Ivanecz, Arpad; Plahuta, Irena; Magdalenic, Tomislav; Ilijevec, Bojan; Mencinger, Matej; Perus, Iztok; Potrc, Stojan
Evaluation of the Iwate Model for Predicting the Difficulty of Laparoscopic Liver Resection: Does Tumor Size Matter?
Teil von
  • Journal of gastrointestinal surgery, 2020-06-03, Vol.25 (6), p.1451-1460
Ort / Verlag
Links zum Volltext
Springer Online Journals Complete
Background This study aimed to externally validate the Iwate scoring model and its prognostic value for predicting the risks of intra- and postoperative complications of laparoscopic liver resection. Methods Consecutive patients who underwent pure laparoscopic liver resection between 2008 and 2019 at a single tertiary center were included. The Iwate scores were calculated according to the original proposition (four difficulty levels based on six indices). Intra- and postoperative complications were compared across difficulty levels. Fitting the obtained data to the cumulative density function of the Weibull distribution and a linear function provided the mean risk curves for intra- and postoperative complications, respectively. Results The difficulty levels of 142 laparoscopic liver resections were scored as low, intermediate, advanced, and expert level in 41 (28.9%), 53 (37.3%), 32 (22.5%), and 16 (11.3%) patients, respectively. Intraoperative complications were detected in 26 (18.3%) patients and its rates (2.4%, 7.5%, 34.3%, and 62.5%) increased gradually with statistically significant values among difficulty levels (P < 0.001). Major postoperative complications occurred in 21 (14.8%) patients and its rates (4.8%, 5.6%, 28.1%, 43.7%; P < 0.001) showed the same trend as for intraoperative complications. Then, the mean risk curves of both complications were obtained. Due to outliers, a new threshold for a tumor size index was proposed at 38 mm. The repeated analysis showed improved results. Conclusions The Iwate scoring model predicts the probability of complications across difficulty levels. Our proposed tumor size threshold (38 mm) improves the quality of the prediction. The model is upgraded by a probability of complications for every difficulty score.
ISSN: 1091-255X
ISSN: 1873-4626
DOI: 10.1007/s11605-020-04657-9

Weiterführende Literatur

Empfehlungen zum selben Thema automatisch vorgeschlagen von bX
Die Universität der Informationsgesellschaft