Right Colon Resection for Colon Cancer: Does Surgical Approach Matter?

Abstract

Background:

Surgical resection with curative intent remains the standard of care for colon cancer. This study aims to compare the 30-day outcomes and oncologic results following open, laparoscopic, and robot-assisted right colon resection for colon cancer using the Targeted Colectomy American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) database.

Materials and Methods:

All patients undergoing elective, right colon resection with primary anastomosis were identified within the targeted colectomy ACS-NSQIP database. Only patients with stage I, II, or III colon cancer were included. The association of surgical approach with oncologic results and 30-day morbidity and mortality outcomes was investigated using a variety of statistical tests.

Results:

A total of 3518 patients met inclusion criteria; 1024 (29.1%) underwent open surgery (OS), 2405 (63.4%) underwent laparoscopic surgery, and 89 (2.5%) underwent robotic surgery. Patients undergoing OS were significantly more likely to have positive resection margins (P < .001). Patients undergoing OS were significantly more likely to experience prolonged intubation (P = .02), deep wound infections (P = .001), wound dehiscence (P = .005), deep venous thrombosis (P = .04), bleeding requiring a blood transfusion (P < .001), a prolonged postoperative ileus (P < .001), and longer length of hospital stay (P < .001), and were more likely to die (P = .02).

Conclusion:

The laparoscopic approach to colon resection for colon cancer has lower 30-day morbidity compared to OS. The robotic approach is equivalent to the laparoscopic approach, and its utilization may increase in the future.

Introduction

There has been a gradual shift toward an increased incidence of right-sided colon cancers compared with transverse and left-sided colon cancers in the United States.¹ Surgical resection for curative intent remains the cornerstone of treatment for colon cancer.^2–4 While the open approach to colon resection has proven to be a durable option, minimally invasive approaches are becoming increasingly popular. The two most commonly used minimally invasive techniques for colon resection are laparoscopy and robot-assisted surgery (RS). Laparoscopy, introduced in the United States in the late 1980s, has shown equivalent pathologic outcomes and has proved advantageous over the open approach to colon resection for colon cancer with respect to postoperative pain, incidence of wound events, length of hospital stay, and cosmesis.^5–7 RS, introduced in the United States in the early 2000s, has several theoretical advantages over laparoscopy to colon resection for colon cancer, including three-dimensional imaging, a stable camera platform, improved instrument mobility, and articulation, as well as ease of performing intracorporeal anastomoses.^8–10 Nonetheless, direct comparisons between these three surgical approaches is lacking in the literature. The purpose of this study is to compare pathologic results and 30-day morbidity and mortality outcomes between the open, laparoscopic, and robot-assisted surgical approaches to right colon resection for colon cancer using the targeted colectomy American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) database.

Materials and Methods

All patients undergoing elective, right colon resection with primary anastomosis for stage I, II, or III colon cancer from the years 2012 through 2014 were identified within the main ACS-NSQIP database using current procedural terminology (CPT) codes. Specifically, those patients undergoing open right colon resection were identified by CPT code 44160, and those patients undergoing laparoscopic right colon resection were identified by CPT code 44205. All patients identified by these two CPT codes were then merged with the targeted colectomy ACS-NSQIP database. To identify those patients who underwent robot-assisted right colon resection, all patients identified by CPT code 44205 were then divided into laparoscopic and robot-assisted groups using the “col-approach” variable within the targeted colectomy database. Finally, within the targeted colectomy database is a “col-indication” variable which describes the primary indication for surgery. Only those patients with the primary indication for surgery of colon cancer were included in our analysis. Patients in the laparoscopic and robotic groups who required conversion to an open procedure, with disseminated/stage IV colon cancer, chemotherapy within 30 days of surgery, radiation within 90 days or surgery, nonelective cases, and those pregnant at the time of surgery were excluded from our analysis.

Preoperative patient variables, intraoperative patient variables, pathology outcomes, and 30-day morbidity and mortality outcomes were investigated. Chi-square and between-groups t-test were used to examine differences between the operative approach groups. A composite outcome variable was created to further elucidate any differences in 30-day morbidity between the three groups. The composite outcome variable included the incidence of prolonged intubation, superficial wound infection, deep wound infection, organ space infection, wound dehiscence, deep venous thrombosis, bleeding requiring transfusion, and unplanned return to the operating room. For 30-day postoperative outcomes that were statistically different between the three surgical approaches, multivariate logistic regression was used to examine the association of operative approach with 30-day morbidity and mortality outcomes. A backward-elimination approach was used in which predictors with P > .10 were dropped at each step. All statistical analyses were done using SAS version 9.3 (SAS Institute, Cary, NC) and P < .05 was considered statistically significant.

