Generic selectors
Exact matches only
Search in title
Search in content
Post Type Selectors
Search in posts
Search in pages
Filter by Categories
Case Report
Clinical research study
Current Issue
Editorial Board
Literature Review
Narrative review
Original Article
Research Article
Review Article
Short Report
Surgical techniques
Generic selectors
Exact matches only
Search in title
Search in content
Post Type Selectors
Search in posts
Search in pages
Filter by Categories
Case Report
Clinical research study
Current Issue
Editorial Board
Literature Review
Narrative review
Original Article
Research Article
Review Article
Short Report
Surgical techniques
View/Download PDF

Translate this page into:

59 (); 97-105
doi:
10.1016/j.jor.2024.07.017

Bone grafting augmentation choices in complex proximal humerus fractures: A systematic review

, , , , , ,
a Division of Orthopaedic Surgery, University of Toronto, Toronto, ON, Canada
b Faculty of Medicine, University of Toronto, Toronto, ON, Canada
c Rothman Orthopaedic Institute, NY, NY, USA
d Sunnybrook Orthopaedic Upper Limb (SOUL), Division of Orthopaedic Surgery, Sunnybrook Health Sciences Centre, Toronto, ON, Canada

⁎Corresponding author: Ali Etemad-Rezaie. ali.etemad@mail.utoronto.ca

Received: , Accepted: ,
The Authors
Disclaimer:
This article was originally published by Reed Elsevier India Pvt. Ltd. and was migrated to Scientific Scholar after the change of Publisher.

Abstract

Abstract

To systematically identify and evaluate different bone graft augmentation techniques in the operative treatment of complex proximal humerus fractures.

Four databases were searched from 1970 to February 2023 for Level I to IV English studies that investigated outcomes of different bone augments in the primary surgical fixation of proximal humerus fractures. The JBI critical appraisal checklist, methodological index for non-randomized studies and cochrane risk of bias tool were used to assess study quality. Descriptive statistics including weighted means are presented where applicable.

Thirty-three articles including 964 patients met the inclusion. Seven bone augments were identified, including fibular strut allograft (693 patients across 21 studies), femoral head allograft (84 patients across 4 studies), iliac crest allograft (54 patients across 3 studies), iliac crest autograft (94 patients across 5 studies), humeral endosteal allograft (6 patients in 1 single study), unspecified cancellous allograft (28 patients in 1 single study) and distal clavicle autograft (3 patients in 1 single study). Mean patient age was 67.1 years, with female patients comprising 65.2 %. Fracture union rates were similar between groups, with an average of 99.6 %. The average Constant Murley Score (CMS) was not reported in the humeral endosteal allograft or the distal clavicle autograft group but was 81.8 (fibular strut allograft), 79.1 (femoral head allograft), 76.8 (iliac crest allograft), 77.7 (iliac crest autograft), and 81.5 (unspecified cancellous allograft) in the remaining groups. Revision surgery was required in 16.7 % of patients receiving humeral endosteal allograft, 7 % of patients with femoral head allograft, 2 % of iliac crest autografts and 1.9 % in the fibular allograft group. Reported complications included avascular necrosis, hardware complications and loss of reduction.

Bone graft augmentation is an effective adjunct to open reduction internal fixation of complex proximal humerus fractures. Fibular strut allograft is the most common bone graft augment used. Majority of patients treated with bone graft augmentation achieved bony union (83%–100 %) and average CMS scores at final follow-up were similar between graft types (76–82). However, no conclusive data suggests superiority of one bone graft type over another. Future studies should aim to compare the outcomes amongst graft types.

Keywords

Autograft
Allograft
Shoulder
Open reduction internal fixation
Osteoporotic
1

1 Introduction

Proximal humerus fractures are the third most common fractures in the elderly.1 These fractures often result from low energy mechanical falls.2 The overall high incidence of these fractures is largely attributable to osteoporosis and poor bone quality seen with increasing age.3 These fractures account for over 150,000 visits to the emergency department in the United States per year significantly affect health and economic burden.4

Despite their high incidence in the clinical setting, the treatment of displaced proximal humerus fractures remains challenging and variable.5 Treatment options are varied and can range from non-operative treatment in a cuff and collar to operative interventions such as open reduction internal fixation (ORIF), humeral intramedullary nailing (IMN), hemiarthroplasty (HA) or reverse total shoulder arthroplasty (RTSA).5 Regardless of the treatment option, many patients continue to suffer poor clinical outcomes with a decrease in their functional outcomes, range of motion and quality of life.3

Currently, ORIF with the use of proximal humeral locking plate remains the most common surgical management used to treat displaced proximal humerus fractures.6 However, given the inherent poor bone stock and quality associated with these fractures, particularly in the elderly, inferior outcomes and complications remain a significant source of surgeon frustration and patient morbidity.7 The complications associated with proximal humerus fractures include intra-articular screw penetration, varus collapse, avascular necrosis (AVN), screw cutout and loss of reduction.7 In locking plate fixation of these fractures without bone graft augmentation, the rate of AVN often quoted in the literature ranges anywhere from 3 to 20 % and has a positive correlation with number of fracture fragments and their displacement, as well as increasing patient age.8,9 A recent randomized controlled trial of Neer 2-, 3- and 4-part proximal humerus fractures demonstrated AVN rates of 4.9 % and 2.6 % for locking plate fixation with and without fibular allograft augmentation, respectively.10 Bone graft augmentation of the proximal humerus either with allograft or autograft has been shown to restore the medial column and minimize complications such as varus collapse and loss of reduction.11–13 The different allograft options for bone grafting are fibular strut, femoral head, iliac crest, humeral and cancellous allografts.14 The most common autograft used is tricortical iliac crest bone graft, but recently distal clavicle has been reported.15

To date, there has been no systematic review evaluating and comparing the outcomes and complications associated with different types of bone graft augmentation used during ORIF of proximal humerus fractures. There is no consensus on optimal bone grafting technique or superior bone graft option. This systematic review aims to identify all the bone grating options available and evaluate their outcomes. We aim to analyze the clinical and radiographic outcomes along with complications associated with various bone grafting options used during the surgical fixation of proximal humerus fractures. The goal is to help surgeons optimize outcomes when surgically treating complex proximal humerus fractures and to identify gaps in the literature on this sparsely studied topic to guide future studies.

