Scand J Urol Nephrol26: 397-401, 1992

LONG TERM MORBIDITY A N D MORTALITY AFTER KIDNEY TRANSPLANTATION Thomas GorlCn', Michael Abdelnoor', Erik Engerl, Stein Halvorsen', Torbjerrn Leivestad', Ole Jacob Malml and Hans Petter Aarseth' From the 'Dtpartments of Nephrology and Surgery and the Institute for Medical Statistics, Ullevd Hospital and the 'Institute of Transplantation Immunology, Rikshospitalet, Oslo, Norway

Scand J Urol Nephrol Downloaded from informahealthcare.com by University of Otago on 01/10/15 For personal use only.

(Submitted June 28, 1991. Accepted for publication December 3, 1991)

Abstract. A cohort of 69 patients received a kidney transplant in the period 1963-1977. The mean observation time was 9.5 years. Accumulated follow-up time was 661.4 patients year. The mean (SE) 10-year survival was U(5.9) Yo. Univariate analysis showed that female patients had poorer survival than male. Patients with a cadaveric donor had lower survival than those with a living donor. Also survival with different HLA-A, B match differed significantly. A multivariate analysis pinpointed nature of donor, cadaveric vs. living, as the sole independent predictor of mortality. Patients receiving a cadaveric kidney were on double (2.2) relative risk of mortality as compared to patients with a living donor. The major causes of death were infections during rejection treatment, and cardiovascular disease. Patients had low rates of morbidity. Our results showed satisfactory outcome of kidney transplantation. Key words: kidney transplantation, mortality, morbidity.

The aims of' kidney transplantation are prolongation of life and improvement of the quality of life. The present study was undertaken to assess the morbidity and mortality in the second decade after transplantation. MATERIAL AND METHODS A continuous series of 69 patients transplanted from 1963 until the end of 1977, was studied. At the time of transplantation the mean (SE) age of 20 females was 39(3) range 15-61 years, that of 49 males were 39(2), range 19-65. Mean duration of dialysis before transplantation was 9 months. Primary kidney diseases were chronic glomerulonephritis (36 patients), chronic pyelonephritis (1 7 patients), polycystic kidney disease (4 patients), hereditary nephritis (3 patients), nephrosclerosis (2 patients) and other (7 patients). There were n o diabetics. The 69 patients received 82 grafts (9 patients had 2 grafts and 2 patients had 3 grafts). Mean (SE) donor age was 41(2) years (range 10-65). Thirty-three patients received a graft from a

living donor, 5 of these were later retransplanted with grafts from cadaveric donors. Of the 36 patients with grafts from a cadaveric donor, 6 were retransplanted. Prospective HLA-A, B typing was performed from 1968. With one exception, those transplanted earlier were typed retrospectively. The actual match grades for cadaveric and living donors are given in Table I. Prednisolone 7.5 mg- 10 mg daily and azathioprine 1-1 112 mg/kg b.w. were given for immunosuppression. One patient (with a graft from a HLA identical sister) discontinued her medication the third year after transplantation. The end point of our observation was death or July 1, 1988. Observation time ranged from 2 to 8 177 days, mean observation time being 9.5 years. Accumulated follow-up time was 661.4 patient years. Thirty-eight patients (55%) survived more than 10 years after first transplantation. At closing date 34 patients were alive, 2 of them were on maintenance dialysis. The data for the present analysis were obtained from hospital journals as well as from interviews and examination of those patients who were alive, with 2 exceptions at our center.

