Clinical Review & Education

JAMA Pediatrics | Review

Musculoskeletal Low Back Pain in School-aged Children

A Review
James MacDonald, MD, MPH; Emily Stuart, MD; Richard Rodenberg, MD

Supplemental content
IMPORTANCE Low back pain (LBP) in children and adolescents is a common problem. The
differential diagnosis of LBP in this population is broad and different from that seen in the
adult population. Most causes of LBP are musculoskeletal and benign in their clinical course.
Clinicians should have an understanding of the relevant anatomy and the most commonly
encountered etiologic factors of LBP in children and adolescents to provide effective care.

OBSERVATIONS Low back pain is rarely seen in youth before they reach school age.
Subsequently, rates of LBP rise until age 18 years, at which age the prevalence of LBP is similar
to that in adults. The differential diagnosis of LBP in this population is broad, and individual
etiologic factors are most often associated with musculoskeletal overuse or trauma. Sinister
etiologic factors are rare. The patient’s history and physical examination are the foundation of
evaluating a child with LBP. The indication for and timing of specific imaging or other studies
Author Affiliations: Division of
will vary depending on the etiologic factor of concern. Most treatment of LBP in this
Sports Medicine, Department of
population is centered on relative rest, rehabilitation, and identification of predisposing risk Pediatrics, Nationwide Children’s
factors. Pharmacologic treatment may be used but is typically a brief course. Orthopedic, Hospital, Columbus, Ohio
rheumatologic, and other subspecialty referrals may be considered when indicated, but most (MacDonald, Rodenberg); The Ohio
State University College of Medicine,
of these patients can be managed by a general pediatrician with a good understanding of the Columbus (MacDonald, Rodenberg);
principles described in this article. Department of Orthopedics,
Children’s Hospital Colorado,
Orthopedic Institute, Aurora (Stuart);
CONCLUSIONS AND RELEVANCE Low back pain in children and adolescents is a common
University of Colorado School of
problem. It is most often nonspecific, musculoskeletal, and self-limiting. Pediatricians should Medicine, Aurora (Stuart).
recognize the importance of a proper history, physical examination, and general knowledge Corresponding Author: James
of the lumbar spine and pelvic anatomy relevant to the child in their evaluation with this MacDonald, MD, MPH, Division of
presenting symptom. Sports Medicine, Department of
Pediatrics, Nationwide Children’s
Hospital, 584 County Line Rd,
JAMA Pediatr. 2017;171(3):280-287. doi:10.1001/jamapediatrics.2016.3334 Westerville, OH 43082
Published online January 30, 2017. (james.macdonald
@nationwidechildrens.org).

L
ow back pain (LBP) in school-aged children is a common oc- previous back injury, and family history of LBP are all potential risk
currence; nevertheless, it is often underappreciated.1 The factors for school-aged children to develop LBP.9-12 Although there
prevalence of LBP rises with age: 1% at age 7 years, 6% at has been concern about a potential association of LBP and back-
age 10 years, and 18% at ages 14 to 16 years.2 By age 18 years, the packs, the evidence pointing to use of backpacks as a risk factor is
lifetime prevalence rates of LBP approach those documented in weak.9 No single risk factor for a first episode of LBP in school-aged
adults, with an estimated yearly prevalence of 20% and a lifetime children has been definitively validated (level of evidence, 1).13
prevalence of 75%.3 More than 7% of adolescents experiencing LBP Historically, it has been taught that most LBP in school-aged chil-
will seek medical attention.1 dren has an identifiable diagnosis. More recent research has chal-
The effect of LBP on this population can be considerable and lenged this thinking. A high-quality prospective study of 73 pediat-
may significantly restrict instrumental activities of daily living for this ric patients with LBP (level of evidence, 2) followed up for 2 years
population, such as attendance at school and gym or sports found that nearly 80% had no definitive diagnosis.14 Most cases of
participation.4 Low back pain in this age group is a significant risk LBP in school-aged children are nonspecific and self-limiting.15,16
factor for developing LBP as an adult.5
Several potential risk factors for developing LBP in school-
aged children have been investigated. The prevalence of LBP cor-
Discussion and Observations
relates with participation in sports and level of competition.4,6,7 There
is a U-shaped association between physical activity and the inci- Relevant Anatomy
dence of LBP in school-aged children, with both low and high levels Pediatricians need a basic but solid understanding of the anatomy
of physical activity associated with a higher risk.8,9 Female sex, of the lumbosacral spine to provide effective care to school-aged chil-
growth acceleration, adverse psychosocial factors, increasing age, dren with LBP. The lumbar spine is composed of 5 vertebrae

280 JAMA Pediatrics March 2017 Volume 171, Number 3 (Reprinted) jamapediatrics.com

Copyright 2017 American Medical Association. All rights reserved.

Downloaded From: by a Kaohsiung Med Univ User on 04/19/2018

 Musculoskeletal Low Back Pain in School-aged Children Review Clinical Review & Education

(L1-L5), with an intervertebral disc between each vertebra and be-

tween L5 and the sacrum. The sacrum articulates with the ilium of Key Points
the pelvis at the sacroiliac (SI) joint, a diarthrodial joint with limited Question What is the most common cause of low back pain in
motion.15 school-aged children?
The most significant components of the vertebra are the body
Findings This review discusses the causes of low back pain in
and the posterior elements, which include the pars interarticularis and
school-aged children and finds that most cases are
the spinous and transverse processes.17 The vertebrae themselves musculoskeletal and have a benign clinical course. Although
articulate posteriorly at the zygapophyseal facet joints. In the pedi- pediatric training has historically emphasized that low back pain in
atric spine, the superior and inferior portions of the vertebral body this population is caused by specific etiologic factors, most cases
and the spinous processes are composed of physeal cartilage and are of low back pain in this age group have no identifiable pain
secondary ossification centers that become radiographically appar- generator.
ent between ages 8 and 12 years and fuse by adulthood.18,19 Before Meaning Pediatricians should be vigilant for identifiable and
that fusion occurs, these areas are more prone to injury than is the serious causes of low back pain, but most presentations are
solid bone of an adult (Figure 1). self-limiting and will respond to conservative treatment.
The spinal nerves will bilaterally exit the foramina of the lum-
bar spine below their corresponding vertebral body level. For ex-
ample, the L2 nerve root exits below L2 through the foramen be- Figure 1. Bony Anatomy of Lumbar Spine
tween L2 and L3. This anatomical feature is important to remember
in the evaluation of potential radiculopathies that may be associ- A L5 Spondylolysis seen on CT scan B Lateral radiograph of lumbar
ated with LBP and their associated findings on neurologic exami- spine with L4 apophyseal ring
fracture
nation (discussed in the Evaluation section).

