DOI:10.2214/AJR.08.1115
AJR 2008; 191:1595-1600
© American Roentgen Ray Society
Value of MRI After Recent Diagnostic or Surgical Intervention in Children with Suspected Osteomyelitis
J. Herman Kan1,
Melissa A. Hilmes1,
Jeffrey E. Martus2,
Chang Yu3 and
Marta Hernanz-Schulman1
1 Department of Radiology and Radiological Sciences, Vanderbilt Children's
Hospital, Vanderbilt University, 2200 Children's Way, Nashville, TN
37232.
2 Department of Pediatric Orthopedics, Vanderbilt Children's Hospital,
Vanderbilt University, Nashville, TN.
3 Department of Biostatistics, Vanderbilt University, Nashville, TN.
Received April 24, 2008;
accepted after revision June 3, 2008.
Statistics supported in part by Vanderbilt CTSA grant 1 UL1 RR024975 from
the National Center for Research Resources, National Institutes of Health.
CME
This article is available for CME credit.
See
www.arrs.org
for more information.
Address correspondence to J. H. Kan
(herman.kan{at}vanderbilt.edu).
Abstract
OBJECTIVE. The purpose of our study was to evaluate the diagnostic
efficacy and impact of emergent MRI after recent intervention in children with
suspected osteomyelitis or septic arthritis.
MATERIALS AND METHODS. This retrospective case-control study in
children with suspected osteomyelitis or septic arthritis consisted of 34
study subjects (mean age, 5.3 years) who underwent MRI after intervention and
96 control subjects (mean age, 8.7 years) who underwent MRI without prior
intervention. Final diagnosis and management were abstracted from medical
records. Consensus MRI review of the study group was performed to evaluate
whether objective MRI criteria of osteomyelitis can be applied to patients who
have undergone prior intervention.
RESULTS. For the study and control groups, no difference was seen in
the final diagnosis of osteomyelitis (26.5% [9/34] and 29.2% [28/96],
p = 0.76), osteomyelitis or septic arthritis (41.2% [14/34] and 37.5%
[36/96], p = 0.70), cellulitis or pyomyositis (20.6% [7/34] and 34.4%
[33/96], p = 0.13), and noninfectious conditions (23.5% [8/34] and
13.5% [13/96], p = 0.17). Objective MRI criteria for osteomyelitis
were present in all nine patients with a final diagnosis of osteomyelitis and
were not present in the remaining 25 who did not have a final diagnosis of
osteomyelitis despite recent intervention. Repeat interventions were necessary
in the study group at a rate not significantly different from single
interventions in the control group (29.4% [10/34] and 27.1% [26/96],
p = 0.79).
CONCLUSION. Iatrogenic soft-tissue and bone edema related to recent
intervention in children with suspected osteomyelitis or septic arthritis does
not affect the diagnostic efficacy of MRI. Performing MRI before intervention
adds efficacy to patient management, prevents unnecessary interventions, and
guides surgical procedures when indicated.
Keywords: children • intervention • MRI • osteomyelitis • pediatric imaging • septic arthritis
Introduction
Untreated septic arthritis and osteomyelitis in children may result in
significant orthopedic morbidity because of potential physeal and epiphyseal
cartilage injury [1,
2]. These patients require
prompt care and potential surgical intervention to prevent complications such
as growth disturbance related to physeal insult, epiphyseal osteonecrosis,
premature arthritis, joint arthrodesis, and sepsis. Musculoskeletal MRI has
been shown to be useful in the management of these patients by confirming and
localizing the diagnosis of infection or establishing an alternative diagnosis
for the symptoms. When septic arthritis or osteomyelitis are present, MRI
further defines which cases should require surgical intervention and which can
be managed medically. If diagnostic or surgical intervention is indicated, MRI
provides a road map by defining the location and size of drainable abscesses
[3,
4].
