Intravenous Tranexamic Acid Reduces the Incidence of Postoperative Delirium in Elderly Patients Undergoing Hip Arthroplasty: A Retrospective Cohort Study

Introduction

With the accelerating pace of global population aging, the prevalence of hip fractures, acetabular deformities, and degenerative joint diseases has surged. Hip arthroplasty stands as the definitive management for these conditions, with over one million procedures performed annually worldwide, and this figure is projected to rise steadily.^1–6 However, this surgical success is often shadowed by postoperative delirium (POD), an acute neuropsychiatric syndrome characterized by fluctuating disturbances in consciousness and cognition. Studies show that the incidence of POD in elderly patients after lower extremity joint replacement surgery often ranges from 12% to 51%,⁷ and can reach up to 56% in elderly patients undergoing hip arthroplasty.⁸ Clinically, POD is associated with a cascade of adverse outcomes, including elevated perioperative complications, prolonged hospitalization, increased readmission rates, and higher mortality.^9,10 While the pathophysiology of POD is multifactorial and remains elusive, emerging evidence implicates perioperative stressors, particularly substantial blood loss, transfusion requirements, surgical inflammation, and oxidative stress, as pivotal contributors to its development.^11–13

Tranexamic acid (TXA), a synthetic lysine analog, exerts its antifibrinolytic effect by competitively inhibiting the activation of plasminogen, thereby blocking hyperfibrinolysis and preventing fibrin dissolution. As a staple in clinical pharmacotherapy, TXA has been extensively validated to effectively minimize intraoperative blood loss via both intravenous and topical administration, without elevating the risk of postoperative thrombotic events.^14,15 Beyond its established hemostatic properties, accumulating evidence suggests that TXA administration also attenuates levels of inflammatory cytokines following procedures such as total knee arthroplasty and cardiac surgery.^16,17 While the precise anti-inflammatory mechanism of TXA remains to be fully elucidated, it is postulated to involve the inhibition of plasmin and complement activation, as well as the downregulation of inflammatory cytokine expression.^18,19

Consequently, we hypothesized that TXA administration could mitigate the risk of POD in elderly patients by reducing blood loss and transfusion requirements and by concurrently suppressing the inflammatory response. However, direct evidence investigating the relationship between perioperative TXA use and the incidence of POD is currently scarce. Therefore, we conducted this retrospective study to investigate the effect of intravenous TXA on the incidence of POD following hip arthroplasty in elderly patients.

Methods

Ethical Approval

This retrospective study was approved by the Ethics Committee of the Affiliated Hospital of Jiaxing University (Approval No. 2025-LP-282) and registered with the Chinese Clinical Trial Registry (Registration No. ChiCTR2500106018). The study was conducted in strict accordance with the Declaration of Helsinki. Given the retrospective nature of the study, the requirement for written informed consent was waived by the Ethics Committee. To protect patient confidentiality, all data extracted from the electronic medical record system were de-identified prior to analysis.

Sample Size Calculation

Based on a preliminary analysis of 100 patients per group, the incidence of POD was observed to be 27.2% in the TXA group and 38.7% in the control group. Sample size calculation was subsequently performed using PASS 15.0 software based on a Chi-square test. Assuming a two-sided significance level (α) of 0.05, a power of 80% (β = 0.20), and a 1:1 allocation ratio, we determined that a minimum of 259 patients per group (518 in total) was required to detect a statistical difference. To account for a potential 10% data exclusion rate, the target sample size was increased to 288 patients per group (576 in total). Ultimately, 627 eligible participants were enrolled in the final analysis.

Participants

Between January 2017 and October 2025, a total of 847 patients undergoing hip arthroplasty at the Department of Orthopedics, Affiliated Hospital of Jiaxing University, were initially screened. Ultimately, the clinical data of 627 patients meeting the eligibility criteria were retrospectively analyzed. Participants were stratified into two cohorts based on intraoperative management: the TXA group (receiving intravenous TXA, n = 336) and the control group (no TXA administration, n = 291). The inclusion criteria were defined as follows: (1) Age≥ 65 years; (2) Undergoing hip arthroplasty; (3) American Society of Anesthesiologists (ASA) physical status I–III; (4) Surgery performed under general anesthesia. Patients meeting any of the following conditions were excluded: (1) Presence of central nervous system infections, stroke, traumatic brain injury, epilepsy, multiple sclerosis, or other major neurological disorders; (2) Presence of mental or psychological disorders, long-term use of psychotropic drugs, steroids, or hormonal agents, or administration of TXA within one week prior to surgery; (3) Preoperative delirium or dementia; (4) History of major surgery within the preceding year; (5) Incomplete medical records or failure to complete follow-up details (see Figure 1). Cognitive function was assessed one day prior to surgery using the Chinese Mini Mental Status score (CMMS). To ensure diagnostic accuracy, exclusion cut-offs were stratified based on educational attainment: scores≤24 for patients with secondary education or higher,≤20 for those with primary education, and≤17 for illiterate individuals.

