Secondary Logo

Share this article on:

Posttransplant Diabetes Mellitus and Acute Rejection: Impact on Kidney Transplant Outcome

Matas, Arthur J.1,3; Gillingham, Kristen J.1; Humar, Abhinav1; Ibrahim, Hassan N.2; Payne, William D.1; Gruessner, Rainer W. G.1; Dunn, Ty B.1; Sutherland, David E. R.1; Najarian, John S.1; Kandaswamy, Raja1

doi: 10.1097/TP.0b013e318160ee42
Original Articles: Clinical Transplantation

Background. The benefits (e.g., low acute rejection [AR] rate) vs. the long-term risk of each immunosuppressive protocol may determine the protocol’s value.

Methods. We studied the long-term impact of new-onset posttransplant diabetes (PTDM) and/or AR in 1,487 adult, primary transplant, nondiabetic recipients. Per Cox regression, donor source, AR, and PTDM were independent risk factors for graft loss (each, p<.0001). Recipients were subdivided by donor source and into these 4 groups: no AR, no PTDM [n=857]; no AR, PTDM [n=134]; ≥1 AR, no PTDM [n=403]; ≥1 AR, PTDM [n=93].

Results. There was a significant difference between groups in 15-yr actuarial graft survival (GS) and death-censored (DC) GS (p<.0001). Importantly, ≥1 AR had more impact on 15-yr GS and DC GS than did PTDM; the worst outcome was for those having both AR and PTDM. In separate analyses, we censored those with >1 AR; and then only compared those developing AR or PTDM in the first year. The results were similar—the AR (no PTDM) group did worse than the PTDM (no AR) group (p<.001).

Conclusions. Determining long-term risks associated with immunosuppressive protocols is important for treating future patients. Our data suggests that 15-year actuarial outcome (GS and DC GS) is worse for those developing AR than for those developing PTDM.

1 Department of Surgery, University of Minnesota, Minneapolis, MN.

2 Division of Renal Diseases and Hypertension, Department of Medicine, University of Minnesota, Minneapolis, MN.

This work was supported by NIH Grant #DK13085.

3 Address correspondence to: Arthur J. Matas, M.D., Department of Surgery, University of Minnesota, MMC 195, 420 Delaware St. SE, Minneapolis, MN 55455.

E-mail: matas001@umn.edu

Received 26 June 2007. Revision requested 13 September 2007.

Accepted 23 October 2007.

Immunosuppressive strategies must take into account the risk and consequences of an acute rejection (AR) episode, of immunosuppressive therapy side effects, and of graft loss. At each transplant center, protocols are typically developed to maximize efficacy (i.e., to minimize the incidence of AR) while minimizing toxicity. Most protocols combine two or more immunosuppressive drugs; for some of these drugs, the side effect profiles are additive and, when used in combination, the risk of a specific side effect increases.

Clinical reports often focus on short-term efficacy (e.g., the incidence of AR) or toxicity (i.e., the incidence of a specific side effect). But what clinicians also want to know is the long-term impact of a protocol for their patient population.

Both AR episodes and new-onset posttransplant diabetes mellitus (PTDM) have been shown to impair long-term posttransplant outcome (1–15). The incidence of AR and of PTDM differs for different immunosuppressive protocols (14, 16–21). In this study, the question was which would be the worst complication for adult, primary transplant, nondiabetic kidney recipients: AR or PTDM.

Back to Top | Article Outline

MATERIALS AND METHODS

From January 1, 1984, through March 1, 2006, a total of 1487 adult, nondiabetic patients underwent a primary kidney transplant at the University of Minnesota (614 deceased donor [DD]; 873 living donor [LD]). The immunosuppressive protocols have been described in detail (22). In brief, DD recipients were treated with sequential therapy: antibody induction; an antiproliferative drug, initially azathioprine (AZA), and then mycophenolate mofetil; and a calcineurin inhibitor, either cyclosporine or tacrolimus. Before 2000, all DD recipients were treated with a prednisone taper and long-term maintenance prednisone; subsequently, they were treated with a rapid discontinuation of prednisone protocol (prednisone given only for the first 5 days posttransplant) (23).

From 1984 through 1996, LD recipients were treated with triple therapy: cyclosporine, AZA, and prednisone. Subsequently, they were treated with sequential therapy using a protocol similar to that for DD transplants. A rapid discontinuation of prednisone protocol for primary LD transplant recipients began in 1999 (23).

Recipients with ≥25% elevation in serum creatinine level underwent percutaneous allograft biopsy. Mild to moderate AR episodes were treated with a recycling of the prednisone taper; steroid-resistant AR episodes and severe AR episodes were treated with antibody.

