Major influence of liver function itself but not of immunosuppression determines glucose tolerance after living-donor liver transplantation

Prediction of New-Onset Diabetes Mellitus within 12 Months after Liver Transplantation-A Machine Learning Approach

BACKGROUND

Liver transplantation (LT) is a routine therapeutic approach for patients with acute liver failure, end-stage liver disease and/or early-stage liver cancer. While 5-year survival rates have increased to over 80%, long-term outcomes are critically influenced by extrahepatic sequelae of LT and immunosuppressive therapy, including diabetes mellitus (DM). In this study, we used machine learning (ML) to predict the probability of new-onset DM following LT.

METHODS

A cohort of 216 LT patients was identified from the Disease Analyzer (DA) database (IQVIA) between 2005 and 2020. Three ML models comprising random forest (RF), logistic regression (LR), and eXtreme Gradient Boosting (XGBoost) were tested as predictors of new-onset DM within 12 months after LT.

RESULTS

18 out of 216 LT patients (8.3%) were diagnosed with DM within 12 months after the index date. The performance of the RF model in predicting the development of DM was the highest (accuracy = 79.5%, AUC 77.5%). It correctly identified 75.0% of the DM patients and 80.0% of the non-DM patients in the testing dataset. In terms of predictive variables, patients' age, frequency and time of proton pump inhibitor prescription as well as prescriptions of analgesics, immunosuppressants, vitamin D, and two antibiotic drugs (broad spectrum penicillins, fluocinolone) were identified.

CONCLUSIONS

Pending external validation, our data suggest that ML models can be used to predict the occurrence of new-onset DM following LT. Such tools could help to identify LT patients at risk of unfavorable outcomes and to implement respective clinical strategies of prevention.

Liver Transplantation: the Role of Metabolic Syndrome

OPINION STATEMENT

The long-term survival in liver transplant recipients (LTRs) is currently at an historical high level stemming from improvement in perioperative care, infection control, and immunosuppression medications. However, compared to the general population, LTRs have decreased survival. Metabolic diseases like hypertension, dyslipidemia, type 2 diabetes, and obesity are key determinants of long-term mortality in LTRs. The incidence and prevalence of these metabolic comorbidities is considerably higher in LTRs and likely results from a combination of factors including exposure to chronic immunosuppression, weight gain, and recurrence of chronic liver disease after liver transplantation (LT). Although there is currently little guidance in managing these metabolic conditions post-LT, recommendations are often extrapolated from non-transplant cohorts. In the current review, we explore the relationship between metabolic syndrome and its comorbidities in LTRs.

β-cell function prior to liver transplantation contributes to post-operative diabetes

Liver cirrhosis and diabetes mellitus (DM) are closely associated. The present study aimed to determine whether liver transplantation (LT) may prevent/cure DM in patients with cirrhosis and whether the degree of glucose tolerance prior to transplantation is associated with the onset of DM after transplantation. Seventy-three patients who received a living donor LT at Nagasaki University Hospital (Nagasaki, Japan) between November 2005 and December 2012 were recruited. Among them, patients were considered diabetic if they had been prescribed diabetes medications or had impaired glucose tolerance, as evidenced by an oral glucose tolerance test (OGTT). Patients were followed up until December 31, 2013 to evaluate glucose tolerance. Patients who had developed DM 2 years after transplantation were found to be older and the incidence of diabetes prior to transplantation (n=73) was higher than in those who did not. Multivariate analysis revealed that DM requiring treatment prior to transplantation was the only independent factor for DM developed at 2 years after transplantation. OGTT results showed that in patients with poor insulin sensitivity indices prior to transplantation (n=45), improvements were seen at 2 years after transplantation, while β-cell function and insulinogenic index had decreased, which may have been the cause of DM after transplantation. In conclusion, the pre-operative β-cell function determined by an OGTT may be a useful predictive tool for the recurrence of DM after LT.

The Origin of New-Onset Diabetes After Liver Transplantation: Liver, Islets, or Gut?

New-onset diabetes is a frequent complication after solid organ transplantation. Although a number of common factors are associated with the disease, including recipient age, body mass index, hepatitis C infection, and use of immunosuppressive drugs, new-onset diabetes after liver transplantation (NODALT) has the following unique aspects and thus needs to be considered its own entity. First, a liver graft becomes the patient's primary metabolic regulator after liver transplantation, but this would not be the case for kidney or other grafts. The metabolic states, as well as the genetics of the graft, play crucial roles in the development of NODALT. Second, dysfunction of the islets of Langerhans is common in cirrhotic patients and would be exacerbated by immunosuppressive agents, particularly calcineurin inhibitors. On the other hand, minimized immunosuppressive protocols have been widely advocated in liver transplantation because of liver tolerance (immune privilege). Third and last, through the "gut-liver axis," graft function is closely linked to gut microbiota, which is now considered an important metabolic organ and known to independently influence the host's metabolic homeostasis. Liver transplant recipients present with specific gut microbiota that may be prone to trigger metabolic disorders. In this review, we proposed 3 possible sites for the origin of NODALT, which are liver, islets, and gut, to help elucidate the underlying mechanism of NODALT.