Differential Diagnosis and Pathophysiological FIndings

The differential diagnosis of LBP in school-aged children is broad and
different from that seen in an adult population (Table 115,18,20-36 and
eTable 1 and eTable 2 in the Supplement).37 The most common cause
of LBP is acute or subacute musculoskeletal pain with no identifi- A
able pain generator.38 A tight thoracolumbar fascia results as a con- A
B
sequence of rapid growth and can be compounded by existing in- B C
flexibility, which results in a hyperlordosis, producing a flat midback
C
and thoracic kyphosis, which can lead to a syndrome known as
hyperlordotic LBP. This term is used synonymously with the terms
posterior overuse syndrome, mechanical LBP, or muscular LBP.7,23
In general, when the clinician is able to identify a specific etio-
logic cause for LBP in school-aged children, it involves injury to the
posterior elements of the spine, such as spondylolysis, which is A, Classic spondylolytic lesion at the left pars interarticularis of the L5 vertebra
much more common than pathologic disc characteristics in this seen on computed tomography (CT) scan (arrowheads); also shown are the
population.15,37 The remainder of this section briefly discusses the vertebral body (A), right pars interarticularis (B), and spinous process (C).
B, Apophyseal ring fracture at the superior vertebral end plate of the L4
pathophysiological features of the more common, specific diagno-
vertebra (arrowhead) seen on lateral radiographs taken while the patient was
ses affecting the posterior elements of the spine; Table 115,18,20-36 standing; also shown are the vertebral body (A), pars interarticularis (B), and
and eTables 1 and 2 in the Supplement contain an in-depth discus- spinous process (C).
sion of other diagnoses, including herniated nucleus pulposus
(HNP), apophyseal ring fractures, atypical Scheuermann disease,
lumbar facet syndromes, compression fractures and transverse Spondylolisthesis may occur with bilateral pars defects, allow-
process fractures, SI joint dysfunction, and benign hypermobility ing a forward translation of a vertebral body over the body subja-
syndrome. cent to it (Figure 2). Spondylolisthesis is graded, with the grade based
Spondylolysis is a condition in which there is a defect in the pars on the percentage of slip of the superior body over the inferior one
interarticularis of a lumbosacral vertebral body.39 There are differ- (grade I, 0%-25%; grade II, 26%-50%; grade III, 51%-75%; grade IV,
ent types of spondylolyses, including congenital: in 1 series of 500 76%-100%; and grade V, >100%, which is also known as spondy-
children, 4.4% of the study group had a pars defect at age 6 years.39 loptosis). The condition is further divided into low-grade slips (grades
Isthmic spondylolysis, an acquired overuse injury of the pars inter- I and II) and high-grade slips (grades III, IV, and V), with high-grade
articularis, is the most common type among school-aged children slips at risk for higher degrees of pain and the presence of radicular
with LBP who present to the pediatrician.40,41 It is often sympto- or neurologic symptoms. Slips can progress, although this is un-
matic, especially in young athletes, and the incidence is higher in the usual after skeletal maturity.39 Spondylolisthesis most often oc-
athletic population.37,42 The most common sites are L5 and, less of- curs at the level of L5 and S1.22,44,45
ten, L4.43 Isthmic spondylolysis is often bilateral, affecting both the Although historically adolescent idiopathic scoliosis was thought
right and left pars of an individual vertebral body and can be acute not to be associated with higher rates of LBP, more recent studies
or chronic (Figure 1 and Figure 2). have found this not to be so and demonstrate as much as a 2-fold

jamapediatrics.com (Reprinted) JAMA Pediatrics March 2017 Volume 171, Number 3 281

Copyright 2017 American Medical Association. All rights reserved.

Downloaded From: by a Kaohsiung Med Univ User on 04/19/2018

 Clinical Review & Education Review Musculoskeletal Low Back Pain in School-aged Children

