Available online at www.sciencedirect.

com

ScienceDirect
Journal of the Chinese Medical Association 78 (2015) 283e286
www.jcma-online.com

Original Article

Measurement of peritoneal fluid urea nitrogen and creatinine levels is useful

to detect iatrogenic urinary tract leakage in colorectal surgery
Jui-Ho Wang a, Ya-Hsin Kung b, Tai-Ming King a, Min-Chi Chang a, Chao-Wen Hsu a,b,*
a
Division of Colorectal Surgery, Kaohsiung Veterans General Hospital, Kaohsiung, Taiwan, ROC
b
Faculty of Medicine, National Yang-Ming University, Taipei, Taiwan, ROC
Received September 2, 2014; accepted November 4, 2014

Abstract

Background: Increased peritoneal drainage after colorectal surgery is a common problem. Measurement of peritoneal fluid urea nitrogen (UN)
and creatinine (Cr) is a diagnostic tool to detect the urinary tract leakage (UTL). We evaluated its application in colorectal surgery.
Methods: We conducted a retrospective chart review study. We enrolled patients with iatrogenic UTL, and measured their UN and Cr levels in
peritoneal fluid and compared them with those in blood and urine. Meanwhile, we assigned patients without UTL to a control group and
compared clinical parameters of both groups.
Results: Twenty-three patients with iatrogenic UTL were recruited. The overall incidence was 0.5%. UN level in peritoneal fluid (322 ± 56 mg/
dL) was significantly higher than that in blood (18.7 ± 4.0 mg/dL, p < 0.001); Cr level in peritoneal fluid (69.7 ± 14.3 mg/dL) was also
significantly higher than that in blood (1.5 ± 0.5 mg/dL, p < 0.001). UN level in peritoneal fluid was significantly higher in the iatrogenic UTL
group than in the control group (322 mL/dL vs. 9.3 mL/dL, p < 0.001); Cr level in peritoneal fluid was also significantly higher (69.7 mg/dL vs.
0.98 mg/dL, p < 0.001).
Conclusion: When increased peritoneal drainage is found postoperatively in colorectal surgery, measurement of UN and Cr levels in peritoneal
fluid can be a useful diagnostic tool to determine intraperitoneal iatrogenic UTL.
Copyright © 2015 Elsevier Taiwan LLC and the Chinese Medical Association. All rights reserved.

Keywords: creatinine; peritoneal fluid; urea nitrogen; urinary tract leakage

1. Introduction ranges from <1% to 8%.2e4 Bladder injuries are rare (<1%)
during colorectal surgery.2 Some iatrogenic UTLs are detected
A common problem after major colorectal surgery is intraoperatively, and repair of the injury can be undertaken
increased clear peritoneal drainage during the postoperative immediately. However, most of the injuries may remain un-
period (>500 mL/d).1 The major differential diagnosis is be- detected intraoperatively and be detected during the post-
tween simple peritoneal drainage and iatrogenic urinary tract operative period.2,5
leakage (UTL), especially the ureteral injury. The incidence of When increased clear peritoneal drainage (>500 mL/d) is
operative ureteral injury with open or laparoscopic surgery detected postoperatively, iatrogenic UTL should be first
considered. Traditionally, abdominal computed tomography
scan, cystography, intravenous pyelography, or uroendoscopy
Conflicts of interest: The authors declare that there are no conflicts of interest is performed to confirm the diagnosis.5 The concept of urea
related to the subject matter or materials discussed in this article. nitrogen (UN)ecreatinine (Cr) disproportion in intraperitoneal
* Corresponding author. Dr. Chao-Wen Hsu, Division of Colorectal Surgery, extravasation of urine was first proposed in 1972.6 Measured
Kaohsiung Veterans General Hospital, 386, Dazhong 1st Road, Kaohsiung
levels of UN and Cr in peritoneal fluid have been compared to
813, Taiwan, ROC.
E-mail address: [email protected] (C.-W. Hsu). their levels in blood, and peritoneal fluid UN and Cr levels

http://dx.doi.org/10.1016/j.jcma.2015.01.006
1726-4901/Copyright © 2015 Elsevier Taiwan LLC and the Chinese Medical Association. All rights reserved.
284 J.-H. Wang et al. / Journal of the Chinese Medical Association 78 (2015) 283e286

