DIAGNOSIS AND TREATMENT
Herniated Lumbar Intervertebral Disk
Richard A. Deyo, MD; John D. Loeser, MD; and Stanley J. Bigos, MD
Low back pain is common, but a herniated intervertebral disk is
the cause in only a small percentage of cases. Most symptomatic
disk herniations result in clinical manifestations (pain, reflex loss,
muscle weakness) that resolve with conservative therapy, and only
5% to 10% of patients require surgery. Sciatica is usually the first
clue to disk herniation, but sciatica may be mimicked by other
disorders that cause radiating pain. Because more than 95% of
lumbar disk herniations occur at the L4-5 or L5-S1 levels, the
physical examination should focus on abnormalities of the L5 and
SI nerve roots. Plain radiography is not useful in diagnosing disk
herniation, but more sophisticated imaging (myelography, com-
puted tomography, or magnetic resonance imaging) should gener-
ally be delayed until a patient is clearly a surgical candidate.
Conservative therapy includes nonsteroidal anti-inflammatory
drugs, brief bed rest (often for less than 1 week), early progressive
ambulation, and reassurance about a favorable prognosis. Muscle
relaxants and narcotic analgesics have a limited role, and their use
should be strictly time-limited. Conventional traction and corsets
are probably ineffective. Except for patients with the cauda equina
syndrome, surgery is generally appropriate only when there is a
combination of definite disk herniation shown by imaging, a
corresponding syndrome of sciatic pain, a corresponding neurologic
deficit, and a failure to respond to 6 weeks of conservative therapy.
Annals of Internal Medicine. 1990;112:598-603.
From the University of Washington School of Medicine and
the Seattle Veterans Affairs Medical Center, Seattle, Washing-
ton. For current author addresses, see end of text.
JLow back pain is a pervasively common problem that
affects up to 80% of adults at some time during their
lives. A herniated disk is the cause of pain in only a
small percentage of patients, and even when a herniated
disk is diagnosed, surgery is only rarely necessary.
Thus, management of the patient with back pain and
sciatica caused by a herniated disk often falls to the
primary care physician, who should be familiar with the
natural history, consevative therapy, and indications for
surgical referral. Disk herniation refers to the protrusion
of the gelatinous nucleus pulposus of the disk through a
weakened anulus fibrosus. Clinically important hernia-
tions most often occur through a posterolateral defect,
but midline herniations may occur. Protruding disk ma-
terial may compress and inflame a spinal nerve root,
resulting in pain and paresthesias, usually perceived in
the sensory distribution of the nerve.
598 © 1990 American College of Physicians
Downloaded From: http://annals.org/ by a University of California San Diego User on 12/26/2016
Epidemiology and Natural History
The natural history of low back pain (in the absence
of underlying neoplastic or infectious causes) is gener-
ally characterized by rapid improvement. Only a small
portion of all affected persons will have back pain that
persists beyond 2 weeks (Table 1). An even smaller
proportion will have features suggesting sciatica, which
is usually the first clue to a possible herniated disk (1).
The incidence of clinically apparent lumbar disk herni-
ation is highest in young adults between the ages of 30
and 40 years. In the typical history, back pain occurs
first, followed by recurrences and eventually the devel-
opment of pain and paresthesias that radiate to the leg,
indicating the onset of sciatica. The back pain often
becomes less severe when leg pain develops. In one
study (2) of patients undergoing surgery for disk herni-
ation, the mean duration of preoperative back symp-
toms was nearly 3 years.
Many risk factors for lumbar disk herniation have
been reported in epidemiologic studies. Some of these
factors are biomechanical and include twisting and
other repetitive motions that occur in the occupational
setting. Sedentary occupations are associated with an
increased risk for disk herniation, possibly related to
muscular deconditioning or the chronic mechanical
stress of the sitting position. Occupations involving pro-
longed driving of motor vehicles may entail a particu-
larly high risk (3). Other apparent risk factors are more
closely related to lifestyle, including obesity (a threefold
risk in the most obese men compared with the least
obese) and cigarette smoking (4, 5). Smoking may in-
crease intradiscal pressure because of chronic coughing,
jeopardize disk metabolism because of vascular effects
of nicotine, or serve as a marker for psychosocial traits
associated with frequent and prolonged pain. Psychoso-
cial factors are clearly associated with back complaints
in general. For example, persons with lower educational
attainment have a higher risk for back pain and also
have worse outcomes (6).
