Outcomes Comparing 19,769 Patients Who Underwent Robotic, Laparoscopic, or Open Colorectal Surgery

Abstract

Background:

This study examines the efficacy of robotic colorectal surgery across a large health system, focusing on factors such as hospital stay, operative time, pain management, and postoperative complications. The objective of this study is to compare outcomes of colorectal surgery by robotic, laparoscopic, and open techniques in a multicenter study.

Methods:

A retrospective cohort study analyzed patients undergoing colorectal surgery from 2016 to 2022 using a clinical database from a large United States hospital system. Multivariable regression adjusted outcomes for various patient factors and institutional practices.

Results:

Among 19,769 patients, robotic surgery was associated with shorter hospital stays (5.6 days vs 7.9 for laparoscopic and 11.2 for open, P < .0001), fewer postoperative complications, and lower pain levels. Despite longer operating room (OR) times, robotic surgery showed favorable trends in mortality, hospice discharge, and readmission rates compared to other approaches.

Conclusion:

Robotic-assisted colorectal surgery may lead to decreased hospital stays, readmission rates, and improved patient outcomes across various healthcare settings.

INTRODUCTION

Laparoscopic surgery is widely regarded as the preferred minimally invasive approach in many surgical fields. In colorectal surgery, it marked a significant technological advancement, offering substantial benefits over traditional open techniques. These include improved postoperative recovery, reduced length of stay (LOS), enhanced patient satisfaction, better cosmetic outcomes, lower complication and readmission rates, and overall cost-effectiveness. Despite these proven advantages, laparoscopic surgery has not always been adopted as the primary approach by some surgeons, with utilization rates estimated at two-thirds of eligible cases and even lower rates for cancer-related procedures.^1–10

The advent of robotic-assisted surgery represents a response to the limitations of laparoscopy and advancements in surgical technology. Robotic-assisted surgery has gained widespread acceptance across various surgical disciplines, including urology, gynecology, and general surgery, and its adoption continues to grow in the United States.^11–13 By 2006, robotic-assisted techniques like low anterior resection (LAR) were being explored for their potential advantages over laparoscopy, such as enhanced visualization, improved instrument control, greater dexterity, and articulating instruments.^20,21 These advancements address significant limitations of laparoscopy, including its two-dimensional imaging, restricted instrument range of motion, and suboptimal ergonomics for surgeons.

The increasing integration of robotic systems into general surgery residency programs and its consideration as a training requirement²² underscore the growing importance of this technology. However, challenges remain. Robotic-assisted surgery has been associated with longer operative times and higher hospital costs compared to laparoscopic approaches.¹¹ Additionally, its adoption in abdominal surgery has lagged behind other specialties, partly due to the advanced laparoscopic skill sets of minimally invasive surgeons.¹³ Nonetheless, robotic colorectal surgery shows promise in addressing the limitations of laparoscopic techniques, particularly in rectal resections, with evidence of favorable outcomes.

Robotic-assisted colorectal surgery has seen a steady rise in utilization, reflecting its growing role in general surgery. Between 2013 and 2018, adoption increased from 2.8% to 11.4%, highlighting its expanding role even in community settings.^23,24 This trend reflects a broader shift toward robotic surgery as an essential component of minimally invasive surgical practice.

The primary aim of this study is to evaluate the comparative advantages of robotic-assisted surgery over laparoscopic and open approaches in a large healthcare network in the United States. Specifically, we assess key outcomes such as length of hospital stay, pain management, and postoperative complications. We hypothesize that with increasing adoption of robotic techniques, patients undergoing robotic-assisted surgery will experience shorter hospital stays, improved pain control, and reduced postoperative complications compared to those undergoing laparoscopic or open surgery.

METHODS

Data Source

Data were obtained from a large hospital system’s enterprise data warehouse (EDW), which includes data from over 180 hospitals across the United States. The study was based on deidentified data and deemed exempt from our hospital system’s institutional review board (IRB) review. Data elements, variables, and the data itself were abstracted by the hospital system’s research division and then provided to the investigators through a secured server.

