Short Pretransplant Dialysis Periods are Clinically Acceptable Without Compromising Kidney Transplant Outcomes: A Japanese Multicenter Study

ABSTRACT

Objectives

Preemptive kidney transplantation (PEKT) is known to have better outcomes than kidney transplantation (KT) after dialysis therapy. However, the effect of pretransplant dialysis duration (PTDD) on KT outcomes under good hemodialysis control remains unclear. We investigated the association between PTDD and KT outcomes in a Japanese cohort with favorable dialysis management.

Methods

Among three institutions, 805 patients who underwent a living donor KT between 2001 and 2021 were enrolled in this study. Those who underwent more than one KT, multiple organ transplantations in addition to KT, or transplantations at < 20 years of age were excluded. Calcineurin inhibitors, mycophenolate mofetil, and steroids were the primary maintenance immunosuppressors.

Results

The allograft survival rates of the non‐PEKT group were significantly lower than those of the PEKT group (hazard ratio [HR] 1.67, 95% confidence interval [CI] 1.04–2.70). Univariate analysis showed that ≥ 2 years of PTDD demonstrated the highest HR for allograft loss when PTDD was separated by year (95% CI 1.11–2.55). Multivariate analysis showed that ≥ 2 years of PTDD (HR 2.61 [95% CI 1.58–4.38]), being a male recipient (HR 2.35 [95% CI 1.30–4.53]), and having diabetic nephropathy (HR 2.94 [95% CI 1.71–4.98]) were suggested to be risk factors for allograft loss. Additionally, < 2 years of PTDD resulted in significantly higher allograft survival rates; the 10‐year allograft survival rates of recipients with ≥ 2 years and < 2 years of PTDD were 82.0% and 91.8%, respectively.

Conclusions

Allograft survival remained unaffected when KT is performed after a short‐term dialysis introduction.

1. Introduction

Kidney transplantation (KT) is the only curative treatment for end‐stage kidney disease among renal replacement therapies. It is considered the first‐line treatment for kidney failure as it is associated with lower mortality compared to maintenance hemodialysis or peritoneal dialysis [1]. In particular, preemptive kidney transplantation (PEKT) is generally associated with more favorable outcomes than KT performed after the initiation of dialysis. It has been reported that prolonged pretransplant dialysis duration (PTDD) adversely affects both transplant recipient and kidney allograft survival rates [2]. Regarding the association between PTDD and transplant outcomes, several studies suggest that short‐term pretransplant dialysis therapy does not adversely affect posttransplant outcomes [3, 4]. For example, Goldfarb et al. demonstrated that a dialysis period of up to 6 months does not impair either transplant patient or allograft survival [3]. Conversely, Prezelin‐Reydit et al. demonstrated that even 6 months of dialysis period prior to KT may lead to inferior outcomes compared with PEKT in French cohort [5]. Other studies indicate that long‐term PTDD could possibly cause lower transplant outcomes [6, 7, 8]. Moreover, even if PTDD negatively affects KT outcomes, the extent to which PTDD correlates with kidney allograft and patient survival rates remains unclear [9, 10]. These correlations may vary across countries depending on the specific circumstances of each society and community regarding dialysis, including the medical insurance system, accessibility, management, and unexplained characteristics.

Dialysis management in Japan is exceptionally advanced compared with that in other countries, resulting in the world's highest long‐term dialysis survival rates [11, 12]. Therefore, we explored the impact of PTDD on KT outcomes in a Japanese cohort to determine whether PEKT yields the best results or, if some degree of pretransplant dialysis is acceptable, to what extent it is tolerable.

2. Materials and Methods

2.1. Study Design

This retrospective cohort study included living donor KT recipients who underwent KT at three institutions: Hokkaido University Hospital (Sapporo, Japan), Sapporo City General Hospital (Sapporo, Japan), and Kushiro City General Hospital (Kushiro, Japan). This study investigated the association between PTDD and KT outcomes. This study was approved by the hospital's institution review board (approval no: 022‐0340).

