Cement‐Augmented Screw Fixation with PHILOS Plating for Osteoporotic Proximal Humeral Fractures: An Observation of Mid‐ and Long‐Term Curative Efficacy

Abstract

Objectives

The mid‐ and long‐term clinical outcomes of cement‐augmented screws in the treatment of osteoporotic proximal humeral fractures have rarely been reported. The aim of this study was to observe the mid‐ and long‐term efficacy of combined cement‐augmented screw fixation and PHILOS plating in the treatment of osteoporotic fractures of the proximal humerus.

Methods

This study retrospectively analyzed data from 19 patients with osteoporotic fractures of the proximal humerus who had undergone internal fixation at the Guizhou Provincial People's Hospital from February 2017 to May 2021. The cohort was comprised of six males and 13 females, aged 75–87 (mean age: 82.52 ± 1.24) years. According to the Neer classification, three, 12, and four patients had two‐part, three‐part, and four‐part fractures, respectively. All patients were treated with open reduction internal fixation with cement‐augmented screws and PHILOS plating. Time until fracture healing was recorded postoperatively. Patients were observed for postoperative complications, including humeral head necrosis, loosening or breaking of the augmented screws, screw perforation of the humeral head, and varus fracture displacement. Visual analog scale and Constant scores of the shoulder joint were compared 1, 3, 6, and 12 months after surgery. Scores at the most recent follow‐up were used to evaluate shoulder joint function. Measured data conforming to a normal distribution were expressed as mean ± SD. Analysis of variance or rank sum tests were used for intergroup comparisons. A value of p < 0.05 was considered significant.

Results

All 19 patients followed up for 1–4 (average: 2.13 ± 0.61) years. Fractures united in all cases, with a healing time of 8–14 (average: 10.25 ± 1.72) weeks. There were no cases of humeral head necrosis, screw loosening, fractures, or perforation of the humeral head. One patient had mild varus fracture displacement with a reduced neck‐shaft angle. There were significant differences in visual analog scale and Constant scores 1, 3, and 6 months after surgery (p < 0.05). The visual analog scale score was 0 at final follow‐up in all cases. The Constant score of the shoulder joint was excellent, good, fair, and poor in two, 12, four, and one case, respectively, yielding an excellent and good rate of 73.68%.

Conclusions

Cement‐augmented screw fixation combined with PHILOS plating of osteoporotic proximal humeral fractures had good mid‐ and long‐term clinical efficacy. It should be considered a new option for fracture treatment in such patients.

Cement‐augmented screws combined with PHILOS plate in the treatment of osteoporotic fractures of the proximal humerus have achieved satisfactory postoperative results.

Introduction

Proximal humerus fractures are among the most common bony injuries, accounting for 4%–5% of all fractures. ¹ As the population ages, osteoporotic proximal humeral fractures have gradually become more common in older adults, although they are less common than distal radius, intertrochanteric, and femoral neck fractures. Proximal humerus fractures are more common in elderly women. ² , ³ Many patients choose surgical treatment to permit early function and a rapid recovery. At present, the common methods for the treatment of proximal humerus fractures in the elderly include: (i) intramedullary nail internal fixation of proximal humerus fractures, which is central fixation of the fracture and has more minimally invasive surgery and better biomechanics ⁴ ; however, for proximal humerus fractures caused by osteoporosis, complications such as screw loosening and humeral head varus may occur, and there is a risk of injury to the rotator cuff, especially for patients with Neer classification of three parts or four parts, the fixation of the greater and lesser tuberosities is relatively difficult. ⁵ (ii) Shoulder arthroplasty may be considered for elderly patients with Neer type three‐part or four‐part osteoporotic proximal humerus fractures, and although shoulder arthroplasty is an effective treatment for osteoporotic proximal humerus fractures that cannot be reconstructed or have high expected complications after reconstruction, this treatment can effectively relieve pain symptoms in patients. However, there are different reports on the recovery of postoperative shoulder function, and the main reasons affecting postoperative shoulder function are poor position or healing of the greater and lesser tuberosities of the humerus, with an incidence as high as 12.5%–50%. ⁶ (iii) Open reduction and plate and screw internal fixation is one of the most commonly used surgical options for these patients. However, osteoporosis may predispose patients to postoperative complications such as screw loosening, humeral head screw penetration, humeral head varus, and non‐union after internal fixation. ⁷ , ⁸ These create additional clinical challenges.

