What are the most disruptive publications in shoulder surgery?

Shoulder surgery literature has demonstrated significant advances over the past century, but a quantitative understanding of which publications truly disrupt the field vs. those that consolidate existing knowledge remains limited. The Disruption Index (DI), a validated bibliometric value, provides a novel approach to assessing whether subsequent research shifts attention away from a publication (disruptive innovation) or builds upon them (consolidation). We hypothesized that the DI analysis would identify distinct shoulder literature compared to publications identified by citation count alone.

A retrospective bibliometric analysis was performed using PubMed and the DI framework to query and analyze shoulder-related publications from 15 prominent orthopedic surgery journals. Disruption scores were calculated as DS = (A-B)/(C + D), where A = future publications citing the focal article without citing its references; B = number of publications citing the focal article and at least one of its references; C = number of times the focal article was cited; and D = number of future publications that cited at least one of the references of the focal article but did not cite the focal article itself. The top 25 most disruptive publications were identified, while also recording author count and temporal trends. Statistical analyses included correlation testing and descriptive statistics.

A total of 5,878 shoulder surgery publications published in 15 journals between 1954 and 2014 were identified and analyzed. The mean DI for all publications was -0.01 ± 0.04. Disruptive scores were found to be poorly correlated with citation count (r = 0.148, P < .001). Shoulder literature published in 1955 and 1967 had the highest mean disruption scores, 0.16 ± 0.27 and 0.15 ± 0.21, respectively. Mean disruption values remained close to zero across team sizes. The most disruptive publication was "A Clinical Method of Functional Assessment of the Shoulder" (DI: 0.77; 1,724 citations). The top 25 publications had a mean DI of 0.37 ± 0.17, citation count of 230.9 ± 395.1, and 2.8 ± 1.8 authors. With the top 25 publications, DI and citation count were strongly correlated (r = 0.81, P = .005).

This study marks the first use of the disruption metric in orthopedic surgery. In contrast to prior studies investigating disruption within other specialties, shoulder literature exhibits a lower level of disruption. These findings offer a glimpse into the trajectory of shoulder literature, potentially underscoring the need to move beyond established paradigms and possibly toward paradigm-shifting topics.