Table 1. Differential Diagnosis of LBP in School-aged Children: Clinical Presentation and Potential Diagnostic Tools
Common History and
Diagnosis Defining Characteristics Examination Findings Diagnostic Tools Comments
Isthmic spondylolysis Stress fracture of the pars interarticularis Pain with extension; AP and lateral More frequent at L5 than at
often owing to repetitive loading or extension-based, twisting radiographs (sensitivity, L4; may be bilateral;
trauma athlete; positive result of 72%-78%), SPECT scan typically insidious onset;
Stork test (sensitivity, (sensitivity, 84%), MRI MRI used more frequently
50%-73%; specificity, (sensitivity, 92%), or CT for diagnosis21
17%-32%)20 scan (sensitivity, 90%)21
Spondylolisthesis Anterior slip of superior vertebrae in Pain with extension; Standing AP and lateral Grading based on slip; in
relation to inferior vertebrae positive result of Stork radiographs; repeat females, high-grade slips,
test; step-off on palpation radiographs every 3-6 and increased growth
(sensitivity, 60%-88%; mo until patient is velocity increase risk of
specificity, 87%-100%)20; skeletally mature to progressive slip
tight hamstrings determine if slip is
progressing22
Atypical Scheuermann disease Schmorl nodes, vertebral end-plate Dull, achy pain at AP and lateral NA
flattening, narrowing of disk spaces23-25 thoracolumbar junction; radiographs will show
tight thoracolumbar characteristic findings
fascia; may have flattening
of lumbar lordosis
Discogenic disease and Protrusion or rupture of a disc; may have Pain worse with flexion or MRI (sensitivity, 75%; Typically at L4 and L5 and
herniated nucleus pulposus degenerative disc disease without true Valsalva maneuver; 5% of specificity, 77%), with L5 and S1; most are
herniation school-aged children have positive predictive value centrolateral26; straight leg
leg pain without LBP26; of 84% and negative raise test has sensitivity of
35% of children have predictive value of 67%-91% and specificity
radiating sciatic pain27 64%28 26%29
Apophyseal ring fracture Fracture of the cartilaginous ring Presents similar to disc May be seen on Occurs in 28% of
apophysis herniation; positive result radiograph or CT scan school-aged children with
of straight leg raise test; but best visualized on CT disc herniations27
tight hamstrings scan30,31
Hyperlordotic back pain Increased lordosis, mechanical LBP Weak core; positive Diagnosis based on NA
Trendelenburg sign; results of clinical
increased lumbar lordosis examination as all results
with thoracic kyphosis of imaging are normal
Facet arthropathy Inflammation of the facet joints Extension-based pain or SPECT scan or CT scan NA
pain with axial loading; will show irregularities of
presents similar to the facet joints or
spondylolysis hypertrophy32
Sacroiliac pain May have a sprain or joint degeneration Pain in the medial Results of radiographs NA
buttocks and posterior are typically normal; MRI
thigh; LBP recreated with may show joint
FABER test degeneration or sacral
stress fracture33
Benign hypermobility Generalized hypermobility; often a Beighton score >4 of Results of all imaging May have instability in
diagnosis of exclusion 934,35 studies will be normal other joints (eg,
subluxation or dislocation
events)
Transitional vertebrae Inflammation of a variant in which the Nonspecific LBP; insidious Radiographs have NA
lumbar transverse process fuses with onset accuracy of 76%-84%;
the sacrum SPECT scan shows
increased uptake of
radiotracers; CT scan will
show any abnormal
anatomy32,36
Inflammatory conditions Spondyloarthropathies, including Morning stiffness >30 min, MRI of the pelvis with May have a family history
conditions positive for HLA-B27, pain that wakes child from intravenous contrast will of autoimmune conditions
enthesitis, or juvenile idiopathic arthritis sleep, alternating buttock show early inflammation
pain, pain that improves of the sacroiliac joints15
with exercise; may have
positive result on modified
Schober test
Tumors Benign (eg, osteochondroma) or Fevers, weight loss, CBC, ESR, CRP, LDH; MRI NA
malignant (eg, leukemia, lymphoma, malaise, night pain, bowel or CT scan may be
or osteosarcoma) or bladder dysfunction required15,18
Infections Discitis, vertebral osteomyelitis, Irritability, limping, fever, CBC, ESR, CRP, blood NA
epidural abscess back pain or abdominal culture; bone scan may
pain; decreased motion show early changes, but
and prefer one position an MRI is more
over another15 specific15,18
Abbreviations: AP, anteroposterior; CBC, complete blood cell count; LBP, low back pain; LDH, lactate dehydrogenase; MRI, magnetic resonance
CRP, C-reactive protein; CT, computed tomography; ESR, erythrocyte imaging; NA, not applicable; SPECT, single-photon emission computerized
sedimentation rate; FABER, Flexion-Abduction-External Rotation; tomography.

increase in LBP in school-aged children with adolescent idiopathic Pediatrician training has traditionally focused the evaluation
scoliosis.46,47 A detailed description of scoliosis is beyond the scope of LBP in children on the need to identify serious pathologic
of this article. conditions, such as infection or malignant neoplasms.15 Although

282 JAMA Pediatrics March 2017 Volume 171, Number 3 (Reprinted) jamapediatrics.com

Copyright 2017 American Medical Association. All rights reserved.

Downloaded From: by a Kaohsiung Med Univ User on 04/19/2018

 Musculoskeletal Low Back Pain in School-aged Children Review Clinical Review & Education

pediatricians must be vigilant in their evaluations to exclude more

Figure 2. Imaging of Spondylolysis and Spondylolisthesis
serious pathologic conditions, such as spondylodiscitis, vertebral os-
teomyelitis, and neoplasm, these conditions are uncommon, and A Bilateral L5 spondylolysis lesions B Severe grade 3-4 spondylolisthesis
most school-aged children will have a mechanical, musculoskeletal (L5 on S1)

etiologic cause for their LBP.15,16,48

Evaluation
Children with LBP require a thorough clinical evaluation based on
the history and physical examination. A complete history including
onset, duration, frequency, location, and severity of pain, as well as
factors that alleviate or aggravate pain, should be elicited.26 Acute-
onset pain is typically caused by trauma, while insidious-onset pain
may be caused by muscular, bony, inflammatory, or biomechanical
issues.49 Pediatricians should inquire about the child’s activities and
sports participation as well as how much the pain is affecting these
activities. Visual analog scales for pain and pediatric-oriented func-
A, Uptake of technetium 99-m as seen classically on acute spondylolytic lesions
tional disability scales may be used to assess the degree to which
with single-photon emission computerized tomography imaging (yellow
LBP is affecting the child’s life.50 Although clinicians should probe arrowheads). B, High-grade spondylolisthesis (grade 3-4) at the L5 on S1
for “red flags,” including pain while sleeping, bowel or bladder dys- vertebrae (red arrowhead) seen on a lateral lumbar radiograph taken while the
function, radicular symptoms, saddle paresthesia, fever, and weight patient was standing.

