Abstract
BACKGROUND: One of the problems in modern obstetrics is how to identify and select pregnant women who are most likely to give premature birth. In the last ten years, due to false-positive test results, i.e., tests with low positive predictive values, there is an increase in unnecessary hospitalization days as well as unnecessary therapy. The probability of preterm birth is 25% in a population of pregnant women with symptoms of preterm birth. OBJECTIVE: The aim was to analyze diagnostic accuracy of tests for the purpose of predicting premature births in< 37(th) and <34(th) week of pregnancy.Incidence of preterm births in < 37(th) week of pregnancy was 28%, while the incidence of preterm births up until 34(th) week of pregnancy, was < 8%. METHODS: We included two groups of pregnant women in a prospective study; one group with the symptoms of threatening preterm birth between 22(nd) and 37(th) week of pregnancyand the other one of the same gestation period with no symptoms. RESULTS: Each pregnant woman underwent test for placental alphamicroglobulin-1, cervical length screening, cervical sampling for microbiological analysis, blood sampling for IL6 and CRP analysis. There were 16% of preterm births, up until 7 days from hospitalization, and they were all PAMG-1 positive; There is 75% of preterm births if PAMG-1 is positive with cervical length under 25mm. Combining tests, we reached the best predictive accuracy with positive PAMG-1 test, cervical length under 15mm along with the increase of CRP values above 15.96%. CONCLUSION: Total number of hospitalization days was 29% with preterm births up to 71% with full term births regardless the symptomatology, which justifies further studies towards releasing the pressure from the health care system and from doctors as well in the process of reaching a decision on treatment of pregnant women with the signs of preterm birth.