Prednisone Withdrawal Late After Adult Liver Transplantation Reduces Diabetes, Hypertension, and Hypercholesterolemia Without Causing Graft Loss

Transcription

1 Transplantation and Surgery Prednisone Withdrawal Late After Adult Liver Transplantation Reduces Diabetes, Hypertension, and Hypercholesterolemia Without Causing Graft Loss MARK D. STEGALL, 1 GREGORY T. EVERSON, 2 GERHARD SCHROTER, 1 FREDERICK KARRER, 1 BAHRI BILIR, 2 TRACY STERNBERG, 1 ROSHAN SHRESTHA, 2 MICHAEL WACHS, 1 AND IGAL KAM 1 We prospectively withdrew prednisone in 28 adult paare plantation are due to cardiovascular disease and another 40% tients who had stable graft function more than 2 years due to infection. 4 after orthotopic liver transplantation (OLTx) and had We previously have observed a high prevalence of cardiobeen on 5 mg/d prednisone for at least 6 months. Prednition. vascular risk factors in patients late after liver transplanta- sone was decreased from 5 mg/d to 2.5 mg/d for 1 month 4 One year after successful liver transplantation in our then stopped completely. Cyclosporine monotherapy patient population, the incidence of diabetes was 13% and was maintained at a level of approximately 200 ng/ml the incidence of hypertension was 69.1%. In addition, 31% of (TDX). Nineteen patients had prednisone withdrawn our patients had serum cholesterol levels ú 240 ng/ml at 1 without complications. Four (14.2%) had modest elevacreased year. Tapering of prednisone from 10 mg/d to 5 mg/d de- tions in liver function tests (two biopsy proven mild resignificantly these metabolic complications in some patients and jections and two were not biopsied). These four were decreased serum cholesterol levels. treated with methylprednisolone boluses and then withthe In this study, we further explore the role of prednisone in drawal of steroids again. Prednisone was restarted in development of metabolic complications after liver trans- five patients because of generalized fatigue and body plantation. We prospectively withdrew prednisone in liver aches (n Å 4) and colitis (n Å 1). Steroids later were transplantation recipients who had survived at least 2 years successfully withdrawn in two of these patients. After after transplantation to determine if the metabolic complicaprednisone withdrawal, three of five insulin-dependent tions such as diabetes, hypertension, and hypercholesterol- diabetic patients were able to discontinue insulin therapy emia could be decreased. and their glycosylated hemoglobin levels improved. Four of fourteen hypertensive patients were able to discontinue PATIENTS AND METHODS antihypertensive medicines. Mean serum cho- Patients included in the study were at least 2 years after orthotopic lesterol decreased from { 43.3 to { 33.3 mg/ liver transplantation (mean, 4.1 { 1.2 years; range, years) dl (P õ.001). The number of patients with serum choles- who had stable hepatic allograft function (serum transaminases õ terol levels ú 220 mg/dl decreased from 13 to 4. A control 50 IU/mL). The mean age was 52.1 { 9.0 years (range, group of 24 patients maintained on 5 mg/d prednisone years). Neither prior rejection episodes nor underlying disease imat least 2 years after liver transplantation also was studcompliance were not asked to participate. Patients with hepatitis B pacted on patient selection. However, patients with a history of non- ied. In this group during the study period, no diabetic also were excluded because they do not receive maintenance prednibecame normoglycemic, no patient decreased their antisone immunosuppression at our center. No children were included hypertensive medicine, and the mean serum cholesterol in this study. levels did not change significantly. We conclude that At the time of the study, 52 patients at our institution were at prednisone withdrawal using cyclosporine monother- least 2 years after OLTx and on 5 mg/d of prednisone. Twelve patients apy late after liver transplantation does not lead to graft were excluded from this initial study because they had a hisapy loss and decreases the prevalence of diabetes, hypertendiagnosis tory of fluctuating serum transaminases that might complicate the sion, and hypercholesterolemia. Symptoms occurring of rejection. Ten patients were excluded because in the during withdrawal may be minimized by earlier or opinions