Developing deep learning-based strategies to predict the risk of hepatocellular carcinoma among patients with nonalcoholic fatty liver disease from electronic health records.

OBJECTIVE: The accuracy of deep learning models for many disease prediction problems is affected by time-varying covariates, rare incidence, covariate imbalance and delayed diagnosis when using structured electronic health records data. The situation is further exasperated when predicting the risk of one disease on condition of another disease, such as the hepatocellular carcinoma risk among patients with nonalcoholic fatty liver disease due to slow, chronic progression, the scarce of data with both disease conditions and the sex bias of the diseases. The goal of this study is to investigate the extent to which the aforementioned issues influence deep learning performance, and then devised strategies to tackle these challenges. These strategies were applied to improve hepatocellular carcinoma risk prediction among patients with nonalcoholic fatty liver disease. METHODS: We evaluated two representative deep learning models in the task of predicting the occurrence of hepatocellular carcinoma in a cohort of patients with nonalcoholic fatty liver disease (n = 220,838) from a national EHR database. The disease prediction task was carefully formulated as a classification problem while taking censorship and the length of follow-up into consideration. RESULTS: We developed a novel backward masking scheme to deal with the issue of delayed diagnosis which is very common in EHR data analysis and evaluate how the length of longitudinal information after the index date affects disease prediction. We observed that modeling time-varying covariates improved the performance of the algorithms and transfer learning mitigated reduced performance caused by the lack of data. In addition, covariate imbalance, such as sex bias in data impaired performance. Deep learning models trained on one sex and evaluated in the other sex showed reduced performance, indicating the importance of assessing covariate imbalance while preparing data for model training. CONCLUSIONS: The strategies developed in this work can significantly improve the performance of hepatocellular carcinoma risk prediction among patients with nonalcoholic fatty liver disease. Furthermore, our novel strategies can be generalized to apply to other disease risk predictions using structured electronic health records, especially for disease risks on condition of another disease.

Developing deep learning-based strategies to predict the risk of hepatocellular carcinoma among patients with nonalcoholic fatty liver disease from electronic health records.

Background: Deep learning models showed great success and potential when applied to many biomedical problems. However, the accuracy of deep learning models for many disease prediction problems is affected by time-varying covariates, rare incidence, and covariate imbalance when using structured electronic health records data. The situation is further exasperated when predicting the risk of one disease on condition of another disease, such as the hepatocellular carcinoma risk among patients with nonalcoholic fatty liver disease due to slow, chronic progression, the scarce of data with both disease conditions and the sex bias of the diseases. Objective: The goal of this study is to investigate the extent to which time-varying covariates, rare incidence, and covariate imbalance influence deep learning performance, and then devised strategies to tackle these challenges. These strategies were applied to improve hepatocellular carcinoma risk prediction among patients with nonalcoholic fatty liver disease. Methods: We evaluated two representative deep learning models in the task of predicting the occurrence of hepatocellular carcinoma in a cohort of patients with nonalcoholic fatty liver disease (n = 220,838) from a national EHR database. The disease prediction task was carefully formulated as a classification problem while taking censorship and the length of follow-up into consideration. Results: We developed a novel backward masking scheme to evaluate how the length of longitudinal information after the index date affects disease prediction. We observed that modeling time-varying covariates improved the performance of the algorithms and transfer learning mitigated reduced performance caused by the lack of data. In addition, covariate imbalance, such as sex bias in data impaired performance. Deep learning models trained on one sex and evaluated in the other sex showed reduced performance, indicating the importance of assessing covariate imbalance while preparing data for model training. Conclusions: Devising proper strategies to address challenges from time-varying covariates, lack of data, and covariate imbalance can be key to counteracting data bias and accurately predicting disease occurrence using deep learning models. The novel strategies developed in this work can significantly improve the performance of hepatocellular carcinoma risk prediction among patients with nonalcoholic fatty liver disease. Furthermore, our novel strategies can be generalized to apply to other disease risk predictions using structured electronic health records, especially for disease risks on condition of another disease.

Unsaturated or saturated dietary fat-mediated steatosis impairs hepatic regeneration following partial hepatectomy in mice.

