 |
 |
 |
|
| home | CME | help | feedback | subscriptions | archive |
|
© 2004 The Journal of Bone and Joint Surgery, Inc.
Low-Back Pain in Athletes1 Department of Orthopaedic Surgery, Boston University Medical Center, 850 Harrison Avenue, Dowling 2 North, Boston, MA 02118. E-mail address: bonocm{at}prodigy.net
Investigation performed at the Department of Orthopaedic Surgery,
Boston University Medical Center, Boston, Massachusetts
While most occurrences of low-back pain in athletes are self-limited sprains or strains, persistent, chronic, or recurrent symptoms are frequently associated with degenerative lumbar disc disease or spondylolytic stress lesions. The prevalence of radiographic evidence of disc degeneration is higher in athletes than it is in nonathletes; however, it remains unclear whether this correlates with a higher rate of back pain. Although there is little peer-reviewed clinical information on the subject, it is possible that chronic pain from degenerative disc disease that is recalcitrant after intensive and continuous nonoperative care can be successfully treated with interbody fusion in selected athletes. In general, the prevalence of spondylolysis is not higher in athletes than it is in nonathletes, although participation in sports involving repetitive hyperextension maneuvers, such as gymnastics, wrestling, and diving, appears to be associated with disproportionately higher rates of spondylolysis. Nonoperative treatment of spondylolysis results in successful pain relief in approximately 80% of athletes, independent of radiographic evidence of defect healing. In recalcitrant cases, direct surgical repair of the pars interarticularis with internal fixation and bone-grafting can yield high rates of pain relief in competitive athletes and allow a high percentage to return to play. Sacral stress fractures occur almost exclusively in individuals participating in high-level running sports, such as track or marathon. Treatment includes a brief period of limited weight-bearing followed by progressive mobilization, physical therapy, and return to sports in one to two months, when the pain has resolved.
An athlete's lower spine usually performs demanding and extreme tasks without problems. The highly mobile lumbar spine and its associated muscles and ligaments, vernacularly called the low back, are an important but underrecognized source of great dynamic power during a golf or baseball swing, a gymnast's landing, a power-lifter's heavy squat, or a boxer's knockout punch. In static mode, it functions to help maintain an infielder's stance, a cycler's tuck, or a ballerina's arabesque. Not infrequently, the low back is revealed, by pain and dysfunction, to be one of the most common reasons for missed playing time by professional athletes1-3. Published rates of low-back pain in athletes range from 1% to >30%4-6 and are influenced by sport type, gender, training intensity, training frequency, and technique7-10. Although most cases are self-limited, many athletes have persistent symptoms8,11-14. Degenerative disc disease and spondylolysis are the most common structural abnormalities associated with low-back pain in athletes. However, despite these patients being highly motivated to return to activity, a specific pain generator is not always found, which often makes diagnosis and treatment challenging15. Thus, awareness of less common causes of low-back pain in athletes, such as sacral or facet stress fractures, is important10,16-18.
It is important to remember that low-back pain is a symptom, not a diagnosis. Most often, it is not associated with an underlying structural abnormality7,15. One must consider this when interpreting epidemiological reports of low-back pain. The lifetime prevalence of low-back pain in the general adult population is estimated to be 85% to 90%19. Between 2% and 5% of people report low-back pain that occurs at least once per year19. With conflicting reports, it is not clear whether athletes are at higher risk for low-back pain. According to one study, the lifetime prevalence of low-back pain in wrestlers (59%, nineteen of thirty-two) was significantly higher than that of age-matched controls (31%, 223 of 716)6. Sward et al.20 found a significantly higher rate of low-back symptoms in elite gymnasts (79%, nineteen of twenty-four) than in a control group (38%, six of sixteen). Likewise, Kujala et al.21 documented that 46% (thirty) of sixty-five adolescent athletes reported low-back pain compared with 18% (six) of thirty-three nonathletes. In contrast, Videman et al.4 found that low-back pain was less common in former elite athletes (present in 275 [29.3%] of 937) than it was in nonathletes (273 [44.0%] of 620). Back pain is a common reason for lost playing time by competitive athletes. McCarroll et al.1 reported that low-back pain accounted for loss of playing time by 30% (forty-four) of 145 college football players. Hainline2 found that 38% of professional tennis players reported low-back pain as the reason for missing at least one tournament. Ninety percent of all tour injuries in professional golfers involve the neck or back3. Low-back pain is more common in some athletes than in others. In a prospective study, Lundin et al.14 found that wrestlers had the highest rate of severe low-back pain (54%, fifteen of twenty-eight), while rates were lower for tennis and soccer players (32%, nine of twenty-eight, and 37%, eleven of thirty, respectively). Granhed and Morelli6 found the lifetime prevalence of low-back pain to be 59% (nineteen of thirty-two) in wrestlers compared with 23% (three of thirteen) in heavyweight lifters. Competitive male and female rowers had a 15% and 25% prevalence of low-back pain, respectively, in a recent study5. In comparison with other athletes, gymnasts appear to be among the most likely to report severe back pain22. Hutchinson23 found that six of seven elite rhythmic gymnasts reported low-back pain over a seven-week period.
