (range, 0-10) tablets and a mean of0.69±1.87 tablets. These small numbers preclude anypowerful statistical analysis.During the diltiazem exercise test, angina occurred

in only one of the 45 patients with ST depressioncompared with 15 of 54 patients during the placebotest (2% versus 28%, p<0.0001).Of the 78 patients evaluated for adverse effects,

including the 18 omitted from the efficacy analysis, oneor more symptoms were reported during the diltiazemphase in 25 (32%) and during the placebo phase in 18(23%). Most of these symptoms were minor and notnecessarily drug related. The most frequently encoun-tered in the diltiazem and placebo phases, respectively,were headache (13 and eight), fatigue (five and two),edema (four and three), nausea (four and three), anddizziness (three and two). During diltiazem therapy,intermittent periods of atrioventricular dissociationwere observed in two patients and of Mobitz type I

second-degree atrioventricular block in two others;these were asymptomatic.

In six patients, serious adverse events requiredtermination of the study. During the placebo periods,two patients experienced a myocardial infarction andanother died suddenly. During diltiazem therapy,one patient developed urticaria, one developed heartfailure, and another reported worsening angina.

DiscussionIn this study, diltiazem used as single therapy re-

duced by one half the frequency of silent ischemicepisodes occurring during daily life. Exercise test pa-rameters were also significantly improved, although themagnitude of the improvement was less than thatobserved during ambulatory ECG monitoring. Dil-tiazem also reduced angina attacks by one half, but thisdifference was not significant because angina was un-common in these patients. Thus, the benefit of dil-tiazem on silent episodes could not be readily inferredfrom the symptomatic response of patients.

Characteristics of Silent IschemiaAs shown in previous studies, most episodes of ST

depression occurring during daily life are not associatedwith symptoms.3'6"4-17 The 95% proportion of silentepisodes in our study was somewhat higher than the75-85% rates usually reported.3'6'14-17 The total num-ber of episodes, as well as the frequency of angina,during placebo treatment was also in the low range ofpreviously reported values.6"1"4-16 The duration ofepisodes of ST depression was, however, near themedian.26'17.'9 Our study population was selected to

by guest on July 14, 2018http://circ.ahajournals.org/

Dow

nloaded from

ThUroux et al Diltiazem and Silent Ischemia 21

include a relatively homogeneous group of patientswith stable coronary artery disease, effort angina, apositive exercise test, and silent ST depression onambulatoryECG monitoring to minimize the possibilityof detecting a differential effect of diltiazem because ofdifferent population subsets. Most of the patients werenonsmoking men with multivessel disease.The methodology and diagnostic criteria were similar

to that used in most previous studies. Defining thebeginning and the end of an episode as the point ofchange from baseline, rather than at 0.5 or 1 mm,affects the duration of an episode but not the numberof episodes. A 72-hour monitoring period was selectedbecause of the variability in the frequency and durationof episodes of ST depression.6-8 Previous studies haveestimated sample size requirements for various degreesof therapeutic efficacy at different levels of power.7Longer monitoring periods with shorter intervals be-tween them reduce sample size requirements. Thelarge number of patients enrolled in our study and thelong monitoring periods thus allow high statisticalpower to detect a beneficial effect.

Effect of DiltiazemPrevious studies indicate that long-acting ni-

trates, 14,17,22 3-blockers,16"18-21 and calcium antago-nists21-23 may be useful to control silent ischemia.

Sustained-release diltiazem was selected for thisstudy because of its favorable effect on heart rate andbecause of our previous finding of a sustained pro-tective effect against spontaneous angina and ofincreased exercise performance 12 hours after doseadministration.29 The improvement in exercise testparameters in this study, with a reduction in heartrate and rate-pressure product at peak exercise, issimilar to what we reported previously.29The 50% reduction in episodes of ST depression is

consistent with the results of the retrospective study byFrishman and Teicher21 and with studies comparingdiltiazem with propranolol20 and with atenolol.2 Dif-ferent results were noted in a preliminary report of arecent double-blind randomized trial using proprano-lol, diltiazem, and nifedipine.27 In that report, thenumber of episodes of ST depression was reduced by10% with diltiazem, a difference that was not signifi-cant, and by 52% with propranolol. Also, nifedipinehad no detectable effect, and only diltiazem signifi-cantly prolonged total exercise duration. Why diltiazemwas less effective in preventing silent ischemic episodes,compared with our study, is uncertain.

