Literature Review: Observations on Proximal Humerus Fractures

Gregory Gilot, MD
Cleveland Clinic Florida


TO REVIEW IS TO OPINE It appears that the large majority of patients who present with complications following locking screw ORIF are those with a high likelihood of vascular disruption of the humeral head (four-part fractures, short metaphysical segment) at the time of injury. While the risks of vascular necrosis, varus collapse with secondary screw cut out are relatively high, under the age of 60 years in patients with adequate bone stock a joint preserving procedure should be attempted. Failure of locking screw ORIF can be recognized as early as nine months and the secondary treatment entertained.

 SUMMARY The purpose of the study was to report the complications, their treatment and outcome of 121 patients referred after primary locking plate ORIF for proximal humerus fractures. The authors recognized the increasing numbers of complications following such surgeries being referred to the tertiary referral center. Among the known complications of AVN, varus collapse and screw cutout being reported, the previously unknown complication of iatrogenic glenoid destruction due to perforated head screws was reported.

 MATERIALS AND METHODS One hundred twenty one patients with complications we recollected. Sixty-seven patients were women and 54 were men. The mean age at initial injury was 59 years. All 121patients were treated with the PHILOS plate (Synthes, Paoli, PA, USA). Of the 114 classified fractures, 17 were 2-part, 373-part, and 60 4-parts. All patients presented with restricted function and pain and were seen after a mean of 15 months after index ORIF.

RESULTS A mean of three complications occurred per patient.


Number ofpatients (%)

Malreduction 67 (55%)
Primary screw cutout 14 (12%)
Malunion 76 (63%)
Nonunion 16 (13%)
AVN 82 (68%)
Infection 5 (4%)
Secondary screw cutout 69 (57%)
Glenoid destruction 40 (33%)

TREATMENT AND OUTCOMES One hundred seven (88 percent) patients underwent revision surgery. A mean of 1.5 surgeries were performed per shoulder. Partial and total hardware removals were the most common revision operations in this study. Overall, over 50percent of patients finally needed secondary arthroplasty.



Re-osteosynthesis              8 3
Arthroscopy                      10 3
Partial hardware removal 16 7
Total hardware removal   41


Secondary arthroplasty    61

CONCLUSIONS In this collective of patients with complications following locking screw ORIF, 50 percent had four-part fractures, 20percent were head splitting or fracture-dislocations, and 80percent had no metaphysical segment attached to the head. These factors are known to be associated with poorer outcomes underscoring that the primary surgery should be definitive if possible. Lastly, Neither secondary hemiarthroplasty, total shoulder arthroplasty or reverse total shoulder arthroplasty yielded results comparable to primary arthroplasty. •


TO REVIEW IS TO OPINE While the radiographic outcomes in terms of tuberosity healing appear comparable, it is the differences in clinical outcomes, forward flexion and revision rate that would lead the reader to believe that rTSA is superior to HA for the treatment of proximal humeral fractures in the elderly. Forward flexion results appear to be independent of tuberosity healing with a predictable low number of patients who do not achieve a minimum of 90 degrees forward flexion following an rTSA. Implant survival is superior for rTSA when compared to HA.

 SUMMARY The surgical treatment of complex proximal humerus fractures in the elderly with poor quality bone remains controversial. These factors negatively impact the results of internal fixation and hemiarthroplasty (HA) has been considered a good surgical option in these patients. Reverse shoulder arthroplasty (rTSA) has the theoretical advantage of independence of tuberosity healing and rotator cuff integrity. Useful clinical conclusions and clear guidelines for treatment currently do not exist. The aim of this study was to compare the outcomes of rTSA and HA for acute proximal humerus fractures in elderly patients.