A new type of cannulated cement‐augmented screw technology has been developed to reduce complications after internal fixation. It has achieved good clinical outcomes in the management of osteoporotic fractures of the proximal humerus. Unger et al. ⁹ studied the biomechanics of humeral head fixation with bone cement‐augmented screws in vitro. They found that cement‐augmented screws had better biomechanical properties than non‐augmented screws, and suggested that cannulated screw augmentation is an effective method for fracture repair. Van Veelen et al. ¹⁰ reported good clinical outcomes using cannulated screw augmentation technology in patients with osteoporotic proximal humeral fractures. This method is a safe and effective solution for treating osteoporotic proximal humeral fractures. A previous study found that cement‐augmented screw fixation with locking plate internal fixation had similar clinical outcomes to patients treated with artificial humeral head replacement, suggesting that it may be a viable alternative to shoulder replacement for some patients. ¹¹

The mid‐ and long‐term efficacy of combined bone cement‐augmented screw fixation and PHILOS plating in the treatment of osteoporotic proximal humeral fractures has not been well reported, and thus remains controversial. The present study sought to evaluate the mid‐ and long‐term clinical efficacy of this combined treatment; to understand the end‐stage complications of cement‐augmented screws through a regular long‐term follow‐up; and to identify patterns in the postoperative functional recovery of these patients to inform patient and rehab education. It also discusses the advantages of cement‐enhanced screws in osteoporotic fractures of the proximal humerus in the elderly, and provides technical suggestions for the widespread use of this technique in clinical practice.

Methods

Subjects

This retrospective case analysis enrolled 19 patients with osteoporotic proximal humeral fractures. The cohort was comprised of six males and 13 females aged 75–87 (mean age: 82.52 ± 1.24) years who had been treated at the Guizhou Provincial People's Hospital between February 2017 and May 2021. All fractures were closed and caused by low‐energy trauma. Concurrent injuries or diseases included a distal radius fracture, rib fractures, hypertension or diabetes, coronary heart disease, chronic bronchitis, and cerebral infarction sequelae in one, two, 11, four, two, and one case, respectively. According to the Neer classification, the fractures were classified as two‐, three‐, and four‐part fractures in three, 12, and four cases, respectively. Dual‐energy X‐ray absorptiometry was used to evaluate bone mineral density after admission; osteoporosis was defined at a T score ≤ −2.5 standard deviations (SD). ¹² Patients underwent surgical treatment after preoperative examination, consultation with relevant departments, and consideration of absolute surgical contraindications. Each enrolled patient underwent open reduction internal fixation with cannulated cement‐augmented screw fixation and PHILOS plating within 4–9 days of the injury.

Diagnostic Criteria

Patients were diagnosed based on the following criteria: (1) a clear history of low‐energy trauma; (2) pain in injured shoulder joint with limitation of motion; (3) imaging results indicative of a proximal humeral fracture; (4) closed injuries; and (5) older patients with a T score ≤ −2.5 SD.

Inclusion Criteria

The following factors were considered for study inclusion: (1) surgically indicated fractures; (2) two‐, three‐, and four‐part fractures according to the Neer classification; (3) age over 65 years; and (4) a fracture to surgery time of <2 weeks. Indications for surgery were: (1) closed osteoporotic fractures of the proximal humerus in patients older than 65 years of age, and (2) Neer classification of fractures as two‐, three‐, and four‐part fractures. Surgical contraindications were (1) pathologic and open fractures; and (2) a split humeral head.

Exclusion Criteria

Patients with any of the following factors were excluded: (1) severe fracture comminution as assessed with a CT scan before surgery that could not be reduced and internally fixed; (2) obvious functional impairment of the shoulder joint before the injury; (3) a general condition unsuitable for tolerating anesthesia and surgery; (4) postoperative follow‐up of less than 1 year.

Surgical Methods

Surgical Procedure

Imaging Evaluation

Preoperative radiographs and plain CT scans of the affected shoulder joint were obtained in all cases to understand the integrity of the humeral head and the presence or absence of head split fractures. Based on these data, the shoulder joint was remodeled three‐dimensionally to evaluate fracture type and displacement and to obtain better insight into the pre‐injury function of the shoulder joint in each patient.