Lumbar intervertebral disks may herniate without
causing symptoms, creating a pitfall for clinicians who
request imaging procedures early in an episode of back
pain or in the absence of clinical findings that suggest
nerve root compression. A study (7) of myelograms in
asymptomatic adults identified 24% with lumbar myelo-
graphy defects suggesting varying degrees of disk her-
niation. Similarly, a recent study (8) of lumbar com-
puted tomography found evidence of a herniated disk in
20% of persons who had no history of back pain or
sciatica. Because herniations do not necessarily result
in nerve root compression or symptoms, the mere ana-
tomic finding does not imply clinical disease.
Further, the presence of severe back pain, leg pain,
or sciatica does not necessarily imply a herniated disk.
 The diagnosis "herniated disk" has probably been over-
used, particularly in labeling persons who have severe
functional impairments associated with back pain. Wad-
dell (9) has called this the ''nominal'' diagnosis of her-
niated disk because it is usually made in the absence of
confirmatory clinical or imaging evidence.
Most symptomatic disk herniations result in clinical
manifestations (pain, reflex loss, sensory changes, and
muscle weakness) that resolve without surgical inter-
vention. It is estimated that only 5% to 10% of patients
with persistent sciatica will require surgery (10). In a
randomized trial (11) that compared surgery with con-
servative treatment for herniated lumbar disks, the re-
covery of foot weakness was equivalent in the two
treatment groups at 4-year follow-up. The major advan-
tage of surgery was more rapid pain relief and a lower
rate of relapse. In randomized trials (12, 13) that com-
pared intradiskal chymopapain injection with saline in-
jection for herniated disks, approximately 50% of pa-
tients receiving placebo reported an improvement in
back pain and sciatica within 6 weeks. Furthermore,
45% of patients showed improvement in reflex abnor-
malities, and 6 1 % had improvement in muscle weakness
(13). Somewhat higher percentages of improvement
were seen with active drug injection. Thus, spontaneous
recovery is part of the natural history of motor and
sensory signs as well as back and leg pain, although
invasive treatments may accelerate their resolution in
highly selected patients.
Unfortunately, recurrence is often part of the natural
history of low back problems, even after complete
symptom resolution. In a prospective study (14) of oc-
cupational back problems (resulting in compensation),
approximately one-third of subjects had a recurrence of
pain within 3 years. In a clinical trial (11) of therapy for
patients with a demonstrated disk herniation, 24% of
conservatively treated patients had recurrent pain
within 4 years, and 15% of surgically treated patients
had relapses.
Despite a reasonably favorable prognosis, back pain
in general (and herniated disks in particular) have an
enormous effect on health care utilization and costs.
One estimate placed the annual cost of direct medical
care for low back pain at 13 billion dollars and for
herniated lumbar disks at 2 billion dollars (15). Back
pain is the second leading cause of office visits to pri-
mary care physicians (16). Among adults under age 65,
medical back problems rank second as a reason for
nonsurgical hospital admissions, and back and neck
problems rank third among all surgical admissions (17).
The indirect costs of disk herniation are also high. In
men of working age, the costs of earnings and produc-
tivity losses attributable to herniated disks are similar to
those of ischemic heart disease (18).
Diagnosis
The differential diagnosis of low back pain is broad
and includes mechanical lesions (degenerative changes,
herniated disks, spinal stenosis, spondylolisthesis); sys-
temic diseases such as metastatic cancer, serious infec-
tions, and inflammatory spondyloarthropathies; and vis-
15 April 1990 • Annals of Internal Medicine • Volume 112 • Number 8
Downloaded From: http://annals.org/ by a University of California San Diego User on 12/26/2016
Table 1. Prevalence of Back Pain and Sciatica in the
Entire Adult Population*
Characteristic Prevalence,
%
Any low back pain 60 to 80
Ever had low back pain persisting at least 2 weeks 14
Low back pain persisting at least 2 weeks at a given 7
time (point prevalence)
Back pain with features of sciatica lasting at least 2 1.6
weeks
Lumbar spine surgery 1 to 2
* Adapted from Deyo and Tsui-Wu (1).
ceral diseases that may present with back pain as a
chief symptom but that do not involve the spine (for
example, nephrolithiasis, endometriosis, and aortic an-
eurysm). The differential diagnosis of sciatica includes
spinal stenosis (usually in older patients) and less com-
mon conditions such as synovial cysts, congenital
anomalies of the lumbar nerve roots, primary neural or
bone tumors, metastatic cancer, and epidural abscesses.
