Comparison of artificial intelligence and human-based prediction and stratification of the risk of long-term kidney allograft failure
BACKGROUND:Clinical decisions are mainly driven by the ability of physicians to apply risk stratification to patients. However, this task is difficult as it requires complex integration of numerous parameters and is impacted by patient heterogeneity. We sought to evaluate the ability of transplant physicians to predict the risk of long-term allograft failure and compare them to a validated artificial intelligence (AI) prediction algorithm. METHODS:We randomly selected 400 kidney transplant recipients from a qualified dataset of 4000 patients. For each patient, 44 features routinely collected during the first-year post-transplant were compiled in an electronic health record (EHR). We enrolled 9 transplant physicians at various career stages. At 1-year post-transplant, they blindly predicted the long-term graft survival with probabilities for each patient. Their predictions were compared with those of a validated prediction system (iBox). We assessed the determinants of each physician's prediction using a random forest survival model. RESULTS:Among the 400 patients included, 84 graft failures occurred at 7 years post-evaluation. The iBox system demonstrates the best predictive performance with a discrimination of 0.79 and a median calibration error of 5.79%, while physicians tend to overestimate the risk of graft failure. Physicians' risk predictions show wide heterogeneity with a moderate intraclass correlation of 0.58. The determinants of physicians' prediction are disparate, with poor agreement regardless of their clinical experience. CONCLUSIONS:This study shows the overall limited performance and consistency of physicians to predict the risk of long-term graft failure, demonstrated by the superior performances of the iBox. This study supports the use of a companion tool to help physicians in their prognostic judgement and decision-making in clinical care.
A Randomized Double-Blinded Placebo Controlled Trial of Clazakizumab for the Treatment of COVID-19 Pneumonia With Hyperinflammation
OBJECTIVES/OBJECTIVE:We designed this study to test whether clazakizumab, a direct interleukin-6 inhibitor, benefits patients hospitalized with severe or critical COVID-19 disease accompanied by hyperinflammation. DESIGN/METHODS:Multicenter, randomized, double-blinded, placebo-controlled, seamless phase II/III trial. SETTING/METHODS:Five U.S. medical centers. PATIENTS/METHODS:Adults inpatients with severe COVID-19 disease and hyperinflammation. INTERVENTIONS/METHODS:Eighty-one patients enrolled in phase II, randomized 1:1:1 to low-dose (12.5â€‰mg) or high-dose (25â€‰mg) clazakizumab or placebo. Ninety-seven patients enrolled in phase III, randomized 1:1 to high-dose clazakizumab or placebo. MEASUREMENTS AND MAIN RESULTS/RESULTS:The primary outcome was 28-day ventilator-free survival. Secondary outcomes included overall survival ,frequency and duration of intubation, and frequency and duration of ICU admission. Per Data Safety and Monitoring Board recommendations, additional secondary outcomes describing clinical status and status changes, as measured by an ordinal scale, were added. Bayesian cumulative proportional odds, logistic, and Poisson regression models were used. The low-dose arm was dropped when the phase II study suggested superiority of the high-dose arm. We report on 152 patients, 74 randomized to placebo and 78 to high-dose clazakizumab. Patients receiving clazakizumab had greater odds of 28-day ventilator-free survival (odds ratio [OR] = 3.84; p [OR > 1] 99.9%), as well as overall survival at 28 and 60 days (OR = 1.75; p [OR > 1] 86.5% and OR = 2.53; p [OR > 1] 97.7%). Clazakizumab was associated with lower odds of intubation (OR = 0.2; p [OR] < 1; 99.9%) and ICU admission (OR = 0.26; p [OR < 1] 99.6%); shorter durations of ventilation and ICU stay (risk ratio [RR] < 0.75; p [RR < 1] > 99% for both); and greater odds of improved clinical status at 14, 28, and 60 days (OR = 2.32, p [OR > 1] 98.1%; OR = 3.36, p [OR > 1] 99.6%; and OR = 3.52, p [OR > 1] 99.8%, respectively). CONCLUSIONS:Clazakizumab significantly improved 28-day ventilator-free survival, 28- and 60-day overall survival, as well as clinical outcomes in hospitalized patients with COVID-19 and hyperinflammation.
