Abstract
Purpose
-
This study sought to determine if the use of tranexamic acid (TXA) in preexisting thromboembolic risk patients undergoing total joint arthroplasty (TJA) was linked to an increased risk of death or postoperative complications.
Methods
-
We conducted a comprehensive search for studies up to May 2023 in PubMed, Web of Science, EMBASE, and the Cochrane Library. We included randomized clinical trials, cohort studies, and case–control studies examining the use of TXA during TJA surgeries on high-risk patients. The Cochrane Risk of Bias instrument was used to gauge the excellence of RCTs, while the MINORS index was implemented to evaluate cohort studies. We used mean difference (MD) and relative risk (RR) as effect size indices for continuous and binary data, respectively, along with 95% CIs.
Results
-
Our comprehensive study, incorporating data from 11 diverse studies involving 812 993 patients, conducted a meta-analysis demonstrating significant positive outcomes associated with TXA administration. The findings revealed substantial reductions in critical parameters, including overall blood loss (MD = −237.33; 95% CI (−425.44, −49.23)), transfusion rates (RR = 0.45; 95% CI (0.34, 0.60)), and 90-day unplanned readmission rates (RR = 0.86; 95% CI (0.76, 0.97)). Moreover, TXA administration exhibited a protective effect against adverse events, showing decreased risks of pulmonary embolism (RR = 0.73; 95% CI (0.61, 0.87)), myocardial infarction (RR = 0.47; 95% CI (0.40–0.56)), and stroke (RR = 0.73; 95% CI (0.59–0.90)). Importantly, no increased risk was observed for mortality (RR = 0.53; 95% CI (0.24, 1.13)), deep vein thrombosis (RR = 0.69; 95% CI (0.44, 1.09)), or any of the evaluated complications associated with TXA use.
Conclusion
-
The results of this study indicate that the use of TXA in TJA patients with preexisting thromboembolic risk does not exacerbate complications, including reducing mortality, deep vein thrombosis, and pulmonary embolism. Existing evidence strongly supports the potential benefits of TXA in TJA patients with thromboembolic risk, including lowering blood loss, transfusion, and readmission rates.
Introduction
Total joint arthroplasty (TJA), encompassing procedures like total knee arthroplasty (TKA) and total hip arthroplasty (THA), has gained popularity because of its success in treating end-stage knee or hip arthritis (1). The prevalence of these procedures is evident in the USA, with over 700 000 TKAs and 300 000 THAs conducted annually, and these numbers are expected to rise substantially in the near future (2). However, the success of TJA is accompanied by a challenge – significant perioperative blood loss, estimated to range from 1450 mL to 1790 mL without specific interventions (3).
Understanding the determinants of preexisting thromboembolic risk is pivotal for individuals undergoing TJA. Patients with a history of venous thromboembolism (VTE), myocardial infarction (MI), cerebrovascular accident, and transient ischemic attack present a unique challenge due to their heightened susceptibility to thromboembolic events. These factors intricately contribute to the preexisting thromboembolic risk profile. It is imperative to delve into these determinants comprehensively, exploring their individual and collective impact on patients undergoing TJA.
Tranexamic acid (TXA), a potent antifibrinolytic agent, has become crucial in mitigating perioperative blood loss during TJA (4). However, its application requires careful consideration, particularly in situations with a heightened risk of thromboembolism. In some instances, the use of TXA may be contraindicated due to the potential increased risk of thrombotic complications. This critical concept underlines the importance of balancing the benefits of reduced blood loss against the potential risks, especially in patients with a history of VTE, MI, cerebrovascular accident, and transient ischemic attack (5).
This systematic review and meta-analysis were designed to evaluate the efficacy and safety of TXA in high-risk patients undergoing TJA, aiming to guide the future TJA surgery clinical treatment program.
Methods
This meta-analysis has been reported per PRISMA and AMSTAR Guidelines (6). The protocol for this study was registered in PROSPERO (CRD:42022372020).
Search strategy
A primary search for relevant studies was conducted in the following electronic databases, including PubMed, Web of Science, EMBASE, and the Cochrane Library, from inception to May 2023. The retrieval was primarily conducted using subject terms and free words, including keywords such as TXA, thromboembolism, arthroplasty, etc. The specific retrieval strategy can be found in Supplementary Table 1 (see section on supplementary materials given at the end of this article). No limitation was applied to language or publication status. Besides, all relevant publications and review, reference lists were manually searched to identify potential studies.
