A come the standard for surgical treatment of endstage

Transcription

1 Cardiac Transplantation With CorticosteroidFree Immunosuppression: LongTerm Results K. Francis Lee, D, Janet D. Pierce, PhD, ichael L. Hess, D, Andrea K. Hastillo, D, Andrew S. Wechsler, D, and Albert J. Guerraty, D Division of Cardiothoracic Surgery, Department of Surgery, and Division of Cardiology, Department of edicine, edical College of VirginiaVirginia Commonwealth University, Richmond, Virginia To assess the longterm safety of an immunosuppressive regimen without corticosteroids, we retrospectively evaluated 42 longterm (>1 year) survivors of orthotopic cardiac transplantation. We determined the incidence of (1) conversion of the immunosuppressive regimen from cyclosporine and azathioprine alone (group I) to cyclosporine, azathioprine, and prednisone (group II), (2) late acute graft rejection (defined as occurring at greater than 1 postoperative year), and (3) major postoperative complications related to corticosteroids. Of the 42 patients who were started on cyclosporine and azathioprine, 48% remained in group I, and 52% converted to group 11. Fortyfive percent of group I1 patients were able to taper and discontinue prednisone in 15.6 & 2.2 months. Among the patients on longterm corticosteroidfree immunosuppression, the incidence of late rejection was 2.1% per endomyocardial biopsy. The incidence of late infectious episodes was not significantly different between the two groups of patients, although diabetes mellitus and hypercholesterolemia were more prevalent in group I1 than in group I. These data suggest that cardiac transplant recipients who chronically remain on corticosteroidfree immunosuppression represent a select group of patients with an acceptably low risk of late graft rejection and associated reduction of potential risk factors of accelerated coronary artery disease. (Ann Thorac Surg 2991;52:2228) lthough orthotopic cardiac transplantation has be A come the standard for surgical treatment of endstage heart failure [l], successful clinical management of longterm survivors is often challenged by the complications of longterm immunosuppression. Corticosteroids in particular have been implicated in the development of many types of postoperative morbidity, eg, infection, hyperlipidemia, hypertension, and diabetes [261. A corticosteroidfree, doubledrug regimen using cyclosporine and azathioprine alone was first reported by Yacoub and associates in 1985 [2]. The purpose of the new regimen was to decrease the risks of postoperative infection and to increase survival. Subsequent clinical reports supported the use of doubledrug immunosuppression without corticosteroids. Katz and colleagues [4] and Renlund and coworkers [3] reported reductions in postoperative infectious episodes by 43% and 54%, respectively. Patients on the corticosteroidfree regimen showed decreased incidence of excess fluid retention and Cushingoid body habitus, as well as improved quality of life, emotional wellbeing, relative freedom from physical distress, and return to fulltime employment (4, 51. Withholding daily corticosteroids did not result in increased mortality [4, 61. The 1year survival rate was reported as greater than 90% for those remaining on nonsteroidal immunosuppression, comparable with those on the con Presented at the Thirtyseventh Annual eetlng of the Southern Thoracic Surgical Association, Dorado, Puerto Rico, Nov 810, Address reprint requests to Dr Lee, Department of Surgery, edical College of Virginia, CV Station 645, Richmond, VA ventional therapycyclosporine, azathioprine, and prednisone. The corticosteroidfree immunosuppressive protocol was not ideal for all patients, however. Within the first postoperative year, recurrent allograft rejection developed in approximately 30% of the patients and necessitated daily prednisone in addition to cyclosporine and azathioprine [3, 4, 61. A question was raised as to whether cyclosporine and azathioprine alone constituted safe, longterm immunosuppression. It was uncertain whether more patients would eventually convert from the doubledrug to the tripledrug therapy beyond the first postoperative year. Furthermore, the incidence of late acute rejection might be greatly increased in the long term if corticosteroid proved to be an essential component of longterm immunosuppression. ost clinical reports have not addressed these concerns because they have been