My watch list  

Osteonecrosis of the jaw

Osteonecrosis of the jaws
Classification & external resources
Osteonecrosis of the jaw of the upper left jaw in a patient diagnosed with chronic venous insufficiency
ICD-10 M87.1
DiseasesDB 1174
eMedicine derm/816 

Osteonecrosis of the jaws (ONj) is a severe bone disease that affects the jaws, including the maxilla and the mandible. Jaw bone (osteo-) damage and death (-necrosis) occurs as a result of reduced local blood supply (ischaemia). The condition is thus included in the general category of ischaemic or avascular osteonecrosis (literally "dead bone from poor blood flow.").

Various forms of ONj have been described over the last 160 years, and a number of causes have been suggested in the literature. In recent years, an increased incidence of ONj has been associated with the use of high dosages of bisphosphonates, required by some cancer treatment regimens, especially when the patient undergoes subsequent dental procedures. The possible risk from lower oral doses of bisphosphonates, taken by patients to prevent or treat osteoporosis, remains uncertain.[1]

Various treatment options have been explored, however severe cases of ONj still require surgical removal of the affected bone.[2] Careful anamnesis (patient history) and assessement of pre-existing systemic problems and possible sites of dental infection are required to help prevent the condition, especially if bisphosphonate therapy is considered.[1]

Additional recommended knowledge



Histopathological alterations

Persons with ONj may have either necrotic bone or bone marrow that has been slowly strangulated or nutrient-starved. Bone with chronically poor blood flow develops either a fibrous marrow since fibres can more easily live in nutrient starved areas, a greasy, dead fatty marrow (wet rot), a very dry, sometimes leathery marrow (dry rot), or a completely hollow marrow space (osteocavitation), also typical of ONj. The blood flow impairment occurs following a bone infarct, a blood clot forming inside the smaller blood vessels of cancellous bone tissue.

Under ischaemic conditions numerous pathological changes in the bone marrow and trabeculae of oral cancellous bone have been documented. Microscopically, areas of "apparent fatty degeneration and/or necrosis, often with pooled fat from destroyed adipose cells (oil cysts) and with marrow fibrosis (reticular fatty degeneration)" are seen. These changes are present even if "most bony trabeculae appear at first glance viable, mature and otherwise normal, but closer inspection demonstrates focal loss of osteocytes and variable micro cracking (splitting along natural cleavage planes). The microscopic features are similar to those of ischaemic or aseptic osteonecrosis of long bones, corticosteroid-induced osteonecrosis, and the osteomyelitis of caisson (deep-sea diver’s) disease".[3]

In the cancellous portion of femoral head it is not uncommon to find trabeculae with apparently intact osteocytes which seem to be "alive" but are no longer synthetizing collagen. This appears to be consistent with the findings in alveolar cancellous bone.[4]

Osteonecrosis can affect any bone, but the hips, knees and jaws are most often involved. Pain can often be severe, especially if teeth and/or a branch of the trigeminal nerve is involved, but many patients do not experience pain, at least in the earlier stages. When severe facial pain is involved, the term NICO, for Neuralgia-Inducing Cavitational Osteonecrosis, is frequently used.

ONj, even in its mild or minor forms, creates a marrow environment that is conducive to bacterial growth. Since many individuals have low-grade infections of the teeth and gums, this probably is one of the major mechanisms by which the marrow blood flow problem can worsen; any local infection / inflammation will cause increased pressures and clotting in the area involved. No other bones have this mechanism as a major risk factor for osteonecrosis. A wide variety of bacteria have been cultured from ONj lesions. Typically, they are the same microorganisms as those found in periodontitis or devitalized teeth. However, according to special staining of biopsied tissues, bacterial elements are rarely found in large numbers. So while ONj is not primarily an infection, many cases have a secondary, very low-level of bacterial infection and chronic non-suppurative osteomyelitis can be associated with ONj. Fungal infections in the involved bone do not seem to be a problem, but viral infections have not been studied. Some viruses, such as the smallpox virus (no longer existent in the wild) can produce osteonecrosis.

