Ninety percent of people have low back pain in their lives, and 40% of these conditions can become chronic(5,24,32). Back pain in the general population has a 5% annual occurrence and 60-90% lifetime incidence(32,55). The economic burden of care for low back pain in the United States is staggering, costing billions of dollars per year. The cost of care, coupled with reports that the majority of cases of back pain and sciatica resolve on their own, has lead many health practitioners to conclude that the need for formal exercise and rehabilitation of the spine is limited. However, while the natural course of back pain may be considered favorable, the resolution of pain does not guarantee freedom from recurrence. In fact, between 70 and 90% of patients have additional episodes of back pain(32,70). Furthermore, long-term follow-up studies indicate that approximately one-third of primary-care acute back pain patients continue to have back pain on at least half the days of the year at 1 and 2 yr of follow-up (70). This strongly suggests that with the initial insult to the spine, certain pathoanatomic and functional changes occur that render the spine vulnerable to further injury.
Back pain has been identified as a major problem in a number of competitive sports. Injuries to the pars inter-articularis is three to four times as common in female gymnasts as in the general population(51). Almost a third of collegiate football players are affected, and in one survey, almost 40% of men's professional tennis players reported having missed at least one tournament in their career due to back pain (30,49). Baseball has also been identified as a sport with spine-related problems, although it has not been as thoroughly investigated as other sports in this regard (32).
For the athlete, the distinction between absence of symptoms and absence of dysfunction is particularly important. Although pain is generally what prompts an athlete to seek medical care, an athlete's performance can suffer in the absence of pain but in the presence of subtle biomechanical flaws and maladaptations which are the product of inadequate training regimens or failure to rehabilitate previous injuries(32,43). This is particularly important for athletes such as baseball pitchers, tennis players, and quarterbacks whose performances are highly dependent upon the coordinated efforts of multiple structures throughout the kinematic chain.
BASIC ANATOMY AND BIOMECHANICS
A full discussion of spine anatomy and biomechanics is beyond the scope of this text. More detailed sources are available (27). The basic motion segment of the spine is the three-joint complex, consisting of one intervertebral disc (sitting between adjacent vertebrae) and the two accompanying zygoapophyseal (facet) joints (Fig. 1). The individual lumbar nerve roots emerge bilaterally below the pedicle of each respective set of vertebrae. The outer one-third of the disc is innervated. The nucleus pulposus and the remainder of the disc are not. The adult disc is an avascular structure, deriving its nutrition via imbibition from loading. The pars interarticularis is the portion of the posterior arch located between the superior and inferior articular processes. The sinuvertebral nerve supplies the posterior margin of the annulus and the posterior longitudinal ligament (PLL). The anterior longitudinal ligament (ALL) and the lateral aspect of the disc are innervated by ventral rami and gray ramus communicans. The facet joints are innervated by up to three levels. Potential pain generators in the lumbar spine include the vertebral body, the outer one-third of the disc, the facet joints, PLL, ALL, supraspinous and interspinous ligaments, muscles, and the spinal nerve root.
The cardinal planes of spine motion are flexion/extension (sagittal translation and rotation), torsion (horizontal rotation and translation), and side bending (coronal rotation and translation). Protection of the three-joint complex from excessive forces requires unrestricted and efficient motions between adjacent vertebral segments. Joint stability is inherently dependent upon the ability to maintain a centralized instantaneous axis of rotation(IAR). Degeneration of discal segments allows for migration of the IAR with repetitive motion leading to injury of the three-joint complex. Pure sagittal motions and compression are relatively well tolerated by the disc and three-joint complex. Torsional stresses and combined motions (i.e., forward flexion and rotation) are the most injurious (67). Because combined motions are commonly encountered in sports, unless the supporting structures (i.e., muscles and connective fascia) are sufficiently strong to offset some of the forces on the spine the three-joint complex will be vulnerable to injury. The consequences of such injury include the potential for development of facet synovitis, breakdown of the annulus, and migration of nuclear material beyond the confines of the disc and surrounding ligaments leading to mechanical or chemical irritation of nerve roots or the spinal cord.
ETIOLOGY OF SCIATICA
The surgical approach to sciatica is highly reliant upon the assumption that a compressive lesion is at fault. This may be very relevant in cases of foraminal or central spinal stenosis, but size of the disc, or the appearance of compression in imaging studies, may not always be relevant. Boden et al. studied 67 persons with no history of low back pain, sciatica, or neurogenic claudication. Imaging studies were interpreted by three neuroradiologists(6). One-third of the subjects were found to have a substantial abnormality, and degenerative or bulging disks were observed in at least one level in 35% of the subjects between the ages of 20 and 39. Thirty-six percent of the subjects in the study by Wiesel et al. had no history of low back pain yet had an abnormal lumbar spine computed tomography(CT) scan (73). Holt found a similar proportion of patients (37%) without a history of low back pain to have“abnormal” discograms (35). One of the more interesting studies was the one performed by Hitselberger and Witten(34). Three hundred patients without a history of low back pain underwent myelography as part of the workup for acoustic neuroma. Given that myelography had long been the “gold standard” for diagnosis of disc abnormalities, the results were striking. There were varying degrees of disc abnormality in 110 (37%) of the patients, with lumbar abnormalities in 24% of the cases. Conversely, even athletes presenting with classic sciatica symptoms may not have a readily identifiable disc herniation that corresponds with the neuroanatomic level of their symptoms(33).
Inflammation has become increasingly recognized as an important cause of sciatica. Numerous biochemical substances have been identified as mediators of pain that can directly stimulate tissue nociceptors, sensitize peripheral nociceptors to other chemicals, lower the response threshold to mechanical stimuli, generate ectopic neuronal discharges, and sensitize the central nervous system via the dorsal horn of the spinal cord at the level of the substantia gelatinosa (12). These substances include inflammogens and neurogenic mediators. Among the inflammogens are cytokines, bradykinin, serotonin, nitric oxide, histamine, and prostaglandins. High concentrations of prostaglandins (i.e., PLA2) are known to be contained in herniated discs (60). Takahashi et al. examined herniated disc specimens and identified increased presence of inflammatory cytokines such as interleukin 1-α (68). Sensitivity of inflammogens to steroids was demonstrated in this study as a secretion of prostaglandin E2, and cytokines were reduced by betamethasone. Similarly, Olmarker had previously shown glucocorticoids to have a beneficial effect upon nuclear induced inflammation (53). Doita et al. suggested that mononuclear cells along the disc margin may be a critical factor as they induce neovascularization and persistence of inflammation(17).
