ABSTRACT
Benign prostatic hyperplasia (BPH) is the non-malignant enlargement of the prostate. Estimation of Prostate volume and dimensions contribute significantly to the management of BPH. Correlations between the trans-abdominal and trans-rectal ultrasound methods in estimating prostate volume and dimensions were studied with variable results. Ninety-one consecutive patients of 50 years or older with were scanned by Trans abdominal and transrectal sonographs (TA&TRUS) at the same session after obtaining the consent. All the scans were performed on a single ultrasound machine. The volume and dimensions of the prostate obtained by both methods were compared and correlated using Pearson correlation coefficient. The data was analysed further in groups based on volumes and ethnicity. Twenty-four patients were also scanned by other consultant radiologist and the data was analysed to compare the interobserver variations. Results: The mean age of the patients was 66.03±10.41 years. The mean prostate volume for ninety one patients by TA & TRUS was 44.4±35.1 ml and 46.2±34.7 ml, respectively (r = 0.965, p<0.001). Among the total patients 42 were of East Indian (EI) origin, 45 were of Caribbean African (CA) origin and 4 were of mixed race. The mean prostate volume of EI race by TA & TRUS was 35.3±23.3 and 38.9±25.9 ml respectively(r = 0.950, p<0.001). The mean prostate volume of CA race by TA & TRUS was 50.8±39.4 and 51.0±38.5 ml, respectively (r = 0.967, p<0.001). The mean prostate volume of observer A and observer B by TA & TRUS was 43.5±28.8 and 45.8±25.9 ml (r = 0.953, P<0.001) and 46.6±39 and 46.9±27.4 ml (r = 0.877, p<0.001) respectively. Conclusion: Strong correlation between TA & TRUS estimation of prostate volume and dimensions for volumes up to 100ml found in our study offers TAUS as a cost effective, less invasive, quick and well tolerable alternative to TRUS. TRUS however may be a reasonable choice for accurate measurements in larger (>100 millilitres) prostates, this needs to be further investigated by a larger sample size.