Aseptic Necrosis of Bone

The idea that part of a person may die while he is still living was first verbalized by Hippocrates. In the section of the Hippocratic corpus “On Fractures” it is written that if a foot is bandaged improperly “there is risk of necrosis of the heel bone; and if there is necrosis the malady may last the patient’s whole life.”lg The Greek word translated as necrosis is more directly translated as “sphacelate,” which is an archaic term for gangrene. The word “necrosis” did not appear in English until 1665, when Needham 37 translated a word in a phrase from a Latin treatise by Thomas Willis. The word was defined as “an inward mortification.” Hunter,3g in the mid-eighteenth century, understood that there was a vital spirit present in living tissues and absent in dead tissues. He recognized that bone died when separated from its milieu.39 Hunter outlined three criteria for the recognition of dead bone: (1) it would not repair; (2) it contained no animal matter; and (3) it would not stain with madder.26 In the latter years of the eighteenth century there were numerous treatises on necrosis of bone, such as the Descriptio Thesauri Ossium Morbosorum Hoviani by Andreae Bonn in 1783,” and Weidmann’s DeNecrosi Ossium in 1793.47 While these authors described dead bone, the clinical sit-

uations were clearly those of acute or chronic osteomyelitis. In his treatise of 1794, A Practical Essay on a Certain Disease of the Bones Termed Necrosis, James described the processes of sequestration (Fig. 1) and how dead bone serves as a scaffolding for new repair bone so that an entire segment of bone may be reconstituted.
It was not until 1845 that John Goodsir17 was able to show that the separation and repair of dead bone was accomplished by cells. With the introduction of Lister's achromatic microscope in 1835,26 there was a revolution in the study of tissues which could then be seen much more clearly. Goodsir was the first to apply this instrument to the study of bone. 26 In 1837, Benjamin Bell7 subdivided necrosis into three different types: superficial, internal, and progressive. While this subdivision is not currently relevant, it was a step toward understanding that bone may die in different ways.
In 1860, Pasteur's discovery of bacteria in abscesses was necessary to develop the idea of aseptic necrosis. Only with the knowledge that bacteria caused the suppuration often associated with necrosis did the idea of a nonbacterial or aseptic necrosis conceptualize. In 1870, Paget3' reported a case of necrosis without the "usually attendent suppuration and other signs of destructive inflammation." Paget was describing a loose body that originated from the femur of a young girl and called the process "quiet necrosis." Also interested in loose bodies was the prominent German surgeon, Franz KOnig.*' In 1888, he described the entity osteochondritis dissecans-loose bodies in joints derived from what he thought was a dissecting, low grade granulation tissue reaction. Konig reported two cases of osteochondritis dissecans of the hip. After his cogent arguments regarding why these patients did not have degenerative arthritis or tuberculosis, the modern reader can clearly recognize that Konig was describing aseptic necrosis of the hip, the first description of this condition.
The study of bone grafts led to important milestones in the history of bone necrosis. The surgery of bone tumors has been a fascination among orthopedic surgeons. A central problem has always been how to replace a bony defect after removing a tumor. To solve this problem, the study of bone grafts evolved. The great research efforts by Ollier, in the 1 8 5 0~,~~ were thwarted by suppuration. However, after Lister's theory of antisepsis, in 1866, MacEwen could continue these experiments.26 One important question that occupied investigators was what happened to the bone graft after transplantation. Ollier26 had said that the grafts remained alive. However, by 1893, Barth' had shown that the graft dies. The histologic features of dead bone grafts awakened the German pathologist, George Axhausen',' to the presence of dead bone in diseased states unassociated with gross suppuration. Axhausen,' who had also studied the histologic factors of bone grafts (Fig. 2), was an important figure in the history of bone necrosis. He was the first to use the term "aseptic necrosis," and, in 1 907,2 described the process of repair by which seams of new bone were laid on dead bone. Axhausen had been strongly influenced by certain medical advancements which occurred in late nineteenth century Germany. In the 187Os, C~h n h e i m~~ had studied the process of embolization, developed the concept of the anemic infarct and, with one of his students (Carl Weigert), defined necrosis at the cellular level. In 1865, Langer3' had shown that bones contained end-arteries. In view of these discoveries and with the ability to recognize dead bone in histopathologic tissue preparations, Axhausen4 postulated that the presence of dead bone in tissue was a manifestation of an anemic infarct of bone. Axhausen first applied his concept of the anemic bone infarct in his study of arthritis deformans, an archaic term for primary and secondary osteoarthritis. He felt that bone ischemia and infarcts were a primary cause of arthritis deformans and that this was the reason that this arthritis occurred in aging patients who have atherosclerosis.
Another medical advancement that influenced Axhausen was the development of the X-ray in 1895. Some diseases were discovered or defined on the basis of their radiographic appearance. Subtle distinctions were made that could not be made on clinical examination alone. For example, in 1909 Legg3' described an "obscure affection of the hip joint" of children. This condition was also described independently in 19 10 by Perthes4 and by Calvt." Also, Kohler" described a condition of the tarsus in 1908, Kienbock*' described a disease of the lunate in 1910, and FreibergI5 described a disease of the metatarsal head in 19 14. These conditions were, at first, only radiologic entities.
The other major contribution of Axhausen was his recognition that the pathology of these particular radiologic lesions was that of bone necro~is.~ He was able to link seemingly different radiologic entities into the broad concept of the anemic bone infarct. Axhausen felt that this was caused by an embolization of low grade bacterial colonies. Many of his cases cultured some organisms, that in retrospect, must have been contaminants. Included in Axhausen's concept of the anemic bone infarct was the condition that is currently recognized as idiopathic femoral head necrosis. This condition was first described by Schmorl in 1924.44 In 1926, FreundI6 wrote a comprehensive article on idiopathic femoral head necrosis that included a detailed pathologic analysis.
