Analysis of risk factors for poor prognosis in conservatively managed early-stage spontaneous osteonecrosis of the knee.

BACKGROUND

Management guidelines for early-stage spontaneous osteonecrosis of the knee (SONK) have not been established. The purposes of this study were to review the outcome of conservative treatment for patients with early-stage SONK and to examine clinical factors affecting the prognosis.

METHODS

Diagnosis of early-stage SONK was made based on the criteria consisting of specific clinical features including magnetic resonance imaging (MRI) findings. During the study period, all patients with this diagnosis underwent standardized conservative treatment. The study population comprised 38 knees in 36 patients with a mean age at presentation of 66.4years. The mean follow-up period was 34.9months. During the treatment course, progressive joint space narrowing or collapse of bony contours identified in serial follow-up radiographs was regarded as indicating a poor prognosis. The significance of potential prognostic factors such as age, gender, obesity, coronal alignment, lesion size, and MRI findings was analyzed using a multivariate logistic regression analysis.

RESULTS

The prognosis was defined to be poor in eight knees (21.1%). The multivariate logistic regression analysis for potential risk factors revealed that only varus alignment with a femorotibial angle (FTA) of 180° or more on the initial radiograph was significantly associated with the poor prognosis (P=0.01, odds ratio 28.1) while no other factors significantly correlated with the prognosis.

CONCLUSIONS

Approximately 80% of patients with early-stage SONK could be managed successfully with conservative treatment without progression of the disease process. The presence of varus deformity (FTA of 180° or more) was significantly associated with poor prognosis complicated with progressive deformity and prolonged disability.

LEVEL OF EVIDENCE

Level IV, case series.