Patient Selection and Study Design

In this study, a retrospective cohort design was employed, utilizing a comprehensive clinical database housed within a prominent hospital system in the United States. The inclusion criteria encompassed patients who underwent colorectal surgery during the extensive time frame spanning from January 1, 2016, to January 1, 2022. Notably, all participating hospitals within this study were integral components of a unified healthcare system and used the same standardized electronic medical record (EMR) software.

A sample size calculation was carried out a priori to estimate the primary outcome of this study, hospital LOS. A minimum of 1,567 patients in each cohort (robotic, laparoscopic, and open) are needed to determine a difference in hospital LOS by 1 day, with a standard deviation (SD) of 15 days, with an 80% power and an α value of 0.05.

The original population from 2016 to 2022 comprised 19,769 deidentified patients who underwent colorectal surgery, including robotic, laparoscopic, and open surgery, both emergent and elective, for benign as well as malignant disease, within the specified study period. Colorectal surgery cases only included laparoscopic or robotic-assisted or open surgery for partial right colectomy, partial left colectomy, lower anterior resection, abdominoperineal resection, total colectomy, subtotal colectomy for both benign disease and malignant cancer cases. Exclusion criteria included patients who did not undergo the aforementioned colorectal operations, those below the age of 18, those with metastatic disease, or those diagnosed with stage 4 colorectal cancer. Emergency surgery status was risk adjusted in the multivariable regression equation described in the statistical analysis section.

Study Cohorts and Variables

For comparison, we stratified the data into 3 cohorts of colorectal surgical techniques: the robotic-assisted surgery cohort, the laparoscopic cohort, and the open surgery cohort. A comprehensive collection of demographic information was performed, encompassing variables such as age, sex, race, insurance status, American Society of Anesthesiologists (ASA) class, smoking status, body mass index (BMI), surgery status (emergent vs nonemergent), and comorbidities, as detailed in Table 1. In addition, we added two institutional variables: (1) Teaching hospital status (Community Hospital Teaching, Community Hospital Nonteaching, Tertiary Referral Hospital Teaching), (2) a hospital-level proxy for cumulative robotic experience was created: ≤5 years versus ≥5 years since robotic colorectal surgery was first recorded in that facility.

Table 1.

Demographics of Patients Who Underwent Colorectal Surgery

	Robotic	Laparoscopic	Open
	N = 8,586	N = 6,192	N = 4,991	P-Value
Age
18–64	56.4%	53.4%	49.2%	<.0001
>65	43.6%	46.6%	50.8%
Gender
Male	49.5%	48.6%	49.5%	.53
Female	50.5%	51.4%	50.5%
Race
White	70.5%	68.7%	66.1%	<.0001
Black	9.2%	10.6%	10.9%	.001
Hispanic	14.7%	15.3%	17.3%	.0002
Other	5.7%	5.4%	5.7%	.77
Insurance
Private	38.5%	34.3%	24.1%	<.0001
Uninsured	3.3%	3.6%	5.6%	<.0001
Medicare	45.3%	48.0%	54.6%	<.0001
Medicaid	5.0%	6.3%	8.5%	<.0001
Other	7.9%	7.9%	7.2%	.31
Smoking status
Current	13.7%	14.3%	17.5%	<.0001
Former	30.2%	28.8%	27.2%	<.0001
Never	49.6%	49.9%	49.4%	.87
Unknown	6.6%	7.0%	6.0%	.09
BMI
Underweight: <18.5	3.3%	4.5%	5.7%	<.0001
Normal: 18.5–24.9	27.2%	29.4%	31.6%	<.0001
Overweight: 25–29.9	32.7%	31.6%	30.4%	.02
Obese: 30–39.9	30.1%	28.1%	25.9%	<.0001
Morbidly obese: >40	6.2%	5.7%	5.6%	.29
ASA class
1—Normal healthy patient	1.1%	1.1%	0.7%	.11
2—Moderate systematic disease	37.5%	31.5%	20.5%	<.0001
3—Severe systemic disease	53.8%	55.2%	54.4%	.25
4—Life threatening disease	5.9%	10.2%	18.9%	<.0001
5—Morbid pt-need surg to survive	0.1%	0.1%	1.5%	<.0001
6—Brain dead for donor organs	0.01%	0.0%	0.0%	.52
7—Level none	1.6%	1.9%	4.0%	<.0001
Surgery status
Elective	90.3%	83.0%	68.3%	<.0001
Emergent	9.7%	17.0%	31.7%
Charlson Comorbidity Index	1.1 (±1.9)	1.4 (±2.2)	2.1 (±2.6)	<.0001