Metabolic complications in liver transplant recipients

The metabolic syndrome (MS), which includes obesity, dyslipidaemia, hypertension and hyperglycaemia according to the most widely accepted definitions now used, is one of the most common post-transplant complications, with a prevalence of 44%-58%. The MS, together with the immunosuppression, is considered the main risk factor for the development of cardiovascular disease (CVD) in transplant recipients, which in turn accounts for 19%-42% of all deaths unrelated to the graft. The presence of MS represents a relative risk for the development of CVD and death of 1.78. On the other hand, non-alcoholic fatty liver disease (NAFLD), considered as the manifestation of the MS in the liver, is now the second leading reason for liver transplantation in the United States after hepatitis C and alcohol. NAFLD has a high rate of recurrence in the liver graft and a direct relation with the worsening of other metabolic disorders, such as insulin resistance or diabetes mellitus. Consequently, it is vitally important to identify and treat as soon as possible such modifiable factors as hypertension, overweight, hyperlipidaemia or diabetes in transplanted patients to thus minimise the impact on patient survival. Additionally, steroid-free regimens are favoured, with minimal immunosuppression to limit the possible effects on the development of the MS.

Risk factors of metabolic disorders after liver transplantation: an analysis of data from fasted patients

BACKGROUND

Metabolic disorders are common complications after orthotopic liver transplantation (OLT) and may lead to increased morbidity and mortality.

METHODS

Fasting glucose and lipid metabolism, and body weight of 81 patients undergoing primary OLT were prospectively analyzed. Patients were investigated preoperatively, on postoperative days 1, 3, 5, 10, 14, 28 as well as 6 months and 1 year after OLT. Data of nonfasted patients were excluded from the analysis. Standardized definitions and classifications for diabetes mellitus (DM), body mass index (BMI), and dyslipidemia were used.

RESULTS

Prevalence of new-onset diabetes after transplantation was 9.3%, and obesity was its only independent risk factor (odds ratio [OR], 16.5). Preoperative impaired glucose homeostasis (OR, 10.8) and initial poor graft function (OR, 6.89) were independent risk factors for postoperative DM. Maximum prevalence of hypertriglyceridemia and hypercholesterolemia was found on postoperative day 10 and 6 months post-OLT, respectively. Risk factors for hypercholesterolemia at 1 year were patient age (OR, 1.17) and postoperative renal dysfunction (OR, 16.33). Higher preoperative BMI was a risk factor for postoperative hypertriglyceridemia (OR, 1.17). Overall body weight and BMI significantly decreased over 1 year (P < 0.05). Prevalence of obesity was 22.2% before and 20.9% after OLT.

CONCLUSION

For the first time, initial poor graft function was identified as a risk factor for post-OLT DM. By ruling out any exogenous factors influencing metabolism, we believe we were able to show the true prevalence of metabolic disorders and therefore provided a valuable contribution to the identification of potential risk factors.

Pre- and postoperative nutritional assessment and health-related quality of life in recipients of living donor liver transplantation

AIM

The nutritional state of living donor liver transplantation (LDLT) recipients is one of the most important factors affecting postoperative outcome. Although the assessment of health-related quality of life (HRQOL) is of increasing importance, few studies have examined this in conjunction with LDLT recipient nutritional state.

METHODS

Ten LDLT recipients with end-stage liver disease were recruited for this study. Measurements of energy expenditure, anthropometrics and laboratory data were performed before and 1, 6 and 12-24 months after LDLT. HRQOL was measured by using the 36-item Short-Form (SF-36) before and 1, 3, 6 and 12-24 months after LDLT.

RESULTS

The preoperative value of non-protein respiratory quotient (npRQ) was 0.796 ± 0.026 and it increased significantly after the operation. Serum non-esterified fatty acid (NEFA) levels were high in the preoperative state, but had significantly decreased 1 month after the operation. A negative correlation between npRQ and NEFA was observed throughout the study period. Cholinesterase and albumin levels improved to normal levels within 6 and 12-24 months, respectively. The recovery of the physical component summary of the SF-36 was observed after the improvement of all domains of laboratory data and energy metabolism based on the nutritional state.

CONCLUSION

This study demonstrated that the recovery of metabolic function, laboratory data and HRQOL in LDLT recipients are variable, and it took more than 6 months to normalize the liver protein synthetic capacity and physical HRQOL score periods. Therefore, long-term nutritional support is required in LDLT recipients.