Delayed onset of graft function (DGF) is one of the determinants of long-term success of renal transplantation from deceased donors. Apart from ischemic and reperfusion injury, the nephrotoxic effects of immunosuppressive and other medications have been implicated as causes of DGF. Since the early 1990s, in our renal transplantation program from deceased donors, the rate of acute rejections has steadily decreased but the rate of DGF has remained at substantially the same level; DGF without doubt augurs significantly worse 1-year graft survival compared to transplantations with early onset of graft function (87% vs. 95%) (1–3). Antibody induction of sequential immunosuppression regimens is a way to avoid exposing the grafts to nephrotoxic agents during the very sensitive early posttransplant phase. These regimens, however, expose the patient to a higher risk of infections and malignancies and thus are not recommended for patients with low immunological risk (4).
Kaden et al. introduced a high-dose single-bolus antithymocyte globulin (ATG) therapy, which was reported to be free from the unwanted side effects of conventional ATG induction. The high-dose bolus of ATG causes a drastic decrease of the T-cell count already at the time of completion of vascular anastomoses, a theoretically perfect timing. Kaden et al. combined single-shot ATG with cyclosporine (CsA) triple immunosuppression and showed a reduced rejection rate and improved 3-year graft survival compared to transplantations without ATG induction (5, 6). Also, in simultaneous pancreas and kidney transplantation, the preconditioning of the recipients with single-bolus ATG was reported to decrease the rate of acute rejections (7). Further, a low rate of DGF and reduced number of acute rejections despite early initiation of CsA was reported with basiliximab, a chimerized interleukin-2 receptor antagonist (8–10) and that medication was not associated with the cytokine-release syndrome and other harmful side effects seen with conventional ATG induction (11).
Having this background we decided to run this prospective, randomized study where our primary objective was to examine whether the administration of single-shot ATG or two doses of basiliximab together with initial low-dose CsA triple regimen would reduce the rate of DGF and shorten the period of posttransplant oliguria as compared to conventional CsA triple therapy. Secondary objectives were the occurrence and timing of acute rejections, patient and graft survival, and safety and tolerability of these regimens. The study protocol included subanalyses on the immediate reperfusion injury and on the alloreactivity of the transplant recipients during the 1-year posttransplant follow up, the results of which have already been reported elsewhere (12–15).
PATIENTS AND METHODS
A total of 168 consecutive recipients of first or repeated deceased donor kidney transplants were recruited between December 1999 and March 2001 into this open, randomized, and controlled study. Patients younger than 16 years, older than 65 years, those with a malignant disease in history, and those with panel reactive antibodies (PRA) higher than 50% or a previous graft lost within 1 year for immunological reasons were not included into the study.
The most common causes of uremia were diabetes mellitus (35%), glomerulonephritis (23%), and polycystic kidney disease (19%). The mean age of the patients (109 men, 59 women) was 46.9 years. All patients had been on maintenance dialysis before transplantation, and they had undergone a blood transfusion program if they had not previously been transfused, transplanted, or pregnant.
Sharing of at least two human leukocyte antigen (HLA)-A/-B antigens and one HLA-DR antigen with the donor as well as a negative cytotoxic cross match test against donor spleen lymphocytes were required for transplantation. Recipient selection and the posttransplant immunological monitoring of the recipients have been previously described in detail (14). All kidney grafts were from heart beating deceased donors with mean age of 41.3 years (range 5 to 67).
Before the transplant operation, originally 140 patients were to be randomized in a ratio of 5:5:4 in blocks of 14 into the two study groups, group A (ATG induction) and group B (basiliximab induction), and into the control group C (CsA triple without induction). After randomization, it emerged 13 patients allotted to group A had to be withdrawn from the study (12 due to a positive hypersensitivity skin test against ATG test serum and one patient due to severe arteriosclerosis precluding transplantation). After a statistician’s consultation, two more blocks in ratio 8:4:2 were randomized to keep the group sizes adequate. Thus, the total number of randomized patients was 168 and the number of patients continuing in the study was 155 (53 in group A, 58 in group B, and 44 in group C). The 12 transplanted patients withdrawn from the study received our standard immunosuppression, similar to that in group C.
