Author + information
- Received April 14, 2000
- Revision received September 11, 2000
- Accepted October 16, 2000
- Published online February 1, 2001.
- Eduardo R Azevedo, MDa,
- Anne M Schofield, RN, BScNa,
- Susan Kelly, RN, BScNa and
- John D Parker, MD, FACCa,* ()
- ↵*Reprint requests and correspondence: Dr. John D. Parker, Mount Sinai Hospital, 600 University Avenue, Suite 1609, Toronto, Ontario M5G-1X5
We sought to determine whether nitroglycerin (NTG) withdrawal contributes to worsening of endothelial dysfunction and development of the rebound phenomenon during intermittent transdermal NTG therapy.
Intermittent transdermal NTG therapy is recommended to avoid the development of tolerance. However, this regimen may precipitate worsening angina in the NTG-free interval.
Twenty patients were randomized to intermittent transdermal NTG (0.6 mg/h; NTG group) or no treatment (control group) five days before angiography. The risk factors for endothelial dysfunction were similar in both groups. After diagnostic angiography, the patients underwent quantitative angiography before and after intracoronary acetylcholine (ACh), 10−4mol/liter. Immediately after the morning study, the patch was removed from the NTG group, and 3 h later, the ACh infusion was repeated in both groups. All patients had mild to moderate coronary artery disease (CAD).
The diameter of the left anterior descending coronary artery at baseline was 2.0 ± 0.1 mm in the control group and 2.6 ± 0.1 mm in the NTG group (p < 0.05). Acetylcholine caused mild vasoconstriction in the control group in the morning and afternoon (2.7 ± 5.3% and 2.4 ± 3.9%, respectively; p = NS). The NTG group demonstrated mild vasoconstriction to ACh in the morning (3.2 ± 2.8%; p = NS vs. control group). After patch removal, there was a significant increase in the magnitude of vasoconstriction in the NTG group (11.6 ± 3.9%, p = 0.04 vs. morning constriction).
These results confirm that NTG withdrawal increases the coronary vasomotor response to ACh in patients with mild CAD and suggests that the rebound phenomena may be secondary to the development of endothelial dysfunction after discontinuation of NTG therapy.
An intermittent dosing regimen with a daily drug-free interval is required to avoid the development of tolerance to transdermal nitroglycerin (NTG) (1–3). A potential complication of this drug-free period is the development of rebound myocardial ischemia during the nitrate-free interval (2,4,5). Indeed, withdrawal after exposure to NTG has been associated with angina pectoris, myocardial infarction and even sudden death (6,7).
Although the cause of nitrate tolerance remains unclear, several different mechanisms have been proposed (8). Recent evidence suggests that nitrate exposure is associated with increased vascular sensitivity to vasoconstrictors, a phenomenon which may be secondary to increased vascular production of free radicals (9)or endothelin, or both (10). In a previous study, we demonstrated that continuous therapy with transdermal NTG was associated with the development of endothelial dysfunction, as assessed by the coronary vasomotor response to acetylcholine (ACh) (11). In this study, withdrawal of NTG therapy for 3 h was associated with even greater evidence of endothelial dysfunction. The cause of NTG-induced endothelial dysfunction remains unknown; however, it may be associated with specific changes in vascular biochemistry, such as increased superoxide anion production or changes in vascular production of endothelin (9,10).
In the present study, we examined the impact of intermittent transdermal NTG therapy and the early nitrate-free interval on coronary endothelial function. We hypothesized that 1) endothelial dysfunction would not develop during the initial 12 h of transdermal NTG therapy; and 2) endothelial dysfunction, as manifested by an increased sensitivity to ACh-induced coronary vasoconstriction, would be observed during the period after removal of transdermal NTG.
Patients referred for diagnostic coronary angiography were eligible to participate in this study. Clinical exclusion criteria included previous coronary artery bypass graft surgery or percutaneous transluminal coronary angioplasty; unstable angina or myocardial infarction within 30 days of randomization; second- or third-degree atrioventricular block; and clinically significant renal or hepatic disease. At the time of catheterization, those with evidence of left main coronary artery (LMCA) atherosclerosis or significant coronary artery disease (CAD; stenosis ≥60%) involving the left anterior descending coronary artery (LAD) or circumflex coronary artery (Cx) were excluded from the study. Of the 48 patients enrolled in this study, 21 were excluded because of significant CAD involving the LMCA, LAD or Cx. Five patients with angiographically normal coronary arteries were also excluded. In two patients, the study was interrupted because of severe bradycardia or second-degree atrioventricular block, and in one patient, the protocol was stopped because of technical difficulties. All complications were treated promptly, and no patient required temporary pacing or atropine.
