Published: August 2010
Low back pain can best be described in terms of specific accompanying features. Low back pain is acute if it has a duration of about 1 month or less. Chronic low back pain is usually defined by symptoms of two months or more. Both acute and chronic low back pain can be further defined by the presence or absence of neurologic symptoms and signs. Nonspecific or nonradicular low back pain is not associated with neurologic symptoms or signs. In general, the pain is localized to the spine or paraspinal regions (or both) and does not radiate into the leg. In general, nonspecific low back pain is not associated with spinal nerve root compression. Nonspecific low back pain might or might not be associated with significant pathology on magnetic resonance imaging (MRI) and is often a result of simple soft tissue disorders such as strain, but it can also be caused by serious medical disorders arising in the bony spine, parameningeal, or retroperitoneal regions. See Table 1 for the differential diagnosis for low back pain.
|Mechanical Causes||Nonmechanical Causes||Causes of Referred Pain|
|Idiopathic (sprain, strain)||Malignancy||Pelvic disease (prostatitis, endometriosis, pelvic inflammatory disease)|
|Spondylosis (disk, annulus, facet)||Infection||Renal disease (kidney stones, pyelonephritis, perinephric abscess)|
|Compression fracture||Inflammatory spondyloarthropathy (ankylosing spondylitis, psoriatic spondylitis, Reiter’s syndrome, inflammatory bowel disease)||Aortic aneurysm|
|Traumatic fracture||Osteochondrosis||Gastrointestinal disease (pancreatitis, cholecystitis, penetrating ulcer)|
|Alignment disorders (kyphosis, scoliosis, spondylolisthesis)||Paget’s disease of bone|
© 2002 The Cleveland Clinic Foundation
Low back pain accompanied by spinal nerve root damage is usually associated with neurologic signs or symptoms and is described as radiculopathy. There is usually pathologic evidence of spinal nerve root compression by disk or arthritic spur, but other intraspinal pathologies may be present and are often apparent on an MRI scan of the lumbosacral spine.
Low back pain is second only to upper respiratory illness as a cause for visiting a physician.1 Up to two thirds of the population has low back symptoms at some time in their lives. In 1995 there were about two worker’s compensation claims for low back pain for every 100 workers. Seventy-five percent of patients with acute low back pain are back to work within 1 month of the onset of symptoms, and only 5% are disabled for more than 6 months.2 However, among those with continuing pain 6 to 10 weeks after onset, most still have some symptoms at 1 year.3
Among persons with chronic low back pain without neurologic deficits, a number of factors play a role in the length of disability. Recurrent low back pain and prolonged disability tend to correlate with prior history of low back pain, advancing age, job dissatisfaction, emotional distress, heavy or repetitive lifting and physical work, prolonged sitting or standing, and the presence of a worker’s compensation claim or pending litigation.4
Lumbosacral radiculopathy and radicular low back pain are less common than nonspecific low back pain. L5 radiculopathy is the most common lumbosacral radiculopathy, usually produced by disk herniation between the fourth and fifth lumbar vertebral bodies. S1 radiculopathy is the next most common, followed by L3 to L4 radiculopathy.
The pathophysiology of nonradicular low back pain is usually indeterminate. In fact, one of the defining features of this disorder is its nonspecific etiology. Pain can arise from a number of sites, including the vertebral column, surrounding muscles, tendons, ligaments, and fascia. Stretching, tearing, or contusion of these tissues can occur after sudden unexpected force applied to the spine from events such as heavy lifting, torsion of the spine, and whiplash injury. Whether muscle spasm is a significant etiology of lumbar spine pain, either as cause or effect of back injury, has not been proved.
The pathophysiology of radicular spine pain and lumbosacral radiculopathy is usually more obvious. Disk herniation through the annulus fibrosis does not in itself produce pain, but compression by disk of the dural lining around the spinal nerve root sleeve is one likely explanation for the back pain associated with acute disk herniation. This is also likely to contribute to the pain from spinal nerve root compression from arthritic spurs at degenerated facet and uncovertebral joints. Compression can directly stretch nociceptors in dura or nerve root sleeve tissues, but ischemia from compression of vascular structures, inflammation, and secondary edema is also likely to play a role in some cases.
