The Professional Athlete Spine Initiative: outcomes after lumbar disc herniation in 342 elite professional athletes.

Abstract

Although clinical outcomes after lumbar disc herniations (LDHs) in the general population have been well studied, those in elite professional athletes have not. Because these athletes have different measures of success, studies on long-term outcomes in this patient population are necessary.
This study seeks to define the outcomes after an LDH in a large cohort of professional athletes of American football, baseball, hockey, and basketball.
Retrospective cohort study.
A total of 342 professional athletes from four major North American sports from 1972 to 2008 diagnosed with an LDH were identified via a previously published protocol. Two hundred twenty-six players underwent lumbar discectomy, and 116 athletes were treated nonoperatively. Only those players who had at least 2 years of follow-up were included.
Functional outcome measures as defined by successful return-to-play (RTP), career games, and years played for each player cohort were recorded both before and after treatment. Conversion factors based on games/regular season and expected career length (based on individual sport) were used to standardize the outcomes across each sport.
Using Statistical Analysis Software v. 9.1, outcome measures were compared in each cohort both before and after treatment using linear and mixed regression analyses and Cox proportional hazards models. A Kaplan-Meier survivorship curve was calculated for career length after injury. Statistical significance was defined as p<.05.
After the diagnosis of an LDH, professional athletes successfully returned to sport 82% of the time, with an average career length of 3.4 years. Of the 226 patients who underwent surgical treatment, 184 successfully returned to play (81%), on average, for 3.3 years after surgery. Survivorship analysis demonstrated that 62.3% of players were expected to remain active 2 years after diagnosis. There were no statistically significant differences in outcome in the surgical and nonoperative cohorts. Age at diagnosis was a negative predictor of career length after injury, whereas games played before injury had a positive effect on outcome after injury. Major League Baseball (MLB) players demonstrated a significantly higher RTP rate than those of other sports, and conversely, National Football League (NFL) athletes had a lower RTP rate than players of other sports (p<.05). However, the greatest positive treatment effect from surgery for LDH was seen in NFL players, whereas for MLB athletes, a lumbar discectomy led to a shorter career compared with the nonoperative cohort (p<.05).
Professional athletes diagnosed with an LDH successfully returned to play at a high rate with productive careers after injury. Whereas older athletes have a shorter career length after diagnosis of LDH, experienced players (high number of games played) demonstrate more games played after treatment than inexperienced athletes. Notably, surgical treatment in baseball players led to significantly shorter careers, whereas for NFL athletes, posttreatment careers were longer than those of the corresponding nonoperative cohort. The explanation for this is likely multifactorial, including the age at diagnosis, respective contractual obligations, and different physical demands imposed by each individual professional sport.

