Does Enhanced Recovery After Surgery Protocols Reduce Complications and Length of Stay After Thoracic Surgery: A Systematic Review of the Literature

Enhanced recovery after surgery (ERAS) has an increasingly important role in the perioperative management of thoracic surgical patients. It has been extensively studied in multiple surgical specialties, particularly colorectal surgery, where ERAS protocols have been shown to reduce postoperative length of stay and postoperative complications. Electronic searches of two research databases were performed: PubMed (1972 to October 2023) and Ovid MEDLINE (1946 to October 2023). The literature search was completed on January 4, 2024. Search terms included: "thoracic surgery" and "ERAS" or "Enhanced Recovery After Surgery". The search was limited to studies evaluating humans undergoing thoracic surgery for any indication. The primary outcome was overall morbidity, with secondary outcomes including mortality, length of stay, and pulmonary complications. The search yielded a total of 794 records, of which 30 (four meta-analyses and 26 observational trials) met the relevant inclusion and exclusion criteria. This review suggested the implementation of ERAS protocols can lead to a reduction in postoperative morbidity; however, this was not a consistent finding. The majority of studies included demonstrated a reduction in the length of stay with the implementation of ERAS. Overall, ERAS/ERATS is an important adjunct to the management of patients requiring thoracic surgery, consistently leading to shorter lengths of stay and likely contributing to reduced rates of postoperative morbidity. Further research will be required to determine the impact of the recently released ERATS guidelines.


Introduction And Background
Enhanced recovery after surgery (ERAS) has an increasingly important role in the perioperative management of thoracic surgical patients.It has been extensively studied in multiple surgical specialities, particularly colorectal surgery, where ERAS protocols have been shown to reduce postoperative length of stay and postoperative complications [1][2][3].ERAS pathways focus on optimising patient outcomes by concentrating on three key domains: preoperative optimisation, perioperative management, and postoperative management [1].The principle is that the application of multiple, evidence-based interventions throughout the patient journey has a synergistic effect greater than the sum of all the parts.The application of ERAS pathways in thoracic surgery has been shown to reduce postoperative length of stay in most studies and potentially reduce postoperative complications [4,5].Furthermore, it has been demonstrated that improved compliance with each part of the ERAS protocol can lead to reduced postoperative morbidity [6].In 2019, the ERAS Society and the European Society of Thoracic Surgeons (ESTS) published guidelines for ERAS in thoracic surgery, known as Enhanced Recovery after Thoracic Surgery (ERATS) [1].These guidelines stipulated 45 ERAS outcomes across the three domains [1].
Since the publication of the 2019 guidelines, there has been limited research assessing the impact of ERATS programs on patient outcomes.The purpose of this review is to summarise the available literature and objectively appraise the influence of ERATS on patient outcomes.
(≥4 studies) the following: (1) preoperative: education and intensive pulmonary physiotherapy; (2) intraoperative: nil ≥ 4; (3) postoperative: epidural analgesia/NSAIDs, IV fluid restriction, early oral intake, and early ambulation The primary outcome measure was overall morbidity, reported in 7/7 studies.The pooled relative risk for this was 0.64 (95% CI: 0.51-0.80),with a decreased overall morbidity with ERAS.No difference in mortality (reported in 4/7 studies).Regarding LOS, unable to complete due to heterogeneity (I 2 =86.5%).Reported in 7/7 studies, with mean LOS significantly shortened in the ERAS cohort in 4 of these studies.3 studies showed no statistical difference.Reduced rates of pulmonary complications with a relative risk of 0.43 (95% CI 0.31-0.60).Overall complication rate was reported in 18 studies.
There was a reduced incidence of postoperative complications in the ERAS cohort with a pooled RR 0.64 (95% CI 0.52-0.78,p<0.001).There was significant heterogeneity (I 2 =63%).In-hospital mortality was reported in 13 studies, with all studies reporting no difference.The LOS was divided into overall LOS (11 studies)    No difference in mortality (only 1 mortality in this study).
There was no overall morbidity rate reported.LOS was reduced in both ERAS cohorts.In the thoracotomy cohort  Perioperative: no sedative premedications, minimised fasting time, multi-modal opioid-sparing analgesia, regional anaesthesia with intercostal blocks, anaesthetic protocols, euvolaemia.No comment on the postoperative pathway.
There was no difference in postoperative LOS between the two cohorts.There were no in-hospital mortalities.
There was no difference in the incidence of postoperative major complications: ERAS cohort 3% vs control cohort 4.7%, p=0.68.When stratified by approach, ERAS had no impact on postoperative LOS.
There was no comment on the overall complication rate.
On subgroup analysis, this difference in pulmonary and cardiac complications was only true for patients who underwent thoracotomy, and not MiV.

