Crowdsourcing the aesthetics of thumb repair in pediatric thumb hypoplasia

Abstract

Pediatric thumb hypoplasia is a group of congenital hand disorders encompassing gross underdevelopment or complete lack of a thumb. Index pollicization and toe transfer are both reasonable reconstructive options, and there is no clear consensus if either method provides a more functional or aesthetically pleasing result. Considering the psychological distress associated with poor cosmetic outcomes, aesthetics must be considered when choosing reconstructive method. A survey was generated using images of reconstructed thumbs from published papers and educational websites and posted on Amazon Mechanical Turk asking participants to rate the images on their overall aesthetic appearance, identify aesthetic features, rank thumbs according to what appears most natural, and answer a series of hypothetical functional questions. Responses were stratified by reconstruction type and analyzed with chi-square, two-sample t test, and linear regression analyses. Participants in the study were diverse across age, sex, ethnicity, and education. The average rating of the thumbs was 5.8 for index pollicization and 5.6 for toe transfer, ranging from 3.85–8.34 across both reconstructive types. Index pollicization was significantly associated with being ranked as “most natural” but neither reconstructive type was associated with being ranked as “least natural.” Aesthetic factors, including length, width, nail, and position of thumb significantly affected overall aesthetic score but only accounted for around 30% of the variation in overall score. There was no significant difference in response to hypothetical functional scenarios based on reconstructive type or dominant vs. non-dominant hand. Limitations of this study include the lack of granularity in crowdsourcing data and the lack of photographic standardization of the thumbs included in the survey. Reconstructive method should continue to be selected on an individual basis based on individual anatomy. Level of evidence: Not ratable