The identification of comorbidities and secondary outcomes was based on the utilization of ICD-10 codes, with the following mappings applied: Coronary Artery Disease (CAD)—I25.10, Congestive Heart Failure (CHF)—I50, Chronic Obstructive Pulmonary Disease (COPD)—J44.9, History of Stroke—I63.0, Chronic Kidney Disease (CKD)—N18.9, Diabetes Mellitus (DM)—E11.9, Obesity (BMI >30)—E66.9, and Tobacco Smoker—F17.200. Emergency surgery status was defined by ICD-9 converted to ICD-10 codes following the guidelines and recommendations of the American Association for the Surgery of Trauma (AAST).²⁵

The primary outcome measures of this study encompassed the evaluation of LOS, operating room (OR) time, and pain control, as detailed in Table 2. OR time was recorded as “wheels-in to wheels-out.” These outcomes were adjusted by confounders that had an independent effect on both the exposure type and outcome of the study. The variables used in our multivariable regression to account for confounders were age, gender, race, insurance, ASA classification (to account for overall patient health before surgery), BMI, smoking status, and Charlson Comorbidity Index (CCI) (to account for prehospital comorbid conditions), and emergent versus nonemergent surgery status (to account for the emergent nature of disease). Additionally, secondary outcomes encompassed occurrence of postoperative complications, as detailed in Table 3. The identification and classification of these complications were obtained by the ICD-10 coding system, which included Pneumonia (J09–J18), Acute Respiratory Distress Syndrome (ARDS) (J80), Surgical Site Infection (L08.9, T81.40), Urinary Tract Infection (UTI) (N30.0, N34.0, N39.0), Deep Vein Thrombosis (DVT) (I82.4x), Pulmonary Embolism (PE) (I26.xx), Postoperative Ileus (K91.89), Cerebral Infarction (I63.9), Myocardial Infarction (MI) (I21.9), Cardiac arrest (I46), Hospice (Z51.5), Mortality (R99), Readmission (Z75.1), Sepsis (A41.9), and Intraoperative bleeding (K91.61, N99.6).

Table 2.

Primary Outcomes

	Robotic	Laparoscopic	Open
	N = 8,586	N = 6,192	N = 4,991
Length of stay (days)	5.7 (±8.6)	7.7 (±13.6)	11.3 (±12.0)
		P < .0001	P < .0001
		P < .0001*	P < .0001*
		P < .0001**	P < .0001**
Or time (minutes)***	253.8 (±93.4)	210.7 (±93.6)	187.5 (±86.2)
		P < .0001	P < .0001
		P < .0001*	P < .0001*
		P < .0001**	P < .0001**
Pain control	93.4%	93.0%	95.7%
		0.93 (0.82, 1.07)	1.58 (1.35, 1.86)
		1.08 (0.92, 1.28)*	1.94 (1.59, 2.38)*
		1.14 (0.97, 1.34)**	2.02 (1.64, 2.47)**
Mortality	0.6%	1.0%	4.2%
		1.59 (1.11, 2.29)	6.81 (5.05, 9.19)
		1.12 (0.75, 1.69)*	2.73 (1.91, 3.92)*
		1.12 (0.75, 1.69)**	2.73 (1.91, 3.92)**
Complication rate	3.7%	4.5%	9.4%
		1.23 (1.04, 1.45)	2.68 (2.31, 3.10)
		1.04 (0.87, 1.25)*	1.59 (1.34, 1.89)*
		1.03 (0.86, 1.24)**	1.58 (1.33, 1.88)**

*Adjusted by age, gender, race, insurance status, ASA class, Charlson Comorbidity Index Score, BMI, smoking status, surgery status (emergency vs nonemergent), reliability adjustment by hospital volume; **also adjusted by hospital/teaching status and institutional robotic experience (≤5 years vs ≥5 years); ***OR time = wheels-in to wheels-out.

