Navigation

American Journal of Kideny Diseases, February 2003, Volume 41, Number 2

Original Investigations: Transplantation

Long-term survival following simultaneous kidney-pancreas transplantation versus kidney transplantation alone in patients with type 1 diabetes mellitus and renal failure

K. Sudhakar Reddy, MD, Don Stablein, PhD, Sarah Taranto, BA, Robert J. Stratta, MD, Thomas D. Johnston, MD, Thomas H. Waid, MD, J. Wade McKeown, MD, Bruce A. Lucas, MD, Dinesh Ranjan, MD

Abstract

Background: Pancreas transplantation improves quality of life and prevents the progression of secondary complications of diabetes. Whether these benefits translate into a long-term survival advantage is not entirely clear. Methods: Using the United Network for Organ Sharing database, we analyzed long-term survival in 18,549 patients with type 1 diabetes and renal failure who received a kidney transplant between 1987 and 1996. Patient survival was calculated using the Kaplan-Meier method. Proportional hazards models were used to adjust for effects of differences in recipient and donor variables between simultaneous kidney-pancreas transplants (SKPTs) and kidney-alone transplants. Results: SKPT and living donor kidney recipients had a significant crude survival distribution advantage over cadaver kidney transplant recipients (8-year survival rates: 72% for SKPT recipients, 72% for living donor kidney recipients, and 55% for cadaver kidney recipients). The survival advantage for SKPT recipients over cadaver kidney recipients diminished, but persisted after adjusting for donor and recipient variables and kidney graft function as time-varying covariates. SKPT recipients had a high mortality risk relative to living donor kidney recipients through 18 months posttransplantation (hazards ratio, 2.2; P < 0.001), but had a lower relative risk (hazard ratio, 0.86; P < 0.02) thereafter. In SKPT recipients, maintenance of a functioning pancreas graft was associated with a survival benefit. Conclusion: The long-term survival of SKPT recipients is superior to that of cadaver kidney transplant recipients with type 1 diabetes. There is no difference in survival of SKPT recipients and living donor kidney recipients with type 1 diabetes at up to 8 years' follow-up; the former have a greater early mortality risk and the latter have a greater late mortality risk. Results of this study suggest that successful simultaneous kidney-pancreas transplantation is not only life enhancing, but life saving. Am J Kidney Dis 41:464-470. © 2003 by the National Kidney Foundation, Inc.

Successful kidney transplantation, in addition to being cost-effective and improving quality of life, increases survival compared with long-term dialysis treatment in patients with end-stage renal disease caused by type 1 diabetes.1 Pancreas transplantation performed simultaneously with kidney transplantation further enhances quality of life and has been reported to stabilize or improve some secondary complications of diabetes mellitus.2-9 Although short-term patient survival is similar for recipients of simultaneous kidney-pancreas transplants (SKPTs) and cadaver kidney transplants,10 the effect of pancreas transplantation on long-term survival has remained uncertain.

In a recent registry report from The Netherlands, mortality after initiating renal replacement therapy was 50% less in the Leiden area compared with another area; the only difference between the two areas was the frequency of simultaneous kidney-pancreas transplantation (73% in the Leiden area compared with 37% in the non-Leiden area).11 Several other single-center reports also suggest that simultaneous kidney-pancreas transplantation results in a substantial decrease in mortality in patients with type 1 diabetes mellitus and renal failure compared with kidney transplantation alone.12-14 However, there generally is a selection bias, with SKPT recipients being younger and healthier and receiving organs from younger donors with shorter cold ischemia times compared with recipients of cadaver kidney transplants, making it difficult to determine the primary effect of pancreas transplantation on survival. In addition, single-center reports are limited by relatively small numbers of patients.

The objective of the present study is to determine the effect of pancreas transplantation on long-term survival of a large cohort of patients with type 1 diabetes with end-stage renal failure.

Methods

Using the United Network for Organ Sharing (UNOS) database, we analyzed the survival of 18,549 patients with type 1 diabetes who received a kidney transplant between 1987 and 1996. Patients who received segmental pancreas grafts from living donors were excluded from this analysis. According to the UNOS Registry, 9,956 patients received cadaver kidney-only transplants, 3,991 patients received living donor kidney transplants, and 4,602 patients received SKPTs during the study period. Mean follow-up in survivors was 4.8 years, with 95% followed up for at least 1 year and 25% followed up for 6.8 or more years.

