Buccal fat applied to transoral robotic lateral oropharyngectomy defects to lessen radical tonsillectomy pain (BOLT): a single-centre, phase II, parallel, randomised control trial study protocol Authors: Michael Xie, John de Almeida, David Goldstein, Rosemary Martino, Yi Fan Liu, Brittany Allen, Wei Xu, Katrina Hueniken, Christopher MKL Yao Published: 2025 ABSTRACT AbstractIntroductionTransoral robotic surgery (TORS) is a minimally invasive technique for surgical removal of tumours of the tonsil and lateral oropharynx. Surgical defects after TORS lateral oropharyngectomy are traditionally left open to heal by secondary intention, resulting in significant postoperative pain and secondarily resulting in delayed swallowing and discharge. Although multimodal analgesia can improve postoperative pain control, no studies to date have assessed the impact of adjunct surgical interventions for reducing postoperative pain after TORS. Buccal fat rotation flap is a regional reconstruction option after TORS lateral oropharyngectomy and provides immediate coverage of the open surgical wound. However, the impact of buccal fat rotation flap reconstruction on postoperative pain and swallowing remains unclear. This trial aims to compare postoperative pain outcomes in patients who undergo TORS lateral oropharyngectomy with and without buccal fat rotation reconstruction.Methods and analysisThis protocol outlines a single centre, parallel, unblinded, phase II, randomised control trial. Inclusion criteria include adult patient (≥18 years) undergoing TORS lateral oropharyngectomy for early to intermediate stage tonsillar squamous cell carcinoma (T1-2N0-1 p16+/−) or early to intermediate stage salivary gland tumours of the palatine tonsils. Exclusion criteria include a history of prior head and neck squamous cell carcinoma, prior head and neck radiotherapy, retropharyngeal lymphadenopathy, bilateral lymphadenopathy, need for bilateral neck dissection, baseline trismus, opioid use or drug addiction, need for open surgery (transcervical lateral oropharyngectomy), free tissue transfer, or alternative regional flap, and pregnancy. All patients are planned for a TORS lateral oropharyngectomy. The intervention group will have a buccal fat rotation flap reconstruction, and the control group will be allowed to heal via secondary intention. The allocation sequence will be created using a computer-generated random sequence with a permuted block strategy. The allocation sequence will be concealed until the time of assignment. The primary outcome is postoperative pain intensity during rest and swallowing using the visualised analogue scale. Secondary outcomes include postoperative complications, other adverse events, patient-reported speech and swallowing, opioid usage, length of hospital stay, feeding tube dependence and blood glucose levels. The trial has a target sample size of 40 patients. Statistical analysis of the primary outcome will be analysed in an intention to treat analysis using a linear mixed effects model.Ethics and disseminationThe study was approved by the University Health Network Coordinated Approval Process for Clinical Research. Study number CAPCR ID: 24-5894. All participants will be required to provide written informed consent to participate. Findings will be presented at national conferences and published in medical journals.Trial registration numberNCT06965738. STRENGTHS AND LIMITATIONS OF THIS STUDY The randomised control trial design will ensure balancing of known and unknown confounders between buccal fat rotation flap reconstruction and secondary intention groups. The randomised control trial design will ensure balancing of known and unknown confounders between buccal fat rotation flap reconstruction and secondary intention groups. Postoperative analgesia for both groups is standardised and clearly defined using a multimodal analgesia regimen to optimise baseline analgesia and minimise differences in co-interventions between groups. Postoperative analgesia for both groups is standardised and clearly defined using a multimodal analgesia regimen to optimise baseline analgesia and minimise differences in co-interventions between groups. The primary outcome of postoperative pain on a visual analogue scale will be measured at multiple time points including the immediate postoperative period and during routine follow-up after wound healing to characterise postoperative pain patterns over time. The primary outcome of postoperative pain on a visual analogue scale will be measured at multiple time points including the immediate postoperative period and during routine follow-up after wound healing to characterise postoperative pain patterns over time. The study sample size will not be adequately powered to make definitive conclusions for rare secondary outcomes including some relevant postoperative complications (eg, postoperative oropharyngeal bleeds). The study sample size will not be adequately powered to make definitive conclusions for rare secondary outcomes including some relevant postoperative complications (eg, postoperative oropharyngeal bleeds). The lack of patient and assessor blinding in the study design may introduce bias. The lack of patient and assessor blinding in the study design may introduce bias. Introduction Transoral robotic surgery (TORS) for oropharyngeal cancers has transformed how oropharyngeal cancers are managed, allowing for minimally invasive surgical access to remove tumours of the tonsil and base of tongue. While large incisions and open approaches can be avoided, oropharyngeal