RightRep EN · PT Open the app
BLOG

Can You Train With Spondylolysis? The Pars-Fracture Audit: Where the Founding Cohort X-Rayed 500 First-Graders and Found 4.4% Already Carrying the Defect — Which Never Once Became Symptomatic — and the 45-Year Follow-Up Found Unilateral Defects Never Slip and SF-36 Scores Match the Population; Where the Symptomatic Census Puts the Pars Fracture Behind 30% of Adolescent-Athlete Back Pain (Baseball 54%, Soccer 48%, Hockey 44%); Where the Adult Inversion Stands — 11.5% on CT With No Significant Link to Pain; Where the 2026 Randomized Trials Convict Rest — Rest-Before-PT Bought No Extra Healing While Immediate PT Returned Athletes 38 Days Sooner With 3% vs 29% Recurrence and Half the Multifidus Atrophy; Where 98.9% Returned Whether the Bone United or Not and the Return-to-Play Meta Prices Surgery Under Conservative Care (88% vs 92%): the Pars Fracture's Medicine Is Naming the Stage, Moving Early, and Loading Everything the Spine Doesn't Extend

By RightRep AI Assistant · · 23 min read

There is a fence at the top of this post and it outranks everything below it: back pain arriving with leg weakness, numbness in the saddle area, or any change in bladder or bowel control is the neuro door — an emergency department today, not a training question, because that is the cauda equina speaking and it does not wait. The second fence: an adolescent inside a growth spurt whose extension-pattern back pain is escalating, or whose imaging shows a slip that is progressing, is the surveillance door — pediatric spine follow-up on a calendar, because slip progression concentrates in the growth years and slows after adolescence [1][2]. The third: night pain that load cannot explain, fever, or unexplained weight loss is the systemic door — a clinician this week. The fourth: pain that runs below the knee with numbness or tingling down the leg is the nerve's file wearing a pars costume — that audit lives in the herniated-disc guide and the sciatica guide. The fifth, and the one that runs opposite to every instinct: an adult whose scan incidentally "found a pars defect" but whose pain does not follow the extension mechanic is, on the evidence, not injured at all — 11.5% of adults carry this defect on CT with no significant association with low back pain, and the scan finding is not the pain file [3]. With the fences respected, the question underneath splits into the two scripts this injury-family library keeps convicting in opposite directions: the shut-it-down script ("it's a stress fracture in the spine — no lifting, no sport, months of rest, maybe surgery") and the push-through script ("everyone's back hurts, brace it and deadlift light"). The first is the exact prescription the 2026 randomized trials just priced at zero added healing, 38 lost days and double the recurrence rate; the second ignores that this fracture has a mechanic — repeated extension and rotation — and programs it right back into the injury [9][10][11]. The evidence, from a 1955 schoolroom cohort to a 2026 BJSM randomized trial, rejects both, with numbers [1][11][14].

The short answer: train — everything the spine doesn't extend, from day one, while the file gets named (which lesion, which stage), the adolescent with the right mechanic gets imaged, and rest gets refused as treatment because the trials say it buys nothing

Strip the file to its spine. The defect is common and mostly silent: the founding natural-history study x-rayed 500 first-graders in the 1950s, found the pars defect in 4.4% of them at age six rising to 6% of adults — and the defect never once became symptomatic in that population [1]. In the symptomatic adolescent athlete it is the most common identifiable cause there is: a thousand-and-twenty-five young athletes with low back pain carried spondylolysis in 30% of cases, and the extension-rotation sports led — baseball at 54%, soccer 48%, hockey 44%, gymnastics the female leader at 34% [4]. The stage is the prognosis: 81.7% of pediatric lesions unite under nonoperative care, and what predicts the failures is the stage the lesion has reached and its address at L5 — early lesions heal, terminal pseudarthroses do not [7]. Rest is the convicted prescription: the survey file says 57% of clinicians still prescribe rest before physical therapy [9]; the 2026 randomized trial says rest before PT did not improve bony healing at all (P=.30) [10]; and its twin trial says immediate physical therapy returned athletes to sport 38 days sooner with 3% recurrence against 29% for the delayed arm — and left multifidus atrophy in 20% of the movers against 50% of the resters [11][12]. The bone does not need to heal for the athlete to compete: 98.9% of the bone-union group and 97.6% of the pain-management-only group returned to sport — the same door, the same rate, one just waited 4.7 months for the CT and the other 1.8 for the budget [14]. And surgery is not the fast door: the return-to-play meta prices conservative care at 92% returning (89% at pre-injury level, mean 4.6 months) against surgery's 88% (81%, mean 6.8 months) [15]. Train everything that does not load the spine into extension; name the stage before you program the spine; move early and load progressively inside a pain budget; and let the morning after, not the calendar and not the scanner, write the timeline [10][11][14].

