Drs Earhart, Roberts, Roc, Gryzlo, and Hsu are from the Department of Orthopaedic Surgery, Feinberg School of Medicine, Northwestern University, Chicago, Illinois.
Drs Earhart, Roberts, and Gryzlo have no relevant financial relationships to disclose. Dr Roc has received funding for research through an honorarium from the Department of Orthopaedic Surgery, Feinberg School of Medicine, Northwestern University. Dr Hsu is a paid consultant for Styker, Medtronic, Pioneer, and Zimmer.
Correspondence should be addressed to: Jeffrey S. Earhart, MD, Department of Orthopaedic Surgery, Feinberg School of Medicine, Northwestern University, 676 N Saint Claire St, Ste 1350, Chicago, IL 60611 (firstname.lastname@example.org).
Lumbar disk herniation is a common cause of work-related disability in the general population, with a substantial economic burden. The outcomes of operative and nonoperative treatment have been examined extensively in the general population.1–5 However, the unique effect of this condition and its treatment on elite athletes has been less well studied. Authors have criticized the Spine Patient Outcomes Research Trial (SPORT)6 studies for providing little information regarding maintenance of activity level after treatment for a lumbar disk herniation.7
Professional athletes place daily stresses on their lumbar spine that are far beyond that of the general population or even the recreational athlete.8 Furthermore, substantial financial implications to an athlete’s recovery period, longevity, and performance after an injury exist. For this reason, common clinical outcome measurements in the peer-reviewed literature, such as visual analog pain scores and return-to-work rates, for sedentary professions are insufficient gauges of successful treatment for elite professional athletes. A better understanding of outcomes after treatment for lumbar disk herniation based on pertinent performance measures would be important to athletes, coaches, team owners, and clinicians.
Major League Baseball (MLB) players have the longest season of any of the major professional sports, with a regular season of 162 games over the course of 6 months. This extensive training and playing period, along with the repetitive stresses of batting and pitching, poses a unique challenge to the successful treatment of a career-threatening injury such as a lumbar disk herniation. Although retrospective studies in professional basketball9 and American football10 players have reported favorable performance-based outcomes after operative treatment for lumbar disk herniation, the expectations after surgery in MLB athletes are less clear.
Recognizing that the stresses of playing professional baseball differ from those of other sports, this study reports the performance-based outcomes after operative and nonoperative treatment for lumbar disk herniation in MLB players.
Materials and Methods
This is a retrospective comparative study of professional MLB athletes diagnosed with symptomatic lumbar disk herniation treated with operative or nonoperative measures. Using a previously established protocol,9–13 players diagnosed with a lumbar disk herniation were identified. Sources such as press releases, newspaper articles, and team injury reports were used. Once a lumbar disk herniation injury was confirmed, personally identifiable player information, such as name and team, were removed prior to entry into our database.
Inclusion criteria included MLB players on an active roster at the time of injury with reported “herniated disk” or “disk herniation” in the lumbar spine, reported radicular symptoms, and a confirmed treatment plan from at least 2 sources. Operative treatment was defined as either a lumbar microdiskectomy and/or laminotomy/foraminotomy, whereas nonoperative treatment was defined as any treatment not involving surgery, such as epidural steroid injections, physical therapy, functional rehabilitation, and activity modification. Standardization of injury type and severity was attempted through exclusion of players with only reported bulging or protruding disk, lack of radicular symptoms, lumbar or back strain, or fracture. Operative patients undergoing lumbar fusion were also excluded. For each eligible player, demographic data such as age, date of MLB debut, height, weight, position, date of injury, category of treatment, and retirement date (if applicable) were recorded. Players were categorized as pitchers (including all starters and relievers) or hitters (all position players, including designated hitters). Hitters were further subcategorized as infielders (first/second/third base, shortstop, and designated hitters), outfielders (left, center, or right fielders), or catchers for separate analysis.
Relevant clinical parameters for successful treatment of lumbar disk herniation in a professional baseball player were identified. Return to play was defined as return to the active roster for at least 1 professional, regular-season game. Career length before and after injury, measured in games and months played, was determined from public MLB data sources.14 Only MLB athletes who retired prior to the end of the 2009 regular season were included in the data analysis for career longevity.
