The vast majority of tissues in the human body will not return to normal after injury, and can only be repaired with granulation and fibrous scar tissue, severe burns and open incisions can only grow scars. However, bone tissue, unlike other tissues, has a strong ability to repair and regenerate. Bones can be joined together when broken, and bones can grow back from bone defects. Now take a fresh fracture as an example to illustrate the process of bone growth and repair: after the fracture, bleeding occurs at the fracture site, forming a hematoma, then a granulation, then a cartilage scab, and finally a hard bone scab, or bone tissue. This bone tissue has a disturbed arrangement of trabeculae, and the bone structure is completely restored to normal by shaping it over a longer period of time. It is an accepted fact that the bone repair process is exactly the same regardless of the cause of necrosis, whether it is a fracture, a drilling, or osteomyelitis-induced osteonecrosis. Because of the special anatomy of the femoral head, what happens after femoral head necrosis is usually ineffective bone repair, and it took a long time for humans to recognize the natural repair process of ineffective bone repair in femoral head necrosis. In 1920, Phemister suggested that this increase in density was due to an external increase in density caused by a decrease in the density of the surrounding bone. In 1958 Bonfiglio and Bardenstein found new bone attached to the surface of necrotic trabeculae in the sclerotic area of femoral head necrosis. In 1965, Bohr et al. conducted a micro-radiological and histological study on 20 resected femoral heads with hip replacements and found that there was no difference between the calcification of the bone matrix in the sclerotic area of the femoral head and normal subjects, and that the necrotic trabeculae were widened by the attachment of new bone on the surface, and that the increase in bone density was proportional to the width of the trabeculae. They highlighted that sclerosis was both a result of necrosis and a clear sign of vascular regeneration and repair in the necrotic area. 1976 Kenzora experimented with 250 rabbits to study the repair pattern after femoral head necrosis. This is a famous experiment, and Kenzora believes that the diagnosis of osteocyte survival by light microscopy is not reliable, because osteocytes can remain intact for a considerable period of time after death, and therefore the determination of osteocyte physiology should be more sensitive and reliable than histomorphology. The most sensitive and reliable indicator of a cell’s survival is its ability to synthesize ribonucleic acid (RNA), and loss of synthesis indicates that the cell is dead. Since the radioactive isotope H3 CVcytidine is the precursor of RNA synthesis, autoradiography can be performed with H3 CVcytidine, and if bone cells cannot be labeled with this isotope, they are dead. Using this method, it was demonstrated that most cells lost their synthesis capacity at 2 hours of ischemia and all cells in the femoral head except cartilage were dead at 12-24 hours. In the conclusion of the article, Kenzora states that the inability of cells to take up the isotope deuterated cytosine was used to confirm the necrosis of the femoral head in adult rabbits. Proliferating capillaries and undifferentiated mesenchymal cells in the bone marrow of the living bone near the osteotomy rapidly filled the marrow spaces of the dead femoral head. The mesenchymal cells gradually took on osteoblastic characteristics as they proliferated toward the surface of the dead bone trabeculae. Finally, they differentiate into functional osteoblasts that cover the surface of the necrotic trabeculae. New bone forms on the surface of the dead bone and expands to fill the trabecular space, resulting in increased bone content per unit volume and increased radiographic density of the femoral head. The core of dead bone in the center of the trabeculae is later resorbed and replaced by living bone. The new trabeculae are thicker than the original ones and are living plate-like bone. The biological response of the subchondral dead dense bone occurs later because of the location far from the starting point of repair. Unlike the rough trabeculae, the primary response here is bone resorption rather than bone formation. The pace of new bone formation cannot keep up with the pace of bone resorption resulting in subchondral bone loss. Capillary penetration and tissue resorption progresses all the way to the cartilage, causing a proliferative response of chondrocytes and changes within the cartilage matrix similar to those seen in osteoarthritis. In addition, destructive synovial opacities form, growing on the cartilage surface and destroying the articular cartilage. The mismatch of the joint, the loss of cartilage is similar to the degenerative osteoarthritis of the femoral head causing similar changes in the acetabular cartilage with subsequent complete destruction of the joint. In simple understanding, different anatomical parts of the rabbit femoral head necrosis repair process have different manifestations: 1 cancellous bone, after necrosis repair new bone is formed on the surface of and between the necrotic trabeculae, which increases bone density per unit volume, and later the necrotic trabeculae are gradually revived. 2 osseous articular surfaces, that is, the subchondral bone dense mass gradually resorbs and disappears. 3 cartilage and joints are gradually destroyed. The bone repair ability of human is far inferior to that of rabbits, and the repair of necrotic cancellous bone in many people cannot be completed for life, and in a few people it takes 10 years to complete, with obvious femoral head deformation and osteoarthritis. Even if a person can reach the repair capacity of a rabbit, he or she cannot avoid joint destruction. It is necessary to exceed the repair capacity of rabbits in order to cure femoral head necrosis. The appearance of hypodense areas in the necrotic femoral head in humans has been explained by atrophy of the trabeculae or resorption during the repair process, but recent studies by Plenk H Jr et al. of the Vienna Institute of Tissue and Embryology on necrotic resected femoral heads have shown that there are three types of repair in human femoral head necrosis: 1. limited repair, in which a sclerotic rim is formed next to the vascularized zone. 2. destructive repair, which results in significant bone resorption and fragmentation of the femoral head. 3. reconstructive repair, which reduces the extent of necrosis and slows or stops the progression of the disease for a certain period of time. The results of this study can better explain the sclerotic edge, low-density areas and high-density areas that appear on plain films after necrosis. In the past, many people mistook the bone resorption produced by destructive bone repair for osteoporosis, which delayed the diagnosis of necrosis. In another article, he also pointed out that all three types of repair are ineffective and that reconstructive bone repair also inevitably decreases the mechanical strength of the femoral head, causing it to collapse. It is generally accepted that femoral head necrosis is incurable once the lesion has progressed to the point where changes appear on plain radiographs. The three challenges in the repair of human femoral head necrosis are the collapse of the femoral head as the repair process progresses, the resorption of the osseous articular surface (subchondral bone dense mass), and the destruction of the articular cartilage. Why does femoral head necrosis collapse? Although the bone cells have died after femoral head necrosis, the inorganic salts of the bone matrix remain unchanged and basically retain the original mechanical strength unchanged. With the development of the repair process, the mechanical strength and elastic modulus of the femoral head gradually decreases, and the mechanical strength is only about half of the original one. In addition, microfractures will appear in the femoral head of normal people, and these microfractures will slowly heal, but once necrosis occurs, these microfractures cannot heal, and time often decreases the loading capacity, and eventually subchondral fractures will inevitably appear, and the femoral head will collapse. In other words, the faster the ineffective repair speed, the faster the collapse speed, so the collapse speed of young people, faster than the elderly, for extensive necrosis, sometimes the collapse will be accelerated after the use of blood-activating drugs. The process of femoral head necrosis repair is summarized in the table below. At present, the natural course of femoral head necrosis and the speed of collapse development are not well understood by human beings, and it is usually believed that the course of femoral head necrosis develops more rapidly. Femoral head collapse usually occurs within two years of the onset of pain, and 50% of patients have to undergo surgery within 3 years after the diagnosis is established. We have found that many patients develop mild collapse within 4-6 months of the onset of symptoms.