Objective: The study is aimed to compare three recently proposed new classification systems for radial polydactyly (the modified Wassel-Flatt classification by Kim et al, the Rotterdam classification, and the modified radiographic classification systems (MCS)) for their practicality and reliability.

Material and Methods: We retrospectively reviewed patients with duplicated thumbs from 2003 to 2023 in a referral center in southern Taiwan. All patients were analyzed by medical records, photographs and radiographs, and classified according to the modified Wassel-Flatt, Rotterdam and MCS classification.

Results: A total of 419 duplicated thumbs from 380 patients were enrolled. All cases could be classified according to the Rotterdam classification, while six and three cases could not well classified by the modified Wassel-Flatt and the MCS classification, respectively. Majority of the unclassifiable cases had combined aberrants including floating, hypoplasia or triphalangism. The Rotterdam classification had the highest reliability when considering type alone and could record the most detailed informations of thumb anomalies. Nevertheless, certain type of anomalies could not be well described (eg, a delta-bone in triphalangism), and we also suggested separating hypoplastic type from floating type for more specific classification.

Conclusions: We recommend Rotterdam classification to classify radial polydactyly due to its high reliability and ability to record the most detailed informations of thumb anomalies as possible. However, certain ambiguity remains over aberrants like hypoplasia, floating duplicates and triphalangism, and further modification and update are needed.