The maxilla is the pivotal structure of the midface,

Transcription

1 ORIGINAL ARTICLE MICROVASCULAR FREE FLAP RECONSTRUCTION VERSUS PALATAL OBTURATION FOR MAXILLECTOMY DEFECTS Mauricio A. Moreno, MD, 1,2 Roman J. Skoracki, MD, 1 Ehab Y. Hanna, MD, 2 Matthew M. Hanasono, MD 1 1 Department of Plastic Surgery, The University of Texas M. D. Anderson Cancer Center, Houston, Texas. mhanasono@mdanderson.org 2 Department of Head and Neck Surgery, The University of Texas M. D. Anderson Cancer Center, Houston, Texas Accepted 3 August 2009 Published online 9 November 2009 in Wiley InterScience ( DOI: /hed Abstract: Background. Palatal obturators and microvascular free flaps are both used to treat patients with maxillectomy defects, however, the optimal technique remains controversial. Methods. A retrospective analysis of 113 patients undergoing maxillectomy for cancer was performed. Seventy-three patients received an obturator and 40 patients were reconstructed with a free flap. Results. Speech intelligibility and postoperative diet were comparable between the obturator and free flap groups, except in cases of extensive (>50%) palatal defects, where free flap reconstruction was superior in both aspects (p ¼.019 and p ¼.043, respectively). The average time for presenting with a local recurrence in advanced cancer involving the palate was comparable in both groups (p ¼.33). Conclusion. Moderate-sized maxillectomy defects involving the palate can be successfully treated with either an obturator or free flap reconstruction. Extensive defects have a better functional outcome with free flaps. Evidence does not suggest that free flap reconstructions delay diagnosis of local recurrences. VC 2009 Wiley Periodicals, Inc. Head Neck 32: , 2010 Correspondence to: M. M. Hanasono This work was presented at the American Society for Reconstructive Microsurgery Annual Meeting, Maui, HI, January 10 13, VC 2009 Wiley Periodicals, Inc. Keywords: head and neck cancer; microvascular free flap; maxillectomies; obturator; prosthesis The maxilla is the pivotal structure of the midface, separating the oral, antral, and orbital cavities, and providing support to the globes, lower eyelids, cheeks, lips, and nose. In addition, the maxillae play a critical role in speech, swallowing, and mastication. Consequently, reconstruction of maxillectomy defects is 1 of the most difficult challenges faced by the head and neck reconstructive surgeon. Maxillectomy defects are typically treated by prosthetic obturation or autologous tissue reconstruction. Each technique has its advantages and disadvantages, and the best approach is a subject of debate. 1 3 Traditionally, rehabilitation with a palatal obturator has been the most common approach for treating maxillectomy defects. The advantages of this technique include a shorter operative time, shorter postoperative hospital stay, and complete visualization of the maxillectomy cavity, which simplifies oncologic surveillance. 4 Unfortunately, there are also numerous disadvantages 860 Reconstruction of Maxillectomy Defects HEAD & NECK DOI /hed July 2010

2 associated with obturators, including the potential for hypernasal speech, regurgitation of foods and liquids into the nasal cavity, difficulty in maintaining hygiene of the maxillectomy cavity, and the need for repeated prosthesis adjustments due to progressive changes in the size and shape of the palatal defect, especially in patients who receive radiation therapy. 5 Larger defects are harder to obturate as the prosthesis may be overly heavy and difficult or impossible to retain, particularly in partially or totally edentulous patients. 6,7 A variety of local and regional flaps have been used to reconstruct maxillary defects with variable success Unfortunately, all these techniques have been limited by a paucity of available tissue, restricted reach of the vascular pedicle, and the frequent need for staged procedures to achieve an optimal result. Maxillary reconstruction changed radically with the advent of microvascular free tissue transfer, which provides abundant tissue for reconstruction, the freedom to orient, shape, and inset the flap as required for the specific defect, and the ability for reconstruction to be performed as a single-stage procedure. 11 Additionally, transfer of vascularized bone provides the option of dental restoration via implantation of osseointegrated implants, even in irradiated tissues. 