|
|
| |||||||||||||||||||||||||||||||||||||||
Background and Methods The safety of long-term immunosuppression with cyclosporine in renal-transplant recipients is not well understood. This drug may cause a progressive toxic nephropathy, but it also preserves renal function because it prevents rejection. To determine the effect of cyclosporine on renal function and graft rejection, we conducted a retrospective analysis of data on 1663 renal-transplant recipients at six centers.
Results The rate of graft survival was 78 percent (median follow-up, 36 months). Grafts were lost in 279 patients (17 percent), mostly because of acute rejection (68 patients) or chronic graft dysfunction that was unresponsive to a reduction in the dose of cyclosporine (125 patients); 92 patients died with functioning grafts. The median change in the serum creatinine concentration in all patients after transplantation was less than 0.001 mg per deciliter per month (<0.09 µmol per liter per month). Patients who had episodes of rejection had decreased rates of long-term graft function and survival. Eight percent of patients with functioning grafts at one year had first episodes of rejection more than one year after transplantation. These late first rejections were associated with noncompliance with therapy (in 34 percent), blood cyclosporine concentrations that were marginally lower than those of patients who had no episodes of rejection, and a low rate of successful reversal of rejection (77 percent, vs. 97 percent in patients with rejection during the first year; P<0.001).
Conclusions The majority of renal-transplant patients tolerate long-term cyclosporine therapy without evidence of progressive toxic nephropathy. Graft failure is most often due to rejection. .
The assumption that cyclosporine causes progressive nephropathy has recently been questioned, in part because many patients receiving long-term therapy have impaired but stable renal function1,20,21. Whether or not treatment with cyclosporine causes progressive renal failure is an important issue, especially because the drug is such a potent immunosuppressive agent when high blood concentrations are maintained and is often responsible for long-term graft survival22,23,24,25. We studied and report here the renal effects of long-term treatment with cyclosporine in patients at six kidney-transplant centers.
Methods
Demographic Characteristics
We retrospectively studied 1663 patients with end-stage renal disease who underwent kidney transplantation at the University of Florida, the University of Wisconsin, Vanderbilt University, Thomas Jefferson Hospital, the Geisinger Medical Center, or Brigham and Women's Hospital. Except at Brigham and Women's Hospital, where the study group was a random sample of 68 patients, consecutive patients were enrolled. The results did not vary among the centers, except as noted below. The information collected on each patient included age, sex, race, type of donor (cadaver or living related donor), number of HLA mismatches, the percentage of panel-reactive antibodies in serum, concurrent diseases, and transplantation history. Data on the patients were evaluated every three months after transplantation, at which times their blood pressure, weight, serum creatinine and cholesterol concentrations, and blood or plasma cyclosporine concentrations were measured and the dose of cyclosporine was recorded. The patients underwent an average of 14 assessments in the first three years after transplantation.
Protocols
At Vanderbilt and the University of Wisconsin, antilymphocyte antibody was given to patients who received cadaveric transplants, and cyclosporine (8 to 10 mg per kilogram of body weight per day) was given when the patient's serum creatinine concentration fell to less than 3.0 mg per deciliter (265 µmol per liter). Patients who received kidneys from living related donors were initially treated with cyclosporine (8 to 10 mg per kilogram per day) at the University of Wisconsin and with antilymphocyte antibody at Vanderbilt University. At four centers, all patients received cyclosporine initially: 8 to 10 mg per kilogram per day orally at Thomas Jefferson Hospital and the Geisinger Medical Center and 4 mg per kilogram per day intravenously at the University of Florida and Brigham and Women's Hospital. Maintenance therapy consisted of cyclosporine, azathioprine, and prednisone at five centers and cyclosporine and prednisone at Thomas Jefferson Hospital.
