Heilman, Raymond

ABSTRACT Predicting delayed graft function (DGF) relies on donor, recipient, and perioperative factors. Despite the growing recognition that DGF duration strongly influences patient outcomes, no models currently address or predict its impact—highlighting an important gap in current research and clinical practice. This study aimed to develop an ensemble‐based machine learning model using perioperative data to predict DGF occurrence and duration. The gradient‐boosted decision trees model was trained and validated on 2725 patients, with k‐fold cross‐validation in an external cohort of 284 patients. Model performance was evaluated based on accuracy, ROC‐AUC, and other metrics using R and Python libraries. The binary DGF prediction model achieved a ROC‐AUC of 0.77, while the DGF duration classification model had an accuracy of 79.2%. DGF duration AUC values by time interval were 0.76 for 0–1 weeks, 0.81 for >1–2 weeks, 0.80 for >2–3 weeks, and 0.87 for >3 weeks; the overall macro‐averaged AUC was 0.81. The mean Brier score for multi‐class predictions was 0.14. External validation showed a 78% accuracy for DGF duration prediction. Acute kidney injury (AKI) and donor donation after circulatory death (DCD) status were key DGF predictors. The use of gradient‐boosted decision trees (GBDT) improves the prediction of both the likelihood and duration of DGF, addressing a current gap in kidney transplant patient care. By facilitating personalized transplant care, this model supports more effective perioperative planning and timely interventions, which may contribute to better patient outcomes.

ABSTRACTBackgroundDespite many people awaiting kidney transplant, kidney allografts from acute kidney injury (AKI) donors continue to be underutilized. We aimed to cluster kidney transplant recipients of AKI kidney allografts using an unsupervised machine learning (ML) approach.MethodsUsing Organ Procurement and Transplantation Network–United Network for Organ Sharing (OPTN/UNOS) data, a consensus clustering cohort analysis was performed in 12 356 deceased donor kidney transplant recipients between 2015 and 2019 in whom donors had a terminal serum creatinine ≥1.5 mg/dL. Significant cluster characteristics were determined, and outcomes were compared.ResultsThe median donor terminal creatinine was 2.2 (interquartile range [IQR] 1.7–3.3) mg/dL. Cluster analysis was performed on 12 356 AKI kidney recipients, and three clinically distinct clusters were identified. Young, sensitized kidney re‐transplant patients characterized Cluster 1. Cluster 2 was characterized by first‐time kidney transplant patients with hypertensive and diabetic kidney diseases. Older diabetic recipients characterized Cluster 3. Clusters 1 and 2 donors were young and met standard kidney donor profile index (KDPI) criteria; Cluster 3 donors were older, more likely to have hypertension or diabetes, and meet high KDPI criteria. Cluster 1 had a higher risk of acute rejection, 3‐year patient death, and graft failure. Cluster 3 had a higher risk of death‐censored graft failure, patient death, and graft failure at 1 and 3 years. Cluster 2 had the best patient‐, graft‐, and death‐censored graft survival at 1 and 3 years. Compared to non‐AKI kidney recipients, the AKI clusters showed a higher incidence of delayed graft function (DGF, AKI: 43.2%, 41.7%, 45.3% vs. non‐AKI: 25.5%); however, there were comparable long‐term outcomes specific to death‐censored graft survival (AKI: 93.6%, 93.4%, 90.4% vs. non‐AKI: 92.3%), patient survival (AKI: 89.1%, 93.2%, 84.2% vs. non‐AKI: 91.2%), and overall graft survival (AKI: 84.7%, 88.2%, 79.0% vs. non‐AKI: 86.0%).ConclusionsIn this unsupervised ML approach study, AKI recipient clusters demonstrated differing, but good clinical outcomes, suggesting opportunities for transplant centers to incrementally increase kidney utilization from AKI donors.

