ABSTRACT

Objective

To present and compare the results of surgical methods used in the treatment of osteoporotic proximal humerus fractures (PHFs).

Methods

A retrospective examination of patients who underwent surgery for osteoporotic PHF between 2009 and 2013 was conducted. The demographic data of the patients, surgical methodology, concomitant injuries, time intervals before surgery, hospitalization time, and follow-up time were recorded. Constant-Murley score and American Shoulder and Elbow Surgeons score were utilized for functional evaluation. Shoulder abduction and flexion ranges of motion were measured as objective evaluation.

Results

Sixteen patients (64%) were operated with plate-screw osteosynthesis, two patients (8%) with percutaneous Kirschner wire fixation and seven patients (28%) with partial shoulder arthroplasty. A significant correlation was identified between treatment and Neer classification (P = 0.011). No significant correlation was observed between functional scores and surgical method (P > 0.05 for each). Objective evaluations revealed a significant difference in shoulder abduction range of motion between patients and surgical method (P = 0.030). Post-hoc analyses showed a significant difference between plate-screw osteosynthesis and hemi-arthroplasty groups (P = 0.010).

Conclusion

Percutaneous techniques, plate-screw osteosynthesis, or arthroplasty methods may be preferred in osteoporotic PHFs, with no superiority over each other. The decision regarding the surgical method for geriatric PHFs should be based on patient -and fracture- related factors.

INTRODUCTION

Proximal humerus fractures (PHFs) can occur even in low-energy trauma, such as simple falls, due to declining bone quality (osteoporosis), especially in older age. PHFs are the most common fracture site after hip and distal radius fractures in the elderly.^{1, 2} Most geriatric PHFs are treated conservatively, given the decreased functional expectation and increased comorbidities. However, for fractures with significant displacement and multiple comminution, surgical treatment come to the fore. As a result of the increasing incidence of fractures in this region and technological developments in orthopedic implants, various surgical methods have been used to PHFs. Despite the numerous studies conducted on the subject, a consensus on the optimal surgical method remains elusive and the relative merits of each method continue to be debated.³ The primary goal of surgery for geriatric PHFs is to allow patients to resume their daily activities as soon as possible. Age, bone quality, fracture pattern, and surgical timing all significantly affect the patient’s functional result. Each PHF is patient-specific, and there is no single universal surgical method that can be used for every patient when conservative treatment is not possible. Therefore, a patient-specific, evidence-based treatment approach should be selected.⁴

This study aims to present and compare the results of surgical methods used to treat osteoporotic PHFs.

MATERIAL AND METHODS

Study Population and Data Collection

Following approval by the Gülhane Military Medical Academy Haydarpaşa Training and Research Hospital Ethics Committee (approval no.: 2013-114, date: 26.12.2013), all patients who underwent surgery for a PHF at the study clinic between October 2009 and December 2013, aged over 60 years, were retrospectively reviewed. Within the specified period, 38 patients were identified who underwent surgery with a diagnosis of osteoporotic PHF. Inclusion criteria were defined as being over 60, having undergone surgical treatment for a PHF in our clinic, and having regular follow-up visits. Seven individuals passed away for a variety of causes, thus they were not included in the study. Three patients were unreachable due to alterations in their contact information. Three patients did not want to take part in the study. Consequently, 13 patients were excluded from the study, whereas 25 patients were included.

Surgical Technique and Rehabilitation

The percutaneous method was used for Kirschner wire (K-wire) fixation.⁵ For plate osteosynthesis (Proximal Humerus Locking Plates, TST Orthopedics^®, TST Medical Tools^®, İstanbul, Türkiye) and arthroplasty (Partial Shoulder Prosthesis, Hipokrat Incorporated Company, İzmir, Türkiye), A deltopectoral technique was used to reach the proximal humerus.⁶ Passive shoulder exercises were initiated on the first postoperative day. Patients who did not have any problems at the wound site were discharged and asked to have dressings every day. All patients were contacted for a follow-up two weeks after the operation, at which point sutures were removed. In the third postoperative week, active assisted shoulder exercises were described in addition to passive shoulder exercises. Postoperative rehabilitation recommendations were obtained for all patients, and home rehabilitation plans were arranged and encouraged.

