Post-operative urinary retention (POUR) is a frequent and potentially serious condition. The prevalence of POUR in recent studies varies from 5 to 70%. The reason for this extensive variation is mainly found in different definitions of POUR, differences in the patient characteristics regarding age and sex and in types of surgery, anaesthesia and fluid therapy.1–7
The evaluation of bladder volumes has become simple and reliable after the development of specially designed ultrasound scanners8 which take into account various anatomic conditions and factors that derive from the surgery.9 As with, for example, post-operative pain, nausea and vomiting, POUR is a condition that may prolong and complicate the post-operative phase. Furthermore, if POUR is not identified and treated in time, it may result in bladder overdistension and detrusor dysfunction.5 The symptoms of POUR vary: tension and pain above the pubic symphysis, restlessness, tachycardia, malaise and even no symptoms at all.10 In order to identify patients with increased risk of developing POUR, and in order to be able to reduce this risk, it is necessary to evaluate factors that pre-dispose to POUR.
The aims of this study were to elucidate the prevalence of POUR in a general surgical population and to identify risk factors for developing POUR. We hypothesised that a number of independent risk factors for the development of POUR in a general surgical population can be identified and that POUR is frequent.
Patients and methods
All acute and elective surgical patients surgically treated at the main operating department at Copenhagen University Hospital Herlev from 1 June to 13 July 2006 were included in the study (n = 334). Exclusion criteria were age below 15 years or surgical procedures in which a urinary catheter was placed. Pre-existing urinary symptom was not an exclusion criterion. The following surgical specialties were represented: abdominal surgery, gynaecological/obstetric surgery, orthopaedic surgery, urology, breast surgery, plastic surgery and eye surgery. A detailed description of the surgical interventions in the first four surgical specialties is given in Table 1.
The study was a prospectively designed cohort study in which the data were extracted from a national anaesthesia database and a database specially made for the study covering mainly post-operative data. The study was reported to the Danish Regional Scientific Ethical Committee. On the 22 March 2006, the chairman of the ethical committee, Erling Tvedegaard, MD, declared the investigation not notifiable (case no. H-KA-06723), the study was reported to the Danish Data Protection Agency also (j. no. 2006–41–6736).
A number of pre-defined perioperative variables were prospectively recorded by The Danish Anaesthesia Database (DAD), a nation-wide clinical database that contains indicators of the pre-operative evaluation, the intraoperative phase and the post-operative phase in the recovery ward. These data were combined with the data collected at the bedside by nurses in the recovery ward using a standardised data extraction form.
POUR: post-operative ultrasonically evaluated bladder volume more than 600 ml and inability to void within 30 min.
Sterile intermittent catheterisation (SIC): intermittent catheterisation using a sterile hydrophilic catheter, sterile gloves and sterile washing.
Shivering: an involuntary oscillatory muscular activity which increases the metabolic energy use by up to 600% above basal levels.11
Duration of anaesthesia: time from the anaesthetist's first handling of the surgical patient (e.g. blood pressure measurement or intravenous line placement) until the patient is transferred to the staff in the recovery room.
Duration of procedure: time from the incision or start of the examination until the operation/examination and bandaging is completed.
Postoperative urinary retention.
A bladder scan was performed on all patients within 30 min of arrival to the recovery room using a custom-made ultrasonic apparatus (Bladder Scan BVI 3000, Verathon Medical, IJsselstein, the Netherlands).12 A re-scan was performed in case of urge to void and at the same time being unable to do so and when the patient was discharged from the recovery room.
SIC was performed according to the department's current guideline. Eighteen patients with less than 600 ml of urine underwent SIC, and thus did not meet the present definition of POUR.
Data were analysed using the statistical program SPSS 17.0 (IBM Corporation, Somers, New York, USA). Unadjusted associations between the recorded variables and POUR were analysed using the χ2-test. Variables with a P-value less than 0.05 (two-sided) were considered statistically significant and were subsequently included in a logistic regression model (five variables). The logistic regression model was generated using the ‘backward stepwise selection’ with POUR as the dependent binary variable. The generated model tested the adjusted association of each variable with the endpoint POUR, hereby controlling for confounding. Hosmer and Lemeshow's goodness-of-fit test was used to validate the generated model. Results are presented as odds ratios with 95% confidence interval.
In total, 334 patients were enrolled in the study: 118 men (35.3%) and 216 women (64.7%). Mean age (SD) was 58.6 years (± 16.6). The prevalence of POUR was 4.8% (16/334).
The distribution of patients in the various surgical specialties is shown in Table 2.