7 For some massive midface skull base lesions, the resection is only possible if the defect can immediately be reconstructed with free tissue transfer, such as when there is exposure of critical neurovascular structures, for example. In addition, there are some defects that are not amenable to rehabilitation with prosthesis unless some form of free tissue transfer reconstruction is performed. Disadvantages of free flaps include longer surgical and recovery times with increased potential for complications compared with prosthetic obturation. Another concern is the possibility that obliteration of the maxillectomy defect by the free flap may delay the diagnosis of a local recurrence. To date, the few studies comparing obturators to free flap reconstructions of maxillectomy defects have not been able to demonstrate a difference in functional outcomes that would favor 1 method over the other, nor have they been able to identify subsets of patients that might benefit from 1 approach or the other. 2,3 One reason is that maxillectomies are relatively uncommon procedures when compared with other head and neck oncologic surgeries, making it difficult to build a large personal or institutional experience with this type of reconstruction. 12 The objectives of the present study were to compare outcomes of prosthetic obturators and microvascular reconstruction of maxillectomy defects with regard to: (1) speech and swallowing function, (2) rate of complications, and (3) cancer surveillance for local recurrence. MATERIALS AND METHODS Institutional review board approval to conduct a retrospective review of reconstruction and rehabilitation after maxillectomy was obtained. A search of a prospectively collected database was performed to identify 201 patients who underwent a maxillectomy at The University of Texas M. D. Anderson Cancer Center between January 1, 2000, and December 31, Inclusion criteria for patients to be included in this study were: (1) a maxillectomy performed at our institution; (2) at least 1 preoperative and 1 postoperative head and neck CT or MRI scan; (3) a minimum of 6 months of postoperative follow-up time; and (4) a speech and swallowing evaluation by a speech pathologist in the postoperative period. Exclusion criteria were: (1) defects that produced no communication between the oral and sinonasal cavities such as medial maxillectomies, suprastructure maxillectomies, and limited tuberosity resections; (2) reconstruction of the defect by grafts, local flaps, or any means other than free tissue transfer or a palatal obturator; and (3) absent or incomplete chart documentation. A total of 113 patients met these criteria and were included in the study. Information retrieved included demographic data, cancer status at presentation (primary vs recurrent), histopathology, status of maxillary and mandibular dentition, margin status, use of adjuvant chemotherapy or radiation therapy, and complications. The surgical defect was classified in the vertical and horizontal (palatal) planes according to the classifications systems described by Okay et al 6 and Brown et al, 13 respectively (Figure 1 and Figure 2). For patients who underwent free flap reconstruction, the information obtained included: type of flap, osseointegrated implant placement, and technique for orbital floor reconstruction, if performed. Reconstruction of Maxillectomy Defects HEAD & NECK DOI /hed July

3 FIGURE 1. Classification of maxillectomy defects based on the extent of palatal resection originally described by Okay et al. 6 Type Ia: Hard palate but no tooth-bearing maxillary alveolus. Type Ib: Premaxilla or any portion of maxillary alveolus posterior to canines. Type II: Any portion of hard palate and tooth-bearing alveolus and only one canine. Transverse palatectomy defects that involve less than 50% of the palate. Type III: Tooth-bearing maxillary alveolus including both canines. Transverse palatectomy defects than involve more than 50% of the palate. Reprinted with permission from Elsevier. As a standard practice in our institution, all patients who received a palatal obturator were assessed preoperatively by the dental team to obtain dental impressions. Upon completion of the ablation, a temporary obturator made of Trusoft (Bosworth Dental, Skokie, IL) was placed intraoperatively and kept in place for approximately 1 week. This was subsequently exchanged for a definitive obturator in the dental clinic, which was then adjusted on an asneeded basis. Patients were evaluated postoperatively by experienced speech pathologists for speech intelligibility defined as the percentage of words understandable to an unfamiliar listener. This information was combined into a 5-point descriptive scale described by Matsui et al 14 : (1) excellent, all speech is understood or patient FIGURE 2. Classification of maxillectomy defects based on the vertical extent of resection originally described by Brown et al. 13 Reprinted with permission from Elsevier. 862 Reconstruction of Maxillectomy Defects HEAD & NECK DOI /hed July 2010

