One hundred and eighty-four patients who underwent dynamic urethroscopy, multichannel urodynamics and physical assessment were prospectively evaluated. Urethroscopy was compared to urodynamics in genuine stress incontinence and detrusor instability. Patients were diagnosed as having genuine stress incontinence if they had loss of urine with a positive cough profile or a positive cough stress test in the absence of vesical contractions, and as having detrusor instability if they exhibited uninhibited detrusor contractions in excess of 15 cmH20 during multichannel urethrocystometry in the erect position. The positive and negative predictive values, false-positive and false-negative rates, and sensitivity and specificity were calculated, comparing urethroscopy to multichannel urodynamics. Urodynamic cough profiles were both highly sensitive (94.3%) and specific (100%) for genuine stress incontinence compared to dynamic urethroscopy, with a sensitivity and a specificity of 60.3 % and 78.9% respectively. In diagnosing vesical instability, supine dynamic urethroscopy yielded a sensitivity and specificity of 60% and 96% respectively, compared to multichannel urethrocystometry which was used as a standard. Dynamic urethroscopy is therefore a relatively insensitive predictor of genuine stress incontinence and vesical instability.