2

2 Materials and methods

2.1

2.1 Literature search

The Preferred Reporting Items for Systematic reviews and Meta-analyses (PRISMA) guidelines was used for this study.16 The following databases were searched from inception to February 2023: OVID MEDLINE, OVID Embase, Cochrane, and PubMed. Key words used in the search strategy include proximal humerus fracture, bone graft, autograft, allograft, shoulder, open reduction internal fixation, bone graft augmentation, and osteoporotic (Table S1). Lastly, the references of eligible studies were reviewed to identify any studies missing from the initial search.

2.2

2.2 Study screening and eligibility

Title and abstract screening were performed by two independent reviewers (SD, VPB) in duplicate using the Covidence online platform. Eligibility conflicts were resolved with a third reviewer. Studies that met the eligibility criteria were: (1) Participants involved patients (≥18 yrs) who underwent primary fixation of a proximal humerus fracture using bone graft augments; (2) Augmentation technique used included allografts or autografts; (3) Study types included case series, case reports, observational cohort studies, case-control studies and randomized controlled trials. Studies were excluded if they investigated revision fixation of proximal humerus fractures, pathologic proximal humerus fractures, fracture dislocations, fracture non-unions or if they were abstracts or reviews.

2.3

2.3 Data abstraction

Data was collected by three reviewers (SD, VPB, AER) and recorded in a Microsoft Excel spreadsheet (Version 2007, Microsoft, Redmond, WA, USA). Abstracted data included the authors, title, journal, year of publication, study design. Demographic data collected included sample sizes, sex ratio, mean age, mechanism of injury, fracture type, graft type and follow-up. Clinical outcomes collected include union percentage, time to union, complications, range of motion (ROM), patient reported outcome scores (PROMs). Radiological outcomes included change in humeral head height (HHH) and neck-shaft angle (NSA).

2.4

2.4 Statistical analysis

Performing a meta-analysis was precluded given the high statistical and methodological heterogeneity. Thus, descriptive analysis was utilized in the form of weighted means to average demographic data and clinical outcome scores based on graft types. Data from all included studies were also pooled to calculate means for “all graft types”.

2.5

2.5 Risk of bias and study quality assessment

The methodological quality of the included non-randomized studies was assessed using the MINORS checklist. This checklist is designed to assess the methodological quality of comparative and non-comparative, non-randomized studies. The Cochrane Risk of Bias Tool was used to evaluate the bias risk in randomized controlled trials (RCT). Case reports were assessed using the Joanna Briggs Institute (JBI) Critical Appraisal checklist. For each bias and quality assessment tool mentioned, two independent reviewers appraised the quality of the included studies and conflicts were resolved by a third independent reviewer.

3

3 Results

3.1

3.1 Study characteristics

The initial search of the online databases resulted in 1048 total studies. A systematic screening and assessment of eligibility identified 33 full-text articles that satisfied the inclusion and exclusion criteria (Fig. 1). The agreement between reviewers, calculated as the kappa statistic, was substantial on title and abstract screening (κ = 0.743, SE 0.035) and near perfect for full text review (κ = 0.833, SE = 0.060). Eligible studies included 1 RCT, 15 cohort studies, 16 case series and 1 case report. Overall, studies reported seven unique bone augments including fibular strut allograft (21 studies), femoral head allograft (4 studies), iliac crest allograft (3 studies), iliac crest autograft (5 studies), humeral endosteal allograft (1 study), unspecified cancellous allograft (1 study) and distal clavicle autograft (1 study) (Table S2). A total of 964 included patients had a mean age of 67.1 years and 65.2 % were female. Table 1 provides a summary of study characteristics and demographic data of patients from each study. There was insufficient data reported on BMI, smoking history, and presence of osteoporosis to analyze these parameters.

A flow chart summarizing the literature search and screening results according to the PRISMA guidelines.
Fig. 1 A flow chart summarizing the literature search and screening results according to the PRISMA guidelines.
Table 1 Study characteristics and patient demographics for included studies.
Study Study Design Graft Type Sample Size Mean Age (yrs) Gender Distribution (%Males) Mean Follow-up (mo)
Avilucea 2020 Retrospective Case Series Fibular Strut Allograft 13 68 15.4 % 13.2
Chen 2015 Retrospective Cohort Fibular Strut Allograft 18 66.9 44.4 % 31.4
Chen 2018 Retrospective Cohort Fibular Strut Allograft 47 68.6 25.5 % 33.5
Cui 2019 Retrospective Cohort Fibular Strut Allograft 25 73.2 28.0 % 31.6
Gardner 2008 Retrospective Case Series Fibular Strut Allograft 7 62 71.4 % NR
Hinds 2015 Retrospective Cohort Fibular Strut Allograft 34 74 29.4 % NR
Kim 2018 Retrospective Cohort Fibular Strut Allograft 84 72 44.0 % 40
Kim 2020 Retrospective Cohort Fibular Strut Allograft 38 69.8 13.2 % 17.2
Lee 2019b Retrospective Cohort Fibular Strut Allograft 45 75.6 26.7 % 13.6
Berkes 2014 Retrospective Case Series Fibular Strut Allograft 11 64 54.5 % 14.6
Matassi 2012 Prospective Case Series Fibular Strut Allograft 17 62 41.2 % 28
Myers 2020 Retrospective Cohort Fibular Strut Allograft 61 62.3 26.2 % 7.8
Neviaser 2011 Retrospective Case Series Fibular Strut Allograft 38 655 NR 12
Tan 2014 Retrospective Case Series Fibular Strut Allograft 9 75.4 22.2 % 3
Tuerxun 2020 Retrospective Cohort Fibular Strut Allograft 41 64. I 29.3 % 19.3
Little 2014 Retrospective Case Series Fibular Strut Allograft 72 64 27.8 % 13.1
Zhao 2019 Retrospective Cohort Fibular Strut Allograft 21 688 42.9 % 12
wang 2019 Retrospective Cohort Fibular Strut Allograft 82 725 36.6 % 19
Sheng 2021 Retrospective Cohort Fibular Strut Allograft 27 64 33.3 % 15.3
Unspecified Cancelllous Allograft 28 64 46.4 % 13.8
wang 2013 Retrospective Case Series Fibular Strut Allograft 3 77 33.3 % NR
Humeral Endosteal Allograft 6 782 33.3 % NR
Chen 2023 Case Report Fibular Strut Allograft 1 60 100.0 % 10.2
Euler 2015 Retrospective Case Series Femoral Head Allograft 10 638 0.2 31.8
Karthik 2011 Prospective Case Series Femoral Head Allograft 12 65 33.3 % 29.7
Robinson 2010 Prospective Cohort Femoral Head Allograft 21 NR NR 24
Zhang 2019 Prospective Cohort Femoral Head Allograft 42 70.9 45.2 % 3 %
Lee 2019a Retrospective Case Series Iliac Crest Allograft 38 688 36.8 % 15.6
Atalar 2014 Retrospective Case Series Iliac Crest Allograft 9 565 NR 24.3
Atalar 2007 Prospective Case Series Iliac Crest Allograft 7 56.7 42.9 % 36.9
Iliac Crest Autograft 3 49 33.3 % 43.3
Kim 2012 Retrospective Case Series Iliac Crest Autograft 19 66.2 % 68.4 % 27.2
Zhu 2014 Prospective Cohort Iliac Crest Autograft 18 51 61.1 % 25.4
Shah 2022 Retrospective Case Series Iliac Crest Autograft 26 47 61.5 % 36.5
Hristov 2022 Retrospective Cohort Iliac Crest Autograft 28 645 10.7 % NR
De Mello Ribeiro Pinto 2022 Prospective Case Series Distal Clavicle Autograft 3 68.3 0.0 % 6.3
3.2

