AIM The study aimed to demonstrate that, owing to a documented luteal phase deficiency, natural progesterone treatment improves expectations for the continuity of pregnancy in patients with a high risk of spontaneous abortion in the first trimester. The endovaginal administration of this hormone in women suffering from multiple abortions during the first trimester may improve the prognosis in terms of the continuation of the pregnancy beyond the 12th week of gestation. METHODS A study was carried out to compare two groups of patients: a group consisting of 19 patients underwent natural progesterone treatment in the form of a vaginal cream (ESOLUT) at a dose of 100 mg a day. The second group, consisting of 15 patients, formed a control group and did not receive any additional progestin treatment. The study was carried out between May 1998 and August 1999 at the 2nd Clinic of Obstetrics and Gynecology at the University of Milan, in the Clinica "L. Mangiagalli", on a total of 34 patients suffering from recurrent spontaneous abortion during the first trimester and luteal phase deficiency. No other risk factors for abortion were noted at the initial screening. The deficiency was confirmed during screening using three blood samples assayed for plasma progesterone during the second phase of the menstrual cycle. In order to make a diagnosis of luteal phase deficiency, at least one of the three values had to be less than 10 ng/ml. These patients (19) received additional endovaginal progesterone therapy. RESULTS A lower incidence of spontaneous abortion during the first trimester was noted at the end of the study in treated patients compared to the control group (21% vs 16%). CONCLUSIONS These results lead us to conclude that natural progesterone treatment in patients with a confirmed hormone deficiency during the luteal phase does not reduce the risk of abortion in the overall population of pregnant women, but improves the prognosis compared to the group of women suffering from multiple abortions, thereby reducing the risk to that in the general population.