The assessment of the gestational sac diameter, crown-rump length, progesterone and fetal heart rate measurements at the 10th gestational week to predict the spontaneous abortion risk.

AIM: The assessment of the first trimester ultrasonographic and progesterone measurements to predict spontaneous abortion risk.

METHODS: Ninety-nine women at the 10th week of pregnancy were included in this prospective study. Their ages, progesterone (P) levels, mean gestational sac diameters (MGSD), crown-rump lengths (CRL), MGSD-CRL measurements and fetal heart rates (FHR) were recorded. These variables were compared by abortion status.

RESULTS: Patients were followed up until the 20th weeks, and 8 (8.08%) aborted. We evaluated the variables by receiver operator characteristic curve to predict abortion. Only the areas under the curve for P levels (0.29) and for MGSD--CRL (0.16) were statistically significant We. also made logistic regression analysis to predict abortion. P level and FHR were statistically significant (P < 0.01) when the threshold value was 50%. Negative predictive value of the model was 98.9%, and positive predictive value was 50%. Overall, this model can correctly classify 94.9% of the groups. We determined threshold values for MGSD-CRL (> or =10 mm) and P (> or =25 ng/mL) to predict abortion, but not for FHR. Interestingly, 14 patients with FHR > or =175 beats/min did not abort. For the MGSD-CRL threshold, we can predict that the pregnancy will continue with 95.78% probability, with 67% sensitivity and 89% specificity. For the P threshold, the pregnancy will continue with 97.85% probability, with 80% sensitivity and 80% specificity.

CONCLUSION: MGSD-CRL and P could predict patients with low abortion risk. However, at the 10th week of pregnancy, FHR > or =175 beats/min should be evaluated for this purpose by future studies with larger sample sizes.