The patients were randomly assigned to receive intermittent transdermal NTG (Ciba-Geigy, Mississauga, Ontario), 0.6 mg/h, or no treatment (we could not obtain placebo NTG patches), in an investigator-blinded fashion. All patients taking long-acting nitrates were asked to stop this medication at the time of randomization. Other vasodilators were discontinued 48 h before the study. Patients were allowed to use sublingual NTG, as required for relief of angina, but no study was carried out within 6 h of NTG use. Starting five to seven days before the date of the study, patients in the transdermal NTG group were instructed to apply the transdermal NTG patch at 10:00 PMand to remove it every morning at 10:00 AM. On the day of the study, patients were admitted to the catheterization laboratory at 8:00 AMfor coronary angiography. In those patients randomized to NTG, the morning study was performed with the transdermal patch left in place. Coronary angiography was performed using 7F diagnostic catheters and the Judkin’s technique. Once angiographic suitability for the study had been confirmed, a single projection best demonstrating the proximal and mid-portion of the LAD was chosen. The study protocol included intracoronary infusions of 5% dextrose in water (D5W) and ACh chloride at 1.25 ml/min for 3 min in the following sequence: 1) control: D5W—the vehicle for ACh infusion; 2) two incremental ACh infusions at concentrations of 10−5and 10−4mol/liter to achieve intracoronary concentrations of 10−7and 10−6mol/liter, respectively, assuming blood flow in the left coronary artery of 125 ml/min; and 3) repeat control: D5W. Coronary angiography was repeated immediately after completion of each infusion. A 3-min period was allowed to elapse between each angiogram and the subsequent infusion to compensate for contrast medium-induced changes in coronary tone (12). After the completion of the morning study, the transdermal patch was removed from patients in the NTG group. The arterial and venous femoral sheaths were left in place, and all patients were transferred to the recovery area. Three hours later, the patients were returned to the catheterization room, where they received an identical intracoronary drug infusion protocol, using the same projection and radiographic technique, as described earlier.
The study protocol was approved by the Committee on Human Subjects Experimentation of the University of Toronto, and written, informed consent was obtained from all patients.
Left coronary artery angiograms were performed by power injection (Medrad, Pittsburgh, Pennsylvania) of 10 to 12 ml (3 to 4 ml/s) of nonionic contrast medium. Quantitation of coronary dimensions was performed using an automated edge-detection system (CMS, Neunen, The Netherlands) and techniques previously reported by our laboratory team (13). A single end-diastolic frame of each angiogram was selected by an investigator who had no knowledge of the study groups. The longest, most clearly visualized segment of the proximal to mid-LAD was selected. The mean lumen diameter of the LAD segment was recorded from each study angiogram. All films were analyzed by a technician who had no knowledge of the patients’ treatment status.
The patients’ baseline characteristics were compared using the unpaired ttest for continuous variables. The Fischer exact test was used to compare binary variables. Changes in the mean lumen diameter, from control values to each dose of ACh, were compared using multivariate analysis of variance within a general linear model procedure, with appropriate contrast statements (SAS version 6.11, SAS Institute Inc., Cary, North Carolina). This model allowed for analysis of intergroup changes in the responses to ACh for both the morning and afternoon angiograms. Furthermore, the differences between the groups in the overall response to ACh during the morning, versus the afternoon, could also be compared. Finally, the procedure also allowed for univariate comparisons at baseline and all subsequent time points between groups. A two-sided p value <0.05 was considered significant.
Ten patients were randomized to the non-NTG arm (nine men, 59 ± 4 years old) and 10 patients to NTG therapy (seven men, age 58 ± 3 years old). The groups were well matched in terms of baseline characteristics and risk factors for endothelial dysfunction (Table 1). All patients had CAD, as manifested by intimal irregularities or mild to moderate obstructive disease.
Heart rate did not change in either group in response to the infusion of ACh. In the non-NTG group, there was no change in blood pressure during the course of the study. A significant increase in systolic (from 133 ± 6 to 146 ± 7 mm Hg, control AMvs. control PM; p < 0.05) and diastolic blood pressure (from 69 ± 3 to 74 ± 3 mm Hg, control AMvs. control PM; p < 0.05) was observed after removal of the patch in the NTG group.
Responses of coronary arteries
At baseline, the mean diameter of the LAD in the non-NTG group was 2.02 ± 0.09 mm, which was significantly smaller than the 2.64 ± 0.10 mm observed in the NTG group (p < 0.05).
