Managing Mandibular Osteoradionecrosis

Abstract

Objective

Mandibular osteoradionecrosis (MORN) is a morbid complication of head and neck radiation therapy. Recent advances in surgical and medical therapies underscore the need for a shift in traditional treatment paradigms and a disease grading system that can guide appropriate management.

Data Sources

Pubmed/MEDLINE.

Review Methods

We conducted a detailed review of publications related to MORN, specifically focusing on its staging and management techniques. Articles meeting inclusion criteria were synthesized into a final narrative review.

Conclusion

There has been a paradigm shift away from hyperbaric oxygen therapy in the management of MORN. Growing evidence for the efficacy of pentoxifylline and tocopherol in early‐stage disease and novel surgical techniques to manage moderate and late‐stage disease warrant an updated staging stratification which is proposed.

Implications for Practice

This review summarizes the clinical efficacy of established and novel therapeutic modalities currently available in treating MORN, emphasizing the significant advances achieved over the last decade. It introduces a contemporary staging and treatment algorithm which incorporates traditional, evidence‐supported surgical and medical management with effective early intervention strategies.

Mandibular osteoradionecrosis (MORN) is a dreaded late complication of radiation therapy utilized in the treatment of head and neck cancer. It carries a progressive course that ultimately leads to significant morbidity and negatively impacts quality of life. Patients are affected by trismus, recurrent swelling, and pain and often require frequent courses of antibiotics and repeated debridement. End‐stage disease manifests as full‐thickness progression resulting in orocutaneous fistulas and nonhealing pathologic fractures with malocclusion and oral motor dysfunction.

To date, treatment of MORN has been challenging and often noncurative. Employed management strategies depend on the presenting disease stage and prior failed therapeutic attempts. Early and moderate‐stage osteoradionecrosis (ORN) have been traditionally managed through various modalities, including prudent oral hygiene, oral or parenteral antibiotics, local sequestrectomy or saucerization, and hyperbaric oxygen therapy (HBOT). Late‐stage ORN requires segmental resection and free tissue microvascular reconstruction with osteocutaneous flaps which, although effective, incur significant morbidity and high complication rates. ¹

Substantial evidence has emerged evaluating newer available treatment options for MORN, including targeted medical therapies and early‐intervention, low‐morbidity surgical techniques. Long‐term efficacy has been demonstrated in arresting ORN progression with the anterolateral thigh fascia lata free flap (rescue flap). ² , ³ , ⁴ , ⁵ Moreover, prospective clinical trials have contested the efficacy of HBOT. These new findings have challenged the traditional approach to treating ORN. The purpose of this study is to conduct a comprehensive review of all existing literature on patients with MORN, and evaluate the trends and evolution in management, both medical and surgical, and their impact on staging and treatment algorithms in the contemporary era.

Methods

A comprehensive search was performed in the PubMed/MEDLINE databases for all existing literature up to September 8, 2024, on any English literature pertaining to MORN. The keywords during initial screening included “osteoradionecrosis,” “osteonecrosis,” “osteomyelitis,” “mandible,” “mandibular,” “jaw,” “head and neck.” Further secondary searches were also performed to identify prominent articles on staging and both medical and surgical management of these conditions. Dedicated searches on specific topics such as “pentoxifylline,” “tocopherol,” “clodronate,” and “hyperbaric oxygen therapy” were also performed. Inclusion criteria were as follows: any literature investigating the pathophysiology, work‐up, and management of MORN. References from the included articles were also evaluated. Studies on ORN of nonmandibular head and neck subsites, studies related exclusively to osteomyelitis, and animal studies were excluded. Articles were initially screened for relevance by K.A., and authors M.A.F. and K.A. jointly performed an independent, in‐depth review of articles based on the above‐noted criteria. All authors reviewed the selected articles and collaborated to create this comprehensive review. Institutional review board approval was not necessary for this study.