Computer-generated numbered randomization slips were sealed into consecutively numbered envelopes by a person not connected with the study, and opened consecutively when the patient had signed the informed consent form.
The study plan included assessment of kidney graft function, immunosuppressive medication, and surgical and infectious complications during the study visits at 3 weeks, 3 months, and 1 year after transplantation. Transplant biopsy was required before the initiation of antirejection therapy. A repeated biopsy was required in rejections not responding to steroids. If no kidney function was noted by 14 days after transplantation a graft biopsy was required to reveal a possible rejection during nonfunction. The histological findings were scored according to the Banff classification (16).
All patients gave their informed consent before entering the study. The study protocol and its amendments were approved by the local Ethics Committee.
Immunosuppressive and Other Medication
In group A, a single 9 mg/kg bolus of ATG (ATG-Fresenius, Fresenius AG, Bad-Homburg, Germany) in NaCl 0.9% was given as an infusion over 30 min during the time of vascular anastomoses and was completed before the graft reperfusion. CsA was given 2.5 mg/kg orally before the transplant operation and continued after the operation with 5 mg/kg/day divided in two doses until day 7, after which the dose was adjusted aiming at a target trough level on 200–300 μg/L. Methylprednisolone was given 250 mg IV at the beginning of surgery (about 1 hour before ATG), and 40 mg/day orally during postoperative days 1–4, tapering down to 20 mg/day by day 16, and to 10–12 mg/day by 3 months. Two 50 mg IV doses of azathioprine were given during the surgery and continued 2 mg/kg/day orally tapering down to 1 mg/kg/day on day 14 posttransplant.
In group B, 20 mg basiliximab (Simulect, Novartis Pharmaceutical Corp.) in NaCl 0.9% was given as an infusion over 30 min before graft reperfusion. A second dose was given on day 4 posttransplant. The administration of CsA, azathioprine, and methylprednisolone was as in group A.
In group C, CsA was started with a 5 mg/kg oral dose before the transplant operation and continued postoperatively with 10 mg/kg/day divided in two doses adjusted thereafter targeting a trough level of 200–300 μg/L. Azathioprine and methylprednisolone were given as in the other two groups.
Other standard medication consisted of 750 mg cefuroxime IV before the transplant operation and 12 hours later, sulfatrimetoprime (135 patients) or Pentamidine inhalation (13 patients) for Pneumocystis jiroveci prophylaxis for the first 6 months, and ganciclovir during antirejection therapy in transplantations from a cytomegalovirus (CMV) seropositive donor to a seronegative recipient.
Treatment of Rejection
First-line rejection therapy was a 5-day intravenous course of methylprednisolone (500 mg+250 mg+250 mg+125 mg+125 mg). Patients were treated with OKT 3 and/or switched to mycophenolate mofetil or tacrolimus if the rejection did not respond to steroids. A plasma exchange program was used when the clinical course and histological evidence suggested vascular rejection.
Assessment of Graft Function
DGF was defined as described by Halloran (17): plasma creatinine concentration higher than 500 μmol/L throughout the first posttransplant week, or need of more than one dialysis session in the first week, or oliguria <1 L/24 hr for more than 2 days. The day of onset of graft function was defined as the first day of spontaneous decrease of plasma creatinine concentration. The graft was considered failed when the graft was removed, the patient returned to maintenance dialysis, or the patient died with a functioning graft.
Kidney function was assessed during the study visits by recording the urinary output and plasma creatinine concentration. Creatinine clearance was measured at three weeks and at one year and it was calculated (Cockroft) annually during the follow up time.
Assessment of Surgical and Infectious Complications
Data on complications and adverse events during the one year study period were collected from the patient hospital records, from the Finnish Kidney Transplant Registry and from the patients during the study visits. Data on complications during the 5-year follow up time were collected from the regular reports of the nephrologists to the Finnish Kidney Transplant Registry.
