# Application of propensity score matching in prognostic analysis of portal hypertension in hepatocellular carcinoma patients

## TL;DR

This study examines how portal hypertension affects outcomes in liver cancer patients who undergo surgery and whether splenic procedures improve survival.

## Contribution

The study introduces the use of propensity score matching to assess the impact of portal hypertension and splenic modulation on hepatocellular carcinoma surgical outcomes.

## Key findings

- PHT patients had worse short-term surgical outcomes, including increased blood loss and longer hospital stays.
- PHT was not linked to worse long-term survival, but splenic modulation improved recurrence-free survival in PHT patients.
- Non-PHT patients undergoing non-anatomical hepatectomy had better recurrence-free survival.

## Abstract

To investigate the impact of portal hypertension on surgical prognosis after hepatectomy for hepatocellular carcinoma and assess the therapeutic value of concomitant splenic modulation procedures.

We retrospectively analyzed HCC patients who underwent open hepatectomy with intraoperative portal venous pressure (PVP) measurement at our center between January 2013 and January 2020. Portal hypertension (PHT) was defined as PVP ≥ 25 cm H2O. Patients were categorized as PHT (n = 88) or non-PHT (n = 642). Propensity score matching (1:1) was performed to balance baseline covariates; matched analyses included 59 pairs. Short-term perioperative outcomes and long-term overall survival (OS) and recurrence-free survival (RFS) were compared between groups. Within the PHT cohort, outcomes were compared between hepatectomy alone and hepatectomy combined with splenectomy or splenic artery ligation.

Post-PSM analysis (59 matched pairs) revealed worse short-term outcomes in the PHT group: shorter surgical duration (p < 0.05) but greater intraoperative blood loss (anatomical/extensive hepatectomy subgroups, p < 0.01), higher postoperative bilirubin levels (p < 0.05), and longer hospital stay (non-anatomical/non-extensive hepatectomy subgroups, p < 0.01). Long-term OS and RFS showed no differences between groups pre- or post-PSM (p > 0.05). However, subgroup analyses demonstrated superior RFS in non-PHT patients undergoing non-anatomical and non-extensive hepatectomy (p = 0.035/0.034). Notably, pre-PSM data indicated improved RFS and OS in PHT patients receiving concomitant splenectomy or splenic artery ligation versus hepatectomy alone (p < 0.001).

Compared with non-PHT patients, PHT was not associated with additional risk factors for poor prognosis after surgery. However, PHT may represent a significant risk indicator for recurrence in HCC patients undergoing non-anatomical or non-extensive hepatectomy. Furthermore, for HCC patients with concomitant PHT, hepatectomy combined with splenic artery ligation or splenectomy was associated with better long-term survival.