Return-to-Play and Competitive Outcomes After Ulnar Collateral Ligament Reconstruction Among Baseball Players: A Systematic Review

Abstract

Background:

Ulnar collateral ligament (UCL) reconstruction (UCLR) is very common in baseball. However, no review has compared the return-to-play (RTP) and in-game performance statistics of pitchers after primary and revision UCLR as well as of position players after UCLR.

Purpose:

To review, synthesize, and evaluate the published literature on outcomes after UCLR in baseball players to determine RTP and competitive outcomes among various populations of baseball players.

Study Design:

Systematic review; Level of evidence, 4.

Methods:

A literature search including studies between 1980 and November 4, 2019, was conducted for articles that included the following terms: ulnar collateral ligament, elbow, medial collateral ligament, Tommy John surgery, throwing athletes, baseball pitchers, biomechanics, and performance. To be included, studies must have evaluated baseball players at any level who underwent UCLR (primary or revision) and assessed RTP and/or competitive outcomes.

Results:

A total of 29 studies with relatively high methodological quality met the inclusion criteria. After primary UCLR, Major League Baseball (MLB) pitchers returned to play in 80% to 97% of cases in approximately 12 months; however, return to the same level of play (RTSP) was less frequent and took longer, with 67% to 87% of MLB pitchers returning in about 15 months. RTP rates for MLB pitchers after revision UCLR were slightly lower, ranging from 77% to 85%, while RTSP rates ranged from 55% to 78%. RTP rates for catchers (59%-80%) were generally lower than RTP rates for infielders (76%) and outfielders (89%). All studies found a decrease in pitching workloads after UCLR. Fastball usage may also decrease after UCLR. Changes in earned run average and walks plus hits per inning pitched were inconclusive.

Conclusion:

Pitchers returned to play after UCLR in approximately 12 months and generally took longer to return to their same level of play. Pitchers also returned to play less frequently after revision UCLR. After both primary and revision UCLR, professional pitchers experienced decreased workloads and potentially decreased fastball usage as well. Catchers may RTP after UCLR less frequently than pitchers, infielders, and outfielders possibly because of the frequency of throwing in the position. These results will help guide clinical decision making and patient education when treating UCL tears in baseball players.

Baseball pitching has been described as the fastest motion that the human body can produce, with the shoulder rotating at over 7000 deg/s. ^12,45 This extremely fast motion has been shown to create very large forces and torques at both the shoulder and the elbow. ¹⁹ Elbow valgus stress has been linked to injuries of the ulnar collateral ligament (UCL) in baseball players. Ultimately, repetitive valgus stress with inadequate dynamic medial elbow stability exposes the anterior band of the UCL to supraphysiological tensile strain. Over time, this can lead to partial tears, full-thickness tears, and the potential need for reconstructive surgery known as UCL reconstruction (UCLR). ²⁹ Mahure et al ³⁸ analyzed 890 patients between 2003 and 2014 in New York State and found a 343% increase in the number of UCLR procedures performed throughout the time period, with patients aged between 15 and 19 years undergoing significantly more UCLR procedures than do other ages. While 30% of athletes, 20% of coaches, and 44% of parents believe that UCLR actually improves athletic performance, the literature does not support this, and revision UCLR is still possible. ^1,30,39,43

Erickson et al ¹⁷ showed that between 2007 and 2011, patients aged between 15 and 19 years accounted for 56.8% of UCLR procedures, while patients aged between 20 and 24 years only accounted for 22.2% of UCLR procedures. The American Sports Medicine Institute documented that only 4 UCLR procedures were performed on high school–aged athletes between 1994 and 1997, whereas the number of patients who underwent UCLR among this population grew to about 30 per year by 2007. ²¹ Increased UCL injury rates among younger populations warrant greater investigation into injury mechanisms, rest allocation, and preventive care. ²¹ Pitch count regulations have been set throughout youth baseball; however, this may not be enough to limit accumulated microdamage, which ultimately weakens ligaments, tendons, and joint capsules. ^21,52 This can lead to significant injuries in older populations, which may affect injury rates described at professional and collegiate levels. ^9,10,21,49

