Intermittent Transdermal Nitroglycerin Therapy in Angina Pectoris
Clinically Effective Without Tolerance or Rebound
Background The objectives of this study were to assess the antianginal and anti-ischemic effects of three dose levels of transdermal nitroglycerin patches applied for 12 hours daily for 30 days. The study also assessed the development of tolerance and rebound. Intermittent transdermal nitroglycerin therapy with a patch-free period of 10 to 12 hours each day has documented clinical benefits during the period of patch application, but studies have failed to clearly document prolonged exercise duration for the entire period of patch application. This study was designed to evaluate the efficacy and duration of action of a range of doses of nitroglycerin. The study also permitted the assessment of the maintenance of initial effects, the development of tolerance, and the presence of rebound.
Methods and Results This study was a multicenter, randomized, double-blind, placebo-controlled parallel design trial with treadmill exercise tests at days 0, 1, 7, 15, and 30. Tests were carried out up to 12 hours after patch application. There was a statistically significant treatment effect with increases in treadmill walking time to moderate angina in each nitroglycerin patch group compared with placebo at various time points up to 12 hours throughout the 30-day study period. Secondary efficacy parameters, including the consistent increase in time to 1-mm ST-segment depression, supported the primary efficacy results. There was no evidence of tolerance or rebound.
Conclusions Intermittent transdermal nitroglycerin therapy increases exercise duration and maintains anti-ischemic effects for 12 hours after patch application, throughout 30 days of therapy, without significant evidence of nitrate tolerance or rebound phenomena.
The organic nitrates continue to be of major therapeutic importance in the management of all phases of coronary artery disease despite the availability of newer anti-ischemic agents. While nitrates have traditionally been administered by the sublingual, oral, or intravenous routes, transdermal nitroglycerin preparations have been used with increasing frequency. Nitroglycerin ointment has been shown to be effective in patients with angina pectoris when administered acutely,1 but studies during sustained therapy are limited.2 Transdermal nitroglycerin patches have been used in North America for more than a decade, and until recently, there was substantial controversy regarding their effectiveness. Continuous nitroglycerin patch application has been shown to have little clinical benefit because of the rapid development of nitrate tolerance.3 4 5 6 Studies during intermittent therapy with a patch-free period of 10 to 12 hours each day have documented clinical efficacy but not for the entire period of patch application.7 8 With greater understanding of the mechanisms responsible for nitrate tolerance and in particular, documentation of important neurohormonal activation9 during nitrate therapy, there has been concern regarding the potential for rebound after patch removal during intermittent therapy. Rebound is a possible explanation for the reported increase in rest angina in some individual patients during the patch-off period while patients were receiving intermittent therapy with nitroglycerin.7 10 The observation in one study that exercise duration before the morning patch application (hour 0) declined throughout the study period in patients receiving active therapy while it increased for patients receiving placebo may represent rebound.7
The present study was designed to assess the antianginal and anti-ischemic efficacy of each of three doses of transdermal nitroglycerin, compared with placebo, during 30 days of intermittent patch therapy. Importantly, the study also provided the opportunity to assess whether tolerance developed during the study period and whether rebound effects occurred during periods of patch removal.
This was a multicenter, randomized, double-blind, placebo-controlled parallel design study of 30 days’ duration.
Male and female patients, 25 years of age or older, with stable angina pectoris present for at least 3 months, were eligible for this investigation. This study was approved by the ethics committees of all participating institutions, and each patient provided written informed consent before enrollment.
The patients had to experience angina during exercise testing. In male patients with a classic history of angina, the presence of horizontal or downsloping ST-segment depression of ≥1 mm when measured 80 milliseconds after the J-point was considered sufficient evidence of coronary disease and myocardial ischemia to be included in the study. Female patients and patients with nonanalyzable ST segments were included if they had angina during treadmill testing and had at least one of the following: angiographic evidence of significant coronary artery disease, previous myocardial infarction, positive thallium stress test, positive exercise radionuclide ventriculogram, or a positive stress echocardiogram.