Anesthesia

Patients were placed under general or nerve block anesthesia and positioned supine on a fluoroscopic operating table with their affected shoulder and limb slightly elevated.

Approach

Step 1: After routine disinfection and draping of the surgical site, a deltopectoral approach was performed. The cephalic vein was exposed and protected and the fracture site was exposed and repaired through the muscle space of the rotator cuff. Limited stripping of the periosteum was performed at the edge of the fracture fragments for fracture reduction. Number 5 Ethibond sutures were used to reduce and stabilize three‐ and four‐part fractures. Soft tissue at the ends of the fracture were removed, and fracture fragments were reduced and temporarily stabilized using Kirschner wires.

Step 2: A PHILOS locking plate of an appropriate length was selected and placed 5–8 mm below the apex of the greater tuberosity of the humerus and 2–4 mm posterior to the intertuberous sulcus. Temporary fixation was achieved with a Kirschner wire. Satisfactory fracture reduction was obtained under C‐arm fluoroscopy, with the locking plate positioned appropriately. The humeral head was stabilized using 2–3 Kirschner wires through the dedicated holes on the steel plate, and the humeral shaft was fixed using one lag screw.

Step 3: In consideration of fracture lines and humeral head size, three cannulated cement‐augmented screws (positioned in a triangular fashion) or four cannulated cement‐augmented screws (distributed in quadrangular fashion) were used. Bone cement was then injected via the screw cannula for cement augmentation. The remaining locking screws were inserted after the bone cement had solidified.

Step 4: Number 5 Ethibond sutures were used to stabilize large and small tubercle fragments and rotator cuff tissue as required. Satisfactory fracture reduction was obtained under C‐arm fluoroscopy, with internal fixation at an appropriate length. There was no shoulder joint abnormality noted during passive movement of the fracture ends. Surgery was completed with a suture closure. All patients were operated on and followed by the same group of surgeons.

Bone Cement Injection

Bone cement for trauma (40% zirconium dioxide [contrast agent] + 15% hydroxyapatite +45% polymethyl methacrylate) was used for all of the patients in this cohort. The bone cement, which has a high initial viscosity, is suitable for long operative times, has better radiographic imaging characteristics, and was mixed and injected using a matching syringe system. Surgeons could follow a familiar routine procedure with only a few additional steps. The consistency of the bone cement was adjusted to achieve a liquid state. Approximately 0.5 mL of bone cement liquid was injected into the screw cannula under C‐arm fluoroscopy, and distributed in a cloud‐shaped configuration at the distal end of the screw. The patient's blood pressure was measured to ensure that it was appropriately elevated before bone cement injection. The blood pressure and respiratory rate were closely monitored during the injection process. Patients were monitored for bone cement leakage into the fracture line or shoulder joint.

Selection of Cement‐Augmented Screws (Figure 1)

FIGURE 1.

Schematic diagrams of the PHILOS plate screw distribution and cannulated cement‐augmented screw configuration. (A–E) The distribution of screws for fixing the humeral head, where A is the screw fixing the marginal humeral head, B and C are the screws fixed in the middle of the humeral head in different directions and planes, D is the screw fixing the central humeral head, and E is a humeral calcar screw.

Augmented screws at levels A–E were selected if the humeral head was cement‐reinforced. This allowed the cement‐supported masses to be widely distributed in the humeral head. However, augmentation could not be performed in patients with a small humeral head because the tip of the screw may be located within the fracture line or level E screws could not be inserted due to the small size of the humeral head. In these cases, level B and/or D screws were used for augmentation. Level C screws are not recommended because the tip of the screw is often at the same level as level A screws. The configuration of the augmented screw was based on the size of the humeral head. In this context, augmentation should proceed at different planes and be placed away from the fracture line. A total of 3–4 screws were generally used.

Precautions

The following precautionary measures were observed: (1) the configuration of the augmented screw was selected according to the size of the humeral head, (2) the distribution and number of screws were based on the configuration of the augmented screw, (3) leakage of bone cement into the shoulder joint space or the fracture line was avoided and a contrast agent was injected for confirmation before cementing, (4) the bone cement injection volume for each screw was generally not more than 0.5 mL, (5) the patient's vital signs were closely monitored during the injection of bone cement to prevent bone cement reaction and embolism, and (6) the distribution of bone cement was closely observed during injection under intraoperative fluoroscopy. If the bone cement leaked into the shoulder joint space, the injection was stopped in time and the leaked cement was removed if necessary.