Full text

Page 1

Clinical Study
The Professional Athlete Spine Initiative: outcomes after lumbar disc
herniation in 342 elite professional athletes
Wellington K. Hsu, MDa,b,*, Kathryn J. McCarthy, MDa, Jason W. Savage, MDa,
David W. Roberts, MDa, Gilbert C. Roc, MDa, Alan J. Micev, MDa, Michael A. Terry, MDa,
Stephen M. Gryzlo, MDa, Michael F. Schafer, MDa
aDepartment of Orthopaedic Surgery, Northwestern University Feinberg School of Medicine, 676 N. St. Clair St, Ste 1350, Chicago, IL 60611, USA
bDepartment of Neurological Surgery, Northwestern University Feinberg School of Medicine, 676 N. St. Clair St, Ste 2210, Chicago, IL 60611, USA
Received 16 May 2010; revised 1 December 2010; accepted 17 December 2010
Abstract BACKGROUND CONTEXT: Although clinical outcomes after lumbar disc herniations (LDHs)
in the general population have been well studied, those in elite professional athletes have not.
Because these athletes have different measures of success, studies on long-term outcomes in this
patient population are necessary.
PURPOSE: This study seeks to define the outcomes after an LDH in a large cohort of professional
athletes of American football, baseball, hockey, and basketball.
STUDY DESIGN/SETTING: Retrospective cohort study.
PATIENT SAMPLE: A total of 342 professional athletes from four major North American sports
from 1972 to 2008 diagnosed with an LDH were identified via a previously published protocol. Two
hundred twenty-six players underwent lumbar discectomy, and 116 athletes were treated nonoper-
atively. Only those players who had at least 2 years of follow-up were included.
OUTCOME MEASURES: Functional outcome measures as defined by successful return-to-play
(RTP), career games, and years played for each player cohort were recorded both before and after
treatment. Conversion factors based on games/regular season and expected career length (based on
individual sport) were used to standardize the outcomes across each sport.
METHODS: Using Statistical Analysis Software v. 9.1, outcome measures were compared in each
cohort both before and after treatment using linear and mixed regression analyses and Cox propor-
tional hazards models. A Kaplan–Meier survivorship curve was calculated for career length after
injury. Statistical significance was defined as p!.05.
RESULTS: After the diagnosis of an LDH, professional athletes successfully returned to sport
82% of the time, with an average career length of 3.4 years. Of the 226 patients who underwent
surgical treatment, 184 successfully returned to play (81%), on average, for 3.3 years after surgery.
Survivorship analysis demonstrated that 62.3% of players were expected to remain active 2 years
after diagnosis. There were no statistically significant differences in outcome in the surgical and
nonoperative cohorts. Age at diagnosis was a negative predictor of career length after injury,
whereas games played before injury had a positive effect on outcome after injury. Major League
Baseball (MLB) players demonstrated a significantly higher RTP rate than those of other sports,
and conversely, National Football League (NFL) athletes had a lower RTP rate than players of other
sports (p!.05). However, the greatest positive treatment effect from surgery for LDH was seen in
NFL players, whereas for MLB athletes, a lumbar discectomy led to a shorter career compared with
the nonoperative cohort (p!.05).
FDA device/drug status: not applicable.
Author disclosures: WKH (speaking and/or teaching arrangements,
Stryker; trips/travel, Stryker; research support: staff and/or materials, Med-
tronic; grants, Pioneer Surgical); MAT (royalties, Smith and Nephew En-
doscopy; consulting, Smith and Nephew, Linvatec; speaking and/or
teaching arrangements, Magellan; scientific advisory board, Smith and
Nephew Endoscopy; research support: staff and/or materials, Arthrex;
grants, AOSSM); MFS (board of directors, AAOS Board of Directors;
other, DePuy, Synthesis, Zimmer, Orthohelix, Biomet, Arthrex).
* Corresponding author. Department of Orthopaedic Surgery, North-
western University Feinberg School of Medicine, 676 N St. Clair St, Ste
1350, Chicago, IL 60611, USA. Tel.: (312) 926-4471; fax: (312) 926-4643.
E-mail address: whsu@nmff.org (W.K. Hsu)
1529-9430/$ - see front matter ? 2011 Elsevier Inc. All rights reserved.
doi:10.1016/j.spinee.2010.12.009
The Spine Journal 11 (2011) 180–186