4 has intelligibility above 80%; (2) good, requires occasional repetition, 65% to 75% intelligibility; (3) fair, can be understood when conversational content is already known, 45% to 60% intelligibility; (4) poor, difficult to understand, requires frequent repetition, 30% to 40% intelligibility, and (5) dismal, extremely difficult to understand or less than 25% intelligibility. The postoperative diet was also recorded by the speech pathologist. When patients had multiple speech pathology visits, the best-documented speech and oral diet were considered for the study. Statistical analysis was performed using the SPSS Statistics version 17.0 software (SPSS Incorporated, Chicago, IL). Proportional data were compared using the chi-square test or Fisher s exact test as appropriate and continuous data were compared using the t test. All analyses were 2-tailed, and p values less than.05 were considered statistically significant. The p values less than.0001 are reported as such; otherwise, actual p values are shown. RESULTS There were 63 men and 50 women in the series with a median age of 54 years (range, 9 88 years). Twenty-five patients presented with recurrent or residual disease while 88 were previously untreated. All patients were treated with curative intent. The mean follow-up time was 27.3 months (range, 7 63 months) and the 5-year overall survival for the series was 61.9%. The histologic diagnoses included squamous cell carcinoma in 43 patients (38%), sarcoma in 22 patients (19.4%), adenoid cystic carcinoma in 20 patients (17.7%), benign tumors in 10 patients (8.9%), melanoma in 6 patients (5.3%), adenocarcinoma in 4 patients (3.5%), verrucoid squamous cell carcinoma in 2 patients (1.7%), ameloblastic carcinoma in 1 patient (0.8%), carcinoma expleomorphic in 1 patient (0.8%), acinic cell carcinoma in 1 patient (0.8%), mucoepidermoid carcinoma in 1 patient (0.8%), adnexal carcinoma in 1 patient (0.8%), and papillary transitional carcinoma in 1 patient (0.8%). Sixty-one patients had complete maxillary dentition while 68 had complete mandibular dentition at the time of presentation. Eighty-one patients underwent radiation therapy, including 12 patients who received preoperative radiation and 69 patients who received postoperative radiation. The total radiation dose ranged from Table 1. Comparison of obturator and free flap groups. Obturator Free flap p value Median age 57.1 (16 88) 50.8 (9 88).0660 Sex, M/F 36/37 26/ Radiation, Y/N 50/23 31/ Average surgical time, min <.0001 Average hospital stay, d <.0001 Recurrent or residual, Y/N 9/64 16/ to 70 Gray (Gy) with an average of 58.8 Gy; 55 patients received a dose of 60 Gy or higher. Thirty-six patients received chemotherapy, which was used as induction therapy in 18 patients and adjuvant chemotherapy in 18 patients. Surgical margins were positive in 11 patients, close (less than 1 mm) in 6 patients, and negative in the remaining 96 patients. Seventy-three patients were rehabilitated with a palatal obturator and 40 patients were reconstructed with a free flap. There were no statistically significant differences in age, sex, use of chemotherapy, use of radiation therapy, or preoperative dentition between these 2 groups (Table 1). Differences in surgical time and hospital stay were significant. In the obturator group, the device was fitted to the remaining teeth in 53 patients, the palatal defect in 17 patients, and osseointegrated in 3 patients. Of the 40 patients in the free flap group, 19 (47.5%) had complete maxillary dentition, 14 (35%) had partial maxillary dentition, and 7 (17.5%) were edentulous before surgery. Upon completion of the reconstruction, 9 patients (22.5%) remained with functional dentition. An obturator was fitted in 27 of the 31 remaining patients; this was implant-borne in 9 cases, implant-retained in 5 cases, and tissue-borne in 13 cases. In 4 cases, the obturator could not be fitted due to bulky soft tissue reconstruction of extensive defects and/or an inability to perform implant placement. Table 2 shows the free flaps utilized for reconstruction and their respective outcome. Of 3 patients that underwent surgical re-exploration for vascular compromise, 1 was successfully salvaged and 2 had complete flap loss. In 1 of these cases, a rectus abdominis myocutaneous free flap was subsequently replaced by an anterolateral thigh free flap without further complications. In the other case, the patient Reconstruction of Maxillectomy Defects HEAD & NECK DOI /hed July

