Mandibular lingual releasing approach SCOTT P. STRINGER, MD, J. RANDALL JORDAN, MD, WILLIAM M. MENDENHALL, MD, JAMES T. PARSONS, MD, NICHOLAS J. CASSISI, MD, DDS, and RODNEY R. MILLION, MD, Gainesville, Florida
The mandibular lingual releasing approach to oral cavity and oropharyngeal tumors provides excellent visualization for resection while integrity of the mandibular arch is preserved. A lingual floor-of-mouth flap is created, which allows delivery of these structures directly into the neck without lip splitting, mandibulotomy, or mandibulectomy. The procedure was carried out on 15 patients between 1987 and 1991, with followup ranging from 2 to 50 months. Nine patients had received previous radiation, whereas planned postoperative radiation was administered to five patients. The visualization afforded by this technique was very good, in that 12 patients had clear margins of resection. Three patients had close margins; recurrent disease developed in one of these patients 18 months later. Twelve of the patients were able to maintain their weight with an oral diet alone. Four postoperative fistulae occurred, three of these were in patients who had not been previously irradiated. The single fistula that did not spontaneously heal occurred in a patient who had received previous radiation and was also on long-term corticosteroids. Mandibular osteoradionecrosis developed in two patients who received postoperative radiation. The complication rate after previous radiation is acceptable; however, there is risk of mandibular osteoradionecrosis after high-dose postoperative radiation. (OTOlARYNGOL HEAD NECK SURG 1992;107:395,)
Mandibular Lingual Releasing Approach
Traditionally, the composite resection was the standard approach for the extirpation of posterior oral cavity and oropharyngeal tumors. In addition to allowing exposure of the tumor, resection of the mandible was deemed necessary in order to remove the lymphatic channels of the mandibular periosteum. However, Marchetta et al. I failed to demonstrate tumor in the periosteal lymphatics except when there was direct invasion of the periosteum. This finding, along with the desire to improve postoperative function, led to the use of mandibular sparing procedures such as median pharyngotomy, lateral pharyngotomy, pull-through procedures, lateral mandibulotomy, mandibular swing, and median labiomandibular glossotomy. These approaches are all subject to some limitation of exposure and/ or
From the Departments of Otolaryngology (Drs. Stringer, Jordan, and Cassisi) and Radiation Oncology (Drs. Mendenhall, Parsons, and Million), University of Florida College of Medicine. Presented at the Annual Meeting of the American Academy of Otolaryngology-Head and Neck Surgery, Kansas City, Mo., Sept. 2226, 1991. Received for publication Oct. 21, 1991; accepted Feb. 12, 1992. Reprint requests: Scott P. Stringer, MD, University of Florida Health Science Center, P.O. Box 100264, Gainesville, FL 32610-0264. 23/1/37326
violate the integrity of the mandible. Stanley? reported the use of the mandibular lingual releasing approach for tumors of the oral cavity and oropharynx. A lingual floor-of-mouth flap is created that allows delivery of these structures directly into the neck without lip splitting, mandibulotomy, or mandibulectomy. This approach provides excellent visualization for resection while the integrity of the mandibular arch is preserved. Since 1987 at the University of Florida, selected tumors of the oral cavity and oropharynx that do not involve the lingual periosteum have been resected using this technique. METHODS
A retrospective analysis of all mandibular lingual releasing procedures at the University of Florida from July 1987 through July 1991 was performed. Fifteen cases were identified and evaluated for demographics, length of hospitalization, tumor stage, tumor site, surgical pathology, use of radiation, reconstructive technique, complications, function, and outcome. Surgical Technique
The surgical technique used has been extensively described by Stanley. 2 The only significant modification was the use of a standard apron flap rather than a visortype flap. Briefly, a tracheotomy is performed and an 395
396 STRINGER et 01.
OtolaryngologyHead and Neck Surgery
Fig. 1. Oral mucoperiosteal incision, either as an alveolar crest incision in edentulous patients or as a sulcular incision adjacent to the teeth.