Although this article focuses on the more common disorders that cause low-back pain in athletes, the evaluating practitioner should consider a broad differential diagnosis at presentation in order to avoid missing less frequent sources of symptoms (Table I).
Warm-up exercises are routinely performed prior to practice and competition to minimize the risk of injury. For the low back, a major focus is increasing flexibility, which in turn might improve the muscles' and ligaments' responses to demands. Despite the widespread use and acceptance of warm-up exercises, there are few data demonstrating that they can decrease the prevalence of low-back pain or the risk of injury in athletes. Athletes frequently have a period of rest between warm-up and play. Interested in the effects of this common scenario, Green et al.24 measured lumbar range of motion in twenty-six volleyball players prior to activity, immediately after a standardized warm-up regimen, and after a standardized warm-up followed by thirty minutes of rest. Although flexion and rotation were not affected, the lumbar spines were stiffer in extension after rest than they were immediately after warm-up. Flexibility immediately after warm-up was not significantly different from pre-warm-up values. These data suggest that bench rest after warm-up exercises can have a detrimental effect on lumbar flexibility. However, the link between the observed degrees of increased stiffness and the subsequent risk of lumbar injury remains unclear. These findings also call into question the commonly held belief that warm-up can improve low-back flexibility, and they suggest that the ability of warm-up to prevent injury, if indeed real, might be due to another mechanism. In support of these findings is the observation by Kujala et al.25, in a three-year longitudinal study, that specifically targeted training did not increase maximal lumbar extension in adolescent athletes. The authors concluded that aggressive attempts at increasing lumbar flexibility could unnecessarily stress structures, such as the intervertebral discs or pars interarticularis. In contrast, Kibler and Chandler26 found a specific conditioning program to be effective in increasing the lumbar range of motion in fifty-nine tennis players. The occurrence of back pain was not measured in either study. These data indicate that, with proper training, lumbar flexibility in competitive athletes reaches a plateau that should be maintained by regular stretching but attempts to push beyond that point in an effort to enhance performance might be detrimental. Others have studied the impact of flexibility on low-back pain. Kujala et al.27 prospectively examined lumbar flexibility in a group of adolescent athletes and nonathlete controls. Neither group had had previous low-back pain. Importantly, lumbar measurements were not performed during episodes of pain. While no differences were detected between male athletes (hockey and soccer players) and controls, female athletes (gymnasts and figure skaters) had a greater overall range of motion (p = 0.014) and range of motion of the low lumbar levels (p = 0.036) than did female nonathletes. Furthermore, a decreased range of motion of the low lumbar levels and decreased maximal extension were predictive of low-back pain in women: those within the lowest quartile had 3.4 times the chance of having pain lasting more than one week. In a study of 116 top male Swedish athletes, Sward et al.28 evaluated lumbar mobility, in addition to various other anthropometric features, in relation to back pain. While wrestlers and gymnasts were more flexible and soccer players were less flexible, there was no correlation between spinal flexibility and back pain, with the numbers available. This finding is in sharp contrast to the findings of Kujala et al.27. Curiously, the strongest predictor of pain was a low sacral inclination angle (p < 0.05), although this did not differ among the different sports. A correlation between lower-extremity function and the risk of low-back pain has been extensively studied. In a prospective examination of 257 college athletes playing various sports, Nadler et al.29 correlated the prevalence of low-back pain with findings related to the lower extremity. Of fifty-seven athletes with a lower-extremity overuse syndrome or acquired ligamentous laxity, fourteen (25%) had low-back pain (p < 0.001). Neither decreased flexibility of the lower extremities or limb-length discrepancy was a risk factor for back pain. However, the primary outcome measure was treatment for low-back pain, and this could have led to an underestimation of the prevalence of low-back pain and it could have affected the statistical analyses. In a later study, Nadler et al.30 linked side-to-side differences in maximum hip extension with the onset of low-back pain in female athletes. As was the case for the previously discussed studies assessing low-back flexibility and back pain, it is not clear whether reversing these so-called risk factors could decrease the chance of low-back injury. In contrast to the findings of Nadler et al., Twellaar et al.31 found no influence of lower-extremity flexibility on the occurrence of low-back pain in 136 physical education students. A history of low-back pain is the greatest predictor of future occurrences in athletes. Greene et al.32 found, in a prospective investigation of 679 college athletes, that those who reported prior low-back injury had three times the risk for subsequent episodes compared with those without prior pain; also, those who had active back pain at the start of the study had six times the risk for subsequent episodes compared with those without prior pain. Supporting these findings was the observation by O'Kane et al.33 that 57.1% (eighty-nine) of 156 competitive rowers with a history of preexisting low-back pain had subsequent occurrences, whereas 36.6% (613) of 1673 rowers without such a history had pain. Possibly because of adaptive measures, rowers with a history of pain before their rowing careers were less likely to quit the sport because of low-back symptoms. Equipment variables can influence the risk for low-back pain. Quinn and Bird34 found that the saddle type influenced the prevalence of low-back pain in 108 equestrians. Use of a traditional (or general purpose) saddle was associated with a 33% and 72% prevalence of pain in men and women, respectively. In comparison, a Western (deep-seated) saddle was associated with rates of only 6% and 33%, respectively. It was speculated that the added cushioning and stability provided by the Western saddle was the critical factor. Salai et al.35 studied the influence of seat angle on the pelvic-lumbar extension angle in recreational cyclists and found lumbar hyperextension to be a risk factor for low-back pain. Adjusting the seat to a neutral lumbar position alleviated back pain in 70% of the cyclists. Footwear can affect force transmission to the low back, which may be important in the understanding of low-back pain in running athletes. Ogon et al.36 compared lumbar paraspinal myoelectric responses in athletes running either barefoot or wearing running shoes with padded insoles. Initial muscle responses were later but the latency to maximal contraction was shorter with shoe wear. On the basis of these data, the authors suggested that running shoes with insoles improved the temporal synchronization between force transmission to the lumbar spine and paraspinal muscle responses.