5 Table 2. Free flaps used and flap outcome. Free flap n Flap outcome Anterolateral thigh 11 1 venous congestion w/o flap loss Osteocutaneous fibula 11 1 TL, 2 PL (<10% cutaneous paddle) Rectus abdominis 10 1 TL Osteocutaneous fibula þ ALT 3 All viable Radial forearm free flap 2 All viable Lateral arm 1 Viable Serratus composite 1 Viable Free rectus þ skin graft 1 Viable Abbreviations: TL, total loss; PL, partial loss; ALT, anterolateral thigh. Table 3. Postoperative complications. Complication Free flap, n % Obturator, n % Delayed nasocutaneous fistula Bleeding/hematoma Cellulitis/abscess Pneumonia Delirium CSF leak Urinary retention Acute renal failure Bowel obstruction Urinary tract infection Total Abbreviation: CSF, cerebrospinal fluid. p ¼.007. Table 4. Classification of the defect in the vertical plane (Brown et al). 13 Obturator Free flap Vertical extension n % n % Type II Type III Type IV Total p < declined another surgery and was rehabilitated with a palatal obturator. A partial flap loss involving less than 10% of the cutaneous paddle was observed in 2 fibula osteocutaneous free flaps, both of which healed without further surgery. One patient who underwent reconstruction with an anterolateral thigh flap was re-explored for venous congestion and developed delayed fat necrosis. This patient did require surgical debridement but no additional reconstructive procedures. Other postoperative complications, summarized in Table 3, were more frequent in the free flap group (p ¼.007). There were no perioperative deaths in this series. There were 59 (52.2%) type II defects, 32 (28.3%) type III defects, and 22 (19.4%) type IV defects based on their vertical extension (Brown et al 13 classification). Since the absence of an oroantral communication was part of the exclusion criteria, there were no patients with a type I defect in this study. Classification of the defect in the horizontal (palatal) plane (Okay et al 6 classification) yielded 2 patients (1.8%) with a type IA defect, 33 patients (29.2%) with a type IB defect, 51 patients (45.1%) with a type II defect, and 27 patients (23.8%) with a type III defect. Reconstruction of the orbital floor with titanium mesh was performed in 5 patients as an associated procedure to free flap reconstruction. A comparison of patients treated with obturators and free flaps demonstrated significantly different distributions of vertical and horizontal defects (p <.0001 and p ¼.004, respectively), with free flaps more frequently used to address more extensive defects in both the vertical and horizontal planes (Table 4 and Table 5). Some degree of functional speech was preserved in all patients, regardless of the defect extent. The extent of the defect in the horizontal (palatal) plane was found to adversely affect speech and swallowing in a statistically significant fashion (Table 6). The vertical extent of the defect did not significantly affect speech and swallowing outcomes (p ¼.66 and p ¼.07 for speech and swallowing, respectively). When comparing the functional outcome between the obturator and free flap groups, no statistically significant difference was found if the data were not stratified (Table 7). When outcomes were stratified by defect size (Table 8), a statistically significant advantage favoring the free flap group in patients with a type III palatal defect was identified. The charts of the 10 patients in the obturator group with poor swallowing Table 5. Classification of the defect in the horizontal plane (Okay et al). 6 Obturator Free flap Horizontal extension n % n % Type Ia/Ib Type II Type III Total p ¼ Reconstruction of Maxillectomy Defects HEAD & NECK DOI /hed July 2010

6 Table 6. Functional results stratified by type of horizontal defect. Type Ia/Ib Type II Type III n % n % n % Diet* Unrestricted Soft Liquid NPO Speech Excellent Good Average Poor *p ¼.003 (combining 2 lower categories). p ¼.007 (combining 2 lower categories). Table 7. Functional results for the obturator and free flap groups, all patients. Obturator Free flap n % n % Diet* Unrestricted Soft Liquid NPO Speech Excellent Good Average Poor Abbreviation: NPO, nothing prescribed orally. *p ¼.68. p ¼.54. Table 8. Summary of p values comparing free flaps versus obturator, stratified by defect extension. Diet, p value Speech, p value Horizontal classification Type Ia/Ib Type II Type III Vertical classification Type II Type III Type IV N/A N/A Abbreviation: N/A, not applicable. outcomes (nothing prescribed orally [NPO]/liquid diet) were reviewed. Six of these patients had extensive (type III) horizontal defects, 2 patients had multiple