Intra and Inter-Observer Reliability of the New Classification System of Progressive Collapsing Foot Deformity

Abstract

Category:Basic Sciences/Biologics; Hindfoot; Midfoot/ForefootIntroduction/Purpose:Over the past three decades, the historical notion that adult acquired flatfoot was attributed to posterior tibial tendon dysfunction (PTTD) has been questioned. Advances in the understanding of arch stabilizers and biomechanics of the midfoot joints have led to the new concept of progressively collapsing foot deformity (PCFD). Most recently, the consensus group proposed a new classification system and recommended renaming PTTD to PCFD. The proposed PCFD classification system incorporates the latest understanding of the condition and provides a concise, standardized description of the deformity. To date, there has been no study reporting the frequency of each subclass of PCFD with various combination of deformity components and evaluating intra- and interrater reliability.Methods:This was a single-center, retrospective study conducted from prospectively collected registry data. 84 patients (92 feet) were assessed between 2014 and 2020. Classification of each patient was made utilizing clinical and radiographic findings by three independent observers. Clinical aspects of the deformity included hindfoot valgus, forefoot/midfoot abduction, forefoot varus deformity, hypermobile medial column, sinus tarsi impingement, peritalar subluxation and valgus tilting of the ankle joint.Radiographic evaluation was focused on Hindfoot valgus (A), increased talar head undercoverage, significant sinus tarsi impingement (B), increased lateral talus-first metatarsal angle, plantar gapping at first TMT/NC joints (C), significant subtalar joint subluxation/subfibular impingement with obliterated joint space (D), and ankle joint valgus tilting with or without arthritic changes (E). Intra- and interrater reliabilities were analyzed with Cohen's Kappa and Fleiss' kappa respectively.Results:Mean age was 54.4, 38% male and 62% female. Mean BMI was 33.6 kg/m2. 1ABC (21 feet, 22.8%) was most common subclass followed by 1AC (12 feet, 13%) and 1ABCD (8 feet, 8.7%). Cumulative percent of frequency of 1ABC, 1AC, and 1ABCD was 44.5%. Only a small percentage of patients had an isolated deformity. 58.7% were flexible, 5.5% were rigid, and 35.8% were combined deformities with flexible and rigid components. A was most frequent component (93.5%) followed by C in 88% and B in 71.7%. D was in 29.4% and E was the least frequently observed component in 23.9%. Moderate inter-rater reliability (Fleiss Kappa=0.561, p<0.001, 95% CI 0.528-0.594) was found. E was the most reliable between raters (91.3%) followed by A (79.4%). D was least reliable between three raters (45.7%). Very good intra-rater reliability was found (Cohen`s Kappa=0.851, P<0.001, 95% CI 0.777-0.926).Conclusion:Most cases predominantly involved hindfoot with various combinations of midfoot and forefoot deformity with/without subtalar joint involvement. This finding suggests most of PCFD exist in combined forms with various deformity components. Despite the limitation of inherent subjectivity, which may account for moderate inter-rater agreement, the new system potentially cover all possible combinations of the deformity in hindfoot, midfoot, forefoot and ankle. This provides a more comprehensive description of PCFD deformity and can guide treatment in a more systematic and individualized manner. Future studies in a larger cohort with advanced imaging are warranted to ascertain reliability and validity of this system.