Enhanced Recovery After Surgery Protocol in Transoral Robotic Surgery: A Case–Control Study Authors: Jonathan Daniel, Raguwinder Bindy Sahota, Mark David Wilkie, Lee Pryor, Ahmed Bassiouni, Jake Jervis‐Bardy, Arun Takhar, Bora Jeong, Oladejo Olaleye, Suren Krishnan, J C Hodge, Andrew Foreman Published: 2026 ABSTRACT ABSTRACTIntroduction‘Enhanced Recovery After Surgery’ (ERAS) protocols have been developed for multiple surgical procedures with favourable outcome measures including Length of Stay (LoS). Trans‐Oral Robotic Surgery (TORS) has similarly facilitated earlier patient discharges, particularly in comparison to open head and neck surgical techniques. There is a dearth of literature on the effect of ERAS protocols in TORS. We sought to comprehensively examine the effect of a TORS ERAS protocol across several key metrics, with a focus on LoS.MethodsWe conducted a retrospective cohort study following implementation of a TORS ERAS protocol. This protocol was developed at our institution and implemented in 2021; it included key features such as pharyngocutaneous risk stratification and thorough perioperative assessments. Patients managed prior to (p‐ERAS) and post implementation (w‐ERAS) were separated into two groups, with key outcomes, including LoS, estimated cost per patient, return to swallow, and complication rate, compared.ResultsA total of 40 TORS cases were included. Patients managed w‐ERAS had a mean adherence rate to protocol of 90%. The w‐ERAS group showed a statistically significant reduction in median length of stay by 3.5 days (p = 0.03) when compared to the p‐ERAS group. There was also a trend towards decreased median cost per patient as well as decreased median time to oral intake.ConclusionOur study demonstrates an easily implementable protocol for the perioperative management of TORS oncological surgery. The use of this ERAS protocol demonstrated a significant reduction in patient length of stay in a small sample size.Keywords: enhanced recovery after surgery protocol, length of stay, return of swallow, surgery, surgical complications, trans oral robotic surgery (TORS) Introduction Length of stay (LoS) has increasingly become a key performance indicator of hospital efficiency and quality of patient care across all medical and surgical fields [1]. As such, a focus in the evolution and innovations within surgery has been to optimise and streamline the post‐operative patient pathway. “Enhanced Recovery After Surgery” (ERAS) protocols have been developed for multiple surgical procedures with favourable results across various outcome measures, including LoS [2, 3]. Within Head and Neck Surgery (HNS), newer minimally invasive approaches such as Trans‐Oral Robotic Surgery (TORS) have also served to facilitate earlier patient discharges when compared to open surgical techniques [4]. There is a dearth of literature when relating ERAS protocols to TORS. Existing studies primarily focus on post‐operative pain management with little emphasis on LoS. An evidence‐based TORS‐specific ERAS protocol was introduced at our institution in 2020; we sought to comprehensively examine the effect of this protocol across several key metrics, with a particular focus on LoS. Methods We completed a retrospective cohort review following the introduction and implementation of an ERAS protocol in TORS in our institution. Criteria for inclusion included adult patients undergoing TORS for p16‐positive oropharyngeal cancer. The study was undertaken in accordance with the Local Health Network Human Research Ethics Committee and reported adhering to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines. Patient data were extracted from both paper and electronic medical records and were kept anonymous and encrypted throughout. Patients were divided into two groups: those managed following the introduction of our ERAS protocol (w‐ERAS) (June 2020 to June 2021) versus those managed prior to its implementation (p‐ERAS) (June 2019–June 2020). Patient factors accounted for included primary cancer site, AJCC/UICC TNM 8 staging, as well as demographic factors including age and gender (Table 1). TABLE 12 cycle audit comparing patients pre and post ERAS implementation (post ERAS patients identified as number followed by ‘a’).PatientGenderAgeType of surgeryTNM 8 stagingUnilateral neck dissection = 1 bilateral neck dissection = 2, no neck dissection = 01F57OropharyngectomyT2N1M011aF62OropharyngectomyT2N1M012M59OropharyngectomyT2N1M012aM58OropharyngectomyT2N1M013M58OropharyngectomyT2N1M013aM58OropharyngectomyT2N1M014M68Tongue Base ResectionT1N1M014aM68Tongue Base ResectionT1N1M015F62OropharyngectomyT2N2M025aM66OropharyngectomyT2N2M126M57Tongue