PTDM was defined as the need for treatment (either oral agents or insulin) >1 month posttransplant (in recipients who were not diabetic pretransplant). Recipients who needed treatment while receiving high-dose prednisone early posttransplant, but who stopped treatment within the first month, were not classified as having PTDM.

All posttransplant information is kept on the database approved by the authors’ institution’s Human Subjects Committee. The impact of AR and PTDM on long-term patient and graft survival rates was studied through four separate analyses of the data. First, the entire recipient cohort was studied. Second, because the previous studies showed that recipients with ≥2 AR episodes have significantly increased late graft loss (vs. those with only 1 AR episode), the outcome in those with only 1 AR episode versus those with PTDM was studied. Third, because AR or PTDM can occur at any time posttransplant, only recipients developing AR or PTDM in the first posttransplant year were studied. Fourth, because those developing PTDM and requiring insulin (vs. being treated with oral hypoglycemic agents) may have a worse prognosis, only recipients who, in the first year, developed AR versus those with PTDM requiring insulin treatment >3 months were studied.

For each cohort, first a Cox regression analysis was performed to determine whether donor source, AR, and PTDM were independent risk factors for graft loss. The recipients were then divided into four groups (no AR, no PTDM; no AR, PTDM; AR, no PTDM; and AR, PTDM); we calculated actuarial patient, graft, and death-censored graft survival rates. Survival rates were compared using Kaplan-Meier life table analyses. To compare differences between groups, we used log-rank and Wilcoxon tests.

Back to Top | Article Outline

RESULTS

Of the 1487 primary kidney transplant recipients in this study, 857 had neither AR nor PTDM, 134 had PTDM but no AR, 403 had AR but no PTDM, and 93 had AR and PTDM. Recipient characteristics are shown in Table 1.

TABLE 1

TABLE 1

There were no differences between groups in the cause of primary kidney disease, gender, or percent panel-reactive antibody (PRA). But there were significant differences in age (P<0.001), ethnicity (P=0.001), and percent receiving a DD transplant (P<0.001). The most striking differences were between the group with neither AR nor PTDM and the group with both AR and PTDM. However, in a comparison between the group with no AR but PTDM and the group with AR but no PTDM, only age at transplant was significantly different—recipients with AR but no PTDM were significantly younger (P<0.001).

Here are the results of the three separate analyses.

Back to Top | Article Outline

All Recipients

Cox regression analysis showed that use of a DD (vs. LD) (hazard ratio [HR]=1.99), the occurrence of ≥1 AR episode (vs. 0) (HR=2.23), and the presence of PTDM (vs. its absence) were independent risk factors for graft loss (each, P<0.0001).

After dividing recipients by donor source and by the presence or absence of AR or PTDM, we compared actuarial outcome. There was a significant difference in the 15-year actuarial graft survival rate between recipients with AR versus PTDM (for all recipients, for LD recipients, and for DD recipients) (P<0.0001) (Fig. 1). Those with both AR and PTDM had the worst outcome. However, of note, the occurrence of ≥1 AR episode had significantly more impact on graft survival than did PTDM.

FIGURE 1.

FIGURE 1.

Similarly, the 15-year actuarial death-censored graft survival rate significantly differed between groups (P<0.0001) (Fig. 2). Again, the occurrence of ≥1 AR episode had more impact than the development of PTDM.

FIGURE 2.

FIGURE 2.

For the entire population (DD recipients and LD recipients), the 15-year patient survival rate differed between those with AR versus PTDM (P<0.0001) (Fig. 3). However, after dividing recipients by donor source, we noted a significant difference in patient survival for LD recipients with AR versus PTDM (P=0.02), but no such difference for DD recipients (P=0.1).

FIGURE 3.

FIGURE 3.

When comparing only recipients with AR (but no PTDM) versus PTDM (but no AR), we found that those with AR (but no PTDM) had a significantly worse graft survival rate (P=0.0002) and death-censored graft survival rate (P<0.0001). But there was no significant difference in patient survival rates.

Back to Top | Article Outline

Only 1 AR Episode Versus PTDM

The analyses were repeated, censoring recipients with >1 AR episode (reasoning that including those with multiple AR episodes might negatively bias results of the AR group). The findings with this cohort were similar—Cox regression analysis showed that use of a DD (HR=2.2, P<0.0001) and the occurrence of an AR episode (HR=1.7, P<0.0001) were independent risk factors for graft loss (each comparison, P<0.001). After dividing recipients by donor source and by the presence or absence of AR or PTDM, we compared actuarial outcome. There was a significant difference in the 15-year actuarial death-censored graft survival rate and in the 15-year actuarial patient survival rate (each comparison, P<0.0001) between groups (data not shown). The 15-year actuarial graft survival rate and the 15-year actuarial death-censored graft survival rate were worse for recipients with only 1 AR episode (but no PTDM), when compared with recipients with PTDM (but no AR). Recipients with both AR and PTDM had the worst outcome.