The type of anaesthesia used was mainly general anaesthesia (Table 2). In total, 281 patients received total intravenous anaesthesia, one patient received an inhalational agent as the only anaesthetic drug and 17 patients had a combination of intravenous and inhalational anaesthesia. The inhalational agent used was desflurane or sevoflurane (and not in combination with nitrous oxide). The most widely used intravenous anaesthesia agents were propofol or thiopental for induction and propofol and remifentanil/fentanyl/alfentanil for maintenance combined with a neuromuscular block with rocuronium (or suxamethonium for rapid sequence induction) when needed for intubation or surgery. For the neuroaxial blocks, plain or hyper-baric bupivacaine without opioids was used. Post-operative pain was treated with acetaminophen, NSAIDS and opioids (morphine, ketobemidone or oxycodone).
Five of the registered variables were univariately associated with POUR: diabetes mellitus type I or II, intraoperative administration of atropine, temperature less than 36.0°C on arrival to the recovery room, duration of surgical procedure for at least 45 min and intraoperative intravenous administration of fluids of at least 750 ml.
After adjusting for confounding, by the logistic regression analysis, diabetes mellitus (type I or II) and intraoperative administration of atropine were statistically significantly associated with POUR (Table 3).
In this cohort study comprising 334 general surgical patients, we found that diabetes and intraoperative administration of atropine were independent risk factors of POUR, and that POUR was present in 4.8% of the patients.
In the literature, the prevalence of POUR varies considerably.1 This could be due to the fact that no standard definition of POUR exists, and by differences in the patient population, the type of surgery performed and the type of anaesthesia used. POUR is usually defined as a urine volume between 400 and 600 ml and the patient not being able to void.9 Others define POUR as a condition in which the patient needs catheterisation post-operatively.4 If the patient's bladder capacity is known, a more physiological/pathophysiological definition of POUR has been proposed: ‘inability of spontaneous urination with a full bladder after surgery’.3,13 The low prevalence of POUR in this study may be due to the relatively high urine volume of 600 ml used in the definition. However, in a French study from 2005 (n = 313), using the same definition of POUR as in this study, 16% developed POUR.2 This may be due to differences in case-mix, including the fact that 51% of the patients were orthopaedic surgical patients.2 In studies comprising mainly orthopaedic surgical patients, a high frequency of POUR is often found.1 In the present study, many surgical specialties were represented, and the proportion of orthopaedic patients was barely 23%. The intraoperative administration of intravenous fluids was also 44% higher in the French study than in our material. Liberal intraoperative fluid administration is often reported as a risk factor of POUR.2,6,7 The mean age of the patients, as well as the duration of procedures and anaesthesia also differs between the two studies. This could explain the discrepancy in the reported incidences of POUR. The shortest duration of surgery in the French study was 10 min,2 whereas eight patients in our study had surgery lasting less than 10 min.
The discovery of diabetes as an independent risk factor of POUR has previously been reported only sporadically. In a study of 2011 consecutive surgical procedures in spinal anaesthesia for benign ano-rectal disorders, Toyonaga et al.6 found that diabetes was an independent risk factor for developing POUR. The prevalence of POUR (defined as need for catheterisation within 24 h post-operatively) was 16.7% in this population. It is a well known fact that diabetes pre-disposes to the development of urinary problems and that this seems to be independent of glycaemic control in patients with diabetes mellitus 1.14 In our study, the risk of developing POUR was increased almost six-fold if the patient had co-existing diabetes. This could indicate that particular attention should be paid to patients with diabetes, and their risk of developing POUR.
The association of intraoperatively administered atropine and POUR has previously been evaluated. In a study by Keita et al.2 from 2005, atropine and atropine-like substances were evaluated, and no association with POUR was identified. Atropine is used – although rarely – as an anti-cholinergic drug during anaesthesia to counteract vagally triggered bradycardia. The half-life is about 4 h, and the substance can potentially relax the detrusor musculature and contract the bladder neck and thereby inhibit the patient's ability to urinate. The effect is well known and atropine shares this characteristic with several other drugs with anti-cholinergic effects, including certain anti-psychotics, class 1 anti-arrhythmic agents, anti-parkinsonian agents and anti-histamines.10 In the present study, intraoperative administration of atropine was administered solely with the purpose of treating bradycardia (heart rate <40 beats/min) and was found to increase the risk of developing POUR six times.
Previous studies have identified a number of pre-disposing factors of POUR not found in this study: age, sex, intraoperative fluid administration and type of operation.2,4,6,7,15,16 Old age is believed to predispose to POUR because of a physiologic age-related neuronal degeneration that has a negative impact on the control of voiding.1 With respect to sex, both male and female sexes have been reported to pre-dispose to POUR.6,7 An association between type of operation and POUR has predominantly been found in studies including many patients operated in the ano-rectal region.7,15 Furthermore, an increased incidence of POUR is seen in surgical procedures affecting the lower urinary tract such as gynaecological operations for urinary incontinence.3 Of the 16 patients having POUR in our patient material, only two represented high-risk patients (one hernia repair and one operation for anal fistula). Of the remaining 14 patients, one was operated in the head and neck region, two had eye surgery, four had operations of the skin on the truncus or lower extremities, one had a reposition of the hip joint, three had mastectomy, two had a revision of a colostomy and one had an arthroplasty of the shoulder. The fact that we did not find significant associations between the above-mentioned risk factors and POUR could partly be explained by our patients' characteristics, with a relatively little representation of high-risk patients and a relatively small number of patients with POUR due to the high bladder volume in the definition compared with many other studies in this field.
