International Journal of Colorectal Disease

, Volume 21, Issue 7, pp 676–682

Postoperative urinary retention after surgery for benign anorectal disease: potential risk factors and strategy for prevention

Authors

    • Department of SurgeryMatsushima Hospital Colo-Proctology Center
  • Makoto Matsushima
    • Department of SurgeryMatsushima Hospital Colo-Proctology Center
  • Nobuhito Sogawa
    • Department of SurgeryMatsushima Hospital Colo-Proctology Center
  • Song Feng Jiang
    • Department of SurgeryMatsushima Hospital Colo-Proctology Center
  • Naomi Matsumura
    • Department of SurgeryMatsushima Hospital Colo-Proctology Center
  • Yasuhiro Shimojima
    • Department of SurgeryMatsushima Hospital Colo-Proctology Center
  • Yoshiaki Tanaka
    • Department of SurgeryMatsushima Hospital Colo-Proctology Center
  • Kazunori Suzuki
    • Department of SurgeryMatsushima Hospital Colo-Proctology Center
  • Junnichi Masuda
    • Department of AnesthesiologyMatsushima Hospital Colo-Proctology Center
  • Masao Tanaka
    • Department of Surgery and OncologyGraduate School of Medical Sciences, Kyushu University
Original Article

DOI: 10.1007/s00384-005-0077-2

Cite this article as:
Toyonaga, T., Matsushima, M., Sogawa, N. et al. Int J Colorectal Dis (2006) 21: 676. doi:10.1007/s00384-005-0077-2

Abstract

Purpose

This study was undertaken to determine the incidence of and risk factors for urinary retention after surgery for benign anorectal disease.

Methods

We reviewed 2,011 consecutive surgeries performed under spinal anesthesia for benign anorectal disease from January through June 2003 to identify potential risk factors for postoperative urinary retention. In addition, we prospectively investigated the preventive effect of perioperative fluid restriction and pain control by prophylactic analgesics on postoperative urinary retention.

Results

The number of procedures and the urinary retention rates were as follows: hemorrhoidectomy, 1,243, 21.9%; fistulectomy, 349, 6.3%; incision/drainage, 177, 2.3%; and sliding skin graft/lateral subcutaneous internal sphincterotomy, 64, 17.2%. The overall urinary retention rate was 16.7%. With hemorrhoidectomy, female sex, presence of preoperative urinary symptoms, diabetes mellitus, need for postoperative analgesics, and more than three hemorrhoids resected were independent risk factors for urinary retention as assessed by multivariate analysis. With fistulectomy, female sex, diabetes mellitus, and intravenous fluids >1,000 ml were independent risk factors for urinary retention. Perioperative fluid restriction, including limiting the administration of intravenous fluids, significantly decreased the incidence of urinary retention (7.9 vs 16.7%, P<0.0001). Furthermore, prophylactic analgesic treatment significantly decreased the incidence of urinary retention (7.9 vs 25.6%, P=0.0005).

Conclusions

Urinary retention is a common complication after anorectal surgery. It is linked to several risk factors, including increased intravenous fluids and postoperative pain. Perioperative fluid restriction and adequate pain relief appear to be effective in preventing urinary retention in a significant number of patients after anorectal surgery.

Keywords

Urinary retentionAnorectal surgeryAnal fistulaHemorrhoid

Introduction

Acute urinary retention is the most common complication after surgery for benign anorectal disease [13]. Urinary retention produces discomfort, and it can lead to urethral injury and urinary tract infection following catheterization. In addition, with increasing financial constraints and the trend toward early discharge, avoidance of this perioperative complication is particularly important.

The exact etiology of urinary retention after anorectal surgery is not clear. Therefore, many different methods have been tried to prevent this complication, including use of parasympathomimetic agents [46], use of alpha-adrenergic blockers [6, 7], use of anxiolytic agents [4], restriction of perioperative fluid intake [8, 9], avoidance of anal packing [10], sitz baths [11], use of local anesthesia [12, 13], use of short-acting anesthesia [1], and outpatient surgery [14].

