The Ottawa knee rules are a validated clinical decision rule that tells clinicians when a knee X-ray is genuinely needed after an acute injury — and, just as importantly, when it can be safely skipped. Fractures are found in fewer than 7% of adults who present to an emergency department with acute knee trauma, yet the majority still undergo radiography. The calculator on this page walks you through the five criteria in under a minute and returns a plain-language answer.
Educational reference only. This page and its calculator support clinical reasoning; they do not replace assessment by a qualified clinician and are not medical advice or a prescription. Always follow your prescriber’s or clinician’s judgement and confirm any figure against the primary source before acting on it.
Clinical decision rule · Emergency & primary care
Ottawa Knee Rule calculator
Examine the knee on the map below. The tool applies the five validated criteria and tells you whether radiography can be safely omitted.
01 Can the rule be used at all?
The rule was derived and validated only in blunt knee trauma sustained within the past 7 days. Tick anything that applies.
02 The examination
Tap a bony landmark on the knee to record tenderness there. Every criterion below is a required input — nothing is assumed.
Right knee · anterior
Criterion 1 is met at 55 years or older.
Criterion 4 is met below 90°. Passive assistance is allowed.
Two steps on each leg. A limp still counts as bearing weight. Criterion 5 is met only if the patient could not do it at both time points.
03 Result
Updates live as you examine. Recomputed from scratch on every change.
Awaiting input
Enter an age and confirm applicability
Age is a required input and is never assumed. Confirm the applicability statement in step 1 to see a result.
0 of 5 criteria met
Show the logic and the worked calculation
The rule is a Boolean OR, not a weighted score. No criterion outranks another.
Waiting for input.
Independent check. Recomputed by counting positives directly from the raw inputs, without reusing the value above:
Waiting for input.
Probability of fracture after this result
Illustrative Bayesian estimate using pooled adult accuracy (sensitivity 99%, specificity 49%). Adjust the pre-test probability to your own population — a fracture is found in roughly 6–7% of adults with acute knee trauma.
Waiting for a result.
These figures describe fracture only. They say nothing about ligament or meniscal injury, which plain radiography does not detect.
Ottawa Knee Rule summary — clinicaltoolslibrary.com
| # | Criterion | How to elicit it |
|---|---|---|
| 1 | Age 55 years or older | Chronological age at presentation. |
| 2 | Isolated tenderness of the patella | The patella is tender and no other knee bone is tender. Any second tender bony site voids this criterion. |
| 3 | Tenderness at the head of the fibula | Palpate the fibular head directly for bone tenderness. |
| 4 | Inability to flex the knee to 90° | Ask the patient to bend the injured knee to a right angle. Passive assistance is allowed. |
| 5 | Inability to bear weight for four steps | Both immediately after the injury and at assessment. Two steps on each leg counts, even with a limp. |
In the derivation studies the “knee” comprised the patella, the head and neck of the fibula, the proximal 8 cm of the tibia and the distal 8 cm of the femur.
| Population | Sensitivity | Specificity | Source |
|---|---|---|---|
| Adults, prospective validation (n≈1,096) | 100% | ≈48% | Stiell et al., JAMA 1996 |
| Adults, pooled systematic review | ≈99% | ≈49% | Bachmann et al., Ann Intern Med 2004 |
| Adults, pooled meta-analysis | High; specificity varies by setting | Heterogeneous | Sims et al., Eur Radiol 2020 |
| Children older than 5 years | 99–100% | ≈43–46% | Bulloch 2003; Vijayasankar 2009 |
Why specificity is low, and why that is acceptable. The rule was engineered as a rule-out instrument. A specificity near 50% means roughly half of those who screen positive will have a normal radiograph. The rule identifies confidently whom not to image; a positive result simply returns the decision to clinical judgement.
Impact. Prospective validation suggested a potential 28% reduction in knee radiography. A multicentre implementation trial achieved a 26% reduction with shorter emergency-department stays and lower cost per visit, and no fractures missed.
Children. Validation supports use above roughly 5 years of age. Below that the evidence is insufficient and the criteria are difficult to elicit reliably. Note that criterion 1 can never be met in a child, so four criteria carry the rule.
Error 1 — treating a limp as inability to bear weight
Weight-bearing means transferring weight twice onto each leg. How the patient looks doing it is irrelevant.
Limps four steps
Weight-bearing is intact. Criterion 5 is not met. A limp is not a positive finding.Cannot take four steps — then and now
Criterion 5 is met. Radiography is indicated.Error 2 — “inability at one time point” is enough
It is not. The criterion requires inability immediately after the injury and at the time of assessment.
