Best Evidence Topics
  • Send this BET as an Email
  • Make a Comment on this BET

The Sensitivity of Conventional Radiography in Acute Knee Trauma

Three Part Question

In [adults with acute knee trauma] what is the sensitivity of [conventional radiography] to [identify fractures]?

Clinical Scenario

A 23-year old man presents to the emergency department because of pain and swelling of the left knee after a motorcycle accident that occurred the day before. Clnical examination suggested the presence of joint effusion in the suprapatellar pouch with sever pain in the affected knee. A radiograph of the knee showed no signs of fracture. Computed tomography (CT) of the left knee revealed an oblique fracture of the tibial plateau. As the treating clinician, you question the sensitivity of plain x-rays to identify fractures in the knee.

Search Strategy

Medline 1966-010/15 using OVID interface, Cochrane Library (2015), and Embase.

("Knee Injuries/radiography"[MAJR]) AND ("Sensitivity and Specificity"[MeSH Terms]). Limit to English language.

Search Outcome

49 papers were identified; two were relevant to the clinical question.

Relevant Paper(s)

Author, date and country Patient group Study type (level of evidence) Outcomes Key results Study Weaknesses
Mustonen AO, et al.
2005
Finland
409 adult patients who had both a knee injury and a MDCT examination of the knee in the acute phase at a level 1 trauma center. Indications for MDCT included preoperative evaluation of known fractures and clinical evidence of knee fracture with normal radiographs.Retrospective observational studyPrimary outcome was descriptive analysis of MDCT findings of fracture location, pattern, and injury mechanism. Secondary outcome was comparison of MDCT findings to conventional radiography findings.The overall sensitivity of radiographs for fracture was 83%. Overall negative predictive value was 49%. Tibial fractures were the most common fracture location. Simple falls and traffic accidents were the most common injury mechanisms.Single-center study, and retrospective study design. Inclusion criteria (referral to level 1 trauma center and performance of MDCT scan of the knee) may represent selection bias and increase the prevalence of fracture in the population studied, altering the test characteristics.
Bonnefoy O, et al.
2006
France
48 consecutive patients with clinical suspicion of knee fracture were enrolled. They subsequently underwent conventional radiography, ultrasonography, and CT.Prospective cohort studyPrimary outcome was to determine diagnostic accuracy of ultrasonography in evaluating for indirect signs of intra-articular knee fracture as compared to conventional radiography and CT.Sensitivity of conventional radiograph for intra-articular fracture was 84%, while negative predictive value was 75%. Sensitivity of ultrasonography for intra-articular fracture was 94%, while negative predictive value was 89%.Small sample size, single-center study.

Comment(s)

This review of two studies examined the sensitivity of conventional radiography in detecting fractures of the knee in patients with acute knee trauma. Because of the morbidity associated with delayed diagnosis of knee fractures, understanding the diagnostic limitations of conventional radiography is important to ensure good patient outcomes for those with trauma to the knee. The data reviewed suggest that conventional radiography alone may be inadequately sensitive to rule out knee fracture in patients for whom there is high clinical suspicion of fracture. Additional imaging studies, such as CT or MRI, should be considered in these patients.

Editor Comment

Trauma J Jason

Clinical Bottom Line

The sensitivity of conventional radiography in adults with acute knee trauma is 83-84%.

References

  1. Mustonen AO, Koskinen SK, Kiuru MJ Acute knee trauma: analysis of multidetector computed tomography findings and comparison with conventional radiography Acta Radiologica 2005;46:866-74
  2. Bonnefoy O, Diris B, Moinard M, Aunoble S, Diard F, Hauger O Acute knee trauma: role of ultrasound European Radiology 2006 Nov;16(11):2542-8