Perioperative Management

Preoperative Management

Preoperative examination and consultation were completed and the patient's blood pressure and blood sugar were brought to within acceptable limits to ensure that the patient's condition was relatively stable. Patients with absolute contraindications to shoulder surgery were identified.

Antibiotics were administered 30 min before and after surgery to prevent infections.

Analgesic Regimen

An individualized analgesic regimen was used for each patient. Non‐steroidal anti‐inflammatory drugs were used before surgery, and low dose opioid analgesics were added if necessary. The aim behind this pain regimen was to keep the patient's visual analog scale (VAS) score below 4 points. Nerve block anesthesia or a local injection of an anesthetic cocktail were used to reduce postoperative pain. The postoperative VAS score was controlled to a level of below 4 using patient‐controlled analgesia and non‐steroidal anti‐inflammatory drugs. Additional low dose opioid analgesics were added if necessary.

Postoperative Management

Patients underwent routine anti‐osteoporosis treatment after surgery, which included oral calcium tablets, alfacalcidol, alendronate or injectable bisphosphonates (that inhibit bone resorption), and teriparatide (that promotes bone formation). In patients with a history of fragility fractures caused by osteoporosis, intravenous calcitonin was used to reduce acute bone loss after the fracture.

For patients with no obvious damage to their rotator cuff, passive shoulder exercises (including anteflexion, rear extension, and abduction) and elbow flexion/extension exercises were initiated within 1 week of surgery. Active shoulder joint exercises were then gradually introduced. For patients with rotator cuff injuries, passive or active shoulder pendulum exercises (anteflexion and rear extension) were initiated within 3 weeks. Functional exercises of the shoulder joint (anteflexion, rear extension, abduction, and lifting) were then gradually introduced.

Postoperative Patient Follow‐Up

Follow‐up visits were scheduled 1, 2, 3, 6, and 12 months after surgery. Fracture healing and shoulder joint function were evaluated at each visit, and subsequent treatment measures were determined according to exam findings. Follow‐up beyond 12 months after surgery was individualized according to the individual's state of recovery.

Outcome Measures

Patients were evaluated for the following indicators: (1) the possibility of a delayed surgical site infection >1 month after surgery; (2) fracture healing time and any collapse of the humeral head or other manifestations of necrosis; (3) screw loosening, fracture, or perforation of the humeral head and varus fracture displacement; (4) VAS and Constant scores of the shoulder joint 1, 3, 6, and 12 months after surgery; and (5) the VAS and Constant scores of the shoulder joint at final follow‐up (for evaluating shoulder function).

Statistical Analysis

All data were processed and analyzed using SPSS 22.0 (IBM, USA). Normally distributed measured data were presented as mean ± SD. Analysis of variance and rank sum tests was used for intergroup comparisons. A value of p < 0.05 was considered statistically significant.

Results

General Results

All operations were successfully performed, and the preoperative length of hospital stay was 5.62 ± 0.82 days, the operation time was 110–150 (average: 122.45 ± 5.35) min. Besides, the mean blood loss was 150.67 ± 9.78 mL (120–230 mL), All 19 patients were followed up for 1–4 (average: 2.13 ± 0.61) years. All fractures had healed during follow‐up, without signs of infection such as fever, incision‐site inflammation, exudate, or tenderness. All patients achieved osseous union, with a fracture healing time of 8–14 (average: 10.25 ± 1.72) weeks.

Complications

No cases of necrosis (including collapse of the humeral head) were observed. There was also no screw loosening, fracture, or perforation of the humeral head. Only one patient presented with mild varus displacement of the fracture and a reduced neck‐shaft angle, however, the patient was satisfied with the postoperative functional recovery and no special treatment was given.

Visual Analog Scale (VAS) Score

There were significant differences in VAS scores 1, 3, and 6 months after surgery (p < 0.05). There was no significant difference in the VAS scores between 6 and 12 months after surgery (p > 0.05). At the last follow‐up, all the 19 patients had no pain affecting sleep, and none had their daily life impacted. AS score of all patients reached 0 (Table 1).

Table 1.