of the investigators, they lived too far away from the center slower tapering. to get weekly laboratory tests done and checked without undo delay. (HEPATOLOGY 1997;25: ) Two patients refused participation. Finally, 28 patients were included in the study. No patient was dropped from the study after enrollment. The 24 patients not included in the steroid withdrawal As more patients enjoy long-term survival after liver trans- study were maintained on 5 mg/d prednisone throughout the study plantation, the late complications become increasingly im- period and continued to be followed with monthly laboratory values. portant in patient management. Late graft loss because of This group constitutes our control group for comparing rejection rejection is rare in liver transplantation recipients, yet 10% and metabolic complications. The control group did not differ from to 15% of patients die over the next 4 years. 1-4 In our adult the study group in terms of age (54 { 8.2 vs { 9.0 years), sex or time after transplantation (3.9 { 1.4 vs. 4.1 { 1.2 years). patient population, 40% of the late deaths after liver trans- Study patients were informed of risks and benefits of steroid withdrawal and consent was obtained. Prednisone was decreased from 5 mg/d to 2.5 mg/d for 1 month, then stopped completely. Sandimmune monotherapy was maintained in all but one patient with 12-hour From the 1 Departments of Surgery and 2 Medicine, The University of Colorado School trough cyclosporine levels maintained greater than 200 ng/ml of Medicine, Denver, CO. (whole blood TDX assay). One patient who already was on azathio- Received March 4, 1996; accepted August 22, Address reprint requests to: Mark D. Stegall, M.D., University of Colorado School of prine and cyclosporine was maintained on double therapy. Liver Medicine, Campus Box C318, 4200 East Ninth Ave., Denver, CO function tests (serum aspartate transaminase; alanine transami- Copyright 1997 by the American Association for the Study of Liver Diseases. nase; bilirubin) were obtained weekly for 4 weeks after steroid with /97/ $3.00/0 drawal, every 2 weeks during the second month, then monthly there- 173 AID Hepa 0006 / 5p1b$$$ :23:42 hptas WBS: Hepatology

2 174 STEGALL ET AL. HEPATOLOGY January 1997 TABLE 1. Results of Prednisone Withdrawal in Liver Transplant The transaminases in all four patients returned to normal Recipients Late After Liver Transplantation and the patients were able to remain off steroids (9-month No. Result follow-up minimum). Protocol biopsies were performed before withdrawal and 2 19 Withdrawn with minimal complications months after withdrawal in the first five patients enrolled in 4 Mild elevation in transaminases the study. None of these patients had any abnormalities in 2 Biopsied: laboratory tests and none of the biopsied showed any evi- 1 Mild Acute Rejection dence of rejection. 1 Minimal Rejection In five other patients, steroids had to be restarted in the 2 Not biopsied: All resolved with bolus steroids* and rewithdrawn first month after withdrawal. Three patients developed se- 2 Developed problems but withdrawn from steroids slowly vere fatigue, loss of appetite, and joint pains. One of these 1 Recurrent colitis-removed after 8 months patients became dehydrated with a rising creatinine. A corti- 1 Severe fatigue-removed after 7 months sol stimulation test showed that this patient had adrenal 3 Remained on steroids insufficiency. The fifth patient developed diarrhea and colonoscopy showed colitis. All five patients were restarted on 5 * Methylprednisolone 500 mg/d IV daily for 3 days. mg/d prednisone with resolution of their symptoms. Two of All had severe fatigue and/or Addisonian-like symptoms. these patients, one with severe fatigue and one with colitis were slowly tapered off prednisone without further problems (1 mg/mo decrease). after. The first five patients who underwent prednisone withdrawal In the control group maintained on 5 mg/d prednisone, had protocol biopsies performed before and at 2 months postwith- one patient (4%) developed an acute rejection episode. The drawal regardless of liver function to rule out subclinical rejection. rejection was associated with a low cyclosporine level (õ200 To determine the effect of prednisone withdrawal on metabolic ng/ml) secondary to a bile duct stricture. No