At our institution, when patients present with characteristic clinical
features of septic arthritis or osteomyelitis, pediatric orthopedic surgeons
may perform diagnostic or surgical intervention without a preintervention MRI
examination. This situation arises when preintervention MRI requires on-call
sedation personnel and thus introduces treatment delay. When procedure
findings validate clinical findings of infection (purulence or positive
culture), these patients will be observed and treated appropriately with
antibiotics. However, a subset of these patients will subsequently be referred
for MRI when a high clinical suspicion of infection remains despite a negative
result after intervention. Unfortunately, iatrogenic injury to the soft
tissues or marrow before the MRI study may potentially confound our ability to
accurately exclude infection or to diagnose alternative causes for the
patient's symptoms.
Therefore, our purpose was to evaluate the diagnostic value and clinical
impact of urgent MRI after recent diagnostic or surgical intervention in
children with suspected osteomyelitis and septic arthritis.
Materials and Methods
Patient Population
This retrospective case-control study was approved by our institutional
review board with waiver of informed consent. A total of 136 emergent
contrast-enhanced MRI examinations were performed from March 2002 through
September 2007 for suspected osteomyelitis or septic arthritis. Patients with
an established diagnosis of osteomyelitis or septic arthritis who had elective
outpatient follow-up MRI studies were excluded. Among the initial 136 exami
nations, six MRI examinations were not included. Five were excluded because
MRI was performed more than 10 days after the initial intervention, and one
was excluded because chart review revealed that the patient was being
evaluated for nonpyogenic inflammatory arthritis despite the clinical
indication of osteomyelitis noted on the radiology requisition.
From the remaining 130 contrast-enhanced MRI examinations, 34 examinations
(mean patient age, 5.3 years; minimum, 0.3 years; maximum, 16.1 years) were
performed within 10 days after intervention (mean, 2.2 days: minimum, 0 days;
maximum, 7.0 days) and constitute the study group. The study group selection
consisted of patients whose MRI examinations were performed within 10 days of
the initial diagnostic or surgical intervention. This cutoff time was chosen
because our purpose was to evaluate the immediate changes in MRI resulting
from the intervention. Our control group was generated during the same period
and consisted of 96 examinations of patients who did not have an intervention
before MRI (mean age, 8.7 years; minimum, 0.08 years; maximum, 16.5
years).
For the study group of 34 subjects, 15 (44%) had lower extremity, 15 (44%)
had pelvis, and four (12%) had upper extremity MRI evaluation. For the control
group of 96 subjects, 53 (55%) had lower extremity, 33 (34%) had pelvis, and
10 (10%) had upper extremity MRI evaluation
(Table 1).
Clinical Data Analysis
Final clinical diagnosis, surgical intervention, and hospitalization were
abstracted from medical records. A patient was considered to have a diagnostic
or surgical intervention if either a diagnostic procedure (joint, marrow, or
soft-tissue aspiration) or treatment (arthrotomy or incision and drainage of
bone or soft tissue) was performed. Final diagnoses were categorized as
osteomyelitis, septic arthritis or osteomyelitis, cellulitis or pyomyositis,
or noninfectious causes on the basis of the discharge diagnosis and clinical
follow-up notes. In addition, the percentage who underwent subsequent
intervention after MRI was calculated. Erythrocyte sedimentation rate and WBC
at the time of MRI were also collected. For the study group, sedimentation
rate and WBC were available in 85% and 100% of patients, respectively. For the
control group, sedimentation rate and WBC were available in 78% and 85% of
subjects, respectively.
Image Analysis
All MRI examinations were performed on a 1.5-T Intera (Philips Healthcare)
or 1.5-T Advantage (GE Healthcare) imager. For all patients and control
subjects, biplane or triplane fluid-sensitive (T2-weighted with fat
saturation, proton density–weighted with fat-saturation, or STIR), and
single-plane T1-weighted non-fat-saturated sequences were obtained. Some
patients also had a single-plane susceptibility (either 2D or 3D
gradient-recalled echo with fat suppression) sequence performed. For all
patients and control subjects, a single-plane or biplane gadolinium-enhanced
T1-weighted fat-saturated sequence was also performed.
