- Email: [email protected]
Predictive factors for surgical indication in adhesive small bowel obstruction
The American Journal of Surgery (2008) 196, 23–27
Clinical Surgery-International
Predictive factors for surgical indication in adhesive small bowel obstruction Shogo Tanaka, M.D.a,*, Takatsugu Yamamoto, M.D.a, Daisuke Kubota, M.D.b, Mitsuharu Matsuyama, M.D.b, Takahiro Uenishi, M.D.a, Shoji Kubo, M.D.c, Koichi Ono, M.D.a a
Department of Surgery, Ishikiriseiki Hospital, 18-28 Yayoicho, Higashiosaka 579-8026, Osaka, Japan; bDepartment of Surgery, Joto Central Hospital, Osaka, Japan; cDepartment of Hepato-Biliary-Pancreatic Surgery, Osaka City University Graduate School of Medicine, Osaka, Japan KEYWORDS: Ileus; Strangulation; Timing of surgery
Abstract BACKGROUND: Small bowel obstruction (SBO) after abdominal surgery is usually treated successfully with intestinal decompression using a long nasointestinal tube (LT), but some cases fail to respond. METHODS: Clinical background and laboratory data on admission were evaluated retrospectively for 53 patients with adhesive SBO to determine predictive factors for failure of LT decompression, and the appropriate timing of laparotomy was investigated. RESULTS: Complete SBO (no evidence of air within the large bowel) and increased serum creatine phosphokinase (ⱖ130 IU/L) were independent predictive factors for LT decompression failure. Laparotomy was indicated in 14 patients (9 and 5 with complete and partial SBO, respectively), whereas successful LT decompression occurred in 39 patients (9 and 30, respectively). Resolution of SBO took significantly longer for complete SBO (6.3 days) than for partial SBO (2.6 days). CONCLUSIONS: Patients with complete SBO or high serum creatine phosphokinase (CPK) may not respond to LT decompression. Laparotomy is appropriate after non-response for 7 and 3 days for complete and partial SBO, respectively. © 2008 Elsevier Inc. All rights reserved.
Small bowel obstruction (SBO) is usually caused by postoperative adhesions that develop in about 95% of adult patients after abdominal surgery.1 At one time, immediate laparotomy was performed in almost all cases of SBO,2,3 but recent studies on the natural history of adhesive SBO have shown that more than 50% of cases resolve with a conservative, non-operative approach using intravenous hydration, decompression with a short nasogastric tube or long nasointestinal tube (LT), or temporary elimination of oral * Corresponding author. Tel.: ⫹81-72-988-3121; fax: ⫹81-72-9883119. E-mail address: [email protected]
0002-9610/$ - see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.amjsurg.2007.05.048
nourishment.4 –7 In particular, LT decompression has been shown to be successful in greater than 70% of patients with SBO,7–10 but surgical treatment may still be required in some patients because of significant complications, such as strangulation, or the failure of conservative management.4 – 6,11 However, predictive factors for failure of LT decompression (leading to surgical indication) and the timing of subsequent laparotomy have not been established. In this study, patients with adhesive SBO after abdominal surgery who were treated with LT decompression were investigated retrospectively to determine predictive factors for a lack of response (surgical indication) and to examine the appropriate timing for surgery.
24
The American Journal of Surgery, Vol 196, No 1, July 2008
Table 1 Characteristic of 53 patients with adhesive small bowel obstruction Variable Background Mean age, y (range) Gender (male/female) Comorbidities Cardiac diseases Cerebral diseases Past postoperative adhesive small bowel obstruction No. of past laparotomies 1 2 3 Past laparotomy Colorectal surgery Appendectomy Cholecystectomy Gastroduodenal surgery Small bowel resection Gynecological surgery Other abdominal surgery Unknown Complete small bowel obstruction Laboratory data on admission White blood cell count (x103/dL)* C reactive protein (mg/dL)* Blood sugar (mg/dL)* Amylase (IU/L)* CPK (IU/L)* Blood urea nitrogen (mg/dL)*
No. of patients or values 66.9 (17–91) 29/24 6 7 21
47 5 1 11 12 4 13 6 6 5 3 18 9.2 0.2 134 109 ⬎56 17.9
(5.5, 17.0) (0.1, 7.1) (94, 216) (60, 164) (28, 134) (9.8, 37.7)
defined as unequivocal evidence of gas in the colon above the level of peritoneal reflection on abdominal radiographs. Based on these definitions, 18 patients were diagnosed with complete SBO and 35 patients with partial SBO. Blood chemistry and laboratory tests, including white blood cell count, blood sugar level, and serum concentrations of C-reactive protein, amylase, creatine phosphokinase (CPK), and blood urea nitrogen, were performed for all 53 patients at admission. Decompression with a LT (Ileus Tube, Sumitomo Bakelite Co, Akita, Japan) was performed within 6 hours from admission. The LT is 300 cm long, Fr 16 (Fig. 1). It is similar to “Gowen decompression tube”.12 The tube has 3 channnels: a main channel, a sump channel, and the balloon channel. The LT was inserted from nose to stomach, and suction of the content under fluoroscopy. Wire was inserted into the main channel to the tips and LT was advanced to duodenum through pyrolic ring, and to Trietz ligament. In jejunum, balloon was inflated with water until it engaged the wall of the bowel; usually 20 mL of water was required. From the balloon channel, a maximum of 30 mL of water can be inflated (30 mm in diameter). The LT was advanced as far as possible. From the sump channel, there was constant bubbling that confirmed the patency of the main channel. Finally, the wire was removed. Decompression with the LT was performed by producing spontaneous negative pressure at the height of the bed, and by forced negative pressure
*Results are given as medians, with 10th and 90th percentiles.