Baseball position players are less likely to undergo UCLR, whereas pitchers are the most prone to undergo this procedure. ^{16,18,23,24,26,29} Among professional baseball players, arm-related injuries accounted for 39% of all injuries between 2011 and 2016 in which pitchers were 3.6 times more likely to be injured than were catchers, 5.1 times more likely than were outfielders, and 5.8 times more likely than were infielders. ⁶ Conte et al ⁸ analyzed the number of Major League Baseball (MLB) players who underwent UCLR between 1974 and 2015 and found that 400 UCLR procedures were performed during this time frame, with nearly one-third (32.8%) of the procedures being performed in the last 5 years of the study. Posner et al ⁴⁹ performed an epidemiological analysis of MLB injuries between 2002 and 2008 and found that pitchers had the most disabled list days (62.4% of the total) and that elbow injuries accounted for 26% of all injuries by anatomic region.

Several systematic reviews have looked at both return-to-play (RTP) and competitive outcomes after UCLR. Coughlin et al ¹¹ found 14 studies assessing RTP and in-game performance statistics after UCLR in pitchers. Rates of return to the same level of play (RTSP) of 79% to 87% were found for MLB pitchers. Additionally, 2 of 5 studies showed an increase in earned run average (ERA), 1 of 5 studies showed a decrease in ERA, 3 of 4 studies showed decreased pitch velocity, and 5 studies found decreased pitching workloads after UCLR. Peters et al ⁴⁶ conducted a systematic review and meta-analysis of RTP and in-game performance statistics after UCL injuries. They found that MLB players returned to play 89% of the time and returned to the same level of play 78% of the time with worse in-game pitching statistics, decreased innings pitched, and decreased fastball velocity after UCL injuries. However, no systematic review has compared RTP and in-game performance statistics of pitchers after primary and revision UCLR and/or position players after UCLR. The purpose of this systematic review was to identify the RTP and competitive outcomes for baseball players who undergo UCLR. The primary goal of this work was to identify the objective efficacy of reconstructive surgery among different baseball populations.

Methods

Literature Search and Study Selection

The PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) statement guided this systematic review. ⁴⁴ An initial systematic search of the literature was completed by a single reviewer (R.W.P.) in MEDLINE (PubMed), the Cumulative Index to Nursing and Allied Health (CINAHL), Embase, and SPORTDiscus between 1980 and November 4, 2019. Studies were pooled and filtered through Zotero reference management software (George Mason University).

A combination of search terms using Boolean search operators included the following: “Elbow” AND [“ulnar collateral ligament” OR “medial collateral ligament” or “Tommy John Surgery”] AND [“throwing athletes” OR “pitchers” AND [“biomechanics” OR “performance”]. Duplicates across databases were removed, followed by a screening of titles and abstracts. Full texts were then extracted and further evaluated for inclusion and exclusion criteria. The final inclusion of articles was then further inspected by all members of the research team, and disagreements regarding inclusion were solved by a consensus.

Inclusion and Exclusion Criteria

To meet inclusion criteria, articles must have included baseball players at any level who had undergone primary or revision UCLR and assessed RTP and/or competitive outcomes. Articles were included if they were written in English and were published in a peer-reviewed journal. Acceptable study designs for inclusion were cohort studies, cross-sectional studies (retrospective and prospective), case studies, and case series. Research studies that assessed/included nonbaseball athletes, did not isolate baseball player position, or did not isolate level of competition (eg, amateur, collegiate, professional) were excluded.

Assessment of Methodological Quality

Because of the inclusion of cohort and cross-sectional studies, the STROBE (STrengthening the Reporting of OBservational studies in Epidemiology) statement was utilized to assess methodological quality. An overall score of a possible 22 was given for each included article. Two reviewers (R.W.P., S.J.W.) assessed each of the articles independently and then met to discuss differences in scoring. Disagreements in STROBE scoring were resolved via a discussion and consensus.

