|
|
| |||||||||||||||||||||||||||||||||||||||
Background Accelerated coronary artery disease is a major cause of late morbidity and mortality among heart-transplant recipients. Because calcium-channel blockers can suppress diet-induced atherosclerosis in laboratory animals, we assessed the efficacy of diltiazem in preventing coronary artery disease in transplanted hearts.
Methods Consecutive eligible cardiac-transplant recipients were randomly assigned to receive diltiazem (n = 52) or no calcium-channel blocker (n = 54). Coronary angiograms obtained early after cardiac transplantation and annually thereafter were used for the visual assessment of the extent of coronary artery disease. The average diameters of identical coronary artery segments were measured on the angiograms obtained at base line and at the first and second follow-up examinations.
Results In the 57 patients who had all three angiograms, the average coronary artery diameter (±SD) decreased in the group that received no calcium-channel blocker from 2.41 ±0.27 mm at base line to 2.19 ±0.28 mm at one year, and to 2.22 ±0.26 mm at two years (P<0.001 for both years). The average diameter in the diltiazem group changed little from the base-line value of 2.32 ±0.22 mm (2.32 ±0.27 mm at one year and 2.36 ±0.22 mm at two years). The average change in the diameter of the segment differed significantly between the two treatment groups (P<0.001), and the estimated effect of treatment changed only negligibly after adjustment for other relevant clinical variables. New angiographic evidence of coronary artery disease developed in 14 patients not given calcium-channel blockers, as compared with 5 diltiazem-treated patients (P = 0.082). Coronary stenoses greater than 50 percent of the luminal diameter developed in seven patients not given calcium-channel blockers, as compared with two patients given diltiazem; death due to coronary artery disease or retransplantation occurred in five patients in the group that did not receive calcium-channel blockers and in none of those who received diltiazem.
Conclusions Our preliminary results suggest that diltiazem can prevent the usual reduction in the diameter of the coronary artery in cardiac-transplant recipients, but further follow-up will be required to determine whether diltiazem can decrease the long-term incidence of symptomatic coronary artery disease.
Methods
Patients
Since September 1986, all adult cardiac-transplant recipients at Stanford University Medical Center have been invited to participate in this study. The study was approved by the medical center's Internal Review Board. Those whose condition was clinically stable and who consented to participate were randomly assigned to receive either diltiazem or no calcium-channel blocker 10 to 14 days after transplantation. By May 1989, a total of 115 adult patients had undergone cardiac transplantation. Seven were not enrolled in the study because of early complications or death, and two declined to participate. Of the remaining 106 patients, 52 were randomly assigned to receive diltiazem and 54 to receive no calcium-channel blocker (Table 1).
|
Base-line and follow-up evaluations, conducted before and one and two years after transplantation, included a medical history, physical examination, routine blood-chemistry measurements, and assessment of medication use, intercurrent illnesses, and risk factors for coronary disease. Diltiazem was initiated at a dose of 30 mg three times a day from 2 to 4 weeks (mean, 22 days) after transplantation. If this dose was tolerated, the dose was increased to 60 to 90 mg three times a day for the remainder of the study. Both groups received antihypertensive therapy other than calcium-channel blockers as required. Compliance was monitored by office visits, medication review, and measurement of trough blood cyclosporine levels every three months. All patients were taking 225 mg of dipyridamole and 80 mg of aspirin daily. Only eight patients were taking lipid-lowering agents during some part of the study period, six in the no-calcium-channel-blocker group and two in the diltiazem group.
Immunosuppressive Therapy
All patients received a triple maintenance regimen for immunosuppression, combining cyclosporine, azathioprine, and prednisone. Cyclosporine was given at a dose of 2 to 8 mg per kilogram of body weight per day, with continuous adjustment to maintain trough blood levels of 7.5 to 22.5 µg per deciliter as measured by radioimmunoassay. Patients received 1 to 2 mg of azathioprine per kilogram per day, adjusted to maintain a white-cell count of 4500 to 6000 per cubic millimeter (4.5 to 6.0 x 109 per liter), and prednisone at a maintenance dose of 0.1 to 0.2 mg per kilogram per day. Episodes of moderate rejection were treated with increased doses of prednisone and, in refractory cases, rabbit antithymocyte globulin or OKT3.
Cineangiography
All patients underwent selective coronary arteriography at base line (median, 19 days after transplantation) and annually thereafter. The diameters of cylindrical metallic markers near the catheter tip provided scaling factors for measuring luminal diameters. The cineangiographic projection angles used in the base-line arteriography were noted and were replicated on subsequent angiograms. Sublingual nitroglycerin (0.4 mg) was administered before the injection of contrast material to minimize variations in vessel tone and to enhance visualization. Diltiazem was withheld for 24 hours (seven half-lives) before each procedure.