In the non-NTG group, the peak dose of ACh caused a very small reduction in the mean LAD diameter, from 2.02 ± 0.09 to 1.97 ± 0.14 mm (p = 0.11), which represents a mean vasoconstriction of only 2.7 ± 5.3% (Fig. 1). In the afternoon, the peak dose of ACh had an almost identical effect on the non-NTG group and caused a change in the mean lumen diameter of the LAD from 2.02 ± 0.10 to 1.98 ± 0.15 mm (p = 0.13), which also represents a nonsignificant vasoconstriction of 2.4 ± 3.9% (Fig. 1). There was absolutely no difference between the morning and afternoon responses to ACh (p = 0.93). In the NTG group, during the morning part of the study, while the transdermal NTG patch was still in place, there was only a very mild effect of the peak dose of ACh on the coronary vasomotor response. The mean lumen diameter of the LAD went from 2.64 ± 0.10 to 2.56 ± 0.12 mm (p = 0.67), which represents a mean vasoconstriction of 3.2 ± 2.8% (Fig. 1). During the afternoon protocol, 3 h after removal of the NTG patch, the mean diameter of the LAD was reduced to 2.17 ± 0.08 at baseline (p < 0.05 vs. morning control, NTG group). The intracoronary infusion of ACh caused a further reduction in the LAD diameter to 1.93 ± 0.12 mm (p < 0.05 vs. afternoon control, NTG group), which represents a vasoconstriction of 11.6 ± 3.9% (p = 0.04 vs. morning constriction, NTG group) (Fig. 1). Multivariate analysis of variance revealed that there was a highly significant difference between the responses of the NTG group and those of the non-NTG treatment group (p < 0.001). The reduction in the LAD diameter after patch removal, as well as the significant vasoconstrictive response to ACh observed in the afternoon, are illustrated in Figure 2.
The development of tolerance to continuous transdermal NTG therapy occurs rapidly after the initiation of therapy and carries important clinical implications. The only effective method of preventing tolerance to NTG has been the use of dosing schedules that provide low or absent plasma NTG levels for a portion of the day. A number of studies have determined that intermittent therapy with transdermal NTG is effective in the prevention of tolerance, and this regimen is approved and indicated for the treatment of angina. However, there are potential problems associated with intermittent NTG therapy. A patient receiving monotherapy with nitrates will have no coverage during the nitrate-free period and may require combined therapy with a beta-blocker or calcium channel blocker. In addition, there is also concern that anginal symptoms may be exacerbated during the nitrate-free period, because of a phenomenon known as “rebound ischemia” (6). Although the cause of rebound ischemia is uncertain, there is evidence to suggest that this is a clinically relevant problem.
For nearly 50 years, several studies have reported what was termed the “withdrawal hazards” of NTG. In most cases, this was manifested as worsening angina, myocardial infarction or sudden death among workers of the explosives industry after they had left the plant for two or three days (6,7). Given the nature of NTG exposure in these reports, it is difficult to extrapolate these observations to the clinical use of this compound. In more recent investigations of intermittent NTG therapy, an increase in angina and myocardial ischemia during the nitrate-free interval was seen (2,4,14). In one multicenter, randomized controlled trial evaluating the efficacy of intermittent transdermal NTG in patients with stable angina, 9 of 138 patients receiving active treatment fulfilled the defined criteria for a significant increase in rest angina after patch removal. None of the 68 patients in the control group developed rest angina during the patch-off interval (2). In another placebo-controlled, crossover design study comparing the effects of continuous versus intermittent NTG therapy in patients with stable angina, a higher incidence of anginal attacks was seen during the NTG-free interval (4). Similar clinical results were documented more recently by Pepine et al. (14). Although these reports are a cause of concern, no such effects were reported in other studies (15–17), including a large-scale trial of intermittent transdermal NTG therapy (3).
Intermittent transdermal NTG therapy has also been shown to have adverse effects on exercise performance during the nitrate-free interval. In patients with stable angina, intermittent transdermal NTG is associated with a decrease in the angina threshold for 4 to 6 h after discontinuation of therapy (5). In a similar study by Pepine et al., an increase in the frequency of angina during the patch-off interval was documented by patients, and this subjective finding was supported by a corresponding trend toward an increase in ambulatory electrocardiographic evidence of ischemia in this same period (14). DeMots and Glasser (2)demonstrated that 12 h after the removal of the transdermal NTG or placebo patch, the placebo group was able to exercise longer than the group receiving active therapy—a phenomenon that was called the “zero hour effect.” Similar results were reported in another large trial of intermittent NTG therapy (3).