Page 2

CONCLUSIONS: Professional athletes diagnosed with an LDH successfully returned to play at
a high rate with productive careers after injury. Whereas older athletes have a shorter career length
after diagnosis of LDH, experienced players (high number of games played) demonstrate more
games played after treatment than inexperienced athletes. Notably, surgical treatment in baseball
players led to significantly shorter careers, whereas for NFL athletes, posttreatment careers were
longer than those of the corresponding nonoperative cohort. The explanation for this is likely mul-
tifactorial, including the age at diagnosis, respective contractual obligations, and different physical
demands imposed by each individual professional sport. ? 2011 Elsevier Inc. All rights reserved.
Keywords: Lumbar disc herniation; Professional athlete; Elite athlete; Clinical outcome; Discectomy
Introduction
Randomized controlled trials have reported excellent
clinical outcomes after surgical treatment for a lumbar disc
herniation (LDH) in the general population [1–6]. Many of
these studies also report high return-to-work and ‘‘satisfac-
tion’’ rates. Historically, the primary and secondary
outcome measures used to follow this condition are self-
reporting questionnaires, such as the Oswestry Disability
Index, Visual Analog Scale, Short Form-36, and leg and
back pain scores. For these patients, satisfaction is often
measured by successfully returning to a ‘‘desk job’’ or be-
ing able to perform activities of daily living without pain.
Asthegeneralpopulationstrivestostayactiveforalonger
period, patient-directed questions regarding treatment for an
LDH often focus on successful return to sporting activities,
suchasrunning,golf,ortennis.Forpatientswholeadanactive
lifestyle and participate in a host of sports, the peer-reviewed
literature falls short in providing this type of data [1,2]. It has
been opined that despite the enormous resources and effort
devotedtotheSpinePatientOutcomesResearchTrialstudies,
it remains unclear what the expected outcomes are for young
active individuals with an LDH after treatment [7].
Professional athletes, who may be predisposed to devel-
oping an LDH because of the unnatural forces imparted on
their spines on a daily basis, are concerned about different
parameters after treatment that are unique to their careers.
Instead of clinical data derived from self-reporting ques-
tionnaires, these patients are preoccupied with other mea-
sures, such as return-to-play (RTP) rates, career length,
and performance-based outcomes after surgical treatment.
These data would be instrumental in guiding the decision-
making process when faced with the dilemma between
operative and nonoperative care. Moreover, athletes of dif-
ferent sports may have substantially different outcomes
based on the respective biomechanical forces on the spine.
Because outcomes data after lumbar discectomy in pro-
fessional athletes are currently limited [8,9], studies that
demonstrate performance-based outcomes in this patient
population would be valuable to not only elite athletes
who are diagnosed with an LDH but also the general pop-
ulation who heed outcomes in these high-profile patients.
This study reports and compares the outcomes after LDH
of elite professional athletes of the four major North Amer-
ican professional sports.
Methods
Elite athletes of the four major professional sports
leagues—NationalFootball
Hockey League (NHL), Major League Baseball (MLB),
and National Basketball Association (NBA)—diagnosed
with an LDH were included in this study. Players with
the diagnosis of LDH were identified via previously re-
ported protocols used in multiple studies [8–10]. In short,
the sources of information included team injury reports,
player profiles, newspaper archives, press releases, and
team medical records. All sources used for player screening
were a part of the public record. Once a player’s criteria
were confirmed, personal information such as name and
team were subsequently removed from the database.
Inclusion criteria were professional athletes reported
with a ‘‘disc herniation’’ or ‘‘herniated disc’’ in the back
or lumbar spine, reported symptoms including lower ex-
tremity pain, on the active roster of a team from any of
the aforementioned leagues at the time of injury, had
a confirmed treatment plan corroborated by at least two
different independent sources, and were listed on at least
one weekly injury report. Players with conflicting medical
information from different sources, diagnosis of a ‘‘bulg-
ing disc,’’ ‘‘protruding disc,’’ ‘‘back strain,’’ ‘‘lumbar
strain,’’ ‘‘spinal cord injury,’’ or fracture were excluded.
Athletes diagnosed with a disc herniation at more than
one time point were also removed from the database. De-
finitive treatment of disc herniations was then recorded
based on the data available within the information sources.
In the operative cohort, only players treated with a lumbar
discectomy and/or laminotomy/foraminotomy were in-
cluded in this study. Any reports of lumbar fusion or lam-
inectomy were excluded. Nonoperative treatment was
defined as epidural steroid injections, physical therapy, ac-
tivity modification, and/or any treatment other than surgical
intervention. Demographic variables including player’s
age, height, weight, position, professional experience, type
of treatment, and date of treatment were collected for each
player.
League (NFL),National
Outcome measures
Three major statistical categories were used as per-
formance outcome measures. First, RTP rates were
181W.K. Hsu et al. / The Spine Journal 11 (2011) 180–186

