- Email: [email protected]
Disease relapse after segmental resection and free flap reconstruction for mandibular osteoradionecrosis
Otolaryngology–Head and Neck Surgery (2010) 142, 586-591
ORIGINAL RESEARCH–FACIAL PLASTIC AND RECONSTRUCTIVE SURGERY
Disease relapse after segmental resection and free flap reconstruction for mandibular osteoradionecrosis Jeffrey D. Suh, MD, Keith E. Blackwell, MD, Joel A. Sercarz, MD, Marc Cohen, MD, Jerome H. Liu, MD, MSHS, Christopher G. Tang, MD, Elliot Abemayor, MD, PhD, and Vishad Nabili, MD, Los Angeles, CA No sponsorships or competing interests have been disclosed for this article. ABSTRACT OBJECTIVE: The objective of this study was to assess the outcomes, complications, and incidence of disease recurrence of mandibular osteoradionecrosis (ORN) after resection and microvascular free flap reconstruction. STUDY DESIGN: Case series with chart review. SETTING: Academic medical center. SUBJECTS AND METHODS: Retrospective patient data review of 40 patients with mandibular ORN who were treated by segmental mandibulectomy and microvascular reconstruction between 1995 and 2009. All patients received radiation therapy for previous head and neck cancer, and 12 of 40 patients received concurrent chemotherapy. All patients failed to respond to conservative management. There were 26 males and 14 females, with a median age of 62 years. Median follow-up was 17.4 months. RESULTS: There were no free flap failures. The incidence of wound-related complications was 55 percent. Median time to complication was 10.6 months. Ten (25%) patients developed symptoms of residual or recurrent ORN, with 70 percent of the recurrences arising in unresected condyles that were adjacent to the segmental mandibulectomy. Statistical analysis revealed that current smokers were at reduced risk to develop residual or recurrent ORN. CONCLUSION: This present study confirms that microvascular free flaps are reliable for treatment of advanced mandibular ORN. Nevertheless, there remains a 55 percent incidence of woundhealing complications. The lack of objective clinical criteria to judge the appropriate amount of mandible resection in patients with ORN remains an unresolved issue that resulted in the development of recurrent ORN in 25 percent of patients. Further investigations are needed to better understand the pathophysiology of ORN to prevent postoperative wound complications and disease recurrence. © 2010 American Academy of Otolaryngology–Head and Neck Surgery Foundation. All rights reserved.
M
andibular osteoradionecrosis (ORN) can be a severe complication of radiation therapy treatment of head and neck cancer. Controversy exists not only in the management of ORN of the mandible, but also in its definition and pathophysiology. Early and limited ORN may respond to conservative management, which usually includes some combination of local wound irrigation, oral hygiene improvement, curettage, debridement, sequestrectomy, bone filling, long-term antibiotic use, and/or hyperbaric oxygen (HBO) therapy. However, cases of advanced ORN with pathologic fractures, large amounts of necrotic tissue, or orocutaneous fistulas are unlikely to respond to conservative treatments. These advanced cases of mandibular ORN are best treated by segmental resection of involved areas.1-3 Mandibular reconstruction has experienced a tremendous evolution with the advent and increasing popularity of microvascular techniques. Vascularized bone grafts are widely recognized to be the most reliable method to achieve singlestage, immediate reconstruction of the mandible.4 Considerable experience has shown that free flaps are extremely reliable despite previous radiation therapy and chemotherapy, with free flap survival rates of over 98 percent4,5 and mandibular bony union rates of over 99 percent6 (Table 1). Studies have shown a higher incidence of early and delayed complications after microvascular reconstruction for ORN compared with non-ORN cases2,7-9 (Table 2). Patterns of failure specific to the reconstruction include hardware extrusion, failure to achieve mandibular continuity, and the development of pathologic fractures. Some of these complications are the direct result of residual or recurrent ORN at the unresected portions of the mandible. The objective of this study is to present our experience in treating mandibular ORN with free flap reconstruction, and to examine the incidence and possible causes for ORN recurrence after segmental mandibulectomy and vascularized bone graft reconstruction.
Received September 10, 2009; revised November 14, 2009; accepted December 3, 2009.
0194-5998/$36.00 © 2010 American Academy of Otolaryngology–Head and Neck Surgery Foundation. All rights reserved. doi:10.1016/j.otohns.2009.12.008
Suh et al
Disease relapse after segmental resection and . . .
587
Table 1 Recent series of microvascular head and neck reconstruction Author
Date
No. of flaps
Flap survival
Prior XRT
Reconstructive complication
Nuara et al5 Pohlenz et al20 Suh et al4 Singh et al21
2009 2007 2003 1999
300 202 400 200
99.7% 97.1% 99.2% 98%
28.7% 40% 37% 33%
17.3% 25.7% 19% 15%
XRT, radiation therapy.
Methods Patients for this study were identified from a database of 1200 consecutive patients who received microvascular free flap reconstruction by a single surgeon (K.E.B) during a 12year period from 1995-2009. This study was approved by the UCLA Institutional Review Board. Inclusion criteria for this study were patients who received microvascular reconstruction of the mandible for the treatment of ORN and had a minimum follow-up of at least three months. All patients had either failed to respond to conservative management, which included HBO, long-term antibiotics, debridement, or had a pathologic fracture of the mandible at the time of their presentation. The free flap database, operative and surgical pathology reports, radiation treatment information, and the office follow-up notes were reviewed. Postoperative complications were defined as the development of orocutaneous fistula, pathologic fracture, plate exposure, hematoma, flap thrombosis, or wound infection. Clinical resolution of ORN was defined as complete restoration of mucosal and cutaneous bone coverage and bony continuity, without clinical evidence of infection. Factors analyzed in this study included age, sex, free flap type, cancer type, history of previous chemotherapy, radiotherapy, surgery, or HBO treatment, American Society of Anesthesiologists (ASA) comorbitity status, history of current alcohol or tobacco use, and length of mandible resected. Mandible plate type was also included in analysis: KLS Threadlock mandible reconstruction plates (KLS Martin, Jacksonville, FL) were used prior to July 2002 and Stryker Universal Locking Mandible reconstruction plates (Stryker Craniomaxillofacial, Portage, MI) were used thereafter. Dental status, including number, health, and position
of teeth, was not available in the medical records and was not included in the analysis. A detailed description of the surgical defect was made by the reconstructive surgeon (K.E.B) in the operative report. Mandible defects were recorded and classified as condyle (C), ramus (R), body (B), or symphyseal (S). Total length of the vascularized bone graft used for defect reconstruction was noted, rounded off to the nearest centimeter. The Kaplan-Meier method was used to estimate the time until reconstructive complication. Complication-free time was measured in months until complication or last recorded follow-up. Both multivariate logistic regression and Cox analysis were used to evaluate independent variables on the risk of wound-related postoperative complications. A multivariate logistic regression was also used to evaluate risk factors for residual ORN following surgery. All statistical analyses were performed with Intercooled Stata 7.0 (StataCorp, College Station, TX).