Discussion

MORN is currently defined as nonhealing irradiated bone exposure of greater than 3 months duration without evidence of neoplasia. ⁶ , ⁷ , ⁸ Several theories have emerged to explain its pathophysiologic mechanism. The earliest sequence described (1) radiation, (2) hypoxic‐hypocellular‐hypovascular tissue, (3) tissue breakdown, and (4) chronic nonhealing wounds. ⁹ Others have further theorized that ORN is a consequence of activation and dysregulation of fibroblastic activity, propagated by free radical production, endothelial dysfunction, microvascular thrombosis, and local inflammation. This leads to fibroatrophic tissue prone to chronic wound formation, termed a radiation‐induced fibroatrophic (RIF) process. ⁷ , ¹⁰ , ¹¹ Still, others have demonstrated the role of bacterial inoculation and infection in ORN pathogenesis. ⁷ , ¹²

Indebted to the advancements in our fundamental knowledge of ORN over the last few decades, the advent of intensity‐modulated radiotherapy, and the routine performance of prophylactic dental extractions, there has been a 20% decrease in the incidence of ORN to about 4% to 8% in the modern era. ¹³ , ¹⁴ Despite this achievement, ORN continues to plague survivors and presents numerous challenges in management. ¹⁵

Current Staging

There are currently at least 9 staging classification systems described for MORN, recently reviewed in depth by Chronopoulos et al. ⁶ , ⁹ , ¹⁶ , ¹⁷ , ¹⁸ , ¹⁹ , ²⁰ , ²¹ , ²² , ²³ The basis upon which each classification strategy is built can be broadly defined as (1) clinical findings only, (2) clinical and radiographic findings, and (3) response to therapy. Within these grading systems, early‐stage and late‐stage classifications are similar, while the variability lies primarily in moderate disease classifications. Early stages often describe areas of bone exposure or sequestra, while late stages commonly involve full‐thickness defects, pathologic fractures, fistulization, and resistance to therapy.

For many years, the Marx classification, based on clinical response to HBOT, was the most widely utilized staging system (Table 1). ⁹ With the effectiveness of HBOT in question, the Marx classification has fallen out of favor. In the early 2000s, Schwartz and Kagan ²² and Notani et al ²³ described the 2 most modern staging systems which are the most utilized at present. Table 1 summarizes these staging systems.

Table 1.

Established Classification Systems for Mandibular Osteoradionecrosis

Classification system	Basis of classification	Stages	Description of stages
Marx 9	Clinical response to HBOT	Stage I	Exposed alveolar bone without pathologic fracture; responds to HBOT
		Stage II	No clinical improvement following HBOT; requires sequestrectomy and saucerization
		Stage III	Full‐thickness disease, orocutaneous fistula, or pathologic fracture; requires segmental resection and reconstruction
Schwartz and Kagan 22	Clinical and imaging findings	Stage I	Minimal soft‐tissue ulceration and limited cortical bone exposure
		Stage II	Localized cortical and underlying medullary bone involvement
		Stage IIa	Minimal soft‐tissue ulceration
		Stage IIb	Orocutaneous fistula and mild soft‐tissue necrosis
		Stage III	Full‐thickness disease and/or pathologic fracture
Notani et al 23	Clinical findings. IANC is used as critical landmark	Stage I	Confined to alveolar bone
		Stage II	Limited to alveolar bone and mandible above level of IANC
		Stage III	Any involvement below level of IANC, including pathologic fracture and orocutaneous fistula
Epstein et al 18	Disease progression	Stage I	Resolved/healed
		Stage Ia	No pathologic fracture
		Stage Ib	Pathologic fracture
		Stage II	Chronic nonprogressive/persistent
		Stage IIa	No pathologic fracture
		Stage IIb	Pathologic fracture
		Stage III	Active progressive
		Stage IIIa	No pathologic fracture
		Stage IIIb	Pathologic fracture

Abbreviations: HBOT, hyperbaric oxygen therapy; IANC, inferior alveolar nerve canal.

Current Treatment

Treatment paradigms are based on the disease stage. In early or moderate disease, options include (1) close observation, (2) early surgical sequestrectomy or saucerization, (3) HBOT, and more recently, (4) targeted medical therapy. ¹⁴ Unfortunately, no robust trials support these options as gold‐standard therapy. When treatment options are ineffective for early disease, progression necessitates segmental resection and microvascular reconstruction with osteocutaneous free flaps. ¹⁴ While effective in controlling ORN and restoring mandibular continuity, this intervention incurs high morbidity.

The management of MORN ranges widely. Traditionally, treatment has leaned toward conservative measures due to evidence that manipulation of the site often leads to further trauma, which further exacerbates an already compromised wound. Adopting a “watch and wait” policy, however, must be practiced with caution for 2 reasons: (1) delaying treatment risks disease progression, and (2) the everchanging definitions of ORN have led to heterogeneity within the literature concerning diagnosis; therefore, previously‐reported spontaneous resolution rates may be overestimated. ⁶ , ⁷ , ¹⁴ To date, no large‐scale studies have evaluated the efficacy of nonsurgical and nonhyperbaric conservative management of MORN.

