STRUCTURED ABSTRACT
Objective
We sought to evaluate how implementing a thoracic Enhanced Recovery After Surgery (ERAS) protocol impacted surgical outcomes after elective anatomic lung resection.
Summary Background Data
The effect of implementing the ERAS Society/European Society of Thoracic Surgery (ESTS) thoracic ERAS protocol on postoperative outcomes throughout an entire healthcare system has not yet been reported.
Methods
This was a prospective cohort study within one healthcare system (1/2019–3/2023). A thoracic ERAS protocol was implemented on 5/1/2021 for elective anatomic lung resections, and postoperative outcomes were tracked using the electronic health record and Vizient data. The primary outcome was overall morbidity; secondary outcomes included individual complications, length of stay (LOS), opioid use, chest tube duration, and total cost. Patients were grouped into pre- and post-ERAS cohorts. Bivariable comparisons were performed using independent t-test, chi-square, or Fisher’s exact tests, and multivariable logistic regression was performed to control for confounders.
Results
There were 1,007 patients in the cohort; 450 (44.7%) were in the post-ERAS group. Mean age was 66.2 years; most patients were female (65.1%), white (83.8%), had a BMI between 18.5–29.9 (69.7%), and were ASA class 3 (80.6%). Patients in the post-implementation group had lower risk-adjusted rates of any morbidity, any respiratory complication, pneumonia, surgical site infection, arrhythmias, infections, opioid usage, ICU use, and shorter postoperative LOS (all p<0.05).
Conclusions
Postoperative outcomes were improved after implementation of an evidence-based thoracic ERAS protocol throughout the healthcare system. This study validates the ERAS Society/ESTS guidelines and demonstrates that simultaneous multihospital implementation can be feasible and effective.
Keywords: ERAS, thoracic ERAS, postoperative outcomes, ESTS guidelines, quality improvement, evidence-based medicine
MINI-ABSTRACT
We sought to evaluate the effect of implementing healthcare system-wide thoracic ERAS program based on established guidelines. After implementation, there were lower complication rates and shorter lengths of stay. This study shows that simultaneous protocol implementation throughout a healthcare system is feasible and effective, and it validates ERAS Society recommendations.
INTRODUCTION
Enhanced Recovery After Surgery (ERAS) protocols are multimodal and multidisciplinary perioperative care pathways aimed at improving patient outcomes after surgery by anticipating and reducing the profound surgical stress response1. Initially developed in the colorectal surgery population, ERAS has now been shown to improve patient outcomes across a wide array of operations2–6. Several thoracic ERAS protocols have been described, and while they consist of similar themes and strategies7–9, the interpretation of specifics and subsequent execution of thoracic ERAS has been inconsistent and varied from hospital-to-hospital. Furthermore, most thoracic ERAS protocols described in the literature have been implemented at individual hospitals. Since many hospitals exist under the umbrella of a larger healthcare system, it is important to delineate whether perioperative care protocols can be effectively implemented throughout entire healthcare systems, despite cultural and logistical differences in the individual hospitals of which they are comprised.
In 2019, the ERAS Society and European Society of Thoracic Surgeons (ESTS) published comprehensive guidelines for the perioperative care of patients undergoing anatomic lung resections10. Prior to implementation of a thoracic ERAS program, the perioperative care of thoracic patients at the University of Colorado Health (UCHealth) system was unstandardized with practice variance between and within hospitals and providers. Some of this variation was not explained by patient characteristics and was identified as a potential source of suboptimal outcomes. The UCHealth system consists of six hospitals where thoracic surgery are performed by seven American Board of Thoracic Surgery-certified general thoracic surgeons. In February of 2020, UCHealth moved to implement a thoracic ERAS protocol based on the ERAS Society/ESTS guidelines throughout the healthcare system as part of a quality improvement initiative. After 13 months of iterative revision, the protocol was universally launched in May of 2021 for all patients undergoing anatomic lung resection. While quality control checks for improving protocol adherence at one hospital have previously been reported11, this represents the first comprehensive analysis of the effect of ERAS implementation on postoperative outcomes throughout the healthcare system.
The purpose of this study was to assess the effect of implementing a universal thoracic ERAS protocol for patients undergoing anatomic lung resection across our diverse multi-hospital healthcare system, specifically examining postoperative outcomes. We hypothesized that despite local differences in hospital culture and logistics throughout the healthcare system, implementation would be effective and overall postoperative outcomes would be improved by implementation of this ERAS protocol. Improved postoperative outcomes after this pilot study would not only validate the recommendations made by the ERAS Society/ESTS but also demonstrate that universal implementation across a diverse healthcare system rather than just a single hospital can be effective.
METHODS
Ethical Oversight
This study was approved by the Colorado Multiple Institute Review Board as a quality improvement project (COMIRB approval #20-3051).
