Validation and applicability of the Tampa Difficulty Score for assessing procedural complexity in robotic liver surgery

BACKGROUND: With the increasing adoption of robotic systems in hepatic surgery, standardized assessment of procedural difficulty has become crucial, particularly for case selection in training and education. The Tampa Difficulty Score (TDS) was developed to classify robotic liver resections into four levels of complexity. This study aimed to externally validate the TDS. METHODS: Seventy-nine consecutive patients undergoing robotic liver resection between 2018 and 2024 were included in this retrospective single-center study. Group comparisons between TDS categories were performed using descriptive and inferential statistics including analysis of variance via the Kruskal-Wallis test, Chi-square test, Mann-Whitney-U test, or Fisher's exact test, as appropriate. Effect sizes were calculated accordingly. Post hoc analyses for intergrupal differences were conducted using the Kruskal-Wallis test. Spearman's rank correlation assessed linear associations between TDS and perioperative variables. RESULTS: Significant intergroup differences between TDS categories were found for major resections (p < 0.001), operative time (p < 0.001), blood loss (p < 0.001), intensive care stay (p = 0.002), hospital stay (p = 0.005), specimen weight (p < 0.001) and tumor size (p = 0.012). Post hoc analyses with Bonferroni-Holm correction confirmed significant differences mainly between TDS groups 2 and 3 for key parameters, but also showed homogenous distribution throughout the other TDS groups. Strong positive correlations were observed between TDS and major resections (|ρ| = 0.651), operative time (|ρ| = 0.715), blood loss (|ρ| = 0.507), and specimen weight (|ρ| = 0.578; all p < 0.001). CONCLUSION: The Tampa Difficulty Score was validated for RLR, showing consistent results in both intergroup comparisons and correlation analyses. TDS proved to be a reliable instrument for assessing procedural difficulty, demonstrating its value for risk assessment, training standardization, as well as future evaluations despite the limited sample size.