Irritation of the sciatic nerve may also occur simply
because of local pressure such as that caused by a
wallet in the back pocket. Finally, pain from other
sources in the lumbar spine such as the facet joints has
been reported to radiate into the upper leg in patterns
that mimic sciatica. Nonetheless, a herniated disk is the
commonest cause of true sciatica.
More than 90% of all episodes of back pain are prob-
ably attributable to mechanical causes, but the precise
pathoanatomic lesion is rarely identifiable (19). Thus,
the early diagnostic evaluation of back pain is devoted
to ruling out systemic disease; identifying and monitor-
ing neurologic abnormalities that may eventually require
surgery (generally resulting from herniated disks or spi-
nal stenosis); and identifying characteristics of the pain
and the patient that may influence conservative therapy
(20).
For most patients with a herniated disk, back pain
precedes the onset of sciatica and the back pain may or
may not abate as pain and paresthesias begin to radiate
down the leg. The onset of sciatic pain may be gradual
or sudden, and frequently there is no specific precipi-
tating event. The patient with a disk herniation often
has difficulty inrisingfrom the sitting or supine position
and typically experiences at least partial pain relief
while supine. This is probably related to the minimiza-
tion of motion and intradiscal pressure that occurs in
the supine position. The sciatica typically radiates down
the posterior or lateral aspect of the leg and is aggra-
vated by coughing or sneezing. It often radiates below
the knee, and this distal radiation suggests that the pain
is probably attributable to a radiculopathy rather than to
other causes of radiating pain.
Because more than 95% of lumbar disk herniations
occur at the L4-5 or L5-S1 levels, the neurologic exam-
ination should focus on the L5 and SI nerve roots (2).
Dysfunction of the L5 nerve root typically results in no
reflex changes, but does result in weakness of the great
toe extensor and other dorsiflexors of the foot and in
sensory loss along the medial aspect of the foot. Sen-
sation is often reduced in the great toe and in the web
599
 space between the first and second toes. Compression
of the SI nerve root typically results in a diminished
ankle reflex, weakness of the plantar flexors of the foot,
and sensory deficits of the posterior calf and lateral
aspect of the foot. Between 80% and 90% of patients
with surgically proven disk herniations will have foot
weakness or impaired ankle reflexes. Knee reflexes are
diminished in only 5% of patients with proven disk
herniations, reflecting the infrequent occurrence of disk
herniation above the L4-5 level (2).
A positive straight leg raising sign helps to confirm
the presence of nerve root irritation. A positive test
results when pain occurs in a radicular distribution with
a straight leg elevation of 60 deg or less. This test is
quite sensitive, and the sign is present in about 95% of
patients with a proven herniated disk at surgery. Thus,
absence of the sign constitutes a reasonable "rule out"
test for lower lumbar disk herniation. However, it is a
relatively nonspecific test; the sign is present in 80% to
90% of surgical patients who undergo surgery but are
found to have no part of the disk protruding beyond
normal anatomic limits (2). The crossed straight leg
raising sign is present when elevation of the contralat-
eral leg produces the sciatica. This sign is less sensitive
(about 25%) but much more specific (88%) than a pos-
itive ipsilateral straight leg raising test (2).
Electromyography is sometimes helpful in confirming
the radicular origin of physical manifestations or leg
pain. It may be particularly useful in discriminating
radicular pain from peripheral neuropathy. However,
because of its limited specificity, electromyography
should not be used as the sole determinant of either the
diagnosis or the need for surgery.
Plain radiography is rarely useful in diagnosing low
back pain and is not helpful in diagnosing a herniated
lumbar intervertebral disk. Nonetheless, plain roentgen-
ograms are appropriate in the patient with neurologic
deficits to rule out other conditions such as tumor,
infection, fracture, and spondylolisthesis. In the ab-
sence of neurologic deficits, plain films should probably
be reserved for patients with pain that persists beyond
3 to 4 weeks, patients with signs of systemic disease,
those with a known history of cancer, and patients with
known drug or alcohol abuse (21).
A clinical diagnosis of a herniated disk can only be
confirmed when appropriate imaging tests such as my-
elography, computed tomography, or magnetic reso-
nance imaging show an abnormality corresponding to
the neurologic deficit. The optimal choice among these
three tests, or the most appropriate sequence of tests,
remains controversial. The sensitivities of computed to-
mographic scanning and metrizamide myelography are
similar (90% to 95%), as are their specificities (68% to
88%) (22, 23). How magnetic resonance imaging will
compare with these other imaging modalities is not yet
clear.