Imaging review of the study subjects was performed to assess whether
objective MRI criteria for osteomyelitis could still be applied to patients
who had undergone recent intervention before MRI. This consensus review was
performed by a board-certified pediatric radiologist who had additional
pediatric musculoskeletal radiology fellowship training with 3 years of
experience and by a board-certified radiologist with pediatric fellowship
training and 2 years of pediatric radiology experi ence. These studies were
reviewed with knowledge of the location of diagnostic or surgical interven
tion based on the operative notes and knowledge of the final discharge
diagnosis. These studies were evaluated for the presence of soft-tissue edema,
joint effusion, and marrow edema.
The studies were also evaluated to determine whether specific MRI features
of osteomyelitis were present. These features were intraosseous abscess,
cortical breach, subperiosteal abscess, and edema in bone or soft tissues that
was unexplained by recent surgical intervention (e.g., juxtacortical muscle or
marrow edema away from the needle aspiration tract) and in which the edema
pattern did not suggest a noninfectious alternative diagnosis. Soft-tissue
fluid collections along the surgical tract were not evaluated because it would
be impossible to distinguish primary or secondary soft-tissue abscess related
to osteomyelitis from iatrogenic fluid collections from recent intervention in
our study group.
View larger version (148K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1A —Postsurgical marrow edema without osteomyelitis in 3-year-old
girl who had complete negative workup for infection. T1-weighted unenhanced,
non-fat-saturated (A) and gadolinium-enhanced fat-saturated (B)
coronal images of pelvis show focal linear marrow edema (arrows)
consistent with marrow aspiration tract. MRI was requested after intervention
because of continued concern for underlying osteomyelitis. When surgical
approach is known, focal marrow edema (arrows) can be attributed to
marrow aspiration only. Notice extensive myositis related to surgical
intervention. This patient had no cause for her presenting symptoms after full
workup.
|
|
View larger version (145K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1B —Postsurgical marrow edema without osteomyelitis in 3-year-old
girl who had complete negative workup for infection. T1-weighted unenhanced,
non-fat-saturated (A) and gadolinium-enhanced fat-saturated (B)
coronal images of pelvis show focal linear marrow edema (arrows)
consistent with marrow aspiration tract. MRI was requested after intervention
because of continued concern for underlying osteomyelitis. When surgical
approach is known, focal marrow edema (arrows) can be attributed to
marrow aspiration only. Notice extensive myositis related to surgical
intervention. This patient had no cause for her presenting symptoms after full
workup.
|
|
Statistical Analysis
Proportions of subjects in both the study and control groups for the study
end points are summarized. Difference in proportions between groups was tested
using either the chi-square test or Fisher's exact test, as appropriate.
Continuous variables such as age, WBC, and erythrocyte sedimentation rate were
compared using a two-sample Student's t test between the groups.
Values for p less than 0.05 were considered statistically
significant. All tests were two-tailed. Statistical analyses were performed
using the statistical package SAS for Windows (version 9, SAS).
View larger version (174K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2A —Postsurgical marrow edema outside osteomyelitis tract in
15-month-old boy with no evidence of osteomyelitis. T2-weighted sagittal
(A) and proton density–weighted axial (B) fat-saturated
images of distal tibia show round focal area of edema (arrows)
consistent with anterior aspiration. When surgical approach for marrow
aspiration is known, imaging findings can be attributed to intervention and
not to early hematogenous osteomyelitis. This patient's final diagnosis was
septic arthritis without concomitant osteomyelitis.
|
|
View larger version (129K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2B —Postsurgical marrow edema outside osteomyelitis tract in
15-month-old boy with no evidence of osteomyelitis. T2-weighted sagittal
(A) and proton density–weighted axial (B) fat-saturated
images of distal tibia show round focal area of edema (arrows)
consistent with anterior aspiration. When surgical approach for marrow
aspiration is known, imaging findings can be attributed to intervention and
not to early hematogenous osteomyelitis. This patient's final diagnosis was
septic arthritis without concomitant osteomyelitis.
|
|
Results
Final clinical diagnoses for the study and control groups are listed in
Table 1.
