INTRODUCTION
As the volume of total shoulder arthroplasty (TSA) continues to grow, it has become increasingly important to characterize perioperative outcomes in medically complex patient populations [
1-
5]. Hemophilia is a rare congenital deficiency of clotting factors VIII or IX that results in prolonged or uncontrolled bleeding. Recurrent intra-articular bleeding can lead to hemophilic arthropathy, a condition characterized by joint remodeling and chronic pain. This condition most often affects the knee (45%) and elbow (30%), and less frequently the shoulder (3%) [
6,
7]. Although routine clotting factor replacement therapy has markedly reduced the incidence of hemophilic arthropathy, advanced disease remains severe and debilitating [
7]. In advanced hemophilic arthropathy, progressive joint destruction may require major orthopaedic procedures, including joint arthroplasty, to prevent irreversible bone loss and restore function [
8,
9].
Although individuals with hemophilia are generally considered to have a low baseline risk for venous thromboembolism (VTE), including deep vein thrombosis (DVT) and pulmonary embolism (PE), major surgery introduces several thrombotic risk factors. These include perioperative immobilization, endothelial injury, and clotting factor replacement, which may paradoxically increase VTE risk in this patient population [
8,
10]. While outcomes following lower extremity arthroplasty in patients with hemophilia have been reported, the perioperative risk profile associated with TSA remains poorly defined [
11-
15].
In this study, we characterized the perioperative complications of patients with hemophilia undergoing total shoulder arthroplasty compared with matched controls. We hypothesized that patients with hemophilia experience higher rates of thromboembolic and medical complications following TSA.
DISCUSSION
The principal findings of this study were that patients with hemophilia undergoing TSA have a significantly higher risk of experiencing thromboembolic events within 90 days postoperatively, including DVT, PE, VTE, and stroke. In addition, hemophilia was associated with increased rates of hospital readmission, revision surgery, bleeding, and medical complications compared with matched controls.
We found that patients with hemophilia had increased risk of thromboembolic events and bleeding complications following TSA. These findings align with prior research and reinforce the importance of perioperative planning, including preparation for potential postoperative blood transfusion and appropriate administration of clotting factor replacement therapy [
11]. Although thromboembolic risk was greater in our hemophilia cohort (3.76%), it remained within the range of VTE rates reported in the general population following TSA (0.2%–16.0%) [
16]. In addition, we found no significant differences in prosthetic joint infection between the groups, which is consistent with prior studies [
11].
Our results challenge the traditional assumption that hemophilia primarily confers bleeding risk while protecting against thrombosis. Instead, our findings demonstrate an association between hemophilia and increased perioperative thromboembolic events. One potential explanation is a perioperative “thrombotic paradox,” in which clotting factor replacement and antifibrinolytic use may shift patients into a transient hypercoagulable state [
8]. However, as our dataset does not include treatment-specific variables such as factor dosing or antifibrinolytic administration, this proposed mechanism remains speculative. Prior studies suggest that high-dose factor replacement can restore or exceed normal coagulation, increasing thrombin generation and thrombotic potential, particularly in the setting of surgical endothelial injury and postoperative immobilization [
19-
22]. Further investigations incorporating detailed perioperative management data are needed to clarify the underlying mechanisms driving this observed association.
The risk profile may be especially relevant in TSA, where postoperative immobilization in adduction and internal rotation differs from early mobilization protocols commonly practiced following hip and knee arthroplasty [
18,
23-
25]. These procedure-specific factors may help explain why thromboembolic risk following TSA in patients with hemophilia appears to differ from that reported in the lower arthroplasty literature [
12,
18,
23].
Existing studies suggest the baseline risk for VTE following TSA may be as high as 13%, despite TSA generally being considered a lower-risk procedure compared to lower extremity arthroplasty [
16,
17]. Several risk factors have been identified, including prolonged hospitalization and postoperative immobilization [
16]. Unlike total hip arthroplasty (THA) and total knee arthroplasty (TKA), which emphasize early postoperative mobilization, postoperative protocols following TSA involve prolonged immobilization [
18,
23]. A cross-sectional study found that 89% of TSA postoperative protocols included sling immobilization with or without an immobilizer for four to six weeks following surgery [
24,
26]. Immobilization is prioritized in shoulder procedures to protect soft tissue structures, with most rehabilitation protocols delaying range-of-motion exercises until six weeks after surgery [
26]. In contrast, early mobilization, weight-bearing, and ambulation are encouraged within 24–48 hours postoperatively following THA or TKA to reduce thromboembolic risk [
18,
25,
27].
