Association between residual urinary volume and urinary tract infection. Prospective study in 225 men

Urinary tract infections can result from bladder outlet obstruction and consecutive post-void residual urine. In a recent publication, a cutoff for post-void residual urine of 180 ml was calculated, revealing sensitivity and specificity of 87 and 98.5%, respectively, regarding occurrence of significant bacteriuria in asymptomatic men. In the present study the association between post-void residual urine volume and urinary tract infection was evaluated, and different cutoff values were validated. A total of 225 asymptomatic patients (median age 66 years) were prospectively evaluated regarding the following criteria: prostate-specific antigen, prostate volume, International Prostate Symptom Score, peak urinary flow rate, urine culture results, urinary test strip, and post-void residual urine volume. By ROC analysis a cutoff predicting significant bacteriuria was calculated, and different cutoff values were validated. The independent influence of several parameters on the incidence of urinary tract infection was measured using multivariate regression analyses. Of the patients, 60% were able to completely empty the bladder (post-void residual urine volume a parts per thousand currency sign10 ml); 31% (n=69) had significant bacteriuria in the urine culture. Escherichia coli was identified in 59 of 69 patients (86%) with positive urine culture. Patients presenting with urinary tract infection had significantly higher mean post-void residual urine volumes than patients with negative urine culture (113 vs 41 ml, p < 0.001). In 29 men (13%) residual volume was 180 ml or greater. Regarding the coincidence of urinary tract infection, this cutoff value showed sensitivity and specificity of 28 and 94%, respectively (AUC: 0.606, p=0.012). By ROC analysis a cutoff value of 150 ml revealed the highest AUC value (0.617). Post-void residual volume had an independent significant influence on detection of urinary tract infection in multivariate regression analysis (urine culture: p=0.006; urinary test strip: p < 0.001). No cutoff value could be determined to predict positive urine culture with sufficient sensitivity and specificity. Based on the results of the present study and currently available data from the literature we are not able to recommend a cutoff value leading to therapeutic consequences. Hence, to establish the indication for treatment further criteria should be taken into consideration.