Eligibility criteria
Inclusion criteria were as follows: (i) patients underwent TKA or THA; (ii) the experimental group receiving TXA as an intervention on top of standard treatment, while the control group may receive a placebo or standard care regimen; (iii) the design of included studies should be randomized controlled trials (RCTs), cohort studies (prospective or retrospective); (iv) patients were with concomitant high-risk diseases (such as VTE, MI, atrial fibrillation, stroke, renal diseases, and so on); and (v) the included study should report at least one of the following outcomes: (i) Primary outcomes: total blood loss, deep vein thrombosis, pulmonary embolism, and all-cause mortality. (ii) Secondary outcomes: transfusion rate, unplanned readmission, MI, and stroke. It is noteworthy that no restrictions were imposed on the following parameters: (i) administration scheme: there are no limitations on the method, timing, or dosage of TXA administration; (ii) outcome evaluation time: the time frame for evaluating study outcomes was not restricted.
Exclusion criteria were: (i) studies provided insufficient data; (ii) case reports, animal experiments, conference papers, and some types of crossover study designs; and (iii) studies with disparate baseline information between research groups and studies with a sample size of less than 50. We applied no restriction on the language of publication.
Study selection
We utilized the online software ‘Rayyan’ for literature screening. After completing automated deduplication and initial literature screening, two researchers (XD and ML) independently reviewed the titles and abstracts of all studies to identify those warranting further assessment. Subsequently, the full texts of initially included studies were scrutinized by evaluators XD and QY to determine the final inclusion criteria. Any disagreements were resolved through discussion with a third reviewer (JT).
Data extraction
Two researchers (QY, JJ) independently extracted data from the included studies. The following information was extracted: basic characteristics of the included studies (authors, publication year, location, study design, and mean age), concomitant high-risk diseases, TXA-related information (TXA dose, administration method, and timing), control group interventions, and surgical information (surgical type). Clinical outcomes were also recorded: total blood loss, transfusions, unplanned readmission, mortality, and thromboembolic events, with a specific focus on the criteria for thrombotic risk events. Any disagreements were resolved by discussion with a third author (JT). All extracted data were carefully entered into a predefined standardized Excel (Microsoft Corporation, USA) file.
Quality assessment
The methodological quality of the included RCTs was meticulously evaluated using the Cochrane Risk of Bias Tool (RoB 2) (7), a widely recognized and standardized instrument developed by the Cochrane Collaboration for assessing the risk of bias in clinical trials. This tool encompasses key domains, including random sequence generation, allocation concealment, blinding of participants and personnel, incomplete outcome data, selective reporting, and other sources of bias. For each domain, the RoB Tool enables a systematic and transparent evaluation, categorizing studies into low, unclear, or high risk of bias. The comprehensive nature of this tool allows reviewers to critically appraise the internal validity of individual studies, providing a robust foundation for evidence synthesis and interpretation.
The methodological quality of cohort studies was assessed using the methodological index for the non-randomized studies (MINORS) (8). The checklist involved 12 components, and each component was graded on a scale from 0 to 2, where the maximum score for a comparative study was 24. A score of 15 or less was considered poor, 15–18 was considered moderate, and 19 or more was deemed high quality. Two reviewers (XD and JJ) independently assessed the methodological quality. Any disagreements were resolved through a discussion with a third reviewer (JT).
In addition, we conducted a comprehensive assessment of the quality of evidence using GRADEpro in accordance with the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) guidelines (9). Each outcome underwent meticulous evaluation, resulting in high, moderate, low, or deficient quality evidence ratings. These ratings were determined based on study limitations, inconsistent results, the indirectness of evidence, imprecision, and the potential impact of publication bias. The assessment process involved two independent researchers (JJ, XD), and any discrepancies were resolved through constructive discussion with a third researcher (YL). When outcomes involved a combination of different study designs, we chose a conservative estimation for the quality level, considering the lower two grades.
Statistical analysis
Review Manager version 5.4 (The Cochrane Collaboration, Oxford, UK) was used for the statistical analyses. The mean difference (MD) with their 95% CIs was summarized for continuous outcomes, and the relative risk (RR) with 95% CIs was for dichotomous outcomes. Based on the background of the specific research outcome indicators and the effectiveness of statistical analysis, we have chosen the appropriate effect model (random-effects model/fixed-effects model) for each outcome indicator (10).
Heterogeneity between trials was evaluated using Cochrane’s Q statistic test and I 2. P < 0.05 or I 2 >50% indicated significant heterogeneity among the studies.
Forest plots were used to graphically represent the difference in outcomes of groups of TXA and control for all included studies. Sensitivity analysis was conducted by removing one of the included studies at a time, followed by re-analysis to determine the robustness of the study. Subgroup analysis to explore the origins of heterogeneity was also planned. P values <0.05 were considered statistically significant. Additionally, if more than ten studies were included for a particular outcome measure, we plan to employ a funnel plot to explore publication bias.