Patient Population
This was a multi-center, prospective cohort study from January 2019 to March 2023 across the University of Colorado Health System (UCHealth), a large healthcare system comprised of several hospitals across the front range of Colorado. The Central Region is comprised of the University of Colorado Hospital (UCH) and Highlands Ranch Hospital (HRH). UCH is a quaternary academic referral medical center associated with the University of Colorado School of Medicine. The general thoracic surgery group at UCH performs about 250 anatomic lung resections annually. Surgeons accrue patients locally, regionally through extensive referrals, and nationally including through a collaboration with the National Jewish Hospital outpatient clinics. Many patients referred from National Jewish Hospital are referred for the surgical treatment of non-tuberculosis Mycobacterial disease with pulmonary manifestations. HRH is a small regional hospital where the same thoracic surgery group performs low-risk thoracic operations. The North Region is comprised of Medical Center of the Rockies (MCR) and Poudre Valley Hospital (PVH), which are both medium, private-practice hospitals without surgical trainees. The South Region is comprised of Memorial Hospital Central (MHC) and Memorial Hospital North (MHN), which are two medium, academic-affiliated hospitals with some surgical trainees, but no cardiothoracic surgery fellows. All thoracic operations at the South Region is performed at MHC. All patients who underwent elective anatomic pulmonary resection (segmentectomy, lobectomy, and pneumonectomy) were targeted for inclusion. Patients who underwent non-anatomic lung resections, emergent resections or who were <18 years old were excluded.
Protocol Implementation and Quality Assurance
A multidisciplinary quality review committee developed and implemented a thoracic ERAS protocol for anatomic lung resections across the healthcare system on May 1, 2021. The protocol was developed during the prior 13 month period using evidence-based tenants published by the ERAS Society/ESTS10 and implemented using structured dissemination and implementation (D&I) techniques12–14. We have previously described in detail both the initial qualitative findings of the D&I process15 and the methods for improving and ensuring minimum ERAS protocol compliance11. The final ERAS protocol implemented throughout the healthcare system can be found in Supplement 1. Quality control for protocol adherence was performed at monthly hospital-level review meetings and quarterly healthcare system-wide review meetings. Protocol items with less than 85% adherence were scrutinized for root cause and action items for improving adherence were planned for implementation by local implementation champions during the review meetings.
Data Extraction
Patients were identified using an automated pull of data from the electronic health record (EHR) where all patients with “segmentectomy,” “lobectomy,” “pneumonectomy,” or “lingulectomy” present in the “log of procedures descriptive list” column were targeted for inclusion. These patients were cross-referenced with billing data to ensure data accuracy. Other data included in this EHR inquiry were unique patient identifiers, operation hospital, operative surgeon, patient demographics and medical comorbidities, medications administered, chest tube-related data, basic ERAS compliance metrics, and intraoperative temperatures. These data were entered directly by clinicians and stored in the EHR, where patient identifiers were used to link individual encounters so that individual outcomes can be tracked. Institutional teams submit data to Vizient, which then packages data into a clinical database to generate postoperative outcomes data, which included complications identified using groups of ICD-10 diagnosis codes, hospital admissions data such as length of stay and intensive care use, and admission related cost information calculated through standardized methods in Vizient. For the purposes of this study, Vizient patient outcomes were checked at random by clinicians for quality control.
Study Design and Statistical Analysis
Patients were grouped into a pre-ERAS cohort, defined by the dates January 1, 2019 – April 30, 2021, and a post-ERAS cohort, defined by the dates May 1, 2021 – March 31, 2023. Patient demographics, medical comorbidities, operative data, admission data, some compliance data, and rates of 30-day postoperative outcomes were tracked for the two cohorts. Compliance items tracked included use of preoperative antibiotic skin prophylaxis, placement of urinary catheter intraoperatively, placement of chest tubes that were no larger than 28 French, multimodal pain regimen use postoperatively, and intensive care use postoperatively. Patient socioeconomic status was compared using the Social Vulnerability Index (SVI) (source), a unique and quantifiable measure of patient sociodemographic status measured from 0 (not vulnerable) to 100 (highly vulnerable) that has been validated extensively in surgical populations16,17, including patients undergoing lung resection18. The primary outcome for the study was any morbidity, which was defined as the occurrence of at least one postoperative complication. Secondary outcomes were individual rates of 30-day complications including prolonged air leak of more than five days postoperatively, chest tube reinsertion, postoperative pneumonia, urinary tract infection (UTI), surgical site infection (SSI), deep venous thrombosis/pulmonary embolism (DVT/PE), bleeding complication (defined as transfusion of red blood cell products), cardiac failure/arrest, cardiac arrhythmias, any infectious complication (defined by the occurrence of pneumonia, surgical site infection, or sepsis), any respiratory complication (defined as the occurrence of prolonged air leak, pneumonia, respiratory failure, or hemothorax), unplanned conversion to open surgery, postoperative ambulation on the day of surgery, intensive care unit (ICU) use, nonhome discharge, postoperative length of stay (LOS), opioid use measured in morphine equivalent daily dose (MEDD), chest tube (CT) duration, and total hospitalization cost. Bivariable comparisons of outcomes for the two groups were compared using unpaired t-tests for continuous variables, and chi-square or Fisher’s exact test for categorical variables. A multivariable regression analysis controlling for patient age, sex, race/ethnicity, body mass index (BMI), American Society of Anesthesiologists physical status classification (ASA class), operation performed, operative approach, procedure duration, presence of chest tube air leak in PACU, and the presence major comorbidities including diabetes mellitus, chronic obstructive pulmonary disease, congestive heart failure, acute renal failure, chronic kidney disease, and preoperative malnutrition was then performed to generate risk-adjusted odds of complications. A false discovery rate (FDR) adjustment was applied to the models to account for the comparison of multiple postoperative outcomes. We also performed subgroup analyses comparing outcomes of both low (<25%ile) and high SVI (>75%ile) patients and at the individual healthcare regions. Outcomes are reported as odds ratios (OR) and 95% confidence intervals (95%CI) for categorical variables and incidence rate ratios (IRR) and 95%CI for continuous variables. Two-sided p-values ≤0.05 were considered statistically significant. Statistical analyses were performed using SAS version 9.4 (SAS Inc, Cary, NC).
