The Ottawa Knee Rule
Avoiding Unnecessary Radiographs in Sports
Paul J. Nugent, DO
THE PHYSICIAN AND SPORTSMEDICINE - VOL 32 - NO. 5 - MAY 2004
In Brief: Physicians are often called upon to evaluate acute knee injuries. Radiographs are frequently ordered, but they may not be helpful for making the diagnosis and guiding treatment of soft-tissue injuries. A set of clinical criteria, called the Ottawa knee rule, has been established to predict when radiographs are truly necessary. The sensitivity and negative predictive value of the rule were both validated at 100% for fractures of the knee in adults. By using the Ottawa knee rule, at least 25% of unnecessary films can be avoided without risking patient safety. The rule has not specifically been studied among sports participants, but indications are that it can be useful in this population.
More than 1.3 million acute knee injuries occur each year1; many result from athletic activity. In the past, patients who had acute knee injuries underwent routine radiography, even though this practice was based on little clinical data. Few knee films yield clinically relevant pathology,2 because the structures of the knee most likely to be injured (the menisci and ligaments) are radiolucent. Fractures are the chief acute structural pathology that can be seen on radiographs.
More than 90% of films ordered to evaluate knee injuries are normal.3 To address the problem of unnecessary knee films, a set of decision criteria known as the Ottawa knee rule was developed in the mid-1990s. Using the criteria does not necessarily exclude major knee injuries, but, if negative, the rule obviates the need for radiographs. This saves considerable time, resources, and money. While not meant to substitute for sound clinical judgment, the Ottawa knee rule is scientifically validated and can be used to augment a good history and physical examination to determine which knee injuries do not need radiographic analysis.4 Many physicians prefer to get x-rays in all sports injuries for legal reasons to make sure nothing is missed, but even in this medical-legal environment, it is important to follow the Ottawa knee rule and use an appropriate rationale for ordering x-rays. In addition, because many athletic venues are removed from x-ray facilities, the rule may be useful in sports settings.
The data in this review were primarily from articles obtained from a MEDLINE search using the key words "Ottawa knee rule" and from a review of knee-injury articles previously published in THE PHYSICIAN AND SPORTSMEDICINE.
Developing the Rule
The criteria (table 1) were established in 1996 subsequent to the better known Ottawa ankle rules.5 Both rules were established to address the issue of low predictive values for joint films after injury.
Validation. To validate prospectively the conclusions of the original work, 1,096 patients were entered into a follow-up study.6 Only 63 (6%) of these individuals had clinically important fractures. Clinically unimportant fractures were defined as avulsion fractures less than 5 mm. All 63 of the patients with fractures were identified. The sensitivity was determined to be 100%, the negative predictive value 100%, and the specificity 49%. Notably, 511 patients without fracture were excluded from films.6 The authors estimated that 28% of all knee films in this study population could be dispensed with if the rule were followed.
A number of other clinical studies4,7 have validated the Ottawa knee rule. In these studies, the number of knee films ordered decreased at least 25% to 50%, and no clinically important knee fractures were missed.
Additionally, a significant time savings was noted. The reported waiting time for patients was reduced by half an hour, with considerable cost savings.8 If hundreds of thousands of unneeded radiographs are not performed annually, our patients and society will see tremendous savings—in the range of tens of millions of dollars per year.
These studies have primarily been done in emergency department populations and reported in the emergency medicine literature, but it is likely that these study results could be generalized to acutely injured adults, whether seen in the emergency department, at a physician office, or at an athletic venue.