Strains occur by disruption of muscle fibers at various locations within the muscle belly or musculotendinous junction37. Acute pain is most intense twenty-four to forty-eight hours after injury. It is often associated with spasm that, after a couple of days, may be localized to a so-called trigger point37. Recurrent muscle strains are denoted by short asymptomatic periods between episodes. Chronic strains are characterized by continued pain attributable to muscle injury. Patients with chronic back strains often undergo extensive radiographic workups, with negative findings. Keene et al.38 found muscle strain to be the most common injury causing low-back pain in 333 college athletes; 59% of the strains were acute and 41% were chronic. Micheli and Wood39 found that muscle strain was the reason for low-back pain in 27% and 6% of 100 adolescent athletes and 100 adult athletes, respectively. Sprains occur by subcatastrophic stretch of one or more of the spinal ligaments. While some individual fibers may be injured, the overall continuity of the ligament is maintained. I found no data delineating the exact tissue injury involved in low-back sprains in athletes in my review of the literature. Although the nociceptive innervation of the spinal ligaments is ill-defined, it is the presumed mode of pain transmission. Keene and Drummond37 thought that the interspinous process ligament is the most commonly affected by sprains. The exact or relative prevalence of lumbar sprains has not been reported, to my knowledge. Most practitioners recommend a short period of rest (one to two days) and intervals of icing in the acute phase after a strain or sprain. Gentle and progressive stretching exercises, preferably under the direction of a qualified trainer or physical therapist, should follow. Unfortunately, I found no clinical series in the literature documenting the effectiveness of a guided rehabilitation program for lumbar sprains or strains in athletes. On the basis of their experience with athletic patients, some practitioners have adapted previously developed programs for back pain typically used for nonathletes. A common link among these programs is the requirement that the individual be pain-free with nearly normal function (strength, flexibility, and endurance) before returning to activity7,12,40,41. George and Delitto41 described a treatment-based classification system for low-back pain in athletes. Nonradicular low-back pain was divided into six different syndromes (extension, flexion, lumbar mobilization, sacroiliac mobilization, immobilization, and lateral shift syndromes) on the basis of exacerbating factors and presumed etiology. Treatment was directed at restricting painful postures and concentrated on exercising the back within a pain-free arc of motion. For example, extension syndrome, characterized by pain that worsened with flexion and improved with extension, is treated with extension exercise and restriction of flexion. According to this method, the type of treatment is ultimately determined by the ability of the practitioner to differentiate responses to various provocative maneuvers. Hopkins and White40 described a three-cycle (level) system for rehabilitation after athletic low-back injuries. Each cycle differs in the relative degrees of rest, therapy, and time until return to play. Cycle I is divided into subsets A, B, and C. A brief summary of their recommendations is outlined in Table II. More succinctly, Dreisinger and Nelson7 guided treatment by categorizing it simply as acute or chronic. Admitting that acute injuries resolve quickly and usually spontaneously, the authors detailed recommendations for a short period of decreased activity and icing, administration of nonsteroidal anti-inflammatory drugs, and stretching followed by strength training and return to sports activity. They recommended that chronic cases be treated with trunk, back, and lower-extremity exercises to restore function.
The exact correlation between a degenerated intervertebral disc and low-back pain remains elusive. High rates of radiographic findings of degenerated discs in asymptomatic patients are evidence against an obligatory cause-and-effect relationship in the general population42. Treatment of discogenic low-back pain in athletes is challenging.