In treating advanced disease, the established gold standard is segmental resection with osteocutaneous free‐flap reconstruction. ⁸ , ¹⁴ , ²⁴ , ²⁵ Necrotic bone is excised to native bleeding bone and replaced with nonirradiated well‐vascularized bone. This surgery introduces a robust blood supply to the region, allowing wound healing and symptomatic resolution. ²⁶ , ²⁷ Although nearly 90% effective, the morbidity of the osteocutaneous flap is not inconsequential. Patients with ORN are at higher risk for flap complications (21%‐60%) and flap loss (as high as 20%). ²⁸ , ²⁹ , ³⁰ , ³¹

Evolution of ORN Management

In 1970, Meyer first described antibiotics with or without surgical debridement as a viable option in managing MORN. ³² This model shifted in 1983 when Marx introduced his pathophysiologic theory and staging system centered around the use of HBOT. ⁹ , ³³ , ³⁴ The paradigm of ORN management remained relatively stagnant until Delanian and Lefaix introduced the RIF theory in 2004. ¹¹ Their findings suggested a role for antioxidant and antifibrotic agents in the nonsurgical management of MORN. ³⁵ As such, several targeted pharmacologic therapies—notably, pentoxifylline, tocopherol (vitamin‐E), and clodronate (PENTOCLO)—have since been investigated for treatment. In addition to expanding available pharmacotherapy, the landscape of MORN has continued to shift concerning the utility of HBOT and novel surgical techniques.

PENTOCLO

Pentoxifylline is a vasodilator with antitumor necrosis factor, anti‐inflammatory, and antiplatelet aggregation properties that increase erythrocyte flexibility and decrease blood viscosity. ³⁶ Tocopherol is a potent antioxidant that prevents oxidative reactions driven by reactive oxidative species such as lipid peroxidation of cell membranes, including the endothelium, which may impact arterial blood flow. It is also a partial inhibitor of transforming growth factor‐β1, which reduces inflammation and tissue fibrosis. ³⁶ , ³⁷ , ³⁸ In isolation, neither of these drugs have demonstrated curative efficacy in ORN. However, due to distinct mechanisms, combinations appear to confer synergism.

Early clinical trials conducted by Delanian et al demonstrated the antifibrotic properties of concurrent pentoxifylline and tocopherol (PENTO) in reversing RIF involving the skin and underlying tissues. ³⁹ Findings were supported with a Phase II prospective trial of 18 patients with early‐stage MORN treated with PENTO. ⁴⁰ These results triggered the emergence of several trials and systematic reviews. Early studies by Delanian et al ⁴¹ (2011, n = 54) and McLeod et al ⁴² (2012, n = 12) included patients with a wide range of disease severity. Patients served as their own controls, and outcomes were assessed by improvement in the Epstein stage (summarized in Table 1) and Subjective Objective Medical Management and Analytic (SOMA) evaluation of injury scores. Given that the basis of the Epstein system relies on clinical disease progression, it is limited in that patients may have a pathologic fracture and still be considered Stage I if there appears to be clinical resolution. ¹⁸ Overall, results were variable and limited by subjective assessment of outcomes. In the McLeod trial, a general improvement in Epstein stage (5 of 12 patients, 41.7%) and SOMA scores (8 of 12 patients, 66.7%) were seen; however, only a few (2 of 12 patients, 16.7%) demonstrated complete resolution. ⁴² Results were more pronounced in the Delanian trial, however, these have yet to be reproduced. ⁴¹

In 2014, Lyons et al evaluated the impact of the extent of bone exposure on response to PENTO therapy. ⁴³ Out of 85 patients, the authors classified patients with less than 2.5 cm of exposed bone as Stage I (n = 28). They noted complete resolution (n = 17, 60.7%), improvement (n = 5, 17.9%), and stability (n = 6, 21.4%). Progression was not seen in patients with Stage I disease. On the other hand, symptomatic patients with >2.5 cm length of bone exposure demonstrated higher rates of progression despite concomitant debridement. Recently, Kolokythas et al conducted the most extensive systematic review of PENTO outcomes comprising 211 patients across 7 studies. ³⁶ Meta‐analysis revealed overall, 62.7% of patients had either full recovery or sufficient improvement in clinical symptoms, bone exposure, and SOMA scores that negated the need for further surgical intervention. A total of 7.5% of patients continued to progress. Studies were limited by a lack of clarity about follow‐up periods and heterogeneity in the concomitant use of steroids, antibiotics, and clodronate in 3 of 7 studies. Most importantly, radiographic outcomes were only reported in 1 study, and staging information was not comprehensive. ⁴⁴