Figure 1 Study Flow Diagram. Abbreviations: TXA Group, intraoperative intravenous tranexamic acid group; Control Group, no TXA administration group.

Anesthesia Management

Upon admission to the operating theater, standard monitoring was initiated for all patients, including the establishment of peripheral venous access and invasive arterial blood pressure monitoring. Additionally, the Bispectral Index (BIS) was utilized to monitor anesthetic depth. Induction of general anesthesia was achieved with the intravenous administration of propofol (1.0–2.0 mg/kg), sufentanil (0.2–0.4 μg/kg), and rocuronium (0.6 mg/kg). Tracheal intubation was performed once adequate jaw relaxation was achieved and consciousness and spontaneous respiration were lost. Mechanical ventilation was maintained using intermittent positive pressure ventilation (IPPV) with a tidal volume of 6–8 mL/kg and a respiratory rate of 14–16 breaths/min. End-tidal carbon dioxide partial pressure (PETCO2) was maintained between 35 and 50 mmHg. Anesthesia maintenance comprised a continuous intravenous infusion of remifentanil (0.08–2.00μg/kg/min) for analgesia, combined with sevoflurane inhalation to maintain BIS values between 40 and 65. All patients received patient-controlled intravenous analgesia (PCIA) postoperatively, and the analgesic agent used was sufentanil.

All patients in the TXA group received intravenous TXA at the start of surgery. The specific regimen was as follows: 1.5 g of TXA was dissolved in 100 mL of 0.9% sodium chloride solution and administered intravenously over 20 minutes.

Data Collection

Data were extracted from the electronic medical record system and the postoperative follow-up database. Comprehensive perioperative clinical data were recorded for all participants. The primary outcome was the incidence of POD within the first 7 postoperative days. Secondary outcomes encompassed intraoperative blood loss, transfusion volumes, and the Numerical Rating Scale (NRS) pain score at 24 hours postoperatively [ranging from 0 (no pain) to 10 (unbearable pain)]. Additionally, laboratory parameters were assessed at the most recent preoperative time point and on postoperative day 2. These included liver function [alanine aminotransferase (ALT), aspartate aminotransferase (AST)], renal function [creatinine (Cr), blood urea nitrogen (BUN)], coagulation profiles [prothrombin time (PT), activated partial thromboplastin time (APTT), D-dimer], and inflammatory markers [high-sensitivity C-reactive protein (hs-CRP), interleukin-6 (IL-6)]. Adverse events included the incidence of coagulation abnormalities (defined as PT prolonged by >3 s or APTT by >10 s beyond the normal range) and postoperative thromboembolic events (specifically pulmonary embolism, cerebral infarction, myocardial infarction, and deep vein thrombosis of the lower extremity). Notably, all instances of lower-extremity deep vein thrombosis were identified through bedside Doppler ultrasound screening to ensure diagnostic accuracy.

Assessment of Postoperative Delirium

Data regarding mental status were retrieved from patient medical records, and the Confusion Assessment Method (CAM) was utilized as the diagnostic criterion.²⁰ A positive diagnosis of POD required the presence of both (1) an acute onset or fluctuating course of mental status and (2) inattention, accompanied by either (3) an altered level of consciousness or (4) disorganized thinking. Due to the pragmatic nature of the clinical workflow in this study, delirium assessments were performed by the duty physicians during their respective shifts. Since the patient cohorts assessed by different physicians did not overlap, independent ratings from multiple physicians for the same patient could not be obtained; consequently, inter-rater reliability was not evaluated. To mitigate potential inter-observer bias, however, all participating medical staff underwent regular, standardized training and proficiency testing in neuropsychological assessment protocols.