AbstractBackgroundUrinary Tract Infections are the most common post‐transplant infection and can have varied presentations. This study aimed to describe the outcomes of kidney transplant recipients with asymptomatic histologic pyelonephritis on allograft biopsy. Histologic Pyelonephritis was defined as neutrophil cast or neutrophilic tubulitis, interstitial infiltrates with predominant neutrophils, and no evidence of rejection or glomerulonephritis on biopsy.MethodsThe study included 123 kidney transplant recipients, of whom 95 underwent protocol biopsies, and 28 had biopsies for elevated creatinine within the first 2 years of a kidney transplant.ResultsThe mean age of the cohort was 55.3 years, with 52% females and 78% deceased donor transplants. The risk factors for asymptomatic histologic pyelonephritis were recipient female sex (OR 1.89, 1.3–2.7, diabetes mellitus (OR 2.479, 1.687–3.645), and deceased donation (OR 1.69, 1.098–2.63). The incidence of asymptomatic pyelonephritis on protocol biopsy was 1.7%, with 52% having positive urine cultures and Escherichia coli being the most common bacteria. Subjects with asymptomatic pyelonephritis had inferior graft survival compared to the matched cohort HR 1.88 (1.06–3.35), p = .0281. In addition, of these 123 subjects, 68 (55%) subsequently developed pyelonephritis, and 34 subjects had pyelonephritis within 6 months after this episode. Subjects with recurrent infections exhibited lower survival HR 2.86 (1.36–6.02) and a trend toward higher rejection risk.ConclusionAsymptomatic histologic pyelonephritis can occur in kidney transplant recipients and is associated with inferior graft survival.

AbstractIntroductionExpedited out‐of‐sequence deceased donor kidney allocation is a strategy to avoid discards after early placement attempts have been unsuccessful. Our study aimed to assess outcomes and characteristics of these transplanted kidneys.MethodsKDPI matching was performed between expedited allocation (EA) and standard allocation (SA) deceased donor kidney transplants performed at our center.ResultsBetween 2018 and 2021, there were 225 EA offers, and 189 (84%) were transplanted. EA recipients were older (p = .007) and had shorter dialysis vintage (p < .0001). EA kidneys were likely to be nationally allocated (p < .001), have AKI (p < .0001) and longer CIT (p < .0001). There were no differences in EA and SA time‐zero kidney biopsies (ci, p = .07; ct, p = .89; cv, p = .95; ah, p = .79). EA kidneys had more DGF (p = .0006), but there were no differences in DGF duration (p = .83), hospital length of stay (p = .43), 1‐ and 2‐year eGFR (p = .16, p = .99), patient (p = .34), or death‐censored graft (p = .66) survival.ConclusionDuring this study period, our center transplanted 189 kidneys through EA following local‐regional declines. These kidneys often came from AKI donors and had more DGF but had similar outcomes to KDPI‐matched SA kidneys. Although it has been suggested that EA has the potential to worsen transplant disparities, transplant center level decisions on organ acceptance contribute to these variations.

Abstract Introduction Significant center‐to‐center variation in attitudes and management of delayed graft function (DGF) remains common. Methods A survey to describe current DGF practices was developed by workgroup members sponsored by the National Kidney Foundation (NKF) and was distributed to both the NKF DGF workgroup members, kidney transplant program directors and the transplant community within the United States and Canada. Seventy‐one percent of NKF workgroup members completed the survey along with 70 unique the United States and three Canadian kidney transplant programs. All Organ Procurement and Transplantation Network (OPTN) regions were represented. Results DGF was reported to occur at rate of 20%–40% for most centers with 3.9% indicating their incidence to be >60%. Most centers reported longer hospital lengths of stay and more frequent outpatient visits. Despite the commonality of DGF, only half of centers reported having an established protocol to manage DGF. Kidney allograft biopsies were the only consistent DGF management strategy observed, although use of machine perfusion was also heavily favored. Other DGF management strategies voiced by a minority included having established outpatient practices to care for DGF patients and administering outpatient community‐based hemodialysis. Conclusion Although approximately a third of survey responders indicated that risk of DGF played a role in their willingness to accept organs, most did not feel that increased cost or clinical impact on outcomes was a deterrent. Future strategies, including broader sharing of best practices, redefining terminology specific to DGF, the establishment of DGF dialysis guidelines and improving access to machine perfusion across OPOs may help reduce discard and improve utilization of kidneys at risk for DGF.

