Returntoplayafter Lumbarspinesurgery: Ralph W. Cook,, Wellington K. Hsu
Returntoplayafter Lumbarspinesurgery: Ralph W. Cook,, Wellington K. Hsu
Returntoplayafter Lumbarspinesurgery: Ralph W. Cook,, Wellington K. Hsu
Lum bar Sp i n e Su r g er y
1
Ralph W. Cook, BS , Wellington K. Hsu, MD*
KEYWORDS
Athlete Return to play Lumbar spine injury Lumbar spine surgery
Performance outcomes
KEY POINTS
Surgical management of selected lumbar spine conditions can produce excellent outcomes
in athletes of all sports.
Microdiscectomy for lumbar disc herniation has been the most well-studied procedure
and leads to favorable outcomes in return to play rates and statistical performance
postoperatively.
Direct pars repair has led to high rates of return to play for a variety of fixation techniques.
There is a paucity of evidence-based return to play criteria, with the majority of literature
based on expert opinion and clinical experience.
INTRODUCTION
Low back pain (LBP) is one of the most common chief complaints encountered in
medicine, affecting 80% of the general population at some point in life. Athletes are
also commonly afflicted, with incidence rates approaching 30% over the course of
a career, accounting for one of the most common reasons for missed playing
time.1,2 In fact, 38% of professional tennis players reported missing at least 1 tourna-
ment owing to LBP at some point during their career.3 A survey of 272 competitive
adolescent athletes involved in 31 different sports found a point prevalence
(within the last 48 hours) of LBP of 14%, a 1-year prevalence of 57%, and a lifetime
Disclosures: Consulting - Stryker, Bacterin, Graftys, Globus, AONA, Synthes, Spinesmith, SI Bone,
Relievant, Ceramtec, Medtronic, Pioneer, Bioventus, LifeNet. Speaking and/or Teaching Arrange-
ments - AONA. Trips/travel - Stryker, Pioneer Surgical, Medtronic, Bioventus, AONA. Board of
Directors - Lumbar Spine Research Society, Cervical Spine Research Society. Scientific Advisory
Board - Bioventus (W.K. Hsu); none (R.W. Cook).
Source of Funding: There was no external source of funding for this project.
Department of Orthopaedic Surgery, Northwestern University Feinberg School of Medicine,
Northwestern University, 676 North Saint Clair Street, Suite 1350, Chicago, IL 60611, USA
1
Present address: 6 Buck Run, Mohnton, PA 19540.
* Corresponding author.
E-mail address: wkhsu@yahoo.com
prevalence of 66%.4 Additionally, a study examining the medical records of 4790 inter-
collegiate athletes competing in 17 varsity sports over a 10-year period revealed a
spine injury rate of 7 per 100 participants.5
Spine problems can be associated with participation in sports involving repetitive
hyperextension, flexion, rotation, and axial loading, such as gymnastics, wrestling,
football, diving, soccer, and dance.2,5–7 Overuse injuries have been found to be
more common than acute ones in a young athletic population.6 Furthermore, the diag-
noses given to athletes differ depending on an athlete’s age. One study compared
adolescent athletes with adults with acute LBP and demonstrated that 47% of adoles-
cents had stress fractures of the pars interarticularis, compared with only 5% of
adults.8 Conversely, discogenic back pain was diagnosed in 48% of adults compared
with only 11% of adolescents.
Initial treatment for a lumbar spine injury consists of conservative management,
including a brief period of rest and cessation of sporting activity for 1 to 2 days.9 Med-
ications may consist of nonsteroidal antiinflammatory drugs and muscle relaxants.
Physical therapy modalities such as ice, heat, compression, and massage may bring
additional pain relief. Certain injuries, such as lumbar disc herniation, may benefit from
lumbar epidural corticosteroid injections. Once pain has been controlled successfully,
activity may be resumed after a short course of flexibility and strengthening exercises.
Return to play, in general, should be considered when an athlete is pain free, has full
active range of motion with all activities, and has normal strength, endurance, and flex-
ibility.2,6,9,10 Although conservative management can often lead to pain relief in the
majority of patients, those who fail this treatment may require surgery.
