Quantitative Evaluation of 99mTc-Galactosyl Human Serum Albumin SPECT/CT for the Prediction of Posthepatectomy Liver Failure

Purpose To evaluate the predictive ability of 99m Tc-galactosyl human serum albumin SPECT/CT quantitative parameters for posthepatectomy liver failure (PHLF). Methods Sixty-eight patients who underwent 99m Tc-galactosyl human serum albumin scintigraphy as a preoperative examination for hepatectomy between July 2021 and December 2023 were prospectively evaluated. The patients were divided into PHLF and non-PHLF groups. The clearance index (HH15) and receptor index (LHL15) were calculated from planar images. Quantitative parameters for the total liver (SUV max , SUV mean , peak SUV [SUV peak ], functional liver volume [FLV], SUV mean × FLV), remnant liver (r-SUV max , r-SUV mean , r-SUV peak , r-FLV, and r-SUV mean × r-FLV), and r-FLV/FLV were calculated from SPECT/CT images. The areas under the curve (AUCs) of parameters from planar images and SPECT/CT parameters of the total and remnant livers were calculated to evaluate the predictive ability of PHLF. Results PHLF occurred in 9 patients. AUC of r-FLV was significantly higher than that of HH15 (0.93; 95% confidence interval [CI], 0.84–0.99 vs 0.66; 95% CI, 0.46–0.84; P = 0.026) and LHL15 (0.93; 95% CI, 0.84–0.99 vs 0.72; 95% CI, 0.55–0.86; P = 0.043). AUC of r-SUV mean × r-FLV was significantly higher than that of HH15 (0.90; 95% CI, 0.80–0.97 vs 0.66; 95% CI, 0.46–0.84; P = 0.037). There were no significant differences between the AUC of other SPECT/CT, r-FLV/FLV, and planar image parameters. Conclusions r-FLV can predict PHLF more accurately than HH15 or LHL15.