Base ResectionT0N1M016aM49Tongue Base ResectionT0N1M017M63Tongue Base resectionT1N0M017aM66Tongue Base ResectionT1N0M018M58Oropharyngectomy + Tongue base ressectionT1N1M018aM60Oropharyngectomy + Tongue base ressectionT1N1M019M55OropharyngectomyT1N1M019aM55OropharyngectomyT1N1M0110M73OropharyngectomyT1N1M0110aM79OropharyngectomyT1N1M0111M62Tongue base ResectionT1N2M0211aM68Tongue base ResectionT1N2M0212M51Tongue base ResectionT1N1M0112aM57Tongue base ResectionT1N1M0113F69OropharyngectomyT2N2M0213aF65OropharyngectomyT2N2M0214M71OropharyngectomyT2N1M0114aM72OropharyngectomyT2N1M0115M62Tongue base ResectionT1N1M0115aF50Tongue Base ResectionT1N1M0116F49Oropharyngectomy + Tongue base resectionT1N0M0116aM47Oropharyngectomy + Tongue base ResectionT1N0M0117M63Tongue Base ResectionT2N2M0217aM70Tongue Base ResectionT2N2M0218M45OropharyngectomyT2N1M0218aM39OropharyngectomyT2N1M0219M62OropharyngectomyT3N2M0219aF62OropharyngectomyT3N2M0220M66OropharyngectomyT3N0M0220aF69OropharyngectomyT3N0M0221M71OropharyngectomyT2N1M0121aM72OropharyngectomyT2N1M0122M69OropharyngectomyT2N2M0222aF79OropharyngectomyT2N2M02 2 cycle audit comparing patients pre and post ERAS implementation (post ERAS patients identified as number followed by ‘a’). The primary outcome of this study was to assess LoS. Secondary outcomes included length of ICU stay, post‐operative complication rates, cost analysis, and day of post‐operative swallow return. It should be noted that the return of swallow was measured by the day of Nasogastric Tube (NGT) removal as this was regarded to reflect an independent and safe swallow although oral diet was introduced prior to this. All data was collated in Excel for Mac 2011 (Microsoft, Redmond, US) and analysed using SPSS version 27 (IBM). Mann Whitney tests were used to examine differences in outcome measure between groups. Eras Protocol Our Institution is among the first in Australia to publish comprehensive ERAS protocols for Head and Neck surgery. A multidisciplinary approach with key stakeholders ranging from Anaesthetists, Intensivists, Otorhinolaryngologists, Allied health and Nursing staff collaborated to generate perioperative protocols using ERAS principles for laryngectomies, free flap reconstructions and TORS. The protocol consists of three major arms: Pre‐operative, intra‐operative and post‐operative management (Figure 1). Post‐operatively, patients are divided into two streams; high and low risk for fistulisation, which is determined intraoperatively by the surgeon. A longer duration of antibiotic therapy, drainage time, and nil‐by‐mouth status is applied to the high‐risk group. All patients received tranexamic acid both intra and post operatively due to its efficacy in intraoperative blood loss and post operative drain volumes [5]. All interventions are communicated effectively to team members by way of a ward checklist attached to each patient documentation (Figure 2). During the time of implementation of protocol, the robotic surgeries occurred on a Friday at a nearby hospital with subsequent transfer to our institution. For this reason, all patients had one night stay at a high dependency unit prior to their transfer. Royal Adelaide Hospital Unit create ERAS TORS protocol. Ward use checklist for post operative recovery. Results In total, 44 patients were included in this study, 22 in each cohort. There was a total of 40 TORS cases post implementation of protocol across a Covid‐19 affected year, 25 of which were malignant oropharyngeal lesions. Primary cancer sites determined surgical resection sites, with 26/44 (59%) cases undergoing lateral oropharyngectomy; 14/44 (32%) cases were base of tongue mucosectomies, with 4/44 (9%) having both oropharyngectomy and tongue base resection. A total of four cases were true carcinomas of unknown primaries. Of the patients managed w‐ERAS, the mean adherence to protocol rate was 90% (Table 2). The low‐risk group had a protocol adherence of 92.6%, the high‐risk; 83.3% (Table 2). When considering LoS, the w‐ERAS group had a median of 6 days (range: 3–33) when compared to 9.5 days (range: 5–22) in the p‐ERAS group. There was a statistically significant reduction in median length of stay by 3.5 days (p = 0.030). Within the w‐ERAS group, the high‐risk stream had a median stay of 9 days (range: 7–33) and the low risk; 5 days (range: 3–12). Additionally, both groups had no difference in their median ICU length of stay (1 day for both streams); this did not show statistical significance (p = 0.24). TABLE 2Adherence rates of the patients managed w‐ERAS.Protocol factors Patients N = 22 (total) n1 = 16 (low risk) n2 = 6 (high risk) N (%)Pre‐operativePre‐operative counselling22/22100%Speech