similar to the levels in blood are believed to be diagnostic of Twenty-three patients had definite diagnosis of iatrogenic
simple peritoneal drainage rather than UTL.7 This diagnostic UTL, and the overall incidence rate was 0.5%. Due to the
tool is not routinely utilized and is mainly applied in patients retrospective nature of the study, parts of the data were
with major abdominal trauma8 or complicated urological missing. The clinical characteristics are shown in Table 1.
surgery and hysterectomy.9 Although this diagnostic tool Seventeen patients (74%) had open surgery, and six patients
could also be used to detect iatrogenic UTL after major (26%) had laparoscopic surgery. The most common pro-
colorectal surgery, little literature was found through a search cedures were low anterior resection (n ¼ 7) and anterior
of MEDLINE and PubMed. Therefore, in this study, we resection (n ¼ 6). Most of the injuries occurred over the left
retrospectively analyzed patients with iatrogenic UTL and ureter (82.6%), especially over the left middle third (M/3) and
evaluated this diagnostic tool for its possible application in lower third. Two patients had bladder injury. One patient had
colorectal surgery. ureteral injury over the right middle third and one patient had
urethral injury. The common mechanisms were laceration
2. Methods (n ¼ 8), ligation (n ¼ 6), and thermal injury (including those
caused during electrocautery and by vessel-sealing device,
At Kaohsiung Veterans General Hospital, Kaohsiung, n ¼ 5). Two patients had ureteroureterostomy leakage, and
Taiwan, we reviewed the medical records and operation notes two patients had leakage from bladder repair.
for all patients receiving colorectal surgery during UN and Cr levels in the peritoneal fluid, blood, and urine of
2000e2013. We used this database for our study, which was a 23 patients with iatrogenic UTL are shown in Table 2. UN/Cr
retrospective caseecontrol analysis with prospective levels in peritoneal fluid (322.1/69.7 mg/dL) were significantly
collection. higher than those in blood (18.7/1.5 mg/dL, p < 0.001) and
In our hospital, we place at least one JacksonePratt drain significantly lower than those in urine (392.6/127.3 mg/dL,
intraperitoneally in all patients receiving colorectal surgery. p < 0.001). Comparisons between patients with iatrogenic
When massive peritoneal fluid (500 mL/d) from Jacksone- UTL and those of the control group are shown in Table 3.
Pratt drain is detected postoperatively, we send it for mea- Peritoneal fluid amount and the levels of peritoneal UN,
surement of UN and Cr levels to exclude UTL. Meanwhile, we peritoneal Cr, blood UN, and blood Cr were significantly
collect blood and urine samples and measure their UN and Cr higher in the iatrogenic UTL group than in the control group,
levels for comparison. If iatrogenic UTL is highly suspected especially peritoneal UN and peritoneal Cr (322 mL vs.
by the analysis result, we perform sonography, abdominal 9.3 mL for UN, and 69.7 mg/dL vs. 0.98 mg/dL for Cr;
computed tomography scan, cystography, intravenous pye- p < 0.001). The ratio of UN levels in peritoneal fluid and
lography, or uroendoscopy to confirm the diagnosis. blood was 17.9 ± 5.1 (ranged from 10 to 29), and the ratio of
In this study, we included patients with iatrogenic UTL
detected postoperatively. During the same period, we also
assigned all patients with simple increased peritoneal fluid Table 1
Clinical characteristics of 23 patients with urinary tract leakage.
(500 mL/d) but without iatrogenic UTL (confirmed by so-
nography, abdominal computed tomography scan, cystog- Clinical characteristics n ¼ 23
raphy, intravenous pyelography, or uroendoscopy) to the Age (y) 60.2 ± 14 (38e78)
control group. Patients with congestive heart disease, chronic Female/male 10/13
Procedure
liver disease with impaired liver function, and chronic renal Low anterior resection 7
disease with impaired renal function were excluded due to Anterior resection 6
possible interference of UN and Cr levels.10 The clinical pa- Abdominoperineal resection 3
rameters included age, sex, procedure name, site and mecha- Right hemicolectomy 1
nism of UTL, peritoneal fluid amount, and UN and Cr levels of Laparoscopic anterior resection 3
Laparoscopic low anterior resection 2
peritoneal fluid, blood, and urine. We also calculated the ratios Laparoscopic abdominoperineal resection 1
of UN and Cr levels in peritoneal fluid and blood in each group Site of urinary tract injury
of patients for comparison. Prevention and management of Left M/3 ureter 12
UTL were not evaluated in this study. Left L/3 ureter 7
All quantitative data were expressed as mean ± standard Bladder dome 2
Urethra 1
deviation. Analysis of variance with post hoc test and two Right M/3 ureter 1
independent sample t test were used for group comparisons of Cause of urinary tract leakage
quantitative data. Statistical analysis was performed using Laceration 8
SPSS version 12.0 for Windows (SPSS Inc., Chicago, IL, Ligation 6
USA). Significance was defined as p < 0.05. Thermal injury (including electrocautery 5
and vessel-sealing device)
Ureteroureterostomy leakage 2
3. Results Leakage of bladder repair 2
Quantitative data are expressed as mean ± standard deviation (range of the
In 2000e2013, 4493 patients receiving major colorectal value).
surgery were registered in our medical record database. L/3 ¼ lower third; M/3 ¼ middle third.
 J.-H. Wang et al. / Journal of the Chinese Medical Association 78 (2015) 283e286 285