loss, a Cochrane Database review of LBP in patients of all ages indi-

cates that there is insufficient evidence regarding the diagnostic ac- when the back is in extension is typically owing to involvement of
curacy of these signs and symtpoms.51 Finally, clinicians should ask the posterior elements (as in spondylolysis), while flexion-based pain
about family history of scoliosis, autoimmune conditions, and ma- is suggestive of HNP.26,49
lignant neoplasms. A complete neurologic assessment including lower extremity
Four questions have been shown to correlate with an inflam- sensation, motor strength, and deep tendon reflexes should be in-
matory cause of LBP in patients younger than 45 years with symp- cluded in the examination. The reflexes of the patellar tendon (L4)
toms that have lasted longer than 3 months: and Achilles tendon (S1) should be elicited, and sensation in the der-
1. Does back stiffness in the morning last more than 30 minutes? matomes of T12 and S1 and motor function of the hip flexors (L2 and
2. Does back pain awaken you during the second half of the night? L3), quadriceps (L3 and L4), and extensor hallucis longus (L5) should
3. Does the pain alternate from one buttock to the other? be assessed.26 Patients who are unable to walk on their heels or toes
4. Does pain improve with exercise but not with rest? and those with abnormal sensation of the medial (L4), dorsal (L5),
When a patient with prolonged symptoms answers 2 of the 4 or lateral (S1) aspect of the foot may have injury to the nerve root in
questions positively, the sensitivity and specificity for an inflamma- the absence of a lower-extremity injury.55 The straight leg raise and
tory etiologic cause reach 70% and 81%, respectively. When an- slump tests can be performed for nerve root compression, and ab-
swers to 3 of 4 questions are positive, the sensitivity drops to 33% normal Babinski or abdominal reflexes suggest a pathologic condi-
while specificity approaches 100%.52,53 tion of the upper motor neurons.26
There should be adequate exposure of the spine during the Finally, the clinician should perform any indicated special tests.
physical examination. It is crucial that clincians directly inspect the The stork test is performed by having the patient stand on 1 leg and
back for signs of significant modifying conditions associated with LBP, hyperextend his or her back (Figure 3A). A positive test result is in-
such as deformities (eg, spondylolisthesis), rashes (eg, psoriasis), hair dicated by re-creation of LBP and suggests spondylolysis.49 The
tufts (eg, spina bifida occulta), or asymmetry of the spine (eg, sco- straight leg raise test is performed by passively flexing a supine pa-
liosis). The patient should be examined both in the coronal and sag- tient’s hip with the knee extended. The test result is positive if the
ittal planes while standing, sitting, and walking.54 Abnormal spinal patient has pain radiating into the posterior thigh and knee and is
alignment, scapular asymmetry, or pelvic obliquity may suggest sco- typically positive in those with HNP.15 The slump test is slightly less
liosis or leg length discrepancy, while a kyphotic deformity may sug- specific but more sensitive than the straight leg test in adults with
gest atypical Scheuermann kyphosis.26 A positive Trendelenburg HNP.56 This test is performed with a seated patient rounding the back
sign, indicated by a downward pelvic tilt to the unaffected side, may while flexing the head forward. The examiner passively extends the
suggest decreased core strength or a neurologic deficit, which could knee with the foot flexed; the test result is positive if the patient ex-
be contributing to the LBP; patients with a positive Trendelenburg periences radicular pain. The Flexion-Abduction-External Rotation
sign should have a complete hip examination to further evaluate any (FABER) test, which suggests a pathologic condition of the SI joint
underlying issues of the lower extremity.49 if the patient develops SI pain, is performed by passively placing the
The clinician should palpate for tenderness over the spinous pro- supine patient’s leg into a figure-4 position while gently pressing on
cesses, paraspinal musculature, and SI joints. Tenderness of the spi- the knee and the opposite anterior iliac crest; the test result is posi-
nous process may suggest fracture or ligamentous injury, while ten- tive if the patient experiences pain in the contralateral SI joint to the
derness of the paraspinal musculature is more indicative of muscle knee in the figure-4 position.15 The modified Schober test is per-
spasm but can also be nonspecific.26 Range of motion is checked in formed by making marks 10 cm proximal and 5 cm distal to the pos-
flexion, extension, lateral bending, and rotation. Pain that is worse terior superior iliac spine on a patient who is standing. The patient

jamapediatrics.com (Reprinted) JAMA Pediatrics March 2017 Volume 171, Number 3 283

Copyright 2017 American Medical Association. All rights reserved.

Downloaded From: by a Kaohsiung Med Univ User on 04/19/2018

 Clinical Review & Education Review Musculoskeletal Low Back Pain in School-aged Children

specificity of computed tomography and magnetic resonance imaging

Figure 3. Stork Test and Thoracolumbar Spinal Orthosis
for common conditions, such as spondylolysis and HNP, there is little
A Stork test performed on the right leg B Rigid thoracolumbar spinal orthosis published about the diagnostic accuracy of advanced imaging tools for
less common causes of pediatric LBP.21,57,58
A complete blood cell count and tests for erythrocyte sedimen-
tation rate and C-reactive protein level may be indicated if there is
high suspicion for an inflammatory, infectious, or malignant
process.19 Laboratories may include tests for antinuclear antibody,
rheumatoid factor, and HLA-B27, but these tests should be or-
dered cautiously and are most helpful when a rheumatologist is con-
firming a suspected autoimmune diagnosis, as up to 20% of the gen-
eral population may have positive results for antinuclear antibody.59
As there are a variety of etiologic causes of LBP in children, there
are notable specific history, examination findings, and imaging re-
sults that are suggestive of certain conditions (Table 115,18,20-36 and
eTable 2 in the Supplement).