When comparing only recipients with AR (but no PTDM) versus PTDM (but no AR), we found that those with AR (but no PTDM) had a significantly worse 15-year actuarial graft survival rate (P=0.003) and 15-year actuarial death-censored graft survival rate (P<0.001). The difference in 15-year actuarial patient survival rate was of borderline significance (P=0.06).

Back to Top | Article Outline

AR in the First Year Versus PTDM in the First Year

The analyses were repeated, comparing recipients with AR in the first year versus PTDM in the first year posttransplant. Again, the results with this cohort were similar to the earlier results. Cox regression analysis showed that use of a DD (HR=2; P<0.0001), occurrence of an AR episode (HR=1.8; P<0.001), and presence of PTDM (HR=1.4; P=0.01) were independent risk factors for graft loss. After dividing recipients by donor source and by the presence or absence of AR or PTDM, we compared actuarial outcome. There was a significant difference in the 15-year actuarial graft survival rate, in the 15-year actuarial death-censored graft survival rate, and in the 15-year actuarial patient survival rate (each comparison, P<0.0001) between groups. The 15-year actuarial graft survival rate and 15-year actuarial death-censored graft survival rate were worse for recipients with only 1 AR episode (but no PTDM), when compared with those developing PTDM (but no AR).

When comparing only recipients with AR (but no PTDM) in the first year versus PTDM (but no AR) in the first year, we found that those with AR (but no PTDM) had a significantly worse 15-year actuarial graft survival rate (P=0.002) and 15-year actuarial death-censored graft survival rate (P<0.0001).

Back to Top | Article Outline

AR in the First Year Versus PTDM Requiring Insulin in the First Year

The analyses were repeated, comparing recipients with AR in the first year versus PTDM requiring insulin >3 months in the first year posttransplant. Again, the results with this cohort were similar to the earlier results. Cox regression analysis showed that use of a DD (HR=1.9; P<0.0001) and occurrence of an AR episode (HR=1.9; P<0.0001) were independent risk factors for graft loss. After dividing recipients by donor source and by the presence or absence of AR or PTDM, we compared actuarial outcome. There was a significant difference in the 15-year actuarial graft survival rate, in the 15-year actuarial death-censored graft survival rate, and in the 15-year actuarial patient survival rate (each comparison, P<0.0001) between groups. The 15-year actuarial graft survival rate and the 15-year actuarial death-censored graft survival rate were worse for recipients with only 1 AR episode (but no PTDM), when compared with those developing PTDM (but no AR); those with AR and PTDM had the worst outcome.

When comparing only recipients with AR (but no PTDM) in the first year versus PTDM requiring insulin >3 months (but no AR) in the first year, we found that those with AR (but no PTDM) had a significantly worse 15-year actuarial graft survival rate (P=0.0002) and 15-year actuarial death-censored graft survival rate (P<0.0001). There was no difference in patient survival.

Back to Top | Article Outline

DISCUSSION

Numerous reports have documented the impact of AR or PTDM on long-term transplant outcome. Yet, we were surprised by this study’s findings. We anticipated that long-term outcome would be worse in recipients with PTDM (vs. AR); in fact, however, long-term outcome for those with AR, even a single AR episode, was worse. Of note, recipient characteristics differed between groups (Table 1). But the only significant difference between recipients developing PTDM (but no AR) and recipients with AR (but no PTDM) was age—those with AR (but no PTDM) were significantly younger.

This study has several limitations. First, the number of recipients followed >12 years was small. It is likely that much of the impact of PTDM occurs late posttransplant. A study with longer follow-up may show that recipients developing PTDM have worse patient and graft survival rates than those with AR. But, clearly, the short- to medium-term findings suggest that the occurrence of AR is worse than PTDM.

Second, the numbers in each of the subgroups were relatively small. Of the 1487 primary transplant recipients during the 12-year study period, 403 had ≥1 AR episode and no PTDM; 134 developed PTDM but did not have an AR episode. To minimize survival bias (i.e., to account for the fact that recipients who develop AR or PTDM late posttransplant have already survived a number of years), a subgroup of recipients with AR (vs. a subgroup who developed PTDM) in the first year posttransplant were studied. It was again found that outcome was worse for recipients with AR (vs. PTDM). But the subgroup with AR had many risk categories (e.g., differing histologic severity). Perhaps recipients with only a single mild AR episode did not fare worse than those developing PTDM.