other rejection complications, the prevalence of diabetes (fasting serum glucose ú episodes were diagnosed during the trial period in the control 120 mg/dl), hypertension (diastolic blood pressure ú 90 mm Hg on group. three occasions), and hypercholesterolemia (fasting serum cholesterol ú 220 mg/dl) was determined before and after prednisone with- Effect of Prednisone Withdrawal on Diabetes. Five patients drawal. The prewithdrawal (5 mg/d prednisone dose) cholesterol level (20.8%) withdrawn from prednisone in this study had insulin- was determined by averaging serum cholesterol levels obtained 1 dependent diabetes on 5 mg/d prednisone. Two patients had and 2 months before withdrawal. The postwithdrawal cholesterol non insulin-dependent diabetes before transplantation and level was determined by averaging the serum cholesterol at 9 and three had developed diabetes after transplantation. All five 10 months after withdrawal. had Type II diabetes as determined by elevated fasting C- Patients in the study were encouraged to monitor their blood glu- peptide levels and all five were able to be withdrawn from coses daily. If their fasting glucose was õ120 mg/dl on 2 successive steroids. Two diabetic patients with posttransplantation diadays, the patients were instructed to decrease their insulin dose. betes were able to discontinue insulin treatment in the first Patients continued to make adjustments if they continued to have low blood glucoses either in the morning or midafternoon readings. month off prednisone and both had improved glucose control Glycosylated hemoglobin levels were obtained in all diabetic patients (HgB A1C levels) 9 months after withdrawal (Table 2). One on the 5 mg/d dose and 9 months following withdrawal of prednisone. patient with pretransplantation diabetes improved more Tapering of antihypertensive medicines was performed on the basis slowly and discontinued insulin treatment 7 months after of weekly blood pressure readings. Patients who developed a diastolic steroid withdrawal. The other two diabetic patients (one with blood pressure below 80 mm Hg on treatment were allowed to taper pretransplantation diabetes and one with posttransplant diatheir medicines and then recheck their blood pressure. In practice, betes) showed no change in their insulin requirements in the almost all insulin and antihypertensive medicine changes were per- 12 months since steroid withdrawal. Both of these patients formed for dramatic clinical changes. had gained more than 15 kg in weight since transplantation. In the control group, laboratory values were obtained at monthly intervals. Beginning serum cholesterol levels were determined by In the control group, none of the three insulin-dependent averaging serum cholesterol levels at 1 and 2 months before the study diabetics noted any change in their glucose control (their HgB period. Ending cholesterol levels were determined by averaging the A1C levels remained the same) and none were able to de- levels at 11 and 12 months into the study period. The presence of crease their insulin dose. hypertension was defined as the use of antihypertensive medicines Effect of Prednisone Withdrawal on Hypertension. Fourand/or a diastolic blood pressure ú 90 mmhg. teen of the 25 patients (56%) were hypertensive on 5 The length of follow-up after steroid withdrawal is at least 12 mg/d prednisone (diastolic blood pressure ú 90 mm Hg). After months in all patients. Statistical analysis was performed using the prednisone withdrawal, 7 of the 14 were able to either disconpaired t test with patients serving as their own controls. tinue their antihypertensive medicines (n Å 4) or decrease RESULTS the amount of antihypertensive medicines used (n Å 3). All remained normotensive at 9-month follow-up. One patient Effect of Prednisone Withdrawal on Graft Function. Of the was able to decrease the number of antihypertensive medi- 28 patients enrolled in the study, 19 had prednisone stopped cines from two to one and two decreased their antihypertensive with minimal complications (Table 1). Most patients complained medicine dose. of mild fatigue and body aches, but these resolved Ten of the 24 patients in the control group (41.7%) were quickly. One female patient developed marked alopecia, but hypertensive at the beginning of the study