Surg Endosc DOI 10.1007/s00464-014-4015-7
and Other Interventional Techniques
Laparoscopic versus open surgical management of adhesive small bowel obstruction: a comparison of outcomes James Byrne • Fady Saleh • Luciano Ambrosini Fayez Quereshy • Timothy D. Jackson • Allan Okrainec
•
Received: 7 February 2014 / Accepted: 9 November 2014 Ó Springer Science+Business Media New York 2014
Abstract Background Laparoscopic management of adhesive small bowel obstruction (SBO) has become an established technique within the domain of acute care surgery. As minimally invasive management of SBO becomes more widely accepted, there is increased need for reporting of outcomes. Objective To compare outcomes of laparoscopic versus open surgery for adhesive SBO. Methods Patients undergoing surgery for adhesive SBO at our institution between 2005 and 2013 were eligible for inclusion. The primary outcome was overall complication rate, while secondary outcomes included operative time, gastrointestinal (GI) function, and postoperative length of stay (LOS). Univariable analysis compared laparoscopic (including conversions) and open groups with regard to patient baseline and perioperative characteristics as well as outcomes of interest. Multivariable analysis was performed comparing the endpoint of overall complications between groups. Sensitivity analysis excluding patients who underwent bowel resection was performed to assess effect
on outcomes. Factors associated with laparoscopic success, as well as impact of conversion to open on postoperative outcomes, are reported. Results A cohort of 269 patients with adhesive SBO was identified: 186 patients (69.1 %) underwent open surgery, 83 (30.9 %) were managed laparoscopically. Within the laparoscopy group, 32 (38.6 %) underwent conversion to open. Operative time was similar between groups (P = 0.506), while laparoscopy was associated with quicker recovery of GI function indicated by removal of nasogastric tube (P = 0.031) and passage of flatus (P = 0.005). Postoperative LOS was shorter (5 vs. 7 days, P = 0.031) with laparoscopy. The overall complication rate was significantly lower in the laparoscopic group (27.7 vs. 43.6 %, P = 0.014), with an adjusted odds ratio (OR) for overall complications of 0.37 (P = 0.002). Following exclusion of bowel resections, secondary outcomes continued to favor laparoscopy, while reduction in overall complications trended toward significance, OR 0.47 (P = 0.050). Conclusion Laparoscopic surgical management of adhesive SBO was associated quicker GI recovery, shorter LOS, and reduced overall complications compared to open surgery.
Presented at the SAGES 2014 Annual Meeting, April 2-5, 2014, Salt Lake City, Utah
Keywords Small bowel obstruction Laparoscopy Adhesiolysis
J. Byrne (&) F. Saleh L. Ambrosini F. Quereshy T. D. Jackson A. Okrainec Division of General Surgery, University Health Network, 399 Bathurst Street, 8-MP 325A, Toronto, ON M5T 2S8, Canada e-mail: [email protected]
Small bowel obstruction (SBO) is a common surgical emergency, accounting for 3.5 % of all surgical hospital admissions [1]. Adhesions are the most frequent etiology, responsible for 65–80 % of SBOs [2–4]. Previous surgery is the most important predisposing factor to the development of adhesions, with incidence reported as greater than 90 % following laparotomy [5, 6]. For patients who require operative management, rates of major complications and 30-day mortality approach 20 % and 4 % respectively [7].
A. Okrainec e-mail: [email protected] F. Quereshy T. D. Jackson A. Okrainec Department of Surgery, University of Toronto, Toronto, ON, Canada
123
Surg Endosc
Following the first reported case of laparoscopic adhesiolysis by Bastug et al. in 1991 [8], widespread acceptance of laparoscopy in the management of SBO was limited in large part by concerns over iatrogenic injury and sequelae of conversion [9, 10]. As surgeon comfort with laparoscopic techniques has grown within the domain of acute care surgery, however, laparoscopic management of SBO has been shown to be safe at centers with expertise in laparoscopic surgery [3, 11–14]. Many retrospective studies have reported benefits of the laparoscopic approach, including less postoperative pain, earlier return of gastrointestinal (GI) function, reduced hospital length of stay (LOS), and fewer complications [4, 7, 15, 16]. Despite these benefits, uptake of laparoscopic surgery for SBO remains low in comparison to the open approach [17]. There is therefore need for continued reporting of outcomes in order to better define the optimal approach. The purpose of this study was to report our institutional experience comparing outcomes in patients with adhesive SBO treated laparoscopically versus those managed with open surgery.
excluded, as these were predominantly multiorgan transplant recipients who underwent open surgery exclusively. Patients classified as ASA 5 were also excluded. Outcomes of interest The primary outcome of this study was the composite endpoint of overall complications. Because some patients presented after discharge to either the emergency department (ED) or outpatient follow-up with predominantly wound complications, all in-hospital complications as well as those occurring during the 30-day follow-up period after discharge were included. Overall complications comprised wound, infectious, cardiac, respiratory, renal, and thromboembolic complications, in addition to blood transfusion, reoperation, and death. Secondary outcomes including mean operative time, surgical blood loss, readmission rates, as well as indicators of postoperative recovery, namely nasogastric (NG) tube removal, passage of flatus, and LOS are reported. Statistical analysis
Methods Data A retrospective review was performed of all patients managed surgically for adhesive SBO at the University Health Network, Toronto, Canada, between April 2005 and September 2013. Data abstracted included patient demographic and baseline characteristics, American Society of Anesthesiologists (ASA) class and comorbidities, as well as preoperative vital signs and lab work. Intraoperative details including surgical approach (laparoscopic versus open), procedure performed, reasons for conversion to open procedure, intraoperative findings, complications, operative time, and blood loss were collected from operative reports and anesthetic records. Postoperative occurrences during the index admission, as well as the 30-day post-discharge period, were collected from the Electronic Patient Record. The study protocol was approved by the University Health Network Research Ethics Board. Study subjects All patients undergoing emergency surgery for the primary diagnosis of acute SBO within the study period were eligible for inclusion in this study. Patients with SBO due to non-adhesional causes such as incarcerated herniae or malignancy were not included in this study due to strong bias toward open surgery. Patients receiving treatment with immunosuppressant medications or systemic steroids were
123
Patients were grouped as having undergone laparoscopic or open surgery. Those who underwent conversion from laparoscopic to open surgery were included within the laparoscopic group with a view to an intention-to-treat analysis. Summary statistics were used to define the study population. Univariable analyses using the Chi-squared statistic or Fisher’s exact test were performed to compare categorical variables and the t test to compare continuous variables. Multiple logistic regression was used to examine the relationship between overall complications (binary dependent variable) and surgical approach (independent variable). Modeling for complications was created using a forward stepwise approach with a P value \0.2 cut-off to identify potential confounders from among the baseline patient characteristics of age, gender, BMI, prior abdominal surgery, ASA, smoking status, hypertension, diabetes, cardiac and pulmonary disease, and the presence of systemic inflammatory response syndrome (SIRS) [18]. To examine the effect of selection bias on outcomes, a sensitivity analysis was performed excluding those patients who underwent bowel resection as these were performed exclusively by open, or converted to open, surgical approach. Preoperative characteristics and operative findings of converted and completed laparoscopic cases were compared to determine factors associated with laparoscopic success. The impact of laparoscopic conversion on patient outcomes was determined by comparing outcomes between converted and open surgical cases. Odds ratios (ORs), P values, and 95 % confidence intervals (CIs) are reported. Statistical significance was set at P \ 0.05.
Surg Endosc
Statistical analyses were performed using Stata/IC (version 12.1, Statacorp, College Station, Texas).
