ASRM Scientific Paper Presentations: Head

January 19, 2016 – 7:30 AM to 9:15 AM

7:30 AM - 7:35 AM Intraoperative Use of Vasopressors Is Safe in Head and Neck Free Tissue Transfer The Johns Hopkins University School of Medicine, Baltimore, MD, USA Edward W. Swanson, MD; Hsu-Tang Cheng; Srinivas M. Susarla; Georgia C. Yalanis; Denver M. Lough; Owen Johnson, III; Anthony P. Tufaro; Paul N. Manson; Justin M. Sacks, MD; The Johns Hopkins University School of Medicine

BACKGROUND: The purpose of this study is to identify whether intraoperative use of vasoactive medications increases the risk of free flap failure or complications through a systematic review and meta-analysis.

METHODS: PubMed/MEDLINE, EMBASE, and SCOPUS databases were searched for studies published through January 2015. English publications that met the following criteria were included: (1) adult patients undergoing head and neck free flap reconstruction; (2) comparison of patients with and without intraoperative vasopressor administration; (3) documentation of flap failure rate and/or flap complications. The primary outcome was the incidence of flap failure. The secondary outcome was the incidence of overall flap complications. Meta-analysis was performed to obtain pooled odds ratios (OR) of the effect of intraoperative use of vasopressors on flap failure and complication rates.

RESULTS: Four cohort studies met inclusion criteria. All studies were of high methodological quality with an average MINORS score of 18.75 (range 16 – 23). A total of 933 patients undergoing head and neck free flap reconstruction were included. Meta-analysis demonstrated no statistically significant difference in the incidence of flap failure (2.9% vs. 3.6%; OR: 0.68, 95% CI: 0.23-1.99; p= 0.48) or incidence of flap complications (16.8% vs. 18.6%; OR: 0.92, 95% CI: 0.60-1.42; p= 0.71).

CONCLUSIONS: Based on the current evidence, intraoperative use of vasopressors has no impact on the incidence of flap failure or flap complications.

7:35 AM - 7:40 AM

Optimizing Outcome of Pharyngoesophageal Reconstruction and Neck Resurfacing: 10-Year Experience Following 294 Cases MD Anderson Cancer Center, Rochester, MN, USA Basel Sharaf, MD, DDS1; Amy Xue, MD2; Mario Solari3; Jun Liu4; Matthew Hanasono5; Peirong Yu6; Jesse Selber, MD, MPH5; 1Mayo Clinic, 2Baylor College of Medicice, 3University of Pittsburgh, 4MD Anderson Cancer Center, University of Texas, 5MD Anderson Cancer Center, 6The University of Texas Introduction: Pharyngoesophageal reconstruction is challenging and demands unique reconstructive considerations. The frequent need for neck skin resurfacing in salvage pharyngoesophageal reconstruction increases the technical complexity. This large series examines factors impacting the need for neck skin resurfacing and evaluates the impact of reconstructive modalities on outcomes.

Methods: A review identified 294 patients who underwent pharyngoesophageal reconstruction from 2002 to 2012. Patients were divided into 2 groups based on neck skin resurfacing requirements. Patients undergoing neck resurfacing were further subdivided into reconstructive technique, including a second skin paddle or muscle component from the same free flap pedicle, a pectoralis major flap, or a second free flap. All groups were compared by co-morbidities, complications, and functional outcomes.

Results: Of 294 cases, 179 (60.9%) required neck skin resurfacing and 115 (38.1%) were closed primarily. In the resurfaced group, there were 90 (50.3%) circumferential defects and 89 (49.7%) partial defects. In the primary closure group, there were 48 (41.7%) circumferential defects and 67 (58.3%) partial defects. In the resurfaced group, 110 (61.4%) were reconstructed with a second skin paddle from the same free flap pedicle, 21 (11.7%) were reconstructed with a muscle component from the same pedicle, and 25 (13.9%) received a pectoralis major flap. There were 5 external paddle flap losses in the resurfaced group (2.8%) that required additional vascularized tissue coverage. There were no internal flap losses. In the neck closure group, there were 3 flap losses (2.6%). Overall complications were similar among groups. In spite of increased complexity and co-morbidity, the resurfaced group had a lower pharyngocutaneous fistula (PCF) rate (4.5%) compared to the primary closure group (11.3%, P=0.026). Prior neck surgery and radiation therapy were strong predictors of neck skin resurfacing (P<0.001). Tracheoesophageal speech quality and post-operative diet were similar.

Conclusions: Pharyngoesophageal reconstruction is often accompanied by neck resurfacing needs. Prior ablative surgery and radiation treatment are strong predictors for neck resurfacing. Multiple techniques are available and an algorithmic approach is useful. Providing additional vascularized tissue over the neo-conduit is predictive of lower PCF rates. Careful surgical planning and refinements in flap design can improve patient outcomes.