Treatment
The specifics of treatment for individual etiologic causes of LBP can
A, Stork test is performed if a clinician suspects a spondylolysis. It is performed
vary widely (eTable 1 in the Supplement). This section discusses gen-
with the patient standing on 1 leg with the clinician guiding the patient’s lumbar
spine in extension. A positive test result is one that reproduces the patient’s eral principles the pediatrician should keep in mind when caring for
pain. B, Rigid thoracolumbar spinal orthosis, a type of brace sometimes used in this population.
the treatment of spondylolysis. The foundations of treatment are accurate diagnosis when pos-
sible and an understanding of the nuances of the spinal anatomy of
flexes forward at the hips with the knees extended and the marks children, who are skeletally immature. Most school-aged children
are remeasured: a distance less than 21 cm suggests a spondyloar- present with nonspecific and self-limiting symptoms and will re-
thropathic condition, although this test may have a positive result spond to conservative treatment, including relative rest from of-
in other disorders in which a patient has limited forward flexion (eg, fending activities causing pain and often some form of physical
HNP).15 Evaluation of overall mobility can be assessed with the Beigh- therapy.7,14-16 Several rehabilitation programs have been described
ton criteria, with a score of 5 or higher suggestive of global joint for children with LBP based on specific diagnoses, but, to our knowl-
hypermobility.34 Although all these examination maneuvers may be edge, there is little evidence in the literature supporting their use.
performed on any patient with LBP, they are specifically indicated These rehabilitative programs tend to be empirically driven.24 Re-
on the basis of suspected diagnoses (Table 115,18,20-36 and eTable 2 habilitation is a multifactorial process and relies first on the treat-
in the Supplement). ment of the effects of the acute injury, including losses in mobility
Anteroposterior and lateral radiography should be considered in and function, as well as recognition of any deficits in biomechanical
children with LBP, especially if pain has been present for more than 3 function leading to alterations in technique or performance of a spe-
weeks.7 Thelateralviewshouldbeperformedwhilethepatientisstand- cific activity that could promote injury (eg, the proper mechanics of
ing, as a spondylolisthesis may not be revealed on a recumbent exami- throwing a baseball).24,60
nation. Owing to increased radiation with little increase in diagnostic Rehabilitation progresses through specific stages that initially
utility, oblique radiography is best avoided.18 If results of radiography focus on preserving and promoting range of motion and strength.
arenonrevealing,advancedimagingmaybeconsidered.Single-photon Hip flexibility is crucial and promoted by emphasizing exercises to
emissioncomputerizedtomographyscan,computedtomography,and stretch the hip flexors and hamstrings.61 Strength and motion re-
magnetic resonance imaging have had changing roles in the workup covery is coupled with proprioceptive training, which then leads to
of pediatric patients with LBP. Although single-photon emission com- correcting deficits noted in the kinetic chain, motion patterns, and
puted tomography scans are useful for identifying subtle bony injuries, neuromuscular control.60 A mainstay of therapy is core stabiliza-
especially in the acute setting, they contain a nuclear isotope (techne- tion, which refers to improving neuromuscular control, strength, and
tium-99) and expose organs throughout the body to higher doses of endurance of the muscles central to maintaining dynamic spinal and
radiationthananyothertypesofimagingstudies.21 Resultsofcomputed trunk stability. These muscle groups include the abdominals, lum-
tomography can provide exquisite bony and cartilage detail, but it also bar multifidi, and erector spinae, as well as other paraspinal, pelvic,
exposes patients’ bone marrow and colon to higher doses of radiation and cervicothoracic musculature.60,61 The literature is unclear as to
than do radiography or magnetic resonance imaging.15,21,57 Although which exercises are best to rehabilitate the core musculature.61
computedtomographyandsingle-photonemissioncomputedtomog- Therapy for issues such as spondylolysis (associated with pain on
raphy scans can be excellent diagnostic tools, increased concerns with back extension) traditionally revolves around a flexion-based therapy
radiation, especially in pediatric patients, have led to decreased use in program (Williams flexion–based therapy program), whereas con-
recentyears.18,21 Magneticresonanceimaging,whichhastypicallybeen ditions such as HNP (associated with pain on back flexion) are treated
used in the evaluation of soft-tissue pathologic conditions, is also now with an extension-based therapy program (McKenzie extension–
used more frequently for evaluation of bony pathologic conditions21 based therapy program). 61,62 Last, the patient focuses on a
(Table2).Althoughthereisgoodliteratureregardingthesensitivityand functional progression aimed at correcting biomechanics and

284 JAMA Pediatrics March 2017 Volume 171, Number 3 (Reprinted) jamapediatrics.com

Copyright 2017 American Medical Association. All rights reserved.

Downloaded From: by a Kaohsiung Med Univ User on 04/19/2018

 Musculoskeletal Low Back Pain in School-aged Children Review Clinical Review & Education

Table 2. Principal Diagnostic Imaging Tools Available and Considerations for Use

Imaging Modality Best Use Advantages Disadvantages

Radiographs Fractures, including Low cost, fast, Radiation; high false-negative
spondylolysis and fractures of readily available rates for many causes of back pain
the spinous process and
transverse process
CT scan Apophyseal ring fracture; bony Fast; specific for Radiation
pathologic features, including bony pathologic
any suspected fractures features
Bone scan or SPECT scan Bony pathologic features, Highly specific for Radiation; injections (pain);
including suspected fractures bony pathologic nuclear medicine imaging
features provides physiological
information but poor anatomical
detail
MRI with intravenous Disc pathologic conditions, No radiation; Costly; time consuming; potential Abbreviations: CT, computed
contrast indicated if masses, ligamentous and detailed soft-tissue need for sedation in younger tomography; MRI, magnetic
inflammatory condition soft-tissue pathologic findings children resonance imaging; SPECT,
suspected conditions; pain associated with single-photon emission
neurologic symptoms computerized tomography.