Results
Literature search
Through the initial literature search in electronic databases, a total of 1125 articles were identified. After removing 287 duplicate reports, the titles and abstracts of the remaining studies were assessed. Following the application of screening criteria and the exclusion of obviously irrelevant studies, a preliminary selection of 27 articles was made for full-text review. During the full-text screening, studies were excluded due to unclear embolism risk (n = 5), severe complications (n = 4), multi-drug comprehensive treatment regimens (n = 6), and unclear study designs (n = 1). Ultimately, 11 relevant studies were included (Fig. 1) (11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21).
The flowchart of literature screening.
Citation: EFORT Open Reviews 9, 6; 10.1530/EOR-23-0140
Characteristics of included studies
The basic characteristics of the 11 studies in the systematic review are presented in Table 1. 812,993 patients were included, 429,166 were administered TXA, and 383,827 received a placebo or standard care. The combined age of the participants at the time of surgery ranged from 62.0 to 80.0 years. Among the included studies, seven (11, 12, 13, 14, 15, 17, 19) were conducted in the USA, three (16, 18, 20) in China, and one (21) in Korea. The date of publication of the included studies varied from 2014 to 2022. Most eligible studies were retrospective cohort studies (11, 12, 13, 14, 15, 18, 19, 21), with two RCTs (16, 20) and one case–control study also included (17). The dosages in three studies (11, 12, 18) were weight-based (including 15 mg/kg, 10 mg/kg, and 10–20 mg/kg). In other studies (13, 14, 15, 16, 17, 19, 20, 21) with fixed doses, the total dosage was controlled within 3 g. The administration was conducted through both intravenous injection and intraarticular delivery. Multiple dosages were primarily administered during skin incision and closure. All studies had a saline solution control group. The procedures for surgery were described in detail in all included studies (11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21). In seven studies, (11, 12, 13, 15, 17, 18, 19) the participants underwent TJA, two (14, 16) underwent THA, and two (20, 21) underwent TKA. The main intervention for the control group was Standard postoperative care and placebo therapy. The criteria for defining thrombotic risk events across studies are generally similar. Low-molecular-weight heparin was used to prevent thromboembolic events in nine studies. The high-risk patients were mainly concomitant with the following diseases: VTE (including DVT, PE) (12, 13, 14, 15, 17, 19, 20, 21), MI (19, 20, 21), atrial fibrillation (13, 14, 18), stroke (14, 20), coronary artery bypass graft (14, 15, 16, 19), and renal diseases (13).
Characteristics of included studies.
| Study | Country | Study design | TXA vs placebo | Surgical procedures | Comorbidities | TXA administration | Transfusion criteria | DVT prophylaxis | |
|---|---|---|---|---|---|---|---|---|---|
| Sample size | Mean age | ||||||||
| Goh et al. (11) | USA | CS | 240/1883 | 65.3/68.6 | TKA, THA | Coagulopathy | i.v. 15 mg/kg | Hb < 7.5 g/dL | LMWH |
| Heller et al. (12) | USA | CS | 1678/1157 | 65.0/66.4 | TKA, THA | VTE | i.v. 10 mg/kg | NR | LMWH |
| Joo et al. (21) | Korea | CS | 298/212 | 70.1/69.6 | TKA | VTE, MI, CVOD | i.a. 3 g | Hb < 8 g/dL | LMWH |
| Poeran et al. (13) | USA | CS | 404974/360037 | 62.0/65.4 | TKA, THA | VTE, renal disease, AF | i.v. 1–3 g | NR | NR |
| Porter et al. (14) | USA | CS | 642/997 | 79.9/80.9 | THA | VTE, MI, stroke, AF, prothrombotic state, atrial flutter, CABG or CAS | i.v. 1–2 g | Hb <10 g/dL | NR |
| Porter et al. (15) | USA | CS | 20501/17719 | 71.3/71.3 | TKA, THA | DVT, PE, MI, CVA, a prothrombotic state | i.v. 2 g | Hb <8 g/dL | LMWH |
| Qiu et al. (16) | China | RCT | 55/47 | 71.2/69.0 | THA | Cardiovascular or cerebrovascular events | i.a. 3 g | Hb <8 g/dL | NR |
| Sabbag et al. (17) | USA | CS | 258/1362 | 68.9/69.6 | TKA, THA | VTE | i.v. 2 g | NR | LMWH |
| Tang et al. (18) | China | CS | 246/216 | NR | TKA, THA | AF | i.v. 10–20 mg/kg | NR | NR |
| Whiting et al. (19) | USA | CS | 240/162 | 69.5 | TKA, THA | DVT, PE, MI, CVA, CABG, or prothrombotic condition | i.v. 1 g | Hb <10 g/dL | Aspirin, warfarin, or LMWH |
| Yen et al. (20) | China | RCT | 34/35 | 69.4/69.7 | TKA | VTE, MI, stroke | i.a. 3 g | Hb <8 g/dL | Exoxaparin |
NR: not reported; LMWH, low molecular weight heparin; Hb, Hemoglobin; DVT, deep vein thrombosis; PE, pulmonary embolus; MI, myocardial infarction; CVA, cerebrovascular accident; CVOD, cerebral vascular ocular disease; CAS, coronary artery stent; AF, atrial fibrillation; CS, cohort study; CABG, coronary artery bypass graft; RCT, randomised controlled trial; TKA, total knee arthroplasty; i.a., intraarticular.