Pathogenesis of Disc Degeneration Stress within the anulus can produce tears within it43. Circumferential tears, representing delamination of the fibers within the tough outer ring, occur first. With continued stress, these can progress to radial tears. Radial tears can be detected as a small zone of increased signal by magnetic resonance imaging (Fig. 1) or as leakage of contrast medium within the posterior aspect of the anulus on a discogram. Next, nuclear desiccation and loss of proteoglycan ensue. At this stage, plain radiographs show mild decreases in disc space height; magnetic resonance imaging can reveal decreased signal intensity in the disc on T2-weighted images. A diminished capacity of the disc to sustain loads places greater demands on the posterior facet joints, causing degeneration of the articular surfaces. It has been proposed that, with time, advanced degenerative changes, such as osteophyte formation in both the disc and the facets, are an attempt at autostabilization.
Various components of the motion segment have been implicated as potential pain generators. Nociceptive microinnervation of the posterior aspect of the anulus, anterior aspect of the anulus, and facet joints has been characterized in anatomical and histological studies44-46. Reproduction of a patient's typical low-back pain with discography suggests that leakage of intradiscal fluid or anular distention is involved in the production of back pain. Despite ever increasing amounts of information, substantial limitations of our diagnostic abilities related to an understanding of disc degeneration and back pain remain.
Disc Mechanics and Sports Gatt et al.50 measured forces in the L4-L5 motion segment during blocking maneuvers in five football linemen. The average peak compressive load was >8600 N, with an average peak sagittal shear force of 3300 N. According to the authors, the magnitude of these forces exceeded the reported in vitro forces necessary to cause fatigue failure of the intervertebral disc. These data suggest that football lineman are at risk for routine repetitive disc microtrauma. Cholewicki et al.51 measured forces in the L4-L5 motion segment in fifty-seven competitive weight lifters. The average compressive loads were >17,000 N. In a similar study, Cappozzo et al.52 found that, when a person performed half-squat exercises with weights approximately 1.6 times body weight, compressive loads across the L3-L4 motion segment were about ten times body weight (approximately 7000 N for an average 70-kg person). Those investigators found that increasing lumbar flexion was the most influential factor affecting compressive loads.
Prevalence of Disc Degeneration in Athletes
Disc degeneration appears to be influenced by the type and intensity of the sport. Videman et al.4 demonstrated that former weight lifters have a higher rate of and more severe degenerative changes in the upper lumbar spine, whereas soccer players have findings almost exclusively in the L4 to S1 levels. While degenerative findings were most common in weight lifters, this group did not have a higher rate of back pain. In a study of Italian volleyball players, Bartolozzi et al.8 found that, of nineteen athletes who used proper technique and did not overtrain, 21% (four) had degenerative changes, whereas, of twenty-six who used improper technique and overtrained, sixteen (62%) had such changes. The frequency of symptoms in these two groups was not reported. Sward et al.22 found male gymnasts to have a higher rate of back pain and a greater number of radiographic degenerative changes than competitors in other sports. Some studies have suggested an association between specific imaging findings and the likelihood of back pain. Lundin et al.14 prospectively examined initial and ten-year follow-up radiographs of a group of athletes. The radiographic finding that most strongly correlated with low-back pain was decreased disc-space height, regardless of whether it was detected on the initial or follow-up examination. Furthermore, the greater the number of levels involved, the more likely the athlete was to have had low-back pain. Sward et al.20 found that decreased signal intensity within the disc on magnetic resonance imaging correlated with low-back pain in both athletes and nonathletes. They also found that an abnormal vertebral configuration (defined as an increased anteroposterior diameter, presumably from osteophyte formation) correlated with the occurrence of low-back pain. Comparing findings on baseline and follow-up magnetic resonance imaging in thirty-one girl athletes, Kujala et al.21 noted that six of eight who had low-back pain had a new radiographic abnormality, the most common of which was a ring apophyseal injury. Ogon et al.54 found that severe anterior end-plate degeneration was associated with a greater risk of low-back pain in 120 adolescent elite skiers. Videman et al.4 reported that former elite athletes with a history of at least monthly low-back pain had significantly higher scores for disc degeneration on magnetic resonance imaging than did those who had pain less frequently than twice a year (p = 0.04). Importantly, low-back pain was more strongly predicted by life dissatisfaction, neuroticism, hostility, extroversion, and poor sleep quality.