year. All remained this resolved in 2 months. hypertensive during the study period although two patients Four patients developed modest elevations in their liver were able to decrease their antihypertensive medicine dostransaminases (aspartate transaminase ú 50 IU/mL) and all age. of these elevations occurred 5 to 7 weeks after withdrawal. Effect of Prednisone Withdrawal on Serum Choles- Two patients had liver biopsies performed. One biopsy terol. Figure 1 shows that fasting serum cholesterol levels showed mild acute rejection. The other biopsy showed only fell in almost all patients who underwent prednisone with- minimal acute rejection. Two other patients lived too far from drawal. The mean serum cholesterol fell significantly (222.6 our center to undergo liver biopsy procedure. All four patients { 43.3 vs { 33.3 mg/dl, P õ.001) and the number of were treated with bolus corticosteroids (500 mg methylpred- patients with serum cholesterol levels ú220 mg/dl decreased nisolone IV daily for 3 days) then immediate rewithdrawal. from 14 to only 4. Two patients were treated with cholesterol-

3 HEPATOLOGY Vol. 25, No. 1, 1997 STEGALL ET AL. 175 TABLE 2. Effect of Prednisone Withdrawal on Insulin Requirements and Glucose Control in Patients With Diabetes Insulin Dose Hemoglobin A1C Prewithdrawal* Postwithdrawal Prewithdrawal Postwithdrawal Patient 1 40U 0U Patient 2 26U 0U Patient 3 40U 0U Patient 4 40U 40U Patient 5 40U 40U not measured *5 mg prednisone/d. 9 mo after being on 0 mg prednisone/d. Total daily insulin dose before and after prednisone withdrawal. Glycosylated hemoglobin level before and after prednisone withdrawal. lowering agents and both had decreased cholesterol levels logic graft loss is rare after liver transplantation combined after steroid withdrawal. Both have been maintained on cho- with the severe side effects of long-term prednisone therapy lesterol-lowering agents, one with a cholesterol of 242 mg/dl supports steroid withdrawal in liver transplantation paand the other with a level of 225 mg/dl. Of the five diabetic tients. patients, four were hypercholesterolemic on 5 mg/d predni- Most steroid withdrawal studies after liver transplantation sone whereas only one was hypercholesterolemic after pred- have been conducted in pediatric recipients. 5-7 In 11% to nisone withdrawal. This patient also remained diabetic. In 26.8% of these children, prednisone was restarted usually to general, serum cholesterol levels fell quickly reaching a new treat rejection. In adults, Padbury et al. have reported on baseline in the first month after withdrawal. In contrast, the 151 liver transplantation recipients who underwent steroid serum cholesterol of the three patients unable to be with- withdrawal 3 months after liver transplantation. 8 Eighty-five drawn from steroids did not change (data not shown). percent of these patients were able to be withdrawn from In the control group maintained on 5 mg/d prednisone, the steroids with a 7.8% incidence of rejection after withdrawal. mean serum cholesterol before the study period was similar Of concern with this early steroid withdrawal protocol was to the cholesterol level at the end of the study year (210 { that 6 of the 12 patients developing acute rejection pro vs. 208 { 38.6 mg/dl). No patient in the control group gressed to ductopenic rejection. Of these six there were three who had a serum cholesterol level greater than 220 mg/dl graft losses and two patient deaths. In our steroid withdrawal decreased their serum cholesterol below this level while being group who were at least 2 years after transplantation, four maintained on 5 mg/d prednisone. patients (14.2%) developed elevations of liver transaminases and all of these resolved with a short course of steroids and DISCUSSION rewithdrawal. The fact that there were no abnormalities in Complete withdrawal of prednisone after liver transplanta- all five protocol biopsies suggested that no subclinical rejection is not a common practice at most transplantation cen- tion occurred in the patients off steroids. In addition, 1 year ters. The possibility of late acute rejection that may lead to after steroid withdrawal we have no biochemical evidence of retransplantation or even death in a patient who was other- late acute rejection or chronic rejection in patients on wise doing well argues against steroid withdrawal in liver cyclosporine monotherapy. transplantation recipients. However, the fact that immuno- Recently, McDiarmid et al. have shown similar success in steroid withdrawal in both adult and pediatric liver transplantation recipients. 