Rejection
Data on acute episodes of rejection were available from all the centers except the University of Florida. Acute rejection was defined by the initiation of antirejection therapy and was confirmed by biopsy in 39 percent of the episodes. Complete reversal of rejection was defined as a return of the serum creatinine concentration (±10 percent) to its level before the episode. Recurrent rejection was defined as rejection occurring at least seven days after the serum creatinine concentration had returned to the base-line value recorded before the first episode. If the serum creatinine concentration did not return to the base-line value, then a rejection episode was considered new only if it occurred more than 30 days after therapy was initiated for the first episode of rejection. A functioning graft was defined by the absence of any need for dialysis, regardless of the patient's serum creatinine concentration. Chronic graft dysfunction was defined as a progressive reduction in renal function that did not respond to reductions in the dose of cyclosporine, that was not due to structural abnormalities such as ureteral obstruction, bladder dysfunction, or renal-artery stenosis, and that was not due to recurrent renal disease or glomerulonephritis of new onset.
Blood or Plasma Cyclosporine Concentrations
To compare the circulating cyclosporine concentrations measured at six institutions with different assays, we developed a five-point scale (Table 1). The values for group 3 represent the target drug level decided on by consensus. Actual blood or plasma cyclosporine concentrations were used in the analyses of the results at each center.
|
Changes in the serum creatinine concentration were analyzed by linear regression for patients for whom three or more values were available. The reciprocal of the serum creatinine values (1/Scr) was analyzed in the same way, with similar results (data not shown). Since the primary goal of the study was to examine long-term renal function, the first three months after transplantation were excluded from the analyses of serial serum creatinine values so that early renal dysfunction would not affect the analysis of long-term renal function. Survival distributions were estimated by the product-limit method. Proportional-hazards analysis with time-varying covariates were used to assess data on rejection. Both the loss of the graft and the death of the patient were regarded as graft failures.
Results
Between January 1, 1983, and January 1, 1990, 1745 patients at the six centers received renal allografts and were given cyclosporine as primary immunosuppressive therapy. We excluded 82 patients without follow-up data on renal function. The mean age of the remaining 1663 patients was 40 years, 80 percent were white, 61 percent were men, and 78 percent had received primary allografts. Graft survival was excellent; 1297 of the 1663 patients (78 percent) were alive with a functioning graft after a median follow-up of 36 months; 361 patients (22 percent) were followed for at least 4 years. A total of 279 grafts were lost. The leading causes of graft loss were acute rejection (68 patients) and chronic graft dysfunction (125 patients), which together accounted for 69 percent of the losses. Technical failures, primary nonfunction, vascular thrombosis, other clinical complications, and recurrence of primary disease accounted for graft failures in another 86 patients. Ninety-two patients died with functioning grafts.
Safety and Renal Function
The mean daily dose of cyclosporine decreased over time; it was 6 mg per kilogram after three to six months of therapy, 4.3 mg per kilogram after one year, 3.7 mg per kilogram after two years, 3.6 mg per kilogram after three years, 3.4 mg per kilogram after four years, and 3 mg per kilogram after five years. Cyclosporine was discontinued in 149 patients; nephrotoxicity was cited as the reason in 30 of these cases. Other reasons included chronic graft dysfunction (29 patients), the physician's preference (24 patients), inability to tolerate the drug (19 patients), noncompliance with the drug regimen (7 patients), infections (7 patients), cost (8 patients), and lymphoma (3 patients). The reason was not determined for 22 patients.
The mean serum creatinine concentration at each evaluation from 6 months to 48 months after transplantation was 1.9 mg per deciliter (168 µmol per liter) (Figure 1). In an analysis of median serum creatinine values, we avoided survivor bias (possibly leading to lower serum creatinine values) by including the results for patients with graft failure at all time points and assigning them an arbitrarily high value (10 mg per deciliter [884 µmol per liter]). According to this analysis the median serum creatinine values were 1.8 to 2.0 mg per deciliter (159 to 177 µmol per liter) throughout the first three years. Thus, graft loss in a few patients (<15 percent) did not create a censoring effect that reduced the mean serum creatinine values among the patients with functioning grafts.