2.2. Study Population

We included consecutive patients aged ≥ 20 years who had received a KT from living donors in between January 2001 and March 2021 at the three institutions. Out of the 1008 recipients, 805 recipients were enrolled in the study based on the following exclusion criteria: recipients who underwent more than one KT, those who underwent multiple organ transplantations in addition to KT, or those who underwent KT despite positive flow cytometric T‐cell crossmatch. A combination of calcineurin inhibitors (tacrolimus or cyclosporine), mycophenolate mofetil, and steroids was administered as primary maintenance immunosuppressive drugs. Of the included recipients, 258 received everolimus at induction. Additionally, recipients who received other immunosuppressants such as azathioprine or mizoribine were excluded from the study.

2.3. Outcome Measurement

Patients were analyzed from the day of KT until death, allograft loss, or the latest follow‐up, that is, until March 2021. In a secondary analysis, the patients were divided into two groups according to the years of PTDD to investigate how prolonged PTDD affects kidney allograft survival. The causes of allograft loss were collected from medical records. Chronic rejection was defined as cases included biopsy‐proven or clinically diagnosed. Death with a functioning graft (DWFG) was defined as death with a functioning allograft without the need for dialysis. Graft loss was defined as the loss of graft function at the point of dialysis induction, or the second KT. PEKT is typically defined as transplantation performed before the initiation of chronic dialysis. In line with previous study, short‐term dialysis (< 1 month) was included in the operational definition of PEKT [13]. In this study, we operationally defined PEKT as transplantation performed before or within 28 days after dialysis initiation.

2.4. Statistical Analysis

Patients' characteristics at KT are described as mean and standard deviation (SD), median and range, or frequency and percentage. The statistical differences in continuous variables between the two groups were compared using the unpaired t‐test or Mann–Whitney test according to the data distribution (parametric or nonparametric). The differences in categorical variables between two or more groups were investigated using Fisher's exact test or the chi‐squared test. Overall and allograft survival curves were estimated using the Kaplan–Meier method, and hazard ratios (HR) were estimated using the Cox proportional hazards model. The confidence level was set to 0.95 (95% CI). In the univariate and multivariate analyses of allograft survivals, risk factor variables were selected based on their clinical relevance in previously reported allograft survivals [14]. For easier interpretation of the results of the univariate and multivariate analyses, continuous variables were categorized based on approximate sample tertiles (age, total ischemic time, and warm ischemic time). Statistical significance was set at p < 0.05. Statistical analyses were conducted using GraphPad Prism ver 9.5.1 (San Diego, CA, USA).

3. Results

3.1. Demographics

The clinical characteristics of the enrolled patients are listed in Table 1. The mean follow‐up period and median PTDD were 7.9 ± 5.3 years and 0.7 (0–34.3) years, respectively. Among the 805 recipients, ABO‐incompatible KT was performed in 29.6%. Data regarding total ischemic time and warm ischemic time were available for 773 and 736, respectively.

TABLE 1.

Recipient characteristics.

	N = 805
Observation period, years a	7.9 ± 5.3
Male recipient, %	522 (64.8)
Recipient age, years a	47.8 ± 13.1
ABO blood type incompatibility, %	238 (29.6)
Diabetic nephropathy, %	147 (18.3)
PEKT, %	261 (32.4)
Everolimus administration, %	258 (32.0)
PTDD, years b	0.7 (0–34.3)
Total ischemic time, minute b , c	99 (45–860)
Warm ischemic time, second b , d	252 (60–2700)
Male donor, %	244 (30.3)
Donor age, years a	57.0 ± 9.9

Abbreviations: PEKT, preemptive kidney transplantation; PTDD, pretransplant dialysis duration.

^a Mean ± SD.

^b Median (range).

^c Including 32 missing data.

^d Including 69 missing data.