Outcome measures used to judge success in the general population may not be
specific enough for professional athletes, who must return to play at a high level for
their livelihood. Validated patient reported outcome measures such as visual analog
scales, the Oswestry Disability Index, and the Short Form-36 may not be as applicable
to athletes, who are interested in returning to their preinjury level of performance and
on career longevity. More recently, clinical studies have focused on return to play and
sport-specific performance based outcome measures after treatment for lumbar spine
injury. Nonetheless, there is substantial variability in published return to play criteria,
which are almost exclusively derived from author’s expert opinion and experience.9,11
This article summarizes the current literature that defines return to play criteria for
various lumbar spine injuries.
For athletes who fail conservative management, lumbar discectomy provides symptom
relief and improved functional outcomes in the vast majority of patients.12 For example, in
14 division I athletes in the National Collegiate Athletic Association from 1988 to 1995,
90% of all athletes undergoing single level microdiscectomy returned to varsity sports,
with all athletes returning to at least recreational sporting activities.13 Recently, a system-
atic review that included 10 studies found that 75% to 100% of elite athletes return to
play after operative treatment for lumbar disc herniation.14 The recovery period after sur-
gery ranged from 2.8 to 8.7 months, with athletes’ postoperative careers ranging from 2.6
to 4.8 years. Notably, elite athletes attained an average of 64% to 104% of preoperative
baseline statistics, with variable performance based on sport. Similarly, a metaanalysis
performed by Overley and colleagues15 evaluated 9 studies representing 558 patients
with lumbar disc herniation who underwent lumbar microdiscectomy. The pooled clinical
success rate (defined as logging playing time in at least 1 regular season game or
Olympic-level event) for return to play after operative treatment was 83.5%.
Return to Play After Lumbar Spine Surgery 3
American Football
Although the physical demands from a collision sport such as American football
should be considered when treating athletes of this sport, in a study of 137 NFL
players from 1978 to 2008, Hsu17 demonstrated the potential beneficial effect of sur-
gery for lumbar disc herniation compared with conservative management. Return to
play rates were not different between operative and nonoperative groups (78% vs
59%, respectively); however, players treated operatively played in significantly more
games postoperatively (n 5 36) than those treated nonoperatively (n 5 20). There
were potential confounding variables; the nonoperative group was significantly older
than the operative cohort. Subsequently, a subgroup analysis evaluated outcomes
in offensive and defensive linemen—players generally considered to be at the greatest
risk for lumbar disc herniation.18 In this group of players that are constantly subjected
to bodily impact in a crouched/stooped position on every play, 81% of surgically
treated players return to play an average of 33 games over 3.0 years compared with
only 29% of nonoperative players for 5.1 games over 0.8 years. In operatively treated
players, 14% of linemen required a revision microdiscectomy, with 6 of 7 athletes
returning to play afterward, suggesting that this procedure is not a contraindication
for return to play in this population. In another subgroup analysis of offensive skill
position players (quarterback, running back, wide receiver, and tight end), 74% of ath-
letes return to play for an average of 36 games over 4.1 years, with no statistically sig-
nificant difference in performance between preinjury and postinjury statistics.19 These
studies suggest that surgical management for lumbar disc herniation in NFL athletes
have generally favorable return to play rates, with little impact on performance.
Baseball
The constant requirements of twisting and axial rotation of baseball athletes during
hitting and pitching may lead to different outcomes after lumbar surgery in this patient
population. In a biomechanical analysis of elite baseball players, the greatest amount
of force generated during batting occurred after ball contact and during pitching near
front foot contact.20 The authors suggested that these forces may lead to greater intra-
discal pressures, which may predispose these athletes to either virgin or recurrent
lumbar spine injuries. In a study of 69 lumbar disc herniation in 64 Major League
4 Cook & Hsu
Baseball athletes, 97% successfully return to play at an average of 6.6 months after
diagnosis.21 Athletes treated operatively required significantly more time to return to
play than those managed conservatively (8.7 vs 3.6 months, respectively).21 In
contrast with studies in other sports,19,22 pitchers and hitters demonstrated signifi-
cantly poorer performance in certain vital statistical categories postoperatively,
whereas the nonoperative cohort returned to play at preoperative performance levels
and with longer careers.16,21
Basketball
Anakwenze and colleagues12 evaluated 24 NBA athletes who underwent lumbar dis-
cectomy for lumbar disc herniation between 1991 and 2007. After surgery, 75% suc-
cessfully return to play, with a significant increase in blocked shots per 40 minutes and
a smaller decrease in rebounds per 40 minutes when compared with age and position
matched controls at 1 year after surgery. More recently, Minhas and colleagues22
evaluated 61 NBA players with lumbar disc herniation, demonstrating that return to
play rates did not differ between operatively and nonoperatively treated players
(78% vs 79%, respectively). Using a novel method for historical, comparable players
as controls, athletes’ careers with lumbar disc herniation were compared with similar
ones without this diagnosis. During the first postindex season, operatively treated ath-
letes played in significantly fewer games and had lower player efficiency ratings
compared with controls; however, no difference was seen at 2 and 3 years after sur-
gery or in postoperative career length. Conversely, whereas athletes managed
conservatively showed no difference at any time point in games played or player effi-
ciency ratings compared with controls, they did tend to play significantly fewer post-
index seasons.
