The Prevalence of Osteonecrosis in 448 SARS Patients: a Screening Study with MRI
X.G. Cheng,H. Qu,W. Liu,X. Liu,K.B. Cheng,T. Zhao,X.S. Li,W. Liang,J. Guo
DOI: https://doi.org/10.1016/j.clinimag.2005.12.021
IF: 2.42
2006-01-01
Clinical Imaging
Abstract:Objective: To investigate the prevalence of osteonecrosis of major joints in patients treated with corticosteroids for severe acute respiratory syndrome (SARS) by screening MRI and discuss the correlation between use of corticosteroids and development of osteonecrosis. Methods: Screening osteonecrosis of major joints with MRI was performed on 448 medical professionals treated with corticosteroids for SARS. The joints surveyed included both hips and both knees in all 448 patients. Additionally, shoulders and ankles were scanned in 192 of 448 patients. To save time, body coils were used to cover both sides of the joint, routinely T1-weighted coronal images of both sides of the joints were acquired, if abnormalities present, STIR coronal images were added. MRI and direct radiographs (DR) of both hips were obtained in 178 patients on the same day. Three senior radiologists interpreted all MR images with regard to the presence of osteonecrosis. Two radiologists read DR and determined if osteonecrosis was present without knowing the results of MRI. Accumulative dosage, average daily dosage, and duration of administrating corticosteroids in 386 patients were collected from the "SARS Clinical Research Database" which was retrieved from the medical records. information on usage of corticosteroids in remaining 62 patients was read from patient's questionnaires during enrolment, All statistical analysis was performed with SPSS. Results: The appearances of osteonecrosis on MRI in patients with SARS were identical to that caused by other etiologies or conditions. The typical findings on MRI were linear or rim-like abnormal signals in the bone marrow of subchondral area, low signal on T1W1 and high on STIR. At least one location with osteonecrosis was found in 138 patients (31%, 138/448). Infarction in the femoral shaft was found in 137 patients treated with corticosteroids (32.6%, 137/420) and one in noncorticosteroids users (3.6%, 1/28). Significant difference (P < .05) in osteonecrosis prevalence between these two groups was analyzed with Chi-square test, 24% (101/420) of corticosteroids users had osteonecrosis in the hips (femoral head), 20% (84/420) involved the knee joints, 11.4% (48/420) involved the femoral head and knee, 12.6% (53/420) occurred in the hip only and 8.6% (36/420) in the knee only. Osteonecrosis of the shoulder (humeral head) was found in 13.5% patients (26/192) and that in the ankles was 6.7% (13/192). The agreement of the diagnosis of osteonecrosis was 57.5% between DR and MRI in 178 patients. The sensitivity and specificity of DR in diagnosing osteonecrosis were 0.11 and 0.97, respectively. The accumulative dose of corticosteroids, the duration, mean daily dose and accumulative dose of prednisolone in the osteonecrosis group were all greater than nonosteonecrosis group determined by Mann–Whitney test (all P's<.05). Logistic analysis revealed that the duration and daily dose of prednisolone were significant factors for femoral head osteonecrosis, while the use of corticosteroids and the accumulative dose were significant factors for development of osteonecrosis at any locations. Conclusion: MRI screening revealed that the prevalence of osteonecrosis was quite high in SARS patients treated with corticosteroids. Osteonecrosis frequently occurred in the femoral head, knee, shoulder, and ankle joints. Multiple joints involvement and bilateral occurrence were common. The development of osteonecrosis was associated with dose and duration of administrating corticosteroids.