Microscopic Features of Maxillofacial Osteonecrosis
ŠThe Maxillofacial Center for Diagnostics & Research
Other Links NICO Clinical Page NICO Follow-up NICO Home Page Home Page		Authors Brief Review Introduction to the Problem Methods & Materials Microscopic Features Discussion Conclusions Reference Tables Photos
Marrow capillary is dilated upstream from a platelet clot (lower right).  Many red blood cells are stained very lightly ("ghost erythrocytes"), a sign of ischemic degeneration.

The Maxillofacial Center, 165 Scott Avenue, Suite 100, Morgantown, WV 26508 USA
Phone: 304-292-4429   Fax: 304-291-5149    Email: MFC@aol.com

Authors: Bouquot JE, Rabinovich S, Griep J, McMahon RE

Ischemic osteonecrosis is a devastating, degenerative condition of the human skeleton. It has a propensity to strike relatively young persons (average age at diagnosis: 35-50 years), the ability to produce a variety of disabling pains, the capacity to destroy large amounts of cancellous bone with minimal radiographic change, and a penchant for destroying multiple bones or multiple areas of the same bone (Figure 1). It has an often unrelenting progression and is now known to be quite a common phenomenon, affecting up to 30% of patients with certain medical conditions, such as lupus erythematosus, and now responsible for 25,000-90,000 total hip arthroplasties annually in the United States and Western Europe.

Ischemic osteonecrosis is not so much a specific disease as it is the logical result of any illness capable of significantly impairing intramedullary blood flow. The marrow microcirculation appears to be especially susceptible to thrombotic events and undiagnosed coagulation disorders have recently been shown to be present in 60-80% of osteonecrosis patients.^7,9,10

Diseased bone resulting from this physiologic problem is now referred to as ischemic osteonecrosis, but other terms have been applied (avascular necrosis, aseptic necrosis, bone infarction, aseptic osteomyelitis). Mild, chronic involvement, with more ischemic than infarctive changes, is called bone marrow edema syndrome but, again, other terms have been applied (transient ischemic osteoporosis, regional osteoporosis, migrating osteoporosis, regional hypertension, compartment disease, intraosseous engorgement-pain syndrome).^7,11-13 To add to this confusing lexicon, there are diagnostic terms based on specific etiologies (corticosteroid-induced osteonecrosis; radiation-induced osteonecrosis, dysbaric osteonecrosis) or involvement of specific bones (Table 1).

Regardless of the diagnostic terminology used, ischemic osteonecrosis is a disease with wide microscopic variation and considerable histopathologic subtly. Few pathologist have provided detailed histological descriptions and very few pathologists have extensive experience in its diagnosis. In the maxillofacial region, especially, there is minimal experience because this is a problem of the marrow spaces, hence, diseased bone is located outside the lamina dura and outside the bony and fibrous encapsulation surrounding apical pathoses and intraosseous neoplasms and cysts. It is not typically present in tissue samples submitted for biopsy of routine oral pathology cases. Because of this, it seemed important to present a review the varied microscopic features of ischemic osteonecrosis of the maxillofacial bones and to explain these features with what is understood of the pathophysiology of the disease.

All cases of osteomyelitis, osteitis and osteonecrosis of the maxillary and mandibular bones (n = 5,645) were retrieved from the archival files of the Latvala Inflammatory Bone Tissue Registry, located in the Maxillofacial Center for Diagnostics and Research, Morgantown, West Virginia. Tissue was received by two national biopsy services and three regional hospitals over a 25 year period. Tissue samples were microscopically reviewed by a single pathologist to confirm the diagnoses and to assess diagnostic parameters for each tissue sample. Microscopic criteria used for the diagnosis of ischemic osteonecrosis were those reported in the literature, especially those established by Arlet (Table 2), Ficat, Massiere, Alimi and Bouquot.^99-99 Representative tissue samples were submitted for consultation and diagnosis confirmation to internationally recognized experts in the field of inflammatory bone disease.

After microscopic review, a total of 4,132 samples of ischemic osteonecrosis from 2,023 patients were identified for review. Of the total, biopsy request forms for 1,333 patients (with 2,867 tissue samples) offered clinical information relative to facial pain. This group presented with the following pain syndromes prior to jaw biopsy:

• 895 with atypical facial neuralgia/pain
• 312 with nonspecific idiopathic facial pain or headache
• 156 with no history of facial pain
• 126 with trigeminal neuralgia

Contributing surgeons did not indicate the manner by which facial neuralgias were diagnosed, nor did they usually indicate whether or not those diagnoses were made by neurologists or neurosurgeons. A previous report of a smaller sample of NICO cases from the Latvala Registry indicated that approximately 76% of cases had been diagnosed by these medical specialists.¹⁵ The authors presume a similar proportion for this larger patient cohort.