based on 1year followups. At the edical College of Virginia, a corticosteroidfree immunosuppressive protocol was instituted in Our database contains information on the longterm outcome of patients who were initially begun on the nonsteroidal regimen. The purpose of the present study was to determine the incidence of late acute rejection among patients on corticosteroidfree immunosuppression. The likelihood of conversion from the doubledrug to the tripledrug regimen at greater than 1 postoperative year was assessed. In addition, we attempted to evaluate whether longterm corticosteroid sparing offered beneficial effects on major postoperative complications associated with by The Society of Thoracic Surgeons /91/$3.50

2 212 LEEETAL Ann Thorac Surg 1991; Table I. Protocol for Immunosuppression Drug Preoperative Cyclosporine Aza thioprine Rabbit ATG lntraoperative ethylprednisolone Postoperative ethylprednisolone Cyclosporine Azathioprine Dose 610 mgikg PO 2 mgikg PO 100 mg I 500 mg IVb 125 mg IV x 3 doses 610 mgikgiday 2 mgikgiday a Dose adjusted to renal function. Administration after cardiopulmonary bypass. Three doses administered every 8 hours during the first 24 hours. Dose adjusted to renal function and trough serum cyclosporine level. Dose adjusted to white blood cell count. ATG = antithymocyte globulin; I = intramuscularly; IV = intravenously; PO = perorally. steroids, eg, infection, hyperlipidemia, diabetes, and hypertension. aterial and ethods The present study was based on a retrospective analysis of all longterm survivors (greater than 1 year) of orthotopic cardiac transplantation performed at edical College of Virginia from June 1985 to ay 1987 (n = 42). All 42 patients initially received nonsteroidal immunosuppression. Patients who continued on the corticosteroidfree, doubledrug protocol (cyclosporine and azathioprine) were designated as group I, and those who eventually required maintenance corticosteroids and received the tripledrug regimen (cyclosporine, azathioprine, and prednisone), group 11. Data on various clinical variables were collected every month for the first 3 postoperative months, then every 3 months up to the first postoperative year, and then every 6 months thereafter. Immunosuppression Table 1 summarizes the protocol for maintenance immunosuppression that was previously reported from our institution [4]. For each postoperative period, the maximum daily dose of cyclosporine and azathioprine were determined and expressed per kilogram of patient weight. For group I1 patients, maximum daily doses of prednisone were also recorded. Reject ion Detection and confirmation of allograft rejection were based on endomyocardial biopsies. Histopathological examination was based on adaptation of the Billingham grading system [7], clinically reported as mild, moderate, or severe. Throughout the study, the term rejection referred to the clinical setting where a biopsy result of moderate or severe histopathological finding was treated with pulse steroid injections. Episodes of mild rejection on endomyocardial biopsy or other untreated histopath ological diagnoses were not counted as rejection. For each postoperative time period, the total number of rejection episodes was tabulated and expressed per patientmonth. A late rejection referred to an episode that occurred after the first postoperative year. The treatment of rejection consisted of pulse intravenous injections of methylprednisolone (total, 2.5 g) administered over 3 days. A followup endomyocardial biopsy was performed in 7 to 10 days. If the rejection persisted, repeated pulse therapies of methylprednisolone or rabbit antithymocyte globulin (100 mg intramuscularly daily for 5 to 7 days), or both, were administered. Patients who exhibited more than two episodes of rejection in the early postoperative months were considered to have failed the doubledrug regimen. These patients were converted to the tripledrug regimen and initially received 0.4 mg * kg * day of oral prednisone in addition to cyclosporine and azathioprine. The dosage of prednisone was tapered slowly and discontinued when clinically appropriate. Clinical Variables Preoperative clinical information was obtained with regard to the patient s age and sex, ischemic time of the allograft, cause of the recipient s cardiac disease, and the HLA match profile. Postoperatively, hypertension, diabetes mellitus, infection, and hypercholesterolemia were