Results Between April 2005 and September 2013, 459 patients underwent emergency surgery for acute SBO, 269 of whom remained following application of exclusion criteria (Fig. 1). Of these, 186 patients (69.1 %) underwent open surgery, while 83 (30.9 %) underwent laparoscopy. Within the laparoscopic group, 32 (38.6 %) underwent conversion to open surgery. Reasons for conversion included dense adhesions (40.6 %), iatrogenic injury (15.6 %), ischemic bowel (15.6 %), inadequate exposure (15.6 %), and need for resection (12.5 %). Of the five cases converted to open surgery due to iatrogenic injury, 4 were due to enterotomy and one due to a trocar injury. Iatrogenic injury occurred more frequently in the open surgery group overall (39.8 vs. 25.3 %, P = 0.027). While a maximum small bowel diameter C4 cm on preoperative CT scan was significantly associated with an increased rate of iatrogenic injury (50.6 vs. 35.3 %, P = 0.021), this risk was not greater for laparoscopic approach compared to open (40.7 vs. 45.1 %, P = 0.700). Wound classification, in general, was higher in the open (clean: 52.2 %, clean-contaminated: 35.0 %, contaminated: 8.1 %, dirty: 4.8 %) compared to laparoscopic
surgery group (clean: 72.3 %, clean-contaminated: 15.7 %, contaminated: 10.8 %, dirty: 1.2 %), P = 0.003. Table 1 illustrates the preoperative characteristics of our cohort. While patients in the laparoscopic group had a slightly higher mean BMI (P = 0.032), those with previous abdominal surgery or higher ASA classification were more likely to undergo open surgery (P = 0.002 and 0.050, respectively). Both groups had similar laboratory values; however, patients in the open surgery group had slightly lower mean serum albumin than those undergoing laparoscopy (33.0 vs. 35.9 g/L, P = 0.008). Table 2 provides a detailed breakdown of postoperative complications. Patients who underwent laparoscopic surgery were significantly less likely to develop pneumonia (P = 0.027), septic shock (P = 0.042), undergo unplanned reintubation (P = 0.045), or require blood transfusion (P = 0.002) in the postoperative period. Rates of early unplanned reoperation were not significantly different between groups. Of five patients in the laparoscopic group requiring early return to the operating room, none were due to enterotomy or missed iatrogenic injury, whereas one patient from the open surgery group required reoperation due to perforation of a serosal tear. In-hospital mortality was significantly lower in the laparoscopic group with one death (1.2 %) compared to 13 (7.0 %) in the open group (P = 0.049). Patients who underwent initial laparoscopy suffered fewer overall complications compared to those
Fig. 1 Patient flow diagram. SBO, small bowel obstruction; ASA, American Society of Anesthesiologists
123
Surg Endosc Table 1 Patient demographics, surgical history, clinical and laboratory findings Total N = 269
Open N = 186 (69.1)
Mean age (years)
66.5 (17.7)
Female Mean BMI (kg/m2)
115 (61.8) **
24.3 (5.2)
Prior abdominal surgery**
160 (87.0)
Laparoscopic N = 83 (30.9)
P
65.1 (16.5)
0.545
42 (50.6)
0.085
26.4 (7.9) 59 (71.1)
ASA
0.032 0.002 0.050
1,2
24 (12.9)
17 (20.5)
3 4
99 (53.2) 63 (33.9)
49 (59.0) 17 (20.5)
Smokera
46 (24.7)
24 (28.9)
Hypertension
88 (47.3)
36 (43.4)
0.550
Diabetes mellitus
23 (12.4)
9 (10.8)
0.722
Cardiac diseaseb
48 (25.8)
16 (19.3)
0.245
Pulmonary diseasec
33 (17.7)
12 (14.5)
0.505
SIRS**
44 (23.8)
12 (14.6)
0.090
CT maximum small bowel diameter (cm)
3.8 (0.8)
3.6 (0.8)
0.065
141.3 (19.9)
143.1 (18.4)
0.501
0.470
Presenting laboratory values (Mean) Hemoglobin (g/L)** 9
White bloodcell count (910 /L) Platelet count (9109/L)** Albumin (g/L)** INR** Creatinine (lmol/L)**
**
11.0 (4.6)
11.6 (5.5)
0.348
280.1 (116.5)
275.5 (124.0)
0.771
33.0 (7.3)
35.9 (6.8)
0.008
1.1 (0.3)
1.1 (0.3)
0.583
108.4 (99.3)
95.4 (52.2)
0.260
Values are N (%) except for small bowel diameter or if specified as mean (± SD) BMI body mass index, ASA American Society of Anesthesiologists, SIRS Systemic inflammatory response syndrome, CT computed tomography, INR International normalized ratio, SD standard deviation **
Values missing from BMI (70), Prior abdominal surgery (2), SIRS (2), Hemoglobin (1), White bloodcell count (1), Platelet count (1), Albumin (61), INR (15), Creatinine (1)
a b c
Smoker defined as any documented history of smoking Cardiac disease defined as history of MI, CHF, angina, or arrhythmia Pulmonary disease defined as history of chronic obstructive pulmonary disease (COPD), asthma, or other chronic respiratory illness
who underwent open surgery (27.7 vs. 43.6 %, P = 0.014). While single band adhesions were significantly more common in the laparoscopy group (41 vs. 26 %, P = 0.0126), the presence of single bands was not itself significantly associated with a reduction in risk of overall complications (31.7 vs. 41.7 %, P = 0.121). The results of our multivariate analysis are shown in Table 3. After controlling for covariates, age, smoking status, and prior abdominal surgery were significant predictors of complication. The presence of SIRS was predictive of increased complications but this was not statistically significant. Laparoscopic surgery was associated with a greater than 60 % reduction in likelihood of complications, OR 0.37 (95 % CI 0.20–0.69, P = 0.002). Table 4 provides a comparison of secondary outcomes between patients undergoing open versus laparoscopic surgery.
123
There was no significant difference in mean operative time between laparoscopic and open groups (131.2 vs. 136.6 min, P = 0.506). Intraoperative blood loss was significantly lower in the laparoscopic group, with a mean volume of 62.5 mL compared to 172.7 mL in the open group (P \ 0.001). There was also a significant reduction in time to recovery of bowel function in the laparoscopic group as indicated by time to NG tube removal (2.3 vs. 3.2 days, P = 0.031) and passage of flatus (3.0 vs. 3.9 days, P = 0.005). Postoperative LOS was significantly lower in patients who underwent laparoscopy with patients being discharged at a median of 5 [interquartile range (IQR) 3–8] days after surgery compared to 7 (IQR 6–13) days after surgery in the open group (P = 0.031). There was no significant difference found in rates of early readmission or ED visits within 30 days of discharge between the open and laparoscopic groups.
Surg Endosc Table 2 Complications during in-hospital stay or 30-day follow-up period in open versus laparoscopic groups Total N = 269
Open N = 186 (69.1)
Superficial surgical site infection
38 (20.4)
Laparoscopic N = 83 (30.9)
P
9 (10.8)
0.056
Wound dehiscence
0 (0)
2 (2.4)
0.094
Urinary tract infection
7 (3.8)
1 (1.2)
0.254 0.716
Acute renal failure
6 (3.2)
2 (2.4)
15 (8.1)
1 (1.2)
0.027
Prolonged ileus ([5 days)
9 (4.8)
5 (6.0)
0.143
Anastomotic leak
6 (3.2)
0 (0)
0.098
Intra-abdominal abscess
8 (4.3)
0 (0)
0.055
Fistula
1 (0.5)
2 (2.4)
0.272
Septic shock
9 (4.8)
0 (0)
0.042
Unplanned reintubation
9 (4.8)
0 (0)
0.042
Pneumonia
Failure to wean from ventilation ([48 h)
14 (7.5)
1 (1.2)
0.037
Blood transfusiona
33 (17.7)
3 (3.6)
0.002
Thromboembolic event
7 (3.8)
1 (1.2)
0.254
Cardiac arrest
4 (2.2)
0 (0)
0.178
Myocardial infarction
6 (3.2)
0 (0)
0.098
Early unplanned reoperationb
8 (4.3)
5 (6.0)
0.543
In-hospital death
13 (7.0)
1 (1.2)
0.049
Overall complicationsc
81 (43.6)
23 (27.7)
0.014
Values are N (%) a
Blood transfusion defined as receipt of at least 1 unit of packed red blood cells within 30 days of surgery
b
Early reoperation defined as further surgery within 30 days of first operation
c
required resection were significantly older (mean age 73.1 vs. 62.4 years, P \ 0.001), of higher ASA classification (P = 0.028), and had lower mean albumin (31.3 vs. 35.3 g/ L, P \ 0.001). SIRS was not significantly associated with need for resection. Furthermore, wound classification was significantly higher in patients who underwent resection (clean: 0 %, clean-contaminated: 71.0 %, contaminated: 19.4 %, dirty: 9.7 %) compared to those who did not (clean: 89.2 %, clean-contaminated: 6.8 %, contaminated: 3.4 %, dirty: 0.6 %), P \ 0.001. After the sensitivity analysis with exclusion of 93 patients who underwent bowel resection, there was no significant difference in ASA classification, serum albumin, or wound classification between groups. There remained no significant difference in mean operative time between laparoscopic and open surgical approach. Intraoperative blood loss, time to flatus, and LOS remained significantly lower in the laparoscopic group (P \ 0.05). The adjusted results comparing laparoscopic to open surgery for overall complications trended strongly toward significance, OR 0.47 (95 % CI 0.22–1.00, P = 0.050). Single band adhesion, documented as the etiology of obstruction in the operative report, was significantly associated with laparoscopic success, present in 51.0 % completed compared to 22.6 % converted laparoscopic operations (P = 0.019). While bowel diameter C4 cm was significantly associated with conversion to open surgery (P = 0.007), patients who underwent conversion were not at increased risk of complications compared to the open group (50.0 vs. 43.6 %, P = 0.498).
