2. Case Presentation
• A previously healthy 4 year old male
presents to the ED with a limp. He has been
reluctant to bear weight on his left leg during
the last 24h and has a temperature of 38.0°C.
• Hip exam: shows decreased ROM due to pain.
• ROS: 2 day hx of rhinorrhea and cough,
resolved 10 days ago.
3. Review of Septic Arthritis
• Septic arthritis (SA)
• The synovium is extremely vascular and contains
no basement membrane, allowing easy access for
bacteria to the synovial space.
• Either by hematogenous spread or direct invasion by
local soft tissue infection, and or penetrating trauma.
4. Review of Septic Arthritis
• Within 24-48 hrs of bacterial invasion:
• Infiltration by neutrophils
• Vascular congestion
• Synovial proliferation
• Within 1 week following bacterial invasion:
• Continual purulent effusion
• Continual synovial proliferation
• Infiltration by mononuclear cells
• Granulation tissue
• Abscess development
• Within 10 days after abscess formation:
• Cytokine induced protelytic enzymes are released
• End result is joint destruction and or systemic sepsis
5. Review of Transient Synovitis
• Transient Synovitis (TS)
• Arthritis secondary to a transient inflammation and
hypertrophy of the synovial membrane.
• Leads to slight effusion that causes bulging of the
anterior joint capsule.
• No definitive cause of TS is known
• Posttraumatic or allergic mechanisms
• Infectious etiology
• Antibody titers
6. Clinical Question
• In children presenting with acute hip pain, is
there a clinical tool or diagnostic test that will
distinguish between septic arthritis and
transient synovitis?
7. PICO
• Patient Population:
• Children with acute hip pain.
• Intervention:
• Clinical feature(s) and or diagnostic test(s) that can
confirm the diagnosis of SA vs TS.
• Comparison:
• The gold standard of joint aspiration.
• Outcome:
• Preventing an invasive procedure with a percentage of
certainty without missing an emergent diagnosis.
10. Validation Study
• The original study published by Kocher, et al. 1999,
described a clinical prediction rule based on four
independent multivariate predictors of SA of the hip:
• A hx of fever
• Non weight bearing
• ESR of 40mm/hr
• WBC >12,000
• The CPR was derived from the data of 82 pts with SA and
86 pts with TS evaluated between 1979-1996. Results
showed statistical significance with a p value of <0.0001 with
an ROC curve of 0.96.
11. Validation Study
• Objective: To evaluate the diagnostic
performance of the previously described
CPR for the differentiation between SA and
TS of the hip in children in a new pt
population.
12. Material and Methods
• Prospective study
• Patient Population
• All patients who presented to a major tertiary care children’s
hospital between 1997-2002 with an acutely irritable hip and a
differential diagnosis of TS or SA.
• 213 eligible consecutive patients
• 24: SA
• 27: Presumed SA
• 103: TS
• 59: Excluded- such as those with immunocompromise, renal
failure, neonatal sepsis, postoperative infection of the hip, later
development of rheumatologic disease, Calve-Perthes disease and
or due to osteomyelitis.
13. Materials and Methods
• True septic arthritis defined:
• Positive culture of joint fluid or
• WBC ≥50,000 in the joint fluid with a positive blood culture.
• Presumed septic arthritis defined:
• WBC ≥50,000 in the joint fluid with negative cultures of the joint
aspirate and blood.
• Transient synovitis defined:
• WBC <50,000 in the joint fluid.
• Negative culture and resolution of symptoms without antimicrobial
therapy and no further development of a disease process as
documented in the medical record.
14. Material and Methods
• Data obtained from all patients:
• Age, gender, date of presentation, duration of sx, hx of fever, hx
of recent infx, hx of recent abx, temp, ESR, WBC, evidence of
hip joint effusion on X-ray, results of gram stain, cell count,
differential, and cx of joint fluid.
• Fever:
• PO temp 38.5°C during wk before presentation.
• Weight bearing status:
• Based on clinical hx and was considered the inability or refusal to
bear weight even with support.
• Effusion:
• As side to side distance of > 2mm from the medial part of the
femoral head to the medial part of acetabulum on anteroposterior
pelvic radiograph.
15. Material and Methods
• Analysis of data
• Univariate analysis by Student T- test.
