The acceptance of older patients into renal replacement therapy programs in industrial countries has increased considerably over the last decade. In Australia, the annual intake of new end-stage renal disease patients over 60 has risen from 30% of all new patients in 1986 to 49% in 1996 (1, 2). Similarly, more than half of all uremic patients in the United States, Canada, Japan, France, and Germany are over 60 at the commencement of renal replacement therapy (3, 4). On the basis of their limited life expectancy in the setting of chronic organ donor shortages, the majority of these patients are treated with dialysis and do not receive renal allografts (5).
Unfortunately, the optimal therapy for end-stage renal disease in patients over 60 remains controversial. A limited number of previous studies comparing older dialysis and renal transplant patient populations have suggested a survival advantage of the latter (2, 4, 6–8). However, these investigations are potentially biased by case selection, because only those patients with minimal or no associated comorbid illnesses are referred for renal transplantation. In the few studies in which comparisons have been made between older patients receiving renal transplants and those receiving dialysis after their inclusion on a transplant waiting list, results have been conflicting, with some reports showing better (9), equivalent (10) or worse (11) survival for renal transplantation. Some of the disparity in these results may be attributable to variable screening for comorbid illness, especially coronary artery disease, before inclusion on the transplant waiting list, and to an excess of infectious deaths in older transplant recipients related to over-immunosuppression (12).
The aim of the present study, therefore, was to compare survival outcomes for renal transplantation and dialysis in a large group of older patients who had been included on a transplant waiting list at a single center with standardized screening protocols and acceptance criteria.
MATERIALS AND METHODS
All patients over 60 accepted on to the Queensland cadaveric renal transplant waiting list between January 1, 1993 and December 31, 1997 were included in the study. Evaluation of the patients before acceptance was standardized and included a medical history, physical examination, routine biochemical and hematological studies, HLA typing, blood grouping, viral studies (hepatitis B, hepatitis C, human immunodeficiency virus, and varicella zoster), cystoscopy and retrograde pyelograms (for patients with analgesic nephropathy) (13), electrocardiogram, chest roentgenogram, echocardiograph, and either a dipyridamole thallium scan or a dobutamine stress echocardiograph. Coronary angiography was additionally performed in any patients with the following features: (a) a history of ischemic heart disease, type II diabetes mellitus or compound analgesic abuse; (b) age over 65; or (c) a significant abnormality detected by either a dobutamine echocardiographic or dipyridamole thallium scintigraphic method. To be accepted on to the transplant waiting list, patients had to have an anticipated 5-year survival in excess of 80%. Patients were not considered suitable for transplantation if they were over 75, had a body mass index >30, currently smoked or had recent (within 5 years) malignancy (other than skin basal cell or squamous cell carcinomas), or had active infection, extra-renal organ impairment, psychiatric illness, or significant coronary, cerebral, or peripheral vascular disease.
Study entry was marked by the date of admission onto the waiting list. The sample was followed until either October 1,1998 or death, at which point data were censored. Patients were assigned to the transplant group once they received a cadaveric renal allograft at any time point during this period, whereas all other patients were assigned to the dialysis group. Participants changing to the transplantation group contributed follow-up to the group from the date of receiving the allograft. Allocation of cadaveric renal allografts was centrally directed and occurred strictly on the basis of a computerized algorithm, which ranked potential recipients primarily according to HLA matching and, to a lesser extent, waiting time.
All transplants were performed at a single center (Princess Alexandra Hospital, Brisbane). Patients who received living donor renal transplants were excluded from the analysis. Standard initial immunosuppressive therapy consisted of cyclosporin (2.5 mg/kg body weight twice daily with diltiazem slow release 180–240 mg daily), prednisolone (0.3 mg/kg omni mane), and either azathioprine (2 mg/kg daily; before 1996) or mycophenolate mofetil (1 g twice daily; 1996 onwards). Cyclosporin dosages were titrated to achieve trough blood concentrations of 180–200 ng/ml (high performance liquid chromatography) in the early posttransplant period and slowly reduced to achieve levels of 75–100 ng/ml after 12 months. Prednisolone dosage was decreased after 1 month at a rate of 1 mg mane every week until a dose of 10 mg mane was attained. Thereafter, dosages were more slowly reduced by 1 mg mane every 2–4 weeks, depending on the presence or absence of previous rejection episodes and steroid side effects (e.g., diabetes mellitus, infection, and osteopenia). Patients receiving mycophenolate mofetil were changed to azathioprine at 6 months, and dosages were gradually reduced to maintenance levels of 1–1.5 mg/kg body weight after 12 months. The majority of patients were changed from triple therapy to dual therapy (cyclosporin and either azathioprine or prednisolone) or cyclosporin monotherapy more than 1 year after the transplant.