pathology review22/22100%Dietetics review22/22100%Intra operativeAntibiotics on induction22/22100%Insertion of NG tube22/22100%Post operativeTranexamic acid 1 g TDS22/22100%H2O2 Gargles 3:1 for 10 days22/22100%IV Cefazolin 2 g TDS, Metronidazole 500 mg BD ×48 h22/22100%SCDs until mobile22/22100%Mobilise out of bed Day 122/22100%IDC out Day 122/22100%Analgesia as per protocol21/2294%Day 3: PO intake (Low risk protocol)12/1675%Day 3 Removal of drains (Low risk Protocol)12/1675%PO antibiotics for further 5 days post IV (High Risk Protocol)6/6100%NET Removed D6 (High Risk Protocol)4/667%Day 6 PO fluids to begin (High Risk Protocol)4/667%Removal drain Day 6 (High Risk Protocol)3/650%Mean overall compliance to Low‐risk Protocol92.6%Mean overall compliance to High‐risk Protocol83.3% Adherence rates of the patients managed w‐ERAS. Patients N = 22 (total) n1 = 16 (low risk) n2 = 6 (high risk) Among patients who had NGTs there was a faster return to swallow among the w‐ERAS group with a median of 4 days post operatively (range: 3–25) compared to a median of 5.5 days in the p‐ERAS group (range: 3–32) (p = 0.27). This includes patients who were discharged with an NGT and subsequently followed up in our outpatient department. Thorough follow‐up with surgeons, speech pathologists, and dietetics facilitated ongoing swallow assessment post‐discharge. Two patients required radiologically inserted gastrostomies as a means of long‐term enteral feeding in the p‐ERAS group versus one patient in the w‐ERAS group. Six (27%) p‐ERAS patients were discharged with a NGT tube. Post‐operative complications graded according to the Clavien‐Dindo classification (Figure 3) showed a similar number of complications comparatively (Table 3). Both groups had a relatively lower number of major complications (Grade 3 and above) compared to minor complications (Grade 1 + 2). The w‐ERAS group fared marginally better, with one fewer Grade 2 and Grade 4 complication. Clavien Dindo classification. TABLE 3Post‐operative outcomes comparing w‐ERAS to p‐ERAS.Post operative outcomesTotalw‐ERAS np‐ERAS np‐valueNo. of patients442222—Median Hospital LOS (range)6.5 (3–33)9.5 (3–33)6 (4–20) 0.03 Mean Hospital LOS ± standard deviation8.54 ± 5.67.6 ± 6.49.5 ± 4.9Median ICU LOS1110.24Mean ICU LOS ± standard deviation1.4 ± 1.41.3 ± 1.31.6 ± 1.5Median day of swallow return5 (1–7)4 (1–7)5.5 (1–5)0.10Mean day of swallow return ± standard deviation6.9 ± 5.96.2 ± 5.17.6 ± 6.7Requiring Gastrostomy feeding tube (n%)2 (4.5%)1 (4.5%)1 (4.5%)—Overall complications (n%)19 (43.2%)9 (40.9%)10 (45.5%)—CD 1 (n%)12 (27.3%)6 (27.3%)6 (27.3%)—CD 2 (n%)1 (2.27%)01 (2.27%)—CD 3 (n%)4 (9.09%)2 (9.09%)2 (9.09%)—CD 4 (n%)2 (4.55%)1 (4.55%)1 (4.55%)—Medical complications (n%)5 (11.4%)3 (13.6%)2 (9.09%)—Cardiac complications (n%)1 (2.27%)1 (4.55%)0—Respiratory complications (Pneumonia, Pulmonary embolism) (n%)3 (6.82%)1 (4.5%)2 (9.09%)—Urinary Tract infection (n%)1 (2.27%)1 (4.5%)0—Surgical complications (n%)5 (11.4%)3 (13.06%)2 (4.55%)—Fistula formation (n%)2 (4.55%)1 (4.55%)1 (4.55%)—Tracheostomy (n%)1 (2.27%)2 (9.09%)1 (4.55%)—Abbreviations: CD = Clavien‐Dindo; Bold = Statistical Significance p < 0.05. Post‐operative outcomes comparing w‐ERAS to p‐ERAS. Abbreviations: CD = Clavien‐Dindo; Bold = Statistical Significance p < 0.05. The median cost of a w‐ERAS patient was AUD $31 259 versus p‐ERAS AUD $39 121 across their length of stay. This was based on a surgical and intensive care bed cost per day figure from our institution's costs and manuals document. Discussion Within the first year of implementing a unit created ERAS protocol for TORS, we observed a significant decrease in median LoS by 3.5 days (p = 0.030) for patients managed w‐ERAS (6 days, range: 3–33) compared to those p‐ERAS (9.5 days, range: 5–22), and by association a decrease in median cost per patient with a saving of AUD $7800 per patient. The w‐ERAS group also demonstrated an earlier return to swallow postoperatively, albeit without statistical significance. ERAS protocols across general surgical fields have shown both practical and economic feasibility as well as a reduction in patient length of stay without increased complication rates [6]. Our study supports these findings when studying ERAS in TORS. Dort et al. was the first consensus document for ERAS principles in open head and neck oncological surgery. Since 2016, several studies have evaluated LoS in open head and neck surgeries after adopting ERAS, with many identifying improvements compared to control groups [7, 8, 9, 10]. Bertazzoni et al. in particular, conducted a similar case‐matched analysis of patients undergoing open head and neck surgeries with and