EMR versus gastrectomy for intramucosal gastric cancer: comparison of long-term outcomes.

BACKGROUND

Limited data exist regarding the long-term outcomes of EMR compared with gastrectomy.

OBJECTIVE

To compare the long-term outcomes after EMR and surgery.

DESIGN

Retrospective analysis with propensity-score matching.

SETTING

Tertiary care center.

PATIENTS

This study involved 215 patients with intramucosal gastric cancer completely removed by EMR and 843 patients who underwent curative surgical resection between January 1997 and August 2002. Propensity-score matching yielded 551 matched patients.

INTERVENTIONS

EMR versus surgery.

MAIN OUTCOME MEASUREMENTS

Death and recurrence.

RESULTS

In the matched cohort, there were no significant between-group differences in the risk of death (hazard ratio [HR] for the EMR group 1.39; 95% CI, 0.87-2.23) or recurrence (HR 1.18; 95% CI, 0.22-6.35). Although patients who underwent EMR had higher risk of metachronous gastric cancers (HR 6.72; 95% CI, 2.00-22.58), all recurrent or metachronous gastric cancers after EMR were successfully re-treated without affecting overall survival. Although complication rates were similar (odds ratio 0.84; 95% CI, 0.41-1.70), there were no mortalities in the EMR group compared with 2 in the surgery group. The EMR group had a significantly shorter hospital stay (median 8 days, interquartile range [IQR] 6-11 days vs 15 days, IQR 12-19 days; P<.001) and lower cost of care ($2049, IQR $1586-2425 vs $4042, IQR $3458-4959; P<.001).

LIMITATIONS

Retrospective, nonrandomized study.

CONCLUSIONS

EMR was comparable to surgery in terms of risk of death and recurrence. Because of its lower medical costs and shorter duration of hospital stay, EMR has advantages over surgery.