Muscle, joint & pain

When a Knee Injury Actually Needs an X-Ray

Save

A knee that has just been twisted or struck feels serious, and the instinct is that a picture will settle it. Emergency clinicians use a short checklist instead, built in the 1990s from more than a thousand knee injuries, designed to catch every fracture worth catching while sparing everyone else the radiation and the wait. Knowing what it checks makes the decision less arbitrary.

Last updated: July 2026

Talk to a clinician

Gale can help you find a clinician in your state and request a visit.

Find care →

What Are the Ottawa Knee Rules?

The Ottawa Knee Rule is a five-criterion checklist that tells a clinician whether an acutely injured knee needs a plain X-ray. It was derived from 1,047 adults seen for knee injury at two university-hospital emergency departments, where researchers recorded 23 standardized clinical findings on every patient and worked out which handful actually predicted a fracture 1. The rule is positive if any one criterion is present.

The five criteria are:

  • Age 55 years or older
  • Isolated tenderness of the patella — the kneecap is sore and nothing else around the joint is
  • Tenderness at the head of the fibula, the bony bump on the outer side just below the joint line
  • Inability to flex the knee to 90 degrees
  • Inability to bear weight for four steps, both immediately after the injury and in the emergency department 1

One criterion is enough. This is not a score and nothing gets added up — a single positive finding indicates radiography.

The name is geographic rather than descriptive: the rule came out of the Ottawa Civic Hospital group, the same researchers behind the better-known ankle version.

What the Rule Was Built to Do

A rule designed to miss nothing buys that by being deliberately unfussy about false alarms, and the Ottawa Knee Rule is built exactly that way. In the derivation study it identified clinically significant knee fractures with a sensitivity of 1.0 (95% confidence interval 0.95 to 1.0) and a specificity of 0.54 (0.51 to 0.57) 1. Applied to that same population, it projected a 28.0% relative reduction in knee radiography, from 68.6% of patients down to 49.4% 1.

Sensitivity 1.0 means every fracture in the derivation sample was flagged. Specificity 0.54 means roughly half the people flagged turned out not to have one.

That asymmetry is the design, not a defect. An unnecessary X-ray costs some minutes, a small radiation dose, and a co-pay. A missed fracture costs considerably more. So the rule is tuned to over-call, which has a consequence worth naming plainly: a positive rule is not a prediction that something is broken. It is a statement that a fracture has not yet been excluded, and most people who get imaged on the strength of it have an intact knee.

The converse matters just as much. The rule's job is ruling out, and it is measured on how reliably it does that — not on responsiveness, reliability, or anything you could track over time. It has no score range and no cutoff in the usual sense, because it is a yes-or-no referral decision rather than a scale.

Why a Normal X-Ray Doesn't Mean a Normal Knee

An X-ray images bone, and most of what gets hurt in a twisted knee is not bone. Ligaments, menisci, cartilage and tendons are effectively invisible on a plain film, so a normal radiograph rules out a fracture and rules out very little beyond it. Meniscus tears are among the most common knee injuries there are, and whether one is treated without surgery, trimmed, or repaired depends on the tear's type and location — questions a plain film cannot answer 2.

Anterior cruciate ligament injuries are the same story. Their management runs along an entirely separate pathway of diagnosis, surgical timing, graft choice, bracing and return-to-sport decisions, set out at length in orthopaedic clinical practice guidelines 3. None of those decisions rests on a radiograph.

So "your X-ray is clear" answers one question: is the bone broken. That is a real and useful answer, and it is not the same as "nothing is wrong with your knee." Anyone still trying to work out why does my knee hurt a week after a clear film is asking a legitimate second question, and it usually needs an examination rather than another picture.

What Bearing Weight Actually Means Here

The weight-bearing criterion is the one people most often get wrong when describing their own injury. The rule asks about four steps, and it asks about two separate moments: immediately after the injury happened, and again in the emergency department 1. Written that way, the criterion is stricter than it sounds — someone who could not stand at the roadside but can take four steps in the department does not meet it on weight-bearing alone, though they may well meet one of the other four.

The tenderness criteria are similarly precise. Isolated tenderness of the patella means the kneecap is the only tender spot; pressing elsewhere around the joint does not reproduce it. Tenderness at the patella plus tenderness along the joint line is a different finding, and it is not what that criterion is asking about 1.

