Early Experience with Dual Kidney Transplantation in Adults using Expanded Donor Criteria

In an effort to overcome the disparity between supply of cadaveric donors and demand, various strategies have emerged to expand the existing donor selection criteria. Kidneys have been used from donors of 50 yr or more, or with a prolonged history of hypertension and diabetes, or other evidence of renal disease (1,2). Although this is technically feasible, there is major concern about the use of such organs based on experimental (3,4) and clinical (5,6) evidence that kidneys with preexisting injury or reduced function may have poor outcomes. Actually, in humans, transplantation of a single kidney with a suboptimal number of nephrons to start with serves to create the clinical counterpart of experimental reduction of renal mass (7). Indeed, in most cases, cold ischemia time, one or more acute rejections, and toxicity of medications—particularly cyclosporin A and tacrolimus—further reduce the already limited number of nephrons that a single suboptimal kidney provides (7). This remnant kidney—now containing 20 to 30% of viable parenchyma or less compared to that in two optimal kidneys—initiates a self-perpetuating program of progressive deterioration, as commonly seen in animals undergoing renal mass ablation (8). That nephron mass is a determinant of chronic allograft failure has been formally tested in animal experiments showing that increasing the number of viable nephron mass by simultaneous transplantation of two kidneys into the same recipient effectively prevented the progressive deterioration in renal function that occurs in control subjects given a single kidney (9,10). Although routine use of two ideal kidneys in humans is unlikely in the short term due to an insufficient number of organs available for all those in need, use of two marginal kidneys in a single recipient might be an option. Most transplant centers currently discard kidneys from donors older than 60 yr, diabetic patients, or those with evidence of other renal diseases (11). Actually, recipients of single kidneys from older donors were at increased risk of posttransplant anuria and relatively poor graft function up to 6 mo follow-up (12). However, if two marginal organs were given to the same recipient, this may result in as many or more functioning nephrons as in a single ideal organ. Thus, use of compromised kidneys, rather than burdening the already limited pool of cadaveric organs, might actually enhance availability. Indeed, analyses of dual marginal versus single ideal transplant outcomes found a comparable 1-yr graft survival and renal function recovery by both of the procedures (13,14). However, these encouraging results could have been influenced to some extent by the nonrandomized, monocenter, retrospective design of the studies and by the use—in as many as one-third of cases—of organs with evidence of acute, rather than chronic, injury that conceivably contributed to improve the outcome of the dual procedure (13).

Thus, the present study was designed to formally compare in a prospective, controlled, multicenter design the short-term outcome of transplantation of two marginal versus single ideal kidneys to extend our experience with this approach and thereby enhance the organ donor pool.

Materials and Methods

In October 1996, the Double Kidney Transplant Group (DKG) was established with the goal of identifying potential donors of suboptimal kidneys and performing double kidney transplants in eligible recipients. Guidelines for donor and recipient selection, the surgical procedures, and the study design were defined in a study protocol finalized by the DKG and approved by the Institutional Review Committees at each participating center and by the Ethics Committee of the Clinical Research Center “Aldo & Cele Daccò,” Villa Camozzi, Ranica, which was also appointed to monitor ethical and statistical issues and, in particular, to evaluate the safety and efficacy profile of the procedures. Each recipient gave written informed consent to enter the study.

Results

Twenty-four cases and 48 control subjects received a double and a single kidney transplant, respectively, from November 1996 to September 1998. Double kidney donors and recipients were significantly older compared with single kidney donors (age: 68.7 ± 6.8 versus 46.7 ± 15.2 yr, respectively, P = 0.0001) and recipients (age: 59.4 ± 9.9 versus 50.2 ± 12.1 yr, P = 0.002). Of the 24 donors, 22 were more than 60 yr old. Of these, 13 were hypertensive and four were hypertensive and diabetic. The two donors less than 60 yr old were diabetic. No donor had clinical proteinuria. Donor creatinine clearance estimated by the Cockcroft-Gault formula was available in 23 cases and 41 control donors and, as expected, was significantly lower in case donors (54.5 ± 16.0 versus 78.1 ± 48.0 ml/min per 1.73 m², P = 0.03). HLA-A, -B, and -DR donor/recipient mismatches (3.9 ± 1.2 versus 3.8 ± 1.2) and recipient baseline characteristics were comparable in the two groups (Table 2). Mean cold (20.3 ± 0.3 versus 20.4 ± 0.6 h) and warm (0.37 ± 0.02 versus 0.38 ± 0.02 h) ischemia times were comparable in cases and control subjects as well.