The tissue from the pain-free patients appeared histopathologically identical to the lesions found in patients with pain. A lack of pain is not an uncommon feature of osteonecrosis in other bones, even when severe destruction of cancellous bone has occurred.^17-20

This is an analysis of retrospective cases and undoubtedly suffers from the usual sample-collection biases of all clinicopathologic reports using the databases of surgical pathology services.²⁶ The sample should not be assumed to represent any particular population.

Early fluid changes. Not all cases of osteonecrosis show severe necrotic changes. One of the earliest significant marrow alterations is simple distention or dilatation of capillaries, veins and sinusoids (Figure 2). These vessels may be engorged with blood (congestion), completely empty, or filled with an eosinophilic serous fluid such as is seen in lymphatic channels. One of the unique features of osteonecrosis is the release of this plasma or serous fluid between fat cells, often separating the cells (Figure 3). This plasmostasis or serous ooze is also seen in the marrow of cachectic or debilitated persons, especially those undergoing chemotherapy.  In the latter setting it may be called serous fat atrophy or gelatinous marrow.

Dilated vessels are a sign of localized intramedullary hypertension, as also is the more rare fibrosis and thickening of arterioles. This hypertension is capable of producing acute and chronic pain (compartment disease, intraosseous engorgement-pain syndrome) and most likely results from and accentuates poor outflow of blood, a situation not unlike trauma-related pulpal changes in which edema or increased hydrostatic pressures collapse apical veins to limit outflow from a tooth. Ischemic osteonecrosis is so uniquely and so consistently associated with poor venous outflow that intramedullary injection of saline is used as a diagnostic test for radiographically normal hip cases because it produces an immediate, severe increase in marrow pressures and local pain.⁹⁹ These events have been measured in the femoral head but can thus far only be inferred for jawbone cases because of the similarity in the microscopic appearance of affected marrow tissues from the two sites.

The other fluid change seen in early or mild involvement is small, multifocal interstitial or intramedullary hemorrhage (Figure 4). This sign has been shown by Ono⁹⁹ and others to be evidence of infarction or, more appropriately, microinfarction. Microinfarctions can be explained by blockage of small medullary vessels by microclots (Figure 5). Two unusual features of microinfarctions in mild cases of osteonecrosis are a lack of inflammatory cell responses to the hemorrhage, and the presence of degenerated, pale-staining degenerated or ghost erythrocytes admixed with other erythrocytes in the hemorrhagic fluid. Hemosiderin deposits may be seen, but are often not pronounced.

Of significance to the pathologist is the presence of aggregated fibrin and platelets within vessel lumina, perhaps completely blocking the vessel (Figure 6). Such thrombi are suggestive of hypofibrinolysis or thrombophilia. Intravascular fat globules and fat cells may also be found. The adipocytes are presumably pushed in because of the elevated intramedullary pressure, but for many years their presence led to the widespread acceptance of fat embolization as the primary cause of osteonecrosis.⁹⁹

Reticular fatty degeneration (ischemic myelofibrosis). Arlet and Ficat⁹⁹ recognized that chronic ischemia produced characteristic marrow changes, usually seen as a variation in the size of adjacent adipocytes and as a wispy fibrosis between adipocytes. Animal studies have proven them correct.⁹⁹ One should be cautious when evaluating size variations, since a variation as great as 3- or 4-fold between the size of adjacent fat cells is considered by many to be normal. In our labs we do not use the term fatty degeneration unless there is at least a 6-fold variation in size between fat cells in the same region of the marrow. Ruptured adipocyte cell walls and a fine granular adipocyte cytoplasm are features of necrosis, not simple degeneration.

The wispy intracellular fibrosis is known by a number of different terms, including fibrofatty marrow, fibrovascular marrow, reticular fatty degeneration, marrow fibrosis, eosinophilic fatty degeneration, and ischemic myelofibrosis. It is the main feature of Arlet Tissue Type I and of bone marrow edema and transient ischemic osteoporosis, the American terms used for Type I tissues (Figure 7).⁹⁹ More established areas of reticular fatty degeneration show more dense and avascular fibrous tissue with fewer remaining adipocytes (Figure 8).