Data Extraction and Analysis

For included studies, information regarding study design, participants, UCL surgical interventions, RTP (the percentage of athletes who were able to compete at any level) and RTSP (the percentage of athletes who were able to compete at their previous level of play) rates, and competitive outcomes were extracted. Workload statistics (innings pitched, pitches thrown), rate statistics (statistics that are divided by workload, such as ERA, walks plus hits per inning pitched [WHIP], batting average, and walks per 9 innings), pitch velocity, and pitch usage were emphasized, while counting statistics (eg, wins, hits, strikeouts, and walks) were not discussed further because of the confounding effect of workload on counting statistics. Additionally, the major results of each of the studies were summarized and compared based on the types of outcomes that each of the studies reported.

Results

Results for the search are shown in Figure 1. After the search, a total of 29 studies ^†† met all inclusion criteria. STROBE scores indicated relatively high methodological quality across the included studies (mean ± SD, 18.9 ± 1.5 [range, 15-22]). After a methodological quality assessment, studies were subdivided by outcome measure assessments into competitive outcomes (n = 24) and RTP outcomes (n = 20).

Figure 1.

Flowchart of articles included in the systematic review. UCLR, ulnar collateral ligament reconstruction.

Competitive Outcomes Related to UCL Injuries

Data extracted related to the 24 studies ^‡‡ that assessed competitive outcomes are found in Appendix Table A1. Overall, 18 studies ^§§ assessed primary UCL surgical interventions, while 2 studies ^30,37 focused on revision UCLR, and 4 studies ^15,32,33,43 assessed both primary and revision UCL surgery. Moreover, 19 of the competitive outcome studies ^∥∥ assessed professional pitchers, with the majority utilizing publicly available data. There were 2 studies ^4,53 that investigated amateur (nonprofessional) pitchers entering the MLB draft, 2 studies ^2,27 that assessed professional position players, and 1 study ⁴⁰ that identified professional catchers. For specific competitive outcome measures, the most commonly reported were ERA (n = 15), WHIP (n = 13), strikeouts (n = 10), walks (n = 10), pitch velocity (n = 10), wins above replacement (WAR; n = 9), and pitch usage (n = 8).

Although most studies ^{16,23,28,34,35,39,51} agreed on a decrease in workload in the first season after primary UCLR, the differences in pitching effectiveness as measured using rate statistics (most frequently ERA and WHIP) were less clear. Erickson et al ¹⁶ found ERA and WHIP to significantly decrease in the season after primary UCLR. However, other studies ^23,39 did not show these significant differences in ERA and WHIP. When comparing ERA and WHIP in MLB pitchers after primary UCLR with controls, pitchers who underwent UCLR performed comparably with ^23,28,34,39 or better than (lower ERA and WHIP) ¹⁶ controls.

Liu et al ³⁷ found that a decrease in workload was the only difference when comparing before and after revision UCLR in MLB pitchers (ERA and WHIP did not differ). Marshall et al ⁴³ observed the same results but also saw an increase in the walk rate from before (4.02 walks per 9 innings) to after (4.79 walks per 9 innings) revision UCLR, while controls only had 3.49 walks per 9 innings after surgery.

There were 3 studies ^35,39,47 that agreed that fastball usage decreased significantly from before to after UCLR. The only study to disagree was Portney et al ⁴⁸ in which no differences in pitch usage or pitch velocity were found. Conclusions about changes in pitch velocity were less clear, with 4 studies ^28,35,39,51 observing decreases in pitch velocity and 2 studies ^31,48 observing no change.

There were 2 studies ^2,27 that looked at competitive outcomes before and after primary UCLR in MLB position players. Jack et al ²⁷ found no differences in performance statistics from before and after primary UCLR for catchers or infielders; however, outfielders had a decrease in WAR from 1.5 before primary UCLR to 0.8 after primary UCLR, while all other performance statistics did not differ. Also, no differences were observed between cases and controls. Begly et al ² only observed a decrease in at-bats, plate appearances, and WAR from before to after primary UCLR in MLB position players, but no differences were observed when performance statistics were compared with those of controls. Marshall et al ⁴⁰ looked at 25 professional catchers only and found a decrease in games played to be the only difference when comparing competitive outcomes before and after primary UCLR. Surprisingly, catchers who underwent UCLR had a significantly higher slugging percentage (0.40) than did controls (0.37).