A common mistake is to request these sophisticated
imaging tests early in the course of back pain or in the
absence of clinical findings that suggest that the patient
is a surgical candidate. Because disk herniation is so
common in asymptomatic persons, the isolated finding
on an imaging test without corresponding clinical signs
may be misleading and may initiate an ill-advised cas-
600 15 April 1990 • Annals of Internal Medicine • Volume 112 • Number 8
Downloaded From: http://annals.org/ by a University of California San Diego User on 12/26/2016
cade of clinical interventions. These tests, in general,
should be limited to the patient who has neurologic
abnormalities suggesting a herniated disk and who has
not responded to 3 to 4 weeks of conservative therapy.
Without such findings, the patient would not be a sur-
gical candidate regardless of the imaging results. The
only indication for imaging earlier in the course of back
pain would be the presence of the cauda equina syn-
drome, a progressive neurologic deficit, or the need for
a more precise diagnosis in a patient with suspected
tumor or infection.
Conservative Therapy
In the absence of the cauda equina syndrome or a
rapidly progressive neurologic deficit, virtually all pa-
tients with a suspected disk herniation should be treated
conservatively for 4 to 6 weeks. If neurologic deficits
still persist at this time, special studies to evaluate any
role for surgical intervention are appropriate. Some sur-
geons believe that delaying surgical treatment in this
circumstance beyond 3 months may be associated with
worse outcomes, although patients with such prolonged
pain may be destined to have a worse prognosis in any
event.
Bed rest has long been a mainstay of conservative
therapy for patients with suspected disk herniation, but
recent studies have led to progressively shorter bed-rest
recommendations. The rationale for bed rest is based on
the clinical observation that many patients with me-
chanical back pain experience symptomatic relief in the
supine position and on the physiologic observation that
intradiskal pressure is minimized in the supine position.
Nonetheless, in patients with back pain or sciatica in
the absence of neurologic deficits, 2 days of bed rest
seems to be as effective as 7 days of bed rest with
regard to pain resolution and functional recovery (24).
In the patient with a neuromotor deficit, bed rest may
need to be longer and stricter, but little evidence is
available to suggest the optimal duration. Because bed
rest promotes muscle weakness, cardiovascular decon-
ditioning, and bone mineral loss, lengthy recommenda-
tions for inactivity (more than a week) are to be
avoided. Furthermore, the standing posture results in
disk pressures that are only slightly higher than those in
the side-lying position and that are lower than those in
the sitting position. Thus, it is generally safe to recom-
mend standing and brief periods of walking to prevent
deconditioning. Patients may often subvert the potential
physiologic benefits of bed rest by sitting in bed to
watch television or read; such a position raises the
intradiscal pressure even above that produced in the
standing position.
Several well-designed clinical trials suggest that non-
steroidal anti-inflammatory drugs are efficacious in the
treatment of back pain in general. In the acute phase,
especially with sciatica, narcotic analgesics may be ap-
propriate. Their use should be relatively brief, however,
and limited by a specific time prescription rather than
contingent on ongoing pain. Evidence supporting the
use of "muscle relaxants" such as diazepam, cycloben-
zaprine, carisoprodol, and methocarbamol is limited.
 Table 2. Randomized, Controlled Trials of Conventional
Study (Reference) Patients, n Patient
Description
Lidstrom et al. (28) 62 Sciatica
Weber (29) 86 Herniated disk
Mathews and Hickling (30) 27 Sciatica
Coxheadet al. (31) 292 Sciatica
Weber et al. (32)t 215 Herniated disk
Pal et al. (33) 41 Sciatica
Mathews et al. (34) 143 Sciatica
* Sham traction consisted of applying the traction apparatus with minimal weight (shown not to produce distraction of the vertebrae in radiographic
studies).
t Used four different methods of traction.
Nonetheless, some clinical trials do suggest that such
agents, especially carisoprodol, are beneficial for treat-
ing back pain. All these drugs have sedating effects, and
it is unclear how to optimally select patients for their
use. It is also unclear whether they offer any advantage
over analgesics or nonsteroidal anti-inflammatory drugs
alone, although a recent trial (25) suggests a modest
benefit for patients receiving combined therapy. Like
narcotic analgesics, their use should be strictly time-
limited, with 1 week being a general guide. Muscle
relaxants and narcotic analgesics should be avoided in
patients with chronic pain syndromes (greater than 3
months in duration).