Overall complications is a binary composite end point incorporating all other complications listed in table, representing the occurrence of at least one complication
Discussion
Table 3 Multivariate analysis: predictors of overall complications Variable
Odds ratio
95 % CI
P
Laparoscopic procedure
0.374
0.20–0.69
0.002
Age
1.041
1.02–1.06
\0.001
Female
0.555
0.32–0.97
0.040
Smoker
2.192
1.15–4.18
0.017
SIRS
1.444
0.75–2.78
0.272
Prior abdominal surgery
2.104
1.00–4.42
0.050
Two patients excluded from analysis due to missing baseline characteristics SIRS systemic inflammatory response syndrome
Bowel resection was performed in 13 patients in the laparoscopic group (15.7 %), all of whom underwent conversion to open surgery, compared to 80 patients in the open group (43.0 %), P \ 0.001. Ischemia was the dominant indication for bowel resection (66.7 %). Patients who
Our study demonstrated a significant reduction in overall complications in patients with adhesive SBO treated by laparoscopy as compared to open surgery (27.7 vs. 43.6 %, P = 0.015), with an adjusted OR 0.37 (95 % CI 0.20–0.69, P = 0.002) based on multivariate analysis. While operative time was comparable between the groups, patients undergoing laparoscopic surgery were subjected to significantly less intraoperative blood loss (62.5 vs. 172.7 mL, P = 0.001). Time to recovery of GI function was lower in patients treated by laparoscopy, as indicated by time to NG removal (2.3 vs. 3.2 days, P = 0.031) and time to flatus (3.0 vs. 3.9 days, P = 0.005), while LOS was shorter, 5 (IQR 3–8) vs. 7 (IQR 6–13) days (P = 0.031). Many studies have compared outcomes between laparoscopic and open surgical management of acute SBO [7, 16, 19–23]. A recent meta-analysis by Li et al. found laparoscopic adhesiolysis to be associated with a reduced overall complication rate (OR 0.42, P \ 0.01) [24]. Our overall complication rate was comparable, though slightly
123
Surg Endosc Table 4 Comparison of secondary outcomes: open versus laparoscopic surgery Total N = 269
Open N = 186 (69.1)
Laparoscopic N = 83 (30.9)
P
Mean operative time (min)
136.6 (64.0)
131.2 (54.6)
0.506
Mean intraoperative blood loss (mL)
\0.001
172.7 (274.4)
62.5 (97.7)
Mean time to NG removal (days)
3.2 (2.9)
2.3 (3.1)
0.031
Mean time to flatus (days)
3.9 (2.4)
3.0 (2.4)
0.005
Median length of stay (days)
7 (6–13)
5 (3–8)
0.031
Early readmissiona
9 (4.8)
8 (9.6)
0.135
20 (10.8)
8 (9.6)
0.782
b
Emergency department visit
Values are N (%) unless otherwise specified as mean (± SD) or median (IQR) NG nasogastric tube, IQR interquartile range, SD standard deviation a
Early readmission defined as readmission to a surgical service within 30 days of discharge
b
Emergency department visit defined as presentation to an emergency department within 30 days of discharge
higher, than that found by Wullstein et al. comparing laparoscopic to open adhesiolysis (19.2 vs. 40.4 %, P = 0.032) [19]. Kelly et al. further demonstrated a lower rate of major complications (10 vs. 22 %, P \ 0.001) [7]. In our own cohort, we found a significantly lower rate of pneumonia, septic shock, unplanned reintubation, need for blood transfusion, and in-hospital mortality in patients treated with laparoscopic surgery. Incisional and respiratory complications are among the most frequent following surgery for SBO [25], and other authors have demonstrated significant reduction in rates of wound infection and pneumonia in patients undergoing laparoscopic compared to open adhesiolysis [7]. Our own results did not show a significant difference in wound infection between laparoscopic and open groups (10.8 vs. 20.4 %, P = 0.056). This appears to be a reflection of our inclusion of converted cases in the laparoscopic group, as 7 of 9 wound infections in the laparoscopic group occurred following conversion to open surgery. Reduced time to recovery of GI function and shorter LOS in patients undergoing laparoscopic management of adhesive SBO have been reported by many authors in retrospective series [19–23]. Khaikin et al. observed a significant reduction in time to first bowel movement in patients treated with laparoscopy (3 vs. 6 days, P \ 0.05) [21], while Wullstein et al. showed earlier resumption of regular diet (5.1 vs. 6.4 days, P = 0.004) [19]. We were able to demonstrate the same benefit through accurate ascertainment from the clinical notes of time to NG tube removal and passage of flatus. Our results echo those of Kelly et al., who showed a significant reduction in mean LOS in patients undergoing laparoscopic adhesiolysis (4.7 vs. 9.9 days, P \ 0.001) compared to open surgery [7], while Mancini et al. also observed a 27 % shorter LOS in their laparoscopic group (P \ 0.01) [16].
123
To address the concern that specific outcomes and complication rates favoring laparoscopic surgery might be due to selection bias, resulting in more severe pathology in the open group, we performed a sensitivity analysis excluding patients who underwent bowel resection. Similar to our findings, Margenthaler et al. demonstrated that overall morbidity is higher in patients undergoing small bowel resection compared to adhesiolysis alone (47 vs. 37 %, P \ 0.001) [25]. After excluding bowel resections, ASA class, albumin, and wound classification were no longer significantly different between cohorts, suggesting that severity of disease in those who require resection accounts for many of the important differences between open and laparoscopic groups. After excluding these patients, operative time remained comparable between groups, with intraoperative blood loss and time to flatus significantly lower in those undergoing laparoscopic surgery. There remained a reduction in the odds of overall complications, OR 0.47 (95 % CI 0.22–1.00; P = 0.050). While this trend was no longer significant, one-third of the patients were excluded in this analysis with a strong reduction in power to detect a difference. Previous criticisms of laparoscopic management of acute SBO have cited longer operative time, iatrogenic injury, and concerns over sequelae of conversion [9, 10, 19, 21, 22]. We found no significant difference between operative time in the laparoscopic compared to open surgery groups (131.2 vs. 136.6 min, P = 0.506). Other studies have also demonstrated comparable [12, 26, 27], or reduced [7, 23] mean operative time with laparoscopy compared to open surgery [25]. Furthermore, iatrogenic injury was significantly more common in the open surgery group, with the only early reoperation attributed to a missed injury occurring following open surgery, suggesting laparoscopic approach may not be associated with increased likelihood of iatrogenic harm.
