The impact of an enhanced recovery after surgery pathway for video-assisted and robotic-assisted lobectomy on surgical outcomes and costs: a retrospective single-center cohort study

To determine the impact of enhanced recovery after surgery (ERAS) pathway implementation on outcomes and cost of robotic- and video-assisted thoracoscopic (RATS and VATS) lobectomy. Retrospective review of 116 consecutive VATS and RATS lobectomies in the pre-ERAS (Oct 2018–Sep 2019) and ERAS (Oct 2019–Sep 2020) period. Multivariate analysis was used to determine the impact of ERAS and operative approach alone, and in combination, on length of hospital stay (LOS) and overall cost. Operative approach was 49.1% VATS, 50.9% RATS, with 44.8% pre-ERAS, and 55.2% ERAS (median age 68, 65.5% female). ERAS patients had shorter LOS (2.22 vs 3.45 days) and decreased total cost ($15,022 vs $20,155) compared with non-ERAS patients, while RATS was associated with decreased LOS (2.16 vs 4.19 days) and decreased total cost ($14,729 vs $20,484) compared with VATS. The combination of ERAS + RATS showed the shortest LOS and the lowest total cost (1.35 days and $13,588, P < 0.001 vs other combinations). On multivariate analysis, ERAS significantly decreased LOS (P = 0.001) and total cost (P = 0.003) compared with pre-ERAS patients; RATS significantly decreased LOS (P < 0.001) and total cost (P = 0.004) compared with VATS approach. ERAS implementation and robotic approach were independently associated with LOS reduction and cost savings in patients undergoing minimally invasive lobectomy. A combination of ERAS and RATS approach synergistically decreases LOS and overall cost.


Introduction
Minimally invasive surgery (MIS) techniques in pulmonary lobectomy have been shown to be safe and offer patients a faster recovery and shorter length of stay (LOS) compared to traditional thoracotomy approach [1][2][3]. Video-assisted thoracoscopic (VATS) lobectomy has been widely adopted across academic centers, while the more recent introduction of robotic-assisted thoracoscopic (RATS) lobectomy adds three-dimensional visualization of the operative field and wristed articulation of the robotic instruments [2]. Despite the theoretic advantages, previous studies have shown RATS lobectomy to be associated with longer operative time [3,4] and higher costs [5], without demonstrating further improvements in LOS compared with VATS lobectomy [6].
Enhanced recovery after surgery (ERAS) pathways are structured, evidence-based care bundles designed to limit physiological stress and expedite patient recovery. The ERAS pathway has been popularized in myriad of surgical procedures including esophagectomy [7], pancreaticoduodenectomy [8], and colorectal surgery [9]. Despite some heterogeneity in protocol, ERAS pathways commonly focus on three aspects of perioperative care: multimodal analgesia, early enteral feeding, and early ambulation [10]. ERAS implementation has been associated with reductions in LOS and costs, and enhanced surgical outcomes [7,11].
More recently, ERAS pathways specifically designed for thoracic surgery have been associated with a reduction in post-operative complications, shorter LOS and cost savings in the open and VATS settings [12]. In the last decade, the utilization of robotic pulmonary resections increased significantly nationwide [2]. To date, the potential impact of ERAS pathway implementation on RATS lobectomy has not been reported. Therefore, after recently instituting an ERAS pathway for all minimally invasive lobectomy patients at our institution, we hypothesized that the combination of ERAS and RATS approach would synergistically optimize outcomes and minimize cost.

Patient cohort and study design
This retrospective study compared VATS lobectomy and RATS lobectomy before and after implementation of a thoracic-specific ERAS pathway at a single tertiary care facility. RATS lobectomy was started on October 2018. A standardized ERAS pathway was implemented on July 2019, utilizing current evidence-based guidelines [13] for optimal management of lobectomies. After obtaining approval from the institutional review board (IRB20-100), a review of a prospectively maintained lung surgery database (IRB 13-087) was performed for all consecutive MIS lobectomies between October 1, 2018 and September 30, 2020. Informed consent was waived by the IRB because of the minimal risk of the study.

Outcomes and data collection
The primary outcomes were post-operative LOS and overall cost. The secondary outcomes included both process measures such as time to chest tube removal, first ambulation and dietary advancement as well as, duration of surgery, pain scores on post-operative day (POD) one, complications and 30-day readmission rates. Costs were further analyzed by per-minute OR cost, supply costs, and costs associated with inpatient stay.