Quality assessment
All studies conducted a risk of bias assessment. The two RCTs (16, 20) had explicit inclusion and exclusion criteria and clearly described the randomization methods. Allocation concealment was reported in only one study (14). Additionally, nine studies were observational studies, including cohort studies and case–control studies (11, 12, 13, 14, 15, 17, 18, 19, 21). Their quality was assessed using the MINOR criteria, with the average score for included observational studies being 19.3 ± 1.73 (range 17–21). The details of the assessment criteria are provided in Table 2.
Methodological quality of each study.
| Study | Criteria* | Total score | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | ||
| Goh et al. (11) | 2 | 2 | 0 | 1 | 2 | 2 | 2 | 0 | 2 | 2 | 2 | 2 | 19 |
| Heller et al. (12) | 2 | 2 | 0 | 2 | 2 | 2 | 2 | 1 | 1 | 2 | 0 | 2 | 19 |
| Joo et al. (21) | 2 | 2 | 0 | 2 | 2 | 2 | 2 | 0 | 2 | 2 | 0 | 2 | 17 |
| Poeran et al. (13) | 2 | 2 | 0 | 2 | 2 | 2 | 2 | 1 | 2 | 2 | 0 | 2 | 19 |
| Porter et al. (14) | 2 | 2 | 0 | 2 | 2 | 2 | 2 | 2 | 2 | 2 | 2 | 2 | 21 |
| Porter et al. (15) | 2 | 2 | 0 | 1 | 2 | 2 | 2 | 0 | 2 | 2 | 2 | 2 | 19 |
| Sabbag et al. (17) | 2 | 2 | 2 | 2 | 2 | 2 | 2 | 0 | 2 | 0 | 2 | 2 | 20 |
| Tang et al. (18) | 2 | 2 | 0 | 1 | 2 | 2 | 2 | 2 | 2 | 2 | 2 | 2 | 21 |
| Whiting et al. (19) | 2 | 2 | 0 | 1 | 2 | 2 | 2 | 0 | 2 | 2 | 2 | 2 | 19 |
*Criteria are as follows: 1, clearly stated aim; 2, inclusion of consecutive patients; 3, prospective collection of data; 4, endpoints appropriate for aim; 5, unbiased assessment of endpoints; 6, appropriate follow-up period; 7, lost to follow-up <5%; 8, prospective calculation of study size; 9, adequate control group; 10, contemporary groups; 11, baseline equivalence of groups; 12, adequate statistical analysis.
Results of meta-analysis
Total blood loss
Four studies (11, 12, 18, 20), including 5489 patients, reported the total blood loss. In total, 2198 patients were treated with TXA and 3291 with placebo. TXA administration was associated with significantly less total blood loss than placebo (MD = –237.33; 95% CI: (–425.44, –49.23); P < 0.01; I 2 = 97%; grade: low) (Fig. 2 and Table 3). The sensitivity analysis illustrated similar conclusions (Supplementary Table 2).
Association between TXA use and total blood loss (11, 12, 18, 20).
Citation: EFORT Open Reviews 9, 6; 10.1530/EOR-23-0140
Quality assessment of study outcomes based on GRADE.
| Outcome index | Combined studies, n | Total sample size | Effect size, RR (95% CI) | Evaluation based on | Overall quality | |
|---|---|---|---|---|---|---|
| OS | RCT | |||||
| Total blood loss | 4 | 5,489 | –237.33* (–425.44, –49.23) | Moderate | Low | Low |
| Transfusion rates | 8 | 124,779 | 0.45 (0.34–0.60) | Moderate | Moderate | Moderate |
| 90-day unplanned readmission | 4 | 129,551 | 0.86 (0.76–0.97) | Moderate | Low | Low |
| Mortality | 5 | 133,351 | 0.53 (0.24–1.13) | Moderate | – | Moderate |
| Deep vein thrombosis | 7 | 132,067 | 0.69 (0.44–1.09) | Moderate | Low | Low |
| Pulmonary embolism | 6 | 131,731 | 0.73 (0.61–0.87) | High | Moderate | Moderate |
| Myocardial infarction | 5 | 130,918 | 0.47 (0.40–0.56) | High | – | High |
| Stroke | 5 | 130,918 | 0.73 (0.59–0.90) | Moderate | – | Moderate |
*Value is mean difference.
