A Randomized Controlled Study on the Impact of Early Urinary Catheter Removal on Postoperative Urinary Retention in Abdominal and Thoracic Surgery Patients with Thoracic Epidural Analgesia

Abstract

Background

Thoracic epidural analgesia (TEA) is a key component of Enhanced Recovery After Surgery protocols for major abdominal and thoracic procedures. Despite its benefits for pain management, TEA has been associated with an increased risk of postoperative urinary retention (POUR). Consequently, it is common practice to maintain a urinary catheter (UC) for the duration of TEA. This study aimed to evaluate the impact of early UC removal in patients receiving TEA through a randomized controlled trial.

Methods

In this randomized controlled trial approved by the Rostock University Medical Center Ethics Board (AZ A2018-0220), patients scheduled for elective major abdominal or thoracic surgery with anticipated TEA within 1 year were enrolled. Participants were randomized into two groups: the early removal group (ERG), where the UC was removed within 48 h post-surgery, and the standard group (SG), where the UC was retained until TEA discontinuation. POUR was defined as a residual urine volume of ≥400 mL measured by ultrasound, and catheter-associated urinary tract infections (CAUTIs) were assessed.

Results

Of the 99 patients initially enrolled, 81 patients were available for analysis. In the ERG (n = 43), the UC was removed within 48 h, whereas in the SG (n = 38), the UC was maintained until TEA cessation. The incidence of POUR was similar between the groups, with 1 patient in each group (2.3% in ERG vs. 2.6% in SG, p = 1) requiring recatheterization. CAUTI developed in 4 patients (4.9%), all of whom were in the SG (10.5%), indicating a statistically significant association between the timing of UC removal and CAUTI incidence (p = 0.044).

Conclusion

Our results suggest that early UC removal under TEA is safe and does not significantly increase the risk of POUR while reducing the incidence of CAUTIs. These findings support the feasibility of early UC removal in this patient population and may inform future guidelines on perioperative UC management in the context of TEA.

Introduction

Thoracic epidural analgesia (TEA) is a common tool for pain management in modern Enhanced Recovery After Surgery (ERAS) protocols, frequently used in major abdominal and thoracic surgeries to provide effective analgesia and improve postoperative recovery. However, its use is associated with an increased risk of postoperative urinary retention (POUR), a complication that can affect recovery [1, 2]. One contributing factor is the use of opioids administered via the epidural route, which may promote POUR by modulating key aspects of bladder function, including the subjective sensation of bladder fullness, detrusor muscle contraction, bladder capacity, and sphincter activity. To manage this risk, many clinics in Germany routinely leave a urethral catheter (UC) in place for the duration of TEA [3]. As recently shown by our group, a retrospective single-center analysis of patients after visceral or thoracic operations using TEA revealed that an early UC removal is feasible without increasing the risk for recatheterization [4].

Despite the clinical relevance of this issue, only a few randomized studies have addressed the optimal timing for UC removal in patients receiving TEA [5–7]. Consequently, no specific recommendations regarding this practice are included in the latest guidelines. In our previous retrospective study, we observed no significant difference in the incidence of POUR between early and late UC removal under TEA [4]. To validate these findings and provide stronger evidence, a randomized controlled trial was conducted to further investigate this critical question.

Methods

This randomized controlled trial was approved by the Ethics Board of the Rostock University Medical Center (AZ A2018-0220) and written informed consent was obtained from all subjects. The study included patients scheduled for elective major abdominal or thoracic surgery with anticipated TEA during January 2020 and January 2021. Patients with known medical conditions, such benign prostatic hypertrophy or surgical conditions, such as low anterior resection, associated with a high risk for POUR were excluded. Using a computer-generated block randomization schedule (“blockrand”-package) [8], patients were assigned to one of two groups: the early removal group (ERG), where the bladder catheter was removed within 48 h, or the standard group (SG), where the bladder catheter was maintained as long as the epidural analgesia was in use. Randomization was not stratified by gender.

An epidural catheter was inserted preoperatively, and all patients received a TEA infusion of ropivacaine 0.2% combined with 10 μg of sufentanil. The bolus dose was 2 mL, with a basal infusion rate between 6 and 8 mL/h, adjusted daily by the pain management team based on the pain scale. General anesthesia was induced with propofol, fentanyl, and rocuronium. After induction, a urinary catheter (UC) was placed, and intraoperative urine output was recorded.