STATISTICAL METHODS Patient survival was calculated. As prognostic factors were postulated: Recipient age and sex, primary kidney disease, time on dialysis prior t o transplantation, donor age, living versus cadaveric donor, HLA haplotype and HLA-A, B match grade. Only data considering the first graft were used for calculations. For each of these factors a preliminary analysis was done by a univariate method, using survival curves according to Kaplan-Meier method (6). Univariate curves were tested according to Breslow (I), a test which is more sensitive t o difference in early survival, and with a Mantel-Cox test which is more sensitive to difference in late survival ( 1 1). Continuous variables were tested using the univariate mode of Cox's hazard model (3). A multivariate run of the Cox's hazard model was used to pinpoint independent risk factors of total mortality. A BMDP statistical software was used in the analysis (4). Scand J Urol Nephrol26

398

T. GorlPn et al. Table 11. Ten-year follow-up of kidney transplantation, analysis of risk factors

60

Scand J Urol Nephrol Downloaded from informahealthcare.com by University of Otago on 01/10/15 For personal use only.

30

90 120 Time i n months

Fig. 1. Patient survival after kidney transplantation. *---* living donor ( n= 33);U-U,cadaveric donor (n=36).

RESULTS

Variables

Wald-test SE(P) pvalue

p*

(a) Univariate analysis of continuous variables using the Cox's model Donor age -0.147 0.074 3.93 0.047 Dialysis time 0.025 0.017 2.1 1 0.147 Recip. age 0.014 0.094 2.10 0.140 (b) Multivariate analysis of the Cox's model Donor source 0.791 0.355 4.96 0.02

* Regression coefficient.

Female patients had a slightly poorer survival

Univariate analysis of mortality than male, at 10 years: 40% vs. 61 O/o. This difThe mean (SE) 10-year patient survival of the ference was, however, not significant (Breslow, total cohort was 55(6)%, 15 years survival 50 p=O. 10; Mantel-Cox, p=0.08). (6)%and 20 years 44 (8)Yo. There was a better The 10-year survival was the same for pa10-year survival for patients with living donor (67%) vs. cadaveric donor (42%), Breslow test p=O.Oll, Mantel-Cox p=0.0244 (Fig. 1). With living donor the 10-year survival was not different comparing recipients of HLA identical ( n = 12) with 1 haplotype mismatched donor grafts ( n= 21). Considering recipients of cadaveric grafts the survival was lower in the 2-3 HLA-A, B antigen mismatched group, but the difference did not reach statistical significance (Table I).

Table I. The distribution of HLA-B, B match and respective patient survival in 69 kidney

tients with chronic glomerulonephritis as for patients with other kidney diseases (56 O/o vs. 5 5 Yo). In this small series, recipient age or pretransplant dialysis time had no prognostic effect. Donor age on the other hand inversely correlated with survival (p=0.047) (Table 11). The multivariate analysis showed that the sole independent predictor of total mortality was living versus cadaveric donor. Our transplanted patients with a cadaveric donor were on a relative risk of death 2.22 times that of patients with a living donor. The two other variables, sex of the patient and the number of HLA-A, B mismatches, derived their prognostic

transplantations Survival

Living donor (n=33)

Table 111. Rates of death according to causes in

kidney transplanted patients

Mismatch

n

lOy

HLA-identical

12

1 Haplotype

21 8

67% . NS 7 I o/o

n x 100/patient years

Cadaver donor 0 A B ag

47 Yo

( n= 36) 1ABag 2ABag

11 14

3ABag

3

NS

29 yo

Scand J Urol NephrolZ6

Total Irrev. rejection CHD Septicemia Stroke Cancer Drug side effect Unknown

5.4 1.4 1.2 1.1

0.7 0.2 0.2 0.4

p-

Ten-yearfollow-up of kidney transplantation Table IV. Causes of death in patients who died in the second decade after transplantation ~

3Cum r z 100%

399

~~

Age at death

Year aft.tx

Cause of death

67 72 13 45

2 1.5 18 12.5 12.5 10.5

Rupture aortic aneurysm Metastatic ca. recti Coronary heart disease Rej. ass. septicemia Ac. myocard. infarction

of re-tx

75

50

Scand J Urol Nephrol Downloaded from informahealthcare.com by University of Otago on 01/10/15 For personal use only.