Anesthesia and Measurement of Physiological Parameters
During transplant surgery, standardized balanced inhalation anesthesia was used. During the study, it turned out that ATG may cause ventilatory and hemodynamic instability. Thus, the anesthesia reports were reviewed retrospectively. The following factors were recorded in groups A and B before and during the period drug infusion, and in group C during the corresponding reference periods: lowest systolic blood pressure, use of vasoactive drugs, lowest hemoglobin oxygen saturation, highest fractional inspiratory concentration of oxygen, highest peak airway pressure adjusted to simultaneous level of positive end expiratory pressure (PEEP), central venous pressure, and end tidal concentration of volatile anesthetic. In addition to absolute values, drug infusion-induced changes of these parameters were also calculated: at least 5% decrease of hemoglobin oxygen saturation, 10% increase of inspiratory oxygen concentration, and 4 cmH2O increase of peak airway pressure.
According to the analysis of our kidney transplant results, the expected rate of DGF for the controls was 31% (1). On the basis on our earlier experience with basiliximab induction with full-dose initial CsA, we expected here a 10% DGF rate in the two induction immunosuppression groups, which was calculated to give a power of 77%. The results below are given for all 155 randomized and transplanted patients continuing in the study (see also Fig. 1) but separate analyses were also done to ensure that the patients randomized in two sets remained comparable.
Statistical methods included χ2 test for contingency tables, median test for medians, and Student’s t test and analysis of variance for comparisons of means.
All transplantations had a 5-year or longer follow-up time. The three groups were similar in terms of the demographic and transplantation data (Table 1). According to the study protocol, immunological high-risk patients were excluded, and accordingly in 118/155 transplantations, the patients were PRA negative. Of the 37 PRA positive patients, only two had a PRA level higher than 20%.
Onset of Graft Function
DGF was significantly less common in group A than in the other groups (Table 2). The patients with DGF in group C were on the average twice as long dialysis dependent as those in group A or B. The average need for posttransplant dialysis service per group was thus significantly different. One graft in group B never started to function. The overall frequency of DGF was 17.4% and the mean length of dialysis dependence in DGF was 12.2 (4–55) days.
Rejection rate was 11.3% in group A, 12.1% in group B, and 20.5% in group C (NS.). There was one steroid-resistant rejection in each group. No significant differences were found in the severity of rejections according to the Banff classification between the groups. The overall rejection frequency within the first posttransplant year was 14.2%.
The time when the rejections occurred was, however, significantly different in the groups. All rejections in group A occurred during the first month starting on the average on day 16 (median 13, range 7–29) posttransplant. In group B, the first posttransplant month was rejection free and the average rejection-free time was 97 (median 46, range 35–267) days posttransplant. In group C, both early and late rejections were seen and they occurred mean 101 (median 35, range 10–364) days posttransplant, with two of them diagnosed at the 1-year protocol biopsy. The difference in timing was significantly different between the groups when divided into early, medium, and late with cutoff points at 30 and 100 days (χ2P<0.005).
Kidney Graft Function and Survival
Kidney graft function as expressed by serum creatinine concentration, 3-week and 1-year measured creatinine clearance, as well as the 2- to 5-year annually calculated (Cockroft) creatinine clearance are given in Table 3. At 5 years, the graft survival (GS) in group B was significantly better than in group C (P<0.05).
During the first posttransplant year, six of the 155 grafts were lost, two of them (group C) by patient death (one in acute myocardial infarction and another in cardiac arrest during OKT-3 treatment for severe rejection). One graft was lost after withdrawal of immunosuppression due to a severe Aspergillus infection (group A) and one to recurrent primary kidney disease (group C). One graft never started to function and one patient returned to dialysis after 2 months of inferior kidney function (group B).
The death censored 5-year graft survivals in groups A, B, and C were 94.3, 96.6, and 90.0%, respectively (P=NS).
In groups A and B, the target CsA blood level was reached already by day 4, whereas in group C the levels overshot the target during the first week merging with the other two groups by day 14 (Fig. 2). The mean daily CsA dose in group C during days 1–3 posttransplant was 641 mg compared to 360 mg in group A and 333 mg in group B (P<0.0001). At 1 week, the mean CsA dose in group C (420 mg) was still significantly higher than in group A (346 mg) and group B (326 mg) (P<0.0001). At 2 weeks, the mean CsA doses were 358 mg (group A), 369 mg (group B), and 341 mg (group C) and at 3 weeks the respective doses were 342 mg, 347 mg, and 321 mg (NS). At the 3 month and 1 year visits, both CsA doses and trough levels were similar in the groups.
