1 Introduction

Proximal humerus fractures, accounting for 4 %–5 % of all fractures, have shown a consistent annual increase of over 13 % in the last three decades.^1–3 Managing these fractures optimally is challenging due to the complexities of preserving bone alignment, joint surface congruity, and humeral head vascularity.⁴ Decision-making for operative treatment involves considering various factors such as age, activity level, patient expectations, medical comorbidities, bone quality, degree of comminution, previous surgeries, associated injuries, and fracture type.⁵ While nonoperative treatment is common for most proximal humerus fractures, there is no consensus on the preferred strategy for 3-part and 4-part fractures.

Historically, open reduction and internal fixation (ORIF) were employed in young patients to preserve bone stock and promote anatomic healing for 3-and 4-part proximal humerus fractures.⁶ However, ORIF has been associated with complication and reoperation rates as high as 35 %,⁷ leading to issues like pain, disability, osteonecrosis, malunion, nonunion, post-traumatic arthritis, and humeral head collapse.⁸ Hemiarthroplasty (HA) outcomes have also been inconsistent, with poor functional results.^9–11 Reverse total shoulder arthroplasty (RTSA) has emerged as an attractive alternative for low-demand and elderly patients, offering more consistent outcomes, faster recovery, reduced postoperative rehabilitation, and less reliance on tuberosity healing.^12,13

An epidemiological analysis spanning from 2008 to 2017 revealed a significant increase in nonoperative management from 67.6 % to 77.2 % (P = 0.001), a decrease in ORIF use from 76.6 % to 72.6 % (P = 0.004), a significant rise in RTSA utilization from 4.1 % to 24.5 % (P < 0.001), and a notable decrease in HA use from 19.3 % to 3 % (P < 0.001).¹⁴ However, studies comparing RTSA use in the acute setting versus post-failed ORIF are limited due to small sample sizes, single institution analyses, and the absence of fracture type (2-part, 3-part, and 4-part) distinction.^15–17 Given the evolving landscape of treatment for 3-and-4-part proximal humerus fractures, an epidemiological analysis provides a means to bridge the gap between understanding and effective intervention. Thus, our objectives were to i) assess the trends in nonoperative and operative management, and ii) compare complication rates between ORIF and conversion ORIF (to RTSA).

2 Methods

We conducted a review of Medicare inpatient and outpatient facility data using the Pearldiver, an all-payer national database based in Warsaw, IN, USA. Pearldiver is a Health Insurance Portability and Accountability Act (HIPAA)-compliant database that encompasses Medicare data spanning from 2005 to 2022. It allows access to Medicare files through Current Procedural Terminology (CPT) codes from the American Medical Association, Chicago, IL, USA, and International Classification of Disease Tenth Edition (ICD-10) codes. To maintain patient privacy, the database does not report groups with 10 patients or fewer.

Included in our study were patients who underwent open reduction and internal fixation (ORIF) for 2-part (n = 2783), 3-part (n = 2170), and 4-part (n = 1087) proximal humerus fractures from April 2016 to December 2021. During the same timeframe, we also considered patients who experienced failed ORIF and subsequently underwent reverse total shoulder arthroplasty (RTSA), including n = 108 for 2-part fractures, n = 123 for 3-part fractures, and n = 128 for 4-part fractures. We collected demographic information and postoperative medical and surgical complications within a 90-day period, along with the time interval between ORIF and RTSA. The Neer classification, which divides proximal humerus fractures into four parts (humeral head, greater tuberosity, less tuberosity, and humeral shaft), guided our classification.

Our statistical analyses primarily employed descriptive methods to capture the incidence of each procedure from 2016 to 2020. Significance was determined at a p-value <0.05, and chi-squared tests were used to compare complication rates. All statistical analyses were performed using SPSS 29 software from IBM Corp, Armonk, NY, USA.

3 Results

During the 5-year study period examined, a total of 17,760 patients who had proximal humerus fractures were treated nonoperatively (Table 1). A total of 19,157, 1,263, and 5209 patients who had proximal humerus fractures were treated with ORIF, HA, and RTSA, respectively in the Medicare population (Tables 2–4). In 2016, 2938 patients had proximal humerus fractures treated nonoperatively (Fig. 1). In 2016, 3,465, 342, and 724 patients who had proximal humerus fractures were treated with ORIF, HA, and RTSA, respectively. In 2020, 2756 patients had proximal humerus fractures treated nonoperatively. In 2020, 3,265,207, and 1464 patients who had proximal humerus fractures were treated with ORIF, HA, and RTSA, respectively (Figs. 1–4). Demographics and baseline characteristics for nonoperative treatment, ORIF, HA, and RTSA are detailed in Tables 1–4 Complications for nonoperative treatment, ORIF, HA, and RTSA are detailed in Tables 5–8.

Close

Fig. 1 Trends of treatment of 2-part proximal humerus fractures.

Export to PPT

Close

Fig. 2 Trends of treatment of 3-part proximal humerus fractures.

Export to PPT

Close

Fig. 3 Trends of treatment of greater tuberosity proximal humerus fractures.

Export to PPT

Close

Fig. 4 Trends of treatment of surgical neck proximal humerus fractures.

