Comprehensive deep learning-based assessment of living liver donor CT angiography: from vascular segmentation to volumetric analysis.

BACKGROUND: Precise preoperative assessment of liver vasculature and volume in living donor liver transplantation is essential for donor safety and recipient surgery. Traditional manual segmentation methods are being supplemented by deep learning (DL) models, which may offer more consistent and efficient volumetric evaluations. METHODS: This study analyzed living liver donors from Samsung Medical Center using preoperative CT angiography data between April 2022 and February 2023. A DL-based 3D residual U-Net model was developed and trained on segmented CT images to calculate the liver volume and segment vasculature, with its performance compared to traditional manual segmentation by surgeons and actual graft weight. RESULTS: The DL model achieved high concordance with manual methods, exhibiting Dice Similarity Coefficients of 0.94±0.01 for the right lobe and 0.91±0.02 for the left lobe. The liver volume estimates by DL model closely matched those of surgeons, with a mean discrepancy of 9.18 mL, and correlated more strongly with actual graft weights (R-squared value of 0.76 compared to 0.68 for surgeons). CONCLUSION: The DL model demonstrates potential as a reliable tool for enhancing preoperative planning in liver transplantation, offering consistency and efficiency in volumetric assessment. Further validation is required to establish its generalizability across various clinical settings and imaging protocols.

Artificial intelligence for predicting survival following deceased donor liver transplantation: Retrospective multi-center study.

BACKGROUND: Previous studies have indicated that the model for end-stage liver disease (MELD) score may fail to predict post-transplantation patient survival. Similarly, other scores (donor MELD score, balance of risk score) that have been developed to predict transplant outcomes have not gained widespread use. These scores are typically derived using linear statistical models. This study aimed to compare the performance of traditional statistical models with machine learning approaches for predicting survival following liver transplantation. MATERIALS AND METHODS: Data were obtained from 785 deceased donor liver transplant recipients enrolled in the Korean Organ Transplant Registry (2014-2019). Five machine learning methods (random forest, artificial neural networks, decision tree, naïve Bayes, and support vector machine) and four traditional statistical models (Cox regression, MELD score, donor MELD score and balance of risk score) were compared to predict survival. RESULTS: Among the machine learning methods, the random forest yielded the highest area under the receiver operating characteristic curve (AUC-ROC) values (1-month = 0.80; 3-month = 0.85; and 12-month = 0.81) for predicting survival. The AUC-ROC values of the Cox regression analysis were 0.75, 0.86, and 0.77 for 1-month, 3-month, and 12-month post-transplant survival, respectively. However, the AUC-ROC values of the MELD, donor MELD, and balance of risk scores were all below 0.70. Based on the variable importance of the random forest analysis in this study, the major predictors associated with survival were cold ischemia time, donor ICU stay, recipient weight, recipient BMI, recipient age, recipient INR, and recipient albumin level. As with the Cox regression analysis, donor ICU stay, donor bilirubin level, BAR score, and recipient albumin levels were also important factors associated with post-transplant survival in the RF model. The coefficients of these variables were also statistically significant in the Cox model (p < 0.05). The SHAP ranges for selected predictors for the 12-month survival were (-0.02,0.10) for recipient albumin, (-0.05,0.07) for donor bilirubin and (-0.02,0.25) for recipient height. Surprisingly, although not statistically significant in the Cox model, recipient weight, recipient BMI, recipient age, or recipient INR were important factors in our random forest model for predicting post-transplantation survival. CONCLUSION: Machine learning algorithms such as the random forest were superior to conventional Cox regression and previously reported survival scores for predicting 1-month, 3-month, and 12-month survival following liver transplantation. Therefore, artificial intelligence may have significant potential in aiding clinical decision-making during liver transplantation, including matching donors and recipients.

Sex Difference in the Tolerance of Hepatic Ischemia-reperfusion Injury and Hepatic Estrogen Receptor Expression According to Age and Macrosteatosis in Healthy Living Liver Donors.