There appears to be a poor understanding of the role of fibers in fatty marrow. Unlike fatty tissues outside the bones, fatty marrow is encased within rigid bony walls and requires no structural support. The only fibers normal to bone marrow are sparse reticulum fibers, which are used more for communication than support. These fibers are usually distributed around blood vessels and adjacent to bone and are typically present only in trace amounts.^99Strauchin Marrow fibrosis is increased in a variety of pathologic states, most of which are related to ischemia.^99Strauchin In the jaws, the ischemic state may be manifested by a fibrous scar tissue filling an old extraction site (residual socket, fibrosed socket), often with little remodeling of the surrounding lamina dura, but in other bones these fibrous islands, or intraosseous fibrous scars, are considered to be secondary to poorly healed infarctions in a chronically ischemic environment. It is important to recognize that intramedullary fibrous scars are distinct and separate from true bone scars, which are composed of newly formed bone, not fibrous tissue.⁹⁹

The susceptibility of bone and marrow cells to ischemia is variable and usually manifested by diminished function and/or numbers.⁹⁹ Hematopoietic cells are the most susceptible, followed in decreasing order of susceptibility by osteoclasts, osteoblasts, osteocytes, adipocytes, and fibroblasts. Under ischemic conditions fibroblasts are created from the multipotential reticular cells. It has been shown repeatedly that marrow fat cells can become fibroblasts and vise versa.

Medullary Infarction. The major damage in ischemic osteonecrosis arises from infarction of cancellous bone. Early recognition of this fact once led orthopedic surgeons to refer to the disease as "coronary disease of the hip."²⁴ While infarction of bone may be a single massive event, most frequently it occurs as a series of small, localized microinfarctions separated by normal marrow and variable amounts of time. Some jawbone cases present with nothing more than dilated marrow sinusoids and focal hemorrhage representing areas of ruptured veins/sinusoids and capillaries. Occasionally fibrin aggregates can be seen plugging the affected vessel in patients with diagnosed or undiagnosed coagulopathies (Figure 9).

Depending on the timing of the biopsy procedure, liquefactive necrosis of the fatty marrow may be seen (Figure 10), perhaps with the appearance of large circular spaces or oil cysts representing coalesced liquid fat from a cluster of adipocytes which died during the infarctive event (Figure 11). Necrosis of hematopoietic marrow is more easily identified, being nothing more than a focal area of hypocellularity with pale, eosin-staining hematopoietic cells, usually admixed with extravasated erythrocytes (Figure 12). Oil cysts may also be seen in hematopoietic marrow. These early events are seldom associated with an inflammatory cell response and are considered by Ono, Arlet, Ficat and others as the minimal diagnostic criteria for the necrotic phase of the disease.^99-99

Fat Necrosis. As already mentioned, ischemic osteonecrosis is a disease with causes cell and tissue death predominantly by microinfarction, perhaps evidenced by nothing more than focal hemorrhage or extravasation of erythrocytes between fat cells. Microinfarctions are often multiple and separated by histologically normal marrow.

The result of infarction in fatty marrow is simple: fat dies. When death occurs to adipocytes supplied by a common feeder vessel, liquefactive necrosis produces oil cysts comprised of coalesced liquid fat (Figures 13). Oil cysts are seen as circular or oval, clear, punched-out spaces with no peripheral condensation or inflammatory cell infiltration. As has been mentioned, these are often large enough to be seen by the naked eye at surgery, usually floating atop hemorrhagic ooze and often mimicking the fat globules seen in chicken soup. The tissue often takes on a foamy or bubbly appearance as the fat is broken down into small, variably-sized globules (Figures 14). Small numbers of chronic inflammatory cells are characteristic of this foamy fat necrosis, as is a wispy fibrosis between fat cells. When histiocytes predominate, the overall appearance is similar to that of a lipogranuloma. Arlet called this Tissue Type II.

Liquid Necrosis. Depending of whether liquefactive or coagulative necrosis dominates, there will be more or less identifiable tissue in an intramedullary biopsy sample. Some lesions contain nothing more than a pool of "dirty motor oil" in the floor of a hollow intramedullary cavitation. This liquid necrosis can be processed in the laboratory by pouring the formalin through a filter and smearing the filtered coagulum onto a microscopic slide prior to routine staining. The resulting slide usually shows numerous fat globules and oil cysts within a background film of fibrin and hemorrhage (Figure 15). There may or may not be fragments of degenerated or necrotic bone and marrow. Very seldom is normal marrow present in any of the tissue fragments.

Desiccated Necrosis. Other cavitations are completely dry or desiccated, i.e. represent "dry rot" of the marrow. They appear as hollow, "air-filled" spaces with walls either as smooth as white marble or as ragged and discolored as wood scorched by fire. In such cases of desiccated necrosis only a small amount of tissue can be scraped from the walls. The presence of the cavitation itself must, therefore, be taken into account when making the pathologic diagnosis, as it must be in making the diagnosis of traumatic (simple) bone cyst. Figure shows the wall of a maxillary cavitation from an autopsy case of maxillofacial osteonecrosis. The smooth, rounded concavities and convexities of the wall are characteristic and may be seen along the edges of curetted tissue fragments in routine biopsy samples.

