• number of patients completing 36 weeks of pregnancy [ Time Frame: 4-8 weeks ] [ Designated as safety issue: No ]
  

• prolongation of gestational weeks [ Time Frame: 1-12 weeks ] [ Designated as safety issue: No ]
  

Patients who attend the antenatal care clinic at Assiut University Hospital were recruited and counseled about participating in the study. A written informed consent was taken. Patients had the right to refuse to participate and/or withdraw from the study at ay time without being denied their regular full clinical care. Personal information as well as data collected were subjected to confidentiality and were not made available to third party. Gestational age was determined on the basis of the last menstrual period, confirmed by an ultrasound between 14 weeks and 20 weeks gestation and calculated in menstrual weeks. Patients were included when they were pregnant +/- 28 to 32 weeks, having a documented history of at least one preterm labour in the previous pregnancy and having no uterine contractions at the time of the study. Patients were excluded if they refused to participate in the study, had prelabour premature rupture of membranes (PPROM), had an incompetent cervical os proved by funneling of the internal os on ultrasound examination or by a documented history with or without a cercelage done in the previous or current pregnancy. Patients with twin pregnancy, intrauterine foetal death, malpresentations, known fetal anomaly, progesterone or heparin treatment during the current pregnancy, hypertension requiring medication, a seizure disorder, Irregular and/or unsure menstrual dates or abortion within 3 months prior to the last menses were also excluded.

Clinical work-up included history taking, examination, obstetric ultrasound evaluation and Bio physical profile (BPP). All Patients were empirically treated for bacterial vaginosis prior to enrollment into one of the study groups by giving them oral metronidazole 250 mgm 3 times/day for one week.

After ending one course of the treatment all patients were subjected to a vaginal swab to exclude bacterial vaginosis. Criteria used for diagnosing BV were those of Amsel et al., 1983 by finding at least three of the four following criteria: 1) thin, dark or dull gray, homogenous, malodorous discharge that adheres to the vaginal wall; 2) elevated vaginal pH of greater than 4.5; 3) positive whiff/amine test, and 4) presence of clue cells on wet-mount microscopic evaluation. Patients with free samples were included in the study. Patients who proved an active or mixed infection were treated accordingly and excluded from the study. Patients were randomly allocated to two treatment groups. Group one, control group: given hydroxyl progesterone caproate (150mgm) every 3 days starting from the 28th week of pregnancy till completed 36 weeks of gestation. Group two patients were given hydroxyl progesterone caproate (150mgm) weekly injections plus NAC 0.6 gm (Sedico, Egypt) orally daily starting from the 28th week of pregnancy till delivery or completed 36th week of pregnancy. Patients were followed up by routine 2 week antenatal clinic visits for the occurrence of uterine contractions, their frequency, intensity and need for seeking medical advise. NAC or 17 hydroxyl progesterone caproate treatment was discontinued in both groups only if patient either completed 36 weeks or entered actively into labour (at least three uterine contractions, 40 seconds each, reaching 50 mmHg on external tocodynamometer monitoring and/or associated with progressive cervical dilatation reaching 5 cms and/or occurrence of rupture of membranes). Once established labour, a second vaginal swab was taken and subjected to microscopical examination and amine test to rule out newly developed BV. Outcomes included occurrence of contractions, prolongation of pregnancy, neonatal outcome in both groups.

Sample size:

Sample size was based on the findings of Iams et al., who found a high incidence of recurrent PTL among women who had positive findings with infection (64%). A reduction in incidence of PTL of 30% was considered acceptable. Sample size was calculated on a basis of 95% confidence interval, 80% power and 49 patients were needed in each arm.