Postoperatively, all patients were monitored daily by the surgical and pain management teams. Pain intensity was evaluated daily using the visual analog scale (VAS) at rest and during movement, with the epidural infusion adjusted to maintain a VAS <3 at rest and <5 during coughing.

The decision to remove or retain the bladder catheter was documented in the nursing chart and communicated to the surgical ward nurse. At the institution, TEA is typically continued for an average of 5 days postoperatively. Patients received oral metamizole (500 mg every 6 h) or paracetamol (1 g every 6 h) during the first five postoperative days. Following TEA discontinuation, patients were transitioned to oral slow-release oxycodone/naloxone (10 mg/5 mg twice daily) with breakthrough oxycodone (5 mg every 4 h as needed).

After catheter removal in both groups, bladder volume was measured using ultrasound. Bladder scanning was performed 6 h post-catheter removal if the patient was unable to void. No routine ultrasound assessment was conducted in patients who voided normally within 6 h. POUR was defined as a residual urine volume of ≥400 mL, acknowledging the lack of consensus in the literature regarding a standard POUR definition. Catheter-associated urinary tract infections (CAUTIs) were diagnosed according to the Guidelines on Urological Infections of the European Association of Urology [9]. The diagnostic criteria included the presence of fever (≥38°C), urinary symptoms (e.g., dysuria, urgency, frequency, suprapubic pain), and a positive urine culture obtained either during catheterization or within 2 days after catheter removal [9]. Empirical broad-spectrum antibiotics were initiated based on local susceptibility patterns, with targeted therapy prescribed following urine culture results.

Data Collection and Statistical Analysis

The collected data included demographic characteristics, type and duration of surgery, intraoperative intravenous fluid administration, duration of bladder catheterization, epidural insertion level, POUR, and CAUTI incidence. The primary outcome was the incidence of POUR, while secondary outcome was the incidence of CAUTI.

Descriptive statistics are presented as mean ± standard deviation or median (Q1–Q3) for continuous variables and frequency (%) for categorical variables. Data distribution was assessed using the Shapiro-Wilk test. Group differences were analyzed with the Mann-Whitney U test for continuous variables and the χ² or Fisher’s exact test for categorical variables. Statistical significance was set at p < 0.05. Statistical analysis was conducted using IBM^® SPSS^® Statistics 29.0 [10].

The sample size was determined based on the pooled incidence of POUR after abdominal, orthopedic, pelvic, or anorectal surgery, as described by Baldini et al. [1]. It was estimated that the POUR rate in the ERG would be approximately 40%, compared to 5% in the SG. To detect a 35% difference in POUR rates between the groups with 80% power, a minimum of 22 patients per group was required.

Results

Patient Enrollment and Study Population

Initially, 99 patients were included in the study. However, 18 patients were excluded postoperatively for the following reasons: in the SG, 11 patients were excluded due to unintentional premature dislodgement of the UC (n = 3), patient request for removal from the study because of discomfort caused by the UC (n = 7), or dislodgement of the epidural catheter within 24 h postoperatively (n = 1); in the ERG, 7 patients were excluded due to medical necessity for retaining the UC (n = 6) or unintentional dislodgement of the epidural catheter within 24 h postoperatively (n = 1). Ultimately, 81 patients were available for analysis.

In the ERG (n = 43), the UC was removed within 48 h post-surgery, while in the SG (n = 38), the UC was retained until the TEA was discontinued. Notably, in the SG, the UC was occasionally left in place slightly longer than the actual duration of the TEA in order to allow for the possibility of prolonged epidural analgesia. Therefore, in some cases the duration of catheterization exceeded the duration of TEA (Table 1). The TEA catheter insertion site was checked daily by the pain management team, and pain intensity was assessed using the VAS. Mobilization began on the first postoperative day under supervision from nurses or physiotherapists. TEA was discontinued based on the decision of the pain management team.

Table 1.