Third, the definition of PTDM included all cases of diabetes posttransplant. Because of the small numbers, it was difficult to differentiate between the severity of diabetes, or between recipients whose treatment (insulin or oral medications) was temporary versus permanent. However, the subset developing PTDM in the first year and requiring insulin for >3 months (and presumably for an extended period of time) (Analysis 4) had better outcome than recipients developing AR in the first year. Still, Cosio et al. showed that hyperglycemia alone (after the first posttransplant month) is a significant risk factor for worse posttransplant outcome (6).

Fourth, >90% of the recipient population is white. Other single-center reports, particularly in areas with a higher proportion of black or Hispanic recipients, have noted a much higher incidence of PTDM.

Finally, in this study, only patient and graft survival rates were assessed. Other outcome measures, such as quality of life, may also differ between groups. For example, if recipients with PTDM had a significantly worse quality of life than those with AR, the findings would be tempered.

Ideally, this study should be replicated using a large multicenter, or national, database. A much larger study could control for other characteristics such as recipient age (Table 1). Such broad findings could help individual centers choose immunosuppressive protocols. Each immunosuppressive drug is associated with individual side effects and with a rate of developing those side effects. Combining the drugs into a protocol might increase the rate of some side effects. For example, the rate of PTDM is higher when tacrolimus is used with prednisone (vs. when tacrolimus is used in a prednisone-free protocol); the rate of nephrotoxicity is higher when cyclosporine is used with sirolimus (vs. when cyclosporine is used with other drugs).

Clinical studies often report only on efficacy (e.g., the rate of AR in the first 6 months) or only on individual side effects of one immunosuppressive protocol versus another. But numerous individual side effects—PTDM (1–8), AR (9–15), hypertension (24–26), and renal dysfunction (27–29)—have been shown to affect long-term outcome for kidney transplant recipients. Other factors affecting survival in the general population (e.g., dyslipidemia) likely also affect posttransplant survival. If the rate of development of each side effect and the impact on long-term outcome could be determined, a formula could be developed to be used when introducing a new immunosuppressive drug or protocol. The formula could take into account the incidence and impact of each side effect (e.g., AR, PTDM) and could be used as a predictor (surrogate marker) of good long-term outcome.

In summary, determining the link between long-term outcome and immunosuppressive protocols is important when deciding how to treat future kidney transplant recipients. This study suggests that the 15-year actuarial graft survival rate and the 15-year actuarial death-censored graft survival rate are worse for recipients with AR than for those with PTDM.

Back to Top | Article Outline

ACKNOWLEDGMENTS

The authors thank Mary Knatterud for editorial assistance and Stephanie Daily for preparation of the article.