Neuropeptides (primarily produced in the dorsal root ganglion, or DRG) such as substance P, vasoactive intestinal peptide (VIP), calcitonin gene-related peptide (CGRP), and somatostatin comprise the majority of the neurogenic group(41). Mechanical stimulation of the DRG and nerve roots increases substance P levels in the DRG and substantia gelatinosa of the dorsal spinal horn (41). In an elegant series of studies utilizing a rat model, Kawakami et al. demonstrated that chemical irritation(chromic gut) and not compression from clamping or silk ligature induced a time dependent reversible thermal hyperalgesia(41,42). Alterations in c-fos gene expression at a spinal cord level was observed with the neurochemical irritant, suggesting a role for this pathway in the persistence of sciatica-type symptoms(41,42). Thus, although spinal nerve compression may be the progenerator of motor loss, the presence of inflammation may be a prerequisite for pain (25,26).
Natural Course of Sciatica
In the strictest sense, there is a lack of information regarding the true natural history of disc herniation in the lumbar spine. The“classic” study often cited in this regard is by Weber(71). From a population of 280 patients with radiculopathy, a subgroup of 126 patients with myelographically diagnosed disc herniation, radicular pain, positive straight leg raises reduced spinal mobility, and in many cases weakness were randomly placed into nonoperative(n = 66) or operative (n = 60 discectomy ± laminectomy) groups. The remaining patients were either assigned to surgery because of “definite” indications or to another conservative care group. The patients were then followed over 10 yr with reevaluations at years 1, 4, and 10. After 1 yr, the surgical group appeared to fare better (80% satisfactory vs 61% satisfactory results), but after 4 yr and at 10 yr no statistical difference existed between groups. Problems with this study include the crossover of 17 nonoperative patients to the operative group when they regressed (although the analysis was done to either include or not include these subjects), the probable initial assignment of “worst cases” to the surgical group, and the lack of “optimal” conservative care in the nonoperative group. One of the most significant findings of this study was that for the 64 patients who originally demonstrated some type of paresis at the outset (32 operative, 31 conservative), restitution of strength was unrelated to type of treatment, with virtually no paretic patients by yr 10. On the other hand, 35% of the subjects still reported some type of sensory dysfunction at yr 10. In another study, Pearce followed 91 patients with sciatica who received complete bed rest for 2-4 wk (54). Fifty-three were interviewed at an average of 8 yr after the initial attack, with 20 answering mailed questionnaires. A 70% satisfactory response and a 30% poor response to absolute bed rest was reported. Even with the limitations of these data, it appears that at least 60% of patients with radiculopathy will improve without either optimal conservative care or surgical intervention(54,56,71).
Indications for Surgery
The athlete presenting with such massive spinal cord compromise that there is evidence of a cauda equina syndrome is in need of decompressive surgery. Most authors are of the opinion that severe motor deficits or intractable pain are also strong indications for surgery (3). However, the majority of patients with disc pathology present with some combination of back and leg pain accompanied by variable amounts of weakness, sensory deficits, and musculoskeletal biomechanical imbalances(32,60). If pain can be controlled by therapy incorporating preferential spine motions, oral medications, or epidural steroid injections, the indication for surgery, even in the face of motor loss, becomes less compelling.
The term “failed conservative treatment” is often used to justify surgical intervention. It is incumbent upon the physician managing the athlete with discogenic pain to ensure that the highest quality conservative care has truly been rendered.
RESULTS OF SURGICAL TREATMENT OF SCIATICA
Despite the ever increasing incidence of spine surgery, indications for surgery and the optimal surgical approach for the treatment of herniated nucleus pulposus (HNP) are still controversial (3). Salenius and Laurent described results of 886 patients who had surgery for sciatica between the years 1960 and 1965, followed by questionnaire in 1971(6-11 yr of follow-up) (63). Of these, 56% reported no pain (cured), 36% were unchanged, and 8% reported that they were worse. Favorable prognostic factors included no previous back operation, presence of neurologic signs, a short period of sciatica, an occupation that did not involve heavy labor or an ability to change to lighter work, a minimal preoperative period of unemployment, a greater level of education, and younger age at the time of surgery. Hakelius attempted to follow a population of 583 patients with myelographically confirmed disc herniation, 330 of whom were treated conservatively over an average of 7 yr (31). Fewer “excellent” results were recorded in the nonoperative group. Long-term satisfaction regarding low back pain was 29 and 52% in the nonoperative and operative groups, respectively. Long-term satisfaction with sciatica reduction was 80 and 88% in the same respective groups. This study suffered from lack of randomization of subjects.
Dvorak et al. retrospectively identified 575 patients who underwent first-time surgery for HNP (19). Questionnaires were sent to all, with 371 responding (65% response rate). Follow-up ranged from 4-17 yr postoperation. Although the authors described imperfections in precise record retrieval, they reported that within this population 70% still had back pain, 23% experienced constant heavy pain, 45% described residual sciatica, 35% were still under treatment of some kind, 14% were receiving disability pensions, and 17% had undergone some repeat operation. Seventy-nine of the patients could not continue working, 64 patients complained of income loss, and 50 patients had a postoperative working capacity of only 50% baseline or less. They were able to identify “good” surgical candidates as those who were young (age 20-35), had symptoms of fewer than 6 mo duration, the presence of clear neurological findings, absence of tendon reflexes, and no abnormalities in psychological assessment. Caution is advised prior to acceptance of their data not only because of the retrospective analysis, but because of the initiation of the study group in the early 1970s when surgical techniques were not as refined as in the 1990s. Spengler's work reported results from lumbar discectomy to vary from 46 to 90% excellent outcomes(64,65). Variation was felt to be due to preoperative selection. Hurme and Alaranta followed 357 consecutive patients admitted for suspected disc herniation (36). Patients were administered preoperative clinical exams, electromyograms (EMGs), profiled for sports activities and psychosocial issues, and had surgery if they exhibited “absolute” indications for surgery, with another subgroup added due to failed conservative care. Best results were obtained when patients had surgery within the first 2 mo of sciatica and exhibited positive psychosocial factors (i.e., desire to return to work, being married). Ironically, these are factors that would portend a good chance for recovery in nonoperative candidates as well. A more recent review of surgical outcomes cites an overall success rate for relief of sciatica from macro- and microdiscectomy of 80-95% in the immediate postoperative period but emphasizes that the long-term outcomes of surgical and nonsurgical care appear to be similar (50).
