Re-interventions Following Laparoscopic Surgery for Colorectal Cancer: Data from 818 Individuals from the Dutch Surgical Colorectal Audit

Abstract

Background:

The surgical procedure of choice for the resection of colorectal cancer has shifted in favor of laparoscopic surgery. Although increasing data prove advantages of elective laparoscopic surgery, less is known about the results in acute indications such as surgical re-interventions following colorectal resections. This study aims to assess the clinical benefits in recovery following laparoscopic re-interventions compared with open re-interventions following laparoscopic colorectal cancer surgery.

Subjects and Methods:

We performed an analysis of data from the Dutch Surgical Colorectal Audit from January 2010 to December 2012. All patients requiring surgical re-intervention after initial laparoscopic colorectal surgery were analyzed.

Results:

Out of 27,448 patients, 11,856 underwent laparoscopic surgery. Following laparoscopic surgery, 159 patients (1.3%) had a laparoscopic re-intervention, and 659 patients (5.6%) had an open re-intervention. In a multivariable analysis adjusting for patients' demographics and risk factors, the length of hospital stay was 17 days (interquartile range, 11–16 days) for the laparoscopic group and 23 days (interquartile range, 14–37 days) for the open group (odds ratio [OR]=0.74; 95% confidence interval [CI], 0.65–0.84). In the laparoscopic group the intensive care unit admission rate was 39% compared with 66% in the open group. The 30-day mortality rate was 7 (4%) in the laparoscopic group compared with 89 (14%) in the open group (OR=0.31; 95% CI, 0.13–0.73).

Conclusions:

Laparoscopic re-intervention following laparoscopic surgery for colorectal cancer is feasible in selected patients. Because of the unknown extent of selection bias, prospective studies are needed to define the exact position and benefits of laparoscopic re-interventions.

Introduction

Laparoscopic surgery for colorectal cancer has gained popularity since the first laparoscopic colectomy was described in the early 1990s.¹ Nowadays, it has become the preferred treatment for both benign and malignant colorectal surgery.^2–4 The rate of laparoscopic surgery for colorectal cancer in The Netherlands increased in the last couple of years from 38% in 2010 to 50% in 2012.^5–7

Despite these increasing rates, laparoscopic surgery for re-interventions following colorectal resections is still under debate. In the acute setting with distended bowel and peritonitis, the theoretical risk of enhanced bacteremia due to the pneumoperitoneum and the fear of incomplete exploration and mechanical damage could be the reasons for the current surgical practice where most complications are treated by conventional laparotomy.^8,9

Elective laparoscopy has shown several benefits compared with open surgery with regard to gastrointestinal recovery, hospital stay, abdominal wall complications, number of re-interventions, and cosmetic results.^10,11 Acute laparoscopic re-interventions could have the same advantages due to less extensive surgical trauma. This is expected to result in shorter hospital stay, fewer additional re-interventions, and conserved abdominal wall integrity with a reduced risk of incisional hernias.^12,13 The benefits of acute laparoscopic surgery are already well established for acute appendicitis, cholecystitis, and perforated diverticulitis.^14,15

We aimed determine whether laparoscopic re-intervention following laparoscopic surgery resulted in faster recovery compared with open re-intervention. Therefore we evaluated the laparoscopic colorectal cancer resections from the Dutch population-based surgical registry (Dutch Surgical Colorectal Audit [DSCA]).^5–7

Patients and Methods

Patients

For this observational cohort study we used prospective collected data from the DSCA. All patients who were registered in this audit between January 2010 and December 2012 and who underwent surgical resection for colorectal cancer were included. The protocol for the data collection of the DSCA has been published previously.¹⁶ In summary, data were collected anonymously in a secured online database by a Web-based form. All 92 participating hospitals (100% participation) assigned a responsible surgeon for the data entry. Data quality was assured by weekly online feedback information on number of registered patients' files and overall completeness of patients' files. Hospitals periodically received data quality reports with feedback on all patients with inconsistent or unusual data combinations. The responsible surgeon was asked to verify the data of these patients and to correct the data when indicated. In addition, benchmarked performance indicators were available for all hospitals concerned.

For this study, we identified all patients who underwent laparoscopic surgery for colorectal cancer from the database. Patients in whom laparoscopy was converted to open surgery during the primary resection were excluded, as were those undergoing an endoscopic resection such as transanal endoscopic microsurgery or endoscopic mucosal resection. As all data in the DSCA database are collected anonymously for auditing, this study did not require additional consent from an ethical committee in The Netherlands.