Patients and Methods The clinical records of 53 patients with adhesive SBO after abdominal surgery who were treated with LT decompression between April 2003 and April 2006 were reviewed retrospectively. SBO was diagnosed by abdominal radiography, including x-ray and computed tomography on admission. Patients who had early postoperative SBO (occurring ⬍30 days after surgery) or SBO in conjunction with Crohn’s disease, documented malignancy, incarcerated hernia, or mesenteric vascular disease were excluded from the study. Patients with SBO who were suspected of harboring an ischemic bowel because of abdominal tenderness, fever, tachycardia, or leukocytosis underwent an immediate operation and were also excluded from the study, as were patients who underwent decompression with a nasogastric tube. The 53 patients (29 males and 24 females) had a mean age of 67 years (range 17 to 91 years). The clinical characteristics of the patients, including comorbidities, prevalence of past SBO, number of past laparotomies, and types of abdominal surgeries, are shown in Table 1. Complete SBO was defined as no clear-cut evidence of air within the large bowel on abdominal radiographs, and partial SBO was
Figure 1 A long nasointestinal tube (Ileus Tube®). The tube has 3 channels: a main channel, a sump channel, and the balloon channel. Drainage occurs proximal and distal to the balloon. (Provided by Sumitomo Bakelite Co., Akita, Japan)
S. Tanaka et al.
Adhesive small bowel obstruction
25
Table 2 Logistic regression analysis of risk factors influencing surgical indication for adhesive small bowel obstruction
This study was conducted in accordance with the Helsinki Declaration and the guidelines of the Ethics Committee of our institution. Written informed consent was obtained from each patient or from parents of patients.
Variable Backgrounds Age (per 1 year) Gender (male) Comorbidities Cardiac diseases Cerebral diseases Past adhesive small bowel obstruction No. of past laparotomies (per 1 operation) Past laparotomy Colorectal surgery Appendectomy Cholecystectomy Gastroduodenal surgery Small bowel resection Gynecological surgery Other abdominal surgery Unknown Complete obstruction Laboratory data on admission White blood cell count (per 103/L) C reactive protein (per 1 mg/dL) Blood sugar (per 1 mg/dL) Amylase (per 1 IU/L) CPK (per 1 IU/L) Blood urea nitrogen (per 1 mg/dL)
P value*
Odds ratio (95% CI)
.90 .86
.997 (.96-1.04) 1.00 (.96-1.04)
.57 .45 .36
.52 (.06-4.91) .42 (.05-3.87) 1.79 (.52-6.14)
.98
-
.17 .39 .29 .26 .68 .68 .97 .98 .008
.22 .48 3.01 2.15 1.46 1.46 6.00
.40
1.00
.37 .25 .31 .08 .16
(.03-1.93) (.09-2.54) (.39-24.3) (.56-8.23) (.24-9.00) (.24-9.00) (1.60-22.5)
.90 (.71-1.14) 1.01 1.01 1.01 1.03
(.995-1.02) (.99-1.02) (.999-1.02) (.99-1.07)
*P values in boldface were entered into multivariate analysis.
through introduction of 500 mL of water using a 50-mL syringe 3 times per day. While under observation, the patients were examined frequently and clinical findings were recorded. Abdominal x-ray was performed every day to document the progress of the tube and the degree of decompression. If the tip progressed down the small bowel, the tube was advanced through the nose as appropriate. The tube was removed when injected contrast medium demonstrated that decompression was complete and the tip with the balloon had passed the stenosis. Laparotomy was performed if LT decompression was judged to have failed, based on no improvement of decompression or on the development of signs of intestinal strangulation; this was defined as surgical indication. Logistic regression analysis was used to identify risk factors for surgical indication. Variables significant at P ⬍.10 by univariate analysis were subjected to stepwise logistic regression analysis to identify independent risk factors (P ⬍.05) for surgical indication. Fisher’s exact test was used to compare the incidence of surgical indication between groups. Statview 5.0 statistical software (Abacus Concepts, Berkeley, CA) was used for these analyses.