RESULTS
A total of 1,115 patients underwent anatomic lung resection during the study period. Of these, 65 patients (5.8%) were excluded for missing data and 43 patients (3.9%) were excluded for undergoing nonelective operations, leaving a total of 1,007 patients (90.3%) in the analytic cohort. Of these, 557 patients (55.3%) were in the pre-ERAS cohort and 450 patients (44.7%) were in the post-ERAS cohort. The majority of patients underwent surgery at UCHealth Central Region (n=670, 66.5%); there were 220 operations (21.9%) performed at UCHealth North and 117 operations (11.6%) performed at UCHealth South in the cohort. Table 1 shows the demographic information and medical comorbidities in the patient cohort. Mean patient age was 65.9 years (standard deviation 12.1 years); most patients were female (63.8%), identified as white (84.8%), had a BMI between 18.5–29.9 (70.8%), and were ASA class 3 (76.4%). Presence of medical comorbidities ranged from 0.8% (acute renal failure) to 9.8% (diabetes mellitus). Patients in the post-ERAS cohort were sicker by preoperative ASA class (p=0.004) and had higher rates of diabetes mellitus (12.9% vs. 7.4%, p=0.003), chronic kidney disease (3.1% vs. 0.9%, p=0.02), congestive heart failure (4.4% vs. 2.0%, p=0.02) and malnutrition (3.8% vs. 1.6%, p=0.03).
Table 1.
Demographic information and medical comorbidities of the patient cohort stratified into pre- and post-ERAS implementation.
| Demographics and Comorbidities | Overall (n=1,007) | Pre-ERAS (n=557) | Post-ERAS (n=450) | p-value | |
|---|---|---|---|---|---|
| n (%) | n (%) | n (%) | |||
| Age, mean [SD] | 65.9 (12.1) | 65.7 (11.7) | 66.3 (12.4) | 0.42 | |
| Sex | 0.91 | ||||
| Female | 642 (63.8) | 356 (63.9) | 286 (63.6) | ||
| Male | 365 (36.3) | 201 (36.1) | 164 (36.4) | ||
| Race/Ethnicity | 0.64 | ||||
| Asian | 63 (6.3) | 35 (6.3) | 28 (6.2) | ||
| Black or African American | 21 (2.1) | 10 (1.8) | 11 (2.4) | ||
| Hispanic | 52 (5.2) | 24 (4.3) | 28 (6.2) | ||
| Multiracial | 17 (1.7) | 10 (1.8) | 7 (1.6) | ||
| White or Caucasian | 854 (84.8) | 478 (85.8) | 376 (83.6) | ||
| BMI | 0.26 | ||||
| <18.5 | 55 (5.5) | 30 (5.4) | 25 (5.6) | ||
| 18.5–24.9 | 404 (40.1) | 235 (42.2) | 169 (37.6) | ||
| 25.0–29.9 | 309 (30.7) | 172 (30.9) | 137 (30.4) | ||
| 30.0–34.9 | 162 (16.1) | 87 (15.6) | 75 (16.7) | ||
| 35.0–39.9 | 53 (5.3) | 22 (4.0) | 31 (6.9) | ||
| 40+ | 24 (2.4) | 11 (2.0) | 13 (2.9) | ||
| ASA Class | 0.004 | ||||
| 1 | 3 (0.3) | 2 (0.4) | 1 (0.2) | ||
| 2 | 193 (19.2) | 129 (23.2) | 64 (14.2) | ||
| 3 | 769 (76.4) | 403 (72.4) | 366 (81.3) | ||
| 4 | 42 (4.2) | 23 (4.1) | 19 (4.2) | ||
| Social Vulnerability Index, mean (SD) | 36.8 (28.9) | 37.6 (28.9) | 35.7 (28.7) | 0.30 | |
| Diabetes | 99 (9.8) | 41 (7.4) | 58 (12.9) | 0.003 | |
| Chronic Kidney Disease | 19 (1.9) | 5 (0.9) | 14 (3.1) | 0.02 | |
| Acute Renal Failure | 8 (0.8) | 3 (0.5) | 5 (1.1) | 0.48 | |
| Congestive Heart Failure | 31 (3.1) | 11 (2.0)) | 20 (4.4) | 0.02 | |
| Chronic Obstructive Pulmonary Disease | 30 (3.0) | 15 (2.7) | 15 (3.3) | 0.55 | |
| Malnutrition | 26 (2.6) | 9 (1.6) | 17 (3.8) | 0.03 | |
Note: Percentages are expressed as column percentages. Abbreviations: ASA Class, American Society of Anesthesiologists physical status classification BMI, body mass index; ERAS, Enhanced Recovery After Surgery; SD, standard deviation.