Intramedullary Cavitations. These cavitations, whether dry or wet, are among the most unique features of intramedullary ischemia and infarction. They may extend from cortex to cortex and may be several centimeters in greatest diameter. It is not unusual for a lesion to extend from the retromolar region to the bicuspid or cuspid region. This phenomenon is so unique that Phemister,⁹⁹ one of the century's premier bone researchers, coined the term "cavitation" to describe it. The father of modern dentistry, G. V. Black,⁹⁹ used the term "cavity" to describe apparent jawbone osteonecrosis several decades prior to Phemister's use of the term. Cavitations are now known to be a part of several ischemia-related disorders, including:^99-99

• osteoporosis (usually small cavitations)
• ischemic osteonecrosis (including NICO)
• simple (traumatic) bone cyst
• osteoarthritis (subchondral bone cyst)
• florid osseous dysplasia with multiple traumatic bone cysts

Only about a third of maxillofacial osteonecrosis cases have hollow cavitations and so these are not a constant feature of ischemic osteonecrosis of the jaws.

Calcific Fat Necrosis. When the microenvironment is amenable, the necrosed adipocyte cell walls remain intact and a fine granular cytoplasm develops in many cells. This combination of coagulation and liquefactive necrosis may remain for a long period of time, but often calcium and other precipitates produce a coalescent mass of hematoxylin-positive material (Figures 15). This is similar to the processes of saponification and adipocere (lipocere) production seen in rancid soaps and fats degenerating under appropriate conditions. A better microscopic term is calcific fat necrosis, seen in Arlet Tissue Types II and III.

Calcific fat necrosis may have a granular or an amorphous appearance. It usually contains several clear spaces representing fat cells which have died without undergoing salt precipitation. It is not unusual for this material to occur as rounded, smudged aggregations, often containing slivers or shards of delaminated bone. Such aggregates are sometimes referred to as NICO globules (Figures 16). This type of calcific debris, called calcific and proteinaceous necrotic marrow detritus can easily be mistaken for the viable "bone dust" which results from the surgical removal of the biopsy sample by rotary instruments. Contributing surgeons should, therefore, be instructed to remove intramedullary tissues only with hand curetting and to place the tissue sample immediately in formalin in order to reduce the difficulty of mistaking bone dust for necrotic bone shards or calcific necrotic debris. When the pathologist is in doubt, a simple and effective means of separating the two types of bone is readily at hand. Necrotic debris is not birefringent under polarizing light, while bone, even small fragments of bone dust, still demonstrates polarizing collagen fibers. This is equally true for the nonviable bone of osteomyelitis sequestra. As with oil cysts, there is only a slight local inflammatory cell response to calcific and proteinaceous necrotic marrow detritus.

Ischemic Bone Changes. Many of the changes seen in long-standing ischemic bone are similar, some say identical, to those seen in osteoporosis.  Involved trabeculae are thin and often are widely spaces, with minimal or no osteoblastic or osteoclastic activity, excess numbers of cement or reversal lines (a number more than 2 is considered a sign of osteoporosis), and prominent or thickened reversal lines or osteoid rimming. Most trabeculae are comprised of a somewhat immature lamellar bone and may show small, spindle-shaped open spaces or microcracks along the cement lines. Complete or partial delamination along multiple cement lines may be seen and is sometimes referred to as an exploded trabecula.  If there has been attempted, albeit poor healing/remodeling, the trabeculae may be composed of woven bone, again with little residual osteoblastic activity. Cortical and lamina dura bone often shows a fish scale or bundle-bone appearance from the attachment of ligaments to bone.  While this seems more pronounced is ischemically damaged bone it may merely be unnoticed in other oral pathology tissue samples because lamina dura is seldom included. The cortex is usually much less involved than the trabeculae, and more mature trabeculae show more of these changes than immature ones.

Ironically, mild ischemia may produce osteoporotic bone or excessively thickened bony trabeculae, i.e. osteosclerosis (Figure 10).  In long bones osteosclerosis occurs in approximately 40% of cases but it is radiographically unusual in maxillofacial osteonecrosis cases except at the microscopic level, i.e. scattered trabeculae are thickened when viewed microscopically but the change is not pronounced enough to be noticed radiographically.  On the other hand, intramedullary viable bone scars are not unusual in the jaws and may be examples of ischemic osteosclerosis, although they are typically diagnosed as focal chronic sclerosing osteomyelitis or condensing osteitis.