apron flap is elevated in the subplatysmal plane to the level of the inferior border of the mandible. Unilateral or bilateral neck dissections are performed as required. Care is taken to preserve the contralateral facial artery to ensure an adequate blood supply to the distal skin flap. A sulcular incision is made through the lingual mucoperiosteum in patients with teeth or on the crest of the mandibular alveolar ridge in edentulous patients (Fig. 1). The periosteum is then incised on the lower border of the mandible from angle to angle with preservation of the contralateral facial artery. The digastric, geniohyoid, genioglossus, and mylohyoid muscles are detached as the periosteum is elevated from the lingual surface of the mandible. The tongue and floor of mouth are then delivered into the neck (Figs. 2 and 3). Resection of the primary tumor and reconstruction of the resultant defect are carried out under direct vision. The periosteum is then reapproximated across the lower mandibular border, and the mucoperiosteum is closed intraorally. RESULTS
Six oropharyngeal and 9 oral cavity tumors were resected by means of the mandibular lingual release in 11 men and four women. The mean age was 61 years (range: 36 to 78 years). The tumors were staged as follows: Stage 1-2, Stage /1-6, Stage 1/1-4, and Stage N - 3. Nine patients had undergone previous radiation with curative intent, either to the same primary or to a second primary in the same field, with doses of 4700 to 7200 cGy. The mean followup of all cases is 24 months, with a range of 2 to 50 months. At the time of surgery, 11 patients underwent unilateral neck dissections, and three patients required bilateral neck dissections. Regional musculocutaneous flaps were used for reconstruction in four patients,
whereas a revascularized cutaneous flap was used in one patient. Eight patients underwent split-thickness skin graft placement, and the remaining two defects were closed primarily. Planned postoperative radiation was administered to five patients. Dosages ranged from 6400 to 7440 cGy, and one patient received 3000 cGy plus an iridium implant. Histopathology was squamous cell carcinoma in 14 cases and adenoid cystic carcinoma in one case. Three cases of close surgical margins were identified. Ofthese cases, a local recurrence developed in one patient at 18 months posttreatment, and a second patient is free of disease at 50 months. The followup of the final patient is inadequate to determine control at this time. The median length of hospitalization was 11 days (range: 8 to 60 days). One patient experienced a cardiopulmonary arrest on postoperative day 2, with resultant anoxic brain damage that required an extended hospital stay of 60 days. All patients were successfully decannulated. Eight patients are able to maintain their weight with a normal oral diet, and five patients require a modified oral diet. Three patients require a gastrostomy or a jejunostomy for the majority of their alimentation. No fistulae occurred at the gingival closure site. A salivary fistula developed in four patients (26%) at the site of their reconstructive closure. Three of the fistulae closed with conservative wound management. Two cases of mandibular osteoradionecrosis developed in patients who received postoperative radiotherapy to doses of 7440 cGy and 6400 cGy, respectively. Both patients began irradiation 4 weeks postoperatively, at which time the surgical sites were well healed. One patient has a small stable area of bone exposure at this time, whereas the second patient required partial mandibulectomy and primary reconstruction.
Volume 107 Number 3 September 1992
Mandibular lingual releasing approach
397
Fig. 2. Delivery of the tongue and floor of mouth into the neck.