The most thorough investigation to include radiographic outcomes was conducted by Patel et al, which included 148 patients who were treated with either PENTO or PENTOCLO. ⁴⁵ The cohort was one‐half Notani I, one‐third Notani III, and the remainder were Notani II. The purported overall resolution rate was 54.4% over a follow‐up of 12.9 months. Medical management was most effective in Notani I and noninfected ORN (about 75% success rate) compared to Notani III (about 25% success). ORN severity was inversely correlated with the likelihood of response to medical management. The authors attributed inconsistent Notani III results to the breadth of disease severity in this group ranging from the inferior alveolar nerve canal (IANC) involvement to full‐thickness fractures and/or fistulas.

Clodronate is a first‐generation oral alkylbisphosphonate that reduces osteoclast activity, inhibits fibroblast and macrophage proliferation, and promotes bone formation by osteoblasts. ⁴⁶ , ⁴⁷ As such, clodronate has been suggested as an addition to PENTO in managing MORN. ⁴⁰ The most prominent clinical trial for PENTOCLO was conducted by Delanian et al in 2011, comprising 54 patients. ⁴¹ The authors reported that all patients experienced a complete recovery in a median of 9 months. Recovery was assessed clinically primarily by SOMA scores, without radiographic correlation or baseline staging. The authors concluded that PENTOCLO was effective, however, they did not recommend its use in fracture or exposed bone greater than 1 cm. Overall, a systematic review in 2018 revealed that studies evaluating PENTOCLO therapy were scarce, heterogeneous, and predominantly qualitative. ⁴⁸

Highlighted in a recent systematic review, both PENTO and PENTOCLO carry minimal adverse side effects. ⁴⁹ Thus far, no trials have focused on directly comparing PENTO to PENTOCLO in MORN. However, as a secondary outcome, Patel and colleagues did find PENTO to be significantly superior to PENTOCLO. ⁴⁵ Paradoxically, bisphosphonates (both oral and intravenous) are associated with medication‐related osteonecrosis of the jaw. ⁵⁰ The incidence of orally administrated bisphosphonate‐induced osteonecrosis of the mandible ranges from 2.5% to 27.3%. ⁵¹ Though this risk is lower in clodronate than in other oral bisphosphonates (eg, aminobisphosphonates), it is not negligible. ⁵²

The most compelling evidence supporting the efficacy of PENTO or PENTOCLO pertains to only earlier stages of the disease corresponding to Lyons and Notani I and II classifications. ⁴¹ , ⁴⁸ , ⁵³ , ⁵⁴ Currently, there are no active prospective investigations on PENTO or PENTOCLO. The only active randomized‐controlled trial investigating these medications is tocopherol in the prophylactic setting at the time of radiation therapy. ⁵⁵ , ⁵⁶

HBOT

In the search for disease‐modifying therapies to halt the progression of MORN, HBOT was introduced in the 1970s. Its mechanism was theorized to be an increase in local tissue oxygen tension, which would improve wound healing. ⁵⁷ Marx and colleagues integrated HBOT into his staging system, and the treatment algorithm centered entirely around HBOT. ⁹ , ¹⁴ , ²⁸ , ³⁴ Initial HBOT trials for treating MORN have been heavily questioned in recent years due to their small cohort sizes, unblinded methodology, and nonreplicability. Initial trials reported a 15% resolution rate at best, with most patients still requiring additional surgical intervention. ¹⁴ , ²⁸ , ³⁴ , ⁵⁷

Over the last 20 years, several high‐quality studies have challenged the efficacy of HBOT. In 2004, a randomized, prospective multicenter, double‐blind, placebo‐controlled trial demonstrated no benefit from HBOT in overt early and moderate‐stage MORN. In fact, the trial was terminated early due to potentially worse outcomes in the HBO group compared to placebo. ⁵⁸ In 2017, Sultan et al produced a multidisciplinary guideline and review that recommended against routine HBOT in the prevention or management of MORN due to the lack of consistent evidence to support it. ⁵⁷ A meta‐analysis of 4 randomized controlled trials and 7 cohort studies corroborated these recommendations, asserting that HBOT is not a substitute for surgical and antibiotic treatment. Moreover, it was not shown to be more effective than antifibrotic medication and antibiotics. ⁵⁹ Most recently, Forner et al combined the data from 2 randomized controlled trials and there was no statistically significant variation in the results among patients who received debridement of necrotic bone followed by HBOT compared to those who only underwent debridement. ⁶⁰ The utility of HBOT in preventing MORN after dentoalveolar surgery has also been refuted via a prospective, randomized, multicenter trial in the United Kingdom. ⁶¹

