(P 32) IMPROVEMENT OF QUALITY AND TEXTBOOK OUTCOMES OF GASTRIC GASTROINTESTINAL STROMAL TUMOR (GIST) RESECTIONS IN THE PAST 13 YEARS; THE INFLUENCE OF MINIMAL INVASIVE SURGERY (MIS)

Objective: The Dutch gastrointestinal stromal tumor (GIST) consortium, comprising five GIST expert centers, established the prospective Dutch GIST Registry (DGR) in 2009 to facilitate research and to enhance the quality of GIST care. Changes in the quality of resections of primary localized gastric GIST has not been described so far. This study aims to describe the differences observed over time for this patient population in indications for surgical resections and surgical outcomes. GIST-specific textbook outcomes (TO) were used as composite measure for surgical outcomes.

Methods: All patients with primary surgically treated localized gastric GISTs recorded in the DGR between January 2009 and January 2022 were included in this study. Metastasized disease at time of surgery was an exclusion criteria. Patient and tumor characteristics were analyzed for all included cases. The analysis of surgical outcomes focused on patients of which the resection was performed in an expert center. To evaluate the variation in quality over time, patients were segregated into four equal periods. TO was defined as clear margins, no perioperative tumor rupture, no surgical complication (Clavien Dindo ≥2), no resurgery or readmission within 90 days after resection, and no prolonged hospital stay (>75th percentile). Multivariate regression analyses were done to search for factors influencing the perioperative outcomes.

Results: A total of 472 patients were included in the study. Patient and tumor characteristics remained consistent over time, except for the median age (62 vs. 65 vs. 68 vs. 68, p = 0.002) and the proportion of minimal invasive surgery (MIS) per period (7% vs. 17% vs. 42% vs. 60%, p < 0.001). Of the included patients, 385 (81.6%) underwent surgery in an expert center. The proportion of achieved TO increased significantly over time (54% vs. 53% vs. 66% vs. 76%, p < 0.001). Multivariate regression analyses showed that achieving TO was significantly correlated with a later period of surgery (OR 1.68 [CI 95% 1.28-2.23], p < 0.001), lower age (OR 0.97 [CI 95% 0.95-0.99], p = 0.004), MIS (OR 0.97 [CI 95% 0.95-0.99], p = 0.004), and lower ASA scores (OR 0.65 [CI 95% 0.44-0.97], p = 0.035). Furthermore, a decrease was seen in the median duration of surgery in minutes (92 vs. 94 vs. 77 vs. 73, p=0.007) and the median length of hospital stay following surgery in days (6 vs. 6 vs. 5 vs. 4, p< 0.001).

Conclusion: The quality of perioperative outcomes for resections of gastric GIST in the five Dutch expert centers has improved over time. Furthermore, its efficiency has increased leading to a decreased duration of surgery and hospital stay.