Table 3.

Secondary Outcomes, Postoperative Complications

	Robotic	Laparoscopic	Open
	N = 8,586	N = 6,192	N = 4,991	P-Value
Pneumonia	0.7%	1.1%	2.4%	<.0001
Acute respiratory distress syndrome	0.03%	0.03%	0.2%	.01
Surgical site infection	3.3%	3.6%	6.7%	.20
Urinary tract infection	0.5%	0.5%	1.2%	<.0001
Deep vein thrombosis	0.06%	0.2%	0.3%	.001
Pulmonary embolism	0.2%	0.3%	0.5%	.004
Postoperative ileus	5.2%	6.2%	10.8%	<.0001
Multisystem organ failure
Cerebrovascular accident	0.1%	0.2%	0.4%	.02
Myocardial infarction	2.2%	2.7%	3.9%	<.0001
Cardiac arrest	0.4%	0.7%	1.7%	<.0001
Hospice	0.6%	1.2%	3.4%	<.0001
ICU admission
Readmission	24.4%	29.0%	32.5%	<.0001
Sepsis	2.4%	3.0%	7.9%	<.0001
Pain scale (mean)	3.08 (±1.9)	3.07 (±2.0)	3.03 (±2.0)
		P = .95	P = .21
		P = .01*	P < .0001*
		P = .01**	P < .0001**
Pain scale (max: worst)	5.87 (2.7)	6.10 (2.7)	6.81 (2.7)
		P < .0001	P < .0001
		P < .0001*	P < .0001*
		P < .0001**	P < .0001**
Intraoperative bleeding	0.01%	0.0%	0.08%	.02

*Adjusted by age, gender, race, insurance status, ASA class, Charlson Comorbidity Index score, BMI, smoking status, surgery status (emergent vs nonemergent), and reliability adjustment by hospital volume; **also adjusted by hospital/teaching status and institutional robotic experience (≤5 years vs ≥5 years).

Statistical Analysis

The data were analyzed between January 2016 to January 2022. Statistical analyses were performed using SPSS (IBM Corporation, Armonk, NY) and SAS (SAS, Cary, NC). P-values of <.05 were deemed statistically significant. Continuous data were expressed as mean with SD or median, and the difference between the two groups was compared. Parametric data expressed as proportions were evaluated by χ² tests and Student t tests for continuous variables. Nonparametric data were evaluated by Fisher’s exact test for proportions and the Wilcoxon-Rank-Sum test for continuous variables. Logistic regression was used for binary outcomes, and linear regression methods were used for continuous outcomes. Both primary and secondary outcomes were adjusted for age, race, gender, insurance status, BMI, tobacco smoking status, American Society of Anesthesia Class, CCI, hospital/teaching status, and emergency surgery status (emergent vs nonemergent).²⁵ Although detailed data on individual surgeons, including their training and experience, were not available, we accounted for variability through reliability adjustments based on hospital volume as well as time of institutional experience for robotic surgery (≤5 years vs ≥5 years). These adjustments aimed to mitigate the potential influence of institutional and surgeon-related factors on the outcomes.²⁹

RESULTS

Patient Demographics and Comorbidities

A total of 19,769 patients met the inclusion criteria. In terms of demographic variables (Table 1), we categorized our study population into two age groups: 18–64 years and 65 years and older. Most of our participants fell within the younger age group. A predominant inclination towards minimally invasive surgical techniques, particularly robotic surgery, was observed in the younger age cohort (below 65 years), in stark contrast to the older age group (above 65 years),¹⁶ where open surgical procedures were more prevalent. Gender-based differentiations exhibited negligible statistical significance, with comparable figures across the board. Of the 8,586 robotic cases, 60.3 % were performed in community teaching hospitals, 2.6 % in teaching tertiary referral hospitals, and 5.5% in nonteaching community hospitals. Fifty-two percent (65/126) of the hospitals that performed at least one robotic case were teaching institutions, versus 49% and 48% for laparoscopic and open cohorts, respectively.