Patients taking β-adrenergic blocking agents were allowed to continue these medications, but the daily dose was withheld until after the final exercise test on each study day. Calcium channel blockers, digoxin, and nitrates other than sublingual nitroglycerin were not permitted during the study period. Patients with myocardial infarction or unstable angina or those who underwent revascularization procedures within 3 months were excluded, as were patients with significant valvular heart disease, uncontrolled hypertension, or congestive heart failure.
Patients who entered the qualification period had to complete a series of treadmill exercise tests (TETs) using either the standard11 or abridged* Bruce protocol. Moderate angina, which was defined as the level of discomfort at which the patient would normally discontinue activity, had to develop within 3 to 7 minutes of the start of the first TET. One hour later, and 5 minutes after taking 0.4 mg of sublingual nitroglycerin, exercise was repeated, and patients had to demonstrate at least a 20% increase in the time to moderate angina. On fulfilling these criteria, the patients continued in the qualifying period, and in a single-blind fashion, wore three placebo patches for 12 hours each day (8 am to 8 pm). Patient diaries were completed during this qualifying period and the subsequent treatment phase to determine the number of angina attacks and the frequency of nitroglycerin consumption occurring during the patch-on and patch-off periods. Patients subsequently were required to have three consecutive TETs, during which exercise duration to moderate angina was within 15% of the preceding test and was reached between 2.5 to 7.5 minutes. This qualified the patient for entry into the double-blind treatment phase of the study.
Upon demonstrating TET reproducibility, the patient had additional TETs at 4, 8, and 12 hours on that day (day 0). During the interval between day 0 and day 1 (1 to 10 days), patients continued to wear placebo patches intermittently. On day 1, after the hour 0 TET, patients were randomized to receive either placebo patches or nitroglycerin patches delivering 0.2 mg/h, 0.4 mg/h, or 0.8 mg/h†, with no individual dose titration. The nitroglycerin patches used were 0.2 mg/h and 0.4 mg/h, and patients wore a combination of active and/or placebo patches to provide the randomized dose. The patches were removed for 12 hours overnight during the double-blind treatment phase, and new patches were applied each morning.
During the double-blind treatment phase, TETs were performed on days 1, 7, 15, and 30. On days 1 and 30, TETs were performed just before patch application (hour 0) and 4, 8, and 12 hours after patch application. On days 7 and 15, TETs were performed just before patch application and at hour 4, the point at which the peak nitrate effect was anticipated.
Sample Size Estimation
Assuming a standard deviation of 60 seconds, it was determined that 60 completed patients per treatment group would be required to provide 80% power to detect a difference of 10% in the time to moderate angina between any given dose of nitroglycerin and placebo. The α-level for the three comparisons of interest in the trial was set at 0.017 to adjust for multiple comparisons. Two-tailed comparisons were planned, and all statistical tests were performed using sas version 6.07.12 13 14
Treatment Group Comparability
Overall treatment differences in patient demographics, cardiovascular history, concomitant β-adrenergic blocker use, and baseline exercise test results were statistically analyzed using two-way ANOVA (treatment and center, proc glm), χ2 (proc freq), or exact probability tests (proc freq). A two-sided P value of .05 or less was considered statistically significant.
The primary measure of efficacy in this study was the treadmill walking time (TWT) to moderate angina. If moderate angina was not experienced during a test, the total exercise duration was substituted for that value. Secondary measures of efficacy included time to the onset of angina, time to the development of 1-mm ST-segment depression, frequency of angina, and the use of sublingual nitroglycerin.
TET parameters at each testing time were compared with hour 0 on each study day. The changes for each treatment group were statistically analyzed using a two-sided, paired t test as generated using proc means.