9 They withdrew steroids more than 1 year after liver transplantation in 31 patients who remained on a cyclosporine and azathioprine regimen. They noted only two episodes of acute rejection (6.5%), both of which resolved with only reinstitution of oral prednisone. At 1-year followup they have observed no chronic or intractable rejections. They also detected two rejection episodes in a control group not taken off steroids. In our study, five patients were not able to be withdrawn from steroids using the initial 1-month taper protocol. Four patients developed severe fatigue and one had documented adrenal insufficiency. Two of these patients were able to be withdrawn from steroids over a 7-month period using a slow taper. Some patients, such as the patient with colitis, may need to be on steroids for other reasons. However, this patient eventually tolerated slow prednisone withdrawal. A slower prednisone tapering schedule (decreasing by 1 mg/mo, for example) may have been better tolerated. Although not statistically significant, women may have greater difficulty with steroid withdrawal than men. All six of the patients who developed complications after prednisone FIG. 1. Comparison of fasting serum cholesterol levels in liver transplanta- withdrawal were women and two of the three that have had tion recipients on 5 mg/d prednisone and 0 mg/d prednisone. The mean serum to remain on prednisone were women. cholesterol decreased from { 43.3 mg/dl to { 33.3 mg/dl (P õ.001, paired t test) and the number of patients considered hypercholesterolemic Taken together, our study and that of McDiarmid et al. 9 (serum cholesterol ú 220 mg/dl, shaded area) decreased from 13 to 4. show that steroid withdrawal late after liver transplantation

4 176 STEGALL ET AL. HEPATOLOGY January 1997 is safe and does not lead to a high incidence of acute rejection the control and study group had been on 5 mg/d prednisone or graft loss at 1-year of follow-up. Our study suggests that for at least 6 months and had shown no improvement in cyclosporine monotherapy may be sufficient immunosuppres- metabolic complications during this time. Second, the control sion late after liver transplantation. Mor et al. have suggested group maintained on 5 mg/d prednisone showed no improve- that late rejection after liver transplantation occurs ment in metabolic complications during the year-long study in patients with cyclosporine trough levels õ200 mg/dl. 10 period. Third, the observation that almost all study patients With this in mind, we strive to maintain cyclosporine levels showed improvements early after streroid withdrawal argues ú200 mg/dl in patients on cyclosporine monotherapy. for a cause-effect relationship. Finally, there were no differ- Our results support the concept that prednisone with- ences in the length of time posttransplantation between the drawal may be easier in liver transplantation recipients than responders and nonresponders. It is our impression that liver in other types of organ recipients. Steroid withdrawal in renal transplantation patients more than 2 years after transplantation transplantation recipients has been extensively studied, but maintained on stable immunosuppression do not rejection rates of up to 50% and the risk of chronic rejection show improvements in hypertension, hypercholesterolemia, have made withdrawal controversial In addition, the results and diabetes. of the Canadian multicenter trial showed a significantly These results suggest that 5 mg/d of prednisone is not a decreased 5-year renal allograft survival in patients undergoing physiological dose of prednisone. The intracellular effect of steroid withdrawal. 