|Home | About | Journals | Submit | Contact Us | Français|
An increased frequency of kidney rejection has been reported in diabetic patients who have simultaneous pancreas and kidney transplantation compared with patients who have a kidney transplant alone. Kidney graft outcome is similar in the two groups. The mechanism for increased kidney graft rejection with a simultaneous pancreas graft is not clear. It is ascribed to the immunogenicity of the exocrine pancreas that initiates migration of activated cells from the peripheral blood that are entrapped in the kidney. Since the volume of the transplanted tissue is less in islet transplantation (usually <2 ml) than in pancreas transplantation, one might not expect an increased frequency of kidney rejection in islet cell recipients. We looked at biopsy-proven kidney rejection episodes in patients who had combined kidney and islet transplants and compared this with the frequency of rejection in diabetic and nondiabetic patients who underwent a kidney transplant alone under the same immunosuppression. Diabetic patients who had kidney islet transplants (n=9) had a higher frequency of rejection (100%) compared with diabetic patients (n=107, 55.1%) and nondiabetic patients (n=327, 65%) who had a kidney transplant alone. The 1-year graft and patient survival rates were not different among the groups. Although the number of patients is small, it would appear that transplantation of a low volume of islet cells with high purity can lead to an increased frequency of kidney rejection. This is unlikely to be explained solely on the basis of fewer antigen matches in these recipients but may reflect the inherent immunogenicity of the purified islet preparations. Alternatively, there may be an effect of their direct infusion into the portal vein.
When a normal kidney is transplanted into a diabetic patient with abnormal glucose metabolism, characteristic changes induced by diabetes occur in the transplanted kidney over a variable period, sometimes faster than the time of onset noted in native kidneys (1). This consequence can be avoided if a successful pancreas transplant is performed restoring euglycemia (2, 3). It has been reported that there is no overall adverse effect of performing pancreas transplant on the outcome of the patient or the transplant kidney (4, 5). With the increasing success of pancreas transplantation, this approach is increasingly accepted as the closest approximation of the ideal of long-term restoration of normal metabolism. While the long-term kidney graft outcome is similar in diabetic patients undergoing combined pancreas and kidney grafts, there are many series that show an increased frequency of acute kidney rejection episodes in this group of patients (6–8). This has not been noted in all series (9, 10). The mechanism of how the pancreas graft might induce rejection in the transplanted kidney is not known; however, on the basis of experimental evidence it has been proposed that activated cells from the circulation migrate and lodge in the transplanted kidney (11).
While this is controversial, exocrine tissue probably contributes significantly to the immunogenicity of islet preparations (12–16). Since the volume of nonislet cells is much lower with an islet graft than with a whole pancreas graft, an increased frequency of kidney rejection might not be expected in patients undergoing islet transplantation. It was therefore of interest to examine the frequency of kidney rejection episodes in patients who underwent combined kidney and purified islet transplantation.
Eight patients aged 29–38 years with long-standing insulin-dependent (type I) diabetes mellitus as evidenced by an absent C-peptide response to either glucagon or Sustacal stimulation received 9 combined cadaveric kidney-islet grafts (one retransplant), with one (n=6), two (n=2), or three (n=1) islet donors. The cadaveric donor ABO types were all compatible with recipient types and HLA matching was random, the antigen match being 0–2 for the kidney and 0–3 for islets (Table 1). All patients had a negative crossmatch. One patient who underwent the procedure died on the fifth postoperative day of aspiration pneumonia and did not have rejection until this time. This patient was not included in the analysis of frequency of rejection, but was included in the calculation of mortality and graft survival.
Human islets were obtained by a modification (17) of the automated method for human islet isolation (18) and purified by centrifugation on discontinuous density gradients (Eurocollins-Ficoll, density=1.108, 1.096, 1.037) using a Cobe 2991 cell separator (Cobe, Lakewood, CO) (19, 20). Immediately after renal transplantation, an upper midline incision was performed and a 16-gauge catheter was placed in a jejunal vein for islet infusion. In one patient (No.3) the catheter was left in place for transplantation of islets from a third-party donor.
Patient 5 received islets via the transhepatic route from a third-party donor, since islets of sufficient quality could not be obtained from the pancreas of the kidney donor. The transhepatic route was also used to transplant islets from a second donor in patient 8. Islets were all transplanted fresh or after overnight culture at room temperature.
All patients received a 100-mg bolus of methylprednisolone during the operation, followed by a decreasing prednisone dose from 200 mg to 20 mg over the course of the first week posttransplant. FK506 was given at a dose of 0.1 mg/kg intravenously administered as a continuous infusion over 24 hr, beginning immediately following the transplantation. When patients resumed a solid diet, an oral dose of 0.15 mg/kg/bid was started. Patients 6–8 also received Imuran 200 mg/day during the first postoperative week in addition to the previously mentioned immunosuppression.