Patient survival distribution estimates were calculated using the Kaplan-Meier method. Because there were differences between cadaver and living donor kidney transplants and SKPTs with respect to covariates known to influence graft and patient survival, proportional hazards models were used to adjust for effects of these variables (recipient age at time of transplantation, recipient race and sex, donor age, donor cause of death, previous kidney transplant, previous transfusions, cold ischemia time, percentage of panel reactive antibody, center volume, and year of transplantation) on the probability of graft and patient survival. Time-varying covariates were used to examine graft function as a predictor of survival and time-varying changes in relative hazard function. Analysis was performed using the SAS statistical program (SAS Institute, Cary, NC).

We also analyzed survival of patients with functioning kidney grafts at the end of the first year to avoid the potential effect of a failed kidney graft as the cause of inferior patient survival. In addition, we compared survival of SKPT recipients with functioning kidney and pancreas grafts at the end of the first year posttransplantation with survival of those with a functioning kidney graft, but failed pancreas graft, within the first year posttransplantation.

Recipients of SKPTs were younger and more often Caucasian, received kidneys from younger donors with a cause of death of head trauma more often as opposed to cardiocerebrovascular in origin, and had shorter cold ischemia times and thereby less often needed dialysis therapy in the first week posttransplantation compared with cadaver kidney transplant recipients (Table 1).

Simultaneous kidney-pancreas transplantation has become more common in recent years, accounting for 10.8% of cadaver donor transplantations in patients with type 1 diabetes in 1987 and 1988 and 54.7% in 1996. Simultaneousfont-family:Arial;color:red'>kidney-pancreas transplantation is performed more frequently in centers that perform larger numbers of kidney transplantations. For example, 87% of simultaneous font-family:Arial;color:red'>kidney-pancreas transplantations were performed at high-volume centers (>50 kidney transplantations in the previous year) as opposed to only 69% of cadaver kidney transplantations.

As listed in Table 2, a total of 4,728 deaths occurred in the study population.

Kidney transplant recipients had a greater frequency of cardiac deaths, whereas SKPT recipients had a greater frequency of deaths from infection (Table 2). This difference in causes of death was even more pronounced when we examined causes of death during the first year; deaths among SKPT recipients were twice as likely to be attributed to infection (19%) relative to kidney-alone recipients (10%), whereas cardiac causes of death were identified in a lower percentage of SKPT recipients (19% versus 33%). However, the high frequency of deaths from unknown causes makes definitive interpretation difficult.

SKPT and living donor kidney recipients had a significant crude survival distribution advantage over cadaver kidney recipients (Fig 1). Fig. 1. Survival distribution estimates posttransplantation for patients with type 1 diabetes with renal failure. Abbreviations: LD, living donor; CAD, cadaveric.

Five- and 8-year patient survival rates were 81% and 72% for SKPT recipients, 84% and 72% for living donor kidney recipients, and 71% and 55% for cadaver kidney recipients, respectively. The living donor transplant group includes 15% HLA-identical transplants. Eight-year survival rates were 71% and 79% for HLA-nonidentical and HLA-identical transplants, respectively. SEs of 5- and 8-year survival estimates were 1% or less. Eight-year survival rates decreased from 73% to 62% to 44% as age groupings increased from younger than 35 years (n = 5,824) to 35 to 49 years (n = 8,952) to older than 49 years (n = 3,773). Simultaneous kidney-pancreas transplantation was performed infrequently in the oldest age group (n = 177), and survival of patients with SKPT and cadaver-only transplants appear similar.

Among SKPT recipients, 5-year survival has improved over time from 78% in the cohort that underwent transplantation through 1992 to 83% among those who underwent transplantation in 1993 and later. When adjusted for pretransplantation variables known to influence graft and patient survival by using the proportional hazards model, the survival benefit for SKPT versus cadaver kidney recipients was diminished (P = 0.03). Because the crude kidney graft survival rate was lower for cadaver than SKPT transplants (Fig 2), we incorporated kidney graft function as a time-varying covariate.

In a model with additional adjustments for recipient demographics, year, center volume, and donor age, simultaneous kidney-pancreas transplantation was observed to have a significant survival advantage (relative risk [RR], 0.92; P = 0.04) relative to cadaver kidney transplantation.

