The study included all renal transplant patients at the Hospital del Mar (Barcelona) from May 1979 to December 2019. Follow-up was performed until graft loss due to patient death or return to RRT (dialysis or retransplantation) or until the end of follow-up on December 31, 2020. All transplantations were performed with negative cytotoxicity crossmatch.

All clinical and research activities included in the present work were carried out in accordance with the principles of the Helsinki and Istanbul Declarations.

Information from all renal graft biopsies (RB) performed in this cohort of patients during follow-up was included. Biopsies were performed according to clinical indication: significant proteinuria, renal function deterioration or detection of donor-specific HLA antibodies (DSA). Since 2014, we have had iterative follow-up RBs at 1, 3 and 5 years post-transplantation. Biopsies were reviewed independently by a pathologist and a nephrologist. The diagnosis of all renal biopsies was performed according to the Banff 2017 classification.10

Anti-HLA antibody screening was performed on the Luminex platform® (Lifecodes LifeScreen Deluxe assay. Immucor Lifecodes Transplant Diagnostics, Nijlen, Belgium). When anti-HLA antibody screening was positive, HLA IgG antibody identification was performed using Lifecodes® LSA class I and/or class II according to the manufacturer's instructions. Analysis was performed with MatchIT® software and the result was considered positive with MFI > 1,000.

The specificity of the anti-HLA antibodies was attributed considering HLA-A, HLA-B, HLA-C, DRB1* and DQB1*. When donor HLA-C or DQB1* typing was not available, anti-HLA antibodies against the donor were assigned considering linkage disequilibrium.

Graft loss was defined as the absence of renal function occurring at any time during post-transplant follow-up due to patient death or irreversible graft damage requiring return to dialysis or retransplantation. The causes of graft loss were classified into 2 groups1: death of the patient with functioning graft2 and loss of graft function.

The cause of graft loss was established after an exhaustive review of the patient's clinical history. The causes of graft failure were classified into 11 groups based on clinical and histological information (Table 1). In some cases, several pathologic processes could have contributed to graft failure. These cases were classified into a single group based on the clinical judgment of the investigators, and according to the most relevant diagnosis.

The causes of graft loss were analyzed according to two transplantation periods: period 1 (from 1979–1999) and period 2 (from 2000–2019).

Variables with a normal distribution were expressed as mean ± standard deviation (SD) and variables with a non-normal distribution with median and interquartile range (IQR). In the comparative analysis of the characteristics between groups, the Chi-square or Fisher's exact test was applied to analyze categorical variables. The Student t test was used for continuous variables with normal distribution and the Mann–Whitney U test for nonparametric continuous variables. ANOVA was performed for comparisons of more than 2 groups.

Graft and patient survival were estimated using Kaplan–Meier curves with the log-rank test. Time to graft loss accounting for death was considered from the date of transplantation to the start of dialysis or retransplantation. Logistic regression was performed for univariate and multivariate analysis of factors associated with graft loss due to thrombosis. For the study of factors associated with graft loss, the hazard ratio (HR) with 95% confidence interval (95% CI) was calculated using Cox regression. The analysis of attrition rates was performed by calculating the absolute rate of loss corrected for the amplitude of the period.7,11

Statistical analysis was performed using the SPSS package® version 25 (IBM). Values of p < 0.05 were considered statistically significant.

Between 1979 and 2019, 1,522 renal transplantations were performed. With a median follow-up of 63 (RIQ: 26–122) months, we observed 722 losses (47.4%).

In 239 (33.1%) the loss was due to death of the recipient with a functioning graft and in 483 (66.9%) due to loss of renal function with return to dialysis or re-transplantation (Fig. 1).

Fig. 1.

Population included in the study.

(0.28MB).

The baseline demographic and clinical characteristics of the study population are shown in Table 2.

Death of the recipient with a functioning graft was the cause of loss in 239 of the 722 (33.1%) grafts lost. The median period of time elapsed until death was 67.5 (RIQ: 35–122) months after transplantation.