History and physical examination are critical to the diagnosis and thus to the formulation of a rational approach to management. The following briefly summarizes the major points.
The medical history should focus on both triggering and alleviating factors, as well as on the character of the pain. Signs and symptoms such as increased pain with Valsalva maneuvers, straight-leg-raising symptoms, the tendency for the pain to radiate into the buttock or leg, the presence of weakness or sensory deficit, and bowel or bladder urgency or incontinence are associated with neurologic causes of low back pain.
The history should also explore factors that increase the likelihood of an underlying systemic disorder as a cause. These include advanced age, history of cancer, unexplained weight loss, use of injected drugs, chronic infection, prolonged duration of pain, pain that does not respond to rest or recumbency, and failure to respond to previous therapy. Box 1 reviews these factors.
|Box 1 Warning Signals of Systemic Disease Underlying Back Pain|
|Prior history of malignancy|
|Unexplained weight loss|
|No pain relief with bed rest|
|Pain duration greater than 4 to 6 weeks|
|Failure to respond to standard therapies|
|History of intravenous drug use|
|Urinary tract infection|
|Prolonged corticosteroid use|
Waddell and colleagues have described a number of historical features that point to nonorganic causes for low back pain, predicting delayed recovery and suggesting the need for a multidisciplinary approach to treatment.5 These are reviewed in Box 2.
|Box 2 Historical Symptoms Suggesting Nonorganic Causes of Back Pain|
|Pain at the tip of the tailbone|
|Whole-leg pain in global distribution|
|Whole-leg numbness in a global distribution|
|Sudden give-way weakness of the leg|
|Absence of even brief periods of relative pain relief|
|Failure or intolerance of numerous treatments|
|Numerous urgent care visits or hospitalizations for back pain|
Data from Waddell G, Bircher M, Finlayson D, Main CJ: Symptoms and signs: Physical disease or illness behaviour? BMJ (Clin Res Ed) 1984:289:739-741.
A general examination should be performed to identify potential systemic disorders, such as rheumatologic disease, skin disease, or bone deformities. The spine should be inspected for alignment, curvature, range of motion, focal tenderness, and overlying skin abnormalities such as a tuft of hair or pore. Mechanical maneuvers to elicit radicular and hip-joint symptoms should be considered, including straight-leg raising, reverse straight-leg raising, Patrick’s test, and Lasègue’s sign.
A careful neurologic examination should be undertaken to exclude motor and sensory deficits. Muscle strength in the L2 through S1 myotomes should be examined. The sensory examination should include soft-touch and pain sensation in the same segmental distributions. Muscle stretch reflexes should be elicited at the knee for the L3 to L4 segment and at the ankle for the S1 segment, and they can also be performed in the posterior thigh at the tendinous insertion of internal hamstrings for the L5 segment.
Waddell and colleagues have also described a number of findings on the physical examination that point to nonorganic causes for low back pain, predicting delayed recovery and suggesting the need for a multidisciplinary approach to treatment (Box 3).6
|Box 3 Signs on the Physical Examination Indicating Nonorganic Causes of Low Back Pain|
|Superficial tenderness over the lumbar region to light touch|
|Exacerbation of pain by applying a few pounds of pressure with the hands to the top of the head|
|Exacerbation of pain by simulated rotation of the spine|
|Ability to sit up straight from a supine position, but intolerance of the straight-leg-raising test|
|Nonanatomic distribution of sensory changes|
Data from Waddell G, McCulloch JA, Kummel E, Venner RM: Non-organic physical signs in low-back pain. Spine 1980;5:117-125.
The appropriate evaluation and management of low back pain rest on a few basic principles. First, it is necessary to determine whether the symptoms are caused by nerve root involvement. Second, it is necessary to determine whether the acute or chronic spine pain is related to a serious underlying medical illness that is manifesting itself as spine pain. After an initial assessment of the likely cause of the symptoms, the spine pain can then be treated.
A number of diagnostic tests can help pinpoint the cause of low back pain. Which test is selected depends on a host of factors uncovered during the history and physical examination. The diagnostic tests are described next.