Surg Endosc
In our cohort of 83 patients managed by initial laparoscopic approach, 32 underwent conversion to open surgery—a conversion rate of 38.6 %. In the literature, frequency of conversion is variable ranging from 0 to 52 % [9]. Two recent case reviews including over 1,000 [28] and 2,000 [29] patients managed with laparoscopy found conversion rates of 33.5 and 29 % respectively. Extent of intra-abdominal adhesions has been emphasized as an important factor affecting successful completion of laparoscopic surgery [26, 30, 31]. In this study, the only factor found to be significantly associated with successful laparoscopic completion was the finding at the time of surgery of a single band adhesion as the etiology of obstruction. As this is an intraoperative finding it cannot be used for clinical decision-making with regard to surgical approach. Conversely, small bowel diameter C4 cm on preoperative CT was significantly associated with conversion to open surgery. Suter et al. has previously reported increased bowel diameter as a predictor of conversion [31]. However, there is no evidence to suggest that attempting laparoscopy in this setting leads to adverse outcome. While we demonstrated that small bowel diameter C4 cm was significantly associated with increased rate of iatrogenic injury overall, this risk was not dependent on surgical approach. Furthermore, when we examined the impact of conversion to open surgery on patient outcomes, we found no significant difference in overall complications following converted compared to open surgery. Therefore, based on our results, while conversion to open surgery may become necessary, it is not associated with worse outcomes. There are important limitations to this study that must be considered. The dominant concern with such retrospective cohort studies is the presence of selection bias toward open surgery in ‘sicker’ patients, thereby selecting fitter candidates for laparoscopy. While multivariable adjustment addresses known potential confounders, it cannot address unknown or unobserved variables such as differences in perioperative care. However, the results of our sensitivity analysis excluding those patients who underwent bowel resection are reassuring. In addition, due to the retrospective nature of this study, deficiencies in data reporting in patient records, and the inability to track patients upon discharge who may have presented to other hospitals were unable to be accounted for. However, we feel the data presented are overall truly representative of our experience with laparoscopy and open surgery for adhesive SBO.
Conclusion Treatment of patients with adhesive SBO by the laparoscopic approach is associated with quicker recovery of GI function, shorter LOS, and a reduction in overall
complications as compared to conventional open surgical management. This must be interpreted within the limitations of a retrospective study. Future prospective studies are needed to more accurately account for these factors, and to better inform patient selection for laparoscopic approach. Disclosure Drs. James Byrne, Fady Saleh, Luciano Ambrosini, Fayez Quereshy, and Timothy Jackson have no conflicts of interest or financial ties to disclose. Dr. Allen Okrainec is a consultant for Covidien and receives honoraria for speaking and teaching. Funding
No external sources of funding were used for this study.
References 1. Sikirica V et al (2011) The inpatient burden of abdominal and gynecological adhesiolysis in the US. BMC Surg 11:13. doi:10. 1186/1471-2482-11-13 2. Tierris I et al (2011) Laparoscopy for acute small bowel obstruction: indication or contraindication? Surg Endosc 25(2):531–535. doi:10.1007/s00464-010-1206-8 Epub 2010 Jul 7 3. Vettoretto N et al (2012) Laparoscopic adhesiolysis: consensus conference guidelines. Colorectal Dis 14(5):e208–e215. doi:10. 1111/j.1463-1318.2012.02968.x 4. Catena F et al (2011) Bologna guidelines for diagnosis and management of adhesive small bowel obstruction (ASBO): 2010 evidence-based guidelines of the World Society of Emergency Surgery. World J Emerg Surg 6(5):21. doi:10.1186/1749-7922-65 5. Menzies D (1992) Peritoneal adhesions. Incidence, cause, and prevention. Surg Annu 24(Pt 1):27–45 6. Luijendijk RW et al (1996) Foreign material in postoperative adhesions. Ann Surg 223(3):242–248 7. Kelly KN et al (2014) Laparotomy for small-bowel obstruction: first choice or last resort for adhesiolysis? A laparoscopic approach for small-bowel obstruction reduces 30-day complications. Surg Endosc 28(1):65–73. doi:10.1007/s00464-013-3162-6 Epub 2013 Sep 4 8. Bastug DF et al (1991) Laparoscopic adhesiolysis for small bowel obstruction. Surg Laparosc Endosc 1(4):259–262 9. Farinella E et al (2009) Feasibility of laparoscopy for small bowel obstruction. World J Emerg Surg. 4:3. doi:10.1186/17497922-4-3 10. Sauerland S et al (2006) Laparoscopy for abdominal emergencies: evidence-based guidelines of the European Association for Endoscopic Surgery. Surg Endosc 20(1):14–29 Epub 2005 Oct 24 11. Leon EL et al (1998) Laparoscopic management of small bowel obstruction: indications and outcome. J Gastrointest Surg 2(2):132–140 12. Strickland P et al (1999) Is laparoscopy safe and effective for treatment of acute small-bowel obstruction? Surg Endosc 13(7):695–698 13. Nagle A et al (2004) Laparoscopic adhesiolysis for small bowel obstruction. Am J Surg 187(4):464–470 14. Zerey M et al (2007) Laparoscopic management of adhesive small bowel obstruction. Am Surg 73(8):773–778 discussion 778–9 15. Gutt CN et al (2004) Fewer adhesions induced by laparoscopic surgery? Surg Endosc 18(6):898–906 Epub 2004 Apr 27 16. Mancini GJ et al (2008) Nationwide impact of laparoscopic lysis of adhesions in the management of intestinal obstruction in the
123
Surg Endosc
17.
18.
19.
20.
21.
22.
23. 24.
US. J Am Coll Surg. 207(4):520–526. doi:10.1016/j.jamcollsurg. 2008.04.026 Epub 2008 Jun 24 Saleh F et al (2014) Laparoscopic versus open surgical management of small bowel obstruction: an analysis of short-term outcomes. Surg Endosc 28(8):2381–2386. doi:10.1007/s00464014-3486-x Epub 2014 Mar 21 Bone RC et al (1992) Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. The ACCP/SCCM Consensus Conference Committee. American College of Chest Physicians/Society of Critical Care Medicine. Chest 101(6):1644–1655 Wullstein C, Gross E (2003) Laparoscopic compared with conventional treatment of acute adhesive small bowel obstruction. Br J Surg 90(9):1147–1151 Johnson KN et al (2012) Laparoscopic management of acute small bowel obstruction: evaluating the need for resection. J Trauma Acute Care Surg 72(1):25–30. doi:10.1097/TA. 0b013e31823d8365 discussion 30-1; quiz 317 Khaikin M et al (2007) Laparoscopic vs. open surgery for acute adhesive small-bowel obstruction: patients’ outcome and costeffectiveness. Surg Endosc 21(5):742–746 Epub 2007 Mar 1 Okamoto H et al (2012) Clinical outcomes of laparoscopic adhesiolysis for mechanical small bowel obstruction. Asian J Endosc Surg 5(2):53–58. doi:10.1111/j.1758-5910.2011.00117.x Epub 2011 Nov 24 Chopra R et al (2003) Laparoscopic lysis of adhesions. Am Surg 69(11):966–968 Li MZ et al (2012) Laparoscopic versus open adhesiolysis in patients with adhesive small bowel obstruction: a systematic
123
25.
26.
27. 28.
29.
30.
31.