45

0

1

2

3

4

5

6 7 8 9 10 Years after transplantation

Fig, 2. Cumulative survival free of retransplantation calculated for the total cohort of 69 kidney transplanted patients.

strength through a secondary correlation with the nature of donor. The early mortality (first 3 months after transplantation) was 4.3 %. In the observation period major causes of death were late rejection (9 pts), coronary heart disease (8 pts), septicemia (7 pts) and cerebral stroke ( 5 pts). Only one patient died from cancer (Table 111). Five patients have died in the second decade after transplantation (Table IV).

Morbidity After exclusion of the first three months following transplantation, the mean number of hospital admissions were 5 (range 0-24) with a total of 67 (0-526) days, or 7 daydpatient-year. For those with functioning grafts at follow-up, mean time of hospitalization had been 45 days, or 2.8 days per patient year. Adverse effects related to immunosuppression are listed in Table V.

.

Table V. Morbidity related to immunosuppression. Reporled reactions related to side-effects of immunosuppression in the 34 patients alive at follow-up spring 88 Reported reaction Cast rointestinal Polar posterior cataract Herpes zoster Aseptic bone necrosis Skin changeslmalignant skin tumours Other malignancies

Rates ( n x 100/ patient years) 1.36 1.06 1.06 0.76 0.66 0.3 1

There were 11 events of first retransplantation, i.e. with a rate 1.66% patient years. Including 2 events of 2nd retransplantation, the rate of retransplantation was 1.96 Yo patient years. The cumulative survival free of retransplantation after 10 years was 83.6% (Fig. 2). DISCUSSION Our material differs from most others in the fact that all but two of the long-time survivors are alive with functioning grafts. SO far, in the second decade only one patient has returned to dialysis. This is in contrast to other series which reports as many grafts lost due to rejection as to death (8, 10, 16), and to the early report by Nylander et al. (1 5) in which 87 O/o of graft losses in second decade were due to rejection. The 1O-year patient survival in our series was 55%. In a series of kidney transplanted patients in Australia, 44 O/o survived more than 10 years (lo), while Lee et al. reported a 50% patient survival at 10 years (8). In these reports, as in the EDTA report ( I 7) and that from Lowrie et al. (9), recipients of grafts from living related donors had a significantly higher survival than those receiving a cadaveric graft. The same was found in our series, further a multivariate analysis showed that the donor's status was the sole independent prognostic factor of survival. When patients with grafts from living or cadaveric donors were studied separately, the degree of HLA-A,B match did not influence survival (Table I). Our groups are however small, and we Scand J Urol Nephrol26

Scand J Urol Nephrol Downloaded from informahealthcare.com by University of Otago on 01/10/15 For personal use only.

400

T. GorlPn et al.

cannot exclude the possibility that the observed difference in survival between recipients of living and cadaveric grafts results from better HLA matching among relatives. Thus the degree of HLA-DR match could not be examined. In our series, the mortality rate seemed to stabilize after 5 years. Several investigators have studied the fate of those surviving more than 10 years. Frisk et al. (5) have calculated a 3% annual mortality rate. In our series 5 patients died in the second decade. The total observation time is 206 patient years, thus the mortality rate is 2.5% per patient year. In our series there was no difference in late mortality comparing living related and cadaveric grafts. However, in their report on only cadaveric grafts, Mahoney et al. (10) found a mortality rate of 4.79/0 per patient year in the second decade. In their report on diabetic recipients, Najarian et al. (14) found no influence of donor source on late mortality or graft loss, although the second decade mortality in this group was very large (50% in 5 years). In most reports, causes of posttransplant death are almost the same. Infections and CVD dominate (2, 5, 12, 13), while cancer is a rare cause of death. In our series only one patient died of cancer (ca. coli). In a report on transplantation in Europe, transplanted patients were compared with a matched population from West-Germany. Cardiovascular deaths were more frequent among transplanted patients, but there was no significant difference in the mean annual death rate from malignant diseases (2). In the second decade after transplantation, cardiovascular events are the dominating cause of death in our series. Death from malignancy accounted for 20% as have also been reported by others (8, 10, 16, 17). In contrast to these reports, however, we have no death from hepatic failure. The number of hospital admissions and days spent in hospital, reflect patient morbidity. Most of the hospitalization is connected with terminal illness as the patients with functioning grafts only have 2.8 days per patient year in hospital. This is in accordance with the data reported by Kirkman et al. (7). At followup, the most debilitating complication of immunosuppression was aseptic bone necrosis in the femoral head which was seen in 149’0 of the &and J Urol NephrolZ6

