In recent years, both dienogest, a fourth-generation progestin, and levonorgestrel-releasing intrauterine device have been found to be effective in terms of endometriosis-related pelvic pain improvement. No data, however, are available about their efficacy in terms of postoperative recurrence prevention in women recently submitted to surgery for endometriosis at every stage. Our objective was to compare two postoperative medical approaches for pain control and reduction of recurrences in patients undergoing surgery for endometriosis. Ninety-two patients undergoing surgery for endometriosis and subsequent treatment by estradiol valerate + dienogest (EP) (group A) or levonorgestrel-releasing uterine device (LNG-IUD) (Group B) between January 2009 and June 2010 were retrospectively analyzed. The primary endpoints were pain relapse and disease recurrence rate at 12 and 24 months. The secondary endpoint was patient satisfaction with the therapy at 24 months. Forty-eight patients in Group A and 44 in Group B were examined. At 12 and 24 months a statistically greater reduction both in Ca125 levels and VAS score was seen in women treated with EP compared to LNG-IUD. The recurrence rate at 12 and 24 months follow-up was slightly lower, but not at a significant level, in Group A than in Group B. Finally, satisfaction with treatment at 24 months was significantly higher in Group B. EP administration is significantly more effective than LNG-IUD in reducing pelvic pain and more effective in reducing recurrence rate but not at a significant level. LNG-IUD, however, has significantly higher patient satisfaction.