Exercise for the treatment of back pain and herniated
disks remains controversial; there are various regimens,
each of which has its advocates. In the case of acute
disk herniation, probably the most important consider-
ation is to avoid prolonged inactivity and debilitation.
Even among patients with severe symptoms, most are
able to resume some standing and walking activities by
the third day of symptoms. It is usually feasible in the
first week to have patients work up to a 20-minute walk
for every 3 hours spent supine. The ability to sit com-
fortably is a sign of improvement and suggests that the
patient can begin controlled endurance training activi-
ties that place a minimal load on the spine and are less
stressful than sitting. Endurance activities may be im-
portant in promoting recovery and preventing future
back problems. Speed walking and stationary bicycling
are commonly recommended, but swimming is also ap-
propriate. Jogging is feasible for younger patients with
milder symptoms. The traditional isometric flexion ex-
ercises (Williams exercises) may be useful for patients
with chronic pain but appear to be ineffective in pa-
tients with acute back problems (26).
Recommendations for rest, medication, and initiating
exercise should be given with an explicit time schedule.
Having patients continue rest or medication "until the
pain is gone" may foster dependence and prolong
symptoms and functional limitations. Similarly, patients
should be reassured that even if they have mild or
moderate symptoms when resuming physical activity,
such symptoms are not indicative of permanent harm. A
randomized clinical trial (27) for patients with acute
back pain compared symptom-limited rest, medication,
and activity with strictly time-limited prescriptions. In
this small trial, patients receiving the time-limited pre-
15 April 1990 • Annals of Internal Medicine • Volume 112 • Number 8
Downloaded From: http://annals.org/ by a University of California San Diego User on 12/26/2016
Traction for Sciatica or Proven Lumbar Disk Herniation
Control Group Blind Outcome Results
Assessment
Heat, rest Yes Negative
Sham traction* No Negative
Sham traction* Yes Negative
Multiple treatments No Negative
Sham traction* Yes Negative
Sham traction* Yes Negative
Heat Yes Negative (except for women
< 45 years old)
scriptions reported less pain and greater functional re-
covery at follow-up than patients who were told to "let
pain be your guide."
Although conventional traction is still widely used,
there is growing evidence against its efficacy in patients
with sciatica or proven disk herniations. Table 2 sum-
marizes seven randomized clinical trials (28-34) of con-
ventional traction using standard weights or motorized
traction devices. None of these trials showed a signifi-
cant benefit for traction compared with the control
treatment. Thus, hospital admission for traction is prob-
ably inappropriate. Indeed, the early management of
patients with herniated disks should generally be at
home to avoid reinforcing illness behavior through in-
stitutionalization and to maximize patient convenience.
There have been no controlled trials of inversion de-
vices or other gravity traction methods, although it is
clear that inversion can have important ocular and car-
diovascular side effects.
As with traction, no studies have convincingly shown
the efficacy of corsets or other spinal orthoses. Some
observers note that use of rigid orthoses may promote
muscle weakness and result in worse long-term out-
comes. Furthermore, to truly limit motion in the lumbar
spine requires a rigid body cast with extension to the
legs. The Quebec Task Force on Spinal Disorders (35)
concluded that although both traction and orthoses are
commonly used, no scientific evidence exists to support
their efficacy.
Longer-term intervention to reduce the frequency and
severity of expected recurrences of back pain or sciat-
ica should include lifestyle changes such as weight loss,
smoking cessation, and regular exercise. Instruction in
proper lifting methods, work positions, and rest posi-
tions may also be helpful, although such instruction is
not sufficient in itself to prevent recurrent back prob-
lems.
Patient reassurance and education is an important
aspect of therapy. The greatest source of patient dissat-
isfaction with care for back pain appears to be the
inadequate explanation of the problems (36). Further-
more, patients need to be reassured that despite recur-
rences, the natural history of back pain, sciatica, and
disk herniation is favorable and that important physical
disability is extremely rare. Physicians should also be
wary of labeling patients with frightening diagnoses.
The term ruptured disk implies a bursting or violent
601
 Table 3. Indications for Surgical Referral in the
Patient with Sciatica
The cauda equina syndrome (a surgical emergency):
characterized by bowel and bladder dysfunction (usually
urinary retention), saddle anesthesia, bilateral leg weakness
and numbness
Progressive or severe neurologic deficit
Persistent neuromotor deficit after 4 to 6 weeks of
conservative therapy
Persistent sciatica, sensory deficit, or reflex loss after 4 to 6
weeks in a patient with positive straight leg raising sign,
consistent clinicalfindings,and favorable psychosocial
circumstances (for example, realistic expectations and no
evidence of depression, substance abuse, or excessive
somatization)
dissolution of tissue, but this may be a common radio-
graphic finding in patients who are asymptomatic. Less
emotionally laden terms such as extruded disk may be
preferable, and physicians should avoid making a nom-
inal diagnosis of disk herniation based simply on the
severity of pain or dysfunction.