Results Forty patients met eligibility requirements for this study. Baseline patient demographics are shown in Table 3. There were 26 males and 14 females, with a median age of 62 years (range 38-86 years). All patients received external beam radiation treatment for previous head and neck cancer, and 12 of 40 patients received concurrent chemotherapy. The most common primary tumor was squamous cell carcinoma (38), followed by adenoid cystic carcinoma (1) and epimyoepithelial carcinoma (1). Flap donor sites included fibula (n ⫽ 36), latissimus dorsi/serratus anterior/rib (n ⫽ 3), and iliac crest (n ⫽ 1). Median length of vascularized
Table 2 Recent series of microvascular reconstruction for mandibular osteoradionecrosis Author
Date
No.
Flap survival
Fibula free flap
Prior HBO
Wound complication
Present series Alam et al1 Hirsch et al2 Curi et al8 Gal et al9
2009 2009 2007 2007 2003
40 33 21 5 30
100.00% 100.00% 86.00% 80.00% 100.00%
87.5% 97% 81.0% 100% 79%
47% 76% 66.7% 40% 10%
55% 24% 50% 40% 43%
HBO, hyperbaric oxygen.
588
Otolaryngology–Head and Neck Surgery, Vol 142, No 4, April 2010
Table 3 Distribution of patient characteristics (n ⴝ 40) Characteristic Age (yrs) Median Range Sex Male Female ASA classification 1 2 3 Current tobacco use Current alcohol use Previous chemotherapy Previous surgery Previous HBO Follow-up time (mo) Median Range Length of mandible removed (cm) Median Range
No.
%
62 38-86 26 14
65 35
2 22 16 26 24 12 22 18
5 56 40 65 40 30 55 38
17.44 5.5-82 8 4-24
ASA, American Society of Anesthesiologists [score]; HBO, hyperbaric oxygen.
bone graft used was 8.0 cm (range 4.0-24.0 cm). All vascularized bone grafts were fixated with locking mandibular reconstruction plates. Median follow-up for patients in this study was 17.4 months (range 5.5-82 months). The extent of mandibular resection was determined on a case-by-case basis according to clinical and radiographic criteria. All patients underwent preoperative assessment to determine the presence and extent of radiographic osteolytic changes via panorex and/or CT scan. All patients underwent resection of all visible necrotic bone and soft tissue as determined by surgical exploration. Areas of mandible that exhibited visible cortical bone erosion were debrided. The presence of active bleeding from the margins of the mandibular resection was used as a criterion to judge adequacy of mandibular resection. The location and length of the
Table 4 Reconstructive complications Complication type
No.
%
Free flap donor site complication Pathologic fracture Hardware extrusion Orocutaneous fistula Skin infection or breakdown Recurrent or residual ORN Number of reconstruction plates removed Overall wound complication
2 2 5 6 7 10 17 22
5 5 13 15 18 25 43 55
ORN, osteoradionecrosis.
Figure 1 (A) Patient #30. Initial postoperative panorex after mandible reconstruction with fibular free flap for osteoradionecrosis showing good alignment of right subcondylar ramus. (B) Fifteen months postoperative; now showing a subcondylar fracture with new angulation (arrow) of the subcondylar ramus.
segmental defect, the nature of the soft tissue defect, and the results of preoperative lower extremity angiography determined the type of free flap used. When possible, a two-team approach was used, with the reconstructive surgeon working simultaneously with the ablative surgeon to minimize operative time and the risk for anesthesia-related complications. Table 4 summarizes the complications that occurred in this series. There were no cases of free flap failure. One patient developed arterial thrombosis and was urgently taken back to the operating room for successful salvage. Twenty-two (55%) of 40 patients had postoperative woundrelated complications during the follow-up period. Seventeen (43%) patients required removal of their mandibular reconstruction plate for hardware-related infections or extrusions. Ten (25%) patients developed residual or recurrent ORN. Of these patients, seven of the recurrences involved the condyle and subcondylar segment that was adjacent to the site of ORN, whereas three recurrences occurred in the contralateral mandibular body. Two patients had pathologic fractures through unresected subcondylar segments of the native mandible (Fig 1). The median time to wound-related complication was 10.6 months. Figure 2 shows the KaplanMeier estimation of time until postoperative wound-related complication following segmental mandibulectomy and free flap reconstruction. No independent predictors of postoperative wound-related complications were identified in multivariate logistic regression or Cox regression after controlling for patient
Suh et al
Disease relapse after segmental resection and . . .
589
Table 6 Multivariate analysis: Factor versus residual/recurrent osteoradionecrosis
Figure 2 Kaplan-Meier estimate—time to complication. KaplanMeier estimation of time until wound-related postoperative complication following reconstructive surgery for patients with osteoradionecrosis. Y-axis: percentage of patients without complication. X-axis: time (months).
age, sex, and ASA status. None of the usual predictors of surgical complications (smoking, alcohol use, prior surgery, history of chemotherapy) had a statistically significant effect on postoperative wound-related complications following free flap surgery for ORN (Table 5). From multivariate logistic regression analysis, current tobacco use was associated with a decreased risk of recurrent ORN (odds ratio 0.07, P ⫽ 0.02) after controlling for patient demographics and clinical covariates (Table 6). No other predictors (alcohol use, ASA status, age, sex, plate type, prior chemotherapy, prior HBO therapy) were statistically significant risk factors for recurrent/residual ORN following free flap surgery for ORN.
Discussion ORN is a late effect of radiation that has been described as exposed bone that fails to heal over three months.3 Common signs and symptoms include pain, drainage, and fistula for-
Table 5 Multivariate analysis: Factor versus wound-related complication Patient factor
Odds ratio
SE
P value
95% CI
ASA HBO Length of mandible Alcohol use Hx of chemotherapy Tobacco use Plate type Sex
0.56 1.70 1.38 0.75 0.70 0.84 1.09 0.94
0.44 1.50 0.98 0.56 0.64 0.63 0.97 0.83
0.46 0.55 0.65 0.70 0.70 0.82 0.92 0.94
0.12-2.61 0.30-9.63 0.34-5.57 0.17-3.22 0.11-4.24 0.19-3.69 0.19-6.27 0.16-5.35
CI, confidence interval; ASA, American Society of Anesthesiologists [score]; HBO, hyperbaric oxygen; Hx, history.