|
The median change in the serum creatinine concentration for all patients was <0.001 mg per deciliter per month (0.09 µmol per liter per month) (Table 2). Among the patients with functioning grafts, the serum creatinine concentration decreased at a rate of 0.003 mg per deciliter per month (0.3 µmol per liter per month). In this subgroup, the slope of serum creatinine over time was less than -0.04 mg per deciliter per year (-3 µmol per liter per year) in half the patients and less than 0.13 mg per deciliter per year (11 µmol per liter per year) in three fourths (a negative slope indicates that renal function in the majority of patients improved over time). The magnitude of the improvement is too small to have clinical relevance, but it does demonstrate that there was no trend toward a progressive loss of renal function. There was no significant difference in the extent of change in the serum creatinine concentration between patients receiving 4 mg or less of cyclosporine per kilogram daily and those receiving more than 4 mg per kilogram daily.
|
Pretransplantation variables (such as the number of HLA mismatches, the number of previous grafts, race, the percentage of panel-reactive antibodies, the type of donor, sex, and the presence or absence of diabetes) were unrelated to the changes in serum creatinine values. The single exception was the patient's age (P = 0.03 by analysis of variance). Among recipients less than 30 years of age, the mean slope was 0.05 mg per deciliter per month (4 µmol per liter per month), as compared with less than 0.001 mg per deciliter per month (0.09 µmol per liter per month) for recipients 30 to 50 years of age and -0.04 mg per deciliter per month (-0.3 µmol per liter per month) for those more than 50 years old.
Rejection
Altogether, 765 patients (46 percent) had episodes of acute rejection, 86 percent of which occurred in the first year. Complete reversal of rejection occurred in 444 of these patients (58 percent) and partial reversal in 268 (35 percent). The mean cyclosporine concentrations during the month preceding the episode of rejection and at the time it occurred were similar in the 222 patients for whom data on both times were available. Thus, there was no evidence that cyclosporine concentrations immediately preceding rejection were predictive of rejection.
Among 160 patients receiving 4 mg or less of cyclosporine per kilogram per day during the first three months, 19 (12 percent) had an episode of rejection each between four and six months after transplantation, as compared with 18 of 365 patients (5 percent) who received more than 4 mg per kilogram (P = 0.009). However, no significant correlation between the dose of cyclosporine and the frequency of rejection was identified for any later time.
Among the 691 patients who had no acute rejection episodes and who had functioning grafts one year after transplantation, 56 each later had an episode of acute rejection. Cyclosporine concentrations measured at the time of rejection in 47 of these patients were low in 83 percent (groups 1 and 2 in Table 1), as compared with 73 percent of the entire population at two years. The response to antirejection therapy was poor in the patients with late first episodes of rejection; only 77 percent had complete or partial reversal of rejection, as compared with 97 percent of patients whose rejection episodes occurred during the first year after transplantation (P<0.001).
Among the 56 patients who had late first rejections, 19 were noncompliant, and in 12 of the 19 the graft was lost. However, the rate of graft loss caused by late first-rejection episodes in compliant patients was also substantial (14 of 37 patients [38 percent]). There were no differences in sex, age, or source of the transplant between the compliant and noncompliant patients who had late first rejections. More blacks were noncompliant than whites (9 of 13 vs. 10 of 43). Women were disproportionately represented in the group with late first-rejection episodes; 32 of 279 women (11 percent) had such episodes, as compared with 24 of 412 men (6 percent). However, survival analysis to evaluate the statistical possibility that the women were followed longer suggested that sex was not significantly related to the occurrence of late first-rejection episodes.
Effects of Rejection
The occurrence of rejection at any time had a significantly negative effect on graft survival. Rejection episodes at any time after three months from transplantation were significantly worse in terms of outcome than early rejection (P<0.001) (Figure 2). For example, 88 percent of patients with early rejections (occurring less than one month after transplantation) had functioning grafts one year after the episodes of rejection, and 71 percent had functioning grafts four years after the episodes. In contrast, 95 percent and 81 percent of patients with no episodes of rejection had functioning grafts one and four years after transplantation, respectively (Figure 2A). The difference was greatest among the patients who had first rejections more than one year after transplantation; only 63 percent had functioning grafts one year after the rejection, and 43 percent at the time of the last follow-up contact (Figure 2B). Late graft loss was increased despite complete or partial reversal of acute rejection in 97 percent of patients who had rejection episodes during the first year. Furthermore, four-year graft survival was essentially the same whether the rejection in the first year was completely or partially reversed (74 percent vs. 66 percent, P = 0.23).
