Non-operative treatment of four-part fractures of the proximal end of the humerus: results of a prospective and retrospective multicentric study

Purpose The aim of this study was to evaluate the short- and medium-term results of non-operative treatment of four-part fractures of the proximal end of the humerus. The initial hypothesis was that non-operative treatment of fractures with little or no displacement is equivalent or superior to surgical treatment, and that non-operative treatment is probably insufficient for displaced fractures. Methods This was a multicentric, prospective and retrospective study, based on 384 four-part proximal humerus fractures, 58 of which involved non-operative treatments - 37 in the prospective study (Pro-CT4) and 21 in the retrospective study (Retro-CT4). The average patient age was 64 +/- 14 years (39-90); 66 % were female and 34 % male. In 88 % of these cases, non-operative treatment was chosen for the fracture, as there was little or no displacement. In 10 % of cases, non-operative treatment was chosen "by default", due to the patient's medical conditions, as surgery was contraindicated, and in 2 % of cases due to the patient refusing surgery. All patients were reviewed clinically and radiologically, with SSV evaluation, absolute and weighted Constant scores and the Quick DASH score all assessed. The main evaluation criterion was the weighted Constant score which was considered a failure when below 70 %. ResultsIn the Pro-CT4 study, the average follow-up period was 11 +/- four months (5-18) with functional scores as follows: average SSV: 72 +/- 26 % (8-100); average Constant score: 65 +/- 21 points (21-95); average weighted Constant score: 86 +/- 26 % (32-130); average Quick DASH: 23 +/- 21 (0-64). 27 % of patients had a weighted Constant score below 70 %. In the Retro-CT4 study, the average follow-up period was 38 +/- 13 months (18-62) with functional scores as follows: average SSV: 73 +/- 17 % (30-100); average Constant score: 68 +/- 18 points (33-95); average weighted Constant score: 88 +/- 27 % (47-133); average Quick DASH: 18 +/- 16 (0-48); 24 % of patients had a weighted Constant score below 70 %. ConclusionThis study confirms our initial hypothesis. When non-operative treatment of four-part proximal humerus fractures is carried out by choice, the results are excellent. However, when this treatment is carried out "by default" - especially because surgery is contraindicated - the results are disappointing.