Table 2 shows the operative characteristics amongst patients included in the cohort. Most patients underwent a robotic assisted (49.4%) or thoracoscopic (39.0%) approach and received a lobectomy (87.7%). The most common indication for operation was for resection of a primary non-small cell lung cancer (63.9%). Median time in the operating room was 193.9 minutes, and about half of patients (n=523, 51.9%) had a chest tube air leak immediately postoperatively. Patients in the post-ERAS cohort more frequently had open operations and less frequently thoracoscopic approach (p=0.008) and had a slightly lower rate of observed postoperative chest tube air leak (48.0 vs 55.1%, p=0.02) than the pre-ERAS cohort.
Table 2.
Operative characteristics of the patient cohort stratified into pre- and post-ERAS implementation.
| Operative Characteristics | Overall (n=1,007) | Pre-ERAS (n=557) | Post-ERAS (n=450) | p-value | |
|---|---|---|---|---|---|
| n (%) | n (%) | n (%) | |||
| Operative Approach | 0.008 | ||||
| Open | 117 (11.6) | 54 (9.7) | 63 (14.0) | ||
| Robotic-Assisted | 497 (49.4) | 264 (47.4) | 233 (51.8) | ||
| Thoracoscopic | 393 (39.0) | 239 (42.9) | 154 (34.2) | ||
| Operation Performed | 0.34 | ||||
| Lobectomy | 883 (87.7) | 481 (86.4) | 402 (89.3) | ||
| Segmentectomy | 77 (7.7) | 48 (8.6) | 29 (6.4) | ||
| Pneumonectomy | 47 (4.7) | 28 (5.0) | 19 (4.2) | ||
| Surgical Indication | 0.40 | ||||
| NTM | 152 (15.1) | 90 (16.2) | 62 (13.8) | ||
| Non-NTM Infection | 15 (1.5) | 9 (1.6) | 6 (1.3) | ||
| Primary non-lung metastasis | 53 (5.3) | 33 (5.9) | 20 (4.4) | ||
| Primary non-small cell lung cancer | 643 (63.9) | 355 (63.7) | 288 (64.0) | ||
| Carcinoid/neuroendocrine tumor | 63 (62.6) | 28 (5.0) | 35 (7.8) | ||
| Structural Pathology | 22 (2.2) | 11 (2.0) | 11 (2.4) | ||
| Solitary Pulmonary Nodule | 59 (58.9) | 31 (5.6) | 28 (6.2) | ||
| Time in operating room, median [IQR] | 193.9 (74.4) | 192.0 (72.2) | 196.3 (77.1) | 0.36 | |
| Air Leak Postoperatively | 523 (51.9) | 307 (55.1) | 216 (48.0) | 0.02 | |
Note: Percentages are expressed as column percentages. Abbreviations: ERAS, Enhanced Recovery After Surgery; IQR, interquartile range; NTM, non-tuberculosis Mycobacteria.
Table 3 shows five select compliance metrics before and after implementation of the ERAS protocol. Compliance after ERAS implementation was significantly increased in areas of low compliance in the pre-ERAS cohort and remained high in the areas of high compliance in the pre-ERAS cohort. Overall compliance was also significantly improved within these five compliance metrics.
Table 3.
Difference in pre- and post-implementation protocol compliance in some select protocol items.
| Compliance Measure | Pre-ERAS (n=557) | Post-ERAS (n=450) | p-value |
|---|---|---|---|
| n (%) | n (%) | ||
| Antibiotic skin prophylaxis | 535 (96.1) | 434 (96.4) | 0.74 |
| No urinary catheter placed | 288 (51.7) | 336 (74.7) | <0.0001 |
| Small chest tube utilized | 540 (96.9) | 447 (99.3) | 0.002 |
| No intensive care use | 469 (84.2) | 396 (88.0) | 0.09 |
| Multimodal pain regimen | 159 (28.5) | 439 (97.6) | <0.0001 |
| Overall | 1991 (71.5) | 2,052 (91.2) | <0.0001 |
Figure 1 shows the differences in adjusted complication rates compared between the pre- and post-implementation cohorts, Figure 2 shows the risk-adjusted differences in complications, and Table 4 provides both complication values. There was an unadjusted 7.0% absolute risk reduction in occurrence of any morbidity after ERAS implementation, with a risk-adjusted OR of 0.65 and 95%CI of 0.46–0.90 in the post-implementation cohort. There were also significant reductions in any respiratory complication (OR 0.58, 95%CI 0.39–0.85), postoperative pneumonia (OR 0.22, 95%CI 0.08–0.61), SSI (OR 0.05, 95%CI 0.01–0.57), rate of cardiac arrhythmias (OR 0.51, 95%CI 0.30–0.86), any infectious complication (OR 0.21, 95%CI 0.08–0.52), and ICU utilization (OR 0.34, 95%CI 0.19–0.61). After controlling for confounders, inpatient opioid use was reduced to 0.50 of pre-ERAS use (95%CI 0.44–0.57) and LOS was reduced by 0.85 days (95%CI 0.78–0.92). There was no difference between the two groups in rates of prolonged air leak (p=0.41), chest tube reinsertion rate (0.41), UTI (p=0.31), DVT/PE (p=0.07), postoperative bleeding (p=0.57), unplanned conversion to open surgery (p=0.24), postoperative ambulation of the day of surgery (p=0.28), nonhome discharge (p=0.82), chest tube duration (p=0.42), and total hospitalization cost (p=0.31).
Figure 1.