Review Results
The initial search yielded 794 records; after removal of duplicates, 374 records were suitable for screening.
The titles and abstracts of these articles were reviewed by two independent authors (JG and LB), with 323 records excluded as they did not fit the inclusion criteria.A total of 51 records were retrieved for review of the full text.After full-text review, a further 21 records were excluded for a final included total of 30 publications, of which four were meta-analyses and 26 were observational cohort studies [10][11][12][13][14][15][16].The PRISMA diagram is shown in Figure 1.

Extent and rigour of ERATS protocol implementation
The breadth of ERAS implementation and the degree of protocol compliance varies significantly among all the studies, including those evaluated within each meta-analysis.Table 2 depicts the components of the 2019 ERATS protocol that each of the included observational studies utilised, categorised under 21 major groups, noting that there are 45 individual components of the ERATS protocol [1].It is important to note that most of the studies either pre-date the release of the 2019 ERATS guidelines or include cohorts of patients which are pre-2019.The range of components implemented was 4-18 with a median of 10.5.The most frequently implemented sections of the protocol were pre-operative education and early mobilisation post-surgery, with 24 of the 26 studies implementing both strategies.The least implemented components were atrial fibrillation prevention (one of 26 studies) and documented alcohol dependency management plan (2 of 26 studies).

Meta-analyses
The four included meta-analyses comprised studies, which largely pre-dated the 2019 ERATS guidelines.Li et al. in 2017 reported that the range of ERAS protocol recommendations implemented was 5-11 in their included studies, with the most frequently implemented recommendations being preoperative education and early ambulation [5].Fiore et al. in 2016 described a total of 15 ERAS elements; within their reviewed literature, the number of elements implemented ranged from 4 to 10 with a median of 6.54.The most commonly implemented elements in their report were preoperative education, prophylactic antibiotics, standardised chest drain management, and early mobilisation [4].Li et al., 2021 reported a range of 0-22 elements with the most common being preoperative education, early ambulation, minimally invasive surgical technique, standardised chest drain management, and standardised anaesthetic protocols [7].Zhang et al., 2021 did not comment on the specifics of ERAS protocols used in each study [8].

Meta-Analyses
Li et al. published a meta-analysis of seven randomised controlled trials (RCTs) with a total of 486 patients in 2017 [5].The authors found that the rate of overall morbidity was reduced by the implementation of ERAS, with a relative risk (RR) of 0.64 (95% confidence interval (CI) 0.51-0.80,i²=1.9%)[5].This reduction was driven by a decrease in pulmonary complications (RR 0.43, 95% CI 0.31-0.60,i²=0%) [5].This finding was not replicated by the earlier systematic review by Fiore et al., which included one RCT and five non-randomised studies with a total of 1,612 patients; the authors found that the implementation of ERAS had no impact on the incidence of post-operative morbidity [4].
Li et al. published a further meta-analysis in 2021, including 21 articles (19 non-RCT and 2 RCT) with a total of 6,480 patients [7].The overall complication rate was reported in 18 of the 21 studies [7].The implementation of ERAS was associated with a reduced incidence in overall post-operative complications with a pooled relative risk of 0.64 (95% CI 0.52-0.78,p<0.001) [7].However, there was significant heterogeneity within the pooled study data, with an i² of 63% [7].
Zheng et al. published a meta-analysis of 23 studies (11 RCT and 12 non-RCT) in 2021, with a total of 8,094 patients included [8].The authors concluded that the introduction of ERAS was associated with a reduction in the incidence of postoperative complications (22 out of 23 studies reporting relevant data) with an odds ratio of 0.48 (95% CI 0.37-0.61,p<0.01) [8].Once again, however, they identified significant heterogeneity within the pooled data (i²=62%) [8].