Nitroglycerin and endothelial dysfunction
In a recent study by our group, we documented an abnormal coronary vasomotor response to the endothelium-dependent vasodilator ACh in patients undergoing continuous therapy with transdermal NTG (11). The abnormal vasoconstriction observed was accentuated after discontinuation of therapy. We hypothesized that these results were secondary to the development of endothelial dysfunction caused by biochemical changes in the coronary vasculature that occurred during NTG therapy. In the current study, we used a similar protocol to test the coronary vasomotor responses to ACh during intermittent therapy with transdermal NTG and after patch removal. This study documents that intermittent NTG therapy is not associated with worsening of endothelial dysfunction during the period of transdermal NTG exposure. Withdrawal of NTG, however, was associated with aggravating endothelial dysfunction. Patients receiving intermittent transdermal NTG demonstrated a significant increase in the magnitude of vasoconstriction 3 h after patch removal, an effect that was not seen in the control group. This interval of 3 h was chosen on the basis of the results of studies that documented a decline in NTG plasma concentrations to undetectable levels within 1 h of patch removal (18,19). This increase in the severity of vasoconstriction after NTG withdrawal was five times larger in the active therapy arm than in the control group (Fig. 1). There was no evidence of tolerance in the NTG arm, given the fact that in this group, the mean LAD diameter at baseline was significantly larger than that in the control group. The mechanism underlying this response is not clear. We hypothesize that during NTG therapy, biochemical changes develop that make the coronary vasculature more sensitive to vasoconstrictors (9,10,20). During the period of NTG exposure, these biochemical changes develop, but are balanced by the dilating effect of the nitrate. After withdrawal of NTG therapy, these biochemical changes persist, but are no longer opposed by the vasodilating properties of the nitrate. The resulting imbalance between dilator and constrictor systems is manifested as a greater sensitivity to the vasoconstrictor actions of ACh.
It is important to consider some limitations of this study. Differences in baseline characteristics can potentially influence our results. A crossover design would not be possible given the need for repeat coronary angiography. We carefully monitored all risk factors for endothelial dysfunction, because endothelium-dependent responses to ACh have been related to risk factors (21). There were no significant differences between the two groups for any of the baseline characteristics commonly reported in previous studies that evaluated the coronary vasomotor response to ACh (11,22).
We believe these findings have clinical significance. Although the incidence of clinically significant episodes of ischemia after transdermal NTG therapy is relatively low (it was seen in only 6.5% of the patients in the report by DeMots and Glasser ), physicians should be aware of this possibility in patients on transdermal therapy. In a recent, large, retrospective study, the use of long-term nitrate therapy appeared to be harmful in a group of patients with CAD. These analyses raise concern about the potential adverse effects of long-acting nitrate therapy in chronic CAD (23). Furthermore, the possibility of rebound ischemia suggests that intermittent transdermal therapy should be used with caution in patients with variable threshold angina or recent unstable ischemic episodes.
The authors thank the staff of the Bayer Cardiovascular Clinical Research Laboratory of the Mount Sinai Hospital: May Chan, RN, Rebecca Allan, RN, Wilson Chan and Thom Benson, RN, for their help in the completion of these studies.
☆ This study was supported by grants-in-aid from Bayer Inc. and the Heart and Stroke Foundation of Canada, Grant #T-3695. Dr. Azevedo holds a research fellowship award from AstraZeneca/Heart and Stroke Scientific Research Corporation of Canada.
- coronary artery disease
- circumflex coronary artery
- 5% dextrose in water
- left anterior descending coronary artery
- left main coronary artery
- Received April 14, 2000.
- Revision received September 11, 2000.
- Accepted October 16, 2000.
- American College of Cardiology
- DeMots H.,
- Glasser S.P.
- Minitran Efficacy Study Group,
- Parker J.O.,
- Amies M.H.,
- Hawkinson R.W.,
- et al.
- Ferratini M.,
- Pirelli S.,
- Merlini P.,
- Silva P.,
- Pollavini G.
- Parker J.D.,
- Parker A.B.,
- Farrell B.,
- Parker J.O.
- Lange R.L.,
- Reid M.S.,
- Tresch D.D.,
- Keelan M.H.,
- Bernhard V.M.,
- Coolidge G.
- Munzel T.,
- Giaid A.,
- Kurz S.,
- Stewart D.J.,
- Harrison D.G.
- Caramori P.R.,
- Adelman A.G.,
- Azevedo E.R.,
- Newton G.E.,
- Parker A.B.,
- Parker J.D.
- Jost S.,
- Rafflenbeul W.,
- Gerhardt U.,
- et al.
- Pepine C.J.,
- Lopez L.M.,
- Bell D.M.,
- Handberg T.E.,
- Marks R.G.,
- McGorray S.
- Holdright D.R.,
- Katz R.J.,
- Wright C.A.,
- et al.
- Bauer J.A.,
- Fung H.L.
- Vita J.A.,
- Treasure C.B.,
- Nabel E.G.,
- et al.
- Multicenter Myocardial Ischemia Research Group,
- Nakamura Y.,
- Moss A.J.,
- Brown M.W.,
- Kinoshita M.,
- Kawai C.