HBOT also presents additional disadvantages. Potential contraindications to HBOT include claustrophobia, seizure disorders, pulmonary disease, and optic neuritis. ⁶² , ⁶³ Ocular, middle ear, pulmonary and neurologic complications are well‐documented. ⁶² , ⁶³ Lastly, HBOT is a time‐consuming and expensive treatment option which may lead to definitive management, ultimately resulting in progression to more advanced disease. ⁶⁴

In 2024, the American Society of Clinical Oncology published a multidisciplinary joint guideline with the Multinational Association of Supportive Care in Cancer‐International Society of Oral Oncology for managing ORN. Their consensus was that there is no conclusive evidence to support HBOT combined with surgery or as single modality. ⁶⁵ Based on these factors and the lack of current compelling evidence for the routine use of HBOT in managing MORN, its incorporation into contemporary evidence‐based treatment algorithms is questionable.

Described Surgical Options

Surgical techniques for reconstructing late‐stage full‐thickness MORN with osteocutaneous free flaps are well‐described and considered gold standard. ⁶⁵ Short of these morbid procedures, surgical options for treating partial thickness disease are limited. Examples of other previously described approaches include periosteal and fasciocutaneous free flaps. Given that the periosteum carries osteogenic and neoangiogenic capacity, its utility has been proposed in MORN. ⁶⁶ , ⁶⁷ , ⁶⁸ In 2019, Bettoni et al published their experience with 13 cases of periosteal free flaps harvested from various locations including inner femoral condylar, iliac crest, external brachial with humeral periosteum, and forearm with radial periosteum. ⁶⁶ Their cohort included Notani I and II stages with greater than 2 cm of bone exposure. This group reports arrest in 11 of 12 patients, over a mean follow‐up period of 89 months, with a 41.7% (n = 5) complication rate which included 3 flap failures necessitating redo and 2 patients with chronic fracture and fistulization. Although hospital stay for patients was not reported, individuals remained NPO for 5 to 7 days after surgical treatment.

In 2020, debridement followed by forearm fasciocutaneous free flap was described in 15 patients with MORN. ⁶⁹ Morlandt and colleagues report a 100% arrest rate over a mean follow‐up of 20.7 months. Patients selected for this treatment were limited to Notani I and II disease, with Stage III disease relegated to segmental resection and reconstruction. The authors described an 18.2% complication rate. The average length of stay was 6 days and patients were kept NPO for 4 to 5 days prior to initiating a liquid diet.

Anterolateral Thigh Fascia Lata (ALTFL) Rescue Flap

The fascia‐only variation of the ALTFL was first described by Koshima et al in 1989 for dural and abdominal reconstruction. ⁷⁰ Its use has since been expanded in the head and neck for various subsite reconstructions, including complex nasal reconstruction, floor of mouth/alveolar defects, anterior skull base/nasopharynx, and hard palate. ⁷¹ , ⁷² , ⁷³ , ⁷⁴ , ⁷⁵ Over the last decade, the authors have gained significant experience with this technique, termed the ALTFL rescue flap, and have published numerous case series highlighting its utility in ORN of various head and neck subsites. ² , ³ , ⁴ In 2011, the first author performed the first ALTFL to treat MORN. The surgery comprises a transoral marginal mandibulectomy with debridement of nonviable necrotic bone, sequestrectomy, dental extraction, or removal of dental hardware as needed, followed by imbricated coverage and obliteration of the defect with vascularized fascia lata. Microvascular anastomosis is achieved via minimal access to donor vessels through a 3‐cm incision at the facial notch, preauricular crease, or nasolabial fold and tunneled subcutaneously. ⁷⁶ An iliac crest bone graft (ICBG) may also be utilized to promote bone formation depending on residual bone height after debridement. Most recently, the authors published their 52‐case cohort of MORN, demonstrating successful arrest of disease, both clinically and radiographically, in 96.2% of cases over a mean follow‐up of 29.3 months. ⁵ Only 2 cases went on to require a fibular free flap reconstruction. Notani Stage III cases had no significant difference in clinical or radiographic progression risk compared to those with Stage I disease, with the caveat that application of this procedure was reserved for those individuals without full‐thickness disease. Morbidity of the rescue flap procedure was low, with a major complication rate of 1.9% and a mean hospital length of stay of 2.7 days. Patients resumed an oral diet immediately after surgery.