BACKGROUND: Partial hepatectomy is a preferred treatment option for many patients with hepatocellular carcinoma however, pre-existing pathological abnormalities originating from hepatic steatosis can alter the decision to perform surgery or postoperative outcomes as a consequence of the impact steatosis has on liver regeneration. AIM: The aim of this study was to investigate the role of a saturated or unsaturated high fat diet-mediated steatosis on liver regeneration following partial hepatectomy. METHODS: Mice were fed a low-fat control diet (CD, 13% fat), lard-based unsaturated (LD, 60% fat) or milk-based saturated high fat diet (MD, 60% fat) for 16 weeks at which time partial hepatectomy (approx. 70% resection) was performed. At days-2 and 7 post hepatectomy, one hour prior to euthanization, mice were injected with 5-bromo-2'-deoxyuridine in order to monitor hepatic regeneration. Serum was collected and assessed for levels of ALT and AST. Resected and regenerated liver tissue were examined for inflammation-indicative markers employing RT-PCR, Western blots, and histological methods. RESULTS: Mice fed LD or MD exhibited higher NAFLD scores, increased expression of inflammatory cytokines, neutrophil infiltration, macrophage accumulation, increased apoptosis, and elevated levels of serum ALT and AST activities, a decrease in the number of BrdU-incorporated-hepatocytes in the regenerated livers compared to the mice fed CD. Mice fed MD showed significantly lower percent of BrdU-incorporated hepatocytes and a higher trend of inflammation compared to the mice fed LD. CONCLUSION: A diet rich in saturated or unsaturated fat results in NASH with decreased hepatic regeneration however unsaturated fat diet cause lower inflammation and higher regeneration than the saturated fat diet following partial hepatectomy in mice.

Long-term clinical outcomes of patients with COVID-19 and chronic liver disease: US multicenter COLD study.

BACKGROUND: COVID-19 is associated with higher morbidity and mortality in patients with chronic liver diseases (CLDs). However, our understanding of the long-term outcomes of COVID-19 in patients with CLD is limited. METHODS: We conducted a multicenter, observational cohort study of adult patients with CLD who were diagnosed with COVID-19 before May 30, 2020, to determine long-term clinical outcomes. We used a control group of patients with CLD confirmed negative for COVID-19. RESULTS: We followed 666 patients with CLD (median age 58 years, 52.8% male) for a median of 384 (interquartile range: 31-462) days. The long-term mortality was 8.1%; with 3.6% experiencing delayed COVID-19-related mortality. Compared to a propensity-matched control group of patients with CLD without COVID-19 (n=1332), patients with CLD with COVID-19 had worse long-term survival [p<0.001; hazards ratio (HR): 1.69; 95% CI: 1.19-2.41] and higher rate of hospitalization (p<0.001, HR: 2.00, 1.62-2.48) over a 1-year follow-up period. Overall, 29.9% of patients reported symptoms of long-COVID-19. On multivariable analysis, female sex (p=0.05, HR: 2.45, 1.01-2.11), Hispanic ethnicity (p=0.003, HR: 1.94, 1.26-2.99), and severe COVID-19 requiring mechanical ventilation (p=0.028, HR: 1.74, 1.06-2.86) predicted long-COVID-19. In survivors, liver-related laboratory parameters showed significant improvement after COVID-19 resolution. COVID-19 vaccine status was available for 72% (n=470) of patients with CLD and history of COVID-19, of whom, 70% (n=326) had received the COVID-19 vaccine. CONCLUSIONS: Our large, longitudinal, multicenter study demonstrates a high burden of long-term mortality and morbidity in patients with CLD and COVID-19. Symptoms consistent with long-COVID-19 were present in 30% of patients with CLD. These results illustrate the prolonged implications of COVID-19 both for recovering patients and for health care systems.

Dosing strategies for de novo once-daily extended release tacrolimus in kidney transplant recipients based on CYP3A5 genotype.

