Written and maintained by CASRAI Editorial Board
Last updated
The Ottawa Ankle Rules are a clinical decision rule that tells an emergency clinician when an ankle or foot radiograph series is not needed after acute blunt injury. They are one of the most heavily validated decision instruments in emergency medicine, and their operational purpose is narrow and specific: they are a rule-out tool. A negative result — no tenderness at the specified bony landmarks and preserved weight-bearing — makes a clinically significant fracture unlikely enough that imaging can reasonably be withheld. A positive result does not diagnose a fracture; it only means the patient falls outside the low-risk group and should be imaged.
For hospital quality, patient-safety and imaging-stewardship staff, this distinction is the whole point. The Ottawa Ankle Rules are usually adopted not as a bedside teaching aid but as an imaging-appropriateness intervention: a documented, auditable justification for the X-rays a department orders and the ones it does not. This page covers the criteria, the two anatomical zones, what the pooled evidence actually shows about sensitivity and specificity, the populations the rule was never derived for, and how to build it into clinical decision support and chart audit without over-claiming what it can do.
What the Ottawa Ankle Rules are, and where they came from
The rules were derived and published by Stiell and colleagues at the University of Ottawa (Stiell IG et al., Annals of Emergency Medicine, 1992;21(4):384–390), specifically to reduce the volume of ankle and foot radiography ordered for acute injuries in which the pretest probability of fracture is low. The core observation behind the derivation is that most patients presenting to an emergency department with an ankle injury do not have a fracture, yet historically nearly all of them were radiographed.
The instrument is actually two paired rules operating on two adjacent anatomical zones, and they are frequently — and unhelpfully — collapsed into a single name:
- The Ottawa Ankle Rule, governing whether an ankle radiograph series is indicated, based on the malleolar zone.
- The Ottawa Foot Rule (often called the Ottawa Midfoot Rule), governing whether a foot radiograph series is indicated, based on the midfoot zone.
A patient can be positive for one and negative for the other, and the correct response is to order only the corresponding series. A department that treats “Ottawa Ankle Rules positive” as a single binary flag will systematically over-order — ordering both ankle and foot films when only one is indicated. This is a common and entirely avoidable source of imaging waste, and it is a documentation problem as much as a clinical one: the chart needs to record which zone was positive, not just that the rule “applied.”
The decision criteria
Both rules share the same structure: pain in the relevant zone plus at least one of three findings — two bone-tenderness landmarks and one functional criterion.
Ankle radiograph series — the malleolar zone
An ankle series is indicated if there is pain in the malleolar zone and any one of the following:
| Finding | Location |
|---|---|
| Bone tenderness at the posterior edge or tip of the lateral malleolus | Distal 6 cm of the posterior fibula |
| Bone tenderness at the posterior edge or tip of the medial malleolus | Distal 6 cm of the posterior tibia |
| Inability to bear weight | Both immediately after the injury and for four steps in the emergency department |
Foot radiograph series — the midfoot zone
A foot series is indicated if there is pain in the midfoot zone and any one of the following:
| Finding | Location |
|---|---|
| Bone tenderness at the base of the fifth metatarsal | Lateral midfoot |
| Bone tenderness at the navicular | Medial midfoot |
| Inability to bear weight | Both immediately after the injury and for four steps in the emergency department |
The two details most often applied wrong
Palpation must be on bone, at the specified landmark. The tenderness criterion is bone tenderness along the posterior edge or tip of the malleolus — not soft-tissue tenderness over the anterior talofibular ligament, which is present in nearly every sprain and, if accepted as a positive, converts the rule into an order-everything rule. Palpating the anterior aspect of the malleolus rather than the posterior edge is the single most common misapplication.
Weight-bearing has a specific operational definition. “Unable to bear weight” means the patient could not transfer weight twice onto each foot — four steps — both at the scene immediately after the injury and again in the department. Limping counts as bearing weight. A patient who can take four steps, however painfully and however slowly, is weight-bearing for the purposes of the rule. Departments that let “won’t walk” substitute for “can’t walk” lose most of the rule’s specificity.
What the evidence actually shows
The Ottawa Ankle Rules have an unusually large validation literature, and the headline figure that gets quoted — “nearly 100% sensitive” — needs to be stated carefully rather than repeated as a slogan.
The most widely cited synthesis is Bachmann et al., BMJ, 2003, “Accuracy of Ottawa ankle rules to exclude fractures of the ankle and mid-foot,” which pooled 27 studies covering 15,581 patients. It reported a pooled sensitivity of approximately 98% (95% CI 96.3–99.3) for the ankle and 99% (95% CI 97.3–100) for the midfoot, against a median specificity in the low-to-mid 30% range. That combination — very high sensitivity, low specificity — is the signature of a well-behaved rule-out instrument.