without ERAS, identifying a significant reduction in both ICU LoS and total LoS in the ERAS group. There are few studies that focus on similar outcomes when studying ERAS in TORS; of the existing studies, most pertain to the efficient and effective multimodal analgesia model that ERAS protocols provide. Both Ganti et al. and Jantali et al. identified decreased mean morphine equivalent doses within 72 h of the post‐operative period. Ganti et al. also reported a decrease in LOS as a secondary outcome, although it did not reach statistical significance in their study. Across the literature regarding HNS two critical factors in determining LoS were early mobilisation [11] and decreased delays in initiating oral intake, with the latter being particularly relevant in TORS [4]. Twomey et al. identified delayed time‐to‐mobilisation as a significant predictor of postoperative complications and prolonged LoS. The ERAS protocol achieved a 100% compliance rate in the removal of IDC and mobilising patients within 24 h of operation and, as such, may have been an important factor in our favourable results. Additionally, early involvement of physiotherapy encouraged this and aided in avoiding potential discharge delays. In relation to swallowing outcomes, TORS in Oropharyngeal SCC has been shown to be superior when compared with open surgical approaches [4], with some data suggesting that outcomes may be favourable when compared with non‐surgical treatment regimens [12, 13]. Nonetheless, TORS, as demonstrated by techniques such as Fibreoptic Endoscopic Evaluation of Swallowing (FEES) in the early post‐operative period, does show a considerable effect on swallowing [13]. As such prophylactic feeding tube placement post TORS and formal swallow assessment are features of this protocol. Prophylactic feeding tubes were common practice in a review by Chia et al. of surgeon preferences across 2015 TORS cases where approximately 71% of surgeons placed an NGT at the time of surgery. Balancing early commencement of oral intake against fistulisation risk in a patient population physiologically affected by cancer with the added insult of surgery was aimed to avoid malnutrition and maximise wound healing. A few days to allow healing of the operative site with a speech pathology assessment may contribute to earlier return to swallow rates and lower RIG dependence in patients managed w‐ERAS. Although there are multiple contributing factors leading to the finding of decreased length of stay, our institution's ERAS protocol promotes a high level of preoperative planning. Early surgical, anaesthetic, speech pathology, and dietetics involvement allow for goal setting and baseline assessments that likely contribute to a clear direction of care for factors such as return of swallow in the post‐operative period. Furthermore, a unique feature of the protocol that differs from other protocols [4, 14] is the post‐operative division of patients determined to be at high or low risk for fistulation. This highlights a somewhat understated advantage of this ERAS protocol, namely its streamlined communication across all personnel involved in patient care (Surgical, Allied Health, Nursing Staff, etc.) on our unit. Coordination of events such as swallow studies, early mobilisation and adequate nutrition are all conducted with ease. Furthermore, patients themselves are aware of upcoming assessments and targets of care, allowing them to participate more readily in the recovery process and ultimately leading to safe, efficient and informed discharge pathways. The TORS program at our institution has been well established, being the first head and neck unit in Australia to undertake robotic transoral surgery in August of 2008. Due to hospital logistics of the location of the robotic platform in a nearby institution followed by patient transfer over a weekend to our hospital, this constitutes a slightly longer length of stay for TORS patients at our hospital in comparison to other institutions. This audit purposes to assess the efficacy of a protocol which kept the aforementioned conditions constant both pre and post implementation of ERAS. Although our study attempted to minimise biases, the significant improvement observed in patient outcomes may have other co‐founders not least of which includes factors such as broader institutional learning and general advancements in perioperative care, in addition to these, there are other limitations to consider. Whilst this study is unique in comprehensively examining the merits of a novel evidence‐based ERAS protocol specific to TORS with LoS as a primary outcome measure, there are several frailties to consider. Firstly, our