Aims of Study:
Pressure-flow studies (PFS) are considered as a golden standard in the diagnosis and quantification of infravesical obstruction due to benign prostatic hyperplasia (BPH). Their invasiveness, however, limits their clinical application where simple uroflowmetry combined with PVR measurement is commonly used as the sole urodynamic study in aging men with lower urinary tract symptoms (LUTS). The aim of this study was to investigate the efficacy of IPSS, uroflowmetric findings and PVR in the prediction of infravesical obstruction which was diagnosed by PFS.

Methods:
A total of 76 men with a mean age of 65.2 (range: 41 - 89 years) who have attended our outpatient clinic with LUTS between the years 1999 and 2000 were analyzed in this study. All patients were initially evaluated with physical examination, IPSS, urine analysis, renal and bladder ultrasonography and simple uroflowmetry with PVR measurement. Within 2 weeks after completing the first evaluation, all patients underwent additional urodynamic studies including a filling cystometrogram and a PFS. All urodynamic tests were performed on UD-2000 (Medical Measurement Systems, The Netherlands) urodynamics unit. Obstruction was defined according to ICS-nomogram and quantified with AG number [1]. The IPSS, PVR values and uroflowmetric parameters such as maximum flow rate (Qmax), average flow rate (Qave), time to initiation of flow and voiding time were correlated with AG number. In addition, the diagnostic accuracy, sensitivity and specificity of different Qmax, Qave and PVR cut-off levels were analyzed.

Results:
In the whole study group, 45 patients (59%) have been found to be obstructed according to ICS nomogram in PFS whereas 24 (32%) were classified as unobstructed and 7 (9%) as equivocal. Seven equivocal patients were further classified as obstructed according to P-Q slope [1]. Linear regression analysis revealed no significant correlation between AG-number, IPSS and following uroflowmetric parameters: Qmax, Qave, time to initiation of flow, voiding time, Qmax/voiding time, time to Qmax/voiding time, Qave/Qmax (All r2 values < 0.1). In the prediction of obstruction, the accuracy, sensitivity and specificity rates of two different Qmax cut off levels, 10 ml/sec and 15 ml/sec have been found to be 57%, 56%, 58% and 67%, 87%, 25%, respectively. Also, the accuracy, sensitivity and specificity rates of two different Qave cut off levels, 5 ml/sec and 8 ml/sec have been found to be 68%, 80%, 42% and 83%, 96%, 32%, respectively. There was a significant difference between obstructed and un-obstructed patients in terms of PVR volumes (Student-t test, p=0.0008). However, the linear regression between PVR measurements and AG-number was found to be statistically not quite significant with a weak correlation (p=0.08, r=0.2). All patients with a PVR higher than 100 ml were found to be obstructed according to ICS-nomogram leading to a specificity and positive predictive value of 100% for PVR. However, the same cut-off level for PVR revealed relatively low accuracy and sensitivity rates with 40% and 13.4%, respectively.

Conclusion:
There is a considerable discrepancy between IPSS, findings on simple uroflowmetry and obstruction as defined by ICS nomogram and AG number in aging men presenting with LUTS. These findings suggest a multi-factorial etio-pathology of voiding dysfunction in the elderly male population and the lack of specificity of the current evaluation criteria.

References:
1. The Abrams-Griffiths nomogram. World J Urol, 13:34-9, 1995.