7:40 AM - 7:45 AM Calf Perforator Flaps - an Ideal Solution for Oral Cavity Reconstruction Guy's & St. Thomas' Hospitals Trust, London, United Kingdom William A. Townley; Natalie Pease; Andrew Davies; Fungayi Chinaka; Luke Cascarini; Guy's &

St. Thomas' NHS Hospitals Trust Calf Perforator Flaps - an Ideal Solution for Oral Cavity Reconstruction

Introduction

Calf perforator flaps are emerging as a popular choice in complex head and neck reconstruction due to their favourable donor site morbidity. There are multiple source vessels in this region. The medial sural artery (MSAP) flap is the most common. We set out to demonstrate that calf perforator flaps have ideal properties for oral cavity reconstruction.

Method

There were three parts to the study. Firstly, a cadaveric dissection study was performed to demonstrate perforator reliability and document variations in source vessels. 20 medial calves from 10 cadavers were dissected (4 males 6 females). The position of the perforators was recorded in relation to fixed landmarks (vertical – intermuscular septum, horizontal – most proximal point on the fibula).

In addition, the quality of donor calf envelope was assessed by comparing relative tissue thickness of the calf with the anterolateral thigh in 50 lower limb CT scans.

Finally, a prospective clinical study was performed to demonstrate clinical effectiveness. Demographic, oncologic and surgical data were collected on sequential calf perforator flaps used in oral cavity reconstruction (floor of mouth, tongue, buccal mucosa) over an18-month period.

Results

In the cadaveric study, musculocuntaneous perforators form the MSA vessels were found in 19 of the 20 cases (mean 2.3/ limb). Perforator location ranged from 20 to 170mm inferior to the fibular head and 0 to 45mm medial to the septum. Septocutaneous perforators from the sural artery system were present in 10% limbs (2 of 20).

The mean thickness of the calf tissue envelope was found to be 49.5% of the thickness of the anterolateral thigh donor site (p<0.05) on review of 50 lower limb CTs (20 male, 30 female).

A total of 14 free calf perforator flaps (12 MSAP, 1 Sural Artery Perforator, 1 Soleal Artery Perforator) were performed over the study period (7 floor of mouth, 6 tongue, 1 buccal mucosa). Mean flap size was 8 x 5cm. Thirteen of 14 donor sites were closed directly. All flaps survived. Early complications included one return to theatre (haematoma) and three wound infections (1 donor, 2 neck) that settled spontaneously with intravenous antibiotics.

Conclusion

Our study suggests that the vascular anatomy of the medial calf is predictable, yielding perforator flaps that are ideal quality and effective in oral cavity reconstruction.

7:45 AM - 7:48 AM Discussion

7:48 AM - 7:53 AM Complex Orofacial Reconstruction with the Intrinsic Chimeric Flap University of Chicago, Chicago, IL, USA Amanda K. Silva; Andres A. Maldonado; Laura S. Humphries; Lawrence J. Gottlieb; University of Chicago Medical Center

Background: Ablation of locally-advanced head and neck cancer often results in composite orofacial defects that require various tissue types inset in different orientations. An additional common issue is limited recipient vessel options due to wide resection and previous radiation. These complex defects lend well to intrinsic chimeric flap reconstruction, which provides a variety of tissue components each on their own vascular leash converging to a single pedicle. The different tissue components of the chimera can be independently oriented and inset based on the reconstructive needs and typically require only one set of recipient vessels for microvascular anastomoses.

Methods: A retrospective chart review was performed for all patients who underwent orofacial reconstruction with an intrinsic chimeric free flap from 2002 to 2015. Information on patient, defect, and flap characteristics, as well as complications and outcomes were analyzed to create an algorithm for flap selection.

Results: Seventy-five patients underwent orofacial intrinsic chimeric free flap reconstruction. Sixty (80%) patients previously underwent radiation therapy and 48% received double radiation. Twenty-four (32%) patients required anastomoses to recipient vessels other than the branches of the carotid. There were 2 (3%) total and 3 (4%) partial flap losses. Average follow-up time was 26 months. Patient, defect, and flap characteristics are summarized in Tables 1 and 2. Results were organized based on defect characteristics to create an algorithm to guide flap selection. The complete algorithm is presented in Figure 1. In general, flap selection for defects with no bony component was determined by total external skin and internal lining surface area. Large defects were reconstructed with a lateral femoral circumflex (LFC) flap (Figure 2) and small defects with a radial forearm flap. Defects with a bony component were further categorized based on the need for external skin, internal lining, dead space fill, and the size of the bony defect. Defects with a large skin, lining, or dead space component were reconstructed with a LFC flap. Moderate-sized skin or lining defects with larger bony defects were reconstructed with peroneal flaps (Figure 3). Small-sized skin or lining defects with smaller bony defects were reconstructed with thoracodorsal (Figure 4) or radial forearm flaps.

Conclusions: Intrinsic chimeric flaps are a versatile and elegant option for reconstruction of complex orofacial defects. The algorithm provided will help guide flap selection in these challenging reconstructions.

