Fever and Upper Back Tenderness
Radiology Cases in Pediatric Emergency Medicine
Volume 7, Case 1
Daniel C.H. Kidani, Medical Student
Loren G. Yamamoto, MD, MPH
Kapiolani Medical Center For Women And Children
University of Hawaii John A. Burns School of Medicine
     The patient is an 8 year old male who presents to 
the emergency department with a chief complaint of a 
two day history of fever.  His maximum temperature at 
home was noted to be 39.8 degrees C, axillary.  He 
also has a slight cough but no coryza.  A brief review of 
systems is unremarkable.
     Examination in the ED:  VS  T 38.7 degrees C, P 
115, RR 20, BP 114/49, oxygen saturation in room air 
97%.  He is comfortable and is in no apparent distress.  
He appears non-toxic.  HEENT unremarkable without 
signs of acute infection.  Neck is supple without 
adenopathy, but his right sternocleidomastoid muscle is 
slightly tender.  He has reproducible point tenderness 
over the thoracic spine, but he demonstrates full range 
of motion of the spine.  Heart regular rhythm with no 
murmurs or extra heart sounds.  Lungs are clear to 
auscultation.  Abdomen soft, non-tender, no masses 
palpable, normal bowel sounds.  No guarding.  
Non-distended.  No hepatosplenomegaly.  Normal male 
genitalia.  His extremities are normal with good 
perfusion.  His sensation and strength are good.  DTRs 
are normal.

     What diagnoses are possible.  What laboratory 
studies or radiographs might be helpful at this point?  

     A CBC, blood culture and ESR are obtained.  
Because of the point tenderness over his thoracic 
spine, thoracic spine radiographs were ordered to 
investigate the possibility discitis.  His WBC is 10.8 with 
66% segs, 28% bands, Hgb 13, Hct 49, platelets 
220,000, ESR 3.

View his thoracic spine radiographs.




     What are the radiographic findings in discitis?  
These studies revealed a narrowing of the T3-T4 
interspace.  This narrowing is best seen on the AP film.  
It is not easily identified on the lateral film.

View a coned down view of this area in some repeat 
radiographs.



     These coned down views of the T3-T4 region show 
the narrowing on the AP and lateral views.

     Based on these radiographs, he is diagnosed with 
suspected discitis.  He is hospitalized and IV antibiotics 
are started.  His fever eventually subsides during his 
stay in the hospital and he is discharged on the fourth 
hospitalization day.  His blood culture does not grow 
any organisms.  His PPD is negative.  A repeat ESR 
during hospitalization is 2.

      
Discussion
     This patient does not have a typical presentation for 
discitis, but his clinical and radiographic findings 
suggest this possibility.
     Although discitis has many clinical presentations, its 
typical presentation is usually a young child, most often 
under the age of 5, who presents with no or low-grade 
fever (1).  The affected disk can be at the lumbar, 
thoracic, or cervical regions; however, the affected disk 
is most often found in the lumbar region of the spine 
which induces limping or the inability to walk.  There is 
also point tenderness around the area of inflammation 
and a marked hesitation to flex the spine, which is 
characteristically held stiff.  Although the patient in this 
case shows point tenderness over the thoracic spine, 
which led the radiographic investigation for the 
possibility of discitis, he did not demonstrate this 
characteristic splinting or hesitation in flexing the spine.
     In a patient with suspected discitis, a CBC, blood 
culture, ESR and radiographs may be helpful (1).  All 
these investigations were done.  Lab findings revealed 
a borderline or normal WBC at 10,800, which is 
common in discitis; however, his ESR, which is typically 
elevated in discitis, was found to be low, at 3.  These 
two tests are nonspecific for discitis and are commonly 
used to track the course of the disease.  His blood 
cultures were found to be negative throughout the 
course of his hospitalization.  This doesn't help to rule 
out discitis since 50%-70% of cultures, both blood and 
disk space, are found to be negative (1).  The patient's 
radiographs are characteristic for discitis, 
demonstrating narrowing of the affected disk space.  
He shows disk narrowing at the T3-T4 disk, which 
corresponds to the point tenderness demonstrated on 
physical examination.  There was no evidence of soft 
tissue abnormalities or associated bone destruction, 
which may help to rule out a secondary infection from a 
paravertebral abscess or osteomyelitis.  Therefore, 
even though the case presentation is rather atypical of 
discitis, the diagnosis of discitis is possible in this case 
given the radiographic evidence and the clinical 
findings of fever and point tenderness over the affected 
disk.  The administration of a PPD test was given to 
rule out tuberculosis infection, which has a higher 
propensity for the spine than other bones (2).