recorded as episodes of major morbidity. Hypertension was defined as diastolic blood pressure of greater than 95 mm Hg during treatment with one or more antihypertensive drugs. The presence of diabetes was determined by the patient s requirement of either oral hypoglycemic agents or insulin. An episode of infection was defined as any serious infection that required either oral or intravenous antibiotic therapy. A clinical condition suggestive of viral infection treated with an antiviral agent was also counted as serious infection. Statistical Analysis All values were expressed as mean * standard error of the mean. Comparisons between group I and group I1 were made using Student s t test for unpaired data or x 2 as appropriate for each clinical variable. Data tested over multiple time periods underwent a repeated measure analysis of variance. Values of p less than 0.05 were considered statistically significant. Results Fortytwo patients with greater than 1year survival were analyzed for a mean followup of months, ranging from 18 to 47 months. Of the 42 patients, 20 (48%) remained on the corticosteroidfree doubledrug regimen (group I), and 22 (52%) eventually required chronic corticosteroid therapy in addition to cyclosporine and azathioprine (group 11). As Table 2 summarizes, there was no significant difference in the clinical backgrounds of the two groups. The solid line in Figure 1 shows the cumulative incidence of conversion from doubledrug to tripledrug

3 Ann Thorac Surg 1991;52:2118 LEEETAL 213 Table 2. Clinical Background Variable All Group I" Group IIb p Value Age (Y) Sex (malelfemale) Cardiac disease Idiopathic Ischemic Valvar Other Ischemic time (min) HLA match (No. loci)' HLAA HLAB HLAC HLADR HLABW Other 44 * (36%) 20 (47%) 3 (7%) 4 (10%) 192 f f f f f f (30%) 11 (55%) 1(5%) 2 (10%) 193 * * f f (41%) 9 (41%) 2 (9%) 2 (9%) 190 * t t t f f 0.22 a Group I = patients on cyclosporine and azathioprine immunosupression alone. Group I1 = patients on cyclosporine, azathioprine, and prednisone. ' Student's t test for unpaired data.,y2 analysis. No. loci = average number of matched loci; NS = not significant. regimen. Of the total 22 patients who eventually required daily prednisone (group 11), 15 patients (68%) converted to tripledrug therapy by the third postoperative month. Twentyone patients (95%) converted by the sixth postoperative month. Only 1 patient (5%) converted to tripledrug therapy at greater than 1 postoperative year. The reasons for requiring the maintenance corticosteroids were as follows. Sixteen of 22 patients (73%) required daily prednisone due to early recurrent rejection. In 3 patients (14%), reduction of cyclosporine dosage due to drug toxicity, ie, renal failure (n = 2) and seizure (n = l), required the addition of daily prednisone. Leukopenia due to azathioprine toxicity was responsible for conversion to tripledrug therapy in 2 patients (9%). In 1 patient (4%), late rejection was the cause. In 10 of the group I1 patients, daily prednisone therapy was eventually tapered and discontinued. The mean duration of prednisone therapy among the 10 patients was months. The dotted line in Figure 1 shows the net percentage of patients who were on maintenance corticosteroids at each time period. At the end of a mean of 33 months of followup, only 29% of total patients remained on the tripledrug therapy. Figure 2 depicts the average daily doses of cyclosporine, azathioprine, and prednisone over time. In addition to daily prednisone, cyclosporine and azathioprine doses were higher in group I1 patients than in group I, especially from the sixth to 24th postoperative months. Figure 3 shows the average number of new rejection episodes per patientmonth over time. There were a total _ ~...~...,, apy conversion from cyclosporinel azathioprine to cyclosporinelazathioprinelprednisone (solid line) and net incidence of therapy conversion (dotted line). See text for explanation. 0 I I I I I I I

4 214 LEEETAL Ann Thorac Surg 1991; Fig 2. Daily doses of cyclosporine (A), azathioprine (B), and prednisorie (C). The statistical difference in doses of cyclosporine and azathioprine between groups 1 and I1 was analyzed by analysis of variance; only the group 11 patients received daily prednisone. See text for explanation. (Tx = therapy). 4 L, l A Gp 1:Doubledrug Tx Gp IkTripledrug Tx P< I I I I I I C Gp 1l:Tripledrug Tx