Nonoperative Treatment Cooke and Lutz13 detailed a five-stage rehabilitation protocol for the treatment of discogenic lumbar pain in athletes. Stage I (early protected mobilization) consists of a brief period of rest followed by various therapeutic modalities (application of heat or ice, nonsteroidal anti-inflammatory drugs, soft-tissue mobilization, and epidural injection). Once pain is controlled, the athlete begins an early exercise program to restore lumbar and lower-extremity range of motion. Stage II (dynamic spinal stabilization) focuses on co-contraction exercises of the abdominal and lumbar extensor muscles to stabilize the injured motion segment. Isometric exercises (contraction of the muscles without changing the length of the muscle) help to retrain muscles to maintain a mechanically neutral position. Stage III focuses on strengthening of the lumbar muscles. Importantly, initial strength gains are derived from improvements in neuromuscular firing as opposed to muscle fiber hypertrophy. In Stage IV, the athlete returns to sports activity. Plyometric exercises (resisted stretch of a muscle, or eccentric contraction, followed by an explosive concentric contraction) are recommended in this stage. The authors' criteria for returning to sports were (1) a full painless range of motion, (2) the ability to maintain a neutral spine position during sports-pecific exercises, and (3) a return of muscle strength, endurance, and control. Stage V includes institution of a maintenance program with regular home and warm-up exercises. Young et al.12 stressed the importance of active participation of the physical therapist in continually modifying the therapeutic regimen as the athlete progresses. They also stressed the importance of not relying solely on an algorithmic approach to rehabilitation. Therapy goals are pain reduction and decreasing the length of symptomatic episodes. This is achieved by (1) targeting abnormal skeletal shifts and posture, (2) reducing abnormally high muscle tone in spastic regions, and (3) reinforcing a comfortable body position, which is more often lumbar extension in patients with discogenic pain. Focus is placed on addressing tight extraspinal muscles, such as the hamstrings, hip flexors, hip rotators, hip extensors, and abdominals.
Minimally Invasive Treatment
Surgical Options Various methods of lumbar fusion have been advocated for the treatment of chronic disabling axial back pain from degenerative disc disease. These include posterolateral, anterior interbody, posterior interbody, transforaminal interbody, and circumferential (anterior and posterior) fusion techniques, with or without instrumentation. A review of the available literature suggests that interbody fusion techniques result in higher fusion rates and possibly better clinical outcomes than do posterolateral fusions57-60. Currently, most surgeons who perform lumbar fusion for discogenic low-back pain prefer an interbody technique rather than a posterolateral fusion alone. This reflects an increasingly popular belief that the disc itself is the main pain generator. Thus, notwithstanding generally higher fusion rates with the former procedure, the critical difference between interbody and posterolateral arthrodeses might be disc excision. The clinical results of interbody fusion for the treatment of back pain have varied, although no reports have specifically addressed the results in athletes, to my knowledge. Good or excellent results have been reported in between 80% and 100% of cases56-61. There is little or no information concerning the optimal time at which the athlete should return to sports activity after lumbar fusion16. Conservatively, however, the athlete should not return until there is radiographic evidence of a solid fusion, complete or nearly complete resolution of pain, and restoration of competitive-level measured functional parameters, such as strength, flexibility, and endurance.
Spondylolysis refers to a defect within the bone of the posterior part of the neural arch. While spondylolyses can develop at various sites62-64, the most common region to be affected is the isthmus of bone between the cephalad and caudad articular processes (Fig. 3). This region, more familiarly known as the pars interarticularis, is most commonly affected at L5 (in 85% to 95% of cases) and L4 (in 5% to 15%)65. While the exact etiology of isthmic spondylolysis is not known, it is widely believed to be a stress fracture caused by repetitive loading66-68, although there may be other contributing factors69-71. The prevalence of spondylolysis in the general population has been estimated to be between 3% and 6%72-74. Supporting a mechanical etiology is the fact that the highest prevalence has been reported in Alaskan Eskimos who sustain crouching postures for long periods of time while skinning whale blubber75. Most cases are asymptomatic. About one-quarter of symptomatic cases are associated with spondylolisthesis74.
The prevalence of spondylolysis in athletes is variable. In general, the prevalence is not higher than that in the general population76. However, some sports appear to be associated with a higher prevalence. In a study of 3132 competitive athletes, Rossi and Dragoni77 reported a rate of 43% in divers, 30% in wrestlers, and 23% in weight lifters. In a study of 3152 competitive athletes, Soler and Calderon76 documented a prevalence of 27% in throwing athletes, 17% in gymnasts, and 17% in rowers. Micheli and Wood39, in a study of 100 adolescent athletes and 100 adult athletes who presented with back pain, found that the adolescents had a higher rate of spondylolysis (47%) than did the adults (5%). Incidentally, these percentages were nearly reversed for the prevalence of degenerative disc disease. In some of the earliest reports of spondylolysis in athletes, young female gymnasts had been identified to be at particular risk. Jackson et al.11 evaluated 100 female gymnasts with radiographs because of back pain. Eleven (11%) demonstrated bilateral spondylolytic pars defects, and six of them had a grade-I slip. Although these prevalences appear substantially lower than those reported more recently, this is likely a result of improvements in radiographic assessment and increased awareness. Pain is usually confined to the low back. If the pain radiates, it does so to the buttocks or the back of the thigh and is more commonly from hamstring tightness than from radiculopathy. Pain is aggravated by extension of the lumbar spine, which is often elicitable during examination. Inspection can demonstrate exaggerated lumbar lordosis from increased sacral inclination without a slip (a possible predisposing factor for slippage78) or from spondylolisthetic deformity. With highergrade spondylolisthesis, the buttocks can appear heart-shaped and a midline step-off between the spinous processes can be palpated. Point tenderness on palpation of the affected spinous process can be present in cases of spondylolysis alone. Straight-leg raising can demonstrate hamstring tightness, but usually it does not reproduce radicular pain that extends below the knee. The single-leg hyperextension test described by Jackson et al.11 is a useful provocative test. It entails the patient standing on one leg while simultaneously extending the low back. This should produce pain on the side of the standing leg in a patient with a symptomatic ipsilateral spondylolytic lesion. To my knowledge, the reliability, sensitivity, and specificity of this test have not been analyzed. Neurologic examination usually reveals normal findings.
