Ma. Oberlander et al., THE ACCURACY OF THE CLINICAL KNEE EXAMINATION DOCUMENTED BY ARTHROSCOPY - A PROSPECTIVE-STUDY, American journal of sports medicine, 21(6), 1993, pp. 773-778
The diagnostic accuracy of the clinical examination for intraarticular
injuries of the knee was documented by arthroscopy over a 6-month per
iod. Two-hundred ninety patients (296 knees) were evaluated by history
, physical examination, and standard radiographs. Supplemental diagnos
tic studies included 41 magnetic resonance images, 2 arthrograms, and
1 previous arthroscopy that had been recently performed. Overall, the
correct diagnosis was made in 165 knees (56%), an incomplete diagnosis
in 92 (31%), and an incorrect diagnosis in 39 (13%). There were only
2 knees (0.07%) with no discernable lesions. When a single lesion was
present in the knee, the diagnosis was made correctly in 72% of cases.
When more than 2 were discovered, the diagnosis was correct in only 3
0%. However, all individual lesions were diagnosed with an accuracy of
greater than 90%. The lesions most difficult to diagnose were chondra
l fractures, fibrotic fat pads, tears in the anterior cruciate ligamen
t, and loose bodies. Knees with acute lesions and those with a single
diagnosis proved to be significantly easier to diagnose (P < 0.01). Th
e variables that proved to be insignificant were age, sex, magnetic re
sonance imaging, surgeon, workers' compensation, or pending litigation
.