According to the study protocol, the immunosuppressive regimen was similar for all patients after the first posttransplant week. By 3 months, 27 patients had discontinued azathioprine. In 10 cases, it had been replaced by mycophenolate.
At the 1-year visit, 112 (75.2%) out of the 149 patients with still-functioning grafts were on the original CsA triple immunosuppression (group A: 79%, group B: 71%, group C: 76%; P=NS). Azathioprine had been replaced by mycophenolate in 22 cases and 11 patients were on CsA and steroids. One patient had stopped taking steroids and three patients had been converted to tacrolimus. The changes in medications did not differ significantly between the groups.
At the end of the follow up of 5 years, 141 patients still had a functioning graft. Of them, 135 (95.7%) were on CsA. Six patients were on tacrolimus, 55 were on azathioprine, and 73 on mycophenolate. Steroids had been discontinued in 64 patients.
Adverse Events During Study Drug Infusions
The infusion of ATG was performed in median 32 min (range 13–150) and that of basiliximab in median 30 min (range 8–89). The infusion was interrupted due to side effects in 3 of 53 cases in group A and in two of them it was resumed at a slower rate after a short interruption.
The study groups were comparable in terms of all recorded physiological parameters before administration of immunosuppressive drugs (data not shown). Despite the fact that vasoactive drugs were used more frequently in groups A and B than group C, systolic blood pressure was lower in group A than groups B or C (LTable 4). Airway pressure was higher in group A than groups B or C. At least 4 cm H20 increase of peak airway pressure was observed far more often in group A than groups B or C. Hemoglobin oxygen saturation was lower in group A than groups B or C. At least a 5% decrease of hemoglobin oxygen saturation was observed more often in group A than groups B or C, despite at least a 10% increase in fractional inspiratory concentration of oxygen in 16 ATG patients, but in none of the basiliximab patients and only in two controls.
Adverse Events During the 1-Year Study Period
Nonfatal complications were noted: Epstein-Barr virus infection (1), tuberculosis (1 pulmonary, 1 uterine), and pulmonary embolism (1). Clinical CMV infection was diagnosed in 22 patients on average 48 days (range 10–180) after transplantation. Three of them had antibodies against CMV at the time of transplantation. The other complications during the first posttransplant year are collected in Table 5. There was no statistically significant difference between the three groups.
Events During the 5-Year Follow-up
Among the 153 survivors at 1-year follow-up, four patients in group A were lost by the end of the 5-year follow-up time. The causes of death were septicemia, cardiovascular complication, and malignant disease in two. The two deaths in group B were one suicide and the other caused by gastrointestinal hemorrhage. In group C, two patients died in cardiovascular disease and one in malignancy. Three patients returned to dialysis. The cause of graft loss was rejection in two and poor donor kidney quality in one. There were no significant differences between the groups in the late complications. Overall there were 10 malignancies, 18 severe infectious complications, 10 cardio/cerebrovascular disturbances, 6 biliary complications necessitating surgical interventions, and one new-onset diabetes.
This study was planned to decrease the rate of DGF after kidney transplantation using induction immunosuppression together with a low initial calcineurin inhibitor protocol. Indeed, in transplantations with single-shot ATG induction, our hypothesis worked fine. Compared to our normal experience, in group A, an exceptionally low number of patients needed posttransplant dialysis and the number of days on dialysis was lower than with our standard medication as given to patients in group C. To our surprise, this was not quite true for the transplantations with basiliximab induction where the rate of DGF was at our usual level, albeit with shorter duration than expected. As these two groups had similar protocols for the first week CsA dosage, different exposure to the nephrotoxic effects of this drug does not explain the significantly different rate of DGF in these two groups. Neither could donor factors nor length of cold ischemia time explain this difference. ATG causes a profound T-cell depletion that can last for years (18, 19). T-cell subsets were not analyzed in this study as a whole, but in a subanalysis of this study, a prompt fall in the peripheral white blood cell count after ATG was noted already before reperfusion, which was not seen in the basiliximab group or controls (12). On the other hand, as demonstrated in the same subanalysis, leukocyte sequestration in the graft was strongly associated with DGF. A leukocyte sequestration might have been prevented by prereperfusion leukocyte depletion, thus diminishing the risk for DGF. This might explain why no difference in the rate of DGF between ATG and basiliximab was seen in the recent paper on high-risk patients with conventional ATG induction (20).