Results are expressed as mean±SEM. The chi-square test was used to compare differences in proportions, whereas comparisons of differences in means between groups were made by Student’s t test. Survival curves, survival probability, and estimated mean survival times were generated according to the Kaplan-Meier method. Differences in the survival curves were evaluated using the log rank test. A Cox’s Proportional Hazard model, with time-varying explanatory variables, was applied for statistical analysis, because exposure to different therapy modalities (dialysis vs. transplant) changed over time (14). The characteristics of the transplant and dialysis groups were compared at the baseline, and any variables on which differences were found, and that could thus be considered potential confounders, were adjusted for in the analysis. Data were analyzed on an intention-to-transplant basis, whereby patients were not excluded if, after initially fulfilling the selection criteria, they were subsequently withdrawn from the transplant waiting list. In view of the fact that an intention-to-transplant design might overestimate the hazard rate of dialysis patients relative to transplant patients, a survival analysis was also performed in which patients no longer considered suitable for transplantation were censored at the time of their removal from the waiting list. Analyses were performed using the software package, SPSS for Windows release 8.0.0 (SPSS Inc., North Sydney, Australia). P values < 0.05 were considered significant.
During the 5-year period between January 1, 1993 and December 31, 1997, 572 patients over 60 were started with renal replacement therapy in Queensland, corresponding to 34 new patients per million inhabitants per year. Of these, 174 (30.4%) met the predefined transplant acceptance criteria and were subsequently placed on the cadaveric renal transplant waiting list and included for analysis in the present study. Mean follow-up time was 2.8±0.1 years (range 0–5.8 years) for the total study population. When censoring was performed in the case of withdrawal from the waiting list, mean follow-up time decreased slightly to 2.4±0.1 years (range 0 - 5.8 years).
There were 67 receiving renal transplants (transplant group) during the study period, whereas the remaining 107 patients did not (dialysis group). These groups were well matched at the commencement of the study in terms of demographic variables, comorbid illnesses, initial dialysis therapies, time undergoing dialysis, and causes of renal failure (Tables 1 and 2). The only significant differences between the groups were a higher proportion of non-Caucasian and diabetic patients in the dialysis group. Australian aborigines and Torres Strait Islanders accounted for both of the non-Caucasian transplant patients and 9 (56%) of the non-Caucasian dialysis patients. The remaining 7 (44%) non-Caucasian dialysis patients were Asian. Race and diabetes status were, therefore, considered potential confounders and were included as covariates in the Cox’s hazard models.
During the follow-up period, 42 (39.3%) dialysis patients and 5 (7.5%) transplant patients died (P <0.001). The overall mortality rate was 0.096 per patient-year (0.131 for dialysis and 0.029 for transplant). Causes of death in the dialysis group included cardiovascular disease (59.5%), infection (19%), cerebrovascular disease (7.1%), malignancy (4.8%), dialysis withdrawal (4.8%), calciphylaxis (2.4%), and dementia (2.4%). In the transplant group, the causes of death included cardiovascular disease (40%), malignancy (40%), and pulmonary embolism (20%). All transplant patients died with functioning grafts and no renal allografts were lost other than because of patient death. Respective 1- , 3- and 5-year survivals were 92%, 62%, and 27% for the dialysis group and 98%, 95%, and 90% (P <0.01) for the transplant group (Fig. 1). Using Cox’s Proportional Hazards model with a time-dependent covariate to control for the time until transplantation, an adjusted hazard ratio of 0.16 (95% confidence interval 0.06 - 0.42) was calculated, in favor of transplant patients.
