1. The authors stress that these fractures differ from "classic" osteoporotic fractures. 2. The authors indicate surgical treatment of these fractures (even incomplete ones) as a method of choice. 3. The authors suggest that the need for further BP therapy should be assessed after 3-5 years of treatment.

An 86-year-old female was operated on due to a comminuted fracture of the right proximal humerus whose configu­ration necessitated over-screwing of the humeral head to improve stability, with the screws directed divergently: anteriorly and poste­riorly, to minimize their conflict with the acetabulum. 3.5 years later she sustained a corresponding fracture of the left shoulder which was stabilized without over-screwing. On both occasions, the fractures were stabilized with titanium interlocking plates (ChM, Poland) and the limbs were immobilized in a shoulder brace for three weeks followed by in­tensive rehabilitation. The patient attended a follow-up visit at 12 months (i.e. 52 months post the right shoulder fracture) which included an assess­ment of radiographs, pain and limb function as well as tests of the range of movement of the shoulder and girdle. Assessments were repeated at 12 and 18 months afterwards. Patient denied limb pain and dysfunction. Abduction was reduced by 20°, which was compensated for by the scapulothoracic joint. Shoulder (girdle) abduction reached 50° (150°) for the right and 70° (170°) for the left extremity. Flexion was reduced by 20°, but extension and rotations were comparable. At 12 months post fracture, no improvement of limb mobility was noted despite continued intensive physiotherapy. X-rays showed satisfactory bone union. The patient scored 87 for the right and 89 for the left shoulder according to the Con­stant score and 6.8 points for each limb according to the QuickDash score. Overscrewing of the humeral head is not the most beneficial method for improving stability of comminuted proximal humeral fractures; however, it may be used when alternative and more suitable methods are unavailable. Moreover, directing screws divergently anteriorly and posteriorly to minimize their conflict with the scapular acetabulum does not interfere with joint function outcomes.

Background. Fractures of the proximal humerus make up 4 to 10% of all fractures. Their incidence increaseswith age, usually affecting individuals over 40 years old, reflecting the mineral status of the bone, Material and methods. Out of a group of 131 patients operated on due to comminuted proximal humeral fractures, 25 cases presenting inappropriate postoperative results were selected for further analysis. Results. Failures were found in 16 cases, but affected the final outcome in seven cases only (lack of anatomical repositioning of bone fragments, including one that was stabilized at the fifth postfracture week, in whom a severe contracture of the supraspinatus muscle dislocated the major tubercle, or conflict of the implant with the acromion), and were negligible in the other nine. In two, properly stabilized major tubercles were dislocated due to their bearing the body weight immediately after the procedure. Another two developed necrosis of the humeral head, and in one the fracture was revealed to be pathological, requiring further oncological treatment. In four cases, ostensible false stabilizations were recorded as a consequence of inappropriate patient positioning for x-rays. Conclusions. 1. Our observations indicate that anatomical repositioning of bone fragments, especially those containing articular surface and muscle attachments, and their firm stabilization as well as proper implant positioning are crucial for the final result. 2. In some cases objective and unpredictable factors influencing the type and time of intervention affect the final results. 3. In those cases the decision to operate seems to be controversial. 4. The possibility of a pathological fracture should also be remembered, as such fractures require an appropriate oncological treatment.