Pre- and posttreatment player performance was compared through the use of common pitching and hitting performance statistics. Based on previous publications reporting performance-based outcomes in pitchers,15,16 total wins, earned run average (ERA), saves, innings pitched, strikeouts, and walks plus hits divided by innings pitched (WHIP) were used as formal posttreatment outcome measures of pitching performance. Runs, home runs, runs batted in (RBI), stolen bases, and batting average were used for the cohort of hitters as well as the subcohorts of infielders, outfielders, and catchers. Although more comprehensive metrics exist, these categories were chosen based on their familiarity and the frequency of their use in quantifying the performance of MLB players.14 When preinjury career length was sufficient, the average for each performance category over the 3 years prior to injury was calculated to standardize performance and level of play at the time of injury for each individual athlete. Players who met the career length criteria had these preinjury averages compared with their posttreatment performance statistics averaged at 1- and 3-year time points. The hitter subcategories of infielders and outfielders were analyzed in the same fashion. Each player cohort served as its own respective control. Although catchers were included in our analysis of all hitters, we also analyzed them separately from all infielders because of the unique demands and forces on the lumbar spine this position requires.
Statistical analysis was performed using the Data Analysis Pack from Microsoft Excel 2007 (Redmond, Washington). Fisher’s exact test was used to analyze categorical data. Continuous data of each cohort was compared using a two-tailed, unequal variance t test for normally distributed data. Statistical significance was set at P<.05.
A total of 69 reported episodes of symptomatic lumbar disk hernation (in 64 players) from 1980 to 2009 met the inclusion criteria. Of these episodes of lumbar disk hernation, 29 occurred in pitchers and 40 in hitters. Based on the available sources, no player was diagnosed or treated for multiple levels simultaneously. Operative treatment was performed in 40 cases (20 pitchers and 20 hitters), whereas nonoperative treatment was used in 29 cases (9 pitchers and 20 hitters).
No significant difference existed between the pitcher and hitter cohorts in terms of age at injury and years played at time of injury; however, the hitter cohort had a significantly higher average body mass index (BMI) than the pitching cohort (26.7 vs 25.4, respectively; P=.04) (Table 1). When stratified for category of treatment, no significant differences in average BMI were noted among the pitcher or hitter cohorts; rather, pitchers managed operatively were found to be older than hitters (32.9 vs 28.1 years, respectively; P=.008) and had longer careers at the time of injury (9.0 vs 4.0 years, respectively; P=.005). This was also found in hitters treated operatively vs nonoperatively (31.8 vs 29.3 years, respectively; P=.04; 9.0 vs 6.2 years played prior to injury, respectively; P=.03). Further stratification of hitters revealed similar differences among outfielders but no significant differences among infielders (Table 2).
Table 1: Patient Demographics by Position
Table 2: Patient Demographics by Position and Treatment
Return to Play and Recovery Time
Overall, the return-to-play rate following treatment of lumbar disk hernation was 97.1% at an average time of 6.6 months from the time of injury. This included 39 of 40 (97.5%) players treated operatively and 28 of 29 (96.6%) treated nonoperatively (Figure 1). Players treated operatively had a significantly longer recovery time than those treated nonoperatively (8.7 vs 3.6 months, respectively; P<.001) (Figure 2).
Figure 1: Return-to-play rate by position.
Figure 2: Mean time to return to play (months).
All pitchers (29 of 29) returned to competitive play after operative or nonoperative treatment of a lumbar disk hernation, whereas 38 of 40 (95%) hitters returned (1 catcher treated nonoperatively and 1 infielder treated operatively did not return to play) (Figure 1). With the limited numbers available, recovery time from the date of injury was not significantly different between pitchers treated operatively and nonoperatively (P=.25). However, hitters demonstrated longer postoperative rehabilitative time compared with those receiving nonoperative care (9.4 vs 2.6 months, respectively; P<.006). When players were stratified by position, this was also true for infielders treated operatively vs nonoperatively (11.4 vs 1.5 months, respectively; P=.006) but not for outfielders (P=.43) (Figure 2). Four catchers, all treated nonoperatively, were identified and analyzed independently from other hitters. Three of the 4 successfully returned to play at an average of 2.3 months after diagnosis (Table 3).
Table 3: Rate and Timing of Return to Play
After operative treatment for lumbar disk hernation, players participated in fewer games than those undergoing nonoperative treatment, with a trend toward significance (432.4 vs 232.8 games, respectively; P=.08). No significant difference existed in the number of months played following treatment (P=.77). Subgroup analysis of athletes based on position played did not show a significant difference in the number of posttreatment games or months played following operative or nonoperative care (Table 4). The absence of an operative cohort for catchers precluded a similar comparative analysis. However, catchers who returned from nonoperative treatment continued their careers for an average of 549.3 games spanning 74.8 months (Table 4).