7:53 AM - 7:58 AM Double-paddle Peroneal Chimeric Flap For Lower Lip Reconstruction Kaohsiung Veterans General Hospital, Kaohsiung, Taiwan Ying-Sheng Lin, MD, MPH; Kuo-Chung Yang, MD; Kaohsiung Veterans General Hospital

Introduction

Lip is one of the most obvious and functional portions in the face. Today's reconstructive surgeons offer an array of options in an attempt to restore the function and appearance of lip defect after trauma affliction or tumor ablation. In this article, we would like to present our experience of using free double-paddle peroneal chimeric flap for lower lip reconstruction.

Patients and Methods

From 1996 to 2014, there are 9 patients having lower lip defects after tumor ablation. Double-paddle peroneal chimeric flap was used to reconstruct the lower lip defect. A part of inner skin paddle was everted to act as the newly reconstructed lower lip. The suture line with outer skin paddle was the new vermillion border (Figure 1). The skin paddle at the superior location of lateral lower leg was used as the inner paddle because of its more thickness, leading to more fullness of the new lip. Medical records were reviewed retrospectively to retrieve information of interest, such as perioperative complications, long-term functional and cosmetic results.

Results

Nine peroneal chimeric flaps were used to reconstruct the lower lip defects of nine patients. Four of them are fasciocutaneous flaps, and the other five are osteocutaneous flaps. The average size of flaps were (74.2+-32)x(61.2+-24) cm. In terms of perioperative complications, neither flap loss nor wound infection was noted. As for long-term functional results, all patients were able to close their mouths tightly without drooling (Figure 2). The cosmetic results were acceptable for all patients.

Conclusion

For lower lip reconstruction, free double-paddle peroneal chimeric flap transfer could not only provide satisfactory oral competency but also acceptable cosmetic result. Therefore, it should be considered one of the options for lower lip reconstruction.

Figure 1.

A: Outer skin paddle B: Inner skin paddle. A portion of inner skin paddle is everted to be the new lower lip. The junction of A and B is the new vermillion border.

Figure 2. (Left) Preoperative (Middle) Immediate postoperative (Right) One-month follow-up

7:58 AM - 8:03 AM Total robotically performed osteotomies for free fibula flap mandible reconstruction The Ohio State University, Columbus, OH, USA Albert H. Chao, MD1; Katie Weimer2; Joerg Raczkowsky, PhD3; Yaokun Zhang, DiplInform3; Mirko Kunze, MSc3; Jesse C. Selber, MD, MPH4; Matthew M. Hanasono, MD4; Roman J. Skoracki, MD1; 1The Ohio State University, 2Medical Modeling Inc, 3Karlsruhe Institute of Technology, 4MD Anderson Cancer Center Background: While technological advances have aided the planning and execution of free fibula flap mandible reconstruction, these have not addressed the human operator component, which can contribute to clinically significant inaccuracy during osteotomies. One potential solution is to eliminate human error through the use of medical robotics. The purpose of this study was to

determine the feasibility of total robotically performed osteotomies, performed without surgeon intervention, for fibula free flap mandible reconstruction, and to measure the accuracy of this approach.

Methods: The KUKA Lightweight Robot (KUKA, Augsburgs, Germany) was utilized, which combines technology used in high precision industries (e.g., automobile manufacturing), with the size and sensitivity required for surgical applications. A standardized anterior mandibular defect was designed on a CAD platform, as well as a corresponding free fibula flap reconstruction of this defect requiring 6 osteotomies. A methodology was developed to translate this virtual surgical planning (VSP) data to the robot. The robot then independently performed osteotomies on fibulas generated through stereolithographic printing with the aid of stereotactic navigation. Using high-resolution computed tomography, the osteotomized segments were compared to the virtually planned segments in order to measure linear and angular variation.

Results: The osteotomy planes created through VSP were translated to the robot using common Euclidean geometry and specification with non-collinear points in a coordinate system. A specialized table mount was created to emulate the positioning and degrees of freedom required in actual surgery. Registration of stereotactic navigation was achieved using fiducial markers on the robot, mount, and fibulas. A total of 18 total robotically performed osteotomies were performed on 3 fibula flaps. The average linear variation was 1.3±0.4 mm. The average angular variation was 4.2±1.7 degrees.

Conclusions: This study demonstrates the feasibility of executing total robotically performed osteotomies for free fibula flap mandible reconstruction. The linear and angular variation associated with total robotically performed osteotomies is low, and the deviation in error less than that associated with the use of VSP and surgical cutting guides with human operators.

8:03 AM - 8:06 AM Discussion

8:06 AM - 8:11 AM Cost-Effectiveness of Microsurgical Reconstruction for Head and Neck Defects after Oncologic Resection University of Pennsylvania Health Systems, Philadelphia, PA, USA Lin Lin Gao1; Marten Basta, BA2; Suhail Kanchwala3; Joseph Serletti3; David Low, MD3; Liza Wu, MD3; 1Hospital of the University of Pennyslvania, 2Perelman School of Medicine at the University of Pennyslvania, 3University of Pennsylvania Background:

Microvascular free tissue transfer has become main technique for head and neck reconstruction. We assessed the cost-effectiveness of free flap reconstruction for head and neck defects after oncologic resection for squamous cell cancer (SCC).