Differences in unadjusted complication rates after anatomic lung resection before and after ERAS implementation (*=statistically significant).
Figure 2.
Risk-adjusted odds ratios and incidence rate ratios for postoperative complications after ERAS implementation.
Table 4.
Unadjusted rates and risk-adjusted odds of postoperative outcomes stratified into pre- and post-ERAS implementation.
| Outcome | Pre-ERAS (n=557) | Post-ERAS (n=450) | OR/IRR (95% CI) | p-value* |
|---|---|---|---|---|
| n (%) | n (%) | |||
| Any morbidity | 144 (25.9) | 85 (18.9) | 0.54 (0.38–0.76) | 0.002 |
| Any respiratory complication | 109 (19.6) | 65 (14.4) | 0.58 (0.39–0.85) | 0.01 |
| Prolonged air leak | 71 (12.8) | 49 (10.9) | 0.80 (0.51–1.26) | 0.41 |
| Chest tube reinsertion | 22 (4.0) | 13 (2.9) | 0.70 (0.32–1.42) | 0.41 |
| Pneumonia | 24 (4.3) | 5 (1.1) | 0.22 (0.08–0.61) | 0.01 |
| UTI | 4 (0.7) | 2 (0.4) | 0.21 (0.02–2.46) | 0.31 |
| SSI | 9 (1.6) | 1 (0.2) | 0.05 (0.01–0.57) | 0.04 |
| DVT/PE | 8 (1.4) | 1 (0.2) | 0.04 (0.01–1.15) | 0.07 |
| Stroke | 2 (0.4) | 0 (0.0) | n/a | 1.00 |
| Bleeding | 8 (1.4) | 9 (2.0) | 1.38 (0.49–3.92) | 0.57 |
| Cardiac failure/arrest | 2 (0.4) | 2 (0.4) | 0.18 (0.01–5.80) | 0.41 |
| Cardiac arrhythmias | 53 (9.5) | 26 (5.8) | 0.51 (0.30–0.86) | 0.03 |
| Infectious complications | 30 (5.4) | 7 (1.6) | 0.21 (0.08–0.52) | 0.003 |
| Unplanned conversion to open | 13 (2.3) | 5 (1.1) | 0.41 (0.13–1.31) | 0.24 |
| Ambulation day of surgery | 215 (38.6) | 192 (42.7) | 1.21 (0.92–1.59) | 0.28 |
| ICU use | 88 (15.8) | 54 (12.0) | 0.34 (0.19–0.61) | 0.002 |
| Nonhome discharge | 15 (2.7) | 11 (2.4) | 1.05 (0.68–1.64) | 0.82 |
| MEDD, mean (SD) | 323.5 (912.0) | 141.5 (120.6) | 0.50 (0.44–0.57) | <0.0001 |
| LOS in days, median [IQR] | 3.0 [2.0–6.0] | 3.0 [2.0–5.0] | 0.85 (0.78–0.92) | 0.001 |
| Chest tube duration, median [IQR] | 2.0 [1.0–4.0] | 2.0 [1.0–4.0] | 0.96 (0.88–1.05) | 0.42 |
| Total cost in US dollars, median [IQR] | 18471 [14636–24388] | 19607 [15965–23906] | 0.97 (0.92–1.02) | 0.31 |
Note: FDR adjustment applied to p-value. Abbreviations: DVT/PE, deep venous thrombosis/pulmonary embolism; ERAS, Enhanced Recovery After Surgery; OR, odds ratio; ICU, intensive care unit; IRR, incidence rate ratio; IQR, interquartile range; LOS, length of stay; MEDD, morphine equivalent daily dose; SD, standard deviation; SSI, surgical site infection; US, United States; UTI, urinary tract infection.
Table 5 shows the subgroup analysis of differences in postoperative outcomes at the individual healthcare regions, and supplemental Table 1 shows the subgroup analysis displaying differences in post-ERAS changes amongst high SVI patients versus low SVI patients. There were statistically significant reductions in opioid use and absolute reduction in complication rates throughout each region and for each level of social vulnerability. Patients who were highly socially vulnerable had a greater absolute reduction in complications than those were less vulnerable. However, for both the regional and sociodemographic analysis, many of these differences did not reach statistical significance due to insufficient power in subset analysis.
Table 5.
Subgroup analysis of pre- and post-ERAS differences in outcomes at the individual healthcare regions.