The ALTFL flap offers several advantages, making it an ideal candidate for managing ORN. It provides a robust blood supply, ensuring efficient oxygenation and nourishment to the transplanted tissue and recipient defect. Consequently, relieving the hypoxic environment promotes the healing process and theoretically reduces the risk of further ORN progression. Additionally, the provision of vascular supply allows for antibiotic delivery that is not achievable in the usual hypovascular milieu of ORN. Surgery also provides the opportunity to obtain deep bone cultures, enabling culture‐driven antibiotic therapy. ⁷⁷ The flap is thin and pliable, allowing easy manipulation to obliterate and contour a wide range of defects in size and location.

Perhaps the most crucial advantage of ALTFL lies in its low morbidity. ² , ³ , ⁷⁰ , ⁷¹ , ⁷² , ⁷³ , ⁷⁴ Of note, this minimal access technique does not preclude or impair future osteocutaneous reconstruction and spares its premature use. This de‐escalation of care is particularly significant as prior studies have not demonstrated substantial improvement in QOL measures in Notani Stage III after segmental resection and reconstruction. ¹⁵ , ⁷⁸ Compared to osteocutaneous reconstruction, the ALTFL has significantly lower postoperative complication rates (1.9% major). ⁵ In stark contrast, a recent single‐institution study of 76 patients who underwent traditional segmental resection and reconstruction underscored this by reporting an overall complication rate of 65% and perioperative mortality of 4%. ¹ Notably, 49% of patients required a return to the operating room, and 11% required a permanent tracheostomy.

Although innovation for disease‐modifying surgical management of MORN continues, the success of the ALTFL rescue flap has offered an assuring option to improve patient outcomes and reduce surgical risks. These findings have ignited a paradigm shift in MORN staging and management.

Implications for Practice

Proposed Staging

Currently, the most utilized ORN staging system is the Notani system. ²³ It is intuitive, efficient, and anatomically based, rendering it ideal for surgical planning. The IANC also serves as a reliable and readily encountered surgical landmark. Although the Notani staging system has its advantages, it primarily falls short in its ability to discriminate between various grades of severity in late‐stage disease. ⁴⁵ For example, a patient with Notani III may range from having an intact mandible with ORN depth marginally past the IANC, to having a fracture with concomitant orocutaneous fistula and widespread bone destruction. Despite the same stage classification, these 2 cases present distinctly divergent challenges in management. Furthermore, previously proposed staging systems have demonstrated substantial variability with respect to the classification of moderate‐staged disease. This has been acceptable to date, as treatment options for moderate disease have been inadequate and thus have often led to disease progression, forcing segmental resection. In other words, one only needed to distinguish between early and late‐stage disease to decide the appropriate treatment strategy. However, with the addition of the ALTFL rescue flap and similar nonosseous free flap techniques, a viable option now exists to treat a range of moderate to severe disease that preserves mandibular continuity. This innovation calls for a staging system that differentiates various phases of ORN progression to guide surgical decision‐making. Furthermore, there is a need for staging that reflects potential differences between presurgical assessment and ultimate extent of disease following intraoperative debridement.

The authors propose a system, adapted from Notani et al ²³ that both describes disease severity and directs management. The proposed distinguishing criteria are estimated based on clinical and radiographic presurgical assessment. However, definitive staging is not assigned until surgical debridement of nonviable bone occurs, and defect characteristics are finalized intraoperatively. The staging system is summarized in Table 2 and illustrated in Figure 1. Preoperative panorex images of patients classified by the proposed staging system are demonstrated in 2, 3, 4, 5, 6.

Table 2.

Contemporary Staging System

Stage I: Confined to DAU and area of exposed bone <2.5 cm

Stage II: Extends beyond the DAU but above the IANC or confined to DAU but area of exposed bone >2.5 cm

Stage III: Extends beyond the IANC but spares >1 cm height of lingual and buccal cortex

Stage IV: Extends beyond the IANC, and there remains ≤1 cm height of lingual and/or buccal cortex with otherwise intact mandible

Stage V: Full‐thickness involvement, presence of pathologic fracture

Abbreviations: DAU, dentoalveolar unit; IANC, inferior alveolar nerve canal.

Figure 1.

Illustration of staging system. Illustration credit: Ava Rose Fritz.

Figure 2.

Bilateral stage I preoperative panorex demonstrating osteoradionecrosis confined to the dentoalveolar unit surrounding teeth #18 and #31.

Figure 3.

Stage II preoperative panorex demonstrating osteoradionecrosis of the left mandibular body with extension past the dentoalveolar unit surrounding tooth #18 but sparing of the inferior alveolar nerve canal.