Table 2
UN and Cr levels of 23 patients with urinary tract leakage.
Peritoneal fluid Blood Urine pa
UN (mg/dL) 322.1 ± 56 (200e403) 18.7 ± 4.0 (12e25) 392.6 ± 55 (290e500) <0.001
Cr (mg/dL) 69.7 ± 14.3 (41e91) 1.5 ± 0.5 (0.77e2.67) 127.3 ± 24 (97e177) <0.001
Quantitative data were expressed as mean ± standard deviation (range of the value).
ANOVA ¼ analysis of variance; Cr ¼ creatinine; UN ¼ urea nitrogen.
a
ANOVA was used.

Cr levels in peritoneal fluid and blood was 49.4 ± 17 (ranged also significantly higher (322 mL vs. 9.3 mL for UN, 69.7 mg/
from 29.7 to 89.8). Both were significantly higher in the iat- dL vs. 0.98 mg/dL for Cr). Therefore, we propose that the
rogenic UTL group than in the control group (Table 3). measurement of UN and Cr levels in peritoneal fluid might be
a valid initial test to determine intraperitoneal UTL. If peri-
4. Discussion toneal fluid levels are equal to blood levels, then one might
follow the patients conservatively. Conversely, if peritoneal
Increased postoperative peritoneal fluid is a common con- fluid levels are higher than blood levels, one might proceed
dition in patients with prolonged or complicated colorectal with more extensive radiologic evaluation to detect possible
surgery. UN and Cr levels in peritoneal fluid can be collected urinary leakage.
from excessive postoperative peritoneal fluid to determine In our study, the Cr ratio of peritoneal/blood was much
intraperitoneal urinary leakage. The concept that intraperito- higher than the corresponding UN ratio (49.4 vs. 17.9, Table
neal urine can be detected by UN and Cr disproportion in 3). Actually, blood Cr is an important indicator of renal
peritoneal fluid was proposed in 19726 and has been applied health because it is an easily measured byproduct of muscle
for clinical practices for decades. The reference values of UN metabolism that is excreted unchanged by the kidneys. Blood
and Cr levels in postoperative peritoneal fluid should be Cr can be measured by a simple test and is the most commonly
equivalent to those in blood.1 Increased drainage after used indicator of renal function.11 Conversely, the range of UN
abdominal surgery can be differentiated from peritoneal fluid in blood, urine, and peritoneal fluid is wide because of normal
by testing the drain contents for UN and Cr levels and variations due to protein intake, endogenous protein catabo-
comparing them with their blood levels. If fluid UN and Cr lism, state of hydration, hepatic urea synthesis, and renal urea
levels are equal to blood UN and Cr levels, the drainage is not excretion.12 Therefore, the Cr ratio of peritoneal fluid/blood is
urine. more sensitive to detect urinary tract injury. Besides, perito-
This is the first study reporting the measurement of peri- neal fluid could be mixed with some postoperative exudates or
toneal fluid UN and Cr levels to detect iatrogenic UTL after fluid. Therefore, this could partially explain why, in the control
colorectal surgery. We clearly demonstrated that levels of UN group, the UN level in peritoneal fluid (9.3 ± 3.3 mg/dL) was
and Cr in peritoneal fluid were significantly greater than those much lower than that in blood (15.4 ± 3.5 mg/dL, Table 3). In
in blood (10e29 times for UN, 29.7e89.8 times for Cr) in Table 3, we can see the blood UN and Cr levels in the UTL
patients with UTL. Additionally, when compared with the group (18.7 mg/dL and 1.5 mg/dL, respectively) are slightly
control group, the levels of UN and Cr in peritoneal fluid were higher than those in the control group (15.4 mg/dL and