Two patients died of intercurrent disease without evidence of recurrence at 23 and 15 months, respectively. One patient is alive with disease at 18 months. Only one patient has died of recurrent disease at 5 months after an unsuccessful salvage attempt by means of mandibular lingual release of a massive recurrence after previous radiation and surgical treatment. DISCUSSION
Those tumors that do not invade the lingual mandibular periosteum may be resected without mandibulectomy. The optimal procedure will provide excellent exposure and postoperative function. A variety of procedures have been described to approach oral cavity and oropharyngeal tumors. Each of these approaches has advantages and disadvantages. While median and lateral pharyngotomy do not always require mandibulotomy, the exposure by means of these routes may be limited. Approaches that require a median or lateral mandibulotomy necessitate an osteotomy and repair. Malunion or non-union of the osteotomy site may occur in 9% to 13% of cases.v' Lateral mandibulotomy or elevation of a cheek flap results in sacrifice of the mental nerve. Lip-splitting incisions may be associated with unsightly scarring. Finally, closure of the paralingual incision used with mandibular swing procedures may be difficult and somewhat tenuous. Stanley? described the mandibular lingual releasing approach and reported the results of this technique with five oral cavity and three base of tongue carcinomas. None of the cases had lingual mucoperiosteal involve-
Fig. 3. Visualization of a posterolateral oral tongue carcinoma afforded by the mandibular lingual reieasing approach.
ment, and no patients had received previous radiation therapy. All of the cases received postoperative radiation therapy. There was a single wound infection, and two salivary fistulae (25%) occurred. This approach provides excellent exposure and keeps the mandibular arch intact. The mental nerves remain intact, and there is no lip-splitting incision. In our series, there were three cases of close margins, with one recurrence and one apparent cure. All of these patients had been previously irradiated. There were no outright positive margins. This is similar to the rate of 13% reported by Stanley.' The mandibular lingual releasing approach provided excellent postoperative function. No permanent tracheotomies were required. Eighty percent of the patients were able to maintain their weight with an oral diet alone. Factors associated with poor function were anoxic brain damage in one patient and previous radiotherapy for a posterior pharyngeal wall lesion in a second patient. An additional patient required a gastrostomy caused by a persistent fistula. The fistula rate in this series was 26%, as compared to 25% in Stanley'S series." No fistulae developed at
398
OlolaryngologyHead and Neck Surgery
STRINGER et 01.
the gingival closure site, but rather they occurred at the site of the reconstruction of the resection defect. Three of the fistulae, all in nonirradiated patients, closed with conservative wound management. One fistula failed to close despite two attempts at repair with pedicled musculocutaneous flaps. This occurred in a patient who had previously been irradiated and who was on continuous corticosteroids for idiopathic pancytopenia. Two cases (13%) of mandibular osteoradionecrosis developed late after postoperative radiotherapy. Both cases had complete healing before the beginning of radiotherapy. The blood supply to the mandible should be adequate because the mandible is not interrupted and the outer periosteum is left intact. No cases of osteoradionecrosis were noted by Stanley"; however, no followup was reported. However, two cases of osteoradionecrosis associated with previous radiotherapy were described by Davidson et al. 3 in 44 cases approached by means of mandibulotomy. Stanley? used the mandibular lingual releasing approach only in nonirradiated cases, but speculated that it should be a viable technique after preoperative radiation or unsuccessful radiation because the mandibular blood supply was preserved. The present series represents the first report of use of the mandibular lingual releasing approach after previous radiotherapy. The wound complication rate was 11% in previously
irradiated cases, with the single complication associated with corticosteroid use. Paradoxically, there was a 40% complication rate in those cases that received postoperative radiation. It therefore appears that it is indeed a viable approach for use with unsuccessful radiation cases or preoperative radiation. SUMMARY
The mandibular lingual releasing approach provides excellent exposure for the resection of oral cavity and oropharyngeal carcinomas that do not involve the mandibular periosteum. The mandibular arch is not violated, and postoperative function and cosmesis are very good. Complication rates after previous radiotherapy are acceptable; however, there is a risk of mandibular osteoradionecrosis after postoperative radiotherapy.
REFERENCES
1. Marchetta FC, Sako K, Murphy JB. The periosteum of the mandible and intraoral carcinoma. Am J Surg 1971;122:711-3. 2. Stanley RB. Mandibular lingual releasing approach to oral and oropharyngeal carcinomas. Laryngoscope 1984;94:596-600. 3. Davidson J, Freeman J, Birt D. Mandibulotomy in the irradiated patient. Arch Otolaryngol Head Neck Surg 1989;115:497-9. 4. DeSanto LW, Whitaker JH, Devine KD. Mandibular osteotomy and lingual flaps. Arch Otolaryngol Head Neck Surg 1975;101:652-5.