BACKGROUND: Tacrolimus extended-release tablets have been Food and Drug Administration-approved for use in the de novo kidney transplant population. Dosing requi rements often vary for tacrolimus based on several factors including variation in metabolism based on CYP3A5 expression. Patients who express CYP3A5 often require higher dosing of immediate-release tacrolimus, but this has not been established for tacrolimus extended-release tablets in the de novo setting. AIM: To obtain target trough concentrations of extended-release tacrolimus in de novo kidney transplant recipients according to CYP3A5 genotype. METHODS: Single-arm, prospective, single-center, open-label, observational study (ClinicalTrials.gov: NCT037 13645). Life cycle pharma tacrolimus (LCPT) orally once daily at a starting dose of 0.13 mg/kg/day based on actual body weight. If weight is more than 120% of ideal body weight, an adjusted body weight was used. LCPT dose was adjusted to maintain tacrolimus trough concentrations of 8-10 ng/mL. Pharmacogenetic analysis of CYP3A5 genotype was performed at study conclusion. RESULTS: Mean time to therapeutic tacrolimus trough concentration was longer in CYP3A5 intermediate and extensive metabolizers vs CYP3A5 non-expressers (6 d vs 13.5 d vs 4.5 d; P = 0.025). Mean tacrolimus doses and weight-based doses to achieve therapeutic concentration were higher in CYP3A5 intermediate and extensive metabolizers vs CYP3A5 non-expressers (16 mg vs 16 mg vs 12 mg; P = 0.010) (0.20 mg/kg vs 0.19 mg/kg vs 0.13 mg/kg; P = 0.018). CYP3A5 extensive metabolizers experienced lower mean tacrolimus trough concentrations throughout the study period compared to CYP3A5 intermediate metabolizers and non-expressers (7.98 ng/mL vs 9.18 ng/mL vs 10.78 ng/mL; P = 0 0.008). No differences were identified with regards to kidney graft function at 30-d post-transplant. Serious adverse events were reported for 13 (36%) patients. CONCLUSION: Expression of CYP3A5 leads to higher starting doses and incremental dosage titration of extended-release tacro limus to achieve target trough concentrations. We suggest a higher starting dose of 0.2 mg/kg/d for CYP3A5 expressers.

Medicaid expansion is associated with a higher likelihood of early diagnosis, resection, transplantation, and overall survival in patients with hepatocellular carcinoma.

BACKGROUND: We examined the association between Medicaid expansion (ME) and the diagnosis, treatment, and survival of patients with hepatocellular carcinoma (HCC). METHODS: We identified patients with HCC <65yrs with Medicaid or without insurance within the National Cancer Database before (2010-2013) or after (2015-2017) ME with early (cT1) or intermediate/advanced (cT2-T4 or M1) disease. RESULTS: We identified 4848 patients with HCC before and 4526 after ME. Prior to ME, there was no association between future ME status and diagnosis of early HCC (34.5% vs. 32.9%). There was no association between future ME status and treating early HCC with ablation, resection, or transplantation. Patients with early HCC in future ME states were less likely to die (HR = 0.81, 95% CI: 0.67-0.98). After ME, patients in ME states were more likely to be diagnosed with early HCC (39.2% vs. 32.1%). Patients with early disease in ME states were more likely to undergo resection (OR=1.78, 95% CI: 1.16-2.75) or transplantation (OR=3.20, 95% CI: 1.40-7.33). There was a further associated decrease in the hazard of death (HR=0.68, 95% CI: 0.54-0.86). CONCLUSION: ME was associated with early diagnosis of HCC. For early HCC, ME was associated with increased utilization of resection and transplantation and improvement in survival.

A nationwide analysis of clinical trial participation for common hepato-pancreato-biliary malignancies demonstrates survival advantages for subsets of trial patients but disparities in and infrequency of enrollment.

BACKGROUND: We describe factors associated with trial enrollment for patients with hepato-pancreato-biliary (HPB) malignancies. We analyzed the association and effect of trial enrollment on overall survival (OS). METHODS: The National Cancer Database (2004-2017) was queried for common HPB malignancies (pancreatic adenocarcinoma [PDAC] & neuroendocrine tumors, hepatocellular carcinoma [HCC], biliary tract cancers [BTC]). Multivariable logistic regression was used to identify factors associated with trial enrollment. OS was analyzed by multivariable Cox regression. Inverse-probability-weighted Cox regression was utilized to determine the effect of trial enrollment on OS. RESULTS: A total of 1573 (0.3%) of 511,639 patients were enrolled in trials; pancreatic malignancy: 1214 (0.4%); HCC: 217 (0.14%); BTC: 106 (0.15%). HCC and BTC were associated with lower likelihood of enrollment compared with pancreatic malignancy. Black and Hispanic patients were less likely to be enrolled compared to White patients. Treatment at academic facilities and metastatic disease were associated with higher likelihood of enrollment. Enrollment was associated with higher OS for PDAC, metastatic HCC, and metastatic BTC. Trial enrollment exhibited an OS advantage for PDAC and metastatic HCC. CONCLUSION: Nationally, fewer than 1% of patients with HPB malignancies were enrolled in clinical trials. There are racial, sociodemographic, and facility-based disparities in trial enrollment.