3.2 Risk of bias

MINORS Criteria, the Cochrane Risk of Bias Tool and the JBI Critical Appraisal Checklist were used to evaluate non-randomized, randomized and case report studies, respectively. After applying the MINORS criteria for non-randomized studies, 2 studies were excellent, 27 studies were fair, and 2 studies were poor (Fig. S1). The Cochrane Risk of Bias Tool for the RCT highlighted some bias in the included randomized controlled trial, particularly pertaining to the lack of blinding included in the study design (Fig. S2). The JBI Critical Appraisal Checklist deemed the included case report of sufficient quality to be included in the review (Fig. S3).

3.3

3.3 All graft types

After pooling data from patients of all graft types included in the study, the mean follow-up period was 20.4 months. The most frequent mechanism of injury was falls (77.4 %), followed by road accidents (15.4 %). Considering surgical methodology, Beach Chair positioning (70.9 %) and a deltopectoral approach (70.7 %) were most commonly used. Union was achieved in 99.6 % of all patients included in the study, while 2.2 % required a revision surgery. The overall complication rate was 10.3 %, with screw complications (3.0 %) and AVN (2.2 %) being the most frequent. For all included patients with any graft types, the mean CMS score was 80.9, HHH was 1.4 mm and NSA was 2.8°.

3.4

3.4 Fibular strut allograft

In the twenty-one studies17–36 evaluating fibular strut allograft, the total number of patients included was 694 with average age of 68.5 years old. 32.2 % were males and 67.8 % were females. The majority of fractures were due to falls (76.3 %). The fibular strut graft was placed in the metaphysis to provide additional medial hinge support. The mean follow-up time was 20.4 months. Union was achieved in 100 % of the fractures post-operatively. The overall complication rate was 9.0 %. The two most common complications were screw complications (i.e. screw cut-out, screw penetration) and AVN at 3.4 %, 3.0 % respectively. There were 1.9 % of patients who had to return to the operating room for revision surgery. The average CMS score was 81.8. Immediate post-operative to final follow-up difference in HHH and NSA was 1.3 mm and 2.6° respectively. Table 2 details the weighted average of demographic data and clinical outcomes of studies evaluating fibular strut allograft.