3.2. Kidney Transplant Outcomes

The overall, allograft, and death‐censored allograft survival rates are shown in Figure 1A–C. Recipient overall survival did not differ significantly between the PEKT and non‐PEKT groups (HR 1.62, 95% CI 0.772–3.41; Figure 2A). In contrast, allograft survival was significantly lower in the non‐PEKT group (HR 1.67, 95% CI 1.04–2.70; Figure 2B). Death‐censored allograft survival did not differ significantly between the groups (HR 1.79, 95% CI 0.981–3.28; Figure 2C). Subsequently, comparing kidney allograft survival rates between the PEKT group and other groups divided by years of PTDD, the recipients with PTDD 2–3 years and those with PTDD ≥ 3 years showed significantly lower allograft survivals. However, there was no significant difference between the PEKT group and groups with PTDD up to 2 years. (Figure 3).

FIGURE 1.

Kaplan–Meier curves of overall survival rates, allograft survival rates and death censored allograft survival rates. (A) Overall survival rates, (B) allograft survival rates and (C) death censored allograft survival rates of this cohort were drawn by Kaplan–Meier curves.

FIGURE 2.

Overall survival curves, allograft survival curves and death censored allograft survival curves of the PEKT and non‐PEKT group. (A) Overall survival rates in the non‐PEKT group were comparable to those in the PEKT group. Conversely, (B) allograft survival rates of the non‐PEKT group were significantly lower than those of the PEKT group. (C) Death censored allograft survival rates in the non‐PEKT group were comparable to those in the PEKT group. Each curve was drawn by the Cox proportional hazards model. HR were analyzed the non‐PEKT vs. PEKT. PEKT; preemptive kidney transplantation.

FIGURE 3.

Allograft survival curves of the PEKT and groups divided by years of PTDD. Allograft survival rates of the groups with PTDD 2–3 years and groups with PTDD ≥ 3 years were significantly lower than those of the PEKT group. Curves were drawn by the Cox proportional hazards model. HR were analyzed the group divided by years of PTDD vs. PEKT. PEKT; preemptive kidney transplantation, PTDD; pretransplant dialysis duration.

3.3. Risk Factors Associated With Allograft Survival: Univariate and Multivariate Analyses

To evaluate the effect of prolonged PTDD on allograft survival and to investigate the potential impact of confounding factors, we compared multiple candidate cutoffs (≥ 1, ≥ 2, ≥ 3, and ≥ 4 years) in univariate and multivariate Cox analyses. In the univariate analysis, HRs of PTDD separated by years were 1.56 (95% CI 1.02–2.43), 1.67 (95% CI 1.11–2.55), 1.49 (95% CI 0.980–2.25), and 1.28 (95% CI 0.824–1.96) in recipients with ≥ 1 year, ≥ 2 years, ≥ 3 years and ≥ 4 years of PTDD, respectively (Table 2). Being a male recipient (HR 1.93 [95% CI 1.22–3.19]), recipients aged ≥ 56 years (HR 1.76 [95% CI 1.05–2.95]), and diabetic nephropathy (HR 2.70 [95% CI 1.69–4.22]) were also indicated as high‐risk factors for allograft loss. Furthermore, multivariate analysis showed that ≥ 2 years of PTDD (HR 2.61 [95% CI 1.58–4.38]), being a male recipient (HR 2.35 [95% CI 1.30–4.53]), and diabetic nephropathy (HR 2.94 [95% CI 1.71–4.98]) were suggested to be risk factors for allograft loss (Table 2). One year or more of PTDD or 3 years or more of PTDD were also risk factors in each multivariate analysis (Tables S1 and S2); however, the highest HR related to PTDD was observed at 2 years or more. Across all centers, PTDD of ≥ 2 years showed the highest hazard ratio. As this threshold was identified through a data‐driven selection process rather than predefined a priori, this finding should be regarded as exploratory and interpreted accordingly.

TABLE 2.

Univariate and multivariate analysis for allograft loss separated by years of PTTD.