A more recent systematic review and meta-analysis in BMC Musculoskeletal Disorders (2022), restricted to adults and to ankle fractures only, pooled 15 studies covering 8,560 patients and reported a lower sensitivity of 0.91 (95% CI 0.89–0.92) with a specificity of 0.25 (95% CI 0.24–0.26), alongside very high statistical heterogeneity between studies (I² above 94%).
Those two estimates are not interchangeable, and a quality department citing one should know why it differs from the other: they use different inclusion criteria, different populations, and different definitions of the target condition. The honest summary for a policy document is that the rule’s sensitivity is high but not perfect, that published estimates range meaningfully across studies, and that its specificity is genuinely poor — roughly a quarter to a third of patients without a fracture will still screen positive. Any local protocol that quotes a single sensitivity figure as though it were a fixed property of the instrument is overstating the evidence.
Why low specificity is a feature, not a defect
A decision rule intended to justify withholding imaging must be tuned so that false negatives — missed fractures — are rare, and it buys that by accepting a large number of false positives. The clinical consequence of a false positive is one unnecessary radiograph; the consequence of a false negative is a missed fracture discharged from the department. The rule is deliberately asymmetric, and it should be governed that way: local audit should focus on missed-fracture events and on whether the rule was applied correctly, not on the false-positive rate.
The stewardship case: imaging volume, radiation, cost and throughput
The reason patient-safety and quality programs adopt the Ottawa Ankle Rules is that ankle and foot injury is a very high-volume emergency presentation with a low fracture yield, so a validated rule-out instrument has leverage that few other decision rules match. Four benefits are commonly cited, and they are worth separating by how well each is evidenced:
- Reduced radiography volume. This is the rule’s primary and best-supported effect. Reported reductions vary substantially by baseline ordering behaviour and by how faithfully the rule is applied, so a department should measure its own before-and-after rate rather than adopting a published percentage as a target.
- Reduced radiation exposure. Real, though modest per patient — extremity radiography is a low-dose study. The stewardship argument here is cumulative and population-level, and it is strongest in patients who present repeatedly or who are young.
- Reduced cost and length of stay. Avoiding a radiograph removes an imaging charge, a transport, and a wait for interpretation. In a boarding-constrained department, the throughput effect can matter more than the cost saving — see CASRAI’s guide to emergency department boarding measures for how that time is actually captured and reported.
- Documentation and defensibility. A charted negative Ottawa assessment converts “no X-ray was ordered” from an unexplained omission into a documented application of a validated rule. This is often the benefit that persuades clinicians who are otherwise reluctant to withhold imaging.
For background on how imaging services are organised and what a radiograph series involves, see CASRAI’s overviews of medical imaging and radiology.
Where the rule does not apply
The Ottawa Ankle Rules were derived in a defined population, and the derivation study applied explicit exclusion criteria. Applying the rule outside that population is not a minor deviation — it is use of an instrument whose validated performance does not transfer. The commonly published exclusions are:
- Age under 18. The derivation cohort was adult, and the 2022 adult meta-analysis explicitly excluded paediatric patients.
- Intoxication or altered mental status. Both the tenderness criterion and the weight-bearing criterion depend on a reliable patient report; neither is trustworthy in an intoxicated or obtunded patient.
- Diminished sensation from a neurological deficit. Peripheral neuropathy, spinal cord injury or regional anaesthesia can abolish the bone tenderness the rule depends on.
- Multiple painful or distracting injuries. A more painful injury elsewhere can mask ankle tenderness and independently prevent weight-bearing, making both criteria uninterpretable.
- Delayed presentation. The rule was derived for acute injury; findings evolve over the following days, and presentations more than roughly a week and a half after injury fall outside the derivation window.
- Isolated skin injury, or return visits for reassessment. Neither situation matches the acute undifferentiated presentation the rule was built for.
- Pregnancy was an exclusion in the derivation, which is worth noting because it is frequently forgotten — though in practice the radiation-avoidance logic points the same direction.
Children specifically
Paediatric use is the most consequential of these caveats, because paediatric ankle injury is high-volume and the temptation to extrapolate is strong. A separate paediatric literature exists and generally supports high sensitivity in children above early school age, but it is a distinct evidence base from the adult validation studies, the open physis changes the injury pattern that matters, and alternative paediatric-specific instruments have been proposed. A hospital that wants to apply an ankle decision rule to children should adopt a rule validated in children and say so explicitly in its protocol — not silently extend the adult rule and cite adult sensitivity figures for it.