Table 2 Summary of data extracted from included studies reporting bone augmentation in the treatment of complex proximal humerus fractures, categorized by graft type. Patient demographics, and information on fracture, surgery and key outcomes are reported. Reported value is expressed as either weighted average [range of means] or number of patients (percentage). “# Patients” refers to the total number of patients that had the parameter reported. Weighted averages were calculated by combining data of each parameter from all graft types and are summarized in the final column. CMS, Constant Murley Score; HHH, humeral head height; NSA, neck shaft angle.
Fibular allograft Femoral head allograft Iliac crest allograft Iliac crest autograft Unspecified cancellous allograft Humeral endosteal allograft Distal clavicle autograft All graft types
# of included studies 21 4 3 5 1 1 1 33
Total # of patients 694 85 54 94 28 6 3 964
Value # of patients Value # of patients Value # of patients Value # of patients Value # of patients Value # of patients Value # of patients Value # of patients
Demographics Age (yrs) 68.5 694 68.7 64 65.2 54 56.9 94 64 28 78.2 6 68.3 3 67.1 943
Sex 656 64 45 94 28 6 3 896
Male 211 (32.2 %) 25 (39.1 %) 17 (37.8 %) 44 (46.8 %) 13 (46.4 %) 2 (33.3 %) 0 (0 %) 34.8 %
Female 445 (67.8 %) 39 (60.9 %) 28 (62.2 %) 50 (53.2 %) 15 (53.6 %) 4 (66.7 %) 3 (100 %) 65.2 %
Follow-up (mo) 20.4 650 15.3 85 19.8 54 31.1 66 13.8 28 NR – 6.3 3 20.4 886
Fracture Mechanism of injury 355 12 7 22 NR 6 NR 402
Falls 271 (76.3 %) 10 (83.3 %) 5 (71.4 %) 19 (86.4 %) NR 6 (100 %) NR 77.4 %
Road accident 55 (15.5 %) 2 (16.7 %) 2 (28.6 %) 3 (13.6 %) NR – NR 15.4 %
Sports-related 15 (4.2 %) – – – NR – NR 3.7 %
Low energy (nonspecified) 9 (2.5 %) – – – NR – NR 2.2 %
High energy (nonspecified) 5 (1.5 %) – – – NR – NR 1.2 %
Neer Classification 653 64 54 94 28 6 3 902
Types 4 258 (39.5 %) 16 (25 %) 18 (33.3 %) 49 (52 %) 12 (42.9 %) 0 (0 %) 2 (66.7 %) 39.4 %
Type 2 246 (37.7 %) 30 (46.9 %) 25 (46.3 %) 34 (36 %) 16 (57.1 %) 2 (33.3 %) 1 (33.3 %) 39.2 %
Type 3 149 (22.8 %) 18 (28.1 %) 11 (20.4 %) 11 (12 %) 0 (0 %) 4 (66.7 %) 0 (0 %) 21.4 %
Surgery Positioning 532 85 54 68 28 NR 3 770
Beach Chair 368 (69.2 %) 43 (50.6 %) 54 (100 %) 50 (74 %) 28 (100 %) NR 3 (100 %) 70.9 %
Supine 164 (30.8 %) 42 (49.4 %) 0 (0 %) 18 (26 %) 0 (0 %) NR 0 (0 %) 29.1 %
Surgical approach 650 85 54 94 28 NR 3 914
Deltopectoral 415 (63.8 %) 52 (61.2 %) 54 (100 %) 94 (100 %) 28 (100 %) NR 3 (100 %) 70.7 %
Deltoid split 235 (36.2 %) 33 (38.8 %) 0 (0 %) 0 (0 %) 0 (0 %) NR 0 (0 %) 29.3 %
Outcomes Union achieved 489 (100 %) 489 83 (97.6 %) 85 54 (100 %) 54 94 (100 %) 94 28 (100 %) 28 5 (83.3 %) 6 3 (100 %) 3 99.6 % 759
Required Revision 9 (1.9 %) 473 3 (7 %) 43 0 (0 %) 45 2 (2 %) 94 0 (0 %) 28 1 (16.7 %) 6 0 (0 %) 3 2.2 % 692
CMS 81.8 507 79.1 31 76.8 54 77.7 57 81.5 28 NR – NR – 80.9 677
ΔHHH (mm) 1.3 282 NR – 1.3 38 NR – 2 28 NR – NR – 1.4 348
ΔNSA (°) 2.6 385 7.4 22 1.1 16 0 3 4 28 NR – NR – 2.8 454
Complication rate 610 85 45 94 28 6 3 871
Total complications 55 (9 %) 8 (9.4 %) 0 (0 %) 24 (25.5 %) 2 (7.1 %) 1 (16.7 %) 0 (0 %) 10.3 %
Deep infection 3 (0.5 %) 3 (3.5 %) – – – – – 0.7 %
Donor site morbidity – – – 7 (7.4 %) – – – 0.8 %
Axillary Nerve injury 1 (0.2 %) – – – – – – 0.1 %
Loss of reduction 4 (0.7 %) 2 (2.4 %) – 4 (4.3 %) – 1 (16.7 %) – 1.3 %
Delayed Union – 3 (3.5 %) – 1 (1.1 %) – – – 0.5 %
Screw complications 21 (3.4 %) – – 4 (4.3 %) 1 (3.6 %) – – 3 %
AVN 18 (3 %) – – – 1 (3.6 %) – – 2.2 %
Symptomatic hardware 1 (0.2 %) – – – – – – 0.1 %
Heterotopic bone formation 3 (0.5 %) – – – – – – 0.3 %
Greater tuberosity migration/resorption 2 (0.3 %) – – 2 (2.1 %) – – – 0.46 %
Adhesive capsulitis 2 (0.3 %) – – 2 (2.1 %) – – – 0.46 %
Subacromial impingement – – – 4 (4.3 %) – – – 0.46 %
3.5

3.5 Femoral head allograft

In the four studies,37–40 evaluating femoral head allograft, the total number of patients included was 85 with average age of 68.7 years old. 39.1 % were male and 60.9 % were female. The majority of fractures were due to falls (83.3 %). The femoral head was either divided into smaller pieces or placed as whole into the metaphysis to provide additional structural support. Patients had a mean follow-up of 15.3 months. Union was achieved in 97.6 % of fractures. The overall complication rate was 9.4 %. The complications reported across the studies included deep infection, loss of reduction and delayed union at 3.5 %, 2.4 %, and 3.5 % respectively. Approximately 7.0 % of patients had to undergo revision surgery with indications including septic nonunion and loss of reduction. The average CMS score was 79.1. Immediate post-operative to final follow-up difference in NSA was 7.4°. Table 2 details the weighted average of demographic data and clinical outcomes of studies evaluating femoral head allograft.

3.6

3.6 Iliac crest allograft

In the three studies41–43 evaluating tricortical iliac crest allograft, the total number of patients included was 54 with average age of 65.2 years old. 37.8 % were male and 62.2 % were female. The majority of fractures were due to falls (71.4 %). The tricortical iliac crest allograft was placed in the metaphysis to provide additional support. Patients had a mean follow-up of 19.8 months. Union was achieved in 100 % of fractures. There were no complications or revision surgeries. The average CMS score was 76.8. Immediate post-operative to final follow-up difference in HHH and NSA was 1.3 mm and 1.1° respectively. Table 2 details the weighted average of demographic data and clinical outcomes of studies evaluating iliac crest allograft.

3.7

3.7 Iliac crest autograft

Five studies43–47 evaluated tricortical iliac crest autograft augmentation of proximal humerus fracture fixation using locking plate. The total number of patients included was 94 patients with an average age of 56.9 years old. 46.8 % were male and 53.2 % were female. The majority of fractures were due to falls (86.4 %). The tricortical iliac crest allograft was harvested and used en-block in the metaphysis to provide structural support. Mean follow-up of the studies was 31.1 months. Union was achieved in 100 % of fractures. The overall complication rate was listed at 25.5 % with donor site morbidity (7.4 %), loss of reduction (4.3 %), screw complication (4.3 %), subacromial impingement (4.3 %), greater tuberosity migration/resorption (2.1 %), adhesive capsulitis (2.1 %) and delayed union (1.1 %). However, given the elevated complication rate, only 2 % of patients had to undergo revision surgery. The average CMS score was 77.7. Table 2 details the weighted average of demographic data and clinical outcomes of studies evaluating iliac crest autograft.

3.8

3.8 Humeral endosteal allograft

In the single study36 evaluating proximal humerus fracture fixation using humeral endosteal allograft and non-locking plate, 6 patients with an average age of 78.2 years old were included. 33.3 % were male and 66.7 % were female. All fractures were due to low energy ground-level falls. Humeral endosteal allograft was shaped and inserted en-block into the metaphysis to provide additional support. Mean follow-up time was not reported. Union was achieved in 83.3 % of fractures. Loss of reduction was observed in one patient. The ASES score improved from 37 pre-operatively to 85 post-operatively. Average range of motion reported was 120° for forward elevation and 42° for external rotation. Average internal rotation was to the level of the 12th thoracic vertebrae. Table 2 details the weighted average of demographic data and clinical outcomes of studies evaluating humeral endosteal allograft.