In the present study, we investigated potential risk factors for postoperative urinary retention after anorectal surgery. In addition, we evaluated the effects of perioperative fluid restriction and prophylactic analgesics on urinary retention.

Patients and methods

Incidence and determinants of urinary retention

A total of 2,011 consecutive patients who underwent surgery under spinal anesthesia for benign anorectal disease were included in a study of the incidence and determinants of urinary retention. All surgeries were performed at our institution during the period January through June 2003. Forty patients who underwent preoperative catheterization were excluded from the study. All procedures were performed after the administration of 3% lidocaine for spinal anesthesia. In all cases, 1,000 to 1,500 ml of intravenous fluids was administered within 24 h after surgery. In some patients, a hemostatic gauze pack was inserted into the anal canal at the end of surgery. Continuous epidural administration of eptazocine (45 mg/day) was used postoperatively to relieve pain in patients who requested it preoperatively [13]. In the ward, the patients were given loxoprofen sodium (120 mg) orally or pentazosine (30 mg) intramuscularly for supplementary analgesia when necessary. Catheterization was performed only when the bladder was palpable or distended after attempts by the nursing staff to assist the patient with voiding (such as by warming the lower abdomen or having the patient stand or walk) were unsuccessful and the patient was clearly uncomfortable. Urinary retention was defined as the need for catheterization within 24 h after surgery.

Factors analyzed as risks for urinary retention were age > or <50 years, sex, presence or absence of preoperative urinary symptoms (prior history of voiding difficulty, including frequent voiding, nocturia, poor stream), presence or absence of diabetes mellitus (DM), presence or absence of privacy during the postoperative stay (type of room; single room, double room vs large room), > or <1 ml 3% lidocaine administered for spinal anesthesia, postoperative epidural anesthesia, the various operative procedures, operation time > or <12 min, anal canal packing, intravenous fluids > or <1,000 ml administered within 24 h, and use of analgesic within 24 h.

The incidence of urinary retention was calculated for the various surgical procedures.

Effect of perioperative fluid restriction on urinary retention

To assess whether perioperative fluid restriction prevents urinary retention, we compared the urinary retention rates between the original 2,011 patients in whom postoperative fluids were not restricted and patients in whom postoperative fluids were restricted (n=291). These 291 patients were operated in May of 2004. There were 204 men and 87 women. The mean age was 47.0 years (range 6 to 85 years). The operations performed on the group were hemorrhoidectomy for 161 patients, fistulectomy for 64, hemorrhoidectomy plus fistulectomy for 17, sliding skin graft (SSG)/lateral subcutaneous internal sphincterotomy (LSIS) for 7, incision/drainage for 32, and other anorectal surgery for 10.

As described earlier, patients in the nonrestricted group were given 1,000 to 1,500 ml of intravenous fluids after surgery. They were also given free access to oral fluids. Patients in the fluid restriction group voided on their way to the operating room, and intravenous fluids were limited to <750 ml. Oral fluids were restricted after surgery until spontaneous urination began.

Effect of prophylactic analgesic treatment on urinary retention

To assess whether prophylactic analgesics prevent urinary retention, we compared the urinary retention rates between patients given an indomethacin suppository (100 mg) at the end of surgery (n=291) and patients not given an indomethacin suppository (n=39). These two groups of patients were given 750 ml of intravenous fluids within 24 h after surgery. The former 291 patients were the same patients who were studied as the fluid restriction group. The latter 39 patients were operated in June of 2004. There were 26 men and 13 women. The mean age was 56.6 years (range 18 to 82 years). The operations performed on the group were hemorrhoidectomy for 22 patients, fistulectomy for 11, hemorrhoidectomy plus fistulectomy for 1, SSG/LSIS for 1, and incision/drainage for 4.

Statistical analyses

Univariate analysis by means of the chi-squared test and multiple logistic regression analysis were performed to identify independent risk factors for urinary retention. The between-group differences in urinary retention rates were analyzed by the chi-squared test. Statistical analyses were carried out with STATVIEW 4.5 software (Abacus Concepts Inc., Berkley, CA). Differences were considered significant when P<0.05.