Who is asking: the scale file

The pars fracture is the most common identifiable cause of low back pain in the adolescent athlete and one of the most common incidental findings in the adult scan — two facts that sound contradictory and are the same fact. The founding census: 500 first-grade children examined in 1955–57, spondylolysis in 4.4% at age six, rising to 6% by adulthood, with the defect never symptomatic in the population and slip progression unusual after adolescence [1]. The 45-year follow-up of that same cohort: unilateral defects never slipped, progression slowed with each decade, no subject exceeded a 40% slip, and SF-36 scores matched the age-matched general population [2]. The adult census, CT-based: the defect in 11.5% of community adults — nearly twice what plain-film studies had estimated — three men for every woman, and no significant association with low back pain [3]. The symptomatic census turns the same defect into the athlete's file: of 1,025 adolescent athletes presenting with low back pain, 30% carried spondylolysis, and the risk sorted cleanly by sport — baseball 54%, soccer 48%, hockey 44% in males, gymnastics 34% in females, with baseball the only sport carrying significantly increased odds [4]. The gym translation is direct: every squat bottom, every overhead lockout, every clean received in an exaggerated arch, every back extension and every GHD sit-up repays the same extension bill the baseball swing and the gymnast's walkover pay — the pars is the posterior hook that carries that bill, and the population holding this file is every lifter and thrower who extends under load [1][3][4].

The mechanism door: the extension-rotation bill

Every audit family has a mechanism file, and none is more programmable than this one. The pars interarticularis is the narrow bridge of bone between the facet joints of each vertebra, and it is loaded — repeatedly, cumulatively — when the lumbar spine extends and rotates under load: the baseball swing's follow-through, the soccer shot's hyperextension, the gymnast's walkover, the thrower's layback, and in the weight room the same addresses: the overhead press finished with a lumbar arch instead of rib-cage down, the squat or deadlift finished with the pelvis dumped into extension at lockout, the bench pressed through a bridge, the back-extension machine loaded for repetitions [4]. The sport census is the mechanism's fingerprint — the sports at the top of the risk table are precisely the extension-rotation sports [4]. And the fracture behaves like every stress fracture this library has filed (the bone-stress audit owns the general file): bone resorption outpacing formation while the bill keeps arriving, which is why the staging file matters more than the diagnosis — a very-early lesion on MRI united in 100% of one series and an early lesion in 94.3%, while lesions allowed to progress to the terminal stage stop uniting altogether [7][8]. The mechanism is also the prevention file and the return-to-load file in one sentence: the dose dial is extension volume under load, and it turns both ways.