Demographic Data and Functional Evaluation

Anteroposterior and lateral radiographs or computed tomography (CT) scans of the shoulder taken on admission were used to classify PHFs according to the Neer classification.² CT scans were used to diagnose and classify cases of multiple comminuted fractures and fracture-dislocations and to determine the surgical method. None of the patients underwent magnetic resonance imaging. Concomitant injuries and the procedures performed for these injuries were recorded. Waiting times for surgery after fracture, reasons, and comorbidities were recorded. The methods used for surgery (K-wire fixation, plate and screw fixation, and arthroplasty) were listed.

Patients were contacted using the contact details in the hospitalization file and the hospital information system. Included patients were invited to our hospital by telephone, provided that a minimum follow-up of at least one year had been achieved. Constant-Murley scoring and American Shoulder and Elbow Surgeons (ASES) scoring were used for functional assessment.^{7, 8} Constant-Murley scoring was used for clinician-based assessment and ASES scoring was used for patient-based assessment. Radiological assessment was performed with direct anterior-posterior and lateral radiographs of the shoulder. Fixation failure and union were assessed on direct radiographs. In terms of objective evaluation, shoulder abduction and flexion ranges of motion were measured on all patients at the last follow-up. Using a standard universal goniometer and the triangulation sites, the same physician assessed the patient’s range of motion.

Statistical Analysis

The International Business Machines (IBM^®) Statistical Package for the Social Sciences (SPSS^®) software, version 26.0 (IBM SPSS Corp.; Armonk, NY, USA), was used to conduct the statistical analysis. Descriptive statistics such as mean, standard deviation, and minimum-maximum values were utilized. Frequency (percentage) were used as descriptive statistics for categorical data. When evaluating the scale data in three-group comparisons the groups were compared using the Kruskal-Wallis test, and post-hoc analyses were carried out using the Mann-Whitney U test. The categorical data were compared using the chi-square test. When the P value was less than 0.05, statistical significance was deemed to exist.

RESULTS

Sixteen patients (64%) were operated with plate-screw osteosynthesis, 2 patients (8%) with percutaneous K-wire fixation and 7 patients (28%) with partial shoulder arthroplasty. The mean age of the patients was 75.08 ± 8.505 years (range: 60-92). According to Neer Classification, 10 patients (40%) had two-part fractures, 13 patients (52%) had three-part fractures and 2 patients (8%) had four-part fractures (Figure 1). Detailed demographic data of the patients are shown in Table 1.

It was determined that 62.5% of the patients treated with plate-screw osteosynthesis had Neer type 2 fractures and 85.7% of the patients treated with arthroplasty had Neer type 3 fractures. There was a significant correlation between the treatment and Neer classification (P = 0.011). There was no significant correlation between the treatment and other fracture and patient characteristics (Table 2).

There was no significant correlation between the functional scores and the surgical method applied in the last follow-up of the patients (P > 0.05 for each). In the objective evaluations, there was a significant difference between the shoulder abduction range of motion of the patients and the surgical method applied (P = 0.030). Post-hoc analyses revealed no significant difference between osteosynthesis with plate-screw and fixation with K-wire (P = 0.941) and between fixation with K-wire and hemi-arthroplasty (P = 0.111), whereas a significant difference was found between osteosynthesis with plate-screw and hemi-arthroplasty groups (P = 0.010). The relationship between the applied surgical method and functional and objective measurements is shown in detail in Table 3.

In one patient treated with plate osteosynthesis, revision plate osteosynthesis was performed in the sixth postoperative week due to loss of reduction. In the follow-up of the same patient, parenteral antibiotherapy was applied for superficial infection due to serous discharge at the wound site and the complaint regressed. In another patient treated with plate osteosynthesis, the fixation materials were removed 1.5 years postoperatively due to pain and abduction limitation despite physiotherapy and the complaints disappeared afterwards.

DISCUSSION

Surgical techniques for PHFs include many options such as minimally invasive techniques, plate and screw applications, hemiarthroplasty, and total shoulder arthroplasty. Our study investigated the superiority of three different surgical techniques described in the literature for osteoporotic PHFs. The most striking finding of this study was that none of the surgical methods investigated was superior to the other in terms of functional scores. Another point that should be emphasized is that the shoulder abduction range was greater in the plate-screw osteosynthesis group.