Surgical treatment is not without complications. Following discectomy, another operation is sometimes necessary. Dvorak et al. reported that 17% of patients required a repeat operation (19). Rish reported recurrence of operation in 18%, while Cauchoix et al. described failed primary disc surgery in 6% of cases (11,59). The international average of recurrence approximates 15%(72). Overall, there appears to be a 10-fold increased risk of repeat surgery for disc prolapse within 10 yr after the first operation (8). Postoperative weakness and deconditioning are common. Rantanen et al. demonstrated type II fiber atrophy in multifidi of postsurgical patients who were still not faring well 5 yr postsurgery(57). Kahanovitz et al. evaluated abdominal and back muscle strength and endurance in 20 patients at least 1 yr after surgical discectomy (39). Almost every parameter on isometric and isokinetic performance showed at least a 30% decrease when compared to normal values. More severe problems in the postoperative period included cerebrospinal fluid (CSF) leak, infection (2.9%), thromboembolism (1.7%), and even death (0.2%) (66).
RESULTS OF NONOPERATIVE TREATMENT OF SCIATICA
The results of aggressive conservative care trials are also hard to interpret. The optimal time course of conservative care, the influence of specific types of disc herniations, the impact of psychosocial factors, the role of the regularity of therapeutic exercise performance, and the frequency with which maintenance programs must be performed have not been clearly established.
Numerous follow-up studies on conservatively treated patients show significant resolution of compression without surgical intervention. Maigne prospectively followed the natural evolution of lumbar disc herniation in 48 patients treated conservatively (47). Initial CT scans were performed during the acute phase of disc herniation, and follow-up scans were obtained after 1-48 mo of clinical “healing.” Radiographic disc bulge/herniation size at these two points in time was then compared. Of the 48 herniations, nine decreased in size by 25-50%, eight from 50 to 75%, and 31 decreased by 75-100%. The authors noted that the largest herniations had the greatest tendency to decrease in size.
The retrospective work by Saal et al. looked at 11 patients with disc extrusion (via CT scan) and radiculopathy, accompanied by leg pain and a positive straight leg raise at an angle of less than 60°(62). Following aggressive conservative care, patients were reimaged; 11% of the discs decreased in size by 0-50%, 36% decreased 50-75%, and 47% decreased by 75-100%. Bozzao et al. followed 98 conservatively treated patients with a magnetic resonance imaging (MRI) diagnosis of disc herniation, 69 of whom were reimaged at an average of 11 mo(7). Overall, 63% of the subjects demonstrated a reduction in herniation size by more than 30%, and 48% showed a reduction of more than 70%. The authors noted that the largest herniation decreased in size the most. Komori in a retrospective review of 77 patients with radiculopathy who underwent interval MRI of the lumbar spine confirmed that larger discs tended to regress in size more than smaller ones (46). The process of shrinkage appears to be associated with macrophage-mediated absorption and not fibrosis at the site of the disc herniation(37). In 1989, Saal and Saal published a retrospective cohort study on patients with disc herniations and radiculopathy(61). Patients had leg and back pain, positive straight leg raises at less than 60°, and positive CT scans. They underwent aggressive treatment including back school, spine stabilization exercise, and when necessary epidural steroid injections. Results were reported as 90% good or excellent, with 92% returning to work. There was 87% good or excellent results in patients who had extruded discs, with an 83% return-to-work rate. Four of the six failures had associated stenosis. A cohort-operated group did not fare better in the 1st yr.
Bush et al. looked at 165 consecutive patients with sciatica initially treated with serial caudal epidural steroid injections and, if not improved, selective nerve root blocks. Subjects underwent CT scans initially and at 1 yr(9). All conservatively treated patients made a satisfactory recovery, with an average decrease in visual analog pain scale of 94%. Partial or complete resolution was seen in 64/84 disc herniations (76%) and 7/27 bulging discs (26%); 86% of the clinical sciatica patients and radiologic nerve root entrapment patients were managed successfully, and 14% had decompressive surgery. Alaranta et al. followed 122 patients with nonoperative treated sciatica and 220 with operatively treated sciatica(2). After 1 yr, 91% of the patients who were operated on were better or much better, and 82% of the nonoperative group who had positive myelography were better or much better. Interestingly, only 52% of those in the nonoperative group who had a negative myelogram were better or much better. Ellenberg et al. followed 18 patients with definite disc herniation, neurological symptoms, and physical examination signs(20). The authors reported 78% resolution or improvement of HNP size over a 6- to 18-mo period. Resolution of discs (via CT scan) was not related to initial disc herniation size.
EXERCISE AND BEST REST FOR LOW BACK PAIN WITH OR WITHOUT SCIATICA
Both exercise and bed rest have been used in the treatment of acute and subacute low back pain. The Agency for Health Care Policy and Research (AHCPR) has established guidelines with respect to the utility and dosing of both bed rest and exercise in the setting of acute low back pain(1). This section highlights and critiques some of these issues.
In general, in patients with nonradicular low back pain, there does not appear to be a proven benefit to prolonged bed rest. The available literature is inadequate to make any definitive statement regarding the efficacy of or the appropriate amount of time of bed rest in patients (athletes or nonathletes) with disc herniations or clear radiculopathy. Given that 70-90% of patients have recurrent episodes of low back pain(5,32), an argument can be made that there are anatomic changes and functional biomechanical deficits that persist following low back injury and that these alterations need to be corrected to lessen the likelihood of further injury. This theory, however, awaits validation with detailed studies.
Studies cited by the AHCPR contribute to misconceptions about the“lack” of efficacy of exercise. It is important to recognize that absence of proof due to methodologic flaws in the available literature is not equivalent to proof of an absence of a beneficial effect of exercise. Studies by Coxhead et al. (1981), Gilbert et al. (1985), Evans et al. (1987), and Koes et al. (1991, 1995) failed to demonstrate a benefit for “exercise” over placebo treatment in acute low back pain(13,21,28,44,45). These studies were cited because they were randomized and prospective. Methodologic and statistical flaws were abundant in these reports, with absence of physical examination data, imaging data, and use of uniform exercise programs irrespective of the ill-defined underlying problem. Malmivaara et al. prospectively compared bed rest of 2 d, mobilizing exercises, and activity as tolerated in the treatment of patients with acute or recently exacerbated chronic low back pain (48). Patients with pain that extended down to the knee were included, but not patients with neurological deficits. The group that tolerated normal activities was used as the control group. Outcome and cost were assessed at 3 and 12 wk. At both the 3- and 12-wk marks the control group fared better with respect to duration of pain, intensity of pain, lumbar flexion, ability to work as measured subjectively, the Oswestry Back Disability Index, and the number of days absent from work(48). Closer examination of the study, though, reveals that only one session of physical therapy was provided for the“intervention group,” the only patients with a prior history of spine surgery were in the exercise group, and there were twice as many subjects with chronic low back pain in the exercise group(48). The work of Faas et al. concluded that exercise therapy for patients with acute low back pain had no advantage over care from the general practitioner (22). Unfortunately, once again, exercise programs were not based on any specific diagnosis and were not in response to physical examination findings. It is unreasonable to expect to observe specific benefits from an exercise program if no specific diagnoses were made prior to initiation of the treatment arm of a study.