Statistical Analysis

SPSS 25.0 (IBM Corp, Armonk, NY, USA) and GraphPad Prism 9.0 (GraphPad Software, San Diego, CA, USA) were employed for the statistical processing of the data. The normality of continuous variables was assessed using the Shapiro–Wilk test, and homogeneity of variances was assessed using Levene’s test. Continuous variables that followed a normal distribution and had equal variances are presented as mean ± standard deviation ( ± s). Intergroup comparisons were performed using the independent samples t-test; if equal variances could not be assumed, Welch’s t-test was used instead. Non-normally distributed data are given as the median (interquartile range [IQR]) and were compared using the Mann–Whitney U-test. Count data are given as number of cases (%) and any differences are evaluated by employing the chi-squared test. Initially, univariate regression analysis was performed to assess the association of all variables with POD. Variables with a P ≤ 0.2 in the univariate regression analysis were tested for multicollinearity using the variance inflation factor (VIF). Variables with a VIF > 5.0 were considered to have multicollinearity and were excluded. Variables with a VIF ≤ 5.0 were included in the logistic regression analysis to adjust for confounding factors. The results were displayed as odds ratio (OR) and 95% confidence intervals (CI). A P-value less than 0.05 was considered to be statistically significant.

Results

Based on the inclusion and exclusion criteria, a total of 627 elderly patients undergoing hip arthroplasty were ultimately included in the final analysis. These participants were stratified into the TXA group (n = 336) and the control group (n = 291) (Figure 1). No statistically significant differences were observed regarding age, gender, body mass index (BMI), ASA classification, history of smoking or alcohol consumption, preoperative comorbidities, or CMMS scores, indicating that the two cohorts were comparable (Table 1).

Table 1 Baseline Characteristics of the Study Population

All patients included in the study completed the delirium assessment within the first 7 postoperative days. We further confirmed the occurrence of POD by reviewing each patient’s electronic medical records. A total of 193 patients developed POD. The incidence was significantly lower in the TXA group compared to the control group (22.6% vs 40.2%; P < 0.001). Comparisons regarding the most recent preoperative laboratory indices [including alanine aminotransferase (ALT), aspartate aminotransferase (AST), creatinine (Cr), blood urea nitrogen (BUN), prothrombin time (PT), activated partial thromboplastin time (APTT), and D-dimer] revealed no statistically significant differences between the two groups (Table 2). Furthermore, high-sensitivity C-reactive protein (hs-CRP) and interleukin-6 (IL-6) were utilized as specific indicators of systemic inflammation. No significant differences were observed in preoperative hs-CRP or IL-6 levels between the two groups. However, on postoperative day 2, the TXA group showed significantly lower hs-CRP (P = 0.006) and IL-6 (P = 0.001) levels compared to the control group (Figure 2). Regarding laboratory findings on postoperative day 2, levels of ALT and AST remained comparable between the groups. Conversely, the TXA group exhibited significantly lower PT (P = 0.019), APTT (P < 0.001), and D-dimer (P < 0.001) levels compared to the control group (Table 3). Key intraoperative variables are illustrated in Figure 3. The TXA group demonstrated significantly reduced intraoperative blood loss (P < 0.0001) and transfusion volume (P = 0.004) compared to the control group. Moreover, upon comparing transfusion rates, a significant difference persisted between the cohorts [9.2% (TXA group) vs 17.5% (control group); P = 0.002]. No significant differences were observed between the two groups in the consumption of sufentanil and remifentanil during anesthesia, or in the incidence of intraoperative hypotension (defined as mean arterial pressure < 65 mmHg lasting at least 1 minute) (Figure 3). Postoperative sufentanil consumption and the incidence of thromboembolic adverse events were comparable between the two groups (Table 3).

Table 2 Comparison of Preoperative Laboratory Test Results in the Two Groups

Table 3 Comparison of Postoperative Data and Adverse Events in the Two Groups

Figure 2 Comparison of Preoperative and Postoperative Inflammatory Indicators Between the Two Groups. (A) Comparison of preoperative IL-6 levels between the two groups. (B) Comparison of postoperative IL-6 levels between the two groups. (C) Comparison of preoperative hs-CRP levels between the two groups. (D) Comparison of postoperative hs-CRP levels between the two groups. ** indicates P < 0.01. Abbreviations: hs-CRP, high-sensitivity c-reactive protein; IL-6, interleukin-6; TXA Group, intravenous tranexamic group; Control Group, no intravenous tranexamic group, ns, no significance.