review and meta-analysis. Am J Surg. 204(5):779–786. doi:10. 1016/j.amjsurg.2012.03.005 Epub 2012 Jul 12 Margenthaler JA et al (2006) Risk factors for adverse outcomes following surgery for small bowel obstruction. Ann Surg 243(4):456–464 Grafen FC et al (2010) Management of acute small bowel obstruction from intestinal adhesions: indications for laparoscopic surgery in a community teaching hospital. Langenbecks Arch Surg 395(1):57–63. doi:10.1007/s00423-009-0490-z Epub 2009 Mar 28 Kirshtein B et al (2005) Laparoscopic management of acute small bowel obstruction. Surg Endosc 19(4):464–467 Epub 2005 Feb 3 Ghosheh B, Salameh JR (2007) Laparoscopic approach to acute small bowel obstruction: review of 1061 cases. Surg Endosc 21(11):1945–1949 Epub 2007 Sep 19 O’Connor DB, Winter DC (2012) The role of laparoscopy in the management of acute small-bowel obstruction: a review of over 2,000 cases. Surg Endosc 26(1):12–17. doi:10.1007/s00464-0111885-9 Epub 2011 Sep 5 Levard H et al (2001) Laparoscopic treatment of acute small bowel obstruction: a multicentre retrospective study. ANZ J Surg 71(11):641–646 Suter M et al (2000) Laparoscopic management of mechanical small bowel obstruction: are there predictors of success or failure? Surg Endosc 14(5):478–483
and Other Interventional Techniques
Laparoscopic versus open surgical management of adhesive small bowel obstruction: a comparison of outcomes James Byrne • Fady Saleh • Luciano Ambrosini Fayez Quereshy • Timothy D. Jackson • Allan Okrainec
•
Received: 7 February 2014 / Accepted: 9 November 2014 Ó Springer Science+Business Media New York 2014
Abstract Background Laparoscopic management of adhesive small bowel obstruction (SBO) has become an established technique within the domain of acute care surgery. As minimally invasive management of SBO becomes more widely accepted, there is increased need for reporting of outcomes. Objective To compare outcomes of laparoscopic versus open surgery for adhesive SBO. Methods Patients undergoing surgery for adhesive SBO at our institution between 2005 and 2013 were eligible for inclusion. The primary outcome was overall complication rate, while secondary outcomes included operative time, gastrointestinal (GI) function, and postoperative length of stay (LOS). Univariable analysis compared laparoscopic (including conversions) and open groups with regard to patient baseline and perioperative characteristics as well as outcomes of interest. Multivariable analysis was performed comparing the endpoint of overall complications between groups. Sensitivity analysis excluding patients who underwent bowel resection was performed to assess effect
on outcomes. Factors associated with laparoscopic success, as well as impact of conversion to open on postoperative outcomes, are reported. Results A cohort of 269 patients with adhesive SBO was identified: 186 patients (69.1 %) underwent open surgery, 83 (30.9 %) were managed laparoscopically. Within the laparoscopy group, 32 (38.6 %) underwent conversion to open. Operative time was similar between groups (P = 0.506), while laparoscopy was associated with quicker recovery of GI function indicated by removal of nasogastric tube (P = 0.031) and passage of flatus (P = 0.005). Postoperative LOS was shorter (5 vs. 7 days, P = 0.031) with laparoscopy. The overall complication rate was significantly lower in the laparoscopic group (27.7 vs. 43.6 %, P = 0.014), with an adjusted odds ratio (OR) for overall complications of 0.37 (P = 0.002). Following exclusion of bowel resections, secondary outcomes continued to favor laparoscopy, while reduction in overall complications trended toward significance, OR 0.47 (P = 0.050). Conclusion Laparoscopic surgical management of adhesive SBO was associated quicker GI recovery, shorter LOS, and reduced overall complications compared to open surgery.
Presented at the SAGES 2014 Annual Meeting, April 2-5, 2014, Salt Lake City, Utah
Keywords Small bowel obstruction Laparoscopy Adhesiolysis
J. Byrne (&) F. Saleh L. Ambrosini F. Quereshy T. D. Jackson A. Okrainec Division of General Surgery, University Health Network, 399 Bathurst Street, 8-MP 325A, Toronto, ON M5T 2S8, Canada e-mail: [email protected]
Small bowel obstruction (SBO) is a common surgical emergency, accounting for 3.5 % of all surgical hospital admissions [1]. Adhesions are the most frequent etiology, responsible for 65–80 % of SBOs [2–4]. Previous surgery is the most important predisposing factor to the development of adhesions, with incidence reported as greater than 90 % following laparotomy [5, 6]. For patients who require operative management, rates of major complications and 30-day mortality approach 20 % and 4 % respectively [7].
A. Okrainec e-mail: [email protected] F. Quereshy T. D. Jackson A. Okrainec Department of Surgery, University of Toronto, Toronto, ON, Canada
123
Surg Endosc
Following the first reported case of laparoscopic adhesiolysis by Bastug et al. in 1991 [8], widespread acceptance of laparoscopy in the management of SBO was limited in large part by concerns over iatrogenic injury and sequelae of conversion [9, 10]. As surgeon comfort with laparoscopic techniques has grown within the domain of acute care surgery, however, laparoscopic management of SBO has been shown to be safe at centers with expertise in laparoscopic surgery [3, 11–14]. Many retrospective studies have reported benefits of the laparoscopic approach, including less postoperative pain, earlier return of gastrointestinal (GI) function, reduced hospital length of stay (LOS), and fewer complications [4, 7, 15, 16]. Despite these benefits, uptake of laparoscopic surgery for SBO remains low in comparison to the open approach [17]. There is therefore need for continued reporting of outcomes in order to better define the optimal approach. The purpose of this study was to report our institutional experience comparing outcomes in patients with adhesive SBO treated laparoscopically versus those managed with open surgery.
excluded, as these were predominantly multiorgan transplant recipients who underwent open surgery exclusively. Patients classified as ASA 5 were also excluded. Outcomes of interest The primary outcome of this study was the composite endpoint of overall complications. Because some patients presented after discharge to either the emergency department (ED) or outpatient follow-up with predominantly wound complications, all in-hospital complications as well as those occurring during the 30-day follow-up period after discharge were included. Overall complications comprised wound, infectious, cardiac, respiratory, renal, and thromboembolic complications, in addition to blood transfusion, reoperation, and death. Secondary outcomes including mean operative time, surgical blood loss, readmission rates, as well as indicators of postoperative recovery, namely nasogastric (NG) tube removal, passage of flatus, and LOS are reported. Statistical analysis
Methods Data A retrospective review was performed of all patients managed surgically for adhesive SBO at the University Health Network, Toronto, Canada, between April 2005 and September 2013. Data abstracted included patient demographic and baseline characteristics, American Society of Anesthesiologists (ASA) class and comorbidities, as well as preoperative vital signs and lab work. Intraoperative details including surgical approach (laparoscopic versus open), procedure performed, reasons for conversion to open procedure, intraoperative findings, complications, operative time, and blood loss were collected from operative reports and anesthetic records. Postoperative occurrences during the index admission, as well as the 30-day post-discharge period, were collected from the Electronic Patient Record. The study protocol was approved by the University Health Network Research Ethics Board. Study subjects All patients undergoing emergency surgery for the primary diagnosis of acute SBO within the study period were eligible for inclusion in this study. Patients with SBO due to non-adhesional causes such as incarcerated herniae or malignancy were not included in this study due to strong bias toward open surgery. Patients receiving treatment with immunosuppressant medications or systemic steroids were
123
Patients were grouped as having undergone laparoscopic or open surgery. Those who underwent conversion from laparoscopic to open surgery were included within the laparoscopic group with a view to an intention-to-treat analysis. Summary statistics were used to define the study population. Univariable analyses using the Chi-squared statistic or Fisher’s exact test were performed to compare categorical variables and the t test to compare continuous variables. Multiple logistic regression was used to examine the relationship between overall complications (binary dependent variable) and surgical approach (independent variable). Modeling for complications was created using a forward stepwise approach with a P value \0.2 cut-off to identify potential confounders from among the baseline patient characteristics of age, gender, BMI, prior abdominal surgery, ASA, smoking status, hypertension, diabetes, cardiac and pulmonary disease, and the presence of systemic inflammatory response syndrome (SIRS) [18]. To examine the effect of selection bias on outcomes, a sensitivity analysis was performed excluding those patients who underwent bowel resection as these were performed exclusively by open, or converted to open, surgical approach. Preoperative characteristics and operative findings of converted and completed laparoscopic cases were compared to determine factors associated with laparoscopic success. The impact of laparoscopic conversion on patient outcomes was determined by comparing outcomes between converted and open surgical cases. Odds ratios (ORs), P values, and 95 % confidence intervals (CIs) are reported. Statistical significance was set at P \ 0.05.
Surg Endosc
Statistical analyses were performed using Stata/IC (version 12.1, Statacorp, College Station, Texas).