ERAS pathway
See Table 1 for a detailed description of the components of our ERAS pathway, based on recommendations from both the ERAS Society and European Society of Thoracic Surgery [13].

RATS technique
Our robotic lobectomy technique utilizes a four-port approach with an additional assist port and positive pressure insufflation to 8 mm Hg. A camera port is placed in the 7th or 8th interspace with two additional 8 mm and one 12 mm port for lower lobectomy and two 12 mm ports for upper and middle lobectomies. Pleural, fissural and nodal dissection is facilitated by a bipolar energy grasper. The non-operative lobes are manipulated atraumatically with a sponge roll. Paravertebral blocks using 0.25% bupivacaine with epinephrine and cryoneurolysis of the T7-T9 intercostal nerves are performed at the end of procedure. The specimen is extracted via enlargement of the assistant port. A single 28-French chest tube (which fits the 8 mm port incision) is inserted through the most anterior port, directed posteriorly and apically and placed to -8 cm H2O pressure using the Medela Topaz electronic drainage system (Medela AG, Switzerland) which applies to all patients. Chest tube removal indications were lack of air leak, and output below 300 ml over 24-h period.

VATS technique
Our VATS lobectomy technique utilizes a uniportal thoracoscopic approach in the 5th or 6th intercostal space along the posterior axillary line. A fissureless technique minimizing the use of cautery during parenchymal dissection is utilized with vascular and parenchymal division by Endo GIA (Medtronic, USA) stapling. A paravertebral block with 0.25% bupivacaine with epinephrine is performed at the end of the procedure. One 32-French chest tube is placed through the same intercostal opening and set to -8 cm H2O pressure via Madela Topaz system (applies to all patients) with identical chest tube removal criteria as the RATS cohort.

Statistical analysis
We tabulated descriptive statistics (mean (SD), median [IQR] and percentages). Characteristics were compared by different MIS approach or by ERAS using chi-square for categorical variables and Students' t tests or Mann-Whitney test for continuous variables. The differences across combination approaches were evaluated by ANOVA or Kruskal-Wallis test. Outcomes comparisons by MIS approach or ERAS were evaluated using generalized liner model with gamma distribution. The multivariable model was adjusted with the patient demographic, pre-operative variables and operative characteristics shown in Table 1 as well as surgeons' factor. All statistical analyses were performed using Stata (version 14.2; StataCorp). Statistical significance was established with two-sided tests as P ≤ 0.05.

Demographics
A total of 116 patients were identified (Table 2). Mean age was 67.58 (± 9.27 years) and 65.5% were female. Indication for lobectomy was lung cancer in 83.6% of the cases with a mean of 15.9 (± 8.6) lymph nodes dissected. The median pre-operative FEV 1 was 2.38 L and the mean pre-operative DLCO was 73.5% of predicted for the entire cohort. The median smoking history was 30 pack-years. Most common payor was Medicare (government national health insurance program in the United States) in 68% of cases. Robotic approach was utilized in 51% of patients, and ERAS pathway was utilized in 55% of patients. No statistical differences were noted between VATS and RATS cohorts, or between pre-ERAS and ERAS cohorts in any of the patient demographic, pre-operative or operative characteristics except for history of thromboembolic event, which was higher in the RATS group than in the VATS group (P = 0.02) ( Table 2).

Impact of operative approach on outcomes and cost
Compared with VATS, the RATS approach decreased operative time by 32.6 min and LOS by 2.03 days. Patients undergoing a robotic approach had a reduction in time to chest tube removal of 42.9 h. There were no 30-day readmissions in the RATS group compared to a 12.3% 30-day readmission rate in the VATS group. The cost analysis showed that the robotic approach resulted in a decrease of $5,755 in total cost, including a $3,369 reduction in inpatient-stay cost and $1,419 lower supply cost. There were no differences between the two groups regarding post-operative pain level, time to first  (Table 3).

Impact of ERAS implementation on outcomes and cost
Compared with the pre-ERAS cohort, patients undergoing minimally invasive lobectomy following ERAS implementation experienced a decrease in LOS of 1.23 days. ERAS implementation reduced time to chest tube removal by 31.7 h and decreased dietary advancement time by 8.3 h. Cost analysis showed that the ERAS group had reduced total cost by $5,133, inpatient-stay cost by $2,361, and supply cost by $1,624 compared with pre-ERAS group. There were no significant differences between the pre-and post-ERAS groups in surgical duration, post-operative pain level, early mobilization, major post-operative complications, 30-day readmission, or OR per-minute cost (Table 3).