AbstractDual kidney transplantation (DKT), utilizing two adult kidneys from the same donor for one recipient, has been used as a way to expand the available donor pool. These kidneys often come from high Kidney Donor Profile Index donors (KDPI > 85%). Data comparing outcomes between high KDPI DKT and single kidney transplants (SKT) remain limited. We assessed outcomes of 336 high KDPI kidney transplants performed at our center; 11.0% (n = 37) were DKT. Recipients of DKT were older (P = .02) and donors had a higher KDPI score (median 96% vs. 91%, P < .0001). DKT operative time was higher compared to SKT (+1.4 hours, P < .0001). There were no differences in delayed graft function (54.1% vs. 51.5%, P = .77) and hospital length of stay (median 4.0 vs. 3.0 days, P = .21) between DKT and SKT. Grade I Clavien‐Dindo complications occurred in 8.1% of DKT and 13.7% of SKT (P = .008). There were no grade IVa, IVb, or V complications in either group. DKT had more glomerulosclerosis (P = .04), interstitial fibrosis (P = .02), tubular atrophy (P = .01), and arterial thickening (P = .03) on 1‐year protocol biopsies. Estimated glomerular filtration was higher for DKT at 1‐ (P = .004) and 2‐years post‐transplant (P = .01). There were no differences in patient (HR 1.3, 95% CI .5‐3.3, P = .58) or graft (HR 1.1, 95% CI .5‐2.3, P = .83) survival. Good outcomes can be achieved with DKT using high KDPI kidneys with moderate chronic changes. DKT is a good option to help further utilize high KDPI kidneys and minimize discard.

Abstract Donation after circulatory death (DCD) liver transplantation (LT) outcomes have been attributed to multiple variables, including procurement surgeon recovery techniques. Outcomes of 196 DCD LTs at Mayo Clinic Arizona were analyzed based on graft recovery by a surgeon from our center (transplant procurement team [TPT]) versus a local procurement surgeon (non‐TPT [NTPT]). A standard recovery technique was used for all TPT livers. The recovery technique used by the NTPT was left to the discretion of that surgeon. A total of 129 (65.8%) grafts were recovered by our TPT, 67 (34.2%) by the NTPT. Recipient age (p = 0.43), Model for End‐Stage Liver Disease score (median 17 vs. 18; p = 0.22), and donor warm ischemia time (median 21.0 vs. 21.5; p = 0.86) were similar between the TPT and NTPT groups. NTPT livers had longer cold ischemia times (6.5 vs. 5.0 median hours; p < 0.001). Early allograft dysfunction (80.6% vs. 76.1%; p = 0.42) and primary nonfunction (0.8% vs. 0.0%; p = 0.47) were similar. Ischemic cholangiopathy (IC) treated with endoscopy occurred in 18.6% and 11.9% of TPT and NTPT grafts (p = 0.23). At last follow‐up, approximately half of those requiring endoscopy were undergoing a stent‐free trial (58.3% TPT; 50.0% NTPT; p = 0.68). IC requiring re‐LT in the first year occurred in 0.8% (n = 1) of TPT and 3.0% (n = 2) of NTPT grafts (p = 0.23). There were no differences in patient (hazard ratio [HR], 1.95; 95% confidence interval [CI], 0.76–5.03; p = 0.23) or graft (HR, 1.99; 95% CI, 0.98–4.09; p = 0.10) survival rates. Graft survival at 1 year was 91.5% for TPT grafts and 95.5% for NTPT grafts. Excellent outcomes can be achieved using NTPT for the recovery of DCD livers. There may be an opportunity to expand the use of DCD livers in the United States by increasing the use of NTPT.