activity-specific techniques, which allows for a controlled and pain- patients with a high-grade spondylolisthesis who have persistent
free progression back to activity (sport, play, or work) or activities radicular or neurologic symptoms.67 Conservative treatment is
of daily living.60 Any program must be reinforced with a home ex- less effective in pediatric vs adult HNP, but it is still first-line treat-
ercise routine that the patient performs during therapy and then as ment owing to fear of the skeletally immature pediatric spine
maintenance after its completion. It is the clinician’s challenge to mo- being more vulnerable to surgical trauma and iatrogenic
tivate the pediatric patient to adhere to this home exercise routine. deformity.69 Long-term success with conservative care is esti-
Bracing can be used in the treatment of LBP and includes soft mated at 25% to 50% for HNP without neurologic deficits.69
lumbar corsets as well as rigid braces, such as thoracolumbar spinal Occasionally, pediatric patients with musculoskeletal LBP will have
orthoses (Figure 3B). One of the major controversies noted in the no identifiable etiologic cause but will have persistent, recalcitrant,
care of school-aged children diagnosed with acute spondylolysis is high levels of pain; consultation with rheumatology and pain spe-
whether to use thoracolumbar spinal orthoses or other rigid cialists may be considered if the clinician suspects amplified mus-
bracing.43 The current evidence does not support the use of rigid culoskeletal pain.15
bracing in spondylolysis: a meta-analysis revealed that most pa-
tients have a successful clinical outcome with conservative treat- Prevention
ment (83.9% treatment success rate) regardless of bracing or no Skeletally immature individuals are more vulnerable to trauma and
bracing (level of evidence, 4).40 Some clinicians will use rigid brac- explosive muscle contractions, especially during periods of rapid
ing or soft lumbar corsets to provide analgesia by further restrict- growth.41 Preventive programs aimed at improving age-associated
ing any extension activity in patients not responding to rest alone, deficits in flexibility have been used to reduce injury, but no causal
although there is no evidence to support this use of bracing.63 relationship between flexibility and risk of injury has been docu-
To our knowledge, there are no specific evidence-based stud- mented, to our knowledge.7,41 More evidence exists to support pre-
ies examining oral medication in the treatment of LBP in school- season sports conditioning programs and neuromuscular training in
aged children. Most populations studied are adults or do not specify reducing injury rates.41 Children should begin strength and condi-
an age. A Cochrane review revealed that nonsteroidal anti- tioning programs several weeks before the start of a sport season,
inflammatory drugs are effective for short-term symptomatic pain allowing for gradual increases in frequency and intensity of training.7
relief for both acute and chronic LBP.64 Another Cochrane review Appropriate rest from training and specific repetitive motions (eg,
assessing the use of muscle relaxants in nonspecific LBP found they tumbling in gymnastics) allows for proper recovery.7,41 Most back in-
are effective in the treatment of pain, but clinicians must take care juries, and overuse injuries in general, can be avoided if the pedia-
in prescribing these medications owing to their associated central trician keeps a simple, evidence-based rule of thumb in mind: young
nervous system adverse effects, including drowsiness and athletes should not participate in more hours of sports in a week than
dizziness.65 Systemic glucocorticoid treatment may provide par- their number of age in years.70 Finally, LBP lingering longer than 2
tial pain relief for select patients with acute lumbosacral radiculopa- to 3 weeks in this population is not normal; if persistent, the child
thy, but existing evidence suggests that systemic glucocorticoid should be evaluated by a pediatrician.7
therapy has limited or no benefit.66
Consultation may be considered by the pediatrician in the
uncommon instance that a patient’s LBP is not responsive to con-
Conclusions
servative treatment. Surgery and other invasive interventions are
rare treatments for the conditions described in this article. Surgical Lowbackpainiscommoninschool-agedchildrenandiscausedbyava-
repair of the pars is infrequently used for a painful nonunion of a riety of individual conditions, most of which are of a benign, musculo-
spondylolytic lesion that has failed to respond to conservative skeletalorigin.Specificsoftheevaluation,workup,andtreatmentofLBP
therapy for a year.67 Watchful waiting of the asymptomatic child in this population will vary with the specific underlying cause. Ongoing
with a high-grade spondylolisthesis is safe and does not lead to research is needed to establish evidence-based best practices for the
complications.68 Surgical fusion may be indicated in individual treatment of many of the diagnoses discussed in this review.

jamapediatrics.com (Reprinted) JAMA Pediatrics March 2017 Volume 171, Number 3 285

Copyright 2017 American Medical Association. All rights reserved.

Downloaded From: by a Kaohsiung Med Univ User on 04/19/2018

 Clinical Review & Education Review Musculoskeletal Low Back Pain in School-aged Children

