Objectives and Material: The authors have evaluated 78 patients with fractures of the proximal humerus that have been interned in the Orthopedics’ Service at Hospital de Santa Maria since the 1st January of 1995 until the 31st December of 1999 with the objective of evaluating different factors influencing the final results and concluding to the best therapeutic conduct in the different cases. Method: The following parameters have been evaluated: sex, age, accident type (low/high energy), associated injuries, fracture’s classification, type of treatment and results achieved. The Constant score have been used for the functional evaluation. Due to reduced number of patients in each class a non-parametric test has been used – Qui-square test (X2). SPSS program has been used to run the calculations.

Results: It has been verified that sixty four percent of the cases happened due to low energy accidents and the three-part fractures prevailed (46%), followed by four-part fractures (34%). The fracture of the proximal humerus occurred isolated and prevailed in women above 60 years old. Patients’ age was not a factor of bad prognostic. The number of bad results depended on the fracture’s type. Open reduction and internal fixation has been the most frequently used technique in all fracture types and the one with the best results. Open reduction and Kirschner pinning should be used only in very specific cases and with limited functional objectives. Humeral head reconstruction has not been a valid option to four-part fractures treatment, allowing concluding that, in many cases of very fragmented fractures, the option for the humeral head substitution by prosthesis should be better than trying its reconstruction.

Objectives: The Authors present a method of patients’ selection based on clinical observation, imaging and arthroscopy for shoulder instability treatment. Open surgery was performed if criteria for arthroscopic treatment were not fulfilled. Material: 58 patients with anterior traumatic shoulder instability, treated between January of 1998 until the December of 2001.

Method: The following parameters have been evaluated: sex, age, accident type (low/high energy), associated injuries, type of treatment and results achieved. The Constant score have been used for the functional evaluation. MRI and arthroscopic criteria’s were also used. The type of surgical treatment was decided on those terms. A non-parametric test has been used – Qui-square test (X2). SPSS program has been used to run the calculations.

Results: The mean Constant Score was 90%, 2 patients had a new episode of shoulder luxation. There were no signs of instability in the remaining patients; mean loss of external rotation was 5° in adduction and 10° at 90° of abduction. There was a statistic significant difference between open and closed surgery in terms of loss of range of motion but not on reluxation.

Conclusion: The patients’ selection method presented is a valid tool for shoulder traumatic instability assessment and treatment.