RESUMO
OBJECTIVE: To study the association of plasma and vaginal levels of matrix metalloproteinases 2 (MMP-2) and 9 (MMP-9), determined between 20+0 and 25+6 weeks of gestation, with preterm birth and its predictive value. STUDY DESIGN: All 129 cases of preterm birth (cases) and 258 randomly selected term patients (control group) were included in a cohort study with a nested case-control design. Inclusion criteria were living in Ribeirão Preto, Brazil, and gestational age 20+0 to 25+6 at enrollment. Exclusion criteria were multiple fetuses, congenital malformations or chromosome syndromes, and loss to follow-up. Maternal age, parity, smoking, and previous preterm birth were included as covariates. A blood sample and vaginal secretion were obtained for the determination of MMP-2 and MMP-9; the patients were screened for urinary tract infection and bacterial vaginosis, and cervical length was measured by ultrasound. The cut-off values for matrix metalloproteinases were calculated using receiver operating characteristic (ROC) curves for logistic regression analysis (crude and adjusted odds ratios). RESULTS: According to the WHO, in this study, preterm subtypes included 3.8 % extremely preterm, 6.9 % very preterm, and 89.2 % late preterm births. The plasma MMP-9 cut-off was 63.25 ng/mL and the area under the ROC curve was 0.725 (standard error 0.03; 95 % confidence interval, 0.677-0.769). The cut-off for plasma MMP-2 was 239.4 ng/mL and the area under the ROC curve was 0.585 (standard error 0.03, 95 % confidence interval, 0.521-0.649). Crude odds ratios showed an increased risk of preterm birth associated with plasma MMP-2 (odds ratio, 1.75; 95 % confidence interval, 1.14-2.68) and plasma MMP-9 (odds ratio, 3.26, 95 % confidence interval, 2.09-5.07); no association was detected for vaginal MMP-2 or 9. For plasma, adjusted odds ratios were 1.42 (95 % confidence interval, 0.80-2.53) for MMP-2 and 2.71 (95 % confidence interval, 1 .52-4.83) for MMP-9, along with an increased risk in other covariates. CONCLUSION: Elevated plasma MMP-9 levels and decreased MMP-2 levels were positively associated with preterm birth. Plasma MMP-9 level increased nearly three times the preterm risk.