Kidney biopsy was performed to confirm a clinical impression of rejection, and the tissue was processed immediately in formalin and stained with hematoxylin and eosin, Jones, trichrome, and PAS stains. The only rejection episodes considered in this report were those confirmed histologically by biopsy. Rejection episodes were graded as mild, moderate, or severe as follows. Mild acute cellular rejection was considered as: small aggregates of mononuclear cells with evidence of tubulitis. Moderate: same as above with some coalescence of mononuclear aggregates. Severe: same as above with evidence of necrosis, interstitial hemorrhage, and acute inflammation or vasculitis.
Data were analyzed by chi square test with continuity correction. Significance of difference was considered as P < 0.05.
Six-month graft survival was 86%, 76%, and 78% in diabetic recipients of a solitary kidney graft (DK),* nondiabetic kidney transplant recipients (NDK), and diabetic recipients of kidney and islets (DKI), respectively (Table 2). One-year graft survival was 82% (DK), 73% (NDK), and 78% (DKI). Mortality rates were not different in the groups. The unexpected finding was the frequency of kidney rejection episodes: 55.1% in DK, 65% in NDK, and 100% in DKI patients (P < 0.02).
The number and grade of biopsy-proven rejection episodes and current graft outcome in DKI patients are shown in Table 3. Most episodes were mild and easily treated with corticosteroids. Only one graft was lost to refractory recurrent rejection episodes. After rejection episodes, 5 patients still have demonstrable islet cell function, but all patients still require exogenous insulin. The best results were obtained in the patients who received islets from more than one donor pancreas.
In order to examine the rejection frequency in more detail, a matched case control study from the original group was done. Diabetic (n=9) and nondiabetic (n=9) controls were randomly selected for each kidney-islet recipient using matching criteria of age, sex, and time of transplant. The frequency of kidney rejection episodes was compared between kidney islet patients and controls using McNemar’s test. Mean and median creatinine, FK dose, and FK levels were compared between the groups using a one-way repeated measure analysis of variance and Friedman’s nonparametric test (21).
Again, the frequency of rejection was higher in DKI recipients—100% vs. 55.6% in DK case controls and 33.3% in NDK case controls. The FK dose was 13.3±5.9 mg/day in DKI, 10.7±7.1 mg/day in DK, and 14.0±9.1 mg/day in NDK (P>0.05). The FK levels were also similar in the groups: 0.6±0.3 ng/ml (DKI), 0.6±0.6 (DK), and 0.8±0.4 (NDK) (P>0.05). Creatinine levels at a similar time point were no different among the groups.
Although the number of patients in our kidney islet group is small, the frequency of rejection episodes seems out of proportion to what was expected for islet transplantation. As with the rejection episodes seen in patients with kidney and pancreas grafts, these rejection episodes are largely manageable, and graft and patient outcomes are similar in the groups studied.
With this small number of patients it was not possible to analyze if significant differences in HLA matches were present in the three groups. In the recipients of kidney transplant alone, 2/3 of the patients had 2-antigen matches or less, similar to our patients.
It has been assumed in whole pancreas transplant that the exocrine gland initiates the recipient response and that reactive cells migrate to the kidney and possibly provide some protection to the pancreas (11). The reactivity to islets was supposed to be minimal.
In light of these data showing an apparent increased frequency of kidney rejection, the immunogenicity of the islet preparation and the intravascular route of administration (portal vein) must be considered as possible determining factors. Despite the small volume and relative apparent purity of the islet preparations, significant numbers of contaminating cells are still present. The potentially immunogenic cells include acini, ducts, lymphoid, and dendritic cells (22).
The intravascular administration of antigen and cells may facilitate antigen recognition and activate the immune response by direct presentation of antigen to circulating recipient lymphoid cells. The site of the intravascular antigen and cell administration may also be a key since in other systems, the portal venous route produces different immune responses than systemic venous administration (23).
Our data also indicate that a single mild rejection episode, even in a patient who receives islets from multiple donors, may be enough to compromise the ability to achieve complete insulin and independence, although diabetes control may be stabilized (17). It is therefore critical to develop effective procedures to decrease the immunogenicity of islet preparations before transplantation or to explore alternative sites or routes of administration. It will be of interest to determine if increased frequency of kidney rejection has been observed following transplantation of kidney and islets that have been cryopreserved or cultured for 1 week at 24 °C. Experimental evidence suggests that these procedures may be effective pretransplant treatments to decrease islet preparation immunogenicity (24, 25).
The authors acknowledge Dr. Mike Nalesnik for review of the biopsies, and the secretarial assistance of Ms. T. Whigham and Ms. K. Murray in the preparation of the manuscript.
1Presented at the 18th Annual Meeting of the American Society of Transplant Surgeons, May 27–29, 1992, Chicago, IL.
2This work was supported by a grant from The Juvenile Diabetes Foundation (Research Grant 1911421) and by a General Clinic Research Center Grant of The University of Pittsburgh Medical Center (NIH Grant 5MO1RR00056).
*Abbreviations: DK, diabetic kidney recipient; DKI, diabetic kidney and islet recipient; NDK, nondiabetic kidney recipient.