Clustered Loss of Osteocytes. Loss and pyknosis of osteocytes are the classical diagnostic signs of ischemic osteonecrosis, in some minds being an absolute requirement for an osteonecrosis diagnosis. Arlet,⁹⁹ Ficat,⁹⁹ Ono,⁹⁹ Hungerford⁹⁹ and others have, however, clearly shown this not to be the case with the more mild examples. Arlet Tissue Type I, in fact, requires no bone changes for its diagnosis. Beginning with Arlet Tissue Type II the missing/pyknotic osteocytes are seen in localized or clustered areas of alveolar bone, with 50% or more cells missing or pyknotic in a cluster. These clusters presumably represent end-stage feeder sites from the same vascular branch. Fast laboratory decalcification can, of course, produce artifactually missing or pyknotic osteocytes, but these are seen as scattered individual cell changes without a clustering effect. We strongly recommend using slow decalcification solutions, such as formic acid or 5% nitric acid, in order to better assess this important diagnostic parameter. With proper handling, very few osteocytes are artifactually lost (Figure 2).

One must use caution when interpreting osteocyte loss, especially in bone from older patients (60 years or more) and especially when the bone has been rapidly decalcified. Bone from older jaws will naturally have more missing osteocytes than bone from younger jaws. Enlow considers this to result from increased bone ischemia in older jaws, but it is not clustered and is not necessarily ischemic osteonecrosis.

One of the more unique features of ischemic bone is a separation or microcracking along cement lines, previously mentioned.⁹⁹ This can be seen in bone without major loss of osteocytes. It may occasionally also be seen in bone affected by osteomyelitis, but usually sequestra in the latter disease have a very characteristic surface resorption with few if any microcracks. In severe cases of maxillofacial osteonecrosis large portions of bone literally delaminate in a fashion similar to untreated plywood kept underwater for a period of time. Arlet and Ficat⁹⁹ used the dramatic term exploded trabecula to describe this microscopic appearance, and it is not unusual for a partially healed area of old infarction to contain residual shards of trabeculae which presumably delaminated during the original infarctive event (Figure 3). Occasional remnants of bone are seen with the delaminated shards remaining in place parallel to one another, an appearance which has led some to refer to it as a log pile.

Poor Bone Healing. In some affected bone, an unusual form of poor bone healing (remodeling?) occurs which has histopathologic similarities to condensing osteitis, except that the trabeculae are less mature and the normal fibrous background stroma is replaced by metaplastic adipocytes with dilated capillaries (Figure 16). Even though the bone is immature, few osteoblasts or osteoclasts are seen, presumably because of the local ischemia. Intramedullary conditions may not be severe enough to force adipocyte metaplasia into fibroblasts, but why the original primitive fibrous stroma disappears is a question yet to be answered.

Osteocyte Death. Occasional cases of maxillofacial osteonecrosis show relatively normal marrow, but have large areas of missing and pyknotic osteocytes. This is more commonly seen in other bones, but when the jaws are involved the cortex is typically less affected than trabeculae. Osteoblasts and osteoclasts are either minimal in numbers or are missing entirely.

Fibrin And Platelet Aggregates. Recent research has discovered that more than 80% of ischemic osteonecrosis patients have one or more heretofore undiagnosed clotting disorders, usually  inherited, resulting in a prothrombotic tendency or hypercoagulable state, i.e. hypofibrinolysis or thrombophilia. This may explain the frequent coating of NICO tissue samples by a thin layer of fibrin, and may also explain the frequent presence of fibrin or platelet aggregates within abnormal marrow or plugging the vessels of the marrow (Figures 17 and 18). Fibrin aggregates can also be found with frequency in the thin film surrounding tissue fragments. Often the tissue samples appear to be literally embedded within a background of fibrin and hemorrhage. In other types of diseases, biopsied tissue fragments are seldom coated by fibrin and aggregates are rarely seen. It is present in only 0.4% (3/700) inflammatory periapical lesions, such as granulomas and cysts. It is in almost 17% of cases of maxillofacial osteonecrosis with enough tissue to make a proper assessment.  The presence of fibrin coating and aggregation is, therefore, suspicious for by certainly not pathognomonic for a hypercoagulable state or osteonecrosis. It is merely a red flag, something to increase suspicion of clotting dysfunction.