Results

Incidence and determinants of urinary retention

The overall urinary retention rate was 16.7%. Urinary retention rates for the various procedures were as follows: hemorrhoidectomy, 21.9%; fistulectomy for anal fistula, 6.3%; SSG/LSIS for anal stenosis following anal fissure, 17.2%; incision/drainage for perianal abscess, 2.3%; hemorrhoidectomy combined with fistulectomy, 19.8%; mucosal plication and anal encircling for rectal prolapse [15], 29.2%; radical surgical excision for hidradenitis suppurativa, 0%; radical surgical excision for pilonidal sinus, 0%; and surgical excision for condyloma acuminatum, 0%.

Results of univariate analysis for the total patients undergoing anorectal surgery are shown in Table 1. In addition to surgical procedures (hemorrhoidectomy and SSG/LSIS), age >50 years, female sex, preoperative presence of urinary symptoms, DM, postoperative epidural anesthesia, operation time >12 min, anal canal packing, and use of postoperative analgesics were found to be risk factors for urinary retention.
Table 1

Potential risk factors for postoperative urinary retention in patients undergoing anorectal surgery (n=2,011)

No.

Risk factor

No. of patients

Urinary retention rate

P value

1

Age

   

 ≦50 years

1,014

0.137

0.0002

 >50 years

997

0.199

2

Sex

   

 Male

1,375

0.114

<0.0001

 Female

636

0.283

3

Urinary symptoms

   

 Absent

1,834

0.156

<0.0001

 Present

177

0.288

4

DM

   

 Absent

1,956

0.164

0.013

 Present

55

0.291

5

Ward

   

 Single room

244

0.184

0.7114

 Double room

54

0.148

 Large room

1,713

0.166

6

Amount of 3% lidocaine

   

 ≦1 ml

825

0.154

0.1818

 >1 ml

1,178

0.177

7

Epidural anesthesia

   

 Absent

569

0.102

<0.0001

 Present

1,442

0.193

8

Operation for

   

 Hemorrhoid

1,243

0.219

<0.0001

 Anal fistula

349

0.063

 Anal fissure

64

0.172

 Perianal abscess

177

0.023

 Hemorrhoid plus anal fistula

96

0.198

 Rectal prolapse

24

0.292

 Hidradenitis suppurativa

26

0

 Pilonidal sinus

6

0

 Condyloma acuminatum

14

0

 Anorectal dysfunction

7

0.143

 Others

5

0

9

Operation time

   

 ≦12 min

1,076

0.149

0.015

 >12 min

935

0.189

10

Anal canal packing

   

 Absent

1,948

0.163

0.0038

 Present

63

0.302

11

Intravenous fluids

   

 ≦1,000 ml

787

0.154

0.1806

 >1,000 ml

1,224

0.176

12

Analgesics requirement

   

 Absent

1,470

0.144

<0.0001

 Present

541

0.233

The chi-squared test was used to determine the statistical significance. All continuous numerous variables were dichotomized into two groups at the median value

Results of univariate analysis for the hemorrhoidectomy group (n=1,243) and the fistulectomy group (n=349) are shown in Tables 2 and 3, respectively. The results of logistic regression analysis in each of these groups are shown in Tables 4 and 5. In the hemorrhoidectomy group, female sex (odds ratio, 2.058; P<0.0001), presence of preoperative urinary symptoms (odds ratio, 2.669; P<0.0001), diabetes mellitus (odds ratio, 2.248; P=0.0414), need for postoperative analgesics (odds ratio, 1.554; P=0.0033), and more than three hemorrhoids resected (odds ratio, 1.304; P=0.0007) were found to be independent determinants of urinary retention. In the fistulectomy group, female sex (odds ratio, 8.264; P=0.0003), diabetes mellitus (odds ratio, 6.129; P=0.0133), and intravenous fluids >1,000 ml (odds ratio, 4.049; P=0.0127) were found to be independent determinants of urinary retention.
Table 2

Potential risk factors for postoperative urinary retention in patients undergoing hemorrhoidectomy (n=1,243)

No.

