[Surgical treatment of humeral diaphyseal fractures]

The aim of the study was to compare two surgical methods of treating diaphyseal fractures of the humerus.

A prospective randomised study on the surgical treatment of humeral diaphyseal fractures was carried out between September 2009 and January 2013. The patients indicated for surgical treatment fell into two groups according to the technique used as follows: minimally invasive plate osteosynthesis (MIPO) with a locking compression plate (LCP; Synthes, Switzerland); depending on the fracture type and course of fracture line, either a straight narrow or a Philos or a metaphyseal LCP was used (group 1); and intramedullary nailing (EHN, Synthes, Switzerland) (group 2). A total of 49 patients were entered into the study. The men-to-women ratio was about 1:1 and the average age was 52 years (18 to 83). Of them, 45 patients with 46 humeral fractures were followed up and evaluated. The injury was caused by a fall from standing or while walking (n=21), traffic accident (n=16), sports activity (n=3) or work-related activity (n=5). A single injury was treated in 72% and multiple trauma in 28% of the patients. Using the AO classification, fractures were diagnosed as types A, B and C in 25, 10 and 11 patients, respectively. The patients were randomised into the groups using a computer programme allowing for the maintenance of group homogeneity. Each patient received information on the method to be used in a sealed envelope.

The average injury-surgery interval was 6 days (range, 0 to 26). The average operative time was 128 minutes (80 to 200). The average intra-operative exposure to X-ray was 8 minutes (3 to 20). The average hospital stay was 20 days (4 to 100) The average rehabilitation time till functional recovery was 17 weeks (3 to 37), the time of bone union by radiographic assessment was 20 weeks (12 to 40). The functional outcome was assessed using the Constant-Murley (CM) and Liverpool Elbow Score (LES) systems. The relative CM/LES score, as compared with the healthy extremity, was 95/100. Excellent and good results were achieved in 89% and complications were recorded in 26% of the patients. The most frequent complication was impingement syndrome or prolonged rehabilitation both in the patients with proximal humerus fractures managed with the Philos locking plate and in those treated by intramedullary nailing, although this was not necessarily due to a technical error, i.e., osteosynthetic material protrusion. Iatrogenic radial nerve injury was found only in one patient treated by intramedullary nailing and was associated with traction during reduction and nail insertion. Pseudoarthrosis was recorded in one patient of each group and there were no infectious complications.

Intramedullary nailing has not yet shown such good outcomes in humeral as in tibial fractures. The results of metaanalysis have indicated that humeral fractures treated by plate osteosynthesis have fewer complications and require repeat surgery less frequently. However, as shown by most recent studies, this difference is getting smaller. The MIPO technique was adopted at our department several years ago and the aim was to compare our results with those of other centres. The studies so far published have show that MIPO and conventional open plate osteosynthesis give comparable outcomes. CONCLUSIONS The statistical evaluation using the unpaired t-test did not show any significant differences in either the functional results or the number of complications between the two methods. Both achieved about 90% of excellent and good results and had 26% of complications. The only significant difference was found in the length of operative time (136 min in MIPO versus 117 min in EHN). With use of the Chi-Square test, a significant correlation between AO fracture type and treatment outcome in the MIPO group was found, i.e., no poor result was recorded for type A fractures, as assessed by the CM/LES score. No such correlation between the fracture type and the functional outcome was seen in the EHN group.