study is limited somewhat by insensitive outcome measures, which may give rise to Type II statistical error. It is possible, therefore, that the absence of any statistically significant associations in return to swallow data between w‐ERAS and p‐ERAS patient groups may be simply a reflection of this. Moreover, the retrospective nature of the study compounds this issue as LoS and return to swallow were quantified in full days as it was difficult to be more precise than this during retrospective data collection. Further limitations inherent to the retrospective study design relate to selection bias. For instance, data were difficult to gain in some cases due to paper‐based records for certain participants, resulting in the omission of some data sets, while return to oral intake was challenging to assess in some cases where there were alternate means of nutrition. As a point of example, there were some patients who remained dependent on an NGT for a prolonged period; in this instance, gastrostomies were considered as a more sustainable long‐term route of enteral nutrition, these patients were excluded in any comparative analysis of swallow return. Despite including 44 patients undergoing TORS, our study may have been limited by relatively small patient numbers with a propensity for both Type I and II statistical error. To a large extent, this relates to the fact that the study period spanned a COVID‐19 affected year, during which there was a significant reduction in TORS at our institution, and as such resulted in a smaller than anticipated sample size. Conclusion Our study demonstrates an easily implementable protocol for the perioperative management of TORS oncological surgery with an adherence to protocol rate of 90%. Features of our protocol included a thorough preoperative assessment as well as two streams of post‐operative management, all of which facilitated streamlined communication across team members. In a small sample size, the use of ERAS protocols in TORS was associated with a reduction in patient length of stay. Author Contributions All authors met IJCE criteria for authorship in the production of this manuscript; parts that had particular contributions were as follows: Andrew Foreman, JC Hodge, Suren Krishnan, Oladejo Olaleye, Bindy Sahota designed the work; Jonathan Daniel, Lee Pryor, Bora Jeong, Bindy Sahota acquired and analysed data; Andrew Foreman, JC Hodge, Mark Wilkie, Bindy Sahota, Arun Takhar, Jonathan Daniel drafted, revised and approved the manuscript; Jonathan Daniel and Andrew Foreman agree to be accountable for all aspects of the work. Funding The authors have nothing to report. Ethics Statement This retrospective cohort review was approved by the Central Adelaide Local Health Network Ethics and Research Governance Office prior to commencement of the study (Reference No. 13407). The study was conducted in accordance with the Declaration of Helsinki, the National Statement on Ethical Conduct in Human Research, and relevant institutional governance requirements. Given the retrospective nature of the study and use of routinely collected, de‐identified clinical data, the requirement for individual informed consent was waived. All data were handled confidentially and analysed in de‐identified form. No identifiable patient information is reported in this manuscript. Conflicts of Interest Raguwinder Bindy Sahota is an Editorial Board member of Clinical Otolaryngology and a co‐author of this article. To minimise bias, they were excluded from all editorial decision‐making related to the acceptance of this article for publication. Mark Wilkie is an Editorial Board member of Clinical Otolaryngology and a co‐author of this article. To minimise bias, they were excluded from all editorial decision‐making related to the acceptance of this article for publication. Andrew Foreman is an Editorial Board member of Clinical Otolaryngology and a co‐author of this article. To minimise bias, they were excluded from all editorial decision‐making related to the acceptance of this article for publication. Oladejo Olaleye is an Editorial Board member of Clinical Otolaryngology and a co‐author of this article. To minimise bias, they were excluded from all editorial decision‐making related to the acceptance of this article for publication. Acknowledgements Open access publishing facilitated by Adelaide University, as part of the Wiley ‐ Adelaide University agreement via the Council of Australasian University Librarians. Data Availability Statement The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions. References Associated Data Data Availability Statement The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.