Comparisons With Day 0
Treatment effect compared with placebo was defined as the difference between within-day changes in TWT on each study day and the within-day changes seen on day 0. This treatment effect was compared among treatment groups using a two-way ANOVA (proc glm) with treatment and center. Adjusted treatment means were determined (lsmeans option in glm), and pairwise comparisons of each dose level of nitroglycerin and placebo were tested using Dunnett’s multiple comparison test. In addition, an intent-to-treat analysis, which included all patients randomized to treatment, also was performed on this primary efficacy measure.
Maintenance of Treatment Effect
The treatment effect at hour 4 on days 1, 7, 15, and 30 was examined using a profile analysis to determine if the magnitude of treatment effect was maintained. Repeated-measures analysis with linear, quadratic, and cubic contrasts of the day effect was used to quantify the change in treatment effect over the 30 days of therapy.
Time to onset of angina and the time to 1-mm ST-segment depression were analyzed as previously described for within-day changes and day 0 comparisons. Analyses of within- and between-treatment group differences were performed for frequency of angina and sublingual nitroglycerin use. Separate analyses were performed for the patch-on and the patch-off periods using the corresponding qualification period as a baseline.
In this study, 325 patients at 32 investigational sites entered the single-blind placebo qualifying period. Of these, 34 (10%) did not qualify. Two hundred ninety-one patients qualified and were randomized to receive double-blind medication (Table 1⇓). The demographics of the treatment groups were similar, with the exception that there was a small but statistically significant difference in age between the 0.4 mg/h (62.7 years) and 0.8 mg/h (59.0 years) nitroglycerin groups. β-Blocker use did not differ significantly among treatment groups, with approximately 60% of patients in each group receiving concomitant β-blockers.
Fifty-one patients were excluded from the efficacy analyses (Table 2⇓). Twenty-one patients were discontinued from the study due to adverse events, the most common of which was headache. Not unexpectedly, a higher proportion of patients receiving 0.8 mg/h were discontinued due to nitroglycerin-related adverse events.
Day 0 (Placebo Day)
All patients received placebo patches in a single-blind manner on day 0. During the course of this placebo day, TWT times increased at hour 4 for all groups and progressively decreased throughout the rest of the day.
After the first double-blind patch application on day 1, all groups increased exercise duration at hour 4 in comparison to hour 0 on that day. Subsequently, the 0.4 mg/h and 0.8 mg/h groups showed significant improvement at 8 and 12 hours, while there was a significant decrease in TWT in those receiving placebo.
Days 7 and 15
On days 7 and 15, significant increases relative to hour 0 occurred at hour 4 in all three active treatment groups (P<.001), while TWT decreased in the placebo group.
On day 30, each of the three active treatment groups showed significant improvement over hour 0 at hour 4 (P<.001), and the 0.4 mg/h and 0.8 mg/h groups showed significant increases at hour 8 (P<.001). At hour 12, the increase in TWT compared with hour 0 was similar for the two higher-dose groups and was significant in the 0.8 mg/h nitroglycerin group and approached significance (P<.056) in the 0.4 mg/h group. In contrast, there was a progressive decrease in TWT in the placebo group that reached significance at hour 12 (P<.001).
Comparisons With Day 0
This measure utilizes the pretreatment placebo reference (day 0) to compare changes seen during the treatment period at days 1, 7, 15, and 30 in each dose group. This forms the basis for comparison with the placebo group. The results demonstrated a significant (P≤.017) treatment effect, with increases in TWT to moderate angina in all three treatment groups compared with placebo at various time points (Fig 2⇓). It is of importance that on day 30, all three nitroglycerin groups showed significant increases in TWT at hour 12, and there was a decrease in TWT in the placebo group.
Maintenance of Treatment Effect
Maintenance of clinical efficacy throughout the 30-day study period was assessed by comparing the magnitude of the treatment effects at hour 4 on days 1, 7, 15, and 30. Over the 30-day treatment period, the therapeutic effect of transdermal nitroglycerin was maintained. Importantly, the significant increases in TWT at 8 and 12 hours on day 1 were maintained on day 30, further supporting the maintenance of treatment effect (Fig 3⇓).