15 In heart transplantation recipients, prednisone may persist for 24 hours and the high peak levels steroid withdrawal has been successful in up to 80% of patients may contribute to the metabolic complications in these pa- late after transplantation, but chronic rejection has tients. Although we did not measure other complications usu- been common ally associated with prednisone treatment in this study (bone In our study of liver transplantation recipients, three of disease, cataracts, infection, and so on) it is possible that the five insulin-dependent diabetic patients were able to discontinue these side effects might be decreased with complete predni- insulin therapy after steroid withdrawal. The fact sone withdrawal. that steroid withdrawal did not completely reverse all of the Our study suggests that prednisone withdrawal late after diabetics is not surprising. Other factors such as pretrans- liver transplantation in adults does not lead to a high incidence plantation diabetes, weight gain, and aging could contribute of rejection and graft loss at 1-year follow-up. Similar to the persistence of insulin-dependent diabetes after predni- to the results obtained in renal transplantation recipients, sone withdrawal. Cyclosporine also might be diabetogenic steroid withdrawal decreases serum cholesterol levels in These results agree with those found in renal transplantation most liver transplantation patients. In addition, steroid withdrawal recipients who also showed a decrease in the prevalence of may allow some diabetic patients to discontinue insurecipients diabetes after steroid withdrawal. 20 lin therapy and reverse hypertension in other patients. Steroids seem to contribute to hypertension in some liver We currently recommend that an attempt should be made transplantation recipients. Steroid withdrawal decreased hy- to discontinue prednisone in selected patients with stable pertension in 4 of 14 patients and allowed three others to graft function 2 years after liver transplantation. We believe decrease the amount of antihypertensive medicines they were that motivated, compliant patients who can be followed taking. In Padbury and colleagues study of early steroid closely are reasonable candidates for a steroid withdrawal withdrawal after liver transplantation, the prevalence of hy- protocol. However, these patients will be on cyclosporine pertension was 26%. 8 The 40% prevalence of hypertension in monotherapy and may be susceptible to rejection if their our patients off steroids may reflect the long-term effects cyclosporine level decreases. Thus we also recommend close of cyclosporine therapy. 21 Steroid withdrawal also has been monitoring of these patients not only during prednisone with- shown to decrease hypertension in renal transplantation re- drawal, but also monthly thereafter. Any increase in transaminase cipients. 22 levels is indication for a liver biopsy procedure and/ Fasting serum cholesterol levels and the number of pa- or the reinstitution of prednisone. We believe that with tients considered hypercholesterolemic (cholesterol ú220 mg/ proper monitoring prednisone can be safely discontinued and dl) decreased after steroid withdrawal. This effect suggests that some of the metabolic complications associated with steroid that corticosteroids are a major causative agent in hypercholesterolemia treatment will improve in many patients. in liver transplantation recipients. McDiarmid et al. showed a trend toward decreased cholesterol levels after REFERENCES steroid withdrawal in liver transplantation recipients, but 1. Belle SH, Beringer KC, Detre KM. Trends in liver transplantation in the this decrease was not statistically significant. 9 Prednisone United States. In: Teraskaki PI, Cecka JM, eds. Clinical Transplants Los Angeles: UCLA Tissue Typing Laboratory, 1994: withdrawal also has been shown to decrease cholesterol levels 2. Abu-Elmagd K, Todo S, Fung J, Demetris J, Rakela J, Rao AS, Iwatsuki in renal transplantation recipients. 23 S, et al. Hepatic transplantation at the University of Pittsburgh: New It should be emphasized that the patients in this study are horizons and paradigms after 30 years of experience. In: Teraskaki PI, a preselect group that do not reflect the entire population Cecka JM, eds. Clinical Transplants Los Angeles: UCLA Tissue Typing Laboratory, 1995: or even the entire liver transplantation population at our 3. Klintmalm GB, Nery JR, Husberg BS, Gonwa TA, Tillery GW. Rejection institution. Even so, the residual incidence of diabetes (8%) in liver transplantation. HEPATOLOGY 1989;10: and hypercholesterolemia (4%) are not significantly different 4. Stegall MD, Everson G, Schroter G, Bilir