Results

A total of 3518 patients met inclusion criteria; 1024 (29.1%) patients underwent open right colon resection, 2405 (68.4%) patients underwent laparoscopic right colon resection, and 89 (2.5%) patients underwent robot-assisted right colon resection (Table 1). With respect to patient demographics, patients who underwent open surgery (OS) were significantly older (P < .0001) and had a lower average preoperative serum albumin level (P < .0001) compared to patients who underwent either laparoscopic surgery (LS) or RS. Patients who underwent LS had a lower average American Society of Anesthesiologists (ASA) class (P < .0001) and were less likely to have experienced greater than 10% weight loss before surgical intervention (P = .002) compared to the OS and RS groups.

Table 1.

Patient Demographics, Stratified by Operative Approach

Patient variable	Open n = 1024	Laparoscopic n = 2405	Robot-assisted n = 89	P
Age, years (mean ± SD)	70.7 ± 12.2	68.3 ± 12.6	68.9 ± 11.8	<.0001
Female gender [n (%)]	576 (56.3)	1276 (53.1)	40 (44.9)	.05
Race (n, %)				.08
Caucasian	751 (73.3)	1792 (74.5)	68 (76.4)
AA	93 (9.1)	249 (10.4)	12 (13.5)
Other	180 (17.6)	364 (15.1)	9 (10.1)
BMI, kg/m² (mean ± SD)	28.6 ± 6.7	28.5 ± 6.3	29.3 ± 6.3	.44
Functional status [n (%)]				.27
Independent	993 (97.0)	2347 (97.6)	96 (96.6)
Partially dependent	26 (2.5)	41 (1.7)	2 (2.3)
Totally dependent	3 (0.3)	5 (0.2)	1 (1.1)
Unknown	2 (0.2)	12 (0.5)	0
ASA class (n, %)				<.0001
1	14 (1.4)	42 (1.8)	0
2	330 (32.2)	1036 (43.1)	32 (36.0)
3	609 (59.5)	1224 (50.9)	52 (58.4)
4	71 (6.9)	103 (4.2)	5 (5.6)
Smoker (n, %)	121 (11.8)	281 (11.7)	7 (7.9)	.53
Chronic steroid use (n, %)	42 (4.1)	77 (3.2)	1 (1.1)	.20
HTN [n (%)]	612 (59.8)	139 (5.8)	5 (5.6)	.25
COPD [n (%)]	69 (6.7)	139 (5.8)	5 (5.6)	.55
DM [n (%)]	197 (19.2)	430 (17.9)	22 (24.7)	.20
ESRD [n (%)]	7 (0.7)	5 (0.2)	0	.10
Serum albumin level, g/dL (mean ± SD)	3.81 ± 0.53	3.92 ± 0.48	3.87 ± 0.47	<.0001
Weight loss >10% [n (%)]	58 (5.7)	77 (3.2)	5 (5.6)	.002

AA, African American; ASA, American Society of Anesthesiologists; BMI, body mass index; COPD, chronic obstructive pulmonary disease; DM, diabetes mellitus; ESRD, end-stage renal disease; HTN, hypertension; SD, standard deviation.

Table 2 details the pathology outcomes and 30-day morbidity and mortality outcomes. Patients who underwent OS had a significantly shorter operative time (P < .001), but were significantly more likely to have positive margins (P < .0001) compared with the LS and RS groups. While all groups had an average lymph node retrieval greater than the minimum recommended 12 lymph nodes, the LS had a significantly larger average number of lymph nodes retrieved (P = .002) compared to the OS and RS groups.^11,12 In terms of 30-day morbidity and mortality outcomes, patients who underwent OS were significantly more likely to experience prolonged intubation (P = .01), deep venous (P = .04), deep surgical site infection (P = .001), wound dehiscence (P = .001), bleeding requiring blood transfusion (P < .001), prolonged ileus (P < .001), have a longer average length of hospital stay (P < .001), and die within 30 days of surgery (P = .001).

Table 2.