Surgical Intervention
The rate of lumbar spine surgery in the United States
is three to eight times higher than in most European
countries, suggesting that it may be overused. Although
it is frequently stated that surgery is becoming more
selective and that we have passed out of the era of
excessive surgery, the rate of lumbar discectomy actu-
ally rose 53% between 1980 and 1985 (National Center
for Health Statistics, Unpublished data). Furthermore,
wide and unexplained variations in the rate of lumbar
spine surgery have been observed among geographic
regions of the United States, suggesting that there are
widely differing practice styles among physicians. The
extensive literature on failed lumbar spine surgery is
testimony to the often ineffective use of this procedure.
Most neurosurgeons and orthopedists agree that sur-
gery for a herniated disk is only appropriate when there
is a combination of definite herniation documented by
some imaging procedure, a corresponding pain syn-
drome, a corresponding neurologic deficit, and failure to
respond to 4 to 6 weeks of conservative therapy. In
practice, these guidelines are often liberalized (to bulg-
ing disks, equivocal neurologic signs, and shorter peri-
ods of conservative care), despite evidence that out-
comes are worse under these circumstances (37, 38).
The presence of a herniated disk on an imaging test
alone is not an indication for surgery, nor is persistent
pain alone, nor is the observation that nothing else has
worked. Several investigators (37, 38) have developed
preoperative scoring indexes to quantify the likelihood
of a successful surgical outcome. These indexes com-
bine imaging results, physical examination findings, his-
tory, and psychosocial features of the patient and have
been shown to have substantial predictive value regard-
ing surgical success.
The major benefit of surgery for herniated lumbar
disks is relief of sciatica. In well-selected patients, ap-
proximately 75% will experience complete relief of sci-
atica and over 90% will have complete or partial relief.
602 15 April 1990 • Annals of Internal Medicine • Volume 112 • Number 8
Downloaded From: http://annals.org/ by a University of California San Diego User on 12/26/2016
The degree of relief declines progressively as the surgi-
cal findings change from complete herniation to bulging
disk to no herniation. Relief of back pain is less con-
sistent, with approximately 70% of patients experienc-
ing relief of back pain. Complete relief of all pain symp-
toms occurs in only half the group of patients
undergoing lumbar discectomy (2). Patients must have
realistic expectations regarding surgery and understand
that they are likely to have recurrent back difficulties
even after successful surgery. Repeat surgery is much
less successful than first-time surgery, unless a new disk
herniation is identified with all the other surgical criteria
described above.
Complications associated with lumbar spine surgery
are infrequent, but include death (0.2%), thromboembo-
lism (1.7%), and infections (2.9%) (2). Neurologic com-
plications due to nerve trauma or postoperative epidural
hematoma may occur but are very infrequent. A perfo-
ration of the dura mater requiring repair occurs in 2% to
5% of operations. Intraoperative complications are com-
moner in repeat operations.
The neurologic deficits resulting from disk herniation
are usually relatively minor and tend to resolve slowly
with or without surgery. Although surgery may slightly
accelerate resolution of neurologic deficits, the major
benefit is pain relief. Given the natural history of symp-
tom resolution and the goals of lumbar discectomy, this
procedure should always be regarded as elective except
in the case of a patient with the cauda equina syn-
drome. Thus, the patient should always be allowed to
make the surgical decision after he or she is given
information about the risks and benefits.
The choice of surgical procedures is controversial.
There are strong advocates for standard laminectomy
and discectomy, for microdiscectomy procedures, for
percutaneous discectomy, and for discectomy with fu-
sion. The indications for spine fusion are particularly
unclear, but it appears to be generally inadvisable for
the patient who has a single herniated disk with other-
wise uncomplicated surgery (39).
Chymopapain injections for herniated disks were in-
troduced in the United States in 1983, following a suc-
cessful randomized, double-blind, placebo-controlled
trial (13). Although chymopapain injection was advo-
cated as a substitute for surgery, in early trials it was
compared with placebo injections rather than with stan-
dard surgical intervention. Two small subsequent ran-
domized trials (40, 41) both suggested that the outcomes
of spine surgery may generally be superior to those of
chymopapain injection. Because of such findings and
because of the occasional occurrence of anaphylactic
reactions and serious neurologic complications, chymo-
papain has waned in popularity.