Page 3

calculated based on successful return to the active roster
for at least one professional regular season game after
treatment. Second, the number of games played after
treatment for an LDH was compiled. Finally, the number
of years played before official retirement from each re-
spective player’s sport was recorded. Data for games
and years played professionally before treatment were
also collected.
Because the number of games in a typical regular sea-
son varied depending on the individual professional sport
(MLB, 160 games; NBA, 82 games; NHL, 82 games; and
NFL, 16 games), a conversion factor was used to normal-
ize the total number of games played for each player. In
addition, because the average career length of a particular
athlete also varied depending on the sport played [11–14],
an additional conversion factor was used in separate data
analyses to standardize outcome across sports. The aver-
age career lengths were identified as 5.6 years for MLB,
5.5 years for NHL, 4.8 years for NBA, and 3.5 years
for NFL.
Statistical analysis
Statistical analysis was performed using Statistical Anal-
ysis Software v. 9.1 (SAS Institute, Inc., Cary, NC, USA).
A Kaplan–Meier survivorship curve with ‘‘retirement from
sport’’ as the end point and 95% confidence intervals was
constructed. A Fisher exact test was used to analyze cate-
gorical data. The continuous variables of each cohort were
compared using a two-tailed Student t test for normally dis-
tributed data. Linear and mixed regression analyses and
Cox proportional hazards models were used to analyze ef-
fects of independent variables, such as age at diagnosis,
sport played, position played, body mass index, profes-
sional experience, and year of injury on performance end
points. Statistical significance was accepted with p!.05.
Results
A total of 342 professional athletes who were treated for
an LDH from 1972 to 2008 were identified with all four
sports well-represented (Table 1). Of these, 226 players
underwent a lumbar discectomy and 116 elected for nonop-
erative treatment. Athletes diagnosed with an LDH success-
fully returned to play 82% of the time, with an average
career length of 3.4 years after injury. There were no statis-
tically significant differences in RTP rates or career length
in the operative and nonoperative cohorts (81% for 3.3
years vs. 84% for 3.5 years, respectively) (Table 2).
Return-to-play rates varied depending on the sport
played (Fig. 1). Major League Baseball athletes demon-
strated a significantly greater rate of return than players
of all other sports, whereas NFL athletes returned to play
at a significantly lower rate (p!.05). Career length in terms
of games and years played after diagnosis of LDH was cal-
culated for all players based on sport. Although substantial
differences were seen in career lengths in different sports,
after these calculations were normalized for the number
of games in a typical regular season and the expected career
lengths for each sport, there were no significant differences
among groups (Table 3).
Regression analysis was performed to account for the ef-
fect of independent variables on career length. Age was
found to be a negative predictor of career length (games
Table 1
Professional athletes who were diagnosed with a lumbar disc herniation
and met the inclusion criteria for this study, stratified by sport played and
treatment arm
League NonoperativeOperativeTotal
NFL
NHL
MLB
NBA
Total
37
32
26
21
116
101
54
42
29
226
138
86
68
50
342
NFL, National Football League; NHL, National Hockey League; MLB,
Major League Baseball; NBA, National Basketball Association.
Context
The determination of a successful outcome for treatment
of any spinal disorder often needs to be individualized
given that patient desires/requirements following treat-
ment vary. While simple pain reduction is adequate
for some, return to challenging physical activities is re-
quired for others, such as competitive athletes. In this
study, the authors assessed outcomes specific to profes-
sional athletes following treatment for lumbar disc
herniations.
Contribution
Using return-to-play, career games played, and years
played as outcome measures, the authors found that
82% of athletes returned to play with an average career
length of 3.4 years following treatment for lumbar disc
herniation.
Implication
As in all case series, the information provided gives in-
sight into how this group of patients has done, but con-
founders place limits on what can be validly applied to
future patients regarding prognosis and what can be con-
cluded by comparing treatments. In the case of profes-
sional athletes, the list of potential confounders
includes additional variables such as extremely high
pre-injury fitness, unique psychosocial factors, physical
demands, financial circumstances, and social pressures
associated with U.S. professional sports.
—The Editors
182W.K. Hsu et al. / The Spine Journal 11 (2011) 180–186