In epidemiological studies, osteoporotic PHFs are more common in women, and fractures usually occur in low-energy trauma, such as falls from the same level. In our study, PHFs were more common in women (92% vs. 8%). Our results are consistent with the literature. In epidemiological studies with larger patient series, fractures are more common in women and the incidence of fractures increases with the aging of the population.⁹

In our study, the mean interval between initial presentation and surgery was calculated to be 18.48 (3-153) days. The large discrepancy between the waiting times was because several patients were initially indicated for conservative management, while surgery was later decided due to loss of reduction and one patient underwent surgery at a late stage due to non-union. The mean length of hospital stay in our study was 14.6 (3-66) days. Prolonged preoperative preparation and increased need for postoperative care due to comorbidities and concomitant fractures were the reasons for the increased length of hospital stay. Eighteen patients had comorbidities.
These included hypertension, diabetes mellitus, heart disease, hyperlipidemia, lung disease, Alzheimer’s disease, hypothyroidism, and chronic renal failure. In addition, six patients in our cohort had concomitant fractures; the first patient had a lateral plateau fracture and osteosynthesis was achieved with two cannulated screws. In addition, a symphysis pubis arm fracture and a fibular shaft fracture were conservatively managed. In the second patient, the concomitant patella fracture was fixed with a traction device. In the third patient, intramedullary nailing was performed for the diaphyseal fracture of the femur. In the fourth patient, the distal radius fracture was treated with closed reduction and a short arm cast. In the fifth patient, partial hip arthroplasty was performed for the collum femoris fracture. The sixth patient underwent proximal femoral nailing for the intertrochanteric femoral fracture and closed reduction with percutaneous pinning for the distal radius fracture.

In our study, 10 patients (40%) had Neer type 2 fractures and 13 patients (52%) had Neer type 3 fractures. Although osteoporotic fractures are expected to be more comminuted due to the fragile bone structure, our results differ from this hypothesis. We believe that the most important reason for this is that almost all the injury mechanisms in our study (92%) were caused by low-energy injuries. Another important reason for the low number of multi-segment Neer type 4 fractures in our cohort may be that conservative treatment of these fractures, especially in the geriatric population, is more prominent both in the literature and in our study group.^{10, 11}

Although higher functional scores were obtained with K-wire fixation and plate-screw osteosynthesis compared to the arthroplasty group, the difference between them was not statistically significant in our study. Similar results have been reported in the literature. However, recent studies have reported that reverse shoulder arthroplasty is preferred to partial shoulder arthroplasty in geriatric multisegment PHFs, and the functional results are similar to those of plate osteosynthesis.^{12, 13}

Our study’s most notable observation was the considerable variation in shoulder abduction range of motion during the last follow-up (P = 0.03). In post-hoc analyses, no significant difference was found between plate-screw osteosynthesis and K-wire fixation (P = 0.941) and between K-wire fixation and hemiarthroplasty (P = 0.111), whereas a significant difference was found between the plate-screw osteosynthesis and hemiarthroplasty groups (P = 0.010). When analyzing the reasons for this situation, it is striking that the groups were not homogeneously distributed. The limited number of patients with percutaneous fixation (K-wire) may have influenced the statistical analyses. Another point to emphasize is that the rate of Neer type 2 fractures was higher in the plate-screw fixation group. Finally, all arthroplasties in our study were partial shoulder arthroplasties. As mentioned above, the number of publications in the literature reporting satisfactory results with reverse shoulder arthroplasty for PHFs is increasing daily.^{13, 14}

Study Limitations

Our study had some limitations. These are;

- The number of patients and surgical techniques used in our study were small and the groups were not homogeneous,

- Absence of reverse shoulder prosthesis cases among the surgical methods used, which are becoming more common today,

- Differences in the time interval between hospital admission and surgery,

- Postoperative bone mineral density (BMD) was not measured and the relationship between BMD age and BMD fracture incidence was not evaluated.

In conclusion, the number of osteoporotic fractures is increasing with the ageing of the population, and PHFs constitute a significant proportion of these. There are many surgical methods for treating geriatric PHFs, and our study showed that these surgical methods are not clearly superior to each other. Although arthroplasty options have become increasingly prevalent in this age group due to technological advancements, satisfactory clinical outcomes can be achieved with plate-screw osteosynthesis in geriatric PHFs.