Prediction of the Total Liver Weight using anthropological clinical parameters: does complexity result in better accuracy?

BACKGROUND: The performance of linear models predicting Total Liver Weight (TLW) remains moderate. The use of more complex models such as Artificial Neural Network (ANN) and Generalized Additive Model (GAM) or including the variable "steatosis" may improve TLW prediction. This study aimed to assess the value of ANN and GAM and the influence of steatosis for predicting TLW. METHODS: Basic clinical and morphological variables of 1560 cadaveric donors for liver transplantation were randomly split into a training (2/3) and validation set (1/3). Linear models, ANN and GAM were built by using the training cohort and evaluated with the validation cohort. RESULTS: The TLW is subject to major variations among donors with similar morphological parameters. The performance of ANN and GAM were moderate and similar to that of linear models (concordance coefficient from 0.36 to 0.44). In 28-30% of cases, TLW cannot be predicted with a margin of error ≤20%. The addition of the variable "steatosis" to each model did not improve their performance. CONCLUSION: TLW prediction based on anthropological parameters carry a significant risk of error despite the use of more complex models. Others determinants of TLW need to be identified and imaging-based volumetric measurements should be preferred when feasible.

Comparison of complete pathologic response and hepatic injuries between hepatic arterial infusion and systemic administration of oxaliplatin in patients with colorectal liver metastases.

BACKGROUND: Whether hepatic arterial infusion (HAI) of oxaliplatin influences the rates of complete pathologic response (CPR) and severe oxaliplatin-related lesions (SOxL) in patients with colorectal liver metastases (CRLM) is unknown. This study aimed to compare the incidence of CPR and SOxL between systemic (intravenous, IV) and HAI administration. METHODS: All patients with initially unresectable CRLM who had undergone hepatic resection in two expert centers between 2004 and 2010 after at least 6 cycles of oxaliplatin-based chemotherapy administered either via HAI (n = 18) or IV (n = 50) were included. The presence of CPR and SOxL were evaluated by two pathologists. A 1:2 case match using a propensity score was used. RESULTS: A CPR was observed significantly more often after HAI (33 vs. 10 %, P = 0.03). However, SOxL had occurred more frequently in patients in the HAI group versus the IV group, 66 and 20 %, respectively (P < 0.001). On a well-balanced cohort, HAI was associated with higher chance of CPR (odds ratio 9.33, 95 % confidence interval 1.59-54.7) but also higher risk of SOxL (odds ratio 13.7, 95 % confidence interval 3.08-61.3). A CPR markedly enhanced overall survival (OS) and disease-free survival (median OS of 114 vs. 42 months, P = 0.02; median disease-free survival of 51 vs. 12 months, P = 0.002). Patients with SOxL did not experience different outcome (median OS of 42 vs. 50 months, respectively; P = 0.92) CONCLUSIONS: HAI of oxaliplatin increases the likelihood of a CPR at the cost of a higher incidence of SOxL in patients with initially unresectable CRLM.