Routine spine radiographs are of limited value because they visualize only bony structures. Guidelines from the U.S. Agency for Health Care Policy and Research (AHCPR) indicated value of routine spine radiographs for acute low back pain in the following settings: acute major trauma, minor trauma associated with risk of osteoporosis, risk of spinal infection, pain that does not respond to rest or recumbency, and history of cancer, fever, or unexplained weight loss.7 They may also be of value in assessing spinal alignment and rheumatologic disorders of bone.
Computed tomography (CT) and magnetic resonance imaging (MRI) are sensitive tools for evaluating neural structures such as spinal nerve roots and the spinal cord, and they can visualize soft tissue structures within the spinal canal. Of the two modalities, MRI resolution for neural structures is superior to CT. In the absence of motor, sensory, or autonomic deficits, and in the absence of significant trauma, infection, or malignancy, the American Academy of Neurology guideline recommends nonsurgical therapy before these techniques are used in patients with uncomplicated acute low back pain of less than 7 weeks’ duration.8 Patients with acute neurologic deficits associated with low back pain should be considered for MRI or CT of the lumbosacral spine unless surgery and invasive therapeutic options are not indicated.
CT myelography employs traditional myelography (intrathecal instillation of a radiocontrast agent, followed by routine spinal radiographs) followed by CT of the spine. It may be of value when MRI studies are inconclusive, especially in the assessment of the relation between spinal nerve roots and the bony neural foramina. Although this technique involves an invasive procedure, it is the study of choice when there are absolute contraindications to the use of MRI (presence of an implanted electronic device such as cardiac pacemaker or metal object in the inner ear, eye, or brain, such as an aneurysm clip).
Electrodiagnostic tests assess the neurophysiologic function of peripheral nerves and can identify the presence of various forms of nerve fiber damage. Electrodiagnostic tests (nerve conduction studies and needle electromyography [EMG]) are most useful in the presence of a motor deficit on neurologic examination. Nerve conduction studies are indicated primarily to exclude other neuromuscular disorders that can mimic radiculopathy, such as peripheral polyneuropathy and mononeuropathies. The H-reflex can be a useful nerve conduction study when assessing for the presence of an S1 radiculopathy. The needle electrode examination is most likely to be useful in the presence of clinical weakness. This procedure will help distinguish weakness due to spinal nerve root damage from other causes of weakness identified on the physical examination, such as other neuromuscular disorders, central nervous system disorders, and non-neurologic causes of weakness (pain, malingering). The needle electrode examination should be performed only after at least 3 weeks have passed since the onset of weakness because fibrillation potentials (the major manifestation of acute denervation) do not reliably develop before that time.
Electrodiagnostic testing may be of value in the assessment of patients with postsurgical deficits, multisegmental neurologic deficits, or multilevel intraspinal structural changes. Such patients present with complicated clinical and neuroimaging evidence, and electrodiagnostic testing might clarify issues of the location, activity, and severity of spinal nerve root disease.
The initial management of acute spine pain must be directed toward determining if a serious neurologic condition exists. If there is a history of recent trauma or serious underlying medical illness, more-aggressive evaluation is warranted. The presence of acute and progressive neurologic dysfunction is an urgent medical problem. This is especially so when there is clinical evidence of bilateral neurologic dysfunction, increasing the likelihood of involvement of the spinal cord or cauda equina. Manifestations include bilateral leg weakness and sensory symptoms, and loss of bowel or bladder control. Such symptoms should trigger an urgent workup that includes MRI studies and possibly neurosurgical consultation.
Acute spine pain is very common, and the likelihood of spontaneous recovery is in the range of 80% to 90%. Prolonged inactivity prolongs recovery. Because there is seldom a recognizable structural cause, treatment regimens tend to be nonspecific.
Patient education is important, and part of the therapeutic effort should include patient education about the nature of the condition, the likelihood of a good outcome, and the approach to be taken to speed recovery and minimize the risk of recurrence. Once these approaches to management have been undertaken, if there is no meaningful response to treatment, it is necessary to explore the possibility that psychosocial issues underlie the symptoms.