Hockey
Schroeder and colleagues23 investigated 87 National Hockey League players with
lumbar disc herniation, of which 31 underwent nonoperative treatment, 48 received
a discectomy, and 8 were treated with a single level fusion. Return to play for all
players was 85% for an average of 136 games over 2.7 years. There was no difference
in return to play rates between those treated surgically (82%) and conservatively
(90%); however, all players had a significant decrease in performance measures after
lumbar disc herniation (games per season, points per game, and performance score)
with no difference between treatment groups. The lumbar fusion group returned to
play 100% of the time for an average of 203 games over 4 years, with no decrease
in performance measures. The authors warned that this difference could be attributed
to the small sample size of the fusion cohort; however, this limited evidence suggests
that lumbar fusion may be compatible with return to play in the National Hockey
League, which remains a relative contraindication in other collision sports. Overall,
this study intimates that National Hockey League athletes with lumbar disc herniation
can expect some decline in performance after injury.
These studies suggest excellent return to play rates and variable posttreatment per-
formance depending on the sport played; however, time to return is not always clear
and differences in methodology make comparing results difficult. Watkins and col-
leagues24 initially evaluated 60 cases of microdiscectomy in 59 Olympic and profes-
sional athletes with lumbar disc herniation between 1984 and 1998. They found that
90% return to play at an average of 5.2 months (range, 1–15). Watkins and associ-
ates25 later attempted to better define the timeline for return in 171 professional ath-
letes between 1996 to 2010 based on in-season eligibility criteria. They found that,
overall, 89% of surgically treated patients return to play at an average of 5.8 months.
Return to Play After Lumbar Spine Surgery 5
Spondylolysis and spondylolisthesis are among the most common causes of LBP in
adolescents, with rates as high as 47% reported in the literature.8 Participation in
sports involving repetitive hyperextension, rotation, axial loading, and torsion against
resistance predispose young athletes to stress fractures of the pars interarticula-
ris.2,7,29 Radiographic analysis of 100 young female gymnasts (average age 14 years)
engaged in high-level competition revealed a prevalence of spondylolysis in 11%.30
Furthermore, prevalence rates are particularly high among athletes engaged in diving
(43%), wrestling (30%), throwing sports (27%), weight lifting (23%), artistic gymnastics
(17%), and rowing (17%).29,31
Nonoperative treatment for acute spondylolysis, consisting of activity restriction
and bracing for up to 6 months, results in successful pain relief in more than 80% of
athletes independent of radiographic evidence of defect healing.2 When conservative
measures fail, direct surgical repair or posterolateral fusion can yield high rates of pain
relief and enable return to play. Direct pars repair may be more advantageous for
athletes and facilitate higher return to play rates, because it preserves spinal motion.9
Several techniques exist, including the Buck screw, Scott wiring technique, and
Morscher hook screw.
A systematic review encompassing 84 young amateur athletes (mean age 20 years)
found that most pars fractures occurred at the L5 vertebral level (96%), and that 84%
of operatively treated athletes return to play at their preinjury level of intensity over the
course of 5 to 12 months after direct pars repair.7 Of the 13 who were unable to return
to play, 7 were able to return to a less strenuous sport. A more recent comprehensive
review reporting on the surgical and conservative treatment of spondylolysis and
low-grade spondylolisthesis in competitive amateur athletes from 1973 to 2014 found
return to play rates of 88% and 85% for surgical and conservative treatment, respec-
tively, over 6 to 12 months.32 Of note, direct pars repair led to 80% to 100% return to
play rates in the studies analyzed, representing an excellent treatment option for
young athletes.
