TACE-induced liver injury in hepatocellular carcinoma: mechanisms, prediction, and prevention in the era of gut microbiota and inflammasome research.

Transarterial chemoembolization (TACE) is a cornerstone locoregional therapy for hepatocellular carcinoma (HCC), yet most candidates also have cirrhosis. That makes TACE a balancing act: controlling tumor progression while protecting a liver with limited functional reserve, often over multiple sessions. This narrative review synthesizes current evidence on TACE-induced liver injury (TACE-LI), with emphasis on its definitions, mechanisms, risk prediction, and prevention, and further contextualizes emerging data on the gut-liver axis and inflammasome signaling. Clinically, TACE-LI spans a spectrum from transient biochemical worsening and post-embolization syndrome (PES; fever, pain, malaise) to clinically meaningful decompensation and post-TACE liver failure (PTLF). A key challenge is that the literature uses heterogeneous definitions and thresholds for both TACE-LI and PTLF, complicating comparisons between studies and weakening clarity around retreatment decisions. Mechanistically, injury is multifactorial: arterial ischemia from embolization, ischemia-reperfusion and oxidative stress, local chemotherapeutic toxicity and material retention, biliary or microvascular injury, and a systemic inflammatory response that can drive both symptoms and laboratory derangement. Our review also highlights how cirrhosis-associated gut barrier dysfunction may prime inflammatory cascades through dysbiosis, bacterial translocation, and endotoxin signaling (e.g., LPS-TLR4), potentially shaping vulnerability after embolization. In this context, inflammasome pathways, especially NLRP3, are discussed as amplifiers that translate danger signals (DAMPs), reactive oxygen species, and microbial cues into IL-1β/IL-18 release and pyroptotic injury, linking sterile ischemic damage to immune activation. For risk stratification, baseline hepatic reserve (Child-Pugh, ALBI), tumor burden, and technical intensity remain central, and dynamic indicators such as post-TACE ALBI deterioration help capture clinically relevant trajectories. Our review also summarizes predictive tools such as nomograms and models for complications (including PES), and notes ongoing exploration of machine-learning-assisted decision support. Prevention ultimately rests on careful patient selection, liver-sparing technique (including superselective approaches when feasible), and structured peri-procedural care to mitigate PES and reduce infectious or biliary complications.