Introduction: The use of dynamic duplex ultrasound of the penis (DDUP) has been proposed as a second-level test throughout the diagnostic work-up of men presenting with erectile dysfunction (ED). We investigated possible predictor factors for DDUP results.
Methods: DDUP data from 188 patients with ED were prospectively collected to rule out vasculogenic alterations (V-ED) at a single academic centre. Baseline data included: age, cardiovascular risk factors (hypertension, diabetes, total cholesterol levels, waist circumference, regular physical activities), BMI, smoking and alcohol habits and the International Index of Erectile Function – erectile function (IIEF-EF) domain scores. DDUP was performed after intracavernous injection of alprostadil 20 ug and sexual stimulation in all patients. V-ED was defined as a peak systolic velocity (PSV) <35 cm/s and a resistance index (RI) <0.8. We relied on Chi-square Automatic Interaction Detectors (CHAID), a recursive machine learning partitioning algorithm in order to identify risk groups for V-ED. ROC curve was applied to test the accuracy of the prediction for each risk group.
Results: Median (IQR) age was 54 (43-60) years and IIEF-EF was 14 (7-20). Of 188, 79 (42%) patients presented with Charlson Comorbidity Index (CCI)=1. Overall, 57 (30%) and 17 (9%) had hypertension and diabetes, respectively. Median waist circumference was 98.5 (91-105) cm. Of all 164 (87%) patients reported moderate alcohol intake. At DDUP, V-ED was detected in 85 (45%) patients. The CHAID identified three groups: i) age =40 years; ii) age >40 years and waist circumference =102cm; iii) age >40 years and waist circumference >102cm. We named those groups as low-, intermediate-, and high-risk for V-ED. The risk of V-ED among the groups was: 16.7%, 41.6%, and 64.9%, respectively. ROC curve revealed that this risk classification has a good predictive ability (AUC: 0.67, 95%CI: 0.56-0.72).
Conclusions: Our study challenges current guidelines regarding DDUP. Pending external validation, we propose performing DDUP only in case of intermediate- and high-risk patients, respectively.
.jpg)
.jpg)