Diagnostic accuracy of ankle x-rays: How often do we miss fractures? How can we improve?

Author: Jeremy Kim, MD (EM Resident Physician, Icahn School of Medicine at Mount Sinai) // Edited by: Alex Koyfman, MD (@EMHighAK, EM Attending Physician, UTSW / Parkland Memorial Hospital) & Justin Bright, MD (@JBright2021)

 Clinical Scenario

A 25-year-old male presents with a left ankle injury while playing basketball. There is swelling and ecchymosis diffusely with tenderness at the tip of the lateral malleolus. Should an x-ray be performed?


Traumatic ankle pain is a common presenting problem to emergency departments. While sprains are the most common injuries (most common = anterior talofibular ligament), the patient still needs to be assessed for fracture.1

Since its derivation and validation in 1992, the Ottawa ankle rules are the most frequently used clinical decision tool when considering to obtain ankle x-rays.2


Ottawa ankle rules: Ankle X-rays if pain in the malleolar zone PLUS one of the following:

  1. Bony tenderness at “A”
  2. Bony tenderness at “B”
  3. Unable to bear weight both immediately and in the ED

How often are fractures missed?

When x-rays are ordered based on the Ottawa ankle rules, multiple studies have shown sensitivity at nearly 100% in adult populations while decreasing the number of x-rays ordered by 20-40%.3,4 Similar sensitivity results are seen in pediatric populations, with near 100% sensitivity when categorizing Salter-Harris Type I as clinically insignificant due to low complication rate.5

One of the original multi-center validation studies for Ottawa ankle rules in 1995 revealed a 0.5% fracture miss rate, mostly because providers misused the clinical decision rule or in one case had difficulty assessing the grossly swollen ankle.3 However, at a similar rate, patients who underwent imaging had the x-rays initially read as negative.

Mortise view
Lateral view

What is the sensitivity of ankle x-rays?

The initial radiographic evaluation of the ankle involves three views (lateral, AP, and mortise). The mortise view is taken in 15-20 degrees of internal rotation with x-ray beams projecting perpendicular to the intermalleolar line. In an effort to save time and resources, some institutions use a two-view approach (lateral PLUS either AP or mortise), but this has a sensitivity of only 85-98%.7,8 Overall, the diagnostic accuracy of ankle x-rays is unknown, but one study suggests that the incidence of occult fractures is about 1% and the degree of effusion correlated with the likelihood of occult fracture.9

In high-energy, poly-trauma, evaluation for ankle injuries may be more difficult due to the presence of distracting injuries or difficulty manipulating the extremity for adequate views on x-ray. In this scenario, when compared to multi-detector CT as a gold standard, sensitivity of x-rays can be as low as 87% for calcaneal fractures, 78% for talar fractures and 25-33% for midfoot fractures.10

Why do we miss fractures?

  • Inadequate physical exam (eg. palpating incorrect areas and misusing the Ottawa ankle rules or gross swelling making palpation difficult)
  • Unreliable patient (ie. altered, distracted, or intoxicated)
  • Misinterpreting x-rays
  • Occult fracture on x-rays, requiring further imaging

How can we improve?

  • Thorough physical exam
    • Palpate posterior edge of distal 6 cm of both lateral and medial malleoli
    • Assess for associated proximal fibular fracture (Maisonneuve fracture): palpate the proximal fibula
    • Assess for syndesmotic ligament injury. Squeeze test (squeeze at mid-calf causes pain just proximal to ankle) or external rotation of foot (with tibia stabilized, causes pain at level of syndesmosis)11
    • Assess for open fracture. Check for skin breakdown.
    • Neurovascular exam
  • Review any imaging you order! You have the advantage of having examined the patient and have additional localization clues to detect fractures.
  • Consider CT imaging if concerning mechanism or if needed for surgical considerations by orthopedic colleagues (eg. Pilon fracture)
  • Instruct patient to follow-up within 1 week if no improvement in pain/ability to walk. Repeat x-rays at that time may reveal fracture
  • Excellent resource on the approach to ankle x-rays: http://radiopaedia.org/articles/ankle-radiograph-an-approach


References / Further Reading

  1. Tintinalli’s Emergency Medicine Manual. 7th Ed. Chapter 175: Ankle and foot injuries.
  2. Stiell I et al. Implementation of the Ottawa ankle rules. JAMA. Mar 1994; 271(11):827-32.
  3. Stiell I et al, Multicentre Ankle Rule Study Group. Multicentre trial to introduce the Ottawa ankle rules for use of radiography in acute ankle injuries. BMJ. Sept 1995; 311:594-7.
  4. Bachmann LM et al. Accuracy of Ottawa ankle rules to exclude fractures of the ankle and mid-foot: a systematic review. BMJ. Feb 2003; 326(7386): 417.
  5. Myers A, Canty K, Nelson T. Are the Ottawa ankle rules helpful in ruling out the need for x-ray examination in children? Arch Dis Child. 2005; 90:1309-11.
  6. Radiology masterclass: Trauma x-ray / Lower limb / ankle. <http://radiologymasterclass.co.uk/tutorials/musculoskeletal/x-ray_trauma_lower_limb/ankle_fracture_x-ray.html#top_second_img>
  7. Brandser EA et al. Contribution of individual projects alone and in combination for radiographic detection of ankle fractures. Am J Roentgenol. Jun 2000; 174(6):1691-7.
  8. De Smet AA, et al. Are oblique views needed for trauma radiography of the distal extremities? Am J Roentgenol. 1999; 172(6):1561-5.
  9. Ho K et al. Using tomography to diagnose occult ankle fractures. Ann Emerg Med. May 1996; 27(5): 600-5.
  10. Haapamaki VV, Kiuru MJ, Koskinen SK. Ankle and foot injuries: analysis of MDCT findings. Am J Roentgenol. Sep 2004; 183(3):615-22.
  11. Busconi BD, Stevenson JH. Sports Medicine Consult: a problem-based approach to sports medicine for the primary care physician. Lippincott: 2009.
  12. Berbaum KS et al. Impact of clinical history on radiographic detection of fractures: a comparison of radiologists and orthopedists. Am J Roentgenol. 1989; 153(6):1221-4.

2 thoughts on “Diagnostic accuracy of ankle x-rays: How often do we miss fractures? How can we improve?”

Leave a Reply

Your email address will not be published. Required fields are marked *