None of this is a self-examination. The findings are checked by someone who can compare both knees, feel where the bone actually is under the swelling, and weigh what a tender spot means alongside everything else in the history. Reading the criteria is useful for understanding a decision that has already been made, not for making it at home.

The Ankle Rules Are a Separate Rule

The ottawa ankle rules came first, are far more widely used, and are a genuinely different instrument — knowing one does not mean knowing the other. The ankle version is built in two zones: a malleolar rule covering the ankle proper and a midfoot rule covering the foot, each pairing bony tenderness at named anatomical landmarks with an inability to bear weight 4.

In the study that refined the rules and then validated them prospectively in a second independent stage, the refined rules were 1.0 sensitive (95% CI 0.93 to 1.0) for malleolar-zone fractures and 1.0 sensitive (95% CI 0.83 to 1.0) for midfoot-zone fractures, with a projected reduction of 34% of ankle series and 30% of foot series 4.

The midfoot confidence interval runs down to 0.83 — a reminder that "100% sensitive" rests on however many fractures happened to be in the sample.

Both rules share one design philosophy: a small number of findings, checked in under a minute, tuned to rule out rather than rule in. That is why they survived where longer scoring systems did not.

Where the Rule Stops

The Ottawa Knee Rule answers a narrow question about a narrow situation: an acute injury, in an adult, presenting for care. It was derived in 1,047 adults 1, and it says nothing at all about knee pain that arrived without an injury, pain that has been building over months, or a joint that swells on its own. A swollen knee with no injury is a different problem with a different workup, and so is knee osteoarthritis, a degenerative joint condition with its own range of nonsurgical and surgical treatment 5.

The rule also does not decide anything on its own. It functions as a floor rather than a ceiling: a clinician who is worried about a particular knee can image it regardless of what the checklist says, and decision rules of this kind are written to support that judgement, not to overrule it. If the account of the injury does not fit — a fall from height, a knee that was struck by a car bumper, a joint that looks visibly wrong — the checklist stops being the interesting part of the encounter.

What to Ask When You're Sent Home Without an X-Ray

Being sent home without imaging is, far more often than not, the rule working correctly rather than a corner being cut. It is still reasonable to leave with three things clear: what the clinician thought the injury was, what would change that assessment, and when to come back. Asking directly whether the Ottawa knee rule was used is a fair question, and it usually gets a straight answer.

Worth having spelled out before leaving:

  • What was ruled out — a fracture, specifically, and not a ligament or meniscus injury
  • What the working diagnosis is — a sprain, a contusion, a suspected meniscal tear
  • The follow-up interval if the knee is no better in a week or two
  • Which changes warrant coming back sooner — new inability to bear weight, a knee that locks or gives way, fever, or swelling that balloons over hours

Most acutely injured knees without a fracture settle with time, and the ones that do not tend to declare themselves within a couple of weeks. That second visit — with the swelling down and the knee examinable — is often more informative than any imaging done on the first day, which is part of why clinicians are comfortable waiting.

Common questions

The rule was derived in adults, and adult decision rules do not transfer automatically to growing skeletons — a child's growth plates create injury patterns that adult criteria were never built to detect. Paediatric knee injuries are generally assessed on their own terms, and a clinician seeing a child will not simply run the adult checklist and stop there.

The criteria are public and readable, but they were validated as findings checked by a clinician during an examination — someone able to locate the fibular head under swelling, compare both knees, and take the history alongside them. Reading the rule helps make sense of a decision already made. It is not reliable as a substitute for being examined after a significant injury.

Age entered the rule because it earned its place statistically: among the 23 clinical findings tested in the derivation study, age 55 and over was one of the few that independently predicted fracture. It is not a judgement about any individual's bone quality — it is a variable that improved how reliably the checklist caught fractures across a large group.

A clear film rules out a fracture and leaves ligament and meniscus injuries entirely unexamined, since neither shows on plain radiography. Instability, locking, or a knee that keeps buckling are examination findings rather than imaging findings, and they usually lead to a repeat physical exam and, where the exam suggests it, soft-tissue imaging such as MRI.

No. Specificity in the derivation study was 0.54, meaning roughly half the people the rule flags turn out to have no fracture at all. That is deliberate: the rule is built to catch every fracture rather than to predict which knees are broken. A positive result means a fracture has not been excluded yet, not that one is likely.