|
Cyclosporine Concentrations and Long-Term Renal Function
Each patient was assigned to one of five groups defined by the cyclosporine concentration at the start of each observation period (Table 1). The mean and median changes in serum creatinine values as a function of the cyclosporine concentrations were determined by comparing the initial serum creatinine value at each successive year of follow-up (Table 3). The patients in the higher-cyclosporine-concentration groups had better renal function in the first and third years after transplantation. We also compared cyclosporine concentrations in three patient groups defined according to the serum creatinine values at the last follow-up contact (
2.2 mg per deciliter, 2.3 to 3.4 mg per deciliter, and 
3.5 mg per deciliter [ 195, 196 to 301, and 302 µmol per liter]); this analysis was similar to those in two previous reports22,24. At three of the six centers, the patients with lower serum creatinine concentrations had higher cyclosporine concentrations.
|
We found that kidney-transplant recipients treated with cyclosporine had stable graft function. Although the dose of cyclosporine decreased over time in most patients, the daily dose was not an independent predictor of graft function. Moreover, there was no correlation between poor graft function and higher early doses of cyclosporine, in contrast to the suggestion in one report that among patients with autoimmune disease, a high early dose of cyclosporine was the best predictor of long-term renal dysfunction26.
The mean follow-up in this study was 36 months, and some patients were followed for 5 years. Consequently, the results may not reflect later deterioration in renal function. In 347 patients treated with cyclosporine for 10 years at one center, however, there was no evidence of progressive nephropathy27. Similarly, in the European Multicentre Trial5 renal function was stable at five years despite an average cyclosporine dose of 4.9 mg per kilogram per day at that time.
Even a single episode of rejection had an adverse effect on long-term graft survival and function. Similarly, data from the United Network for Organ Sharing Registry indicated that one-year graft survival was reduced from 85 percent to 67 percent by a single rejection episode28. Data from the UCLA registry also indicated that rejection reduced one-year graft survival, and even in patients with complete responses to antirejection therapy, graft survival was 8 to 15 percent lower than in patients with no rejection episodes29.
Several reports have suggested that higher blood concentrations of cyclosporine are associated with better long-term renal function22,23,24,30. In a study of 566 patients who received kidney transplants, the frequency of chronic rejection was higher in patients who received less than 5 mg of cyclosporine per kilogram per day31. In our study, changes in serum creatinine concentrations were correlated with cyclosporine concentrations: patients who received higher doses had better renal function for the first three years after transplantation (Table 3). This correlation could reflect a selection process in which patients with better renal function continue to receive higher doses of cyclosporine, whereas patients with declining renal function receive reduced doses. In any case, patients with higher cyclosporine concentrations did not have any evidence of progressive nephropathy; this finding is consistent with the results of the European Multicentre Trial,5 in which the mean dose of cyclosporine was higher than in our study (4.9 vs. 3.0 mg per kilogram per day).
The current tendency to reduce progressively both the dose and the concentration of cyclosporine has been based on the assumption that the drug can cause progressive nephropathy. Therefore, patients with stable renal function are given lower doses to avert renal damage, and patients with declining renal function (indicated by elevated serum creatinine values) are given lower doses to minimize toxicity. We found no evidence of progressive nephropathy even in patients who were given relatively high doses or had high cyclosporine concentrations. Thus, it is reasonable to conclude that the dose of cyclosporine should not be reduced solely on the assumption that the drug causes progressive renal damage. Moreover, an elevated serum creatinine concentration six months or more after transplantation should not be assumed to reflect drug toxicity. Chronic graft dysfunction was the most common cause of graft loss in our study and others,32 and it may be caused by inadequate immunosuppression22,31.