Table 3
Comparisons between patients with urinary tract leakage (n ¼ 23) and control group (n ¼ 20).
Urinary tract leakage (n ¼ 23) Control group (n ¼ 20) pa
Peritoneal fluid
Fluid amount (mL/d) 1187 ± 616 (590e2900) 773 ± 178 (500e1100) 0.005
UN (mg/dL) 322 ± 56 (200e403) 9.3 ± 3.3 (4e16) <0.001
Cr (mg/dL) 69.7 ± 14.3 (41e91) 0.98 ± 0.28 (0.48e1.64) <0.001
Blood
UN (mg/dL) 18.7 ± 4.0 (12e25) 15.4 ± 3.5 (10e22) 0.006
Cr (mg/dL) 1.5 ± 0.5 (0.77e2.67) 0.89 ± 0.33 (0.45e1.67) <0.001
Urine
UN (mg/dL) 392 ± 55 (290-500) 410 ± 89 (250e589) 0.44
Cr (mg/dL) 127 ± 24 (97-177) 138 ± 31 (98e200) 0.182
UN ratio
Peritoneal/blood 17.9 ± 5.1 (10-29) 0.6 ± 0.1 (0.35e1.0) <0.001
Cr ratio
Peritoneal/blood 49.4 ± 17 (29.7-89.8) 1.1 ± 0.2 (0.54e1.69) <0.001
Quantitative data were expressed as mean ± standard deviation (range of the value).
Cr ¼ creatinine; UN ¼ urea nitrogen.
a
Two independent sample t test.
286 J.-H. Wang et al. / Journal of the Chinese Medical Association 78 (2015) 283e286