Page 4

and years) after diagnosis of an LDH (p!.05). Notably,
however, a player’s professional experience (games played)
before diagnosis was a positive predictor for career length
(games and years) and RTP rate after treatment (p!.05).
The positions of infielder and outfielder in baseball were
a negative predictor in career length (games and years).
The positions of quarterback and punter were a positive
predictor for career length; however, this could likely be
explained by the longer expected career lengths of these
positions at baseline [15]. Other variables such as the date
of injury and type of treatment had no significant effect on
outcomes. With survivorship defined as official retirement
from the sport, Kaplan–Meier analysis demonstrated that
at the 2-year time point after treatment for an LDH,
62.3% of athletes were still expected to be on active rosters
in their respective professional leagues (Fig. 2).
Although the surgical and nonoperative cohorts demon-
strated no significant difference in outcomes for all players,
subgroup analysis revealed notable differences. In NFL ath-
letes, those who underwent a lumbar discectomy demon-
strated significantly longer careers than the nonoperative
group (p!.05) (Table 4). The greatest positive effect from
surgery for an LDH was seen in NFL players when com-
pared with all other sports. Athletes in the nonoperative
cohort were significantly older than those in the operative
group (p!.05).
On the other hand, in MLB athletes, although the RTP
rate was comparable for both operative and nonoperative
cohorts, those players who underwent a discectomy had
significantly shorter careers in both games and years played
than the nonoperative cohort after treatment (Table 4).
Kaplan–Meier analysis demonstrated similar survivorship
rates shortly after treatment in MLB athletes; however, over
time (250 games after treatment), the rate of retirement dif-
fered significantly (Fig. 3). Players in the nonoperative co-
hort were significantly younger than those in the operative
group (p!.05).
Discussion
Clinical outcomes after an LDH in the general popula-
tion are excellent. With a 90% satisfaction rate, 75%
return-to-work rate, and successful return to activities of
daily living, patients are generally happy with outcomes af-
ter a lumbar discectomy [1,3]. However, for those active
patients who have important questions about return to rig-
orous sporting activities, such as downhill skiing, running,
basketball, and other impact activities, the evidence-based
literature fails to effectively guide expectations after treat-
ment for an LDH. Moreover, for the subset of patients
who play sports on a competitive and professional level,
it is unknown what the outcomes are after either operative
or nonoperative treatment.
The data in this study suggest that professional athletes,
who have unique job demands distinct from the general
population, have excellent outcomes after the diagnosis of
an LDH. Of the four major North American sports (foot-
ball, basketball, baseball, and hockey), 82% of athletes in-
cluded in our study successfully returned to play at
a professional level for an average of 3.4 years after
Fig. 1. Return-to-play rates after diagnosis of a lumbar disc herniation in athletes of four major North American professional sports. Major League Baseball
(**) players returned to play at a significantly higher rate than athletes of all other sports (p!.05), whereas National Football League (*) athletes returned to
the field at a significantly lower rate than the average of all other players (p!.05). RTP, return-to-play; NFL, National Football League; NBA, National Bas-
ketball Association; NHL, National Hockey League; MLB, Major League Baseball.
Table 2
Performance outcomes in all professional athletes who either underwent
nonoperative or operative treatment after diagnosis with a lumbar disc
herniation
TreatmentTotalRTP RTP rate (%) Career length (y)
Nonoperative
Operative
Total
116
226
342
97
184
281
84
81
82
3.5
3.3
3.4
RTP, return-to-play.
Elite athletes returned to play at high rates while sustaining long ca-
reers after treatment for a lumbar disc herniation. Overall, there were no
significant differences in the nonoperative and operative cohorts.
183 W.K. Hsu et al. / The Spine Journal 11 (2011) 180–186