The effects of persistent ischaemia on bone cells

Cortical bone is well vascularized by the surrounding soft tissues thus less susceptible to ischaemic damage. Cancellous bone, with its mesh like structure and spaces filled with marrow tissue is more susceptible to damage by bone infarcts, leading to anoxia and premature cell apoptosis.[5][6][7][8] The mean life-span of osteocytes has been estimated to be 15 years in cancellous bone,[9] and 25 years in cortical bone.[10] while the average lifespan of human osteoclasts is about 2 to 6 weeks and the average lifespan of osteoblasts is approximately 3 months.[11] In healthy bone these cells are constantly replaced by differentiation of bone marrow mesenchymal stem cells (MSC).[12] However in both non-traumatic osteonecrosis and alcohol-induced osteonecrosis of the femoral head, a decrease in the differentiation ability of mesenchymal stem into bone cells has been demonstrated,[13][14] and altered osteoblastic function plays a role in ON of the femoral head.[15] If these results are extrapolated to ONj the altered differentiation potential of bone marrow mesenchymal stem cells (MSC) combined with the altered osteoblastic activity and premature death of existing bone cells would explain the failed attempts at repair seen in ischaemic-damaged cancellous bone tissue in ONj.

The rapidity with which premature cell death can occur depends on the cell type and the degree and duration of the anoxia. haematopoietic cells , in bone marrow, are sensitive to anoxia and are the first to die after reduction or removal of the blood supply. In anoxic conditions they usually die within 12 hours. Experimental evidence suggests that bone cells composed of osteocytes, osteoclasts, and osteoblasts die within 12-48 hours, and marrow fat cells die within 120 hours.[16] The death of bone does not alter its radiographic opacity nor it’s mineral density. Necrotic bone does not undergo resorption; therefore, it appears relatively more opaque.

Attempts at repair of ischaemic-damaged bone will usual occur in 2 phases. First, when dead bone abuts live marrow, capillaries and undifferentiated mesenchymal cells grow into the dead marrow spaces, while macrophages degrade dead cellular and fat debris. Second, mesenchymal cells differentiate into osteoblasts or fibroblasts. Under favorable conditions, layers of new bone form on the surface of dead spongy trabeculae. If sufficiently thickened, these layers may increase the radiopacity of the bone; therefore, the first radiographic evidence of previous osteonecrosis may be patchy sclerosis resulting from repair. Under unfavorable conditions repeated attempts at repair in ischaemic conditions can be seen histologically and are characterized by extensive delamination or microcracking along cement lines as well as the formation of excessive cement lines.[17] Ultimate failure of repair mechanisms due to persistent and repeated ischaemic events is manifested as trabecular fractures that occur in the dead bone under functional load. Later followed by cracks and fissures leading to structural collapse of the area involved (osteocavitation).[16]

Other contributing factors

Other factors such as toxicants can adversely impact bone cells. Infections, chronic or acute, can affect blood flow by inducing platelet activation and aggregation, contributing to a localized state of excess coagulability (hypercoagulability) that may contribute to clot formation (thrombosis), a known cause of bone infarct and ischaemia. Exogenous estrogens, also called hormonal disruptors, have also been linked with an increased tendency to clot (thrombophilia) and impaired bone healing.[18]

Heavy metals such as lead and cadmium have been implicated in osteoporosis. Cadmium and lead also promotes the synthesis of plasminogen activator inhibitor-1 (PAI-1) which is the major inhibitor of fibrinolysis ( the mechanism by which the body breaks down clots ) and shown to be a cause of hypofibrinolysis.[19] Persistent blot clots can lead to congestive blood flow (hyperemia) in bone marrow, impaired blood flow and ischaemia in bone tissue resulting in lack of oxygen (hypoxia), bone cell damage and eventual cell death (apoptosis). Of significance is the fact that the average concentration of cadmium in human bones in the 20th century has increased to about 10 times above the pre-industrial level.[20]