Primary care physicians should identify a neurosur-
geon or an orthopedist who operates in a highly selec-
tive manner and who uses criteria such as those de-
scribed here. Table 3 provides an explicit set of
recommendations for surgical referral.
Acknowledgments: The authors thank David Buchner and Stephan Fihn
for reviewing an earlier draft of this manuscript and Chris Morrison and
Kathy Minotto for helping to prepare the manuscript.
Grant Support: In part by the Northwest Health Services Research and
Development Field Program, Seattle Veterans Affairs Medical Center,
 and by grant HS06344-01 from the National Center for Health Services
Research and Technology Assessment.
Requests for Reprints: Richard A. Deyo, MD, Seattle Veterans Affairs
Medical Center (152), 1660 South Columbian Way, Seattle, WA 98108.
Current Author Addresses: Dr. Deyo: Seattle Veterans Affairs Medical
Center (152), 1660 South Columbian Way, Seattle, WA 98108.
Dr. Loeser: Department of Neurological Surgery, RI-20, University of
Washington School of Medicine, Seattle, WA 98195.
Dr. Bigos: Department of Orthopaedics, RK-10, University of Wash-
ington School of Medicine, Seattle, WA 98195.
References
1. Deyo RA, Tsui-Wu YJ. Descriptive epidemiology of low-back pain
and its related medical care in the United States. Spine. 1987;
12:264-8.
2. Spangfort EV. The lumbar disc herniation: a computer-aided anal-
ysis of 2,504 operations. Acta Orthop Scand Suppl. 1972;142:5-95.
3. Kelsey JL, White AA 3d. Epidemiology and impact of low-back
pain. Spine. 1980;5:133-42.
4. Heliovaara M. Body height, obesity, and risk of herniated lumbar
intervertebral disc. Spine. 1987;12:469-72.
5. Kelsey JL, Githens PB, O'Conner T, et al. Acute prolapsed lumbar
intervertebral disc. An epidemiologic study with special reference to
driving automobiles and cigarette smoking. Spine. 1984;9:608-13.
6. Deyo RA, Tsui-Wu YJ. Functional disability due to back pain. A
population-based study indicating the importance of socioeconomic
factors. Arthritis Rheum. 1987;30:1247-53.
7. Hitselberger WE, Witten RM. Abnormal myelograms in asympto-
matic patients. J Neurosurg. 1968;28:204-6.
8. Wiesel SW, Tsourmas N, Feffer HL, Citrin CM, Patronas N. A study
of computer-assisted tomography. I. The incidence of positive CAT
scans in an asymptomatic group of patients. Spine. 1984;9:549-51.
9. Waddell G. A new clinical model for the treatment of low-back
pain. Spine. 1987;12:632-44.
10. Frymoyer JW. Back pain and sciatica. N Engl J Med. 1988;318:
291-300.
11. Weber H. Lumbar disc herniation. A controlled, prospective study
with ten years of observation. Spine. 1983;8:131-9.
12. Fraser RD. Chymopapain for the treatment of intervertebral disc
herniation. A preliminary report of a double blind study. Spine.
1982;7:608-12.
13. Javid MJ, Nordby EJ, Ford LT, et al. Safety and efficacy of chy-
mopapain (Chymodiactin) in herniated nucleus pulposus with sciat-
ica. Results of a randomized, double-blind study. JAMA. 1983;
249:2489-94.
14. Abenhaim L, Suissa S, Rossignol M. Risk of recurrence of occupa-
tional back pain over three year follow-up. Br J Ind Med. 1988;
45:829-33.
15. Holbrook TL, Grazier KL, Kelsey JL, Stauffer RN. The Frequency
of Occurrence, Impact, and Cost of Musculoskeletal Conditions in
the United States. Chicago, Illinois: American Academy of Ortho-
paedic Surgeons; 1984:154-7.
16. Cypress BK. Characteristics of physician visits for back symptoms:
a national perspective. Am J Public Health. 1983;73:389-95.
17. Graves EJ. Diagnosis-related groups using data from the National
Hospital Discharge Survey: United States-1985. Hyattsville, Mary-
land: U.S. Department of Health and Human Services; 1987.
DHHS publication no (PHS) 87-1250. (Advance Data from Vital and
Health Statistics, No 137).