Meta-Analyses
All four of the included meta-analyses commented on the impact of ERAS on LOS, with all demonstrating an overall reduction in LOS either based on pooled analysis or on qualitative grounds, according to the included studies.However, the quality of the pooled analyses was generally poor due to significant heterogeneity among the included studies.Li et al. did not perform a pooled analysis due to significant heterogeneity (I² = 86.5%);among their seven included studies, four demonstrated that ERAS led to reduced LOS, while three showed no difference [5].Fiore et al. found that LOS was significantly reduced in the included non-randomised trials, but not in the one included randomised controlled trial [4].
Li et al. divided their analysis for LOS into overall LOS (11 studies) and postoperative LOS (12 studies) [7].
Most publications showed that the implementation of ERAS was associated with reduced postoperative and overall LOS.Pooled analysis was performed for seven of the studies, with patients in the ERAS cohort having a significantly shorter LOS.This analysis had the important limitation of significant heterogeneity (I² = 98%).Zhang et al. reported a pooled analysis of 20 studies examining LOS, finding that LOS was reduced in the included ERAS patients by 2.7 days (p < 0.01).Again, this pooled analysis was subject to significant heterogeneity (I² = 97%) [7].

Observational Studies
Of the included 26 observational studies, 25 reported on the impact of ERAS on LOS [3,.Twenty of these studies found that ERAS was associated with reduced LOS after thoracic surgery [9-  [32].This included a pre-ERAS cohort (n = 1,615), a transitional period (n = 929), and the ERAS cohort (n = 342) [32].They reported a reduction in mean LOS from 5 days (pre-ERAS) to 4 days (transitional) to 4 days (ERAS, p < 0.001) [32].Wang et al. reported another large retrospective cohort study including 1,749 patients [16].This cohort comprised patients who underwent anatomical resection for non-small cell lung cancer with ERAS protocols (2016 to 2017) compared to those who had comparable surgery under the previous standard of care [16].They found the mean postoperative LOS was reduced from six days to four days following the introduction of ERAS (p < 0.001) [16].

Secondary outcome: mortality
Due to the low event rate for perioperative mortality after thoracic surgery and the small sample size of the majority of the included studies, it is difficult to conclude the impact of ERAS on in-hospital or 30-day mortality.

Meta-Analyses
All four meta-analyses commented on in-hospital or perioperative mortality, with no trial showing a difference between patients treated under ERAS and those without [4,5,7,8].Zhang et al. reported an overall low event rate, with an odds ratio of 1.15 (95% CI 0.6 to 2.22, p>0.05) [8].

Observational Trials
Of the 26 included studies, 14 reported on mortality [12,13,16,18,19,22,23,26,27,29,[30][31][32].The only study to report a significant difference in mortality was the study by Alwarti et al., a retrospective cohort study published in 2021 [13].This large study, which included 10,021 patients, encompassed all patients who underwent open lobectomy between 2005 and 2019, divided into three temporal cohorts as described previously [13].The overall mortality rate for the study was 2.2% [13].The authors reported a decreasing trend in mortality over time: from 2.8% in the pre-ERAS period to 2% during the transitional period, and finally to 1.9% in the ERAS cohort, with a p-value of 0.05 [13].While the introduction of ERAS may have contributed to this improvement in mortality, the analysis clearly contains significant confounding factors, with multiple other potential contributors to the downtrending mortality over a 15-year period.