Oncologic Outcomes and 30-Day Morbidity and Mortality Outcomes, Stratified by Surgical Approach

Outcome	Open n = 1024	Laparoscopic n = 2405	Robot-assisted n = 89	P
Operative time, min (mean ± SD)	135.9 ± 89.2	142.5 ± 63.3	187.2 ± 81.4	<.001
Positive margins [n (%)]	24 (2.3)	22 (0.9)	0	<.0001
Lymph nodes retrieved (mean ± SD)	18 ± 12	19 ± 11	18 ± 9	.002
Prolonged intubation [n (%)]	22 (2.2)	24 (1.0)	0	.01
Pulmonary embolism [n (%)]	7 (0.7)	15 (0.6)	0	.73
Deep venous thrombosis [n (%)]	19 (1.9)	22 (0.9)	0	.04
Superficial SSI [n (%)]	61 (6.0)	115 (4.8)	2 (2.3)	.17
Deep SSI [n (%)]	15 (1.5)	8 (0.3)	1 (1.1)	.001
Organ space SSI [n (%)]	39 (3.8)	57 (2.3)	2 (2.3)	.06
Wound dehiscence [n (%)]	11 (1.1)	5 (0.2)	0	.001
Bleeding requiring transfusion [n (%)]	156 (15.2)	168 (7.0)	10 (11.2)	<.001
Unplanned return to the OR [n (%)]	48 (4.7)	79 (3.3)	2 (2.3)	.10
Prolonged ileus^a [n (%)]	198 (19.3)	235 (9.8)	11 (12.4)	<.001
Total LOS (mean ± SD)	7.9 ± 7.7	5.2 ± 4.7	4.4 ± 2.4	<.001
30-day mortality [n (%)]	15 (1.5)	8 (0.3)	1 (1.1)	.001

Defined as prolonged NPO status or nasogastric decompression for more than 3 days after surgery or reinsertion of nasogastric tube or reinstating NPO status later than postoperative day 4.^22,23

LOS, length of stay; NPO, nil per os; OR, operating room; SD, standard deviation; SSI, surgical site infection.

Table 3 details the variables associated with the composite morbidity outcome. Older patients were significantly more likely to experience the composite outcome (P = .0003) while female patients and those without end-stage renal disease were significantly less likely to experience the composite event (P = .003 and P = .02, respectively). With respect to surgical approach, patients who underwent OS were significantly more likely to experience the composite event than either those who underwent LS or RS (P < .001 and P = .049, respectively), while there was no difference between the LS and RS groups (P = .87).

Table 3.

Multivariate Logistic Regression Predicting Composite Outcome Event

Variable	OR (95% CI)	P
Age	1.01 (1.001, 1.02)	.0003
Female gender	0.77 (0.65, 0.92)	.003
Without ESRD	0.26 (0.08, 0.84)	.02
Laparoscopic versus open approach	0.54 (0.45, 0.64)	<.001
Robot-assisted versus open approach	0.56 (0.32, 1.00)	.049
Laparoscopic versus robot-assisted approach	0.95 (0.54, 1.68)	.87

CI, confidence interval; ESRD, end-stage renal disease; OR, odds ratio.

Discussion

The ideal approach to right-sided colon resection for colon cancer remains unknown, with a paucity of literature available to surgeons to make an informed decision. Our study shows that patients who undergo open right colon resection are older, are more likely to have positive margins, and experience higher 30-day morbidity and mortality compared to patients who undergo either laparoscopic or robot-assisted right colon resection. With respect to laparoscopic and robot-assisted right colon resection, there was no significant difference between the two groups with respect to the composite morbidity outcome.

Patients who underwent open right colon resection were older and malnourished as evidenced by a lower preoperative average serum albumin level and a significantly higher number of patients who lost greater than 10% of their weight preoperatively.^2,13 While these patients may be considered at higher risk, which is why they underwent an open operation, our data demonstrate that the benefits gained from a potentially shorter operative time do not outweigh the postoperative morbidity and mortality experienced in this group. Not unexpectedly, patients who underwent OS were significantly more likely to experience prolonged intubation, experience deep wound infections, experience a prolonged ileus, and had a longer average length of hospital stay. These outcomes may be all related to the midline incision most commonly used during open colon resection, which has been shown to lead to increased postoperative pain and an increased risk of postoperative wound infection compared to minimally invasive approaches.^14–16 Furthermore, the increased bowel manipulation performed during OS has been previously shown to be associated with prolonged postoperative ileus, as seen in our study.¹⁷

With respect to cancer operations, the most important factors to consider are the pathology and oncologic results as well as the morbidity and mortality associated with the surgical approach. Our study showed that the average lymph nodes retrieved were higher than the minimum recommendation of 12 lymph nodes regardless of operative approach. Therefore, while the LS group had a higher average number of lymph nodes retrieved, this may not be clinically significant. On the contrary, patients who underwent OS were significantly more likely to have positive resection margins than either the LS or the RS group. This finding, in combination with the increased risk of 30-day morbidity and mortality following OS, leads us to recommend that all patients with right-sided colon cancer undergoing right colon resection with primary anastomosis should have surgery through a minimally invasive approach unless contraindicated by an inability to tolerate pneumoperitoneum.