Could not walk then, walks four steps now
Criterion 5 is not met.Could not walk then, still cannot now
Criterion 5 is met.Error 3 — forgetting the word “isolated”
Criterion 2 is isolated patellar tenderness. Tenderness of the patella plus a tender femoral condyle is not a positive criterion 2 — this calculator disarms it automatically and tells you.
Patella tender, nothing else
Criterion 2 is met.Patella and tibial plateau tender
Criterion 2 is not met — though such a patient will almost always trigger another criterion, and widespread bone tenderness warrants independent clinical judgement.Error 4 — reading a negative result as “no injury”
A negative rule makes fracture very unlikely. Ligament tears, meniscal injuries and cartilage damage are invisible to both the rule and a plain radiograph, and still need examination and follow-up.
Sources
- Stiell IG, Greenberg GH, Wells GA, et al. Prospective validation of a decision rule for the use of radiography in acute knee injuries. JAMA. 1996;275(8):611–615. PubMed 8594242
- Bachmann LM, Haberzeth S, Steurer J, ter Riet G. The accuracy of the Ottawa knee rule to rule out knee fractures: a systematic review. Ann Intern Med. 2004;140(2):121–124. PubMed 14734335
- Sims JI, Chau MT, Davies JR. Diagnostic accuracy of the Ottawa Knee Rule in adult acute knee injuries: a systematic review and meta-analysis. Eur Radiol. 2020;30:4438–4446. European Radiology
- Vijayasankar D, Boyle AA, Atkinson P. Can the Ottawa knee rule be applied to children? A systematic review and meta-analysis. Emerg Med J. 2009;26(4):250–253. PubMed 19307383
- Physiopedia. Ottawa Knee Rules. physio-pedia.com
Full disclaimer
This calculator is published by clinicaltoolslibrary.com as an educational reference and decision aid. It does not provide medical advice, a diagnosis, or a recommendation to image or to treat. It computes and informs; the clinical decision remains with the treating clinician.
Output is generated entirely from the values entered and is only as reliable as those values. The Ottawa Knee Rule is a peer-reviewed and widely adopted decision rule; it is not investigational. Nothing here is off-label or unapproved. Local imaging protocols vary, and reference figures should be verified against the primary literature and current institutional guidance before they are relied upon.
The rule addresses fracture only. It does not exclude ligamentous, meniscal or cartilaginous injury.
Reporting problems. Errors, unexpected behaviour or content concerns in this tool can be reported to clinicaltoolslibrary.com. Adverse events involving a medical device, medicine or diagnostic in the United States can be reported to the FDA through MedWatch; in the United Kingdom through the MHRA Yellow Card scheme.
Last reviewed: July 2026 Next review due: July 2027 Published by clinicaltoolslibrary.com
What are the Ottawa knee rules?
The rule was derived by Ian Stiell and colleagues at the University of Ottawa and published in Annals of Emergency Medicine in 1995, then prospectively validated in JAMA in 1996. It is a rule-out instrument: if none of the five findings is present, a clinically significant fracture is highly unlikely and radiography can usually be omitted. If one or more is present, an X-ray is indicated — but a positive result is a prompt to image, not a diagnosis of fracture.
In the validation cohort of more than a thousand adults the rule was 100% sensitive for fracture and would have reduced knee radiography by roughly 28%. A later implementation trial published in JAMA showed a 26% fall in knee films, shorter emergency-department stays and lower cost per visit, with no missed fractures.
The five Ottawa knee rule criteria for imaging
Radiography is indicated when any one of the following is present. The criteria are deliberately simple, require no equipment, and were designed for bedside use in a busy department.
| # | Criterion | How to assess it |
|---|---|---|
| 1 | Age 55 years or older | Chronological age at presentation. |
| 2 | Isolated tenderness of the patella | The patella is tender and there is no bone tenderness anywhere else in the knee. |
| 3 | Tenderness at the head of the fibula | Palpate the fibular head directly for bone tenderness. |
| 4 | Inability to flex the knee to 90° | Ask the patient to bend the injured knee to a right angle; passive assistance is allowed. |
| 5 | Inability to bear weight for four steps | Both immediately after the injury and at assessment. Two steps on each leg counts, even with a limp. |
Who the Ottawa knee rules apply to — and who they do not
The rule was derived and validated in adults with blunt knee trauma sustained within the previous seven days. In the original studies the “knee” comprised the patella, the head and neck of the fibula, the proximal 8 cm of the tibia and the distal 8 cm of the femur.