Risk factor

No. of patients

Urinary retention rate

P value

1

Age

   

 ≦50 years

518

0.207

0.3473

 >50 years

725

0.229

2

Sex

   

 Male

722

0.172

<0.0001

 Female

521

0.286

3

Urinary symptoms

   

 Absent

1,130

0.205

<0.0001

 Present

113

0.363

4

DM

   

 Absent

1,213

0.215

0.0157

 Present

30

0.4

5

Ward

   

 Single room

147

0.238

0.5968

 Double room

40

0.175

 Large room

1,056

0.219

6

Amount of 3% lidocaine

   

 ≦1 ml

518

0.195

0.0769

 >1 ml

721

0.237

7

Epidural anesthesia

   

 Absent

233

0.185

0.1516

 Present

1,010

0.228

8

Operation time

   

 ≦12 min

589

0.211

0.4623

 >12 min

654

0.228

9

Anal canal packing

   

 Absent

1,196

0.215

0.0415

 Present

47

0.34

10

Intravenous fluids

   

 ≦1,000 ml

425

0.219

0.9606

 >1,000 ml

818

0.22

11

Analgesics requirement

   

 Absent

865

0.194

0.001

 Present

378

0.278

12

Number of hemorrhoids resected

   

 1

349

0.158

<0.0001

 2

448

0.203

 3

388

0.278

 4

58

0.328

The chi-squared test was used to determine the statistical significance

Table 3

Potential risk factors for postoperative urinary retention in patients undergoing fistulectomy (n=349)

No.

Risk factor

No. of patients

Urinary retention rate

P value

1

Age

   

 ≦50 years

253

0.063

0.9798

 >50 years

96

0.062

2

Sex

   

 Male

319

0.05

0.0012

 Female

30

0.2

3

Urinary symptoms

   

 Absent

322

0.062

0.8066

 Present

27

0.074

4

DM

   

 Absent

336

0.057

0.0111

 Present

13

0.231

5

Ward

   

 Single room

45

0.044

0.3982

 Double room

5

0.2

 Large room

299

0.064

6

Amount of 3% lidocaine

   

 ≦1 ml

141

0.043

0.1884

 >1 ml

206

0.078

7

Epidural anesthesia

   

 Absent

87

0.023

0.0765

 Present

262

0.076

8

Operation time

   

 ≦12 min

204

0.054

0.4073

 >12 min

145

0.076

9

Anal canal packing

   

 Absent

341

0.062

0.4671

 Present

8

0.125

10

Intravenous fluids

   

 ≦1,000 ml

159

0.031

0.0263

 >1,000 ml

190

0.089

11

Analgesics requirement

   

 Absent

266

0.056

0.3618

 Present

83

0.084

The chi-squared test was used to determine the statistical significance

Table 4

Logistic regression model for the patients undergoing hemorrhoidectomy (n=1,243)

Group variable

Odds ratio

95% CI

P value

Female sex

2.058

1.546∼2.732

<0.0001

Urinary symptom

2.669

1.734∼4.109

<0.0001

DM

2.248

1.032∼4.898

0.0414

Anal canal packing

1.834

0.956∼3.52

0.068

Analgesic requirement

1.554

1.158∼2.085

0.0033

More than 3 hemorrhoids resected

1.304

1.118∼1.522

0.0007

CI Confidence interval

Table 5

Logistic regression model for the patients undergoing fistulectomy (n=349)

Group variable

Odds ratio

95% CI

P value

Female sex

8.264

2.652∼25.641

0.0003

DM

6.129

1.458∼25.767

0.0133

Intravenous fluids >1,000 ml

4.049

1.348∼12.157

0.0127

CI Confidence interval

Because intravenous fluids >1,000 ml and the need for postoperative analgesics were each shown to be independent risk factors, we supposed that increased perioperative fluids and increased postoperative pain play important roles in the occurrence of urinary retention. That is why we examined prospectively whether perioperative fluid restriction and/or pain control by prophylactic analgesics would decrease the incidence of postoperative urinary retention.