Variable	Univariate analysis		Multivariate analysis
	Hazard ratio	95% CI	Hazard ratio	95% CI
PTDD ≥ 1 year	1.56	1.02–2.43
PTDD ≥ 2 years	1.67	1.11–2.55	2.61	1.58–4.38
PTDD ≥ 3 years	1.49	0.980–2.25
PTDD ≥ 4 years	1.28	0.824–1.96
Male recipient	1.93	1.22–3.19	2.35	1.30–4.53
Recipient's age, years
≤ 40	1.0 (Reference)
41–55	1.41	0.850–2.35	1.38	0.743–2.61
≥ 56	1.76	1.05–2.95	1.18	0.619–2.26
ABO blood type incompatibility	1.10	0.673–1.72	0.792	0.454–1.33
Diabetic nephropathy	2.70	1.69–4.22	2.94	1.71–4.98
Everolimus administration	0.621	0.282–1.23	0.541	0.230–1.13
Total ischemic time, minutes
≤ 90	1.0 (Reference)
91–120	1.05	0.624–1.72	0.923	0.521–1.62
≥ 121	0.776	0.441–1.32	0.583	0.305–1.07
Warm ischemic time, seconds
≤ 240	1.0 (Reference)
241–300	1.13	0.617–2.01	1.17	0.620–2.14
≥ 301	1.17	0.677–1.99	1.43	0.810–2.49
Male donor	0.574	0.332–0.938	0.702	0.360–1.29
Donor's age, years
≤ 50	1.0 (Reference)
51–60	0.984	0.591–1.67	1.17	0.625–2.22
≥ 61	1.23	0.724–2.11	1.40	0.724–2.73

Abbreviation: PTDD, pretransplant dialysis duration.

3.4. Kidney Transplant Outcomes of the Two Groups Divided by 2 years of PTDD

The recipients were divided into two groups, as the highest HR related to PTDD was observed at 2 years or more: group A (n = 515), which included those with < 2 years of PTDD, including recipients with PEKT, and group B (n = 290), which included recipients with ≥ 2 years of PTDD. The overall survival of group B was not significantly lower than that of group A (HR 1.82, 95% CI 0.965–3.45; Figure 4A). In contrast, both allograft survival (HR 1.68, 95% CI 1.11–2.55; Figure 4B) and death‐censored allograft survival (HR 1.73, 95% CI 1.03–2.89; Figure 4C) were significantly lower in group B than in group A.

FIGURE 4.

Overall survival curves, allograft survival curves and death censored allograft survival curves of group A and group B. (A) Overall survival rates of the group B were comparable to those of the group A. Conversely, (B) allograft survival rates of the group B were significantly lower than those of the group A. Similarly, (C) death censored allograft survival rates of the group B were significantly lower than those of the group A. Curves were drawn by the Cox proportional hazards model. HR was analyzed the group B vs. group A. PTDD; pretransplant dialysis duration. Group A: Recipients with PTDD < 2 years, including preemptive kidney transplant recipients. Group B: Recipients with PTDD ≥ 2 years.

3.5. Comparison of Patient Characteristics of Two Groups

Patient characteristics were compared to evaluate the confounding factors between the two groups (Table 3). Although the mean age at KT tended to be lower in Group A than in Group B (46.8 ± 13.6 years vs. 49.5 ± 12.1 years, respectively), we do not consider this difference to be clinically significant. The incidence of everolimus administration in Group A tended to be higher than that in Group B (36.9% vs. 23.4%, respectively). No significant differences were observed in other variables between the two groups, including proportion of patients with diabetic nephropathy.

TABLE 3.

The comparison of recipient characteristics between group A and group B.

	Group A (n = 515)	Group B (n = 290)	p
Male recipient, %	340 (66.0)	182 (62.8)	0.3572 a
Recipient age, years b	46.8 ± 13.6	49.5 ± 12.1	0.0045 c
ABO blood type incompatibility, %	149 (28.9)	89 (30.7)	0.6295 a
Diabetic nephropathy, %	102 (19.8)	45 (15.5)	0.1539 a
Everolimus administration, %	190 (36.9)	68 (23.4)	< 0.0001 a
Total ischemic time, minutes d , e	98 (45–610)	99 (54–860)	0.7581 f
Warm ischemic time, seconds d , g	250 (60–946)	260 (60–2700)	0.3309 f
Male donor, %	148 (28.7)	96 (33.1)	0.2020 a
Donor age, years b	57.1 ± 9.8	57.0 ± 10.2	0.9106 c

Abbreviation: PTDD, pretransplant dialysis duration.