Black race is independently associated with underutilization of transplantation for clinical T1 hepatocellular carcinoma.

BACKGROUND: Hepatocellular carcinoma (HCC) is a leading cause of cancer mortality. Operative management of early disease includes ablation, resection, and transplantation. We compared the operative management of early-stage HCC in patients stratified by race. METHODS: We identified patients with cT1 HCC and Charlson-Deyo score 0-1 in the National Cancer Database (2004-2016). We compared operative/non-operative management by race, adjusting for clinicodemographic variables. We performed marginal standardization of logistic regression to ascertain adjusted probabilities of resection or transplantation in patients under 70 years of age with insurance. RESULTS: A total of 25,029 patients were included (White = 20,410; Black = 4619). After adjusting for clinico demographic variables, Black race was associated with a lower likelihood of undergoing operative intervention (OR 0.89,p = 0.009). Black patients were more likely to undergo resection (OR 1.23,p < 0.001) and less likely to undergo transplantation (OR 0.60,p < 0.001). Marginal standardization models demonstrated Black race was associated with increased probability of resection in patients >50yrs, with private insurance/Medicare, and lower probability of transplantation regardless of age or insurance payor. CONCLUSION: Black race is associated with lower rates of hepatic transplantation and higher rates of hepatic resection for early HCC regardless of age or insurance payor. The etiology of these disparities is multifactorial and correcting the root causes represents a critical area for improvement.

Outcomes of Surgical Mitral and Aortic Valve Replacements Among Kidney Transplant Candidates: Implications for Valve Selection.

Background Limited literature exists that evaluated outcomes of kidney transplant-eligible patients who are having dialysis and who are undergoing valve replacement. Our main objective in this study was to compare mortality, reoperation, and bleeding episodes between bioprosthetic and mechanical valve procedures among kidney transplant-eligible patients who are having dialysis. Methods and Results We studied 887 and 1925 dialysis patients from the United States Renal Data System, who underwent mitral valve replacement and aortic valve replacement (AVR) after being waitlisted for a kidney transplant (2000-2015), respectively. Time to death, time to reoperation, and time to bleeding requiring hospitalizations were compared separately for AVR and mitral valve replacement. Kaplan-Meier survival curves, Cox proportional hazards model for time to death, accelerated time to event model for time to reoperation, and counting process model for time to recurrent bleeding were used. There were no differences in mortality (hazard ratio [HR], 0.92; 95% CI, 0.77-1.09) or risk of reoperation or risk of significant bleeding events between bioprosthetic and mechanical mitral valve replacement. However, mechanical AVR was associated with a modestly significant less hazard of death (HR, 0.83; 95% CI, 0.74-0.94) compared with bioprosthetic AVR. There were no differences in time to reoperation, or time to significant bleeding events between bioprosthetic and mechanical AVR. Conclusions For kidney transplant waitlisted patients who are on dialysis and who are undergoing surgical valve replacement, bioprosthetic and mechanical valves have comparable survival, reoperation rates, and bleeding episodes requiring hospitalizations at both mitral and aortic locations. These findings emphasize that an individualized informed decision is recommended when choosing the type of valve for this special group of patients having dialysis.

A comparison of surgical resection and liver transplantation in the treatment of intrahepatic cholangiocarcinoma in the era of modern chemotherapy: An analysis of the National Cancer Database.