     Intervertebral discitis, the association of back pain, 
progressive loss of intervertebral disk height, and 
erosion of adjacent vertebral end plates (3), is an 
uncommon entity.  It is estimated that there are 1 to 2 
cases of childhood discitis in a hospital that evaluates 
32,500 patients a year.  Discitis has a biphasic age 
distribution, with a higher incidence early in childhood 
and a subtler peak during adolescence (1).
     The pathogenesis of discitis is not clearly 
understood.  Its propensity to develop in the pediatric 
population is believed to be due to anatomical changes 
that occur with age.  In early childhood, vascular 
channels within the cartilaginous region of the 
intervertebral disk space as well as intraosseous 
arterial anastomoses allow for a hematogenous route of 
infection.  With age, the intervertebral disk gradually 
devascularizes.  After the age of 30, it is thought that 
disk inflammation is secondary to direct invasion (i.e. 
paravertebral abscess, vertebral osteomyelitis).  
Although it widely held that discitis is caused by an 
infection, typically a blood borne bacterial or viral 
infection of low virulence (4), the infectious agent is not 
always clear since most cultures are sterile and many 
patients recover without antimicrobial therapy (1).  
Culture of the intervertebral disk by needle biopsy are 
positive more often than blood cultures.  However, all 
cultures, blood or disk space, are negative in most 
patients (50%-70%).  This has led some to believe that 
discitis may be caused by a partial dislocation of the 
epiphysis, secondary to a flexion injury (5).  When 
cultures are positive, the most common organism is 
Staphylococcus aureus.
     Discitis occurs most often in patients under the age 
of 5 who have no or low-grade fever.  Discitis is 
uncommon in children 8 years or older (1).  The 
affected disk is usually in the lumbar region, which 
causes a progressive limp or the inability to walk.  The 
affected disk shows evidence of decreased height 2-4 
weeks into the illness with variable degrees of damage 
to the vertebral end plates (1).  However, some patients 
present with severe back pain, high fever, and 
bacteremia, which is commonly seen in osteomyelitis.  
Other patients show much milder symptoms even 
though bacteria is isolated from blood cultures; still 
others, show few signs and symptoms and no evidence 
of infection (3).
     The clinical manifestations may vary with age.  
Infants may refuse to eat and be fussy, toddlers may 
refuse to walk, while adolescents typically complain of 
back, abdominal, or pelvic pain.  However, the spine is 
characteristically held stiff or in a splinted position, with 
a tendency to avoid flexing the lumbar spine.  The area 
of inflammation should be evident on physical 
examination.  The lumbar lordosis is commonly 
reversed and may be associated with paravertebral 
muscle spasms.  The presence of fever is variable (3).

     Differentiation between discitis and vertebral 
osteomyelitis is difficult early in the disease course as 
the two entities are considered by some to be different 
stages of the same disease spectrum.  However, 
vertebral osteomyelitis typically occurs in older 
children, with a peak incidence in adolescence, who 
present with fever and complain of lumbar, thoracic, or 
cervical back pain.  The patient is more likely to be ill 
appearing and have a higher, longer lasting fever at the 
time of diagnosis compared to the patient with discitis.  
Radiographic evidence, which also lags behind clinical 
symptoms, shows rarefaction of the involved vertebral 
body followed by deterioration of bone, typically from 
the anterior region of the vertebral body.  MRI is the 
diagnostic study of choice for patients with suspected 
vertebral osteomyelitis since it is 96% sensitive and 
93% specific on evaluation for this disease (1). 
     Radiographic evidence for discitis lags behind 
clinical signs and symptoms.  Radiographs are usually 
interpreted as normal shortly after the onset of clinical 
symptoms; however, subtle changes that are evident 
early in the disease process (i.e. end plate 
demineralization or irregularity) are often detectable in 
retrospect.  Two to four weeks into the disease 
process, progressive narrowing of the intervertebral 
disk is evident.  If the radiographs demonstrate the 
characteristic changes associated with discitis, this 
supports the diagnosis of discitis.  MRI may be more 
definitive.
     Radionuclide scans may be useful in the early 
detection of vertebral infections, such as discitis and 
osteomyelitis.  One study demonstrated that bone 
scans allowed for an earlier diagnosis of disc space 
infection relative to non-scanned patients, with an 
average time after admission of 3 days compared to 11 
days for the non-scanned group (5).  However, bone 
scans cannot differentiate between different disease 
states and must be interpreted alongside other clinical 
and diagnostic studies.  It has been suggested, in the 
advent of MRI, that nuclear bone scans should only be 
used in very young patients in whom localization of 
symptoms to the spine cannot be done solely on 
physical examination (1).
     CT can facilitate the diagnosis of discitis by 
confirming the presence of disc narrowing and vertebral 
end plate destruction earlier in the course of the 
disease (4).  However, like nuclear bone scans and lab 
studies, CT typically provides nonspecific information 
(1).  Thus, it is especially useful when clinical evidence 
of infection associated with back pain and neurologic 
symptoms are present.  CT is also better at 
demonstrating areas of inflammation than a 
conventional radiograph.
     MRI is the imaging study of choice for discitis.  It is 
more sensitive, 96%, than either bone scans or CT in 
the diagnosis of disc space infections.  Characteristic 
changes include irregularity and destruction of the 
vertebral end plates and body, increased signal 
intensity on T2 weighted images, bone destruction 
within adjacent vertebrae which are evident as low 
intensity signals on T1 weighted images and high 
intensity signals on T2 weighted images, and soft 
tissue changes on all images (6).
     If the patient shows constitutional symptoms and is 
suspected of having a bacterial cause of infection, 
initial treatment with immobilization and appropriate 
antibiotics is indicated.  The most common bacterial 
isolate in discitis is Staph aureus; however, cultures 
should guide antibiotic therapy.  A minimum of 4-6 
weeks of antibiotic therapy should be initiated if a 
patient has a positive culture, elevated WBC, or 
persistently elevated ESR.  In patients who show 
minimal symptoms and who have normal labs, the use 
of antibiotic therapy is less clear.  Immobilization in 
such cases may be sufficient treatment (3).  
Recommendations for immobilization include the use of 
strict immobilization with a spica cast or modified 
immobilization with bed rest with early mobilization in a 
halo vest (7).  It is recommended that a brace be worn 
for 3 months following mobilization (5).
     Operative procedures are indicated in patients who 
do not respond to initial therapy, when there are 
negative blood or closed biopsy cultures, neurologic 
deficit, paravertebral abscess, and progressive spinal 
deformity (7).  In children, aggressive surgical 
treatment is rarely needed, except in the cases of 
tuberculosis or other caseating diseases that are not 
responding to antibiotics (8).  Patients treated 
surgically typically have a more rapid resolution of 
symptoms which may be secondary to decompression 
of the infected disk space (9).  Gebhard and Brugman 
have shown that percutaneous discectomy of the 
infected disc is effective in relieving symptoms, 
obtaining a bacterial diagnosis, and helped to eliminate 
the infection (9).
     Discitis usually carries a good prognosis.  Mortality 
is extremely low, approaching zero, and recurrence is 
uncommon.  However, residual back pain, limited spinal 
mobility and neurologic deficit can occur (7).  In a study 
by Bernard et al., which followed up 35 children (an 
average of 17 years after they had intervertebral 
discitis), 43% had residual back pain, 91% had normal 
flexion of the lower back, 86% had markedly restricted 
lumbar extension, and 80% had a narrowing of the 
vertebral canal.  These results did not seem to be 
influenced by the type of treatment the patients 
received (10).  It was also suggested that the incidence 
of back pain increased with age.       

