Factors in Successful Elimination of Elective Tracheotomy in Mandibular Reconstruction With Microvascular Tissue

Importance  Limited data exist on performing major oral cavity resections and reconstructions without elective tracheotomy.

Objectives  To describe a group of patients who successfully underwent major microvascular mandibular reconstruction without an elective tracheotomy and to perform a literature review to identify commonalities between our group and the available literature to identify potential common factors that might contribute to the success of this approach.

Design, Setting, and Participants  Case series with retrospective medical chart review of 15 patients who underwent fibula microvascular free flap reconstruction of mandibular defects without tracheotomy between 2000 and 2014 (the most common indication was osteoradionecrosis) conducted at a tertiary referral hospital center.

Exposures  Mandibular reconstruction with fibula free flap without elective tracheotomy.

Main Outcomes and Measures  Perioperative morbidity and mortality with a focus on airway management and perioperative complications.

Results  The median age of 15 patients (11 males and 4 females) at the time of reconstruction was 42 years (range, 10-64 years). The indication for surgery was nonmalignant pathologic abnormalities in 8 patients (53%), osteosarcoma in 4 patients (27%), and oral cavity squamous cell carcinoma in 3 patients (20%). All patients were intubated endonasally and, excepting 1 pediatric case, were extubated the day following surgery. The fibula spanned the parasymphysis and/or symphysis in 2 patients, and was limited to the mandibular body in others (mean length, 7.6 cm [range, 4.0-15.0 cm]). The flap design was osteocutaneous in 3 patients and osseous in the rest. No patient required reintubation or tracheotomy. No complications due to endonasal intubation occurred. The average hospital length of stay was 11 days.

Conclusions and Relevance  Major oral cavity resection and reconstruction with microvascular free flaps can be performed safely without elective tracheotomy in a select group of patients. Limited data exist on patient selection criteria. Further studies are needed to identify favorable factors and develop protocols for safe patient selection.

Tracheotomy is electively performed for most patients undergoing head and neck microvascular free flap reconstruction to secure the airway. The necessity of routine tracheotomy has been questioned,^1,2 however, owing in part to potential added complications associated with tracheotomy, such as tracheal stenosis, bleeding, fistula formation, external scars, and pneumonia.³ Additionally, tracheotomy has been shown to negatively impact quality of life⁴ and increase hospital length of stay.⁵

The use of tracheotomy in mandibular reconstruction is of particular significance due to the potentially large oral defects associated with these procedures and the ensuing implications in maintaining a secure and safe airway in the immediate postoperative period.⁶ However, studies addressing outcomes associated with the omission of tracheotomy in head and neck reconstructions have not specifically examined patients undergoing repair of large composite mandibular defects with free tissue transfer procedures.^5,7,8 Consequently, the decision for elective tracheotomy in mandibular reconstruction has been predominantly operator dependent, because definitive outcomes are unknown and protocols for guidance do not yet exist. We hypothesize that in a select group of patients, mandibular reconstruction can be safely attempted without elective tracheotomy. Herein, we describe our experience with this approach and investigate perioperative morbidity and mortality with a focus on airway management and perioperative complications. We compared our cases with the available literature to identify common characteristics that might influence the success of this paradigm.

Approval for this study was obtained from the New York University School of Medicine institutional review board. From January 2000 to December 2014, all head and neck microvascular free flap reconstructions performed at the NYU Langone Medical Center by the senior author (M.D.D.) were reviewed. Patients who underwent osseous reconstruction for segmental mandibular defects using microvascular free flaps were identified. Patients who did not receive elective tracheotomy at the time of the operation were included. The decision to eliminate elective tracheotomy was at the discretion of the operating surgeon. In brief, all patients were assessed and examined by the operating surgeon and anesthesiologist. Only patients who were deemed to be at low risk for oral reintubation based on the anatomy, size, and location of their defects, as well as lack of complicating factors, such as trismus and extensive fibrosis, were selected.

Data were collected on demographics, indication for surgery, intraoperative findings, pathologic characteristics, perioperative complications, and hospital length of stay. Mandibular defects were described by anatomical location (body, ascending ramus, angle, symphysis) and length of the osseous defect, as determined by the length of resected mandible in the pathology report. Surgical approaches and any associated soft-tissue defects were documented. Outcomes, including hospital length of stay and perioperative complications, were abstracted. Perioperative complications were defined as any adverse events occurring within 30 days of the surgery. Data were analyzed between February and March of 2015 using IBM SPSS statistical software (version 20).