BACKGROUND: The main surgical approach to patients with localized intrahepatic cholangiocarcinoma (ICC) is hepatectomy, but transplantation has been described. A comparison of outcomes between these surgical approaches is necessary to determine if one is preferable. METHODS: Patients with ICC were identified using the National Cancer Database (2010-2016). Patients were grouped based on operation and matched 1:1 by propensity score. Pathologic and postoperative outcomes, as well as overall survival were analyzed. RESULTS: There were 1879 hepatectomy and 74 liver transplantation patients. Before matching, transplantation patients were younger and more often treated at academic centers. More patients who underwent a transplantation received neoadjuvant chemotherapy (70.3% vs. 12.8%). Patients who underwent transplantation had more pathologic T0 (7.7% vs. 0.4%) and T1 (47.7% vs. 42.1%) tumors (p < .001). There were no differences in length of stay, unplanned readmissions, 30/90-day mortality, or median survival between groups (36.1 vs. 36.1 months, p = .34). After matching (n = 57/group), there were no differences in postoperative outcomes or survival between transplantation or hepatectomy (36.1 vs. 33.6 months, p = .57). CONCLUSION: Among patients with ICC, hepatectomy and liver transplantation were associated with similar postoperative outcomes and survival. In light of the resources and chronic immunosuppression required for transplantation, hepatectomy seems preferable for localized ICC.

A Novel Method to Improve Perfusion of Ex Vivo Pumped Human Kidneys.

OBJECTIVE: To determine if addition of the S-nitrosylating agent ethyl nitrite (ENO) to the preservation solution can improve perfusion parameters in pumped human kidneys. BACKGROUND: A significant percentage of actively stored kidneys experience elevations in resistance and decreases in flow rate during the ex vivo storage period. Preclinical work indicates that renal status after brain death is negatively impacted by inflammation and reduced perfusion-processes regulated by protein S-nitrosylation. To translate these findings, we added ENO to the preservation solution in an attempt to reverse the perfusion deficits observed in nontransplanted pumped human kidneys. METHODS: After obtaining positive proof-of-concept results with swine kidneys, we studied donated human kidneys undergoing hypothermic pulsatile perfusion deemed unsuitable for transplantation. Control kidneys continued to be pumped a 4°C (ie, standard of care). In the experimental group, the preservation solution was aerated with 50 ppm ENO in nitrogen. Flow rate and perfusion were recorded for 10 hours followed by biochemical analysis of the kidney tissue. RESULTS: In controls, perfusion was constant during the monitoring period (ie, flow rate remained low and resistance stayed high). In contrast, the addition of ENO produced significant and sustained reductions in resistance and increases in flow rate. ENO-treated kidneys had higher levels of cyclic guanosine monophosphate, potentially explaining the perfusion benefits, and increased levels of interleukin-10, suggestive of an anti-inflammatory effect. CONCLUSIONS: S-Nitrosylation therapy restored the microcirculation and thus improved overall organ perfusion. Inclusion of ENO in the renal preservation solution holds promise to increase the number and quality of kidneys available for transplant.

Predictors of Outcomes of COVID-19 in Patients With Chronic Liver Disease: US Multi-center Study.

BACKGROUND & AIMS: Chronic liver disease (CLD) represents a major global health burden. We undertook this study to identify the factors associated with adverse outcomes in patients with CLD who acquire the novel coronavirus-2019 (COVID-19). METHODS: We conducted a multi-center, observational cohort study across 21 institutions in the United States (US) of adult patients with CLD and laboratory-confirmed diagnosis of COVID-19 between March 1, 2020 and May 30, 2020. We performed survival analysis to identify independent predictors of all-cause mortality and COVID-19 related mortality, and multivariate logistic regression to determine the risk of severe COVID-19 in patients with CLD. RESULTS: Of the 978 patients in our cohort, 867 patients (mean age 56.9 ± 14.5 years, 55% male) met inclusion criteria. The overall all-cause mortality was 14.0% (n = 121), and 61.7% (n = 535) had severe COVID-19. Patients presenting with diarrhea or nausea/vomiting were more likely to have severe COVID-19. The liver-specific factors associated with independent risk of higher overall mortality were alcohol-related liver disease (ALD) (hazard ratio [HR] 2.42, 95% confidence interval [CI] 1.29-4.55), decompensated cirrhosis (HR 2.91 [1.70-5.00]) and hepatocellular carcinoma (HCC) (HR 3.31 [1.53-7.16]). Other factors were increasing age, diabetes, hypertension, chronic obstructive pulmonary disease and current smoker. Hispanic ethnicity (odds ratio [OR] 2.33 [1.47-3.70]) and decompensated cirrhosis (OR 2.50 [1.20-5.21]) were independently associated with risk for severe COVID-19. CONCLUSIONS: The risk factors which predict higher overall mortality among patients with CLD and COVID-19 are ALD, decompensated cirrhosis and HCC. Hispanic ethnicity and decompensated cirrhosis are associated with severe COVID-19. Our results will enable risk stratification and personalization of the management of patients with CLD and COVID-19. Clinicaltrials.gov number NCT04439084.