Table 4: Career Longevity Preinjury and Posttreatment
For pitchers in the operative and nonoperative cohorts, the number of wins, saves, innings pitched, and strikeouts did not change from average preinjury values at either 1 or 3 years posttreatment. However, pitchers treated operatively had a significantly poorer ERA and WHIP at 1 (P<.05 and P<.04, respectively) and 3 years after injury (P<.04 and P<.03, respectively) when compared with preinjury levels (Table 5). With the limited numbers available, pitchers treated nonoperatively had no significant change in any measured performance parameter at 1 or 3 years when compared with preinjury statistics.
Table 5: Pitchers: Preinjury and Posttreatment Performance
Hitters in both the operative and nonoperative cohorts had no change in the number of home runs, stolen bases, or batting average at 1 or 3 years posttreatment when compared with preinjury average values. However, the operative cohort demonstrated a significant decrease in runs (P=.008) and RBIs (P=.009) 1 year postoperatively, which was not seen in the nonoperative group. No significant differences were seen in either of these performance statistics at the 3-year time point postoperatively. Similar to pitchers, nonoperative treatment for hitters did not cause a significant change in any statistical category of production after 1 or 3 years (Table 6).
Table 6: All Hitters: Preinjury and Posttreatment Performance
Subgroup analysis demonstrated that infielders had no statistically significant change in any production category following operative or nonoperative treatment; however, a decrease in runs and RBIs produced 1 year postoperatively approached significance (P=.10 and P=.08, respectively) (Table 7). In contrast, outfielders produced significantly fewer runs (P<.05) and had a poorer batting average (P<.03) 1 year postoperatively. A trend also existed toward fewer RBIs produced after 1 year (P=.07). Each of these decreases in postoperative production was not significant when compared with the 3-year posttreatment averages (Table 8). Nonoperative treatment among the infielder and outfielder cohorts was not associated with a significant change in any production category at either time point (Tables 7, 8).
Table 7: Infielders: Preinjury and Posttreatment Performance
Table 8: Outfielders: Preinjury and Posttreatment Performance
Decades of clinical research on the treatment of herniated lumbar disks have focused on validated outcome measures such as the visual analog scale, Short Form-36 questionnaire, Oswestry Disability Index, Low Back Outcome Score, and return-to-work status. Although these outcome measures have proven pertinent to the general population, the unique physical demands of the professional athlete are such that these are inadequate. Thus, landmark prospective studies of lumbar disk herniation in the general population are not necessarily helpful to guide decision making for this group of patients.2,4,5
The data from this study suggest high overall return-to-play rates after operatively or nonoperative lumbar disk hernation treatment (97.1%), with a significantly longer recovery time in the operative cohort. The operative group showed a trend toward fewer games played after treatment than the nonoperative cohort. Finally, using several commonly used statistical measures of performance, pitchers and hitters demonstrated significantly poorer performance in certain categories at 1 year postoperatively, with the maintenance of differences in pitchers after 3 years. Notably, no significant changes in any performance statistic were seen in pitchers or hitters at 1 or 3 years following nonoperative treatment.
This study had several inherent limitations. Although our method of data acquisition in this study has precedence in the literature,9–13 the use of public records introduces the possibility of reporting errors. Furthermore, without medical records available for every player, the indications for operative treatment were not standardized across the population. In addition, although we used common performance statistics for pitching and hitting, these statistics are influenced by a player’s teammates and managerial decisions and do not reflect defensive play. Finally, several confounding variables could have affected the data, such as the age difference and preinjury playing time in our operative cohorts; limited numbers in subgroups, such as pitchers treated nonoperatively; contract stipulations related to injury or performance; personal reasons for retirement; and other concomitant injuries in addition to lumbar disk hernation.
Nonetheless, our findings suggest that further study is necessary to identify athletes that may be predisposed to poor outcome after lumbar disk herniations. The reasons for the operative group having lower statistical performance than the nonoperative group could be explained by a number of reasons. First, the average age of athletes in the operative group at the time of treatment was significantly older than the nonoperative cohort. Although this difference was small, this fact could have affected postoperative career length after treatment. Second, MLB athletes enjoy guaranteed contracts, unlike National Football League (NFL) players, which may create potential incentive differences after injury. Finally, the demands of MLB are likely unique compared with other sports. The regular season is longer and the expectation of repetitive, high-torque hitting and pitching motions hundreds of times a day may accelerate chronic degenerative processes of the lumbar spine after operative treatment. The paraspinal musculature can be weakened by surgery, which could affect the outcome.