Plowman, in her analysis of retrospective studies, indicated that there might be a link between trunk strength and low back pain(55). This conclusion was made with caution inasmuch as assessment of maximal strength in patients with pain is almost impossible, and apparent loss of strength in patients with long-term or previous episodes of low back pain could be attributed to deconditioning. The work of Donochin et al. examined hospital employees with greater than three annual episodes of low back pain, enrolling participants in lumbar flexion and pelvic tilt training versus back school and a control group (18). Those who participated in the exercise arm of the study did report fewer subsequent months of back pain and demonstrated increased abdominal“strength.” Gundewall et al., in a prospective randomized controlled trial using patients with and without back pain, reported fewer days of low back pain in patients given six exercise treatments a month emphasizing trunk-strengthening exercises (29). Bierring-Sorensen reported trunk extension endurance testing to be predictive for development of low back pain in previously asymptomatic individuals but not in those with previous low back pain (5). In a nonrandomized trial, Troup et al. were unable to demonstrate a positive predictive value of trunk endurance exercises in British workers(69). Johannsen et al. randomized 40 consecutive chronic low back pain patients to 3 months of either “endurance” exercises or “training including coordination” type exercise(38). Improvement in isokinetic back extension strength was significantly greater in the endurance group, and there was no intergroup difference with regard to disability score, although both groups improved(38). No control group was used and the study groups were poorly described, particularly with reference to how active the participants were on their own. In addition, 13 of the 40 participants dropped out of the study.
An additional problem when reviewing the “strength” literature relates to whether or not isokinetic measurements are useful. An extensive review by Newton and Waddell concluded that there was inadequate evidence to support preemployment isokinetic testing and that isokinetic tests did not help predict which initially asymptomatic individuals would be more likely to endure low back injury in the future (52). On the other hand, Reimer et al. suggested that the use of isokinetics to measure total work as part of a fitness test battery is a more useful tool in employee screening (58). That an endurance-type measure is more valuable intuitively makes more sense, but caution is advised as this study was retrospective and an actual statistical analysis of the variables examined was omitted from the body of the paper (58).
A final area of concern with the available literature is that the lack of an accurate definition of the subject populations' underlying spine problems has lead to the apparent inability to identify either flexion or extension exercises as particularly useful in the treatment of low back pain(22,28,44,45). This problem is exemplified in a recent paper by Dettori et al. (15). In this prospective study, soldiers with acute low back pain were randomized to flexion exercises and posture, extension exercises and posture, or to no exercise for 8 wk. Subjects were assessed at 1, 2, 4, and 8 wk after initiation of treatment. After 1 wk, both exercise groups exhibited decreased disability scores, fewer subjects with a positive straight leg raise, and a greater percentage of subjects able to return to work than the control group. However, a follow-up questionnaire administered 6-12 mo after subjects entered the study did not detect a significant difference between groups for recurrence of back pain. The authors concluded that either exercise was“slightly” more effective than none, but that it made no difference if flexion or extension exercises were utilized. Unfortunately, randomization of acute low back pain patients without regard to pathology or individual position/posture preferences probably leads to subjects who would have fared better with extension exercises (i.e., posterolaterally protruding disc with increased dural tension in flexion) being placed in the flexion group, and some of those who would have fared better with flexion exercises(i.e., facet synovitis) being assigned to the extension group. Both of these randomizations would lead to dilution of the potential treatment effect. Delitto et al. attempted to identify acute and subacute low back pain patients with or without radicular symptoms who might benefit from an extension-mobilization program (14). These patients were then randomized to either an extension motion with a sacroiliac mobilization or a flexion-type exercise program. A control group was not included. Modified Oswestry Low Back Questionnaires were administered at days 3 and 5 of treatment with the extension group improving more rapidly. However, study design did not permit distinction between the effect of the mobilizations versus the effect of the exercises (14).
In general, use of the available literature is inadequate to establish guidelines for the role of strength training in the prevention or treatment of acute low back pain. Variability in operational definitions as cited above, lack of specificity or distinction between flexion versus extension-based exercises, the use of patients with and without back pain in the same studies, and their inclusion of industry-based populations reveal a need for more carefully designed studies. Gearing “strength”- or“endurance”-type exercises toward the specific needs of the subject populations or the use of entirely pain-free populations followed longitudinally for the development of low back pain may also be more beneficial.
The literature surrounding “flexibility” is even worse than that for cardiovascular and strength training. Definitions vary greatly from study to study, and measurements are extremely hard to interpret. Cady et al. reported that workers who had poorer flexibility cost seven times more than the most “flexible” subjects with regard to work-related low back injury (10). The work of Battie et al. was unable to demonstrate a significant difference between the flexibility of those with and without low back pain (4). Bierring-Sorensen actually reported that those with better mobility were more likely to develop low back pain in the next year (5). Thus, the available literature is inadequate to determine guidelines and recommendations for flexibility training in acute low back pain.
EXAMPLES OF SPINE REHABILITATION STRATEGIES
In the 1990s it is unusual for a team of professionals treating low back pain to use only one type of treatment program. This section describes some of the popular options available today, such as spine stabilization, Williams, and McKenzie programs.
In spine stabilization exercises, the goal is to teach the patient how to find and maintain a “neutral spine” during activities of daily living(16,32,40,74). The neutral spine position is individual-specific and is the pelvic and spine posture that places the least stress upon the elements of the spine and supportive structures (16,40). In classic discogenic pain associated with a posterolateral disc herniation, the neutral spine will have an extension bias (32,40). In classic posterior element pain or central spinal stenosis, the neutral spine may have a mild flexion bias.