Patient factor
Odds ratio
SE
P value
95% CI
Tobacco use HBO Length of mandible ASA Alcohol use Plate type Hx of chemotherapy Sex
0.07 0.23 0.41 0.39 2.29 0.41 2.52 1.69
0.08 0.28 0.41 0.41 2.36 0.52 0.35 2.00
0.02 0.23 0.37 0.38 0.42 0.48 0.51 0.66
0.01-0.67 0.02-2.57 0.06-2.85 0.05-3.10 0.30-17.28 0.03-4.96 0.16-38.55 0.17-17.20
CI, confidence interval; HBO, hyperbaric oxygen; ASA, American Society of Anesthesiologists [score]; Hx, history.
mation. Historically, ORN occurs most commonly in the mandible with an incidence of two to 22 percent.10 However, there has been a decrease in the reported rate of mandibular ORN, presumably from improvements in radiation delivery and improved dental care. Although ORN can occur at any time after radiation therapy, the majority of cases develop within the first three years.3 There appears to be a consensus in the literature that the treatment of advanced ORN requires segmental resection of the affected bone with free flap reconstruction. However, despite aggressive treatment and resection of all necrotic and nonviable bone, a subset of patients will nonetheless develop symptoms of recurrent ORN. There has been a significant evolution in the understanding of mandibular ORN since Marx’s description in 1983.11 Marx described a sequence of events starting with radiation, formulation of hypoxic-hypocellular-hypovascular tissue, then tissue breakdown, and finally a chronic nonhealing wound as the pathophysiology of ORN. This differed from the previously accepted theory that ORN was due to radiation, followed by trauma and infection. Marx showed that ORN is a problem of wound healing that is due to a complex metabolic and tissue homeostatic deficiency, and advocated the use of HBO for treatment of irradiation complications. However, inconsistent treatment results with HBO raised questions that the hypoxic-hypocellular-hypovascular theory fully explains ORN. Radiation-induced fibrosis is a more recent theory for the pathophysiology of ORN.12 This theory suggests that the progression of ORN is due to the activation and dysregulation of fibroblastic activity that leads to atrophic tissue within a previously irradiated field. A cascade of cytokines and free radicals leads to destruction of endothelial cells and vascular thrombosis, which leads to necrosis of microvessels, local ischemia, and tissue loss. The combination of osteoblast death in the mandible after irradiation, failure of the osteoblasts to repopulate, and the excessive proliferation of myofibroblasts results in the bony matrix being replaced
590
Otolaryngology–Head and Neck Surgery, Vol 142, No 4, April 2010
by fibrous tissues. On the basis of this theory, some early promising results are being reported with vascular and antioxidant therapy using pentoxifylline and tocopherol.13 A greater understanding of bisphosphonate-related osteonecrosis of the jaw (BRONJ) has provided additional insight into the pathophysiology of ORN. Work by Marx14 and Ruggiero15 has found an association between bisphosphonate use and the development of mandibular avascular necrosis. Histological analysis of both BRONJ and ORN mandibles has shown an increase in the number of osteoclasts compared with controls.16,17 Other similarities include increasing data supporting the role of Actinomyces as an important factor in ORN and BRONJ.18 ORN specimens with Actinomyces are associated with higher levels of osteoclasts and osteoclast activity. Microscopic analysis suggests that progressive resorption of osteoclasts leads to extensive demineralization of bone and disease progression.19 The evidence suggests that Actinomyces infection and increased osteoclast activity could be factors in the development of nonhealing tissues and bony destruction seen in advanced ORN. A disruption in bone homeostasis favoring osteoclastogenesis may be an ongoing disease process leading to recurrent ORN after extensive free flap reconstruction. Microvascular reconstruction for mandibular ORN has an increased incidence of wound-related postoperative complications compared with free flaps for other indications (Table 2). Some would argue that the impact of previous radiotherapy could account for the increased number of reconstructive complications. Radiotherapy has well-known deleterious effects on local tissue, can damage small vessels, and has the potential to adversely affect microvascular anastomoses. In an analysis of 202 free flap reconstructions, Pohlenz et al20 found that preoperative radiation therapy was one of the main factors associated with increased risk of recipient site complications. Singh et al21 also found a statistically significant association of prior radiotherapy and recipient site complications. However, in an earlier study, our group found that radiation therapy was not a risk factor for free flap reconstructive complications.4 There was a 55 percent incidence of reconstructive complications in this study, usually occurring after the perioperative period. This represents a substantially higher rate of complications compared with microvascular reconstruction for non-ORN cases, which in most series is less than 20 percent (Table 1). These data suggest that ORN itself might be a risk factor for increased free flap complications. Our study did not demonstrate any statistically significant risk factors for postoperative wound-related complications. Our statistical analysis is limited by our small sample size and, therefore, limited power to detect statistically significant associations. However, current smokers appeared to be at reduced risk to develop residual/recurrent ORN after multivariate analysis in this study. This could be secondary to the effects of tobacco use on bone metabolism in ORN patients that are currently unidentified and poorly
understood. For instance, the effects of nicotine on bone metabolism are controversial and unclear. Yuhara et al22 showed that nicotine reduces osteoclastogenesis and stimulates bone deposition in vitro. Rothem et al23 demonstrated that the effects of nicotine on bone metabolism were dosedependent. Low nicotine concentrations were found to cause an increased osteoblast proliferation and bone metabolism, whereas high nicotine concentrations lead to the opposite effect. Both of these studies give insight into the effects of nicotine on bone metabolism. Further research by way of in vitro and in vivo studies is currently underway by the senior author (V.N.) to investigate the osteoclast-driven theory of mandibular ORN. The fibula free flap is the most commonly used method of mandibular reconstruction for the treatment of advanced ORN (Table 2). The fibula free flap offers up to 25 cm of vascularized bone, which is an adequate length to span any segmental defect of the mandible. However, despite aggressive mandibular resection and sufficient vascularized bone for any size defect, 10 (25%) patients in this study developed symptoms of residual or recurrent ORN in unresected portions of the mandible. Of the 10 patients with residual or recurrent ORN, seven of the recurrences were at the condyle and subcondylar region that was adjacent to the segmental resection. Clinically, it is very difficult to assess bleeding from the condyle, as it is dense cortical bone that contains no marrow space. If bone hypovascularity does contribute to the pathogenesis of ORN, then it is possible that resection of the ramus and body may contribute to the development of recurrent ORN in the condyle, through sacrifice of the inferior alveolar neurovascular bundle and loss of condylar periosteum during placement of fixation hardware. Finally, our poor understanding of the pathophysiology of ORN may directly contribute to the lack of objective clinical criteria to judge the appropriate amount of mandible resection in patients with ORN. For instance, the commonly used criteria of intraoperative assessment of bone vascularity and the presence of bleeding from the margins of the resection may be irrelevant to the assessment of the adequacy of mandibular resection if perturbations of bone metabolism and the cellular functions of fibroblasts, osteoclasts, and osteoblasts are more relevant than vascularity in the development of ORN. In light of some reports of complications and increased morbidity after total condyle resection and reconstruction, it is not surprising that surgeons will try to spare the condyle when possible.24 However, there is increasing evidence that total replacement of the condyle can achieve acceptable long-term functional and esthetic results.25 It is now our belief that the condyle should be removed in those cases in which the condyle received radiation doses equivalent to those of adjacent parts of the mandible that are clinically and radiographically involved by ORN, even when there is no visible evidence of condylar cortical bone necrosis or radiographic evidence of condylar ORN.