5 Ann Thorac Surg 1991;52:2118 LEEETAL Per Pat ienton t h Fig 3. Total rejection episodes for both groups I and 11 are represented for each time period. Arrows indicate late rejection episodes (at >1 year). Vertical lines above the bars indicate standard error of the mean O f Postoperative onths of six late rejection episodes in 6 patients, detected among 228 total endomyocardial biopsies performed at greater than 1 year. The incidence of late rejection (at >1 year) was found to be 2.6% per endomyocardial biopsy. Table 3 summarizes the rejection episodes and relevant clinical information. Only 2 of the 20 group I patients (10%) on corticosteroidfree immunosuppression encountered rejection at greater than 1 year. In both patients, the rejection episodes were strongly associated with inadequate cyclosporine dosing. Late rejection developed in 4 of the 22 group I1 patients on tripledrug therapy (18%) ( p = not significant by x analysis). Three rejection episodes were associated with graft coronary artery disease. Hypertension was highly prevalent in both groups, 95% and 100% in groups I and 11, respectively ( p = not significant by x 2 analysis). There was no difference in the preoperative incidence of hypertension. Diabetes mellitus, preoperatively absent among patients, was present postoperatively in only l patient in group I (5%). In group 11, only 1 patient (5%) had development of postoperative diabetes before receiving daily prednisone. However, diabetes developed in an additional 3 patients during the course of their chronic corticosteroid therapy, thus elevating the total incidence of diabetes to 18% in group 11. Compared with group I, the difference was not statistically significant ( p < 0.09); however, this may have been due to small sample size. The mean incidence of late infectious episodes (at >1 year) was 1.7% * 0.4% per patientmonth in group I. In group 11, the incidence while on daily prednisone was 1.5% t 0.4% per patientmonth (p = not significant by Student s t test). The postoperative serum cholesterol levels are depicted in Figure 4. Group I1 patients had higher serum cholesterol levels compared with group I during the second and Table 3. Cases of Late Rejection Age Time to LR Group (y) Sex (mo) Relevant Clinical History I I I1 I1 I1 I Cr = creatinine level; CyA = cyclosporin A; LVEF = left ventricular ejection fraction. Diarrhea, malabsorption, CyA level = 59 mg/dl Depression, medical noncompliance, CyA discontinuation Difficulty with insurance, inadequate funds to buy CyA, subtherapeutic selfdosing of CyA Frequent early rejection (n = 3 at 1, 3, 6 mo), renal failure (Cr = 247 Fmol/ L [2.8 mg/dl]), refractory hypertension, tapering of both CyA and prednisone, angiographic evidence of coronary artery disease Frequent early rejection (n = 3 at 1, 2, 6 mo), heart failure (LVEF = 0.20), angiographic evidence of coronary artery disease Acute heart failure, percutaneous coronary angioplasty for coronary artery disease, sepsis, death LR = late rejection;

6 216 LEEETAL Ann Thorac Surg 1991;52:2118 Fig 4. Serum cholesterol levels for groups I (CYNAZA) and I1 (CYN AZNPRED). Dark vertical lines on top of bars represent standard errors of the means. The statistical differences between groups 1 and ll was analyzed by analysis of variance. (NS = not significant.) Pc0.05 Pc I I I I I (CYA/AZA) * II (CYA/AZA/PRED) third postoperative years, 6.96? 0.49 mmovl (269 * 19 mg/dl) versus 5.38 * 0.21 mmol/l ( mg/dl) ( p < 0.05 by analysis of variance) and 6.21? 0.67 mmol/l ( mg/dl) versus mmovl ( mg/dl) (p < 0.05 by analysis of variance), respectively. Overall there were three late deaths beyond the first postoperative year, all of which occurred in group 11. The causes of death were intrathoracic lymphoma, sepsis, and coronary atherosclerosis. Comment Our data suggest that once a patient survived to the sixth postoperative month on cyclosporine and azathioprine, the likelihood that he or she would require maintenance steroids at a later date was very low (5%). In addition, even if a patient converted from a doubledrug to a tripledrug regimen in early postoperative months, there was a 50% chance that he or she would be able to taper and eventually discontinue the daily prednisone in the future. Overall, about one third of the patients who initially began the doubledrug therapy were at risk of requiring chronic maintenance corticosteroidsa rate similar to those reported in the literature [3, 4, 61. In regard to late rejection, believed