Results SBO was successfully treated with LT decompression in 39 patients (74%), whereas LT decompression failed in the remaining 14 patients (26%), who subsequently underwent laparotomy. Resolution was achieved in 30 of 35 patients with partial SBO, but only 9 of 18 patients with complete SBO. In univariate analysis, complete SBO and an increased serum concentration of CPK were possible risk factors for surgical indication (Table 2), and these remained as independent risk factors for surgical indication in multivariate analysis (Table 3). Successful resolution of SBO in 39 patients required LT decompression for a mean of 3.5 days (95% confidence interval [CI] 2.5– 4.5 days) and no patient died. Resolution for the 9 patients with complete SBO took a significantly longer time (mean 6.3 days; 95% CI 3.6 –9.0 days) than for the 30 patients with partial SBO (mean 2.6 days; 95% CI: 1.8 –3.4 days; P ⫽ .0011). Mean length of hospital stay of the 39 patients was 17.2 days (95% CI 12.8 –21.5 days). Recurrence of SBO was present in 7 patients (18%). For the 14 patients in whom LT decompression was unsuccessful, the mean interval until laparotomy was 8.4 days (95% CI 5.1–11.8 days). The mean intervals to laparotomy for 9 patients with complete SBO and for 5 patients with partial SBO were 7.2 days (95% CI 2.9 –11.6 days) and 10.6 days (95% CI 5.4 –15.8 days), respectively (P ⫽ .37). Five patients underwent adhesiolysis alone and the remaining 9 patients underwent adhesiolysis with concomitant limited resection of the small bowel. Small bowel resection was performed because of ischemic changes in 4 cases and due to irreversible stenosis caused by an adhesive band in 5 patients (the balloon of the LT with at least 10 mL of air could not pass the stenosis of the small bowel). The mean interval until laparotomy for the 9 patients who required bowel resection was 5.4 days (95% CI 1.9 – 8.4 days), which was significantly shorter than that for the 5 patients who underwent adhesiolysis alone (13.8 days; 95% CI 9.9 –17.7 days, P ⫽ .012). Of the 9 patients with complete SBO who underwent operation, 3 underwent operation within 48 hours and 6 after 48 hours. All 3 patients (100%) operated
Table 3 Multiple logistic regression analysis of risk factors influencing surgical indication for adhesive small bowel obstruction Variables
P value
Odds ratio (95% CI)
Complete obstruction CPK (per 1 IU/L)
.0016 .0137
24.8 (3.4-181) 1.02 (1.004-1.034)
26
The American Journal of Surgery, Vol 196, No 1, July 2008
Table 4 Frequency of surgical indication associated with types of small bowel obstruction and concentration of CPK Small bowel obstruction Concentration of CPK
Complete (n ⫽ 18)
Partial (n ⫽ 35)
Total
ⱖ 130 (n ⫽ 6) ⬍ 130 (n ⫽ 47)
2/2 (100%) 7/16 (40%) 9/18 (50%)
2/4 (50%) 3/31 (10%) 5/35 (14%)
4/6 (67%) 10/47 (21%) 14/53 (26%)
on within 48 hours and one (17%) of the 6 patients operated on after 48 hours had bowel ischemia (P ⫽ .048). Of the 14 patients who underwent laparotomy, 5 had postoperative complications, such as intractable diarrhea, wound infection, methicillin-resistant staphylococcus aureus colitis, renal failure, and leakage from anastomosis. Of the 14 patients who underwent laparotomy, 2 died of brain infarction probably due to hydration by short bowel syndrome and sepsis 124 and 4 days after surgery, respectively. Lengths of total hospital stay and postoperative hospital stay of the 14 patients were 35.6 days (95% CI 20.7–50.5 days, P ⫽ .003 vs those of 39 conservatively treated patients) and 33.6 (95% CI 17.4 – 49.7 days), respectively. Of 5 patients with partial SBO, 1 underwent laparotomy within 3 days (early group) and the remaining 4 did so after more than 3 days (delay group). Of 9 patients with complete SBO, 6 underwent laparotomy within 7 days (early group) and the remaining 3 after more than 7 days (delay group). Postoperative complications occurred in 4 (57%) of the 7 patients in early group and 1 (14%) of the 7 patients in the delay group (P ⫽ .27). The rate of surgical indication was significantly higher in patients with a serum CPK concentration of ⱖ130 IU/L, compared to patients with a serum CPK concentration of less than 130 IU/L (4/6, 67% vs 10/47, 21%; P ⫽ .036). LT decompression failed in both patients (2/2, 100%) with complete SBO and a serum CPK concentration of ⱖ130 IU/L (Table 4). However, the rate of surgical indication in patients with partial SBO and a serum CPK concentration of less than 130 IU/L was only 10% (3 of 31 patients, P ⫽ .019).
