[Supracondylar humerus fracture in childhood--an efficacy study. Results of a multicenter study by the Pediatric Traumatology Section of the German Society of Trauma Surgery--I: Epidemiology, effectiveness evaluation and classification].

In this retrospective study of the pediatric trauma group of German trauma society, issued to investigate the state of the art treatment of the supracondylar fracture of the humerus, 13 clinics took part. In this first part of our study we tested the epidemiology and effectivity of therapeutic interventions based on the classification of v. Laer. 886 fractures were included with an average patients age of 5.8 years (+/- 2.9). Causes of trauma was in 45% playing, followed by school/kindergarden and sports injuries. Fractures were initially classified according to v. Laer and showed following displacement: 35.4% Type I, 21.9% Type II, 18.1% Type III and 24.6% Type IV. 10 of the 886 cases (1.1%) were open fractures. Damages to nerves were described in 45 patients (5.1%) and only 7 (0.7%) had primary vessel lesions. 476 patients were treated by reduction of fragments, 72% using a closed technical approach and 28% using an open approach. 6% underwent a second resposition-maneuver, which was mainly observed after crossed Kirschner-wire in type-III-and-IV-fractures. Therapy was changed in 5.1% mostly of the cases were initially closed reduced and then fixed with a collar and cuff sling. 540 patients were seen at follow-up (61%). 81.1% of these patients showed symmetrical axis compared to the uninjured arm. A varus-deformity was noted in 11.7%, a valgus-deformity in 7.2%. Analysis of effectivity showed that the primarily used classification was not sufficient for prediction of the outcome after reposition and retention. Therefore the classification was modified based on 4 groups: Type I undisplaced, Type II displacement in one plane, Type III displacement in two planes and Type IV displacement in three spatial planes. Using this classification we could found that in group II 25% of reduction an 7% of retentions were ineffective. For group III and IV we found that > 20% of the retention proofed to be ineffective.