Fifth metacarpal fractures are frequently managed non-operatively with removable splint immobilization. Adherence to prescribed splint wear is variable, and patients injured by punching are often perceived as less adherent. This assumption has not been specifically evaluated. The primary objective was to compare splint adherence between punching and non-punching injuries.

This prospective single-center observational study included 113 adults treated non-operatively for an isolated fifth metacarpal fracture between November 2023 and May 2025. All patients received a custom thermoformed splint and standardized instructions. Follow-up adherence was assessed with a study-specific, non-validated, self-administered questionnaire. Strict adherence was defined as patient-reported wear for the full prescribed duration. A secondary lenient definition classified discontinuation no more than one week before the prescribed end date as adherent. Follow-up analyses used complete cases only. Continuous and ordinal variables were compared with Mann-Whitney U tests when non-normal or ordinal, and categorical variables with Pearson chi-square or Fisher exact tests as appropriate.

Of 113 enrolled patients, 107 (94.7%) completed follow-up and 6 did not attend scheduled follow-up visits. Strict adherence was reported by 64 of 107 patients (59.8%). Adherence was 58.8% (30/51) after punching injuries and 60.7% (34/56) after non-punching injuries, an absolute difference of -1.9 percentage points (95% CI, -19.9 to 16.1; p = 0.84). With the lenient definition, adherence was 82.4% (42/51) and 83.9% (47/56), respectively (p = 0.83). In exploratory univariate analysis, strictly adherent patients were older than non-adherent patients (median 32.5 vs 26.0 years, p = 0.028); no other demographic or fracture-related characteristic differed significantly.

In this cohort, no statistically significant difference in splint adherence was detected between punching and non-punching fifth metacarpal fractures. The confidence interval was wide, so these data do not establish equivalence and do not exclude a clinically relevant difference. Injury mechanism alone should therefore not be used to presume poor adherence. The absence of functional outcome assessment prevents conclusions about the clinical consequences of early splint discontinuation.