Introduction: The purpose of this study was to evaluate the relationship between the beta-hCG cervicovaginal secretion and preterm delivery.Material & Methods: In this study a cross-sectional analysis was used and cervicovaginal secretion specimens were obtained from women in their 24th- 36thweeks (each 6 days) of pregnancy who visited the emergency clinic with preterm labor symptoms. Exclusion factors include: ruptured preterm (PROM), poly hydroamnios, multiple gestation, abruptio placental, placenta previa, cervical cerglege, amniocentesis, bleeding, coitus during the past 24 hours, smoking, systemic diseases, gestational hypertension or pre eclampsia, history of ovarian cyst, chorioamnionitis, use of corticosteroid before sampling, fetal abnormalities, intra uterine growth restriction, use of lubricant or vaginal cream during past 24 hours and dilatation more than 3 cm. Samples were frozen at -200 centigrade and transferred to laboratory immediately and examined with Radio immunoassay method in 72 hours. Then, the patients were explored for the occurrence of preterm or term delivery. Results: The results showed that mean of Beta hCG level in cervicovaginal secretion was higher in group 1 (preterm labor and delivery) than group 2 (preterm labor and term delivery) and group 3 (control). The mean± standard error for human chorionic gonadotropin level in 24th- 26th weeks in groups 1, 2, 3 were (33.5 ±10.16), (7.12±6.46) and (0), in 27th - 29th weeks in groups 1,2,3 was (30.28±8.23) (23±13.44) and (7.80±4.74), 30th - 32nd weeks in group 1, 2, 3 was (30.20±2.23), (18.85±5.65) and (16.25±8.57), 33th - 35th weeks in group 1, 2, 3 was (36.90±5.98), (22.32±6.35) and (10.70±4.67), 36th weeks in group 1,2,3 was (38.76±10.99), (20.05±19.73) and (8.42±5.89), respectively. According to the results obtained in weeks gestation, the range of values for human chorionic gonadotropin in three pregnant. groups was extensive and Beta hCG Level mean became more extensive in 24th -26th weeks. ROC curve showed that Beta hCG cervicovaginal level, equal to or more than 22.5milli unit per milliliter, is associated with preterm delivery (97% sensitivity, 76% specifity, 81% positive predictive value, 96% negative predictive value). However the cut off for groups 2 and 3 was 18 mlu/ml (100% sensitivity, 93% specifity, 94% positive and 100% negative predictive value, respectively). However, the cut off for the second and third groups was 18 mlu/ml (48% sensitivity, 93% sepcifity, 87% positive and 62% negative, respectively). Conclusion: Measuring the b-hCGlevel in cervicovaginalsecretionsof the patientscan predictpreterm delivery.