Proximal humerus fractures (PHFs) have bimodal distribution involving young patients with high-energy trauma or, more commonly, older patients with low-energy falls. There is an increasing trend in the use of arthroplasty, specifically the reverse shoulder arthroplasty (RSA), for surgical treatment of proximal humeral fractures especially in elderly patients, due to the increased risk of non-union and avascular necrosis.

A 50-year-old Southeast Asian male presented with a right shoulder 4-part right proximal humerus posterior fracture-dislocation with head split (AO Classification 11-C3). Open reduction internal fixation was performed with consideration of the patient’s age, activity level, and aim for early mobilisation. On review at postoperative 3 months, the patient regained right shoulder range of motion. Bony union was observed on radiographs taken at postoperative 5 months.

Conclusion

Open reduction internal fixation is a feasible surgical option for select 4-part proximal humerus fracture-dislocation, especially in the young, high functional demand patients. 

Proximal humerus fractures (PHFs) account for 4 to 10% of all fractures [1-3] and is the seventh most frequent fracture in adults [4-6]. There is a bimodal distribution involving young patients with high-energy trauma or, more commonly, older patients with low-energy falls [7]. There is an increasing trend in the use of arthroplasty, specifically the reverse shoulder arthroplasty (RSA), for surgical treatment of proximal humeral fractures especially in elderly patients [8-10]. Management of four-part proximal humerus fracture-dislocation (PHFD) is extremely difficult with high incidence of complications including avascular necrosis (AVN) [11]. This fracture pattern tends to have the worst prognosis because there are only very few or no soft-tissue links to the head (destroyed hinge periosteum). Elderly patients with this devastating fracture pattern are best treated by arthroplasty, however, the best treatment for younger patients (aged under 60 years) with significant loss of humeral head perfusion remains controversial [12]. The Proximal Fracture of the Humerus: Evaluation by Randomisation (Pro FHER) trial sought to evaluate the clinical and cost-effectiveness of surgical compared with non-surgical treatment for proximal fracture of the humerus in adults; and concluded that surgery does not result in a better outcome for most patients with displaced fractures of the proximal humerus involving the surgical neck and is not cost-effective [7,13,14]. Despite supporting non-operative treatment, an inherent limitation of the ProFHER trial lies in the exclusion of PHFs in which the authors deemed operative intervention necessary [7]. By extension, salvage surgery following initial non-operative treatment has been proven to yield poorer outcomes in functionality and pain as compared to initial operative treatment [15]. Neer et al. [16] first reported that four-part proximal humerus fractures are associated with a greater risk of necrosis. Hertel et al. then proposed a morphological classification based on the presence of a medial hinge, calcar length and significant displacement in evaluating humeral head ischemia [17]; in which patterns [2, 9, 10, 11, and 12]. and fractures with posteromedial head extension less than or equal to 8 mm, or diaphysis deviation greater than 2 mm (injury to the medial hinge), were at increased risk for Humeral Head Osteonecrosis [18]. However, later studies demonstrated that humeral head ischemia does not accurately predict for subsequent avascular necrosis [19,20]

There is ambiguity with regards to the most suitable treatment for middle aged patients who have high functional demands. Patient age is an important factor that influences the management of proximal humerus fractures [21]. Younger patients typically have better bone quality and may be able to tolerate surgical intervention better than older patients [22,23]. Humeral head-preserving surgical treatment is often prioritised over replacement (humeral head-sacrificing arthroplasty) since young patients may outlive finite implants, necessitating revision surgery. Yet, Zhao Y et al [24]. observed high implant survival rate with lasting pain relief and stable functional outcomes in younger patients after humeral head replacement (HHR) for proximal humeral fractures (PHFs) during long-term follow-up; in comparison to the increased risk for revision and functional deterioration over time after reverse shoulder arthroplasty in younger patients. However, evidence-based treatment recommendations for younger populations are limited by the lack of studies comparing treatment modalities and patient-reported outcomes [21,25]. The gap in knowledge exists surrounding the optimal treatment modality (head-preserving vs. replacement). Numerous studies have demonstrated no difference in outcomes between non-operative treatment versus common operative options such as open reduction internal fixation (ORIF), hemiarthroplasty (HA), and reverse total shoulder arthroplasty (rTSA) [26]. Hertel et al. [18] sought to evaluate the predictors of fracture-induced humeral head ischemia through anatomic considerations. Good predictors of ischemia were the length of the metaphyseal head extension, integrity of the medial hinge, and the basic fracture pattern. Moderate and poor predictors of ischemia were fractures consisting of four fragments, angular displacement of the head, the amount of displacement of the tuberosities, glenohumeral dislocation, head-split components, and fractures consisting of three fragments. When the above criteria (anatomic neck, short calcar, disrupted hinge) were combined, positive predictive values of up to 97% could be obtained. The most relevant predictors of ischemia were the length of the dorsomedial metaphyseal extension, the integrity of the medial hinge, and the basic fracture type determined with the binary description system. 

