Selecting patients for penile doppler ultrasound in clinical practice

Abstract

ABSTRACT Introduction The routine use of dynamic duplex ultrasound of the penis (DDUP) to detect vasculogenic erectile dysfunction (V-ED) has been questioned because of a related patient’ discomfort and costs issues. Objective to investigate clinical factors which could guide the physician in selecting candidates for DDUP in routine practice. Methods Data were collected for 119 patients with ED who underwent DDUP regardless of their clinical history, to rule out V-ED at a single centre. Baseline data included: age, comorbidities (as scored with the Charlson Comorbidity Index (CCI)), BMI, recreational habits and the International Index of Erectile Function–6 (IIEF-6). DDUP was performed after intracavernous injection of alprostadil 20 ug and self sexual stimulation. V-ED was defined as a peak systolic velocity (PSV)<30 cm/s ± end diastolic velocity (EDV) of >3 cm/s and a resistance index (RI)<0.8. Logistic regression analysis tested clinical predictors of V-ED. The accuracy of the predictive model was assessed and decision curve analyses (DCA) tested their clinical benefit. Results Median (IQR) age was 54(43-60) years and IIEF-6 was 15(8-21). Of 119, 44 (37%) patients had comorbidities (CCI≥1), with cardiovascular disease (CVDs) and diabetes in 13(11%) and 16(14%) patients. Of all, 18(15%) patients were current smokers and 58(49%) had no regular physical activity. At DDUP, median right and left PSV were 36.7(25.7-43.9) and 36.1(25.3-44.9) cm/s. V-ED was detected in 50(42%) patients. Age (OR:1.05;95%CI 1.01–1.08; p=0.004) and smoking history (OR:3.33; 95%CI 1.13–9.8; p=0.02) were associated with a higher risk of V-ED, while physical activity was a protective factor (OR:0.43; 95%CI:0.20-091; p=0.03). Conversely, BMI, IIEF-6, CVDs, diabetes and CCI were not associated with V-ED. The model including age, smoking and physical activity had a fine accuracy to predict V-ED (AUC:0.70;95%CI:0.60- 0.79). At DCA (fig 1), the use of those clinical factors to select candidates for DDUP would result in the best clinical strategy if a threshold risk of V-ED≥35% is considered as a reasonable value to trigger the investigation. Conclusions Clinical factors including age and lifestyle habits could be used in the everyday clinical practice to better select patients for DDUP, if the use of this test is considered clinically reasonable for patients with ≥35% risk of V-ED. Disclosure Work supported by industry: no.