Background
Vertebral fractures, hip fractures, distal forearm fractures and humeral fractures are the most common osteoporotic fractures. The average lifetime risk in a 50 year old Caucasian of experiencing a humeral fracture has been estimated at 12.9% for women and at 4.1% for men[
1].
There is increasing incidence of such fractures due to the increasing mean age of the population and the higher levels of activity among the elderly. Palvanen et al. showed that in the over 60 age group in Finland the incidence of proximal humerus fractures tripled between 1970 and 1998 and they anticipate that this trend will continue until 2030[
2].
Possibilities for treating for such fractures range from conservative treatment to operative options such as the plate or nail fixation, humerus-block or other k-wire based systems, hemiarthroplasty or the reversed shoulder arthroplasty (RSA)[
3‐
5].
Grammont invented the RSA for rotator cuff tear arthropathies in 1985[
6]. The indication for PHF in the elderly as an alternative to hemiarthroplasty and plate fixation has broadened in recent years as the number of complications associated with osteosynthesis with unfavourable functional outcome is high in this group of patients due to osteoporotic bone structure as well as the high risk of avascular necrosis of the humeral head and the lack of sufficient rotator cuff[
7].
There is a growing tendency for complex PHF in osteoporotic patients to be treated with RSA. It has been proved that patients treated with RSA are easier to mobilize, require less time in hospital and have a better functional outcome after 6 months compared with patients who have undergone other forms of treatment[
8].
The unsolved questions of RSA are the need of refixation of the tubercles, the lack of retroversion and the lack of further solutions in case of failure, as we do not have many long-term results.
We set out to investigate preoperative characteristics as well as clinical and radiological outcomes in patients who had undergone a primary RSA as treatment for complex 3 or 4-part PHF in our department in the period from January 2008 to December 2011.
Discussion
The treatment of complex 3- and 4-part fractures of the proximal humerus with high risk of avascular necrosis represents a difficult problem for the surgeon. In our experience and according to the literature, methods such as plate or nail fixation may have a significant complication rate and unpredictable functional outcome.
Our mid-term follow-up shows satisfying results for the treatment of severe displaced fractures in elderly patients using RSA. According to our results, RSA can provide immediate relief and good shoulder function for elderly patients with severe PHF.
Nevertheless, the question of longevity of these implants remains to be observed. Long term follow-up studies for RSA as the primary treatment of PHF are necessary in order to assess the incidence of late or long-term complications with respect to component longevity, loosening, duration of pain relief and reduction of strength and ROM.
In a study by Favard et al.[
9] the authors found in a population with a mean age of 73 years a decrease in the relative CMS from 88% at under five years to 78% after more than 9 years. Additionally, the authors found that 72% of the patients had a CMS of less than 30 (defined as an end-point) after 10 years. In another study by Guery et al.[
10] a survivorship of 58% after 10 years is reported.
In patients younger than 65 years improved function is reported to be maintained for up to 10 years[
11].
Patients with severely displaced 3- or 4-part proximal humerus fractures are at high risk of suffering avascular necrosis of the humeral head[
12].
Open reduction and internal fixation (ORIF) is often difficult to achieve, as fragment displacement, comminution and osteoporotic bone quality act as limiting factors.
For these reasons, humeral head replacement is indicated when ORIF is not possible, or in cases where there is a high risk of avascular necrosis of the humeral head according to Hertel’s studies[
7,
13,
14].
Satisfactory shoulder function after hemiarthroplasty can be inadequate due to the problem of tuberosity fixation and preexisting rotator cuff disease with or without arthopathy especially in older patients. Poor shoulder function after hemiarthroplasty is often associated with nonunion, displacement and resorption of tuberosity fragments, which often leads to revision arthroplasty using an inverse design[
15].
While hemiarthroplasty has been seen to produce satisfactory pain control there are problems with limited range of motion and shoulder function when it is used as a treatment for severe proximal humerus fractures[
4,
16].
In a study by Boons et al.[
17] the authors investigated in a randomized controlled trial the outcome after hemiarthroplasty was used to treat four-part fractures in patients older than 60 years. In 25 patients treated with hemiarthroplasty they established a mean CMS of 64, a mean abduction of 77° and anterior flexion of 98°. In two patients they found postoperative tuberosity resorption. However, they noted that there may have been more disrupted tuberosities which were not observed on plain radiographs.