Surgical Duct-to-Duct Reconstruction: an Alternative Approach to Late Biliary Anastomotic Stricture After Deceased Donor Liver Transplantation.

BACKGROUND: Bilio-enteric diversion is the current surgical standard in patients after deceased donor liver transplantation (DDLT) with a biliary anastomotic stricture failing interventional treatment and requiring surgical repair. In contrast to this routine, the aim of this study was to show the feasibility and safety of a duct-to-duct biliary reconstruction. PATIENTS: Between 2012 and 2019, we performed a total of 308 DDLT in 292 adult patients. The overall biliary complication rate was 20.5%. Patients with non-anastomotic or combined strictures were excluded from this analysis. Out of 273 patients after a primary duct-to-duct reconstruction, 20 (7.3%) developed late isolated AS. Seven of these patients failed interventional biliary treatment and required a surgical repair. RESULTS: Duct-to-duct reconstruction was feasible and successful in all patients. Liver function tests fully normalized and no patient required any form of biliary intervention after surgery. One patient with intraoperative cholangiosepsis was ICU bound for 5 days, and another patient with a subhepatic abscess required percutaneous drainage. There was no perioperative death. The median length of hospital stay was 8 (5-17) days. The median time of follow-up after relaparotomy was 1593 (434-2495) days. CONCLUSION: Duct-to-duct reconstruction is a feasible and safe option in selected patients requiring surgical repair for isolated AS after DDLT. This approach preserves the biliary anatomy and avoids the potential side effects of a bilio-enteric diversion.

Management of early hepatocellular carcinoma: results of the Delphi consensus process of the Americas Hepato-Pancreato-Biliary Association.

BACKGROUND: There are many potential treatment options for patients with early stage hepatocellular carcinoma (HCC) and practice patterns vary widely. This project aimed to use a Delphi conference to generate consensus regarding the management of small resectable HCC. METHODS: A base case was established with review by members of AHPBA Research Committee. The Delphi panel of experts reviewed the literature and scored clinical case statements to identify areas of agreement and disagreement. Following initial scoring, discussion was undertaken, questions were amended, and scoring was repeated. This cycle was repeated until no further likelihood of reaching consensus existed. RESULTS: The panel achieved agreement or disagreement consensus regarding 27 statements. The overarching themes included that resection, ablation, transplantation, or any locoregional therapy as a bridge to transplant were all appropriate modalities for early or recurrent HCC. For larger lesions, consensus was reached that radiofrequency ablation and microwave ablation were not appropriate treatments. CONCLUSION: Using a validated system for identifying consensus, an expert panel agreed that multiple treatment modalities are appropriate for early stage HCC. These consensus guidelines are intended to help guide physicians through treatment modalities for early HCC; however, clinical decisions should continue to be made on a patient-specific basis.

Management of disappearing colorectal liver metastases: an international survey.

BACKGROUND: Improved chemotherapy response rates have lead to "disappearing" colorectal liver metastases (dCRLM). We aim to assess management patterns of dCRLM from an international body of hepatobiliary surgeons. METHODS: A survey was designed, tested for item relevance, readability and content validity, and distributed to the AHPBA, IHPBA and ANZHPBA. RESULTS: The majority of 226 respondents were <15 years from training (69%), practiced in academia (82%) and devoted >50% of their practice to hepatobiliary (75%). Surgeons utilize CT(45%) or MRI(47%) for preoperative planning with a preferred imaging interval of <6 weeks. Nearly all have experienced dCRLM (99%) and 63% of surgeons have waited a few months to assess for durability of response prior to definitive surgical/ablative therapy. Only 24% place fiducial markers for lesions <1-cm prior to neoadjuvant chemotherapy. Intra-operatively, 97% of surgeons perform ultrasound, and 71% ablation. When a tumor has "disappeared," 49% elect for observation and 31% resect if the dCRLM is superficial. Of those electing observation, 87% believe there is effective treatment with progression on surveillance imaging. CONCLUSIONS: Nearly all surgeons have experienced dCRLM with half choosing observation over intervention due to the belief that these lesions may be re-addressed in the future.