Mechanism ofActionThe pathophysiological mechanisms involved in

asymptomatic ST segment depression occurring dur-ing daily life, and the respective roles of increasedmyocardial oxygen demand or of decreased supplyare as yet incompletely defined. The study does notallow an accurate characterization of myocardialoxygen demand because blood pressure and theinotropic state of the left ventricle were not contin-

may be relevant to the understanding of the mecha-nisms of silent ischemia.

First, ST segment depression occurred at a slowerheart rate with diltiazem than with placebo treatment,and no circadian variation in the efficacy of diltiazemwas found; the drug reduced both heart rate andischemic episodes in a parallel and sustained fashionthroughout the day. Second, the ischemic-type STdepression on ambulatory ECG monitoring was muchmore responsive to the drug than were any of theparameters measured during exercise testing. Third,the frequency of ST depression appeared to be modu-lated more by changes in heart rate than by theabsolute heart rate. An accelerating heart rate was

associated with a higher frequency of ischemic epi-sodes, and the reverse was also true. When heart ratewas relatively constant, episodes were less frequent.These observations suggest that one of the mechanismsinvolved in the pathogenesis of silent ischemia is a rapidrate of change in myocardial oxygen demand and thatdiltiazem has a beneficial effect by attenuating rapidincreases in demand. Similar observations were re-

cently made with ST depression occurring during tread-mill exercise testing; with lower intensity exercise pro-

tocols, the onset of ST segment depression was delayedand occurred at a lower heart rate, documenting theimportance of work load.31,32The relation between heart rate reduction and

efficacy is, however, not clear. In the study by Khurmiet a120 using propranolol and in the study by Bonettiet a130 using atenolol, both diltiazem and the/3-blocker were equally effective in preventing isch-emia even though their effect on heart rate wasmarkedly different. These results may reflect theimportance of myocardial oxygen supply and alsoindicate the complexity of the interaction betweensupply and demand in response to various stimuli. Itcan be hypothesized that a common pathophysiolog-ical mechanism could lead to both an increase indemand and a reduced supply and that diltiazemcould interact with this mechanism at different levels.

Exercise-induced ST depression was less frequentand, as judged by the maximum depth of ST depres-sion, less severe during diltiazem compared withplacebo exercise tests. Furthermore, during diltiazemexercise tests, ST depression was almost always silent(44 of 45 patients). There are two potential explana-tions for this finding. First, Margonato et a133 sug-gested in a preliminary report that diltiazem mayhave analgesic properties because it increased thepain threshold to three different types of stimuli. Asecond more likely explanation is that when diltiazemdoes not completely eliminate exercise-induced isch-emia, it at least decreases its extent and severity andthat the severity of ischemia is an important deter-minant of whether an episode is silent or painful.

Implications of the StudyThis study demonstrates that sustained-release dil-

tiazem can greatly reduce the frequency of silentuously monitored. Some of the findings, however, ischemic episodes, thus extending the previously

by guest on July 14, 2018http://circ.ahajournals.org/

Dow

nloaded from

22 Circulation Vol 84, No 1 July 1991

well-documented antianginal activity of the drug.Whether medical treatment of silent ischemia im-proves prognosis has not been established but iscurrently being investigated in clinical trials.Whether the goal of therapy should be the elimina-tion of all episodes, or whether a reduction in thenumber of episodes is helpful, is also unknown.These issues remain to be elucidated before recom-mending routine treatment of silent ischemia.

AcknowledgmentWe thank Luce Begin for her excellent secretarial

work.

References1. Langow RA, Huang EK, Kelley MJ, Cohen LS: Predictive

accuracy of coronary artery calcification and abnormal exer-

cise test for coronary artery disease in asymptomatic men.