Results Between April 2005 and September 2013, 459 patients underwent emergency surgery for acute SBO, 269 of whom remained following application of exclusion criteria (Fig. 1). Of these, 186 patients (69.1 %) underwent open surgery, while 83 (30.9 %) underwent laparoscopy. Within the laparoscopic group, 32 (38.6 %) underwent conversion to open surgery. Reasons for conversion included dense adhesions (40.6 %), iatrogenic injury (15.6 %), ischemic bowel (15.6 %), inadequate exposure (15.6 %), and need for resection (12.5 %). Of the five cases converted to open surgery due to iatrogenic injury, 4 were due to enterotomy and one due to a trocar injury. Iatrogenic injury occurred more frequently in the open surgery group overall (39.8 vs. 25.3 %, P = 0.027). While a maximum small bowel diameter C4 cm on preoperative CT scan was significantly associated with an increased rate of iatrogenic injury (50.6 vs. 35.3 %, P = 0.021), this risk was not greater for laparoscopic approach compared to open (40.7 vs. 45.1 %, P = 0.700). Wound classification, in general, was higher in the open (clean: 52.2 %, clean-contaminated: 35.0 %, contaminated: 8.1 %, dirty: 4.8 %) compared to laparoscopic
surgery group (clean: 72.3 %, clean-contaminated: 15.7 %, contaminated: 10.8 %, dirty: 1.2 %), P = 0.003. Table 1 illustrates the preoperative characteristics of our cohort. While patients in the laparoscopic group had a slightly higher mean BMI (P = 0.032), those with previous abdominal surgery or higher ASA classification were more likely to undergo open surgery (P = 0.002 and 0.050, respectively). Both groups had similar laboratory values; however, patients in the open surgery group had slightly lower mean serum albumin than those undergoing laparoscopy (33.0 vs. 35.9 g/L, P = 0.008). Table 2 provides a detailed breakdown of postoperative complications. Patients who underwent laparoscopic surgery were significantly less likely to develop pneumonia (P = 0.027), septic shock (P = 0.042), undergo unplanned reintubation (P = 0.045), or require blood transfusion (P = 0.002) in the postoperative period. Rates of early unplanned reoperation were not significantly different between groups. Of five patients in the laparoscopic group requiring early return to the operating room, none were due to enterotomy or missed iatrogenic injury, whereas one patient from the open surgery group required reoperation due to perforation of a serosal tear. In-hospital mortality was significantly lower in the laparoscopic group with one death (1.2 %) compared to 13 (7.0 %) in the open group (P = 0.049). Patients who underwent initial laparoscopy suffered fewer overall complications compared to those
Fig. 1 Patient flow diagram. SBO, small bowel obstruction; ASA, American Society of Anesthesiologists
123
Surg Endosc Table 1 Patient demographics, surgical history, clinical and laboratory findings Total N = 269
Open N = 186 (69.1)
Mean age (years)
66.5 (17.7)
Female Mean BMI (kg/m2)
115 (61.8) **
24.3 (5.2)
Prior abdominal surgery**
160 (87.0)
Laparoscopic N = 83 (30.9)
P
65.1 (16.5)
0.545
42 (50.6)
0.085
26.4 (7.9) 59 (71.1)
ASA
0.032 0.002 0.050
1,2
24 (12.9)
17 (20.5)
3 4
99 (53.2) 63 (33.9)
49 (59.0) 17 (20.5)
Smokera
46 (24.7)
24 (28.9)
Hypertension
88 (47.3)
36 (43.4)
0.550
Diabetes mellitus
23 (12.4)
9 (10.8)
0.722
Cardiac diseaseb
48 (25.8)
16 (19.3)
0.245
Pulmonary diseasec
33 (17.7)
12 (14.5)
0.505
SIRS**
44 (23.8)
12 (14.6)
0.090
CT maximum small bowel diameter (cm)
3.8 (0.8)
3.6 (0.8)
0.065
141.3 (19.9)
143.1 (18.4)
0.501
0.470
Presenting laboratory values (Mean) Hemoglobin (g/L)** 9
White bloodcell count (910 /L) Platelet count (9109/L)** Albumin (g/L)** INR** Creatinine (lmol/L)**
**
11.0 (4.6)
11.6 (5.5)
0.348
280.1 (116.5)
275.5 (124.0)
0.771
33.0 (7.3)
35.9 (6.8)
0.008
1.1 (0.3)
1.1 (0.3)
0.583
108.4 (99.3)
95.4 (52.2)
0.260
Values are N (%) except for small bowel diameter or if specified as mean (± SD) BMI body mass index, ASA American Society of Anesthesiologists, SIRS Systemic inflammatory response syndrome, CT computed tomography, INR International normalized ratio, SD standard deviation **
Values missing from BMI (70), Prior abdominal surgery (2), SIRS (2), Hemoglobin (1), White bloodcell count (1), Platelet count (1), Albumin (61), INR (15), Creatinine (1)
a b c
Smoker defined as any documented history of smoking Cardiac disease defined as history of MI, CHF, angina, or arrhythmia Pulmonary disease defined as history of chronic obstructive pulmonary disease (COPD), asthma, or other chronic respiratory illness
who underwent open surgery (27.7 vs. 43.6 %, P = 0.014). While single band adhesions were significantly more common in the laparoscopy group (41 vs. 26 %, P = 0.0126), the presence of single bands was not itself significantly associated with a reduction in risk of overall complications (31.7 vs. 41.7 %, P = 0.121). The results of our multivariate analysis are shown in Table 3. After controlling for covariates, age, smoking status, and prior abdominal surgery were significant predictors of complication. The presence of SIRS was predictive of increased complications but this was not statistically significant. Laparoscopic surgery was associated with a greater than 60 % reduction in likelihood of complications, OR 0.37 (95 % CI 0.20–0.69, P = 0.002). Table 4 provides a comparison of secondary outcomes between patients undergoing open versus laparoscopic surgery.
123
There was no significant difference in mean operative time between laparoscopic and open groups (131.2 vs. 136.6 min, P = 0.506). Intraoperative blood loss was significantly lower in the laparoscopic group, with a mean volume of 62.5 mL compared to 172.7 mL in the open group (P \ 0.001). There was also a significant reduction in time to recovery of bowel function in the laparoscopic group as indicated by time to NG tube removal (2.3 vs. 3.2 days, P = 0.031) and passage of flatus (3.0 vs. 3.9 days, P = 0.005). Postoperative LOS was significantly lower in patients who underwent laparoscopy with patients being discharged at a median of 5 [interquartile range (IQR) 3–8] days after surgery compared to 7 (IQR 6–13) days after surgery in the open group (P = 0.031). There was no significant difference found in rates of early readmission or ED visits within 30 days of discharge between the open and laparoscopic groups.
Surg Endosc Table 2 Complications during in-hospital stay or 30-day follow-up period in open versus laparoscopic groups Total N = 269
Open N = 186 (69.1)
Superficial surgical site infection
38 (20.4)
Laparoscopic N = 83 (30.9)
P
9 (10.8)
0.056
Wound dehiscence
0 (0)
2 (2.4)
0.094
Urinary tract infection
7 (3.8)
1 (1.2)
0.254 0.716
Acute renal failure
6 (3.2)
2 (2.4)
15 (8.1)
1 (1.2)
0.027
Prolonged ileus ([5 days)
9 (4.8)
5 (6.0)
0.143
Anastomotic leak
6 (3.2)
0 (0)
0.098
Intra-abdominal abscess
8 (4.3)
0 (0)
0.055
Fistula
1 (0.5)
2 (2.4)
0.272
Septic shock
9 (4.8)
0 (0)
0.042
Unplanned reintubation
9 (4.8)
0 (0)
0.042
Pneumonia
Failure to wean from ventilation ([48 h)
14 (7.5)
1 (1.2)
0.037
Blood transfusiona
33 (17.7)
3 (3.6)
0.002
Thromboembolic event
7 (3.8)
1 (1.2)
0.254
Cardiac arrest
4 (2.2)
0 (0)
0.178
Myocardial infarction
6 (3.2)
0 (0)
0.098
Early unplanned reoperationb
8 (4.3)
5 (6.0)
0.543
In-hospital death
13 (7.0)
1 (1.2)
0.049
Overall complicationsc
81 (43.6)
23 (27.7)
0.014
Values are N (%) a
Blood transfusion defined as receipt of at least 1 unit of packed red blood cells within 30 days of surgery
b
Early reoperation defined as further surgery within 30 days of first operation
c
required resection were significantly older (mean age 73.1 vs. 62.4 years, P \ 0.001), of higher ASA classification (P = 0.028), and had lower mean albumin (31.3 vs. 35.3 g/ L, P \ 0.001). SIRS was not significantly associated with need for resection. Furthermore, wound classification was significantly higher in patients who underwent resection (clean: 0 %, clean-contaminated: 71.0 %, contaminated: 19.4 %, dirty: 9.7 %) compared to those who did not (clean: 89.2 %, clean-contaminated: 6.8 %, contaminated: 3.4 %, dirty: 0.6 %), P \ 0.001. After the sensitivity analysis with exclusion of 93 patients who underwent bowel resection, there was no significant difference in ASA classification, serum albumin, or wound classification between groups. There remained no significant difference in mean operative time between laparoscopic and open surgical approach. Intraoperative blood loss, time to flatus, and LOS remained significantly lower in the laparoscopic group (P \ 0.05). The adjusted results comparing laparoscopic to open surgery for overall complications trended strongly toward significance, OR 0.47 (95 % CI 0.22–1.00, P = 0.050). Single band adhesion, documented as the etiology of obstruction in the operative report, was significantly associated with laparoscopic success, present in 51.0 % completed compared to 22.6 % converted laparoscopic operations (P = 0.019). While bowel diameter C4 cm was significantly associated with conversion to open surgery (P = 0.007), patients who underwent conversion were not at increased risk of complications compared to the open group (50.0 vs. 43.6 %, P = 0.498).
