Rapidly destructive arthrosis of the hip an uncommon subset of osteoarthritis and was first described by Forestier in 1957  and subsequently labelled atrophic osteoarthritis, rapidly progressive osteoarthritis, destructive osteoarthritis, and Postel's osteoarthritis . Lequesne defined it as a greater than 2 mm/year rate of joint space narrowing, i.e. loss of more than 50% of the joint space within 1 year .
Over the next three decades, subsequent series and reports emphasized the clinical, radiographic and pathologic findings of this unusual condition. The series of cases in our region with rapid severe progressive hip joint destruction within few months drew our attention, which we think need to be prioritised with regard to surgical intervention and are unsuitable for un-reviewed placement on long waiting lists. This is mainly due to the intraoperative technical difficulty due to significant acetabular bone loss; increased blood loss during surgery coupled with increased operating times and need for special implants all making joint reconstruction challenging in these patients. This is associated with compromised outcome following surgery and researchers have indicated that overall results and survivorship are adversely affected by any acetabular defect, and this is certainly the main technical problem with rapidly progressive cases . There is therefore clearly a need for early diagnosis of this rapidly destructive hip OA.
Initially, radiographs show either normal anatomy or mild osteoarthritic changes. Follow-up radiographs in a few months of the onset of symptoms, demonstrate destruction of the femoral head and acetabulum with sclerosis, subchondral cysts, and minimal or no osteophytes. RDHD may radiographically be confused with other disease entities such as primary osteonecrosis with secondary arthritis, rheumatoid arthritis, neuropathic arthropathy or septic arthritis. In most cases, however, clinical history and radiographic findings are sufficient for excluding the latter entities. However, the rapid progression of this disease makes it difficult to obtain sequential radiographs in its early stages .
The average age at onset is greater than that of patients with ordinary coxarthrosis with primarily unilateral involvement, and severe pain but relatively preserved range of motion. One characteristic of rapidly progressive cases, however, is that they appear radiologically to be atrophic rather than hypertrophic with an almost complete absence of osteophytes . They are also characterized by the presence of lateral disease [8, 9]. Overweight, elderly women with lateral disease and minimal osteophyte formation appear to represent the group most "at risk" .
The pathologic findings, however, are consistent with osteoarthritis. Our study supports the features of RDHD described previously in literature. The short duration of symptoms as well as the retrospective nature of study may explain the absence of sequential radiographs and MR imaging in our patients. The referring clinician in primary care may not be aware of this entity and as the patient is already on waiting list for hip replacement, repeat radiographs may not be deemed necessary. But it has to be emphasized that patients suffering rapid deterioration cannot reliably be picked up by questionnaire and do not always alert the clinicians to the worsening symptoms . In addition, x-ray and pathologic progression may approach , but not exactly match pain and disability . If the clinician is made aware of RDHD, the need for extensive investigation to exclude sepsis may be obviated. Furthermore, placement of a total joint prosthesis in a patient with neuropathy can often lead to failure .
The precise pathogenesis of rapidly destructive hip osteoarthritis remains unclear. Direct drug toxicity and nonsteroidal anti-inflammatory drugs were first incriminated but their contributory effect was subsequently challenged [2, 12]. Subchondral bone ischemia and cell necrosis recently have been emphasized as major factors in the development of RDHD by Mitrovic and Riera . Pathologically, many investigators have reported that osteoclast count is significantly greater in active areas of the hip in RDHD than in osteoarthritic patients along with vascular-rich granulation, suggest ing that vascularity is an important factor in the osteolysis and destruction of the bone [14, 15]. Others have demonstrated elevated levels of interleukin-6 (IL-6), IL-1β in the joint fluid of affected patients as well as increased secretion of matrix metalloproteinases(MMP) by fibroblasts from the affected synovium and subchondral cysts [16, 17]. Although implicated as a causative factor, no genetic analysis such as that using human leukocyte antigen for rapidly destructive arthropathy has been reported.
Histologically, partial necrosis of the subchondral bone has been recognized, but it is unknown whether osteonecrosis is the primary cause of RDHD or secondary to degenerative changes because it is difficult to determine the pathogenesis from late stage specimens obtained at total hip arthroplasty. Histologically, RDHD has elements characteristic of ordinary coxarthrosis or idiopathic avascular necrosis (AVN) of the femoral head (or both): Destruction of the femoral head and the acetabulum is severe and widespread, articular cartilage disappears completely, and the synovial membrane is hypervasculated and slightly inflamed. However, no new bone formation has been observed and no osteophytes have been found, which are typical of ordinary coxarthrosis. Findings such as more rapid and more pervasive invasion, lack of a line of demarcation between necrotic and healthy tissue, and lack of a recurrent necrosis differentiate the disease from idiopathic AVN of the femoral head. Immunohistochemical analysis of articular cartilage and the synovium also demonstrates a different pattern from that of idiopathic AVN and coxarthrosis. However, the pattern is similar to that seen in rheumatoid arthritis. This immunologic response of the articular cartilage explains the chondrolysis associated with the disease, which is usually identified roentogenographically in its early stage .
The histologic features observed in our series – nonspecific severe degenerative changes with few marginal osteophytes and no evidence of primary osteonecrosis, pannus formation, and crystal deposition are similar to those observed by previous authors [13, 19, 20]. The eburnated areas showed partial osteonecrosis, which was consistent with secondary osteonecrosis in osteoarthritis . Furthermore, none of our patients exhibited chondrocalcinosis (pelvis, knees) radiographically or histopathologically. Moreover, pathologic confirmation of osteoarthritis was available in 9 cases.
MR imaging can be valuable in the evaluation of such disorders. Therefore, radiologists should be aware of MR imaging findings in patients with rapidly destructive hip osteoarthritis that can overlap other diagnostic entities. The key MR imaging features include an extensive bone marrow edema like pattern in the femoral head and neck, femoral head flattening, and cyst like subchondral defects.
Our study has a major limitation for being a retrospective study with an inherent selection bias and the study group is already selected by virtue of having come to the need for a hip replacement. However, distinction of RDO from an infectious process or neuropathic osteoarthropathy is of paramount importance and familiarity with this entity may obviate unnecessary diagnostic tests and lengthy delay in treatment with total hip replacement.