Secondary Efficacy Parameters
Onset of Angina
TWT to the onset of angina increased at every time point on every study day for all three active treatment groups. These treatment effects were significant at hour 4 on all days except for the 0.4 mg/h group on day 1.
Comparisons with day 0 in the TWT to the onset of 1-mm ST-segment depression are shown in Fig 4⇓. Throughout the study period, the majority of time points shows significant differences from placebo (P≤.012). It is of importance that after 30 days, all three nitroglycerin groups showed significant prolongation of TWT to 1-mm ST-segment depression.
Angina Frequency and Nitroglycerin Use
Only patients with stable, reproducible, effort-induced angina were eligible to enter the study, and consequently, these patients had reasonably low frequencies of angina (mean, 1.7 attacks per week) and sublingual nitrate use (mean, 0.8 tablets per week) during the patch-on phase of the qualifying placebo period. During double-blind treatment, the frequency of angina attacks during patch-on periods was decreased significantly relative to the qualification phase in all four treatment groups (means, −0.9, −0.7, −0.8, and −0.4 attacks per week in the placebo, 0.2 mg/h, 0.4 mg/h, and 0.8 mg/h treatment groups, respectively). There were no significant differences detected between nitroglycerin treatment groups and placebo. Corresponding decreases in sublingual nitroglycerin use also were shown (means, −0.3, −0.4, −0.4, and −0.3 tablets per week in the placebo, 0.2 mg/h, 0.4 mg/h, and 0.8 mg/h treatment groups, respectively), with no significant differences between the nitroglycerin and placebo groups.
As expected, the number of anginal attacks (mean, 0.6 attacks per week) and frequency of sublingual nitroglycerin use (mean, 0.4 tablets per week) was lower in the overnight patch-off period than the daytime patch-on periods for all groups in the qualifying period, and this continued during the double-blind study periods. The changes in anginal attacks (means, −0.2, 0.3, 0.2, and 0.2 attacks per week in the placebo, 0.2 mg/h, 0.4 mg/h, and 0.8 mg/h treatment groups, respectively) and sublingual nitroglycerin use (means, −0.04, 0.2, 0.2, and 0.4 tablets per week in the placebo, 0.2 mg/h, 0.4 mg/h, and 0.8 mg/h treatment groups, respectively) during patch-off periods showed no significant differences between the placebo and active treatment groups.
Hour 0 Results
During the course of the double-blind treatment phase, there was an increase in TWT at the end of the patch-off period (hour 0) on days 1, 7, 15, and 30 relative to that observed at hour 0 on day 0. This increase was seen for each treatment group, with a somewhat greater increase seen in the placebo group. The 0.2 mg/h group showed significant differences from the placebo group. Apart from day 7 in the 0.4 mg/h group, the 0.4 mg/h and 0.8 mg/h groups showed no significant changes in TWT at hour 0 from that seen in the placebo group.
Transdermal nitroglycerin patches were well tolerated during the study period. Twenty-one (7.2%) patients who received double-blind treatment were discontinued due to adverse advents. These consisted primarily of headache. Two patients, while receiving placebo therapy, had a myocardial infarction. One patient in the 0.4 mg/h group had unstable angina. Another patient in the 0.4 mg/h group had hypotension and bradycardia, probably related to nitroglycerin administration. A total of 103 patients (35%) experienced headaches, but in the majority of patients, this was successfully treated with acetaminophen, and these patients carried on with the study. Seventeen patients (6%) had local skin reactions at the application site, but in no case did this lead to patient discontinuation. There were no patient deaths during the study.
This study confirms that intermittent therapy with transdermal nitroglycerin is effective in improving exercise performance in patients with chronic stable angina pectoris. Each of the nitroglycerin groups was superior to placebo with regard to efficacy, while greater treatment effects were seen with the higher doses (Fig 2⇑). A significant treatment effect was observed for each group for 12 hours at day 30. Exercise tests carried out 4 hours after patch application, the time of maximal drug effect, showed no significant change in the magnitude of treatment effect over the 30-day study period. These findings confirm continued clinical efficacy throughout the 12 hours of application during intermittent transdermal therapy and the absence of significant tolerance during the treatment period.