ARTICLE INFORMATION 12. Balagué F, Skovron ML, Nordin M, Dutoit G, Pol nerve root compression. Arch Phys Med Rehabil.
Accepted for Publication: August 27, 2016. LR, Waldburger M. Low back pain in schoolchildren: 2007;88(7):840-843.
a study of familial and psychological factors. Spine 30. Peh WC, Griffith JF, Yip DK, Leong JC. Magnetic
Published Online: January 30, 2017. (Phila Pa 1976). 1995;20(11):1265-1270.
doi:10.1001/jamapediatrics.2016.3334 resonance imaging of lumbar vertebral apophyseal
13. Hill JJ, Keating JL. Risk factors for the first ring fractures. Australas Radiol. 1998;42(1):34-37.
Author Contributions: Dr MacDonald had full episode of low back pain in children are
access to all the data in the study and takes 31. Kumar TV R, Rao R, Gadi D, Grover A. Lumbar
infrequently validated across samples and apophyseal ring fracture—a case report. J Clin Diagn
responsibility for the integrity of the data and the conditions: a systematic review. J Physiother. 2010;
accuracy of the data analysis. Res. 2015;9(5):RD01-RD02.
56(4):237-244.
Study concept and design: MacDonald, Rodenberg. 32. Trout AT, Sharp SE, Anton CG, Gelfand MJ,
Acquisition, analysis, or interpretation of data: 14. Bhatia NN, Chow G, Timon SJ, Watts HG. Mehlman CT. Spondylolysis and beyond: value of
Stuart, Rodenberg. Diagnostic modalities for the evaluation of pediatric SPECT/CT in evaluation of low back pain in children
Drafting of the manuscript: All authors. back pain: a prospective study. J Pediatr Orthop. and young adults. Radiographics. 2015;35(3):819-834.
Critical revision of the manuscript for important 2008;28(2):230-233.
33. Gottschlich LM, Young CC. Spine injuries in
intellectual content: All authors. 15. Houghton KM. Review for the generalist: dancers. Curr Sports Med Rep. 2011;10(1):40-44.
Administrative, technical, or material support: All evaluation of low back pain in children and
authors. adolescents. Pediatr Rheumatol Online J. 2010;8:28. 34. Remvig L, Jensen DV, Ward RC. Are diagnostic
Study supervision: MacDonald. criteria for general joint hypermobility and benign
16. Shah SA, Saller J. Evaluation and diagnosis of joint hypermobility syndrome based on
Conflict of Interest Disclosures: None reported. back pain in children and adolescents. J Am Acad reproducible and valid tests? a review of the
Additional Contributions: We thank the patient Orthop Surg. 2016;24(1):37-45. literature. J Rheumatol. 2007;34(4):798-803.
illustrated in Figure 3 for granting permission to 17. Denis F. The three column spine and its 35. Fabry G. Clinical practice: the spine from birth
publish this information. significance in the classification of acute to adolescence. Eur J Pediatr. 2009;168(12):
thoracolumbar spinal injuries. Spine (Phila Pa 1976). 1415-1420.
REFERENCES 1983;8(8):817-831.
36. Jancuska JM, Spivak JM, Bendo JA. A review of
1. Olsen TL, Anderson RL, Dearwater SR, et al. 18. Rodriguez DP, Poussaint TY. Imaging of back symptomatic lumbosacral transitional vertebrae:
The epidemiology of low back pain in an adolescent pain in children. AJNR Am J Neuroradiol. 2010;31(5): Bertolotti’s syndrome. Int J Spine Surg. 2015;9:42.
population. Am J Public Health. 1992;82(4):606-608. 787-802.
37. Micheli LJ, Wood R. Back pain in young
2. Taimela S, Kujala UM, Salminen JJ, Viljanen T. 19. Kim HJ, Green DW. Adolescent back pain. athletes: significant differences from adults in
The prevalence of low back pain among children Curr Opin Pediatr. 2008;20(1):37-45. causes and patterns. Arch Pediatr Adolesc Med.
and adolescents:a nationwide, cohort-based 20. Alqarni AM, Schneiders AG, Cook CE, Hendrick 1995;149(1):15-18.
questionnaire survey in Finland. Spine (Phila Pa 1976). PA. Clinical tests to diagnose lumbar spondylolysis
1997;22(10):1132-1136. 38. Auerbach JD, Ahn J, Zgonis MH, Reddy SC,
and spondylolisthesis: a systematic review. Phys Ecker ML, Flynn JM. Streamlining the evaluation of
3. Jeffries LJ, Milanese SF, Grimmer-Somers KA. Ther Sport. 2015;16(3):268-275. low back pain in children. Clin Orthop Relat Res.
Epidemiology of adolescent spinal pain: 21. Miller R, Beck NA, Sampson NR, Zhu X, Flynn 2008;466(8):1971-1977.
a systematic overview of the research literature. JM, Drummond D. Imaging modalities for low back
Spine (Phila Pa 1976). 2007;32(23):2630-2637. 39. Beutler WJ, Fredrickson BE, Murtland A,
pain in children: a review of spondyloysis and Sweeney CA, Grant WD, Baker D. The natural
4. Fritz JM, Clifford SN. Low back pain in undiagnosed mechanical back pain. J Pediatr Orthop. history of spondylolysis and spondylolisthesis:
adolescents: a comparison of clinical outcomes in 2013;33(3):282-288. 45-year follow-up evaluation. Spine (Phila Pa 1976).
sports participants and nonparticipants. J Athl Train. 22. Lonstein JE. Spondylolisthesis in children: 2003;28(10):1027-1035.
2010;45(1):61-66. cause, natural history, and management. Spine 40. Klein G, Mehlman CT, McCarty M.
5. Hestbaek L, Leboeuf-Yde C, Kyvik KO, Manniche (Phila Pa 1976). 1999;24(24):2640-2648. Nonoperative treatment of spondylolysis and grade
C. The course of low back pain from adolescence to 23. d’Hemecourt PA, Gerbino PG II, Micheli LJ. I spondylolisthesis in children and young adults:
adulthood: eight-year follow-up of 9600 twins. Back injuries in the young athlete. Clin Sports Med. a meta-analysis of observational studies. J Pediatr
Spine (Phila Pa 1976). 2006;31(4):468-472. 2000;19(4):663-679. Orthop. 2009;29(2):146-156.
6. Schmidt CP, Zwingenberger S, Walther A, et al. 24. Standaert CJ. Low back pain in the adolescent 41. DiFiori JP, Benjamin HJ, Brenner J, et al.
Prevalence of low back pain in adolescent athlete. Phys Med Rehabil Clin N Am. Overuse injuries and burnout in youth sports:
athletes—an epidemiological investigation. Int J 2008;19(2):287-304, ix. a position statement from the American Medical
Sports Med. 2014;35(8):684-689. Society for Sports Medicine. Clin J Sport Med. 2014;
25. Waicus KM, Smith BW. Back injuries in the
7. Purcell L, Micheli L. Low back pain in young pediatric athlete. Curr Sports Med Rep. 2002;1(1): 24(1):3-20.
athletes. Sports Health. 2009;1(3):212-222. 52-58. 42. Iwamoto J, Takeda T, Wakano K. Returning
8. Skoffer B, Foldspang A. Physical activity and 26. Haus BM, Micheli LJ. Back pain in the pediatric athletes with severe low back pain and
low-back pain in schoolchildren. Eur Spine J. 2008; and adolescent athlete. Clin Sports Med. 2012;31(3): spondylolysis to original sporting activities with
17(3):373-379. 423-440. conservative treatment. Scand J Med Sci Sports.
9. Jones GT, Watson KD, Silman AJ, Symmons DP, 2004;14(6):346-351.
27. Lavelle WF, Bianco A, Mason R, Betz RR,
Macfarlane GJ. Predictors of low back pain in British Albanese SA. Pediatric disk herniation. J Am Acad 43. Kim HJ, Green DW. Spondylolysis in the
schoolchildren: a population-based prospective Orthop Surg. 2011;19(11):649-656. adolescent athlete. Curr Opin Pediatr. 2011;23(1):
cohort study. Pediatrics. 2003;111(4, pt 1):822-828. 68-72.
28. Wassenaar M, van Rijn RM, van Tulder MW,
10. Troussier B, Davoine P, de Gaudemaris R, et al. Magnetic resonance imaging for diagnosing 44. Cavalier R, Herman MJ, Cheung EV, Pizzutillo
Fauconnier J, Phelip X. Back pain in school children: lumbar spinal pathology in adult patients with low PD. Spondylolysis and spondylolisthesis in children
a study among 1178 pupils. Scand J Rehabil Med. back pain or sciatica: a diagnostic systematic and adolescents, I: diagnosis, natural history, and
1994;26(3):143-146. review. Eur Spine J. 2012;21(2):220-227. nonsurgical management. J Am Acad Orthop Surg.
11. Harreby M, Neergaard K, Hesselsøe G, Kjer J. 2006;14(7):417-424.
29. Rabin A, Gerszten PC, Karausky P, Bunker CH,
Are radiologic changes in the thoracic and lumbar Potter DM, Welch WC. The sensitivity of the seated 45. Ruiz-Cotorro A, Balius-Matas R,
spine of adolescents risk factors for low back pain in straight-leg raise test compared with the supine Estruch-Massana AE, Vilaró Angulo J. Spondylolysis
adults? a 25-year prospective cohort study of 640 straight-leg raise test in patients presenting with in young tennis players. Br J Sports Med. 2006;40
school children. Spine (Phila Pa 1976). 1995;20(21): magnetic resonance imaging evidence of lumbar (5):441-446.
2298-2302.

