Currently there is no consensus on the standard work-up of recurrent urinary tract infections (UTIs) in women. Most guidelines do not recommend specific investigations that should be included in the primary analysis. In our centre, we routinely use the following investigations (besides urinalysis): renal ultrasound (to rule out hydronephrosis), plain abdominal X-ray (to rule out urolithiasis),  cystoscopy (to rule out malignancy) and flowmetry and ultrasound of the bladder to measure post void residual urine. However, the utility of these investigations is unclear. In this study, we evaluated the outcome of these investigations and their clinical consequences.

We evaluated all women with recurrent UTIs who were referred to our department between January 2015 and February 2016. Only women >18 years were included. Patients with neurogenic bladder dysfunction or urological malignancy were excluded. Data were retrospectively collected from electronic patient charts. Mean post void residual urine  was calculated and with student t-test the mean number of UTIs between patients with and without post void residual urine was calculated.A chi-square analysis was done to determine a possible relation between menopausal or continence status and the presence of post void residual urine. All findings on ultrasound and abdominal X-ray were confirmed by a urologist and clinical consequences of these findings were scrutinized.

In total, 193 women with a mean age of 56.3 years, were included in the analysis. The mean number of UTI in the previous year was 5.2 (SD 3.3) The majority of the patients were referred by the general practitioner (79.1%), 5% were referred by other urologists and 8% and 5.9% were referred by internal medicine specialist and gynecologist respectively. Previous treatment consisted of antibiotics during an episode of infection (77.8%) and low dose maintenance antibiotics (18%), only a small minority of patients were treated with local estrogens (1.4%). Most women were post-menopausal (67.2%).
We have complete flowmetry data of 161 women. 75 patients did not have any post void residual urine, the mean post void urine residual was 35.5 ml (SD 63 ml). There was no difference in post void residual urine between pre and post-menopausal women.  No difference in number of infections between women with and without post void residual urine was found. In addition there was no relation between the presence of post void residual urine and continence status. 
All patients underwent routine evaluation with cystoscopy, ultrasound and X-ray (table 1) Of all cystoscopies performed, 66% were without abnormalities and 24% showed signs of inflammation (e.g. redness, edema, cystititis cystica). None showed signs of a malignancy (e.g. urothelial carcinoma).  
Of all renal ultrasounds performed, 8 (4%) showed unilateral hydronephrosis. Additional CT was performed in all patients of which 4 showed extrarenal pyelum, 1 showed parapelvic cysts, 1 showed ureteropelvic juntion stenosis and 2 showed no abnormalities.
Of all abdominal X-rays, 10 (5%) showed signs of urolithiasis. Five of these patients received a CT: in four cases no abnormalities were found and in one case small stones were found in the upper pole. The remaining 5 patients did not receive a CT and were treated conservatively. In order to further determine the results of the ultrasounds and abdominal X rays a total of 21 CT scans were made. 
None of the investigations showed to have consequences in the eventual treatment of recurrent UTIs.

The post void residual urine in this cohort of women with recurrent UTIs was small. None of the findings on ultrasound, abdominal X-ray or cystoscopy showed to have consequences in the treatment of recurrent UTIs.

According to our findings, the routine use of cystoscopy and radiological investigations such as renal ultrasound or abdominal X-ray should not be recommended in the work-up of women with recurrent urinary tract infections. It could lead to additional costs of extra investigations. Further evaluation in a larger cohort should be awaited to justify this conclusion.