Perioperative parameters

Parameter	SG, n = 38	ERG, n = 43	p value
Type of surgery	​	​	0.174
Abdominal	27	24	​
Thoracic	11	19	​
Access	​	​	0.605
Open surgery	35	39	​
Endoscopic surgery	3	4	​
Operation time in minutes, median (Q1-Q3)	238.5 (154.2–341)	191 (111.5–267)	0.646
Intraoperative fluid volume administered in mL, median (Q1-Q3)	3,755 (2,482–5,544)	3,100 (1,657–4,400)	0.383
Epidural catheter duration, days	5.14 (±1.60)	5.58 (±1.56)	0.205
Level of epidural catheter insertion	​	​	0.251
T 6-8	21	29	​
T 9-12	17	13	​
Time to bladder catheter removal, days	6.24 (±3.9)	2 (±0.46)	<0.001

Fisher’s exact test or chi-square test.

SG, standard group; ERG, early removal group.

Postoperative Urinary Retention

In the SG, 1 patient (2.6%) required recatheterization due to POUR, while in the ERG, 1 patient (2.3%) also needed recatheterization for the same reason. No statistically significant difference was found between the groups in terms of POUR incidence (p = 1). Both affected patients were male, aged 79 and 69 years, exceeding the median age of the study population (63 years). The duration of surgery was 321 min in the first patient and 254 min in the second, compared with a study population median of 203 min. Intraoperative fluid administration was 5,100 mL and 3,600 mL, respectively, compared with a median of 3,200 mL. Residual urine volumes 6 h after catheter removal were 420 mL and 560 mL, respectively.

Patient Characteristics and Perioperative Parameters

The distribution of age, body mass index, and gender is shown in Table 2. No significant differences were observed between the ERG and SG in demographic or baseline characteristics. A total of 65 (69.1%) male and 25 (30.9%) female patients were included, with a median age of 63 years (range: 19–82). The body mass index did not differ significantly between the groups (25.3 vs. 25.6, p = 0.754). No statistically significant differences were found in preoperative comorbidities, ASA classification, or Charlson Comorbidity Index (CCI).

Table 2.

Characteristics and comorbidities of the 81 prospectively examined patients

Characteristic	SG, n = 38	ERG, n = 43	p value
Age, years	64 (57–74.75)	62 (49–70)	0.154
Sex	​	​	0.812
Male	27	29	​
Female	11	14	​
BMI	25.3 (±4.46)	25.6 (±4.79)	0.754
Comorbidities
Diabetes mellitus	10	4	0.75
Coronary heart disease	4	6	0.74
Arterial hypertension	18	19	0.826
Chronic kidney disease	2	4	0.679
Chronic obstructive pulmonary disease	10	11	1
Chronic heart failure	3	2	0.661
History of neoadjuvant therapy	6	7	1
Nicotine addiction	9	10	1
Obesity	3	0	0.099
Rheumatoid arthritis	3	2	0.661
ASA score (II/III/IV)	11/26/1	9/28/6	0.169
CCI	5.16 (±2.21)	4.3 (±2.96)	0.142

Fisher’s exact test or chi-square test.

SG, standard group; ERG, early removal group; BMI, body mass index; ASA, American Society of Anesthesiologists Physical Status Classification System; CCI, Charlson comorbidity index.

All surgeries were elective, with 63% (n = 51) of patients undergoing laparotomy for colorectal, gastric, pancreatic, hepatic, or other abdominal procedures, while 37% (n = 30) underwent thoracic surgery, including lobectomy, pneumonectomy, segmentectomy, decortication, enucleation, and esophagectomy. No significant differences were observed between the groups in the type of surgery, duration of surgery, intraoperative fluid administration, or the epidural insertion level (Table 1).

Catheter-Associated Urinary Tract Infection

A total of 4 patients (4.9%) developed CAUTIs, all of whom were in the SG (10.5%). A statistically significant association was found between the timing of UC removal and the incidence CAUTIs (p = 0.044).

None of the 4 patients who developed CAUTI required recatheterization due to POUR after the initial UC removal. In the subgroup analysis of the SG, no significant difference was observed in the incidence of CAUTIs based on gender, with 2 out of 27 men and 2 out of 27 women affected (p = 0.326). Additionally, no significant association was found between the type of surgery (laparoscopic vs. open) or surgical site (thoracic vs. abdominal) and CAUTI incidence (p = 1 and p = 0.7791, respectively).