defects following TORS are traditionally left as an open wound to heal by secondary intention, resulting in significant postoperative pain, and possibly dysphagia and dehydration.14 While this has been highlighted as a key outcome measure, studies on acute pain following TORS have not had any standardisation regarding the timing of pain measurement and pain assessment.5 Only one study has assessed pain during ‘movement-related pain’, such as swallowing, which has been advocated as a key metric.1 6 7 Most studies used a visual analogue scale (VAS); however, the majority failed to describe the timing of pain assessment relative to administration of analgesia, time of day or diet. Overall, the variation between studies in pain ratings and opioid utilisation are reflected by pain ratings that range widely from 2.5 to 8 in the first few postoperative days.5 Based on the current evidence, multimodal analgesia resulted in the lowest pain ratings following TORS, with those that included gabapentin, a COX-2 inhibitor and acetaminophen having excellent pain control in the early postoperative period.1 8 9 Using these regimens, Castellanos et al found that postoperative opioid consumption significantly decreased in the first three postoperative days (PODs).9 Clayburgh et al identified that addition of perioperative corticosteroids was safe and significantly improved diet consistency and decreased length of hospital stay, though pain scores were minimally affected.10 No study thus far has assessed the impact of surgical interventions on postoperative pain and swallowing. Here, we hope to evaluate the impact of the buccal fat rotational flap on the postoperative pain following TORS. The buccal fat rotation flap was first described for closure of an oroantral communication following oncologic resection in 1977 by Egyedi.11 It was subsequently noted to epithelialise in the oral cavity within 3–4 weeks, negating the need for further reconstruction.12 13 While the maximum size of reconstructed defects ranged from 4 to 5 cm, recent reports have indicated its use in mandibular and maxillary bone coverage, and various oral cavity and oropharyngeal defects.1416 Complications include haematoma, partial necrosis or arterial bleeding, though rare. Following a transoral lateral oropharyngectomy, the ipsilateral posterior pharyngeal wall can be released from the prevertebral plane and sutured to the velum and remnant uvula with 3–0 vicryl sutures. The buccal fat pad can then be harvested from the buccal space by extending a horizontal incision along the cheek mucosa, and through the buccinator muscle adjacent to the posterior molars. The fat pad is gradually released using gentle traction with forceps and sutured in place along the oropharyngeal wall defect and closed over the rotated buccal fat pad.16 Poorly controlled pain can affect swallowing and prolong postoperative hospitalisation. It is understood that in the acute period, dysphagia is significantly worsened after TORS, with up to 23% of patients having moderate-severe dysphagia.17 While the majority of patients do return to baseline by a year, it would be critical to understand how our added intervention would impact swallowing in the short term.18 The literature regarding postoperative pain following TORS of the lateral oropharynx lacks standardisation with analgesic regimen and pain measurement. While it has been recognised as an important metric, lack of standardisation and lack of pain measurement during swallowing leaves room for further understanding. The primary purpose of this study is to evaluate the impact of adding a surgical intervention, the buccal fat rotation flap, on lateral oropharyngeal defects. Specifically, we hope to understand its impact on postoperative pain and swallowing. Secondary results of the study will help to elucidate its safety and risk of complications. Results from the study may help identify approaches to minimise treatment-related morbidity. Summary of the problem Patients presenting with an oropharynx tumour requiring a transoral robotic lateral oropharyngectomy are at risk of significant postoperative pain, which results in delayed swallowing and prolonged discharge. While prior studies have focused on varying postoperative pain management, there is significant variation and lack of standardisation to the measurement timing and instrument used. This study is aimed at assessing the impact of a surgical intervention with the rotation of buccal fat into the wound bed compared with no buccal fat for reducing postoperative pain and opioid usage in a single centre, phase II, randomised control trial. We hypothesise that patients who undergo buccal fat reconstruction will have diminished pain needs. Methods and analysis Trial start and anticipated end date 26 May 2025 to 1 June 2027. Participants This study involves a single academic centre in Toronto, Ontario, Canada. This study includes patients who require an ipsilateral transoral robotic lateral oropharyngectomy of the tonsil. The most common conditions requiring this surgery include early-intermediate risk oropharyngeal squamous cell carcinomas, salivary gland cancers and select benign salivary gland tumours. Inclusion criteria: Age >18. Age >18. Require a transoral robotic lateral oropharyngectomy as a result of the following conditions:Early—intermediate stage tonsil oropharyngeal squamous cell carcinoma (T1-2N0-1 p16+/−SCC of the palatine tonsils).Benign minor salivary gland tumours or early to intermediate stage salivary gland tumours of the palatine tonsils.Based on