286 JAMA Pediatrics March 2017 Volume 171, Number 3 (Reprinted) jamapediatrics.com

Copyright 2017 American Medical Association. All rights reserved.

Downloaded From: by a Kaohsiung Med Univ User on 04/19/2018

 Musculoskeletal Low Back Pain in School-aged Children Review Clinical Review & Education

46. Sato T, Hirano T, Ito T, et al. Back pain in 55. Bratton RL. Assessment and management of drugs for low back pain: an updated Cochrane
adolescents with idiopathic scoliosis: acute low back pain. Am Fam Physician. 1999;60 review. Spine (Phila Pa 1976). 2008;33(16):1766-1774.
epidemiological study for 43,630 pupils in Niigata (8):2299-2308. 65. van Tulder MW, Touray T, Furlan AD, Solway S,
City, Japan. Eur Spine J. 2011;20(2):274-279. 56. Majlesi J, Togay H, Unalan H, Toprak S. The Bouter LM; Cochrane Back Review Group. Muscle
47. Smorgick Y, Mirovsky Y, Baker KC, Gelfer Y, sensitivity and specificity of the slump and the relaxants for nonspecific low back pain:
Avisar E, Anekstein Y. Predictors of back pain in straight leg raising tests in patients with lumbar disc a systematic review within the framework of the
adolescent idiopathic scoliosis surgical candidates. herniation. J Clin Rheumatol. 2008;14(2):87-91. Cochrane Collaboration. Spine (Phila Pa 1976).
J Pediatr Orthop. 2013;33(3):289-292. 57. Delavan JA, Stence NV, Mirsky DM, Gralla J, 2003;28(17):1978-1992.
48. Brown R, Hussain M, McHugh K, Novelli V, Fadell MF. Confidence in assessment of lumbar 66. Levin K, Hsu PS, Armon C. Acute lumbosacral
Jones D. Discitis in young children. J Bone Joint Surg spondylolysis using three-dimensional volumetric radiculopathy: treatment and prognosis. https:
Br. 2001;83(1):106-111. T2-weighted MRI compared with limited field of //www.uptodate.com/contents/acute-lumbosacral
49. Taxter AJ, Chauvin NA, Weiss PF. Diagnosis and view, decreased-dose CT. Sports Health. 2016;8(4): -radiculopathy-treatment-and-prognosis?source
treatment of low back pain in the pediatric 364-371. =search_result&search=Acute lumbosacral 14
population. Phys Sportsmed. 2014;42(1):94-104. 58. van Rijn RM, Wassenaar M, Verhagen AP, et al. radiculopathy: Treatment and prognosis&
Computed tomography for the diagnosis of lumbar selectedTitle=2~150. Updated November 9, 2016.
50. d'Hemecourt PA, Zurakowski D, d'Hemecourt Accessed December 21, 2016.
CA, et al. Validation of a new instrument for spinal pathology in adult patients with low back
evaluating low back pain in the young athlete. Clin J pain or sciatica: a diagnostic systematic review. 67. Radcliff KE, Kalantar SB, Reitman CA. Surgical
Sport Med. 2012;22(3):244-248. Eur Spine J. 2012;21(2):228-239. management of spondylolysis and
59. Lane SK, Gravel JW Jr. Clinical utility of spondylolisthesis in athletes: indications and return
51. Henschke N, Maher CG, Ostelo RW, de Vet HC, to play. Curr Sports Med Rep. 2009;8(1):35-40.
Macaskill P, Irwig L. Red flags to screen for common serum rheumatologic tests. Am Fam
malignancy in patients with low-back pain. Physician. 2002;65(6):1073-1080. 68. Lundine KM, Lewis SJ, Al-Aubaidi Z, Alman B,
Cochrane Database Syst Rev. 2013;2(2):CD008686. 60. Standaert CJ, Herring SA, Pratt TW. Howard AW. Patient outcomes in the operative and
Rehabilitation of the athlete with low back pain. nonoperative management of high-grade
52. Daniels JM, Pontius G, El-Amin S, Gabriel K. spondylolisthesis in children. J Pediatr Orthop.
Evaluation of low back pain in athletes. Sports Health. Curr Sports Med Rep. 2004;3(1):35-40.
2014;34(5):483-489.
2011;3(4):336-345. 61. Sampsell E. Rehabilitation of the spine
following sports injury. Clin Sports Med. 2010;29(1): 69. Dang L, Liu Z. A review of current treatment for
53. Braun J, Inman R. Clinical significance of lumbar disc herniation in children and adolescents.
inflammatory back pain for diagnosis and screening 127-156.
Eur Spine J. 2010;19(2):205-214.
of patients with axial spondyloarthritis. Ann Rheum 62. Selhorst M, Fischer A, Graft K, et al. Timing of
Dis. 2010;69(7):1264-1268. physical therapy referral in adolescent athletes with 70. Jayanthi NA, LaBella CR, Fischer D, Pasulka J,
acute spondylolysis: a retrospective chart review. Dugas LR. Sports-specialized intensive training and
54. Dizdarevic I, Bishop M, Sgromolo N, Hammoud the risk of injury in young athletes: a clinical
S, Atanda A Jr. Approach to the pediatric athlete Clin J Sport Med. 2016;(Jun):22.
case-control study. Am J Sports Med. 2015;43(4):
with back pain: more than just the pars. Phys 63. Standaert CJ. Spondylolysis in the adolescent 794-801.
Sportsmed. 2015;43(4):421-431. athlete. Clin J Sport Med. 2002;12(2):119-122.
64. Roelofs PD, Deyo RA, Koes BW, Scholten RJ,
van Tulder MW. Nonsteroidal anti-inflammatory

jamapediatrics.com (Reprinted) JAMA Pediatrics March 2017 Volume 171, Number 3 287

Copyright 2017 American Medical Association. All rights reserved.

Downloaded From: by a Kaohsiung Med Univ User on 04/19/2018