From 2000 to 2014, 120 patients underwent head and neck microvascular free flap reconstructions at the NYU Langone Medical Center by the senior author. Of these patients, 29 (24%) had undergone reconstruction for segmental mandibular defects. Fifteen patients (12%) did not receive elective tracheotomy.

The data are summarized in Table 1. All 15 patients (11 males, 4 females) had undergone reconstruction using a fibula microvascular free flap, 3 (20%) of which were osteocutaneous and 12 (80%) osseous. The median age of patients at the time of reconstruction was 42 years (range, 10-64 years). Indications for surgery were osteoradionecrosis or pathologic fracture in 4 patients (27%), oral cavity squamous cell carcinoma in 3 (20%), osteosarcoma in 4 (27%), benign mandibular tumors (ameloblastoma, odontogenic myxoma) in 3 (20%), and extraoral exposed hardware in 1 (7%). A total of 8 (53%) patients had undergone surgery or radiotherapy prior to reconstruction. Three patients (20%) had undergone both radiotherapy and surgery, 3 (20%) had undergone radiation therapy alone, and 2 (13%) had undergone surgery alone. All patients were intubated endonasally. A transcervical approach was used in 10 patients (67%), and a translabial approach was used in 5 (33%). A soft-tissue defect was present in 5 (33%), and involved the floor of the mouth in 3 (20%), the buccal mucosa in 1 (7%), and the skin of the neck in 1 (7%). All mandibular defects included the body of the mandible. The parasymphysis was involved in 7 patients (47%), and the ramus in 2 (13%). There was a bilateral parasymphyseal defect in 2 (13%) (Figure). Mean osseous defect length was 7.6 cm (range, 4.0-15.0 cm) based on the measured length of the resected specimen on the pathology report.

All patients, except 1, were extubated the day following the surgery over tube exchangers in a monitored environment with the exception of the pediatric patient (case 12), for whom extra caution by managing pediatric intensive care unit staff mandated an extra day of intubation. No patient required tracheotomy or emergent reintubation due to airway compromise. The mean hospital length of stay was 11 days (range, 8-15 days). Length of stay data were unavailable for 3 patients. One patient (case 9) underwent reconstruction after experiencing a flap failure from an attempted primary reconstruction at a different hospital. This patient experienced a failure of the second flap 19 days after surgery. One patient (case 14) with a history of extensive head and neck surgerical procedures experienced an episode of arterial hemorrhage 6 days after the operation, the short-term treatment of which required sacrificing the otherwise viable flap. One patient (case 13) had developed a hematoma 1 day after surgery day, which was drained a week after the primary surgery in the operating room. There were no other complications associated with this adverse event, and this patient did not require tracheotomy. All patients were discharged from the hospital without a tracheotomy.

Tracheotomy is one of the most commonly performed operations in head and neck surgery. Though generally safe, the morbidity associated with tracheotomy is well described^3,9 and includes tracheal stenosis, bleeding, fistula formation, and pneumonia. The most common indications for tracheotomy are airway compromise and expected prolonged respiratory failure and dependence on mechanical ventilation. Tracheotomy is also performed electively in patients undergoing major head and neck reconstruction in cases of difficult airway (trismus, fibrosis), in anticipation of possible airway compromise in the immediate postoperative period, or in high-risk cases for complications and return to the operating room. The risk of airway compromise after these cases, however, is unknown. To our knowledge, no prospective, controlled study has ever investigated the impact of eliminating elective tracheotomy in the outcome of these patients. Only a limited collection of retrospective reports exist in the literature, and most describe a variety of patients undergoing different ablative and reconstructive procedures.^{1,5,6,8,10-12} In addition, no criteria have been identified for patients who are at lower risk for experiencing airway complications after most upper aerodigestive tract reconstruction. The decision to eliminate the elective tracheotomy is mostly based on the clinical judgment of the operating team. In contrast, established criteria exist for patients who undergo tracheal resection and reconstruction, and most of these patients do not receive elective tracheotomy.¹³