| Outcome | Region 1 | Region 2 | Region 3 | ||||||
|---|---|---|---|---|---|---|---|---|---|
| Pre-ERAS (n=112) | Post-ERAS (n=108) | p-value | Pre-ERAS (n=363) | Post-ERAS (n=307) | p-value | Pre-ERAS (n=82) | Post-ERAS (n=35) | p-value | |
| n (%) | n (%) | n (%) | n (%) | n (%) | n (%) | ||||
| Overall | 26 (23.2) | 13 (12.0) | 0.03 | 105 (28.9) | 72 (23.5) | 0.11 | 13 (15.9) | 0 (0.0) | 0.006 |
| Respiratory | 19 (16.7) | 7 (6.5) | 0.02 | 82 (22.6) | 58 (18.9) | 0.24 | 8 (9.8) | 0 (0.0) | 0.08 |
| MEDD | 102 [61–161] | 76 [44–125] | 0.02 | 210 [68–425] | 89 [49–119] | <0.0001 | 106 [65–195] | 45 [27–73] | <0.0001 |
| LOS in days | 3 [2.5–5] | 3 [2–4] | 0.0005 | 4 [3–6] | 4 [3–6] | 0.41 | 2 [1–3] | 2 [1–2] | 0.06 |
| CT duration | 2 [2–4] | 2 [2–3] | 0.04 | 3 [2–4] | 3 [1–4] | 0.89 | 1 [1–3] | 1 [1–2] | 0.20 |
Definitions: ERAS, Enhanced Recovery After Surgery; LOS, length of stay; MEDD, morphine equivalent daily dose, CT, chest tube. Note: MEDD, LOS, and CT duration expressed in median with interquartile ran
DISCUSSION
The results of our prospective study of more than one thousand anatomic lung resection patients demonstrate that a thoracic ERAS protocol can be universally implemented across a diverse multi-hospital healthcare system despite differences in the local cultures, referral patterns, and settings of individual hospitals. Implementation of a thoracic ERAS protocol based on the guidelines published by the ERAS Society/ESTS resulted in improved postoperative outcomes including decreasing overall morbidity, any respiratory complication, postoperative pneumonia, SSI, infectious complications and cardiac arrythmias. The authors also believe this is the largest study on thoracic ERAS implementation and the first true pilot study that validates the recommendations made in the ERAS Society/ESTS guidelines.
This study builds on existing literature describing the success of a thoracic ERAS program in improving postoperative outcomes. Prior to this, studies have utilized varying surgical populations, specifically focusing on one approach or one extent of resection, and have had selective utilization of specific ERAS intervention components19. Additionally, many existing studies are limited in their analysis because they do not contain a comparison group, report unadjusted outcomes or were from centers outside the United States. One commendable study from the University of California at San Francisco demonstrated improved postoperative outcomes, decreased length of stay, and lower hospitalization cost after implementing an ERAS protocol with similar tenants as those published by the ERAS Society20. Their impressive results were obtained in a cohort of 295 propensity matched patients using an ERAS protocol that was implemented over a year before the ERAS Society/ESTS guidelines were published, so it was not a true pilot of these guidelines. However, given the similarities of their program with the ERAS Society/ESTS guidelines and the results of our study, there is enough evidence to confirm that the published guidelines for the care of lung resection patients lead to improved postoperative outcomes.
While our study observed no difference in operation-related outcomes such as rates of prolonged air leak, chest tube reinsertion, postoperative bleeding or unplanned conversion to open surgery, we saw distinct individual reductions in rates of postoperative pneumonia, respiratory complications, SSI, infections and cardiac arrythmias. This indicates that the benefits observed in the patient population were likely related to preoperative and postoperative care bundles rather than other potential confounders, like changes in operative technique over time. While patients in the post-ERAS cohort in general had more comorbidities, more open surgeries, and were sicker by ASA class, unadjusted postoperative outcomes were still improved. Additionally, patients benefitted from ERAS implementation regardless of their sociodemographic background. Our data continues to support the notion that ERAS protocols improve postoperative outcomes through the mitigation of the surgical stress response, a speculation backed by several studies examining the association of ERAS compliance with a downregulation in well-known stress markers21,22.
We were surprised to note that despite decreased postoperative morbidity, intensive care utilization and overall length of stay, the overall hospital cost was not reduced after ERAS protocol implementation. In this sense, our results are different from the well demonstrated cost-savings reported in the literature4,23. However, our cost-effectiveness analysis is unique in that many of post-ERAS patients underwent surgery during or after the COVID-19 pandemic. During the pandemic, there was a significant increase in healthcare operating expenses due to the growing volume of COVID-19 patients overwhelming the healthcare system and the subsequent resumption of deferred health services24. Synergistic with the effects of the COVID-19 pandemic, healthcare costs continued to inflate with labor costs growing 25% between 2019 and 2022, closely followed by pharmaceuticals growth of 21% and supplies at 18%25.
Finally, to the authors’ knowledge, this is the first study to demonstrate that a thoracic ERAS protocol can be implemented effectively across a diverse healthcare system. Of particular importance, we showed that patient outcomes in each of the three healthcare regions benefitted from ERAS implementation, rather than one region driving the benefits observed without any other region benefitting from implementation. Prior critique of the dissemination and implementation of ERAS programs has centered on the lack of generalizability, as most data is derived from single institutional studies19,26. Variation in performance amongst clinicians and facilities is well documented in the medical literature27, and cultural and organization-specific barriers may hinder the adoption of ERAS protocols into daily practice due to the necessary logistical and structural adjustments required28. We have described in detail the experience in dissemination and implementation of the program across a diverse healthcare system, which is comprised of academic, academic-affiliated and private practice facilities with different hospital resources and culture15. Despite the inherent challenges, this study showed that simultaneous implementation across large healthcare systems can be effective if buy-in is achieved from key players throughout the healthcare system. This information is important to report as large healthcare systems become more and more common in healthcare.