Comments Our results showed that LT decompression successfully resolved adhesive SBO in 39 of 53 patients (74%), consistent with other reports.7–10 Many surgeons recommend an initial trial of intestinal decompression, particularly with partial SBO, since this approach may avoid the need for surgery in the majority of the patients;9,11,13,14 our results support the position that LT decompression is effective treatment for partial SBO. Conversely, other surgeons prefer early operative intervention, especially for complete SBO, because of the difficulty in distinguishing simple from
strangulated obstruction and the high complication rate associated with delayed operative intervention.15–17 In our study, complete SBO was an independent risk factor for surgical indication, but complete SBO was resolved by LT decompression in 50% of patients (9 of 18); previous reports have indicated that 20% to 50% of patients with complete SBO respond to LT decompression.9,10 Collectively, our findings suggest that LT decompression should be performed on admission for patients with either complete or partial SBO, and that urgent operative intervention is unnecessary in most cases. The serum concentration of CPK in patients with strangulated SBO is well known to be higher than in those without strangulated SBO; this indicates that the serum concentration of CPK reflects intestinal necrosis.18,19 We found that an increased serum CPK level was an independent predictive factor for surgical indication in patients with adhesive SBO, although the cutoff value was within the normal reference range (normal: ⬍180 IU/L). Increased intraluminal pressure is thought to cause ischemic necrosis of the bowel,10 and we suggest that an increase in serum CPK (especially to ⱖ130 IU/L) may indicate the severity of adhesive SBO, even without strangulation. Therefore, careful observation is needed for patients with a serum CPK concentration of ⱖ130 IU/L, even in cases of partial SBO. However, we note that the number of patients with a serum CPK concentration of ⱖ130 IU/L was small in the current study (6 patients), and more data are needed to establish the relationship of serum CPK with surgical indication. The timing of operative intervention is probably most controversial in patients with adhesive SBO. Some authors recommend that if a patient with SBO fails to respond in 24 to 72 hours, laparotomy should be considered.6,7,11,20 However, we propose that this timing should be dependent on the type of SBO, because the duration until resolution differed for partial and complete SBO in our study. Specifically, we suggest that laparotomy should be performed if patients with partial SBO fail to respond in 3 days or if those with complete SBO fail to respond in 7 days. In conclusion, LT decompression is the most appropriate initial treatment for all patients with adhesive SBO, provided that there is no suspicion of an ischemic bowel on admission. Although most cases of SBO are successfully treated with LT decompression, some fail to respond, and the risk factors for non-response and surgical indication for laparotomy are complete SBO and a serum CPK concentration of ⱖ130 IU/L. For patients with such risk factors, correction of fluid and electrolyte deficits and preoperative examinations, including chest roentgenogram and electrocardiogram, can be completed simultaneously with tube advancement through the small bowel.
References 1. Menzies D, Ellis H. Intestinal obstruction from adhesions— how big is the problem? Ann R Coll Surg Engl 1990;72:60 –3.
S. Tanaka et al.
Adhesive small bowel obstruction
2. Davidson AT Sr. Early operation in the treatment of small bowel obstruction. J Natl Med Assoc 1981;73:245– 6. 3. Fabri PJ, Rosemurgy A. Reoperation for small intestinal obstruction. Surg Clin North Am 1991;71:131– 46. 4. Seror D, Feigin E, Szold A, et al. How conservatively can postoperative small bowel obstruction be treated? Am J Surg 1993;165:121–5. 5. Sosa J, Gardner B. Management of patients diagnosed as acute intestinal obstruction secondary to adhesions. Am Surg 1993;59:125– 8. 6. Cox MR, Gunn IF, Eastman MC, et al. The safety and duration of non-operative treatment for adhesive small bowel obstruction. Aust N Z J Surg 1993;63:367–71. 7. Gowen GF. Long tube decompression is successful in 90% of patients with adhesive small bowel obstruction. Am J Surg 2003;185:512–5. 8. Quatromoni JC, Rosoff L Sr, Halls JM, et al. Early postoperative small bowel obstruction. Ann Surg 1980;191:72– 4. 9. Wolfson PJ, Bauer JJ, Gelernt IM, et al. Use of the long tube in the management of patients with small-intestinal obstruction due to adhesions. Arch Surg 1985;120:1001– 6. 10. Fleshner PR, Siegman MG, Slater GI, et al. A prospective, randomized trial of short versus long tubes in adhesive small-bowel obstruction. Am J Surg 1995;170:366 –70. 11. Hofstetter SR. Acute adhesive obstruction of the small intestine. Surg Gynecol Obstet 1981;152:141– 4.
27 12. Gowen GF. Rapid resolution of small-bowel obstruction with the long tube, endoscopically advanced into the jejunum. Am J Surg 2007;193: 184 –9. 13. Brolin RE, Krasna MJ, Mast BA. Use of tubes and radiographs in the management of small bowel obstruction. Ann Surg 1987;206:126 –33. 14. Bizer LS, Liebling RW, Delany HM, et al. Small bowel obstruction: the role of nonoperative treatment in simple intestinal obstruction and predictive criteria for strangulation obstruction. Surgery 1981; 89:407–13. 15. Sarr MG, Bulkley GB, Zuidema GD. Preoperative recognition of intestinal strangulation obstruction. Prospective evaluation of diagnostic capability. Am J Surg 1983;145:176 – 82. 16. Becker WF. Acute adhesive ileus; a study of 412 cases with particular reference to the abuse of tube decompression in treatment. Surg Gynecol Obstet 1952;95:472– 6. 17. Turner DM, Croom RD 3rd. Acute adhesive obstruction of the small intestine. Am Surg 1983;49:126 –30. 18. Graeber GM, O’Neill JF, Wolf RE, et al. Elevated levels of peripheral serum creatine phosphokinase with strangulated small bowel obstruction. Arch Surg 1983;118:837– 40. 19. Mukai M, Tamaki T, Noto T, et al. A new mechanism of serum creatine phosphokinase elevation in strangulated small bowel obstruction: an experimental rat model. J Int Med Res 1995;23:184 –90. 20. Brolin RE. Partial small bowel obstruction. Surgery 1984;95:145–9.