BACKGROUND: Hepatic estrogen signaling, which is important in liver injury/recovery, is determined by the level of systemic estrogen and hepatic estrogen receptor. We aimed to evaluate whether females' advantage in the tolerance of hepatic ischemia-reperfusion injury decreases according to the age of 40 y (systemic estrogen decrease) and macrosteatosis (hepatic estrogen receptor decrease). METHODS: We included 358 living liver donors (128 female and 230 male individuals). The tolerance of hepatic ischemia-reperfusion injury was determined by the slope of the linear regression line modeling the relationship between the duration of intraoperative hepatic ischemia and the peak postoperative transaminase level. Estrogen receptor content was measured in the biopsied liver samples using immunohistochemistry. RESULTS: In the whole cohort, the regression slope for aspartate transaminase was comparable between female and male individuals (P = 0.940). Within the subgroup of donors aged ≤40 y, the regression slope was significantly smaller in female individuals (P = 0.031), whereas it was comparable within donors aged >40 y (P = 0.867). Within the subgroup of nonmacrosteatotic donors aged ≤40 y, the regression slope was significantly smaller in female individuals in univariable (P = 0.002) and multivariable analysis (P = 0.006), whereas the sex difference was not found within macrosteatotic donors aged ≤40 y (P = 0.685). Estrogen receptor content was significantly greater in female individuals within nonmacrosteatotic donors aged ≤40 y (P = 0.021), whereas it was not different in others of age >40 y or with macrosteatosis (P = 0.450). CONCLUSIONS: The tolerance of hepatic ischemia-reperfusion injury was greater in female individuals than in male individuals only when they were <40 y and without macrosteatosis. The results were in agreement with the hepatic estrogen receptor immunohistochemistry study.

New-onset diabetes after adult liver transplantation in the Korean Organ Transplantation Registry (KOTRY) study.

BACKGROUND: New-onset diabetes after transplantation (NODAT) is a serious complication following liver transplantation (LT). The present study aimed to investigate the incidence of and risk factors for NODAT using the Korean Organ Transplantation Registry (KOTRY) database. METHODS: Patients with history of pediatric transplantation (age ≤18 years), re-transplantation, multi-organ transplantation, or pre-existing diabetes mellitus were excluded. A total of 1,919 non-diabetic adult patients who underwent a primary LT between May 2014 and December 2017 were included. Risk factors were identified using Cox regression analysis. RESULTS: NODAT occurred in 19.7% (n=377) of adult liver transplant recipients. Multivariate analysis showed steroid use, increased age, and high body mass index (BMI) in recipients, and implantation of a left-side liver graft was closely associated with NODAT in adult LT. In living donor liver transplant (LDLT) patients (n=1,473), open donor hepatectomy in the living donors, steroid use, small for size liver graft (graft to recipient weight ratio ≤0.8), increased age, and high BMI in the recipient were predictive factors for NODAT. The use of antimetabolite and basiliximab induction reduced the incidence of NODAT in adult LT and in adult LDLT. CONCLUSIONS: Basiliximab induction, early steroid withdrawal, and antimetabolite therapy may prevent NODAT after adult LT. High BMI or advanced age in liver recipients, open donor hepatectomy in living donors, and small size liver graft can predict the occurrence of NODAT after adult LT or LDLT.

Survival Outcomes in Split Compared With Whole Liver Transplantation.

Split-liver transplantation (SLT) should be cautiously considered because the right trisection (RTS) graft can be a marginal graft in adult recipients. Herein, we analyzed the outcomes of RTS-SLT in Korea, where >75% of adult liver transplantations are performed with living donor liver transplantation. Among 2462 patients who underwent deceased donor liver transplantations (DDLTs) from 2005 to 2014, we retrospectively reviewed 86 (3.5%) adult patients who received a RTS graft (RTS-SLT group). The outcomes of the RTS-SLT group were compared with those of 303 recipients of whole liver (WL; WL-DDLT group). Recipient age, laboratory Model for End-Stage-Liver Disease (L-MELD) score, ischemia time, and donor-to-recipient weight ratio (DRWR) were not different between the 2 groups (P > 0.05). However, malignancy was uncommon (4.7% versus 36.3%), and the donor was younger (25.2 versus 42.7 years) in the RST-SLT group than in the WL-DDLT group (P < 0.05). The technical complication rates and the 5-year graft survival rates (89.0% versus 92.8%) were not different between the 2 groups (P > 0.05). The 5-year overall survival (OS) rate (63.1%) and graft-failure-free survival rate (63.1%) of the RTS-SLT group were worse than that of the WL-DDLT group (79.3% and 79.3%; P < 0.05). The factors affecting graft survival rates were not definite. However, the factors affecting OS in the RTS-SLT group were L-MELD score >30 and DRWR ≤1.0. In the subgroup analysis, OS was not different between the 2 groups if the DRWR was >1.0, regardless of the L-MELD score (P > 0.05). In conclusion, a sufficient volume of the graft estimated from DRWR-matching could lead to better outcomes of adult SLTs with a RTS graft, even in patients with high L-MELD scores.

