Adherence to Initial PDE-5 Inhibitor Treatment: Randomized Open-Label Study Comparing Tadalafil Once a Day, Tadalafil on Demand, and Sildenafil on Demand in Patients with Erectile Dysfunction


  • Financial support: The study was funded by Eli Lilly and Company, Indianapolis, IN, USA.

Corresponding Author: Hartwig Büttner, MD, Lilly Deutschland GmbH, Medical Department, Werner-Reimers Straße 2-4, 61352 Bad Homburg, Germany. Tel: +49-6172-273-2059; Fax: +49-6172-273-2520; E-mail:



Phosphodiesterase type 5 (PDE-5) inhibitor treatment for erectile dysfunction (ED) is frequently discontinued; adherence may vary depending on the initial regimen.


To evaluate the effects of initiating treatment with tadalafil once a day (OaD), tadalafil on demand (pro re nata [PRN]), or sildenafil PRN on treatment adherence.


In this multicenter, open-label study, men (≥18 years) with ED, naïve to PDE-5 inhibitors, were randomized (1:1:1) to tadalafil 5 mg OaD, tadalafil 10 mg PRN, or sildenafil 50 mg PRN. An 8-week randomized treatment (RT) period (dose adjustment possible) was succeeded by 16 weeks of pragmatic treatment (switches between PDE-5 inhibitors allowed).

Main Outcome Measures

Treatment adherence was measured as time to discontinuation of RT (any cause), estimated by Kaplan–Meier product-limit method. Treatment-group differences were estimated as hazard ratio (HR; Cox proportional hazards).


Seven hundred seventy patients (mean age 53 years) were randomized to tadalafil OaD (N = 257), tadalafil PRN (N = 252), and sildenafil PRN (N = 261). Kaplan–Meier estimates for patients discontinuing RT were 52.2, 42.0, and 66.7%, respectively. Median time to discontinuation of RT was significantly longer for tadalafil OaD and PRN (130 and >168 days) compared with sildenafil (67 days) (HR [97.5% confidence interval]: 0.66 [0.51, 0.85] and 0.49 [0.37, 0.65]; P < 0.001). Reasons for discontinuation with significant differences between groups (P < 0.05) included “lack of efficacy (duration of erection)” (sildenafil 9.2% vs. tadalafil OaD 4.3%, PRN 2.8%), “time constraints due to short window of action” (sildenafil 4.2% vs. tadalafil OaD 0%, PRN 0.4%), and “feel medication controls my sexual life” (sildenafil 2.7% vs. tadalafil OaD 0%). No between-group differences were found in International Index of Erectile Function-Erectile Function domain change from baseline to end of RT (least squares mean: 9.4–10.0, P = 0.359) or discontinuations due to adverse events (1.2–1.6%). The most common adverse event (≥4%) was headache.


ED patients assigned to tadalafil OaD or PRN adhered significantly longer to initial treatment than patients assigned to sildenafil PRN. Improvement of erectile function and safety profiles were similar in all three treatment groups.