There is general agreement that patients with acute nonspecific spine pain or nonlocalizable lumbosacral radiculopathy (without neurologic signs or significant neurologic symptoms) require only conservative medical management. Patients should abstain from heavy lifting or other activities that aggravate the pain. Bed rest is not helpful and has been shown to delay recovery.9 Bed rest may be recommended for the first few days for patients with severe pain with movement. Recommended medications include nonsteroidal anti-inflammatory drugs such as ibuprofen or aspirin. If there are complaints of muscle spasm, muscle relaxants such as cyclobenzaprine may be used in the acute phase of pain. Narcotic analgesia should be avoided, in general, but it can be prescribed in cases of severe acute pain.
A study by Cherkin and coworkers compared standard physical therapy maneuvers and chiropractic spinal manipulation for the treatment of acute low back pain and found that both provide small short-term benefits and improve patient satisfaction, but they increase the cost of medical care and do not decrease the recurrence of back pain.10 Although patients were somewhat less satisfied with reassurance and an education booklet (the third group in that study), this group fared no worse than the groups receiving therapy.
The initial treatment of the patient with lumbosacral radiculopathy presenting with sensory symptoms and pain without significant neurologic deficits is not different from the approach for the patient with uncomplicated low back pain. However, such patients require observation for possible worsening of their neurologic status.
The treatment plan should fit the severity of the symptoms and signs. The management approach for radiculopathy covers the gamut from avoidance of heavy lifting to laminectomy and fusion. In acute radiculopathy, the goals of treatment should be the reduction of pain and the stabilization or amelioration of neurologic deficits.
Even patients with neurologic deficits such as segmental distributions of weakness, segmental loss of sensation, and reflex changes are likely to have significant spontaneous recovery. The initial approach to their treatment need not be different from that outlined for the patient with radicular sensory symptoms only. Reliable outcome studies that establish guidelines for medical versus surgical treatment in this patient group are not available. However, the risk is clearly greater in this group for progression of the neurologic deficits and residual neurologic impairment if spinal nerve root compression persists.
With a significant motor deficit, it is necessary to identify lesions that are amenable to surgical correction and to exclude the additional (and at times subclinical) presence of spinal cord or cauda equina compression. Thus, MRI studies are appropriate in this setting. With a very large disk protrusion or concomitant spinal cord compression, surgical intervention becomes a more important consideration, especially if neurologic deficits continue to worsen over time or if pain persists.
Management should consist of the avoidance of provocative activities, the use of non-narcotic analgesics, and the use of muscle relaxants if symptoms suggest that spasm is a component. Prolonged inactivity is not beneficial, and mobilization should be encouraged as soon as symptoms stabilize. Gentle exercises, massage, and mobilization are beneficial, but spinal manipulation is not. If there is a strong suspicion of nerve root impingement by disk protrusion or spondylosis, a high-dose, fast-taper course of corticosteroids can be used. A typical course would be prednisone 60 to 80 mg daily for 5 to 7 days, followed by a fast taper to discontinuation over the next 7 to 14 days, but there are no prospective, carefully controlled trials that confirm the value of corticosteroids. Prophylaxis against gastritis is recommended, and special precautions are needed in patients with diabetes; otherwise, the short course of treatment is not likely to produce complications.
When symptoms of spine pain extend beyond 4 to 8 weeks, the condition has moved from the acute to the chronic phase. At this point it is appropriate to reassess the patient’s symptoms and examination. If no neuroimaging was performed in the acute phase of the illness, the need for studies at this time should be reassessed. In the face of true radiculopathy with new or worsening neurologic deficits, a surgical opinion should be considered. Depending on the full clinical picture, a number of alternative nonsurgical approaches may be considered at this point, although in general their efficacy has not been proved.
The standard approach to the patient with nonspecific chronic spine pain is physical therapy. By 3 to 4 weeks after onset of symptoms, unless there is serious underlying structural disease, there is no reason the patient should not be enrolled in an aggressive program of mobilization, postural improvement, and increased endurance. Yoga techniques provide useful stretching maneuvers that the patient can learn by video instruction. In the treatment of subacute and chronic spine pain, osteopathic physicians and chiropractors provide spinal manipulation techniques, such as thrust, muscle energy, counter-strain, articulation, and myofascial release. A study by Andersson and associates in patients with nonradicular lumbar spine pain of 3 to 26 weeks’ duration compared a medical program that included physical therapy with a program that included active spinal manipulation.11 At 12 weeks, there was no significant difference in the degree of improvement between the two groups, although the group that received manipulation required significantly less analgesia, anti-inflammatories, and muscle relaxants, and they used less physical therapy. More than 90% of the patients in both groups were satisfied with their care.