Suh et al
Disease relapse after segmental resection and . . .
Conclusion This present study confirms that microvascular free flaps are reliable in the treatment of advanced mandibular ORN. Nevertheless, there remains a 55 percent incidence of wound-healing complications, usually occurring in a delayed fashion after the immediate perioperative period. The incidence of hardware-related complications is increased in ORN patients, compared with patients undergoing microvascular flap reconstruction of the mandible for other indications. The lack of objective clinical criteria to judge the appropriate amount of mandible resection in patients with ORN remains an unresolved issue that resulted in the development of recurrent ORN in 25 percent of patients. Further research into the pathophysiology of mandibular ORN is needed and may objectify our treatment criterion.
Author Information From the Division of Head and Neck Surgery (Drs. Suh, Blackwell, Sercarz, Cohen, Tang, Abemayor, and Nabili), and Plastic and Reconstructive Surgery (Dr. Liu), University of California, Los Angeles, Los Angeles, CA. Corresponding author: Jeffrey D. Suh, 62-132 CHS, UCLA Medical Center, Los Angeles, CA 90095-1624. E-mail address: [email protected]. This article was presented at the 2009 AAO–HNSF Annual Meeting & OTO EXPO, San Diego, CA, October 4-7, 2009.
Author Contributions Jeffrey D. Suh, study design, data collection, writer, statistical analysis; Keith E. Blackwell, study design, data collection, writer; Joel A. Sercarz, study design, data collection; Marc Cohen, data collection; Jerome H. Liu, statistical analysis, writer; Christopher G. Tang, data collection; Elliot Abemayor, study design, data collection; Vishad Nabili, study design, data collection, writer.
Disclosures Competing interests: None. Sponsorships: None.
References 1. Alam DS, Nuara M, Christian J. Analysis of outcomes of vascularized flap reconstruction in patients with advanced mandibular osteoradionecrosis. Otolaryngol Head Neck Surg 2009;141:196 –201. 2. Hirch DL, Bell RB, Dierks EJ, et al. Analysis of microvascular free flaps for reconstruction of advanced mandibular osteoradionecrosis: a retrospective cohort study. J Oral Maxillofac Surg 2008;66:2545–56. 3. Teng MS, Futran ND. Osteoradionecrosis of the mandible. Curr Opin Otolaryngol Head Neck Surg 2005;13:217–21. 4. Suh JD, Sercarz JA, Abemayor E, et al. Analysis of outcome and complications in 400 cases of microvascular head and neck reconstruction. Arch Otolaryngol Head Neck Surg 2004;130:962– 6.
591
5. Nuara MJ, Sauder CL, Alam DS. Prospective analysis of outcomes and complications of 300 consecutive microvascular reconstructions. Arch Facial Plast Surg 2009;11:235–9. 6. Knott PD, Suh JD, Nabili V, et al. Evaluation of hardware-related complications in vascularized bone grafts with locking mandibular reconstruction plate fixation. Arch Otolaryngol Head Neck Surg 2007; 133:1302– 6. 7. Sandel HD, Davison SP. Microsurgical reconstruction for radiation necrosis: an evolving disease. J Reconstr Microsurg 2007;23:225–30. 8. Curi MM, Santos MO, Feher O, et al. Management of extensive osteoradionecrosis of the mandible with radical resection and immediate microvascular reconstruction. J Oral Maxillofac Surg 2007;65: 434 – 8. 9. Gal TJ, Bevan Y, Futran ND. Influence of prior hyperbaric oxygen therapy in complications following microvascular reconstruction for advanced osteoradionecrosis. Arch Otolaryngol Head Neck Surg 2003; 129:72– 6. 10. Store G, Boysen M. Mandibular osteoradionecrosis: clinical behavior and diagnostic aspects. Clin Otolaryngol 2000;25:378 – 84. 11. Marx RE. Osteoradionecrosis: a new concept of its pathophysiology. J Oral Maxillofac Surg 1983;41:283– 8. 12. Delanian S, Lefaix JL. The radiation-induced fibroatrophic process: therapeutic perspective via the antioxidant pathway. Radiother Oncol 2004;73:119 –31. 13. Delanian S, Depondt J, Lefaix JL. Major healing of refractory mandible osteoradionecrosis after treatment combining pentoxifylline and tocopherol: a phase II trial. Head Neck 2005;27:114 –23. 14. Marx RE. Pamidronate (Aredia) and zoledronate (Zometa) induced avascular necrosis of the jaws: a growing epidemic. J Oral Maxillofac Surg 2003;61:1115– 8. 15. Ruggiero SL. Bisphosphonate-related osteonecrosis of the jaw (BRONJ): initial discovery and subsequent development. J Oral Maxillofac Surg 2009;67(suppl 1):13– 8. 16. Hansen T, Kunkel M, Weber A, et al. Osteonecrosis of the jaws in patients treated with bisphosphonates— histomorphologic analysis in comparison with infected osteoradionecrosis. J Oral Pathol Med 2006; 35:155– 60. 17. Hansen T, Kirkpatrick CJ, Walter C, et al. Increased numbers of osteoclasts expressing cysteine proteinase cathepsin K in patients with infected osteoradionecrosis and bisphosphonate-associated osteonecrosis—a paradoxical observation? Virchows Arch 2006;449:448 –54. 18. Hansen T, Kunkel M, Kirkpatrick CJ, et al. Actinomycosis in infected osteoradionecrosis— underestimated? Hum Pathol 2006;6:61–7. 19. Lyons A, Ghazali N. Osteoradionecrosis of the jaws: current understanding of its pathophysiology and treatment. Br J Oral Maxillofac Surg 2008;46:653– 60. 20. Pohlenz P, Blessmann M, Heiland M, et al. Postoperative complications in 202 cases of microvascular head and neck reconstruction. J Craniomaxillofac Surg 2007;35:311–5. 21. Singh B, Cordeiro PG, Santamaria E, et al. Factors associated with complications in microvascular reconstruction of head and neck defects. Plast Reconstr Surg 1999;103:403–11. 22. Yuhara S, Kasagi S, Inoue A, et al. Effects of nicotine on cultured cells suggest that it can influence the formation and resorption of bone. Eur J Pharmacol 1999;383:387–93. 23. Rothem DE, Rothem L, Soudry M, et al. Nicotine modulates bone metabolism-associated gene expression in osteoblast cells. J Bone Miner Metab 2009;27:555– 61. 24. Mercuri LG. Considering total temporomandibular joint replacement. Cranio 1999;17:44 – 8. 25. Marx RE, Cillo JE Jr, Broumand V, et al. Outcome analysis of mandibular condylar replacements in tumor and trauma reconstruction: a prospective analysis of 131 cases with long-term follow-up. J Oral Maxillofac Surg 2008;66:2515–23.
