Abstract
Objectives:
Three biomechanical traits which are concerning for potential injury to the ulnar collateral ligament (UCL) in baseball players include: the Tommy John Twist, an open front (landing) foot position, and early swinging gate (opening) of the hips. Our objective was first to determine if these risk factors were more commonly found in those who had undergone an ulnar collateral ligament reconstruction (UCLR) versus controls and second to determine if viewing standard video and/or broadcast video would allow a predictable identification of these three factors. Our hypothesis was the three aforementioned factors would be present in a higher frequency in those with previous UCLR and that the reviewers would have at least a substantial level of agreement of identification.
Methods:
Twenty (20) consecutive professional pitchers who underwent UCLR (height=188.0±6.9cm; weight=96.5±8.1kg; throw left=5; throw right=15) were analyzed and compared to 20 controls (height=188.5±6.0cm; weight=94.5±9.1kg; throw left=4; throw right=16) with similar height and weights and handedness (p≥0.46). 3 fastballs from 3 separate games were obtained for each pitcher from publicly available video and adjusted to a very low speed for review. The 3 factors analyzed included: Tommy John Twist (at the maximum abducted, externally rotated position of the shoulder, the palm (and ball) faces center field), open front foot position (front foot not aligned with the catcher or closed [open to first base side for a right hander or third base side for a left hander]), and early swinging gate of the hips (at the maximum abducted, externally rotated position of the shoulder, the hips have already rotated and are facing home plate). 6 reviewers with experience and current positions in professional baseball and familiarity with pitching biomechanics performed the video reviews. Both the number of subjects and the number of reviewers was appropriately powered. Each pitch from each of the pitchers was graded: yes, indeterminant, or no, for each of the biomechanical factors of interest. Reviews were conducted on two separate occasions by each reviewer 1 month apart. The frequency of responses was calculated and compared between the UCLR and control groups. Inter-rater agreement was calculated via Fleiss’ Kappa (K). Agreement (K) was interpreted as: ≤ 0=poor, 0.01-0.20=slight, 0.21-0.40=fair, 0.41-0.60=moderate, 0.61-0.80=substantial, and 0.81-1.00=almost perfect.
Results:
There was significantly greater percentages of “yes” responses (p<0.01) in the UCLR group versus the control group for Tommy John Twist (9%), open front foot position (9%), and early swinging gate of the hips (14%) for both rating sessions (Table 1). Agreement was similar (fair to moderate) for control and UCLR pitchers for identifying Tommy John Twist (K=0.28-0.48). However, the agreement for UCLR pitchers was moderate (K=0.41-0.46) for all 9 pitches of Early Swinging Gate, while the agreement for control pitchers was fair (K=0.29-0.38) for 8 of 9 pitches. Finally, the agreement for UCLR pitchers was moderate to substantial (K=0.50-0.67) for Open Front Foot while agreement for control pitchers was fair to moderate (K=0.27-0.46).
Conclusions:
The identification of particular, concerning high-risk biomechanical factors in those who have had UCLR was more common and identifiable versus controls. Presence of any of these factors in a pitcher could warrant a discussion of correctability. However, based on the range of indeterminant responses of 5-25%, along with Kappa values of fair to substantial, the shortcomings of video quality and position of the camera angle suggest an improved and reproducible method for video analysis should be employed.