Impact of combinations of ERAS and RATS on outcomes and cost
Evaluating combinations of pre-and post-ERAS pathway and surgical approach, patients in the post-ERAS cohort undergoing a robotic approach had the shortest LOS (1.35 days, P < 0.0001) compared with all other combinations, including the shortest time to chest tube removal (20.35 h, P < 0.0001) as well as shortest surgery duration (147.5 min, P = 0.0003). This resulted in the lowest total cost ($13,588, P = 0.0001), inpatient-stay cost ($2,247, P < 0.0001), and supply cost ($4,165, P = 0.002) ( Table 5).
In multivariate analysis, compared with the reference group pre-ERAS + VATS, pre-ERAS + Robotic group and ERAS + Robotic group had significantly reduced LOS (P = 0.005 and P < 0.001, respectively) and chest tube removal time (P = 0.001 and P < 0.001, respectively) ( Table 6). However, the combination of ERAS + Robotic approach impacted LOS the most, suggesting synergistic effect of ERAS and robotic approach on reducing LOS for patients underwent lobectomy.
ERAS + Robotic group was the only group that significantly reduced the total cost compared to the reference group (P < 0.001). Both pre-ERAS + Robotic group (P = 0.001) and ERAS + Robotic group (P < 0.001) had decreased the inpatient-stay cost significantly (Table 6); however, ERAS + Robotic combination had the largest impact.