As in Fig 1, Table 3 shows simultaneous kidney-pancreas transplantation to have a high mortality risk relative to living donor transplantation in the early posttransplantation period. Adjusted for demographics and kidney graft function, the hazard ratio of simultaneous kidney-pancreas transplantation relative to the living-donor group is estimated to be 2.2 (P < 0.001) through 18 months posttransplantation, but reverses to a decreased RR (RR, 0.86; P = 0.02) thereafter. At the 10-year point, the SKPT group has a small survival advantage (67.0% versus 64.1%) compared with the living-donor group. However, this long-term survival advantage is not statistically significant because 95% confidence intervals are each approximately 6% points wide.

Five-year survival rates were 89.8%, 87.8%, and 79.7% for SKPT, living donor, and cadaver donor transplant recipients, respectively. At 8 years, respective rates were 79.3%, 76.2%, and 61.6%. Relative hazards (adjusted for baseline risk factors) for cadaver kidney transplant and SKPT recipients compared with living donor kidney transplant recipients were 1.4 (P < 0.001) and 0.98 (P = 0.78), respectively.

Maintenance of a functioning pancreas graft appears to be associated with a survival benefit (Fig 4).

Of 403 patients with a functioning kidney transplant, but failed pancreas transplant, at 1 year, 18.6% died versus 9.8% of 3,445 patients with both organs functioning at 1 year. Mortality for patients with failed kidney allografts (functioning pancreas grafts; N = 105) or both organs failed (N = 122) is substantial, with death rates greater than 33% for each group. This highlights the substantial hazard increase associated with kidney graft failure. Five- and 8-year patient survival rates for SKPT recipients with a functioning kidney graft, but failed pancreas graft, at the end of the first year were 81.8% and 61.7%, significantly inferior to the 90.5% and 81.2% rates for those with functioning kidney and pancreas grafts, respectively (P < 0.0001; Fig 4). Relative hazards, adjusted for baseline risk factors, are 4.6 for failed kidney, 1.9 for failed pancreas, and 5.0 for both organs failed (each P < 0.001 relative to both organs functioning).

Discussion

We found that long-term survival of SKPT and living donor kidney transplant recipients was similar and superior to that of cadaver donor kidney transplant recipients. A functioning pancreas graft appears to provide a survival advantage because: (1) survival of SKPT recipients is better than that of cadaver kidney transplant recipients, even after adjusting for recipient and donor characteristics; and (2) survival of SKPT recipients who have functioning kidney and pancreas grafts is superior to that of recipients with functioning kidney graft, but failed pancreas graft, at 1 year. Although SKPT recipients have lower mortality than living donor kidney transplant recipients after the first year (3.4 versus 3.8 deaths/100 years' follow-up), survival was similar at 8 years because of greater mortality for SKPT recipients in the first year (hazards ratio for the SKPT relative to the living-donor group, 2.2).

A few other studies also reported that SKPT recipients have a survival advantage over cadaver kidney transplant recipients. Smets et al11 compared the mortality of patients with type 1 diabetes with renal failure in two geographic areas that differed in the frequency of simultaneous kidney-pancreas transplantation: the Leiden area (73% simultaneous kidney-pancreas transplantation) and non-Leiden area (37% simultaneous kidney-pancreas transplantation). Survival was adjusted for age, sex, and duration of dialysis therapy pretransplantation. The hazard ratio for mortality after the initiation of renal replacement therapy was 0.4 (95% confidence interval, 0.20 to 0.77; P = 0.008) in the Leiden area compared with the non-Leiden area.

Tyden et al12 reported an 80% patient survival rate at 10 years in 14 SKPT recipients compared with a 20% patient survival rate in 15 cadaver kidney recipients (10 patients who lost the pancreas after simultaneous kidney-pancreas transplantation and 5 patients who opted for a kidney alone, although eligible for simultaneous kidney-pancreas transplantation).

In the largest single-center report from the University of Wisconsin, Becker et al14 showed that SKPT recipients (n = 335) had an increased observed/expected life span compared with cadaver kidney (n = 147) and living donor kidney (n = 160) recipients. The annual mortality rate was 1.5% for SKPT recipients compared with 6.3% for cadaver kidney recipients and 3.7% for living donor kidney recipients.