The starting position for the patient in acute and subacute pain is usually supine but may also be a side-lying position. Neutral spine is found by contracting abdominal and pelvic musculature to rock the pelvis anteriorly or posteriorly through the pain available range. The most functional position is in the midpoint of the range. Postural corrections incorporate the neutral spine concept. “Stabilization” is accomplished via teaching proper posture and mechanics and utilizing these potential stabilizing mechanisms:
- Intraabdominal pressure. Increases in intraabdominal pressure (utilization of a closed glottis) may transmit throughout the torso and reduce pressure placed upon the spine. This does not respond to exercise training(32,40).
- Thoracolumbar fascial (TLF) support system. The TLF has fibers that are derived from the latissimus dorsi, abdominal oblique, and gluteus maximus muscles. Strengthening of these muscular attachments helps to stabilize the spine as the TLF spans many spine levels and is activated by flexion of the spine (32,40).
- Hydraulic amplifier mechanism. The TLF is enhanced by the contraction of the erector spinae which it overlies (32,40).
- Posterior ligamentous system. The posterior ligaments engage when the spine is flexed. This reduces compressive forces placed upon the other supportive structures (32,40).
- Muscular support mechanism. This mechanism utilizes the small intersegmental muscles between consecutive vertebrae(32,40).
Progression of treatment is a function of addressing inflexibilities and weaknesses while attempting to teach these “new” exercises. More and more dynamic challenges are added as the patient demonstrates the ability to maintain a neutral spine without worsening of lumbar spine symptoms(32,61,62).
McKenzie exercises are techniques that identify postures and motions that “centralize” radicular/referred low back pain(32,74). Although extension-biased exercises are commonly used in classic radiculopathy associated with a posterolaterally protruding disc, these exercises are not solely extension exercises. Potential mechanisms of improvement with this type of therapy include reduction of the disc herniation and reduction of dural tension. Prior to incorporating extension-based exercises, lateral trunk shifts must be corrected. McKenzie's patient groupings include those problems due to “posture,”“dysfunction,” and “derangement.” Patients in the postural group develop pain from prolonged end range positioning, in the presence or absence of true pathology. Those in the dysfunctional group have shortened connective tissue and experience pain when it is stretched (i.e., scar tissue). Those in the derangement group have symptoms in response to both static loading and repeated movements. Derangement of the disc is felt to be responsible. Centralization of pain is considered to be a positive response, whereas persistence of radiculopathy or worsening of symptoms necessitates reevaluating the patient. Patients with central disc herniations may not fare well with extension bias exercises.
Williams' flexion exercises emphasize a different approach(32,74). Flexion bias exercises (i.e., posterior pelvic tilt, double knee to chest) were often prescribed with the thought that discogenic symptoms could be reduced in this manner. Flexion was theorized to decrease loading of the posterior portion of the disc (typically protruded) and to widen the neuroforamen. However, patients with acute disc herniations may have exacerbation of symptoms from increased intradiscal pressure. Alternatively, patients with posterior element pain and central spinal stenosis may benefit from this technique. Patients with presumed spinal stenosis who do not improve with flexion-based exercises should be evaluated carefully for foraminal stenosis.
Rehabilitation may be broken into phases. During the acute phase, focus is upon reducing pain from injured or inflamed tissue. This may be accomplished via relative rest, medications, and/or sparing use of modalities. At this point, physical or occupational therapy is utilized sparingly. One to two sessions focused upon finding positions of comfort, correction of trunk shifts, teaching neutral spine position (see later sections), and pain-free methods of position change, i.e., lie to sit, sit to stand, and vice versa, are suggested. The initial movement patterns are based upon presumed pathology and pain pattern and pain centralization (i.e., the McKenzie method). Extension bias is most commonly used in a discogenic process with reduction of symptoms during repetitive extension motion pattern testing and centralization of pain with extension. Extension exercises may reduce intradiscal pressure and allow anterior migration of the nucleus pulposus. However, they can exacerbate the symptoms from posterior element irritation or from a central disc herniation. Flexion bias is most commonly used in posterior element pain with pain reduction in symptoms with repetitive flexion. Flexion exercises may reduce compressive forces on the facet joints. They may increase intradiscal pressure and increase radicular symptoms in the posterolaterally protruding disc but may actually be more comfortable for the patient with a centrally located disc or an upper lumbar spine disc protrusion.
During the subacute phase, connective tissue restrictions are addressed. Manual techniques are used to increase soft tissue distensibility along the planes of physiologic stress to promote proper alignment of connective tissue fibers during the healing and remodeling process. Myofascial techniques are utilized to apply pressure and shear forces to the fascial layers in order to improve elasticity and produce less restricted, painful movement. The theory behind the need for these techniques assumes that fascia absorbs shock, separates and supports muscles, and helps in the transmission and attenuation of mechanical forces (32,40). It also assumes that loss of normal fascial gliding and increased cross-linking of fibers result in loss of myofascial system mobility with secondary loss of segmental bony mobility and flexibility. The purpose of mobilization is to restore optimal joint mobility by applying forces at individually targeted specific non-soft tissue motion segment levels. These are graded I-IV, depending upon the force and depth of the applied load (23,32). Grade I and II are referred to as oscillations. Grade III and IV represent larger amplitude forces that move a joint into its restricted range and provide stretch. Grade V manipulations are of a high-velocity, low-amplitude nature, taking the joint to its end range of physiologic motion(32,40).
During the recovery phase (typically the most lengthy portion of the rehabilitation program), emphasis is on restoration of function. This phase combines both supervised physical therapy and a home program, with the majority being home based. Appropriate tissue loading is emphasized in the exercises. Strengthening of the abdominal and gluteal muscle groups (muscles which attach to the thoracolumbar fascia) is stressed. Goals include development of range of motion on the more affected side to be within 10% of the unaffected side and strength to be within 20% of the unaffected side.
The maintenance phase represents the final phase and is the basis of the prevention program as well. Eccentric (lengthening) muscle-strengthening exercises, including dynamic conditioning exercises with inflatable gym ball exercises, are emphasized, and sports-specific training is incorporated.
“Completion” of a spine rehabilitation program is signified by the patient's demonstration of absence of signs or symptoms of the original injury, full pain-free range of motion, normal strength and flexibility, normal mechanics for sports activities, and the ability to perform sport-specific skills. Under any circumstance, the philosophy adhered to is that resolution of symptoms alone is inadequate and that restoration of optimal function is the goal.