Discussion

In this study, we aimed to determine whether early UC removal following major abdominal or thoracic surgery under TEA impacts the incidence of POUR. To achieve this, we conducted a prospective, randomized study over 1 year, involving consecutive patients undergoing abdominal or thoracic surgical procedures.

Our findings demonstrate that early UC removal is not associated with an increased incidence of POUR. On the contrary, when the bladder catheter was retained throughout the entire TEA period, the incidence of CAUTI was significantly higher. In the ERG, where the catheter was removed within 48 h after surgery, no CAUTIs were observed among the 43 patients.

Both cases of POUR in our trial occurred in male patients. Although we did not stratify by gender during randomization, the gender distribution in the two groups was similar (approximately 70% male in each). We performed a subgroup analysis by gender within the SG and found no significant difference in CAUTI incidence between men and women (as noted above). However, the absence of POUR in female patients is noteworthy and suggests that men may have a higher predisposition to urinary retention under TEA. This observation is consistent with prior reports that male patients appear to be at higher risk of POUR than female patients [1, 6], a difference attributed to anatomical factors and the increasing prevalence of benign prostatic hyperplasia with age [11]. In addition to sex, the likelihood rises substantially with age, with patients over 50 years having a 2.4-fold higher risk compared with younger individuals [12]. Taken together, these findings highlight the importance of considering patient age and sex in risk stratification for POUR.

Because 60% of patients in our retrospective study who developed POUR had undergone rectal resection, patients who underwent rectal surgery or deeper pelvic dissection were excluded from the prospective study as this was considered an independent risk factor for POUR [4]. It is postulated that bladder function recovers more slowly in patients undergoing rectal resection than in those undergoing colon resection. This may be due to the proximity of the pelvic autonomic nerves to the incision plane during rectal resection [13, 14]. A meta-analysis published in 2021 by McIntosh et al. [15] showed that early removal of an indwelling catheter within the first two postoperative days after colorectal surgery with pelvic dissection was associated with an increased risk of urinary retention compared to late removal (26.4% vs. 8.8% retention; p < 0.001).

The optimal timing for UC removal under TEA remains a topic of debate. In the “Guidelines for enhanced recovery after lung surgery: recommendations of the Enhanced Recovery After Surgery Society and the European Society of Thoracic Surgeons (ESTS) 2019” it is mentioned that TEA is associated with POUR and that it makes sense to place a transurethral catheter in these patients. A recommendation for the timing of removal could not be given [16]. The latest German guideline on the perioperative management of gastrointestinal tumors highlights the scarcity of randomized studies and the lack of consensus on best practices. Existing literature offers only a handful of randomized studies addressing this question [17]. Many perioperative protocols recommend early removal of the UC after colorectal or thoracic surgeries. These protocols are based on data demonstrating an increased risk of urinary tract infections in postoperative patients undergoing cardiac, visceral, vascular, and orthopedic surgeries when catheterization lasts more than 2 days [18]. However, it should be noted that these protocols do not specifically target surgical patients receiving TEA as early removal of UCs in this context may be associated with a potential increased risk of POUR and its related complications [19–22]. The decision to use 48 h as the threshold for early removal of the indwelling catheter in our study was based on weighing the risk of developing POUR under TEA, as demonstrated in the aforementioned studies, against evidence that patients with indwelling catheters for more than 48 h are at a higher risk for urinary tract infections compared to those catheterized for less than 2 days [18].

For instance, Zaouter et al. [5] conducted a randomized study on patients undergoing abdominal and thoracic surgeries with TEA. They observed a marginally higher incidence of POUR with early UC removal (8.2% vs. 1.8%, p = 0.09), though this was not statistically significant. Similarly, Coyle et al. [6] conducted a prospective randomized controlled clinical pilot study on 44 patients after colorectal surgery. These patients were randomized into two groups: In group 1, the UC was removed within 48 h after surgery, while in group 2, the UC was removed after completion of the TEA. POUR occurred in three men (20%) in group 1 and in two men (22.2%) in group 2, whereas no POUR occurred in women. Interestingly, all 5 patients who developed POUR underwent rectal resection.