^a Fisher's exact test.

^b Mean ± SD.

^c Unpaired t‐test.

^d Median (range).

^e Including 32 missing data.

^f Mann–Whitney test.

^g Including 69 missing data.

3.6. Comparison of Causes of Allograft Loss

Next, we sought to investigate the causes of allograft loss in order to figure out a rational explanation to make a difference in allograft survival rates between the two groups. The causes of allograft loss are shown in Figure 5. The main causes of allograft loss were DWFG (38% in group A and 33% in group B) and chronic rejection (25% in group A and 33% in group B). There were no obvious differences in any variables regarding causes of allograft loss.

FIGURE 5.

The comparison of causes of allograft loss between group A and group B. PTDD; pretransplant dialysis duration. Group A: Recipients with PTDD < 2 years, including preemptive kidney transplant recipients. Group B: Recipients with PTDD ≥ 2 years.

4. Discussion

In this study, we showed that the allograft survival rates of the non‐PEKT group were significantly lower than those of the PEKT group. Furthermore, the allograft survival rates of the group with ≥ 2 years of PTDD were significantly lower than those of the group with < 2 years of PTDD. Multivariate analysis suggested that ≥ 2 years of PTDD was one of risk factors for allograft loss, though there were no huge differences in the causes of allograft loss between the two groups.

The question of whether the induction of short‐term pretransplant dialysis affects the allograft survival rate is inconclusive. Some studies have reported that PTDD of ≤ 6 months does not impair the outcomes of KT [3]. Similarly, the current study suggested that short‐term PTDD does not decrease the allograft survival rate under good hemodialysis management. In contrast, although there are many previous reports of PTDD affecting overall survival [7, 9, 15], this study found no significant difference in the overall survival rate after the induction of dialysis. The lower prevalence of cardiovascular disease in Japanese dialysis patients compared with Europe or the US [11] may explain why this phenomenon occurred, as cardiovascular disease is the most common cause of death in dialysis patients.

Furthermore, it is unclear why shorter PTDD leads to better allograft survival and how prolonged PTDD affects KT outcomes. Bueti et al. reported that KT ameliorates the vascular endothelial cell damage caused by prolonged dialysis [16], and Hotta et al. reported that successful KT decreases systolic and diastolic blood pressure by measuring pulse wave velocity [17]. This mechanism may explain why a shorter PTDD resulted in better kidney allograft survival. Furthermore, prolonged dialysis is known to cause dialysis‐related complications such as vascular calcification and atrophic bladder, which are known to be risk factors for intraoperative vascular and urological complications [18, 19]. Additionally, intraoperative complications in KT are known to adversely affect allograft outcomes [18, 20]. Therefore, prolonged dialysis treatment may impair allograft survival following an increase in intraoperative complications. Since intraoperative complications were not examined in this study, further investigation of the relationship between PTDD and intraoperative complications of KT is required.

To elucidate the relationship between prolonged PTDD and poor allograft outcomes, we investigated the causes of allograft loss in detail. Long‐term dialysis has been reported to increase the incidence of acute rejection [21] and cardiovascular events [13, 22] for the following reasons: maintenance dialysis is associated with the rapid progression of cardiovascular changes, especially left ventricular hypertrophy and vascular calcification [23]. Moreover, multiple factors contribute to increased cardiovascular morbidity and mortality, such as increased inflammation [24], increased FGF‐23 levels, hyperphosphatemia [25, 26], and alterations in the lipid profile and concentrations of advanced glycosylation end‐products [24, 27]. Therefore, we had hypothesized that allograft loss due to acute rejection or DWFG, which is associated with cardiovascular events, was common in recipients with prolonged PTDD. However, this study found no differences in the causes of allograft loss between the groups, which were divided into those who received ≥ 2 years of PTDD and those who received < 2 years of PTDD. Further investigations are required to elucidate the background differences that impair allograft survival rates in recipients with long PTDD.