A knee radiograph is a low-dose study, well below the dose of a CT scan of the same region, which is part of why an unnecessary knee film is a modest cost rather than a serious one. The stronger arguments for a decision rule are time, expense, and avoiding incidental findings that lead to further tests without changing the plan.

Related

Say it back

How would you explain this to someone you love?

Two or three sentences, just as you’d say it. Gale reflects back what you focused on — a mirror, not a quiz.

Talk to a clinician

Gale can help you find a clinician in your state and request a visit.

Find care →

When an Injured Knee Needs Same-Day Assessment

  • —A knee or leg that looks visibly deformed or angulated after the injury, or a kneecap sitting somewhere it clearly should not be.
  • —An open wound over the joint, or a puncture or laceration that reaches down toward the knee — an open fracture or a joint that has been breached needs urgent care, not a wait-and-see.
  • —Numbness, pins and needles, a foot that has gone cold or pale, or an inability to lift the foot after a knee injury — findings that raise a nerve or blood-vessel problem.
  • —A hot, swollen knee with fever and severe pain on any movement, particularly when there was no real injury — a joint infection is treated urgently and is not what an injury rule is looking for.

Visible deformity, an open wound over the joint, or a cold, pale or numb foot after a knee injury is an emergency-department visit now — call 911 if the leg cannot be moved or the person cannot be transported safely.

This explains how a published clinical decision rule works. It is not an assessment of any particular knee, and it is not a substitute for being examined by a clinician who can see and handle the joint.

References

  1. 1.Stiell IG, Greenberg GH, Wells GA, McKnight RD, Cwinn AA, Cacciotti T, McDowell I, Smith NA (1995). Derivation of a Decision Rule for the Use of Radiography in Acute Knee Injuries. Annals of Emergency Medicine, 1995;26(4):405-413. doi:10.1016/S0196-0644(95)70106-0Supports the five criteria of the Ottawa Knee Rule (age 55 or older, isolated patellar tenderness, fibular head tenderness, inability to flex to 90 degrees, inability to bear weight for four steps both immediately and in the emergency department), the derivation cohort of 1,047 adults assessed on 23 standardized findings at two university-hospital emergency departments, the derivation-phase sensitivity of 1.0 (95% CI 0.95-1.0) and specificity of 0.54 (0.51-0.57) for clinically significant knee fracture, and the projected 28.0% relative reduction in radiography use from 68.6% to 49.4%.
  2. 2.American Academy of Orthopaedic Surgeons (OrthoInfo) (2024). Meniscus Tears. OrthoInfo — AAOS. link ✓Supports the lay-education claims that meniscus tears are among the most common knee injuries and that treatment depends on the tear's type and location, ranging from nonsurgical management to meniscectomy or repair.
  3. 3.American Academy of Orthopaedic Surgeons (AAOS) (2022). Management of Anterior Cruciate Ligament Injuries — Clinical Practice Guideline. AAOS. link ✓Supports the claim that ACL injury management is addressed by an orthopaedic-society clinical practice guideline covering diagnosis, surgical timing, graft choice, bracing and return-to-sport considerations.
  4. 4.Stiell IG, Greenberg GH, McKnight RD, Nair RC, McDowell I, Reardon M, Stewart JP, Maloney J (1993). Decision rules for the use of radiography in acute ankle injuries. Refinement and prospective validation. JAMA. 1993 Mar 3;269(9):1127-32. doi:10.1001/jama.1993.03500090063034 ✓Supports the two-zone structure of the Ottawa Ankle Rules (a malleolar-zone ankle rule and a midfoot-zone foot rule, each combining bony tenderness at named landmarks with inability to bear weight), the refinement-and-prospective-validation design, and the refined rules' sensitivities of 1.0 (95% CI 0.93-1.0) malleolar and 1.0 (95% CI 0.83-1.0) midfoot with projected reductions of 34% of ankle series and 30% of foot series.
  5. 5.American Academy of Orthopaedic Surgeons (OrthoInfo) (2024). Arthritis of the Knee. OrthoInfo — AAOS. link ✓Supports the lay-education claim that knee osteoarthritis is a distinct degenerative joint condition with its own range of nonsurgical and surgical treatment options, separate from acute injury assessment.

5 sources, numbered by first appearance. General health information, not medical advice. AI-assisted editorial content — citations link their sources. Editorial policy

← Back to the health library