CI, confidence interval; RR, relative risk; RCT, randomized controlled trial; OS, observational study.
Transfusion rate
Eight studies (11, 12, 13, 16, 18, 19, 20, 21) provided available data concerning transfusion rate. Of which, 60 468 patients were treated with TXA and 64 311 with placebo. The pooled analysis enlightened that the TXA group had a significantly lower transfusion rate than the placebo group (RR = 0.45; 95% CI (0.34, 0.60); P < 0.01; I 2 = 87%, grade: moderate) (Fig. 3 and Table 3). The results of the sensitivity analysis are relatively stable (Supplementary Table 3).
Ninety-day unplanned readmission
Four studies (11, 13, 15, 16) involving 129 551 patients reported the rates of 90-day unplanned readmission. Of these, 409 patients (0.66%) were reported to have an incidence of unplanned readmission in the TXA group, whereas 639 patients (0.95%) in the placebo group had an incidence of unplanned readmission. TXA was associated with decreased risk of unplanned readmission (RR = 0.86; 95% CI (0.76, 0.97); P = 0.01; I 2 = 16%; grade: Llow) (Fig. 4 and Table 3). The sensitivity analysis indicates that even after excluding the RCT (16), the study results remain stable (Supplementary Table 4).
Association between TXA use and 90-day unplanned readmission (11, 13, 15, 16).
Citation: EFORT Open Reviews 9, 6; 10.1530/EOR-23-0140
Postoperative complications
Mortality
Five studies (11, 12, 13, 14, 15) involving 133 351 patients reported death events occurring postoperatively. There were 111 (0.17%) deaths in the TXA group compared to 296 (0.43%) in the control group. The meta-analysis results showed no significant difference in the mortality risk between the two groups (RR = 0.53; 95% CI (0.24, 1.13); P = 0.10; I 2 = 85%; grade: moderate) (Fig. 5 and Table 3). The sensitivity analysis illustrated similar conclusions (Supplementary Table 5). However, after excluding the study by Poeran et al. (13), heterogeneity significantly decreased.
Deep vein thrombosis
Seven studies (12, 13, 14, 15, 16, 19, 21) involving 64 533 patients reported deep vein thrombosis (DVT) rates. Of these, 244 patients (0.38%) were reported to have DVT in the TXA group, whereas 393 patients (0.58%) in the placebo group had DVT. No significant difference was found in the occurrence of DVT between the two groups (RR = 0.69; 95% CI (0.44, 1.09); P = 0.11; I2 = 66%; grade: low) (Fig. 6 and Table 3). The sensitivity analysis indicates that, even after excluding the RCT (16), the study results remain stable (Supplementary Table 6).
Pulmonary embolism
Six studies (12, 13, 14, 15, 16, 19) involving 131 731 patients reported PE rates. In total, 208 patients (0.32%) were reported to have an incidence of PE, whereas 304 patients (0.45%) in the placebo group had an incidence of PE. There was no difference between TXA and placebo groups in the occurrence of PE (RR 0.73; 95% CI (0.61, 0.87); P < 0.01; I 2 = 18%; grade: moderate) (Fig. 7 and Table 3). The sensitivity analysis indicates that, even after excluding the RCT (16), the study results remain stable (Supplementary Table 7).
Myocardial infarction
Five studies (11, 13, 14, 15, 19) reported the rate of MI. There were 205 (0.33%) MIs in the TXA group compared to 646 (0.95%) in the control group. We found a significantly reduced risk of MI in the TXA users than in nonusers (RR = 0.47; 95% CI (0.40–0.56); P < 0.01; I 2 = 0%; grade: high) (Fig. 8 and Table 3). The sensitivity analysis attests to the stability of the results (Supplementary Table 8).
Stroke
The stroke rate was mentioned in five studies (11, 13, 14, 15, 19). In total, 145 (0.23%) stroke events were in the TXA group compared to 225 (0.33%) in the control group. We found a significantly reduced risk of stroke in the TXA users than in nonusers (RR = 0.73; 95% CI (0.59–0.90); P < 0.01; I 2 = 39%; grade: moderate) (Fig. 9 and Table 3). The sensitivity analysis suggests that, after excluding the study by Poeran et al. (13), the reduction in the incidence of MI with TXA is no longer significant (Supplementary Table 9).
