Constructing model for predicting progression of low-grade subchondral insufficiency fracture of the knee based on the first imaging manifestations
Objective To construct a model for predicting progression of low-grade subchondral insufficiency fractures of the knee(SIFK)based on imaging findings at the first visit.Methods Data of 60 low-grade SIFK patients were retrospectively analyzed.The patients were divided into progressed group(progressed to grade 3 or 4,n=30)and non-progressed group(still in grade 1 or 2,n=30)according to SIFK grades of follow-up 1 year later,and clinical data and imaging findings at the first visit were compared between groups.Then a multivariate logistic regression model was constructed based on factors being significantly different between groups for predicting progression of low-grade SIFK.The receiver operating characteristic curves were drawn,the area under the curve(AUC)was calculated to assess the predicting value of the model,which was compared with that of each predictive factor alone.Results Significant differences of the tibial varus angle,posterior tibial slope(PTS),grading of condyle cartilage injury,medial meniscus extrusion distance,also of the root injury and tear types of the medial meniscus were observed between groups(all P＜0.05).The multivariate logistic regression model established based on PTS,grading of condyle cartilage injury and medial meniscus extrusion distance at the first visit for predicting progression of low-grade SIFK was as follows:Logit(P)=-0.561+0.300XPTS(°)+1.702 X grading(grading of condyle cartilage injury)+0.874 X distance(medial meniscus extrusion distance in mm),with AUC of 0.962,higher than that of each predictive factor alone(all P＜0.05).Conclusion The model established based on images findings of PTS,grading of condyle cartilage injury and medial meniscus extrusion distance at the first visit was valuable for predicting progression of low-grade SIFK.
NETL
NSTL
万方数据