Export to PPT

The malunion and nonunion rates for ORIF between different types of proximal humerus fractures were similar (2-part fractures: 1.8 %, 4.7 %; 3-part fractures: 1.8 %, 3.5 %; 4-part fractures: 2.4 %, 3.7 %). The conversion rates of failed ORIF to RTSA were 1.9 %, 2.8 %, and 5.9 % for 2-part, 3-part, and 4-part fractures, respectively (Table 8). The time interval from failed ORIF to RTSA was 190 days for 2-part fractures, 169 days for 3-part fractures, and 129 for 4-part fractures (Table 9).

4 Discussion

Despite the rising incidence of proximal humerus fractures in the elderly, a consensus on the optimal treatment strategy remains elusive.¹⁴ Furthermore, existing studies on US epidemiological trends related to proximal humerus fractures often lack emphasis on fracture types, relying on small sample sizes and single-institution analyses.^15–17 Our key findings indicate: (1) RTSA experienced the most significant increase from 2016 to 2020, irrespective of fracture type; and (2) the conversion rates of failed ORIF to RTSA for 4-part fractures were twice as high as the malunion and nonunion rates of ORIF.

We acknowledge certain limitations in our study. Despite third-party review of all patient records, the possibility of medical billing and coding errors exists. The reported nationwide billing/coding error rate of 1.0 % by the United States Department of Human and Health Services further mitigates this risk.¹⁸ Exclusion of specific CPT codes for specificity may have led to an underestimation of the number of procedures. Additionally, the SARS-CoV-2 (COVID-19) pandemic might have influenced the number of procedures performed in a given year. Certain CPT codes amalgamate multiple procedures (e.g., HA, TSA, and RTSA), hindering the distinction between them. While our study focuses on the Medicare population representing the elderly in the United States, its applicability to other countries may be limited. Given the nature of our study, inferences can only be drawn based on the identified trends. Notably, our study boasts strengths such as a large sample size, the classification of fractures by type, and exclusive use of ICD-10 codes.

Several studies highlight an increasing rate of RTSA in the US for proximal humerus fractures. Patel et al. reported a 1841.4 % increase (P < 0.001) in RTSA from 2010 to 2019, attributed to high prosthesis survival, satisfactory functional outcomes, and greater tolerance for tuberosity positioning and healing compared to ORIF and HA.^19–21 Hasty et al. found a 406 % increase in RTSA, a 47 % decrease in HA, and constant ORIF rates from 2005 to 2012, with an overall rise in operative vs. nonoperative management for all fractures.²² However, these studies did not delve into trends based on fracture type. Bahrs et al. found no significant increase (p = 0.09) in fracture severity from 2006 to 2011 but observed a higher occurrence of complex proximal humerus fractures in older women with comorbidities.²³ Iglesias-Rodriguez et al. linked the increase in complex fractures to advancements in RTSA options and the preference for surgical interventions.²⁴

Our study reveals high malunion and nonunion rates in 4-part fractures undergoing ORIF, prompting consideration of salvage ORIF to RTSA efficacy compared to acute RTSA. Studies show higher complication rates and worse outcomes in RSA after failed ORIF compared to acute RTSA, potentially due to better tuberosity repair in acute RTSA versus malpositioned union in late RTSA.^25–27 Patient demographic characteristics influence RTSA outcomes, with older age, dependent functional status, and higher American Society of Anesthesiologists designation being factors. These factors, coupled with the acuity of proximal humerus fractures, may contribute to surgeons opting for RTSA in select patients.^28–30 In contrast, ORIF remains a viable option with good functional outcomes in young patients.³¹

Nonoperative treatment for 2-part, 3-part, and 4-part fractures remained unchanged from 2016 to 2020, while RTSA increased during this period. A meta-analysis suggested an initial trial of nonoperative treatment in the elderly, reserving RTSA for cases where nonoperative treatment fails, potentially resulting in fewer patients ultimately requiring RTSA. Du et al. compared studies and found that RTSA had significantly higher Constant scores and lower reoperation rates than nonoperation, HA, and ORIF for 3-or 4-part proximal humeral fractures.³² Lopiz et al.'s prospective trial indicated less pain perception in RTSA patients, although functional outcomes were similar at 12 months postoperatively.³³

Our epidemiological analysis of proximal humerus fractures by type revealed an increase in RTSA for 2-part, 3-part, and 4-part fractures, while nonoperative treatment showed no change from 2016 to 2020. Further research is needed to determine the most effective operative treatments while optimizing outcomes, especially in the context of complex proximal humerus fractures and varied patient demographic characteristics.

Funding

None.

Data availability

Available in a respository upon request.

Patient consent

No patient consent needed due to retrospective nature and public database.

Ethical approval

IRB exemption due to retrospective nature and public database.

Use of AI tool

No use of AI tool.

CRediT authorship contribution statement

Sandeep S. Bains: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Roles, Writing – original draft, and, Writing – review & editing. Jeremy A. Dubin: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Roles, Writing – original draft, and, Writing – review & editing. Ethan A. Remily: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Roles, Writing – original draft, and, Writing – review & editing. Ruby Gilmor: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Roles, Writing – original draft, and, Writing – review & editing. Daniel Hameed: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Roles, Writing – original draft, and, Writing – review & editing. Rubén Monárrez: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Roles, Writing – original draft, and, Writing – review & editing. John V. Ingari: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Roles, Writing – original draft, and, Writing – review & editing.