25 patients with a partial or complete collapse of the humeral head caused by post-traumatic avascular necrosis underwent clinical and radiologic evaluation at an average of 7.5 years (range 2.3 to 17.6 years) after having an underlying proximal humeral fracture. Posttraumatic humeral head necrosis was always associated with disability. The overall shoulder function as assessed with the Constant score was 46 points, corresponding to a functional shoulder value of 51% of an age- and sex-matched normal control group. The clinical outcome was significantly related to the anatomic alignment of the fragments of the humerus by the time of healing. In 13 patients (group 1) treatment resulted in an anatomic or nearly anatomic healing of the fracture, and in 12 other patients (group 2) avascular necrosis and collapse ensued in addition to malunion of 1 or more of the fracture fragments. Subjective overall outcome (P < .0001) and pain (P < .0001) were significantly better in group 1. Active anterior elevation averaged 125 degrees in group 1 and 80 degrees in group 2 (P = .0007), and abduction averaged 110 degrees in group 1 and 63 degrees in group 2 (P = .007). The relative shoulder score according to Constant was 65% of an age- and sex-matched normal population for group 1 and 41% for group 2 (P = .001). The results obtained in group 1 were comparable to those reported after hemiarthroplasty for complex humeral fractures. A proximal humeral fracture that is at risk for avascular necrosis has to be reduced anatomically if joint-preserving treatment is selected. If anatomic reduction cannot be obtained, other treatment options such as arthroplasty should be considered.
Copyright remains with the authors and publisher.
25 patients with a partial or complete collapse of the humeral head caused by post-traumatic avascular necrosis underwent clinical and radiologic evaluation at an average of 7.5 years (range 2.3 to 17.6 years) after having an underlying proximal humeral fracture. Posttraumatic humeral head necrosis was always associated with disability. The overall shoulder function as assessed with the Constant score was 46 points, corresponding to a functional shoulder value of 51% of an age- and sex-matched normal control group. The clinical outcome was significantly related to the anatomic alignment of the fragments of the humerus by the time of healing. In 13 patients (group 1) treatment resulted in an anatomic or nearly anatomic healing of the fracture, and in 12 other patients (group 2) avascular necrosis and collapse ensued in addition to malunion of 1 or more of the fracture fragments. Subjective overall outcome (P < .0001) and pain (P < .0001) were significantly better in group 1. Active anterior elevation averaged 125 degrees in group 1 and 80 degrees in group 2 (P = .0007), and abduction averaged 110 degrees in group 1 and 63 degrees in group 2 (P = .007). The relative shoulder score according to Constant was 65% of an age- and sex-matched normal population for group 1 and 41% for group 2 (P = .001). The results obtained in group 1 were comparable to those reported after hemiarthroplasty for complex humeral fractures. A proximal humeral fracture that is at risk for avascular necrosis has to be reduced anatomically if joint-preserving treatment is selected. If anatomic reduction cannot be obtained, other treatment options such as arthroplasty should be considered.
Copyright remains with the authors and publisher.