There were 51 patients in the dialysis group withdrawn at some time point after initial inclusion on the transplant waiting list. The indications for withdrawal included cardiovascular disease (45%), general poor health (15.7%), malignancy (7.8%), peripheral vascular disease (5.9%), cerebrovascular disease (3.9%), dementia (3.9%), patient preference for dialysis (3.9%), obesity (2.0%), depression (2.0%), amyloidosis (2.0%), noncompliance (2.0%), light chain deposition disease (2.0%), cirrhosis (2.0%), and progressive systemic sclerosis (2.0%). Withdrawals were intended to be permanent in 41 (78.8%) cases. Before withdrawal, only 22 (20.6%) dialysis patients had died (mortality rate=0.097 per patient-year). If patients were censored at the time of their withdrawal from the transplant waiting list, the adjusted hazard ratio was 0.24 (95% confidence interval 0.09 - 0.69), again favoring transplant patients (Fig. 2).
The results of the present study provide strong evidence that renal transplantation confers a substantial survival advantage over dialysis in older patients with end-stage renal disease who are considered “fit” enough to undergo transplantation. Specifically, older renal transplant recipients were 8 times less likely to die than waiting-listed, older dialysis patients. Even when dialysis patients who were no longer considered suitable for transplantation were censored after their removal from the waiting list, the hazard rate for transplantation was still 4 times lower than for dialysis. To our knowledge, this is the first analysis to demonstrate a propitious effect of renal transplantation on the survival of older patients, by applying a Cox regression model in a single-center study.
These results contrast with those of several earlier studies that failed to find a survival benefit of renal transplantation compared with dialysis (15–17). Sommer et al. (15) and Lundgren et al. (16) reported lower survival probabilities for older cadaveric renal transplant recipients compared with dialysis patients, whereas Hutchinson and associates (17) found no differences between the two treatments after adjustment for comorbid factors. Kyllonen and Ahonen (11) and Bonal and co-workers (10) both observed initially lower survival rates in older transplant populations compared with waiting-list dialysis populations at 12 months, followed by marginally better survival after 5 years. Most of these studies were notable for their high mortality rates in the early posttransplant period as a result of postoperative infectious and cardiovascular complications. It is possible, therefore, that some of the disparities between the results of these earlier studies and those of the present investigation reflect improvements in peri-operative management and immunosuppressive therapy that have outpaced concomitant advances in the delivery of dialysis (18, 19). Improved selection of renal transplant candidates may also have contributed (20, 21), because all of the previous studies had more liberal selection criteria for admission to their transplant waiting lists. Finally, “center effects” have been shown to be an important determinant of patient survival in both the transplant and dialysis settings (7, 22, 23) and could conceivably have accounted for some of the contrasting results between studies.
In keeping with the results of the current study, Schaubel et al. (7) reported greater 5-year survival rates for transplant recipients over 60 (81%) compared with matched, randomly-selected dialysis patients (51%). However, the findings in this report were potentially confounded by case selection bias because renal transplant candidates exhibit less comorbidity than the average dialysis patient (24). Thus, by comparing renal transplant recipients with the general dialysis population, rather than with transplant candidates, it is likely that the survival benefit of renal transplantation was overestimated. Fauchald and co-workers (9) attempted to address this issue by comparing the survival of older renal transplant recipients and transplant waiting-list dialysis patients. They observed superior survival in the transplant group at 6 months (93% vs. 64%), 12 months (87% vs. 44%), and 4 years (62% vs. 7%). However, the fact that the 4-year survival in the waiting-list dialysis population was actually worse than in the remainder of the dialysis population (7% vs. 13%, respectively) again raised the concern that the demonstrated benefit of transplantation was biased by differential overall health in the two populations. Although it is difficult to entirely exclude the possibility of such bias also existing in the present study, it is noteworthy that the two study populations were very well matched at the baseline for demographic variables, comorbid illness, and time undergoing dialysis. The higher proportion of non-Caucasians in the dialysis population most likely reflected a greater prevalence of uncommon major histocompatibility antigens not shared with those of an almost exclusively Caucasian donor pool. Diabetes was also more frequent in the dialysis population, but adjusting for this comorbid condition using multivariate regression only slightly reduced the impressive survival benefit of renal transplantation.