Figure 3 Comparison of Intraoperative Conditions Between the Two Groups. (A) Comparison of operation duration between the two groups. (B) Comparison of intraoperative blood loss between the two groups. (C) Comparison of intraoperative transfusion volume between the two groups. (D) Comparison of intraoperative transfusion rate between the two groups. (E) Comparison of sufentanil consumption during anesthesia between the two groups. (F) Comparison of remifentanil consumption during anesthesia between the two groups. (G) Comparison of the incidence of intraoperative hypotension between the two groups. **indicates P < 0.01, ****indicates P < 0.0001. Abbreviations: TXA Group, intravenous tranexamic group; Control Group, no intravenous tranexamic group; ns, indicates no significance.

Univariate regression analysis identified TXA administration as a significant protective factor against POD (OR 0.435; 95% CI [0.307–0.615]; P < 0.001). Conversely, postoperative NRS scores (OR 1.317; 95% CI [1.096–1.582]; P = 0.003) and preoperative comorbid COPD (OR 1.618; 95% CI [1.057–2.475]; P = 0.027) were significantly associated with an elevated risk of POD (Table 4). Established literature indicates that smoking history, alcohol consumption, and postoperative hs-CRP levels influence the development of POD.^21–23 Variables with P ≤ 0.2 in the univariate regression were subjected to multicollinearity analysis. Using a variance inflation factor (VIF) threshold of > 5.0 to identify potential multicollinearity, the NRS score (VIF = 5.294) was found to exhibit multicollinearity and was subsequently excluded (Table 5). Other variables with P ≤ 0.2 were entered into a multivariable binary logistic regression model. After adjusting for potential covariates, including COPD, smoking history, alcohol consumption, and gender, these factors showed no significant association with POD. However, preoperative comorbid COPD remained an independent risk factor for POD (OR, 1.643; 95% CI; 1.043–2.590; P=0.032). Multivariable analysis also confirmed that TXA administration was a protective factor against POD (OR, 0.441; 95% CI, 0.304–0.639; P < 0.001) (Figure 4).

Table 4 Univariate Regression Analysis of Factors Associated with Postoperative Delirium

Table 5 Multivariate Logistic Regression Analysis and Collinearity Diagnostics for Postoperative Delirium

Figure 4 Association of COPD and TXA with POD after Multivariate Regression Analysis. Abbreviations: COPD, Chronic Obstructive Pulmonary Disease; TXA, tranexamic; OR, odds ratio; P-value, the P-value of multivariate logistic regression.

Discussion

In this retrospective study involving 627 elderly patients aged 65 years or older undergoing hip arthroplasty at a major medical center in China, the incidence of POD was 22.6% in the group receiving intraoperative intravenous TXA. In comparison, the incidence in the control group was 40.2%. After adjusting for confounders, intraoperative intravenous TXA administration was significantly associated with a lower incidence of postoperative delirium in elderly patients undergoing hip arthroplasty.

Concomitant with population aging, the incidence of orthopedic pathologies such as femoral neck fractures, osteonecrosis of the femoral head, and hip osteoarthritis remains persistently high. Hip arthroplasty represents a pivotal surgical intervention for these conditions, effectively ameliorating pain, restoring joint function, and enhancing quality of life. Nonetheless, given that POD is a prevalent postoperative complication in geriatric patients, the implementation of effective preventive strategies remains an indispensable component of clinical practice. POD is widely characterized as a multifactorial pathology driven by a convergence of mechanisms, including neuroinflammation, neurotransmitter imbalances, cerebral network dysfunction, and endocrine stress responses.^24,25 The overall incidence of POD in our retrospectively reviewed cohort was 30.8%, which aligns with the reported range of 17.8%–56.0% in existing literature.^8,26 Specifically, the incidence of POD within 7 days post-surgery was 22.6% in patients receiving intravenous TXA intraoperatively, compared to 40.2% in patients who did not receive TXA (P < 0.05). We posit that this finding may be attributable to a combination of underlying mechanisms.