A study of surface electromyographic (EMG) activity in both the paraspinal and trunk musculature during the pitching motions of professional baseball pitchers showed that these muscle groups are vital for stability in the uncoiling phase, assisting in the transfer of power from the legs and trunk to the throwing arm. This process of muscle activation was shown to be consistently asymmetric, with a respective 40%, 50%, and 85% increase in EMG activity of the nondominant rectus abdominis, oblique, and paraspinal muscles compared with the dominant side.8 This asymmetry highlights the significant amount of torque that is generated and resisted by the lumbar spine of pitchers, making a strong trunk and low back critical to proper pitching mechanics and injury prevention. These demands could contribute to a difference in outcome after lumbar disk herniation compared with athletes of other professional sports such as American football and basketball.
The findings of our study contrast with those reported in players of the National Basketball Association (NBA)9 and NFL,10 where data were collected using the same methodology. Whereas we discovered several significant decreases in MLB player performance postoperatively, NBA players treated operatively had small but statistically significant improvements in performance when compared with a nonoperative cohort.9 National Football League players treated operatively for lumbar disk herniation, unlike our MLB cohort, played in significantly more games (P=.002) and for more months (P=.03) post-treatment, while maintaining equivalent performance scores when compared with nonoperative controls.10
Although definitive conclusions cannot be made from this study, we believe that the physical demands of professional baseball athletes may eventually prove to contribute to differences in outcome after operative treatment of lumbar disk hernation in comparison with NFL and NBA players. Prospective outcome studies in this patient population are warranted to answer this important question.
- Atlas SJ, Deyo RA, Keller RB, et al. The Maine Lumbar Spine Study, Part II. 1-year outcomes of surgical and nonsurgical management of sciatica. Spine (Phila Pa 1976). 1996; 21(15):1777–1786. doi:10.1097/00007632-199608010-00011 [CrossRef]
- Atlas SJ, Keller RB, Wu YA, Deyo RA, Singer DE. Long-term outcomes of surgical and nonsurgical management of sciatica secondary to a lumbar disc herniation: 10 year results from the Maine Lumbar Spine Study. Spine (Phila Pa 1976). 2005; 30(8):927–935. doi:10.1097/01.brs.0000158954.68522.2a [CrossRef]
- Tosteson AN, Skinner JS, Tosteson TD, et al. The cost effectiveness of surgical versus nonoperative treatment for lumbar disc herniation over two years: evidence from the Spine Patient Outcomes Research Trial (SPORT). Spine (Phila Pa 1976). 2008; 33(19):2108–2115. doi:10.1097/BRS.0b013e318182e390 [CrossRef]
- Weber H. Lumbar disc herniation. A controlled, prospective study with ten years of observation. Spine (Phila Pa 1976). 1983; 8(2):131–140. doi:10.1097/00007632-198303000-00003 [CrossRef]
- Weinstein JN, Lurie JD, Tosteson TD, et al. Surgical versus nonoperative treatment for lumbar disc herniation: four-year results for the Spine Patient Outcomes Research Trial (SPORT). Spine (Phila Pa 1976). 2008; 33(25):2789–2800. doi:10.1097/BRS.0b013e31818ed8f4 [CrossRef]
- Birkmeyer NJ, Weinstein JN, Tosteson AN, et al. Design of the Spine Patient Outcomes Research Trial (SPORT). Spine (Phila Pa 1976). 2002; 27(12):1361–1372. doi:10.1097/00007632-200206150-00020 [CrossRef]
- O’Neill C. Dream from our President: Barack Obama and evidence-based medicine. Spine J. 2010; 10(2):105–107. doi:10.1016/j.spinee.2009.12.001 [CrossRef]
- Watkins RG, Dennis S, Dillin WH, et al. Dynamic EMG analysis of torque transfer in professional baseball pitchers. Spine (Phila Pa 1976). 1989; 14(4):404–408. doi:10.1097/00007632-198904000-00011 [CrossRef]
- Anakwenze OA, Namdari S, Auerbach JD, et al. Athletic performance outcomes following lumbar discectomy in professional basketball players. Spine (Phila Pa 1976). 2010; 35(7):825–828.
