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

Review of Classification Systems for Adult Acquired Flatfoot Deformity/Progressive Collapsing Foot Deformity and the Novel Development of the Triple Classification Delinking Instability/Deformity/Reactivity and Foot Type

Background: Classifications of AAFD/PCFD have evolved with an increased understanding of the pathology involved. A review of classification systems helps identify deficiencies and respective contributions to the evolution in understanding the classification of AAFD/PCFD. Methods: Using multiple electronic database searches (Medline, PubMed) and Google search, original papers classifying AAFD/PCFD were identified. Nine original papers were identified that met the inclusion criteria. Results: Johnson's original classification and multiple variants provided a significant leap in understanding and communicating the pathology but remained tibialis posterior tendon-focused. Drawbacks of these classifications include the implication of causality, linearity of progression through stages, an oversimplification of stage 2 deformity, and a failure to understand that multiple tendons react, not just tibialis posterior. Later classifications, such as the PCFD classification, are deformity-centric. Early ligament laxity/instability in normal attitude feet and all stages of cavus feet can present with pain and instability with minor/no deformity. These may not be captured in deformity-based classifications. The authors developed the 'Triple Classification' (TC) understanding that primary pathology is a progressive ligament failure/laxity that presents as tendon reactivity, deformity, and painful impingement, variably manifested depending on starting foot morphology. In this classification, starting foot morphology is typed, ligament laxities are staged, and deformity is zoned. Conclusions: This review has used identified deficiencies within classification systems for AAFD/PCFD to delink ligament laxity, deformity, and foot type and develop the 'Triple classification'. Advantages of the TC may include representing foot types with no deformity, defining complex secondary instabilities, delinking foot types, tendon reactivity/ligament instability, and deformity to represent these independently in a new classification system. Level of Evidence: Level V.

Radiographic Outcomes of Titanium Augment vs Bone Graft in Lateral Column Lengthening for Adult-Acquired Flatfoot Deformity

Background

Lateral column lengthening (LCL) is a surgical procedure used to manage forefoot abduction and, in theory, also increases the longitudinal arch by plantarflexion of the first ray through tensioning the peroneus longus for patients with stage IIB adult acquired flatfoot deformity (AAFD). This procedure utilizes an opening wedge osteotomy of the calcaneus, which is then filled with autograft, allograft, or a porous metal wedge. The primary aim of this study was to compare the radiographic outcomes of these different bone substitutes following LCL for stage IIB AAFD.

Methods

We conducted a retrospective review of all patients who underwent LCL from October 2008 until October 2018. Preoperative weightbearing radiographs, initial postoperative radiographs, and 1-year weightbearing radiographs were reviewed. The following radiographic measurements were recorded: incongruency angle, talonavicular coverage angle (TNCA), talar-first metatarsal angle (T-1MT), and calcaneal pitch.

Results

A total of 44 patients were included in our study. The mean age of the cohort was 54 (range, 18-74). The study cohort was divided into 2 groups. There were 17 (38.7%) patients who received a titanium metal wedge and 27 (61.5%) that received autograft or allograft. Patients that underwent LCL with the autograft/allograft group were significantly older (59 vs 47 years old, P = .006). Patients who underwent LCL with a titanium wedge had a significantly higher preoperative talonavicular angle (32 vs 27 degrees, P = .013). There were no significant differences in postoperative TNCA, incongruency angle, or calcaneal pitch at 6 months or 1 year.

Conclusion

At 6 months and 1 year, no radiographic differences were found between autograft/allograft bone substitutes vs titanium wedge in LCL.

Level of Evidence

Level III, retrospective cohort study.

Influence of Weightbearing Computed Tomography in the Progressive Collapsing Foot Deformity Classification System

BACKGROUND

The objective of this study was to compare progressive collapsing foot deformity (PCFD) classifications performed using clinical and conventional radiographs (CR) with classifications established using clinical and weightbearing computed tomography (WBCT).

METHODS

This retrospective comparative study evaluated 89 consecutive PCFD feet (84 patients). Three readers performed chart reviews and CR evaluations, determining PCFD classifications that were previously published. After a washout period, the sequence was randomized, and a new classification was executed using clinical and WBCT assessment. One of the readers repeated the WBCT evaluation for intrarater reliability.

RESULTS

Interrater reliability for the WBCT was found moderate (0.55) and intrarater excellent (0.98). Evaluation using WBCT produced 29.6% of 1ABC (CR: 25.4%, P = .270), 11.6% of 1ABCD (CR: 6.9%, P = .081), and 6.4% of BC (CR: 3.3%, P = .090) as most prevalent. Class A was presented in 83.9% (CR: 89.5%, P = .55), class B in 89.9% (CR: 76.4%, P < .001), class C in 93.6% (CR: 86.2%, P = .004), class D in 46.4% (CR: 34.8%, P = .006), and class E in 27.7% (CR: 22.5%, P = .158) of the classifications performed by WBCT.