0.89 mg/dL), with a significant difference. We postulated that There are some limitations of this study. First, this was a
patients in the UTL group had more complicated situations, retrospective chart review study with limited case number
such as advanced cancer with adjacent urinary tract organ (n ¼ 23), and parts of the data were missing. Hence, the
adhesion. As a result, during the operation, we had to reliability and validity of this diagnostic tool could not be
manipulate the ureter, bladder, or even kidney, and we needed completely assessed. Second, some minor urinary tract injuries
more time to complete the operation, which might have without obvious leakage and increased peritoneal drainage
compromised partial renal function with slightly elevated were not included in our study. There might be a selection
blood UN and Cr levels. bias.
In a retrospective review of >2.1 million colorectal pro- In conclusion, when increased peritoneal fluid is found
cedures identified from a US nationwide database, the risk of postoperatively in colorectal surgery, initial measurement of
ureteral injury was estimated to be 0.28%.4 In our study, 23 UN and Cr levels in peritoneal fluid could be a useful diag-
patients had iatrogenic UTL among a total of 4493 patients nostic tool to determine possible intraperitoneal iatrogenic
reviewed during 2000e2013, including ureter, bladder, and UTL. Of course, this finding deserves further randomized
urethral injury, and the overall incidence rate was 0.5%. Injury controlled trials and their validation in large-scale studies.
to the ureter usually occurs during high ligation of the inferior
mesenteric artery; mobilization of the upper mesorectum near References
the sacral promontory; dissection deep in the pelvis in the
plane between the lower rectum, pelvic sidewall, and bladder 1. Manahan KJ, Fanning J. Peritoneal fluid urea nitrogen and creatinine
base; and dissection of the most cephalad portion of the reference values. Obstet Gynecol 1999;93:780e2.
2. Rose J, Schneider C, Yildirim C, Geers P, Scheidbach H, Kockerling F.
perineal dissection in an abdominoperineal resection.13 In this Complications in laparoscopic colorectal surgery: results of a multicentre
study, most injuries occurred over the left ureter (82.6%), trial. Tech Coloproctol 2004;8(Suppl. 1):s25e8.
especially over the left middle third (during the high ligation 3. Redan JA, McCarus SD. Protect the ureters. JSLS 2009;13:139e41.
of inferior mesenteric artery) and the lower third (over pelvic 4. Halabi WJ, Jafari MD, Nguyen VQ, Carmichael JC, Mills S, Pigazzi A,
sidewall and deep pelvic dissection). One patient had right et al. Ureteral injuries in colorectal surgery: an analysis of trends, out-
comes, and risk factors over a 10-year period in the United States. Dis
ureteral injury caused by electrocautery during right Colon Rectum 2014;57:179e86.
hemicolectomy. 5. Delacroix Jr SE, Winters JC. Urinary tract injuries: recognition and
In 2012, Palaniappa et al14 reported a significant increase in management. Clin Colon Rectal Surg 2010;23:104e12.
the incidence of iatrogenic ureteral injuries during laparoscopy 6. Sullivan MJ, Lackner LH, Banowsky LH. Intraperitoneal extravasation of
compared with that during open colectomies. In our study, the urine. BUNeserum creatinine disproportion. JAMA 1972;221:491e2.
7. Wong MH, Lim SK, Ng KL, Ng KP. Pseudo-acute kidney injury with
incidence was also higher in laparoscopy (1.4%, 6/401) than in recurrent ascites due to intraperitoneal urine leakage. Intern Med J
the open group (0.4%, 17/4092; p ¼ 0.01). Two patients had 2012;42:848e9.
total transection of the ureter by laparoscopic vessel-sealing 8. Rubin MJ, Blahd WH, Stanisic TH, Meislin HW. Diagnosis of intraper-
device (LigaSure, 5 mm Blunt Tip 44 cm; Covidien Inc.). itoneal extravasation of urine by peritoneal lavage. Ann Emerg Med
One had total transection of the left middle ureter during the 1985;14:433e7.
9. Wu HH, Yang PY, Yeh GP, Chou PH, Hsu JC, Lin KC. The detection of
high ligation of inferior mesenteric artery during laparoscopic ureteral injuries after hysterectomy. J Minim Invasive Gynecol
anterior resection, and the other had bilateral ureteral tran- 2006;13:403e8.
section during perineal dissection in laparoscopic abdomi- 10. Nguyen-Khac E, Thevenot T, Capron D, Dharancy S, Paupart T,
noperineal resection. Thabut D, et al. Are ascitic electrolytes usable in cirrhotic patients?
Bladder injury usually occurs when dissecting densely Correlation of sodium, potassium, chloride, urea, and creatinine concen-
trations in ascitic fluid and blood. Eur J Int Med 2008;19:613e8.
adherent rectosigmoid tumors or diverticular disease from the 11. Howard TE. Clinical chemistry. New York: John Wiley and Sons; 1989.
bladder wall.2 Two patients in our study had bladder injury p. 58e62.
due to leakage from the repair. One of them had sigmoid 12. Hosten AO. BUN and creatinine. In: Walker HK, Hall WD, Hurst JW,
cancer with invasion to the bladder dome, and the other was a editors. Clinical methods: the history, physical, and laboratory exami-
case of sigmoid diverticulitis with colovesicle fistula forma- nations. 3rd ed. Chapter 193. Boston: Butterworths; 1990. Available at,
http://www.ncbi.nlm.nih.gov/books/NBK305 [accessed 21.10.14].
tion. One patient with bulky lower rectal cancer undergoing 13. Robin Boushey LJW. Management of intra-abdominal, pelvic, and geni-
neoadjuvant concurrent chemoradiation therapy had urethra tourinary complications of colorectal surgery. 2014 UpToDate®. Available
injury during perineal dissection during an abdominoperineal at: http://www.uptodate.com. [accessed 21.10.14].
resection. The failure rate of ureteroureterostomy was around 14. Palaniappa NC, Telem DA, Ranasinghe NE, Divino CM. Incidence of
10% in most studies.5 Two patients in our study had leakage iatrogenic ureteral injury after laparoscopic colectomy. Arch Surg
2012;147:267e71.
following ureteroureterostomy, probably due to devasculari-
zation of the ureteral vessels.