Demographic differences were observed in the selection of surgical approaches. For instance, patients with private insurance were more likely to undergo robotic surgery, while those with Medicare or Medicaid were more likely to have open procedures. Similarly, racial disparities were evident, with Black and Hispanic patients more likely to undergo open surgery compared to robotic or laparoscopic approaches. Table 1 illustrates these racial differences in surgical approach; specifically, Whites had the highest proportion undergoing robotic surgery (70.5%). In comparison, although Whites made up most open surgery, there was a decreasing trend from robotic to laparoscopic to open (70.5% to 68.7% to 66.1%, P-value .0001). Conversely, individuals of Black and Hispanic racial ethnicities displayed an increasing trend towards open surgery as opposed to minimally invasive procedures. A similar pattern emerged concerning insurance status, wherein patients with private insurance were more inclined towards robotic surgery compared to open (38.5% vs 24.1%), while uninsured patients, as well as those with Medicare or Medicaid, were more likely to opt for open surgery rather than laparoscopic or robotic surgery.

Among current smokers, there was a greater prevalence of open surgery (17.5%) relative to laparoscopic or robotic surgery (14.3% and 13.7%, respectively, P-value <.0001). As anticipated, former smokers exhibited a higher incidence of minimally invasive surgery, with robotic surgery being more prominent (30.2% vs 28.8% and 27.2%, P-value .0001).

Patients classified as underweight based on BMI demonstrated a higher likelihood of undergoing open surgery as opposed to robotic surgery (5.7% vs 3.3%, P-value <.0001). A similar trend was observed for patients with normal BMI, where open surgery was more prevalent (31.6% vs 27.2%, P-value <.0001). However, a contrasting pattern emerged for overweight (32.7%) and obese (30.1%) BMI categories. For operative technique, the open surgery group demonstrated higher rates of certain complications compared to minimally invasive approaches. For instance, the incidence of surgical site infections (SSIs) was higher in the open group (6.7%) than in the robotic (3.2%) and laparoscopic (3.7%) groups (P < .0001). Similarly, venous thromboembolism (VTE) occurred more frequently in the open group (0.3%) compared to robotic (0.06%) and laparoscopic (0.2%) groups (P = .001). Postoperative ileus was also more common in the open group (10.8%) compared to robotic (5.2%) and laparoscopic (6.3%) groups (P < .0001). All P-values have been clarified to ensure accurate reporting of statistical differences.

Overall, robotic surgery consistently demonstrated lower rates of complications, including SSIs, VTE, and ileus, compared to the open and laparoscopic approaches, and these findings were statistically significant where reported.

Analyzing the ASA classification, patients classified as ASA 1 displayed comparable outcomes for minimally invasive and open surgeries, albeit lacking statistical significance. In contrast, ASA 2 patients demonstrated a higher incidence for robotic surgery over open surgery (37.5% vs 20.5%, P-value <.0001). No statistically significant variations were observed for ASA 3 patients. Conversely, ASA 4 and ASA 5 patients exhibited higher percentages of open surgery, aligning with expectations.

Patients with comorbidities exhibited a notable predilection for open surgery. Among patients with prior MI, CHF, peripheral vascular disease, cerebrovascular disease, dementia, chronic pulmonary disease, rheumatologic disease, peptic ulcer disease, diabetes, cerebrovascular events, mild and moderate to severe liver disease, lymphoma, AIDS, and moderate to severe renal disease, open surgery prevailed as the primary choice, showcasing statistically significant outcomes as detailed in Table 1. Furthermore, a higher CCI correlated with a greater likelihood of opting for open surgery, while robotic surgery emerged as the least favored approach (2.1 [±2.6], 1.1 [±1.9]). As anticipated, elective surgeries predominantly leaned towards robotic procedures (90.3%), followed by laparoscopic surgeries (83.0%), and open surgeries (68.3%). Conversely, emergent cases exhibited a contrasting pattern, with open surgery prevailing (31.7%) over laparoscopic surgery (17.0%) and robotic surgery (9.7%).