Figure 4.

Stage III preoperative panorex demonstrating osteoradionecrosis of the left mandibular body with involvement of the inferior alveolar nerve canal but sparing greater than 1 cm of mandibular cortex.

Figure 5.

Stage IV (“borderline”) preoperative panorex demonstrating osteoradionecrosis of the right mandible with an intact but less than 1‐cm height of the remaining lingual or buccal cortex.

Figure 6.

Stage V preoperative panorex demonstrating full‐thickness osteoradionecrosis of the left mandibular body.

Treatment Algorithm and Rationale

Preoperative assessment is estimated through clinical and imaging work‐up but finalized intraoperatively. The proposed treatment algorithm is depicted in Figure 7. Stage I disease is treated conservatively with a 3‐month course of PENTO. If the site appears objectively infected, a concurrent trial of oral antibiotics is pursued. In symptomatic patients or those with obvious necrotic bone, debridement is also offered; patients are counseled that this carries about risk of consequential upstaging as the defect size may increase and require subsequent escalation in treatment. Stage II patients are treated as Stage I plus definitive debridement with sequestrectomy and/or saucerization. All Stage II patients are also offered ALTFL rescue flap as PENTO is less effective in this group, and the size of the defect meets the minimal requirement for the feasibility of fascia flap inset. ⁴³ Stage II patients who defer rescue flap and opt for conservative management require close clinical and radiographic observation every 3 months. Stages I and II with progression on follow‐up are managed with rescue flaps. If patients are followed closely, delaying definitive management is acceptable given the large extent of viable mandible remaining and low likelihood of rapid progression to full‐thickness disease.

Figure 7.

Contemporary staging and treatment algorithm. Asterisk (*) denotes that other similar nonosseous free flap may be offered instead of rescue flap. Illustration credit: Mary Reagan. CT, computed tomography; DAU, dentoalveolar unit; IANC, inferior alveolar nerve canal; ICBG, iliac crest bone graft; ORN, osteoradionecrosis; PENTO, pentoxifylline and tocopherol.

Stage III and IV patients are treated with the rescue flap procedure, and those presenting clinically in Stage IV are classified as “borderline” rescue flap candidates. Patients who meet Stage IV criteria intraoperatively postdebridement (less than 1 cm inferior cortex remaining) are managed with ALTFL rescue flap and concurrent iliac crest cancellous bone grafting. Stage V disease is routinely treated with segmental extirpation and reconstruction in a single stage. All patients undergoing surgery are consented to procedures 1 stage above their presurgical classification, as postdebridement defect size can result in upstaging intraoperatively. This consideration is especially important in Stage III patients who may require ICBG and Stage IV borderline patients who may require an intraoperative pivot toward segmental resection.

Notably, this treatment algorithm does not recommend that the rescue flap replace osseous free flaps for full‐thickness disease but rather serve as a treatment option for partial‐thickness disease that has to date been treated with either HBOT or observation until further progression necessitating a radical resection.

All rescue flap patients are treated with culture‐directed antibiotics for 6 weeks postoperatively as based on the standard length of treatment for osteomyelitis. ⁷⁷ , ⁷⁹ Posttreatment, our patients undergo panoramic X‐ray (panorex) at 3 months, and both CT imaging and panorex at 6 and at 12 months. Thereafter, patients are followed with an annual panorex.

Of note, the authors' recommendations in this treatment algorithm are to utilize ALTFL perforator flaps due to minimal morbidity, short hospital stays, and high efficacy if these are within the armamentarium of the treating facility. However, experience and clinical evidence suggest that aggressive debridement and defect reconstruction with similar free vascularized tissue would likely confer similar benefits. Others' experience with radial forearm free flaps in the setting of Notani I and II disease supports this view. ⁸⁰

To the authors' knowledge, the ALTFL rescue flap is currently the only available low‐morbidity option that is definitively effective against Stages I to IV of MORN (Notani I to III excluding full‐thickness or fracture). ² , ³ , ⁴ , ⁵ Furthermore, the growing evidence against HBOT efficacy, along with its high cost and nonnegligible risk of complications, has resulted in the omission of its routine use in the proposed treatment algorithm. Studies have demonstrated that it is more cost‐effective to proceed with surgical resection and osteocutaneous reconstruction with a 7‐day postoperative hospital stay than with conservative and hyperbaric therapies. ⁸¹ An even greater economic advantage is suspected with the rescue flap technique as the hospital stay is, on average, only between 2 and 3 days. Rescue flap costs are estimated to be 26.2% of that incurred from a typical 40‐dive course of HBOT. ² Conservative management is primarily replaced by PENTO due to rising evidence to support its use in early‐stage ORN. Due to the aforementioned equivocal literature, potential risks, and high costs, we have not routinely added clodronate. Moreover, the cost of clodronate can be burdensome. Although exact costs are difficult to estimate, in 2004, a single 400 mg tablet of clodronate cost $1.92 Canadian dollars. ⁸² At the standard dose of 1600 mg daily, a 30‐day supply would cost about $230. This is significant given the long course of therapy required for ORN.