Median (range) follow-up was 6 (3 to 6) and 6 (4 to 6) mo in cases and control subjects, respectively. At last follow-up, patient and organ survival was 100% with both of the procedures. The incidence of posttransplant anuria was identical in cases and control subjects. Overall, five cases (20.8%) versus 10 control subjects (20.8%) required posttransplant chronic dialysis for a median (range) of 5 (2 to 12) and 7 (2 to 13) d, respectively. Time to chronic dialysis discontinuation was remarkably comparable in the two groups (Figure 1). Time to normal serum creatinine was comparable in the two study groups as well (Figure 2). Overall, mean serum creatinine values at each time visit and, on average, during the whole follow-up period were almost identical (4.2 ± 3.4 versus 4.3 ± 3.3 mg/dl, P = 0.16), but with significantly lower levels in cases compared with control subjects (1.56 ± 0.65 versus 1.74 ± 0.73 mg/dl, P = 0.04) from month 2 to last follow-up (Figure 3). At 6 mo, serum creatinine was numerically lower in cases than in control subjects (1.48 ± 0.54 versus 1.77 ± 0.93 mg/dl), but the difference failed to achieve statistical significance (P = 0.24). At each visit, estimated creatinine clearance was comparable in cases and control subjects and, in parallel with the decline in serum creatinine, progressively increased throughout the follow-up period in both groups (data not shown). At last follow-up, cases had a mean creatinine clearance virtually identical to that of corresponding donors (51.1 ± 22.2 versus 54.5 ± 16.0 ml/min, P = 0.55). On the contrary, at last follow-up control subjects had significantly lower creatinine clearances than corresponding donors. However, since control subjects received only one kidney from corresponding donors, we repeated the analyses comparing recipient creatinine clearances with halved donor creatinine clearances and found significantly higher values in recipients than in donors (53.6 ± 16.2 versus 41.0 ± 22.8 ml/min, P = 0.006). In other words, the filtration power of donor kidneys increased by about 30% after they were singly engrafted in the recipient. Overall, 13 kidneys were found to have a histology score <3 and were transplanted as single kidneys at the institutions involved in the double kidney transplant study. Of note, final serum creatinine of recipients of these kidneys was 1.5 ± 0.4 mg/dl and did not differ from that of other control subjects (1.9 ± 0.7 mg/dl, P = 0.09). Six kidneys were discarded because they had a histology score >7.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

. Percentage of cases and control subjects on chronic dialysis at different times.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

. Percentage of cases and control subjects with serum creatinine >1.4 mg/dl at different times.

• Download figure
• Open in new tab
• Download powerpoint

Figure 3

. Time course of serum creatinine concentration and systolic/diastolic BP in cases and control subjects. Absolute numbers of cases and control subjects on follow-up at different times are given.

In the majority of the posttransplant visits, systolic and diastolic BP values were comparable in the two groups, with significantly lower values in cases compared with control subjects only at days 0 and 1 (systolic) and at days 0 and 9 (diastolic) after transplant (Figure 3). Overall, average follow-up systolic BP values were comparable in the two groups (147.5 ± 22.5 versus 148.9 ± 21.6 mmHg in cases and control subjects, respectively, P = 0.32). On the contrary, average follow-up diastolic BP values were significantly lower in cases than in control subjects (83.2 ± 11.5 versus 85.1 ± 12.5 mmHg, P = 0.008).

In the entire study group, among the considered donor and recipient baseline characteristics, donor/recipient body weight was the only variable significantly associated at univariate (P = 0.0001) and multivariate (P = 0.001) analysis with last available serum creatinine concentrations (Table 3). Of note, the model found a similar trend even for donor/recipient BMI ratio at univariate (P = 0.01) and at multivariate (P = 0.07) analysis. Furthermore, both donor/recipient body weight (r = -0.549, P = 0.0001, Figure 4) and BMI (r = -0.416, P = 0.009) ratios negatively correlated with serum creatinine values at last available visit. At univariate and multivariate analyses, the type of transplant and recipient anuria were not associated with last available serum creatinine values (Table 3). Donor creatinine clearance had no predictive value as well (Table 3), and did not correlate with final serum creatinine (r = -0.12, P = 0.38). When recipient creatinine clearance was considered as outcome variable, again donor/recipient body weight and BMI ratios were the only variables with a significant predictive value at univariate and at multivariate analyses (data not shown).

• Download figure
• Open in new tab
• Download powerpoint

Figure 4

. Correlation between donor/recipient body weight ratio and final serum creatinine concentration in the whole study group.