Comparison of shoulder joint visual analog scale (VAS) scores and Constant scores over the course of postoperative follow‐up (n = 19, $\overline{\mathrm{x}}$ ± s).

Score	1 month	3 month	6 month	12 month
VAS score	2.09 ± 0.73	0.85 ± 0.13	0.08 ± 0.01	0.00 ± 0.00
Constant score	70.97 ± 4.63	80.12 ± 5.78	86.73 ± 7.46	87.58 ± 7.12

Note: All data are presented as mean ± SD (n = 19). For both VAS score and Constant score, there were significant differences in 1, 3, and 6 months after surgery (p < 0.05). There was no significant difference in the between 6 and 12 months after surgery (p > 0.05).

Abbreviation: VAS, Visual analog scale.

Constant Score

There were significant differences in Constant scores 1, 3, and 6 months after surgery (p < 0.05). There was no significant difference in the Constant score measured 6 and 12 months after surgery (p > 0.05). At the last follow‐up, 19 cases of shoulder elevation was 100°–150°, reaching 121.25° ± 7.34°on average; external rotation was 50–70°, with an average of 58.12° ± 4.13; internal rotation was 40–60°, with 53.45° ± 2.41° on average (Table 1).

At final follow‐up, the VAS score was 0 in all cases and the Constant score of the shoulder joint was excellent, good, fair, and poor in two, 12, four, and one case, respectively (excellent and good score rate of 73.68%).

Typical Cases

Case 1

An 87‐year‐old female patient was admitted for right shoulder pain and limited mobility due to low‐energy trauma. Imaging revealed a fracture of her right proximal humerus. Her T score was < −3.6 SD. She was diagnosed with an osteoporotic fracture of the right proximal humerus (a three‐part fracture as defined by the Neer classification). The surgical plan was for open reduction of the right proximal humeral fracture followed by internal fixation with cement‐augmented screws and a PHILOS plate. Three cement‐augmented screws were placed in a triangular fashion. The patient recovered well and achieved osseous union. At her last follow‐up, her VAS score was 0 and her Constant score of the shoulder joint was ranked as excellent. Her imaging and functional performance data are shown in Figure 2A–I.

FIGURE 2.

A 87‐year‐old female patient with right proximal humerus fracture (a three‐part fracture as defined by the Neer classification) caused by low‐energy trauma. (A, B) X‐rays and CT showed a fracture of the right proximal humerus. (C, D) The X‐ray films of the AP view, showed the right proximal humerus fracture immediately after open reduction with cannulated bone cement‐augmented screw combined with PHILOS plate internal fixation. Satisfactory fracture reduction and internal fixation were obtained, without evidence of bone cement leakage. (E, F) X‐ray re‐examination 12 months after surgery. There was no evidence of necrosis, including collapse of the humeral head. There was also no screw loosening or perforation of the humeral head. (G, I) functional observation of the shoulder joint 4 years after surgery.

Case 2

A 78‐year‐old female patient was admitted with left shoulder pain and limited mobility due to low‐energy trauma. Imaging revealed a fracture of her left proximal humerus. Her T score was < −3.1 SD. The patient was diagnosed with an osteoporotic fracture of her left proximal humerus (a four‐part fracture as defined by the Neer classification). She underwent open reduction of her left proximal humerus fracture with internal fixation using cement‐augmented screws and PHILOS plating. A rotator cuff injury was repaired during surgery and three cement‐augmented screws distributed in a triangular fashion were used for fixation. The patient recovered well and achieved osseous union. At her last follow‐up, her VAS score was 0 and her Constant score of the shoulder joint was ranked as excellent. Her imaging and functional performance data are shown in Figure 3A–I.

FIGURE 3.

A 78‐year‐old female patient with left proximal humerus fracture (a four‐part fracture as defined by the Neer classification) caused by low‐energy trauma. (A, B) X‐rays and CT showed a fracture of the left proximal humerus. (C, D) X‐ray observation of the left proximal humeral fracture immediately after open reduction with cannulated bone cement‐augmented screws combined with PHILOS plate internal fixation. Satisfactory fracture reduction and internal fixation were obtained, without bone cement leakage. (E, F) X‐ray re‐examination 24 months after surgery. There was no evidence of necrosis, including collapse of the humeral head, screw loosening, perforation of the humeral head. (G–I) Functional observation of the shoulder joint 24 months after surgery.