A PROPOSED APPROACH TO REHABILITATION OF ACUTE DISCOGENIC LOW BACK PAIN
Although research trials have failed to stand up to scrutiny, there is mounting clinical evidence that there is a purpose to rehabilitation and conditioning exercises in patients with low back pain, even in the face of disc herniation with radiculopathy(32,60-62). Exercise and physical and occupational therapy regimens can be beneficial and cost effective when they are constructed to address specific conditions associated with accurate clinical diagnoses. A multifaceted approach appears to be more effective(32,61,62,74). Therapists who have a broad range of skills in addressing low back pain have a greater likelihood of helping patients who have different problems. Precisely prescribed rehabilitation programs tailored toward specific anatomic and biomechanical deficits are needed to optimize function following low back injury. Clinicians who are capable of modifying and upgrading regimens on the basis of changes in the patient's clinical picture rather than merely following algorithmic flow charts are needed to direct rehabilitation programs.
Goals of physical therapy include reduction of pain intensity and length of episodes. This is typically accomplished by correction of abnormal skeletal shifts and posture, reduction of the associated heightened muscle tone, and establishment of a comfortable body position(32,40,74). For typical discogenic pain(i.e., from a posterolaterally herniated disc), comfort incorporates an extension-based spine posture. For a midline disc, a mild flexion-based or neutral program is typically utilized. The concept of the “neutral spine” is also utilized, as the patient is taught how to control hip, pelvic, and abdominal musculature so as to reduce forces on the supportive muscles and ligaments about the spine(32,40,74). The purpose of these activities is to facilitate return to work. A key to achieving a rapid progression in the exercise program is managing pain. This is accomplished via medications and or judicious use of pain-relieving modalities. This permits the treating physician and therapist to begin actively correcting associated biomechanical flaws. For example, typical problem areas include tightness/inflexibility of the hamstrings, hip flexors, and hip rotators and weakness of the hip extensors, abdominal oblique muscles, and lower rectus abdominis. At the same time the patient is instructed in proper ergonomics and correct lifting techniques (i.e., use of legs rather than flexing/extending lumbar spine emphasizing utilization of gluteal muscles and attachments of the thoracolumbar fascia) and educated to prevent reinjury.
The initial period is also critical for identification of those patients who do not benefit from exercise alone. Those acute/subacute low back pain patients with persistent radicular symptoms following 6 wk of appropriately prescribed physical therapy need to be reevaluated to assess the need for radiologic imaging (x-ray, CT, MRI, etc), the need for alteration of an oral medication regimen, the need for diagnostic/therapeutic injection, or the need for surgical evaluation (3). Those patients exhibiting persistent “centralized” low back pain following 6 wk of appropriately prescribed therapeutic exercise also require reevaluation to assess the need for radiologic evaluation, alteration of the oral medication regimen, and in particular whether pharmacologic intervention is needed to restore sleep. For those exhibiting chronic nonradicular low back pain without clinical improvement following a 6-wk trial of therapeutic exercise, the clinician needs to be vigilant for the presence of signs of nonorganic features and to be prepared to assess for “pain” issues rather than prescribe more therapy.
Provision of true aggressive conservative care is essential to determine if a patient falls into a surgical or nonsurgical category. The majority of spine surgeries follow failed aggressive care rather than development of cauda equina syndromes. Nonspecific physical therapy programs lead to poorer outcomes (22) and, by default, more spine surgeries. When patients appear to have “plateaued” in their progress, the therapist and physician must reestablish the goals and objectives of the program. This may lead to further diagnostic testing (ie EMG, MRI) to further define pathology. Fluoroscopically guided, contrast enhanced injections can provide diagnostic and therapeutic benefit and will help patient progress in many cases.
Low back rehabilitation may be described as an orderly progression of therapeutic activities designed to reduce pain, inflammation, and facilitate healing of injured tissues, to reestablish flexibility, range of motion, strength and endurance, normal or near normal biomechanics of spine, pelvic, and hip motion and allow the individual to return to his or her activities of daily living including gainful employment, through development of work/sports specific regimens and maintenance programs. Exercises are based upon scientific evidence when available. An assumption is that a complete biomechanical and neurological evaluation will be completed prior to initiation of the program. A kinetic chain approach should be adhered to. Unless proven otherwise, all muscle strength and flexibility imbalances need to be considered relevant. Although algorithmic approaches to spine rehabilitation are attractive to many practitioners, the best programs are ultimately tailored to meet the patient's individual needs. Therapists who have a broad range of skills in addressing low back pain have a greater likelihood of helping patients who have different problems. Therapeutic spine injections (eg. lumbar epidural steroid injections) can also be a valuable component of spine rehabilitation, but in most instances are performed to facilitate the therapeutic exercise program(32,62).
CONCLUSIONS AND AREAS FOR FUTURE INVESTIGATION
- Relief from sciatica-type symptoms associated with lumbar disc herniation may be achieved more rapidly with surgical intervention than with conservative care. However, there is no evidence that there is an advantage of surgical treatment in the long run.
- The available literature does not provide evidence that neurologic recovery in particular motor strength, is hastened by surgical disc excision.
- Immediately after surgery, results are highly satisfactory, particularly if the patient has a predominance of sciatic pain and has experienced pain for less than 6 mo time. Thus, avoiding protracted pain is essential from a patient management standpoint. However, who should have surgery and at what time cannot be definitively answered by the available literature(3).
- The decision to commit to aggressive conservative or surgical management of discogenic pain in the elite athlete is complicated by numerous factors. The most pressing issue is the length of time the athlete can allow for conservative care. Younger athletes have more of their career ahead of them and may be willing to dedicate 6-9 mo of time conditioning their spine and learning appropriate and protective spine postures and mechanics. Collegiate or professional athletes who are on the verge of championship years or large economic reward may view surgery as the fastest way to try to secure the future. Older athletes may not be able to return from significant single-level or multilevel discal injury regardless of the treatment mode. The likelihood of operative success with intention to return to sport must be looked at carefully and on an individual athlete by athlete basis.
- The patient with a lumbar disc herniation and concomitant pathology, such as spondylolysis or spinal stenosis, must be viewed differently than the patient with an isolated lumbar disc herniation.
- Under any circumstance, surgery should not be pursued unless the long-term consequences of such intervention are thoroughly discussed with the athlete(and family when appropriate) and the athlete comes to the understanding that spine surgery does not constitute a restoration of normal anatomy and biomechanics.
- There is no proven benefit of prolonged bed rest (i.e., greater than 4 d) in acute low back pain without radiculopathy. The literature does not provide adequate information to establish guidelines for the amount of rest an athlete requires in the face of proven acute disc injury with or without radicular symptoms.