Primarily, the neuroinflammatory hypothesis offers a plausible explanation. This theory postulates that physiological stressors, such as trauma, surgery, and infection, trigger an acute systemic inflammatory response. Peripheral inflammatory mediators can compromise the integrity of the blood-brain barrier (BBB), thereby increasing its permeability. These mediators subsequently infiltrate the central nervous system (CNS) through the disrupted BBB, causing neuronal injury and synaptic dysfunction, which ultimately precipitate the onset of POD.²⁷ Surgical trauma and nociceptive stimulation elicit a systemic stress response, characterized by immune activation and an elevated risk of postoperative complications.²⁸ As an antifibrinolytic agent, TXA primarily functions by inhibiting fibrinolysis and stabilizing clot formation to achieve hemostasis.^29,30 However, recent studies have highlighted its additional capabilities, including the attenuation of inflammatory responses, preservation of endothelial barrier integrity, inhibition of apoptosis, and reduction of blood-brain barrier permeability.^31–33 Utilizing a rat model of traumatic hemorrhagic shock, Yue et al³⁴ demonstrated that intravenous TXA administration effectively suppressed the levels of pro-inflammatory cytokines, specifically interleukin-6 (IL-6) and tumor necrosis factor-alpha (TNF-α), in both alveolar lavage fluid and serum. This suggests that TXA may modulate the inflammatory response by regulating cytokine profiles. Consistent with these preclinical findings, clinical investigations involving lower limb arthroplasty have observed that patients treated with TXA exhibited significantly reduced plasma levels of inflammatory markers, including C-reactive protein (CRP) and IL-6.^16,35 In the present study, hs-CRP and IL-6 were utilized as surrogate markers to assess the systemic inflammatory status. Our results indicated that on postoperative day 2, patients receiving intraoperative intravenous TXA exhibited significantly lower concentrations of inflammatory markers compared to patients where TXA was not used (P < 0.05). Specifically, the respective concentrations were: hs-CRP [39.90 (24.60, 47.20) mg/L vs 39.90 (31.80, 58.10) mg/L] and IL-6 [71.36 (35.65, 128.68) pg/mL vs 89.56 (56.82, 135.69) pg/mL]. This indicates that elderly patients treated with TXA had lower levels of systemic inflammatory response after undergoing hip arthroplasty. Furthermore, Ormseth et al³¹ demonstrated that TXA reduces blood-brain barrier (BBB) permeability by modulating complement activation and leukocyte function. Given that BBB hyperpermeability is established as a critical pathophysiological mechanism driving POD,²⁷ the decreased POD risk in the TXA group might be associated with such protective actions. Regarding alternative administration routes, a randomized controlled trial confirmed that the oral administration of TXA is similarly effective in attenuating postoperative inflammatory levels.¹⁵ This finding substantiates the theoretical basis for the more flexible clinical application of TXA.

Secondly, TXA is a relatively inexpensive and safe conventional antifibrinolytic agent. Its primary mechanism of action involves the competitive inhibition of the lysine-binding sites on plasminogen, thereby exerting its antifibrinolytic effect. For this reason, it has long been clinically employed for the treatment and prevention of hemorrhage in settings such as trauma, obstetrics, and major surgical procedures.³⁶ In terms of administration routes, various methods, including intravenous, topical, and oral administration, have been confirmed to effectively attenuate the risk of perioperative bleeding. Furthermore, a meta-analysis indicated that TXA demonstrates significant hemostatic efficacy, regardless of whether a single route or a combined multimodal approach is employed.³⁷ According to the findings of DeFrancesco et al,³⁸ among patients aged > 45 years undergoing non-cardiac surgery, the TXA-treated cohort demonstrated a lower risk of perioperative bleeding, with no significant difference in thrombotic risk compared to patients who did not receive TXA. Parallel to these observations, our study demonstrated that intraoperative blood loss and transfusion volumes were significantly curtailed in patients receiving intraoperative intravenous TXA. Notably, perioperative massive blood loss and blood transfusion are well-established risk factors precipitating the onset of POD.^12,39,40 The underlying mechanisms are primarily associated with the accumulation of lipid peroxidation products, elevated levels of inflammatory cytokines, and reduced cerebral perfusion.^41,42 TXA was included in the World Health Organization’s Model List of Essential Medicines in 2011. However, notwithstanding its potent antifibrinolytic efficacy, concerns persist within the academic community regarding potential thromboembolic events, such as deep vein thrombosis, myocardial infarction, pulmonary embolism, and cerebral embolism. Even the formation of postoperative cerebral microthrombi could potentially elevate the incidence of POD. However, a growing body of recent evidence corroborates that the perioperative administration of TXA significantly diminishes total blood loss, without exerting a notable impact on the incidence of postoperative thrombotic or vascular occlusive adverse events.^43–45 Our study revealed that patients receiving TXA exhibited lower postoperative D-dimer levels and no increase in the incidence of thromboembolic events compared to patients where TXA was not used. These findings verify that intraoperative TXA administration does not elevate the risk of postoperative thrombosis, thereby supporting its safety in clinical application.^14,15