ORIGINAL RESEARCH–FACIAL PLASTIC AND RECONSTRUCTIVE SURGERY
Disease relapse after segmental resection and free flap reconstruction for mandibular osteoradionecrosis Jeffrey D. Suh, MD, Keith E. Blackwell, MD, Joel A. Sercarz, MD, Marc Cohen, MD, Jerome H. Liu, MD, MSHS, Christopher G. Tang, MD, Elliot Abemayor, MD, PhD, and Vishad Nabili, MD, Los Angeles, CA No sponsorships or competing interests have been disclosed for this article. ABSTRACT OBJECTIVE: The objective of this study was to assess the outcomes, complications, and incidence of disease recurrence of mandibular osteoradionecrosis (ORN) after resection and microvascular free flap reconstruction. STUDY DESIGN: Case series with chart review. SETTING: Academic medical center. SUBJECTS AND METHODS: Retrospective patient data review of 40 patients with mandibular ORN who were treated by segmental mandibulectomy and microvascular reconstruction between 1995 and 2009. All patients received radiation therapy for previous head and neck cancer, and 12 of 40 patients received concurrent chemotherapy. All patients failed to respond to conservative management. There were 26 males and 14 females, with a median age of 62 years. Median follow-up was 17.4 months. RESULTS: There were no free flap failures. The incidence of wound-related complications was 55 percent. Median time to complication was 10.6 months. Ten (25%) patients developed symptoms of residual or recurrent ORN, with 70 percent of the recurrences arising in unresected condyles that were adjacent to the segmental mandibulectomy. Statistical analysis revealed that current smokers were at reduced risk to develop residual or recurrent ORN. CONCLUSION: This present study confirms that microvascular free flaps are reliable for treatment of advanced mandibular ORN. Nevertheless, there remains a 55 percent incidence of woundhealing complications. The lack of objective clinical criteria to judge the appropriate amount of mandible resection in patients with ORN remains an unresolved issue that resulted in the development of recurrent ORN in 25 percent of patients. Further investigations are needed to better understand the pathophysiology of ORN to prevent postoperative wound complications and disease recurrence. © 2010 American Academy of Otolaryngology–Head and Neck Surgery Foundation. All rights reserved.
M
andibular osteoradionecrosis (ORN) can be a severe complication of radiation therapy treatment of head and neck cancer. Controversy exists not only in the management of ORN of the mandible, but also in its definition and pathophysiology. Early and limited ORN may respond to conservative management, which usually includes some combination of local wound irrigation, oral hygiene improvement, curettage, debridement, sequestrectomy, bone filling, long-term antibiotic use, and/or hyperbaric oxygen (HBO) therapy. However, cases of advanced ORN with pathologic fractures, large amounts of necrotic tissue, or orocutaneous fistulas are unlikely to respond to conservative treatments. These advanced cases of mandibular ORN are best treated by segmental resection of involved areas.1-3 Mandibular reconstruction has experienced a tremendous evolution with the advent and increasing popularity of microvascular techniques. Vascularized bone grafts are widely recognized to be the most reliable method to achieve singlestage, immediate reconstruction of the mandible.4 Considerable experience has shown that free flaps are extremely reliable despite previous radiation therapy and chemotherapy, with free flap survival rates of over 98 percent4,5 and mandibular bony union rates of over 99 percent6 (Table 1). Studies have shown a higher incidence of early and delayed complications after microvascular reconstruction for ORN compared with non-ORN cases2,7-9 (Table 2). Patterns of failure specific to the reconstruction include hardware extrusion, failure to achieve mandibular continuity, and the development of pathologic fractures. Some of these complications are the direct result of residual or recurrent ORN at the unresected portions of the mandible. The objective of this study is to present our experience in treating mandibular ORN with free flap reconstruction, and to examine the incidence and possible causes for ORN recurrence after segmental mandibulectomy and vascularized bone graft reconstruction.
Received September 10, 2009; revised November 14, 2009; accepted December 3, 2009.
0194-5998/$36.00 © 2010 American Academy of Otolaryngology–Head and Neck Surgery Foundation. All rights reserved. doi:10.1016/j.otohns.2009.12.008
Suh et al
Disease relapse after segmental resection and . . .
587
Table 1 Recent series of microvascular head and neck reconstruction Author
Date
No. of flaps
Flap survival
Prior XRT
Reconstructive complication
Nuara et al5 Pohlenz et al20 Suh et al4 Singh et al21
2009 2007 2003 1999
300 202 400 200
99.7% 97.1% 99.2% 98%
28.7% 40% 37% 33%
17.3% 25.7% 19% 15%
XRT, radiation therapy.
Methods Patients for this study were identified from a database of 1200 consecutive patients who received microvascular free flap reconstruction by a single surgeon (K.E.B) during a 12year period from 1995-2009. This study was approved by the UCLA Institutional Review Board. Inclusion criteria for this study were patients who received microvascular reconstruction of the mandible for the treatment of ORN and had a minimum follow-up of at least three months. All patients had either failed to respond to conservative management, which included HBO, long-term antibiotics, debridement, or had a pathologic fracture of the mandible at the time of their presentation. The free flap database, operative and surgical pathology reports, radiation treatment information, and the office follow-up notes were reviewed. Postoperative complications were defined as the development of orocutaneous fistula, pathologic fracture, plate exposure, hematoma, flap thrombosis, or wound infection. Clinical resolution of ORN was defined as complete restoration of mucosal and cutaneous bone coverage and bony continuity, without clinical evidence of infection. Factors analyzed in this study included age, sex, free flap type, cancer type, history of previous chemotherapy, radiotherapy, surgery, or HBO treatment, American Society of Anesthesiologists (ASA) comorbitity status, history of current alcohol or tobacco use, and length of mandible resected. Mandible plate type was also included in analysis: KLS Threadlock mandible reconstruction plates (KLS Martin, Jacksonville, FL) were used prior to July 2002 and Stryker Universal Locking Mandible reconstruction plates (Stryker Craniomaxillofacial, Portage, MI) were used thereafter. Dental status, including number, health, and position
of teeth, was not available in the medical records and was not included in the analysis. A detailed description of the surgical defect was made by the reconstructive surgeon (K.E.B) in the operative report. Mandible defects were recorded and classified as condyle (C), ramus (R), body (B), or symphyseal (S). Total length of the vascularized bone graft used for defect reconstruction was noted, rounded off to the nearest centimeter. The Kaplan-Meier method was used to estimate the time until reconstructive complication. Complication-free time was measured in months until complication or last recorded follow-up. Both multivariate logistic regression and Cox analysis were used to evaluate independent variables on the risk of wound-related postoperative complications. A multivariate logistic regression was also used to evaluate risk factors for residual ORN following surgery. All statistical analyses were performed with Intercooled Stata 7.0 (StataCorp, College Station, TX).