References
     1.  Fernandez M, Carrol CL, Baker CJ.  Discitis and 
Vertebral Osteomyelitis in Children: An 18-Year 
Review.  Pediatrics 2000;105(6):1299-1304.
     2.  Maguire JH.  Osteomyelitis.  In:  Braunwald E, 
Fauci AS, Hauser SL, Isselbacher KJ, Kasper DL, 
Longo DL, Martin JB, Wilson JD (eds).  Harrison's 
Principles of Internal Medicine, 14th Edition CD-ROM. 
McGraw-Hill, New York, 1998, pp824-827.
     3.  Wood GW.  Infections of the Spine.  In:  
Behrman RE, Kliegman RM, Jenson HB (eds).  Nelson 
Textbook of Pediatrics, 16th ed.  W.B. Saunders Co.,  
Philadelphia, 2000, pp3094-3119. 
     4.  Garcia FF, Semba CP, Sartoris CC, Sartoris DJ.  
Diagnostic Imaging of Childhood Spinal Infections. 
Orthopaedic Review 1993;22(3):321-327.
     5.  Glazer PA, Hu SS.  Pediatric Spinal Infections.  
Orthopedic Clinics of North America 
1996;27(1):111-123.
     6.  Ozuna RM, Delamarter RB.  Pyogenic Vertebral 
Osteomyelitis and Post Surgical Disc Space Infections.  
Orthopedic Clinics of North America 1996;27(1):87-94.
     7.  Harris LF, Haws FP.  Disc Space Infection. 
Alabama Medicine 1994;63(7):12-14. 
     8.  Wood II GW.   Infections of the Spine.  In:  
Canale ST (ed).  Campbell's Operative Orthopaedics, 
9th ed., Mosby Inc., St. Louis, 1998, pp3094-3119.
     9.  Gebhard JS, Brugman JL.  Percutaneous 
Discectomy for the Treatment of Bacterial Discitis.  
Spine 1994;19(7):855-857.
     10.  Jansen BR, Hart W, Schreuder O.  Discitis in 
Childhood, 12-35-year Follow-up of 35 Patients.  Acta 
Orthop Scand 1993;64(1):33-36.

Return to Radiology Cases In Ped Emerg Med Case Selection Page

Return to Univ. Hawaii Dept. Pediatrics Home Page

Web Page Author:
Loren Yamamoto, MD, MPH
Professor of Pediatrics
University of Hawaii John A. Burns School of Medicine
Loreny@hawaii.edu