Early experience with donation after circulatory death heart transplantation using normothermic regional perfusion in the United States
OBJECTIVE:This pilot study sought to evaluate the feasibility of our donation after circulatory death (DCD) heart transplantation protocol using cardiopulmonary bypass (CPB) for normothermic regional reperfusion (NRP). METHODS:Suitable local DCD candidates were transferred to our institution. Life support was withdrawn in the operating room (OR). On declaration of circulatory death, sternotomy was performed, and the aortic arch vessels were ligated. CPB was initiated with left ventricular venting. The heart was reperfused, with correction of any metabolic abnormalities. CPB was weaned, and cardiac function was assessed at 30-minute intervals. If accepted, the heart was procured with cold preservation and transplanted into recipients in a nearby OR. RESULTS:Between January 2020 and January 2021, a total of 8 DCD heart transplants were performed: 6 isolated hearts, 1 heart-lung, and 1 combined heart and kidney. All donor hearts were successfully resuscitated and weaned from CPB without inotropic support. Average lactate and potassium levels decreased from 9.39Â Â±Â 1.47Â mmol/L to 7.20Â Â±Â 0.13Â mmol/L and 7.49Â Â±Â 1.32Â mmol/L to 4.36Â Â±Â 0.67Â mmol/L, respectively. Post-transplantation, the heart-lung transplant recipient required venoarterial extracorporeal membrane oxygenation for primary lung graft dysfunction but was decannulated on postoperative day 3 and recovered uneventfully. All other recipients required minimal inotropic support without mechanical circulatory support. Survival was 100% with a median follow-up of 304Â days (interquartile range, 105-371Â days). CONCLUSIONS:DCD heart transplantation outcomes have been excellent. Our DCD protocol is adoptable for more widespread use and will increase donor heart availability in the United States.
Next steps for the xenotransplantation of pig organs into humans
Regarding normothermic regional perfusion: Arguing by insistence is not a strong argument [Letter]
Results of Two Cases of Pig-to-Human Kidney Xenotransplantation [Case Report]
BACKGROUND:Xenografts from genetically modified pigs have become one of the most promising solutions to the dearth of human organs available for transplantation. The challenge in this model has been hyperacute rejection. To avoid this, pigs have been bred with a knockout of the alpha-1,3-galactosyltransferase gene and with subcapsular autologous thymic tissue. METHODS:We transplanted kidneys from these genetically modified pigs into two brain-dead human recipients whose circulatory and respiratory activity was maintained on ventilators for the duration of the study. We performed serial biopsies and monitored the urine output and kinetic estimated glomerular filtration rate (eGFR) to assess renal function and xenograft rejection. RESULTS:in Recipient 2. In both recipients, the creatinine level, which had been at a steady state, decreased after implantation of the xenograft, from 1.97 to 0.82 mg per deciliter in Recipient 1 and from 1.10 to 0.57 mg per deciliter in Recipient 2. The transplanted kidneys remained pink and well-perfused, continuing to make urine throughout the study. Biopsies that were performed at 6, 24, 48, and 54 hours revealed no signs of hyperacute or antibody-mediated rejection. Hourly urine output with the xenograft was more than double the output with the native kidneys. CONCLUSIONS:Genetically modified kidney xenografts from pigs remained viable and functioning in brain-dead human recipients for 54 hours, without signs of hyperacute rejection. (Funded by Lung Biotechnology.).
Response to American College of Physician's statement on the ethics of transplant after normothermic regional perfusion
This paper responds to the position statement released by the American College of Physicians (ACP) entitled "Ethics, Determination of Death, and Organ Transplantation in Normothermic Regional Perfusion (NRP) with Controlled Donation after Circulatory Determination of Death (cDCD): American College of Physicians Statement of Concern." The ACP's statement engages with critical ethical issues surrounding cDCD NRP, but several of their conclusions are flawed. Contrary to the statement, the practice respects the dead donor rule and the legal definition of death while honoring the wishes of the deceased and their loved ones to help save the lives of those in need of organ transplants. cDCD NRP is well established in many countries, it can enhance trust in medical practice and organ donation, and will increase the availability of optimal organs for life-saving transplants.
Immunogenicity after heterologous third dose COVID-19 vaccination in a heart transplant recipient [Letter]
First Report of Xenotransplantation from a Pig to Human Recipient [Meeting Abstract]