unsuccessful revisions of the device, 1 patient presented with severe trismus, and 1 patient had a cerebrovascular event with severe aspiration. There were no differences in the use of external beam radiation or chemotherapy in this group when compared with the rest of the series (p ¼.66 and p ¼.32, respectively). In the same fashion, there were no identifiable additional defects, such as pharyngeal or soft palate resection that could explain their worse functional outcome. To determine whether free flap reconstruction hampers oncologic surveillance, we analyzed 11 patients primarily treated for a T4 squamous cell carcinoma of the maxillary gingiva/hard palate that developed local recurrence. This histology and stage was chosen because of the high prevalence in the series and comparable number of patients reconstructed with free flaps and palatal obturation. The average time to diagnose the local recurrence was 7.6 months for the free flap group and 13 months for the obturator group; this difference was not statistically significant (p ¼.33). Additionally, the charts of all patients who had a local recurrence were analyzed to document if the diagnosis was first made by physical examination or routine imaging. We found that the diagnosis was most frequently made by physical examination in both the free flap and obturator groups (Table 9), and no significant difference was found between the groups (p ¼.694). DISCUSSION Most studies comparing functional outcomes of prosthetic rehabilitation and microvascular free flap reconstruction after maxillectomy have not shown significant differences between these techniques. 2,3 One of the problems with evaluating these techniques is the lack of a universally accepted nomenclature for maxillectomy defects, which vary greatly in their size and extent. 15,16 In the present study, we decided to compare palatal obturation to free flap reconstruction using Table 9. Method used for diagnosis of local recurrence. Clinical Images Free flap 9 4 Obturator 8 6 p ¼.694. Reconstruction of Maxillectomy Defects HEAD & NECK DOI /hed July

7 2 systems that are based on expected functional outcomes. Our goal was not only to compare these 2 techniques, but also to identify subsets of patients who might benefit more from 1 technique over the other. The extent of the maxillectomy defect in the horizontal plane (ie, the palatal component) was classified according to the system described by Okay et al. 6 This classification system not only considers the extent of the palatal defect but also utilizes biomechanic concepts to predict the ability of successful obturator retention and, therefore, the likelihood of successful prosthetic rehabilitation. In the present study, we found that this system was highly correlated to postoperative speech and swallowing outcomes and is, therefore, a valuable tool to evaluate the success of obturation and free flap reconstruction. In contrast, the extension of the defect in the vertical plane according to the classification system described by Brown et al 13 did not correlate with functional outcome. This probably indicates that once the palatal surface and alveolar arch have been adequately restored, the continuity of the maxillary superstructure has a limited effect on speech and swallowing, although they may have an effect on cosmesis or orbital support. This was acknowledged by Brown et al, 13 who also proposed a subclassification based on the palatal defect. We opted to use the horizontal classification proposed by Okay et al 6 because it takes into account the ability to stabilize and retain an obturator. When the extent of the horizontal component of the defect (Okay et al 6 classification) was considered, a significant difference favoring free flap reconstruction in patients with a type III defect was observed. This was confirmed by reviewing the charts of patients with poor functional outcome in the obturator group. In these cases, no identifiable factors other than the extensive nature of the defect appeared as plausible explanations for this observation in the majority of these patients. These results imply that patients with a resection of the hard palate greater than 50% of the total area, or of the anterior palate, in which both canines are included in the resection, should undergo reconstruction with a free flap whenever feasible, while smaller defects could be addressed equally well by either a palatal obturator or a free flap. Free flaps probably give better functional results in extensive or anterior defects because obturators cannot be adequately stabilized by the remaining alveolus or teeth. The majority of