The clinical efficacy of each dose of nitroglycerin was confirmed by the finding that the time to 1-mm ST-segment depression was consistently prolonged throughout the 30-day study in each active treatment group.
The doses used in this study were similar to those of a previous large multicenter trial with intermittent transdermal nitroglycerin therapy.7 The treatment effect in that study was documented on day 1, but by day 29, efficacy could only be shown at 8 hours, and only in the larger doses.
It has been suggested that rebound phenomena may occur after patch removal during intermittent therapy. This was based in part on a study where some patients reported an increased frequency in rest angina during the patch-off period.7 Because of difficulties of the definition and documentation of true rest angina by patient diary, the total frequency of angina in the patch-off period was evaluated in the present study. The frequency of angina in the patch-off period was less than during the day in all study groups. There were no significant differences in angina frequency during patch-off periods in the active and placebo groups.
The other finding in a previous study suggestive of rebound was the so-called “zero-hour effect,” where exercise duration before the morning patch application declined slightly in the patients receiving active therapy while it increased in patients receiving placebo. In the present study, all groups showed improvement in the hour 0 exercise test throughout the 30-day study period. The improvement was similar for the 0.4 mg/h, 0.8 mg/h, and placebo groups, with the placebo group having a statistically significant difference from the 0.2 mg/h group. Because this difference did not occur with the larger doses, it is believed that this does not present a rebound phenomena. This conclusion is supported by the observation regarding the frequency of angina and nitroglycerin consumption during the patch-off period.
The results of this large multicenter study document that intermittent transdermal nitroglycerin is effective in the management of patients with chronic stable angina pectoris. There was prolongation of exercise duration to moderate angina and to 1-mm ST-segment depression in each dose group over a 12-hour period throughout the duration of the 30-day study. The study also showed that this treatment regime was not associated either with the development of tolerance or evidence of rebound during the period of patch removal.
Minitran Clinical Trial Study Group Study Centers
University of Ottawa Heart Institute, Ottawa, Ontario: Michael G. Baird, MD; Margaret Fraser, RN.
Oklahoma Cardiovascular and Hypertension Center, Oklahoma City, Okla: Steven Chrysant, MD; Catherine Chrysant, MD.
Portland Veterans Administration Medical Center, Portland, Ore: Henry DeMots, MD; Barbara Keeton.
Harbor-UCLA Medical Center, Torrance, Calif: Robert Detrano, MD, PhD; Barbara Vaitovas.
Medical Research Center, New Orleans, La: C. Andrew De Abate, MD.
Clinical Research, Internal Medicine Specialists, Brea, Calif: Michael Dougherty, MD; Barbara Manning.
Cardiovascular Center of Sarasota, Sarasota, Fla: Mahfouz A. El Shahawy, MD, FACC; June Wheeler.
Hospital of Albert Einstein, College of Medicine, Bronx, NY: William H. Frishman, MD; Suzanne Furia, RN, BSN.
Lawrence Clinical Research, Lawrenceville, NJ: W. Thomas Garland, MD; Jane Alliprandini.
Doctors’ Clinic Research, Vero Beach, Fla: Arthur L. Glaser, MD; Betsy A. Screws, RN.
Charleston Cardiology Associates, Charleston, SC: William J. Grossman, MD; Kathleen Fletcher, RN.
Centre Epic, Montreal, Quebec, Canada: Martin Juneau, MD; Lucie Larivee.
Osteopathic Medical Center of Philadelphia, Philadelphia, Pa: Joseph S. Kenney, DO.
The Christ Hospital, Cardiovascular Research Center, Cincinnati, Ohio: Dean J. Kereiakes, MD; Linda Martin, RN; Linda Anderson, RN.