Methods:

We developed a Markov model of the cost, quality of life, survival and incremental cost-effectiveness of reconstruction with free tissue transfer compared to locoregional flaps. Health state probabilities and quality of life scores were determined from literature. Costs were determined from institutional experience. Outcomes included quality-adjusted life years, costs and incremental cost-effectiveness ratio.

Results:

Free flap reconstruction was more costly than pedicled flap but associated with greater quality of life with no survival benefit. A value less than $50,000 per QALY was defined as cost-effective. The incremental cost-effectiveness for head and neck free flap reconstruction was below the threshold and therefore free flap reconstruction is cost effective. Reconstruction was more cost effective for patients with lower stage cancers: $4643 per QALY for stage I SCC, $8226 for Stage II, $17,269 for stage III and $23,324 for Stage IV. Univariate sensitivity analysis showed the cost-effectiveness would remain below $50,000 for all stages of SCC for all variable except for QALY after locoregional reconstruction without complications.

Conclusion:

Microsurgical head and neck reconstruction is cost effective compared to locoregional flaps, even more so in patients with early stage cancer. This finding supports the current practice of free flap head and neck reconstruction. Screening and early detection are important to optimize costs.

8:16 AM - 8:21 AM Endoscopic-Assisted Radial Forearm Free Flap Harvest: Long-Term Results and Complications Albany Medical Center, Albany, NY, USA Emily Van Kouwenberg, BS, MD; Ash Patel, MBChB; Richard L. Agag, MD; Albany Medical Center Background

Endoscopic-assisted radial forearm free flap (ERFFF) harvest was previously described by the authors as a novel technique aimed to decrease donor site morbidity. Based on initial findings, ERFFF was found to be a safe and effective technique, with decreased donor site morbidity, a short learning curve, and similar operative time. With a greater sample size and duration of follow up, ERFFF patients were reexamined to assess long-term results and complications.

Methods

A retrospective case series was conducted following patients undergoing ERFFF for intraoral reconstruction by a single surgeon at Albany Medical Center between 11/2013 and

06/2015. Data recorded includes patient characteristics, duration of harvest, methods of donor site reconstruction, and postoperative complications.

Similar technique was employed in all patients, beginning with design of the skin paddle followed by proximal incision and isolation of the radial artery and venae comitantes. Through the incision the vessels were visualized endoscopically. If deemed necessary, the flap design was altered to incorporate the cephalic vein. Endoscopic dissection of the vascular pedicle was performed using a 5mm angled endoscope and the “VascuClear” ®Endoscopic Vessel Harvesting kit (SORIN GROUP Inc.). Vessel branches were ligated with Harmonic ACE 5mm Diameter Shears (Ethicon Inc.). The extremity was then exsanguinated and the fasciocutaneous flap raised in a conventional manner. Upon completion of flap elevation, the proximal pedicle was ligated endoscopically.

Results

Fourteen ERFFF harvests were performed to reconstruct patients 22-79 years old (mean 56) after intraoral ablations, including floor of mouth (50 %), pharyngolaryngeal (14.3%), tongue (14.3%), palatal (7.1%), nasomaxillary (7.1%), and retromolar trigone (7.1%), treating 13 squamous cell carcinomas (92.9%) and 1 pleomorphic adenocarcinoma (7.1%). Four (28.6%) patients were female and 10 (71.4%) male. Documented endoscopic pedicle harvests ranged 21-25 minutes (mean 22.67), with subsequent flap harvest ranging 50-102 minutes (mean 74.5). All flaps were single paddle, with skin paddles ranging 20-84cm² (mean 38.7cm²). Donor sites were reconstructed with primary closure (7.1%), STSG (35.7%), or Integra™ meshed bilayer wound matrix and subsequent STSG (57.1%). Four patients experienced complications, including sialocele, hematoma, venous occlusion requiring revision, and graft loss in a noncompliant patient. No long-term complications occurred. Follow-up ranged 0-372 days (mean 120.9). Five patients (35.7%) have had local recurrences since reconstruction.

Conclusion

ERFFF is a safe and effective technique to obviate the need for lengthy incisions, with decreased donor site morbidity and no observed long-term complications. Given high recurrence rates among patients, it’s crucial to minimize reconstructive morbidity.

8:21 AM - 8:24 AM Discussion

8:24 AM - 8:29 AM Free Fibula Flap Versus Rotational Trunk Flap for Head and Neck Cancer: Comparison of 30-day Surgical Outcomes Duke University, Durham, NC, USA Jared Blau, BS, MEd; Yash J. Avashia, MD; Ronnie Shammas, BS; Kelli Aibel; Scott T. Hollenbeck, MD; Duke University Medical Center

Introduction

In patients with locally advanced head and neck cancer involving the mandible, the use of a free fibula flap is the mainstay of reconstructive therapy after oncologic resection. However, in some patients, a trunk muscle, such as the pectoralis, may be rotated on a vascular pedicle to provide coverage when a microsurgical procedure is not indicated. The purpose of this study is to compare surgical outcomes of these two procedures.