FNA, core biopsy, punch biopsy. Require a transoral robotic lateral oropharyngectomy as a result of the following conditions: Early—intermediate stage tonsil oropharyngeal squamous cell carcinoma (T1-2N0-1 p16+/−SCC of the palatine tonsils). Early—intermediate stage tonsil oropharyngeal squamous cell carcinoma (T1-2N0-1 p16+/−SCC of the palatine tonsils). Benign minor salivary gland tumours or early to intermediate stage salivary gland tumours of the palatine tonsils. Benign minor salivary gland tumours or early to intermediate stage salivary gland tumours of the palatine tonsils. Based on FNA, core biopsy, punch biopsy. Based on FNA, core biopsy, punch biopsy. Ability to understand and willing to sign a written informed consent. Ability to understand and willing to sign a written informed consent. Exclusion criteria: History of prior head and neck squamous cell carcinoma, or prior head and neck radiotherapy. History of prior head and neck squamous cell carcinoma, or prior head and neck radiotherapy. Presence of retropharyngeal lymphadenopathy. Presence of retropharyngeal lymphadenopathy. Presence of bilateral lymphadenopathy, or patients requiring bilateral neck dissections. Presence of bilateral lymphadenopathy, or patients requiring bilateral neck dissections. Patients with trismus at baseline. Patients with trismus at baseline. Patients with psychological risk factors for persistent opioid use or drug addiction. Patients with psychological risk factors for persistent opioid use or drug addiction. Need for open surgery (transcervical lateral oropharyngectomy), free tissue transfer or alternative regional flap. Need for open surgery (transcervical lateral oropharyngectomy), free tissue transfer or alternative regional flap. Pregnancy. Pregnancy. Interventions Both arms will undergo a standard transoral robotic lateral oropharyngectomy with the Davinci Xi (Intuitive Surgical, Sunnyvale, CA, USA). Patients will be intubated nasotracheally and positioned in a standard fashion. The Davinci Xi will then be appropriately docked for exposure of the tumour. Standard mucosal incisions around the tumour will be made for tumour extirpation with removal of at least the constrictor muscle as the deep margin. Haemostasis will be achieved in a standard fashion with bipolar, surgical clips and suction monopolar as needed. Mucosal margins from the specimen or patient will then be sent for frozen section analysis to guide need for further ablation. Patients undergoing intervention will undergo buccal fat reconstruction. Following a transoral lateral oropharyngectomy, the ipsilateral posterior pharyngeal wall can be released from the prevertebral plane and sutured to the velum and remnant uvula with 3–0 vicryl sutures. The buccal fat pad can then be harvested from the buccal space by extending a horizontal incision along the cheek mucosa, and through the buccinator muscle adjacent to the posterior molars. The fat pad is gradually released using gentle traction with forceps and sutured in place along the oropharyngeal wall defect and closed over the rotated buccal fat pad.16 Patients in the control group will be allowed to heal via secondary intention without further reconstructive interventions. Relevant concomitant care and interventions that are permitted or prohibited during the trial There will be no restrictions on intraoperative anaesthetic medications. The use of long-acting narcotics, steroids and other analgesia will be at the discretion of the attending anaesthesiologist. Local anaesthetic, including topical and infiltrative, will not be permitted during the trial. A standard postoperative pain regimen will be applied to all patients (figure 1). Standardised postoperative pain regimen. Primary, secondary and other outcomes Primary outcome The primary objective for this study is to determine the effect of a buccal fat rotation flap on postoperative pain following transoral robotic lateral oropharyngectomy. The primary outcome will be determined using VASs at rest and swallowing following surgery. These questionnaires will be acquired daily following surgery during hospitalisation, once between days 7–10, then at POD 21–28. Patients will score their pain in their pain journal (online supplemental I) after taking their oral analgesia per the pain regimen to measure optimised postoperative pain. Secondary outcome To compare between arms the following: Postsurgical complications and adverse events within 30 days of surgery. This outcome will be determined using the National Cancer Institute Common Terminology Criteria for Adverse Events (NCI CTC-AE V.4.0). This includes postoperative length of stay, bleeding, need for re-operation within 30 days, re-admission within 30 days, visit to the Emergency Room within 30 days. Postsurgical complications and adverse events within 30 days of surgery. This outcome will be determined using the National Cancer Institute Common Terminology Criteria for Adverse Events (NCI CTC-AE V.4.0). This includes postoperative length of stay, bleeding, need for re-operation within 30 days, re-admission within 30 days, visit to the Emergency Room within 30 days. Postoperative opioid consumption. This outcome will be captured daily during the inpatient stay, first 7 days, first 14 days and first 30 days and quantified in morphine equivalent units. Postoperative opioid consumption. This outcome will be captured daily during the inpatient stay, first 7 days, first 14 days and first 30 days and quantified in morphine equivalent units. Speech and swallowing. Swallowing-related quality of life (performance status scale head and neck (PSS-HN)) (online supplemental II) and MD Anderson Symptom Inventory (MDASI) assessed preoperatively, days 7–19, days 21–28. Patients will be assessed for their dietary consistency and speech at discharge and follow-ups. Speech and swallowing. Swallowing-related quality of life (performance status scale head and neck (PSS-HN)) (online supplemental II) and MD Anderson Symptom Inventory (MDASI) assessed preoperatively, days 7–19, days 21–28. Patients will be assessed for their dietary consistency and speech at discharge and follow-ups. Weight: baseline weight and postoperative weights captured at the first follow-up visit between day 21–28. Weight: baseline weight and postoperative weights captured at the first follow-up visit between day 21–28. Postoperative haemorrhage within 30 days postoperatively. Postoperative haemorrhage within 30 days postoperatively. Feeding tube placement within 30 days postoperatively. Feeding tube placement within 30 days postoperatively. Blood glucose levels during the 30 days postoperatively. Blood glucose levels during the 30 days postoperatively. Other measured variables Post-resection defect size in centimetres. Post-resection defect size in centimetres. Buccal fat rotation flap size in centimetres and percentage of surgical wound coverage (in the buccal fat rotation flap group). Buccal fat rotation flap size in centimetres and percentage of surgical wound coverage (in the buccal fat rotation flap group). Study flow diagrams Patients will be screened, recruited and then randomised into either arm (figure 2). Standard preoperative and postoperative outcome assessments will be completed based on established timepoints (figure 3). Study flow diagram. MDASI, MD Anderson symptom inventory; PSS, performance status scale; TORS, transoral robotic surgery. Standardised outcome assessment timeline. MDASI, MD Anderson symptom inventory head and neck; POD, postoperative day; PSS, performance status scale head and neck; VAS, visual analogue scale. Sample size From prior clinical trials on postoperative pain after maximal pain regimens with (dexamethasone, Celebrex, gabapentin, Tylenol and opioids), the pain scores range from 6.8, 6.4 to 6.7.10 From our own experience, the minimal clinically significant pain difference would be at least a decrease by 1.0% or 10%.19 A parallel two-group design will be used to test whether the means between groups will be different (H0: μ1-μ2=0 vs H1: μ1-μ2≠0). The comparison will be made using a two-sided, two-sample equal-variance t-test, with a Type I error rate (α) of 0.05. The common SD for both groups is assumed to be 2.1. To detect a difference in means of μ1-μ2=6.7–4.7=2 with 80% power. Conservatively, we anticipate that we would need 40 patients. Strategies for achieving adequate participant enrolment to reach target sample size Strategies to achieve target enrolment rates and sample sizes include an educational presentation to all attending surgeons and relevant staff regarding the trial start date, eligibility criteria and study contacts. The study coordinators will release bi-annual recruitment target emails to summarise enrolment progress. Study information and updates will also be posted on the University of Toronto Otolaryngology Head and Neck Surgery website. Assignment of interventions The allocation sequence will be created using a computer-generated random sequence with a permuted block strategy by an independent statistician. The allocation sequence will be implemented and assignments will be executed using Research Electronic Data Capture (REDCAP) by an independent statistician. The allocation sequence will be concealed until the time of assignment. Surgeons and research coordinators will enrol participants. After enrolment, interventions will be assigned by an independent statistician and the attending surgeon, research team and patient will be notified. There will be no blinding as part of this trial. Given that patient care and documentation will be completed through the standard electronic health record, surgery group assignment will be readily accessible and outcome assessment blinding cannot be guaranteed based on the current trial design and is a limitation of the study. Data collection, management and analysis The following types of data will be collected from participants. Pain (measured by numeric scale both in the morning/night/at rest and on dynamic movement) will also be measured during the hospitalisation, day 7–10, and day 21–28. Pain (measured by numeric scale both in the morning/night/at rest and on dynamic movement) will also be measured during the hospitalisation, day 7–10, and day 21–28. Total opioid usage will also be measured during the first 5 days, then by day 7, and 21–28. Total opioid usage will also be measured during the first 5 days, then by day 7, and 21–28. PSS-HN is a 3-item instrument rated via semi-structured interview for normalcy of diet, feeling towards public eating and understandability of speech. This will be captured preoperatively, POD 7–10 and POD 21–28. PSS-HN is a 3-item instrument rated via semi-structured interview for normalcy of diet, feeling towards public eating and understandability of speech. This will be captured preoperatively, POD 7–10 and POD 21–28. MDASI is a multi-symptom patient-reported measure that assesses symptom severity and interference with daily living. This will be captured preoperatively, POD 7–10 and POD 21–28. MDASI is a multi-symptom patient-reported measure that assesses symptom severity and interference with