Effect of perioperative fluid restriction on urinary retention

Urinary retention rates of patients with and without perioperative fluid restriction are given in Table 6. Among the total patients undergoing anorectal surgery, the incidence of urinary retention was significantly lower in the fluid restriction group than in the control group (P<0.0001). Similarly, in the subgroup of patients who underwent hemorrhoidectomy, fluid restriction significantly decreased the urinary retention rate (P=0.0028).
Table 6

Urinary retention rates of patients with and without perioperative fluid restriction

 

Nonrestricted group

Fluid restriction group

P value

N

Retention rate

N

Retention rate

Anorectal surgery

2,011

0.167

291

0.079

<0.0001

Hemorrhoidectomy

1,243

0.219

162

0.117

0.0028

Fistulectomy

349

0.063

64

0.031

0.318

The chi-squared test was used to determine the statistical significance

Effect of prophylactic analgesic treatment on urinary retention

Urinary retention rates of patients with and without prophylactic analgesic treatment are given in Table 7. Among the total patients undergoing anorectal surgery, the incidence of urinary retention was significantly lower among patients given the indomethacin suppository at the end of surgery than among those not given the suppository (P=0.0005). Similarly, in the subgroups of patients undergoing hemorrhoidectomy and fistulectomy, use of the prophylactic analgesic significantly decreased the urinary retention rate (P=0.002 and P=0.0406, respectively).
Table 7

Urinary retention rates of patients with and without prophylactic analgesic treatment

 

Untreated group

Treated group

P value

N

Retention rate

N

Retention rate

Anorectal surgery

39

0.25

291

0.079

0.0005

Hemorrhoidectomy

22

0.364

162

0.117

0.002

Fistulectomy

11

0.182

64

0.031

0.0406

The chi-squared test was used to determine the statistical significance

Discussion

The reported incidence of urinary retention after anorectal surgery ranges widely from less than 1% to more than 50% [13, 8, 9, 14]. This variation may be the result of differences between studies in the definition of urinary retention, exclusion criteria, types of operative procedures, and type of anesthesia. Our overall urinary retention rate of 16.7% is within the reported range.

The exact cause of urinary retention after anorectal surgery is not clearly understood. However, previous studies have suggested that retention is due either to inhibited bladder contraction or bladder outlet obstruction [10, 16]. Pompeius showed the major factor leading to urinary retention to be inhibition of the detrusor muscle [10]. Inhibition of the detrusor muscle is the result of a reflex involving afferent fibers of the pudendal nerve, sacral spinal cord, and efferent pelvic sympathetic nerves. Perianal pain, dilatation of the anal canal, and overdistension of the bladder trigger this reflex.

Barone and Lummings, however, suggested that retention following anorectal surgery is due to bladder outlet obstruction [16]. In the acute postoperative setting, sympathetic nerve discharge causes catecholamine release and alpha-adrenergic-mediated contraction of the bladder neck, resulting in functional obstruction of the bladder outlet. Pain, discomfort caused by anal canal packing, anxiety, and an overfilled bladder are thought to cause sympathetic nerve discharge. If the bladder muscle cannot overcome the added resistance at the bladder neck, retention occurs. Thus, postoperative urinary retention seems to involve a combination of factors. For this reason, few reported studies were able to identify independent risk factors by multivariate analysis [3]. By examining a large number of patients undergoing anorectal surgery, we were able to identify several independent risk factors for urinary retention.

Two surgical procedures (hemorrhoidectomy and SSG/LSIS) were found to be the most important determinants of urinary retention. In addition, in patients who underwent hemorrhoidectomy, urinary retention was related to the disease severity, measured as the number of resected hemorrhoids. As the number of resected hemorrhoids increases, the extent of dissection also increases. This may lead to increased postoperative edema and pain and thus give rise to detrusor inhibition and bladder outlet obstruction, resulting in urinary retention. Patients undergoing SSG/LSIS for anal stenosis following anal fissure commonly suffer from anal pain and anal sphincter spasms during the preoperative period. This may lead to severe postoperative pain, resulting in a high rate of urinary retention. No urinary retention occurs in patients in whom the surgery does not invade the anal canal such as in radical excision of hidradenitis suppurativa, pilonidal sinus, or condyloma acuminatum.