3.9

3.9 Unspecified cancellous allograft

In the single study48 evaluating proximal humerus fracture fixation using unspecified cancellous allograft, 28 patients with an average age of 64 years old were included. 46.4 % were male and 53.6 % were female. Cancellous allograft was morselized and packed into the comminuted metaphyseal defect after ORIF. Mean follow-up time was 13.8 months. Union was achieved in 100 % of fractures. One patient developed screw penetration and one patient developed AVN. Neither of the two complications required revision surgery. The average CMS score was 81.5. Immediate post-operative to final follow-up difference in HHH and NSA was 2 mm and 4°, respectively. Table 2 details the weighted average of demographic data and clinical outcomes of studies evaluating unspecified cancellous allograft.

3.10

3.10 Distal clavicle autograft

In a single study evaluating proximal humerus fracture fixation using distal clavicle autograft, 3 patients whom were all female, with a mean age of 68.3 years and fractures classified as either Neer Type 3 or Type 4 were included.15 This appears to be the first published description of this technique for augmenting proximal humerus fixation. A 1.5 cm length tricortical distal clavicle autograft was taken from the ipsilateral clavicle to fill the central metaphyseal defect at the fracture site. Although the data is limited, Pinto et al. reported a union rate of 100 %. No complications were reported (for either the donor or fracture site), although the mean follow up period was only 6.3 months post-operatively (11, 5 and 3 months respectively).

4

4 Discussion

The management of proximal humerus fractures is challenging and variable, and despite the known benefits of using bone graft augmentation in ORIF of these fractures, there exists a dearth of literature comparing the different graft options. This systematic review investigates primary fixation of proximal humerus fractures using various bone graft augments. The key finding of this study is that union rates were high and comparable between graft types and that fibular strut allograft is the most common graft type utilized to augment proximal humerus fixation.

Akin to common clinical practice, allografts reported were fibular strut, femoral head, iliac crest, humeral endosteal and unspecified cancellous allografts, as well as tricortical iliac crest autografts. A pilot study of 3 patients reported a new autograft not previously described, where the ipsilateral distal clavicle was harvested for proximal humerus ORIF augmentation. Interestingly, clinical outcomes were quite similar between the graft options reported in the 33 studies meeting inclusion criteria. Most notably, fracture union rates were not only similar between the groups, but 100 % with most graft choices other than humeral endosteal and femoral head allografts, with union rates of 83.3 % and 97.6 %, respectively. However, we note that the data for the use of humeral endosteal allograft is only taken from one single retrospective case series of only 6 patients, and also used non-locking plates in the fixation construct which could have contributed to the lower union rates seen.49

In 2-, 3- and 4-part proximal humerus fractures undergoing ORIF (locking vs non-locking plate not specified) without bone grafting, Yahuaca et al. observed malunion, nonunion or screw complications in 25 % of patients, as well as a 17.5 % reoperation rate.50 These outcomes are similar to those seen elsewhere in the literature with 39 % complication rate and 13 % nonunion rate demonstrated by Boesmueller et al. in a prospective study with a final follow up at 6 months post-operative.51 These values are much higher than the average complication rate of 10.3 % (range 0.0%–25.5 %) and non-union rate 0.4 % (range 0.0–7.0 %) amongst all patients who received bone graft augmentation included in this review.

Fibular strut allografts were the most commonly used bone graft augmentation (693 patients in 20 studies) in the studies meeting our inclusion criteria. One notable benefit of these grafts is the robust structural support they provide to the construct which is especially valuable in the proximal humerus fracture patient population, where osteoporotic bone often poses a legitimate challenge to the surgeon. Additionally, numerous biomechanical studies have shown increased load to failure thresholds when locking plate fixation is augmented with a fibular strut allograft.52,53 One factor that should be considered however, is the difficulty encountered during a revision surgery if needing to convert from an ORIF with fibular strut allograft to a stemmed arthroplasty component, since the medullary canal gets filled with strong fibular cortical bone.54 Compared to those without a previous fibular allograft, Polisetty et al. found a 22 % increase in operative time when converting to a rTSA in patients with a fibular allograft in situ from previous ORIF.55

Although it is difficult to definitively conclude which bone graft augment is the best for the fixation of proximal humerus fractures, fibular strut allograft appears to be the most popular in clinical practice. However in a revision and/or non-union setting, iliac crest autograft remains the gold standard.56 The advantages of autogenous bone grafting, both cancellous and cortical, are well described and replicated through basic science, biomechanical and clinical studies alike.56 However in this review, iliac crest autograft was associated with the highest complication rate (25.5 %) with the most frequent complication being donor site morbidity (7.4 %). Furthermore, the limitations on the size of the graft available for harvest are described as another disadvantage of iliac crest autografts.57 Similarly, the recommended size of distal clavicle autograft is limited to avoid disrupting the stability of the acromioclavicular joint, as demonstrated in numerous existing studies.58

The Constant-Murley Score (CMS) is used to gauge a patient's shoulder function based on two objective and two subjective evaluations.59 Shoulder range of motion and strength are measured by a clinician (maximum score 65) and the level of pain and ability to perform activities of daily living (ADLs) are self-reported by the patient (maximum score 35), with a maximum score of 100 representing the best shoulder function. Although the applicability of this tool to all shoulder pathologies is controversial, it remains a commonly used scale and is often reported even in the context of proximal humerus fractures. In this review, the weighted mean of the CMS scores were calculated for each group, other than the humeral endosteal allograft and the distal clavicle autograft groups as the studies by Wang et al. and Pinto et al. did not report CMS scores as an outcome measure.15,36 CMS scores were quite similar between the remaining groups with average scores of 81.8 (fibular strut allograft), 77.7 (iliac crest autograft), 81.5 (unspecified cancellous allograft), 79.1 (femoral head allograft), 76.8 (iliac crest allograft). This demonstrates good clinical outcomes with the various graft types, however, comparative prospective studies are needed to draw more accurate conclusions about their differences.