Glycemic responses to intermittent hepatic inflow occlusion in living liver donors.

The occurrence of glycemic disturbances has been described for patients undergoing intermittent hepatic inflow occlusion (IHIO) for tumor removal. However, the glycemic responses to IHIO in living liver donors are unknown. This study investigated the glycemic response to IHIO in these patients and examined the association between this procedure and the occurrence of hyperglycemia (blood glucose > 180 mg/dL). The data from 154 living donors were retrospectively reviewed. The decision to perform IHIO was made on the basis of the extent of bleeding that occurred during parenchymal dissection. One round of IHIO consisted of 15 minutes of clamping and 5 minutes of unclamping the hepatic artery and portal vein. Blood glucose concentrations were measured at predetermined time points, including the start and end of IHIO. Repeated hyperglycemic episodes occurred after unclamping. The mean maximum intraoperative blood glucose concentration was greater in donors who underwent ≥3 rounds of IHIO versus those who underwent 1 or 2 rounds (169 ± 30 versus 149 ± 31 mg/dL, P = 0.005). The incidence of intraoperative hyperglycemia was also greater in donors who underwent ≥3 rounds of IHIO versus those who underwent 1 or 2 rounds (38.7% versus 7.7%, odds ratio = 7.1, 95% confidence interval = 2.5-20.4, P < 0.001). Donors who did not undergo IHIO and those who underwent 1 or 2 rounds of IHIO exhibited similar maximum glucose concentrations and similar incidence rates of hyperglycemia. In conclusion, IHIO induced repeated hyperglycemic responses in living donors, and donors who underwent ≥3 rounds of IHIO were more likely to experience intraoperative hyperglycemia. These results provide additional information on the risks and benefits of IHIO in living donors.

Time of hepatocellular carcinoma recurrence after liver resection and alpha-fetoprotein are important prognostic factors for salvage liver transplantation.

Salvage liver transplantation (LT) is considered a feasible option for the treatment of recurrent hepatocellular carcinoma (HCC). We performed this multicenter study to assess the risk factors associated with the recurrence of HCC and patient survival after salvage LT. Between January 2000 and December 2011, 101 patients who had previously undergone liver resection (LR) for HCC underwent LT at 3 transplant centers in Korea. Sixty-nine patients' data were retrospectively reviewed for the analysis. The recurrence of HCC was diagnosed at a median of 10.6 months after the initial LR, and patients underwent salvage LT. Recurrences were within the Milan criteria in 48 cases and were outside the Milan criteria in 21 cases. After salvage LT, 31 patients had HCC recurrence during a median follow-up period of 24.5 months. There were 24 deaths, and 20 were due to HCC recurrence. The 5-year overall survival rate was approximately 54.6%, and the 5-year recurrence-free survival rate was 49.3%. HCC recurrence within the 8 months after LR [hazard ratio (HR) = 3.124, P = 0.009], an alpha-fetoprotein level higher than 200 ng/mL (HR = 2.609, P = 0.02), and HCC outside the Milan criteria at salvage LT (HR = 2.219, P = 0.03) were independent risk factors for poor recurrence-free survival after salvage LT. In conclusion, the timing and extent of HCC recurrence after primary LR both play significant roles in the outcome of salvage LT.

The prognosis and treatment outcomes of patients with recurrent hepatocellular carcinoma after liver transplantation.

Liver transplantation (LT) is performed in patients with hepatocellular carcinoma (HCC), but recurrent HCC after LT remains a problem. We retrospectively reviewed the data from 63 patients with recurrent HCC who underwent LT at a single institution between September 1996 and March 2011 to determine the prognosis of patients with recurrent HCC after LT. A survival analysis was performed with the preoperative data, histological findings, patterns of recurrence, and treatment methods. Univariate and multivariate analyses were performed to determine the factors associated with early (<1 yr) cancer-related death. The independent prognostic factors, according to the multivariate analysis, were recurrence within six months (hazards ratio [HR] = 4.557, p = 0.021) and initial multiple-organ involvement (HR = 5.494, p = 0.015). The survival rates of patients differed according to the treatment type. The combined treatment with local and systemic treatment resulted in increased survival even in patients with HCC recurrences involving multiple organs.