Therapeutic level III. See instructions for authors for a complete description of levels of evidence.

Level III-retrospective study.

When knee bone marrow edema is observed on magnetic resonance imaging (MRI), it usually follows a pattern that can be explained by certain etiologies. This article describes a series of unusual knee bone marrow edemas in soldiers presumed to represent self-inflicted trauma.Ten soldiers (9 men and 1 woman; age range, 19-24 years) underwent knee MRI. None reported recent trauma or stress, and all presented with nonspecific pain or failure to respond to therapy. All showed a similar unusual pattern of bone marrow edema in the medial femoral condyle. Three observers evaluated the location of the bone marrow edema within the medial femoral condyle and its distance from the articular surface, dimensions, overlying soft tissue abnormality, and internal derangements. The edema was always subcortical and located in the middle aspect (n=7) or mid-anterior aspect (n=3) of the medial femoral condyle but was never centered subarticularly. Edema size ranged between 8 × 10 × 8 and 32 × 46 × 40 mm. Overlying soft tissue abnormalities were common (n=4) and included organizing (n=1) and residual hematoma (n=3). Concomitant MRI abnormalities were seen in 3 patients, usually minor. Eight patients reported longstanding pain with no antecedent trauma, and 2 reported remote trauma. One patient had a negative 4-month follow-up MRI, and another had a negative arthroscopy. Poor correlation existed between MRI findings and the absence of stress and trauma. Soldier chat rooms were found that describe how to induce fractures at this location.

Background: Unstable pertrochanter femur fractures are common in orthopedic practice. They pose a surgical challenge in both reduction and fixation. The fixation devices used are based on hip intramedullary nailing with femur head lag screw or blade. The aim of this paper is to compare different types of unstable pertrochanter fractures. Materials & Methods: We retrospectively reviewed 386 unstable pertrochanter femur fractures surgically treated in our hospital from 2000 to 2009. These included 62 (16.1%) unstable pertrochanter fractures with fractured lesser trochanter (31.A2-2, 31.A2-3); 63 (16.3%) reverse oblique fractures (31.A3-1), 51 (13.2%) transverse fractures (31.A3-2), 145 (37.6%) comminuted fractures (31.A3-3) and 65 (16.8%) subtrochanter fractures. We compared survival rates between fracture types. The clinical characteristics, surgery immediate outcome (e.g., tip apex distance, reduction quality), and long term results, i.e., complications were also compared between fracture types. Results: Survival analysis showed that the fracture types can be grouped into low and high risk fracture types. The former group included, reverse oblique and comminuted fractures. Lesser trochanter, transverse and subtrochanter fractures were included in the high risk group. The survival estimates for five years were 64.6% and 49.3% for the low and high risk fracture types, respectively (p value = 0.008). Multivariate survival analysis showed that the hazard ratio for the high risk fracture group was 1.9 (95% CI = 1.37 -2.67). No differences were found between unstable pertrochanter femur fractures with regards to clinical and epidemiology characteristics. Optimal tip apex distance (TAD) of less than 25 mm was found in 66.7%, 57.1% and 66.7% of lesser trochanter, reverse oblique and sutrochanter fractures, respectively. TAD of less than 25 mm was found in 81.2% of both transverse and comminuted fractures (p value = 0.032). No statistically significant difference was found between fracture types, in regards to complication or revision rates. Conclusions: Survival rates were higher in patients suffering from reverse oblique or comminuted pertrochanteric fractures. No differences were found between fracture types, in regards to clinical and other outcome parameters.

No abstract