Our study has several limitations. Because it was based on a retrospective chart review, we could not control for variables such as a tendency to reduce the dose of cyclosporine in patients with elevated serum creatinine values. Also, because of the retrospective design, the criteria for the diagnosis and treatment of rejection and compliance at different centers were not standardized. Although it seems reasonable to summarize the results of cyclosporine assays, as we did, doing so could obscure valid results by converting a range of concentrations into categories. Of course, we acknowledge that renal function is better assessed by measuring clearance rather than serum creatinine. Thus, we cannot exclude the possibility that renal reserve decreased with time. Nonetheless, we believe that the stability of serum creatinine concentrations in this group of patients, the association of rejection with diminished graft function, and the indications of a beneficial relation among higher cyclosporine concentrations, better renal function, and reduced risk of rejection argue against cyclosporine as a cause of progressive nephropathy. These results should influence the way cyclosporine is used for immunosuppression in patients undergoing kidney transplantation.
Supported by a research grant from Sandoz Pharmaceuticals.
We are indebted to the following for their assistance: Bruce E. Jarrell, M.D., Elizabeth Mallon-Hunter, Shawn Moyer, Anita Cole, Ellen Stefanosky, Steve Dunn, and Dylan Burke at the Thomas Jefferson University Hospital; Sally Kohl, R.N., Anthony D'Alessandro, M.D., Munci Kalayoglu, M.D., Hans Sollinger, M.D., Ph.D., and Folkert Belzer, M.D., at the University of Wisconsin Transplant Program; Charles B. Carpenter, M.D., Terry B. Strom, M.D., Edgar L. Milford, M.D., Nicholas L. Tilney, M.D., Robert L. Kirkman, M.D., Amy Chang, R.N., Marta Tillman, and Terese Taylor at the Brigham and Women's Hospital Renal Transplant Program; William W. Pfaff, M.D., Richard J. Howard, M.D., John C. Peterson, M.D., Matthew E. Brunson, M.D., Todd Pickard, Joel T. Van Sickler, M.D., Robert Thompson, C.J. Parris, P.A., and Pam Patton, P.A., at the University of Florida Kidney Transplant Program; Robert E. Richie, M.D., H. Keith Johnson, M.D., Robert Mac Donnell, M.D., William Nylander, M.D., J. Harold Helderman, M.D., Lisa Trussler, R.N., Kim Byrum, R.N., and Alice Blanton, B.A., at the Vanderbilt University Transplant Center; Ruth Latsha, R.N., Joan Miller, R.N., Stephen E. Kelley, B.A., Joseph Bisordi, M.D., Fred Brown, M.D., Mildred Fleetwood, M.D., John Gerig, M.D., Jill Largent, M.D., Oscar Oberkircher, M.D., and Elizabeth Squiers, M.D., at the Geisinger Medical Center Transplant Section; and Anthony Venditti and Cynthia Romney at the Sandoz Pharmaceuticals Corporation.
Source Information
From the Thomas Jefferson University Hospital Transplant Program, Philadelphia (J.F.B.); the University of Wisconsin Transplant Program, Madison (J.D.P.); the Brigham and Women's Hospital Renal Transplant Program, Boston (E.L.R.); the University of Florida Kidney Transplant Program, Gainesville (D.R.S.); the EMMES Corporation, Potomac, Md. (D.M.S.); the Vanderbilt University Transplant Center, Nashville (D.H.V.B.); and the Transplant Section, Department of Surgery, Geisinger Medical Center, Danville, Pa. (J.C.W.).
Address reprint requests to Dr. Salomon at the Department of Molecular and Experimental Medicine -- SBR5, Scripps Research Institute, 10666 N. Torrey Pines Rd., La Jolla, CA 92037.
References
| |||||||||||||||||||||||||||||||||||||||
Related Letters:
Efficacy and Safety of Cyclosporine in Renal-Transplant Recipients
Singh A. K., Bennett W. M., Burke J. F., Pirsch J. D., Salomon D. R.
Extract |
Full Text
N Engl J Med 1994;
331:1777-1778, Dec 29, 1994.
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