The survival probabilities of older transplant recipients at 1 year (98%), 3 years (95%), and 5 years (90%) are similar to or better than in previous reports (3, 7, 11, 15, 16, 18, 25–40) and are comparable to those of patients between 30 and 50 years who receive transplants at our center (97%, 94% and 92%, respectively) (12). Graft survival was identical to patient survival in our older patients because the only cause of graft loss was patient death. Numerous other studies have also reported that patient death is a far more significant cause of graft loss than rejection in older patients (12, 18, 35, 36, 41, 42). However, this should not necessarily be viewed as a waste of precious renal allografts because graft survival in older recipients was comparable with that reported in younger transplant patients (12). Other groups have described similar findings (32, 35, 36, 43, 44). Thus, the increase in graft loss because of patient death in the over 60 age group seems to be offset by a reduction in loss because of rejection.
It is also interesting to note in the present study that the annual mortality rate of patients in the waiting-list dialysis group (13%) was only marginally better than that reported for the general older dialysis population in Australia (25%) (2). This was so despite the fact that the former population was deemed to represent the healthiest tertile of older Queensland dialysis patients on the basis of a standardized and exhaustive screening protocol involving clinical, laboratory, and radiological assessments (including coronary angiography in many cases). We have previously demonstrated that the institution of this screening protocol has resulted in a profound improvement in the survival of patients over 60 after transplantation (12), suggesting that these “fitter” patients do only substantially better, provided they subsequently receive a renal transplant. Schaubel and associates (7) similarly observed that the effect of transplantation on survival was greatest among the healthiest group of older end-stage renal disease patients with no comorbid illnesses.
As found in other studies (9, 25, 45), the major cause of death in both the dialysis and transplant populations was cardiovascular disease (59.5% and 40% of deaths, respectively). It is interesting that infection accounted for 19% of deaths in the dialysis group, but did not contribute to the demise of any transplant patient. Because aging is itself associated with a reduced activity of cell-mediated immunity (3), transplantation in older patients is generally characterized by lower rejection rates and higher rates of viral, bacterial, and fungal infections (3, 9, 18, 25, 29, 46). Consequently, less immunosuppressive therapy is administered to older transplant recipients that young recipients in our unit, and this may have obviated the occurrence of fatal infectious events in this study.
The observation in the present study that older dialysis patients have improved longevity after kidney transplantation complements previous reports of significantly enhanced quality of life in such patients. In a prospective study of patients undergoing renal transplantation at three Canadian hospitals, Laupacis et al. (47) demonstrated that transplantation in individuals over 60 was significantly more cost-effective than dialysis and was associated with marked improvement in health-related quality of life, as measured by the Sickness Impact Profile and the Time Trade-Off Technique. Bonal and associates (10) also showed that patients older than 55 who received a renal transplant had superior functional autonomy compared with their waiting-list dialysis counterparts. Moreover, Hestin et al. (35) observed that functional rehabilitation and quality of life were as good in renal allograft recipients over 60 as in younger recipients. Thus, denying transplantation on the basis of age alone seems to be difficult to justify.
In conclusion, the present study suggests that renal transplantation seems to confer a substantial survival advantage over dialysis in older patients with end-stage renal failure who are rigorously screened and considered suitable for renal transplantation. Patient and graft survivals in such individuals are comparable to those in younger patients, indicating that older patients should not be rejected for transplantation purely on the basis of their age. Indeed, in patients over 60 who have minimal comorbidity, transplantation should be recommended as the renal replacement therapy of first choice.
The authors gratefully acknowledge the contributions of their nephrologist colleagues, Dr. Michael Falk, Dr. Ed Meagher, Dr. Alan Parnham, Dr. Amir Alamir, Dr. Peter Craswell, Dr. Phil Boyle, Dr. Peter de Jersey, Dr. Tim Furlong, Dr. Emlyn Jones, Dr. Simon Fleming, Assistant Professor Zoltan Endre, Dr. David Saltissi, and Dr. Helen Healy, who co-operated with the screening protocol and provided ongoing patient care after the early posttransplant phase. The authors also thank the nursing staff of our unit (particularly Sister Joan Allen) for their expert patient care.