• Predictors with p<0.20 in univariate analysis were
entered into multivariate logistic regression using
backward selection to identify independent clinical
predictors of SA and TS groups.
• Receiver operating characteristic curve was than
constructed to assess the diagnostic performance of the
group of multivariate predictors in identifying SA.
16. Results of Multivariate Analysis
• Showed the same four independent multivariate
predictors of SA in the current population as they
had in the original population.
17. Results of Algorithm for
Probability of SA
PREDICTORS ORIGINAL
STUDY
NEW STUDY
Zero < 0.2% 2.0%
One 3.0% 9.5%
Two 40.0% 35.0%
Three 93.1% 72.8%
Four 99.6% 93.0%
20. Study Limitations
• CRP not included in new study.
• Authors admitted CRP shows greater benefit than ESR
for detecting SA in children. However, at the beginning
of the study CRP testing was limited at the hospital as the
study progressed it became routine testing. Therefore,
to avoid biases associated with incomplete data analysis
CRP was not incorporated into the CPR.
• Not an external validation.
21. Study Limitations
• Diagnostic cut off <50,000 WBC in joint fluid was
considered TS and >50,000 considered SA or
presumed SA. Other sources state 60,000 or
greater.
• When comparing SA to TS in multivariate analysis,
the study does not specify if its true SA and
presumed SA or just true SA.
• Variables might be different in a community based
population vs this tertiary care hospital’s
population.
22. Are the results of this diagnostic study
valid?
1. Was there an independent, blind comparison with a
reference (“gold”) standard of diagnosis?
• yes
1. Was the diagnostic test evaluated in an appropriate
spectrum of patients (like those in whom it would be used
in practice)?
• yes
1. Was the reference standard applied regardless of the
diagnostic test result?
• maybe
1. Was the test (or cluster of tests) validated in a second,
independent group of patients?
• yes
23. Are the valid results of this
diagnostic study important?
• Yes
24. Can we apply this valid, important evidence
about a diagnostic test in caring for our
patient?
1. Is the diagnostic test available, affordable, accurate, and
precise in our setting?
• yes
1. Can we generate a clinically sensible estimate of our
patients pre-test probability (from personal experience,
prevalence statistics, practice databases or primary
studies?
• yes
A. Are the study patients similar to our own?
• yes
A. Is it unlikely that the disease possibilities or
probabilities have changed since this evidence was
gathered?
• yes
25. Caring for our patient cont.
3. Will the resulting post test probabilities affect our
management and help our patient?
• yes
A.Could it move us across a test-treatment
threshold?
• yes
A.Would our patient be a willing partner in
carrying it out?
• yes
26. Conclusions
• In my patient’s case:
• My patient had 3 of the predictors (fever, elevated WBC
count, and inability to bear weight, CRP was not
elevated. I chose not to aspirate and dx him with TS.
• How would I make a better study?
• CRP
• External validation
• Strictly SA versus TS
27. Resources
Kocher MS., Zurakowski D., Differentiating between septic arthritis and
transient synovitis of the hip in children: an evidence based clinical
prediction algorithm. J Bone Joint Surg Am. 1999; 81: 1662-70.
Raheem B., Shojani, et al. Case-based review : Septic arthritis in patients
with pre-existing inflammatory arthritis. Canadian Medical Association
Journal. May 22, 2007; 176 (11).
Electronic based medical databases. E-medicine and UpToDate.
And…. Harriet :)
Leads to damage to the joint cartilage and increase in pressure within the joint, impairing bloody supply and leading to avascular necrosis of the femoral head, dislocation or osteomyelitis.
32-50% of pts have had a recent URI
About 50% have had elevated titers to M. pneumoniae or a range of viruses (sometimes more than one), including parvovirus B19
UTD and emedicine
Narrowed down to 5 that specifically discussed the clinical prediction rule developed by Kocher. Two of the studies were prospective studies while the other two had small patient populations. The one chosen was a prospective study that was a validation study by Kocher, the original author of the CPR.
ROC= receiver operating characteristic curve.
Mean duration of follow up was 11.8 months (5.9 to 23.7 months).
The predicted probability of septic arthritis of the hip from the prediction rule was similar to the actual distributions in the new patient population. The area under the ROC curve for the new population was 0.86 compared with 0.96 in the original population. Both close to 1 and indicating still very good diagnostic performance.
These are the data values for the original population and the new population to form the ROC curve.