Methods

Patients less than 80 years of age with head and neck cancer were identified by ICD-9 codes in the NSQIP database from 2005 through 2013. Free fibula patients were first compared with those who have only received a muscle, myocutaneous, or fasciocutaneous flap coverage from the trunk, as identified by Current Procedural Terminology codes. Both groups were propensity matched using a 1:1 nearest-neighbor algorithm matching for age, sex, race, BMI category, ASA class, and disseminated cancer status. Preoperative characteristics and postoperative outcomes were compared, using a t-test for continuous variables and Pearson chi-square for nominal variables.

Results

A total of 137 patients were identified with significant differences in sex, BMI category, ASA class, hematocrit, and disseminated cancer status. Free fibula cases required more operative time (634.61 min vs 442.91 min, p<.01). However, the overall incidence of surgical complications, defined by a surgical site infection, wound dehiscence, or flap failure, was similar between the two groups (32.65% vs 27.27%, p=0.51). [Tables 1 and 2] Propensity matching identified 49 patients receiving a free fibular flap and 49 receiving pedicled flap coverage from the trunk. After matching, there were no significant differences in patient demographic or health characteristics between the two groups. Free fibula flaps still required more operative time, though the rate of surgical complications was still similar between the two groups (32.65% vs 28.57%, p=0.66). [Figure 1] There were no significant differences between the rate of additional surgical and medical complications seen over the 30-day period. [Table 3]

Conclusions

For locally advanced head and neck cancers, the patient characteristics of those undergoing free fibula flaps versus pedicled flaps from the trunk are different. After controlling for these factors, free fibula flaps were associated with longer operative times but 30-day morbidity was the same. These results imply the relative safety of free fibula flaps and the versatility of pedicled flaps from the trunk for head and neck reconstruction.

8:29 AM - 8:34 AM Computer-Aided Surgical Simulation in Head and Neck Reconstruction: A Cost Comparison among Traditional, In-House, and Commercial Options Inova Fairfax Hospital, Falls Church, VA, USA Sean S. Li, BA1; Alexander J. Kaminsky, MD, MPH2; Libby R. Copeland-Halperin, MD2; Fahad K. Lodhi, BS1; Jihui Li, PhD2; Reza Miraliakbari, MD3; 1Virginia Commonwealth University, 2Inova Fairfax Hospital, 3Private Practice

Background

Computer-aided surgical simulation (CASS) has redefined surgery, improving precision and reducing the reliance on intraoperative trial-and-error manipulations. CASS is provided by third-party services; however hospitals may develop in-house programs that may be more cost-effective. This study provides the first cost analysis and comparison among traditional (no CASS), commercial, and in-house CASS for head and neck reconstruction.

Methods

The costs of three-dimensional (3D) pre-operative planning for mandibular and maxillary reconstructions were obtained from an in-house CASS program, as well as a commercial provider. Start-up costs were also obtained for our in-house CASS program. The personnel costs are based on ten cases per year for the early development of the program and reflect the learning curve (over-estimation).

Operative times are expressed as averages of the most common types of reconstructions performed by the senior author, with an effort to consider all levels of complexity (i.e. mandibular body reconstruction vs. near total mandibular or maxillary reconstruction).

A cost comparison was then performed among these modalities and extrapolated in-house CASS costs were derived. The calculations are based on CASS use in ten clinical cases annually.

Results

Average sentinel operating room utilization time was estimated at ten hours, with an average of two hours of time saved with the use of CASS. The hourly cost for the use of the operating room at our hospital (including anesthesia and other ancillary costs) is estimated at $4614/hour.

Conclusion

CASS minimizes the trial-and-error manipulations at the initial surgery, resulting in significant time savings in the operating room. In comparing the in-house option to the traditional one, the program proves financially viable if performing at least ten cases per year (breaks even at $40,000). In comparing the in-house option to the commercial option, assuming ten clinical cases per year, there is an added cost to the hospital of $12,000. In light of comparable operative time between the in-house and commercial options, the in-house option will in time prove financially beneficial as the number of clinical cases move beyond 15 cases per year and the learning curve is mastered by all involved. In addition, there is the vast benefit of having the immediate availability of an institution's own unit, and not being limited by the commercial option's time restrictions and turnaround times. Our data demonstrate that hospitals performing greater than 10 cases of 3D head and neck reconstructions per year should consider developing an in-house CASS program.

8:34 AM - 8:39 AM Quality of Life on Head and Neck Reconstruction: Surgical Perspective Memorial Sloan Kettering Cancer Institute, New York, NY, USA Claudia R. Albornoz; Wess Cohen; Shantanu Razdan; Elizabeth Encarnacion; Meghan Lee, BS; Mike Cavalli; Peter Cordeiro; Snehal Patel; Andrea Pusic; Evan Matros, MD; Memorial Sloan Kettering Cancer Center

Introduction

Despite advances in the treatment of head and neck(H&N) malignancies, the rate of disease-specific morbidity and mortality remains high. Improved understanding and management of the health-related quality of life (HR-QOL) issues perhaps represents the greatest unmet need for these patients. Previous studies have evaluated HR-QOL combining a variety of defects, therefore limiting the validity of the results. No study has classified surgical resections into common, discrete anatomical locations while investigating associated patient reported outcomes

(PROs) prospectively. Such information can be used for preoperative counseling and patient selection.