Ethanol both from exogenous and endogenous sources and, its more toxic metabolite, acetaldehyde, have also been implicated in both osteoporosis and osteonecrosis. Acetaldehyde, a highly toxic metabolite of ethanol, can play a role in hypoxia and inhibit the osteoblastogenic potential of the bone marrow.[21] Ethanol has been shown to alter the epithelial barrier through ethanol oxidation into acetaldehyde by the colonic microflora and downstream mast cell activation. Such alterations that remain for longer periods could result in excessive endotoxin passage into the vascular network.[22] Intracolonic acetaldehyde may also be an important determinant of the blood acetaldehyde level and a possible hepatotoxin.[23] High serum antibody titers against acetaldehyde-protein adducts have been found not only in alcoholics but also in patients with nonalcoholic liver disease, suggesting a contribution of acetaldehyde derived from sources other than exogenous ethanol.[24] In a study on rats the role of intestinal bacterial overgrowth on the production and metabolism of ethanol, rats with a jejunal self-filling diverticulum (blind-loop) were compared to controls with a self-emptying diverticulum. Both endogenous ethanol and acetaldehyde were found in the blind-loop contents. Intragastric administration of sucrose produced a marked increase in acetaldehyde and acetate in the portal venous blood, with only a modest elevation of ethanol. It was concluded that the resulting high concentrations of acetaldehyde, both in the intestinal lumen and the portal blood, may have deleterious effects on the gastrointestinal(GI) mucosa and the liver.[25] Another experimental in-vitro study showed the potential of certain bacteria representing normal human colonic flora to produce acetaldehyde under various atmospheric conditions that may prevail in different parts of the GI tract. This bacterial adaptation may be an essential feature of the bacteriocolonic pathway to produce toxic and carcinogenic acetaldehyde from either endogenous or exogenous ethanol.[26] Many species of gut bacteria, yeast and fungal organisms such as Candida albicans found in the human GI tract and involved in gut dysbiosis, an imbalance in the microbial flora, have been shown to significantly increase blood ethanol levels, post-mortem, in individuals who had not consumed any alcohol before death.[27][28]

The effects of chronic gut dybiosis and long term exposure to low levels of endogenous acetaldehyde on bone tissue and hepatic function is not yet fully understood. However Cordts et al suggested in 2001 that gut dysbiosis (as indicated by stool yeast) and hepatic detoxification challenge pathway exhaustion may lead to subclinical, systemic inflammation and chronic venous insufficiency (CVI). CVI is a pathological condition caused either by the congenital absence of or damage to venous valves in the superficial and communicating systems. Venous incompetence due to thrombi and formation of thrombi favoured by the Virchow triad (venous stasis, hypercoagulability, endothelial trauma) also can cause CVI.[29]

Bisphosphonates may alter the disease process

In the past few years, thousands of cases of ONj in patients on bisphosphonate therapy have been diagnosed usually following lack of healing after a dental extraction but also in cases of spontaneous exposure of the cortical bone tissue through the gingiva and mucosa.[30][31]

The recent increase of such cases has been linked with a major emphasis on the therapeutic use of bisphosphonates for osteoporosis, especially since hormone replacement therapy has been shown to increase the risk of breast cancer, clots and cardiovascular disease in women following the 2003 findings of the U.S. Women’s Health Initiative study.[32] Two classes of bisphosphonates are presently prescribed:

  • Non-nitrogen containing bisphosphonates such as etidronate (Didronel®, Procter & Gamble Pharmaceuticals)
  • Nitrogen containing such as alendronate (Fosamax®, Merck), pamidronate (Aredia®, Novartis), zoledronate (Zometa®, Novartis), risedronate (Actonel®, Procter & Gamble) and ibandronate (Boniva®, Roche Laboratories).

The nitrogen containing bisphosphonates are the most potent inhibitors and no case of ONj associated with etidronate has been reported yet. The main pharmacological action of bisphosphonates is inhibition of the osteoclast driven bone resorption. This is achieved by shortening osteoclast lifespan via apoptosis and by inhibiting osteoclast activity and recruitment on the bone surface (61). When a bisphosphonate binds to bone mineral, osteoclast resorb both bone and the bound bisphosphonate. During bone formation, if any, bisphosphonate remaining on the surface of the bone is covered and remains there until future osteoclastic bone resorption at the site. This explains why inhibition of bone resorption continues long after bisphosphonate treatment has been discontinued.[33]

This form of therapy has been shown to prevent loss of bone mineral density (BMD) as a result of a reduction in bone turnover. However bone health is a lot more than BMD.