1. Disney APS. Demography and survival of patients receiving treatment for chronic renal failure in Australia and New Zealand: report on dialysis and renal transplantation from the Australia and New Zealand Dialysis and Transplant Registry. Am J Kidney Dis 1995; 25:165.
2. Disney APS, Russ GR, Walker R, Sheil AGR. ANZDATA Registry Report 1997. 1997.
3. Ismail N. Renal replacement therapy in the elderly: an old problem with young solutions. Nephrol Dial Transplant 1997; 12:873.
4. US Renal Data System. USRDS 1995 Annual Data Report. 1995.
5. Helderman JH. Transplantation and the elderly. Nephrol Dial Transplant 1995; 10:773.
6. Taube DH, Winder EA, Ogg CS, et al. Successful treatment of middle aged and elderly patients with end stage renal disease. Br Med J Clin Res Ed 1983; 286:2018.
7. Schaubel D, Desmeules M, Mao Y, Jeffery J, Fenton S. Survival experience among elderly end-stage renal disease patients: a controlled comparison of transplantation and dialysis. Transplantation 1995; 60:1389.
8. Okiye SE, Engen DE, Sterioff S. Primary renal transplantation in patients 50 years of age and older. Transplant Proc 1983; 1046 [Abstract].
9. Fauchald P, Albrechtsen D, Leivestad T, Berg KJ, Talseth T, Flatmark A. Renal replacement therapy in elderly patients. Transpl Int 1988; 1:131.
10. Bonal J, Cleries M, Vela E. Transplantation versus haemodialysis in elderly patients. Renal Registry Committee. Nephrol Dial Transplant 1997; 12:261.
11. Kyllonen L, Ahonen J. Kidney transplantation in the elderly in Finland. Transplant Proc 1990; 22:163.
12. Hawley CM, Parnham AP, Petrie JJB, et al. Improved patient and graft survival in elderly renal allograft recipients following the introduction of cardiovascular screening [Abstract]. J Am Soc Nephrol 1997; 8:684A.
13. Swindle P, Falk M, Rigby R, Petrie J, Hawley C, Nicol D. Transitional cell carcinoma in renal transplant recipients: the influence of compound analgesics. Br J Urol 1998; 81:229.
14. COX DR. Regression models and life-tables. J Stat Soc 1972; 34:187.
15. Sommer BG, Ferguson RM, Davin TD, et al. Renal transplantation in patients over 50 years of age. Transplant Proc 1981; 13:33.
16. Lundgren G, Fehrman I, Gunnarsson R, et al. Cadaveric renal transplantation in patients over 55 years of age with special emphasis on immunosuppressive therapy. Transplant Proc 1982; 14:601.
17. Hutchinson TA, Thomas DC, Lemieux JC, Harvey CE. Prognostically controlled comparison of dialysis and renal transplantation. Kidney Int 1984; 26:44.
18. Ismail N, Hakim RM, Helderman JH. Renal replacement therapies in the elderly: Part II. Renal transplantation. Am J Kidney Dis 1994; 23:1.
19. Shaw BW. Transplantation in the elderly patient. Surg Clin North Am 1994; 74:389.
20. Kasiske BL, Ramos EL, Gaston RS, et al. The evaluation of renal transplant candidates: clinical practice guidelines. Patient Care and Education Committee of the American Society of Transplant Physicians. J Am Soc Nephrol 1995; 6:1.
21. Jordan ML, Novick AC, Steinmuller D, et al. Renal transplantation in the older recipient. J Urol 1985; 134:243.
22. Khan IH, Campbell MK, Cantarovich D, et al. Survival on renal replacement therapy in Europe: is there a “centre effect”? Nephrol Dial Transplant 1996; 11:300.
23. Taylor RM, Ting A, Briggs JD. Renal transplantation in the United Kingdom and Ireland: the centre effect. Lancet 1985; 1:798.
24. Gaylin DS, Held PJ, Port FK, et al. The impact of comorbid and sociodemographic factors on access to renal transplantation. JAMA 1993; 269:603.
25. Ost L, Groth CG, Lindholm B, Lundgren G, Magnusson G, Tillegard E. Cadaveric renal transplantation in patients of 60 years and above. Transplantation 1980; 30:339.
