Recent reports reveal that some threatened premature labors (tpl) are caused by intrauterine infections. And the prognosis of tpl due to intrauterine infections are worse than tpl due to other factors because of lower sensitivity to the beta-stimulants. So we examined the relationship between tpl and intrauterine infections retrospectively by obtaining intracervical cultures of inpatients who suffered from tpl, and prospectively by getting intracervical cultures from healthy pregnant women at a gestational age around 20 weeks, and divided them into two groups--the infectious group and non-infectious group--and studied the tpl rate. The patients who had positive bacterial cervical cultures were again divided into two groups. One group had antibiotics administered orally and the other had none. The tpl rate was observed. The results are: 1. Thirty two point six percent of tpl inpatients were found to be infected intracervically. No statistical significance was noted in the microbodies detected. 2. Three cases of premature labor at less than 35 gestational weeks were observed in the infected group. 3. The majority of infected group showed high CRP levels, and the study of Preterm Labor Score (PLS) showed that most cases over 4 points of PLS were infected. 4. In a prospective study, 18 of 100 healthy pregnant women showed positive intracervical bacterial cultures. The tpl rates were 44.4% in all of the infected group. Twenty two point two percent of tpl rate in the antibiotic administration group and 66.7% in the non-administration group (p less than 0.01). And the non-infected tpl rate was 11.0% (p less than 0.01), or only 9 cases out of 82.(ABSTRACT TRUNCATED AT 250 WORDS)