MATERIALS AND METHODS This blinded, randomized, controlled prospective study included patients 70 years or older whose complex fractures were not amenable to reconstruction. To avoid selection bias patients found to have irreparable cuff tears were not excluded. Sixty-two patients were randomized to rTSA (31) and HA (31). The mean follow-up was 28.5 months. The modular shoulder replacement system (SMR; Lima, Udine, Italy) was used in both groups. Surgery was performed similarly in both groups. Tuberosities were handled identically. Postoperative regimens were similar in both groups. In the HA group, a rotator cuff tear was found in 11 cases; in three cases the tear was irreparable. In the rTSA group, a rotator cuff tear was found in 14 cases; in five cases the tear was irreparable. Clinical evaluation included the Constant-Murley score, theUCLA score, the DASH score, range of motion and strength. Radiographic evaluation included the status and position of the tuberosities, implant loosening, proximal migration in the HA group and inferior scapular notching in the rTSA group.

There were no statistically preoperative differences between the groups.

Outcome rTSA Group HA Group
Constant score (p=0.001) 56.1 40.0
UCLA score (p=0.001) 29.1  


DASH score (p=0.001 17.5 24.4
Forward flexion (p=0.001) 120.3 79.8
Forward flexion <90 degrees 1 (3.2%) 10 (33.3%)
External rotation (p=0.023) 4.7 3.3
Internal rotation (p=0.914) 2.7 2.6
Tuberosity healing 20 (64.5%) 17 (56.7%)
Tuberosity resorption 5 (16.2%) 9 (30.0%)
Revisions 1 6
40-month implant survival 96.8% 80.0%

 CONCLUSIONS In this study, rTSA was superior to HA with respect to pain, functional outcome, and revision rate. While forward flexion and abduction were significantly better in the rTSA group there was no significant difference in internal rotation. There was one (3.2 percent) case of scapular notching observed. The six revisions within the HA group were due to proximal migration resulting in severe pain and limited function. There was one deep wound infection within the rTSA group requiring a two-stage revision. All revised patients suffered poor Constant scores and unsuccessful functional outcomes. Revision from HA to rTSA did not appear to improve outcomes. •



 TO REVIEW IS TO OPINE The management of complex proximal humeral fractures in the elderly continues to be a difficult problem to solve with a predictable solution. This study is one of the first to report short term results. What is well demonstrated is the need for reliable tuberosity reconstruction to optimize outcomes. These conclusions argue strongly for disease specific implants that allow for anatomical tuberosity.

SUMMARY The use of reverse shoulder arthroplasty in complex proximal humerus fractures cannot be routinely recommended due to the paucity of clinical studies. While mid-term results appear to be encouraging, to date its use in trauma has been described in only small series. The aim of this study was to describe the author’s experience with this technique in the short term in elderly patients.

 MATERIALS AND METHODS Forty-three patients with a mean age of 78 years with an acute proximal humerus fracture were treated with a Delta reversed shoulder prosthesis (Depuy, Saint Priest, France).Operative treatment and post-operative care were similar for all patients. While the tuberosities were repaired, the supra-spinatus, when present, was divided and removed. Mean follow-up was 22 months. Clinical outcomes included Constant and Murley score, the ASES score, the DASH score, and mobility. Radiographic evaluation included recording inferior scapular notching, assessing the position of the centre of the shoulder, the glenoid inclination angle, heterotopic ossification and healing of the tuberosities.

RESULTS Complications occurred in 12 patients. While greater tuberosity healing did improve external rotation this was not significant. The shoulder centre medialization had a mean of 21mm and a mean of 9mm below the centre of rotation of the contralateral side. Patients with a lower center had better results but this difference was not significant. Of the scapular notching that was observed, only one was Sirveaux grade 3.

Clinical & Radiographic Outcomes rTSA Group
Constant score 44
ASES score 9
DASH score 44
Anterior elevation 97
External rotation 8
Tuberosity displacement 19 (53%)
Tuberosity malunion 5 (13.8%)
Tuberosity nonunion 14 (3.8%)
Scapular notching 10 (25%)
Glenoid component inclination 15 degrees
Shoulder center medicalization 21mm
Heterotopic ossification 36 (90%)

 CONCLUSIONS In this study, satisfactory mobility was obtained with the use of a reversed shoulder arthroplasty in complex proximal humerus fractures despite a 53 percent rate off displacement of the tuberosities. When anatomical reconstruction of the tuberosities was achieved (41.5 percent) the effect on the Constant score appeared to be moderate. •