The mandibular lingual releasing approach to oral cavity and oropharyngeal tumors provides excellent visualization for resection while integrity of the mandibular arch is preserved. A lingual floor-of-mouth flap is created, which allows delivery of these structures directly into the neck without lip splitting, mandibulotomy, or mandibulectomy. The procedure was carried out on 15 patients between 1987 and 1991, with followup ranging from 2 to 50 months. Nine patients had received previous radiation, whereas planned postoperative radiation was administered to five patients. The visualization afforded by this technique was very good, in that 12 patients had clear margins of resection. Three patients had close margins; recurrent disease developed in one of these patients 18 months later. Twelve of the patients were able to maintain their weight with an oral diet alone. Four postoperative fistulae occurred, three of these were in patients who had not been previously irradiated. The single fistula that did not spontaneously heal occurred in a patient who had received previous radiation and was also on long-term corticosteroids. Mandibular osteoradionecrosis developed in two patients who received postoperative radiation. The complication rate after previous radiation is acceptable; however, there is risk of mandibular osteoradionecrosis after high-dose postoperative radiation. (OTOlARYNGOL HEAD NECK SURG 1992;107:395,)
Mandibular Lingual Releasing Approach
Traditionally, the composite resection was the standard approach for the extirpation of posterior oral cavity and oropharyngeal tumors. In addition to allowing exposure of the tumor, resection of the mandible was deemed necessary in order to remove the lymphatic channels of the mandibular periosteum. However, Marchetta et al. I failed to demonstrate tumor in the periosteal lymphatics except when there was direct invasion of the periosteum. This finding, along with the desire to improve postoperative function, led to the use of mandibular sparing procedures such as median pharyngotomy, lateral pharyngotomy, pull-through procedures, lateral mandibulotomy, mandibular swing, and median labiomandibular glossotomy. These approaches are all subject to some limitation of exposure and/ or
From the Departments of Otolaryngology (Drs. Stringer, Jordan, and Cassisi) and Radiation Oncology (Drs. Mendenhall, Parsons, and Million), University of Florida College of Medicine. Presented at the Annual Meeting of the American Academy of Otolaryngology-Head and Neck Surgery, Kansas City, Mo., Sept. 2226, 1991. Received for publication Oct. 21, 1991; accepted Feb. 12, 1992. Reprint requests: Scott P. Stringer, MD, University of Florida Health Science Center, P.O. Box 100264, Gainesville, FL 32610-0264. 23/1/37326
violate the integrity of the mandible. Stanley? reported the use of the mandibular lingual releasing approach for tumors of the oral cavity and oropharynx. A lingual floor-of-mouth flap is created that allows delivery of these structures directly into the neck without lip splitting, mandibulotomy, or mandibulectomy. This approach provides excellent visualization for resection while the integrity of the mandibular arch is preserved. Since 1987 at the University of Florida, selected tumors of the oral cavity and oropharynx that do not involve the lingual periosteum have been resected using this technique. METHODS
A retrospective analysis of all mandibular lingual releasing procedures at the University of Florida from July 1987 through July 1991 was performed. Fifteen cases were identified and evaluated for demographics, length of hospitalization, tumor stage, tumor site, surgical pathology, use of radiation, reconstructive technique, complications, function, and outcome. Surgical Technique
The surgical technique used has been extensively described by Stanley. 2 The only significant modification was the use of a standard apron flap rather than a visortype flap. Briefly, a tracheotomy is performed and an 395
396 STRINGER et 01.
OtolaryngologyHead and Neck Surgery
Fig. 1. Oral mucoperiosteal incision, either as an alveolar crest incision in edentulous patients or as a sulcular incision adjacent to the teeth.