Furthermore, an association between acute postoperative pain and POD is widely acknowledged. Research suggests that postoperative pain may disrupt sleep patterns, thereby inducing or exacerbating POD.⁴⁶ Notably, a dose-response meta-analysis identified a correlation between the intensity of postoperative pain and the risk of both the onset and progression of POD.⁴⁷ Multiple prior investigations have demonstrated that the intravenous injection of TXA reduces postoperative incision drainage and local hematoma formation, subsequently improving early postoperative pain scores.^48–50 In the present study, the NRS was employed to assess postoperative pain. Our analysis suggests that intraoperative intravenous TXA may indeed exert a perceptible analgesic effect in patients undergoing hip arthroplasty. This finding aligns with the results reported by Goldstein et al⁵¹ Although the precise underlying mechanisms warrant further exploration, these data highlight the potential value of TXA in optimizing perioperative pain management strategies.

As a common postoperative complication in elderly patients, POD is critical to prevent for improving prognosis. Currently, pharmacological interventions (such as dexmedetomidine) and regional anesthesia (such as peripheral nerve blocks) represent the primary focuses of clinical investigation. Dexmedetomidine, a sedative α2-adrenoceptor agonist, exerts neuroprotective effects through mechanisms such as attenuating neuroinflammation, inhibiting apoptosis, and mitigating blood-brain barrier (BBB) disruption within the central nervous system.⁵² Peripheral nerve blocks can effectively alleviate postoperative pain and reduce opioid consumption, similarly demonstrating preventive efficacy against POD.^53,54 Compared with the aforementioned strategies, the advantages of TXA lie in its pleiotropic effects, including antifibrinolytic, anti-inflammatory, and analgesic properties, as well as its relatively low administration cost. Currently, there is a lack of comparative studies between TXA and dexmedetomidine or regional anesthesia regarding their efficacy in reducing POD incidence, an area that warrants further exploration in future research.

Several limitations of the present study must be acknowledged. First, As a single-center retrospective study lacking randomization, although multivariate logistic regression was employed to adjust for known confounders, we still cannot entirely exclude the potential impact of unmeasured or unknown factors on the results. For example, preoperative anxiety is a known risk factor for postoperative delirium.^55,56 Due to the retrospective nature of this study, standardized quantitative assessment data for preoperative anxiety could not be obtained. Therefore, preoperative anxiety, as an unmeasured confounder, may still have some influence on the study findings. Furthermore, the triggering criteria for TXA administration were primarily determined by the attending surgeons’ clinical preferences and individualized assessments of contraindications, rather than pre-established objective indicators. In clinical practice, the surgeon’s decision was influenced by multiple factors, including personal experience, concern about bleeding risk, and fear of potential adverse drug reactions. These factors represent the real‑world basis for triggering the use of hemostatic agents. Consequently, the trigger rules for TXA administration in this study exhibited considerable irregularity and heterogeneity. As a retrospective study, we could not prospectively define or standardize these trigger rules, and this limitation may introduce potential bias into the results. Secondly, our findings are derived from clinical data analysis and lack further basic experiments to verify the specific mechanism by which TXA influences POD. Thirdly, to maintain data integrity, this study only evaluated inflammatory marker levels on postoperative day 2, thus lacking continuous tracking of the longitudinal dynamic changes in postoperative inflammation. Additionally, caution should be exercised when directly extrapolating the obtained results to other surgeries in elderly patients. Finally, the follow-up duration of this study was relatively short, limiting the assessment to the incidence of delirium within the first 7 postoperative days. Consequently, we failed to track critical clinical endpoints, including the duration of POD, long-term cognitive outcomes, and long-term survival. Despite these limitations, we advocate for future large-scale, multi-center, prospective randomized controlled trials to further validate the efficacy and underlying mechanisms.

Conclusion

Our retrospective analysis revealed that in elderly patients undergoing hip arthroplasty, intraoperative intravenous administration of TXA was associated with reduced inflammation, less blood loss, lower pain scores, and a decreased incidence of postoperative delirium. These findings shift the traditional clinical perception of TXA; it should no longer be viewed solely as an antifibrinolytic agent, but rather as a drug with potential value in mitigating inflammation and improving cognitive outcomes in the perioperative period. These findings offer novel mechanistic insights and therapeutic strategies for the prevention of POD in elderly patients, warranting further attention from both clinicians and researchers.