Baseline variables and outcome

The following variables were available and evaluated: age, gender, body mass index, overall comorbidity, cardiac disease, pulmonary disease, diabetes, previous abdominal surgery, American Society of Anesthesiologists (ASA) score, surgical technique, urgency for surgery (“acute” [i.e., within 2 hours from diagnosis], “urgent” [i.e., within 24 hours from diagnosis], or “elective”), type of resection, anastomosis constructed during primary surgery, and, in the case of a primary anastomosis, the presence of a deviating stoma. The types of resections were categorized as a right-sided segmental colectomy (ileocecal resection or right-sided hemicolectomy), left-sided segmental colectomy (transverse resection, left-sided hemicolectomy), rectosigmoid resection (sigmoid resection, anterior resection, low anterior resection), abdominoperineal resection, or other/unspecified.

A re-intervention was defined as a re-operation (open surgery or laparoscopic surgery). The timing of the re-intervention after primary surgery, type, and technique of re-intervention were available in the dataset. No data were available on the number of additional re-interventions as only the most severe re-intervention was asked for to be recorded in the database. For morbidity, only the most severe event was scored, and as this was the re-intervention for this group, no additional information was available. The assessed outcomes were combined 30-day mortality and in-hospital mortality, total length of hospital stay, intensive care unit (ICU) admission rate, and length of ICU stay.

Statistical analysis

Descriptive statistics are presented as mean±standard deviation values, as medians (interquartile range), or as numbers (percentages). Patients' demographics and characteristics of the intervention and re-intervention were compared between the open and laparoscopic groups using the chi-squared test for categorical variables and the independent-samples t test for continuous variables. For the two-sided P value, the level of significance was set at <.05.

The main outcomes are combined 30-day and in-hospital mortality and total length of hospital stay (in days). Additional outcomes are rate of ICU admission and length of ICU stay (in days). Univariate analysis was performed to identify factors associated with the two main outcomes. Age, overall comorbidity, ASA classification, type of resection, anastomosis with or without deviating stoma, and indication for re-intervention were independent predictors for mortality and length of hospital stay in a univariate analysis.

Separate multivariable logistic regression models and multivariable linear regression models were built for mortality, ICU admission rate, and length of hospital stay to retrieve odds ratios (ORs) and 95% confidence intervals (CIs). Length of hospital stay was log-transformed to meet the assumption of a normal distribution. Length of ICU stay could not be analyzed in a linear model because it was not normally distributed even after log transformation. Models were built to adjust for possible confounders in the association between operating technique and the clinical outcome. Two levels of adjustment were used. Model 1 is built with only basic adjustment for the most clinically relevant preoperative risk factors for mortality and that could affect the choice for open or laparoscopic surgery. These included the patients' demographics age, gender, ASA classification, overall comorbidity, and previous abdominal surgery. In Model 2, a more extensive adjustment was performed, and the type of primary procedure, the presence of an anastomosis with or without deviating stoma, and the indication for re-intervention were added to the previous model.

Statistical analyses were performed using Statistical Package for the Social Sciences for Windows version 21.0 software (2012; IBM Corp., Armonk, NY).

Results

Patients and enrollment

From January 2010 until December 2012, 27,448 patients were registered in the DSCA database. Of these patients, 11,856 patients had laparoscopic surgery, and 15,320 patients had open surgery. Overall, in total, 3628 re-interventions were recorded in the database, including 2701 surgical re-interventions (9.8%). Other re-interventions were radiological re-interventions (343 patients) or unspecified. Thus 818 patients (6.9%) who underwent a laparoscopic or open surgical re-intervention after laparoscopic colorectal surgery for a colorectal malignancy between 2010 and 2012 were included in the analyses. Of these 818 patients, 159 patients underwent a laparoscopic re-intervention, and 659 underwent an open re-intervention as shown in the flowchart in Figure 1.

FIG. 1.

Flowchart of patient selection from the Dutch Surgical Colorectal Audit (DSCA) database, 2010–2012.

Patient demographics

The demographics of the included patients are shown in Table 1. The mean age was 69 years in the open group compared with 67 years in the laparoscopic group. No difference in gender, body mass index, previous abdominal surgery, and overall comorbidity was seen. Fewer patients in the laparoscopic group had cardiac comorbidity (P=.039). The ASA classification was higher for the open re-intervention group (24.3% ASA III–IV in the open group compared with 17.0% in the laparoscopic group; P=.048). Over 95% of all patients had an initial elective procedure (P=.122). Compared with the laparoscopic group, more patients in the open group had a right-sided colectomy (28.7% versus 20.1%), and fewer patients had a rectogismoid resection (51.0% versus 66.7%; P=.018). An anastomosis was constructed in 87% of the patients in both groups, but in the open group fewer patients had a deviating stoma compared with the laparoscopic group (P=.023).