Meta-Analyses
Of the four meta-analyses, three reported on the effect of ERAS protocols with respect to postoperative pulmonary complications [4,5,7].The subgroup analysis performed by Li et al. demonstrated that patients undergoing thoracic surgery under ERAS protocols had a reduced incidence of postoperative pulmonary complications, with a relative risk of 0.43 (95% CI 0.31-0.60)[5].In the study by Fiore et al., no pooled analysis was performed; however, the included RCT demonstrated a reduction in postoperative pulmonary complications [4].Finally, Li et al. also reported a reduction in postoperative pulmonary complications associated with ERAS, with a relative risk of 0.58 (95% CI 0.45-0.75,p<0.001) [7].

Discussion
ERAS, or more specifically ERATS (after the publication by the ERAS society in conjunction with ESTS in 2019), will play an increasingly important role in the care of patients undergoing thoracic surgery.It provides a framework for a standardised pathway of care, to streamline processes, and improve perioperative outcomes.This standardised approach to patient care is importantly multidisciplinary beginning preoperatively with adequate education and risk factor modification [3].
The most consistent finding from this systematic review was that the implementation of ERAS protocols for thoracic surgery is associated with reduced hospital length of stay.Most studies show that this reduction in length of stay is not accompanied by an increased rate of readmission.Not only is this important from a health economic perspective, but it is also vital from a patient recovery perspective [3].Expedited postoperative recovery, with a faster return to a 'new' baseline function, gives a patient the best possible chance to receive additional oncological treatment such as chemoimmunotherapy [33].Paci et al. published a study in 2017 looking at the economic impact of the implementation of ERAS for patients undergoing elective lung surgery [34].They found that patients undergoing surgery with ERAS protocols had a reduced length of stay (four vs six days, p < 0.01) with no increase in the rate of readmission [34].Interestingly, they also evaluated the caregiver burden on discharge, finding that patients who underwent their surgery with ERAS protocols had a trend toward a lower caregiver burden (53 +/-90 hours vs 101 +/-252 hours, p = 0.17) [34].Economic costs to the healthcare system were also reduced with ERAS, with a mean difference of -$4,396 Canadian dollars [34].
Three of the four meta-analyses included reported a reduction in postoperative morbidity with ERAS.This outcome was also mirrored in seven of the 26 observational trials.It is known that postoperative complications have significant deleterious effects on patients and the healthcare system.Shinohara et al. in 2019 published a retrospective study of 345 patients who underwent lobectomy for NSCLC [35].They found that patients who had postoperative complications, in particular pulmonary complications, had reduced five-year overall and disease-free survival [35].Postoperative complications can also delay the time to receiving adjuvant therapy or reduce the ability of patients to tolerate adjuvant therapy.Nelson et al. in 2019 demonstrated that the implementation of ERAS for patients with clinical stage I or II NSCLC without induction therapy resulted in not only a shorter time to commencing adjuvant therapy but importantly increased rates of receiving adjuvant therapy (40% vs 62%, p<0.001) [36].
A significant limitation of any systematic review is the quality of evidence provided by the included studies.Most of the evidence comes from meta-analyses of small randomised controlled trials and observational trials and from individual prospective and retrospective observational trials with varying sample sizes.Furthermore, there is significant heterogeneity in the design of ERAS protocols between the reported studies, and the compliance with each individual component of the protocol within each study.From Rogers et al., 2018, and Forster et al., 2020, improved compliance with ERAS protocols leads to improved outcomes with lower postoperative morbidity and shorter lengths of stay.There is not yet sufficient evidence with which to comment specifically on the impact of the recently released ERATS guidelines; this will be an important avenue for further research [1].One benefit of a more standardised approach to ERAS in thoracic surgical patients is greater comparability of findings between studies, in addition to greater transferability of results across patient populations and centres.