Clinical Surgery-International
Predictive factors for surgical indication in adhesive small bowel obstruction Shogo Tanaka, M.D.a,*, Takatsugu Yamamoto, M.D.a, Daisuke Kubota, M.D.b, Mitsuharu Matsuyama, M.D.b, Takahiro Uenishi, M.D.a, Shoji Kubo, M.D.c, Koichi Ono, M.D.a a
Department of Surgery, Ishikiriseiki Hospital, 18-28 Yayoicho, Higashiosaka 579-8026, Osaka, Japan; bDepartment of Surgery, Joto Central Hospital, Osaka, Japan; cDepartment of Hepato-Biliary-Pancreatic Surgery, Osaka City University Graduate School of Medicine, Osaka, Japan KEYWORDS: Ileus; Strangulation; Timing of surgery
Abstract BACKGROUND: Small bowel obstruction (SBO) after abdominal surgery is usually treated successfully with intestinal decompression using a long nasointestinal tube (LT), but some cases fail to respond. METHODS: Clinical background and laboratory data on admission were evaluated retrospectively for 53 patients with adhesive SBO to determine predictive factors for failure of LT decompression, and the appropriate timing of laparotomy was investigated. RESULTS: Complete SBO (no evidence of air within the large bowel) and increased serum creatine phosphokinase (ⱖ130 IU/L) were independent predictive factors for LT decompression failure. Laparotomy was indicated in 14 patients (9 and 5 with complete and partial SBO, respectively), whereas successful LT decompression occurred in 39 patients (9 and 30, respectively). Resolution of SBO took significantly longer for complete SBO (6.3 days) than for partial SBO (2.6 days). CONCLUSIONS: Patients with complete SBO or high serum creatine phosphokinase (CPK) may not respond to LT decompression. Laparotomy is appropriate after non-response for 7 and 3 days for complete and partial SBO, respectively. © 2008 Elsevier Inc. All rights reserved.
Small bowel obstruction (SBO) is usually caused by postoperative adhesions that develop in about 95% of adult patients after abdominal surgery.1 At one time, immediate laparotomy was performed in almost all cases of SBO,2,3 but recent studies on the natural history of adhesive SBO have shown that more than 50% of cases resolve with a conservative, non-operative approach using intravenous hydration, decompression with a short nasogastric tube or long nasointestinal tube (LT), or temporary elimination of oral * Corresponding author. Tel.: ⫹81-72-988-3121; fax: ⫹81-72-9883119. E-mail address: [email protected]
0002-9610/$ - see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.amjsurg.2007.05.048
nourishment.4 –7 In particular, LT decompression has been shown to be successful in greater than 70% of patients with SBO,7–10 but surgical treatment may still be required in some patients because of significant complications, such as strangulation, or the failure of conservative management.4 – 6,11 However, predictive factors for failure of LT decompression (leading to surgical indication) and the timing of subsequent laparotomy have not been established. In this study, patients with adhesive SBO after abdominal surgery who were treated with LT decompression were investigated retrospectively to determine predictive factors for a lack of response (surgical indication) and to examine the appropriate timing for surgery.
24
The American Journal of Surgery, Vol 196, No 1, July 2008
Table 1 Characteristic of 53 patients with adhesive small bowel obstruction Variable Background Mean age, y (range) Gender (male/female) Comorbidities Cardiac diseases Cerebral diseases Past postoperative adhesive small bowel obstruction No. of past laparotomies 1 2 3 Past laparotomy Colorectal surgery Appendectomy Cholecystectomy Gastroduodenal surgery Small bowel resection Gynecological surgery Other abdominal surgery Unknown Complete small bowel obstruction Laboratory data on admission White blood cell count (x103/dL)* C reactive protein (mg/dL)* Blood sugar (mg/dL)* Amylase (IU/L)* CPK (IU/L)* Blood urea nitrogen (mg/dL)*
No. of patients or values 66.9 (17–91) 29/24 6 7 21
47 5 1 11 12 4 13 6 6 5 3 18 9.2 0.2 134 109 ⬎56 17.9
(5.5, 17.0) (0.1, 7.1) (94, 216) (60, 164) (28, 134) (9.8, 37.7)
defined as unequivocal evidence of gas in the colon above the level of peritoneal reflection on abdominal radiographs. Based on these definitions, 18 patients were diagnosed with complete SBO and 35 patients with partial SBO. Blood chemistry and laboratory tests, including white blood cell count, blood sugar level, and serum concentrations of C-reactive protein, amylase, creatine phosphokinase (CPK), and blood urea nitrogen, were performed for all 53 patients at admission. Decompression with a LT (Ileus Tube, Sumitomo Bakelite Co, Akita, Japan) was performed within 6 hours from admission. The LT is 300 cm long, Fr 16 (Fig. 1). It is similar to “Gowen decompression tube”.12 The tube has 3 channnels: a main channel, a sump channel, and the balloon channel. The LT was inserted from nose to stomach, and suction of the content under fluoroscopy. Wire was inserted into the main channel to the tips and LT was advanced to duodenum through pyrolic ring, and to Trietz ligament. In jejunum, balloon was inflated with water until it engaged the wall of the bowel; usually 20 mL of water was required. From the balloon channel, a maximum of 30 mL of water can be inflated (30 mm in diameter). The LT was advanced as far as possible. From the sump channel, there was constant bubbling that confirmed the patency of the main channel. Finally, the wire was removed. Decompression with the LT was performed by producing spontaneous negative pressure at the height of the bed, and by forced negative pressure
*Results are given as medians, with 10th and 90th percentiles.