At 1 and 5 years, there was a trend for slightly better GS in groups A and B. As the study was not planned to be powered to show differences in graft survival, this effect can only be speculated on. The differences between the groups in this respect did not reach statistical significance at 1 or 5 years. At 5 years, only the difference in GS between groups B and C was statistically significant. However, in death-censored GS, the differences between the groups were minimal, only showing a trend for better survival in groups A and B.
The low number of rejections and also the delay of the first rejection in patients treated with basiliximab were in line with earlier reports (8, 21). The potentially increased rejection risk of low dose initial CsA seems to be well obviated by the high dose single bolus ATG and the two dose basiliximab induction. However, the postponing of rejections by basiliximab may be treacherous as it sets particularly high demands on patient surveillance during the time patients may already be out of the hands of the transplant unit. The trend towards more CMV infections in both antibody groups was noted and perhaps could have been avoided by a more ambitious prophylactic medication than the one used at the time of this study.
Hemodynamic or pulmonary instability has not previously been reported when this particular ATG preparation has been administered in a single-shot scheme (22, 23). Compared to previous publications, there are two differences in ATG administration that may explain the hemodynamic and pulmonary disturbances in our patients. Instead of 60-minute duration of ATG infusion as described by Kaden (22), drug infusion in our patients was intended to last 30 minutes, but in some patients it was even shorter than that. Further, the dose of methylprednisolone to avoid the cytokine release syndrome was only half on that used by Kaden (22).
The study was originally not planned to detect possible changes in hemodynamics or ventilatory mechanics. All now reported physiological data are based on retrospective review of the anesthesia reports and should be interpreted with caution. Thus, the fact that vasoactive drugs were used more often and oxygen was supplied in higher inspiratory concentrations in patients receiving ATG does not necessarily indicate that the patients were in worse condition than the patients of the other groups. Importantly, however, it implies that ATG patients were treated (i.e. blood pressure maintained and oxygen supply secured) at least as actively as the other patient groups. Despite this, ATG patients suffered from reduced blood pressure and oxygenation and increased airway pressure during ATG infusion. All in all, the collected data indicate the ATG infusion was related to the impairment of physiological homeostasis. In some patients, the deterioration was severe.
On the basis of this study, we can conclude that basiliximab was safe and effective against rejection but did not decrease the rate of DGF, which was the primary aim of this study. The high-dose single-shot ATG was superior in terms of onset of graft function and rejections and it did not carry any increased risk for severe infectious or malignant complications as seen after the conventional ATG induction. However, in order to be able to recommend high-dose single-shot ATG induction, the hemodynamic and ventilatory adverse events during ATG infusion seen in some of our patients urge certain modifications in the adjuvant immunosuppressive medication, and the transplant team must be well prepared for undesirable side effects possible with this kind of medication.
1. Kyllönen LEJ, Salmela KT, Eklund BH, et al. Long-term results of 1047 cadaveric kidney transplantations with special emphasis on initial graft function and rejection. Transpl Int
2000; 13: 122.
2. Kyllönen L, Salmela K. Transplantation of both kidneys from 408 donors; Comparison of results. Transpl Int
2000; 13 Suppl 1: S95.
3. Salmela K, Kyllönen L. Two decades of experience with cyclosporine in renal transplantation in Helsinki. Transplant Proc
2004; 36 Suppl: 94S.
4. Opelz G, Dohler B. Lymphomas after solid organ transplantation: A collaborative transplant study report. Am J Transplant
2004; 4: 222.