View larger version (72K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3 —Osteomyelitis after initial surgical procedure in 6-month-old
girl. T1-weighted fat-saturated axial image of right foot after IV
administration of gadolinium shows large soft-tissue abscess (arrows)
and intraosseous calcaneal abscess (arrowhead). This child initially
underwent soft-tissue aspiration that yielded cellulitis with abscess
(arrows). After MRI, patient underwent second intervention to drain
intramedullary calcaneal abscess.
|
|
View larger version (168K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4A —Osteomyelitis after surgical procedure in 5-year-old girl who
initially underwent elbow joint aspiration that was negative. T2-weighted
(A) and T1-weighted (B) fat-saturated axial images of distal
humerus show intramedullary and subperiosteal abscess (arrows). After
MRI, patient was given diagnosis of osteomyelitis, and antibiotic regimen was
changed accordingly. A second operation was not performed because child was
improving clinically with antibiotics alone.
|
|
View larger version (178K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4B —Osteomyelitis after surgical procedure in 5-year-old girl who
initially underwent elbow joint aspiration that was negative. T2-weighted
(A) and T1-weighted (B) fat-saturated axial images of distal
humerus show intramedullary and subperiosteal abscess (arrows). After
MRI, patient was given diagnosis of osteomyelitis, and antibiotic regimen was
changed accordingly. A second operation was not performed because child was
improving clinically with antibiotics alone.
|
|
Before MRI, the 34 patients in the study population had undergone the
following procedures: 20 joint aspirations only, five both joint aspiration
and marrow aspiration (Figs.
1A,
1B and
2A,
2B), four arthrotomies, three
marrow aspirations only, and two soft-tissue abscess aspirations
(Fig. 3). For those who had
undergone joint aspiration only, six (30%) were subsequently diagnosed with
osteomyelitis (Fig. 4A,
4B) and one patient required
intraosseous débridement (Fig.
5A,
5B).
View larger version (123K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5A —Osteomyelitis with abscess after initial joint aspiration in
3-year-old boy who had undergone knee joint aspiration yielding clear bloody
fluid that was negative for infection 1 day before MRI was performed.
Unenhanced proton density–weighted fat-saturated sagittal (A) and
T1-weighted axial fat-saturated gadolinium-enhanced (B) images show
subperiosteal abscess (arrows) and diffuse small intramedullary
abscesses. Small size of joint effusion (arrowheads) likely
contributed to false-negative joint aspiration. After MRI, patient underwent
second intervention to drain intramedullary and subperiosteal abscesses.
|
|
View larger version (127K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5B —Osteomyelitis with abscess after initial joint aspiration in
3-year-old boy who had undergone knee joint aspiration yielding clear bloody
fluid that was negative for infection 1 day before MRI was performed.
Unenhanced proton density–weighted fat-saturated sagittal (A) and
T1-weighted axial fat-saturated gadolinium-enhanced (B) images show
subperiosteal abscess (arrows) and diffuse small intramedullary
abscesses. Small size of joint effusion (arrowheads) likely
contributed to false-negative joint aspiration. After MRI, patient underwent
second intervention to drain intramedullary and subperiosteal abscesses.
|
|
In the study group, MRI findings led to a need for additional intervention
in 29.4% of patients (10/34). Despite prior intervention, this rate for a
second intervention was similar to that of surgical intervention in the
control group, 27.1% (26/96; p = 0.79). In detail, for the 10 study
subjects requiring additional intervention, four subjects underwent
arthrotomy; three, bone incision and drainage; one, arthrotomy, bone incision,
and drainage; one, diagnostic joint aspiration; and one, soft-tissue abscess
incision and drainage. In detail, for the 26 control subjects requiring
intervention after MRI, 13 subjects underwent bone débridement; six,
soft-tissue abscess incision and drainage; three, arthrotomy, bone incision,
and drainage; one, arthrotomy; one, diagnostic joint aspiration; one,
diagnostic bone aspiration; and one, diagnostic bone and joint aspiration.
Nine subjects in the study population (intervention performed before MRI)
had a final diagnosis of osteomyelitis (Figs.