- Based upon literature review, aerobic fitness may be mildly protective against low back injury and low back pain. Acute low back pain results in reduction of physical activity which leads to reduction of aerobic fitness. This problem is accentuated if low back pain occurs on a recurrent basis. This may have significant consequences with regard to sports performance and, in the long run, cardiovascular risk. On this basis alone, it is recommended that aerobic exercises be incorporated in the spine rehabilitation program as early as possible. Longitudinal studies that include asymptomatic cohort groups need to be performed to see if the presence of back pain is associated with greater deterioration of aerobic fitness over time.
- Operational definitions of strength, power, and endurance are inconsistent across studies, making generalized statements about the efficacy of“strength” training or “strength” testing difficult at best. Nevertheless, there appears to be at least a mild relationship between trunk muscle “strength” and low back pain.
- The current literature neither supports the notion that flexibility training confers protection against the development of low back pain nor proves that it is essential in the treatment of acute low back pain.
- Although back pain and sciatica superficially appear to be episodic, they actually represent points along a continuum and require lifelong management to ensure long-term sports participation.
1. Agency for Health Care Policy and Research.Clinical Practice Guidelines for Acute Low Back Pain
2. Alaranta, H., M. Hurme, S. Einola, B. Falck, et al. A prospective study of patients with sciatica. Spine
3. Andersson, G. B. J., M. D. Brown, J. Dvorak, R. Herzog, et al. Consensus summary on the diagnosis and treatment of lumbar disc herniation. Spine
4. Battie, M. C., S. M. Bigos, L. D. Fisher, et al. A prospective study of the role of cardiovascular risk factors and fitness in industrial back pain complaints. Spine
5. Bierring-Soerersen, F. Physical measurements as risk indicators for low back trouble over a one year period. Spine
6. Boden, S. D., D. O. Davis, T. S. Dina, N. J. Patronas, and S. Wiesel. Abnormal magnetic resonance scans of the lumbar spine
in asymptomatic subjects. J. Bone Joint Surg. Am.
7. Bozzao, A., M. Gallucci, C. Masciocchi, I. Aprile, A. Barile, and R. Passariello. Lumbar disk herniation: MR imaging assessment of natural history in patients treated without surgery
8. Bruske-Holhfeld, I., J. L. Merritt, B. M. Onofrio, H. H. Stockington, et al. Incidence of lumbar disc surgery
9. Bush, K., N. Cowan, D. E. Katz, and P. Gishen. The natural history of sciatica associated with disc pathology. Spine
10. Cady, L. D., P. C. Thomas, and R. J. Karwasky. Program for increasing health and physical fitness of fire fighters. J. Occup. Med.
11. Cauchoix, J., C. Ficat, and B. Girard. Repeat surgery
after disc excision. Spine
12. Cavanaugh, J. M. Neural mechanisms of idiopathic low back pain. In: Low Back Pain: A Scientific and Clinical Overview
, J. N. Weinstein and S. L. Gordon (Eds.). Rosemont, IL: American Academy of Orthopedic Surgeons, 1996, pp. 583-605.
13. Coxhead, C. E., T. W. Meade, H. Inskip, and W. R. S. North. Multicentre trial of physiotherapy in the management of sciatic symptoms. Lancet
14. Delitto, A., M. T. Cibulka, R. E. Erhard, R. W. Bowling, and J. A. Tenhula. Evidence for use of an extension-mobilization category in acute low back pain syndrome: a prospective validation pilot study.Phys. Ther.
15. Dettori, J. R., S. H. Bullock, T. G. Sutlive, R. J. Franklin, and T. Patience. The effects of spinal flexion and extension exercises and their associated postures in patients with acute low back pain.Spine
16. DeWerd, J. Stabilization exercises for the aging athlete. J. Back Musculoskel. Rehabil.
17. Doita, M., T. Kanatani, T. Harada, and K. Mizuno. Immunohistologic study of the ruptured intervertebral disc of the lumbar spine
18. Donochin, M., O. Woolf, L. Kaplan, and Y. Floman. Secondary prevention of low-back pain: a clinical trial. Spine
19. Dvorak, J., M.-H. Gauchat, and L. Valach. The outcome of surgery
for lumbar disc herniation. A 4-17 years' follow-up with emphasis on somatic aspects. Spine
20. Ellenberg, M. R., M. L. Ross, J. C. Honet, M. Schwartz, G. Chodoroff, and S. Enochs. Prospective evaluation of the course of disc herniations in patients with proven radiculopathy. Arch. Phys. Med. Rehabil.
21. Evans, C., J. R. Gilbert, W. Taylor, and A. Hildebrand. A randomized controlled trial of flexion exercises, education and bed rest for patients with acute low back pain. Physiother. Can.
22. Faas, A., A. W. Chavannes, J. T. M. van Eijk, and J. W. Gubbels. A randomized, placebo-controlled trial of exercise
therapy in patients with acute low back pain. Spine
23. Farrell, J. P. and G. M. Jensen. Manual therapy: a critical assessment of role in the profession of physical therapy.Phys. Ther.
24. Frymoyer, J. W. Back pain and sciatica. N. Engl. J. Med.
25. Garfin, S. R., B. L. Rydevik, R. A. Brown, and D. J. Saratoris. Compressive neuropathy of spinal nerve roots. A mechanical or biological problem? Spine
26. Garfin, S. R., B. Rydevik, B. Lind, and J. Massie. Spinal root compression. Spine
27. Gargan, M. F. and J. C. T. Fairbank. Anatomy of the spine. In: The Spine in Sports
, R. G. Watkins (Ed.). St. Louis: Mosby Year Book, 1996, pp. 2-12.
28. Gilbert, J. R., D. W. Taylor, A. Hildebrand, and C. Evans. Clinical trial of common treatments for low back pain in family practice. Br. Med. J.
29. Gundewall, B., M. Liljeqvist, and T. Hansson. Primary prevention of back symptoms and absence from work. Spine
30. Hainline, B. Low back injury. Clin. Sports Med.
31. Hakelius, A. Prognosis in sciatica. Acta Orthop. Scand. Suppl.
32. Herring, S. A. and S. M. Weinstein. Assessment and nonsurgical management of athletic low back injury. In: The Lower Extremity & Spine in Sports Medicine
, J. A. Nicholas and E. B. Hershman(Eds.; 2nd Ed.). St. Louis: Mosby-Year Book, 1995, pp. 1171-1197.