Discussion

In this clinical study, the use of cement‐augmented screws with PHILOS plating in the treatment of osteoporotic proximal humeral fractures had good clinical results through mid‐ and long‐term follow‐up. Over 1–4 years of follow‐up, all 19 patients did not develop an infection. All fractures healed without screw loosening, fracture, or penetration of the humeral head, and no osteonecrosis, such as humeral head collapse, occurred. Postoperative functional recovery was relatively slow in all of the older patients. There was still room for improvement in their shoulder function within 6 months after surgery, but recovery plateaued more than 6 months postoperatively. All 19 patients had a pain‐free shoulder joint after surgery, and the excellent rate of shoulder Constant score was 73.68%. No patients required postoperative fracture revision or a joint replacement.

Osteoporotic fractures of the proximal humerus in the elderly are a clinical challenge. Treatment options include conservative treatment, open reduction and internal fixation, and joint replacement. These options remain controversial. ¹³ Kmithof et al. ¹⁴ found that most patients with proximal humeral fractures who are managed conservatively achieve good long‐term functional recovery and quality of life. However, the long‐term immobilization required for conservative treatment precludes early functional exercise and may lead to shoulder stiffness. This may negatively impact patient quality of life, especially among those elderly patients with limited lower limb mobility. Surgery may be a better alternative in patients with unstable fractures, those who are unable to tolerate long‐term immobilization, those requiring early functional exercises, or those with glenohumeral mismatch after a fracture. For those patients with proximal humeral fractures who are indicated for surgery, open reduction and internal fixation is generally considered the treatment of choice. Doshi et al. ¹⁵ suggested that open reduction and internal fixation with a PHILOS plate and screws offers good clinical results. The results of this clinical trial indicate that cement‐augmented screws combined with PHILOS plating can achieve good clinical results in elderly patients with osteoporotic proximal humeral fractures requiring surgery.

Some authors ¹⁶ , ¹⁷ believe that open reduction and internal fixation of proximal humeral fractures are associated with a high possibility of complications; three‐ and four‐part osteoporotic proximal humeral fractures in particular have a high rate of non‐union, humeral head varus, and humeral head necrosis following open reduction and internal fixation. Some clinicians use a fibular graft for medial support during internal fixation of the humeral head or shoulder joint replacement for these fractures. Shu et al. ¹⁸ used the PHILOS plate combined with fibular support to repair proximal humeral fractures with a medial calcar fracture, reporting a lower complication rate compared with internal fixation and good clinical outcomes. Furthermore, the clinical use of autologous fibular grafts is considerably limited due in elderly patients with osteoporotic proximal humeral fractures due to their poor bone quality and the substantial surgical trauma involved during harvest. However, fibular allograft carries a risk of rejection and infection, which increases treatment costs. Joint replacement is required in elderly patients with osteoporotic distal humeral fractures if humeral head necrosis occurs after internal fixation. A joint prosthesis is difficult or impossible to implant in patients with fibular grafts because the fibular graft occupies the medullary cavity of the humerus. Fibular grafts therefore cannot be widely used in clinical practice and additional solutions need to be found. Our findings indicate that in elderly patients with three‐ or four‐part osteoporotic fractures of the proximal humerus, fracture healing can be achieved with satisfactory clinical results through open reduction with internal fixation using cement‐augmented screws combined with PHILOS plating, with a lower incidence of non‐union, humeral head inversion, and humeral head necrosis.