In healthy bone tissue there is a homeostasis between bone resorption and bone apposition. Diseased or damaged bone is resorbed through the osteoclasts mediated process while osteoblasts form new bone to replace it, thus maintaining healthy bone density. A process commonly called remodelling.

However osteoporosis is essentially the result of a lack of new bone formation in combination with bone resorption in reactive hyperemia, related to various etiological and contributing factors, and bisphosphonates do not address these factors at all.

An individual who is already having problems with osteoporosis/ osteonecrosis of the jaws due to the effects of these etiological factors will be more susceptible to the adverse effects of bisphosphonates. In theory, by suppressing osteoclastic activity and bone resorption, any ischaemic-damaged bone will be left in situ instead of being resorbed. The damaged bone will not be repaired either if the factors already inhibiting osteoblastic activity are still present. Therefore the amount of osteonecrotic tissue should be expected to increase until it reaches a level when any trauma or insult to this necrotic bone will result in extremely poor healing, exposed necrotic bone to the oral environment, development of pain, and increased risks of microbial infection, as effectively seen in bisphosphonates associated cases of ONj.

In a systematic review of cases of bisphosphonates associated ONj up to 2006, it was concluded that the mandible is more commonly affected than the maxilla (2:1 ratio), and 60% of cases are preceded by a dental surgical procedure. According to Woo, Hellstein and Kalmar, oversuppression of bone turnover is probably the primary mechanism for the development of this form of ONj, although there may be contributing co-morbid factors (as discussed elsewhere in this article). It is recommended that all sites of potential jaw infection should be eliminated before bisphosphonate therapy is initiated in these patients to reduce the necessity of subsequent dentoalveolar surgery. The degree of risk for osteonecrosis in patients taking oral bisphosphonates, such as alendronate (Fosamax®), for osteoporosis is uncertain and warrants careful monitoring.[1]

History in dental medicine

ONj is not a new disease, around 1850 various forms of "chemical osteomyelitis" resulting from environmental pollutants, such as lead and the white phosphorus used in early (non-safety) matches (Phossy jaw), as well as from popular medications containing mercury, arsenic or bismuth, were reported in the literature.[34][35][36][37][38][39][40] This disease apparently did not often occur in individuals with good gingival health, and usually targeted the mandible first.[35]It was associated with localized or generalized deep ache or pain, often of multiple jawbone sites. The teeth often appeared sound and suppuration was not present. Even so, the dentist often began extracting one tooth after another in the region of pain, often with temporary relief but usually to no real effect.[36]

Today a growing body of scientific evidence indicate that this disease process, in the cancellous bone and bone marrow, is caused by bone infarcts mediated by a range of local and systemic factors. Bone infarcts as well as damage to the deeper portion of the cancellous bone is an insidious process. It is certainly not visible clinically and routine imaging techniques such as radiographs are not effective for that sort of damage. "An important and often incompletely understood principle of radiography is the amount of bone destruction that goes undetected by routine x-rays procedures; this has been demonstrated by numerous investigators. Destruction confined to the cancellous portion of the bone cannot be detected radiographically, ad radiolucencies appear only when there is internal or external erosion or destruction of the bone cortex."[41] In fact no radiographic findings are specific for bone infarction / osteonecrosis. A variety of pathologies may mimic bone infarction, including stress fractures, infections, inflammations, and metabolic and neoplastic processes. The limitations apply to all imaging modalities, including plain radiography, radionuclide studies, CT scans, and magnetic resonance imaging (MRI). Through-transmission alveolar ultrasound, based on quantitative ultrasound (QUS) in combination with panoramic dental radiography (orthopantomography) is helpful in assessing changes in jawbone density.[42][43] When practitioners have an up to date understanding of the disease process and a good anamnesis is combined with detailed clinical findings and course of events, the diagnosis, with the help of various imaging modality, can be achieved earlier, in most patients.