Primary Outcomes

In our study, our primary outcomes of interest were hospital LOS, OR time, and pain control. The open surgery approach was associated with a longer LOS, as measured in days, with an average of 11.2 (±11.7) days. In comparison, laparoscopic surgery had a shorter LOS at 7.9 (±13.8) days (P < .0001), and robotic surgery had the shortest LOS at 5.6 (±8.5) days (P < .0001). These values were adjusted to account for factors such as age, gender, race, insurance, CCI, ASA classification, BMI, smoking status, and emergency surgery status.

Conversely, the trend for OR time, measured in minutes, displayed a different pattern. Robotic surgery had the longest OR time, with an average of 253.3 (±92.9) minutes (P < .0001), followed by laparoscopic surgery at 211.7 (±94.7) minutes (P < .0001). Open surgery had the shortest OR time at 186.0 (±84.2) minutes (P < .0001). These values were also risk-adjusted. After additional adjustment for institutional variables and surgeon experience, effect estimates changed by ≤3 % and remained statistically significant (Table 2).

Regarding pain control, as measured by the highest reported pain value on the first postoperative day, the open surgery arm had a higher value (6.81 ± 2.7) compared to laparoscopic surgery (6.10 ± 2.7), with the lowest pain value in the robotic surgery arm (5.87 ± 2.7), P < .0001. It indicates that patients undergoing open surgery were more likely to report higher pain levels than those undergoing minimally invasive surgery. Open surgery had significantly higher mortality and complication rates compared to robotic and laparoscopic surgeries. Robotic surgery had overall lower mortality and complication rates, but they were not significant after risk adjustment.

Secondary Outcomes

Our analysis of postoperative complications, as outlined in Table 3, revealed noteworthy findings. Open procedures exhibited a higher likelihood of developing pneumonia compared to minimally invasive surgery, with the lowest incidence observed in the robotic surgery group (2.4% vs 0.7%). A similar trend was observed for ARDS, with open surgery having the highest incidence. Surgical site infection was more prevalent in the open surgery group, although this difference did not reach statistical significance (P-value .20).

Conversely, several postoperative complications demonstrated statistically significant differences favoring the robotic surgery arm. These included UTIs, with a higher incidence in the open surgery group compared to both robotic and laparoscopic groups (1.2% vs 0.5%). DVT, PE, and postoperative ileus were also more common in the open surgery group (10.8%) compared to the robotic arm (5.2%) and laparoscopic arm (6.2%). Additionally, cerebrovascular accidents (CVAs), MIs, and cardiac arrests were more prevalent in the open surgery group, highlighting statistically significant differences.

Patients undergoing open surgery were more likely to be discharged to hospice care (3.4%) compared to those in the robotic arm (0.6%) and laparoscopic arm (1.2%), with these differences being statistically significant. Similarly, the mortality rate was highest in the open surgery group (4.2%), while the robotic group had the lowest mortality rate (0.6%), and the laparoscopic group had a mortality rate of 1.0%. Readmission rates to the hospital were also higher after open surgery (32.5%) compared to the robotic group (24.4%). Furthermore, sepsis and intraoperative bleeding were more frequently observed in the open surgery group, with the lowest incidence found in the robotic surgery group.

Overall, these findings suggest that open surgery was associated with a higher risk of certain postoperative complications, greater discharge to hospice care, higher mortality rates, increased readmissions, sepsis, and intraoperative bleeding, while robotic surgery demonstrated more favorable outcomes in several of these measures.

DISCUSSION

With the increasing demand for less invasive procedures aimed at enhancing pain control, reducing hospital stays, and improving cosmetic outcomes, robotic surgery has emerged as a promising technology in colorectal minimally invasive surgery. The adoption of robotic surgery has surged not only in colorectal procedures but across various surgical specialties.^14,15 This study aimed to evaluate several outcomes, focusing primarily on hospital LOS.