Subgroup analysis
To investigate the sources of heterogeneity, we conducted multiple subgroup analyses, comprising of population (US vs Asia), study design (cohort vs RCT), surgical approach (THA vs TKA), and TXA dosage (1–2 g vs ≥3 g). The subgroup disparities were evaluated using the chi-square test. The subgroup analyses demonstrated that TXA's effectiveness in reducing the risk of transfusion was dose-dependent (P = 0.016). However, its dose had no impact on other outcome variables. By contrast, the subgroup analyses of study design, surgical approach, and population revealed that variables related to outcomes, including transfusion rate, unplanned discharge, deep vein thrombosis, MI, or stroke, were not influenced by it. Table 4 presents the results of the subgroup analyses.
Subgroup analysis. Values are presented as OR (95% CI) and comparisons are ROR (96% CI).
| TR | UPR | DVT | MI | Stroke | |
|---|---|---|---|---|---|
| Country/region | |||||
| USA | 0.43 (0.37, 0.48)* | 0.86 (0.76, 0.98)* | 0.62 (0.38, 1.01) | 0.47 (0.40, 0.56)* | 0.73 (0.59, 0.90)* |
| Asia | 0.35 (0.17, 0.70)* | 0.22 (0.03, 1.88) | 1.33 (0.56, 3.14) | – | – |
| USA vs Asia† | 1.23 (0.6,2.52) | 3.91 (0.49,31.07) | 0.47 (0.17, 1.26) | – | – |
| P | 0.562 | 0.178 | 0.295 | – | – |
| Study design | |||||
| RS | 0.46 (0.34, 0.62)* | 0.86 (0.76, 0.98)* | 0.71 (0.44, 1.13) | 0.47 (0.40, 0.56)* | 0.73 (0.59, 0.90)* |
| RCT | 0.32 (0.14, 0.73)* | 0.22 (0.03, 1.88) | 0.44 (0.04, 4.65) | – | – |
| RS vs RCT† | 1.44 (0.6,3.46) | 3.91 (0.49,31.07) | 1.61 (0.14,18.22) | – | – |
| P | 0.401 | 0.178 | 0.82 | – | – |
| Surgical procedure | |||||
| THA | 0.32 (0.14, 0.75)* | 0.67 (0.24, 1.88) | 0.80 (0.43,1.51) | 0.55 (0.27,1.41) | 1.30 (0.77, 2.22) |
| TKA | 0.54 (0.47, 0.64)* | – | 1.57 (0.62, 3.96) | – | – |
| THA vs TKA† | 0.59 (0.25,1.39) | – | 0.51 (0.17, 1.56) | – | – |
| P | 0.173 | – | 0.39 | – | – |
| Dose of TXA | |||||
| 1–2 g | 0.36 (0.26, 0.50)* | 0.86 (0.74, 1.00) | 0.64 (0.30, 1.37) | 0.50 (0.32, 0.76)* | 0.87 (0.60, 1.27) |
| ≥3 g | 0.54 (0.46, 0.63)* | 0.22 (0.03, 1.88) | 1.33 (0.03, 1.88) | 0.45 (0.08, 2.66) | 1.35 (0.12, 14.77) |
| 1–2 g vs ≥3 g† | 0.67 (0.46, 0.96) | 3.91 (0.49, 31.12) | 0.48 (0.05, 4.36) | 1.11 (0.18, 6.75) | 0.64 (0.06, 7.36) |
| P | 0.016 | 0.179 | 0.206 | 0.94 | 0.898 |
†Values are ROR (96% CI); *Significant difference in intra-group comparisons for each subgroup, P < 0.05.
DVT, deep vein thrombosis; MI, myocardial infarction; PE, pulmonary embolus; RS, retrospective study; TR, transfusion rate; UPR, unplanned readmission.
Discussion
In the context of patients undergoing total joint arthroplasty with preexisting thromboembolic risk, the administration of TXA emerges as a safe and effective approach. Our meta-analysis, incorporating data from eight studies, which are retrospective cohort studies (11, 12, 13, 14, 15, 18, 19, 21), along with two RCTs (16, 20) and one case–control study (17), underscores that TXA use in this population does not elevate the risk of death or thromboembolic complications, including deep vein thrombosis (DVT), pulmonary embolism (PE), mortality, MI, and stroke.
