Massive Humeral Bone Loss



Massive Humeral Bone Loss


Bradley S. Schoch, MD

Jean-David Werthel, MD



INTRODUCTION

Humeral bone loss represents a challenging problem for shoulder surgeons, especially in the setting of arthroplasty reconstruction. Bone loss can occur secondary to trauma, prior surgery, infection, osteolysis, or oncologic destruction. Newer humeral implants for shoulder arthroplasty have trended toward uncemented humeral fixation with certain stem designs causing stress shielding and loss of lateral humeral or calcar bone. Removal of these well-fixed uncemented stems can lead to iatrogenic humeral bone loss with the associated destruction of both cortical and cancellous bone. In rare cases, a humeral osteotomy may even be required.1 Loss of the proximal humerus architecture can then lead to loss of function and glenohumeral instability.

There are many techniques available to help restore or replace humeral bone stock and restore shoulder function for patients who present with massive humeral bone loss. However, understanding the pattern of bone loss and treatment options available are vital to being adequately prepared for this complex revision surgery. This chapter is designed to describe the classification of humeral bone loss, review the reconstructive treatment options for shoulder arthroplasty in the setting of massive humeral bone loss, and outline tendon transfer techniques that may assist in improving functional outcomes.


CLASSIFICATION

Historically, classification of proximal humeral bone loss has been described in the setting on oncologic resection. Multiple systems have been described, often in relation to sacrificed anatomic structures and the type of oncologic resection performed. In the setting of the shoulder girdle, Malawer et al2 proposed a classification system for shoulder-girdle resections in patients undergoing limb-sparing procedures. Bone loss following oncologic resections can be very similar to the bone loss encountered in patients undergoing revision arthroplasty. However, of the six resection types described, only one involves the humerus in isolation (Type I). It is then subdivided into Type 1A (preservation of the deltoid insertion and/or rotator cuff) and Type 2B (resection of the deltoid tuberosity or rotator cuff). With this classification system, it is difficult to describe the variations of proximal humeral bone loss which can be treated with humeral implants ranging from standard primary components to a total humerus.

Furthermore, in the setting of revision arthroplasty, oncologic classification systems fail to take into account the quality of remaining bone, which has significant clinical implications when considering reconstructive techniques. Given the limitations with previous classification systems, Chalmers et al3 proposed a classification system for humeral bone loss in the setting of revision shoulder arthroplasty (FIGURE 45.1). This has been termed the Proximal Humeral Arthroplasty Revision Osseous inSufficiency (PHAROS) classification. The authors designed this system around both the presence and absence of critical bony structures (greater tuberosity, calcar, deltoid insertion) as well as the quality of the remaining cortical bone. The proximal humerus was then divided into three sections: epiphysis (tuberosities and calcar bone), metadiaphyseal bone (between the epiphysis and deltoid insertion), diaphysis (below the deltoid insertion). Type 1 epiphyseal lesions were subdivided into deficiencies at the calcar (Type 1C) and greater tuberosity (Type 1G). The clinical photo in FIGURE 45.2 is an example of a Type 1C defect. Both metadiaphyseal (Type 2) and diaphyseal (Type 3) compromise was further classified as thinning of greater than 50% of the cortex (Type 2A or 3A) or frank loss of cortical bone (Type 2B or 3B).


IMPLANT OPTIONS

When performing revision shoulder arthroplasty, surgeons must be aware of both existing and potential iatrogenic bone loss, which can occur at the time of surgery. Prior to revision surgery, surgeons should consider the level of cementation (when present), the on-growth properties and metaphyseal filling of the humeral stem to be removed, and cortical bone quality. During extraction, metaphyseal bone loss can be anticipated. Iatrogenic bone loss patterns differ based on both the implanted stem and extraction techniques. Depending on the quality of cortical bone, surgeons may be able to anticipate potential cortical bone loss at the time of stem extraction. Prior to stem
extraction, it is important to measure the length of the humerus by measuring from the humeral calcar to a structure which is likely to be preserved after humeral component explant or bone resection. This measurement can then be used during the reconstruction to ensure restoration of appropriate humeral length. Depending on the extent of anticipated bone loss, surgeons must be adequately prepared at the time of surgery, which often may involve having multiple implant options available.


