This study has several important limitations to consider. First, there was no analysis of how individual adherence affected outcomes, which is common among prior ERAS literature. However, a large meta-analysis of ERAS protocols across various specialties was not able to identify any significantly beneficial individual compliance metric29, which ultimately suggests a synergistic effect of all protocol components. Secondly, our cohort excluded wedge resections and was dominated by lobectomies, so there may be more or less impact on improvement for pneumonectomies or segmentectomies. However, the number of patients who received these operations was too small to power subgroup analyses to detect important differences. The patient population in this cohort had a lower proportion of patients who were racial or ethnic minorities and overweight or obese than what would be expected based on national demographics, aligned with the predominantly Colorado-based patient population. This limits the generalizability of the study and our ability to analyze the effects of implementation on highly vulnerable patients for more discrete outcomes. However, our subgroup analysis suggests that even the most sociodemographically vulnerable patients benefit from ERAS implementation. We additionally saw a difference in ASA class between pre- and post-ERAS cohorts. Given the subjective nature of the classification system it is important to recognize that under-scoring of the pre-ERAS cohort versus over-scoring of the post-ERAS cohort may have occurred, as practices and scoring personnel (typically, anesthesiologists) may have changed over time. Our cost analysis used Vizient data, which only provides an overview of total and direct costs. A more in-depth cost analysis may have helped determine where additional cost savings could have been achieved. Finally, our implementation was concurrent with the COVID-19 pandemic. Given the disruption of the national healthcare landscape during this time30, especially for lung cancer, it is unclear how this timing affected outcomes in this study.
In conclusion, an ERAS protocol based on the ERAS Society/ESTS guidelines was implemented across a large, diverse healthcare system, and its implementation resulted in decreased rates of postoperative complications, shorter inpatient length of stay, and fewer ICU admissions throughout the healthcare system. This study not only validates the comprehensive recommendations made by the ERAS Society/ESTS regarding the perioperative care of patients undergoing lung resection but shows that simultaneous enterprise implementation of perioperative protocols is feasible and can be effective. The strategies used during the implementation process can serve as the blueprint for other institutions wishing to implement ERAS or other perioperative care protocols.
Supplementary Material
ACKNOWLEDGMENTS
Funding/Support:
This work was supported by an internal grant from the Department of Surgery, University of Colorado School of Medicine and supported by the National Institutes of Health, under Ruth L. Kirschstein National Research Service Award T32CA17468. The funding organizations had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication. Drs. Dyas, Stuart and Meguid had full data access to all the data in the study and take full responsibility for its integrity and the accuracy of analysis.
Footnotes
Credit Author Statement
Authors Dyas, Stuart, Kelleher, and Meguid conceived the study and design. Authors Dyas, Erickson, Blum, Barker, Funk, Sack, and Meguid contributed to project implementation. Authors Bata, Cumbler, and Vizena contributed to data collection, and author Bronsert performed the formal analysis. Authors Dyas, Stuart, and Meguid interpretated the data. Authors Abrams, Randhawa, David, Mitchell, Weyant, Scott, and Meguid provided resources and supervision. Dyas wrote the original draft. All authors contributed to critical revisions.
Conflict of Interest Disclosures:
The authors report no conflicts of interest.
REFERENCES
- 1.Melnyk M, Casey RG, Black P, Koupparis AJ. Enhanced recovery after surgery (ERAS) protocols: Time to change practice? Can Urol Assoc J. 2011;5(5):342–348. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Eskicioglu C, Forbes SS, Aarts MA, Okrainec A, McLeod RS. Enhanced recovery after surgery (ERAS) programs for patients having colorectal surgery: a meta-analysis of randomized trials. J Gastrointest Surg. 2009;13(12):2321–2329. [DOI] [PubMed] [Google Scholar]
- 3.Lassen K, Soop M, Nygren J, et al. Consensus review of optimal perioperative care in colorectal surgery: Enhanced Recovery After Surgery (ERAS) Group recommendations. Arch Surg. 2009;144(10):961–969. [DOI] [PubMed] [Google Scholar]
- 4.Noba L, Rodgers S, Chandler C, Balfour A, Hariharan D, Yip VS. Enhanced Recovery After Surgery (ERAS) Reduces Hospital Costs and Improve Clinical Outcomes in Liver Surgery: a Systematic Review and Meta-Analysis. J Gastrointest Surg. 2020;24(4):918–932. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Ripolles-Melchor J, Abad-Motos A, Diez-Remesal Y, et al. Association Between Use of Enhanced Recovery After Surgery Protocol and Postoperative Complications in Total Hip and Knee Arthroplasty in the Postoperative Outcomes Within Enhanced Recovery After Surgery Protocol in Elective Total Hip and Knee Arthroplasty Study (POWER2). JAMA Surg. 2020;155(4):e196024. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Murr AT, Sweeney C, Lenze NR, Farquhar DR, Hackman TG. Implementation and Outcomes of ERAS Protocol for Major Oncologic Head and Neck Surgery. Laryngoscope. 2023. [DOI] [PubMed] [Google Scholar]