to be a rare complication on conventional immunosuppression, our data showed that a patient who survived the first postoperative year on cyclosporine and azathioprine alone would have a minimal chance of the development of late rejection. The incidence of late rejection among group I patients was 2.1% per endomyocardial biopsy, lower than the reported rates of 5% to 7% per endomyocardial biopsy [8, 91. The absence of chronic steroids did not appear to increase the incidence of late rejection. In our series, the 2 cases of late rejection episodes on nonsteroidal immunosuppression were both preceded by acute disruptions of daily cyclosporine therapy (see Table 3). One patient showed medical noncompliance and stopped taking cyclosporine altogether before the allograft rejection. The other patient had malabsorption of cyclosporine after a 2week bout of diarrhea. The admission cyclosporine level was 59 mg/dl. Thus, the development of late acute rejection on corticosteroidfree immunosuppression depended more on the inadequacy of cyclosporine maintenance than on withholding the daily steroids. These data strongly suggest that a patient who was successfully managed on corticosteroidfree immunosuppression for the first postoperative year would have an excellent chance of remaining on the same regimen on a longterm basis without a significant risk of late graft rejection. Because of an inherent limitation in our study design, it was difficult to assess whether the differences in the major postoperative complications between the two groups were directly due to the corticosteroid factor. Group I1 consisted of patients who "failed" the corticosteroidfree doubledrug protocol. As a group they had more rejection episodes and required more immunosuppression (see Fig 2). They were selected from group I patients with a bias, and therefore, they did not represent a fair comparison group. A proper control group would have been patients who were initially started on tripledrug therapy, visavis doubledrug therapy. However, such a group of patients without other selection bias did not exist in our database; therefore, we attempted to draw conclusions based on historical comparisons with the literature data. The incidence of late infection beyond the first postoperative year was not significantly different between groups I and I1 in our series. This was in contrast to the earlier reports [3, 4, 61 which demonstrated a significant decrease in infectious episodes in early postoperative months (at <1 year) for patients on corticosteroidfree immunosuppression. The control group in those reports received an average daily prednisone dose of 0.2 to 0.4

7 Ann Thorac Surg 1991; LEEETAL 217 mg * kg' day' [4], 0.12 to 0.30 mg * kg' * day' [2], and 0.2 to 1.0 mg kg' * day' [3]. At such doses in the early postoperative period, corticosteroids might have had a deleterious effect on resistance to infection. In our study, however, the average daily prednisone dose at greater than 1 postoperative year ranged from 0.07 & 0.02 to 0.13? 0.04 mg. kg' * day'. It is possible that infection associated with corticosteroids may not be an important clinical factor at such a reduced daily dose of prednisone. Our data suggested that the effect of corticosteroid sparing on reducing infectious episodes might be much more critical during the early postoperative period when the daily corticosteroid dose was relatively high (see Fig 2). Beyond the first postoperative year, the maintenance corticosteroid therapy is at such a low dose that administering or withholding corticosteroids might not affect the infectious rate significantly. The serum cholesterol level, however, appears to have been affected by the steroids. As previously reported from our institution [lo], serum cholesterol levels were not significantly different between group I and group I1 during the first postoperative year. However, the average serum cholesterol level of group I was significantly lower than that of group I1 during the second and the third postoperative years, 5.38 versus 6.96 and 4.63 versus 6.21 mmol/l, respectively. This might have been due to the patient population differences between the two groups, although our data were almost identical to those of the report by Renlund and associates [ll], which demonstrated that during the second postoperative year (at 18 months), the serum cholesterol levels of the doubledrug and the tripledrug groups