Primary Arthroplasty Stems

Primary arthroplasty has historically been performed with humeral stems that require both canal reaming and metaphyseal broaching to achieve rotational stability. However, new generations of humeral implants have employed shorter stems and broach-only techniques in an effort to preserve humeral bone stock and prevent stress shielding.4

The use of primary implants at the time of revision surgery is possible in the setting of Type 1 deficits, where the metadiaphyseal bone remains intact.

Caution should be taken when using these techniques with significant Type 1G defects, where the greater tuberosity is absent. In a study of 420 patients undergoing reverse total shoulder arthroplasty (RTSA) for proximal humerus fractures, Ohl et al compared a group of patients with iatrogenic bone loss (tuberosity excision) to a group with healed or malunited/resorbed/nonunited greater tuberosities. Postoperative instability was significantly higher in shoulders with resected tuberosities (12.5%).5 This is likely due to loss of the deltoid wrapping angle and the rotator cuff deficiency, which both provide a compressive force across the glenohumeral joint and reduce postoperative instability.6 Therefore, surgeons should be cautious when using standard medialized glenoid/medialized humerus7 RTSA implants in the setting of greater tuberosity bone loss. In the future, metal augmentation of the greater tuberosity will be available in conjunction with standard stem fixation options (FIGURE 45.3).

Fixation of standard stems in proximal humeri with Type 1 bony deficits can be performed in both uncemented and cemented fashion, depending on the system selected.8 It is important to achieve rotational stability regardless of fixation type. When using cement, this can be performed using traditional techniques or cementing into a previous cement mantle at the time of revision when no infection is suspected (cement-within-cement technique).9 When cancellous bone is compromised in the metaphyseal region, rotational stability can also be obtained with impaction grafting of auto- or allograft bone.10 This technique has been used successfully in the setting of revision surgery, where bone loss is isolated to the epiphyseal/metaphyseal cancellous bone. Lastly, rotational stability can be achieved by simply increasing the size of the humeral stem until adequate cortical contact is achieved at the metaphyseal portion of the implant. However, fixation using this technique has led to significant stress shielding and longer term loss of humeral bone in the area of cortical contact.11 (FIGURE 45.4).






Some surgeons have also attempted to treat Type 2a bone defects with standard stems. In a study of 32 revision RTSA with (16) and without (16) proximal humeral bone loss, Stephens et al12 documented similar improvements in pain and patient-reported outcome measures. However, in those with proximal humeral bone loss, patients demonstrated significantly less forward elevation (100° vs 135°, P = 0.022) and external rotation (19° vs 34°, P = 0.009). Furthermore, the authors noted
higher rates of aseptic humeral loosening in shoulders with proximal humeral bone loss (19% vs 0).12







Long-Stem Monoblock

In the setting of bone loss proximal to the deltoid insertion (Type 1 and Type 2) or with smaller defects distally, arthroplasty can be performed with long-stemmed monoblock humeral components. Similar to reconstruction with standard stems, consideration should be given to greater tuberosity bone loss and the deltoid wrapping angle. When reconstructing with long-stemmed humeral components, these are often cemented into the remaining diaphyseal bone. Cementing can be performed the entire length of the stem or only at the most proximal aspect of the bone-implant interface. During preparation, care must be taken to avoid a distal cortical perforation.13 It is important to remember that these long-stemmed components are straight and do not accommodate for the natural bow of the humerus. Intraoperative fluoroscopy may be useful during humeral preparation for these implants to ensure that diaphyseal reaming remains within the intramedullary canal.

In addition to intraoperative complications, previous reports on these prostheses have raised concern about high rates of aseptic humeral loosening. In a study of 124 revision shoulder arthroplasties performed with long-stem cemented humeral components, the authors noted a 10% rate of aseptic humeral loosening.14 The failure rates with this prosthesis are likely related to the high torsional loads, which occur with use of the forearm during daily activity. Following reconstruction in the setting of humeral bone loss, humeral implants experience increased rotational micromotion, which may lead to failure at the bone cement interface.15 This method of failure likely explains the higher rates of aseptic loosening when standard humeral stems are used to treated proximal humeral bone loss in revision RTSA.12 For these reasons, we prefer not to use long-stemmed components in isolation in the setting of proximal humeral bone loss, which involves complete loss of the greater tuberosity.