Schreiber et al. [23] reported in a prospective, non-randomized study that early removal of the UC during TEA after colorectal surgery is a risk factor for the development of urinary retention. In this study, 7.8% of patients in whom the UC was removed on the first postoperative day required recatheterization due to POUR vs. 2.6% in the control group. In contrast, Chia et al. [7] conducted a randomized study on 78 patients undergoing thoracic surgery and found no cases of POUR requiring recatheterization, nor any urinary tract infections, highlighting the variability in POUR incidence based on surgical type and population.

The incidence of CAUTIs in our study was significantly higher (10%) when the UC was left in place for the entire duration of TEA. In contrast, no CAUTIs occurred in patients whose catheter was removed on the second postoperative day. As demonstrated by Maki et al. [24] the relative risk of CAUTIs increases 5.1-fold when the duration of urinary catheterization exceeds 5 days.

Zaouter et al. [5] conducted a prospective study on 215 patients undergoing abdominal and thoracic surgeries and reported a significantly higher CAUTI rate (14%) in the control group, where the UC was removed only after the completion of TEA, compared to patients whose catheter was removed on the morning following surgery (2%). Schreiber et al. [23] also demonstrated a significant reduction in CAUTIs from 30.4% to 13.8% when the UC was removed on the first postoperative day following colorectal surgeries under TEA, compared to the control group, in which the catheter was removed only after the completion of epidural analgesia.

It is well established that urinary catheterization is a common cause of UTIs as it increases the risk of infection by 5–10% per day of use [25]. Wald et al. published an analysis of postoperative UTIs in 35,904 hospitalized patients who underwent major surgical procedures. Their study showed that the cumulative probability of developing a UTI requiring treatment was twice as high (9.4%) in patients who were catheterized for more than 2 days compared to those who were catheterized for 2 days or less, where the UTI rate was only 4.5% [17].

Despite our study showing no significant difference in CAUTI incidence regarding the type of surgery, a study by Regenbogen et al. [26] found that a large-scale analysis involving over 54,000 cases of colorectal procedures showed a higher rate of postoperative UTIs following colorectal surgery (4.1%) compared to non-colorectal gastrointestinal surgeries (1.8%). Therefore, targeted measures are necessary to minimize nosocomial UTIs after colorectal resections.

This study has several limitations that should be named. The most significant limitation is the considerable heterogeneity within the patient cohort, which may affect the rates of POUR, CAUTIs, and other outcomes. Additionally, while the study was sufficiently powered, the relatively small sample size and single-center design represent potential constraints. Furthermore, no baseline bladder function tests were performed prior to patient inclusion, which could have provided valuable context for interpreting the results. We also note that because our trial excluded patients undergoing rectal or deep pelvic surgeries (who may have a higher inherent risk for POUR), our findings may not be directly applicable to those populations. Another limitation is that bladder volume after catheter removal was measured only in patients with clinical signs of retention (i.e., those unable to void within 6 h). This selective assessment could have led to an underestimation of the true POUR incidence if any patients had asymptomatic retention. However, it is likely that any significant retention would have become symptomatic and thus been detected by our protocol.

Conclusion

Our results suggest that early UC removal under TEA is safe and does not significantly increase the risk of POUR while it reduces the incidence of CAUTIs. These findings support the feasibility of early UC removal in this patient population and may inform future guidelines on perioperative urinary management in the context of TEA.

Statement of Ethics

The study was conducted in accordance with the ethical principles outlined in the Declaration of Helsinki. This study protocol was reviewed and approved by the Rostock University Medical Center Ethics Committee, including an exception of obtaining written informed consent to participate, Approval No. A 2018-0220.

Conflict of Interest Statement

E.K. was a member of the journal’s Editorial Board at the time of submission.

Funding Sources

The authors report no specific funds or grants for this study.

Author Contributions

Ahmed Alwali: conception of the work, data acquisition, interpretation of the data, writing of the first draft of the manuscript, and revision of the manuscript. Clemens Schafmayer, Ernst Klar, and Mark Philipp: interpretation of the data and revision of the manuscript. Matthias Leuchter: statistical analysis, writing of parts of the manuscript, and revision of the manuscript. Eberhard Grambow: conception of the work, interpretation of the data, and revision of the manuscript. All authors have read and approved the manuscript.

Funding Statement

The authors report no specific funds or grants for this study.

Data Availability Statement

All data generated or analyzed during this study are included in this article and its online supplementary material files. Further inquiries can be directed to the corresponding author.