3,
4A,
4B,
5A,
5B). Features of osteomyelitis
were present in all nine subjects (Table
2) on consensus review by applying objective MRI criteria for the
diagnosis of osteomyelitis, as outlined in the Materials and Methods section,
as well as knowledge of the location and extent of intervention. Among these
subjects, eight of nine had one or more imaging criteria of osteomyelitis
present, including intraosseous abscess, cortical breach, or subperiosteal
abscess. The diagnosis in one subject who had acetabular osteomyelitis was
based on the presence of marrow edema with juxtacortical soft-tissue edema in
the obturator internus muscle away from the aspiration site.
Twenty-five subjects in the study population (intervention performed before
MRI) did not have a final diagnosis of osteomyelitis (Figs.
1A,
1B and
2A,
2B).
Table 2 lists the MRI features
in these 25 patients on consensus review. With knowledge that these patients
had previously undergone intervention, the diagnosis of osteomyelitis could be
excluded in all patients, including the five subjects who had marrow edema
present, on the basis of objective MRI criteria for the diagnosis of
osteomyelitis. None of these patients had characteristic imaging features of
osteomyelitis present.
Noninfectious causes for symptoms in the study population included
nonpyogenic inflammatory arthritis (3), toxic synovitis (3), left ankle
contusion (1), and osteonecrosis (1). No cause was found to explain symptoms
in four patients.
Noninfectious causes for symptoms in the control population included
cat-scratch disease (1), toxic synovitis (1), leukemia (1), nonpyogenic
inflammatory arthritis (3), rhabdomyosarcoma (1), osteonecrosis (2), stress
reaction medial sesamoid (1), muscle strain (1), intramuscular ganglion (1),
and tenosynovitis (1). No cause was found for 15 patients.
Discussion
Iatrogenic marrow edema on MRI from recent surgical intervention has been
reported to persist for as long as 13 months
[5]. In the immediate
postoperative setting, the presence of juxtacortical soft-tissue edema may
further complicate the diagnosis, thereby decreasing the specificity of
musculoskeletal MRI in excluding underlying infection. However, with correct
clinical and surgical history, we found that the patterns of soft-tissue and
marrow edema can be explained, and that MRI plays an important role in the
management of these patients because of its ability to evaluate underlying
osteomyelitis despite recent intervention (both diagnostic or therapeutic).
Thus, the impact of MRI in guiding appropriate care was not obviated by recent
intervention and was similar when compared with patients who were undergoing
MRI for suspected osteomyelitis or septic arthritis without recent
intervention.
MRI has been shown to be helpful in altering the medical and surgical
management in patients with a clinical diagnosis of osteomyelitis or septic
arthritis, which has been studied in patients with pelvic osteomyelitis
[3,
4]. Ideally, and as our results
show, MRI should be performed before intervention. Although performing MRI may
delay definitive treatment, require additional sedation in young children, and
add additional initial cost, its importance in determining the need for
diagnostic or surgical intervention could be seen in both our study patients
and our control subjects, and as we have found, can guide management, prevent
unnecessary surgery, and obviate a second procedure. MRI before intervention
will decrease the need for a second procedure, as seen in 29% of our study
population. When intervention is necessary, a surgical road map provided by
preoperative MRI may potentially decrease operative morbidity by decreasing
operative time and extent of surgical exposure required in comparison with
patients who do not undergo preoperative MRI.
There are accepted clinical practice guidelines for the workup and
diagnosis of septic arthritis
[6]. Current literature
advocates the diagnosis of septic arthritis based on clinical evaluation,
laboratory parameters, and joint aspiration without routinely including
preintervention MRI in the algorithm
[7,
8]. We believe that if patients
meet criteria for joint aspiration for suspected septic arthritis, they may
also benefit from preintervention MRI, provided that the study can be
performed in a timely fashion. This is because osteomyelitis may be clinically
indistinguishable from and may coexist with septic arthritis
[9]. Preintervention MRI may
help guide additional treatment and offer alternative diagnoses in the workup
of patients with septic arthritis, including coexisting pyomyositis or
osteomyelitis (as seen in 29% of patients in our selected study population who
underwent joint aspiration only before MRI) and may indicate noninfectious
causes for symptoms.
