Fastball Pitch Velocity Helps Predict Ulnar Collateral Ligament Reconstruction in Major League Baseball Pitchers

Abstract

Background:

Ulnar collateral ligament injury and its subsequent surgical reconstruction are some of the most common issues among Major League Baseball (MLB) players.

Purpose/Hypothesis:

The purpose of this study was to determine factors predictive of ulnar collateral ligament reconstruction (UCLR) among MLB pitchers. The hypothesis was that pitchers who underwent UCLR would have higher preinjury peak fastball pitch velocity.

Study Design:

Case-control study; Level of evidence, 3.

Methods:

Data on pitch velocity, number, and type (fastball, curveball, etc) for every pitcher and game within MLB from April 2, 2007 to April 14, 2015 were gathered from the publically available PitchFx database. Pitcher demographic information was also recorded. Data from after 2012 were excluded to avoid lead-time bias. Using publically available information, the names and approximate dates of surgery for every MLB pitcher who ever underwent UCLR, including those before 2007 and after 2012, were collected. Each pitcher-game was then classified as “control,” “preinjury,” or “postoperative.” Control and preinjury pitchers were then compared to determine risk factors for UCLR.

Results:

Overall, 1327 pitchers were included, of whom 309 (26.8%) had undergone UCLR. Of these, 145 had preinjury velocity data. Peak pitch velocity was significantly higher among preinjury pitchers than control pitchers (mean [95% CI], 93.3 mph [92.8-93.8] vs 92.1 mph [91.9-92.3]; P < .001), as was mean pitch velocity (87.8 mph [87.3-88.3] vs 86.9 mph [86.7-87.1]; P = .001). Both demonstrated a dose-response relationship. Although height did not differ (P = .934), weight was significantly higher for preinjury pitchers than controls (P = .005). Pitch counts per year were significantly lower for preinjury pitchers compared with control pitchers, although preinjury pitchers threw more breaking pitches (P = .003). On multivariate regression, peak pitch velocity was the primary independent predictor of whether a pitcher underwent UCLR (P < .001), with mean velocity (P = .013), body mass index (P = .010), and age (P = .006) being secondary predictors. However, a model constructed with these variables only explained 7% of the variance in UCLR rates. Pitch counts were not significant predictors.

Conclusion:

Higher pitch velocity is the most predictive factor of UCLR in MLB pitchers, with higher weight and younger age being secondary predictors, although these factors only explained 7% of the variance in UCLR rates.