5. Kaden J, May G, Mueller P, et al. Intraoperative high-dose anti-T-lymphocyte globulin bolus in addition to triple-drug therapy improves kidney graft survival. Transplant Proc
1995; 27: 1060.
6. Kaden J, Strobelt V, May G. Short and long-term results after pretransplant high-dose single ATG-Fresenius bolus in cadaveric kidney transplantation. Transplant Proc
1998; 30: 4011.
7. Schoulz T, Konzack J, Buesing M. Mycophenolate mofetil/prednisolone/single-shot ATG with tacrolimus or cyclosporine in pancreas/kidney transplantation: First results of an ongoing prospective randomised trial. Transplant Proc
1999; 31: 591.
8. Sollinger H, Kaplan B, Pescowitz MD, et al. Basiliximab versus antithymocyte globulin for prevention of acute renal allograft rejection. Transplantation
2001; 72: 1915.
9. Ponticelli C, Cambi V, Shapira Z, et al. A multicenter double-blind placebo controlled study of basiliximab (Simulect) in combination with triple therapy including azathioprine for the prevention of acute rejection episodes in renal allograft patients. Transplantation
1999; 67: S151.
10. Mulloy LL, Wright F, Hall ML, Moore M. Simulect (basiliximab) reduces acute cellular rejection in renal allografts from cadaveric and living donors. Transplant Proc
1999; 31: 1210.
11. Kovarik J, Wolf P, Cisterne JM, et al. Disposition of basiliximab, an interleukin-2 receptor monoclonal antibody, in recipients of mismatched cadaver renal allografts. Transplantation
1997; 64: 1701.
12. Turunen AJ, Lindgren L, Salmela KT, et al. Association of graft neutrophil sequestration with delayed graft function
in clinical renal transplantation. Transplantation
2004; 77: 1821.
13. Turunen AJ, Fernandez JA, Lindgren L, et al. Activated protein C reduces graft neutrophil activation in clinical renal transplantation. Am J Transplant
2005; 5: 2204.
14. Matinlauri IH, Kyllönen LEJ, Eklund BH, et al. Weak humoral posttransplant alloresponse after a well-matched cadaveric kidney transplantation. Transplantation
2004; 78: 198.
15. Matinlauri IH, Kyllönen LEJ, Salmela KT, et al. Serum sCD30 in monitoring of alloresponse in well HLA-matched cadaveric kidney transplantations. Transplantation
2005; 80: 1809.
16. Racusen L, Solez K, Colvin RB, et al. The Banff 97 working classification of kidney allograft pathology. Kidney Int
1999; 55: 713.
17. Halloran PF, Aprile MA, Farewell V, et al. Early function as the principal correlate of graft survival. A multivariate analysis of 200 cadaveric renal transplants treated with a protocol incorporating antilymphocyte globulin and cyclosporine. Transplantation
1988; 46: 223.
18. Esposito L, Kamar N, Tkaczuk J, et al. Long-term evolution of lymphocytes subsets after induction therapy based on continuous versus discontinuous administration of anti-thymocyte globulins in renal transplant patients. Transplant Proc
2005; 37: 785.
19. Weimer R, Staak A, Suesal C, et al. ATG induction therapy: Long-term effects on Th1 but not Th2 responses. Transpl Int
2005; 18: 226.
20. Brennan D, Daller J, Lake K, et al. Rabbit antithymocyte globulin versus basiliximab in renal transplantation. N Engl J Med
2006; 355: 1967.
21. Mourad G, Rostaing L, Legendre C, et al. Sequential protocols using basiliximab versus anti-thymocyte globulins in renal-transplant patients receiving mycophenolate mofetil and steroids. Transplantation
2004; 78: 584.
22. Kaden J. Optimal management of induction therapy with ATG in kidney allograft recipients. Int J Immunotherapy
1999; 15: 115.
23. Schulz T, Papapostolou G, Schenker P, Kapischke M. Single-shot antithymocyte globulin (ATG) induction for pancreas/kidney transplantation: ATG-Fresenius versus Thymoglobulin. Transplant Proc
2005; 37: 1301.