The algorithm foregoes the routine prescription of PENTO in Stages III to V due to previously discussed evidence of poor response in more advanced disease. Given the dramatically higher morbidity and increased complication rates of managing full‐thickness disease, and therefore a distinct advantage in intervention before disease progression, stepwise management (ie, medical followed by surgical) of advanced disease is not recommended. ⁸³

Borderline rescue flap candidates (Stage IV) warrant special consideration. The authors' experience with employing ICBG and managing these complex patients is growing, with uniformly positive outcomes thus far. ⁵ The rationale behind bone grafting in Stage IV is for improved mandibular stability, ⁸⁴ creation of bone stock for potential future dental rehabilitation, ⁸⁵ and osteoconductivity. ⁸⁶ The ALTFL rescue flap may have significant advantage in the setting of deeper defects that require complete coverage of bone grafts with minimal intraoral bulk as evidenced by our success in managing grafted Notani III defects (no difference between Notani I).

The employed treatment algorithm carries significant potential to benefit patients with moderate‐stage disease who are currently being offered standard care (segmental resection and fibula free flap reconstruction). For example, 1 institution reports that of 76 patients who underwent fibula free flap for MORN, only 49% and 38% had a fracture or fistula, respectively. ¹ Although patient‐level data is unavailable to discern the exact number of patients who had neither a fracture nor a fistula, it is safe to assume a significant number of these patients would have qualified for an ALTFL rescue flap.

Future Direction

Innovative treatment strategies that have emerged over the last decade, encompassing both medical and surgical, hold the potential to revolutionize the landscape of ORN treatment. These new tools may open numerous avenues for future investigations. For example, is there a role for including concentrated growth factor with iliac cancellous bone? Early evidence suggests that CGF promotes bone regeneration, increases the rate of successful take, and diminishes inflammation within the affected area. ⁸⁷ , ⁸⁸ , ⁸⁹ Similarly, although studies are not controlled, there is developing evidence that platelet‐rich fibrin may serve as a beneficial adjunct to surgical treatment of ORN. ⁹⁰ Furthermore, advances in regenerative medicine have ignited inquiry into the role of stem cell therapy in preventing or treating mandibular ORN. Specifically, mesenchymal stem cells facilitate tissue regeneration and angiogenesis, especially in tissue with compromised healing. As a result, these cells have shown promising preclinical results. ⁹¹ , ⁹² Moreover, as the search for improved biomaterials and scaffolds ensues, their implication in the reconstruction of mandibular ORN defects also continues to unfold, both in isolated synthetic materials and in combination with allograft and autografts. ⁹³ The era of personalized medicine may demand specimen‐driven therapy. For example, polymerase chain reaction testing has unveiled a broad spectrum of noncultivable microorganisms that may play a role in ORN pathogenesis and, as such, ultimately impact antibiotic selection in the future. ⁹⁴

Conclusion

MORN is a morbid disease entity that is difficult to treat. Although management options have plateaued for many years, the recent emergence of evidence to support innovative medical and surgical techniques warrants a re‐evaluation of modern staging and treatment. This narrative review suggests a new outlook focused on the omission of HBOT, the inclusion of PENTO for early‐stage therapy, and the utilization of the ALTFL rescue flap or similar vascularized free flaps to arrest disease progression. The result is a treatment protocol that reserves osteocutaneous free flap reconstruction as a last resort to minimize its inherent risks and morbidity.

Author Contributions

Michael A. Fritz, design, conduct, analysis, writing, critical revisions; Khashayar Arianpour, design, conduct, analysis, writing, figure design, critical revisions; Sara W. Liu, design, writing, critical revisions; Eric D. Lamarre, design, analysis, critical revisions; Dane J. Genther, design, analysis, critical revisions; Peter J. Ciolek, design, analysis, critical revisions; Patrick J. Byrne, design, analysis, critical revisions; Brandon L. Prendes, design, analysis, critical revisions.

Disclosures

Competing interests

None.

Funding source

None.