All follow-up coronary arteriograms were analyzed by a single reviewer for the presence or absence of coronary artery disease in the transplanted heart; if an abnormality was detected, the presence or absence of disease was determined by consensus by two angiographers. The reviewers were blinded to the patients' treatment assignments and drug therapy, but not to the sequence of the arteriograms. Serial cineangiographic films were compared side by side, with use of two projectors. A diagnosis of coronary artery disease in the transplanted heart was made if there was visible evidence of coronary artery disease, including minor irregularities in the lumen. Of the 106 base-line angiograms, 3 showed abnormalities. For these three patients, the development of disease was defined as evidence of new abnormalities in a different location.
The luminal diameter of the coronary artery was measured by computer-assisted, computerized edge detection21 of digitized images from a 35-mm cineangiographic film-transport mechanism mounted on a movable stage. End-diastolic frames were identified, and the frame was magnified (x3.5) and digitized with a high-resolution camera connected to a video frame grabber installed in a personal computer. All adequately visualized portions of the right and left coronary arteries larger than 1.0 mm in diameter were identified on the initial angiogram on the basis of coronary artery branch points. Only these segments, in identical lengths (range, 2.5 to 30.0 mm), were measured on subsequent angiograms. Hand-drawn sketches and magnified photographic prints of all measured segments were used to ensure that identical segments were measured on serial angiograms.
Computer-generated representations of vascular boundaries were identified as the points of maximal change in the density of the image along a series of profiles perpendicular to the long axis of the vessel. When the computer algorithm could not resolve vessel boundaries in areas of noise or vessel crossing, short segments of boundary were edited manually. The mean diameter of each segment was calculated from perpendicular lines constructed through the length of a computer-generated center line. The average diameter of all measured segments was computed for each angiogram of each patient (those obtained at base line, at one year, and at two years). The change in the average diameter for each patient was adjusted for the actual length of time between the base-line and follow-up angiograms.
Statistical Analysis
Differences between the groups in characteristics at base line and during the study were assessed with chi-square tests for categorical variables and two-tailed t-tests for continuous variables. A P value of less than 0.05 was considered to indicate statistical significance. No explicit adjustment was made for multiple comparisons, but the issue must be considered when interpreting the P values.
Changes in coronary artery diameter from the base-line angiogram to the one-year or two-year angiogram were analyzed for each patient and for each coronary artery segment. In the analysis of the measurements for each patient, we used the average diameter of all the segments evaluated, with each segment weighted equally because earlier pathological studies have documented a diffuse vascular process22. For the analysis of the changes in individual segments we used the method of Rosner,23 which permits more precise treatment comparisons by using the larger number of individual segments as the basic sample size, but only after measuring and adjusting for the correlation of changes in the segments in individual patients.
In paired and unpaired t-tests assessing differences between the groups, the dependent variable was the average luminal diameter for each patient. The sample size reflects the fact that the rate of change in the averaged mean (±SD) diameters during the first year after transplantation was known to be -0.227 ±0.190 mm per year24. On the basis of this rate of progression, a projected 50 percent effect of the intervention, an alpha of 0.05, and a one-year interval between the base-line and follow-up angiograms, we calculated that a sample of 48 patients per group would be required for the study to have a power of 80 percent to detect a difference of this magnitude, on the basis of a two-tailed t-test of no change. Univariate analyses were performed with adjustment for treatment-group assignment, and multivariate linear regression analysis was used to identify correlates of the reduction in the luminal diameter, with both patients and segments as the unit of observation.
Statistical analyses were performed with SAS and S software25 (S was used for the Rosner model, which for purposes of implementation was regarded as a gaussian, intraclass-correlation special case of the code of Liang and Zeger26,27). "Robust" standard errors were computed with use of theorem 226. Variables with an absolute z score (coefficient divided by the standard error) greater than 1.6 in the analysis of segments were chosen as candidates for the multivariate regression analysis.
Results
Characteristics of the Patients
Of the 106 patients randomly assigned to the two treatment groups, 75 have completed one or two years of follow-up, with annual arteriograms. Table 1 shows the clinical characteristics of these 75 patients before transplantation, as compared with the 31 patients who have not had a follow-up angiogram. There were no significant differences, except that among those who completed the study, the donors were older in the diltiazem group than in the no-calcium-channel-blocker group.