Although robotic surgery is known to require longer operative times, previous studies suggest that its postoperative outcomes may outweigh those of laparoscopic and open techniques.^19,26 Our findings indicate that robotic surgery prolonged colorectal operations on average by 42 minutes. Despite this, LOS was significantly shorter in the robotic cohort by 2 days compared to laparoscopic surgery and over 5 days compared to open surgery. This difference remained significant after adjusting for pre-existing comorbidities, emergency surgery status, perioperative risk (ASA class), BMI, hospital volume, and other patient-specific factors.

Several factors potentially contributed to the shorter hospital stays observed in the robotic cohort. Patients undergoing robotic surgery reported lower maximum pain levels and were less likely to experience postoperative complications such as pneumonia and postoperative ileus. These results align with previous studies demonstrating robotic surgery’s benefits in reducing postoperative ileus and LOS among colorectal cancer patients.²⁵ Smaller studies have also shown that robotic colorectal surgery is associated with lower postoperative pain compared to laparoscopic techniques.²⁶

Our study encompassed a substantial cohort of 19,769 patients from 2016 to 2022, revealing a notable shift towards robotic-assisted procedures, which accounted for 43.4% of colorectal operations. This trend suggests increasing acceptance of robotic surgery as an accepted approach for colorectal disease. In contrast, a study on Medicare beneficiaries from 2010 to 2016 reported significantly fewer robotic procedures compared to laparoscopic surgeries.²⁷ This difference may reflect varying patient demographics, with our study highlighting a bias for robotic surgery among younger patients, former smokers of white ethnicity with private insurance, while patients without insurance who identified as Black or Hispanic were more likely to undergo open surgery.

Our analysis highlights demographic disparities, such as differences by race and insurance status, in the utilization of surgical techniques, highlighting inequities of access to clinical care.¹⁷ Patients with private insurance and those identifying as White were more likely to undergo robotic or laparoscopic procedures, whereas patients with Medicare, Medicaid, or no insurance and those identifying as Black or Hispanic were more likely to undergo open surgery. These disparities may reflect variations in access to minimally invasive surgical techniques, referral patterns, or other unmeasured factors such as socioeconomic status. To mitigate bias from demographic factors, we performed risk adjustment by including race and insurance status as covariates in our multivariable regression analysis. While this approach helps control for some variability, residual confounding may still persist. Addressing these disparities through further investigation and targeted interventions is essential for promoting equitable access to advanced surgical care.

Incorporating a hospital-level proxy for surgeon experience (≤5 years vs ≥5 years) did not attenuate the advantage of robotic surgery for LOS or complications, suggesting that the observed benefits are not confined to high-volume experts. However, access to robotic surgery remains an issue as only 2–3% of tertiary hospital admissions involved robotic surgery despite comparable outcomes after risk adjustment. This skewed availability may contribute to the racial and insurance disparities¹⁸ we observed and supports targeted investment in robotic platforms and training at under-resourced centers.

Our findings also indicate a higher likelihood of open surgery among patients with multiple comorbidities such as CAD, COPD, and others, reinforcing previous observations that patients without such comorbidities are more likely to undergo robotic-assisted procedures.¹⁴

Robotic surgery’s advantages were particularly evident in obese patients (BMI >25), where it was associated with shorter hospital stays and lower readmission rates compared to laparoscopic and open approaches.¹⁹ Despite longer OR times for robotic procedures compared to open surgery, our analysis did not find an increased incidence of postoperative complications. On the contrary, robotic surgery was associated with lower rates of specific complications such as pneumonia, UTI, DVT, PE, and postoperative ileus, whereas open surgery showed higher rates of CVAs, MIs, and cardiac arrests. Finally, patients undergoing robotic surgery exhibited favorable outcomes in terms of mortality rates, discharge to hospice care, and readmission rates compared to open surgery, suggesting improved survival and reduced need for postoperative care.