The microscopic features of ischemic osteonecrosis and especially its more nascent or less "severe" forms, bone marrow edema and regional ischemic osteoporosis are constantly being refined and are now known to many oral pathologists.  Visible changes are often not striking and individual are usually not diagnostic, but taken together they will typically allow an appropriate diagnosis to be made.  The key to diagnosis is awareness and experience.  Without awareness the pathologist will simply describe the marrow changes under such generic terms as fibrosis, or will perhaps take it one step further and provide a logical alternative diagnosis such as chronic fibrosing osteomyelitis.  This was certainly our own experience in the early days of our understanding of IO.  Even today there are comorbid cases in which it is impossible to separate ischemic from inflammatory changes.  These are usually given a dual diagnosis of IO and osteomyelitis.

1.Ono K, ed. Symposium: recent advances in avascular necrosis. Clin Orthopaedics Related Res 1992; 277:2-138.

2. Steinberg ME, Steinberg DR. Osteonecrosis. In: Kelly WN, Harris ED Jr, Ruddy S, Sledge CB. Textbook of rheumatology, 4th ed. Philadelphia: W.B. Saunders; 1993:1628-1650.

3. Mazieres B. Osteonecrosis. In: Klippel JH, Dieppe PA, eds. Rheumatology. St. Louis: Mosby; 1994:41.1-41.8.

4. Sweet DE, Madewell JE. Osteonecrosis: pathogenesis. In: Resnick D (ed). Diagnosis of bone and joint disorders, 3rd ed. Philadelphia: W. B. Saunders, 1995: 3445-3494.

5. Bullough, PG. Orthopaedic pathology. 3rd edition. Baltimore; Mosby-Wolfe, 1997.

6. Jones JP Jr. Osteonecrosis. In: Koopman WJ (ed). Arthritis and allied conditions; a textbook of rheumatology, 13^th edition. Baltimore: Williams & Wilkins, 1997:1923-1942

7. Urbaniak JR, Jones, JP Jr (eds). Osteonecrosis -- etiology, diagnosis, and treatment. American Academy of Orthopaedic Surgeons; Chicago, Illinois, 1997.

8. Mankin HJ. Nontraumatic necrosis of bone (osteonecrosis). N Eng J Med 1992; 326:1473-1479.

9. Van Veldhuizen PJ, Neff J, Murphey MD, et al. Decreased fibrinolytic potential in patients with idiopathic avascular necrosis and transient osteoporosis of the hip. Am J Hematol 1993; 44:243-248.

10. Glueck CJ, Freiberg R, Glueck HI, et al. Hypofibrinolysis; a common, major cause of osteonecrosis. Am J Hematol 1994; 45:156-168.

11. Arlet J, Mazieres B, eds. Bone circulation and bone necrosis. Heidelberg: Springer-Verlag, 1990.

12. Guerra JJ, Steinberg ME. Distinguishing transient osteoporosis from avascular necrosis of the hip. J Bone Jnt Surg 1995; 77:616-624.

Brigden ML. The hypercoagulable state; who, how, and when to test and treat. Postgrad Med 1997; 101:249-262.

2. Tabernero MD, Tomas JF, Alberca I, et al. Incidence and clinical characteristics of hereditary disorders associated with venous thrombosis. Am J Hematol 1991; 36:249-254.

3. Svensson PJ, Dahlback B. Resistence to activated protein C as a basis for venous thrombosis. N Engl J Med 1994; 330:517-522.

4. Griffin JH, Evatt B, Wideman C, et al. Anticoagulant protein C pathway defective in majority of thrombophilic patients. Blood 1993; 82:1989-1993.

5. Arnoldi CC. Intraosseous engorgement-pain syndromes. The pathomechanism of pain. In::253-259.

6. Simkin PA, Downey DJ, Lanzer WL. Hydraulic resistance: a measure of outflow resistance in osteonecrosis. In: Arlet J, Mazieres B, eds. Bone circulation and bone necrosis. Heidelberg: Springer-Verlag, 1990: 273-276.

7. inject saline--pain.

8. Glueck CJ, McMahon RE, Bouquot JE, et al. Thrombophilia, hypofibrinolysis and osteonecrosis of the jaws. Oral Surg Oral Med Oral Pathol 1996; 81:557-566.

9. Gruppo R, Glueck CJ, McMahon RE, et al. The pathophysiology of osteonecrosis of the jaw: anticardiolipin antibodies, thrombophilia, and hypofibrinolysis. J Lab Clin Med 1996; 127:481-488.

10. Ficat PR, Arlet J. Ischemia and necrosis of bone. Williams & Wilkins, Baltimore, 1980.

11. Jones JP Jr. Osteonecrosis. In: McCarty D, ed. Arthritis and allied conditions: a textbook of rheumatology, 11th ed. Philadelphia: Lea & Febiger, 1989: 1545-1562.