Table 2 compares the postoperative clinical features and laboratory results of the patients in the two groups. There were no significant differences in clinical or laboratory characteristics between the two groups with the exception that the mean oral dose of cyclosporine was significantly lower in the diltiazem group; this finding is consistent with the known tendency of diltiazem to elevate blood levels of cyclosporine, thus reducing cyclosporine requirements. The trough blood cyclosporine levels were the same in the two groups. Antihypertensive therapy was generally similar in the two groups, although angiotensin-converting-enzyme inhibitors were used more often in the no-calcium-channel-blocker group (P = 0.03). Blood pressures were the same in both groups. Plasma high-density lipoprotein cholesterol levels were lower in the diltiazem group. No patient in either group smoked.
|
Table 3 shows the reasons for the absence of follow-up coronary angiograms in the two groups. Qualitative assessment was based on all available angiograms through March 1, 1992, at which time 17 patients in the diltiazem group and 14 in the no-calcium-channel-blocker group had undergone neither one-year nor two-year follow-up angiography. Nineteen of the 31 patients without follow-up angiography died before the angiogram could be obtained.
|
|
Freedom from angiographic evidence of coronary artery disease is plotted for both groups in Figure 1 as a survival-curve analysis (P = 0.082 for the comparison of the groups by the Mantel-Haenszel log-rank test for equality). To date, coronary artery disease has developed in 14 patients who did not receive calcium-channel blockers, as compared with 5 in the diltiazem group. Lesions resulting in stenosis of more than 50 percent have developed in seven patients in the no-calcium-channel-blocker group, as compared with two patients in the diltiazem group; death related to coronary artery disease or retransplantation occurred in five patients and no patients, respectively. As can be seen from the number of subjects at risk in Figure 1, follow-up for clinical events is not yet complete.
|
In a univariate analysis of individual coronary artery segments, the mean reduction in the segment diameter was 0.23 mm greater in the group that did not receive calcium-channel blockers than in the diltiazem group (P<0.001). The correlation among the changes in the segments for individual patients, estimated in the Rosner model,23 was 0.3 (SE = 0.04, P<0.001). Other clinical variables -- including the systolic blood pressure, the incidence of symptomatic cytomegalovirus infection after transplantation, the number of HLA-A and B mismatches, the number of rejection episodes, the serum creatinine concentration, and the mean diameter of the segments on the base-line angiogram -- were statistically significant predictors of the change in the diameter of the segment. These variables were included in a multivariate regression analysis (data not shown). However, their effects were small -- less than half the effect of the treatment group in all cases. Furthermore, the estimated effect of treatment (a rate of change of -0.257 mm) was changed only negligibly by adjustment for these variables. The results of regression analyses of data on individual patients were similar but less strongly significant. Multivariate regression analyses of changes in individual patients and in individual coronary artery segments were consistent in identifying the absence of treatment with a calcium-channel blocker as highly correlated with advancing coronary disease.
Discussion
This preliminary report suggests that diltiazem may attenuate the usual reduction in the coronary artery diameter during the first year after cardiac transplantation and may inhibit the development of visually evident coronary artery disease. In a retrospective review of 106 angiograms in cardiac-transplant recipients, it was found that 52 percent of those taking a calcium-channel blocker for hypertension had no coronary artery disease, as compared with only 27 percent of those who were not taking a calcium-channel blocker (Vetrovec G: personal communication). In a randomized study of coronary disease in patients who had not received a transplant, Lichtlen et al.29 reported that nifedipine, as compared with placebo, reduced the number of new lesions in 348 patients who underwent repeat angiography, and Waters et al.30 reported a decrease in the progression of stenosis measured at 
20 percent at base line in a randomized study of nicardipine. Similar results were reported by Loaldi et al.31 using nifedipine and Kober et al.32 using verapamil.
We selected diltiazem as the calcium-channel blocker for this study because of its antihypertensive efficacy and its excellent record regarding adverse side effects. At the beginning of the study, some patients taking diltiazem early after transplantation had sinus bradycardia. Lowering the dose of diltiazem until sinus-node function fully recovered eliminated this problem. It is important to monitor trough blood levels of cyclosporine carefully during the initiation of diltiazem therapy because most patients require a 30 to 50 percent lower oral dose to maintain similar therapeutic blood levels during therapy33.
The angiographic results are consistent with those of many experimental studies in rabbit models of atherosclerosis, which have demonstrated that calcium-channel blockers may prevent or retard atherosclerosis induced by diet or injury6,7,8,9,10,11,12,13,14,15,16,17,18,19,20. The mechanism by which diltiazem attenuates the narrowing of the coronary artery lumen is unclear. In laboratory animals, calcium-channel blockers reduce proliferative lesions in the carotid arteries induced by balloon-catheter injury13 or by electrical stimulation and cholesterol feeding34. The anti-atherogenic effects of calcium-channel blockers may involve the regulation of lipoprotein-receptor synthesis,35 lipoprotein uptake or degradation,36,37 cholesterol-ester hydrolytic activity,38,39,40 or arterial-matrix synthesis41.