apron flap is elevated in the subplatysmal plane to the level of the inferior border of the mandible. Unilateral or bilateral neck dissections are performed as required. Care is taken to preserve the contralateral facial artery to ensure an adequate blood supply to the distal skin flap. A sulcular incision is made through the lingual mucoperiosteum in patients with teeth or on the crest of the mandibular alveolar ridge in edentulous patients (Fig. 1). The periosteum is then incised on the lower border of the mandible from angle to angle with preservation of the contralateral facial artery. The digastric, geniohyoid, genioglossus, and mylohyoid muscles are detached as the periosteum is elevated from the lingual surface of the mandible. The tongue and floor of mouth are then delivered into the neck (Figs. 2 and 3). Resection of the primary tumor and reconstruction of the resultant defect are carried out under direct vision. The periosteum is then reapproximated across the lower mandibular border, and the mucoperiosteum is closed intraorally. RESULTS
Six oropharyngeal and 9 oral cavity tumors were resected by means of the mandibular lingual release in 11 men and four women. The mean age was 61 years (range: 36 to 78 years). The tumors were staged as follows: Stage 1-2, Stage /1-6, Stage 1/1-4, and Stage N - 3. Nine patients had undergone previous radiation with curative intent, either to the same primary or to a second primary in the same field, with doses of 4700 to 7200 cGy. The mean followup of all cases is 24 months, with a range of 2 to 50 months. At the time of surgery, 11 patients underwent unilateral neck dissections, and three patients required bilateral neck dissections. Regional musculocutaneous flaps were used for reconstruction in four patients,
whereas a revascularized cutaneous flap was used in one patient. Eight patients underwent split-thickness skin graft placement, and the remaining two defects were closed primarily. Planned postoperative radiation was administered to five patients. Dosages ranged from 6400 to 7440 cGy, and one patient received 3000 cGy plus an iridium implant. Histopathology was squamous cell carcinoma in 14 cases and adenoid cystic carcinoma in one case. Three cases of close surgical margins were identified. Ofthese cases, a local recurrence developed in one patient at 18 months posttreatment, and a second patient is free of disease at 50 months. The followup of the final patient is inadequate to determine control at this time. The median length of hospitalization was 11 days (range: 8 to 60 days). One patient experienced a cardiopulmonary arrest on postoperative day 2, with resultant anoxic brain damage that required an extended hospital stay of 60 days. All patients were successfully decannulated. Eight patients are able to maintain their weight with a normal oral diet, and five patients require a modified oral diet. Three patients require a gastrostomy or a jejunostomy for the majority of their alimentation. No fistulae occurred at the gingival closure site. A salivary fistula developed in four patients (26%) at the site of their reconstructive closure. Three of the fistulae closed with conservative wound management. Two cases of mandibular osteoradionecrosis developed in patients who received postoperative radiotherapy to doses of 7440 cGy and 6400 cGy, respectively. Both patients began irradiation 4 weeks postoperatively, at which time the surgical sites were well healed. One patient has a small stable area of bone exposure at this time, whereas the second patient required partial mandibulectomy and primary reconstruction.
Volume 107 Number 3 September 1992
Mandibular lingual releasing approach
397
Fig. 2. Delivery of the tongue and floor of mouth into the neck.
Two patients died of intercurrent disease without evidence of recurrence at 23 and 15 months, respectively. One patient is alive with disease at 18 months. Only one patient has died of recurrent disease at 5 months after an unsuccessful salvage attempt by means of mandibular lingual release of a massive recurrence after previous radiation and surgical treatment. DISCUSSION
Those tumors that do not invade the lingual mandibular periosteum may be resected without mandibulectomy. The optimal procedure will provide excellent exposure and postoperative function. A variety of procedures have been described to approach oral cavity and oropharyngeal tumors. Each of these approaches has advantages and disadvantages. While median and lateral pharyngotomy do not always require mandibulotomy, the exposure by means of these routes may be limited. Approaches that require a median or lateral mandibulotomy necessitate an osteotomy and repair. Malunion or non-union of the osteotomy site may occur in 9% to 13% of cases.v' Lateral mandibulotomy or elevation of a cheek flap results in sacrifice of the mental nerve. Lip-splitting incisions may be associated with unsightly scarring. Finally, closure of the paralingual incision used with mandibular swing procedures may be difficult and somewhat tenuous. Stanley? described the mandibular lingual releasing approach and reported the results of this technique with five oral cavity and three base of tongue carcinomas. None of the cases had lingual mucoperiosteal involve-
Fig. 3. Visualization of a posterolateral oral tongue carcinoma afforded by the mandibular lingual reieasing approach.