Other treatments have been used with varying results.12 Transcutaneous electrical nerve stimulation has been used in patients with subacute and chronic spine pain, with varying results. A number of factors appear to influence success, including chronicity of the pain, electrode-pad placement, and prior treatments. Corticosteroid injections in facet joints and epidural locations have been advocated by some clinicians. The value of therapeutic corticosteroid injection in the setting of chronic nonspecific back pain without established radiculopathy has not been proven.
The long-term outlook is good for significant spontaneous recovery in patients with lumbosacral radiculopathy. However, the pain of acute radiculopathy can persist beyond 3 or 4 weeks, becoming chronic, at which point acute remedies such as rest, analgesics, and cervical traction may be less effective and other therapeutic options must be sought. When the manifestations of radiculopathy are primarily sensory or when the motor deficits are stable, a number of nonsurgical options have come into general use, although their true effectiveness has not yet been proved. The therapies used in patients with chronic nonspecific spine pain may be used in this setting as well, but several other procedures may be considered after conservative maneuvers have failed.
Chronic lumbar radiculopathy can also result from lumbar canal stenosis (LCS). This condition refers to narrowing of the intraspinal (central) canal, which can be associated with narrowing of lateral recesses and neural foramina. The primary symptoms of LCS include discomfort, sensory loss, and weakness in the legs, reflecting dysfunction of multiple spinal nerve roots within the lumbar spinal canal. Lateral recess and neural foraminal stenosis can give rise to lumbar radiculopathy. Neurogenic claudication—the tendency for exacerbation of symptoms due to walking, standing, and maintaining certain postures—is the hallmark of the condition. Neurogenic claudication may be described by patients as discomfort in the buttocks, thighs, or legs on standing or walking, which is relieved by sitting or lying. In one study, 94% of patients with LCS described symptoms of neurogenic claudication.13 Patients might demonstrate a combination of claudication symptoms and symptoms of more focal lumbar radiculopathy.
Anecdotal reports and retrospective analyses indicate that there is value in epidural corticosteroid injection for chronic radiculopathy, but the few placebo-controlled, prospective studies that have been performed for lumbar procedures have been criticized for design flaws.14 Of theoretical value, based on the assumption that radiculopathic pain is, at least in part, a result of inflammatory factors such as cytokinins in the vicinity of spinal nerve root compression, epidural corticosteroid injection aims to alleviate pain enough to allow more active physical therapy.
A randomized, double-blind trial performed by Carette and colleagues studied 158 patients with lumbar radiculopathy of 4 to 52 weeks’ duration who had evidence of radicular deficits on clinical examination and CT evidence of disk herniation.15 Six weeks after three epidural injections of either corticosteroids or saline, patients who had received corticosteroids had somewhat more improvement in leg pain, but at 3 months there was no significant difference between the two groups. Twenty-five percent of patients in both groups eventually went on to lumbar spine surgery.
Another local injection procedure, selective nerve root block, has been used for diagnostic and therapeutic purposes at the lumbosacral and cervical levels. This diagnostic technique has been used when there is lack of agreement between clinical and neuroimaging findings, when there is atypical limb pain, and when there is a history of failed surgery at the level in question. Nerve root blocks are contraindicated in the presence of systemic infection, local infection, or bleeding diathesis.
For the therapeutic procedure at the cervical level, it is standard practice to use a combination of 0.5 mL of 1% lidocaine and a long-acting corticosteroid. The therapeutic injection is preceded by a localization procedure under fluoroscopic guidance using a nonionic contrast medium to outline the selected nerve root. Pretreatment contrast injection is critical to exclude intravascular needle positioning. The needle may also be misplaced in the epidural space or at the sinuvertebral nerve, in which case selective nerve root block and the selectiveness of the results of injection cannot be verified.