Circulation 1980;62:1196-12032. Mulcahy D, Keegan P, Crean P, Quyyumi A, Shapiro L,

Wright C, Fox K: Silent myocardial ischemia in chronic stableangina: A study of its frequency and characteristics in 150patients. Br Heart J 1988;60:417-423

3. Cohn PF: Silent myocardial ischemia. Ann Intern Med 1988;109:312-317

4. Deanfield JE, Shea M, Kensett M, Horlock P, Wilson RA,DeLandsheere C, Selwyn AP: Silent myocardial ischemia dueto mental stress. Lancet 1984;2:1001-1005

5. Chierchia S, Lazzari M, Freedman B, Brunelli C, Maseri A:Impairment of myocardial perfusion and function duringpainless myocardial ischemia. J Am Coll Cardiol 1983;1:924-930

6. Tzivoni D, Gavish A, Benhorin J, Bonai S, Keren A, Stern S:Day-to-day variability of myocardial ischemia episodes incoronary artery disease. Am J Cardiol 1987;60:1003-1005

7. Nabel EG, Barry J, Rocco MB, Campbell S, Mead K, FentonT, Orav EJ, Selwyn AP: Variability of transient myocardialischemia in ambulatory patients with coronary artery disease.Circulation 1988;78:60-67

8. Nademanee K, Christenson PD, Intarachot V, Robertson HA,Vaghaiwalla M: Variability of indexes for myocardial isch-emia: A comparison of exercise treadmill test, ambulatoryelectrocardiographic monitoring and symptoms of myocardialischemia. JAm Coll Cardiol 1989;13:574-579

9. Gottlieb SO, Gottlieb SH, Achuff SC, Baumgardner R, MellitsED, Weisfeldt ML, Gerstenblith G: Silent ischemia on Holtermonitoring predicts mortality in high-risk postinfarctionpatients. JAMA 1988;259:1030-1035

10. Gottlieb SO, Weisfeldt ML, Ouyang P, Mellits ED, Gersten-blith G: Silent ischemia as a marker for early unfavorableoutcomes in patients with unstable angina. N Engl J Med1986;314:1214-1219

11. Rocco MB, Nabel EG, Campbell S, Goldman L, Barry J,Mead K, Selwyn AP: Prognostic importance of myocardialischemia detected by ambulatory monitoring in patients withstable coronary artery disease. Circulation 1988;78:877-884

12. Hebdlad B, Juul-Moller S, Svensson K, Hanso BS, IsacssonSO, Janzon L, Lindell SE, Steen B, Johansson BW: Increasedmortality in men with ST-segment depression during 24-hourambulatory long-term ECG recording: Results from prospec-tive population study "men born in 1914" from Malmo,Sweden. Eur Heart J 1989;10:149-158

13. Deedwania PC, Carbajal EV: Silent ischemia during daily lifeis an independent predictor of mortality in stable angina.Circulation 1990;81:748-756

14. Schang SJ, Pepine C: Transient asymptomatic ST-segmentdepression during daily life activity. Am J Cardiol 1977;39:396-402

15. Checchi AC, Dovellini EV, Marchi F, Pucci P, Santoro GM,Fazzini PJ: Silent myocardial ischemia during ambulatoryelectrocardiographic monitoring in patients with effort angina.JAm Coll Cardiol 1983;1:934-939

16. Quyyumi AA, Wright C, Mockus L, Shackell M, Sutton GC, FoxKM: Effect of combined alpha- and beta-adrenoreceptor block-ade in patients with angina pectoris: A double-blind studycomparing labetolol with placebo. Br Heart J 1985;53:47-52

17. Shell WE, Kirowitz CF, Rubine SB, See J: Mechanisms andtherapy of silent myocardial ischemia: The effect of transder-mal nitroglycerin. Am Heart J 1986;212:222-229

18. Khurmi NS, Bowles MJ, O'Hara MJ, Raffery EB: Effect ofpropranolol on indices of intermittent myocardial ischemia,assessed by exercise testing and ambulatory ST-segment mon-itoring. Clin Cardiol 1986;9:391-397

19. Imperi GA, Lambert CR, Coy K, Lopez L, Pepine CJ: Effectsof titrated beta blockade (metoprolol) on silent myocardialischemia in ambulatory patients with coronary artery disease.Am J Cardiol 1987;60:519-524

20. Khurmi NS, O'Hara MJ, Bowles MF, Raffery EB: Effect ofdiltiazem and propranolol on myocardial ischemia duringunrestricted daily life in patients with effort-induced chronicstable angina pectoris. Eur J Clin Pharmacol 1987;32:443-447