Allograft Prosthetic Composite

In cases of proximal humeral bone loss exceeding 5 cm, reconstruction using standard or long-stem implants is not possible even with the use of eccentric glenospheres or augmented humeral trays. One option for this more distal bone loss is reconstruction with an allograft prosthetic composite (APC). This has historically been used for Type 2 and Type 3 bone loss (FIGURE 45.5). This procedure involves removing proximal poor quality bone and replacing the bony defect with a frozen allograft. Our general preference is to use a size-matched proximal humerus allograft, but when unavailable, alternative grafts such as a distal tibia can be utilized. A standard or long-stemmed humeral component is then placed into/through the allograft bone and then mated with the remaining native bone. Fixation into the host bone has been described using both cemented and uncemented techniques. The junction between the allograft and native bone can be supplemented with plate fixation and local bone grafting.


Previous authors have recommended reconstruction with an APC when bone loss exceeds 5 cm from the native humeral calcar. However, this remains as level 5 evidence (expert opinion) based on surgical experience.16,17 Advantages of APC reconstruction include restoration of bone stock, reproduction of the deltoid wrapping angle, and ability to repair soft tissues to the allograft with the potential for biologic healing. In a study of 14 APCs undergoing revision surgery, all or a portion of the allograft was able to be retained in 64% of cases.18

Multiple studies have reported clinical success with APCs used in both the hemiarthroplasty and RTSA configurations. However, long-term success of shoulder reconstruction with an APC hemiarthroplasty remains unclear. In a study of 21 APC hemiarthroplasties following tumor resection, El Beaino et al reported 48% of patients showed evidence of superior subluxation (>1 cm) at 1-year follow-up despite reattachment of the rotator cuff musculature.19 The authors do not report functional outcomes, but the presence of superior subluxation is concerning for failure of the rotator cuff repair, which would place the patient at high risk for meaningful loss of shoulder function.20 In addition to rotator cuff failure, resections that sacrifice the deltoid insertion have also been shown to have poorer range of motion and function following APC reconstruction with a hemiarthroplasty.21

Because of concern about the long-term function of hemiarthroplasty with rotator cuff repair, APC reconstruction with a reverse shoulder configuration has become increasingly popular.22,23 Sanchez-Sotelo et al24 reported on 26 patients treated with reverse APC reconstructions for massive proximal humeral bone loss. At a mean follow-up of 4 years, mean forward elevation had improved from 41° to 98° (P < 0.0001). Patients also demonstrated acceptable function, with a mean American Shoulder and Elbow Surgeons (ASES) functional score of 66.1. Implant survivorship at 5 years was 92%.

Despite successes described with APC reconstruction, surgeons should expect a higher level of complications following these surgeries compared to primary arthroplasty. Long-term success of this construct depends on the junction of the host bone to the humeral allograft. Compression plating has become increasingly popular to secure the APC construct to host bone to improve healing rates.19,24 Described techniques to increase bony union include transverse cuts with compression plating, step-cut mating with cable and/or plate fixation, and dome osteotomy with compression plating.24,25,26 However, nonunion at the junction of the host/allograft remains the most common complication.21 Other complications include aseptic loosening, instability, periprosthetic fracture, graft osteolysis, and infection.


Modular Reconstruction Prostheses

As humeral bone loss progresses to involve the deltoid insertion, substantial structural loss of the proximal humerus architecture is present. As bone loss increases, reconstruction with standard components becomes impossible, and long-stem components are at increased risk of mechanical failure. Based on the concern with traditional components, modular reconstruction prostheses have been developed, which allow for replacement of humeral bone stock with metallic components. These devices traditionally involve fixation of a humeral stem into the remaining diaphysis, with multiple-sized proximal body attachments, which can be used to reconstruct the proximal humeral bone loss. Attachment sites for soft tissues are often incorporated into the design in an attempt to restore function. However, long-term stability of the remaining repaired soft tissues to hydroxyapatite-coated metal components remains in question. Early reports on the use of these types of prosthesis have shown encouraging results.27,28

Similar to APC reconstruction, modular reconstruction prostheses can be implemented in both a hemiarthroplasty or RTSA configuration. When used for tumor surgery, these are often used as an RTSA, as long as the deltoid remains intact and functional.27 In the setting of revision surgery, the amount of remaining glenoid bone stock may dictate whether or not a glenosphere can be securely placed. Strategies to restore humeral length and offset are important. Soft tissue repairs and/or transfers may also help to minimize postoperative dead space.

Only gold members can continue reading. Log In or Register to continue

Jun 23, 2022 | Posted by in ORTHOPEDIC | Comments Off on Massive Humeral Bone Loss
Premium Wordpress Themes by UFO Themes