daily living. This will be captured preoperatively, POD 7–10 and POD 21–28. Complications data will be assessed at 30 days, including:Trismus.Dimpling of skin.Haematoma.Bleeding.Additional complications per CTC-AE. Complications data will be assessed at 30 days, including: Trismus. Trismus. Dimpling of skin. Dimpling of skin. Haematoma. Haematoma. Bleeding. Bleeding. Additional complications per CTC-AE. Additional complications per CTC-AE. The following data will be completed by the study team: Demographics. Demographics. Medical history and medications. Medical history and medications. Pathology. Pathology. Initial staging. Initial staging. Baseline pain requirements. Baseline pain requirements. Baseline trismus measurements. Baseline trismus measurements. Baseline. Baseline. Disease status. Disease status. Complications. Complications. G tube dependence. G tube dependence. Tobacco and alcohol. Tobacco and alcohol. Work status. Work status. Body mass index. Body mass index. Participant retention and complete follow-up, data entry, coding, security, storage There are no additional plans to promote participant retention as this is a relatively short perioperative/follow-up period. If any patients decide to withdraw from the study, they will be marked as withdrawn/discontinued and no data will be collected/used. Study data will be stored using REDCAP on site at the Toronto General Hospital in the Department of Otolaryngology. The database will be located on a network server hospital computer at TGH in a password-protected file. Data will be entered, coded and quality checked by a single statistician. Statistical analysis Final analysis will be performed in a blinded fashion. Baseline characteristics (age, sex, tumour stage, resection extent, smoking history and preop pain characteristics) will be summarised by groups using means±SD, and median (IQR) for continuous variables and frequencies (%) for categorical variables. Pain scores at each time point will be summarised by group and presented as mean±SD or median with IQR (depending on distribution). A linear mixed effects model will be used to test whether the change in pain over time differs between groups, with fixed effects including (flap vs control), time, and random effects including (patient level intercept). As secondary analyses, time-point specific comparisons will be performed with independent t-tests or Mann-Whitney U tests; total opioid consumption as measured by oral morphine equivalents will be compared at multiple time points with independent t-tests. Primary and secondary outcomes will be reported using an intention to treat analysis. Monitoring This trial will not have a data monitoring committee or planned interim analysis as it is anticipated to be of short duration, minimal risks and does not involve developmental interventions. Medical and surgical complications will be documented during the study follow-up period and reported in the final manuscript. Ethics and dissemination The present study was approved by the University Health Network (UHN) CAPCR ethics committee, study number CAPCR ID: 24-5894. All participants in this study will be asked to provide written informed consent. The findings will be presented at conferences (eg, Canadian Otolaryngology Society, American Head and Neck Society). The trial results will be published in a peer-reviewed medical journal. Participant consent Eligible participants for the study will be screened during routine clinic lists by a provider in the patient’s circle of care. Should the patient express interest in participating, a research team member will then approach the participant in person to go over the study and obtain written informed consent (online supplemental III), if the patient remains interested in enrolling. Data storage and privacy Routine practices for ensuring the confidentiality and privacy of all participants will be followed in this study. Data will be stored in a secure database software on site at the Toronto General Hospital in the Department of Otolaryngology. The database will be located on a network server hospital computer at TGH in a password-protected file. Participants will be de-identified and assigned a study number. The identity log will be kept separately in the department with the study coordinator. Participant names and other personal identifiers will not be used in any reports, presentation or published work. Data will be stored for approximately 10 years from the date of study closure. All data will remain with the PI in the Department of Otolaryngology. Only investigators, study team members and authorities to do monitoring, audits, research ethics board review and regulatory inspection will have access to data. In the event of an inappropriate release of personal health information, any information that can be retrieved will be retrieved, the UHN Privacy Office and research ethics board will be notified, and the appropriate further actions will be taken. Patient and public involvement Patients or the public were not involved in the design, or conduct, or reporting, or dissemination plans of our research. Trial protocol and dataset access The protocol will be published for review. Participant level data and statistical analyses will be available on request by contacting the principal investigator. Trial data will be available at request by contacting the principal investigator. Trial result communication The results of the trial will be communicated via journal article publication and trial status updated in trial registries. Supplementary material