The changing demographics of knee dislocation: a retrospective database review

Abstract

Background: Knee dislocations are uncommon but devastating orthopaedic injuries. Little is known about their frequency and the types of patients who are affected.

Questions/purposes: Using a large national insurance database, we determined (1) the incidence of knee dislocation in patients with orthopaedic injuries and examined the incidence as a function of (2) year of diagnosis, (3) dislocation type (open versus closed, direction), and (4) patient demographic factors (sex, age).

Methods: We searched the PearlDiver database, a national database of private insurance records consisting of 11 million patients with orthopaedic diagnoses, using diagnosis (ICD-9-CM) codes for knee dislocation between the years 2004 and 2009. The PearlDiver database does not include Medicare, Medicaid, or uninsured patients. Patients were stratified by age, sex, and year of diagnosis. Incidence was defined as the number of dislocation events per 100 patient-years.

Results: We identified 8050 dislocations, representing an incidence of 0.072 events per 100 patient-years between 2004 and 2009. Annual dislocation incidence did not increase during the 6-year study period. Of the 8050 dislocations, 1333 (17%) were open and 6717 (83%) were closed, representing an incidence of 0.060 per 100 for closed dislocations and 0.012 per 100 for open dislocations. The most common direction of dislocation was unspecified or other (65%), followed by anterior (13%), lateral (11%), posterior (6%), and medial (5%). Of the patients sustaining dislocations, 4172 (52%) were female and 3878 (48%) were male. Males displayed an increased risk of knee dislocation compared to females (odds ratio = 1.09). The mean patient age was 35 years, and patient age was inversely correlated to the incidence of knee dislocation (10-year odds ratio = 0.77).

Conclusions: Our data suggest that knee dislocation might represent a significantly larger burden among orthopaedic injuries than previously thought. The finding that males and females have a nearly equal risk of knee dislocation enhances the diagnosing physician's clinical suspicion of this injury. Future large prospective studies analyzing the various causes of knee dislocation could provide insight into the changing demographics of this injury.

Level of evidence: Level IV, prognostic study. See Instructions for Authors for a complete description of levels of evidence.

Fig. 1

A pie chart shows the incidence of knee dislocations as a function of direction of dislocation. Dislocations of unspecified direction and other direction were the most common.

Fig. 2

A graph shows the incidence of knee dislocation as a function of age group. The 10- to 19-year age group had the highest incidence.