Methods

This was a prospective study of HR-QOL in H&N cancer patients undergoing surgical resection with flap reconstruction for stage II or III H&N malignancies at Memorial Sloan Kettering Cancer Center from 2008 to 2013. Patients completed the European Organization for Research and Treatment of Cancer Core Quality-of-Life Questionnaire 30 (EORTC 30) and EORTC QLQ H&N Cancer Module 35 (EORTC 35) preoperatively, and at 3, 6, and 12 months postoperatively. Scores were compared with t test and linear regression.

Results

Seventy-five patients were included. Mandibulectomy was the most common defect(31%) followed by partial glossectomy(25%), oral lining(15%), maxillectomy(12%), total glossectomy(9%), and laryngectomy(8%). The radial forearm free flap was used most frequently for reconstruction(40%), followed by the rectus abdominis(25%), anterolateral thigh(11%) and fibula(9%) flaps.

At one year postoperatively maxillectomy, partial glossectomy, and oral lining defects had worse function and more symptoms in 4, 7, and 10 of 33 items, respectively. In contrast, mandibulectomy, laryngectomy, and total glossectomy had worse function and more symptoms in 28, 26, and 17 of 33 items, respectively. for Partial glossectomy and oral lining defects were associated with higher levels of HR-QOL at 6 and 12 months compared to laryngectomy defects (p<.05)(Figure 1).

Physical, role, and social functioning scores at 3 months were significantly lower than preoperative values (p<.05). At 12 months postoperatively, none of the function or global QOL scores differed significantly from preoperative levels.

Conclusion

HR-QOL after H&N cancer resection is associated with the anatomical location of the surgical defect. Surgery negatively impacts HR-QOL in the immediate post-operative period with a return to baseline function at 12 months. Preoperative teaching and should be targeted for each ablative procedure with postoperative expectations adjusted appropriately.

Figure 1. Evolution of EORTC global QOL by defect type

8:39 AM - 8:42 AM Discussion

8:42 AM - 8:47 AM The Posteromedial Thigh Flap for Head and Neck Reconstruction Yen-Chou Chen, Kaohsiung City, Taiwan Yen-Chou Chen, MD1; Mario Scaglioni2; yur-Ren Kuo1; Johnson Chia-Shen Yang1; 1Kaohsiung Chang Gung Memorial Hospital and Chang Gung University College of Me, 2University Hospital Zurich, Zurich, Switzerland

Background: The posterior medial thigh is a neglected donor site for head and neck reconstruction. The profunda femoris artery, it nourishes the adductor magnus muscle and overlying skin, and supplies a number of septocutaneous and musculocutaneous perforators that can potentially be used as pedicles for fasciocutaneous flaps. The authors present the "posteromedial thigh" (PMT) flap as an alternative source, and the perforator patterns and vascular anatomy of this flap were further investigated.

Patients and Methods: From March 2014 to June 2015, 60 patients underwent head and neck reconstructions with 60 PMT flaps. The defect locations were buccal mucosa(26), tongue(13), lower gum(4), palate(4), hypopharynx(3), oropharynx(2), trigone(2), upper gum(2), neck(1), parotid(1), lip(1) and orbit(1). The skin paddle can be designed as transverse or vertical pattern with the potential of flap split to fit the requirement of reconstruction. The adductor magnus muscle and the gracilis muscle can be incorporated for dead space obliteration or soft tissue augmentation. A new double flaps combination consist of the PMT flap and the fibula flap is

available for the composite oromandibular defects reconstruction. The numbers, locations, and types of perforators were measured. The results and complications after reconstruction were evaluated.

Results: Most perforators were located 8-10 cm away from the pubic crease on the reference line between the perineum and the insertion of the semitendinosus muscle. The average perforators number was 2.0 (range: 1 to 4), and the average pedicles length was 10.2 cm (range: 7 to 14 cm). The pedicles of the PMT flap originated from originated from the profunda femoris artery(91.7%, 55/60) or the medial circumflex femoral artery(8.3%, 5/60). The flap survival rate was 98.3%; one flap suffered from pedicle thrombosis and was not salvaged successfully. The donor sites were all closed primarily with minimal morbidities.

Conclusion: The location of the perforators of the PMT flap is consistent, and the pedicle length is sufficient to reach the neck region. Different reconstruction demands can be accomplished by incorporating various soft tissue components. The donor site scar is well concealed in the posterior medial thigh region with minimal morbidity. The above advantages make the PMT flap an excellent option for head and neck reconstruction.

Figure 1. The posteromedial thigh flap design

Figure 2. Different flap modification

Single flap(above, left), Split flap(above, right), Skin flap with adductor magnus muscle(below, left), Skin flap with gracilis muscle(below, right).

Figure 3. For hypopharynx with neck defect reconstruction.

