AbstractBackground and AimPost‐hepatectomy liver failure (PHLF) after major hepatopancreatoduodenectomy (HPD) is a challenge to overcome. However, the appropriate target proportion of the future liver remnant (pFLR) to prevent severe PHLF in major HPD remains uncertain. This study aimed to determine the minimum pFLR required for safe major HPD.MethodsThis retrospective study involved 48 major HPD patients. We assessed pFLR and remnant liver function scores (pFLR × albumin‐bilirubin [ALBI] / albumin‐indocyanine green evaluation [ALICE]/plasma clearance rate of indocyanine green [KICG]) as predictors for Grade B/C PHLF and established safety criteria.ResultsGrade B/C PHLF occurred in 40% of the patients (n = 19), leading to severe morbidity and two in‐hospital deaths. pFLR was a good predictor of Grade B/C PHLF [area under the curve (AUC) 0.80, p < 0.01] with a 45% optimal cutoff. While all remnant liver function scores predicted PHLF, the remnant ALICE demonstrated the best predictability (AUC 0.85, p < 0.01), with the sensitivity and specificity at 89% and 83%, respectively, using −0.86 as the cutoff. Independent risk factors for Grade B/C PHLF were remnant ALICE ≥−0.86 and blood loss ≥1500 mL. Grade B/C PHLF developed in 14% with pFLR ≥45% but reached 64% with pFLR <45%. However, the rate could be reduced to 33% with remnant ALICE <−0.86.ConclusionTo prevent Grade B/C PHLF, a pFLR ≥45% is recommended. Nevertheless, major HPD may be considered in patients with good remnant liver function.