patients who underwent a free flap reconstruction for this type of defect had bony restoration of the palatal arch with a fibula flap, resulting in a stable occlusal plane. Larger maxillary defects often preclude the use of a conventional denture for dental restoration, particularly in edentulous patients. In some cases, a denture-obturator can make use of the maxillary defect to improve retention. However, in very large defects, such an obturator can be overly heavy, may require multiple magnet-retained parts, and the retention may be unstable. For large defects, our current strategy is to perform bony reconstruction followed by dental restoration using implant-retained prostheses whenever possible. The reconstructive plan in these cases obviously requires close cooperation and pretreatment planning with a prosthodontist and with the patient. The complication rate was higher in the free flap group than in the obturator group (Table 3). This was not entirely unexpected as free flap procedures are longer and were generally used for more extensive defects in both the vertical and horizontal plane. Interestingly, a nasocutaneous fistula along the Weber- Ferguson incision was the most common surgical complication in both groups. Six of the 8 patients who developed a nasocutaneous fistula received full-dose radiation, which emphasizes the role of this treatment in the development of this complication. One of the major concerns about free flap reconstruction after maxillectomy for cancer is the risk of delaying the diagnosis of a local recurrence. To date, there are no studies in the literature documenting any delay in the detection of recurrent tumors in patients undergoing free flap reconstruction after maxillectomy, but this is still an area of debate. 17 To address this question, we compared the average time for presentation of local recurrence in a group of patients matched for cancer stage and histology and found no statistically significant differences between the free flap group and the obturator group. Also, we found that the diagnosis of recurrence was more frequently made by physical examination in both groups. This contrasts with results from other series in which CT and MRI scan techniques were shown to identify local recurrences earlier than physical examination alone. 18 Overall, our findings support the hypothesis that free flaps are an oncologically 866 Reconstruction of Maxillectomy Defects HEAD & NECK DOI /hed July 2010

8 sound option for addressing the maxillectomy defect in patients with cancer. Despite the fact that palatal obturation provides good functional results in reconstruction of small to medium-sized palatal defects, there are other limitations that are inherent to prosthetic rehabilitation. These include difficulties with keeping the maxillectomy cavity clean, residue buildup on the obturator even with vigilant cleaning, the inability to eat or communicate effectively without the device, and the need for repeated readjustment of the obturator as the size and shape of the palatal defect changes over time, which may negatively affect the patient s quality of life. In a recent series, Genden et al 19 demonstrated that free flap reconstruction improves the patient s quality of life over prosthetic rehabilitation even in patients with small to medium-sized defects. Thus, in patients who are good candidates for microvascular reconstruction based on their medical status and overall prognosis, free flap reconstruction may be preferable even when the defect size is modest. Multiple microvascular free flaps have been described for maxillary reconstruction, including the iliac crest, 20 fibula, 7 radial forearm, 21 anterolateral thigh, 22 rectus abdominis myocutaneous, 23,24 and scapula 12,17 free flaps. Each of these alternatives has its own advantages and disadvantages in maxillary reconstruction. An evaluation of the various free flap types used in maxillary reconstruction is beyond the scope of this study. Ultimately, the decision of which flap to use is dependent on the extent and location of the defect, potential for dental restoration, and donor site availability, as well as the surgeon s personal preference. CONCLUSIONS The size and extent of the palatal defect best predicts the speech and swallowing outcome of patients undergoing maxillectomy. While functional results are comparable in small to medium-sized palatal defects, reconstruction with free flaps provides better speech and swallowing results than palatal obturation in extensive or anterior defects. Microvascular free flaps, in our experience, were associated with a 95% success rate without significant