Results Forty patients met eligibility requirements for this study. Baseline patient demographics are shown in Table 3. There were 26 males and 14 females, with a median age of 62 years (range 38-86 years). All patients received external beam radiation treatment for previous head and neck cancer, and 12 of 40 patients received concurrent chemotherapy. The most common primary tumor was squamous cell carcinoma (38), followed by adenoid cystic carcinoma (1) and epimyoepithelial carcinoma (1). Flap donor sites included fibula (n ⫽ 36), latissimus dorsi/serratus anterior/rib (n ⫽ 3), and iliac crest (n ⫽ 1). Median length of vascularized
Table 2 Recent series of microvascular reconstruction for mandibular osteoradionecrosis Author
Date
No.
Flap survival
Fibula free flap
Prior HBO
Wound complication
Present series Alam et al1 Hirsch et al2 Curi et al8 Gal et al9
2009 2009 2007 2007 2003
40 33 21 5 30
100.00% 100.00% 86.00% 80.00% 100.00%
87.5% 97% 81.0% 100% 79%
47% 76% 66.7% 40% 10%
55% 24% 50% 40% 43%
HBO, hyperbaric oxygen.
588
Otolaryngology–Head and Neck Surgery, Vol 142, No 4, April 2010
Table 3 Distribution of patient characteristics (n ⴝ 40) Characteristic Age (yrs) Median Range Sex Male Female ASA classification 1 2 3 Current tobacco use Current alcohol use Previous chemotherapy Previous surgery Previous HBO Follow-up time (mo) Median Range Length of mandible removed (cm) Median Range
No.
%
62 38-86 26 14
65 35
2 22 16 26 24 12 22 18
5 56 40 65 40 30 55 38
17.44 5.5-82 8 4-24
ASA, American Society of Anesthesiologists [score]; HBO, hyperbaric oxygen.
bone graft used was 8.0 cm (range 4.0-24.0 cm). All vascularized bone grafts were fixated with locking mandibular reconstruction plates. Median follow-up for patients in this study was 17.4 months (range 5.5-82 months). The extent of mandibular resection was determined on a case-by-case basis according to clinical and radiographic criteria. All patients underwent preoperative assessment to determine the presence and extent of radiographic osteolytic changes via panorex and/or CT scan. All patients underwent resection of all visible necrotic bone and soft tissue as determined by surgical exploration. Areas of mandible that exhibited visible cortical bone erosion were debrided. The presence of active bleeding from the margins of the mandibular resection was used as a criterion to judge adequacy of mandibular resection. The location and length of the
Table 4 Reconstructive complications Complication type
No.
%
Free flap donor site complication Pathologic fracture Hardware extrusion Orocutaneous fistula Skin infection or breakdown Recurrent or residual ORN Number of reconstruction plates removed Overall wound complication
2 2 5 6 7 10 17 22
5 5 13 15 18 25 43 55
ORN, osteoradionecrosis.
Figure 1 (A) Patient #30. Initial postoperative panorex after mandible reconstruction with fibular free flap for osteoradionecrosis showing good alignment of right subcondylar ramus. (B) Fifteen months postoperative; now showing a subcondylar fracture with new angulation (arrow) of the subcondylar ramus.
segmental defect, the nature of the soft tissue defect, and the results of preoperative lower extremity angiography determined the type of free flap used. When possible, a two-team approach was used, with the reconstructive surgeon working simultaneously with the ablative surgeon to minimize operative time and the risk for anesthesia-related complications. Table 4 summarizes the complications that occurred in this series. There were no cases of free flap failure. One patient developed arterial thrombosis and was urgently taken back to the operating room for successful salvage. Twenty-two (55%) of 40 patients had postoperative woundrelated complications during the follow-up period. Seventeen (43%) patients required removal of their mandibular reconstruction plate for hardware-related infections or extrusions. Ten (25%) patients developed residual or recurrent ORN. Of these patients, seven of the recurrences involved the condyle and subcondylar segment that was adjacent to the site of ORN, whereas three recurrences occurred in the contralateral mandibular body. Two patients had pathologic fractures through unresected subcondylar segments of the native mandible (Fig 1). The median time to wound-related complication was 10.6 months. Figure 2 shows the KaplanMeier estimation of time until postoperative wound-related complication following segmental mandibulectomy and free flap reconstruction. No independent predictors of postoperative wound-related complications were identified in multivariate logistic regression or Cox regression after controlling for patient
Suh et al
Disease relapse after segmental resection and . . .
589
Table 6 Multivariate analysis: Factor versus residual/recurrent osteoradionecrosis
Figure 2 Kaplan-Meier estimate—time to complication. KaplanMeier estimation of time until wound-related postoperative complication following reconstructive surgery for patients with osteoradionecrosis. Y-axis: percentage of patients without complication. X-axis: time (months).
age, sex, and ASA status. None of the usual predictors of surgical complications (smoking, alcohol use, prior surgery, history of chemotherapy) had a statistically significant effect on postoperative wound-related complications following free flap surgery for ORN (Table 5). From multivariate logistic regression analysis, current tobacco use was associated with a decreased risk of recurrent ORN (odds ratio 0.07, P ⫽ 0.02) after controlling for patient demographics and clinical covariates (Table 6). No other predictors (alcohol use, ASA status, age, sex, plate type, prior chemotherapy, prior HBO therapy) were statistically significant risk factors for recurrent/residual ORN following free flap surgery for ORN.
Discussion ORN is a late effect of radiation that has been described as exposed bone that fails to heal over three months.3 Common signs and symptoms include pain, drainage, and fistula for-
Table 5 Multivariate analysis: Factor versus wound-related complication Patient factor
Odds ratio
SE
P value
95% CI
ASA HBO Length of mandible Alcohol use Hx of chemotherapy Tobacco use Plate type Sex
0.56 1.70 1.38 0.75 0.70 0.84 1.09 0.94
0.44 1.50 0.98 0.56 0.64 0.63 0.97 0.83
0.46 0.55 0.65 0.70 0.70 0.82 0.92 0.94
0.12-2.61 0.30-9.63 0.34-5.57 0.17-3.22 0.11-4.24 0.19-3.69 0.19-6.27 0.16-5.35
CI, confidence interval; ASA, American Society of Anesthesiologists [score]; HBO, hyperbaric oxygen; Hx, history.
