Confirmation of the safety and accuracy of physical examination in the evaluation of knee dislocation for injury of the popliteal artery: a prospective study.

BACKGROUND: Knee dislocation, which poses a significant risk for injury of the popliteal artery, prompts many surgeons to evaluate these patients with arteriography routinely. Our hypothesis was that physical examination alone (without arteriography) accurately confirms or excludes surgically significant vascular injuries associated with knee dislocation.
METHODS: All patients diagnosed with a knee dislocation by an attending orthopedic surgeon between January 1990 and January 2000 were prospectively managed by protocol at our Level I trauma center according to their physical examination. Those with hard signs (active hemorrhage, expanding hematoma, absent pulse, distal ischemia, bruit/thrill) underwent arteriography followed immediately by surgical repair if indicated. Patients with no hard signs (negative physical examination) were admitted for 23 hours, underwent serial physical examination, and then followed as outpatients.
RESULTS: There were 35 knee dislocations in 35 patients during this 10-year period. The average age was 31 years; 18 dislocations were on the right knee and 17 were on the left. Two patients died from closed head injuries and multisystem trauma. Eight patients were found to have hard signs (positive physical examination) either at presentation (six patients) or during their hospitalization after reduction of their dislocation (two patients). All eight patients demonstrated a loss of pulses only. Six of these patients showed occlusion of the popliteal artery on arteriography and underwent surgical repair without complication (five vein grafts, one primary repair), one demonstrated spasm of the popliteal artery, and one showed a normal artery that required no treatment. None of the 27 patients with negative physical examination during their hospitalization ever developed limb ischemia, needed an operation for vascular injury, or experienced limb loss. Sixteen patients were available for follow-up (46%). Twelve patients with negative physical examination (44%) were contacted (mean, 13 months; range, 2-35 months), and four of the eight patients with positive physical examination (50%) and surgical repair were contacted (mean, 19 months; range, 6-49 months). None of the patients in either group developed any vascular-related symptoms or suffered from a vascular repair complication over the follow-up interval.
CONCLUSION: This limited series suggests that the presence or absence of an injury of the popliteal artery after knee dislocation can be safely and reliably predicted, with a 94.3% positive predictive value and 100% negative predictive value. Arteriography appears to be unnecessary when physical examination is negative but may avert negative vascular exploration when physical examination is positive. This approach substantially reduces cost and resource use without adverse impact on the patient.