The complications commonly associated with proximal humerus fracture ORIF include hardware complications (such as screw cut-out, intra-articular screw penetration or symptomatic hardware), varus collapse, avascular necrosis, loss of reduction, malunion and nonunion. Previously described risk factors for loss of reduction include osteoporosis, increasing age, varus malposition and medial calcar comminution.60 The humeral head height (HHH) is a serial radiographic measurement calculated postoperatively following a proximal humerus ORIF. Greater than or equal to 5 mm of HHH loss is often used to define a loss of reduction.60 In this review, changes in HHH were reported for fibular strut allograft, iliac crest allograft and unspecified cancellous allograft with an average of 1.3 mm, 1.7 mm and 2 mm respectively, all of which fall below the threshold for defining loss of reduction. Data for the other graft options did not include HHH measurements. Restoration of medial calcar support has been shown to be an important factor for maintaining reduction. Gardner et al. showed an average change in HHH of only 1.2 mm in a group of patients determined to have adequate medial column support, versus 5.8 mm in patients without.10 Most clinically relevant is that only 5.6 % of patients with adequate medial column support experienced intra-articular screw penetration, meanwhile 29.4 % of patients experienced screw penetration when there was a change of HHH of greater than or equal to 5.8 mm (inadequate medial column support). To address this issue, bone grafts can be used to restore the medial calcar, maintain HHH and prevent loss of reduction in order to improve the outcomes in these difficult fractures.

Another radiographic measurement used to define a loss of reduction over time is the change in the humeral neck shaft angle (NSA). NSA is the angle between a line bisecting the humeral shaft and a line crossing the anatomic neck. This measurement can be used initially to evaluate the adequacy of the reduction as a predictor of eventual failure or can be compared between the initial postoperative radiograph and the radiographs at later time points (including final follow up) to determine if there was a loss of reduction over time. Yewlett et al. concluded that to successfully fix a proximal humerus fracture with a locking plate, an appropriate NSA must be restored and is the best prognostic factor for avoidance of screw penetration and implant failure.60 Where NSA measurements were available in our review, the mean changes in angulation were less than 5° in each of the bone grafting groups, other than the femoral head allograft at 7.4°.

One key factor that was not reported, and therefore not summarized in our study, was the surgical time associated with each graft option. This comparison could certainly impact graft choice as it may relate to infection rates, blood loss, operating room scheduling and hospital resources, and other key aspects of clinical practice. For instance, femoral head allograft must be cut and shaped to appropriately fit into the medullary canal of the proximal humerus, which may add operative time when compared to the fibular strut allograft which does not require much preparation. Similarly, the iliac crest autograft harvest may require additional operative time if patient positioning or surgical team limitations prevent graft harvest and exposure of the proximal humerus fracture site from happening simultaneously. Differences in operative time is another factor that can be evaluated in future comparative studies.

A recently published randomized controlled trial compared standard proximal humerus ORIF with a locking plate, to those augmented with a fibular allograft in adult patients with a comminuted medial column (≥3 fragments).61 Despite its relevance, the results of this trial had to be excluded from our review, as proximal humerus fracture dislocations were included in their study but excluded from ours. Wang et al. used the Disabilities of the Arm, Shoulder and Hand (DASH) questionnaire (also scored out of 100) to quantify patient-reported outcomes. The DASH score at the 12 month postoperative mark was used as the primary outcome measure and showed no significant difference between the groups. Their secondary outcomes included changes in HHH and NSA, complications and the CMS. These also did not demonstrate any significant difference between standard ORIF and ORIF augmented with fibular allograft. Of note, one confounding factor is that the operating surgeons were allowed to also augment their fixation with iliac crest autograft or bone substitutes as needed.

Rischen et al. performed a large retrospective review of 41,216 proximal humerus fractures treated with locking plate fixation between the years 2010 and 2018, as determined by diagnostic and procedural codes claimed to one of Germany's largest insurers.62 Of these, 840 fixations were augmented by either allogenic or autologous bone grafting. Use of augmentation (including bony and cemented) increased by 25 % when comparing the first 3 years of the study dates, to the final 3, with the rate of only locking plate fixation alone decreasing over the study period. In the augmentation group, overall intra-hospital complication rate, mechanical surgical complications and rate of reoperation were all statistically significant (p < 0.001), however the authors comment on the undeniable role selection bias plays, as more complex fractures are more likely to undergo augmentation.

4.1

4.1 Limitations

The robust search strategy and methodology used is a major strength of this systematic review as it allowed us to capture data for many different graft types used, from a large number of patients with a wide geographic distribution across multiple continents. This also allowed us to reduce bias while reporting on a novel topic that has not yet been thoroughly studied.

This study has several limitations. As mentioned previously, descriptive statistics were used to summarize and compare results by graft type, as the heterogeneity of the data precluded a meta-analysis. In particular, length of final follow up was variable both within, and between studies which is a challenge especially when comparing outcomes that are inevitably dependent on time from the surgical intervention. In the study reporting outcomes of humeral endosteal allograft for example, no mean follow-up duration was reported. Additionally, the number of patients in each graft group was quite variable, from as many as 694 patients receiving the fibular strut augmentation, to only 6 and 3 patients receiving the humeral endosteal and distal clavicle autografts, respectively. Finally, the lack of standardized outcome measures poses a challenge when comparing bone graft options. Further comparative studies would be most beneficial in determining which graft choice is associated with the best clinical and radiologic outcomes in the fixation of complex displaced proximal humerus fractures.

5

5 Conclusion

Bone graft augments are an effective adjunct to open reduction internal fixation of complex proximal humerus fractures. Fibular strut allograft is the most common bone graft augment used. The majority of patients treated with bone graft augmentation achieved bony union (99.6 %, range 83.3%–100 %) and average CMS scores at final follow-up were similar between graft types (range 76.8–81.8). However, no conclusive data suggests superiority of one particular bone graft type over another as comparative studies are lacking. Future comparative prospective and randomized studies should aim to investigate and compare the outcomes associated with each graft type relative to controls.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Ethics

This study did not need any ethics approval as it was a systematic review of the literature and did not involve any human subjects.

Guardianpatient statement

No guardian/patient consent was needed for this study as this study was a systematic review.

CRediT authorship contribution statement

Ali Etemad-Rezaie: Data curation, Writing – review & editing. Serena Dienes: Data curation, Writing – review & editing. Chetan Gohal: Conceptualization, Methodology. Valerie Politis-Barber: Data curation, Writing – review & editing. Stephanie Searle: Data curation, Writing – review & editing. Diane Nam: Conceptualization, Writing – review & editing. Ujash Sheth: Conceptualization, Methodology, Writing – review & editing.