Conclusions
Overall, ERAS/ERATS is an important adjunct to the management of patients requiring thoracic surgery, consistently leading to shorter lengths of stay and likely contributing to reduced rates of postoperative morbidity.Further research will be required to determine the impact of the recently released ERATS guidelines as these are more comprehensive with forty-five total elements.Furthermore, it will be important to critically analyse each of the forty-five elements to ensure that they are effective without becoming cumbersome, thereby increasing uptake across the board.
The importance of ERAS/ERATS in shortening the length of stay and potentially lowering postoperative morbidity will continue to become apparent as patients become older, more comorbid, and healthcare resources become more constrained.By consistently reducing the length of stay without increasing the rate of readmission, the application of ERATS/ERAS may contribute to a more efficient use of scarce healthcare resources.Furthermore, by reducing postoperative morbidity, it may improve access to adjuvant oncoimmunotherapy, further improving overall-and disease-free survival.It will be vital, therefore, that thoracic surgery encompasses a multidisciplinary, guideline-based approach to streamline patient care as much as possible, with a view to optimising outcomes.
studies.Non-RCT all had a high risk of bias in the majority of domains of the Cochrane tool.Included a total of 1612 patients, with 821 in the ERAS cohort and 791 in the control. 2 studies only included patients undergoing lobectomy.The remaining 4 included any pulmonary resection.The proportion of VATS was higher in the ERAS cohort.Varied by study.Described a total of 15 elements, with the number used ranging from 4 to 10, a median of 6.5.The most frequently cited included: (1) preoperative: education (n=5) and prophylactic antibiotics (n=4); (2) intraoperative: epidural (n=4); (3) postoperative: standard chest tube management (n=5) and early mobilisation (n=4) Results included: (1) LOS: significantly reduced in the non-RCT (difference 1.2-9.1 days), with no difference in the RCT; (2) Readmission rates: Variably reported (3 non-RCT) with rates 1-10%.Note that 1 study spanned 7 days, 1 study spanned 30 days, and no time period was specified in 1 study; (3) Complications Reported in 3/6 studies.No difference in the overall complication rate.RCT showed a significant reduction in pulmonary complications with ERAS; (4) Mortality reported in 3/6 studies with no difference.Li et al., 2021 [7] Systematic review and metaanalysis included 21 articles in total: 19 cohort studies and 2 RCTs.The range was 2006 to 6,480 patients total, with 2,617 patients in ERAS cohorts and 3,863 patients in the control.The number utilised in the ERAS group ranged from 5 to 22 and that in the non-ERAS cohort ranged from 0 to 10.The most used included: preadmission education (19/21 studies), early ambulation (18/21 studies), VATS (12/21 studies), rational use of analgesics postop (17/21 studies),

