Abstract
Background and Aims
Histological findings that meet criteria for borderline changes “suspicious” for acute T cell mediated rejection (BR) as defined by the Banff Classification (2017) are frequently seen. However, its clinical significance, as well as the appropriate clinical management, are still controversial. Our goal was to compare clinical outcomes of kidney transplant recipients with biopsy proven BR versus acute T cell mediated rejection (aTCMR) and the influence of treating BR rejection in graft outcomes.
Method
A retrospective cohort study was performed in all kidney transplant recipients with biopsy proven BR and aTCMR between January 2012 and December 2018. Data related to donor and recipient demographics, treatment and subsequent evolution of serum creatinine, proteinuria and graft survival were collected. Mean time at follow up was 31.2 ± 29.1 months.
Results
We included 91 patients with biopsy proven T cell rejection of which 34 (37.4%) had a BR and 57 (62.6%) aTCMR: 39 (68.4%) IA, 9 (15.8%) IB, 7 (28.1%) IIA and 2 (3.5%) IIB. There was no difference between groups (BR vs aTCMR) regarding age (45,5 vs 48,1, p=0,38), sex (male 73% vs 60%, p=0,27) or race (Caucasian 100% vs 93%, p=0.114). For both groups, deceased donor was more frequent (82% vs 95%, p=0.074), and there was no difference in cytotoxic PRA (mean 4.5 ± 9.2 vs 3.7 ± 12.8, p=0.762) or number of compatibilities (mean 2.2 ± 1.2 vs 2.4 ± 1.3, p=0.539).
At the time of rejection diagnosis, the mean time of transplant was similar between groups (32.9 ± 43.6 vs 42.3 ± 67.4 months, p=0.467), but estimated glomerular filtration rate (GFR) was significantly higher in patients with BR when compared to aTCMR (32.0 ± 22.5 vs 19.9 ± 13.1 ml/min/1.73m2, p=0.009). We found no significant difference in proteinuria at the time of biopsy between the 2 groups.
Treatment with steroids was started in 20 (58.8%) patients with BR and all the patients with aTCMR were treated with steroids with or without thymoglobulin, depending on the Banff class. Fourteen (41.2%) patients with BR were followed closely with no acute interventions.
At 1-year post biopsy, graft survival was 70%, and we found no significantly statistical difference between the two groups (79.4% vs 64.3%, p=0.129). In patients with preserved graft, there was no difference in GFR (41.9 ± 17.7 vs 37.7 ± 19.8, p=0.401) at 12 months post biopsy for both groups. When performing Kaplan-Meyer survival curves at follow-up, we also found no difference between BR and aTCMR (57.6 ± 7.1 vs 43.6 ± 5.5 months, p=0.157) (Figure 1).
When analyzing the BR group (N=34) and comparing the patients that were treated (N=20) versus the patients with conservative approach (N=14), we found no difference in demographic features, sCr at biopsy (3.0 ± 1.1 vs 2.8 ± 2.1 mg/dl, p=0.696) and time post-transplant (28.1 ± 43.2 vs 39.7 ± 44.8 months, p=0.454). Graft survival at 1-year was 80% for treated patients and 79% for non-treated patients, p=0.919 and GFR for patients with preserved graft was not different between groups (43.9 ± 21.0 vs 39.7 ± 13.5 ml/min/1.73m2, p=0.572). When performing survival curves, we found that treated patients had almost the double time with functioning graft compared to non-treated patients (71.9 ± 8.5 vs 41.3 ± 6.2 months, p=0.104), although not statistically different probably due to the small sample size (figure 2).
Conclusion
Our study showed that despite having better GFR at time of biopsy, patients with BR (overall and treated) did not present better graft survival nor graft function at 1 year post biopsy or at follow up, compared with aTCMR. We also found a tendency to better graft survival in patients with BR treated with steroids compared with conservative approach. These results reinforce the importance of borderline rejection in graft outcomes and that the decision of whether to treat or not can influence long-term outcomes.
Living Donor Kidney Transplantation Improves Graft and Recipient Survival in Patients with Multiple Kidney Transplants