26. Barry JM, Lemmers MJ, Meyer MM, DeMattos A, Bennett WM, Norman DJ. Cadaver kidney transplantation in patients more than 65 years old. World J Urol 1996; 14:243.
27. Berthoux FC, Jones EH, Mehls O, Valderrabano F. Transplantation Report. 1: renal transplantation in recipients aged 60 years or older at time of grafting. The EDTA-ERA Registry. European Dialysis and transplant Association-European Renal Association. Nephrol Dial Transplant 1996; 11 Suppl 1: 37.
28. Pretagostini R, Berloco P, Poli L, et al. Kidney transplantation in elderly patients. Transplant Proc 1996; 28:192.
29. Babinet F, Hourmant M, Cantarovich D, Karam G, Bouchot O, Soulillou JP. Kidney transplantation after 60 years of age: survival of patients and functional results. Presse Med 1990; 19:201.
30. Fehrman I, Brattstrom C, Duraj F, Groth CG. Kidney transplantation in patients between 65 and 75 years of age. Transplant Proc 1989; 21:2018.
31. Pirsch JD, Stratta RJ, Armbrust MJ, et al. Cadaveric renal transplantation with cyclosporine in patients more than 60 years of age. Transplantation 1989; 47:259.
32. Lauffer G, Murie JA, Gray D, Ting A, Morris PJ. Renal transplantation in patients over 55 years old. Br J Surg 1988; 75:984.
33. Konigsrainer A, Wohlfahrter T, Spielberger M, et al. Experiences with kidney transplantation in elderly patients. Wien Klin Wochenschr 1988; 100:318.
34. Westlie L, Umen A, Nestrud S, Kjellstrand CM. Mortality, morbidity, and life satisfaction in the very old dialysis patient. Trans Am Soc Artif Intern Organs 1984; 30:21.
35. Hestin D, Frimat L, Hubert J, Renoult E, Huu TC, Kessler M. Renal transplantation in patients over sixty years of age. Clin Nephrol 1994; 42:232.
36. Cantarovich D, Baatard R, Baranger T, et al. Cadaveric renal transplantation after 60 years of age. A single center experience. Transpl Int 1994; 7:33.
37. Nyberg G, Nilsson B, Hallste G, Haljamae U, Norden G, Blohme I. Renal transplantation in elderly patients: survival and complications. Transplant Proc 1993; 25:1062.
38. Vivas CA, Hickey DP, Jordan ML, et al. Renal transplantation in patients 65 years old or older. J Urol 1992; 147:990.
39. Andreu J, de la Torre M, Oppenheimer F, et al. Renal transplantation in elderly recipients. Transplant Proc 1992; 24:120.
40. Velez RL, Brinker KR, Vergne Marini PJ, et al. Renal transplantation with cyclosporine in the elderly population. Transplant Proc 1991; 23: 1749.
41. Mallick NP, Dyer P, Johnson RW. Transplantation in the elderly. Nephrol Dial Transplant 1996; 11 (suppl) 9: 95.
42. Takemoto S, Terasaki P. Donor and recipient age. In: Terasaki PI, ed. Clinical transplants, Los Angeles: UCLA Tissue Typing Laboratory, 1995: 345.
43. Nyberg G, Nilsson B, Norden G, Karlberg I. Outcome of renal transplantation in patients over the age of 60: a case-control study. Nephrol Dial Transplant 1995; 10:91.
44. Morris GE, Jamieson NV, Small J, Evans DB, Calne R. Cadaveric renal transplantation in elderly recipients: is it worthwhile? Nephrol Dial Transplant 1991; 6:887.
45. Kjellstrand CM, Koppy K, Umen A, Nestrud S, Westlie L. Haemodialysis of the elderly. In: Oreopoulos DG, ed. Geriatric nephrology. Dordrecht, Nijhoff, 1986: 135.
46. Murie JA, Lauffer G, Gray D, Ting A, Morris PJ. Renal transplantation in the older patient. Transplant Proc 1989; 21:2024.
47. Laupacis A, Keown P, Pus N, et al. A study of the quality of life and cost-utility of renal transplantation. Kidney Int 1996; 50:235.