Virtual Organ Computer-aided Analysis (VOCAL) technique. Logistic regression models using PET as the outcome were determined for each PVI. Covariates included: study group, age, BMI, smoking status, aspirin use, parity, material deprivation of area of residence and Mean Arterial Pressure (MAP). Results: The overall rate of PET was 12%. No significant differences in age, smoking status, deprivation or MAP were seen between groups. Significant differences in parity (P=0.002), BMI (P<0.001) and aspirin use (P<0.001) were noted. In the logistic regression model, a 1% increase in VI from a mean of 14.2 (SD 7.7) to 15.2% was associated with a 10% reduction in the odds of PET (OR 0.90, 95% CI 0.82-0.98, P=0.02) and a 1 unit increase in FI (mean 42.1, SD 10.3) was associated with a 6% reduction in the odds of PET (OR 0.94, 95% CI 0.89-1.0, P=0.04). A doubling in VFI (median 5.1, IQR 3.4-8.3) was associated with a 57% reduction in the odds of PET (OR 0.43, P=0.001). Conclusions: Using logistic regression, VI, FI and VFI were predictive of PET in separate models controlling for confounding variables. This technique may be a useful adjunct for clinicians wishing to refine first trimester screening models to predict PET.

No abstract

No abstract