Back to Top | Article Outline

REFERENCES

1. Miles AM, Sumrani N, Horowitz R, et al. Diabetes mellitus after renal transplantation: As deleterious as non-transplant-associated diabetes? Transplantation 1998; 65: 380.
2. Revanur VK, Jardine AG, Kingsmore DB, Jaques BC, Hamilton DH, Jindal RM. Influence of diabetes mellitus on patient and graft survival in recipients of kidney transplantation. Clin Transplant 2001; 15: 89.
3. Cosio FG, Pesavento TE, Kim S, Osei K, Henry M, Ferguson RM. Patient survival after renal transplantation: IV. Impact of posttransplant diabetes. Kidney Int 2002; 62: 1440.
4. Kasiske BL, Snyder JJ, Gilbertson D, Matas AJ. Diabetes mellitus after kidney transplantation in the United States. Am J Transplant 2003; 3: 178.
5. Gonzalez-Posada JM, Hernandez D, Bayes GB, Garcia Perez J, Rivero SM. Impact of diabetes mellitus on kidney transplant recipients in Spain. Nephrol Dial Transplant 2004; 19(suppl 3): iii57.
6. Cosio FG, Kudva Y, van der Velde M, et al. New onset hyperglycemia and diabetes are associated with increased cardiovascular risk after kidney transplantation. Kidney Int 2005; 67: 2415.
7. Hjelmesaeth J, Hartmann A, Leivestad T, et al. The impact of early-diagnosed new-onset posttransplantation diabetes mellitus on survival and major cardiac events. Kidney Int 2006; 69: 588.
8. Porrini E, Delgado P, Bigo C, et al. Impact of metabolic syndrome on graft function and survival after cadaveric renal transplantation. Am J Kidney Dis 2006; 48: 134.
9. Basadonna GP, Matas AJ, Gillingham KJ, et al. Early versus late acute renal allograft rejection: Impact on chronic rejection. Transplantation 1993; 55: 993.
10. Almond PS, Matas A, Gillingham K, et al. Risk factors for chronic rejection in renal allograft recipients. Transplantation 1993; 55: 752.
11. Lindholm A, Ohlman S, Albrechtsen D, Tufveson G, Persson H, Persson NH. The impact of acute rejection episodes on long-term graft function and outcome in 1347 primary renal transplants treated by 3 cyclosporine regimens. Transplantation 1993; 56: 307.
12. Matas AJ, Gillingham KJ, Payne WD, Najarian JS. The impact of an acute rejection episode on long-term renal allograft survival (t1/2). Transplantation 1994; 57: 857.
13. Cole E, Naimark D, Aprile M, et al. An analysis of predictors of long-term cadaveric renal allograft survival. Clin Transplant 1995; 9: 282.
14. Tanabe K, Takahashi K, Toma H. Causes of long-term graft failure in renal transplantation. World J Urol 1996; 14: 230.
15. Humar A, Hassoun A, Kandaswamy R, Payne WD, Sutherland DE, Matas AJ. Immunologic factors: The major risk for decreased long-term renal allograft survival. Transplantation 1999; 68: 1842.
16. Pascual J, van Hooff JP, Salmela K, Lang P, Rigotti P, Budde K. Three-year observational follow-up of a multicenter, randomized trial on tacrolimus-based therapy with withdrawal of steroids or mycophenolate mofetil after renal transplant. Transplantation 2006; 82: 55.
17. Romagnoli J, Citterio F, Nanni G, et al. Incidence of posttransplant diabetes mellitus in kidney transplant recipients immunosuppressed with sirolimus in combination with cyslosporine. Transplant Proc 2006; 38: 1034.
18. Ciancio G, Burke GW, Gaynor JJ, et al. A randomized long-term trial of tacrolimus/sirolimus versus tacrolimus/mycophenolate versus cyclosporine/sirolimus in renal transplantation: Three-year analysis. Transplantation 2006; 81: 845.
19. Araki M, Flechner SM, Ismail HR, et al. Posttransplant diabetes mellitus in kidney transplant recipients receiving calcineurin or mTOR inhibitor drugs. Transplantation 2006; 81: 335.
20. van Hooff JP, Christianns MH, van Duijnhoven EM. Tacrolimus and posttransplant diabetes mellitus in renal transplantation. Transplantation 2005; 79: 1465.
21. Bouchta NB, Ghisdal L, Abramowicz D, et al. Conversion from tacrolimus to cyclosporin is associated with a significant improvement of glucose metabolism in patients with new-onset diabetes mellitus after renal transplantation. Transplant Proc 2005; 37: 1857.
22. Matas AJ, Sutherland DE, Najarian JS. Evolution of immunosuppression at the University of Minnesota. Transplant Proc 2004; 36(suppl 2): 64S.
23. Matas AJ, Kandaswamy R, Gillingham KJ, et al. Prednisone-free maintenance immunosuppression—A 5-year experience. Am J Transplant 2005; 5: 2473.
24. Opelz G, Wujciak T, Ritz E. Association of chronic kidney graft failure with recipient blood pressure. Collaborative Transplant Study. Kidney Int 1998; 53: 217.
25. Opelz G, Dohler B; Collaborative Transplant Study. Improved long-term outcomes after renal transplantation associated with blood pressure control. Am J Transplant 2005; 5: 2725.
26. Fernandez-Fresnedo G, Palomar R, Escallada R, et al. Hypertension and long-term renal allograft survival: Effect of early glomerular filtration rate. Nephrol Dial Transplant 2001; 16(suppl 1): 105.
27. Hariharan S, McBride MA, Cherikh WS, Tolleris CB, Bresnahan BA, Johnson CP. Post-transplant renal function in the first year predicts long-term kidney transplant survival. Kidney Int 2002; 62: 311.
28. Salvadori M, Rosati A, Bock A, et al. One-year posttransplant renal function is a strong predictor of long-term kidney function: Results from the Neoral-MOST Observational Study. Transplant Proc 2003; 35: 2863.
29. Tonelli M, Wiebe N, Culleton B, et al. Chronic kidney disease and mortality risk: A systematic review. J Am Soc Nephrol 2006; 17: 2034.
Keywords:

Kidney transplant; Acute rejection; Diabetes mellitus

© 2008 Lippincott Williams & Wilkins, Inc.