Patients and Methods The clinical records of 53 patients with adhesive SBO after abdominal surgery who were treated with LT decompression between April 2003 and April 2006 were reviewed retrospectively. SBO was diagnosed by abdominal radiography, including x-ray and computed tomography on admission. Patients who had early postoperative SBO (occurring ⬍30 days after surgery) or SBO in conjunction with Crohn’s disease, documented malignancy, incarcerated hernia, or mesenteric vascular disease were excluded from the study. Patients with SBO who were suspected of harboring an ischemic bowel because of abdominal tenderness, fever, tachycardia, or leukocytosis underwent an immediate operation and were also excluded from the study, as were patients who underwent decompression with a nasogastric tube. The 53 patients (29 males and 24 females) had a mean age of 67 years (range 17 to 91 years). The clinical characteristics of the patients, including comorbidities, prevalence of past SBO, number of past laparotomies, and types of abdominal surgeries, are shown in Table 1. Complete SBO was defined as no clear-cut evidence of air within the large bowel on abdominal radiographs, and partial SBO was
Figure 1 A long nasointestinal tube (Ileus Tube®). The tube has 3 channels: a main channel, a sump channel, and the balloon channel. Drainage occurs proximal and distal to the balloon. (Provided by Sumitomo Bakelite Co., Akita, Japan)
S. Tanaka et al.
Adhesive small bowel obstruction
25
Table 2 Logistic regression analysis of risk factors influencing surgical indication for adhesive small bowel obstruction
This study was conducted in accordance with the Helsinki Declaration and the guidelines of the Ethics Committee of our institution. Written informed consent was obtained from each patient or from parents of patients.
Variable Backgrounds Age (per 1 year) Gender (male) Comorbidities Cardiac diseases Cerebral diseases Past adhesive small bowel obstruction No. of past laparotomies (per 1 operation) Past laparotomy Colorectal surgery Appendectomy Cholecystectomy Gastroduodenal surgery Small bowel resection Gynecological surgery Other abdominal surgery Unknown Complete obstruction Laboratory data on admission White blood cell count (per 103/L) C reactive protein (per 1 mg/dL) Blood sugar (per 1 mg/dL) Amylase (per 1 IU/L) CPK (per 1 IU/L) Blood urea nitrogen (per 1 mg/dL)
P value*
Odds ratio (95% CI)
.90 .86
.997 (.96-1.04) 1.00 (.96-1.04)
.57 .45 .36
.52 (.06-4.91) .42 (.05-3.87) 1.79 (.52-6.14)
.98
-
.17 .39 .29 .26 .68 .68 .97 .98 .008
.22 .48 3.01 2.15 1.46 1.46 6.00
.40
1.00
.37 .25 .31 .08 .16
(.03-1.93) (.09-2.54) (.39-24.3) (.56-8.23) (.24-9.00) (.24-9.00) (1.60-22.5)
.90 (.71-1.14) 1.01 1.01 1.01 1.03
(.995-1.02) (.99-1.02) (.999-1.02) (.99-1.07)
*P values in boldface were entered into multivariate analysis.
through introduction of 500 mL of water using a 50-mL syringe 3 times per day. While under observation, the patients were examined frequently and clinical findings were recorded. Abdominal x-ray was performed every day to document the progress of the tube and the degree of decompression. If the tip progressed down the small bowel, the tube was advanced through the nose as appropriate. The tube was removed when injected contrast medium demonstrated that decompression was complete and the tip with the balloon had passed the stenosis. Laparotomy was performed if LT decompression was judged to have failed, based on no improvement of decompression or on the development of signs of intestinal strangulation; this was defined as surgical indication. Logistic regression analysis was used to identify risk factors for surgical indication. Variables significant at P ⬍.10 by univariate analysis were subjected to stepwise logistic regression analysis to identify independent risk factors (P ⬍.05) for surgical indication. Fisher’s exact test was used to compare the incidence of surgical indication between groups. Statview 5.0 statistical software (Abacus Concepts, Berkeley, CA) was used for these analyses.