4
days vs 5.5 days, p=0.01 and in VATS cohort 2 vs 3 days, p<0.01.Multivariate analysis found year of surgery and VATS were associated with a reduction in atelectasis requiring bronchoscopy (OR 0.83, p=0.048),AF (OR 0.74, p<0.001) and urinary retention (OR 0.84, p<0.001).Lee et al., 2021 [3] Prospective longitudinal study over 21 months 2017 to 2019.This spanned the introduction of the ERATS program.There were three periods: 9 months prior to ERATS, 3-month transition phase and 9 months post-ERATS.The post-ERATS cohort was further divided into 3, 3monthly cohorts.The 3-month transition phase was excluded from the study.All patients ambulation, early diet, standardised chest drain management.There was no change in the median LOS (3 days) across the study period.However, over the study period, there was a reduction in LOS for those undergoing VATS lobectomy (4.3 to 3.3 days, p=0.34) and VATS sublobar resection (3.4 to 2.2 days, p=0.13).The incidence of minor complications decreased from 18.2% in the pre-ERATS to 7.9% in the 7-9-month post-ERATS cohort (p=0.009).Major complications also decreased from 13.6% pre-ERATS to 4.4% in the 7-9-month post-ERATS cohort (p=0.007).Brunelli et Retrospective cohort study on prospectively collected data between 2014 and 2017.600 patients were included in total, with 365 patients pre-protocol was implemented in 2016 which formalised the processes.Previous processes continued.Preoperative: education, There was no difference in LOS in the two cohorts (pre-ERAS 5 days vs post-ERAS 4 days, p=0.44).There was no difference in in-hospital mortality between the two cohorts (pre-ERAS 2.2% vs post-ERAS 3.8%, p=0.31).2024 Goldblatt et al.Cureus 16(5): e59918.DOI 10.7759/cureus.: minimal fasting, normothermia.Postoperative: early mobilisation, early oral diet, PONV prevention, preventing fluid overload, multimodal analgesia.There was also no difference in 90-day mortality: pre-ERAS 3% vs post-ERAS 4.7%, p=0.34.No difference in the incidence of cardiovascular and pulmonary complications (pre-ERAS 22.6% vs post-ERAS 22.ERAS protocol developed from the best available evidence.Preoperative: MDT education, smoking cessation, nutrition management, prehabilitation, correction of anaemia.Perioperative: minimise fasting, minimise premedications, antibiotics, DVT prophylaxis, multi-modal analgesia, VATS, and avoid urinary catheters.Postoperative: PONV prophylaxis, multimodal analgesia, standard chest drain management, early mobilisation.LOS only outcome which was reduced in the ERAS cohort: 5.2 days vs 11.7 days, p<0.0001. ERAS protocols.Preoperative: education, smoking cessation, pulmonary rehabilitation, nutritional assessment Perioperative: antibiotics, analgesia with epidural OR intercostal, euvolaemia, VTE prophylaxis, PONV prevention Postoperative: early mobilisation, early diet, multi-modal analgesia.Group 3 (i.e. with ERAS) was associated with significantly shorter postoperative LOS: 5.5 vs 7 vs 8.5 days, p<0.001.Group 3 had a lower incidence of overall complications (no specific values, but p<0.001).developed protocol, based on ERAS principles.Preoperative: education.Perioperative: preference for epidural anaesthesia Postoperative: early mobilisation, standardised chest drain management.LOS was significantly reduced with ERAS: 6 vs 7 days (p<0.01).The incidence of overall morbidity was reduced in the ERAS cohort: 37% vs 50%, p=0.03.This difference however was driven by a reduction in urinary tract infections (3% vs 12%, p<0.01).There was a difference in the incidence of any pulmonary complications: 25% vs 31%, p=0.38.There was no difference in mortality, with only 1 mortality in the entire into three cohorts: pre-ERAS (2006-11), transitional period (2012-15) and post-ERAS (2015-2016).Included all patients undergoing anatomic and non-anatomic resection for primary lung cancer.USA 2,886 patients, with 1,615 in the pre-ERAS cohort, 929 in the transitional cohort and 342 in the ERAS cohort.Interestingly, there was an improvement in preoperative ECOG scores from pre-ERAS to transitional to ERAS.The proportion of MiV surgery also increased across the study period, with 32.9% VATS/RATS in the pre-ERAS cohort compared to 46.8% in the transitional compared to 51.1% in the ERAS cohort Standard of care included prophylactic antibiotics and VTE prophylaxis.ERAS protocol developed within the hospital.Preoperative: education.Perioperative: minimise fasting time, antibiotics, VTE prophylaxis, PONV prevention, minimise sedative pre-medications, standard anaesthetic protocol, euvolaemia Postoperative: early ambulation, early diet, standardised chest drain management LOS reduced across the time of the study: 5 days (pre-ERAS) vs 4 days (transitional) vs 4 days (ERAS), p<0.001.There was no difference in overall mortality

FIGURE 1 :
FIGURE 1: Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flowchart.

TABLE 2 : ERAS/ERATS Protocols Utilised in Each Study
[13]3-22,24-26,28- 30-32].The largest study included was by Alwarti et al., which encompassed a total of 10,021 patients[13].This study examined three epochs of time: pre-ERAS (2005 to 2011), a transition period (2012 to 2015), and post-ERAS (2016 to 2019)[13].Alwarti et al. found that the mean LOS progressively reduced from a pre-ERAS mean of 8.1 days to a transitional LOS of 7.1 days, down to an ERAS LOS of 6.6 days (p < 0.01)[13].Similar findings were reported by another large retrospective cohort published byVanHaren et al. in 2018; this study included 2,886 patients divided into three cohorts between 2006 and 2016