The sinuvertebral nerve (derived from sympathetic fibers from the ramus communicans and fibers from either the primary anterior or posterior ramus) provides innervation over several segments to adjacent dura, posterior longitudinal ligament, and annulus. Although the efficacy of selective nerve root block at the lumbosacral levels has been studied, a further concern regarding the value of this procedure is the placebo response rate, which is reported to be as high as 38%.16
Several multicenter trials have compared surgical and medical treatment for lumbar radiculopathy. The Spine Patient Outcomes Research Trial (SPORT) compared surgical and medical therapies for lumbar disk herniation and found that over the short term, surgical patients had fewer symptoms, but over a 2-year period there was no difference between the groups.17,18 Likewise, in a Dutch study of severe sciatica for 6 to 12 weeks with lumbosacral radiculopathy confirmed by a neurologist, patients were randomized to either conservative treatment (with or without eventual surgery) or early minimal unilateral transflaval surgery with magnification.19 Patients with severe weakness and cauda equina syndrome were excluded. At 1 year, outcomes were similar for the two groups, although the rates of pain relief and of perceived recovery were faster for those assigned to early surgery.
In another study, patients with lumbar spondylolisthesis, lumbar canal stenosis, and neurogenic claudication with neurologic signs were randomized to conservative treatment or decompressive surgery with spinal fusion.20 This study demonstrated significantly greater improvement in pain and function among the surgically treated patients, but the analysis was complicated by an unexpectedly high rate of crossover from the conservative treatment group to the surgery group. Back pain did not improve as much as leg pain, indicating that the greatest benefit was likely to have occurred for nerve-root related symptoms.21 A study of patients with LCS without spondylolisthesis more convincingly demonstrated a benefit for surgically treated patients compared to medically treated patients over a 2-year period.22
Factors that should influence the likelihood that surgery will be performed for radiculopathy or LCS include an obvious neurologic deficit, progression of the deficits over time, unresolved pain, and identification of an anatomic lesion that corresponds with the neurologic picture. For patients with a lumbar disk herniation, surgery versus medical treatment may be a lifestyle decision, depending on whether they can alter their lifestyle to accommodate the pain or can obtain surgery to attempt to achieve a faster return to a former lifestyle.23
A small number of patients fail to respond to all the previously discussed therapeutic interventions. Some patients have chronic spine pain without evidence of structural intraspinal pathology, others have had previously treated structural lesions, and some have had multiple previous surgical interventions, a condition described as the failed-back syndrome. The goal in these patients is to improve the ability to perform activities of daily living and to diminish pain perception.
It is well established that a number of nonmedical factors play a role in the triggering and perpetuation of pain behavior. These include psychosocial issues such as job dissatisfaction, family stresses, and underlying psychiatric disorders. In other cases, patients develop a behavior of pain avoidance and fear of pain. Patients with chronic pain are best treated in programs dedicated to the rehabilitation of patients with multifactorial pain syndromes. Functional rehabilitation programs and pain programs concentrate on re-educating the patient to diminish fear of activities of daily living through graded exercise programs, the exploration of psychosocial stressors, and the non-narcotic treatment of pain.24
Determination of the degree of impairment and disability is required in patients with chronic disorders who are seeking worker’s compensation or permanent disability status. The American Medical Association has published guidelines for the evaluation of permanent impairment, incorporating both an injury model and a range-of-motion model for rating lumbar impairment.25
Recurrences of low back pain are common. Studies have shown recurrence rates between 30% and 75% within 3 years of the first episode. A Canadian study showed the highest recurrence rate among nurses and truck drivers and the lowest among office workers.26 Another Canadian study compared two interventions in the rehabilitation of workers who had missed 4 weeks of work. Patients were randomized to occupational intervention (occupational physician, ergonomist, and employer working to improve the work environment and the patient’s relation to it) or to clinical intervention (back pain specialist, back school, and functional rehabilitation after 12 weeks off work). The patients fared much better with the occupational intervention.27
For patients with radiculopathy, there is less likelihood of early recovery; however, without surgery, about 50% of patients can return to work after 4 to 6 weeks. A study of postal workers who had had lumbar laminectomy showed an almost six-fold greater likelihood for recurrent occupational low back injury.28