Imaging
When plain radiographs of a patient with persistent symptoms reveal negative findings, a bone scan, computerized tomography scan, single-photon-emission computed tomography scan, or magnetic resonance imaging scan can be made. A bone scan detects areas of bone turnover; i.e., bone deposition. Uptake can represent impending stress fractures, also known as stress reactions67. Jackson et al.67 utilized bone scans to detect stress reactions in thirty-seven young athletes. Importantly, initial bone scans were negative in seven athletes who subsequently had positive uptake within the pars on repeat examination one month later. Single-photon-emission computed tomography has been touted as the most sensitive test to detect a pars lesion. Bellah et al.80 performed plain radiography, bone scans, and single-photon-emission computed tomography scans for 162 adolescent athletes with low-back pain. In thirty-nine of seventy-one patients, the single-photon-emission computed tomography scan demonstrated increased uptake in the pars articularis when bone scans were negative. Whenever the bone scan was positive, the single-photon-emission computed tomography scan was also positive. The utility of single-photon-emission computed tomography for differentiating between symptomatic and asymptomatic pars lesions has also been studied. In a series of nineteen patients with radiographically confirmed lesions, Collier et al.81 found that single-photon-emission computed tomography scanning showed positive findings in eleven of thirteen symptomatic patients but in none of six asymptomatic patients. Additional prospective studies using single-photonemission computed tomography scans are needed to demonstrate more clearly their ability to predict symptoms. Computed tomography scans are more sensitive than plain radiographs. Spondylolytic lesions demonstrate a characteristic appearance that resembles an arthritic facet joint at the level of the pedicle on axial images. Some believe computed tomography to be the most sensitive test for spondylolysis82,83. Congeni et al.82 used computed tomography images to differentiate chronic nonhealing and acute-healing fractures. In their group of forty athletes with back pain, all had positive bone scans. Forty-five percent had a chronic lesion demonstrated by computed tomography, 40% had an acute lesion, and 15% had no obvious lesion on computed tomography. Unfortunately, the radiographic criteria for chronic and acute lesions were not detailed, and the authors did not attempt to correlate findings with prognosis. The role of magnetic resonance imaging in detecting or classifying spondylolysis is unclear. In a recent study, increased bone edema (hypointensity) within the pars on T1-weighted images in seven symptomatic patients with initially negative computed tomography scans was associated with the subsequent development of a detectable pars defect84. After fracture-healing, as demonstrated by follow-up computed tomography at five months, the findings on the magnetic resonance imaging had normalized. Kujala et al.9 prospectively followed a group of young athletes with low-back pain who had initial and follow-up bone scans as well as magnetic resonance imaging. Importantly, the magnetic resonance image was negative for eight patients who had a positive bone scan. Notably, the magnetic resonance imaging was performed with a low-field unit, and only standard T1 and T2-weighted image sequences were made. The investigators suggested that higher-strength magnets, similar to those currently available, with fat-suppression and STIR (short tau inversion recovery) sequences might increase sensitivity. To my knowledge, there have been no comparisons of magnetic resonance imaging and single-photon-emission computed tomography for the diagnosis of spondylolysis.
Natural History and Risk of Progression Ikata et al.86 compared the radiographs and magnetic resonance imaging scans of seventy-seven adolescent athletes with high-grade isthmic spondylolisthesis with those of eighty-eight adolescent athletes with spondylolysis alone. Slips in younger patients were more likely to progress. Furthermore, the authors found wedging of the L5 vertebra and rounding of the superior end plate of S1 in all patients who had a slip but in none of the patients who did not. It was not clear if these morphological changes were a cause or result of spondylolisthesis.
