Session II - Pediatrics/Spine

Pediatric Supracondylar Humerus Fractures: The Anterior Approach

M. Loudstaal, MD; Victor A. de Ridder, MD; S. DeLange, MD; C. Ulrich, MD, Westeinde Hospital, Va Den Haap, Holland

Purpose: To evaluate the efficacy of the anterior approach for displaced supracondylar pediatric humeral fractures.

Method: A retrospective analysis of all children treated operatively for a displaced supracondylar humerus fracture through an anterior approach, with the result compared to a historical group treated through a lateral or a combined lateral and medial approach.

The anterior approach group consisted of 26 patients, age range 16 months to 13 years, operated on between 1995 - 1998, follow-up longer than 6 months, average 18 months. All had a completely displaced extension-type supracondylar humerus fracture (type III). All were treated within 6 hours of admission. The fractures were operated on through a ventral approach: incision in the fossa cubitalis of about 1 inch of only the skin and subcutis as you then tumble into the space created by the anterior dislocation of the humeral shaft, reduction with the thumb and control of interpostion by muscle or vascular and neural structures. Often the M. coracobrachialis is partly severed by the dislocation of the humeral shaft. Both medial and lateral condyle can be palpated and the rotation corrected. A retractor should not be used, only the blunt thumb is sufficient. If vascular or neural structures are injured, the incision can then be enlarged into a lazy S. fixation with crossed percutaneous K-wires under image intensifier, followed by splinting with plaster of Paris for two week.

The historical group, operated on through a lateral or a combined lateral and medial approach, consisted of 32 patients, age range 18 months to 14 years, between 1992 - 1996, follow-up average of 4.5 years. All had a completely displaced extension-type supracondylar humerus fracture (type III). All were operated on within 6 hours of admission. The fractures were operated through a "classical" lateral or medial approach. Fixations were done with crossed percutaneous K-wires under image intensifier, followed by splinting with plaster of Paris for two weeks.

All patients were contacted; 25 of the 26 and 29 of the 32 patients were re-examined at the outpatient clinic. The other patients were contacted by telephone. Medical records and all X-rays were evaluated for quality of reduction and clinical results. Examination of both upper extremities was performed for neurovascular status, function, and strength. The results were graded according to Flynn's criteria.

Results: In both groups no early complications such as compartment syndrome or Volkmann's ischaemic contracture were seen. In the anterior approach group two associated brachial artery injuries, and in the control group one associated brachial artery injury were diagnosed. In the latter patient, an additional ventral approach was used. Operation time (skin to skin) in the lateral approach group averaged 65 minutes, in the lateral and medial group 80 minutes. In the anterior approach group the average operating time was 25 minutes; reduction and placement of the K-wires was notably easier in the anterior approach group.

Early postoperative fracture displacement occurred in one of the 26 anterior-approached patients and in four of the 32 lateral/medial approach group. Displacement occurred in all instances due to suboptimal placement of the K-wires, and the K-wires were not drilled through the humeral cortex. Rotational or gunstock deformity was diagnosed in physical examination and confirmed by X-ray with malreduction of greater than 10 degrees in both groups. One outpatient in each group had a comprised function due to the deformity: 50 and 100 elbow extension limitation and 1200 and 1250 degrees elbow flexion.

None of the above mentioned differences were statistically significant due to the small sized patient groups. No specific complication related to the anterior approach was noted.

Discussion and conclusions: The results of this study show that the anterior approach is safe, simple and easy to perform. The anterior approach produced no additional complications and is even advised in neurovascular compromised patients. The lateral or combined lateral and medial approach is not used any more. The anterior approach seems to produce better results in these limited groups: good and excellent results by Flynn's criteria in 84% compared to 75% in the lateral and combined lateral and medial approach group.