The incidence of acute allograft rejection was comparable in the two groups. Five cases (20.8%) versus 9 control subjects (18.8%) had one episode of acute rejection responsive to intravenous steroids (P = 1.00). Rejection episodes occurred at posttransplant days 9, 13, 60 (two episodes), and 90 in cases, and at posttransplant days 4 (two episodes), 5, 7, 8, 9, 10, 13, and 150 in control subjects (P = 0.94).

The overall incidence of major surgical complications was comparable in the two groups, but with a trend to an increased risk (P = 0.06) of urinary tract fistulas requiring surgical correction among cases compared with control subjects (Table 4). At univariate analysis, recipient age was the only covariate associated with the risk of urinary tract fistula (P = 0.04). At multivariate analysis, however, none of the considered covariates (including double versus single transplant) was significantly associated with this risk. No renal artery or vein thrombosis was reported in either group.

Discussion

This is the first study formally designed to compare in a prospective, controlled, and multicenter manner the short-term outcome of double transplants of marginal kidneys with that of single transplants of optimal kidneys. Overall, we found that double and single kidney transplants were associated with remarkably good short-term patient and organ survival, which reached 100% with both of the procedures. Furthermore, the overall incidence of surgical or infectious complications was comparable. These findings corroborate preliminary evidence from uncontrolled (13,14,17) or small (18,19) series, and from the United Network of Organ Sharing Registry data (H. S. Mackenzie, G. M.Chertow, B. M. Brenner, E. L. Milford. Dual kidney transplantation in the United States, 1987-1995: A review of the United Network of Organ Sharing (UNOS) database, submitted for publication) showing a relatively good safety profile of the procedure, and would indicate that fear of more surgical complications or more infections should not prevent the systematic use of marginal kidneys in double procedure protocols.

The incidence and duration of posttransplant anuria requiring continued renal replacement therapy were virtually identical in cases and control subjects. The rate of renal function recovery was comparable as well, as documented by the similar course of serum creatinine concentrations and the comparable incidence of patients achieving normal renal function at each posttransplant evaluation. It is of note, however, that cases compared with control subjects had lower serum creatinine values over the last 5 mo follow-up and lower diastolic BP levels averaged during the entire study period. These findings could not be attributed to bias in donor/recipient selection, because all relevant clinical and laboratory parameters that could affect renal function recovery were accurately matched in the two groups. There also were no differences in general medical support or in immunosuppressive protocols between cases and control subjects. Cold and warm ischemia times were comparable in the two groups and likewise could not explain the better functional recovery in cases compared with control subjects. On the contrary, a worse short-term outcome would have been predicted on the basis of the older age of donors and recipients of double marginal kidneys compared with donors and recipients of single kidneys (20). Indeed, the finding that cases showed a trend to a better recovery would suggest that the increased number of nephrons provided by the double procedure might have contributed to enhanced filtration power of the graft despite the older age of the donors. Actually, findings that higher donor/recipient body weight and BMI ratios were both significantly associated with a better renal function recovery suggested that in our series better outcomes actually reflected conditions of more nephrons. This would indicate that the greater the renal mass to start with, the higher the number of viable nephrons that can survive a given ischemic, toxic, or early immune insult and therefore continue to provide enough filtration power to sustain graft function. This enhanced functional reserve might have contributed to the lower diastolic BP we found in double compared with single kidney recipients during the entire follow-up period. This might be of major clinical relevance, since it is well established that high BP is one of the strongest predictors and determinants of accelerated renal function deterioration and eventual failure of the graft (21). Evidence that kidneys after a single transplant, but not after a dual transplant, increased their filtration power by about 30% may have some relevance as well for long-term graft outcome. Of interest, single transplants from donors more than 60 yr old to age- and gender-matched recipients performed as well as kidneys from younger donors, since the excess of graft failure in the old-donor-age group was due entirely to recipient deaths (22).

Of note, the incidence of acute rejections was virtually identical in the two groups. This finding, therefore, suggests that the enhanced antigenic load provided by more nephron mass supplied with the double transplant does not enhance the immune response of the host. Actually, experimental data suggest that at least in some circumstances the opposite is true (23).

We conclude that transplanting two marginal kidneys otherwise destined to be discarded is as safe and tolerated as a single transplant, and possibly offers an improved filtration power without exposing the recipient to enhanced risk of delayed renal function recovery, acute allograft rejection, or kidney loss because of major surgical complications.

In the present report, we showed the main results of a short-term study. Follow-up data, which will become available in the next few years, will reveal whether transplanting two marginal kidneys is an effective way to expand the limited donor pool, thereby offering new hope for transplantation to a substantial percentage of patients on dialysis. Further investigations will also show whether double kidney transplantation, by providing more nephron mass, will serve to delay or prevent chronic allograft dysfunction in the long-term.

Appendix