Nonoperative Treatment Jackson et al.67 treated a group of young athletes with pars stress reactions by limiting movements or activities that aggravated pain. This treatment was individualized to each athlete, and none discontinued playing sports. The treatment included a short period of initial bed rest. While the authors reported using a form-fitting brace intended to limit hyperextension of the lumbar spine, they did not report the duration of use or the criteria for discontinuation of such treatment. Blanda et al.87 reported the results of nonoperative care of sixty-two athletes with symptomatic spondylolysis. Defects were documented by radiographs and, when radiographs were negative, by bone scans. Treatment included restriction of activity and bracing for two to six months. No sports or exercise was permitted during the entire treatment period, and there was no description of rehabilitative exercises. Notably, the brace was designed to maintain lumbar lordosis. Fifty-two patients (84%) were reported to have an excellent result; eight (13%), a good result; and two (3%), a fair result. The rate of radiographic healing (independent of clinical outcome) was higher for unilateral defects (78% [eighteen] of twenty-three such defects healed) than bilateral defects (8% [three] of thirty-seven healed). The fact that eight patients underwent a posterolateral fusion because of the development of presumably asymptomatic or minimally symptomatic slip progression is of concern, since 98% of patients had either a good or an excellent result with regard to pain relief after bracing. The average duration of follow-up was 4.2 years, with a minimum of two years. While the authors concluded that nonoperative care with lordotic bracing was an effective treatment, it appears that this approach might predispose to the development of spondylolisthesis. In the same study, twenty athletes were treated for spondylolisthesis with the same protocol. Twelve had a grade-I slip; six, grade-II; and two, grade-III. Eighteen had pars defects, and two had an elongated pars. Seventeen (85%) had an excellent result, and one each had a good, fair, or poor result. Interpretation of these findings is obfuscated by the fact that the majority (twelve) of the twenty patients eventually underwent posterolateral fusion for progression of the slip (five), persistent pain (five), or a neurological deficit (two). The average duration of follow-up was 3.2 years, with a minimum of two years. Again, these data suggest reconsideration of the proposed regimen of nonoperative care including lordotic bracing. Steiner and Micheli88 used a modified, overlapping brace to treat sixty-seven young athletes with symptomatic spondylolysis or grade-I spondylolisthesis. The antilordotic brace was designed to hold the lumbar spine in relative flexion (in distinction to that used by Blanda et al.87). Seventy-eight percent (fifty-two) of the patients demonstrated a good or excellent result with no pain and returned to full sports activity. Nine (13%) had continued mild pain, and six (9%) underwent a posterolateral fusion for pain relief. The average duration of follow-up was 2.5 years. In a later study from the same institution, d'Hemecourt et al.89 evaluated the results of antilordotic brace treatment in seventy-three young athletes with spondylolysis or grade-I spondylolisthesis. Importantly, thirty-three patients had negative findings on plain radiographs and computed tomography images but had detectable pars lesions on either a bone scan or a single-photon-emission computed tomography scan. The treatment regimen included brace wear for twenty-three hours per day for six months followed by a weaning period of several months. A physical therapy program with a focus on flexion exercises was also instituted. Athletes returned to sports as early as four to six weeks after the initiation of treatment if they (1) had no pain with extension on physical examination, (2) had worn the brace full-time, and (3) remained pain-free. Results were very similar to those in the previous study88, with fifty-six (77%) of the athletes having a good or excellent result. In this series, the fate of the remaining 23% (seventeen) of the patients was not detailed; it was not reported whether they eventually underwent surgery. Sys et al.90 documented the results of nonoperative treatment of twenty-eight elite athletes (age range, twelve to twenty-seven years) with a pars lesion. All patients had negative findings on plain radiographs. Bone scans, single-photonemission computed tomography, and computed tomography were used to confirm the diagnosis. Treatment included bracing for a mean of sixteen weeks and subsequent follow-up for an average of thirteen months. A second computed tomography scan was made at the time of final follow-up to assess healing of the defect. Results were subcategorized according to whether the patient had a unilateral, bilateral, or so-called pseudobilateral defect. (A pseudobilateral defect was defined as asymmetrical signal within the pars bilaterally, indicating a confirmed unilateral lesion with a questionable or developing contralateral one.) All eleven unilateral lesions and five of the nine bilateral lesions healed. However, none of the eight pseudobilateral lesions had healed at the time of final follow-up. Independent of healing, 82% (twenty-three) of the athletes had an excellent outcome, 11% (three) had a good outcome, and 7% (two) had a fair result. The rate of return to sports activity did not differ among the three groups. The authors concluded that an unhealed defect does not preclude a good clinical result or a return to athletic pursuits.
Operative Treatment
Fusion Recently, interbody fusion for isthmic spondylolisthesis in adult patients has been reported more frequently94-98. Despite higher fusion rates and better maintenance of sagittal alignment, these methods have not been demonstrated to have clinical advantages compared with posterolateral fusion94,97. The role of this surgical technique for the treatment of adult athletes remains unclear. With the exception of patients in whom reduction of a high-grade slip has been elected (Fig. 4-B), adolescent athletes are not usually candidates for interbody fusion.