CONCLUSION

WBCT showed a different rate of deformity recognition, which increased the incidence of all classes, especially B, C, and D. An excellent intrarater agreement was found, which infers assessment reliability combining clinical and WBCT evaluation. The obtained information could enhance disease understanding and supply patients with more precise care.

LEVEL OF EVIDENCE

Level III, retrospective comparative study.

Diagnostic Accuracy of the Progressive Collapsing Foot Deformity (PCFD) Classification

BACKGROUND

A consensus group recently proposed the term progressive collapsing foot deformity (PCFD) and a new classification with 2 stages plus 5 classes to describe the complex array of flatfoot deformities. This study aimed to validate different diagnostic accuracy rates of the PCFD classification.

METHODS

This was a survey-based study distributed among 13 foot and ankle fellowship programs for 3 groups of participants with varied experience in practice (group 1: fellows in training, group 2: surgeons in practice for 1-4 years, and group 3: surgeons in practice for ≥5 years). Each participant was asked to assign 20 different cases of flatfoot deformity to the appropriate classes and stages using the PCFD classification. The overall diagnostic accuracy, class, and stage diagnostic accuracy rates for the 20 cases were calculated first for the entire cohort and then compared among the 3 groups. The misdiagnosis rate for each class of deformity (the sum of overdiagnosis and underdiagnosis rates) of the entire cohort was calculated and compared with the other classes. Mean and standard evidence were used to describe numerical data. One-way analysis of variance was used to compare values among the 3 groups and the 5 classes. P <.05 was considered statistically significant.

RESULTS

For the whole cohort, the overall diagnostic accuracy, class diagnostic accuracy, and stage diagnostic accuracy rates were 71.0%, 78.3%, and 81.7%, respectively There was a statistically significant difference between group 1 and 2, and group 1 and 3, in overall diagnostic accuracy and class diagnostic accuracy, with no significant difference among the 3 groups regarding stage diagnostic accuracy. Class B had a significantly higher overdiagnosis rate than the rest of the classes, whereas class D was significantly underdiagnosed than others. The misdiagnosis rates for classes A to E were 3.3%, 17.5%, 11.1%, 26.0%, and 3.7%, respectively.

CONCLUSION

The PCFD classification showed overall fair diagnostic accuracy rates. The highest diagnostic accuracy was for "hindfoot valgus deformity" and "ankle instability." Further content validation of the PCFD classification is needed to examine the terminology and interpretation of those classes with low diagnostic accuracy including "midfoot/forefoot abduction deformity," "forefoot varus deformity/medial column instability," and "peritalar subluxation/dislocation."Level of Evidence: Level II, prospective comparative study.

Early results and complication rate of the LapiCotton procedure in the treatment of medial longitudinal arch collapse: a prospective cohort study

INTRODUCTION

Instability/collapse of the medial column has been associated with many conditions, particularly progressive collapsing foot deformity (PCFD), hallux valgus (HV), and midfoot arthritis (MA). Restoration of first ray length and sagittal plane alignment to restore the foot tripod is essential when treating these deformities. This study aimed to assess early results, healing, and complication rate of a distraction dorsal opening plantarflexion wedge allograft first tarsometatarsal joint fusion (LapiCotton Procedure) in patients with collapse/instability of the medial column.

METHODS

In this prospective cohort study, we included PCFD, HV, and MA patients that underwent a LapiCotton procedure. Fusion site healing was defined by > 50% bone bridging in both interfaces between allograft wedge and host bone using weight-bearing computed tomography (WBCT) after 3 months. First ray collapse radiographic correction and minor and major complications (deep dehiscence, deep infection, and reoperation) were assessed.

RESULTS

A total of 22 patients (22 feet) were included (11 PCFD, 6 MA, and 5 of HV patients). Mean follow-up was 5.9 months (range 3-12) and median allograft size was 8 mm (range 5-19 mm). Bone healing was observed in 91% of cases. Two minor complications (9%, both superficial dehiscence) and one major complication (4.5%, deep infection) were observed. Statistically significant improvement of the sagittal plane talus-first metatarsal angle was observed, with mean improvement of 9.4° (95% CI 6.7-12.1°; p < 0.0001).

CONCLUSION

In this prospective cohort study of 22 patients treated with the LapiCotton procedure for medial longitudinal arch collapse/instability, we observed a low complication rate (9% minor, 4.5% major), high healing rate after 3 months (91%), one clinically stable radiographic non-union (4.5%) and one unstable non-union (4.5%) needing reoperation. Our results demonstrate promising initial results for LapiCotton technique in treating collapse of the medial longitudinal arch in patients with PCFD, MA and HV deformities. Long-term results are needed to confirm these promising results.

LEVEL OF EVIDENCE

Level II, prospective cohort study.