It is probable that the reduction in the luminal diameter at the end of the first year in the control patients (those not given calcium-channel blockers) reflected diffuse and concentric intimal proliferation that was not apparent on angiography42,43. The lag in the development of visual evidence of arteriographic abnormalities reflected the late appearance of focal and typical atherosclerotic lesions, a course consistent with pathological features seen on microscopy in postmortem examinations43. The observation that a quantitative benefit of diltiazem was observed only in the first year after transplantation suggests that this drug inhibits an early, immunologically mediated, diffuse vascular insult associated with allograft implantation that moderates substantially during the second year. To determine whether the absence of luminal narrowing in diltiazem-treated patients during the first year after transplantation may make them less likely to have typical atherosclerotic disease later will require further follow-up.
Univariate and multivariate regression analysis confirmed that the lack of treatment with a calcium-channel blocker was the most important determinant of a reduction in the diameter of the lumen. These quantitative measurements are supported by a trend toward a lower prevalence of visually apparent and clinically important coronary lesions among the patients who received diltiazem. Furthermore, the magnitude of the treatment effect remained essentially unchanged after adjustment for other variables associated with the progression of coronary artery disease. The difference in the incidence of visually evident coronary artery disease and of accompanying clinical events between diltiazem-treated patients and those not so treated was sufficiently compelling that we are continuing to follow all patients in the prospectively assigned groups.
Limitations of the Study
This prospective randomized study was neither blinded nor placebo-controlled, in order to allow maximal patient compliance and permit optimal antihypertensive therapy in the group that did not receive a calcium-channel-blocking agent. Nonetheless, the qualitative and quantitative angiographic assessments were made by investigators and staff members who were blinded to the patients' treatment assignments. Although residual pharmacologic effects of diltiazem at the time of the arteriogram could theoretically influence vascular tone, this is unlikely, since all patients were taking the short-acting form of diltiazem (half-life, 3.5 hours). The drug was discontinued before angiography, and sublingual nitroglycerin was given in both groups to produce maximal coronary vasodilatation. Deaths, withdrawals from the study, and refusals to participate, by preventing angiographic follow-up, may have been a potential source of bias.
Prophylaxis
Therapeutic approaches to accelerated coronary artery disease in transplanted hearts, once disease is angiographically apparent, have been frustrating. Antiplatelet agents are used routinely in most transplantation programs because they have few side effects, but no controlled clinical trial has demonstrated that they reduce the incidence of coronary artery disease among heart-transplant recipients. The diffuse distribution of the disease43,44 limits the usefulness of conventional medical interventions and surgical techniques for revascularization. Among patients with coronary artery disease in the transplanted heart, survival after retransplantation has not been encouraging, with survival rates of 55 percent at one year and 25 percent at three years. Of patients who survive more than one year after receiving a second transplant, coronary artery disease subsequently develops in the second transplant in 41 percent45.
The results of this study suggest that during the first year after cardiac transplantation, diltiazem prevents or slows the usual decline in the average coronary-vessel diameter. However, determining the long-term clinical benefit of inhibiting intimal proliferation early after transplantation requires further follow-up.
Dr. Schroeder has been a consultant to Marion Laboratories and Marion Merrell Dow Laboratories, distributors of the diltiazem used in the study.
We are indebted to Anne Mullin, R.N., Mary Nejedly, R.N., and Lisa Prikazsky, R.N., for their assistance in conducting the study; to Anne Schwarzkopf and Irene Hill, Ph.D., for their analytic assistance; and to Stanley Azen, Ph.D., and Wendy Mack, M.S., for their discussions of Rosner's method and for providing us with access to the computer code.
Source Information
From the Division of Cardiovascular Medicine (J.S.S., S.-Z.G., E.L.A., S.A.H.), the Department of Statistics (I.J., D.B.B.), and the Department of Cardiothoracic Surgery (V.W., E.B.S.), Stanford University School of Medicine, Stanford, Calif.
Address reprint requests to Dr. Schroeder at the Division of Cardiovascular Medicine, CVRC, Stanford University School of Medicine, Stanford, CA 94305.
References
| |||||||||||||||||||||||||||||||||||||||
Related Letters:
Diltiazem in the Prevention of Coronary Artery Disease in Heart-Transplant Recipients
Pollak R., Fabrega A. J., Schroeder J.S., Gao S.-Z., Alderman E.L.
Extract |
Full Text
N Engl J Med 1993;
328:1851-1852, Jun 24, 1993.
Correspondence
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2008 Massachusetts Medical Society. All rights reserved. |
Previous