Results SBO was successfully treated with LT decompression in 39 patients (74%), whereas LT decompression failed in the remaining 14 patients (26%), who subsequently underwent laparotomy. Resolution was achieved in 30 of 35 patients with partial SBO, but only 9 of 18 patients with complete SBO. In univariate analysis, complete SBO and an increased serum concentration of CPK were possible risk factors for surgical indication (Table 2), and these remained as independent risk factors for surgical indication in multivariate analysis (Table 3). Successful resolution of SBO in 39 patients required LT decompression for a mean of 3.5 days (95% confidence interval [CI] 2.5– 4.5 days) and no patient died. Resolution for the 9 patients with complete SBO took a significantly longer time (mean 6.3 days; 95% CI 3.6 –9.0 days) than for the 30 patients with partial SBO (mean 2.6 days; 95% CI: 1.8 –3.4 days; P ⫽ .0011). Mean length of hospital stay of the 39 patients was 17.2 days (95% CI 12.8 –21.5 days). Recurrence of SBO was present in 7 patients (18%). For the 14 patients in whom LT decompression was unsuccessful, the mean interval until laparotomy was 8.4 days (95% CI 5.1–11.8 days). The mean intervals to laparotomy for 9 patients with complete SBO and for 5 patients with partial SBO were 7.2 days (95% CI 2.9 –11.6 days) and 10.6 days (95% CI 5.4 –15.8 days), respectively (P ⫽ .37). Five patients underwent adhesiolysis alone and the remaining 9 patients underwent adhesiolysis with concomitant limited resection of the small bowel. Small bowel resection was performed because of ischemic changes in 4 cases and due to irreversible stenosis caused by an adhesive band in 5 patients (the balloon of the LT with at least 10 mL of air could not pass the stenosis of the small bowel). The mean interval until laparotomy for the 9 patients who required bowel resection was 5.4 days (95% CI 1.9 – 8.4 days), which was significantly shorter than that for the 5 patients who underwent adhesiolysis alone (13.8 days; 95% CI 9.9 –17.7 days, P ⫽ .012). Of the 9 patients with complete SBO who underwent operation, 3 underwent operation within 48 hours and 6 after 48 hours. All 3 patients (100%) operated
Table 3 Multiple logistic regression analysis of risk factors influencing surgical indication for adhesive small bowel obstruction Variables
P value
Odds ratio (95% CI)
Complete obstruction CPK (per 1 IU/L)
.0016 .0137
24.8 (3.4-181) 1.02 (1.004-1.034)
26
The American Journal of Surgery, Vol 196, No 1, July 2008
Table 4 Frequency of surgical indication associated with types of small bowel obstruction and concentration of CPK Small bowel obstruction Concentration of CPK
Complete (n ⫽ 18)
Partial (n ⫽ 35)
Total
ⱖ 130 (n ⫽ 6) ⬍ 130 (n ⫽ 47)
2/2 (100%) 7/16 (40%) 9/18 (50%)
2/4 (50%) 3/31 (10%) 5/35 (14%)
4/6 (67%) 10/47 (21%) 14/53 (26%)
on within 48 hours and one (17%) of the 6 patients operated on after 48 hours had bowel ischemia (P ⫽ .048). Of the 14 patients who underwent laparotomy, 5 had postoperative complications, such as intractable diarrhea, wound infection, methicillin-resistant staphylococcus aureus colitis, renal failure, and leakage from anastomosis. Of the 14 patients who underwent laparotomy, 2 died of brain infarction probably due to hydration by short bowel syndrome and sepsis 124 and 4 days after surgery, respectively. Lengths of total hospital stay and postoperative hospital stay of the 14 patients were 35.6 days (95% CI 20.7–50.5 days, P ⫽ .003 vs those of 39 conservatively treated patients) and 33.6 (95% CI 17.4 – 49.7 days), respectively. Of 5 patients with partial SBO, 1 underwent laparotomy within 3 days (early group) and the remaining 4 did so after more than 3 days (delay group). Of 9 patients with complete SBO, 6 underwent laparotomy within 7 days (early group) and the remaining 3 after more than 7 days (delay group). Postoperative complications occurred in 4 (57%) of the 7 patients in early group and 1 (14%) of the 7 patients in the delay group (P ⫽ .27). The rate of surgical indication was significantly higher in patients with a serum CPK concentration of ⱖ130 IU/L, compared to patients with a serum CPK concentration of less than 130 IU/L (4/6, 67% vs 10/47, 21%; P ⫽ .036). LT decompression failed in both patients (2/2, 100%) with complete SBO and a serum CPK concentration of ⱖ130 IU/L (Table 4). However, the rate of surgical indication in patients with partial SBO and a serum CPK concentration of less than 130 IU/L was only 10% (3 of 31 patients, P ⫽ .019).
