Radiographic parameter(s) influencing functional outcomes following angular stable plate fixation of proximal humeral fractures

Abstract

Abstract Purpose Radiographic parameters which correlate with poor clinical outcome after proximal humeral fractures could be helpful indicators to answer the question which patients should be followed up closer. Moreover, during surgery, radiographic parameters correlating with unfavourable outcome should be avoided. The primary aim of the study was to compare radiographic measurements between the injured and the contralateral, uninjured shoulder. The secondary aim was to correlate these radiographic parameters with post-operative shoulder function. Methods Fifty-eight patients (age: 55.6 ± 14.4 years, age at surgery) following angular stable plate fixation of a proximal humeral fracture (2-part fractures according to Neer: 24, 3-part: 25, 4-part: 9) were included in this retrospective cohort study. All patients were followed up at least six years (7.9 ± 1.4 after surgical intervention). During follow-up examination, the Constant score (CS) was assessed, and radiographs of both shoulders were taken. Radiographs were analyzed regarding lateral humeral offset, distance between tuberculum and head apex, head diameter, head height, perpendicular height, perpendicular center, vertical height, and angles between head and humeral shaft (CCD and HSA). These parameters were compared between the injured and uninjured shoulder. The cohort was divided in two groups: patients with a CS category of excellent/good and satisfying/worse. Both groups were tested regarding differences of demographic and radiographic parameters. Results The distance between tuberculum and head apex (2.6 ± 3.4 mm vs. 4.3 ± 2.1 mm; p = 0.0017), the CCD (123.1 ± 12.9° vs. 130.1 ± 7.3°; p = 0.0005), and the HSA (33.1 ± 12.8° vs. 40.1 ± 7.3°; p = 0.0066) were significantly smaller on the treated shoulder compared to the uninjured side. Patients reached a Constant score of 80.2 ± 17.4 (95% CI 75.6–84.8) points. Regarding outcome categories of the Constant score, 46 patients had a good to excellent outcome, and 12 patients had a satisfying or bad outcome. The comparison of these groups revealed that patients with inferior outcome in the long-term follow-up were older, female, had a more complex fracture type (AO classification), smaller lateral humeral offset, smaller head diameter and height, lower perpendicular height, and lower CCD and HSA angles. Conclusion If the abovementioned parameters cannot be restored sufficiently during surgery, (reversed) shoulder arthroplasty might be a better solution to reach good post-operative outcome. Moreover, patients presenting these radiographic characteristics in the follow-up, older patients, and patients with a more complex fracture type should be followed up closer to possibly prevent poor shoulder function. Trial registration: 83 250/2011BO2