Overall, cardiovascular disease (CVD) (25.1%) and neoplasms (25.1%) were the most frequent causes of death, followed by infectious causes, which accounted for 21.8% of deaths. The cause of death could not be determined in 19.2% of cases (Fig. 2A).

Fig. 2.

Causes of graft loss. (A) Causes of loss due to death of the recipient with functioning graft. (B) Causes of graft failure with return to dialysis or re-transplantation.

(0.26MB).

The mean age at death was 62.3 ± 13.8 years. No differences were found in the ages of recipients at death according to causes of death (CVD: 64.5 ± 15.9 years; neoplastic: 62 ± 13.5; infections: 62.5 ± 12.3; p = 0.52).

Graft failure that required initiation of RRT was the leading cause of graft loss (66.9%) with an overall median time to failure of 52 (RIQ: 43–61) months posttransplant. The median time to early loss (before 1st year post-transplant) was 1 (IQR: 0–3) month and to late loss (more than 1 year post-transplant) was 89 (IQR: 51–153) months.

Patients in whom the graft failed were younger than those who lost the graft due to death (47.2 vs. 55.3 years; p < 0.01). Moreover, in the former group, there were more women (42 vs. 30.8%; p < 0.01) and patients had fewer comorbidities: diabetes mellitus (DM) (14.1 vs. 20.1%; p = 0.04), ischemic heart disease (5 vs. 12.1%; p < 0.01), cerebrovascular accident (CVA) (3.5 vs. 6.7%; p = 0.05) and peripheral artery disease (PA) (21.1 vs. 35.1%; p < 0.01).

Regarding donor characteristics, grafts lost due to donor failure were from younger donors than those lost due to recipient death (48.7 vs. 52 years; p = 0.03). Grafts from donors with expanded criteria were more frequently lost due to recipient death (42.7 vs. 33%; p < 0.01). Cold ischemia time (CIT) was slightly longer in grafts lost due to graft failure relative to those lost due to death (17 vs. 16 h; p = 0.06) (Table 2).

During follow-up, 483 grafts failed, requiring a return to dialysis or re-transplantation. The most frequent cause of graft loss was chronic graft dysfunction without histological study (32.3%), followed by antibody-mediated rejection (15.9%) and IF/AT (12.6%). Losses due to T-cell-mediated rejection and glomerular disease (recurrence/de novo) were 7.9% and 5.6%, respectively. Vascular thrombosis accounted for 10.1% of losses, followed by medical-surgical complications (9.9%) and never functioning grafts (4.3%) (Fig. 2B).

During period 1 (years 1979–1999) and period 2 (years 2000–2019), 445 (29.2%) and 1,077 (70.8%) transplants were performed, respectively.

Of the transplants in period 1, 399 (89.7%) were lost. Of the total losses in that group, 300 (75.2%) were due to graft failure with return to dialysis/retransplantation and 99 (24.8%) to death of the recipient. Losses were more frequent in diabetic recipients (0 vs. 6.8%; p < 0.01) and in grafts from older donors (33.4 vs. 40.4 years; p = 0.05) and with worse renal function (creatinine at extraction 1.0 vs. 1.2 mg/dl; p < 0.01). In this period, CIT and delayed graft function (DFR) was similar between lost and functioning grafts.

Of the transplants performed during the second period, 323 (30%) grafts were lost during the follow-up period. Of the total losses, 183 (56.6%) were due to graft failure and 140 (43%) due to death of the recipient (data not shown). These grafts were from older donors (603 vs. 55.2 years; p = 0.02) and with SCD (58 vs. 48.1%; p < 0.01). Grafts from living donors were more frequent among functioning grafts (14.2 vs. 5.9%; p = 0.02). Moreover, losses occurred more frequently in older recipients (60.5 vs. 54.7 years; p < 0.01), diabetics (27.7 vs. 22.4%; p = 0.04), with more previous comorbidity (CI: 12.7 vs. 9.2%; p = 0.05 and AP: 48.8 vs. 24.1%; p < 0.01) and more re-RT (1.86 vs. 12.3%; p = 0.05). In this period, lost grafts had higher CIT (15 vs. 13 h; p < 0.01) and higher delayed graft function rate (53.8 vs. 27.3%; p < 0.01) (Table 3).