Comments Our results showed that LT decompression successfully resolved adhesive SBO in 39 of 53 patients (74%), consistent with other reports.7–10 Many surgeons recommend an initial trial of intestinal decompression, particularly with partial SBO, since this approach may avoid the need for surgery in the majority of the patients;9,11,13,14 our results support the position that LT decompression is effective treatment for partial SBO. Conversely, other surgeons prefer early operative intervention, especially for complete SBO, because of the difficulty in distinguishing simple from
strangulated obstruction and the high complication rate associated with delayed operative intervention.15–17 In our study, complete SBO was an independent risk factor for surgical indication, but complete SBO was resolved by LT decompression in 50% of patients (9 of 18); previous reports have indicated that 20% to 50% of patients with complete SBO respond to LT decompression.9,10 Collectively, our findings suggest that LT decompression should be performed on admission for patients with either complete or partial SBO, and that urgent operative intervention is unnecessary in most cases. The serum concentration of CPK in patients with strangulated SBO is well known to be higher than in those without strangulated SBO; this indicates that the serum concentration of CPK reflects intestinal necrosis.18,19 We found that an increased serum CPK level was an independent predictive factor for surgical indication in patients with adhesive SBO, although the cutoff value was within the normal reference range (normal: ⬍180 IU/L). Increased intraluminal pressure is thought to cause ischemic necrosis of the bowel,10 and we suggest that an increase in serum CPK (especially to ⱖ130 IU/L) may indicate the severity of adhesive SBO, even without strangulation. Therefore, careful observation is needed for patients with a serum CPK concentration of ⱖ130 IU/L, even in cases of partial SBO. However, we note that the number of patients with a serum CPK concentration of ⱖ130 IU/L was small in the current study (6 patients), and more data are needed to establish the relationship of serum CPK with surgical indication. The timing of operative intervention is probably most controversial in patients with adhesive SBO. Some authors recommend that if a patient with SBO fails to respond in 24 to 72 hours, laparotomy should be considered.6,7,11,20 However, we propose that this timing should be dependent on the type of SBO, because the duration until resolution differed for partial and complete SBO in our study. Specifically, we suggest that laparotomy should be performed if patients with partial SBO fail to respond in 3 days or if those with complete SBO fail to respond in 7 days. In conclusion, LT decompression is the most appropriate initial treatment for all patients with adhesive SBO, provided that there is no suspicion of an ischemic bowel on admission. Although most cases of SBO are successfully treated with LT decompression, some fail to respond, and the risk factors for non-response and surgical indication for laparotomy are complete SBO and a serum CPK concentration of ⱖ130 IU/L. For patients with such risk factors, correction of fluid and electrolyte deficits and preoperative examinations, including chest roentgenogram and electrocardiogram, can be completed simultaneously with tube advancement through the small bowel.
References 1. Menzies D, Ellis H. Intestinal obstruction from adhesions— how big is the problem? Ann R Coll Surg Engl 1990;72:60 –3.
S. Tanaka et al.
Adhesive small bowel obstruction
2. Davidson AT Sr. Early operation in the treatment of small bowel obstruction. J Natl Med Assoc 1981;73:245– 6. 3. Fabri PJ, Rosemurgy A. Reoperation for small intestinal obstruction. Surg Clin North Am 1991;71:131– 46. 4. Seror D, Feigin E, Szold A, et al. How conservatively can postoperative small bowel obstruction be treated? Am J Surg 1993;165:121–5. 5. Sosa J, Gardner B. Management of patients diagnosed as acute intestinal obstruction secondary to adhesions. Am Surg 1993;59:125– 8. 6. Cox MR, Gunn IF, Eastman MC, et al. The safety and duration of non-operative treatment for adhesive small bowel obstruction. Aust N Z J Surg 1993;63:367–71. 7. Gowen GF. Long tube decompression is successful in 90% of patients with adhesive small bowel obstruction. Am J Surg 2003;185:512–5. 8. Quatromoni JC, Rosoff L Sr, Halls JM, et al. Early postoperative small bowel obstruction. Ann Surg 1980;191:72– 4. 9. Wolfson PJ, Bauer JJ, Gelernt IM, et al. Use of the long tube in the management of patients with small-intestinal obstruction due to adhesions. Arch Surg 1985;120:1001– 6. 10. Fleshner PR, Siegman MG, Slater GI, et al. A prospective, randomized trial of short versus long tubes in adhesive small-bowel obstruction. Am J Surg 1995;170:366 –70. 11. Hofstetter SR. Acute adhesive obstruction of the small intestine. Surg Gynecol Obstet 1981;152:141– 4.
27 12. Gowen GF. Rapid resolution of small-bowel obstruction with the long tube, endoscopically advanced into the jejunum. Am J Surg 2007;193: 184 –9. 13. Brolin RE, Krasna MJ, Mast BA. Use of tubes and radiographs in the management of small bowel obstruction. Ann Surg 1987;206:126 –33. 14. Bizer LS, Liebling RW, Delany HM, et al. Small bowel obstruction: the role of nonoperative treatment in simple intestinal obstruction and predictive criteria for strangulation obstruction. Surgery 1981; 89:407–13. 15. Sarr MG, Bulkley GB, Zuidema GD. Preoperative recognition of intestinal strangulation obstruction. Prospective evaluation of diagnostic capability. Am J Surg 1983;145:176 – 82. 16. Becker WF. Acute adhesive ileus; a study of 412 cases with particular reference to the abuse of tube decompression in treatment. Surg Gynecol Obstet 1952;95:472– 6. 17. Turner DM, Croom RD 3rd. Acute adhesive obstruction of the small intestine. Am Surg 1983;49:126 –30. 18. Graeber GM, O’Neill JF, Wolf RE, et al. Elevated levels of peripheral serum creatine phosphokinase with strangulated small bowel obstruction. Arch Surg 1983;118:837– 40. 19. Mukai M, Tamaki T, Noto T, et al. A new mechanism of serum creatine phosphokinase elevation in strangulated small bowel obstruction: an experimental rat model. J Int Med Res 1995;23:184 –90. 20. Brolin RE. Partial small bowel obstruction. Surgery 1984;95:145–9.