Although PEKT has been reported to have better outcomes than non‐PEKT, some issues regarding PEKT persist. First, as pretransplant evaluation required substantial time, patients undergoing PEKT must be stable enough to remain dialysis‐free throughout evaluation period [28]. Some patients with chronic kidney disease insist on undergoing KT without dialysis induction even when it is clinically inevitable [29]. Occasionally, other patients require insertion of a vascular access catheter due to an unexpected decline in kidney function, which can lead to susceptibility to perioperative infection.

This study had a few limitations. First, this study was a retrospective analysis of the data obtained from three institutions. Therefore, to align recipient profiles, exclusion criteria, including number of KTs undergone, multiple organ transplants during KT, high immunological risk, and immunosuppressant regimen, were applied. Second, the ≥ 2‐year PTDD threshold was not predefined but identified through a data‐driven, exploratory comparison of multiple candidate cutoffs, introducing a risk of threshold optimization bias that may overestimate the association and limit generalizability. External validation in independent cohorts is warranted. Third, this study was conducted in a cohort of patients who underwent dialysis therapy in Japan, where dialysis outcomes are extraordinary compared to those in other countries. This interpretation is largely based on nationwide data and may not necessarily be applicable to the three centers included in this study. Fourth, the cutoff for PTDD is expected to vary depending on the dialysis facility and recipient background as well as medical insurance system in each regional community. Therefore, we recommended that the threshold of ≥ 2 years of PTDD was obtained in an exploratory manner and should not be universally applied to other facilities. We place emphasis on the contention that short‐term PTDD does not hamper allograft survival. Fifth, this study spans a long period from 2001 to 2021. In Japan, the proportion of preemptive kidney transplantation has increased in recent years, and the use of everolimus differed among PTDD groups. However, when outcomes were compared according to dialysis duration stratified by 5‐year intervals, no significant differences were observed (data not shown). In addition, it has been reported that kidney transplant outcomes in Japan have remained relatively stable since 2001 [30]. Finally, donor‐specific antibodies could not be assessed in all recipients because of historical limitations; therefore, flow cytometric T‐cell crossmatch was used as an alternative immunological assessment.

In conclusion, PEKT remains an ideal strategy for achieving long‐term allograft survival in patients with kidney failure. Induction of dialysis does not seem to lower allograft survival if administered not long before kidney transplantation. Therefore, if necessary, dialysis therapy should be initiated without hesitation before kidney transplantation. Moreover, efforts should be made to shorten the waiting period for donated kidney transplantation.

Author Contributions

Keisuke Maeda: writing – original draft, investigation, data curation, visualization. Kiyohiko Hotta: conceptualization, investigation, validation, writing – review and editing, supervision, methodology. Takayuki Hirose: writing – original draft, data curation, investigation, visualization. Nobuo Shinohara: supervision. Tatsu Tanabe: data curation. Keita Takahashi: formal analysis, validation. Naoya Iwahara: data curation. Hajime Sasaki: data curation. Haruka Higuchi: data curation. Shigeru Harada: data curation. Yusuke Takada: data curation. Keita Kawashiro: data curation. Ken Morita: data curation.

Funding

The authors have nothing to report.

Ethics Statement

This study protocol was reviewed and approved by the hospital's institution review board (approval number: 022‐0340).

Consent

The opt‐out method was applied to obtain consent from participants via websites.

Conflicts of Interest

Kiyohiko Hotta is an Editorial Board member of International Journal of Urology and a co‐author of this article. To minimize bias, he was excluded from all editorial decision‐making related to the acceptance of this article for publication. The other authors declare no conflicts of interest.

Supporting information

Table S1: Multivariate analysis for allograft loss separated by PTTD of 1 year.

Table S2: Multivariate analysis for allograft loss separated by PTTD of 3 years.

Data Availability Statement

The data that supports the findings of this study are available in the Supporting Information of this article.