RTP Outcomes Related to UCL Injuries

Data extracted related to the 20 studies ^¶¶ that assessed RTP outcomes are found in Appendix Table A2. Overall, 12 studies ^## assessed primary UCLR in pitchers, 4 studies ^5,30,37,43 focused on revision UCLR in pitchers, 1 study ¹⁵ assessed both primary and revision UCLR in pitchers, 3 studies ^2,3,27 analyzed primary UCLR in position players, and 1 study ⁴⁰ looked at catchers after primary UCLR. All of the included RTP outcome studies assessed professional players. For specific RTP outcome measures, the most commonly reported were RTP rate (n = 13) and RTSP rate (n = 16).

RTP rates for MLB pitchers after primary UCLR ranged from 80% to 97% with a mean of 12 months until RTP, ^{16,22,23,25,28,34,41,42} while RTSP rates ranged from 67% to 87% with almost 15 months until RTP. ^{16,22,23,25,28,34,39,41,42} RTP rates for MLB pitchers after revision UCLR were slightly lower, ranging from 77% to 85%, ^5,43 while RTSP rates ranged from 55% to 78%. ^5,30,37,43

There were 3 studies ^2,3,27 that analyzed RTP of MLB position players, but only Camp et al ³ had a sufficient sample size after separating the position players into infielders (n = 34), outfielders (n = 35), and catchers (n = 24). For position players, infielders returned to play the fastest (294 days), outpacing outfielders (375 days) and catchers (363 days). Catchers had an RTP rate of 59% and an RTSP rate of 50%, infielders had an RTP rate of 76% and an RTSP rate of 78%, and outfielders had an RTP rate of 89% and an RTSP rate of 74%. ³ Marshall et al ⁴⁰ looked at MLB and Minor League Baseball (MiLB) catchers and found higher RTP (80%; 20/25) and RTSP (76%; 19/25) rates, with 12.9 months until RTP.

Discussion

The purpose of this systematic review was to identify the RTP and competitive outcomes for baseball players who underwent UCLR. RTP rates were lower for pitchers after revision UCLR than after primary UCLR. RTP rates for catchers were significantly lower than those for pitchers, infielders, and outfielders. Pitching workloads decreased after UCLR, and most studies found that fastball usage also decreased after UCLR. Surprisingly, after surgery, pitchers who underwent UCLR performed equal to or better than controls. For position players, the only statistics that differed before and after UCLR were workload related (at-bats, plate appearances, WAR), and no differences were seen when compared with controls.

The results of our methodology quality assessment using STROBE scores demonstrated that, on average, the included studies were of high quality. Looking at the results more closely, it was found that 3 of the 29 studies met the “study design” category (presented key elements of the study design early in the article). It was also found that 13 of 29 met the “main results” category (gave unadjusted estimates and/or confounder-adjusted estimates and their precision; reported category boundaries; and translated estimates of the relative risk into absolute risk). Title/abstract, background/rationale, objectives, setting, participants, study size, quantitative variables, statistical methods, key results, and limitations were met by all 29 studies.

MLB pitchers returned to play after primary UCLR at rates of 80% to 97% at a mean of 12 months. ^{16,22,23,25,28,34,40,41} RTSP was less frequent, with 67% to 87% of MLB pitchers returning in about 15 months. ^{16,22,23,25,28,34,40,41} These results do demonstrate variability in the RTP and RTSP rates. When returning to play at a professional level, several things need to be considered that could contribute to this variability: age of the pitcher, role in the organization (starter, reliever, and closer), additional injuries (bone spurs, chondral defects, valgus extension overload, etc), and comorbidities. In addition, extra time is required to return to the MLB level, as players can spend weeks or months in MiLB preparing for Major League play.

In an attempt to limit the injury risk of pitchers after UCLR, the pitching workload was often decreased during the first season after surgery. ^16,32,33 Even after a structured throwing program is completed throughout rehabilitation, some tissues still may not be prepared to return to normal workloads. Disregarding the need to decrease workloads could result in worse pitching performance ³⁴ and injuries. ^32,53 However, when compared controls, pitchers who underwent UCLR had ERA and WHIP statistics equal to ^23,28,34,39 or better than ¹⁶ controls. This may be caused by several factors, such as the integrity of the surgical procedure, decreased workloads after UCLR allowing pitchers to limit fatigue, or strengthening during the rehabilitation process.