- 7.Brunelli A, Thomas C, Dinesh P, Lumb A. Enhanced recovery pathway versus standard care in patients undergoing video-assisted thoracoscopic lobectomy. J Thorac Cardiovasc Surg. 2017;154(6):2084–2090. [DOI] [PubMed] [Google Scholar]
- 8.Dinic VD, Stojanovic MD, Markovic D, Cvetanovic V, Vukovic AZ, Jankovic RJ. Enhanced Recovery in Thoracic Surgery: A Review. Front Med (Lausanne). 2018;5:14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Semenkovich TR, Hudson JL, Subramanian M, Kozower BD. Enhanced Recovery After Surgery (ERAS) in Thoracic Surgery. Semin Thorac Cardiovasc Surg. 2018;30(3):342–349. [DOI] [PubMed] [Google Scholar]
- 10.Batchelor TJP, Rasburn NJ, Abdelnour-Berchtold E, et al. Guidelines for enhanced recovery after lung surgery: recommendations of the Enhanced Recovery After Surgery (ERAS(R)) Society and the European Society of Thoracic Surgeons (ESTS). Eur J Cardiothorac Surg. 2019;55(1):91–115. [DOI] [PubMed] [Google Scholar]
- 11.Dyas AR, Kelleher AD, Cumbler EU, et al. Quality Review Committee Audit Improves Thoracic Enhanced Recovery After Surgery Protocol Compliance. Accepted for publication. Journal of Surgery Research, September/2023. [DOI] [PubMed] [Google Scholar]
- 12.Bauer MS, Damschroder L, Hagedorn H, Smith J, Kilbourne AM. An introduction to implementation science for the non-specialist. BMC Psychol. 2015;3:32. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Albright K, Gechter K, Kempe A. Importance of mixed methods in pragmatic trials and dissemination and implementation research. Acad Pediatr. 2013;13(5):400–407. [DOI] [PubMed] [Google Scholar]
- 14.Powell BJ, Waltz TJ, Chinman MJ, et al. A refined compilation of implementation strategies: results from the Expert Recommendations for Implementing Change (ERIC) project. Implement Sci. 2015;10:21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Dyas AR, Kelleher AD, Erickson CJ, et al. Development of a universal thoracic enhanced recover after surgery protocol for implementation across a diverse multi-hospital health system. J Thorac Dis. 2022;14(8):2855–2863. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Diaz A, Hyer JM, Barmash E, Azap R, Paredes AZ, Pawlik TM. County-Level Social Vulnerability is Associated with Worse Surgical Outcomes Especially Among Minority Patients. Ann Surg. 2020;Publish Ahead of Print. [DOI] [PubMed] [Google Scholar]
- 17.Dyas AR, Carmichael H, Bronsert MR, et al. Social vulnerability is associated with higher risk-adjusted rates of postoperative complications in a broad surgical population. Am J Surg. 2023. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Stuart CM, Dyas AR, Byers S, et al. Social vulnerability is associated with post-operative morbidity following robotic-assisted lung resection. J Thorac Dis. 2023;15(11):5931–5941. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Khoury AL, McGinigle KL, Freeman NL, et al. Enhanced recovery after thoracic surgery: Systematic review and meta-analysis. JTCVS Open. 2021;7:370–391. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Haro GJ, Sheu B, Marcus SG, et al. Perioperative Lung Resection Outcomes After Implementation of a Multidisciplinary, Evidence-based Thoracic ERAS Program. Ann Surg. 2021;274(6):e1008–e1013. [DOI] [PubMed] [Google Scholar]
- 21.Crippa J, Mari GM, Miranda A, Costanzi AT, Maggioni D. Surgical Stress Response and Enhanced Recovery after Laparoscopic Surgery - A systematic review. Chirurgia (Bucur). 2018;113(4):455–463. [DOI] [PubMed] [Google Scholar]
- 22.Crippa J, Calini G, Santambrogio G, et al. ERAS Protocol Applied to Oncological Colorectal Mini-invasive Surgery Reduces the Surgical Stress Response and Improves Long-term Cancer-specific Survival. Surg Laparosc Endosc Percutan Tech. 2023;33(3):297–301. [DOI] [PubMed] [Google Scholar]
- 23.Thanh NX, Chuck AW, Wasylak T, et al. An economic evaluation of the Enhanced Recovery After Surgery (ERAS) multisite implementation program for colorectal surgery in Alberta. Can J Surg. 2016;59(6):415–421. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.McClellan M, Rajkumar R, Couch M, et al. Health Care Payers COVID-19 Impact Assessment: Lessons Learned and Compelling Needs. NAM Perspect. 2021;2021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Swanson Erik. National Hospital Flash Reports: January 2019 – June 2022, Kaufman, Hall & Associates, 2019–22. [Google Scholar]
- 26.Kehlet H, Wilmore DW. Evidence-based surgical care and the evolution of fast-track surgery. Ann Surg. 2008;248(2):189–198. [DOI] [PubMed] [Google Scholar]
- 27.Tsai TC, Joynt KE, Orav EJ, Gawande AA, Jha AK. Variation in surgical-readmission rates and quality of hospital care. N Engl J Med. 2013;369(12):1134–1142. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Lambert-Kerzner AC, Aasen DM, Overbey DM, et al. Use of the consolidated framework for implementation research to guide dissemination and implementation of new technologies in surgery. J Thorac Dis. 2019;11(Suppl 4):S487–S499. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Nicholson A, Lowe MC, Parker J, Lewis SR, Alderson P, Smith AF. Systematic review and meta-analysis of enhanced recovery programmes in surgical patients. Br J Surg. 2014;101(3):172–188. [DOI] [PubMed] [Google Scholar]
- 30.Dyas AR, Bronsert MR, Stuart CM, et al. Analyzing the impact of the Coronavirus disease 2019 pandemic on initial oncologic presentation and treatment of non-small cell lung cancer in the United States. J Thorac Cardiovasc Surg. 2023. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.