33. Herzog, R. J. The radiologic assessment for a lumbar disc herniation. Spine
34. Hitselberger, W. E. and R. M. Witten. Abnormal myelograms in asymptomatic patients. J. Neurosurg.
35. Holt, E. P., Jr. The question of lumbar discography.J. Bone Joint Surg. Am.
36. Hurme, M. and H. Alaranta. Factors predicting the result of surgery
for lumbar intervertebral disc herniation. Spine
37. Ito, T., M. Yamanda, F. Ikuta, T. Fukuda, et al. Histologic evidence of absorption of sequestration-type herniated disc.Spine
38. Johanssen, F., L. Remvig, P. Kyger, et al. Exercises for chronic low back pain: a clinical trial. J. Sports Phys. Ther.
39. Kahanovitz, N., K. Viola, and M. Gallagher. Long-term strength assessment of postoperative diskectomy patients. Spine
40. Kaul, M. P. and S. A. Herring. Rehabilitation of lumbar spine
injuries in sports. Phys. Med. Rehabil. Clin. North Am.
41. Kawakami, M. and J. N. Weinstein. Associated neurogenic and nonneurogenic pain mediators that probably are activated for nociceptive input. In: Low Back Pain: A Scientific and Clinical Overview
, J. N. Weinstein and S. L. Gordon (Eds.). Rosemont, IL: American Academy of Orthopedic Surgeons, 1996, pp. 265-274.
42. Kawakami, M., J. N. Weinstein, K. F. Spratt, et al. Experimental lumbar radiculopathy: Immunohistochemical and quantitative demonstrations of pain induced by lumbar nerve root irritation of the rat.Spine
43. Kibler, W. B. Clinical aspects of muscle injury.Med. Sci. Sports Exerc.
44. Koes, B. W., L. M. Bouter, H. Beckerman, G. J. M. G. van der Heijden, and P. G. Knipschild. Physiotherapy exercises and back pain: a blinded review. Br. Med. J.
45. Koes, B. W., L. M. Bouter, and G. J. M. G. van der Heijden. Methodological quality of randomized clinical trials on treatment efficacy in low back pain. Spine
46. Komori, H., K. Shinomiya, O. Nakai, I. Yamaura, S. Takeda, and K. Furuya. The natural history of herniated nucleus pulposus with radiculopathy. Spine
47. Maigne, J-Y., B. Rime, and B. Deligne. Computerized tomographic follow-up study of forty-eight cases of nonoperatively treated lumbar intervertebral disc herniation. Spine
48. Malmivaara, A., U. Hakkinen, T. Aro, et al. The treatment of acute low back pain-bed rest, exercises, or ordinary activity?N. Eng. J. Med.
49. McCarrol, J. R., J. M. Miller, and M. Ritter. Lumbar spondylosis and spondylolisthesis in college football players. Am. J. Sports Med.
50. McCulloch, J. A. Focus issue on lumbar disc herniation: macroand microdiscectomy. Spine
51. Micheli, L. J. Back injuries in gymnastics. Clin. Sports Med.
52. Newton, M. and G. Waddell. Trunk strength testing with isomachines: part I: review of a decade of scientific evidence.Spine
53. Olmarker, K., G. Byrod, M. Cornefjord, C. Nordborg, and B. Rydevik. Effects of methylprednisolone on nucleus polposus induced nerve root injury. Spine
54. Pearce, J., and J. M. H. Moll. Conservative treatment and natural history of acute lumbar disc lesions. J. Neurol. Neurosurg. Psychiat.
55. Plowman, S. A. Physical activity, physical fitness and low back pain. In Holloszy J. O. (Ed.): Exercise
and Sport Sciences Reviews 1992;20:222-242.
56. Postacchini, F. Spine Update: Results of surgery
compared with conservative management for lumbar disc herniations.Spine
57. Rantanen, J., M. Hurme, B. Falck, H. Alaranta, et al. The lumbar multifidis five years after surgery
for a lumbar intervertebral disc herniation. Spine
58. Reimer, D. S., B. D. Halbrook, P. H. Dreyfuss, and C. Tilbetti. A novel approach to preemployment worker fitness evaluations in a material-handling industry. Spine
59. Rish, B. L. A critique of the surgical management of lumbar disc disease in private neurosurgery practice. Spine
60. Saal, J. A. Natural history and nonoperative treatment of lumbar disc herniation. Spine
61. Saal, J. A. and J. S. Saal. Nonoperative treatment of herniated lumbar intervertebral disc with radiculopathy. Spine
62. Saal, J. A., J. S. Saal, and R. J. Herzog. The natural history of lumbar intervertebral disk extrusions treated nonoperatively.Spine
63. Salenius, P. and L. E. Laurent. Results of operative treatment of lumbar disc herniation. Acta Orthop. Scand.
64. Spengler, D. M. Results with limited excision and selective foraminotomy. Spine
65. Spengler, D. M. and C. W. Freeman. Patient selection for lumbar discectomy. An objective approach. Spine
66. Sprangfort, V. Lumbar disc herniation. A computer-aided analysis of 2504 operations. Acta Orthop. Scand. Suppl.
67. Stith, W. J. Exercise
and the intervertebral disc.Spine: State Art Rev.
68. Takahashi, H., T. Soguro, Y. Okazima, M. Motegi, Y. Okada, and T. Kakiuchi. Inflammatory cytokines in the herniated disc of the lumbar spine
69. Troup, J. D. G., T. K. Foreman, C. E. Baxter, and D. Brown. The perception of back pain and the role of physiological tests of lifting capacity. Spine
70. Von Korff, M. and K. Saunders. The course of back pain in primary care. Spine
71. Weber, H. Lumbar disc herniation. A controlled, prospective study with ten years of observation. Spine
72. White, A. A. Overview of and clinical perspective on low back pain syndrome. In: Spine Update 1984
, H. K. Genant (Ed.). San Francisco: Radiology Research and Education Foundation, 1984, pp. 127-130.
73. Wiesel, S. W., N. Tsourmas, H. L. Feffer, C. M. Citrin, and N. Patronas. A study of computer-assisted tomography. I. The incidence of positive CAT scans in an asymptomatic group of patients. Spine
74. Young, J. L., J. M. Press, and A. J. Cole. Physical therapy options for lumbar spine
pain. In: A. J. Cole and Herring, S. A.(Eds.). The Low Back Pain Handbook
. Philadelphia: Hanley & Belfus, 1996, pp. 125-140.