Table 1.

Patient Demographics and Primary Surgical Parameters

	Open (n=659)	Laparoscopic (n=159)	P value
Age	68.7±11.3	67.0±10.5	.071
Gender (male/female)	440 (66.8)/219 (33.2)	107 (67.3)/52 (32.7)	.899
BMI (kg/m²)	26.3±4.5	26.6±4.1	.438
Comorbidity	502 (76.2)	113 (71.1)	.181
Cardiac	213 (32.3)	38 (23.9)	.039
Pulmonary	106 (16.1)	24 (15.1)	.759
Diabetes	101 (15.3)	23 (14.5)	.786
Previous abdominal surgery	190 (28.9)	45 (28.5)	.913
ASA classification			.048
ASA I	133 (20.2)	38 (23.9)
ASA II	365 (55.5)	94 (59.1)
ASA III	156 (23.7)	24 (15.1)
ASA IV	4 (0.6)	3 (1.9)
Urgency of surgery			.122
Elective	625 (95.4)	156 (98.1)
Urgent or acute	30 (4.6)	3 (1.9)
Type of resection			.018
Right-sided colectomy	189 (28.7)	32 (20.1)
Left-sided colectomy	58 (8.8)	8 (5.0)
Rectosigmoid resection	336 (51.0)	106 (66.7)
Abdominoperineal resection	48 (7.3)	10 (6.3)
Other/unknown	28 (4.3)	3 (1.8)
Anastomosis			.023
None	86 (13.1)	20 (12.6)
With stoma	88 (15.4)	35 (25.2)
Without stoma	483 (84.6)	104 (74.8)

Data are mean±standard deviation values or n (%).

ASA, American Society of Anesthesiologists; BMI, body mass index.

Indication for re-intervention

The most common indication for surgical re-intervention in both groups was anastomotic leakage or anastomotic abscess, as shown in Table 2. These accounted for 56% of the indications in the open group and 43% in the laparoscopic group. Anastomotic leakage was more frequently the indication for re-intervention in the open group (P=.005). In the laparoscopic group, small bowel obstruction or an abdominal abscess other than an anastomotic abscess was the main indication in 23% of the patients, versus 15% in the open group. The mean interval between primary surgery and re-intervention was 8 days in both groups.

Table 2.

Interval and Indications for Re-intervention

	Open (n=659)	Laparoscopic (n=159)	P value
Interval to re-intervention (days)	8.1±11.9	7.9±8.0	.885
Indication for re-intervention			.073
Anastomotic leak or abscess	368 (55.8)	69 (43.4)	.005
Small bowel obstruction	68 (10.3)	20 (12.8)
Abdominal abscess	35 (5.3)	16 (10.3)
Bleeding	30 (4.6)	8 (5.1)
Bowel perforation	27 (4.1)	5 (3.2)
Urogenital damage	8 (1.2)	3 (1.9)
Unspecified/other	123 (18.7)	38 (23.9)

Data are mean±standard deviation values or n (%).

Clinical outcomes

Clinical outcomes are shown in Table 3. The recorded 30-day and in-hospital mortality was lower in the laparoscopic group with 7 deaths (4.4%) compared with 89 deaths (13.6%) in the open group (P=.001). The median length of hospital stay was shorter for the laparoscopic compared with the open group (17 days versus 23 days; P<.001). ICU admission and stay were higher in the open re-intervention group, with 66% admittance and a stay of 5 days versus 39% admittance and a stay of 2 days in the laparoscopic group (P<.001). In multivariable regression analysis the mortality rate (OR=0.31; 95% CI, 0.13–0.73) remained lower in the laparoscopic group, as did the length of hospital stay (OR=0.74; 95% CI, 0.65–0.84) and the ICU admission rate (OR=0.34; 95% CI, 0.21–0.54), as shown in Table 4.

Table 3.

Clinical Outcomes

	Open (n=659)	Laparoscopic (n=159)	P value
Mortality (30-day and in-hospital)	89 (13.6)	7 (4.4)	.001
In-hospital stay (in days)	23 (14–37)	17 (11–26)	<.001
ICU admission rate	289 (66.4)	45 (39.1)	<.001
ICU length of stay (in days)	5 (2–13)	2 (2–6)	<.001

Data are median (interquartile range) values or n (%).