Comparative literature contrasting the functional outcome of RSA to hemiarthroplasty in the management of PHF is relatively limited. However, functional outcomes showed no significant difference between these two methods and were even higher in the RSA group according to CMS[
18,
19].
In a comparative study by Garrigues et al.[
20] the authors investigated 23 patients of whom 12 had undergone hemiarthroplasty and 11 had undergone reverse total shoulder arthroplasty for proximal humeral fractures in elderly patients. The mean ASES score of the RSA group was 81.1 (range 75-88) and was significantly better than the hemiarthroplasty group with a score of 47.4 points (range 30-81) (p < 0.05). The mean forward elevation was 121° (range 90°-145°) for the RSA group and 91° (range 30-140°) for the hemiarthroplasty group, respectively (p < 0.05).
In our study, results for RSA compare favourably with than the functional outcome for hemiarthroplasty in such patients. Our own results are encouraging and correlate with previously published reports from other trauma centers.
Proximal humeral head fractures often occur in low-demand patients, who can have multiple comorbidities and reduced requirements with regard to shoulder function. In this population group, the primary goal of treatment is pain-free shoulder mobility that provides good function for daily activities and personal hygiene requirements.
Use of primary RSA in the treatment of complex 3- and 4-part fractures of the humeral head remains controversial in Austria. It was previously not routinely performed in our trauma center. There are also certain limitations associated with RSA - the age and the comorbidities of the patients, for example, must of course always be taken into account.
Since 2008, after experiencing unsatisfying functional outcomes for treatment with hemiarthroplasty or plate fixation, we decided to treat patients over 65 years of age or low-demand patients with complex 3- and 4-part PHF and high risk for avascular necrosis and/or with preexisting comorbidities with primary RSA. Our preferred form of treatment would otherwise be ORIF using a plate or alternative methods such as humerus-block in selected cases.
As we have only operated on the worst forms of fractures in elderly or low-demand patients, we achieved a relatively satisfying functional outcome score of 54.8 in CMS, which is comparable to those described in the published literature[
4,
12,
18,
21].
We attempted to repair the tuberosities in all cases as there is evidence that healing of the tuberosities is associated with better functional outcome[
5,
22‐
24]. Satisfactory outcome also depends on the positioning of the glenosphere in the lower part of the glenoid in order to prevent scapular notching[
21,
25,
26].
The difference in the functional outcome by comparing the ROM, the grip strength and the pain relief in the non-affected and the affected shoulder did not have a major influence on day-to-day activities of the patients as the CMS and the DASH scores showed good results. Our results for pain reduction and range of motion correlate with results observed in previously published literature for this procedure[
10,
21,
27,
28].
We observed good functional results during rehabilitation with an early start of passive and active motion although a sling fixation was used for 4-6 weeks. This can also be associated with improved acceptance of the implant by the patient as RSA provides immediate stability of the shoulder.
We encountered some system related complications including a case of transient axillary nerve impairment, which resolved completely after 1 month without additional therapy. A case of dislocation of the shoulder lead to revision with a thicker inlay. The patient did not experience any further instability of the shoulder. We also observed 2 cases of infection. One was a superficial infection that was treated with debridement and did not lead to further complications. The other was a deep infection related to the prosthesis that led to explantation of the prosthesis in a multimorbid patient.
The study has several limitations. Firstly, not all of our operated patients fulfilled the inclusion criteria and we therefore had a highly selected group of patients. Secondly, the average follow up time of 20 months is relatively short. Thirdly, there was a lack of comparison group for comparison of results for ORIF or hemiarthroplasty for example.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
GM was responsible for conception, design, acquisition of data, analysis and interpretation of data and drafting the manuscript. LM has been involved in drafting the manuscript and interpretation of data. RO has been involved in critical revision of the manuscript and has improved the discussion section regarding complications associated with RSA. PW participated in the design of the study and was responsible for clinical management of the patients. RK has been involved in critical revision of the manuscript with regard to important intellectual content. AK provided general supervision of the research group and was involved in critical revision of the manuscript. All authors read and approved the final manuscript.