Only 1 recently published case series by Moubayed et al,¹⁴ has described a patient population comparable with ours. Similar to our study, major airway complications were exceedingly rare or nonexistent, with only 1 major airway complication in their report.¹⁴ In the current study, we present a group of patients undergoing mandibular reconstruction who did not receive elective tracheotomy. None of our patients experienced complications related to airway management. Endonasal intubation was adequate and safe during the operation, and all of our patients, excepting 1, were successfully extubated on postoperative day 1. We reviewed all available literature on omission of elective tracheotomy to identify potential factors in the success of this approach (Table 2). We compared our cases with prior reports to identify commonalities. Because not enough data exist at this time and no prospective study is available, we could only suggest potential influential factors.

None of our patients were assessed as having a difficult airway or to be at high risk for intubation. Specifically, none of our patients were diagnosed as having trismus. We believe this is an important factor in the success of this method. Since most of the patients that could be considered for this approach are those with osteoradionecrosis and pathologic fractures, attention to this factor is particularly important. Most patients with osteoradionecrosis have a history of radiotherapy or chemoradiotherapy to the oral cavity or oropharynx which can cause clinically significant fibrosis, trismus, and difficulty with intubation. Consequently, a difficult airway or clinically significant trismus is most likely a contraindication for eliminating elective tracheotomy. In addition, patients with prior radiotherapy or chemoradiotherapy are at increased risk for postoperative complications, such as wound breakdown, fistulas, and infections that might require return to the operating room and reintubation.¹⁵ These factors should be carefully considered before a decision to eliminate elective tracheotomy is made.

Most patients in our study underwent resection for nonmalignant pathologic abnormalities (53.3%). Only 1 patient with a nonmalignant disease required repair of a soft-tissue defect. In contrast, 3 of the 5 patients with malignant tumors had a concurrent soft tissue defect requiring reconstruction. Most patients undergoing mandibular reconstruction are diagnosed as having oral cavity squamous cell carcinomas (SCC). However, SCC primary tumors were limited to only 20.0% of our patient population and 10.0% of the patient population in the report by Moubayed et al.¹⁴ While we do not believe the type of malignant neoplasm is a decisive factor, the size of the soft-tissue defect may be an important factor. In general, patients with SCC might require larger soft-tissue reconstructions, which could make airway management more difficult postoperatively.⁸ In a study of oropharyngeal tumors, Kruse-Losler et al⁷ reviewed 152 patients undergoing tumor ablation and different methods of reconstruction without tracheotomies. Tumor location and size were the most important factors influencing placement of a postoperative tracheotomy. Tumors located at the base of the tongue, floor of the mouth, palate, or oropharynx represented a significantly increased likelihood of tracheotomy placement postoperatively. Thus, we consider extensive soft-tissue defects, particularly when located in the posterior oral cavity and oropharynx, to be a relative contraindication to the omission of elective tracheotomy. Conversely, patients undergoing reconstruction for nonmalignant pathologic abnormalities may be better suited for the omission of elective tracheotomy, assuming that these defects are not associated with extensive soft-tissue involvement or a history of radiotherapy or chemoradiation.

We were able to reconstruct long mandibular defects, with an average defect length of 7.6 cm, and defects as long as 15.0 cm. The longer defects are usually included under the category of patients needing a tracheotomy by the scoring system created by Kruse-Losler et al⁷; however, all patients had oropharyngeal primary neoplasms that could cause larger soft-tissue defects in a more central and posterior location. The large range of osseous defect sizes reconstructed in both the present study and the report by Moubayed et al¹⁴ suggest that osseous defect size might not be a limiting factor in patients undergoing reconstruction for defects limited to the oral cavity with a more lateral location. In addition, maintaining genioglossus and geniohyoid attachments are important factors in maintaining a patent airway. By preserving these structures, the anterior attachment of tongue and floor of mouth are maintained, which reduces the risk of airway blockage due to posterior displacement of these structures. In our series, unilateral anterior attachments were preserved in all cases but 2. Therefore, we believe that while osseous defect size is a nonlimiting factor in patient selection, bilateral central defects may yet represent a possible factor that could compromise the airway patency in the postoperative period. Additional data are needed to further determine this risk.