Table 4.

Effect of Laparoscopy on Mortality, Length of Stay, Intensive Care Unit (ICU) Admission, and Length of ICU Stay

	Laparoscopic re-intervention
Outcomes	OR_unadj (95% CI)	OR_adj-1 (95% CI)	OR_adj-2 (95% CI)
Mortality (30-day and in-hospital)	0.30 (0.13–0.65)	0.33 (0.14–0.77)	0.31 (0.13–0.73)
Length of hospital stay	0.72 (0.63–0.82)	0.72 (0.64–0.83)	0.74 (0.65–0.84)
ICU admission rate	0.33 (0.21–0.50)	0.31 (0.20–0.49)	0.34 (0.21–0.54)

CI, confidence interval; OR_adj-1, odds ratio for Model 1, adjusted for age, gender, American Society of Anesthesiologists classification, previous abdominal surgery, and comorbidity; OR_adj-2, odds ratio for Model 2, adjusted for age, gender, American Society of Anesthesiologists classification, previous abdominal surgery, comorbidity, urgency, type of procedure, anastomosis and stoma, and indication for re-intervention; OR_unadj, unadjusted odds ratio.

Discussion

Laparoscopic re-intervention following laparoscopic colorectal resection resulted in less mortality and faster recovery compared with open re-intervention in the DSCA database. The differences found are clinically relevant, with a 10% lower in-hospital mortality, a shorter hospital stay by 6 days, and 26% fewer ICU admissions. After correction for the known baseline differences in a regression analysis, these differences remained statistically significant.

Six reports have been published on the topic of laparoscopic re-interventions following colorectal surgery, and they reported on a total of 98 patients (range, 4–38 patients).^{12,13,17–20} Half of these studies published a case series and stated laparoscopic re-intervention after laparoscopic or open abdominal surgery to be safe and feasible.^17–19 The other three compared open re-intervention with laparoscopic re-intervention after colorectal surgery.^12,13,20 All reported a high success rate, faster recovery, and fewer deaths in the laparoscopic group. When we compare our results with these studies, we have to note that Wind et al.¹³ compared their laparoscopic re-interventions with patients with open re-intervention after open surgery instead of after laparoscopic surgery. Patient characteristics, indication for re-intervention, and also outcomes are quite similar between these studies with regard to length of hospital stay and mortality rates.

This study is the largest study on laparoscopic re-intervention up until now and describes more patients than the others in total. The prospective and nationwide nature of the DSCA warrants a complete and valid dataset. However, the DSCA is a short-term oncological audit, and therefore not all re-intervention details and related outcomes are available in the dataset. Variables of interest such as patients' condition prior to re-intervention, the presence of organ failure, additional re-interventions, and morbidity after re-intervention were not recorded.

Without a scoring system such as an APACHE score or an intraoperative peritonitis score, it is impossible to determine whether these patients were in a similar condition prior to re-intervention. At the same time, as the exact indication and findings are unknown, re-intervention could equally mean resection of the anastomosis as a negative diagnostics procedure. It is likely that hemodynamic unstable patients had open surgery and patients with severe leakage during diagnostic laparoscopy had conversion to open surgery and were recorded as open surgery, whereas negative diagnostic laparoscopies would be recorded as a laparoscopic re-intervention. These examples are likely to result in worse outcomes for the open group.

In addition, the small number of laparoscopic re-interventions compared with the open re-interventions is likely to be the result of an extensive selection bias. However, during this 3-year period, there were increasing numbers of laparoscopic re-interventions following laparoscopic colorectal surgery, with 15% of all recorded surgical re-interventions in 2010 to 19% in 2011 and 21% in 2012.

Finally, without any data on conversions and intraoperative morbidity, we cannot conclude on the safety of laparoscopic re-intervention. In particular, the fear for incomplete inspection and a high risk of mechanical damage cannot be contradicted by the lack of these data. However, previous studies have not shown this to be more than a theoretical concern.^12,13

Conclusions

This 3-year population-based database of all laparoscopic colorectal cancer resections in The Netherlands shows a lower mortality and faster recovery for selected patients following laparoscopic re-intervention compared with open re-intervention. These results show laparoscopic re-intervention to be a feasible approach for selected patients. However, these results should be handled with caution because of the unknown and potentially large extent of selection bias.

These results should initiate prospective studies to define the exact position and benefits of laparoscopic re-intervention for complications following laparoscopic colorectal surgery.