What a negative result does and does not exclude
The rule is calibrated to clinically significant fracture, not to every radiographic abnormality and certainly not to every injury. A negative Ottawa assessment does not exclude a severe ligamentous injury, a syndesmotic injury, an osteochondral lesion, an occult stress fracture, or a soft-tissue injury requiring follow-up. Patients discharged on the basis of a negative rule still need explicit return precautions, and a local protocol should say so in the same paragraph that authorises withholding the film.
Implementing the rule: decision support, order sets and audit
Adopting the Ottawa Ankle Rules as a stewardship intervention means treating them as a structured, recorded assessment rather than an informal heuristic. Practically, that involves:
- A discrete, structured chart element capturing the four data points per zone — zone pain, each tenderness landmark, and weight-bearing status — rather than free text. Free-text documentation cannot be abstracted, which means the intervention cannot be measured. See CASRAI’s guide to quality measure chart abstraction for why discrete capture matters to any measurement programme.
- Decision support at the point of order entry, firing on the ankle/foot radiograph order rather than at triage, so the prompt arrives at the moment the decision is actually made. An alert that fires too early is dismissed reflexively.
- Explicit exclusion handling. The rule’s exclusions must be selectable, not implicit — a clinician overriding the rule because the patient is intoxicated should be able to record that reason, so overrides can be separated from non-compliance during review.
- Zone-specific ordering. The order set should let a positive malleolar-zone assessment order an ankle series alone, without bundling a foot series.
- Missed-fracture review. The safety metric that matters is the false negative. Any fracture identified on a return visit after a documented negative Ottawa assessment should be reviewed — not necessarily as an error, since the rule is not 100% sensitive, but to confirm the rule was applied correctly and that return precautions were given.
- Handoff discipline. Where the assessment is performed by one clinician and the disposition made by another, the assessment and its exclusions need to travel with the patient; see SBAR handoff communication.
Related decision rules and structured bedside instruments
The Ottawa Ankle Rules sit in a family of Ottawa-derived emergency decision rules from the same research group, including the Ottawa Knee Rule and the Canadian C-Spine Rule, all built on the same design philosophy: high sensitivity, explicit exclusions, and a defined target condition. Departments generally find that adopting one successfully makes the next easier, because the documentation and decision-support scaffolding is shared.
Within CASRAI’s patient safety cluster, related structured assessment instruments include the Wells Criteria for DVT and PE (the closest analogue — another pretest-probability rule that gates a diagnostic test), the Glasgow Coma Scale, the NIH Stroke Scale, the ASA Physical Status Classification, the SOFA and qSOFA scores, the Modified Early Warning Score, the Braden Scale and the RASS scale.
Frequently asked questions
What are the Ottawa Ankle Rules in one sentence?
They are a validated clinical decision rule indicating that an ankle radiograph is needed only if there is malleolar-zone pain plus bone tenderness at the posterior edge or tip of either malleolus or inability to bear weight for four steps, with an equivalent rule for the midfoot zone based on the base of the fifth metatarsal and the navicular.
How sensitive are the Ottawa Ankle Rules?
High, but not perfect, and estimates vary. The widely cited 2003 BMJ meta-analysis reported pooled sensitivity of about 98% for the ankle and 99% for the midfoot across 15,581 patients; a 2022 adult-only meta-analysis of ankle fractures reported 0.91 across 8,560 patients with high between-study heterogeneity. Specificity is low in every analysis — roughly 25–35% — which is expected for a rule-out instrument.
Does a positive result mean there is a fracture?
No. A positive result means the patient is not in the low-risk group and should be radiographed. Because specificity is low, most positive patients will not have a fracture.
What counts as being able to bear weight?
Transferring weight twice onto each foot — four steps — both immediately after the injury and in the emergency department. Limping counts. Only a patient who cannot complete four steps meets the inability-to-bear-weight criterion.
Can the Ottawa Ankle Rules be used in children?
Not by direct extrapolation. The derivation cohort was adult and major adult meta-analyses exclude paediatric patients. A separate paediatric evidence base exists, and a hospital applying an ankle rule to children should adopt and cite a rule validated in that population rather than quoting adult sensitivity figures.
When should the rules not be applied at all?
In intoxicated or altered patients, patients with diminished sensation from a neurological deficit, patients with multiple distracting injuries, delayed presentations, isolated skin injuries, return visits for reassessment, and — per the original derivation — pregnancy and age under 18.
Do the rules exclude ligament injury?
No. They are calibrated to clinically significant fracture only. A negative assessment does not rule out significant ligamentous or syndesmotic injury, osteochondral lesions or occult stress fractures, so discharged patients still need return precautions and follow-up guidance.
Should a positive ankle assessment trigger a foot X-ray too?
No. The ankle rule and the foot rule are separate and govern separate radiograph series. Ordering both when only one zone is positive is a common and avoidable source of imaging waste, which is why order sets should keep the two zones distinct.