Liver Injury in Liver Transplant Recipients With Coronavirus Disease 2019 (COVID-19): U.S. Multicenter Experience.

BACKGROUND AND AIMS: Coronavirus disease 2019 (COVID-19) is associated with liver injury, but the prevalence and patterns of liver injury in liver transplantation (LT) recipients with COVID-19 are open for study. APPROACH AND RESULTS: We conducted a multicenter study in the United States of 112 adult LT recipients with COVID-19. Median age was 61 years (interquartile range, 20), 54.5% (n = 61) were male, and 39.3% (n = 44) Hispanic. Mortality rate was 22.3% (n = 25); 72.3% (n = 81) were hospitalized and 26.8% (n = 30) admitted to the intensive care unit (ICU). Analysis of peak values of alanine aminotransferase (ALT) during COVID-19 showed moderate liver injury (ALT 2-5× upper limit of normal [ULN]) in 22.2% (n = 18) and severe liver injury (ALT > 5× ULN) in 12.3% (n = 10). Compared to age- and sex-matched nontransplant patients with chronic liver disease and COVID-19 (n = 375), incidence of acute liver injury was lower in LT recipients (47.5% vs. 34.6%; P = 0.037). Variables associated with liver injury in LT recipients were younger age (P = 0.009; odds ratio [OR], 2.06; 95% confidence interval [CI], 1.20-3.54), Hispanic ethnicity (P = 0.011; OR, 6.01; 95% CI, 1.51-23.9), metabolic syndrome (P = 0.016; OR, 5.87; 95% CI, 1.38-24.99), vasopressor use (P = 0.018; OR, 7.34; 95% CI, 1.39-38.52), and antibiotic use (P = 0.046; OR, 6.93; 95% CI, 1.04-46.26). Reduction in immunosuppression (49.4%) was not associated with liver injury (P = 0.156) or mortality (P = 0.084). Liver injury during COVID-19 was significantly associated with mortality (P = 0.007; OR, 6.91; 95% CI, 1.68-28.48) and ICU admission (P = 0.007; OR, 7.93; 95% CI, 1.75-35.69) in LT recipients. CONCLUSIONS: Liver injury is associated with higher mortality and ICU admission in LT recipients with COVID-19. Hence, monitoring liver enzymes closely can help in early identification of patients at risk for adverse outcomes. Reduction of immunosuppression during COVID-19 did not increase risk for mortality or graft failure.

Toothpick in the porta: Recurrent liver abscesses secondary to transgastric migration of a toothpick with successful surgical exploration retrieval.

We present a rare case of a 72-year-old man with recurrent hepatic abscesses secondary to transgastric migration of a toothpick into the liver parenchyma and left portal venous branch. Prior to identification of the foreign body, the patient received multiple courses of antibiotics and underwent image-guided catheter placement without resolution of infection. Given his refractory abdominal pain, fevers, and chills, a repeat abdominal CT was obtained and demonstrated a radio-opaque object extending through the prepyloric gastric submucosa into the liver parenchyma and left portal vein. EGD confirmed a pre-pyloric fistula tract with purulent discharge. The patient subsequently underwent exploratory laparotomy, cholecystectomy, porta hepatis exploration, removal of foreign body, and ligation of porto-enteric fistula tract. A wooden toothpick was removed in its entirety. Interval CT demonstrated resolution of hepatic abscesses and no evidence of persistent porto-enteric fistula. This exceptional case demonstrates the value of multidisciplinary care, hypervigilance for patients with refractory pyogenic liver abscesses of unknown origin, and the importance of careful preoperative planning.

Discovery and validation of a novel blood-based molecular biomarker of rejection following liver transplantation.