In the modern dental profession, it is only recently when severe cases associated with bisphosphonates came to light, that the issue of ONj has been brought to the attention of a majority of dentists. At present, the focus is mostly on bisphosphonates associated cases, and is sometimes referred to colloquially as "phossy jaw", a similar, earlier occupational disease.[44][45] However, the pharmaceutical manufacturers of bisphosphonates drugs such as Merck and Novartis have stated that ONj in patients on this class of drug, can be related to a pre-existing condition, coagulopathy, anemia, infection, use of corticosteroids, alcoholism and other conditions already known to be associated with ONj in absence of bisphosphonate therapy. The implication is that bisphosphonates may not be the initiating cause of ONj and that other pre-existing or concurrent systemic and/or local dental factors are involved.[46][47]

Since ONj has been diagnosed in many patients who did not take bisphosphonates, it is thus logical to assume that bisphosphonates are not the only factor in ONj. While the oversuppression of bone turnover seems to play a major role in aggravating the disease process, other factors can and do initiate the pathophysiological mechanisms responsible for ONj. In non-bisphosphonate cases of ONj, it is mainly the cancellous portion of the bone and it’s marrow content that are involved in the disease process. The first stage is an oedema of the bone marrow initiated by a bone infarct, which is itself modulated by numerous etiological factors, leading to myelofibrosis as a result of hypoxia and gradual loss of mineral bone density characteristic of ischaemic osteoporosis. Further deterioration can be triggered by additional bone infarcts leading to anoxia and a localized areas of osteonecrosis within the osteoporotic cancellous bone. Secondary events such as dental infection, injection of local anaesthetics with vasoconstrictors, such as epinephrine, and trauma can add further complications to the disease process and chronic non-pus forming bone infection osteomyelitis can also be associated with ONj. [48][49][50]

However, in patients on bisphosphonates, the cortical bone is also frequently involved as well. Spontaneous exposure of necrotic bone tissue through the oral soft tissues or following non-healing bone exposure after routine dental surgery, characteristics of this form of ONj, may be the result of late diagnosis of a disease process that has been masked by the oversuppression of osteoclastic activity, allowing pre-existing etiological factors to further aggravate bone damage.


The treatment should be tailored to the individual patient according to the etiological factors involved and the severity of the disease process. With oral osteoporosis the emphasis should be on good nutrient absorption and metabolic wastes elimination through a healthy gastro-intestinal function, effective hepatic metabolism of toxicants such as exogenous estrogens, endogenous acetaldehyde and heavy metals, a balanced diet, healthy lifestyle, assessment of factors related to potential coagulopathies, and treatment of periodontal diseases and other oral and dental infections.

In cases of advanced oral ischaemic osteoporosis and/or ONj that are not bisphosphonates related, clinical evidence has shown that surgically removing the damaged marrow, usually by curettage and decortication, will eliminate the problem (and the pain) in 74% of patients with jaw involvement.[2] Repeat surgeries, usually smaller procedures than the first, may be required, and almost a third of jawbone patients will need surgery in one or more other parts of the jaws because the disease so frequently present multiple lesions, i.e. multiple sites in the same or similar bones, with normal marrow in between. In the hip, at least half of all patients will get the disease in the opposite hip over time; this pattern occurs in the jaws as well. Recently, it has been found that some osteonecrosis patients respond to anticoagulation therapies alone. The earlier the diagnosis the better the prognosis. Research is ongoing on other non-surgical therapeutic modalities that could alone or in combination with surgery further improve the prognosis and reduce the morbidity of ONj. A greater emphasis on minimizing or correcting known etiological factors is necessary while further research is conducted on chronic ischaemic bone diseases such as oral osteoporosis and ONj.

In patients with bisphosphonates-associated ONj, the response to surgical treatment is usually poor.[51] Conservative debridement of necrotic bone, pain control, infection management, use of antimicrobial oral rinses, and withdrawal of bisphosphonates are preferable to aggressive surgical measures for treating this form of ONj.[52] Although an effective treatment for bisphosphonate-associated bone lesions has not yet been established,[53] and this is unlikely to occur until this form of ONj is better understood, there as been clinical reports of some improvement after 6 months or more of complete cessation of bisphosphonate therapy.[54]

This article is licensed under the GNU Free Documentation License. It uses material from the Wikipedia article "Osteonecrosis_of_the_jaw". A list of authors is available in Wikipedia.
Your browser is not current. Microsoft Internet Explorer 6.0 does not support some functions on Chemie.DE