This study holds a number of limitations.
Data in the study was collected from the DAD version 2. A large number of anaesthetists enter data into this database which increases the risk of erroneous entries. This may affect the validity of the data. To minimise this risk, the data have been carefully reviewed for double entries and obvious mistakes.
Despite adjusting for confounding by using multivariate logistic regression, there is a risk of confounding by non-measured risk factors – residual confounding.
Additional risk factors for POUR, other than those examined in the present study, could exist, for example factors related to the type of anaesthesia used. This might have been possible to disclose if a larger number of patients had been included.
Outliers, for example patients with very short duration of surgery, might bias the results. Bladder volume measured ultrasonically may, although the method has been validated,9 be uncertain.
Different definitions of POUR exist.
Finally, there is a non-negligible uncertainty in the point estimates, especially for atropine.
In the present large cohort of general surgical patients, the prevalence of POUR was approximately 5%. Co-existing diabetes and intraoperative administration of atropine increased the adjusted risk of POUR nearly six-fold.
In an effort to reduce the prevalence of POUR in surgical patients, an individual risk assessment, that is based on age, sex and co-existing diseases including diabetes, is recommended. In the presence of any risk factors, it may be considered to place a urethral catheter or to carry out a routine post-operative bladder scan. When administering atropine intraoperatively, the increased risk of POUR should be taken into account.
Further studies, evaluating the long-term effects of POUR, are needed.
This study was carried out with assistance from the nurses on duty in the recovery ward during the period of inclusion of the patients. There has been no financial support or sponsoring, and there are no conflicts of interests.
1 Baldini G, Bagry H, Aprikian A, Carli F. Postoperative urinary retention
: anesthetic and perioperative considerations. Anesthesiology 2009; 110:1139–1157.
2 Keita H, Diouf E, Tubach F, et al
. Predictive factors of early postoperative urinary retention
in the postanesthesia care unit. Anesth Analg 2005; 101:592–596, table.
3 Bodker B, Lose G. Postoperative urinary retention
in gynecologic patients. Int Urogynecol J Pelvic Floor Dysfunct 2003; 14:94–97.
4 Koch CA, Grinberg GG, Farley DR. Incidence and risk factors for urinary retention
after endoscopic hernia repair. Am J Surg 2006; 191:381–385.
5 Lamonerie L, Marret E, Deleuze A, et al
. Prevalence of postoperative bladder distension and urinary retention
detected by ultrasound measurement. Br J Anaesth 2004; 92:544–546.
6 Toyonaga T, Matsushima M, Sogawa N, et al
. Postoperative urinary retention
after surgery for benign anorectal disease: potential risk factors and strategy for prevention. Int J Colorectal Dis 2006; 21:676–682.
7 Zaheer S, Reilly WT, Pemberton JH, Ilstrup D. Urinary retention
after operations for benign anorectal diseases. Dis Colon Rectum 1998; 41:696–704.
8 Pedersen LM, Mantoni T, Lynggaard MD, et al
. Postoperative urinary retention
. Clinical assessment versus ultrasound examination. Ugeskr Laeger 2007; 169:605–608.
9 Rosseland LA, Stubhaug A, Breivik H. Detecting postoperative urinary retention
with an ultrasound scanner. Acta Anaesthesiol Scand 2002; 46:279–282.
10 Verhamme K, Sturkenboom M. Drug-induced urinary retention
: incidence,management and prevention. Drug Saf 2008; 31:373–388.
11 Bhattacharya PK, Bhattacharya L. Post anaesthesia
shivering. Indian J Anaesth 2003; 47:88–93.
12 Ghani KR, Pilcher J, Rowland D, et al
. Portable ultrasonography
and bladder volume accuracy: a comparative study using three-dimensional ultrasonography
. Urology 2008; 72:24–28.
13 Bjarnesen J, Lose G. Postoperative urinary retention
. Ugeskr Laeger 1991; 153:1920–1924.
14 Van Den Eeden SK, Sarma AV, Rutledge BN, et al
. Effect of intensive glycemic control and diabetes complications on lower urinary tract symptoms in men with type 1 diabetes. Diabetes Control and Complications Trial/Epidemiology of Diabetes Interventions and Complications (DCCT/EDIC) study. Diabetes Care 2009; 32:664–670.
15 Lau H, Lam B. Management of postoperative urinary retention
: a randomized trial of in-out versus overnight catheterization. ANZ J Surg 2004; 74:658–661.
16 Lingaraj K, Ruben M, Chan YH, Das SD. Identification of risk factors for urinary retention
following total knee arthroplasty: a Singapore hospital experience. Singapore Med J 2007; 48:213–216.