Overall complications is a binary composite end point incorporating all other complications listed in table, representing the occurrence of at least one complication
Discussion
Table 3 Multivariate analysis: predictors of overall complications Variable
Odds ratio
95 % CI
P
Laparoscopic procedure
0.374
0.20–0.69
0.002
Age
1.041
1.02–1.06
\0.001
Female
0.555
0.32–0.97
0.040
Smoker
2.192
1.15–4.18
0.017
SIRS
1.444
0.75–2.78
0.272
Prior abdominal surgery
2.104
1.00–4.42
0.050
Two patients excluded from analysis due to missing baseline characteristics SIRS systemic inflammatory response syndrome
Bowel resection was performed in 13 patients in the laparoscopic group (15.7 %), all of whom underwent conversion to open surgery, compared to 80 patients in the open group (43.0 %), P \ 0.001. Ischemia was the dominant indication for bowel resection (66.7 %). Patients who
Our study demonstrated a significant reduction in overall complications in patients with adhesive SBO treated by laparoscopy as compared to open surgery (27.7 vs. 43.6 %, P = 0.015), with an adjusted OR 0.37 (95 % CI 0.20–0.69, P = 0.002) based on multivariate analysis. While operative time was comparable between the groups, patients undergoing laparoscopic surgery were subjected to significantly less intraoperative blood loss (62.5 vs. 172.7 mL, P = 0.001). Time to recovery of GI function was lower in patients treated by laparoscopy, as indicated by time to NG removal (2.3 vs. 3.2 days, P = 0.031) and time to flatus (3.0 vs. 3.9 days, P = 0.005), while LOS was shorter, 5 (IQR 3–8) vs. 7 (IQR 6–13) days (P = 0.031). Many studies have compared outcomes between laparoscopic and open surgical management of acute SBO [7, 16, 19–23]. A recent meta-analysis by Li et al. found laparoscopic adhesiolysis to be associated with a reduced overall complication rate (OR 0.42, P \ 0.01) [24]. Our overall complication rate was comparable, though slightly
123
Surg Endosc Table 4 Comparison of secondary outcomes: open versus laparoscopic surgery Total N = 269
Open N = 186 (69.1)
Laparoscopic N = 83 (30.9)
P
Mean operative time (min)
136.6 (64.0)
131.2 (54.6)
0.506
Mean intraoperative blood loss (mL)
\0.001
172.7 (274.4)
62.5 (97.7)
Mean time to NG removal (days)
3.2 (2.9)
2.3 (3.1)
0.031
Mean time to flatus (days)
3.9 (2.4)
3.0 (2.4)
0.005
Median length of stay (days)
7 (6–13)
5 (3–8)
0.031
Early readmissiona
9 (4.8)
8 (9.6)
0.135
20 (10.8)
8 (9.6)
0.782
b
Emergency department visit
Values are N (%) unless otherwise specified as mean (± SD) or median (IQR) NG nasogastric tube, IQR interquartile range, SD standard deviation a
Early readmission defined as readmission to a surgical service within 30 days of discharge
b
Emergency department visit defined as presentation to an emergency department within 30 days of discharge
higher, than that found by Wullstein et al. comparing laparoscopic to open adhesiolysis (19.2 vs. 40.4 %, P = 0.032) [19]. Kelly et al. further demonstrated a lower rate of major complications (10 vs. 22 %, P \ 0.001) [7]. In our own cohort, we found a significantly lower rate of pneumonia, septic shock, unplanned reintubation, need for blood transfusion, and in-hospital mortality in patients treated with laparoscopic surgery. Incisional and respiratory complications are among the most frequent following surgery for SBO [25], and other authors have demonstrated significant reduction in rates of wound infection and pneumonia in patients undergoing laparoscopic compared to open adhesiolysis [7]. Our own results did not show a significant difference in wound infection between laparoscopic and open groups (10.8 vs. 20.4 %, P = 0.056). This appears to be a reflection of our inclusion of converted cases in the laparoscopic group, as 7 of 9 wound infections in the laparoscopic group occurred following conversion to open surgery. Reduced time to recovery of GI function and shorter LOS in patients undergoing laparoscopic management of adhesive SBO have been reported by many authors in retrospective series [19–23]. Khaikin et al. observed a significant reduction in time to first bowel movement in patients treated with laparoscopy (3 vs. 6 days, P \ 0.05) [21], while Wullstein et al. showed earlier resumption of regular diet (5.1 vs. 6.4 days, P = 0.004) [19]. We were able to demonstrate the same benefit through accurate ascertainment from the clinical notes of time to NG tube removal and passage of flatus. Our results echo those of Kelly et al., who showed a significant reduction in mean LOS in patients undergoing laparoscopic adhesiolysis (4.7 vs. 9.9 days, P \ 0.001) compared to open surgery [7], while Mancini et al. also observed a 27 % shorter LOS in their laparoscopic group (P \ 0.01) [16].
123
To address the concern that specific outcomes and complication rates favoring laparoscopic surgery might be due to selection bias, resulting in more severe pathology in the open group, we performed a sensitivity analysis excluding patients who underwent bowel resection. Similar to our findings, Margenthaler et al. demonstrated that overall morbidity is higher in patients undergoing small bowel resection compared to adhesiolysis alone (47 vs. 37 %, P \ 0.001) [25]. After excluding bowel resections, ASA class, albumin, and wound classification were no longer significantly different between cohorts, suggesting that severity of disease in those who require resection accounts for many of the important differences between open and laparoscopic groups. After excluding these patients, operative time remained comparable between groups, with intraoperative blood loss and time to flatus significantly lower in those undergoing laparoscopic surgery. There remained a reduction in the odds of overall complications, OR 0.47 (95 % CI 0.22–1.00; P = 0.050). While this trend was no longer significant, one-third of the patients were excluded in this analysis with a strong reduction in power to detect a difference. Previous criticisms of laparoscopic management of acute SBO have cited longer operative time, iatrogenic injury, and concerns over sequelae of conversion [9, 10, 19, 21, 22]. We found no significant difference between operative time in the laparoscopic compared to open surgery groups (131.2 vs. 136.6 min, P = 0.506). Other studies have also demonstrated comparable [12, 26, 27], or reduced [7, 23] mean operative time with laparoscopy compared to open surgery [25]. Furthermore, iatrogenic injury was significantly more common in the open surgery group, with the only early reoperation attributed to a missed injury occurring following open surgery, suggesting laparoscopic approach may not be associated with increased likelihood of iatrogenic harm.