It should not be applied when any of the following is present:
- Injury more than seven days old, or a return visit for reassessment of the same injury
- Isolated superficial skin injury with no underlying bone or soft-tissue involvement
- Altered level of consciousness, intoxication, or paraplegia
- Multiple trauma or a distracting fracture elsewhere
- Pregnancy, or referral from another facility with radiographs already taken
Ottawa knee rules PDF, pocket cards and mobile calculators
Because the rule has only five items, most clinicians carry it as a laminated card or a phone shortcut rather than a PDF. If you want a printable version, the five criteria and exclusions in the chart above are the complete rule — nothing else is required to apply it. The calculator on this page reproduces the same logic that appears in tools such as MDCalc, so results should agree exactly. For the source description, the Physiopedia summary of the Ottawa knee rules is a compact, referenced overview.
Sensitivity and specificity: how accurate is the rule?
The Ottawa knee rules were engineered for near-perfect sensitivity at the cost of specificity — which is exactly what a rule-out tool requires. A systematic review in Annals of Internal Medicine (Bachmann et al., 2004) pooled the adult evidence and reported approximately 99% sensitivity and 49% specificity for knee fracture. A more recent meta-analysis in European Radiology (Sims et al., 2020) confirmed high sensitivity with pooled specificity varying substantially between settings.
| Population | Sensitivity | Specificity | Key source |
|---|---|---|---|
| Adults, prospective validation | 100% | ~48% | Stiell et al., JAMA 1996 |
| Adults, pooled systematic review | ~99% | ~49% | Bachmann et al., 2004 |
| Children older than 5 years | 99–100% | ~43–46% | Bulloch et al. 2003; Vijayasankar et al. 2009 |
What the numbers mean in practice. A specificity near 50% tells you that roughly half of patients who screen positive will have a normal X-ray. That is an accepted trade-off: the rule tells you confidently whom not to image. A positive result simply returns the decision to clinical judgement.
Using the Ottawa knee rules in children
Multicentre validation and a later meta-analysis in the Emergency Medicine Journal support use in children older than about five years, with sensitivity comparable to adults. Below that age the evidence is thin and the criteria — particularly reliable four-step weight-bearing and localisation of tenderness — are harder to elicit, so most paediatric services default to clinical judgement. Note that the age ≥55 criterion is, by definition, never met in a child; the other four carry the rule.
Ottawa knee and ankle rules: a paired approach to limb trauma
The Ottawa ankle rules came first, in 1992, and share the same design philosophy: a handful of bedside findings, tuned for sensitivity, aimed at cutting low-yield radiography. Clinicians frequently apply both in the same encounter, because a fall or twisting injury can be ambiguous about where the pain originates. The two rules are, however, entirely separate instruments with different criteria — a negative ankle rule says nothing about the knee.
Two related decision tools are worth knowing. The Pittsburgh decision rules restrict knee radiography to patients with blunt trauma or a fall who are either under 12 or over 50, or who cannot take four weight-bearing steps; comparative studies suggest similar sensitivity with somewhat higher specificity, but far less external validation. The Canadian C-Spine rule, also from the Ottawa group, applies the same logic to cervical-spine imaging.
Explore related tools on this site: the Ottawa ankle rules calculator, the Pittsburgh knee rules calculator, and the Canadian C-Spine rule calculator.
Ottawa knee rules: frequently asked questions
What are the five Ottawa knee rule criteria?
Knee radiography is indicated if the patient is 55 or older, has isolated patellar tenderness, has tenderness at the head of the fibula, cannot flex the knee to 90 degrees, or cannot bear weight for four steps both immediately after the injury and at assessment. Any single positive criterion is sufficient.
What is the sensitivity and specificity of the Ottawa knee rules?
Pooled adult data give a sensitivity of roughly 99% and a specificity of roughly 49% for clinically significant knee fracture. The rule is designed to rule fracture out, not to rule it in.
Can the Ottawa knee rules be used in children?
Yes, in children older than about five years, where reported sensitivity is 99–100%. The evidence in younger children is insufficient, and the criteria are less reliable to elicit, so clinical judgement should govern.
Does a negative Ottawa knee rule mean nothing is injured?
No. It means a clinically significant fracture is very unlikely. Ligament tears, meniscal injuries and other soft-tissue damage are not detected by the rule — or by a plain X-ray — and still require examination and follow-up.
Are the Ottawa knee and ankle rules the same thing?
No. They are two distinct decision rules from the same research group. The knee rule has five criteria; the ankle rule assesses malleolar and midfoot tenderness plus weight-bearing. Applying one does not substitute for the other.
How much imaging do the Ottawa knee rules actually save?
Prospective validation suggested a potential reduction in knee radiography of about 28%, and a multicentre implementation trial achieved a 26% reduction alongside shorter emergency-department stays and lower cost per visit, with no fractures missed.
Is a knee X-ray still reasonable if the rule is negative?
Sometimes. The rule informs but does not override clinical judgement. Escalating pain, an unreliable examination, high-energy mechanism, or an inability to follow up may all justify imaging despite a negative result. Discuss the decision with your clinician.