Despite these considerations, the literature investigating outcomes of TSA in patients with hemophilia remains limited [
28]. A 2025 literature review identified only four articles evaluating shoulder arthroplasty in this patient population [
10]. One study reported that patients with hemophilia may have an increased risk of bleeding complications and VTE following TSA, while others describe TSA as a reasonable surgical option for patients with hemophilic arthropathy [
10,
11]. TSA performed in patients with hemophilia represents an understudied and distinct clinical scenario in which bleeding risk, thrombosis risk, and prolonged postoperative immobilization intersect.
From a clinical perspective, these findings have important implications for preoperative counseling and perioperative planning. Surgeons should counsel patients with hemophilia undergoing TSA about the elevated risk for bleeding and the increased risk for thromboembolic and medical complications. Early involvement of hematology is essential for clotting factor replacement, including preoperative factor optimization and postoperative dose adjustments to maintain adequate hemostasis, while avoiding excessive hypercoagulability. Consideration of thromboprophylaxis strategies, including mechanical or chemoprophylaxis, should also be individualized to balance the risk for postoperative bleeding with the increased thromboembolic risks associated with surgery and postoperative immobilization. When feasible, early and structured postoperative mobilization may help mitigate thromboembolic risk, while protecting the surgical repair. Finally, close postoperative monitoring for bleeding and thromboembolic complications is warranted given the competing hemostatic risks in this patient population. The observed increases in readmission, revision surgery, and medical complications further highlight the need for multidisciplinary perioperative management to optimize outcomes following TSA in patients with hemophilia [
29].
Beyond thrombotic events, patients with hemophilia experienced higher rates of bleeding-related complications, revision surgery, and postoperative opioid prescription utilization. Chronic hemophilic arthropathy and recurrent hemarthroses may contribute to joint degeneration and increased technical complexity, potentially increasing revision risk [
6,
7]. Higher rates of systemic complications such as pneumonia, UTI, and AKI may partially explain the increased readmission observed in this cohort. Notably, rates of PJI did not differ between groups, highlighting the heterogeneity of complication risk in this population.
This study has several limitations. We utilized data from the TriNetX research database, which compiles de-identified electronic health records across multiple institutions. While this provides a large sample size, it introduces potential inconsistencies in coding, documentation, and treatment practices, which could affect data reliability. In particular, the hemophilia cohort included patients with hemophilia A, hemophilia B, and related coagulation disorders identified through diagnostic coding, which precluded stratification by specific disorder type, disease severity, or factor replacement regimen. Hemophilic conditions were grouped to include von Willebrand disease and other clotting factor deficiencies. As these disorders differ in pathophysiology and bleeding risk, evaluating them as a single cohort may have introduced clinical heterogeneity and potentially confounded the interpretation of our results. Additionally, due to the retrospective design of our study and inherent limitations within the TriNetX database, hemophilia subgroup analyses were not performed. Variations in disease subtype and severity may have an impact on postoperative complication rates.
Next, the dataset lacked granularity in key clinical variables. Similarly, data regarding surgical indications for TSA, including fracture, arthritis, or rotator cuff arthropathy, were unavailable, which may have also influenced complication rates. Differences between primary and secondary TSA may be associated with different intraoperative and postoperative complication profiles, representing a limitation of this study. Further, TSA implant types were not accounted for due to limitations in the TriNetX database. As implant types have evolved substantially between 2005 and 2025, this factor may have influenced postoperative outcomes. In addition, some outcomes (e.g. PJI and bleeding complications) have very low event counts. Although ORs are reported, these outcomes have limited statistical power, and the results should be interpreted with appropriate caution.
Data regarding important hemophilia-specific factors, such as clotting factor replacement regimen, inhibitor status, use of antifibrinolytics, and intraoperative blood loss, were not available in the TriNetX database. As these factors could significantly influence perioperative bleeding and thromboembolic risk, the absence of this information may have impacted our results. Furthermore, some outcomes, including PJI and bleeding, had fewer than 10 reported incidences in the control group. Although the limited sample size likely reduced statistical power, several comparisons with confidence intervals were given to inform the reader of precision for estimated effects. Finally, this study only captured data within a 90-day postoperative period and did not account for long-term complications, delayed VTE events, or late-onset infections. Future prospective studies with longer follow-up periods are warranted to better characterize the perioperative risks in this unique patient population.