Patient factors influence the choice of surgery. Younger patients possess better healing, better prognosis [27], are not suitable for arthroplasty and farebetter with fixation [28]. The typical age cut off to decide between operative versus nonoperative management is 60 years of age. [25,48] non-operative treatment in the elderly with osteoporotic bone leads to compromised outcomes with early failures, especially in three- and four-part fractures [15,29]. Critchley O et al. [30] demonstrated that rTSA is superior to ORIF and HA for three- and four-part fractures in elderly patients with osteoporotic bone However, with patients increasingly being active in sports into their later years, many patients fall into the age group where arthroplasty, while being predictable, may not demonstrate the longevity and functional performance that patients demand. Rehabilitation for these patients is also challenging. We present a challenging case of a proximal humerus fracture which falls in this ambiguous zone, and document our experience, especially in rehabilitation and recovery, along with some technical tips. 

We present a 50-year-old Southeast Asian male who presented to the emergency department with right shoulder pain and deformity. This closed injury was sustained after colliding with a rock while cycling, in which the patient fell off his bike and landed onto his right shoulder. Neurovascular status was intact, and radiographs of the right shoulder revealed a 4-part right proximal humerus posterior fracture-dislocation with head split (AO Classification 11-C3) (Image 1). During the consultation, the options of surgical fixation and hemiarthroplasty (HA) were both explained to the patient. The risk of AVN leading to salvage arthroplasty was discussed for fixation, whereas the concerns of implant longevity and dislocation risk was discussed for arthroplasty. In view of the displaced fracture, shared decision making between surgeon and patient was made for surgery to facilitate early mobilisation. Considering the patient’s age and activity level, fixation will be attempted. However, the patient was counselled on the possibility of shoulder hemiarthroplasty should the intraoperative findings reveal a fragmented fracture pattern to prevent AVN, or an unfavourable fracture pattern in which fixation is unlikely to be successful. One key limitation of Hertel’s study to predict the risk of AVN is the presence of few young patients (mean age of the study participants were 60 years old) [18]. This also influenced the surgeon’s decision to offer joint preserving surgery to the patient. A computed tomography scan was not necessary since the severity of the fracture-dislocation is evident on radiographs. 

Diagnostic studies

                                                                                                        A. Anteroposterior (AP) view

                                                                                                                         B. Valpeau view

                                                                                                        C.  Lateral/ Y scapula view

        Image 1: Preoperative imaging of the right shoulder showing a right proximal humerus posterior fracture-dislocation with head split.

The patient underwent open reduction internal fixation (ORIF) on post-injury day 11. Following a deltopectoral approach and biceps tenodesis, the 4-part PHFD was seen. The humeral head was posteriorly dislocated and impacted upon the posterior glenoid with a nondisplaced head split. The greater tuberosity fragment was identified with an intact supraspinatus However there was an avulsion of the subscapularis off the lesser tuberosity.  The humeral head was reduced with judicious soft tissue management and assessed in-situ to prevent secondary injury and ischemia. The articular block was secured with 2 headless compression screws. The neck-shaft angle was restored using plate-assisted reduction and an angular stable construct was achieved with an anatomical locking compression plate (Synthes PHILOSTM). To prevent varus malunion, the height of the plate was determined on fluoroscopy (Image 2) to enable the inferomedial ‘kickstand’ screw [31]. The medial calcar was reduced and cerclaged to the plate for bony apposition and increased fixation strength. Cement augmentation (Stryker Hydroset) was performed. Transosseous repair of the Subscapularis was done with closure of the rotator interval.