Page 5

LDH. Furthermore, for those athletes who required a lumbar
discectomy as a definitive treatment option, 81% returned
to play for an average of 3.3 years after surgery. There were
no significant differences in outcome in the operative and
nonoperative cohorts. But because this was a retrospective
study, it is important to note that the algorithm in which
players were chosen for surgery could not be standardized.
Career longevity after LDH was affected by a number of
variables. As anticipated, increasing age was a negative
predictor for career length after treatment for LDH, as older
athletes have a higher likelihood of retirement because of
declining athletic skill, concomitant injuries, or personal
reasons. Conversely, a players’ experience was a positive
predictor for clinical outcome. The more games an athlete
played before the diagnosis of LDH, the longer the career
after treatment. Thus, data in our study suggest that the
ideal outcome after LDH occurs in young, reliable, and pro-
ductive athlete, whereas the poor outcomes are seen in the
older athlete who plays fewer games for any number of rea-
sons, such as noteworthy injury history. Experienced
players may have additional skills that lead to a better out-
come after injury when compared with inexperienced ath-
letes. Furthermore, injury-prone athletes may have more
difficulty in rehabilitation after a similar injury in injury-
free players.
Although there were no significant differences in the op-
erative and nonoperative groups as a whole, Cox and
Kaplan–Meier analyses of the surgical cohort data demon-
strated variability in outcome depending on the sport
played. The greatest treatment effect occurred in NFL
players, where surgical treatment led to longer careers.
On the other hand, MLB players demonstrated shorter ca-
reers after surgery when compared with the nonoperative
cohort. There are a number of reasons that could explain
these findings. First, the relative age of each cohort could
have affected these outcomes, as the older cohort from each
group (ie, nonoperative in the NFL group and operative in
the MLB group) demonstrated shorter careers after treat-
ment (Table 4). Second, the different realities inherent to
each sport, such as the nature of each contract, could lead
to a different approach after injury. In MLB, contracts are
typically guaranteed for the duration of the agreement,
whereas in the NFL, any contract can be terminated if
a player is released from that particular team. This can lead
to different incentives or motivation that could affect the
clinical outcome after injury. Furthermore, the culture and
environment of each sport can also affect the behavioral re-
sponses to an LDH. Whereas NHL athletes have returned to
play minutes after a root canal and NFL quarterbacks after
Table 4
Subgroup analysis of MLB and NFL athletes who underwent operative and
nonoperative treatment for a lumbar disc herniation
League Operative Nonoperativep Value
NFL
RTP rate
Games
Years
Age (y)
0.78
36
3.1
27.5
0.59
20
2.0
29.4
.26
.002
.03
.02
MLB
RTP rate
Games
Years
Age (y)
0.96 0.971.0
.05
.18
.006
256 471
4.1
31.7
5.3
29.1
MLB, Major League Baseball; NFL, National Football League; RTP,
return-to-play.
For NFL athletes, patients treated with a lumbar discectomy played in
more games than the nonoperative cohort (p#.002). Conversely, MLB ath-
letes demonstrated a significantly shorter career after surgical treatment for
a lumbar disc herniation (p#.05). Notably, players in the NFL operative
and MLB nonoperative cohorts had a small but significantly younger
age than their counterparts (p#.02). No significant differences were dem-
onstrated in either cohort in the NHL and NBA athletes.
Fig. 2. Kaplan–Meier survivorship curve after surgical (red dashed line)
and nonoperative (blue solid line) treatment for a lumbar disc herniation
with end point defined as athlete’s retirement from sport. Each data point
represents the percentage of athletes who were expected to remain active
as a professional athlete normalized to the average career length of each
respective sport (years after Dx adjusted). The shaded bands represent
95% confidence intervals. Dx, diagnosis.
Table 3
Demographicdataandperformance-basedoutcomesinprofessionalathletes
after treatment for a lumbar disc herniation, stratified by sport played
League Age (y) BMI Games Years Games adjusted Years adjusted
NFL
NBA
NHL
MLB
27.9
28.0
29.2
30.7
32.1
24.6
26.4
26.2
29
219
135
321
2.8
3.7
3.1
4.6
465
498
269
321
4.4
4.4
3.1
4.6
BMI, body mass index; NFL, National Football League; NBA, Na-
tional Basketball Association; NHL, National Hockey League; MLB,
Major League Baseball.
Number of games and years played after treatment were normalized to
the total number of games in a typical regular season and the expected ca-
reer lengths for players of each individual sport (Games adjusted and Years
adjusted, respectively). The expected career lengths were used as follows:
5.6 years for MLB, 5.5 years for NHL, 4.8 years for NBA, and 3.5 years
for NFL [11–14]. There were no significant differences in adjusted career
length after treatment for lumbar disc herniation between sports (pO.05).
184W.K. Hsu et al. / The Spine Journal 11 (2011) 180–186