Patient factor
Odds ratio
SE
P value
95% CI
Tobacco use HBO Length of mandible ASA Alcohol use Plate type Hx of chemotherapy Sex
0.07 0.23 0.41 0.39 2.29 0.41 2.52 1.69
0.08 0.28 0.41 0.41 2.36 0.52 0.35 2.00
0.02 0.23 0.37 0.38 0.42 0.48 0.51 0.66
0.01-0.67 0.02-2.57 0.06-2.85 0.05-3.10 0.30-17.28 0.03-4.96 0.16-38.55 0.17-17.20
CI, confidence interval; HBO, hyperbaric oxygen; ASA, American Society of Anesthesiologists [score]; Hx, history.
mation. Historically, ORN occurs most commonly in the mandible with an incidence of two to 22 percent.10 However, there has been a decrease in the reported rate of mandibular ORN, presumably from improvements in radiation delivery and improved dental care. Although ORN can occur at any time after radiation therapy, the majority of cases develop within the first three years.3 There appears to be a consensus in the literature that the treatment of advanced ORN requires segmental resection of the affected bone with free flap reconstruction. However, despite aggressive treatment and resection of all necrotic and nonviable bone, a subset of patients will nonetheless develop symptoms of recurrent ORN. There has been a significant evolution in the understanding of mandibular ORN since Marx’s description in 1983.11 Marx described a sequence of events starting with radiation, formulation of hypoxic-hypocellular-hypovascular tissue, then tissue breakdown, and finally a chronic nonhealing wound as the pathophysiology of ORN. This differed from the previously accepted theory that ORN was due to radiation, followed by trauma and infection. Marx showed that ORN is a problem of wound healing that is due to a complex metabolic and tissue homeostatic deficiency, and advocated the use of HBO for treatment of irradiation complications. However, inconsistent treatment results with HBO raised questions that the hypoxic-hypocellular-hypovascular theory fully explains ORN. Radiation-induced fibrosis is a more recent theory for the pathophysiology of ORN.12 This theory suggests that the progression of ORN is due to the activation and dysregulation of fibroblastic activity that leads to atrophic tissue within a previously irradiated field. A cascade of cytokines and free radicals leads to destruction of endothelial cells and vascular thrombosis, which leads to necrosis of microvessels, local ischemia, and tissue loss. The combination of osteoblast death in the mandible after irradiation, failure of the osteoblasts to repopulate, and the excessive proliferation of myofibroblasts results in the bony matrix being replaced
590
Otolaryngology–Head and Neck Surgery, Vol 142, No 4, April 2010
by fibrous tissues. On the basis of this theory, some early promising results are being reported with vascular and antioxidant therapy using pentoxifylline and tocopherol.13 A greater understanding of bisphosphonate-related osteonecrosis of the jaw (BRONJ) has provided additional insight into the pathophysiology of ORN. Work by Marx14 and Ruggiero15 has found an association between bisphosphonate use and the development of mandibular avascular necrosis. Histological analysis of both BRONJ and ORN mandibles has shown an increase in the number of osteoclasts compared with controls.16,17 Other similarities include increasing data supporting the role of Actinomyces as an important factor in ORN and BRONJ.18 ORN specimens with Actinomyces are associated with higher levels of osteoclasts and osteoclast activity. Microscopic analysis suggests that progressive resorption of osteoclasts leads to extensive demineralization of bone and disease progression.19 The evidence suggests that Actinomyces infection and increased osteoclast activity could be factors in the development of nonhealing tissues and bony destruction seen in advanced ORN. A disruption in bone homeostasis favoring osteoclastogenesis may be an ongoing disease process leading to recurrent ORN after extensive free flap reconstruction. Microvascular reconstruction for mandibular ORN has an increased incidence of wound-related postoperative complications compared with free flaps for other indications (Table 2). Some would argue that the impact of previous radiotherapy could account for the increased number of reconstructive complications. Radiotherapy has well-known deleterious effects on local tissue, can damage small vessels, and has the potential to adversely affect microvascular anastomoses. In an analysis of 202 free flap reconstructions, Pohlenz et al20 found that preoperative radiation therapy was one of the main factors associated with increased risk of recipient site complications. Singh et al21 also found a statistically significant association of prior radiotherapy and recipient site complications. However, in an earlier study, our group found that radiation therapy was not a risk factor for free flap reconstructive complications.4 There was a 55 percent incidence of reconstructive complications in this study, usually occurring after the perioperative period. This represents a substantially higher rate of complications compared with microvascular reconstruction for non-ORN cases, which in most series is less than 20 percent (Table 1). These data suggest that ORN itself might be a risk factor for increased free flap complications. Our study did not demonstrate any statistically significant risk factors for postoperative wound-related complications. Our statistical analysis is limited by our small sample size and, therefore, limited power to detect statistically significant associations. However, current smokers appeared to be at reduced risk to develop residual/recurrent ORN after multivariate analysis in this study. This could be secondary to the effects of tobacco use on bone metabolism in ORN patients that are currently unidentified and poorly
understood. For instance, the effects of nicotine on bone metabolism are controversial and unclear. Yuhara et al22 showed that nicotine reduces osteoclastogenesis and stimulates bone deposition in vitro. Rothem et al23 demonstrated that the effects of nicotine on bone metabolism were dosedependent. Low nicotine concentrations were found to cause an increased osteoblast proliferation and bone metabolism, whereas high nicotine concentrations lead to the opposite effect. Both of these studies give insight into the effects of nicotine on bone metabolism. Further research by way of in vitro and in vivo studies is currently underway by the senior author (V.N.) to investigate the osteoclast-driven theory of mandibular ORN. The fibula free flap is the most commonly used method of mandibular reconstruction for the treatment of advanced ORN (Table 2). The fibula free flap offers up to 25 cm of vascularized bone, which is an adequate length to span any segmental defect of the mandible. However, despite aggressive mandibular resection and sufficient vascularized bone for any size defect, 10 (25%) patients in this study developed symptoms of residual or recurrent ORN in unresected portions of the mandible. Of the 10 patients with residual or recurrent ORN, seven of the recurrences were at the condyle and subcondylar region that was adjacent to the segmental resection. Clinically, it is very difficult to assess bleeding from the condyle, as it is dense cortical bone that contains no marrow space. If bone hypovascularity does contribute to the pathogenesis of ORN, then it is possible that resection of the ramus and body may contribute to the development of recurrent ORN in the condyle, through sacrifice of the inferior alveolar neurovascular bundle and loss of condylar periosteum during placement of fixation hardware. Finally, our poor understanding of the pathophysiology of ORN may directly contribute to the lack of objective clinical criteria to judge the appropriate amount of mandible resection in patients with ORN. For instance, the commonly used criteria of intraoperative assessment of bone vascularity and the presence of bleeding from the margins of the resection may be irrelevant to the assessment of the adequacy of mandibular resection if perturbations of bone metabolism and the cellular functions of fibroblasts, osteoclasts, and osteoblasts are more relevant than vascularity in the development of ORN. In light of some reports of complications and increased morbidity after total condyle resection and reconstruction, it is not surprising that surgeons will try to spare the condyle when possible.24 However, there is increasing evidence that total replacement of the condyle can achieve acceptable long-term functional and esthetic results.25 It is now our belief that the condyle should be removed in those cases in which the condyle received radiation doses equivalent to those of adjacent parts of the mandible that are clinically and radiographically involved by ORN, even when there is no visible evidence of condylar cortical bone necrosis or radiographic evidence of condylar ORN.