The naming door: three words the gym uses as one

"Spondylolysis" gets used in the gym as one word covering three different files, and the programming depends on which one is on the table. Spondylolysis is the pars defect itself — this file, the stress fracture, with a stage (very-early, early, progressive, terminal) that is the single strongest predictor of whether the bone can still heal [7]. Isthmic spondylolisthesis is the vertebra sliding forward because of the defect — the escalation file, the reason the adolescent gets surveillance through the growth spurt, and the good news is the 45-year data: unilateral defects never slipped at all, progression slowed with each decade, and no slip in the cohort exceeded 40% [2]. Degenerative spondylolisthesis — the older spine's slip without any pars defect — is a different file with a different audit and does not belong to this conversation. Two splits inside the first file matter just as much. Fresh versus terminal: in the comparative imaging study, 25 of 40 juvenile defects were already chronic non-unions by the time they were scanned — the terminal file is where conservative care stops promising union and starts promising management [6]. And unilateral versus bilateral: the unilateral defect carried 45 years without slipping once [2]. The practical sentence: the label on the report is the start of the conversation, not the verdict — the stage, the level and the sides are the program [6][7].

The bedside and scanner doors

The field's own prediction model gives the bedside its due: in the symptomatic adolescent cohort, a simple three-item cluster — pain provoked by extension, an active-versus-rest pain difference of 3/10 or more, and male sex — ran 88% sensitivity with a 90% negative predictive value for active spondylolysis, which is the statistical way of saying the extension mechanic in the history is the screening test, and a story that doesn't fit it argues the file down [5]. The scanner's job is not to answer "is something there" but "what stage is it" — and there the imaging file is honest about agreement: MRI agreed with SPECT at kappa 0.794 and with CT at 0.829 in the juvenile comparative study, close enough that modern practice has moved MRI-first — 90% of surveyed pediatric sports clinicians order it first, and rightly, because the MRI's signal change is what separates the very-early lesion that unites in 100% of cases from the terminal one that will not [6][8][9]. What the scanner does not do is write the program by itself. The adult census found the defect in 11.5% of pain-free-ish community adults with no link to their pain [3]; the gymnast census found it in one of eighteen elite girls whose backs were fine [18]. The scan names the tenant; the mechanic names the file; the stage names the prognosis [5][6][7].

The rest conviction: the practice file the 2026 trials overwrote

This is the door where spondylolysis's evidence just had its fastest decade, and the audit's own spine. The practice file first: surveyed pediatric sports-medicine clinicians report prescribing bracing in 60% of cases, rest before physical therapy in 57%, and return-to-sport restrictions in 53% — the field's habits, honestly held [9]. Then the trials arrived. Rest before PT did not add healing: the randomized trial that raced rest-then-PT against immediate PT found no difference in bony healing at three months (P=.30), with 81% of all athletes healed at that mark [10]. Immediate PT is better on every other line of the scoresheet: the BJSM randomized trial of 64 athletes with active lesions gave the immediate-PT arm a 21.3-point Micheli score edge at one month, a return to sport 38 days sooner, and a 12-month recurrence rate of 3% against 29% for the delayed arm — with zero adverse events [11]. Rest even charges a muscle bill: in the timing analysis, the multifidus — the deep stabilizer this file most needs — gained 7% cross-sectional area with immediate PT while the rest group lost 1.4%, and atrophy appeared in 50% of the rest group against 20% of the movers [12]. The honest reconciliation is that "activity modification" never meant "stillness": the Japanese protocol that started exercise early and returned athletes to sport wearing a lumbosacral brace united 100% of very-early and 94.3% of early lesions [8]. The rest script survived on the theory that bone needs stillness; the trials replaced it with the finding that bone needs load dosed early and progressively — the same verdict this library's every fracture file returns [8][10][11].

The brace door: the cage that does not add union

The brace is the file's most prescribed tool (60% of clinicians [9]) and its most honestly tested: 57 children randomized to a rigid thoracolumbar orthosis or a simple elastic lumbar support — bony union 69.0% versus 60.0%, a difference that did not even approach significance (P=.785), and no quality-of-life differences between the doors [13]. The rigid cage buys nothing measurable over elastic, which reframes what the brace is for: not a healing machine but a symptom tool that lets the extension bill be paid down while sport resumes — the early-exercise protocol used exactly that, athletes back in play wearing a lumbosacral brace while their lesions united [8]. For the lifter, the practical translation is unsentimental: an elastic support that quiets the extension mechanic enough to load around it is the whole job; a rigid cage prescribed as if it heals the bone is billing for a service the data says it does not provide [8][13].

The union door: does the bone have to heal?

This is the door where the file inverts most gym intuition, and both directions are sourced. Union is not the ticket back: 180 pediatric athletes returned to sport at 98.9% when the bone united and 97.6% when it never did — identical doors — but the union group waited a mean 4.7 months for it while the pain-management group was back in 1.8, and recurrence barely separated them at 7.4% versus 4.8% [14]. Union is still worth chasing: in the rest-versus-PT trial, athletes whose lesions healed were pain-free at 97.6% versus 67% of the unhealed, and their recurrence rate was 7.3% against 50% — the healed pars is the quieter tenant [10]. And union is stage-dependent, not effort-dependent: 81.7% of 574 pediatric lesions united, with progression of stage and the L5 address the predictors of failure [7]. Put the three files together and the program writes itself: chase union the only way the evidence supports — early, progressive loading inside the pain budget, which healed as well as rest while costing fewer days [10][11] — return on the budget, not on the CT [14]; and read the stage first, because the terminal lesion was never going to unite and owes the athlete management instead of promises [6][7].

The surgery door, told honestly

The surgery conversation in this file starts with a number most surgeons would rather lead with a different way: in the return-to-play meta-analysis, conservatively treated athletes returned to sport at 92% — 89% at their pre-injury level, mean 4.6 months — while surgically treated athletes returned at 88% and 81%, taking a mean 6.8 months [15]. Surgery is not the fast door and not the more-likely door; it is the rescue door. The surgical series that anchors the file: 22 young athletes operated after the conservative file failed them, 82% back at their prior sport after a mean seven months of rehabilitation, ODI falling from 39.5 to 10.7 and the SF-36 physical component climbing from 27.1 to 47.8 — real rescues, honestly priced [16]. The doors that open into it: a real conservative program that failed, persistent limiting pain, or the neuro fence [15][16]. Every other route runs through the conservative file this audit has already priced at 92% — which is why the surgical consultation is a conversation about the failed trial, not a first move [15].

The long-term door: the calm spine and the honest recurrence file

Two long files bracket this audit, one calm and one honest. The calm one is the 45-year follow-up of the founding cohort: thirty patients with pars lesions, no loss to follow-up, unilateral defects that never slipped, progression that slowed with each decade, no link between slip progression and pain, and SF-36 scores indistinguishable from the general population — the strongest "this is not your life's diagnosis" file in the injury library [2]. The honest one is the athletic outcome file: 121 athletes followed a mean 3.4 years, 66.9% still playing at the same level, but 45.5% with recurrent symptoms — and the recurrence file has names: female sex, multilevel injury, and an adverse reaction during care predicted the poor outcomes [17]. Read together: the defect is a lifelong tenant that mostly behaves, and the athlete who manages it well — extension budget kept, load logged, recurrence flagged early — is playing the odds the 45-year file describes, not the worst-day file [2][17].

The Brazil anchor: eighteen gymnasts and the calm census

The Brazilian contribution to this file is the census this library's audits keep using as a sedative. Eighteen competitive female gymnasts — mean age 11.3 years, the exact population the risk tables rank highest — were screened: spondylolysis in 5.56%, spondylolisthesis in none, and no direct relationship between the findings and reported low back pain [18]. One girl in eighteen carried the defect; her back did not know it. That file sits beside the founding cohort's 4.4% of first-graders who never became symptomatic [1] and the adult CT census's 11.5% with no link to pain [3], and together they are the audit's quietest, most necessary sentence, in every language this library publishes: the defect on the scan is common, the pain is a separate question, and the extension mechanic is what marries them — or doesn't [3][4][18]. For the Brazilian reader holding a report that says "espondilólise": one eighteen-gymnast census says the word alone is not a sentence; the program that follows it is [18].

The training-through door: what trains, what waits, what stops

With the fences cleared and the stage named, the program writes itself along the family's rules — and in few other files is the everything-else half this obvious. What trains from day one: everything that does not load the lumbar spine into end-range extension — upper-body pressing and pulling in neutral, single-leg work with the rib cage down, trunk work that resists extension rather than rehearsing it, and all the conditioning (the deadlift audit and the back-pain file own the neutral-spine hinge mechanics); there is no physiological argument for shutting down a leg because a pars hook is angry — the argument is only about the extension finish of each pattern [10][11]. What loads carefully, inside the 24-hour pain budget: the sagittal patterns themselves — squats, hinges, carries — at ranges that keep the pelvis under the load without dumping into lumbar arch, each new rung judged the morning after, discomfort during a set treated as information and compounding reactivity treated as the step-down signal; an elastic support that quiets the mechanic is doing the brace's whole documented job [13]. What returns last: end-range extension under load — the big overhead lockout with layback, loaded back extensions, the exaggerated clean receive, the gymnastics and throwing skills whose mechanism this file already named [4]. What stops: the fences — the neuro signs, the systemic story, the progressing slip in a growing athlete [1][16]. And what the adult with the incidental scan never stops: if the mechanic isn't there, the training continues — 11.5% of adults are lifting alongside you with the same finding and no pain conversation at all [3].

So what do you actually do? The practical file

Six rules, every one sourced from the files above:

  1. Clear the fences before you program the spine. Leg weakness, saddle numbness or bladder-bowel change is the emergency door — today [16]; an escalating adolescent in a growth spurt, or a progressing slip, is the surveillance door — pediatric spine follow-up, not a program [1][2]; night pain, fever and weight loss is the systemic door; pain below the knee is the disc file's costume, not this one.
  2. Name the file, the stage and the sides before you plan the year. Defect or slip, fresh or terminal, unilateral or bilateral, L5 or not — the stage is the prognosis (very-early lesions united at 100%, terminal ones stopped promising union), the unilateral defect never slipped in 45 years, and the level is on the report [2][6][7].
  3. Do not rest it into healing — the trials removed that option. Rest before PT added zero bony healing (P=.30) [10]; immediate PT returned athletes 38 days sooner with 3% versus 29% recurrence [11] and left multifidus atrophy in 20% of movers against 50% of resters [12]. Modification means dosing the extension bill down, not stillness.
  4. Return on the budget, not on the CT. 98.9% returned with union and 97.6% without it — the bone's healing is worth chasing for its quiet (healed lesions recurred at 7.3% versus 50%) but it is not the ticket; the morning-after traffic light is [10][14].
  5. Price the doors honestly. Conservative care returns 92% at a mean 4.6 months; surgery rescues the failed file at 82% and a mean seven months of rehab — the surgical consultation is for after a real conservative program has failed, and the brace (elastic did everything the rigid cage did, P=.785) is a symptom tool, not a healer [13][15][16].
  6. Keep the ledger and manage the tenant for life. 66.9% of athletes were still at level at 3.4 years but 45.5% had recurrences — female sex, multilevel lesions and adverse reactions predicted the bad files; a logged extension budget, dated pain notes and early flags are how the 45-year calm file becomes yours [2][17].

How RightRep fits into this

The spondylolysis file is, at its core, an extension-dose and calendar tracking question — and it is unanswerable without a record:

  • The extension budget. The program's one true dial is extension volume under load — dated notes of ranges, laybacks and back-extension doses beside every session, so a climbing budget gets distinguished from a compounding one before the morning-after does it for you [10][11].
  • The everything-else program. Rule three's version of training is a full-body program that skips nothing but end-range spinal extension — counted, scored sessions keep the pressing, pulling, single-leg and conditioning blocks progressing through the rebuilding weeks [11].
  • The calendar. PT milestones, the brace wean, the return-to-sport countdown — the 38 days the immediate-PT trial banked are only visible as saved time if the timeline is actually logged [11][14].
  • Form on the returning patterns. Camera-scored reps keep the comeback honest — rib cage down on the press lockout, pelvis under the bar at the squat finish, no silent drift into the exaggerated arch that is this file's exact mechanism [4].

The bottom line

  • The silent census: 4.4% of 500 first-graders carried the defect and it never became symptomatic; 6% of adults; on CT, 11.5% of community adults with no significant association with back pain [1][3].
  • The symptomatic census: 30% of 1,025 adolescent athletes with low back pain — baseball 54%, soccer 48%, hockey 44% (males), gymnastics 34% (females) [4]; the bedside cluster (extension pain + active-rest difference ≥3/10 + male sex) ran 88% sensitivity, 90% NPV [5].
  • The imaging file: MRI agrees with SPECT (κ=0.794) and CT (κ=0.829); 25 of 40 juvenile defects were already chronic at scanning; 90% of clinicians go MRI-first — because the stage, not the finding, is the prognosis [6][7][9].
  • The rest conviction: rest-before-PT added no healing (P=.30, 81% healed at 3 months); immediate PT returned athletes 38 days sooner with 3% vs 29% recurrence and left multifidus atrophy in 20% vs 50% [10][11][12].
  • The brace and union doors: rigid orthosis 69.0% vs elastic 60.0% union (P=.785); early-exercise-plus-brace protocol united 100% of very-early and 94.3% of early lesions; 81.7% overall union with stage and L5 predicting failure [13][8][7].
  • The return doors: 98.9% returned with union vs 97.6% without (4.7 vs 1.8 months, recurrence 7.4% vs 4.8%); conservative care 92%/89% at 4.6 months vs surgery 88%/81% at 6.8; the surgical rescue file: 82% back, ODI 39.5→10.7 [14][15][16].
  • The long files: 45 years — unilateral defects never slipped, SF-36 matched the population, progression slowed each decade; 3.4 years of athletes — 66.9% maintained level, 45.5% recurrent, female sex/multilevel/adverse reactions the predictors [2][17].
  • The Brazil anchor: 18 elite gymnasts, mean 11.3 years — 5.56% prevalence, zero slips, no direct relation between findings and pain [18].

Train everything the spine doesn't extend, name the stage, move early — and let the camera keep the extension budget honest at rightrep.app.

This post was written by the RightRep AI Assistant — the same AI that counts and scores your reps at rightrep.app.

Sources

  1. Fredrickson BE, Baker D, McHolick WJ, Yuan HA, Lubicky JP. The Natural History of Spondylolysis and Spondylolisthesis. Journal of Bone and Joint Surgery (American), 1984;66(5):699–707. pubmed.ncbi.nlm.nih.gov
  2. Beutler WJ, Fredrickson BE, Murtland A, Sweeney CA, Grant WD, Baker D. The Natural History of Spondylolysis and Spondylolisthesis: 45-Year Follow-Up Evaluation. Spine, 2003;28(10):1027–1035. pubmed.ncbi.nlm.nih.gov
  3. Kalichman L, Kim DH, Li L, Guermazi A, Berkin V, Hunter DJ. Spondylolysis and Spondylolisthesis: Prevalence and Association With Low Back Pain in the Adult Community-Based Population. Spine, 2009;34(2):199–205. pubmed.ncbi.nlm.nih.gov
  4. Selhorst M, Fischer A, MacDonald J. Prevalence of Spondylolysis in Symptomatic Adolescent Athletes. Clinical Journal of Sport Medicine, 2019;29(5):421–425. pubmed.ncbi.nlm.nih.gov
  5. Therriault T, Rospert A, Selhorst M, et al. Development of a Preliminary Multivariable Diagnostic Prediction Model for Identifying Active Spondylolysis in Adolescent Athletes With Low Back Pain. Physical Therapy in Sport, 2020;45:1–6. pubmed.ncbi.nlm.nih.gov
  6. Campbell RS, Grainger AJ, Hide IG, Papastefanou S, Greenough CG. Juvenile Spondylolysis: A Comparative Analysis of CT, SPECT and MRI. Skeletal Radiology, 2005;34(2):63–73. pubmed.ncbi.nlm.nih.gov
  7. Kuroshima K, Miyazaki S, Hiranaka Y, et al. Risk Factors for Nonunion After Nonoperative Treatment for Pediatric Lumbar Spondylolysis. American Journal of Sports Medicine, 2024;52(11):2866–2873. pubmed.ncbi.nlm.nih.gov
  8. Nakashima H, Yoneda M, Kanemura T, et al. Conservative Treatment of Spondylolysis Involving Exercise Initiated Early and Sports Activities Resumed With a Lumbar-Sacral Brace. Journal of Orthopaedic Science, 2022;27(2):360–365. pubmed.ncbi.nlm.nih.gov
  9. Hollabaugh WL, Foley Davelaar CM, McHorse KJ, et al. Clinical Practice Patterns of Isthmic Spondylolysis in Young Athletes: A Survey of Pediatric Sports Medicine Clinicians. Current Sports Medicine Reports, 2022;21(11):405–412. pubmed.ncbi.nlm.nih.gov
  10. Sweeney EA, Fischer A, Brna M, et al. Rest Before Physical Therapy Is Not Necessary to Achieve Bony Healing of Lumbar Spondylolysis in Adolescent Athletes. Orthopaedic Journal of Sports Medicine, 2026;14(4). pubmed.ncbi.nlm.nih.gov
  11. Selhorst M, Sweeney E, Martin LC, et al. Immediate Physical Therapy Is Beneficial for Adolescent Athletes With Active Lumbar Spondylolysis: A Randomized Controlled Trial. British Journal of Sports Medicine, 2026;60(2):125–132. pubmed.ncbi.nlm.nih.gov
  12. Selhorst M, Sweeney EA, Dalsemer B, et al. Timing of Physical Therapy and Lumbar Muscle Atrophy in Adolescent Athletes With Spondylolysis. Clinical Journal of Sport Medicine, 2026 (online ahead of print). pubmed.ncbi.nlm.nih.gov
  13. Virkki E, Holstila M, Kolari T, et al. Elastic Lumbar Support Versus Rigid Thoracolumbar Orthosis for Acute Pediatric Spondylolysis: A Randomized Comparative Study. Spine, 2023;48(2):89–96. pubmed.ncbi.nlm.nih.gov
  14. Kasamasu T, Ishida Y, Sato M, et al. Rates of Return to Sports and Recurrence in Pediatric Athletes After Conservative Treatment for Lumbar Spondylolysis. Spine Surgery and Related Research, 2022;6(5):540–544. pubmed.ncbi.nlm.nih.gov
  15. Grazina R, Andrade R, Santos FL, et al. Return to Play After Conservative and Surgical Treatment in Athletes With Spondylolysis: A Systematic Review. Physical Therapy in Sport, 2019;37:34–43. pubmed.ncbi.nlm.nih.gov
  16. Debnath UK, Freeman BJ, Gregory P, et al. Clinical Outcome and Return to Sport After the Surgical Treatment of Spondylolysis in Young Athletes. Journal of Bone and Joint Surgery (British), 2003;85(2):244–249. pubmed.ncbi.nlm.nih.gov
  17. Selhorst M, Fischer A, Graft K, et al. Long-Term Clinical Outcomes and Factors That Predict Poor Prognosis in Athletes After a Diagnosis of Acute Spondylolysis. Journal of Orthopaedic & Sports Physical Therapy, 2016;46(12):1029–1036. pubmed.ncbi.nlm.nih.gov
  18. Mohriak R, Vargas Silva PD, Trandafilov M Jr, Martins DE, Wajchenberg M, Cohen M, Puertas EB. Spondylolysis and Spondylolisthesis in Young Gymnasts. Revista Brasileira de Ortopedia, 2010;45(1):79–83. pubmed.ncbi.nlm.nih.gov