Noninvasive biomarker profiles of acute rejection (AR) could affect the management of liver transplant (LT) recipients. Peripheral blood was collected following LT for discovery (Northwestern University [NU]) and validation (National Institute of Allergy and Infectious Diseases Clinical Trials in Organ Transplantation [CTOT]-14 study). Blood gene profiling was paired with biopsies showing AR or ADNR (acute dysfunction no rejection) as well as stable graft function samples (Transplant eXcellent-TX). CTOT-14 subjects had serial collections prior to AR, ADNR, TX, and after AR treatment. NU cohort gene expression (46 AR, 45 TX) was analyzed using random forest models to generate a classifier training set (36 gene probe) distinguishing AR vs TX (area under the curve 0.92). The algorithm and threshold were locked and tested on the CTOT-14 validation cohort (14 AR, 50 TX), yielding an accuracy of 0.77, sensitivity 0.57, specificity 0.82, positive predictive value (PPV) 0.47, and negative predictive value (NPV) 0.87 for AR vs TX. The probability score line slopes were positive preceding AR, and negative preceding TX and non-AR (TX + ADNR) (P ≤ .001) and following AR treatment. In conclusion, we have developed a blood biomarker diagnostic for AR that can be detected prior to AR-associated graft injury as well a normal graft function (non-AR). Further studies are needed to evaluate its utility in precision-guided immunosuppression optimization following LT.

Discovery and Validation of a Biomarker Model (PRESERVE) Predictive of Renal Outcomes After Liver Transplantation.

BACKGROUND AND AIMS: A high proportion of patients develop chronic kidney disease (CKD) after liver transplantation (LT). We aimed to develop clinical/protein models to predict future glomerular filtration rate (GFR) deterioration in this population. APPROACH AND RESULTS: In independent multicenter discovery (CTOT14) and single-center validation (BUMC) cohorts, we analyzed kidney injury proteins in serum/plasma samples at month 3 after LT in recipients with preserved GFR who demonstrated subsequent GFR deterioration versus preservation by year 1 and year 5 in the BUMC cohort. In CTOT14, we also examined correlations between serial protein levels and GFR over the first year. A month 3 predictive model was constructed from clinical and protein level variables using the CTOT14 cohort (n = 60). Levels of ß-2 microglobulin and CD40 antigen and presence of hepatitis C virus (HCV) infection predicted early (year 1) GFR deterioration (area under the curve [AUC], 0.814). We observed excellent validation of this model (AUC, 0.801) in the BUMC cohort (n = 50) who had both early and late (year 5) GFR deterioration. At an optimal threshold, the model had the following performance characteristics in CTOT14 and BUMC, respectively: accuracy (0.75, 0.8), sensitivity (0.71, 0.67), specificity (0.78, 0.88), positive predictive value (0.74, 0.75), and negative predictive value (0.76, 0.82). In the serial CTOT14 analysis, several proteins, including ß-2 microglobulin and CD40, correlated with GFR changes over the first year. CONCLUSIONS: We have validated a clinical/protein model (PRESERVE) that early after LT can predict future renal deterioration versus preservation with high accuracy. This model may help select recipients at higher risk for subsequent CKD for early, proactive renal sparing strategies.

Peroxisomal footprint in the pathogenesis of nonalcoholic steatohepatitis.

Nonalcoholic steatohepatitis (NASH) is a form of fatty liver disease where benign hepatic steatosis leads to chronic inflammation in the steatotic liver of a patient without any history of alcohol abuse. Mechanisms underlying the progression of hepatic steatosis to NASH have long been investigated. This review outlines the potential role of peroxisomal dysfunctions in exacerbating the disease in NASH. Loss of peroxisomes as well as impaired peroxisomal functions have been demonstrated to occur in inflammatory conditions including NASH. Because peroxisomes and mitochondria co-operatively perform many metabolic functions including O2 and lipid metabolisms, a compromised peroxisomal biogenesis and function can potentially contribute to defective lipid and reactive oxygen species metabolism which in turn can lead the progression of disease in NASH. Impaired peroxisomal biogenesis and function may be due to the decreased expression of peroxisomal proliferator-activated receptor-α (PPAR-α), the major transcription factor of peroxisomal biogenesis. Recent studies indicate that the reduced expression of PPAR-α in NASH is correlated with the activation of the toll-like receptor-4 pathway (TLR-4). Further investigations are required to establish the mechanistic connection between the TLR-4 pathway and PPAR-α-dependent impaired biogenesis/function of peroxisomes in NASH.