Suh et al
Disease relapse after segmental resection and . . .
Conclusion This present study confirms that microvascular free flaps are reliable in the treatment of advanced mandibular ORN. Nevertheless, there remains a 55 percent incidence of wound-healing complications, usually occurring in a delayed fashion after the immediate perioperative period. The incidence of hardware-related complications is increased in ORN patients, compared with patients undergoing microvascular flap reconstruction of the mandible for other indications. The lack of objective clinical criteria to judge the appropriate amount of mandible resection in patients with ORN remains an unresolved issue that resulted in the development of recurrent ORN in 25 percent of patients. Further research into the pathophysiology of mandibular ORN is needed and may objectify our treatment criterion.
Author Information From the Division of Head and Neck Surgery (Drs. Suh, Blackwell, Sercarz, Cohen, Tang, Abemayor, and Nabili), and Plastic and Reconstructive Surgery (Dr. Liu), University of California, Los Angeles, Los Angeles, CA. Corresponding author: Jeffrey D. Suh, 62-132 CHS, UCLA Medical Center, Los Angeles, CA 90095-1624. E-mail address: [email protected]. This article was presented at the 2009 AAO–HNSF Annual Meeting & OTO EXPO, San Diego, CA, October 4-7, 2009.
Author Contributions Jeffrey D. Suh, study design, data collection, writer, statistical analysis; Keith E. Blackwell, study design, data collection, writer; Joel A. Sercarz, study design, data collection; Marc Cohen, data collection; Jerome H. Liu, statistical analysis, writer; Christopher G. Tang, data collection; Elliot Abemayor, study design, data collection; Vishad Nabili, study design, data collection, writer.
Disclosures Competing interests: None. Sponsorships: None.
References 1. Alam DS, Nuara M, Christian J. Analysis of outcomes of vascularized flap reconstruction in patients with advanced mandibular osteoradionecrosis. Otolaryngol Head Neck Surg 2009;141:196 –201. 2. Hirch DL, Bell RB, Dierks EJ, et al. Analysis of microvascular free flaps for reconstruction of advanced mandibular osteoradionecrosis: a retrospective cohort study. J Oral Maxillofac Surg 2008;66:2545–56. 3. Teng MS, Futran ND. Osteoradionecrosis of the mandible. Curr Opin Otolaryngol Head Neck Surg 2005;13:217–21. 4. Suh JD, Sercarz JA, Abemayor E, et al. Analysis of outcome and complications in 400 cases of microvascular head and neck reconstruction. Arch Otolaryngol Head Neck Surg 2004;130:962– 6.
591
5. Nuara MJ, Sauder CL, Alam DS. Prospective analysis of outcomes and complications of 300 consecutive microvascular reconstructions. Arch Facial Plast Surg 2009;11:235–9. 6. Knott PD, Suh JD, Nabili V, et al. Evaluation of hardware-related complications in vascularized bone grafts with locking mandibular reconstruction plate fixation. Arch Otolaryngol Head Neck Surg 2007; 133:1302– 6. 7. Sandel HD, Davison SP. Microsurgical reconstruction for radiation necrosis: an evolving disease. J Reconstr Microsurg 2007;23:225–30. 8. Curi MM, Santos MO, Feher O, et al. Management of extensive osteoradionecrosis of the mandible with radical resection and immediate microvascular reconstruction. J Oral Maxillofac Surg 2007;65: 434 – 8. 9. Gal TJ, Bevan Y, Futran ND. Influence of prior hyperbaric oxygen therapy in complications following microvascular reconstruction for advanced osteoradionecrosis. Arch Otolaryngol Head Neck Surg 2003; 129:72– 6. 10. Store G, Boysen M. Mandibular osteoradionecrosis: clinical behavior and diagnostic aspects. Clin Otolaryngol 2000;25:378 – 84. 11. Marx RE. Osteoradionecrosis: a new concept of its pathophysiology. J Oral Maxillofac Surg 1983;41:283– 8. 12. Delanian S, Lefaix JL. The radiation-induced fibroatrophic process: therapeutic perspective via the antioxidant pathway. Radiother Oncol 2004;73:119 –31. 13. Delanian S, Depondt J, Lefaix JL. Major healing of refractory mandible osteoradionecrosis after treatment combining pentoxifylline and tocopherol: a phase II trial. Head Neck 2005;27:114 –23. 14. Marx RE. Pamidronate (Aredia) and zoledronate (Zometa) induced avascular necrosis of the jaws: a growing epidemic. J Oral Maxillofac Surg 2003;61:1115– 8. 15. Ruggiero SL. Bisphosphonate-related osteonecrosis of the jaw (BRONJ): initial discovery and subsequent development. J Oral Maxillofac Surg 2009;67(suppl 1):13– 8. 16. Hansen T, Kunkel M, Weber A, et al. Osteonecrosis of the jaws in patients treated with bisphosphonates— histomorphologic analysis in comparison with infected osteoradionecrosis. J Oral Pathol Med 2006; 35:155– 60. 17. Hansen T, Kirkpatrick CJ, Walter C, et al. Increased numbers of osteoclasts expressing cysteine proteinase cathepsin K in patients with infected osteoradionecrosis and bisphosphonate-associated osteonecrosis—a paradoxical observation? Virchows Arch 2006;449:448 –54. 18. Hansen T, Kunkel M, Kirkpatrick CJ, et al. Actinomycosis in infected osteoradionecrosis— underestimated? Hum Pathol 2006;6:61–7. 19. Lyons A, Ghazali N. Osteoradionecrosis of the jaws: current understanding of its pathophysiology and treatment. Br J Oral Maxillofac Surg 2008;46:653– 60. 20. Pohlenz P, Blessmann M, Heiland M, et al. Postoperative complications in 202 cases of microvascular head and neck reconstruction. J Craniomaxillofac Surg 2007;35:311–5. 21. Singh B, Cordeiro PG, Santamaria E, et al. Factors associated with complications in microvascular reconstruction of head and neck defects. Plast Reconstr Surg 1999;103:403–11. 22. Yuhara S, Kasagi S, Inoue A, et al. Effects of nicotine on cultured cells suggest that it can influence the formation and resorption of bone. Eur J Pharmacol 1999;383:387–93. 23. Rothem DE, Rothem L, Soudry M, et al. Nicotine modulates bone metabolism-associated gene expression in osteoblast cells. J Bone Miner Metab 2009;27:555– 61. 24. Mercuri LG. Considering total temporomandibular joint replacement. Cranio 1999;17:44 – 8. 25. Marx RE, Cillo JE Jr, Broumand V, et al. Outcome analysis of mandibular condylar replacements in tumor and trauma reconstruction: a prospective analysis of 131 cases with long-term follow-up. J Oral Maxillofac Surg 2008;66:2515–23.