Our analysis showed that the need for postoperative analgesia was an independent risk factor for urinary retention. Furthermore, we showed that pain control by prophylactic analgesic treatment significantly suppressed the incidence of postoperative urinary retention. Thus, urinary retention seemed to be related to the degree of pain and not to the side effects of analgesics. Therefore, control of postoperative pain by prophylactic analgesic treatment is considered effective for preventing urinary retention in a significant number of patients following anorectal surgery. Postoperative epidural anesthesia also relieves postoperative pain. However, it simultaneously may impair bladder sensation and the strength of detrusor contraction, leading to retention of urine. Postoperative use of continuous epidural anesthesia did not affect the incidence of urinary retention in our study.

When patients are under spinal anesthesia, which impairs bladder sensation and inhibits the voiding reflex, administration of a large amount of fluids causes overdistension of the bladder, resulting in irreversible dysfunction of the detrusor muscle [17]. Many investigators have shown that increased fluid intake is an important determinant of urinary retention [1, 3, 5]. Bailey and Ferguson reported a reduction in postoperative urinary retention from 15 to 4% in patients for whom both intravenous and oral fluids were restricted [8]. Our analysis showed that intravenous fluid intake of >1,000 ml was an independent risk factor for urinary retention after surgery for anal fistula. In addition, we showed that perioperative fluid restriction significantly decreased the incidence of urinary retention after anorectal surgery. Therefore, perioperative fluid restriction is recommended for prevention of urinary retention after anorectal surgery.

The in vitro study of Tammela and Arjamaa showed that when a strip of rat detrusor muscle was stretched 300% in length and released immediately, there was no reduction in the amplitudes of active tension, but when the strip was stretched to the maximum length for 3 h, a significant reduction was observed [18]. Thus, the duration of stretching is very important in the impaired contractility of the detrusor muscle [17]. Therefore, once urinary retention occurs (especially in cases in which the volume of urine in the bladder is 500 ml or more), intermittent sterile catheterization is needed to avoid irreversible damage to the detrusor muscle, prolonging bladder acontractility and urinary retention.

The influence of sex on urinary retention is controvertible. Some investigators have reported that male sex is a risk factor [3, 19], and others have reported that there is no statistical difference between the sexes [1, 5, 9, 20]. However, surprisingly, female sex was shown in our study to be an independent risk factor. There are several possible reasons for this result. Voiding was hindered in the women because they were lying supine on the bed, immobilized by the intravenous line and hampered by the gauze over the anal region. Women might not object to urinary catheterization as much as men do. The men attempted to void on their own and escape from catheterization, fearing that the procedure will be painful. Women, especially in Japan, might feel more reluctant than men to void in an unfamiliar surrounding without privacy. Further investigation is warranted to clarify the influence of sex on postoperative urinary retention.

Tammela reported previous infravesical obstruction and neuropathy as important predisposing factors for postoperative urinary retention [17]. In accordance with this, previous urinary symptoms and DM were shown in our study to be the independent risk factors. Therefore, preoperative assessment of the patient’s urinary condition and glucose metabolism is needed to predict the occurrence of postoperative urinary retention.

Conclusions

Urinary retention is a common complication after anorectal surgery. Urinary retention occurs most often in patients undergoing hemorrhoidectomy and SSG/LSIS. Furthermore, female sex, preoperative urinary symptoms, DM, large amount of intravenous fluid administered perioperatively, and postoperative pain are independent risk factors for urinary retention in selected cases of anorectal surgery such as hemorrhoidectomy and fistulectomy. Perioperative fluid restriction and adequate pain relief are recommended to prevent such urinary retention. Once urinary retention occurs, intermittent sterile catheterization is required to avoid irreversible damage to the detrusor muscle, which prolongs bladder acontractility and urinary retention.

Copyright information

© Springer-Verlag 2006