18. Salkever DS. Morbidity Costs: National Estimates and Economic
Determinants. Rockville, Maryland: U.S. Department of Health and
Human Services; 1985. DHHS publication no (PHS) 84-3393.
15 April 1990 • Annals
Downloaded From: http://annals.org/ by a University of California San Diego User on 12/26/2016
19. White AA 3d, Gordon SL. Synopsis: workshop on idiopathic low-
back pain. Spine. 1982;7:141-9.
20. Deyo RA. Early diagnostic evaluation of low back pain. J Gen
Intern Med. 1986;1:328-38.
21. Deyo RA, Diehl AK. Lumbar spine films in primary care: current
use and effects of selective ordering criteria. J Gen Intern Med.
1986;1:20-5.
22. Hudgins WR. Computer-aided diagnosis of lumbar disk herniation.
Spine. 1983;8:604-15.
23. Haughton VM, Eldevik OP, Magnaes B, Amundsen P. A prospective
comparison of computed tomography and myelography in the diag-
nosis of herniated lumbar disks. Radiology. 1982;142:103-10.
24. Deyo RA, Diehl AK, Rosenthal M. How many days of bed rest for
acute low back pain? A randomized clinical trial. N Engl J Med.
1986;315:1064-70.
25. Basmajian JV. Acute back pain and spasm. A controlled multicenter
trial of combined analgesic and antispasm agents. Spine. 1989;
14438-9.
26. Gilbert JR, Taylor DW, Hildebrand A, Evans C. Clinical trial of
common treatments for low back pain in family practice. Br Med J
[Clin Res]. 1985;291:791-4.
27. Fordyce WE, Brock way J A, Bergman J A, Spengler D. Acute back
pain: a control-group comparison of behavioral vs traditional man-
agement methods. / Behav Med. 1986;9:127-40.
28. Lidstrom A, Zachrisson M. Physical therapy on low back pain and
sciatica. An attempt at evaluation. Scand J Rehabil Med. 1970;
2:37-42.
29. Weber H. Traction therapy in sciatica due to disc prolapse (does
traction treatment have any positive effect on patients suffering
from sciatica caused by disc prolapse?) J Oslo City Hosp. 1973;
23:167-76.
30. Mathews JA, Hickling J. Lumbar traction: A double-blind controlled
study for sciatica. Rheumatol Rehabil. 1975;14:222-5.
31. Coxhead CE, Inskip H, Meade TW, North WR, Troup JD. Multi-
centre trial of physiotherapy in the management of sciatic symp-
toms. Lancet. 1981:1:1065-8.
32. Weber H, Ljunggren AE, Walker L. Traction therapy in patients
with herniated lumbar intervertebral discs. / Oslo City Hosp. 1984;
34:61-70.
33. Pal B, Mangion P, Hossain MA, Diffey BL. A controlled trial of
continuous lumbar traction in the treatment of back pain and sci-
atica. Br J Rheumatol. 1986;25:181-3.
34. Mathews J A, Mills SB, Jenkins VM, et al. Back pain and sciatica:
controlled trials of manipulation, traction, sclerosant and epidural
injections. Br J Rheumatol. 1987;26:416-23.
35. Spitzer WO, LeBlanc FE, Dupuis M, et al. Scientific approach to the
assessment and management of activity-related spinal disorders. A
monograph for clinicians. Report of the Quebec Task force on
Spinal Disorders. Spine. 1987;12(Suppl l):Sl-59.
36. Deyo RA, Diehl AK. Patient satisfaction with medical care for low-
back pain. Spine. 1986;11:28-30.
37. Herron LD, Turner J. Patient selection for lumbar laminectomy and
discectomy with a revised objective rating system. Clin Orthop.
1985;199:145-52.
38. Finneson BA, Cooper VR. A lumbar disc surgery predictive score
card. A retrospective evaluation. Spine. 1979;4:141-4.
39. White AH, von Rogov P, Zucherman J, Heiden D. Lumbar laminec-
tomy for herniated disc: a prospective controlled comparison with
internal fixation fusion. Spine. 1987;12:305-7.
40. Ejeskar A, Nachemson A, Herberts P, et al. Surgery versus chemo-
nucleolysis for herniated lumbar disks. A prospective study with
random assignment. Clin Orthop. 1983;174:236-42.
41. Crawshaw C, Frazer AM, Merriam WF, Mulholland RC, Webb JK.
A comparison of surgery and chemonucleolysis in the treatment of
sciatica. A prospective randomized trial. Spine. 1984;9:195-8.
of Internal Medicine • Volume 112 • Number 8 603