were approximately 5.17 mmol/l (200 mg/dl) versus 6.98 mmol/l (270 mg/dl), respectively. Although far from being conclusive, such similarities support the view that even a lowdose daily corticosteroid therapy might be responsible for elevating the serum cholesterol level. The difference in the incidence of diabetes between the two groups was not statistically significant (p < 0.09), perhaps due to small sample size. The association of maintenance corticosteroids with postoperative diabetes was suggested by the fact that group I1 patients showed the same incidence of diabetes (5%) as group I before the onset of daily corticosteroid therapy. However, the disease developed in an additional 13% of the patients in group I1 once they began steroid intake. Others have also reported an increased incidence of diabetes [3, 61, thus strengthening our association between postoperative diabetes and chronic corticosteroid therapy. Neither posttransplantation diabetes mellitus nor hypercholesterolemia has definitively been shown to correlate with accelerated coronary artery disease. But there is cause for concern, based on the evidence that hypercholesterolemia in conjunction with allograft rejection may facilitate the pathogenesis of coronary atherosclerosis [ 12, 131. Posttransplantation diabetes mellitus has also been associated with accelerated coronary artery disease [9]. It is clearly in the best interest of patients to avoid maintenance corticosteroids as long as immunosuppression is provided adequately. Our database presently does not contain adequate information on the longterm coronary atherosclerotic changes. Based on our study, however, it is reasonable to conclude that cardiac transplant recipients who remain on corticosteroidfree immunosuppression represent a select group of patients who are at an acceptably low risk of late graft rejection and have significant reductions of potential risk factors of accelerated coronary artery disease. Supported in part by a grantinaid from the National Institutes of Health (RR00065) to the Clinical Research Center of edical College of Virginia. References 1. Kriett J, Kaye P. The registry of the International Society for Heart Transplantation: seventh official report1990. J Heart Transplant 1990;9: Yacoub, Alivizatos P, Khaghani A, itchell A. The use of cyclosporine, azathioprine and antithymocyte globulin with or without low dose steroids for immunosuppression of cardiac transplant patients. Transplant Proc 1985;17: Renlund DG, O'Connell JB, Gilbert E, Watson FS, Bristow R. Feasibility of discontinuation of corticosteroid maintenance therapy in heart transplantation. J Heart Transplant 1987;6: Katz R, Barnhart GR, Szentpetery 5, et al. Are steroids essential for successful maintenance of immunosuppression in heart transplantation? J Heart Transplant 1987;6: Jones B, Taylor FJ, Wright O, et al. Quality of life after heart transplantation in patients assigned to double or tripledrug therapy. J Heart Transplant 1989;9: Esmore DS, Spratt P, Keogh A, Chang VP. Cyclosporine and azathioprine immunosuppression without maintenance steroids: a prospective randomized trial. J Heart Transplant 1989;8: Billingham E. Dilemma of variety of histopathologic grading systems for acute cardiac allograft rejection by endomyocardial biopsy. J Heart Transplant 1990;9: Warnecke H, Schueler 5, Hetzer R. Late acute rejection after cardiac transplantation: incidence and treatment. Transplant Proc 1987;19:250&5. 9. Narrod J, Kormos R, Armitage J, Hardesty R, Ladowski J, Griffith B. Acute rejection and coronary artery disease in longterm survivors of heart transplantation. J Heart Transplant 1988;8:41% Taylor DO, Thompson ]A, Hastillo A, et al. HDLcholesterol after cardiac transplantation: comparing prednisone vs. nonprednisone antirejection protocols [Abstract]. J Heart Transplant 1988;7: Renlund DG, Bristow R, Crandall BG, et al. Hypercholesterolemia after heart transplantation: amelioration by corticosteroidfree maintenance immunosuppression. J Heart Transplant 1989;8: Alonso DR, Starek PK, inick CR. Studies on the pathogenesis of atheroarteriosclerosis induced in rabbit cardiac allografts by the synergy of graft rejection and hypercholesterolemia. Am J Pathol 1977;87: Hess L, Hastillo A, ohanakumar DV, et al. Accelerated atherosclerosis in cardiac transplantation: role of cytotoxic Bcell antibodies and hyperlipidemia. Circulation 1983; 68(Suppl 2):9&101.