Patients with diabetes with renal failure have a high mortality on dialysis therapy. Wolfe et al1 recently reported that death rates per 100 patient-years were 19.9 for all patients with diabetes on dialysis therapy and 10.8 for patients with diabetes on the waiting list for kidney transplantation. The risk for death decreased significantly after kidney transplantation (5.6 deaths/100 patient-years). However, patients with diabetes have a higher mortality after kidney transplantation compared with patients without diabetes. Ojo et al15 reported that death with a functioning graft was more common after kidney transplantation for diabetes compared with transplantation for other causes of renal failure (RR, 1.93; 95% confidence interval, 1.82 to 2.05). In the present study, the risk for death after simultaneous kidney-pancreas transplantation was 4.5/100 patient-years (8.2 in the first year and 3.4 thereafter).

How does a functioning pancreas graft prolong the survival of patients with end-stage renal disease caused by diabetes? One possibility is by decreasing cardiac causes of death. It has been shown that all groups of patients with end-stage renal disease have severely compromised cardiac autonomic function, and patients with diabetes have the greatest degree of dysregulation.16 Autonomic neuropathy, with abnormal cardiorespiratory reflexes, is associated with high mortality, and several investigators reported that successful pancreas transplantation may not only prevent the progressive decline, but also improve the function of cardiorespiratory reflexes.17

Jukema et al18 recently reported that progression of coronary atherosclerosis, analyzed using quantitative coronary angiography, is reduced in patients with a functioning pancreas graft compared with patients with pancreas graft failure. Other studies have noted ongoing improvements in echocardiographic parameters of cardiac function in SKPT recipients compared with kidney-alone transplant recipients with diabetes.7,19 La Rocca et al19 and Tyden et al12 reported decreased death rates from cardiovascular causes in SKPT recipients compared with kidney-alone transplant recipients. In a recent study, Larsen et al20 reported that carotid intima media thickness, shown to correlate with cardiovascular events, is reduced after successful pancreas transplantation. A favorable effect on lipid profiles also was shown in SKPT recipients.21,22 In the present study, the proportion of deaths from cardiac causes is less in SKPT recipients compared with kidney-only recipients. However, the high rate of deaths in the unknown category in both groups makes it difficult to derive a definitive conclusion from the present study.

Although shortage of donor organs is the major factor limiting more widespread use of other solid-organ transplants, pancreas transplantation remains underused for other reasons: noncoverage by third-party payers, lack of routine pancreas procurement, increased morbidity with the surgical procedure compared with kidney transplantation alone, and the procedure being viewed as enhancing quality of life, rather than improving life expectancy. With Medicare coverage effective July 1, 1999, for simultaneous kidney-pancreas transplantation and sequential pancreas after kidney transplantations, improving pancreas graft survival rates approaching or exceeding those of other solid-organ transplants,23 a continued decrease in surgical complications,24 and results of long-term studies showing a survival advantage with simultaneous kidney-pancreas transplantation, we believe pancreas transplantation should be considered a life-enhancing and life-prolonging procedure in patients with type 1 diabetes with end-stage renal disease.

In summary, long-term survival of SKPT recipients is superior to that of cadaveric kidney transplant recipients with type 1 diabetes because of donor organ quality, recipient selection, and a functioning pancreas graft. This observation must be tempered because the conclusion is made after analysis and adjustment for known risk factors, but other important selection factors may remain unaccounted for. There is no difference in survival of SKPT recipients and living donor kidney transplant recipients at up to 8 years' follow-up, with the former having a greater early mortality risk and the latter having a greater late mortality risk. Results of this study suggest that durable benefits of successful simultaneous kidney-pancreas transplantation may become even more pronounced with longer follow-up.

