EAES Academy

Aims: Although a high-quality lymphadenectomy during gastrectomy for gastric cancer is essential, clear definitions of high-quality and accurate assessment standardization methods are not widely acknowledged. This side-study assessed the surgical quality of the D2-lymphadenectomy for gastric cancer in the LOGICA-trial.
Methods: The LOGICA-trial is a multicenter randomized trial comparing laparoscopic versus open gastrectomy with D2-lymphadenectomy for resectable (cT1-4aN0M0) gastric cancer patients. Each lymph node station (no. 1-9, 11 and 12a) was collected in separate pathology containers for pathological assessment. Intraoperative photographs (figure 1) of the D2-lymphadenectomy were prospectively scored by two expert surgeons independently during inclusion on a weekly basis to provide feedback to the operating center. The dissection quality of stations no. 8+9, 11 and 12a was scored on a 4-point scale ranging from optimal (3) to unevaluable (0). Consensus was reached in case of disagreement. Pathological lymph node yield, as counted in the resected specimen, was compared to the prospective lymphadenectomy scoring with multivariable linear regression analyses. X2-tests were used to compare the quality scores between open and laparoscopic photographs. The interobserver variability was calculated using weighted Kappa (κw) to measure degree of disagreement.
Results: Between 2015–2018, 111 (52%) of the 212 LOGICA-patients had assessable intraoperative photographs. Initial scores of the photographs showed a moderate agreement between both expert surgeons for stations no. 8+9 (κw=0.522; 95% confidence interval (CI): 0.346-0.699; p<0.0005), station 11 (κw=0.446; 95% CI: 0.293-0.600; p<0.0005) and station 12 (κw=0.447; 95% CI 0.280-0.614; p<0.0005). For those cases that were discrepant, the 2 observers scored again through a consensus meeting. Multivariable linear regression analyses revealed no correlation (p>0.05) between surgical quality scores and lymph node yield. Laparoscopic photographs could be assessed better compared to open photographs (station 8+9; p=0.006, station 11; p=0.004 and station 12a; p=0.011).
Conclusion(s): Overall, moderate agreement (κw=0.41-0.60) on surgical quality of the D2-lymphadenectomy was scored by two expert surgeons independently. The intraoperative photograph assessment of the lymphadenectomy was not correlated with lymph node yield. This 4-point photograph scoring system may be a method to standardize assessment of the surgical quality of the lymphadenectomy, especially for laparoscopic approach.