8:47 AM - 8:52 AM The Deep Temporal Nerve: An Anatomical Feasibility Study with Possible Implications for Dynamic Brow Reanimation UT Southwestern Medical Center, Dallas, TX, USA Phillip Dauwe, MD1; Austin Hembd, MS1; Erika De La Concha-Blankenagel, MD2; Salim Saba, MD3; Charles White, MD1; Alexander Cardenas-Mejia, MD2; Shai Rozen, MD1; 1UT

Southwestern Medical Center, 2Hospital General Dr. Manuel Gea Gonzalez, 3American University of Beirut Introduction: Dense peripheral facial paralysis has a profound impact on the brow, and currently static procedures are the mainstay to restore its position. The deep temporal branches of the trigeminal nerve, given their proximity to the brow and frontal branch of the facial nerve, may serve as possible donor nerves for both potential innervation of a free muscle transfer in cases of prolonged facial palsy or nerve transfers in cases of acute or sub-acute facial palsy. As such, we present the detailed surgical anatomy of the deep temporal nerve assessing feasibility for both functional muscle transfer and nerve transfer including a proposed surgical technique.

Methods: Thirty cadaver hemi-faces were dissected to establish deep temporal nerve location, anatomical variability, and design of a surgical approach. The proposed surgical approach was performed on two hemi-faces.

Results: Two (53%) or three (47%) branch divisions of the deep temporal nerve were noted with the most consistent division being the posterior division in case of two branches or first posterior division in cases of three branches (30/30 specimens). This division was consistently found approximately 4.1 (3.7-4.5) cm anterior to the tragus at the level of the zygomatic arch, coursing intramuscularly and parallel to the fibers of the temporalis muscle. At 2cm cranial to the arch, the nerve measures 4.7 (4.2-5.2) cm from the tragus; at 4cm cranial to the arch, the nerve measures 5.5 (5.3-5.7) cm from the tragus; at 6cm cranial to the arch, the nerve measures 6.8 (6.5-7.1) cm from the tragus. Effectively, for each 1cm cranial to the arch, the nerve courses approximately 1mm posteriorly.

Conclusion: Given its location, anatomic consistency, and expendability, the posterior or first posterior division of the deep temporal nerve is a viable donor nerve for dynamic brow reanimation with either functional muscle transfer or nerve transfer depending on the length of facial palsy.

8:52 AM - 8:57 AM Predictive and Protective Factors for Reconstructive Plate Failure in Free Fibula Flaps for Mandibular Reconstruction University of Texas MD Anderson Cancer Center, Houston, TX, USA Alice S. Yao, MD; Jun Liu; Patrick B. Garvey, MD; The University of Texas MD Anderson Cancer Center

Background:

The free fibula osteocutaneous flap is the favored option for oncologic mandible reconstruction. Rigid fixation to the native mandible is most commonly achieved using a reconstruction plate/screw system. Reconstruction plate fracture following mandible reconstruction with a vascularized fibula flap bone is a rare, poorly understood, and devastating complication that typically necessitates major operative revision. We hypothesized that certain factors such as

radiation, smoking, and precision of the surgical osteotomies with computer-assisted virtual planning predispose to or protect against reconstruction plate fracture.

Methods:

We retrospectively reviewed all consecutive patients who received free fibula flaps for oncologic mandible reconstruction at a single institution over 10 years (2005-2014). The primary outcome measure was the development of a postoperative reconstruction plate fracture event, which was defined as plate fracture combined with a fibula non-union. Secondary outcomes included recipient and donor site complications and major complications requiring operative management within 30 days. We performed univariate and multivariate logistic regression analysis to determine the relationship between specific patient and surgical characteristics and the development of a plate fracture or other complication.

Results:

A total of 505 consecutive free fibula flaps were performed for mandibular reconstruction with an average follow-up of 27 months. We identified 17 (3.4%) plate fractures, 208 (41%) recipient site complications, and 67 (13%) plate events including fractures and exposures. The flap failure rate was 1.4%. Major predictive factors for a plate fracture included post-operative radiation (OR 5.26, p=0.015) and two or more osteotomies (OR 3.74, p=0.034). The single predictive factor for complications that were serious enough to require operative management within 30 days was active smoking (OR 1.68, p=0.041). Virtual planning with computer-assisted design appeared to protect against recipient site complications (OR 0.46, p=0.025). Neither virtual planning nor surgeon experience was significantly protective against plate fractures or total plate events.

Conclusion:

Post-operative radiation and multiple osteotomies were independent predictors for plate fractures after oncologic mandible reconstruction with free fibula flaps. Neither surgeon experience nor virtual planning was protective against plate fractures, although virtual planning (but not surgeon experience) was an independent protective factor against recipient site complications. We believe this information may better guide surgeons’ decision-making when planning and performing free fibula flaps for oncologic mandible reconstruction.