Surg Endosc
In our cohort of 83 patients managed by initial laparoscopic approach, 32 underwent conversion to open surgery—a conversion rate of 38.6 %. In the literature, frequency of conversion is variable ranging from 0 to 52 % [9]. Two recent case reviews including over 1,000 [28] and 2,000 [29] patients managed with laparoscopy found conversion rates of 33.5 and 29 % respectively. Extent of intra-abdominal adhesions has been emphasized as an important factor affecting successful completion of laparoscopic surgery [26, 30, 31]. In this study, the only factor found to be significantly associated with successful laparoscopic completion was the finding at the time of surgery of a single band adhesion as the etiology of obstruction. As this is an intraoperative finding it cannot be used for clinical decision-making with regard to surgical approach. Conversely, small bowel diameter C4 cm on preoperative CT was significantly associated with conversion to open surgery. Suter et al. has previously reported increased bowel diameter as a predictor of conversion [31]. However, there is no evidence to suggest that attempting laparoscopy in this setting leads to adverse outcome. While we demonstrated that small bowel diameter C4 cm was significantly associated with increased rate of iatrogenic injury overall, this risk was not dependent on surgical approach. Furthermore, when we examined the impact of conversion to open surgery on patient outcomes, we found no significant difference in overall complications following converted compared to open surgery. Therefore, based on our results, while conversion to open surgery may become necessary, it is not associated with worse outcomes. There are important limitations to this study that must be considered. The dominant concern with such retrospective cohort studies is the presence of selection bias toward open surgery in ‘sicker’ patients, thereby selecting fitter candidates for laparoscopy. While multivariable adjustment addresses known potential confounders, it cannot address unknown or unobserved variables such as differences in perioperative care. However, the results of our sensitivity analysis excluding those patients who underwent bowel resection are reassuring. In addition, due to the retrospective nature of this study, deficiencies in data reporting in patient records, and the inability to track patients upon discharge who may have presented to other hospitals were unable to be accounted for. However, we feel the data presented are overall truly representative of our experience with laparoscopy and open surgery for adhesive SBO.
Conclusion Treatment of patients with adhesive SBO by the laparoscopic approach is associated with quicker recovery of GI function, shorter LOS, and a reduction in overall
complications as compared to conventional open surgical management. This must be interpreted within the limitations of a retrospective study. Future prospective studies are needed to more accurately account for these factors, and to better inform patient selection for laparoscopic approach. Disclosure Drs. James Byrne, Fady Saleh, Luciano Ambrosini, Fayez Quereshy, and Timothy Jackson have no conflicts of interest or financial ties to disclose. Dr. Allen Okrainec is a consultant for Covidien and receives honoraria for speaking and teaching. Funding
No external sources of funding were used for this study.
References 1. Sikirica V et al (2011) The inpatient burden of abdominal and gynecological adhesiolysis in the US. BMC Surg 11:13. doi:10. 1186/1471-2482-11-13 2. Tierris I et al (2011) Laparoscopy for acute small bowel obstruction: indication or contraindication? Surg Endosc 25(2):531–535. doi:10.1007/s00464-010-1206-8 Epub 2010 Jul 7 3. Vettoretto N et al (2012) Laparoscopic adhesiolysis: consensus conference guidelines. Colorectal Dis 14(5):e208–e215. doi:10. 1111/j.1463-1318.2012.02968.x 4. Catena F et al (2011) Bologna guidelines for diagnosis and management of adhesive small bowel obstruction (ASBO): 2010 evidence-based guidelines of the World Society of Emergency Surgery. World J Emerg Surg 6(5):21. doi:10.1186/1749-7922-65 5. Menzies D (1992) Peritoneal adhesions. Incidence, cause, and prevention. Surg Annu 24(Pt 1):27–45 6. Luijendijk RW et al (1996) Foreign material in postoperative adhesions. Ann Surg 223(3):242–248 7. Kelly KN et al (2014) Laparotomy for small-bowel obstruction: first choice or last resort for adhesiolysis? A laparoscopic approach for small-bowel obstruction reduces 30-day complications. Surg Endosc 28(1):65–73. doi:10.1007/s00464-013-3162-6 Epub 2013 Sep 4 8. Bastug DF et al (1991) Laparoscopic adhesiolysis for small bowel obstruction. Surg Laparosc Endosc 1(4):259–262 9. Farinella E et al (2009) Feasibility of laparoscopy for small bowel obstruction. World J Emerg Surg. 4:3. doi:10.1186/17497922-4-3 10. Sauerland S et al (2006) Laparoscopy for abdominal emergencies: evidence-based guidelines of the European Association for Endoscopic Surgery. Surg Endosc 20(1):14–29 Epub 2005 Oct 24 11. Leon EL et al (1998) Laparoscopic management of small bowel obstruction: indications and outcome. J Gastrointest Surg 2(2):132–140 12. Strickland P et al (1999) Is laparoscopy safe and effective for treatment of acute small-bowel obstruction? Surg Endosc 13(7):695–698 13. Nagle A et al (2004) Laparoscopic adhesiolysis for small bowel obstruction. Am J Surg 187(4):464–470 14. Zerey M et al (2007) Laparoscopic management of adhesive small bowel obstruction. Am Surg 73(8):773–778 discussion 778–9 15. Gutt CN et al (2004) Fewer adhesions induced by laparoscopic surgery? Surg Endosc 18(6):898–906 Epub 2004 Apr 27 16. Mancini GJ et al (2008) Nationwide impact of laparoscopic lysis of adhesions in the management of intestinal obstruction in the
123
Surg Endosc
17.
18.
19.
20.
21.
22.
23. 24.
US. J Am Coll Surg. 207(4):520–526. doi:10.1016/j.jamcollsurg. 2008.04.026 Epub 2008 Jun 24 Saleh F et al (2014) Laparoscopic versus open surgical management of small bowel obstruction: an analysis of short-term outcomes. Surg Endosc 28(8):2381–2386. doi:10.1007/s00464014-3486-x Epub 2014 Mar 21 Bone RC et al (1992) Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. The ACCP/SCCM Consensus Conference Committee. American College of Chest Physicians/Society of Critical Care Medicine. Chest 101(6):1644–1655 Wullstein C, Gross E (2003) Laparoscopic compared with conventional treatment of acute adhesive small bowel obstruction. Br J Surg 90(9):1147–1151 Johnson KN et al (2012) Laparoscopic management of acute small bowel obstruction: evaluating the need for resection. J Trauma Acute Care Surg 72(1):25–30. doi:10.1097/TA. 0b013e31823d8365 discussion 30-1; quiz 317 Khaikin M et al (2007) Laparoscopic vs. open surgery for acute adhesive small-bowel obstruction: patients’ outcome and costeffectiveness. Surg Endosc 21(5):742–746 Epub 2007 Mar 1 Okamoto H et al (2012) Clinical outcomes of laparoscopic adhesiolysis for mechanical small bowel obstruction. Asian J Endosc Surg 5(2):53–58. doi:10.1111/j.1758-5910.2011.00117.x Epub 2011 Nov 24 Chopra R et al (2003) Laparoscopic lysis of adhesions. Am Surg 69(11):966–968 Li MZ et al (2012) Laparoscopic versus open adhesiolysis in patients with adhesive small bowel obstruction: a systematic
123
25.
26.
27. 28.
29.
30.
31.
review and meta-analysis. Am J Surg. 204(5):779–786. doi:10. 1016/j.amjsurg.2012.03.005 Epub 2012 Jul 12 Margenthaler JA et al (2006) Risk factors for adverse outcomes following surgery for small bowel obstruction. Ann Surg 243(4):456–464 Grafen FC et al (2010) Management of acute small bowel obstruction from intestinal adhesions: indications for laparoscopic surgery in a community teaching hospital. Langenbecks Arch Surg 395(1):57–63. doi:10.1007/s00423-009-0490-z Epub 2009 Mar 28 Kirshtein B et al (2005) Laparoscopic management of acute small bowel obstruction. Surg Endosc 19(4):464–467 Epub 2005 Feb 3 Ghosheh B, Salameh JR (2007) Laparoscopic approach to acute small bowel obstruction: review of 1061 cases. Surg Endosc 21(11):1945–1949 Epub 2007 Sep 19 O’Connor DB, Winter DC (2012) The role of laparoscopy in the management of acute small-bowel obstruction: a review of over 2,000 cases. Surg Endosc 26(1):12–17. doi:10.1007/s00464-0111885-9 Epub 2011 Sep 5 Levard H et al (2001) Laparoscopic treatment of acute small bowel obstruction: a multicentre retrospective study. ANZ J Surg 71(11):641–646 Suter M et al (2000) Laparoscopic management of mechanical small bowel obstruction: are there predictors of success or failure? Surg Endosc 14(5):478–483