Royal Alexandra Hospital, Edmonton, Alberta, Canada: W. Peter Klinke, MD; Linda Kvill.
University Hospital, London, Ontario, Canada: William Kostuk, MD; Rita Kennedy.
Royal University Hospital, Saskatoon, Saskatchewan, Canada: J.F. Lopez, MD; Patricia Kuny, RN.
Cardiology Consultant Laboratory, Calgary, Alberta, Canada: Patrick Ma, MD; Neil G. Filipchuk, MD; Mag Grose, RN.
Queen Elizabeth Hospital, Montreal, Quebec, Canada: Claude Maranda, MD; Elizabeth Graham.
Kingston General Hospital, Kingston, Ontario, Canada: John O. Parker, MD; Bernice Farrell, RN.
Center for Clinical Research, Austin, Tex: Thomas Parker, MD; Tammy Watkins, RN.
Beverly Hills Cardiology Research, Beverly Hills, Calif: William E. Shell, MD; Angila Kapit.
Louisiana Cardiovascular Research Center, New Orleans, La: William B. Smith, MD; Tina Serpas, RN.
University of Alabama at Birmingham, Birmingham, Ala: James Taylor, MD; Cynthia Olsson.
Cardiology Center, Marrero, La: Craig Unger, MD; Phyllis Ponti.
St Luke’s Hospital, Denver, Colo: Nampalli Vijay, MD; Melinda Washam, RN.
Foothills Provincial Hospital, Calgary, Alberta, Canada: J. Wayne Warnica, MD; Brenda Smith.
University of Miami School of Medicine, Miami, Fla: Donald J. Weidler, MD, PhD; Nader S. Jallad, PhD.
Cardiology Research Associates, Ormond Beach, Fla: David L. Williams, MD; Diane Tracy, RN; Diane Martin, RN.
Clinical Physiology Associates, Fort Myers, Fla: Stephen R. Zellner, MD; David D. Michie, PhD; Glenda Younger.
This work was sponsored by a grant from 3M Pharmaceuticals, St Paul, Minn. The authors would like to acknowledge the dedication, writing, and editorial support provided by Altha Edgren, BS, as well as the editorial support provided by Carin Verduyn, MD, and graphic design work by Randy Bunde. Additionally, we wish to recognize the untiring work provided by the people at Bio-Pharm Clinical Services, Inc.
1 In the abridged protocol, stages 1 and 2 of the Bruce protocol were shortened to 90 seconds each. If the abridged protocol was used, all subsequent TETs were done with that technique.
2 Minitran (nitroglycerin) Transdermal Delivery System, 3M Pharmaceuticals, 6.7 cm2 (0.2 mg/h) and 13.3 cm2 (0.4 mg/h) patches.
- Received June 27, 1994.
- Revision received October 5, 1994.
- Accepted October 14, 1994.
- Copyright © 1995 by American Heart Association
Riechek N, Goldstein RE, Redwood DR, Epstein SE. Sustained effects of nitroglycerin ointment in patients with angina pectoris. Circulation. 1974;50:348-352.
Parker JD, Farrell B, Fenton T, Cohanim M, Parker JO. Counter-regulatory responses to continuous and intermittent therapy with nitroglycerin. Circulation. 1991;84:2336-2345.
Ferratini M, Pirelli S, Merlini P, Silva T, Pollavini G. Intermittent transdermal nitroglycerin monotherapy in stable exercise-induced angina: a comparison with a continuous schedule. Eur Heart J. 1989;10:998-1002.
SAS Institute Inc. sas® Language: Reference, Version 6, ed 1. Cary, NC: SAS Institute Inc; 1989:846.
SAS Institute Inc. sas/stat® User’s Guide, Version 6, ed 4, vol 1. Cary, NC: SAS Institute Inc; 1989:943.
SAS Institute Inc. sas/stat® User’s Guide, Version 6, ed 4, vol 2. Cary, NC: SAS Institute Inc; 1989:846.