Voiding disorders of the bladder may occur after any pelvic surgery. Among patients with deep infiltrating endometriosis (DIE) it is unclear if bladder dysfunction occurs due to the disease itself or as sequelae of operating interventions. Several prospective studies conclude preoperative urodynamic multichannel testing (UD) could obtain useful information and suggested routine examination, even for asymptomatic patients. Nevertheless, UD is an invasive examination that is time, personnel, and cost-consuming: the benefit on the therapy decision must be considered when determining indications. UD measures parameters relevant to the function of the lower urinary tract.
DIE is found in approximately 20% of patients with endometriosis, causing pelvic pain and infertility. When medical treatment is ineffective laparoscopic excision of endometriotic lesions is necessary. Although DIE is a benign disease, the growth is often infiltrative to neighbouring structures, in the pelvis often into the inferior hypogastric plexus (IHP). Therefore, after surgical resection of DIE, bladder dysfunction occurs in up to 40% and may be caused by surgical damage to the IHP. Alternatively, the bladder function may already be impaired before surgery: DIE impairs bladder storage and voiding abilities, causing peripheral damage to the bladder and detrusor innervations in women without urinary symptoms. Therefore, we aimed to evaluate preoperative and postoperative UD among patients with DIE to identify pre-existing or surgically induced voiding dysfunction.

We performed a single-center prospective observational cohort study between September 2015 and December 2022. Our centre is a certified endometriosis center for treating endometriosis, with surgical teams experienced in following the laparoscopic "nerve-respecting" surgery for DIE. Whenever possible, the IHP and lower plexus are spared. We determined endometriosis classification based on the #ENZIAN classification of DIE and rASRM classification.
We invited eligible premenopausal women with clinical or imaging diagnoses of DIE requiring surgical treatment to participate.
We excluded patients with bladder endometriosis requiring bladder surgery, prior extensive colorectal surgery, pelvic malignancies, radiation therapy of the pelvis, recurrent urinary tract infections, history of psychiatric or neurologic disease, and concomitant vaginal prolapse >I° stage according to the Pelvic Organ Prolapse-Quantification system.
Maximal urinary flow rate—"uroflow"—was the primary outcome. Various cystomanometric and uroflowmetry parameters, subjective micturition disturbance rated by the International Prostate Symptom Score (IPSS), and pain score by Visual Analogue Scale (VAS) were secondary outcomes. All patients were evaluated with multichannel urodynamics before and 6 weeks after surgery, using a standardized protocol by the Good Urodynamic Practices Guidelines of the International Continence Society. 
Additionally, symptoms were recorded prospectively before and 6 weeks after surgery using the Visual Analogue Scale (VAS) questionnaire and the International Prostate Symptom Score (IPSS). The IPSS is validated for women with lower urinary tract symptoms (LUTS) and comprises 8 questions (incomplete emptying, frequency, intermittency, urgency, weak stream, straining, nocturia, and QoL). It objectively measures the subjective burden of a patient's LUTS through a 10-minute self-administered questionnaire. LUTS were classified according to absent/minimal (IPSS: 0–7), moderate (IPSS: 8–19), or severe (IPSS: >20). The VAS questionnaire comprises 5 questions and assesses the degree of dysmenorrhea, abdominal pain, dyspareunia, dysuria, and dyschezia during the previous 4 weeks on a scale from 1 to 10, with higher values corresponding to increased symptom severity.
We performed statistical analysis using Stata 16 (Stata Corporation, College Station, TX). We calculated median, range, mean, and standard deviation for continuous variables and percentages for the qualitative variables. We used paired t-test and Wilcoxon-signed rank test for the parametric and non-normally distributed continuous variables to carry out the study objectives. A significant correlation existed if p <.05. We performed an a priori statistical power analysis based on data from the published study of Kitta et al. (23). A possible bladder denervation effect caused by a prolapse surgery would be similar to that caused by surgery for DIE. Based on a clinically relevant effect size (ES) of 0.6. With an alpha = .05 and power = .90, the projected sample size needed with this ES was approximately N =33.

Maximal urinary flow "uroflow" was the primary outcome: The cohort's uroflow was 22.1ml/s preoperatively and decreased postoperatively to 21.5ml/s (p=.56, 95%CI -1.5 - 2.71). Further UD parameters, subjective micturition disturbance rated by the International Prostate Symptom Score (IPSS), and pain score by Visual Analogue Scale (VAS) were secondary outcomes: The mean bladder contractility index (BCI) dropped significantly from 130.4 preoperation to 116.6 postoperation, respectively (p=.046, 95%CI 0.23 - 27.27). Dysmenorrhea, abdominal pain, dyspareunia, and dyschezia improved significantly after surgery (p=<.001). IPSS was in the lower moderate range pre- and postoperatively (mean 8.37 vs. 8.51, p=0.893, 95%CI -2.35 - 2.05). Subgroup analysis revealed previous endometriosis surgery as a significant preoperative risk factor for elevated residual volume (43.6 ml, p=.026, 95%CI 13.89 – 73.37). Residual volume increased significantly postoperative among participants with DIE on the rectum to 54.39 ml (p=.078, 95%CI 24.06 – 84.71). Participants who underwent hysterectomy had significantly decreased uroflow (16.4 ml/s, p=.014, 95%CI 12 – 20) and BCI (75.1, p=.036, 95%CI 34.9 – 115.38).

The current study shows that patients with DIE have voiding dysfunction in 20%. There is no significant clinical or urodynamic deterioration of bladder function postoperatively. Even those with pathological UD did not get worse postoperative, even the contrary. The patient-reported outcomes in this cohort showed that bladder symptoms remained unchanged after surgical therapy, and were not clinically significant. Additionally, all endometriosis-typical complaints on the VAS were significantly improved.The rate of preoperative pathological urodynamics was 20%. The localization of the DIE showed no significant influence on the uroflowmetric or cystomanometric parameters. A small extension of the endometriosis affected all urodynamic parameters favorably. Postoperatively, there was a trend towards lower urinary flow without clinical significance. In particular, performing a hysterectomy simultaneously with endometriosis surgery resulted in a significant decrease in uroflow and BCI. After bowel resections, there were increased amounts of postvoid residual volume.

This study shows no significant difference in uroflow in patients with DIE and surgery. Impaired bladder function is found among 20% of patients with DIE, both preoperatively and postoperatively. However, UD does not show isolated, clinically relevant findings, and therefore, post-void residual volume alone is sufficient to detect impaired bladder function. Urodynamic testing is not routinely indicated among patients undergoing surgery for DIE, but preoperative and postoperative measurement of residual urine and ultrasound of the kidneys is recommended