12. Arlet J, Mazieres B, eds. Bone circulation and bone necrosis. Heidelberg: Springer-Verlag, 1990.

13.Hungerford DS. Diagnosis and treatment of ischemic necrosis of the femoral head. In: McC Evarts C (editor). Surgery of the musculoskeletal system, 2nd edition. New York: Churchill Livingstone, 1990: vol 3: 2757-2794.

14.

16.

17.

18.

19. Ratner EJ, Person P, Kleinman DJ, et al. Jawbone cavities and trigeminal and atypical facial neuralgias. Oral Surg Oral Med Oral Pathol 1979; 48:3-20.

20. Roberts AM, Person P. Etiology and treatment of idiopathic trigeminal and atypical facial neuralgias. Oral Surg Oral Med Oral Pathol 1979; 48:298-308.

21. Shaber EP, Krol AJ. Trigeminal neuralgia -- a new treatment concept. Oral Surg Oral Med Oral Pathol 1980; 49:286-293.

22. Roberts AM, Person P, Chandran NB, et al. Further observations on dental parameters of trigeminal and atypical facial neuralgias. Oral Surg Oral Med Oral Pathol 1984; 58: 121-129.

23. Ratner EJ, Langer B, Evins ML. Alveolar cavitational osteopathosis -- manifestations of an infectious process and its implication in the causation of chronic pain. J Periodontol 1986; 57:593-603.

24. Bouquot JE, Roberts AM, Person, P, et al. The histopathology of neuralgia-inducing cavitational osteonecrosis (NICO). J Dent Res 1989; 68:952.

25. Segall RO, Del Rio CE. Cavitational bone defect: a diagnostic challenge. J Endo 1991; 17:396-400.

26. Shankland WE. Osteocavitational lesions (Ratner bone cavities): frequently misdiagnosed as trigeminal neuralgia-a case report. J Craniomand Pract 1993; 11:232-234.

27. Bouquot J, Roberts A. NICO (neuralgia-inducing cavitational osteonecrosis): radiographic appearance of the "invisible" osteomyelitis. Oral Surg Oral Med Oral Pathol 1992; 74:600.

20. Ingle JI, Bakland LK. Endodontics, 4th ed. Baltimore: Williams & Wilkins, 1994; 581-582.

21. Byron MA. A clinicopathologic review of 2278 NICO cases (neuralgia-inducing cavitational osteonecrosis). Master's Thesis (Endodontics). Morgantown, WV: West Virginia University, 1994.

22. Langlais RP, Langland OE, Nortje CJ. Diagnostic imaging of the jaws. Baltimore: Williams & Wilkins; 1994:393-395.

23. Neville B, Damm D, Allen C, Bouquot JE. Oral & maxillofacial pathology. Philadelphia: W. B. Saunders, 1995; 631-632.

24. coronary disease of hip

25. Glueck--hip paper

Glueck CJ, McMahon FR, Bouquot J, et al. Preliminary pilot study of the treatment of thrombophilia and hypofibrinolysis and the amelioration of the pain of osteonecrosis of the jaws. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1998; 85:64-73.

Strauchen, JA. Diagnostic histopathology of the bone marrow. New York: Oxford University Press, 1996.

Black GV. A work on special dental pathology, 2nd edition. Chicago: Medico-Dental Publ Co, 1915.

Chuong R, Piper MA. Avascular necrosis of the mandibular condyle: pathogenesis and concepts of management. Oral Surg Oral Med Oral Pathol 1993; 75:428-432.

Chuong R, Piper MA, Boland TJ. Osteonecrosis of the mandibular condyle, pathophysiology and core decompression. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1995; 79:539-545.

Eversole LR. Facial pain. In: Cohen S, Burns RC. Pathways of the pulp, 6th ed. St. Louis: Mosby; 1994; 57.

Bouquot JE, Roberts AM, Person P, et al. NICO (neuralgia-inducing cavitational osteonecrosis): osteomyelitis in 224 jawbone samples from patients with facial neuralgias. Oral Surg Oral Med Oral Pathol 1992; 73:307-319.

McMahon RE, Adams W, Spolnik K. Diagnostic anesthesia for referred trigeminal pain, Parts I & II. Compendium Cont Educ Dent 1992; 11:870-881, 980-997.

McMahon RE, Griep J, Marfurt CP, et al. Local anesthetic effects in the presence of chronic osteomyelitis/necrosis of the mandible: implications for localizing the etiologic sites of referred trigeminal pain. J Craniomand Pract 1995; 13:212-226.

Brown RS, Hinderstein B, Reynolds DC, et al. Using anesthetic localization to diagnose oral and dental pain. J Amer Dent Assoc 1995; 126: 633-641.