The analysis of the risk factors associated with graft loss globally show that only donor age (HR: 1.010; 95% CI: 1.00–1.01; p = 0.04) resulted as an independent risk factor. The time period of transplantation did not turn out to be a risk factor for loss in the adjusted analysis (Appendix BT Supplementary Table S1).

• 1)

  Losses due to death with functioning graft according to the transplantation period.

The causes of death of the recipient were similar in both transplantation periods. Only the number of patients dying from an unknown cause was lower in the current period (29.6 vs. 12.1%; p < 0.01) (Fig. 3A).

• 2)

  Graft loss with return to renal replacement therapy according to the transplantation period.

Fig. 3.

(A) Causes of recipient death according to transplantation period. (B) Causes of graft loss according to transplantation period.

(0.72MB).

The most frequent cause of graft loss in period 1 (years 1979−1999) was graft dysfunction without histological verification (41%), whereas in period 2 (years 2000−2019) it was vascular thrombosis (24.6%).

T-cell-mediated rejection was more frequent among lost grafts in period 1 (10 vs. 4.4%; p = 0.02), whereas never-functioning grafts were more frequent in period 2 (8.2 vs. 2%; p = 0.02). BK virus nephropathy was also higher in this period (0 vs. 2.2%). The remaining causes of graft loss were similar in both periods (Fig. 3B).

• 3)

  Causes of early and late loss (>1 year post-transplantation) according to transplantation period.

During period 1 (years 1979−1999), the most frequent cause of early loss (<1 year post-transplant) was T-cell-mediated rejection (31.3 vs. 6.5%; p < 0.01). However, in period 2 (years 2000−2019) vascular thrombosis was the most frequent reason for graft loss in the first year post-transplant (46.7 vs. 6.3%; p < 0.001).

The analysis of causes of late graft loss (1−3 years post-transplantation) showed differences in losses due to T-cell mediated rejection (16.7 vs. 2.8%; p = 0.04) which was more frequent in the older stages of the transplantation program. We found no differences in the causes of graft loss at years 3–5 post transplantation between both transplantation periods. Beyond 5 years post-transplantation, loss due to IF/TA was more frequent in period 2 (Table 4).

Univariate analysis of the 48 recipients who lost the transplant due to thrombosis showed that renal grafts came from donors (OR: 1.05; 95% CI: 1.03−1.07; p < 0.001) and older recipients (OR: 1.04; 95% CI: 1.02−1.06; p < 0.001). Thrombosis was more frequent in grafts from donors with HTN (OR: 2.51; 95% CI: 1.4−4.49; p < 0.001) and in controlled asystole donation (OR: 5.57; 95% CI: 2.59−11.9; p < 0.00).

However, in the multivariate analysis we did not find any independent risk factor associated with graft loss due to vascular thrombosis (Table 5).

Comparative analysis of graft survival and patient survival between both transplantation periods showed superior 10-year survival in grafts from the years 2000−2019 with respect to those from the years 1979−1999 (62 vs. 44%; p < 0.01). However, there was no difference in 10-year patient survival between the two periods (73 vs. 70%; p = 0.39) (Fig. 4A and B).

Fig. 4.

(A) Patient survival. (B) Graft survival. (C) Graft loss rates (attrition rate).

(0.54MB).

Analysis of loss rate showed a significantly higher loss rate during the first year that has improved over the last decades. However, loss rates are similar in the periods 1−3 years, 3−5 years and more than 5 years in all transplantation periods (Fig. 4C).