Application of Cement‐Augmented Screw Technology

A new approach for managing osteoporotic proximal humeral fractures involving cannulated cement‐augmented screws combined with locking plates has been developed and applied clinically. This technique is advantageous because trabecular osteoporosis can reduce the holding force of screws and impact the integration force between the screws and bone surface. This technique offers good clinical outcomes because it can increase the strength of the screws and the bone surface, reduce the incidence of fixation failure including screw loosening and pull‐out, and reduce the possibility of humeral head collapse, necrosis, and varus displacement. Bone cement is directly injected into the screw tip through the screw cannula. Injected bone cement can then penetrate between the cancellous bone trabeculae through a hole at the screw tip and the lateral holes to strengthen the connection between the screws, the bone cement, and the bone interface. A stable complex similar to a tree root forms between them, allowing the screw to be more firmly anchored to the humeral head and thereby increasing the stability of the screw. Siebenbürger et al. ¹⁹ compared a proximal humerus locking plate combined with cement‐reinforced screws with common locking screws in the treatment of proximal humerus fractures in the elderly, reporting that the use of cement‐reinforced screws significantly reduced the rate of postoperative loss of reduction, humeral head necrosis, and internal fixation failure. Foruria et al. ²⁰ found that the cement‐enhanced screws combined with proximal humerus locking plates in the treatment of proximal humerus fractures in the elderly significantly reduced the failure rate of internal fixation. Cement augmentation can effectively increase the primary stability of the internal fixation of osteoporotic proximal humeral fractures. Cement screw augmentation therefore appears to be a safe and effective approach for treating osteoporotic proximal humeral fractures, and is worthy of more widespread implementation. Röderer et al. ²¹ used screw augmentation technology for locking plate fixation of osteoporotic proximal humeral fractures in vitro in their biomechanical analysis. They found that the use of bone cement‐augmented screws can effectively hinder varus displacement of the humeral head. Further, the biomechanical performance of fixation with two bone cement‐augmented screws was comparable to that of four ordinary screws. Van Veelen et al. ¹⁰ used cannulated cement‐augmented screws to stabilize osteoporotic proximal humeral fractures, finding that cement augmentation reduced the incidence of internal fixation failure, including screw loosening and pull‐out, and lowered the possibility of bone cement reaction and leakage. Helfen et al. ²² compared cement‐enhanced screws combined with a PHILOS plate and a multiloc intramedullary nail in the treatment of two‐part proximal humerus fractures in the elderly, reporting similar clinical results to the present work and satisfactory clinical outcomes at 2‐year follow‐up. Although cannulated cement‐augmented screw technology has been used clinically, there are few reports on its long‐term outcomes. In 2017, our team used cannulated cement‐augmented screw fixation with locking plates in the treatment of osteoporotic proximal humeral fractures in the elderly for the first time. By 2021, 19 patients had been successfully treated with satisfactory clinical outcomes over 1–4‐year follow‐up. Such mid‐ and long‐term outcomes may provide the clinical basis for the widespread use of this technology.

Clinical Follow‐Up Analysis

All of the 19 patients who were included in this study followed up for 1–4 years. None of the patients developed any symptom of an infection, all received regular anti‐osteoporosis treatment, and all achieved osseous union after surgery. There was no screw loosening or breakage or perforation of the humeral head, indicating that cannulated cement‐augmented screws offer good anchorage to the surrounding bone. Cement augmentation increases screw stability and holding force and provides additional support to the humeral head. Bone cement does not need to be injected through the fracture end. This reduces damage to the soft tissues and bone of the humeral head. All patients achieved osseous union with no delayed union, justifying the reduced incidence of screw breakage. One patient had mild varus fracture displacement and a reduced neck‐shaft angle. This may have been due to the lack of effective support for the comminuted humeral calcar. There were significant differences in the shoulder joint VAS and Constant scores at 1, 3, and 6 months after operation, Shoulder function recovered gradually, it suggests that there is still room for improvement of shoulder joint function in elderly patients within 6 months after surgery, which may be related to degeneration of shoulder ligaments and bone in elderly patients, as well as the affordability and compliance of postoperative rehabilitation in elderly patients, making postoperative functional recovery slower. But plateaued 6 to 12 months after surgery, it indicated that the shoulder joint function of the patients did not improve significantly after 6 months. The function of the who had a longer follow‐up period of up to 4 years postoperatively remained similar to that measured 6 months after surgery. This suggests that functional recovery after shoulder surgery is slow for elderly patients. This may be related to bone and ligament degeneration in the shoulder and to poor tolerance and compliance with postoperative rehabilitation among elderly patients. The excellent and good patient‐reported outcome rate after surgery was only 73.68%. This rate may be due to shoulder joint degeneration, fractures combined with rotator cuff injuries, postoperative functional exercises, and osteoporosis in elderly patients. Although the excellent and good outcome score rate was not high, postoperative satisfaction was high in these patients.