Table 1
Recommendations for RTP after discectomy and microdiscectomy
With the Buck screw technique in 25 competitive athletes, there was a 76% return to
play rate at a mean of 6 months (range, 3–10) after surgery.33 Similarly, of 16 adoles-
cent patients with 29 pars defects treated using Buck screw placement, 94% had
symptom resolution, with a 97% overall fusion rate.34 Of the 8 athletes in this cohort,
100% successfully returned to play. Finally, Debnath and colleagues35 evaluated 19
young athletes who underwent operative treatment for lumbar spondylolysis with
Buck screw fixation, reporting 95% return to play at a mean of 7 months (range, 4–10).
Using a Scott wire fixation technique, Nozawa and colleagues36 reported outcomes
in 20 athletes after surgery for spondylolysis. All of these athletes were able to return to
play, although to varying degrees after surgery, demonstrating that excellent results
can be obtained with this method of fixation. In 1 study of 43 adolescent athletes
treated with the Morscher hook screw construct for spondylolysis or grade I spondy-
lolisthesis, 100% were able to return to play after 4 months after surgery.37 Similarly, in
a study of 5 intercollegiate athletes with acute lumbar spondylolysis undergoing direct
pars repair with the hook screw construct, 100% return to play within 6 months after
diagnosis.38 A recent trend using novel, minimally invasive techniques to treat pars
fractures has become more prevalent.39 One such method involves placement of
pedicle screws connected to a radiolucent cord placed under tension through 2 bilat-
eral stab incisions 3 cm paramedian to the midline. In 1 study using this technique, 8
athletes of varying levels of competition with 16 pars defects underwent minimally
invasive surgery to repair the defect. Of these, 75% returned to play at their previous
level of competition after 6 months, and all patients reported overall mean improve-
ments in validated patient-reported outcome measures.
Evidence-based return to play criteria are also limited in the present literature for
spondylolysis and spondylolisthesis, with all available guidelines based exclusively
on expert opinion (Tables 2 and 3). Although authors may disagree on the timing
and possibility of return for certain sports, there is a general consensus that those
who return to play should be pain free with nearly normal strength, flexibility, and
endurance.2,10,26 Furthermore, those who undergo lumbar fusion should have radio-
graphic evidence of solid bony fusion; however, this is not necessarily as critical for
those requiring direct pars repair.2,10,26,36
There is also general agreement that before return to play, athletes should complete
a postoperative physical therapy and rehabilitation program.9 Radcliff and col-
leagues10 described a rehabilitation protocol for patients recovering from direct
pars repair. In it, they recommend a program of supervised core strengthening, flex-
ibility work for the extremities, and water exercises performed with a neutral spine
beginning at 2 weeks postoperatively. Nonimpact aerobic activities may commence
2 to 4 weeks postoperatively, with a neutral spine for up to the first 3 months. Gradual
impact and dynamic exercises were added at 3 months, with sport-specific training
Table 2
Recommendations for RTP after direct pars repair
Table 3
Recommendations for RTP after lumbar fusion
following at 4 to 6 months. The ultimate goal of this protocol is for return to play to con-
tact sports at 6 to 12 months after surgery. Similarly, Gillis and colleagues39 recom-
mended return to high impact activities at 6 months, and Nozawa and colleagues36
recommend that patients should be allowed to return to contact and collision sports
after surgery if they show evidence of rigid bony union on plain radiographic films or
computed tomography scan.
For posterolateral fusion for spondylolysis/spondylolisthesis, there are many sur-
geons who would allow eventual return to noncontact but not collision sports.40
Conversely, others have supported their patients return to contact sports after lumbar
fusion.26,28,32,40,41 Rubery and Bradford40 conducted a poll of 261 Scoliosis Research
Society members to determine return to play recommendation patterns for different
lumbar spine pathologies. The data suggested that a majority of surgeons would
recommend returning to contact sports at 1 year after lumbar fusion for both low-
and high-grade spondylolisthesis (56% and 51%, respectively). Interestingly, 12%
to 15% of respondents would advise against and 2% to 6% would forbid return to
contact sports in this population. In a separate study, Bouras and Korovessis32
allowed return to contact sports after lumbar fusion for spondylolisthesis within
1 year of surgery; however, they strictly forbade return to sports such as gymnastics,
football, rugby, wrestling, weight lifting, skydiving, and bungee jumping. In contrast,
Eck and Riley28 were more conservative, recommending against return to contact
sports.
Degenerative disc disease (DDD) is a common finding in athletes with LBP.2 Older ath-
letes are at an increased risk of developing back pain secondary to DDD with 48% of
patients given this diagnosis in one study.8 Although conservative management is
always the first line of treatment for this patient population, it is possible that chronic
pain recalcitrant to nonoperative modalities may be treated successfully with surgery
in selected athletes.2
Although there are no studies that evaluate lumbar fusion for DDD in an athletic pop-
ulation, recent literature has looked at outcomes of total disc replacement (TDR) in the
active competitive athlete and military population.42,43 One study evaluated 39 ath-
letes (average age 39.8 years) treated with TDR for DDD.42 These athletes return to
play 95% of the time to all sports, with subjective full recovery and peak fitness noted
8 Cook & Hsu
after 5.2 months. The authors returned their patients to all contact and/or extreme
sports without restrictions. In 8% of patients, persistent LBP limited physical activity.
Another study compared TDR with anterior lumbar interbody fusion in 24 active duty
military personnel (average age 37.3 years).43 TDR patients returned to full active duty
83% (10/12) of the time, at an average of 22.6 weeks, whereas anterior lumbar inter-
body fusion patients returned 67% (8/12) of the time at an average of 32.4 weeks post-
operatively (P 5 .156). The authors note that return-to-action criteria may have been
affected by confounding factors, such as radiographic criteria after fusion. These
studies suggest that TDR may be a viable alternative to lumbar fusion for the treatment
of DDD in this population.
At best, there is level V evidence to guide return to play recommendations after sur-
gical treatment for lumbar DDD (see Table 3; Tables 4 and 5). Many authors suggest
that, before considering return to play, an athlete should have radiographic evidence
of a solid fusion, resolution of preoperative pain, and restoration of strength, flexibility,
and endurance.2 A survey of North American Spine Society members revealed that a
majority of respondents would recommend a return to golf no sooner than 6 months
after lumbar spine fusion.41 Although many respondents allowed significantly shorter
recovery times for professional compared with noncompetitive golfers, they stressed
the importance of fusion healing and evidence of radiographic stability. Siepe and col-
leagues42 recommended athlete participation in noncontact sports within 3 months of
TDR, and that contact and even extreme sports may be resumed after 4 to 6 months.
Tumialan and colleagues43 support nonimpact training starting at 3 months, with no
activity limitations after 6 months after TDR surgery. After a laminectomy alone for spi-
nal stenosis, Abla and colleagues41 suggest that athletes may resume golf after 4 to
8 weeks, whereas 2 other studies suggest that participation in contact sports may
commence 4 to 6 months postoperatively.11,28
Adolescent idiopathic scoliosis can cause LBP in the athletic population, and in certain
situations, surgery is indicated to halt the progression of the curvature. One study to
date has evaluated return to play outcomes in 42 athletically active adolescents with
scoliosis who underwent posterior spinal fusion.44 The authors report that 60% of chil-
dren returned to play contact and noncontact sports at an equal or higher level of
activity postoperatively. There was a significant relationship between the distal level
of fusion and the rate of return to play—the lower the level of instrumentation from
T11 to L4, the lower the return to play rate, with calculated odds ratios indicating
that for each level fused distally, patients were 36.7% less likely to return to play at
or above the same level of preoperative activity (P 5 .039). Furthermore, higher Lenke
Table 4
Recommendations for RTP after laminectomy
Table 5
Recommendations for RTP after lumbar total disc replacement
classification curve types and lower final Scoliosis Research Society-22 scores were
also negative predictors of return to play rates. The data in this study suggests that
fusion for adolescent idiopathic scoliosis may have negative consequences on return
to play for young athletes, which also depends on the extent of the surgery.
As far as time to return to play is concerned, some have recommended return to
play at 4 months after surgery if patients were pain free with radiographic evidence
that the implants and curve correction remained unchanged44 (see Table 3). In this
particular study, clearance was granted at an average of 7.4 months postoperatively.
A Scoliosis Research Society survey study reported that a majority of respondents
supported return to noncontact sports at 6 months after fusion for adolescent idio-
pathic scoliosis, with 61% of respondents supporting return to contact sports at
1 year.40 Interestingly, collision sports were allowed by 32% of respondents, with
the vast majority advising against (36%) or forbidding (24%) them.
SUMMARY
REFERENCES
1. Dreisinger TE, Nelson B. Management of back pain in athletes. Sports Med 1996;
21(4):313–20.
2. Bono CM. Low-back pain in athletes. J Bone Joint Surg Am 2004;86A(2):382–96.
3. Hainline B. Low back injury. Clin Sports Med 1995;14(1):241–65.
4. Schmidt CP, Zwingenberger S, Walther A, et al. Prevalence of low back pain in
adolescent athletes - an epidemiological investigation. Int J Sports Med 2014;
35(8):684–9.
5. Keene JS, Albert MJ, Springer SL, et al. Back injuries in college athletes. J Spinal
Disord 1989;2(3):190–5.
6. Purcell L, Micheli L. Low back pain in young athletes. Sports Health 2009;1(3):
212–22.
10 Cook & Hsu
28. Eck JC, Riley LH 3rd. Return to play after lumbar spine conditions and surgeries.
Clin Sports Med 2004;23(3):367–79, viii.
29. Soler T, Calderon C. The prevalence of spondylolysis in the Spanish elite athlete.
Am J Sports Med 2000;28(1):57–62.
30. Jackson DW, Wiltse LL, Cirincoine RJ. Spondylolysis in the female gymnast. Clin
Orthop Relat Res 1976;(117):68–73.
31. Rossi F, Dragoni S. Lumbar spondylolysis: occurrence in competitive athletes.
Updated achievements in a series of 390 cases. J Sports Med Phys Fitness
1990;30(4):450–2.
32. Bouras T, Korovessis P. Management of spondylolysis and low-grade spondylo-
listhesis in fine athletes. A comprehensive review. Eur J Orthop Surg Traumatol
2015;25(Suppl 1):S167–75 [A systematic review or a meta-analysis].
33. Menga EN, Kebaish KM, Jain A, et al. Clinical results and functional outcomes
after direct intralaminar screw repair of spondylolysis. Spine 2014;39(1):104–10.
34. Snyder LA, Shufflebarger H, O’Brien MF, et al. Spondylolysis outcomes in adoles-
cents after direct screw repair of the pars interarticularis. J Neurosurg Spine
2014;21(3):329–33.
35. Debnath UK, Freeman BJ, Gregory P, et al. Clinical outcome and return to sport
after the surgical treatment of spondylolysis in young athletes. J Bone Joint Surg
Br 2003;85(2):244–9.
36. Nozawa S, Shimizu K, Miyamoto K, et al. Repair of pars interarticularis defect by
segmental wire fixation in young athletes with spondylolysis. Am J Sports Med
2003;31(3):359–64.
37. Durrani AA, Desai R, Chavanne A, et al. 125. Outcome of direct PARS repair in
competitive athletes. Spine J 2008;8(Suppl 5):64S.
38. Sutton JH, Guin PD, Theiss SM. Acute lumbar spondylolysis in intercollegiate
athletes. J Spinal Disord Tech 2012;25(8):422–5.
39. Gillis CC, Eichholz K, Thoman WJ, et al. A minimally invasive approach to defects
of the pars interarticularis: Restoring function in competitive athletes. Clin Neurol
Neurosurg 2015;139:29–34.
40. Rubery PT, Bradford DS. Athletic activity after spine surgery in children and
adolescents: results of a survey. Spine 2002;27(4):423–7.
41. Abla AA, Maroon JC, Lochhead R, et al. Return to golf after spine surgery.
J Neurosurg Spine 2011;14(1):23–30.
42. Siepe CJ, Wiechert K, Khattab MF, et al. Total lumbar disc replacement in ath-
letes: clinical results, return to sport and athletic performance. Eur Spine J
2007;16(7):1001–13.
43. Tumialan LM, Ponton RP, Garvin A, et al. Arthroplasty in the military: a preliminary
experience with ProDisc-C and ProDisc-L. Neurosurg Focus 2010;28(5):E18.
44. Fabricant PD, Admoni S, Green DW, et al. Return to athletic activity after posterior
spinal fusion for adolescent idiopathic scoliosis: analysis of independent predic-
tors. J Pediatr Orthop 2012;32(3):259–65.