Discussion
This study suggests that implementation of an ERAS pathway in patients undergoing robotic lobectomy improved surgical outcomes and reduced cost. It highlights the importance of perioperative care and the evolution of operative techniques, both of which are directed by the same team of thoracic surgeons, nurses, nutritionists, and physiotherapists. RATS lobectomy piqued the interest of thoracic surgeons when it was first introduced in 2002 [14] due to its purported advantages of smaller incisions, less postoperative pain, and faster recovery time. More importantly, RATS has the advantage of advanced features such as 3-D visualization, increased instrument range of motion and degrees of freedom to facilitate meticulous dissection; and console design that enhances ergonomic comfort for surgeons operating in a rigid chest cavity [15,16]. VATS, on the other hand, still presents some limitations, such as 2D visualization, non-ergonomic rigid instruments, and the loss of the eye-hand-target axis, which introduces limitations to conducting atraumatic dissection in a closed space [17]. Although there is still debate about whether RATS is superior to VATS in surgical outcomes and costs, the trend in use of RATS for both segmentectomy and lobectomy is increasing significantly across the country, while VATS segmentectomy and lobectomy volume is decreasing [2].
We observed that RATS results in less air leaks compared with VATS. As such, chest tube removal time is a surrogate marker to indicate air leak duration. Our data show that RATS technique significantly reduces time to chest tube removal when compared to VATS. RATS may have contributed to this result through technical modifications such as cautious handling of the lung with the use of sponge rolls only to avoid direct grasping. Improved visualization and precise dissection of the fissure with bipolar cautery compared to monopolar dissection with VATS, as well as improvement in atraumatic angles of dissection and stapling vectors secondary to articulation advantages of the robotic instruments, may also contribute to reduce the risk of air leak by RATS [18].
ERAS pathways for thoracic surgery have recently been described [19][20][21]. The implementation of such multimodal pathways has resulted in fewer complications and a shorter LOS. Our institution has a long-standing history of driving ERAS implementation in esophageal surgery [7]. As such, implementing ERAS pathway in lung surgery was aided by the existing culture of enhanced recovery elements within our nursing staff.
An important ERAS element that we adopted was the utilization of a digital drainage system (Medela Tropez) for chest tube management. The ability to objectively quantify the volume of air leak, as well as store information and display trends in air leak over time, allows for more informed decision-making about chest tube removal and reduces interobserver and clinical practice variability [22]. Our result demonstrates that ERAS implementation based on the utilization of a digital drainage system significantly reduces the chest tube removal time. Other randomized clinical trials have also illustrated the advantage of digital over traditional devices in both chest tube duration and LOS [23,24]. When combined, the RATS technique and ERAS management implementation have a synergistic effect on chest tube management, resulting in the shortest time to chest tube removal.
Optimization of intraoperative and perioperative measures in our institution demonstrated reduction of LOS, cost, and readmission. VATS has previously showed a clear recovery benefit over conventional thoracotomy. One study demonstrated decreased LOS and hospital costs by utilizing their institution's advancement of the ERAS protocol [25]. Brunelli and colleagues, on the other hand, did not demonstrate an improvement in outcomes with ERAS implementation applied to VATS cohort compared to standard care [26]. Similarly, our findings show that ERAS implementation does not significantly improve outcomes on VATS. The potential explanation is that prior to the implementation of ERAS in our institution, we had already adopted some elements of ERAS as our standard care, such as post-operative multimodal analgesia and intraoperative paravertebral analgesia, and early mobilization on postoperative day one. A similar rationale was provided by Brunelli et al. to explain a lack of improvement after ERAS implementation [26]. However, one aspect that we pushed more aggressively after implementing ERAS at our institution was early post-operative oral intake. Nursing staffs are instructed to encourage oral intake as soon as our patients are transferred from PACU to the thoracic ward. Following the implementation of ERAS, we were able to significantly decrease time to first oral intake on both VATS and RATS settings. The combination of early return to oral intake and early removal of the chest tube results in a shortened LOS. Furthermore, a significant reduction in hospital stays leads to a significant reduction in patient costs. This demonstrates the synergistic benefit of the ERAS protocol in conjunction with improvements in operative technique (RATS).
The evolution of the robotic system, along with the increasing experience of the surgeons, has an impact on surgical outcomes and the adoption of this technology [27]. Although several authors have conducted studies on the feasibility, safety, surgical, and oncologic outcomes of robotic pulmonary resection, Kent and colleagues found a trend but not a statistically significant reduction in LOS and complications in RATS lobectomy compared to VATS lobectomy by analyzing the State Inpatient Databases of 8 states from 2008 to 2010 [2]. However, the most recent report from Kent et al. which examined 21 centers in the United States from 2013 to 2019, demonstrated a significant reduction of operative time and LOS for RATS versus VATS [28]. The differences between these two studies from two different era were attributed to surgeons collectively having more experience with robotic procedures, and the most recent generation robotic platform was introduced in 2014 and has several technological advancements over the previous platform [28]. Our own data demonstrate a significant reduction in both operative time and LOS, which eventually contributes to the greater reduction in patient's total cost.
The surgeon's learning curve is directly related to better outcomes and shorter operative time. Meyer and colleagues reported an overall learning curve of 18 ± 3 cases of robotic lobectomy from multiple surgeons. They also reported a trend toward lower morbidity and shorter LOS with increased experience [29]. Similarly, in a systemic review study, Soomro et al. reported a range of 15-20 cases to overcome the learning curve for RATS lobectomy [30]. Our institution is a high-volume RATS lobectomy center that performs more than 100 cases annually. Notably, our institution has a long track record of successfully implementing ERAS pathways in variety of settings, including esophagectomy, colectomy, and pancreatoduodenectomy. Before we implemented ERAS in the lobectomy setting, the nurse team, nutritionists, and physiotherapists were well informed and equipped about each component of the system. This is another reason why we saw a rapid decrease in LOS after implementing the ERAS pathway. Overall, our findings confirm that, with proper training for surgeons and care teams, RATS lobectomy can achieve reproducible overnight inpatient recovery with no or low readmission and the lowest overall cost as a result of the combination of RATS and ERAS implementation.

Limitations
Certain limitations of this study merit consideration. First, the presence of unmeasured confounding variables is possible, as it is in all observational studies. However, we attempted to reduce this bias using a comprehensive prospective dataset that included patient demographics, surgical details, post-op benchmarks, and costs. We do not specifically choose simple cases for RATS. Indeed, patients from all groups are well balanced from the start. Second, this report comes from a single high-volume institution where robotic surgeons have already completed the learning curve. These findings may not be applicable to other clinical settings. Thirdly, our study focused on patients who had minimally invasive thoracic surgery. The results may not be generalizable to units with a lower proportion of minimally invasive cases verses open cases. Another constraint is the sample size. Larger studies will be needed to validate the findings in a different cohort. Additionally, even though this study compares direct costs of care, it does not account for initial fixed cost of a robotic system acquisition. Finally, additional prospective clinical trials are warranted to validate this finding.

Conclusions
Robotic approach and ERAS implementation are independently associated with shorter LOS and cost savings for minimally invasive lobectomy without increasing postoperative complications or 30-day readmission compared to VATS approach. The synergy of ERAS and robotic approach has the potential to set a new standard for LOS and overall cost compared to other strategies.