Abstract
BACKGROUND: Hepatoblastoma is the most common primary pediatric liver malignancy. Indocyanine green (ICG) has been described as an adjunct to resection in small series. Its utility remains undefined in larger cohorts. METHODS: Records for 29 patients diagnosed with hepatoblastoma who received ICG prior to surgical resection from 2017 to 2020 at a single institution were retrospectively reviewed. The primary outcome was correlation between intraoperative ICG-avidity and histologic presence of hepatoblastoma. A secondary outcome included the histologic margin designation for resected liver specimens. RESULTS: ICG sensitivity was 91% for 120 resected thoracic specimens from 21 patients. Specificity was 57%. In 10% of operations, HB-positive specimens were resected solely on ICG-avidity. In an additional 40% of cases, ICG assisted in localizing a preoperatively diagnosed lesion. ICG sensitivity during thoracotomy and thoracoscopic surgery was 95 and 74%, respectively; primary and relapsed disease demonstrated sensitivity of 94 and 73%, respectively. Sensitivity was 92% for 25 resected liver specimens from nine patients with all parenchymal margins grossly negative for disease. Four multifocal lesions were identified with two resected solely by ICG-avidity. CONCLUSIONS: ICG is a sensitive adjunct for identifying local and metastatic hepatoblastoma, including lesions not visualized on preoperative imaging, and delineating margins during liver resection. False positives limit specificity; however, there were no adverse outcomes from additional resections. We noted that thoracoscopic surgery can be completed safely in patients with less significant disease burden, and conversion to thoracotomy, if necessary, is straightforward.