What is the most appropriate evaluation for a patient with an acute knee injury?
Traditionally, physical examination maneuvers, such as the Lachman test, the pivot shift, the anterior drawer, and the McMurray test, have been recommended for patients with acute or subacute knee injury. A recent systematic review1 identified 35 studies that used results of arthroscopic surgery as the reference standard; however, in most of these studies, the arthroscopists were not blinded to the physical examination findings, and most studies had other design flaws. Nevertheless, they still provide important guidance regarding the relative accuracy of the most widely used maneuvers.1 Data for the physical examination are summarized in Table 1. A positive Lachman test or pivot test is strong evidence of an existing anterior cruciate ligament (ACL) tear, and a negative Lachman test is fairly good evidence against that injury. Although widely used, the anterior drawer is the least helpful maneuver for diagnosing an ACL tear. Joint line tenderness is not very helpful at ruling in or ruling out meniscal injury, while a positive McMurray test is most helpful for confirming the diagnosis.
|Probability of specific injury if examination maneuver is:†|
|Maneuver||Positive LR*||Negative LR*||Positive (%)||Negative (%)|
|Anterior drawer test||3.7||0.6||29||6|
|Joint line tenderness||1.1||0.8||11||8|
Radiography also is widely used, but is unhelpful in many cases. Several clinical decision rules have been developed to assist the physician by identifying patients who are at very low risk of bony injury and so do not require a radiograph. The Pittsburgh Knee Rule2 recommends obtaining a radiograph for patients with a recent fall or blunt-trauma mechanism, those who are younger than 12 years or older than 50 years, and patients who are unable to take four weight-bearing steps in the emergency department or primary care office. In a prospective validation3 conducted by the developers of the Pittsburgh Knee Rule, the rule was 99 percent sensitive and 60 percent specific for diagnosing acute knee injury in a convenience sample of 934 patients between six and 96 years of age. In this group, 25 percent of patients with a positive Pittsburgh Knee Rule evaluation had a fracture, and 99.7 percent with a negative evaluation had no fracture.
The Ottawa Knee Rule considers five items: (1) age 55 years or older; (2) tenderness at the head of the fibula; (3) isolated tenderness of the patella (no bone tenderness of knee other than patella); (4) inability to flex knee to 90 degrees; and (5) inability to bear weight for four steps both immediately and in the examination room regardless of limping. The presence of any of these items is an indication for radiography. The Ottawa Knee Rule has been more extensively validated in a greater variety of adult populations4 than other rules, and, therefore, was recommended in a 2003 systematic review1 as the preferred clinical decision rule for acute knee injury. A study3 that included adults and children, and a study5 of only children showed lower sensitivity for the Ottawa Knee Rule; therefore these rules should not be used in pediatric populations. The Pittsburgh Knee Rule found adequate sensitivity in a mixed population of adults and children by ordering radiography for children younger than 12 years.3
The accompanying patient encounter form for patients presenting with acute knee injury includes the four most accurate clinical examination maneuvers and guidelines for ordering radiography based on the Ottawa Knee Rule. It also reminds physicians always to consider performing radiography in children younger than 12 years given the results of the Pittsburgh Knee Rules. The back side of the form illustrates the physical examination maneuvers.
Applying the Evidence
A 38-year-old man experienced a sudden severe pain in his left knee as he was carrying a couch up some stairs while pivoting on that leg. He initially is able to ambulate, but later develops locking relieved by shaking his leg gently. On examination, he has a small effusion, no erythema, nearly normal range of motion, and slight joint line tenderness medially. There is no tenderness of the patella or head of the fibula.
Using the Ottawa Knee rule, a radiograph is not indicated. While he has negative results for anterior drawer, Lachman, and pivot tests for an ACL tear, he has a positive result for the McMurray test. Although his magnetic resonance imaging is negative for ligamentous or meniscal tear, a tear of the medial meniscus is discovered during arthroscopic exploration.
editor’s note: This case was the author’s experience with his own knee injury.