TXA proves to be a valuable tool in reducing blood loss, transfusion rates, and the likelihood of 90-day unplanned readmission, with implications extending to knee and hip surgeries. Recent evidence consistently advocates for the use of TXA in controlling blood loss during TJA surgery, primarily attributing postoperative blood loss to the instantaneous activation of the fibrinolytic cascade (22). TXA's mechanism of action involves preventing fibrinolysis by blocking the lysine-binding site of plasminogen, thereby displacing plasminogen from the surface of fibrin (23). This synthetic fibrinolytic inhibitor competitively inhibits plasmin, plasminogen, and fibrin from combining, directly inhibiting plasmin activity (24). Despite the well-established efficacy and safety of TXA in TJA, its application in patients with preexisting thromboembolic risk has been a subject of limited attention. Our study addresses this gap and brings to light that TXA administration in this high-risk cohort significantly reduces total blood loss and transfusion rates (25).
The study by Poeran et al. utilized a large retrospective approach, comparing 20 051 TJA patients in the TXA group with 852 365 patients in the non-TXA group (18). Notably, they found no association between TXA use and increased thromboembolic events or renal failure. It's important to highlight that the focus of the Poeran et al. study did not center specifically on high-risk patients (26). In recent investigations, other studies have delved into the safety of TXA in relatively healthy TJA patients, demonstrating a lack of an elevated risk of thromboembolic complications (12, 27). The perioperative use of TXA aligns with the recommendations outlined in the American Society of Anesthesiologists (ASA) Practice Guidelines for Perioperative Blood Management, particularly in reducing transfusion requirements for patients at an increased risk of bleeding (28).
Despite the increasing body of literature on TXA, scrutiny of its safety is becoming more stringent, especially in patients at higher risk for thrombotic complications, including VTE (15) and coronary artery disease (18). While existing meta-analyses have not conclusively identified a higher risk of thromboembolic complications in total TJA patients treated with TXA (29), it is crucial to note that these analyses have not specifically targeted high-risk patients. Many studies deliberately excluded individuals at high risk, with a history of DVT, PE, MI, or stroke (30). This systematic review and meta-analysis, however, specifically address the safety of TXA in TJA patients with preexisting thromboembolic risk, filling a notable gap in the literature.
This meta-analysis sheds light on the potential of TXA to reduce total blood loss and transfusion rates in patients with preexisting thromboembolic risk undergoing TJA. The effectiveness of TXA in minimizing blood loss during TJA has been consistently demonstrated in both clinical trials and a previous meta-analysis (18). In a randomized trial involving individuals with preexisting thromboembolic risk, Yen et al. reported a significantly lower total blood loss in the TXA group (645 mL) compared to the placebo group (1103 mL) (P < 0.001) (20). Similarly, Tang et al. observed that high-risk patients in the TXA group experienced less total blood loss and lower transfusion rates than those in the non-TXA group (P < 0.001) (19). These findings provide additional confirmation of the efficacy of TXA in patients with preexisting thromboembolic risk undergoing TJA.
Sensitivity analysis overall reveals relatively stable results. It is noteworthy that the study by Poeran et al. (13), being a pivotal large-sample cohort study, significantly influences the overall direction of the combined results. Exclusion of this study would lead to inconclusive combined outcomes in four key outcome measures: 90-day unplanned readmission, deep vein thrombosis, pulmonary embolism, and MI. We have emphasized the impact of this study on the overall direction of outcomes. In the future, we hope for more large-sample cohort studies to substantiate the limited conclusions of our research further.
In our subgroup analyses, we observed a dose-dependent effect of TXA on transfusion rates. Higher TXA doses correlated with a reduction in transfusion needs, consistent with the pharmacological effects of TXA. However, escalating the dosage did not significantly impact outcomes such as unplanned readmission, deep vein thrombosis, MI, and stroke. These findings align with those of Mary K. Richardson et al. (31), supporting the notion that TXA administration does not correlate with an increased risk of pulmonary embolism, stroke, or MI following arthroplasty. Our meta-analysis further extends its scope to clinical practice guidelines endorsed by the 2018 AAHKS/AAOS/ASRA/AKS/AHS, emphasizing that administering TXA to patients with a high prevalence of comorbidities, including venous thrombosis, MI, cerebrovascular accident, transient ischemic attack, or a history of vascular stenting, does not elevate the risk of primary total joint replacement surgery. The incidence of unfavorable perioperative thromboembolic events adds robust support to the accuracy of our findings (32).
Moreover, our investigation delves into the influence of ethnic background on the clinical effectiveness of TXA. Contrary to expectations, we discovered no significant association between ethnic backgrounds and the efficacy of TXA, reinforcing the generalizability of our findings across diverse patient populations. Additionally, the observed similarity in outcomes between patients undergoing TKA and THA highlights the comparable nature of these procedures (33). Both involve replacing diseased joints with prostheses, and our analysis found no notable distinctions in outcome variables between Cohort and RCT studies pertaining to TXA. This suggests that to some extent, Cohort studies hold the same evidential validity as RCT studies.