Attempted refinement. An attempt was made to further refine the Ottawa knee rule by substituting the inability to flex to 60° for the inability to flex to 90°. This refinement of the original criteria, when applied to the original study population, increased the specificity to 56% but with a small, but significant, increase in unidentified, clinically important fractures. However, physicians believed that the tradeoff was not worth it, and 60° flexion did not become part of the rule.6
More About the Criteria
The five criteria were established from a set of 23 clinical variables using multivariate analysis applied to a population of 1,047 adults who had knee injuries. The original analysis identified the variables most highly correlated with fracture, and these are reflected in the Ottawa knee rule.9
Age. Patients younger than 18 were not included in the knee rule. Because children have incompletely fused physes, they differ anatomically from adults. Injuries through the proximal tibial and distal femoral epiphyseal plates occur more often in children after knee trauma. While one retrospective study10 validated the rule with 100% sensitivity in children, another prospective study11 exhibited 1 missed fracture (a nondisplaced fracture of the proximal fibula) from a relatively small sample size of 234 patients. The authors' conclusion was that the rule should not be used in children, pending further evaluations.
However, a more recent prospective study12 found the Ottawa knee rule to be as consistent when applied to children as to adults. It found a 100% sensitivity and a 43% specificity in a large series of pediatric patients. The authors noted that if the rule was used in a pediatric population, 31% of radiographs could be eliminated, without any missed fractures. Their recommendation was made to strongly consider using the rule for injured children.
Patients older than 55 are susceptible to more fractures on average, likely due to osteoporosis and degenerative joint disease. This age-group has not specifically been studied in regard to the rule, and, generally, these patients should undergo radiography.
Isolated pain over the fibula and patella. Isolated pain over the fibular head and patella were both found to be independent predictors of fracture. The fibula bears only 15% of the weight of the leg; therefore, a fracture at the fibular head might allow for weight bearing. The fibular head should be palpated, and if the examination causes pain, the clinician should order radiographs. Note also that the common peroneal nerve is adjacent to the fibula, and it is important to evaluate the nerve for possible injury. If the fibular head is injured and the common peroneal nerve is damaged, foot drop may occur.
The patella itself is commonly fractured, accounting for about 1% of all skeletal fractures,13 and pain associated with palpation directly over the patella requires radiographs.
Inability to bear weight or flex the knee. The inability to bear weight for four steps and the inability to flex the knee to 90° were both significant variables in determining the clinical criteria for x-rays. A patient is said to be able to bear weight if two strides can be made on each leg. Even if the patient limps, as long as weight is transferred, that is sufficient.6 A number of significant injuries, with and without fracture, will result in an inability to flex the knee less than 90°
Implications for Sports Medicine
Many of the forces that injure the bones and disrupt the soft tissues around the knee occur commonly during athletics. The cruciate ligaments, the collateral ligaments, and both menisci are susceptible to injury. Except for calcified cartilages in older patients, none of these structures are radiopaque.
The most dreaded single sports-related knee injury is probably the anterior cruciate ligament (ACL) tear. Even with a complete rupture, the radiograph is often not helpful, because this structure is not seen. Rarely, an indirect indication of ligamentous injury is observed, such as an avulsion at the insertion of the ACL at the anteromedial tibial spine or a small avulsion of the lateral tibial plateau (called a Segond fracture, seen in less than 3% of ACL tears).14 The Segond fracture is a sign of a possible ACL tear, but it is not the primary pathology. With appropriate clinical suspicion and follow-up, it is unlikely that a significant knee injury will be missed if one remembers that only radiodense structures can be seen on radiographs. A normal finding with the Ottawa knee rule does not preclude a significant knee injury.
Athletes who have ACL tears, even if extensive, are sometimes able to ambulate after the injury, and an effusion may not develop for some time, resulting in a benign exam. Note that even with internal derangements and marked ligamentous instability of the knee, radiographs are likely to be of little value, because the ACL, medial collateral ligament, and menisci are radiolucent.
Regardless of the mechanism of injury, whether twisting, a direct blow, hyperflexion, or hyperextension, the Ottawa knee rule is appropriate to use to evaluate the injury. The presence of an effusion is not one of the criteria, although a significant effusion would likely make ambulation and flexion to 90° very difficult.