Technical Recommendations

For elderly patients with Neer three‐ and four‐part osteoporotic fractures of the proximal humerus, certain authors recommend shoulder replacement. However, there is no unified consensus on the relative clinical efficacy of this strategies. If a reconstruction is possible, it is important to reconstruct fracture integrity and restore shoulder joint function as much as possible. The opportunity for internal fixation may therefore be available for some patients who may otherwise require a shoulder replacement. In a previous study, we found that cannulated cement‐augmented screw fixation combined with locking plates offers similar clinical efficacy to humeral head replacement. For proximal humerus fractures with non‐supported calcar fractures, level D or E cement‐augmented screws can be used for augmentation to reduce the incidence of postoperative complications such as humeral head varus and fracture non‐union. This method may be an effective alternative for patients who would otherwise have required medial support with fibular grafts. However, it is not recommended for young patients or those without osteoporosis. Further, a shoulder replacement is still recommended for patients with osteoporotic fractures of the proximal humerus that cannot be reconstructed or who have a higher likelihood of expected complications after fracture fixation. Angiography is required before bone cement injection to reduce intraoperative bone cement leakage. However, in this study, to reduce intraoperative blood loss, reduce the number of fluoroscopy, and shorten the operation time, no angiography was used during surgery in any of the 19 patients, and no cement leakage occurred. To prevent cement leakage, our experience offers certain valuable insights. First, the configuration of the augmented screws should be kept away from the fracture line, the humeral head should not be penetrated by the electric drill, and the bone cement should be injected slowly under intraoperative fluoroscopy. Bone cement should be injected while it is in a liquid state. This allows the bone cement to exit distal end of the screw through the screw cannula, permitting a cloud‐shaped configuration. Second, it is necessary to closely observe the bone cement under C‐arm fluoroscopy to prevent leakage into the shoulder joint space, especially in patients with fractures of the humeral head. Third, three (in a triangular distribution) or four (in a quadrangular distribution) cannulated cement‐augmented screws are generally selected depending on fracture anatomy. Our findings indicate that both configurations can yield good stability and effective support. Fourth, the greater and lesser tubercles or rotator cuff tissue may be suture repaired during surgery if necessary. This plays an important role in postoperative functional recovery. Finally, regular anti‐osteoporosis treatment and follow‐up are important to achieving good postoperative functional recovery.

Limitations

There were some limitations to this study. First, it included a small number of cases and had a relatively short follow‐up time. The mid‐ and long‐term follow‐up conclusions drawn from this cohort may therefore be biased and require further validation prior to clinical implementation. Statistical analysis of a large number of cases is warranted to support the conclusions of this study. Second, it is difficult to measure long‐term follow‐up in this cohort as some patients were aged over 90 years. Subsequent studies need to enroll more patients who are suitable for our treatment strategy, pursue a longer follow‐up period, and produce relatively reliable conclusions. Such works may provide more effective guidance for clinical decision‐making.

Conclusions

This study describes an internal fixation method for osteoporotic proximal humeral fractures in elderly patients. This internal fixation technique uses bone cement to strengthen the tip of the screw in the humeral head. The bone cement provides excellent anchorage of the screw to the osteoporotic bone and also provides supplemental support to the humeral head, especially for Neer three‐ or four‐part osteoporotic proximal humeral fractures. This strategy resulted in a reduced incidence of fracture non‐union, humeral head inversion, and humeral head necrosis compared with internal fixation alone, and functional recovery was satisfactory for all patients. No complications of internal fixation failures or humeral head necrosis were noted, and good functional recovery of the shoulder joint was achieved. Cement‐augmented screws combined with PHILOS plating can achieve good mid‐ and long‐term clinical efficacy in the treatment of osteoporotic proximal humeral fractures, providing a new option for the treatment of such patients. This method can be promoted in clinical practice.

Author Contributions

She Rongfeng: Research design and implementation, data organization and statistical analysis, paper writing, and research funding support. Zhang Bin, Jiang Kundou, Yang Shuaiqi: Case collection and patient follow‐up; Zhang Yi: Research design and implementation, research guidance, paper revision, and research funding support.

Funding Information

Supported by the Guizhou Provincial Science and Technology Projects (Qian ke he cheng guo‐LC[2021]008). Supported by the Guizhou Provincial Science and Technology Projects (Qian ke he ji chu[2019]1210; GZWKJ2021‐253).

Conflict of Interest Statement

The authors declare no conflicts of interest.

Ethics Statement

This study was approved by the review board of Guizhou Provincial People's Hospital, China (Review No. 2022‐69).