The limitations of this study in addition to its retrospective review
include that the study and control populations had dissimilar clinical
presentations. Nevertheless, the rate of intervention after MRI was similar in
both groups. Therefore, its clinical impact on management was similar. The
younger age of the study group compared with the control group likely reflects
the difficulty at our institution of scheduling MRI coordinated with sedation.
Older children, who are less likely to require sedation, are more likely to
undergo MRI before any intervention. Conversely, younger children who are more
likely to require sedation, tend not to undergo preintervention MRI because of
the difficulty and potential treatment delay of scheduling a sedated MRI.
Second, our consensus review for the evaluation of osteomyelitis in
postintervention patients was inherently biased because the final diagnosis
was known. The two radiologists involved with consensus review were too
familiar with the known osteomyelitis cases in the study group to perform a
truly blinded review. However, the intent of the consensus review was to
determine whether objective MRI criteria for osteomyelitis could still be
applied in the postintervention setting, and whether these facts would be
known during clinical practice. It was not possible to assess diagnostic
efficacy of MRI in our study group for cellulitis or septic arthritis because
iatrogenic soft-tissue edema or joint fluid related to recent intervention
were present. Gadolinium is used in our clinical practice setting to assess
for the presence of abscess and not as an adjunct to the diagnosis of
osteomyelitis. Therefore, unenhanced T1-weighted fat-suppressed images were
not obtained, and any specific value of these images in the diagnosis of
osteomyelitis is not addressed in this study.
In conclusion, iatrogenic soft-tissue and bone changes related to recent
intervention in children with suspected osteomyelitis or septic arthritis do
not affect the diagnostic efficacy of MRI. In the workup of musculoskeletal
infection, the efficacy of MRI in guiding further management of patients who
have recently undergone intervention is similar to that in patients who have
not undergone MRI before intervention. However, performing MRI before
intervention adds efficacy to patient management, guides the surgical
procedure, and prevents additional surgery in children with suspected pelvic
or appendicular osteomyelitis or septic arthritis.
References
- Offiah AC. Acute osteomyelitis, septic arthritis and discitis:
differences between neonates and older children. Eur J
Radiol 2006; 60:221
–232[CrossRef][Medline]
- Perlman MH, Patzakis MJ, Kumar PJ, Holtom P. The incidence of joint
involvement with adjacent osteomyelitis in pediatric patients. J
Pediatr Orthop 2000; 20:40
–43[CrossRef][Medline]
- Connolly SA, Connolly LP, Drubach LA, Zurakowski D, Jaramillo D.
MRI for detection of abscess in acute osteomyelitis of the pelvis in children.
AJR 2007; 189:867
–872[Abstract/Free Full Text]
- Karmazyn B, Loder RT, Kleiman MB, et al. The role of pelvic
magnetic resonance in evaluating nonhip sources of infection in children with
acute nontraumatic hip pain. J Pediatr Orthop2007; 27:158
–164[Medline]
- Ledermann HP, Kaim A, Bongartz G, Steinbrich W. Pitfalls and
limitations of magnetic resonance imaging in chronic posttraumatic
osteomyelitis. Eur Radiol 2000;10
:1815
–1823[CrossRef][Medline]
- Kocher MS, Mandiga R, Murphy JM, et al. A clinical practice
guideline for treatment of septic arthritis in children: efficacy in improving
process of care and effect on outcome of septic arthritis of the hip.
J Bone Joint Surg Am 2003;85
: 994–999[Abstract/Free Full Text]
- Kocher MS, Zurakowski D, Kasser JR. Differentiating between septic
arthritis and transient synovitis of the hip in children: an evidence-based
clinical prediction algorithm. J Bone Joint Surg Am1999; 81:1662
–1670[Abstract/Free Full Text]
- Jaramillo D, Treves ST, Kasser JR, Harper M, Sundel R, Laor T.
Osteomyelitis and septic arthritis in children: appropriate use of imaging to
guide treatment. AJR 1995;165
: 399–403[Abstract/Free Full Text]
- Kocher MS, Lee B, Dolan M, Weinberg J, Shulman ST. Pediatric
orthopedic infections: early detection and treatment. Pediatr
Ann 2006; 35:112
–122[Medline]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Top
Abstract
Introduction