Laparoscopy has proved equivalent to open colon resection in terms of pathology and oncologic outcomes over both the short and long term.^7,18 What remains to be determined is the role of the robotic platform. Interestingly, de'Angelis et al. investigated the learning curve associated with adoption of laparoscopy and RS for right colon resection for colon cancer and found that proficiency in RS required only 16 cases, while proficiency in laparoscopy required 31 cases.⁵ While our study included a significantly larger number of patients in the LS group, this finding is consistent with the fact that RS is currently being promoted for more technically challenging pelvic operations.^19,20 Indeed, the adoption of RS for rectal cancer increased from 5.5% to 13.3% from 2005 through 2013, while the use of laparoscopy for rectal cancer only increased from 33.5% to 38.1% over the same time period.¹⁹ Nevertheless, our study shows that the robotic approach may prove beneficial for operations outside of the pelvis with respect to both pathology outcomes as well as early patient morbidity and mortality outcomes.

One of the biggest obstacles to the adoption of robotic surgery is the perceived, or even real, costs associated with this surgical approach. While it is true that the short-term direct costs associated with robotic surgery are more expensive than the costs associated with laparoscopy, the way that these costs are reported can be misleading. For example, most studies report only the costs associated with the robot and the robotic instruments rather than including the total costs associated with patient length of stay, readmission rates, and long-term reinterventions specific to colon cancer patients. While the ACS-NSQIP database does not include cost information, we would argue that additional studies are needed to more effectively determine differences in the cost of these approaches over the long term and that the adoption of either laparoscopy or RS should be based on the ability of these platforms to minimize patient morbidity and to produce oncologically sound results.

Despite our results, our study does have limitations that are worth mentioning. The 2014 main ACS-NSQIP database contains information from 517 hospitals throughout the United States, while the targeted colectomy ACS-NSQIP database contains information from 203 hospitals throughout the United States.²¹ This creates the potential for underreporting of events in our study due to less hospitals participating in the targeted colectomy ACS-NSQIP database. This may at least partially explain why only 89 RS cases were available for analysis. Furthermore, the ACS-NSQIP database only includes outcomes within the first 30 days postoperatively, which limits our analysis to early pathology and patient morbidity and mortality outcomes only. Finally, the ACS-NSQIP database does not include cancer staging information, and therefore, we cannot determine if any particular cancer stage would benefit more from one surgical approach over the other using this database.

Conclusion

The incidence of right-sided colon cancer is increasing in the United States compared to transverse and left-sided colon cancers. Determining the ideal surgical approach to right-sided colon cancer in terms of both patient morbidity and mortality and pathology outcomes is necessary. Our study demonstrates that a minimally invasive approach to right-sided colon resection for colon cancer leads to improved pathologic outcomes with decreased 30-day patient morbidity and mortality. Additional studies are needed to determine differences in minimally invasive approaches, including laparoscopy and RS, over the long term.

Authors' Contributions

I.N.H.: study concept, data acquisition, data analysis, data interpretation, and article writing; T.J.: data acquisition, data analysis, data interpretation, and article writing; M.S.: data interpretation and writing; X.K.: data acquisition, data interpretation, and final article review; R.L.A.: data acquisition, data interpretation, and final article review; F.B.: data interpretation and final article review; V.O.: data interpretation and final article review; S.A.: study concept, data interpretation, and final article review.

Footnotes

Disclosure Statement

T.J., M.S., X.K., R.L.A., F.B., and S.A. have no conflicts of interest. Ivy N. Haskins has no conflicts of interest relevant to this publication but does have an active grant with the Americas Hernia Society. Vincent Obias has no conflicts of interest relevant to this publication, but he is a consultant for Intuitive Surgical, Inc. but has not received any personal fees from Intuitive Surgical, Inc. for production of this article.

References

Macrae

. Colorectal cancer: Epidemiology, risk factors, and protective factors. In: UpToDate, Goldberg

, Lipman

(eds). Waltham, MA: UpToDate, 2017, Accessed January 19.

