Radiographic Cut-Off Values for Predicting Lateral Bony Impingement in Progressive Collapsing Foot Deformity: A Correlation Analysis with Weightbearing Computed Tomography

Abstract

Category: Hindfoot; Midfoot/Forefoot Introduction/Purpose: Lateral bony impingement, one of the main causes of lateral foot pain in progressive collapsing foot deformity (PCFD), reflects disease severity and may affect surgical planning. Weightbearing computed tomography (WBCT) is known to provide better sensitivity in detecting impingement over simple radiographs, however, many Orthopaedic centers have not yet acquired WBCT imaging. This study aimed to (1) investigate the correlation of common radiographic parameters measured on standard weightbearing radiographs with talocalcaneal and calcaneofibular distance assessed with WBCT and (2) establish cut-off values for radiographic measurements on standard radiographs to detect lateral bony impingement as identified on WBCT. Methods: Ninety-one adult patients (mean age and standard deviation, 54.1 +- 17.2 years) treated for PCFD with standard preoperative radiographs and WBCT within 6 months of each other were retrospectively identified. Patients with previous ipsilateral foot and ankle surgery or asymmetric ankle arthritis (talar tilt > 2 degrees) were excluded. The talocalcaneal distance at the sinus tarsi and subfibular calcaneofibular distance were measured in multiplanar reconstructed WBCT images. Bony impingement was defined as direct contact between the structures. The relationships between WBCT measurements and four common parameters (Talonavicular coverage angle [TNC], Talo-1st metatarsal angle, Calcaneal pitch, and hindfoot moment arm [HMA]) in standard radiographs were assessed with Pearson correlations. Receiver operating characteristic (ROC) curve analysis was performed to determine the ability of radiographic parameter thresholds to predict sinus tarsi or calcaneofibular bony impingement, and the area under curve (AUC), sensitivity, specificity, negative and positive predictive value were calculated. Results: Talocalcaneal distance narrowing at the sinus tarsi on WBCT was strongly correlated with TNC (r = 0.64, p < 0.001), and the calcaneofibular distance narrowing on WBCT correlated with the HMA moderately yet best among the parameters (r=0.55, p<0.001). Interrater and intrarater reliability of WBCT measurements was excellent. TNC (AUC=0.837, 95% confidence interval [CI], 0.745-0.906) and HMA (AUC=0.959, 95%CI, 0.895-0.989) provided the best predictive ability for sinus tarsi and calcaneofibular bony impingement, respectively (Figure 1). The cut-off value for TNC for predicting sinus tarsi bony impingement was 25.4 degrees, with a sensitivity of 80.4% and a specificity of 72.5%. The cut-off value for HMA for predicting calcaneofibular bony impingement was 25.4mm, with a sensitivity of 100% and a specificity of 81.2% (Table 1). Conclusion: This study provides evidence that common radiographic parameters in standard radiographs can be potentially used to detect lateral bony impingement in PCFD. Narrowing of talocalcaneal distance at the sinus tarsi was best correlated with abduction deformity of the foot, and the narrowing of calcaneofibular distance was best correlated with valgus hindfoot deformity. TNC and HMA on standard radiographs may be used for detecting sinus tarsi and calcaneofibular bony impingement, respectively.