Although it is often assumed that fastball velocity increases after primary UCLR, 4 ^28,35,39,51 of 7 studies ^{28,31,35,39,42,48,51} found that it actually decreased. Interestingly, following primary UCLR fastball usage was also shown to decrease in 4 ^35,39,42,47 of 5 studies ^{35,39,42,47,48} . Fastballs account for the largest forces of any pitch type, ²⁰ and therefore, a decrease in fastball usage may help lessen the load frequency to the newly reconstructed ligament, allowing for integration and maturation of the graft. Another possible explanation may be that decreased fastball velocity results in less utilization of the pitch because of it not being as effective. Further research will need to be conducted to definitively confirm this.

Catchers may also require close monitoring after UCLR, with catchers (59%-80%) ^3,40 returning to play after primary UCLR less frequently than pitchers (80%-97%), ^{16,22,23,25,28,34,40,41} infielders (76%), ³ and outfielders (89%). ³ Catchers throw much more often than other position players and play in more games than pitchers, possibly explaining their low RTP rates. Catchers are also sometimes required to throw out base runners, requiring maximal effort and accurate throws. Some of these throws are from their knees, requiring more force to be generated from the arm instead of the legs. Adjustments to workload and throwing mechanics may be warranted for catchers after UCLR.

RTP rates were lower after revision UCLR ^5,43 than after primary UCLR. ^{16,22,23,25,28,34,40,41} Keller et al ³² found that pitchers who underwent revision UCLR pitched in more games and threw more pitches in the season after primary UCLR than pitchers who did not require revision surgery. Another study by Keller et al ³³ had contradicting results, finding that pitchers who required revision UCLR threw fewer total pitches their first season after primary UCLR compared with pitchers who did not require revision surgery; however, the revision surgery group threw a higher percentage of fastballs, showing the effect that fastball usage has on UCL injury rates. ^7,33,50 Suboptimal pitching mechanics and/or poor conditioning that are not corrected after primary UCLR may also increase the chances of needing revision UCLR. Also, pitchers who undergo revision UCLR are frequently older than pitchers who undergo primary UCLR, giving these pitchers a shorter career length and lower chances of RTP. Pitchers who undergo revision UCLR should be monitored similarly to how pitchers are monitored after primary UCLR; however, extra time for workload and throwing intensity progression should be given to a pitcher after revision surgery because of the surgical complications surrounding revision.

This study is not without limitations. First, the majority of studies that included professional players collected data from the MLB database and therefore included overlapping populations (2000-2015, 1990-2010, etc) and likely duplicated participants at times. As we would have liked to conduct a meta-analysis, this was a significant barrier, as any effect size calculated would have had substantial overlap, violating the assumptions of independence in observations within meta-analyses. ³⁶ Second, there was a lack of data on RTP rates and competitive outcomes for younger populations. With such a rise in UCLR procedures performed on youth populations, these data would be valuable for understanding this trend. Because the only included studies that looked at the amateur population focused on players entering the MLB draft, these data cannot be extrapolated to the rest of the youth population. Third, there were inconsistencies in timelines used to collect means. For example, some studies averaged 3 years before and 3 years after UCLR, while some averaged 1 year before and 1 year after UCLR. Finally, only 4 studies featured position players after primary UCLR, only 4 studies featured pitchers after revision UCLR, and no studies featured position players after revision UCLR. Therefore, the effects of UCLR on RTP and competitive success in pitchers and position players who underwent revision UCLR require additional research.

Conclusion

Pitchers returned to play after UCLR in approximately 12 months and generally took longer to return to their same level of play likely because of rehabilitation in lower levels of competition. Pitchers also returned to play less frequently after revision UCLR. After both primary and revision UCLR, professional pitchers experienced decreased workloads and potentially decreased fastball usage as well. Catchers may return to play after UCLR less frequently than pitchers, infielders, and outfielders possibly because of the frequency of throwing in the position. UCLR may only affect offensive workloads (at-bats, plate appearances) and statistics related to at-bats, such as WAR. These results will help guide clinical decision making and patient education when treating UCL tears in baseball players.