Image 2: Intraoperative fluoroscopy demonstrating satisfactory anatomical reduction of the 4-part proximal humeral posterior fracture-dislocation.

                                                                Image 3: Immediate postoperative radiographs of the right shoulder.

                                                                                                     A. Anteroposterior (AP) view

                                                                                                                       B. Valpeau view

                                                                                                               C. Lateral/ Y scapula view

                                   Image 4. 5: Month follow-up radiograph of the right shoulder demonstrating bridging callus and bony union.

                                                                                                               A. Anteroposterior (AP) view

                                                                                                                   B. Valpeau view

                                                                                                                   C. Lateral/ Y scapula view

                                            Image 5: 24-month follow-up radiograph of the right shoulder demonstrating bony union.

                                                                                                             A. Anteroposterior (AP) view

                                                                                                                        B. Valpeau view

                                                                                                             C. Lateral/ Y scapula view

Postoperative course 

Physiotherapy Rehabilitation and Postoperative Recovery

Postoperative rehabilitation following ORIF focused on progressive joint protection, mobility restoration, and functional strengthening. [33,34] The rehabilitation process was structured into distinct phases to facilitate a safe return to function while minimizing complications.

The patient followed a structured rehabilitation program that progressed in alignment with key postoperative milestones (Table 1). Pendulum exercises were initiated in the early phase (week 3 post-op) to promote gentle mobility while respecting non-weight-bearing precautions.[33] By week 7, the patient progressed to passive and active-assisted ROM exercises, with particular emphasis on external rotation (Image 6) and flexion (Image 7) using cane-assisted techniques. [33,34] At this stage, the patient also returned to light-duty work. Due to reported stiffness during daily activities, external rotation mobility (Image 8) was further emphasized at the 3-month review. At week 12, the patient initiated active ROM with light resistance, including anterior press and pulley-based exercises. To address his persistent motion limitations, glenohumeral joint mobilization was introduced [35]. Radiographs at 5 months post-operation (Image 4) confirmed bony union, and by week 23, the patient had achieved full functional recovery, resumed cycling and daily activities, with continued education to avoid high overhead movements in view of rotator cuff muscle repair.

Table 1: Postoperative Rehabilitation Guideline [33,34]

                                                                             Image 6: External rotation exercise using cane-assisted techniques

                                                                                     Image 7: Flexion exercise using cane-assisted techniques

                                       Image 8: External rotation mobility exercise was further emphasized at the 3-month review.

This rehabilitation process reflected a progressive, individualized approach that prioritized joint protection, functional mobility, and long-term return to activity.

Postoperatively, the patient was started on pendular exercises for 6 weeks. before increasing to passive ROM at the postoperative 7 week. He was seen regularly at postoperative 2 weeks, 7 weeks, 3 months and 5 months. Patient returned to work while being placed on light duty at postoperative 7 weeks. On review at postoperative 3 months, there was an increased focus on external rotation range of motion exercises due to shoulder stiffness reported by the patient during daily activities. Bony union was observed on radiographs taken at postoperative 5 months (Image 4).

                                                                                                                         A. Abduction

                                                                                                                     B. External rotation

                                                                                                                     C. Internal rotation

                                Image 9: Shoulder ROM at postoperative 5 months showing excellent forward flexion, external rotation and internal rotation.

                                                                                                                           A. Abduction

                                                                                                                      B. External rotation

                                                                                                                       C. Internal rotation

                         Image 10: Shoulder ROM at postoperative 24 months showing excellent forward flexion, external rotation and internal rotation.

PHFD are challenging to treat (Table 2). Our patient had a four-part proximal humeral posterior fracture-dislocation with several risk factors for humeral head ischemia. His risk factors include (1) anatomical neck fracture (2) 4-part severely displaced fracture-dislocation (3) loss of medial calcar (4) presence of head split and (5) subacute timing to surgery (about 2 weeks).

