Search

BACKGROUND Several groups have previously developed logistic regression models for predicting delayed graft function (DGF). In this study, we used an automated machine learning (ML) modeling pipeline to generate and optimize DGF prediction models en masse. METHODS Deceased donor renal transplants at our institution from 2010 to 2018 were included. Input data consisted of 21 donor features from United Network for Organ Sharing. A training set composed of ~50%/50% split in DGF-positive and DGF-negative cases was used to generate 400 869 models. Each model was based on 1 of 7 ML algorithms (gradient boosting machine, k-nearest neighbor, logistic regression, neural network, naive Bayes, random forest, support vector machine) with various combinations of feature sets and hyperparameter values. Performance of each model was based on a separate secondary test dataset and assessed by common statistical metrics. RESULTS The best performing models were based on neural network algorithms, with the highest area under the receiver operating characteristic curve of 0.7595. This model used 10 out of the original 21 donor features, including age, height, weight, ethnicity, serum creatinine, blood urea nitrogen, hypertension history, donation after cardiac death status, cause of death, and cold ischemia time. With the same donor data, the highest area under the receiver operating characteristic curve for logistic regression models was 0.7484, using all donor features. CONCLUSIONS Our automated en masse ML modeling approach was able to rapidly generate ML models for DGF prediction. The performance of the ML models was comparable with classic logistic regression models.

Background The vast majority of polyomavirus nephropathy (PVN) is due to BK virus, but rare cases result from JC virus reactivation. To date, only a handful of biopsy-proven JC-PVN cases have been reported. Here, we describe the clinical and pathologic findings in 7 patients with biopsy-proven JC-PVN. Methods Search of the pathology archives at 2 institutions found 7 cases of JC-PVN. Clinical data were extracted from the electronic medical records, and the biopsies were reviewed. Results Four cases were diagnosed at 6 y posttransplant or later. The remaining 3 cases presented within approximately 2 y posttransplant, of which 2 showed subclinical JC-PVN on surveillance biopsy. Two early presenting patients were treated for acute rejection just before acquiring JC-PVN. Late presenting patients had higher chronicity, which correlated to worse outcome. All but 1 biopsy showed nonspecific inflammation within areas of interstitial fibrosis without significant inflammation in unscarred cortex. The earliest presenting patient was the exception and showed active inflammation with tubulitis. Viral cytopathic changes were detected in all cases with moderate or high-histologic viral load (pvl), showing preference for the distal tubules and medulla. The 2 cases with low pvl did not demonstrate cytopathic changes but were SV40 positive. Conclusions JC-PVN can be insidious in presentation, which may cause delayed or missed diagnosis. Unlike BK-PVN, which typically occurs early in the posttransplant period, JC-PVN can occur both early and late following transplant. Overreliance on negative plasma and urine BK viral loads to exclude PVN can be a pitfall.

International audience; BACKGROUND: Treatment decisions in kidney transplantation requires patients and clinicians to weigh the benefits and harms of a broad range of medical and surgical interventions, but the heterogeneity and lack of patient-relevant outcomes across trials in transplantation makes these trade-offs uncertain, thus, the need for a core outcome set that reflects stakeholder priorities. METHODS: We convened 2 international Standardized Outcomes in Nephrology-Kidney Transplantation stakeholder consensus workshops in Boston (17 patients/caregivers; 52 health professionals) and Hong Kong (10 patients/caregivers; 45 health professionals). In facilitated breakout groups, participants discussed the development and implementation of core outcome domains for trials in kidney transplantation. RESULTS: Seven themes were identified. Reinforcing the paramount importance of graft outcomes encompassed the prevailing dread of dialysis, distilling the meaning of graft function, and acknowledging the terrifying and ambiguous terminology of rejection. Reflecting critical trade-offs between graft health and medical comorbidities was fundamental. Contextualizing mortality explained discrepancies in the prioritization of death among stakeholders-inevitability of death (patients), preventing premature death (clinicians), and ensuring safety (regulators). Imperative to capture patient-reported outcomes was driven by making explicit patient priorities, fulfilling regulatory requirements, and addressing life participation. Specificity to transplant; feasibility and pragmatism (long-term impacts and responsiveness to interventions); and recognizing gradients of severity within outcome domains were raised as considerations. CONCLUSIONS: Stakeholders support the inclusion of graft health, mortality, cardiovascular disease, infection, cancer, and patient-reported outcomes (ie, life participation) in a core outcomes set. Addressing ambiguous terminology and feasibility is needed in establishing these core outcome domains for trials in kidney transplantation.

Background Limited data are available on whether donor kidneys with diffuse glomerular fibrin thrombi (GFT) are safe to use. In this study, the clinicopathologic characteristics of allografts with diffuse donor-derived GFT were examined. Methods All deceased donor kidney transplant implantation biopsies from our institution between July 2011 and February 2018 with diffuse GFT were included. A control group for comparison consisted of all cases with implantation biopsies obtained during the study period without diffuse GFT. Clinical data were extracted from electronic medical records for all study patients, including donor information. Results Twenty-four recipients received kidneys with diffuse GFT from 16 deceased donors. All donors died from severe head trauma. On average, 79% of glomeruli contained fibrin thrombi. Nineteen cases had subsequent biopsy; all revealed resolution of GFT. Compared with the control group, kidneys with diffuse GFT had longer cold ischemia time (34 versus 27 h), were more frequently pumped using machine perfusion (100% versus 81%), and recipients experienced a higher frequency of delayed graft function (58% versus 27%). Only 2 grafts with diffuse GFT failed within the first year. Overall graft survival was similar between the diffuse GFT group and control group. Conclusions Deceased donor kidneys with diffuse GFT appear to be safe to use given that nearly 92% of recipients in this cohort who received such allografts experienced good clinical outcomes. Histologically, GFT demonstrated rapid resolution following transplantation. Interestingly, diffuse GFT only occurred in donors who suffered severe head trauma in this cohort, which may be a predisposing factor.