Haskins

, Baginsky

, Amdur

, Agarwal

. Preoperative hypoalbuminemia is associated with worse outcomes in colon cancer patients. Clin Nutr, 2017; 36:1333–1338.

Rodriguez-Bigas

, Lin

, Crace

. Surgical management of colorectal cancer. In: Kufe

, Pollock

, Weichselbaum

, et al. (eds). Holland-Frei Cancer Medicine. 6th Ed. Hamilton, ON: BC Decker, 2003.

Tokunaga

, Sakamoto

, Nakagawa

, et al. Prognostic nutritional index predicts severe complications, recurrence, and poor prognosis in patient with colorectal cancer undergoing primary tumor resection. Dis Colon Rectum, 2015; 58:1048–1057.

Miller

, Dao

, Paluvoi

, et al. Comparison of 30-day postoperative outcomes after laparoscopic vs. robotic colectomy. J Am Coll Surg, 2016; 223:369–373.

de'Angelis

, Lizzi

, Azoulay

, et al. Robotic versus laparoscopic right colectomy for colon cancer: Analysis of the initial simultaneous learning curve of a surgical fellow. J Laparoendosc Adv Surg Tech A, 2016; 26:882–892.

Guillou

, Quirke

, Thorpe

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

Witkiewicz

, Zawadzki

, Rzaca

, et al. Robotic-assisted right colectomy: Surgical technique and review of the literature. Wideochir Inne Tech Maloinwazjne, 2013; 8:253–257.

Corcione

, Esposito

, Cuccurullo

, et al. Advantages and limits of robot-assisted laparoscopic surgery: Preliminary experience. Surg Endosc, 2005; 19:117–119.

10.

Delaney

, Lynch

, Senagore

, et al. Comparison of robotically performed and traditional laparoscopic colorectal surgery. Dis Colon Rectum, 2003; 46:1633–1639.

11.

Wong

. Lymph node counts and survival rates after resection for colon and rectal cancer. Gastrointest Cancer Res, 2009; 3:S33–S35.

12.

Choi

, Park

, Lee

, et al. Variability in the lymph node retrieval after resection of colon cancer: Influence of operative period and process. Medicine (Baltimore), 2016; 95:e4199.

13.

Bozzetti

, Gianotti

, Braga

, et al. Postoperative complications in gastrointestinal cancer patients: The joint role of nutritional status and nutrition support. Clin Nutr, 2007; 26:698–709.

14.

Haskins

, Amdur

, Vaziri

. The effect of smoking on bariatric surgical outcomes. Surg Endosc, 2014; 28:3074–3080.

15.

Masoomi

, Reavis

, Smith

, et al. Risk factors for acute respiratory in bariatric surgery: Data from the nationwide inpatient sample, 2006–2008. Surg Obes Relat Dis, 2013; 9:277–281.

16.

Kang

, Park

, Baik

, et al. A comparison of open, laparoscopic, and robotic surgery in the treatment of right-sided colon cancer. Surg Laparosc Endosc Percutan Tech, 2016; 26:497–502.

17.

Keller

, Stein

. Facilitating return of bowel function after colorectal surgery: Alvimopan and gum chewing. Clin Colon Rectal Surg, 2013; 26:186–190.

18.

Green

, Marshall

, Collinson

, et al. Long-term follow-up of the medical research council CLASSIC trial of conventional versus laparoscopically assisted resection in colorectal cancer. Br J Surg, 2013; 100:75–82.

19.

Schootman

, Hendren

, Ratnapradipa

, et al. Adoption of robotic technology for treating colorectal cancer. Dis Colon Rectum, 2016; 59:1011–1018.

20.

Liao

, Zhao

, Lin

, et al. Robotic-assisted versus laparoscopic colorectal surgery: A meta-analysis of four randomize controlled trials. World J Surg Oncol, 2014; 12:122.

21.

ACS NSQIP Participant Use Data File. https://www.facs.org/quality-programs/acs-nsqip/program-specifics/participant-use (accessed January 19, 2017).

22.

Tevis

, Carchman

, Foley

, et al. Postoperative ileus-more than just prolonged length of stay?. J Gastrointest Surg, 2015; 19:1684–1690.

23.

Haskins

, Fleshman

, Amdur

, Agarwal

. The impact of bowel preparation on the severity of anastomotic leak in colon cancer patients. J Surg Oncol, 2016; 114:810–813.