The next important figure in the history of bone necrosis was Dallas P h e m i~t e r~' .~~ who, as many American surgeon-pathologists, studied in Europe where he was influenced by Axhausen's work. During his many years in Chicago, first at Rush Medical College, then at the University of Chicago, his major work was related to the many facets of bone necrosis and grafting.
Phemister's great contribution was his correlations of gross and microscopic pathology with clinical and radiologic patterns. In one of his first important articles,'" he established the important concept funda- mental to bone pathology-pathologic features must correlate with X-ray findings.
Phemister also broadened the understanding of bone necrosis. He studied changes in the femoral head after fracture of the femoral neck and changes in patients who had caisson disease. Caissons were first used in 1851 to build the bridge over the Medway river in Rochester, England." Many workers who came up out of caissons became ill. In 187 1, Paul Bert8 explained that effervesced nitrogen bubbles caused tissue ischemia. In 1888, T~y n a m~~ reported a case of a man who had a bone lesion associated with decompression illness, but clinically, the man appears to have had an osteomyelitis. In 1913, Peter Bassoe6 tabulated 161 cases of compressed air sickness, including 11 men who had chronic joint pain. One year earlier, Borstein and Plate" published radiographs of compressed air workers, the first published radiographs of aseptic necrosis of bone in adults (Fig. 3). Kahlstrom, Phemister and Burton" demonstrated that these bone lesions were infarctions.
Phemister not only broadened Axhausen's concept of the ischemic bone infarct, he also thought specifically about each case, recognizing that there may be definable causes or specific associations. For example, currently, the association of bone necrosis with lupus, gout, alcoholism, renal dialysis, etc., is recognized.
Another contribution of Phemister was his development of a surgical procedure to treat femoral head necrosis, i.e., drilling and tibia1 bone grafting. 42 The treatment was based on the principal pathophysiology of the initial insult and the repair process. It was founded on the surgical maxim of Hunter," which states that the surgeon should try to assist natural healing processes rather than heal the patient himself.
Dr. Henry Jaffe also contributed to the study of bone infarction. As early as 1934, Jaffe and P~m e r a n z~~ described infarcts in the bones of extremities amputated for ischemia. Currently, Jaffe's'' classic analysis of the pathology of bone infarcts remains one of the most detailed and comprehensive. Fremont Chandler,14 who was a crosstown acquaintance of Phemister, was interested in femoral head necrosis as related to the vasculature of the hip, particularly the ligamentum teres. In 1935, Chandler recognized the distinctiveness of femoral head necrosis and how, often no underlying or associated disease was found. He postulated that the infarctions must be due to specific vascular occlusions and likened the condition to coronary artery disease. Because of his interest, idiopathic femoral head necrosis has been occasionally referred to as "Chandler's disease." In 1962, Mankin and B r~w e r~~ could collect only 27 cases of idiopathic femoral head This enabled clinicians to diagnose bone necrosis early when treatment might be most effective. Also, this work implied the concept that certain people are at risk to develop bone necrosis and that these might be benefited by diagnostic screening.
The application of radionucleotide scanning to the management and experimental study of bone necrosis has been a recent milestone in our understanding of this disease. In 1969, CameronI3 reported his studies on the use of strontium-85 scintimetry in the management of nontraumatic bone necrosis. He found that bone scan changes may precede radiologic changes by as much as 18 months. More recently, Gregg and Walder" have shown experimentally that a bone scan can become positive three weeks after a vascular insult. On the basis of such studies, it is possible to diagnose bone necrosis at an early stage when treatment may be most effective. Bone scintigraphy currently is a routine tool in the management and diagnosis of bone necrosis at many orthopedic centers. Further advancements in the understanding of bone necrosis must develop from a multidisciplinary approach, using bone scintigraphy and hemodynamic studies in addition to histopathologic analyses.
Also, complications of bone infarction other than joint disease may occur. Malignant fibrous h i s t i o c y t~m a s~~ and osteogenic sarcomas35 have occasionally been shown to be associated with bone infarcts. These tumors presumably arise in the chronic granulation tissue around infarcts.
The cause for most cases of aseptic necrosis is unknown. Although there have been many theories, e.g., fat e m b~l i z a t i o n ,~~ hypercoagulability,' increased intraosseous pressure,*' and fat cell swelling,46 no theory is incontrovertible.
Reviewing the historic perspective of bone necrosis, as with any disease, is both encouraging and humbling. It is encouraging because it reminds one that research is at the brink of the unknown, as was Franz Konig; the history is humbling because it reminds one that current expertise and knowledge are based on the hard work of predecessors. Truth is not necessarily a pet theory. Most great medical advancements have resulted from cooperation among investigators and idea sharing.

SUMMARY
The history of aseptic bone necrosis includes important contributions by Hunter, Russell, Goodsir, Paget, Konig, Axhausen, Phemister, and Chandler. Only after Pasteur discovered bacteria in abscesses could a nonbacterial or aseptic necrosis be conceptualized. Techniques and information gained from the study of bone grafts led to important steps in the recognition of aseptic necrosis. The use of the X-ray and careful pathologic correlation enabled a wide group of radiographically distinct lesions to be recognized as bone necroses. Although many identifiable diseases can underlie aseptic necrosis, the cause of most cases is unknown.