Bouquot JE, Christian J. Long-term effects of jawbone curettage on the pain of facial neuralgia; treatment results in neuralgia-inducing cavitational osteonecrosis. J Oral Maxillofac Surg 1995; 53:387-397.

Bouquot JE, Roberts AM, Person, P, et al. The histopathology of neuralgia-inducing cavitational osteonecrosis (NICO). J Dent Res 1989; 68:952.

Bouquot J, McMahon R. Ischemic osteonecrosis of the jaws in 2,023 patients with facial pain. J Oral Pathol Med 1996; 25:271.

Bouquot JE, Roberts AM, Person P, et al. NICO (neuralgia-inducing cavitational osteonecrosis): osteomyelitis in 224 jawbone samples from patients with facial neuralgias. Oral Surg Oral Med Oral Pathol 1992; 73:307-319.

Enlow DH. Osteocyte necrosis in normal bone. J Dent Res 1966; 45:213.

Froberg PK, Braunstein EM, Buckwalter KA. Osteonecrosis, transient osteoporosis, and transient bone marrow edema: current concepts. Radiol Clin North Am 1996; 34:273-291.

Brown CR. Pain management: NICO. Pract Perio Aesth Dent 1996; 8:916.

Nicol K, Klingman J, Holt J, et al. Ultrasonic gum/jaw bone detection instrument. Thesis. Socorro, New Mexico; New Mexico Institute of Mining & Technology, 1996.

Haley B. Characterization and identification of chemical toxicants isolated from cavitational material and infected root canaled teeth: in situ testing of teeth for toxicity and infection. Proceedings of annual meeting, American Biological Dentistry; Carmel, California; March, 1997.

Frost HM. In-vivo osteocyte death. J Bone Joint Surg 1960; 42A:138-143.

Glueck CJ, McMahon RE, Bouquot JE, et al. Thrombophilia, hypofibrinolysis and osteonecrosis of the jaws. Oral Surg Oral Med Oral Pathol 1996; 81:557-566.

Graff-Radford SB, Simmons M, Fox L, et al. Are bony cavities exclusively associated with atypical facial pain and trigeminal neuralgia? Proceedings, Annual Meeting, Western USA Pain Society, 1988.

Gruppo R, Glueck CJ, McMahon RE, et al. The pathophysiology of osteonecrosis of the jaw: anticardiolipin antibodies, thrombophilia, and hypofibrinolysis. J Lab Clin Med 1996; 127:481-488.

Haley B. Characterization and identification of chemical toxicants isolated from cavitational material and infected root canalled teeth: in situ testing of teeth for toxicity and infection. Proceedings of annual meeting, American Biological Dentistry; Carmel, California; March, 1997.

Jones JP Jr. Intravascular coagulation and osteonecrosis. Clin Orthop Related Res 1992; 277:41-53.

Neville B, Damm D, Allen C, Bouquot JE. Oral & maxillofacial pathology. Philadelphia: W. B. Saunders, 1995; 631-632.

Oikarinen VJ. The inferior alveolar artery: a study based on gross anatomy and arteriography, supplemented by observations on age changes. Suomen Hammaslaakariseuran Toimituksia, Finska Tandlakarsallskapets Forhandlingar 1965; 61:11-131.

Phemister, DB. Repair of bone in the presence of aseptic necrosis resulting from fractures, transplantations, and vascular obstruction. J Bone Joint Surg 1930; 12:769-778.

Van Veldhuizen PJ, Neff J, Murphey MD, et al. Decreased fibrinolytic potential in patients with idiopathic avascular necrosis and transient osteoporosis of the hip. Am J Hematol 1993; 44: 243-248.

Wannfors K, Gazelius B. Blood flow in jawbones affected by chronic osteomyelitis. Br J Oral Maxillofac Surg 1991; 29:147-153..

Weksler BB. Altered hemostasis in inflammation. In: Gallin JI, Goldstein IM, Snyderman R, eds. Inflammation: basic principles and clinical correlations. New York: Raven Press, 1988;43-557.

Yokochi C, Rohen JW, Weinreb, EL. Photographic anatomy of the human body. 3rd ed. New York; Igaku-Shoin, 1989.

Bennington JL. Surgical pathology of bone marrow--core biopsy diagnosis. Philadelphia: W. B. Saunders, 1985.

Rywlin AM. Histopathology of the bone marrow. Boston: Little, Brown & Co., 1996:153-190.

Strauchen, JA. Diagnostic histopathology of the bone marrow. New York: Oxford University Press, 1996.

Bullough, PG. Orthopaedic pathology. 3^rd edition. Baltimore: Mosby-Wolfe, 1997