8:57 AM - 9:00 AM Discussion

9:00 AM - 9:05 AM Approaches and Outcomes of Facial Reanimation in Patients over 60 years - A Multicenter Study UT Southwestern Medical Center, Dallas, TX, USA Bridget Harrison, MD1; Khalil Chamseddin, MS1; Erika De La Concha-Blankenagel, MD2; Daniel Labbé, MD3; Alexander Cardenas-Mejia, MD2; Shai Rozen, MD1; 1UT Southwestern

Medical Center, 2Hospital General Dr. Manuel Gea Gonzalez, 3University of West Paris

Background: Dynamic reanimation of the paralyzed face is the optimal goal. While the literature is abundant in techniques, it is sparse in regard to reanimation of older patients. Between 1968 to 2013 only 57 papers mention merely 45 patients over 60, severely lacking detailed outcomes. Due to age alone, this population is mostly neglected and delegated to static procedures aiming to improve symmetry in repose but not reanimation. With this in mind, we present the largest series to date from three centers detailing 26 patients over the age of 60, who had undergone facial reanimation between 1997- 2014 using three techniques – Lengthening Temporalis Myoplasty (LTM), Free Functional Muscle Transfer (FFMT), and Nerve Transfers (NT).

Methods: 26 patients over age 60 at time of surgery were reviewed for demographics including age, gender, pathology, type of reconstruction, time from facial palsy to surgery, time from surgery to motion, complications, additional procedures, and excursion measured with MEII – FACE gram system.

Results: Seven males and nineteen females with an average age of 65 years at the time of surgery (Range 60-86) were included. Twelve were reconstructed with LTM, ten with FFMT, and four with NT. The average pre op HB score was 5.6 in the LTM, 4.9 in the FMT , and 5.3 in the NT group. The mean time interval between palsy to reanimation was 181 months in the LTM, 62 months in the FFMT, and 7.5 months in the NT group. The average time from surgery until observed motion was 10.6 days in the LTM, 4.4 months in the FMT, and 6 months in the NT group. Average follow up was 48 months in the LTM, 34 months in the FFMT, and 20 months in the NT group. From the FFMT group, one patient died from an aggressive disease before motion was observed, while no motion was observed in another. The ability to smile without clenching the teeth was 100% (12/12) in the LTM group, 75% (6/8) in the FMT group, and 25% (1/4) in the NT group. An average 3.5 ancillary procedures were performed on each patient beyond dynamic reanimation. Using the FACE system, the average excursion was 3.1 millimeter in the LTM, 6.2 mm in FFMT, and 6.7 mm in the NT group.

Conclusion: Successful dynamic reanimation can be performed in properly selected patients over 60 years old. Age alone should not eliminate surgical options for dynamic reanimation.

9:05 AM - 9:10 AM Preemptive Mandibulectomy and Free Fibula Flap Reconstruction for Osteoradionecrosis is Not Superior to Delayed Reconstruction Following Orocutaneous Fistula or Pathologic Fracture The University of Texas MD Anderson Cancer Center, Houston, TX, USA Alice S. Yao, MD; Steven B. Chinn; Patrick Garvey, MD; The University of Texas MD Anderson Cancer Center

Background:

Radiation therapy is a core element of the treatment of most head and neck cancers. Osteoradionecrosis (ORN) of the mandible has become an increasingly prevalent consequence of the expanding indications and more complex protocols of radiotherapy. Severe ORN can progress to complications such as orocutaneous fistulae and/or pathologic fracture of the mandible. It is unclear whether the appropriate management of severe ORN is conservative treatment or preemptive mandibulectomy and reconstruction with a vascularized bone flap before the development of complications. We hypothesized that preemptive mandibulectomy and reconstruction with vascularized fibula flaps would lead to improved outcomes for patients with ORN of the mandible in comparison to mandibulectomy and reconstruction for only those presenting with advanced symptoms.

Methods:

We retrospectively reviewed consecutive free fibula flaps for mandibular reconstruction by multiple surgeons at a single institution over 10 years (2005-2014). Outcomes for the ORN reconstructions (experimental group) were compared to outcomes for oncologic mandibulectomy (control group). Subset analysis compared outcomes between two groups of reconstructions for ORN based on the timing of surgery. We defined the early group as patients managed with resection and reconstruction prior to the development of fistula or fracture and the late group as those treated after the development of the above symptoms. The primary outcome measures included overall and specific postoperative complications, including recipient site, donor site, and major complications requiring operative management within 30 days.

Results:

505 consecutive free fibula flaps were performed for mandibular reconstruction with an average follow-up of 27 months, of which 89 (17%) were for ORN. There was a trend toward increased recipient site complications in ORN patients (OR 1.42, p=0.154) as well as major operative returns within 30 days (OR 1.54, p=0.134) that did not reach statistical significance. Subset analysis of the ORN reconstructions alone included 42 patients in the early management group and 47 patients in the late group. Overall and specific complications were also statistically equivalent between the early and late ORN reconstruction groups.

Conclusion:

Outcomes for mandibular reconstruction with free fibula flaps were similar for the ORN and oncologic mandibulectomy groups. Contrary to our hypothesis, we did not see a difference in outcomes between the early versus late treatment groups. We believe this data supports both the strategy of non-operative, expectant management of early mandible ORN as well as mandibulectomy with vascularized free fibula flap reconstruction for symptomatic mandible ORN presenting with orocutaneous fistula or pathologic fracture.

9:10 AM - 9:12 AM Discussion