Quality control of lymph node dissection in the Dutch randomized trial of D1 and D2 lymph node dissection for gastric...

BACKGROUND: Variability among surgeons and reduced protocol adherence threaten the conduct and outcome of surgical multicenter trials. We introduced, in the Dutch Gastric Cancer Trial of D1 and D2 (extended) lymph node dissection for gastric cancer, a novel way of managing instruction, quality control, and evaluation of protocol adherence.METHODS: Of 1078 patients entered in the Dutch trial, 711 patients with potentially curative resections were evaluated. Numbers and locations of lymph nodes detected at pathological investigation were compared according to the guidelines of the Japanese Research Society for the Study of Gastric Carcer. Non-compliance indicated inadequate removal of lymph node stations, whereas contamination indicated that lymph nodes were detected outside the intended level of dissection. Protocol adherence during the course of the trial, and the impact on complications, hospital mortality, and survival were evaluated.RESULTS: Major non-compliance was noted in 15.3% of D1 and 25.9% of D2 patients. Contamination was present in 22.9% of D1 and 23.5% of D2 patients, and was limited to one or two lymph node stations only. Intensification of quality control resulted in only a marginal improvement in protocol adherence and in the number of lymph nodes detected. There was no association between protocol adherence and the occurrence of complications or long term survival.CONCLUSIONS: Contamination proved an important parameter to substantiate protocol adherence by the surgeon, whereas non-compliance had a multifactorial cause. Non-adherence to the protocol did not lead to increased hospital morbidity and mortality, but also had no impact on long term survival.