Ultrasonographer experience vs. power Doppler index for ovarian tumor discrimination

Objectives: To assess the performance of the power Doppler index (PDI) compared to subjective judgment of the ultrasonographer in the preoperative diagnosis of ovarian malignancy. Methods: Eighty-two successive complex adnexal masses were examined prospectively with power Doppler before surgical treatment. The investigation was multicentric and included in a large contrast-enhanced ultrasound European study. A vascularity index based on the number of colored pixels, was estimated on selected frames (defined region of interest covering the entire tumor) of the tumors using an in-house color quantifying program. The sensitivity and specificity of PDI, the resistance index and subjective visual scoringwere compared using receiver operating characteristic (ROC) curves. Results: Histology identified 34 malignant tumors including 11 borderline tumors and 48 benign tumors. PDI was considerably higher in malignant than in benign lesions: 21.88 (95% CI, 19.19 24.57) vs. 8.11 (95% CI, 6.09 10.19) for the benign tumors and 11.63 (95% CI, 1.20 22.06) for borderline tumors. The PDI cut-off value for differential diagnosis was set at 0.107 (10.7% of the tumor colored). However, the sensitivity and specificity of the PDI were quite low: 75% (95% CI, 72.9 90.7) and 60% (95% CI, 56.3 92.5) respectively. Using logistic regression, visual Doppler scoring alone (OR 6.14 (95% CI, 2.30 16.38), P = 0.0003) or associated with gray-scale examination (OR 10.5 (95% CI, 3.19 34.54), P = 0.0001) performed better than PDI (OR 2.61 (95% CI, 1.06 6.43), P = 0.04). Conclusions: The power Doppler index obtained using our color quantifying software has discriminating power for adnexal masses, but visual Doppler scoring performed best.