Footnotes

Disclosure Statement

No competing financial interests exist.

References

Jacobs

, Verdeja

, Goldstein

. Minimally invasive colon resection (laparoscopic colectomy). Surg Laparosc Endosc, 1991; 1:144–150.

Buunen

, Veldkamp

, Hop

, et al. Survival after laparoscopic surgery versus open surgery for colon cancer: Long-term outcome of a randomised clinical trial. Lancet Oncol, 2009; 10:44–52.

Kuhry

, Schwenk

, Gaupset

, Romild

, Bonjer

. Long-term results of laparoscopic colorectal cancer resection. Cochrane Database Syst Rev, 2008;(2):CD003432.

Phillips

, Franklin

, Carroll

, Fallas

, Ramos

, Rosenthal

. Laparoscopic colectomy. Ann Surg, 1992; 216:703–707.

DSCA. Dutch Surgical Colorectal Audit Annual Report. 2010. Available at www.clinicalaudit.nl (accessed July 1, 2014).

DSCA. Dutch Surgical Colorectal Audit Annual Report. 2011. Available at www.clinicalaudit.nl (accessed July 1, 2014).

DSCA. Dutch Surgical Colorectal Audit Annual Report. 2012. Available at www.clinicalaudit.nl (accessed July 1, 2014).

Agresta

, Ciardo

, Mazzarolo

, et al. Peritonitis: Laparoscopic approach. World J Emerg Surg, 2006; 1:9.

Neudecker

, Sauerland

, Neugebauer

, et al. The European Association for Endoscopic Surgery clinical practice guideline on the pneumoperitoneum for laparoscopic surgery. Surg Endosc, 2002; 16:1121–1143.

10.

Veldkamp

, Kuhry

, Hop

, et al. Laparoscopic surgery versus open surgery for colon cancer: Short-term outcomes of a randomised trial. Lancet Oncol, 2005; 6:477–484.

11.

Guillou

, Quirke

, Thorpe

, et al. Short-term endpoints of conventional versus laparoscopic-assisted surgery in patients with colorectal cancer (MRC CLASICC trial): Multicentre, randomised controlled trial. Lancet, 2005; 365:1718–1726.

12.

Vennix

, Abegg

, Bakker

, et al. Surgical re-interventions following colorectal surgery: Open versus laparoscopic management of anastomotic leakage. J Laparoendosc Adv Surg Tech A, 2013; 23:739–744.

13.

Wind

, Koopman

, van Berge Henegouwen

, Slors

, Gouma

, Bemelman

. Laparoscopic reintervention for anastomotic leakage after primary laparoscopic colorectal surgery. Br J Surg, 2007; 94:1562–1566.

14.

Sartelli

, Viale

, Catena

, et al. 2013 WSES guidelines for management of intra-abdominal infections. World J Emerg Surg, 2013; 8:3.

15.

Agresta

, Ansaloni

, Baiocchi

, et al. Laparoscopic approach to acute abdomen from the Consensus Development Conference of the Societa Italiana di Chirurgia Endoscopica e nuove tecnologie (SICE), Associazione Chirurghi Ospedalieri Italiani (ACOI), Societa Italiana di Chirurgia (SIC), Societa Italiana di Chirurgia d'Urgenza e del Trauma (SICUT), Societa Italiana di Chirurgia nell'Ospedalita Privata (SICOP), and the European Association for Endoscopic Surgery (EAES). Surg Endosc, 2012; 26:2134–2164.

16.

Van Leersum

, Snijders

, Henneman

, et al. The Dutch Surgical Colorectal Audit. Eur J Surg Oncol, 2013; 39:1063–1070.

17.

O'Riordan

, Larkin

, Mehigan

, McCormick

. Re-laparoscopy in the diagnosis and treatment of postoperative complications following laparoscopic colorectal surgery. Surgeon, 2013; 11:183–186.

18.

Pera

, Delgado

, Garcia-Valdecasas

, et al. The management of leaking rectal anastomoses by minimally invasive techniques. Surg Endosc, 2002; 16:603–606.

19.

Rosin

, Zmora

, Khaikin

, Bar Zakai

, Ayalon

, Shabtai

. Laparoscopic management of surgical complications after a recent laparotomy. Surg Endosc, 2004; 18:994–996.

20.

Rotholtz

, Laporte

, Lencinas

, Bun

, Aued

, Mezzadri

. Is a laparoscopic approach useful for treating complications after primary laparoscopic colorectal surgery?. Dis Colon Rectum, 2009; 52:275–279.