21. Frishman WH, Teicher M: Anti-anginal drug therapy for silentmyocardial ischemia. Med Clin North Am 1988;72:185-196

22. Von Arnim TV, Erath A: Nitrates and calcium antagonists forsilent myocardial ischemia. Am J Cardiol 1988;61:15E-18E

23. Cohn PF, Vetrovec GW, Nesto R, Gerber FR, and the TotalIschemia Awareness Program Investigators. Am J Cardiol1989;63:534-539

24. Oakley GDG, Fox KM, Dargie HF, Selwyn AP: Objectiveassessment of treatment in severe angina. Br Med J 1979;1:1540-1541

25. Deedwania PC, Carbajal EV: Prevalence and patterns of silentmyocardial ischemia during daily life in stable angina patientsreceiving conventional antianginal drug therapy. Am J Cardiol1990;65:1090-1096

26. Josephson MA, Nademanee K, Intarachot V, Lewis H, SinghBN: Abolition of Holter monitor-detected silent myocardialischemia after percutaneous transluminal coronary angiogra-phy. JAm Coll Cardiol 1987;10:499-503

27. Stone PH, Gibson RS, Glasser SP, DeWood MA, Parker JD,Kawanishi DT, Crawford MH, Messineo FC, Shook TL, RabyK, Curtis DG, Hoop RS, Young PM, Braunwald E, and theASIS Study Group: Comparison of propranolol, diltiazem, andnifedipine in the treatment of ambulatory ischemia in patientswith stable angina: Differential effects on ambulatory isch-emia, exercise performance, and anginal symptoms. Circula-tion 1990;82:1962-1972

28. Conover WJ, Iman RL: Rank transformations as a bridgebetween parametric and non parametric statistics. Am Statist1981;35:124-133

29. Klinke WP, Juneau M, Grace M, Kostuk WJ, Pflugfelder P,Maranda CR, Warnica W, Chin C, Annable L, Dempsey E,Waters D: Usefulness of sustained-release diltiazem for stableangina pectoris. Am J Cardiol 1989;64:1249-1252

30. Bonetti F, Pancaldi A, Rossetti A, Margonato SL, ChierchiaSL: Differential action of betablockers and Ca-antagonists inpreventing "daily-life" ischemia. EurHeart J 1989;10(suppl):53

31. Heller GV, Ahmed I, Tilkemeier P, Warren LS, Carleton RA,Garber CE: Silent myocardial ischemia: A function of work-load? (abstract) JAm Coll Cardiol 1990;15:45A

32. Panza JA, Diodati J, Callahan TS, Quyyumi AA: Prediction ofthe magnitude of silent ischemia during daily life by stress testin patients with coronary artery disease: Importance of theexercise protocol (abstract). JAm Coll Cardiol 1990;15:81A

33. Margonato A, Fragasso G, Rossetti E, Macchi A, Carandente0, Costantino S, Chierchia SL: Intravenous diltiazem modifiessomatic pain perception in man (abstract). Circulation 1988;78(suppl II):II-453

KEY WORDS * diltiazem * silent myocardial ischemia -

ambulatory electrocardiographic monitoring

by guest on July 14, 2018http://circ.ahajournals.org/

Dow

nloaded from

Chin and E DempseyP Théroux, M Baird, M Juneau, W Warnica, P Klinke, W Kostuk, P Pflugfelder, E Lavallée, C

occurring during daily life and during exercise.Effect of diltiazem on symptomatic and asymptomatic episodes of ST segment depression

Print ISSN: 0009-7322. Online ISSN: 1524-4539 Copyright © 1991 American Heart Association, Inc. All rights reserved.

is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation doi: 10.1161/01.CIR.84.1.15

1991;84:15-22Circulation. 

http://circ.ahajournals.org/content/84/1/15the World Wide Web at:

The online version of this article, along with updated information and services, is located on

  http://circ.ahajournals.org//subscriptions/

is online at: Circulation Information about subscribing to Subscriptions: 

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

  document. Permissions and Rights Question and Answer information about this process is available in the

located, click Request Permissions in the middle column of the Web page under Services. FurtherEditorial Office. Once the online version of the published article for which permission is being requested is

can be obtained via RightsLink, a service of the Copyright Clearance Center, not theCirculationpublished in Requests for permissions to reproduce figures, tables, or portions of articles originallyPermissions:

by guest on July 14, 2018http://circ.ahajournals.org/

Dow

nloaded from