References

  1. , , , , , . Changing incidence and residual lifetime risk of common osteoporosis-related fractures. Osteoporos Int. 1993;3(3):127-132.
    [Google Scholar]
  2. , , , . The epidemiology of proximal humeral fractures. Acta Orthop Scand. 2001;72(4):365-371.
    [Google Scholar]
  3. , , , et al . A scoping review of the proximal humerus fracture literature. BMC Muscoskel Disord. 2015;16:112.
    [Google Scholar]
  4. , , , . Epidemiology of humerus fractures in the United States: nationwide emergency department sample, 2008. Arthritis Care Res. 2012;64(3):407-414.
    [Google Scholar]
  5. , , , . Interventions for treating proximal humeral fractures in adults. Cochrane Database Syst Rev. 2012;12
    [Google Scholar]
  6. , , , , . Plate fixation for management of humerus fractures. Injury. 2018;49(Suppl 1):S33-S38.
    [Google Scholar]
  7. , , , , , . Avoiding complications of locking plating for proximal humerus fractures. J Long Term Eff Med Implants. 2022;32(3):73-81.
    [Google Scholar]
  8. , , , , , . Factors that influence reduction loss in proximal humerus fracture surgery. J Orthop Trauma. 2015;29(6):276-282.
    [Google Scholar]
  9. , , , et al . Postoperative evaluation of reduction loss in proximal humeral fractures: a comparison of plain radiographs and computed tomography. Orthop Surg. 2017;9(2):167-173.
    [Google Scholar]
  10. , , , , , , . The importance of medial support in locked plating of proximal humerus fractures. J Orthop Trauma. 2007;21(3):185-191.
    [Google Scholar]
  11. , , , . 501 Lorich DG. Indirect medial reduction and strut support of proximal humerus fractures using an endosteal implant. J Orthop Trauma. 2008;22(3):195-200.
    [Google Scholar]
  12. , , , , . Evaluation of mushroom-shaped allograft for unstable proximal humerus fractures. Arch Orthop Trauma Surg. 2022;142(3):409-416.
    [Google Scholar]
  13. , , , , . Valgus impakte Neer tip 4 humerus üst uç kiriklarinda açik redüksiyon, yükseltme ve greftleme [Treatment of Neer type 4 impacted valgus fractures of the proximal humerus with open reduction, elevation, and grafting] Acta Orthop Traumatol Turcica. 2007;41(2):113-119.
    [Google Scholar]
  14. , , , , . Technique of open reduction and internal fixation of comminuted proximal humerus fractures with allograft femoral head metaphyseal reconstruction. Am J Orthoped. 2015;44(10):471-475.
    [Google Scholar]
  15. , , , et al . Distal clavicle autograft in fractures of the proximal humerus: surgical technique. Tech Orthop. 2022;37(4):245-248.
    [Google Scholar]
  16. , , , et al . The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. 2021;372
    [Google Scholar]
  17. , , , , . Modified use of a fibular strut in the reduction and stabilization of 2-Part Osteoporotic proximal humerus fractures. J Am Acad Orthop Surg Glob Res Rev. 2020;4(10)
    [Google Scholar]
  18. , , , , , . Clinical outcomes of allograft with locking compression plates for elderly four-part proximal humerus fractures. J Orthop Surg Res. 2015;10:114.
    [Google Scholar]
  19. , , , et al . The augment of the stability in locking compression plate with intramedullary fibular allograft for proximal humerus fractures in elderly people. BioMed Res Int. 2018;2018
    [Google Scholar]
  20. , , , , , . Fibular strut allograft influences reduction and outcomes after locking plate fixation of comminuted proximal humeral fractures in elderly patients: a retrospective study. BMC Muscoskel Disord. 2019;20(1):511.
    [Google Scholar]
  21. , , , , . Indirect medial reduction and strut support of proximal humerus fractures using an endosteal implant. J Orthop Trauma. 2008;22(3):195-200.
    [Google Scholar]
  22. , , , , , , . Geriatric proximal humeral fracture patients show similar clinical outcomes to non-geriatric patients after osteosynthesis with endosteal fibular strut allograft augmentation. J Shoulder Elbow Surg. 2015;24(6):889-896.
    [Google Scholar]
  23. , , , , . Which additional augmented fixation procedure decreases surgical failure after proximal humeral fracture with medial comminution: fibular allograft or inferomedial screws? J Shoulder Elbow Surg. 2018;27(10):1852-1858.
    [Google Scholar]
  24. , , , . Comparison between MIPO and the deltopectoral approach with allogenous fibular bone graft in proximal humeral fractures. Clin Shoulder Elb. 2020;23(3):136-143.
    [Google Scholar]
  25. , , , , . Outcomes of locking plate fixation with fibular allograft augmentation for proximal humeral fractures in osteoporotic patients: comparison with locking plate fixation alone. Bone Joint Lett J. 2019;101-B(3):260-265.
    [Google Scholar]
  26. , , , et al . Intramedullary allograft fibula as a reduction and fixation tool for treatment of complex proximal humerus fractures with diaphyseal extension. J Orthop Trauma. 2014;28(3):e56-e64.
    [Google Scholar]
  27. , , , et al . Locking plate and fibular allograft augmentation in unstable fractures of proximal humerus. Injury. 2012;43(11):1939-1942.
    [Google Scholar]
  28. , , , , , , . Improved outcomes using a fibular strut in proximal humerus fracture fixation. Orthopedics. 2020;43(5):262-268.
    [Google Scholar]
  29. , , , , , . Endosteal strut augment reduces complications associated with proximal humeral locking plates. Clin Orthop Relat Res. 2011;469(12):3300-3306.
    [Google Scholar]
  30. , , , . Early outcomes of proximal humerus fracture fixation with locking plate and intramedullary fibular strut graft. Orthopedics. 2014;37(9):e822-e827.
    [Google Scholar]
  31. , , , , , . Locking Plate combined with endosteal fibular allograft augmentation for medial column comminuted proximal humeral fracture. Orthopedics. 2020;43(6):367-372.
    [Google Scholar]
  32. , , , et al . The impact of preoperative coronal plane deformity on proximal humerus fixation with endosteal augmentation. J Orthop Trauma. 2014;28(6):338-347.
    [Google Scholar]
