Citations

Abstract Disclosure: B. Bahrani Fard: None. N. Nikravangolsefid: None. J. Maroun: None. A. Cross: None. T. Foster: None. T. McKenzie: None. W. Young Jr: None. B.M. Dy: None. M. Lyden: None. I. Bancos: None. Background: Giant pheochromocytomas (PHEO), defined based on size > 10 cm, are rare. Data on clinical presentation, management, and prognosis of giant PHEOs are scarce. Objectives: to determine prevalence of giant PHEO and compare presentation, management, and outcomes of patients with giant vs non-giant PHEOs. Methods: We conducted a single-center retrospective cohort study of consecutive patients with giant PHEO and randomly chosen patients with non-giant PHEO (referents) at 6:1 ratio, evaluated between 2000-2023. All variables were extracted from the electronic medical record. Results: Of 828 patients with PHEO evaluated between 2000 and 2023, 31 (3.7%) had giant PHEO (median size 12 cm, IQR 10.0-13.5). In comparison to referents (n=186, median size 4 cm, IQR, 2.9-5.0), patients with giant PHEO were similar in age (median age 53 vs 55 years, P=.55), more frequently discovered based on symptoms (52% vs 32%, P=.04), and less likely to be bilateral (3% vs 12%, P=.31). Genetic testing was performed in 103 (48%) of patients, without group differences, however, patients with giant PHEO were less likely to have an associated germline mutation detected (7% vs 47%, P=.004). Patients with giant PHEO had more symptoms of catecholamine excess (median of 2 vs 1, P=.04) and presented with a higher prevalence of severe metanephrine excess >10 times above upper normal range (76% vs 30%, P<.0001). Adrenalectomy was performed in 94% of patients with giant PHEOs and in 100% referents, with higher rates of open adrenalectomy in giant PHEO (86% vs 14%, P<.0001). In addition to preoperative alpha-adrenergic blockade (89%), metyrosine was used in 14 (7%) patients, mostly in patients with giant PHEO (26% vs 3%, P<.0001). Patients with giant PHEO had a longer hospital length of stay (laparoscopic: median 5 vs 2 days, P<.0001; open: median 6 vs 5 days, P=.24), and a higher complication rate (31% vs 10%, P=.006).Median follow up was 3 years (IQR 1-8) years, similar in patients and referents. Metastases developed at a much higher rate in patients with giant PHEOs (45% vs 4% in referents, P<.0001). Disease specific mortality was 7% in patients with giant PHEOs and 0% in referents. Conclusion: Patients with giant PHEO were more symptomatic, had a higher degree of catecholamine excess, and a higher rate of perioperative complications. Almost half of patients with giant PHEO developed metastases, warranting a close follow-up. Presentation: 6/2/2024

Abstract Disclosure: S. Paknikar: None. D. Toro Tobon: None. J. Durski: None. H. Alkhateeb: None. M.M. Ryder: None. Introduction: Radioactive iodine (RAI) therapy-induced myelosuppression is a dose-limiting toxicity for RAI avid metastatic differentiated thyroid cancers. We describe a patient with exceptionally RAI avid metastatic follicular thyroid cancer (FTC) who underwent autologous stem cell cryopreservation (ASCC) to enable high dose I-131 treatments. Case: A 60-year-old woman with T1bN0M0 FTC underwent two-stage thyroidectomy and three doses of I-131 between 2002-2007 (434 mCi cumulative activity) due to incomplete biochemical response. She presented to our institution in 2015 with rising thyroglobulin (Tg) levels and lung nodules that were intensely RAI avid. She enrolled in a clinical trial of selumetinib versus placebo (for 1 month) followed by 150 mci I-131. She had a profound response with an 80% reduction in Tg and 50% tumor shrinkage. The following year her Tg rose to 8381 ng/mL and she developed biopsy proven, well differentiated metastatic FTC in the hip and liver. Lenvatinib treatment resulted in partial tumor and Tg response but was halted due to grade 3/4 toxicities. Tumor genomic interrogation revealed NRAS, PIK3CA, and TSH-R activating mutations, providing a causal explanation of her intensely RAI avid and responsive disease. Repeat I-131 therapy was facilitated by GM-CSF stimulated ASCC (8.57 x10(6) CD34+ cells/kg for two cycles) as a tool to help mitigate future potential RAI-induced myelosuppression. The patient received 295 mci I-131 followed by a 98% decrease in Tg and 20-25% shrinkage of liver metastases. Transient pancytopenia ensued but did not require autologous stem cell transplantation (ASCT). Due to insurance issues, the patient was lost to follow-up for two years. She returned with cachexia from severe, untreated thyrotoxicosis due to widespread functional, thyroxine producing metastatic disease. Anti-thyroid drug therapy (ATD) restored euthyroid status with significant clinical improvement. A non-stimulated I-123 whole body scan showed diffuse RAI avid disease with 88% whole body radioiodine retention at 48h. She received 125 mci I-131 with initial tumor response but with progression of bone marrow, calvaria, and retinal metastases. Six months later, a final dose of 253 mCi I-131 (cumulative activity 1,370 mCi) resulted in mixed tumor responses. Unfortunately, resection of a RAI refractory calvaria lesion postoperatively resulted in serious complications. ASCT and salvage treatments could not be pursued due to her impaired functional status. The patient pursued hospice care and passed away 20 years after her diagnosis. Conclusion: We characterize a patient with NRAS, PIK3CA and TSH-R mutant metastatic, thyroxine-producing FTC whose recurrences over 20 years retained intense RAI avidity and response to I-131 therapy. In cases with potential benefit from multiple high dose RAI therapies, ASCT may be a tool to mitigate I-131-induced myelosuppression. Presentation: 6/1/2024

Abstract Disclosure: S. Mohan: None. O.A. Shariq: None. J.C. Andrews: None. W.F. Young: None. Background: Adrenal vein sampling (AVS) is performed in patients with primary aldosteronism (PA) to subtype surgically curable unilateral disease. During AVS, iodinated contrast is used to confirm catheterization of adrenal veins for accurate sampling. However, in patients with severe contrast allergy, the optimal method of safely and effectively confirming localization is debatable. Premedication with oral corticosteroids has raised concern for suppression of cortisol and confounding of AVS results. We therefore investigated our experience of AVS after corticosteroid premedication in patients with PA and iodine contrast allergy. Methods: Patients with PA who underwent AVS at our center under continuous cosyntropin infusion (50 mcg/hr started 30 minutes before sampling) between September 1990 - October 2023 were identified. Individuals with documented allergy to iodinated contrast who received corticosteroid premedication were matched 1:1 for age and sex with control patients without contrast allergy or premedication. Baseline demographics, biochemical variables, AVS results, and follow-up data were compared between groups. Results: 1273 AVS studies were performed in 1255 patients with PA and 35 (2.8%) patients received corticosteroid premedication for contrast allergy. Methylprednisolone was used in 32 (91.4%) patients while dexamethasone was used in the rest. No patients experienced a severe breakthrough allergic reaction. Mean age was similar in cases vs. controls (53.9 vs. 54.1 years, P=0.92) with 16 females (45.7%) per group. There were no significant differences in baseline aldosterone-to-renin ratios (57.7 vs. 49.3, P=0.57) or daily defined doses of antihypertensives (5.5 vs. 4.4, P=0.21) between the two groups. Absolute cortisol levels (mcg/dL) from both adrenal veins (AV) and the inferior vena cava (IVC) were similar between cases and controls (right AV: 752.5 vs. 813.7, P=0.48; left AV: 527.2 vs. 462.7, P=0.33; IVC: 23.1 vs. 24.5, P=0.18). The ratio of cortisol in the corresponding AV to the IVC (selectivity index) was >5 in all patients, confirming appropriate cannulation, with similar mean selectivity indices between groups (right AV: 33.0 vs. 33.7, P=0.86; left AV: 22.8 vs. 19.0 P=0.17). Unilateral disease (indicated by a lateralization index >4) was present in 20 (57.1%) patients who received premedication compared with 16 (45.7%) controls (P=0.34). Surgery for unilateral disease was performed in 16 cases and 12 controls, with similar post-op day 1 aldosterone levels in both groups (3.8 vs. 3.3, P=0.36), consistent with surgical cure. Conclusion: In patients with PA and iodine contrast allergy, premedication with corticosteroids prior to cosyntropin-stimulated AVS does not significantly suppress cortisol levels or affect interpretation of AVS results, with comparable rates of disease subtyping and surgical outcomes when compared to a control group. Presentation: 6/3/2024

Abstract Disclosure: M.D. Figueiredo: None. J. Wagoner: None. W.F. Young: None. L.M. Fishbein: None. M.T. McDermott: None. Introduction: Primary aldosteronism (PA) accounts for at least 5% of hypertension cases and up to 20% of resistant hypertension. Underdiagnosis leads to delayed treatment and increases cardiovascular morbidity and mortality. Our objective was to assess the impact of an educational series for primary care providers (PCPs) on increasing screening rates for PA. Methods: Educational intervention consisted of a 5-day, 2-minute email series for 111 PCPs on PA screening, confirmation, and treatment (trained group). Following the training, EMR data were extracted for patients with PAC (plasma aldosterone concentration) or PRA (plasma renin activity), new hypertension diagnosis code, or a new MRA prescription within 1 year pre- or post-intervention and data pre- and post- intervention were compared. A proper screen included sending both PAC and PRA simultaneously. IRB approval was obtained. Results: In the pre-intervention year, 57 patients were properly screened for PA by the 111 PCPs. Post-intervention, 151 patients were screened for PA showing a 2.65-fold increase in screening with the educational intervention. Using a paired T-test across quartiles pre- and post-intervention, this increase was statistically significant (P=0.0148). We further demonstrated a sustained effect throughout the 12 months post-intervention per quartile (Q+1 n=29, Q+2 n=45, Q+3 n=33, Q+4 n=42). Of the total cohort screened (both pre- and post-intervention), 28% had a positive screen for PA (defined liberally as PAC >5 ng/dL with PRA <1 ng/mL/h). Discussion: The educational intervention drove an increase in PCP screening for PA in the year post-intervention and this increase was sustained across all quartiles of the year. Of the screened cohort (pre- and post-intervention), 28% screened positive for PA, and those patients were more frequently referred to specialists post-intervention. Our study has some limitations, including the relatively small sample size of PCPs educated and patients screened, and the 2020 start to the COVID-19 pandemic impact during the post-intervention period. Conclusion: Our findings advocate for continued education to reinforce PA screening criteria using an easy, not time consuming and effective educational tool. Continuous education is crucial to reinforce next steps in confirmation and treatment. Presentation: 6/3/2024

Abstract Disclosure: M. Salama: None. A.N. Lteif: None. Introduction: Intravenous propofol is widely employed for procedural sedation, monitored anesthesia care, and induction of general anesthesia. It can be administered as a bolus or an infusion. While extensively used, reports in adults and animal studies suggest potential implications of propofol on adrenal steroidogenesis. In this study, we present findings detailing transient adrenal suppression in two pediatric patients following propofol infusion. Case: The first case involves a 5-year-old boy diagnosed with left cerebellar high-grade glioma, having undergone tumor resection and proton beam irradiation. Morning cortisol levels measured twice post-propofol sedation (20 mg bolus followed by infusion at a rate of 275 mcg/kg/min prior to radiation therapy were 3.1 and 1.2 mcg/dl (ref range: 2.4-15 mcg/dl), accompanied by an undetectable ACTH level. Subsequent non sedated cortisol testing done four days later revealed a robust reading of 24 mcg/dl, with an ACTH level of 46 pg/ml.The second case is a 4-year-old boy diagnosed with right cerebellar high-grade astrocytoma glioma, who underwent total resection two months before his initial cortisol assessment. Following proton beam irradiation and propofol sedation (20 mg bolus followed by propofol infusion at a rate of 300 mcg/kg/min) for therapy, his morning cortisol obtained 45 minutes after starting propofol infusion was low at 2 mcg/dL (ref range: 2.4-15 mcg/dl). A repeat cortisol test eight days later showed an increase to 5.6 mcg/dL. Notably, neither patient was on exogenous steroids or medications impacting the adrenal axis. Discussion: The transient decline in cortisol and ACTH observed in these cases gains significance as the patients were not on any medication known to potentially induce adrenal suppression. Additionally, the gliomas were not parasellar in location and radiation had just begun. These findings align with prior reports linking propofol administration with the suppression of the hypothalamic-pituitary-adrenal axis. The impact of propofol on adrenal function needs to be considered especially in pediatric patients undergoing procedures or treatments. Conclusion: Our cases shed light on the potential transient influence of propofol on adrenal function in pediatric patients. Further studies are warranted to comprehensively evaluate the effects of propofol injections and infusions on adrenal function in pediatric cohorts. Presentation: 6/3/2024

Abstract Disclosure: A.Y. Chang: None. D.Z. Erickson: None. W.F. Young: None. K. Grennan: None. J.C. Andrews: None. E. Kapoor: None. Background: While mild androgen excess is frequently encountered in endocrine practices, severe symptoms, rapid onset or markedly abnormal biochemical testing lead to further evaluation for rare ovarian androgen-producing tumors, ovarian hyperthecosis or androgen-secreting adrenal tumors. Combined ovarian and adrenal vein sampling (COAVS) might be necessary to identify the source of androgen hypersecretion and guide treatment. The utility of COAVS has been studied in limited case series, combining case reports and different centers. There is no consensus regarding the testosterone gradient that localizes the androgen source or the added value of COAVS when a pathologic source of androgen excess is suspected. The additional clinical benefit of adrenal vein sampling (AVS) when combined with ovarian vein sampling (OVS) has also not been examined. Objectives: We sought to determine: a) the optimal total testosterone (TT) gradient for identifying unilateral vs bilateral source of hyperandrogenism and b) the utility of AVS in addition to OVS. Methods: We performed a retrospective chart review of all adult women who had COAVS at Mayo Clinic, Rochester from 1996 to 2023. Pathologic sources of androgen excess were defined as ovarian hyperthecosis (OH) as well as ovarian and adrenal tumors. We calculated the ratio of dominant OV or AV divided by the nondominant OV or AV. Results: 22 COAVS procedures from 7 premenopausal and 15 postmenopausal women were included in the study: 12 (55%) women had adrenal abnormalities seen on computed cross sectional imaging and 5 (23%) had unilateral ovarian imaging abnormalities. Of the 22 participants in the case series, successful 4-vessel catheterization was completed in 20 (1 right ovarian vein could not be cannulated, and 1 had bilateral oophorectomy). Nine (41%) women had unilateral and 10 (45%) bilateral ovarian androgen hypersecretion. Eleven (50%) women had pathologic causes for elevated TT (5 ovarian tumors, 6 cases of OH), 5 lacked follow-up pathology and 5 had benign pathology. AVS did not diagnose an adrenal source of androgen hypersecretion. TT concentrations were significantly greater in patients with unilateral compared to those with bilateral ovarian androgen hypersecretion (median 262.5 ng/dL [245- 401.5 IQR] vs 104 ng/dL [82-190], P = 0.01). In addition, TT concentrations were higher in those women with a pathologic source of TT hypersecretion compared to those with benign ovaries (349 ng/dL [251-538] vs 115 ng/dL [93-195], P = 0.03). All cases of unilateral ovarian disease had a dominant OV gradient of TT ≥ 2.1 compared to the nondominant OV. Conclusion: In a series of successful COAVS in women with androgen hypersecretion, an ovarian side-to-side gradient of ≥ 2.1 diagnosed all sources of unilateral ovarian pathology. AVS did not diagnose any pathologic source of adrenal androgen excess in this patient cohort. Presentation: 6/3/2024

Abstract Disclosure: A.Y. Chang: None. D.Z. Erickson: None. W.F. Young: None. J.C. Andrews: None. E. Kapoor: None. Background: While mild androgen excess is frequently encountered in endocrine practices, severe symptoms, rapid onset or markedly abnormal biochemical testing lead to further evaluation for rare ovarian androgen-producing tumors, ovarian hyperthecosis or androgen-secreting adrenal tumors. Combined ovarian and adrenal vein sampling (COAVS) might be necessary to identify the source of androgen hypersecretion and guide treatment. The utility of COAVS has been studied in limited case series, combining case reports and different centers. There is no consensus regarding the testosterone gradient that localizes the androgen source or the added value of COAVS when a pathologic source of androgen excess is suspected. The additional clinical benefit of adrenal vein sampling (AVS) when combined with ovarian vein sampling (OVS) has also not been examined. Objectives: We sought to determine: a) the optimal total testosterone (TT) gradient for identifying unilateral vs bilateral source of hyperandrogenism and b) the utility of AVS in addition to OVS. Methods: We performed a retrospective chart review of all adult women who had COAVS at Mayo Clinic, Rochester from 1996 to 2023. Pathologic sources of androgen excess were defined as ovarian hyperthecosis (OH) as well as ovarian and adrenal tumors. We calculated the ratio of dominant OV or AV divided by the nondominant OV or AV. Results: 22 COAVS procedures from 7 premenopausal and 15 postmenopausal women were included in the study: 12 (55%) women had adrenal abnormalities seen on computed cross sectional imaging and 5 (23%) had unilateral ovarian imaging abnormalities. Of the 22 participants in the case series, successful 4-vessel catheterization was completed in 20 (1 right ovarian vein could not be cannulated, and 1 had bilateral oophorectomy). Nine (41%) women had unilateral and 10 (45%) bilateral ovarian androgen hypersecretion. Eleven (50%) women had pathologic causes for elevated TT (5 ovarian tumors, 6 cases of OH), 5 lacked follow-up pathology and 5 had benign pathology. AVS did not diagnose an adrenal source of androgen hypersecretion. TT concentrations were significantly greater in patients with unilateral compared to those with bilateral ovarian androgen hypersecretion (median 262.5 ng/dL [245- 401.5 IQR] vs 104 ng/dL [82-190], P = 0.01). In addition, TT concentrations were higher in those women with a pathologic source of TT hypersecretion compared to those with benign ovaries (349 ng/dL [251-538] vs 115 ng/dL [93-195], P = 0.03). All cases of unilateral ovarian disease had a dominant OV gradient of TT ≥ 2.1 compared to the nondominant OV. Conclusion: In a series of successful COAVS in women with androgen hypersecretion, an ovarian side-to-side gradient of ≥ 2.1 diagnosed all sources of unilateral ovarian pathology. AVS did not diagnose any pathologic source of adrenal androgen excess in this patient cohort. Presentation: 6/3/2024

Abstract Disclosure: J. Saini: None. R. Nathani: None. J. Larson: None. V. Fell: None. I. Bancos: None. Background: Surgical menopause due to bilateral salpingo-oophorectomy (BSO) at a younger age is associated with an increased risk of cardiovascular disease, cognitive decline, and increased mortality. Aging is associated with abnormalities in steroid metabolome, most notably an increased glucocorticoid to androgen ratio (“adrenal aging”). We hypothesized that in postmenopausal women of the same age, the duration of menopause is associated with adrenal aging. Objective: To determine the differences in the adrenal steroid metabolome of women with a history of BSO as compared to age-matched women following natural menopause. Methods: We conducted a single-center cross-sectional study in postmenopausal women with or without a history of BSO, between December 2020 to December 2022 Each participant collected 24-hour urine, which was analyzed for steroid profiling by liquid chromatography-tandem mass spectrometry (25 steroids). Exclusion criteria were inability to collect a 24h urine sample, exogenous glucocorticoid use, advanced malignancy, and chronic opioid use. Primary endpoints were glucocorticoid/androgen ratio, 11β-hydroxysteroid dehydrogenase type 2 activity (cortisol/cortisone, Tetrahydrocortisol (THF) + 5 α-tetrahydrocortisol (5 α-THF)/ Tetrahydrocortisone, THE) and 5 α-reductase type activity (Tetrahydrocortisol (THF)/5 α-tetrahydrocortisol (5 α-THF) and Etiocholanolone/Androsterone, Et/An). Results: A total of 245 women were enrolled: 124 women (median age of 66 years, IQR 63-72) with a history of BSO and 121 women (median age of 68 years, IQR 64-71) who underwent natural menopause (P=0.214). Median age at menopause was 44 years (IQR 41-46) in the BSO group, and 50 years (IQR 48-53) in the natural menopause group. History of hormone replacement therapy with estrogen was reported in 161 (66%) women, more frequently (92% vs 39%, P<0.001) and for a longer duration (median 12 years (IQR 6-16) vs 6 years (IQR 3-9), P<0.001) in in the BSO group, as compared to the natural menopause group. Although both groups had similar levels of urinary androgens (median 1387 vs 1686 mcg/24h, P=0.063) and total glucocorticoids (median 10877 vs 10277 mcg/24h, P=0.354), women with a history of BSO demonstrated a higher glucocorticoid/androgen ratio (median 7.9 (IQR 5.0-11.1) vs 6.7 (4.3-9.9), P=0.014), and a lower Et/An ratio (median 1.4 vs 1.6, P=0.045), but not THF/5aTHF ratio (median 1.7 vs 1.8, P=0.380) in comparison to women with natural menopause. No differences were found in other ratios. Conclusion: Women with a history of BSO demonstrated a higher glucocorticoid/androgen ratio in comparison to women undergoing natural menopause, suggestive of BSO-induced accelerated “adrenal aging”. Future studies need to examine associations between the abnormal steroid metabolome and increased cardiovascular morbidity, frailty, and mortality reported following BSO. Presentation: 6/3/2024

ABSTRACT Introduction: ASPS is a rare, soft tissue tumor which represents 0.2-1 % of soft tissue sarcomas. A characteristic nonreciprocal translocation is found in nearly all cases. The resulting ASPSCR1-TFE3 fusion gene is postulated to be involved in the pathogenesis of this tumor. Case report: 29 years-old female presented with abdominal pain for few months. MRI and CT revealed a large necrotic abdominopelvic mass, which is extra-uterine, extra-ovarian, and peritoneal nodules. Exploratory laparotomy revealed a large mass arising from the distal part of the descending colon involving the mesentery and the surrounding peritoneum. She underwent left colectomy with primary anastomosis, and resection of the involved adjacent peritoneal and mesenteric surface nodules. Histopathology revealed ASPS involving the outer serosal layer of the large bowel wall negative margins. It also involved the resected mesenteric fat, greater omentum, and peritoneum. No further treatment was received as the patient traveled to home country. Discussion: The treatment of early-stage ASPS involves radical surgical excision of the primary tumor. Traditionally, patients with advanced ASPS have limited therapeutic options because of the resistance of the tumor to conventional cytotoxic chemotherapy, however, it is a targetable sarcoma. The response rate of metastatic ASPS to targeted therapy is clinically meaningful, and comparable to that of first-line chemotherapy. Most patients who received targeted therapy achieved disease control. Patients who had refractory or progressive disease to one targeted agent demonstrated a response to other agents. More randomized trials are warranted to expand treatment options and compare them to standard care regimens. Conclusion: In conclusion, while radical surgical excision remains crucial for early-stage ASPS, the landscape for advanced cases is evolving with targeted therapies showing promising results. However, further randomized trials are necessary to broaden the treatment options available and to evaluate these new therapies against existing standard care regimens. This ongoing research is vital to improving outcomes for patients with ASPS and tailoring more effective treatment strategies. Highlights

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.