I found no data concerning the appropriate time after which an athlete may return to sports following lumbar fusion for spondylolysis or spondylolisthesis. In my opinion, the criteria should be similar to those for an athlete's return following nonoperative care: the athlete should be pain-free and have nearly normal function (strength, flexibility, and endurance) along with the added requirement of a solid fusion radiographically7,12,40,41.
Direct Pars Repair
Various fixation methods have been used successfully, including wiring, interfragmentary screws, pedicle screw-rod constructs, and pedicle screw-rod-hook constructs99,102,104,105,107. Biomechanical evidence suggests that pedicle screw-rod-hook constructs allow the least motion across the defect site (Figs. 6-A and 6-B). A critical portion of the repair, regardless of fixation type, is resection of the fibrous tissue within the defect, decortication to a bleeding surface, and ample autogenous bone-grafting.
Debnath et al.100 performed a prospective study of twenty-two competitive athletes (fifteen to thirty-four years of age) treated with a direct pars repair by either Scott wiring (passed around the transverse process and spinous process in a figure-of-eight pattern) or the Buck (translaminar interfragmentary) screw technique. All patients had defects documented by either single-photon-emission computed tomography or computed tomography. Two of the three patients treated with Scott wiring did not have healing of the defect, had revision to a posterolateral fusion, and did not return to sports. The other patient treated with Scott wiring had a healed defect but did not return to sports. In contrast, eighteen of the nineteen patients who underwent Buck screw fixation returned to sports, after an average of seven months, and demonstrated significant improvements in the Oswestry disability index and Short Form-36 scores (p < 0.001 for both). The numbers were too small to allow any meaningful statistical comparison between the two techniques, although the clinical failures in all three patients with the Scott wiring are troubling. Nozawa et al.103 documented the outcomes of a wiring technique in twenty competitive young athletes (average age, 23.7 years old). Bone healing was reported in all patients, and the Japanese Orthopaedic Association scores were significantly improved at up to an average of 3.5 years postoperatively (p < 0.0001). Furthermore, all patients returned to the same sport, but not all returned to the same level of competition. It is difficult to reconcile these contrasting findings with those of Debnath et al.100. Other investigators have reported surgical results in recreational athletes. Gillet and Petit102 reported the results of treatment with a rod-screw construct in ten patients. Six had an excellent result, returning to participation in recreational sports. One patient each reported a good and fair result. Two patients were considered to have a failure of treatment, and one of them later underwent interbody fusion. No postoperative brace was used. Although the authors stated that they assessed union with plain radiographs and tomograms, they did not report the healing rate. In a study of a similar patient population, Roca et al.99 reported that thirteen of fifteen patients were able to return to recreational sports activities one year after translaminar screw repair.
Stress fractures of the sacrum are an uncommon cause of low-back pain in athletes. The prevalence is unknown. Although such fractures appear to be more common in female athletes10,17,108, they have been reported in male athletes as well18. These fractures almost exclusively affect running athletes involved in sports such as cross-country, track, or marathon10,18. The presentation commonly includes an insidious onset of asymmetric low-back or gluteal pain that develops over a period of weeks, usually with no history of an acute incident. Physical findings are paramedian point tenderness of one side of the sacrum or sacroiliac joint. The faber test (figure-of-four test of the lower extremity) can be positive on the ipsilateral side. Delvaux and Lysens18 described a "hopping test" in which pain is reproduced by bouncing on the leg on the affected side. In the one case that they reported, this sign was negative after the fracture healed. The flamingo test (patient standing on the ipsilateral leg) may also be positive. Female patients should be questioned about eating habits and menstrual history to rule out the so-called terrible triad in women athletes. A positive history of amenorrhea or an eating disorder should prompt a bone mineral density test. If this reveals a decreased bone density in association with a sacral lesion, an insufficiency fracture, rather than a stress fracture, is more likely. Treatment of the underlying cause of osteoporosis should be initiated in conjunction with psychological counseling. Plain radiographs usually reveal negative findings, necessitating advanced imaging studies for diagnosis. Magnetic resonance imaging, computed tomography, single-photon-emission computed tomography, and bone scans can be diagnostic. Johnson et al.10 used various combinations of these tests to detect lesions. In all cases in which magnetic resonance imaging was performed, it confirmed the presence of a fracture that was detectable on bone scan. In both of their case reports, Shah and Stewart108 and Featherstone17 used magnetic resonance imaging alone to confirm the diagnosis. Treatment is always nonoperative, consisting of rest and protected or non-weight-bearing. This is followed by progressive mobilization, weight-bearing, and activity as symptoms permit. The overall prognosis is favorable, with the athletes returning to sports activity in an average of about one and a half months108. The athlete should be adequately rehabilitated before returning to full activity. Most patients, however, report persistent mild or intermittent pain10.
|
Top
Abstract
Introduction