Many sports medicine physicians take care of athletic patients who have chronic knee problems, such as osteochondritis dissecans, patellar tracking syndrome, or jumper's knee. The Ottawa knee rule is meant to help clinicians decide which injuries are amenable to radiography. Physicians should use sound clinical judgment when evaluating acute-on-chronic conditions and ordering radiographs to study them.
It is of great value to be able to decide reliably when knee films are not indicated after a sports-related injury. The Ottawa knee rule allows clinicians to make decisions based on the likelihood of finding radiographic evidence of knee injuries, essentially fractures. While most knee films will still be noncontributory even when using these rules, we can reduce unnecessary films by at least 25%.
Athletes often compete or are examined in areas far removed from radiographic facilities.15 Using the Ottawa knee rule will allow physicians to reassure patients when films are not needed. This may lead athletes to remain at the competition or at practice, perhaps even returning to action sooner. While one must use good clinical judgment when determining when an athlete can return to play, the use of this rule can save time, money, and inconvenience. Further, it is likely that a frank discussion of the need for x-rays will improve communication between the athlete, coach, and physician.
If an athlete is badly injured but the Ottawa knee rule is negative, it would be safe to skip emergent x-rays, because the results are consistent with an absence of fracture. Should the athlete continue to have pain or instability, follow-up and further evaluation are, of course, appropriate. Additionally, a normal radiograph should never preclude subsequent care of an injured knee with splinting, crutches, or pain medication, as necessary.
The Ottawa knee rule has been formulated and studied in the emergency medicine venue. It has not undergone rigorous study in a strictly sports medicine application. Further studies are needed to evaluate the application of this rule to acutely injured athletes, but it seems amenable to evaluating and treating athletes.
While some positive information has been generated about using the rule with children, more study is necessary to state positively that this rule has the same high negative predictive value that it does with adults who are between 18 and 55 years old. However, it seems likely that the Ottawa knee rule will prove to be valid in the pediatric population.
Physicians see many patients who still compete after serious knee injuries and surgeries, even with orthopedic appliances and bony postoperative changes. While patients who had previous knee injuries or surgeries were not evaluated in the original literature, it seems appropriate for clinicians to individualize care for these patients also. This practice is another area in which further study is indicated.
Unfortunately, as in all fields of medicine, the specter of litigation for missed diagnoses hangs over sports medicine. Potential lawsuits are often used as a rationale for ordering tests and radiologic studies that are not indicated by clinical judgment and evidence-based medicine, often to the detriment of performing good histories and physical exams. The justification for routinely ordering knee films after every injury can be overcome by noting the 100% negative predictive value of the Ottawa knee rule and the scientific basis for using it. Just as important, though, is the fact that good follow-up and appropriate communication with the athlete should never allow a major injury to pass unappreciated. Should a patient continue to have problems, further workup (eg, magnetic resonance imaging, arthroscopy) would likely be in order.
The Ottawa knee rule is scientifically validated and easy to apply. It uses the results of exam criteria for injured patients to make rational diagnostic and treatment decisions. The sensitivity and negative predictive value have both been validated at 100%, meaning that a clinically relevant injury will not be missed if all the clinical criteria are negative. Unfortunately, the specificity is much lower, yielding a large number of normal films, but this is still an improvement over not using the clinical criteria at all.
The Ottawa knee rule predicts only which patients can forego radiographs for pathology that is radiopaque (ie, fractures). While injuries may exist to the menisci or ligaments and further evaluation may be required, a large portion (at least 25%) of those patients who are seen for acute knee injuries do not need radiographs. The Ottawa knee rule benefits both the patient and the physician.
Dr Nugent is assistant residency director for the Bethesda Family Practice Residency Program at Bethesda Hospital in Cincinnati. Address correspondence to Paul J. Nugent, DO, Assistant Residency Director, Bethesda Family Practice Residency, 4411 Montgomery Rd, Cincinnati, OH 45212; e-mail to [email protected].
Disclosure information: Dr Nugent discloses no significant relationship with any manufacturer of any commercial product mentioned in this article. No drug is mentioned in this article for an unlabeled use.