B, Karrer F, Kam I. Metabolic from the general population suggesting that prednisone may complications after liver transplantation: diabetes, hypercholesterolemia, hypertension and obesity. Transplantation 1995;60: be a major contributor to the high prevalence of these compli- 5. Dunn SP, Falkenstein K, Lawrence JP, Meyers R, Vincur CD, Billmire cations seen after liver transplantation. In contrast, the 40% DF, Weintraub WH. Monotherapy with cyclosporine for chronic immunosuppression in pediatric liver transplant recipients. Transplantation 1994; incidence of hypertension, however, is still much higher than in the general population and suggests that other factors, 57: Andrews WS, Shimaoka S, Sommerauer J, Moore P, Hudgins P. Steroid such as cyclosporine, contribute to hypertension in this pawithdrawal after pediatric liver transplantation. Transplant Proc 1994; tient population. 26: The possibility exists that the observed reversal of diabenance 7. Margarit C, Martinez-Ibanez V, Tormo R, Infante D, Iglesias H. Maintetes, hypertension, and hypercholesterolemia in this study immunosuppression without steroids in pediatric liver transplanta- tion. Transplant Proc 1989;21: may have occurred over time if this group of patients had 8. Padbury RT, Gunson BK, Dousset B, Hubscher SG, Buckels JAC, Neuremained on 5 mg/d prednisone. However, we believe that berger JM, Elias E, et al. Steroid withdrawal from long-term immunosuppression in liver allograft recipients. Transplantation several points argue against this. First, all patients in both 1993;55:

5 HEPATOLOGY Vol. 25, No. 1, 1997 STEGALL ET AL McDiarmid SV, Farmer D, Goldstein LI, Martin P, Vargas J, Tipton JR, N, Randrich R, Drinkwater DC, et al. Initial success of steroid weaning Simmons F, et al. A randomized prospective trial of steroid withdrawal late after heart transplantation. J Heart Lung Transplant 1992;11:428- after liver transplantation. Transplantation 1995;60: Mor E, Gonwa TA, Husberg BS, Goldstein RM, Klintmalm GB. Late-onset 17. Miller LW, Wolford T, McBride LR, Peigh P, Pennington DG. Successful acute rejection in orthotopic liver transplantation associated risk factors withdrawal of corticosteroids in heart transplantation. J Heart Lung and outcome. Transplantation 1992;54: Transplant 1992;11: Stratta RJ, Armbrust MJ, Oh CS, Pirsch JD, Kalayoglu M, Sollinger HW, 18. Robertson RP. Cyclosporine-induced inhibition of insulin secretion in isolated Belzer FO. Withdrawal of steroid immunosuppression in renal transplant rat islets and HIT cells. Diabetes 1986;35: recipients. Transplantation 1988;45: Roth D, Milgrom M, Esquenazi V, Fuller L, Burke G, Miller J. Post-transplant 12. Hricik DE, O Toole MA, Schulak JA, Herson J. Steroid-free immunosuppression hyperglycemia: increased incidence in cyclosporine-treated renal al- in cyclosprorine-treated renal transplant recipients: a meta-anal- lograft recipients. Transplantation 1989;47: ysis. J Am Soc Nephrol 1993;4: Hricick DE, Bartucci MR, Moir EJ, Mayes JT, Schulak JA. Effects of 13. Ratcliffe PJ, Firth JC, Higgins RM, Smith D, Gray DW, Morris PJ. Ran- steroid withdrawal on posttransplant diabetes mellitus in cyclosporinetreated domized controlled trial of complete steroid withdrawal in renal transplant renal transplant recipients. Transplantation 1991;51: recipients receiving triple immunosuppression. Transplant Proc 1993;25: 21. Bennett WM, Porter GA. Cyclosporine-associated hypertension. Am J Med ;85: Hricik DE, Whalen CC, Lautman J, Bartucci MR, Moir EJ, Mayes JH, 22. Hricick DE, Lautman J, Bartucci MR, Moir EJ, Mayes JT, Schuk JA. Schulak JA. Withdrawal of steroids after renal transplantation clinical Variable effects of steroid withdrawal on blood pressure reduction ion predictors of outcome. Transplantation 1992;53: cyclosporine-treated renal transplant recipients. Transplantation 1992;53: 15. Sinclair NR. Low-dose steroid therapy in cyclosporine-treated recipients with well-functioning grafts. The Canadian Multicentre Transplant Study 23. Hricick DE, Bartucci MR, Mayes JT, Schulak JA. The effects of steroid Group. Can Med Assoc J 1992;147: withdrawal on the lipoprotein profiles of cyclosporine-treated kidney and 16. Kobashigawa JA, Stevenson LW, Brownfield ED, Morigucki JC, Kawata kidney-pancreas transplant recipients. Transplantation 1992;54: