One of the most common conditions of low back pain in school-age athletes is spondylolysis.'"3 Spondylolysis is a stress fracture of one or both of the pars interarticularis and most commonly occurs in the lower lumbar spine. Predisposing factors include trauma,4,5 genetics,6 and developmental defects.7
In athletes, fracture typically occurs as a result of repetitive microtrauma, mostly with hyperextension combined with rotation and impact loading, rather than as a result of a single episode of macrotrauma. This finding is based on experimental studies in which authors were unable to induce a defect through isolated hyperextension or flexion,8,9 whereas the fracture of the pars interarticularis could be produced when repetitive cortical fatigue stresses were applied.410 Sports such as gymnastics, dance, figure skating, and football place children at a higher risk for developing these lesions.1 M4
Treatment for spondylolysis is a matter of debate. Although most specialists agree on an initial conservative treatment course, modalities are diverse and outcome measures are poorly reported. i5~ls Restricting sports activity for up to 6 months is a common part of the treatment regimen.19 Bracing has been variably reported, ranging from a corset to a molded lumbosacral orthosis.17,19,20 A nonspecific physical therapy regimen often is recommended.
The primary goal of treatment is to achieve a stable, pain-free union of the fracture. A bony union is preferred. However, a stable, pain-free fibrous union that enables full activity has been deemed acceptable by many authors.20,21 Treatment should be defined precisely with respect to the type and duration of immobilization and should allow the young athlete early return to sports participation.
Reports from our institution have described a specific approach to treat symptomatic individuals with spondylolysis or minimal grade spondylolisthesis using an antilordotic lumbosacral orthosis, the modified Boston Overlap Brace.20,22-23
In a study of young athletes by Steiner and Micheli,20 78% showed excellent or good results with no pain and a return to full activities, 13% continued to have mild symptoms, and 9% required later operative fusion. Diagnosis was based on clinical fmdings and radiographs, including Tc99 bone scans. Since this report, single photon emission computed tomography (SPECT) was added to our evaluation of low back pain24 and has increased the likelihood of early detection of stress injuries.24,25 Single photon emission computed tomography (CT) differentiates chronic nonunion fractures from biologically active lesions and improves localization by spatial separation of overlapping bony structures. Recently, CT has been used to grade and determine fracture healing potential.17 Computed tomography was not routinely used during this study period for radiographic assessment of healing.
Type of Sport Precipitating Symptoms
This study evaluated clinical outcome in young athletes with spondylolysis or low-grade spondylolisthesis treated with an antilordotic lumbosacral orthosis, a physical therapy program, and a return to sports in 4-6 weeks.
MATERIALS AND METHODS
Between 1988 and 1995, 161 children, adolescents, or young adults were treated at our clinic for low back pain for at least 6 months. Seventy-three (45%) patients (39 boys and 34 girls) were diagnosed with spondylolysis and underwent 6 months of treatment with followup of at least 1 year. Mean patient age was 15.7±2 years (range: 9-19 years).
Pain at diagnosis was rated as moderate or severe in 93% of patients and was apparent during exercise alone (38%) or all the time (62%). Pain was always located in the lower back, and 20% of patients experienced additional pain in the buttocks, hips, thighs, or lower legs. Thirty (41%) patients experienced pain of <2 months' duration. No patient had been treated with a brace.
Radiographic Lesions at Initial Presentation
All patients were evaluated by a sports medicine physician. Personal history, clinical evaluation, radiographic assessment, and treatment was obtained from the patient charts. Each patient was interviewed by telephone an average of 4 years following diagnosis (range: 2-9 years). Sports precipitating symptoms were divided into high-risk and low-risk sports (Table 1). The high-risk group included sports with intense repetitive hyperextension and impact such as figure skating, ballet or dance, football, soccer, and gymnastics. In addition to the type of sports activity, the amount of hours trained per week during the 6 months prior to the initial clinic visit were evaluated.
Evaluation included anteroposterior (AP), lateral, and oblique radiographs, and SPECT Tc"-scintigraphy. Lesions were classified according to location. If the lesion was not seen on initial radiographs, CT was performed (Table 2). Forty patients had positive visible fractures on CT or radiographs (26 with positive bone scans and 14 with negative bone scans). Thirty-three patients had negative visible fractures on CT or radiographs (25 with intense focal uptake in one or both pars interarticularis at a single lumbar level and 8 with diffuse posterior element uptake ?? the SPECT bone scan).
All patients received an antilordotic lumbosacral orthosis to wear for 23 hours a day for 6 months with a weaning-off period of several months. The brace is a rigid, anteriorly overlapping module of polyethylene designed to reduce lordosis with anterior cavitation of 15° and posterior alignment of 0° of lordosis. Patients underwent a specific physical therapy program with a flexion bias and pelvic flexibility. Athletes having no pain on lumbar extension at 4-6 weeks in the brace were allowed to return to sports provided they wore the brace and remained pain-free.
Treatment results were classified in four groups according to pain, brace requirements, and activity level. The "excellent" outcome group included patients with no pain and no brace requirements who were back to foil activities. The "good" outcome group experienced occasional aching with vigorous activity, no brace requirements, and were back to full activities. The "fair" outcome group included pain with vigorous activity, occasional use of the brace, but no pain in activities of daily living. The "poor" outcome group included patients with pain during daily activities or candidates for fusion.
Clinical outcome was analyzed as the primary outcome. Multiple stepwise logistic regression analysis was used to establish the independent multivariate risk factors and to control for possible confounding. Variables included age at diagnosis, gender, maturity, participation in high- or low-risk sports, hours of weekly training, duration of pain before treatment, occurrence of an acute macrotraumatic episode, hamstring tightness, bracing compliance, location and laterality of the lesion, and initial radiographic diagnosis. The likelihood ratio Chisquare test was used to assess significance of each variable, and the odds ratio with a 95% confidence interval was constructed as measures of association.26
Fisher's exact test was used to compare simple proportions. Two-tailed values of P<.05 were considered statistically significant. Continuous data are presented as mean ±standard deviation. Statistical analysis was performed using the PROC LOGISTIC procedure in the SAS software package (version 6.12, SAS Institute, Cary, NC).
The main clinical finding was pain on lumbar hyperextension (93%). Additionally, 77% of patients demonstrated tight hamstrings with a straightleg raise <90°. Twenty (28%) patients had persistent pain on flexion. Weekly training averaged 15 ±7 hours (range: 3-30 hours) in the sport that precipitated the low back pain.
The lumbosacral orthosis was applied within 2 weeks of diagnosis and worn for 6.9±3.1 months. Compliance was high; 85% of patients wore the lumbosacral orthosis for at least 80% of the recommended 23 hours.
There was a predominance of L5 spondylolysis (66%), with 4 patients showing a grade I spondylolisthesis (Table 2). Clinical outcome was not affected by coexisting spinal pathology (Table 3). In approximately 50% of patients, spondylolysis was associated with a codiagnosis such as spina bifida occulta. Twenty-seven (71%) of 38 patients with comorbtd spine pathologies had good or excellent clinical outcomes, which was not significantly lower than the 29 (83%) of 35 patients with no additional diagnosis of spinal pathology (P=.28, Fisher's exact test).
Logistic regression analysis identified several multivariate predictors of clinical outcome. Male athletes fared better than females, and high-risk sports were associated with poorer outcome than low-risk sports (P<.01). In addition, hamstring tightness at diagnosis and acute duration of pain were associated with a less favorable clinical outcome 0P=.O3). Other variables including age at diagnosis, training intensity, bracing time, location of the lesion, and occurrence with trauma were not significant predictors of clinical outcome (P>.20). Similar to previous studies,*1,20,27 athletes in our study participated in a wide range of sports. Girls or boys who participated in high-risk sports were five times more likely to have an unfavorable clinical outcome than those who participated in low-risk sports (odds ratio=5, 95% confidence interval=2.4-7.5, P=.003).
Clinical Outcome and Associated Spinal Pathology
The radiographic evaluation of spondylolysis has recently evolved into a specific algorithm at our institution (Figure 1 ). However, the radiographic evaluation during this retrospective study was less defined. Nonetheless, the treatment was identical to our current regimen. Subsequent studies to this review have shown plain radiograph insensitivity to the detection of fractures seen on bone scans. Congeni et al27 reported CT demonstrated spondylolytic fractures in 85% of athletes with back pain and a positive bone scan and normal plain radiographs. Saifuddin et al28 explained this apparent discrepancy by showing that only 32% of these fractures are aligned within the beam of the oblique radiograph. Using CT, these fractures may be classified as early, progressive, or terminal.17 Therefore, one limitation of this study is the lack of specificity in classifying the presenting and final fracture types by CT criteria. However, the SPECT imaging data obtained in this study gives an indication of the chronicity and stage of fracture healing.
Treatment regimens for young athletes with spondylolysis or minimal grade spondylolisthesis have varied from corset bracing to full lumbosacral orthoses.1719 However, the timing for return to sports participation has not been well established. This study demonstrates treatment with a full-time brace, and a specific physical therapy program allows the adolescent athlete an early return to sports participation. This approach led to an 80% good to excellent clinical outcome. Our results are consistent with previous reports.19,20
The lumbosacral orthosis brace treatment may seem demanding. However, through stabilization of the lumbar spine, children and adolescents usually are able to return to full activity after A6 weeks. Previous authors have demonstrated similar results with brace treatment and physiotherapy, but restricted the young athlete from sports activities during the full course of treatment.19
One could hypothesize that our approach prevents the athlete from losing muscle mass as an important stabilizer of the spine and therefore enhances rehabilitation. The documented high compliance of wearing the brace (85% of athletes wore the brace for at least 80% of the recommended 23 hours) suggests children and adolescents want to rerum to sports participation quickly. Additionally, most athletes reported that they felt protected and experienced pain relief while wearing the lumbosacral orthosis, particularly during physical activities.
Figure: Imaging protocol for symptomatic lumbar hyperextension. The brace protocol was 46 months of full-time lumbosacral orthosis followed by CT. Short-term brace was full-time lumbosacral orthosis until asymptomatic. Abbreviations: AP=anteroposterior, PT=physical therapy, and SPECT=single photon emission computed tomography.
Despite the 80% favorable clinical outcome with the lumbosacral orthosis treatment, radiographic healing was not evaluated. However, current CT techniques offer a more accurate assessment of fracture union. In addition, further research is needed to determine the optimal bracing time based on CT classification.
A higher incidence of spondylolysis has been reported for patients with scoliosis,29,30 atypical Scheuermann's disease,31 and spina bifida occulta.7,32 This observation appears tenable as many of these structural abnormalities contribute to biomechanical stress on the lumbar spine. However, in our study, a good or excellent clinical outcome was not significantly lower in patients with comorbid spinal pathologies compared to those with no additional diagnosis of spinal pathology. The incidence of spina bifida occulta was lower in our population than previously reported.7 However, the majority of our patients were performing a high-risk sport at a high intensity level, which increased the risk for an isthmic rather than a dysplastic lesion.
The risk of spondylolisthesis appears low. Only four athletes presented with a grade I spondylolisthesis, and no progression of slip occurred throughout the study period. There also was no progression or a new appearance of süppage. This finding is in accordance with Blackburne and Velikas33 and Laurant and Oestarman34 who stated that if the slip is <30% (equivalent to spondylolisthesis grade I at presentation), further dislocation >30% is unlikely. The incidence of unilateral lesions was similar to the study by Congeni et al27 but higher than that reported by Blanda et al.19
Given previously reported theories that repetitive stress is a cause of spondylolysis particularly when hyperextension is combined with rotation or impact loading,810 the poorer outcome of pars fractures in the high-risk sport group is not surprising. These young athletes might have stressed their lumbar spine with a higher repetitive load and in a more extreme position, thereby causing more severe fractures. In accordance with others,8,35,36 hamstring tightness at diagnosis was apparent in 67% of the population and was a predictor of poor clinical outcome. Tight hamstrings contribute to lumbar hyperlordosis by tilting the pelvis forward, thus increasing the permanent stress on the inferior articular facets of the lumbar vertebra.9,37
Our results demonstrate that antilordotic lumbosacral orthosis bracing for 6 months is an effective treatment for young athletes with spondylolysis. Bracing allows the healing of growth tissue. Clinical improvement can be anticipated in patients appropriately managed with lumbosacral orthosis treatment.38 We believe bracing often is a preferred approach for young, active athletes because it allows them to return to activities 4-6 weeks after initiation of treatment.
1. Arvidson EB, Micheli LJ. Spine and trunk problems in athletes. Current Opinion in Orthopedics. 1990; 1:361-364.
2. Micheli LJ. Low back pain in the adolescent: differential diagnosis. Am J Sports Med. 1979; 7:362-364.
3. Micheli LJ. Back injuries in dancers. Clin Sports Med. 1983; 2:473-484.
4. Cyron BM, Hutton WC. The fatigue strength of the lumbar neural arch in spondylolysis. J Bone Joint Surg Br. 1978; 60:234-238.
5. Troup JD. Mechanical factors in spondylolisthesis and spondylolysis. Clin Orthop. 1976; 117:59-67.
6. Wynne-Davies R. Scott JH. Inheritance and spondylolisthesis: a radiographic family survey. J Bone Joint Surg Br. 1979; 61 :30l -305.
7. Wiltse LL. The etiology of spondylolisthesis. J Bone Joint Surg Am. 1962; 44:539-560.
8. Letts M, Smallman T, Afanasiev R, Gouw G. Fracture of the pars interarticularis in adolescent athletes: a clinical-biomechanical analysis. J Pediatr Orthop. 1986: 6:40-46.
9. Farfan HF, Osteria V, Lamy C. The mechanical etiology of spondylolysis and spondylolisthesis. Clin Orthop. 1976; 117:40-55.
10. Cyron BM, Hutton WC, Troup JD. Spondylolytic fractures. / Bone Joint Surg Br. 1976; 58:462-466.
11. Micheli LJ. Back injuries in gymnastics. Clin Sports Med. 1985;4:85-93.
12. Micheli LJ, Micheli ER. Back injuries in dancers. In: Shell CG, ed The i984 Olympic Scientific Congress Proceedings. Vol 8: The Dancer as Athlete. Champaign. Ill: Human Kinetics; 1986:91-94.
13. Micheli LI, Kasser JR. Case report 6: painful spondylolysis in a young football player. In: Hochschuler SH, ed. The Spine in Sports. Philadelphia, Pa: Hanley and Belfus; 1990:327-330.
14. Micheli U, McCarthy C. The spine in sports. In: Watkins RG, ed. Figure Skating. St Louis, Mo: Mosby Ine; 1996:557-564.
15. Gramse RR, Sinaki M, Ilstrup DM. Lumbar spondylolisthesis: a rational approach to conservative treatment. Maw Clin P roc. 1980: 55:681-686.
16. Jackson DW. Wiltse LL. Dingeman RD. Hayes M. Stress reactions involving the pars interarticularis in young athletes. Am J Sports Med. 1981:9:304-312.
17. Monta T. Ikata T, Katoh S, Miyake R, Lumbar spondylolysis in children and adolescents. J Bone Joint Surg Br. 1995; 77:620-625.
18. Turner RH, Bianco AJ Jr. Spondylolysis and spondylolisthesis in children and teenagers. J Bone Joint Surg Am. 197 1 : 53: 1 298- 1 306.
19. Blanda J, Bethem D, Moats W, Lew M. Defects of pars interarticularis in athletes: a protocol for nonoperative treatment. J Spinal Disord. 1993:6:406-411.
20. Steiner ME. Micheli U. Treatment of symptomatic spondylolysis and spondylolisthesis with, the modified Boston brace. Spine. 1985; 10:937-943.
21 . Muscbik M, Hahnel H, Robinson PN, Perka C, Muschik C. Competitive sports and the progression of spondylolisthesis. J Pediatr Orthop. 1996; 16:364-369.
22. Micheli LI. The use of the modified Boston brace system for back pain: clinical indications. Orthotics and Prosthetics. 1985; 39:4146.
23. Micheli LJ. Hall JE, Miller ME. Use of modified Boston brace for back injuries in athletes. Am J Sports Med. 1980; 8:351-356.
24. Bellah RD, Summerville DA, Treves ST, Micheli LI. Low-back pain in adolescent athletes: detection of stress injury to the pars interarticularis with SPECT. Radiology. 1 99 1 : 1 80:509-5 1 2.
25. Bodner RJ, Heyman S, Drummond DS, Gregg JR. Tbe use of single photon emission computed tomography (SPECT) in the diagnosis of low-back pain in young patients. Spine. 1988; 13:1155-1160.
26. Hosmer DW, Lemeshow S. Applied Logistic Regression. New York. NY: John Wiley & Sons; 1989.
27. Congeni J. McCuI loch J. Swanson K. Lumbar spondylolysis. A study of natural progression in athletes. Am J Sports Med. 1997; 25:248-253.
28. Saifuddin A, White J. Tucker S. Taylor BA. Orientation of lumbar pars defects: imphcations for radiological and surgical management. J Bone Joint Surg Br. 1998: 80:208-21 1.
29. Hutchinson MR. Low back pain in elite rhythmic gymnasts. Med Sci Sports Exerc. 1999; 31:1686-1688.
30. Fisk JR, Moe JH, Winter RB. Scoliosis, spondylolysis and spondylolismesis. Their relationship reviewed in 539 patients. Spine. 1978; 3:234-245.
31. Ogilvie JW, Sherman J. Spondylolysis in Scheuermann's disease. Spine. 1987; 12:251-253.
32. Heftì F, Brunazzi M, Morscher E. Natural course in spondylolysis and spondylolismesis [Germán]. Orthopäde. 1994; 23:220-227.
33. Blackburne JS, Velikas EP. Spondylolisthesis in children and adolescents. / Bone Joint Surg Br. 1977;59:490-494.
34. Laurant LE, Oesterman K. Spondylolisthesis in children and adolescents. Acta Orthop BeIg. 1969;35:717-727.
35. Ciullo JV, Jackson DW. Pars interarticularis stress reaction, spondylolysis, and spondylolisthesis in gymnasts. Clin Sports Med. 1985: 4:95-110.
36. Seitsalo S, Osterman K. Hyvarinen H, Tallroth K, Schlenzka D, Poussa M. Progression of spondylolismesis in children and adolescents. A long-term follow-up of 272 patients. Spine. 1991; 16:417-421.
37. Hensinger RN. Spondylolysis and spondylolisthesis in children and adolescents. J Bone Joint Surg Am. 1989; 71 : 1098-1 107.
38. d'Hemecourt PA, Gerbino PG II, Micheli LI. Back injuries in the young athlete. Clin Sports Med. 2000: 19:663-679.
Type of Sport Precipitating Symptoms
Radiographic Lesions at Initial Presentation
Clinical Outcome and Associated Spinal Pathology