Our patient population is younger than most patients undergoing mandibular reconstruction.^16,17 This is probably reflective of the pathologic abnormalities found in our group and patient selection process. The significance of age as a prognostic factor is unknown. Kruse-Losler et al⁷ demonstrated no significantly increased need for postoperative tracheotomy with respect to the cohort age. Yet, Cameron et al⁸ found that a younger patient population was more likely to require a tracheotomy. However, in their series, the younger patients had received more preoperative radiotherapy. Moubayed et al¹⁴ reported a median age of 58 years, which was not different from the median age for all patients undergoing mandibular reconstruction with osteocutaneous free flaps in their institution. Considering differing conclusions from available studies, the significance of age as a factor is unknown at this time. Intuitively, younger patients might be considered better candidates for elimination of the tracheotomy, but whether older patients are at a disadvantage is yet to be determined.

Major complications occurring in our cohort were limited to 2 flap failures. Both patients had a history of head and neck surgerical procedures and reconstruction, which made their secondary reconstruction technically challenging. Whether performing a tracheotomy could have changed this outcome is unknown. In general, patients with prior operations and treatments are at higher risk of surgical complications. There was only 1 complication (hematoma) in a patient without prior treatments. Owing to small numbers, we could not perform a meaningful statistical analysis to investigate if prior treatments will add to the risk of perioperative complications. It seems that in patients with a history of treatments the risk of complications might be higher and therefore a more traditional, conservative approach might be more appropriate.

Based on our review, we classified different factors as favorable, unfavorable, or unknown (Box). Many of these factors corroborate with our patient selection criteria as well. Not enough data exist at this time to guide clinicians with certainty. Therefore, lists such as ours are potentially helpful in guiding investigators in designing further studies. If standard criteria are established for eliminating elective tracheotomy in head and neck reconstruction, a significant portion of patients with head and neck cancer can benefit from that paradigm with an earlier ability to talk, shorter length of stay, and avoidance of the complications of tracheotomy. The real challenge for head and neck surgeons is to balance the risk of postoperative airway compromise vs morbidity of elective tracheotomy. Our study is a small step in further stratifying this risk.

Box Section Ref ID

We acknowledge the limitations of our study. Similar to all case reports, the patients were accrued over a long period of time, and their medical records were reviewed retrospectively. The decision for omission of elective tracheotomy was operator dependent, and chosen based on consideration of many of the factors discussed in the literature, in addition to clinical judgment. Consequently, we could only provide a summary of possibly influential factors, because retrospective analysis of this case series is unable to yield novel factors for consideration in this approach. However, our findings share similarities with other reported series; many of the reported factors suggested to influence the elimination of elective tracheotomy were corroborated by factors identified in our study. The absence of any airway complications in our patient population might suggest that many of these factors are applicable to mandibular reconstruction with free tissue transfer. Future studies involving larger sample sizes with matched group analysis could serve to establish more definitive criteria for a protocol-driven selection of patients for eliminating elective tracheotomy.

We present the second largest series of microvascular mandibular reconstruction without elective tracheotomy. We found no airway-related adverse events. Mandibular resection and reconstruction can be performed safely without elective tracheotomy in a selected group of patients. Further prospective studies are needed to identify favorable factors in the success of this paradigm to develop patient selection protocols.

Corresponding Author: Babak Givi, MD, Department of Otolaryngology–Head & Neck Surgery, New York University, Langone Medical Center, 550 First Ave, NBV 5E5, New York, NY 10016 (Babak.Givi@nyumc.org).

Published Online: December 10, 2015. doi:10.1001/jamaoto.2015.2845.

Author Contributions: All authors had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Givi, DeLacure.

Acquisition, analysis, or interpretation of data: Lapis.

Drafting of the manuscript: Lapis.

Critical revision of the manuscript for important intellectual content: Givi, DeLacure.

Conflict of Interest Disclosures: None reported.

Submitted for Publication: June 15, 2015; final revision received September 4, 2015; accepted October 15, 2015.

Previous Presentation: This study was presented as a poster at the Fifth International Federation of Head and Neck Oncologic Societies World Congress/American Head and Neck Society meeting; July 30, 2014; New York, New York.