ment, and no patients had received previous radiation therapy. All of the cases received postoperative radiation therapy. There was a single wound infection, and two salivary fistulae (25%) occurred. This approach provides excellent exposure and keeps the mandibular arch intact. The mental nerves remain intact, and there is no lip-splitting incision. In our series, there were three cases of close margins, with one recurrence and one apparent cure. All of these patients had been previously irradiated. There were no outright positive margins. This is similar to the rate of 13% reported by Stanley.' The mandibular lingual releasing approach provided excellent postoperative function. No permanent tracheotomies were required. Eighty percent of the patients were able to maintain their weight with an oral diet alone. Factors associated with poor function were anoxic brain damage in one patient and previous radiotherapy for a posterior pharyngeal wall lesion in a second patient. An additional patient required a gastrostomy caused by a persistent fistula. The fistula rate in this series was 26%, as compared to 25% in Stanley'S series." No fistulae developed at
398
OlolaryngologyHead and Neck Surgery
STRINGER et 01.
the gingival closure site, but rather they occurred at the site of the reconstruction of the resection defect. Three of the fistulae, all in nonirradiated patients, closed with conservative wound management. One fistula failed to close despite two attempts at repair with pedicled musculocutaneous flaps. This occurred in a patient who had previously been irradiated and who was on continuous corticosteroids for idiopathic pancytopenia. Two cases (13%) of mandibular osteoradionecrosis developed late after postoperative radiotherapy. Both cases had complete healing before the beginning of radiotherapy. The blood supply to the mandible should be adequate because the mandible is not interrupted and the outer periosteum is left intact. No cases of osteoradionecrosis were noted by Stanley"; however, no followup was reported. However, two cases of osteoradionecrosis associated with previous radiotherapy were described by Davidson et al. 3 in 44 cases approached by means of mandibulotomy. Stanley? used the mandibular lingual releasing approach only in nonirradiated cases, but speculated that it should be a viable technique after preoperative radiation or unsuccessful radiation because the mandibular blood supply was preserved. The present series represents the first report of use of the mandibular lingual releasing approach after previous radiotherapy. The wound complication rate was 11% in previously
irradiated cases, with the single complication associated with corticosteroid use. Paradoxically, there was a 40% complication rate in those cases that received postoperative radiation. It therefore appears that it is indeed a viable approach for use with unsuccessful radiation cases or preoperative radiation. SUMMARY
The mandibular lingual releasing approach provides excellent exposure for the resection of oral cavity and oropharyngeal carcinomas that do not involve the mandibular periosteum. The mandibular arch is not violated, and postoperative function and cosmesis are very good. Complication rates after previous radiotherapy are acceptable; however, there is a risk of mandibular osteoradionecrosis after postoperative radiotherapy.
REFERENCES
1. Marchetta FC, Sako K, Murphy JB. The periosteum of the mandible and intraoral carcinoma. Am J Surg 1971;122:711-3. 2. Stanley RB. Mandibular lingual releasing approach to oral and oropharyngeal carcinomas. Laryngoscope 1984;94:596-600. 3. Davidson J, Freeman J, Birt D. Mandibulotomy in the irradiated patient. Arch Otolaryngol Head Neck Surg 1989;115:497-9. 4. DeSanto LW, Whitaker JH, Devine KD. Mandibular osteotomy and lingual flaps. Arch Otolaryngol Head Neck Surg 1975;101:652-5.