References

1. Wolfe RA, Ashby V, Milford EL, et al: Comparison of mortality in all patients on dialysis, patients on dialysis awaiting transplantation, and recipients of a first cadaveric transplant. N Engl J Med 341:1725-1730, 1999
2. Adang EMM, Engel GL, Van Hooff JP, Koostra G: Comparison before and after transplantation of pancreas-kidney and pancreas-kidney with loss of pancreas—A prospective controlled quality of life study. Transplantation 62:754-758, 1996
3. Najarian AJ, McHugh L, Payne W, et al: Long-term quality of life after kidney and simultaneous pancreas-kidney transplantation. Clin Transplant 12:233-242, 1998
4. Hathaway DK, Abell T, Cardoso S, Hartwig MS, El Gebely S, Gaber AO: Improvement in autonomic and gastric function following pancreas-kidney versus kidney alone transplantation and the correlation with quality of life. Transplantation 57:816-822, 1994
5. Allen RDM, Al-Harbi IS, Morris JGL, et al: Diabetic neuropathy after pancreas transplantation: Determinants of recovery. Transplantation 63:830-838, 1997
6. Navarro X, Sutherland DER, Kennedy WR: Long-term effects of pancreatic transplantation on diabetic neuropathy. Ann Neurol 42:727-736, 1997
7. Wicks MN, Hathaway DK, Shokouh-Amiri MH, et al: Sustained improvement in cardiac function 24 months following pancreas-kidney transplant. Transplant Proc 30:333-334, 1998
8. Fioretto P, Steffes MW, Sutherland DER, Goetz FC, Mauer M: Reversal of lesions of diabetic nephropathy after pancreas transplantation. N Engl J Med 339:69-75, 1998
9. Cheung AT, Perez RV, Chen PC: Improvements in diabetic microangiopathy after successful simultaneous pancreas-kidney transplantation: A computer-assisted intravital microscopy study on the conjunctival microcirculation. Transplantation 68:927-932, 1999
10. US Renal Data System: USRDS 1992 Annual Report: Simultaneous kidney-pancreas transplantation versus kidney transplantation alone: Patient survival, kidney graft survival, and post-transplant hospitalization. Am J Kidney Dis 20:S61-S67, 1992 (suppl 2)
11. Smets YFC, Westendrop RGJ, Van der Pijl JW, et al: Effect of simultaneous pancreas-kidney transplantation on mortality of patients with type-1 diabetes mellitus and end-stage renal failure. Lancet 353:1915-1919, 1999
12. Tyden G, Bolinder J, Solders G, Brattstrom C, Tibell A, Groth CG: Improved survival in patients with insulin-dependent diabetes mellitus and end-stage diabetic nephropathy 10 years after combined pancreas and kidney transplantation. Transplantation 67:645-648, 1999
13. Rayhill SC, D'Alessandro AM, Odorico JS, et al: Simultaneous pancreas-kidney transplantation and living related donor renal transplantation in patients with diabetes: Is there a difference? Ann Surg 231:417-423, 2000
14. Becker BN, Brazy PC, Becker YT, et al: Simultaneous pancreas-kidney transplantation reduces excess mortality in type I diabetic patients with end-stage renal disease. Kidney Int 57:2129-2135, 2000
15. Ojo AO, Hanson JA, Wolfe RA, Leichtman AB, Agodoa LY, Port FK: Long-term survival in renal transplant recipients with graft function. Kidney Int 57:307-313, 2000
16. Hathaway DK, Cashion AK, Milstead EJ, et al: Autonomic dysregulation in patients awaiting kidney transplantation. Am J Kidney Dis 32:221-229, 1998
17. Navarro X, Kennedy WR, Aeppli D, Sutherland DER: Neuropathy and mortality in diabetes: Influence of pancreas transplantation. Muscle Nerve 19:1009-1016, 1996
18. Jukema JW, Smets YFC, Van Der Pijl JW, et al: Impact of simultaneous pancreas and kidney transplantation on progression of coronary atherosclerosis in patients with end-stage renal failure due to type 1 diabetes. Diabetes Care 25:906-911, 2002
19. La Rocca E, Fiorina P, Di Carlo V, et al: Cardiovascular outcomes after kidney-pancreas and kidney-alone transplantation. Kidney Int 60:1964-1971, 2001
20. Larsen JL, Ratanasuwan T, Burkman T, et al: Carotid intima media thickness decreases after pancreas transplantation. Transplantation 73:936-940, 2002
21. Larsen J, Larsen C, Hirst K, et al: Lipid status after combined pancreas-kidney transplantation and kidney transplantation alone in type 1 diabetes mellitus. Transplantation 54:992-996, 1992
22. Foger B, Konigsrainer A, Palos G, et al: Effect of pancreas transplantation on lipoprotein lipase. Postprandial lipemia, and HDL cholesterol. Transplantation 58:899-904, 1994
23. Sollinger HW, Odorico JS, Knechtle SJ, D'Alessandro AM, Kalayoglu M, Pirsch JD: Experience with 500 simultaneous pancreas-kidney transplants. Ann Surg 228:284-296, 1998
24. Humar A, Kandaswamy R, Granger D, Gruessner RW, Gruessner AC, Sutherland DER: Decreased surgical risks of pancreas transplants in the modern era. Ann Surg 231:269-275, 2000