Dataset Information

An explainable supervised machine learning predictor of acute kidney injury after adult deceased donor liver transplantation.

ABSTRACT:

Background

Early prediction of acute kidney injury (AKI) after liver transplantation (LT) facilitates timely recognition and intervention. We aimed to build a risk predictor of post-LT AKI via supervised machine learning and visualize the mechanism driving within to assist clinical decision-making.

Methods

Data of 894 cases that underwent liver transplantation from January 2015 to September 2019 were collected, covering demographics, donor characteristics, etiology, peri-operative laboratory results, co-morbidities and medications. The primary outcome was new-onset AKI after LT according to Kidney Disease Improving Global Outcomes guidelines. Predicting performance of five classifiers including logistic regression, support vector machine, random forest, gradient boosting machine (GBM) and adaptive boosting were respectively evaluated by the area under the receiver-operating characteristic curve (AUC), accuracy, F1-score, sensitivity and specificity. Model with the best performance was validated in an independent dataset involving 195 adult LT cases from October 2019 to March 2021. SHapley Additive exPlanations (SHAP) method was applied to evaluate feature importance and explain the predictions made by ML algorithms.

Results

430 AKI cases (55.1%) were diagnosed out of 780 included cases. The GBM model achieved the highest AUC (0.76, CI 0.70 to 0.82), F1-score (0.73, CI 0.66 to 0.79) and sensitivity (0.74, CI 0.66 to 0.8) in the internal validation set, and a comparable AUC (0.75, CI 0.67 to 0.81) in the external validation set. High preoperative indirect bilirubin, low intraoperative urine output, long anesthesia time, low preoperative platelets, and graft steatosis graded NASH CRN 1 and above were revealed by SHAP method the top 5 important variables contributing to the diagnosis of post-LT AKI made by GBM model.

Conclusions

Our GBM-based predictor of post-LT AKI provides a highly interoperable tool across institutions to assist decision-making after LT.

SUBMITTER: Zhang Y

PROVIDER: S-EPMC8317304 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:Adult-to-adult living donors and recipients were studied to characterize patterns of liver growth and identify associated factors in a multicenter study. Three hundred and fifty donors and 353 recipients in the Adult-to-Adult Living Donor Liver Transplantation Cohort Study (A2ALL) receiving transplants between March 2003 and February 2010 were included. Potential predictors of 3-month liver volume included total and standard liver volumes (TLV and SLV), Model for End-Stage Liver Disease (MELD) score (in recipients), the remnant and graft size, remnant-to-donor and graft-to-recipient weight ratios (RDWR and GRWR), remnant/TLV, and graft/SLV. Among donors, 3-month absolute growth was 676 ± 251 g (mean ± SD), and percentage reconstitution was 80% ± 13%. Among recipients, GRWR was 1.3% ± 0.4% (8 < 0.8%). Graft weight was 60% ± 13% of SLV. Three-month absolute growth was 549 ± 267 g, and percentage reconstitution was 93% ± 18%. Predictors of greater 3-month liver volume included larger patient size (donors and recipients), larger graft volume (recipients), and larger TLV (donors). Donors with the smallest remnant/TLV ratios had larger than expected growth but also had higher postoperative bilirubin and international normalized ratio at 7 and 30 days. In a combined donor-recipient analysis, donors had smaller 3-month liver volumes than recipients adjusted for patient size, remnant or graft volume, and TLV or SLV (P = 0.004). Recipient graft failure in the first 90 days was predicted by poor graft function at day 7 (HR = 4.50, P = 0.001) but not by GRWR or graft fraction (P > 0.90 for each). Both donors and recipients had rapid yet incomplete restoration of tissue mass in the first 3 months, and this confirmed previous reports. Recipients achieved a greater percentage of expected total volume. Patient size and recipient graft volume significantly influenced 3-month volumes. Importantly, donor liver volume is a critical predictor of the rate of regeneration, and donor remnant fraction affects postresection function. Liver Transpl 21:79-88, 2015. © 2014 AASLD.

Project description:Socioeconomically deprived individuals with renal disease are less likely to receive a live-donor kidney transplant than less-deprived individuals. This qualitative study aimed to identify reasons for the observed socioeconomic disparity in live-donor kidney transplantation.A qualitative study using face-to-face in-depth semistructured interviews.A UK tertiary renal referral hospital and transplant centre.Purposive sampling was used to select deceased-donor transplant recipients from areas of high socioeconomic deprivation (SED) (19 participants), followed by a low SED comparison group (13 participants), aiming for maximum diversity in terms of age, gender, ethnicity, primary renal disease and previous renal replacement therapy.Participants were interviewed following their routine transplant clinic review. Interviews were digitally audio-recorded and transcribed verbatim. Transcripts were coded using NVivo software and analysed using the constant comparison method described in Grounded Theory.Themes common and distinct to each socioeconomic group emerged. 6 themes appeared to distinguish between individuals from areas of high and low SED. 4 themes were distinct to participants from areas of high SED: (1) Passivity, (2) Disempowerment, (3) Lack of social support and (4) Short-term focus. 2 themes were distinct to the low SED group: (1) Financial concerns and (2) Location of donor.Several of the emerging themes from the high SED individuals relate to an individual's lack of confidence and skill in managing their health and healthcare; themes that are in keeping with low levels of patient activation. Inadequate empowerment of socioeconomically deprived individuals by healthcare practitioners was also described. Financial concerns did not emerge as a barrier from interviews with the high SED group. Interventions aiming to redress the observed socioeconomic inequity should be targeted at both patients and clinical teams to increase empowerment and ensure shared decision-making.