  Table 2: Evidence surrounding proximal humerus fractures (PHFs).

Table 3: Our case report shows that fixation of comminuted PHFD is a robust joint preserving alternative to shoulder replacement. Although the results of a secondary arthroplasty following failure of a primary head-conserving procedure may be reportedly less satisfactory than a primary arthroplasty, fixation is still important and can yield good results in PHFD.

In young people, every attempt should be made to reduce the fracture to obtain the original anatomy of the humeral head [11]. Solioman et al. reported good outcomes in their prospective study of 39 young patients with PHFD. Two-thirds of patients were pain-free, 9 had mild pain and 4 had moderate pain [11]. This is supported by Robinson et al. who reported a relatively low risk of postoperative complications and generally favourable functional outcomes [36].

Although fixation is generally preferred in younger patients, it is not without its risks. If AVN develops, secondary arthroplasty can be performed. However, the results of a secondary arthroplasty following failure of a primary head-conserving procedure may be less satisfactory than a primary arthroplasty [11]. In a cohort study of 40 PHFD patients, Haupt S et al. [37] studied 33 patients (mean age: 50) with fixation performed and 7 patients (mean age: 68) with primary arthroplasty. There are functional outcome deficits in patients with AVN compared to no AVN, revision surgery compared to no revision surgery and in hemiarthroplasty (HA) compared to RSA (reverse shoulder arthroplasty). Revision rates in fixation are high (18%) leading to secondary conversion into arthroplasty, thereby decreasing outcome scores significantly [37]. This is also supported by Demirhan et al. [38] who concluded that primary replacement is the superior treatment considering the high risk of AVN and severe collapse of the humeral head. A recent study by Gavaskar et al. [39] further asserted that internal fixation resulted in poor shoulder function and complications in high number of patients, although fracture union was achieved in most patients. The lack of a predictive criteria for AVN makes decision making difficult [40] between ORIF and arthroplasty. The basis of morphological classifications of proximal humerus fractures is due to blood supply of the humeral head, with the posterior circumflex being the predominant blood supply [41,42]. While the Hertel classification is a reliable classification system [42]. Hertel’s criteria is unable to predict the development of AVN after proximal humerus fracture [44]. Malcherczyk D et al. [45] assessed humeral head back bleeding intraoperatively using proxy indicators of blood loss and transfusion rate and found that the rate of blood transfusion and amount of blood loss are higher in patients receiving shoulder arthroplasty than in patients with plate fixation for proximal humerus fractures. However, ischemia may not necessarily lead to AVN [43,46]. Barlow JD et al. [47] reported that despite the fracture being united in ORIF, AVN may still develop. Both scoring systems predicting patients who fail primary ORIF and the reasons surrounding the development of AVN late into the injury have yet to be established. While our patient presented with multiple risk factors relating to high risk of ischemia, it can be tempting at the point of surgery to undergo primary arthroplasty. Yet, our case study shows that patients with primary proximal humerus ORIF and absence of AVN can yield superb results. Therefore, surgeons should not jump into primary arthroplasty straight away, which may be what some are advocating for.

The management of proximal humerus fracture-dislocation (PHFD) is difficult with high incidence of complications such as avascular necrosis (AVN). The decision for operative versus nonoperative management of proximal humerus factors is multifactorial. The typical age cut off to decide between operative versus nonoperative management is 60 years of age.Since younger patients possess better healing and better prognosis, they are not suitable for arthroplasty and fair better with fixation. Humeral head-preserving surgical treatment is often prioritised over replacement (humeral head-sacrificing arthroplasty) since young patients may outlive finite implants, necessitating revision surgery.  Given that patients are being increasingly active in sports into their later years, many patients fall into the age group where arthroplasty, while being predictable, may not demonstrate the longevity and functional performance that patients demand. Besides age, other patient factors that are important in influencing the decision to pursue surgery is included for.

No funding was received in support of this work.

Proximal humerus fracture dislocation.