Aim: The objective of this study is to determine the predictive role of cervical length, cervicovaginal irrigation liquid vitronectin andprolactin levels in preterm birth.Material and Methods: A total of 73 pregnant women between 24-34th gestational week were included and the study populationwas divided into two groups as: term birth group (n=54) and preterm birth(n=19). Birth weeks, weight and methods were noted. Theirrigation fluid has been collected from posterior fornix and vitronectin levels are detected by Enzyme-linked-immunosorbent-assaymethod.Results: Vitronectin were higher in preterm group as compared to controls [90 (78-98) vs 16 (9-41) ng/ml, p<0.001]. Positive predictivevalue for vitronectin was 78 %, negative predictive value was 100 % and accuracy ratio was 91.8%. The sensitivity of vitronectin was100 %, specifity was 89 % with a cut-off value of 59 ng/mL. Area under curve for vitronectin was 0.96 which was compatible with‘perfect’. When average prolactin levels have been found as 0.14 ng/mL in term and 2 ng/mL in preterm birth group, cervical lengthhave been found as 34.5 and 24.2 mm respectively. Cut-off value for prolactin has been found as 0.088 ng/dL. Positive predictivevalue was 58.3 % for prolactin and 51.5 % for cervical length. Negative predictive value was 89.8% for prolactin and 95 % for cervicallength. Vitronectin was strongly positively correlated with prolactin levels while it was negatively correlated with cervical length(r=0.691 and r=-0.348 respectively).Conclusion: Many trials have been done to make the most accurate prediction of preterm delivery risk. By this study, it makes usthink about that vitronectin could be a valuable marker for preterm birth, and also could be independent from the gestational week.