Our data show that robotic colectomy added approximately 45 minutes of OR time yet still delivered a 2-day shorter hospital stay and a one-third lower complication rate versus laparoscopy. The same trade-off appears in some large registry studies. In a matched NSQIP study robotic cases ran longer, yet they achieved a 1.5-day LOS saving, fewer septic events and fewer discharges to skilled-nursing facilities.²⁹ The most recent six-year NSQIP cohort confirmed the same pattern where operative time rose, but LOS, overall morbidity and 30-day mortality all fell with the robotic approach.³⁰ Taken together, these results suggest that robotic surgery’s economic value hinges less on time in OR than on what happens afterwards. When recovery is faster and complications are less, these savings may offset the up-front capital and instrument premium.

Conversely, we confirm a persistent access gap where uninsured and Medicaid patients in our system were significantly more likely to undergo an open operation than a robotic one (5.6 % vs 3.3 %, P ≤ .001). A study using a national registry reported the same disparity, reporting that public or absent insurance halves the odds of receiving a robotic colectomy even after adjustment for tumor stage and hospital volume.¹⁷ To ensure the observed clinical advantages translate into population-level value rather than selective benefit, we advocate a move towards bundled or episode-based reimbursement that rewards total-care efficiency rather than penalizing the initial use of robotic equipment.

Several limitations of our study should be acknowledged, including its retrospective nature, which precludes establishing causality. In addition, complication risks may reflect patient selection and confounders like comorbidities. Our analysis adjusted for ASA class, CCI, and emergent status to mitigate variability. This would predominantly bias against the open cohort, as the robotic and laparoscopic cohort were comparable in acuity. In addition, while our study utilized a unified healthcare system with standardized EMRs, ensuring consistent data collection across hospitals, the distribution of procedures at specific hospitals was not analyzed. Thus, variations in perioperative pathways or discharge criteria may have influenced outcomes like LOS. While reliability adjustment accounted for hospital-level variations,²⁸ some residual confounding may persist. It is also important to acknowledge that surgeon selection and preference for specific surgical techniques were not directly captured in our dataset. Differences in individual surgeon experience, skill, and decision-making may have influenced the choice of surgical approach and, consequently, the observed outcomes. To address potential variability in patient acuity and comorbidities, the CCI was included as a covariate in our multivariable regression analysis, mitigating its influence on the results. The inability to distinguish console generation (eg, S/Si vs ξ) remains a limitation. However, as older models or robotic consoles are phased out newer models should continue to improve surgeon performance.

As previously mentioned, we employed reliability adjustment methods to account for hospital-specific practice variations, including perioperative pathways, resource availability, and discharge protocols, which may have contributed to differences in outcomes such as LOS and complications. Despite these adjustments, we recognize that unmeasured factors, such as prior surgeries, surgeon experience, and institutional practices toward certain techniques such as enhanced recovery after surgery (ERAS), may still have influenced the findings.

Despite these limitations, the study’s strengths include its large sample size, allowing robust evaluation of LOS as a primary outcome, and comprehensive adjustment for confounding clinical variables not typically found in large administrative dataset studies. Our findings contribute to the expanding literature on robotic surgery’s efficacy in colorectal procedures.

CONCLUSION

Our study underscores the growing role of robotic surgery in colorectal minimally invasive procedures. Despite the longer operative times associated with robotic surgery, our findings demonstrate potential significant advantages in terms of shorter hospital stays and lower postoperative complication rates compared to laparoscopic and open techniques. This suggests that while robotic procedures may extend operative duration, these additional minutes may also translate into improved patient outcomes.

The increasing adoption of robotic technology, as evidenced by our data showing a substantial proportion of colorectal surgeries performed robotically, reflects a shifting paradigm in surgical practice towards minimally invasive approaches. This trend is particularly notable among younger patients and those with favorable demographic profiles, emphasizing the need for equitable access to advanced surgical techniques across diverse patient populations.

In conclusion, our findings support the continued integration of robotic-assisted surgery as a viable option in colorectal procedures, offering improved patient outcomes and contributing to the evolution of surgical practice towards enhanced minimally invasive techniques. Future research should focus on prospective studies to confirm these findings and explore additional factors influencing surgical outcomes across different patient demographics and healthcare settings.