  33. , , , et al . Comparison of the effects of proximal humeral internal locking system (PHILOS) alone and PHILOS combined with fibular allograft in the treatment of neer three- or four-part proximal humerus fractures in the elderly. Orthop Surg. 2019;11(6):1003-1012.
    [Google Scholar]
  34. , , , , , , . Locking Plate use with or without strut support for varus displaced proximal humeral fractures in elderly patients. JBJS Open Access. 2019;4(3)
    [Google Scholar]
  35. , , , et al . Cancellous bone allograft is comparable to fibular strut allograft for augmentation in three- or four-part proximal humeral fractures. J Shoulder Elbow Surg. 2021;30(9):2065-2072.
    [Google Scholar]
  36. , , , et al . Treatment of osteopenic or non-united fractures of the humerus with structure bone allograft and non-locking plates. J Med Sci. 2013;33:67-74.
    [Google Scholar]
  37. , , , , , . Allogenic bone grafting for augmentation in two-part proximal humeral fracture fixation in a high-risk patient population. Arch Orthop Trauma Surg. 2015;135(1):79-87.
    [Google Scholar]
  38. , , , et al . Restoration of medial hinge in acute complex proximal humeral fractures using intra-medullary strut allograft. Shoulder Elbow. 2011;3
    [Google Scholar]
  39. , , , et al . Proximal humeral fractures with a severe varus deformity treated by fixation with a locking plate. J Bone Joint Surg Br. 2010;92(5):672-678.
    [Google Scholar]
  40. , , , et al . Efficacy of proximal humeral internal locking system combined with allogeneic femoral head bone grafts for complex osteoporotic proximal humeral fractures. Int J Clin Exp Med. 2019;12(7):8847-8854.
    [Google Scholar]
  41. , , , . Strut support with tricortical iliac allografts in unstable proximal humerus fractures: surgical indication and new definition of poor medial column support. Clin Shoulder Elb. 2019;22(1):29-36.
    [Google Scholar]
  42. , , , , . Results of surgical management of valgus-impacted proximal humerus fractures with structural allografts. Acta Orthop Traumatol Turcica. 2014;48(5):546-552.
    [Google Scholar]
  43. , , , , . Valgus impakte Neer tip 4 humerus üst uç kiriklarinda açik redüksiyon, yükseltme ve greftleme [Treatment of Neer type 4 impacted valgus fractures of the proximal humerus with open reduction, elevation, and grafting] Acta Orthop Traumatol Turcica. 2007;41(2):113-119.
    [Google Scholar]
  44. , , , et al . Outcomes for four-part proximal humerus fractures treated with a locking compression plate and an autologous iliac bone impaction graft. Injury. 2012;43(10):1724-1731.
    [Google Scholar]
  45. , , , et al . Locking plate fixation combined with iliac crest bone autologous graft for proximal humerus comminuted fracture. Chin Med J. 2014;127(9):1672-1676.
    [Google Scholar]
  46. , , , , , , . Clinical and functional outcomes of displaced proximal humerus fractures treated with PHILOS plate system and autologous bone graft. J Pakistan Med Assoc. 2022;72(11):2291-2294.
    [Google Scholar]
  47. , , , et al . A novel technique for treatment of metaphyseal voids in proximal humerus fractures in elderly patients. Medicina (Kaunas). 2022;58(10):1424.
    [Google Scholar]
  48. , , , et al . Cancellous bone allograft is comparable to fibular strut allograft for augmentation in three- or four-part proximal humeral fractures. J Shoulder Elbow Surg. 2021;30(9):2065-2072.
    [Google Scholar]
  49. , , , . Fractures and fracture-dislocations of the proximal humerus: a retrospective analysis of 82 cases treated with the Philos(®) locking plate. Injury. 2014;45(Suppl 6):S43-S48.
    [Google Scholar]
  50. , , , et al . Acute surgical management of proximal humerus fractures: ORIF vs. hemiarthroplasty vs. reverse shoulder arthroplasty. J Shoulder Elbow Surg. 2020;29(7S):S32-S40.
    [Google Scholar]
  51. , , , et al . Risk factors for humeral head necrosis and non-union after plating in proximal humeral fractures. Injury. 2016;47(2):350-355.
    [Google Scholar]
  52. , , , , , , . Biomechanical analysis of proximal humeral fixation using locking plate fixation with an intramedullary fibular allograft. Clin Biomech. 2010;25(7):642-646.
    [Google Scholar]
  53. , , , , , , . Proximal humeral fracture fixation: locking plate construct ± intramedullary fibular allograft. J Shoulder Elbow Surg. 2012;21(7):894-901.
    [Google Scholar]
  54. , , , , , , . Failed proximal humerus osteosynthesis using intramedullary fibular strut allograft conversion to reverse shoulder arthroplasty. Tech Hand Up Extrem Surg. 2020;24(1):7-12.
    [Google Scholar]
  55. , , , , , . Reverse shoulder arthroplasty for failed proximal humerus osteosynthesis with intramedullary allograft: a case series. Journal of Shoulder and Elbow Arthroplasty. 2020;4
    [Google Scholar]
  56. , , . Autologous iliac crest bone graft: should it still be the gold standard for treating nonunions? Injury. 2007;38(Suppl 1):S75-S80.
    [Google Scholar]
  57. , , , , , . Complications of iliac crest bone graft harvesting. Clin Orthop Relat Res. 1996;329:300-309.
    [Google Scholar]
  58. , , , , , . Ligamentous anatomy of the distal clavicle. J Shoulder Elbow Surg. 2003;12(4):355-359.
    [Google Scholar]
  59. , , . A clinical method of functional assessment of the shoulder. Clin Orthop Relat Res. 1987;214:160-164.
    [Google Scholar]
  60. , . What is the single most important technical aspect when fixing a proximal humeral fracture with A PHILOS plate?”. MOJ Orthopedics & Rheumatology.. 2016;6
    [Google Scholar]
  61. , , , et al . Effect of fibular allograft augmentation in medial column comminuted proximal humeral fractures: a randomized controlled trial. J Bone Joint Surg. February 15, 2023;105(4):302-311.
    [Google Scholar]
  62. , , , et al . Treatment reality of proximal humeral fractures in the elderly-trending variants of locking plate fixation in Germany. J Clin Med. 2023;12(4):1440.
    [Google Scholar]

Fulltext Views
390

PDF downloads
107
View/Download PDF
Download Citations
BibTeX
RIS
Show Sections

Suggested read for related articles: