Background: Humeral osteotomies for cubitus varus have a notoriously high complication rate. Pitfalls of this difficult procedure are highlighted.
Methods: A 50-year experience of 68 consecutive surgeries was reviewed. Factors such as surgical approach and fixation technique were compared for complication incidence and type.
Results: Seventeen patients (25%) had 23 (34%) clinically remarkable complications. Nine postoperative nerve palsies occurred in 8 patients. Loss of reduction requiring revision or manipulation was seen in 3 patients. The following complications were noted in 2 patients each: nonunion, loss of flexion, lateral prominence, and unsatisfactory scar. Growth arrest, osteomyelitis, and under-correction requiring revision each occurred once. A lateral, triceps-sparing approach was associated with an overall prevalence of complications of 24% (5 of 21) equivalent to the posterior, triceps splitting approach of 24% (10 of 42). An olecranon osteotomy was used in 2 patients both with complications. No nerve injuries occurred in patients who underwent a lateral approach, whereas nerve palsies occurred in 14% (6 of 42) of the patients where a posterior approach was used. An olecranon osteotomy was used in 2 patients with nerve injury occurring in both. A medial approach in 2 patients and a combined medial-lateral approach in 1 patient were used with no complications. Plate and screw fixation was implemented in 29 cases with complications occurring in 6 of them; pin fixation, in 30 cases, 7 of which had complications. There was a higher incidence of under-correction requiring additional surgery with plate fixation (1 of 29) compared with pin fixation which had no under correction but had loss of fixation in 2 of 30. The average correction obtained was similar in the group with complications (32 degrees) versus those without (27 degrees).
Conclusions: Supracondylar humeral osteotomy is a technically demanding procedure fraught with complications. Plate fixation and pin fixation techniques resulted in similar complication rates, but the surgical approach used appeared to make a difference. The posterior, triceps splitting, approach resulted in a high incidence of nerve palsies versus none with the lateral, triceps-sparing approach.
Level of evidence: This is a retrospective case series, Level IV.