patients. Gastrointestinal complications were the most frequently reported complication in our series, but among these only one was serious, one patient had a life-treatening bleeding from oesophageal varices of unknown etiology. In conclusion: This study has shown that in patients who have had their grafts for more than 10 years, the morbidity is low and there are few serious complications to the immunosuppressive regime. Cardiovascular events cause most of the deaths in the second decade after transplantation. REFERENCES 1. Breslow NA. Generalized Kruskal-Wallis test for

2. 3. 4.

5. 6. 7. 8.

9.

LO. 11.

12.

13. 14.

comparing k samples, subject to unequal censorship. Biometrica 1970; 57: 579-593. Brunner FP, ed. Combined report on regular dialysis and transplantation in Europe, XVIII, 1987. London: Springer, 1988. Cox DR. Regression model and life table. J R Stat 1972; 34; 187-201. Dixon JWED. BMDP. Statistical software. Los Angeles: University of California Press, 1985. Frisk B, Persson H, Wedel N et al. Study of 172 patients at 10 to 12 years after renal transplantation. Transplant Proc 1987; 19; 5; 2769-3771. Kaplan EL, Meier P. Non-parametric estimation from incomplete observation. J Am Stat Assoc 1958; 53; 457-481. Kirkman RL, Strom TB, Weir MR, Tihey NL. The natural history of long-term renal transplantation. Transplant Proc 1983; 15; 1; 1043-1045. Lee HM, Kenyon N, Mendez-Picon G, Posner MP. Long term survivors of kidney transplantation, mortality, rehabilitation, and immunological reactivities. Transplant Proc 1987; 19; 1; 2 120-2 12 1. Lowrie EG, Lazarus JM, Mocelin AJ et al. Survival of patients undergoing chronic hemodialysis and renal transplantation. N Engl J Med 1973; 288; 17; 863-867. Mahony JF, Savdie E, Caterson RJ. The natural history of cadaveric renal allografts beyond ten years. Transplant Proc 1986; 18; 1 ; 135-1 37. Mantel N. The evaluation of survival data and two rank order test statistics, arising in its considerations. Cancer Chemother Rep 1966; 50; 163-1 70. McGeown MG, Douglas JF, Donaldson RA et al. Ten-year results of renal transplantation with azathioprine and prednisolone as only immunosuppression. Lancet 1988; i; 983-985. Murray JE, Tihey NL, Wilson RE. Renal transplantation: A twenty-five year experience. Ann Surg 1976; 184; 565-573. Najarian JS, Kaufman DB, Fryd DS et al. Longterm survival following kidney transplantation in

Ten-yearfollow-up of kidney transplantation

16. Rao KV, Andersen RC. Long-term results and complications in renal transplant recipients. Transplantation 1988; 45;45-52. 17. Wing AJ. Combined report on regular dialysis and transplantation in Europe, X, 1979. Proc Eur Dial Transpl 1980; 17; 4-86.

Scand J Urol Nephrol Downloaded from informahealthcare.com by University of Otago on 01/10/15 For personal use only.

100 type I diabetic patients. Transplantation 1989; 4’7; 1; 106-113. 15. Nylander WA, Sutherland DER, Bentley FR, Najarian JS. Fifteen- to twenty-year follow-up of renal transplants performed in the 1960s. Transplant Proc 1985; 17; 1; 104-105.

40 1

Scand J Urol Nephrol26

Long term morbidity and mortality after kidney transplantation.

A cohort of 69 patients received a kidney transplant in the period 1963-1977. The mean observation time was 9.5 years. Accumulated follow-up time was ...
355KB Sizes 0 Downloads 0 Views