- Hsu WK. Performance-based outcomes following lumbar discectomy in professional athletes in the National Football League. Spine (Phila Pa 1976). 2010; 35(12):1247–1251.
- Carey JL, Huffman GR, Parekh SG, Sennett BJ. Outcomes of anterior cruciate ligament injuries to running backs and wide receivers in the National Football League [published online ahead of print July 26, 2006]. Am J Sports Med. 2006; 34(12):1911–1917. doi:10.1177/0363546506290186 [CrossRef]
- Hsu WK. Outcomes following nonoperative and operative treatment for cervical disc herniations in National Football League athletes. Spine (Phila Pa 1976). 2011; 36(10):800–805. doi:10.1097/BRS.0b013e3181e50651 [CrossRef]
- Savage J, Hsu WK. Statistical performance in National Football League athletes after lumbar discectomy. Clin J Sport Med. 2010; 20(5):350–354. doi:10.1097/JSM.0b013e3181efc259 [CrossRef]
- Baseball statistics and history. Baseball-Reference Web site. http://www.baseball-reference.com/. Updated November 30, 2011. Accessed June 10, 2010.
- Cerynik DL, Ewald TJ, Sastry A, Amin NH, Liao JG, Tom JA. Outcomes of isolated glenoid labral injuries in professional baseball pitchers. Clin J Sport Med. 2008; 18(3):255–258. doi:10.1097/JSM.0b013e318170ae86 [CrossRef]
- Ricchetti ET, Weidner Z, Lawrence JT, Sennett BJ, Huffman GR. Glenoid labral repair in Major League Baseball pitchers [published online ahead of print February 10, 2010]. Int J Sports Med. 2010; 31(4):265–270. doi:10.1055/s-0030-1247596 [CrossRef]
Patient Demographics by Position
|Mean age at injury, y||31.1||31.4||30.5||.43|
|Mean years played at injury||7.5||7.5||7.6||.94|
Patient Demographics by Position and Treatment
| Mean BMI||25.4||25.6||24.9||.07|
| Mean age at injury, y||31.4||32.9||28.1||.008|
| Mean years played at injury||7.5||9.0||4.0||.005|
| Mean BMI||26.7||26.5||26.9||.20|
| Mean age at injury, y||30.5||31.8||29.3||.04|
| Mean years played at injury||7.6||9.0||6.2||.03|
| Mean BMI||26.4||26.5||26.3||.64|
| Mean age at injury, y||30.4||30.9||29.5||.51|
| Means years played at injury||7.5||8.2||6.5||.46|
| Mean BMI||26.6||26.6||26.6||.065|
| Mean age at injury, y||30.9||33.4||28.6||.001|
| Mean years played at injury||7.8||10.5||5.5||.008|
Rate and Timing of Return to Play
|N||No. Players Returned to Play (%)||Mean Time to Return, mo||P|
| Total||69||67 (97.1)||6.6|
| Operative||40||39 (97.5)||8.7|
| Nonoperative||29||28 (96.6)||3.6||<.001|
| Total||29||29 (100)||7.3|
| Operative||20||20 (100)||8.0|
| Nonoperative||9||9 (100)||5.7||.25|
| Total||40||38 (95)||6.0|
| Operative||20||19 (95)||9.4|
| Nonoperative||20||19 (95)||2.6||.006|
| Total||21||20 (95.2)||7.4|
| Operative||13||12 (92.3)||11.4|
| Nonoperative||8||8 (100)||1.5||.006|
| Total||15||15 (100)||4.8|
| Operative||7||7 (100)||5.9|
| Nonoperative||8||8 (100)||3.8||.43|
| Total||4||3 (75)||2.3|
| Operative||0||0 (0)||0|
| Nonoperative||4||3 (75)||2.3||n/a|
Career Longevity Preinjury and Posttreatment
|Mean No. Games Played||P||Mean No. Months Played||P||Mean No. Games Played||P||Mean No. Months Played||P|
Pitchers: Preinjury and Posttreatment Performancea
|Treatment (N)||3 y Pre||1 y Post||P||3 y Post||P|
All Hitters: Preinjury and Posttreatment Performancea
|Treatment (N)||3 y Pre||1 y Post||P||3 y Post||P|
Infielders: Preinjury and Posttreatment Performancea
|Treatment (N)||3 y Pre||1 y Post||P||3 y Post||P|
Outfielders: Preinjury and Posttreatment Performancea
|Treatment (N)||3 y Pre||1 y Post||P||3 y Post||P|