8 218 LEEETAL Ann Thorac Surg 1991; DISCUSSION DR LENOX D. BAKER (Norfolk, VA): I congratulate you for providing us with some data that are new and exciting. As you know, this work presented today was done in these patients by Drs Richard Lower, Szabolcs Szentpetery, and Glenn Barnhart when they were at Richmond. I am happy to report that they have continued with this protocol in the Transplant Program at Norfolk and that of our first 31 transplants, 24 are now past the 6month period. We have been initiating a steroidfree protocol and have actually been able to maintain it in 18 of these 24 patients (75%). These patients are certainly easier to handle off steroids, and it is nice to see that these data are holding up long term. The 6 patients who have required steroids require them for the usual reasons. Primarily these are associated with the renal problems with cyclosporine and the occasional immunologically complex patient who has multiple rejections. I have one question for youare you starting these patients on a steroidfree protocol from day l? Some of these programs, as you know, have been starting them out initially on tripledrug therapy and then trying to wean them off steroids after the first 3 months. This has been difficult and has resulted in increased acute rejection at the time of trying to get them off steroids. We believe it is best to start them off from day 1 on a steroidfree protocol. DR LEE: Except for the induction protocol, these patients were started on day 1 without corticosteroids. Of course, if acute rejection develops they do require Sohedrol (methylprednisolone) pulse therapy. But the maintenance protocol is cyclosporine and azathioprine without steroids. DR JOHN D. OSWALT (Austin, TX): I enjoyed your paper very much. Unfortunately we have tried, as Dr Baker suggested, the weaning of steroids after a 6month period. Initially we began our patients on tripledrug therapy and have beencontrary to what you showedvery unsuccessful with weaning these. I would like you to address the levels of cyclosporine in your patients. Are these levels different in your doubledrug therapy and your tripledrug therapy patients? ost of our patients have had one to two episodes of rejection when we tried to wean them off prednisone. We have been able to get them down to a very low maintenance level, however. DR LEE: We did not have to increase cyclosporine or azathioprine doses to compensate for corticosteroid sparing. It is clear that these two groups of patients are different groups of patients. Although our analysis of their background variables, such as age, sex, ischemic time, and HLA matching, did not show any significant difference, that may possibly be because of the small number of patients. I think these two groups of patients are different, and in a sense the corticosteroidfree protocol represents a sort of selection process. Actually, the second group of patients required higher doses of cyclosporine simply because they had more rejection episodes and there were attempts to try to control their allograft rejection. During the third to twelfth months, the cyclosporine levels were higher in the second group of patients. This tells us that the group on maintenance corticosteroids, on the whole, is more difficult to manage; they are different from the first group of patients, who do not require maintenance steroids. DR WILLIA A. BAUGARTNER (Baltimore, D): I would also like to congratulate you on a very clear and nice presentation. I think your summary statement really is the crux of this type of therapy; that is, do steroid or steroidsparing protocols lead to less coronary artery disease, whether it is due to a decrease in diabetes or less cholesterol, which is clearly unknown at this time. ost of your patients are, by definition, out beyond 2 years. In most studies at 2 years there is about a 20% to 25% incidence of coronary artery disease in the population. Is that your incidence, or is it different between the two groups? DR LEE: The incidence of accelerated coronary disease in our group of patients is rather difficult to determine, because the angiographic data on these patients have been selective and retrospective. A comparison of the atherosclerosic coronary disease in the two groups of patients is not possible at present.