long-term complications. In addition, we found no difference in the rate of detecting local recurrences that would favor either free flap reconstruction or palatal obturation. Based on the results of this study, microvascular free flap reconstruction should be strongly considered in patients undergoing maxillectomy, particularly when the defect is large or anterior. Less extensive defects can be rehabilitated successfully with either an obturator or reconstructed with free tissue transfer, although free flap reconstruction eliminates problems with prosthetic retention, maxillectomy cavity hygiene, and the need for repeated obturator adjustment to prevent oral-nasal escape. REFERENCES 1. Kornblith AB, Zlotolow IM, Gooen J, et al. Quality of life of maxillectomy patients using an obturator prosthesis. Head Neck 1996;18: Rogers SN, Lowe D, McNally D, Brown JS, Vaughan ED. Health-related quality of life after maxillectomy: a comparison between prosthetic obturation and free flap. J Oral Maxillofac Surg 2003;61: Eckardt A, Teltzrow T, Schulze A, Hoppe M, Kuettner C. Nasalance in patients with maxillary defects reconstruction versus obturation. J Craniomaxillofac Surg 2007;35: Davison SP, Sherris DA, Meland NB. An algorithm for maxillectomy defect reconstruction. Laryngoscope 1998; 108: Gillespie CA, Kenan PD, Ferguson BJ. Hard palate reconstruction in maxillectomy. Laryngoscope 1986;96: Okay DJ, Genden E, Buchbinder D, Urken M. Prosthodontic guidelines for surgical reconstruction of the maxilla: a classification system of defects. J Prosthet Dent 2001;86: Futran ND, Wadsworth JT, Villaret D, Farwell DG. Midface reconstruction with the fibula free flap. Arch Otolaryngol Head Neck Surg 2002;128: Kostrubala JG. Repair of extensive palatal defects with skin tubes. Plast Reconstr Surg (1946) 1950;5: Bakamjian VY, Poole M. Maxillo-facial and palatal reconstructions with the deltopectoral flap. Br J Plast Surg 1977;30: Colmenero C, Martorell V, Colmenero B, Sierra I. Temporalis myofascial flap for maxillofacial reconstruction. J Oral Maxillofac Surg 1991;49: Shestak KC, Schusterman MA, Jones NF, Johnson JT. Immediate microvascular reconstruction of combined palatal and midfacial defects using soft tissue only. Microsurgery 1988;9: Clark JR, Vesely M, Gilbert R. Scapular angle osteomyogenous flap in postmaxillectomy reconstruction: defect, reconstruction, shoulder function, and harvest technique. Head Neck 2008;30: Brown JS, Rogers SN, McNally DN, Boyle M. A modified classification for the maxillectomy defect. Head Neck 2000;22: Matsui Y, Ohno K, Yamashita Y, Takahashi K. Factors influencing postoperative speech function of tongue cancer patients following reconstruction with fasciocutaneous/myocutaneous flaps a multicenter study. Int J Oral Maxillofac Surg 2007;36: Spiro RH, Strong EW, Shah JP. Maxillectomy and its classification. Head Neck 1997;19: Reconstruction of Maxillectomy Defects HEAD & NECK DOI /hed July

9 16. Cordeiro PG, Santamaria E. A classification system and algorithm for reconstruction of maxillectomy and midfacial defects. Plast Reconstr Surg 2000;105: ; discussion Uglesić V, Virag M, Varga S, Knezević P, Milenović A. Reconstruction following radical maxillectomy with flaps supplied by the subscapular artery. J Craniomaxillofac Surg 2000;28: Urken ML, Catalano PJ, Sen C, Post K, Futran N, Biller HF. Free tissue transfer for skull base reconstruction analysis of complications and a classification scheme for defining skull base defects. Arch Otolaryngol Head Neck Surg 1993;119: Genden EM, Wallace DI, Okay D, Urken ML. Reconstruction of the hard palate using the radial forearm free flap: indications and outcomes. Head Neck 2004;26: Brown JS. Deep circumflex iliac artery free flap with internal oblique muscle as a new method of immediate reconstruction of maxillectomy defect. Head Neck 1996;18: Chepeha DB, Moyer JS, Bradford CR, Prince ME, Marentette L, Teknos TN. Osseocutaneous radial forearm free tissue transfer for repair of complex midfacial defects. Arch Otolaryngol Head Neck Surg 2005;131: Amin A, Rifaat M, Civantos F, Weed D, Abu-Sedira M, Bassiouny M. Free anterolateral thigh flap for reconstruction of major craniofacial defects. J Reconstr Microsurg 2006;22: Olsen KD, Meland NB, Ebersold MJ, Barley GB, Garrity JA. Extensive defects of the sino-orbital region. Results with microvascular reconstruction. Arch Otolaryngol Head Neck Surg 1992;118: ; discussion Cordeiro PG, Santamaria E. The extended, pedicled rectus abdominis free tissue transfer for head and neck reconstruction. Ann Plast Surg 1997;39: Reconstruction of Maxillectomy Defects HEAD & NECK DOI /hed July 2010