Crucially, our meta-analysis dispels concerns about the safety of TXA in patients with preexisting thromboembolic risk undergoing TJA. No significant increase in mortality risk or thromboembolic complications was associated with TXA administration. Fillingham et al.'s retrospective study, focusing on high-risk TJA patients based on the American Society of Anesthesiologists (ASA) physical status score, found that for patients with a higher ASA score, TXA administration did not escalate the risk of thromboembolic complications (34). Similarly, Porter et al., who included 38 220 TJA patients based on prothrombotic conditions, reported no significantly higher risk of thromboembolic complications among patients with and without high-risk diseases (19). Duncan et al. (35) found no association between TXA use and mortality risk, and Shakur et al. (36) reported a reduction in mortality risk in bleeding trauma patients with TXA administration. This comprehensive systematic review and meta-analysis reinforce the safety profile of TXA, enlightening its efficacy without an increased risk of adverse events.
The use of TXA in hip and knee arthroplasty patients at high risk of thrombosis has been widely reported in the literature, However, a comprehensive retrospective analysis of the available evidence to provide clinical guidance still needs to be provided. This meta-analysis suggests that the risk of bleeding is reduced with low (1–2 g) and high (≥3 g) doses of TXA, and that the risk of transfusion is higher with low doses of TXA compared with high doses, which may be due to a dose-response or limited number of studies. Therefore, TXA can be used in patients with TJA at high clinical risk of thrombotic events to reduce bleeding and transfusion rates, providing evidence supporting the safe use of TXA in patients at high risk of thrombotic events.
Clinical implication
Although the efficacy and safety of TXA in reducing blood loss and the need for blood transfusion have been well established, preexisting thromboembolic risk patients with significant comorbidities associated with hypercoagulable or prothrombotic states are still challenging for clinicians regarding medical decision-making. Previously, contraindications to TXA use include thromboembolic events, coagulopathies or ischemic events, renal failure, and known allergy to the medication. The inclusion of preexisting thromboembolic risk patients in our study comprised VTE (including DVT, PE), MI, atrial fibrillation, stroke, coronary artery bypass graft, and renal diseases. This meta-analysis adds evidence to the recent clinical guide on TXA administration in TJA patients, precisely these preexisting thromboembolic risk patients (32).
Strengths and limitations
To our knowledge, this is the first systematic review evaluating the use of TXA in patients undergoing TJA surgery with pre-existing thromboembolic risk. An important strength of this systematic review is its strict adherence to the PRISMA guidelines in the implementation and reporting of the review. Additionally, it has compiled a large sample dataset through aggregation to validate the impact of TXA on thromboembolic risk and associated complications.
This study has some limitations. First, only 2 out of the 11 included studies were randomized trials. Evidence from retrospective or prospective studies carries a lower level of certainty, which may partially constrain the quality of this study. Future research with high-quality RCTs is still needed to confirm the results of this systematic review and meta-analysis. Secondly, heterogeneity in the route and dosage of TXA administration persists, and due to limited data, it is challenging to address this issue through heterogeneity tests and sensitivity analysis. Finally, we included various study designs in our forest plot, which may be a better way to integrate data. However, given the limited evidence, we can only provide relatively prioritized evidence and suggest appropriate recommendations for future research.
Conclusion
In conclusion, this systematic review and meta-analysis provides an up-to-date overview of the impact of TXA on postoperative benefits and complications in patients with TKA or THA at high risk of thrombotic events. We found that the administration of TXA did not increase the incidence of complications, including reduced mortality, deep vein thrombosis and pulmonary embolism. The use of TXA in TJA patients at risk of thromboembolism is also beneficial, including reduced blood loss, transfusion rates and readmission rates.
Supplementary materials
This is linked to the online version of the paper at https://doi.org/10.1530/EOR-23-0140.
ICMJE Conflict of Interest Statement
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the study reported.
Funding Statement
This work was supported by the Natural Science Foundation of Gansu Province, China (22JR5RA982), and the CuiYing Scientific and Technological Innovation Program of Lanzhou University Second Hospital, Gansu province, China (20CX9ZA113).
References
- 1↑
Schwartz AM, Farley KX, Guild GN, & Bradbury TL Jr. Projections and epidemiology of revision hip and knee arthroplasty in the United States to 2030. Journal of Arthroplasty 2020 35 S79–S85. (https://doi.org/10.1016/j.arth.2020.02.030)
- 2↑
Johnson CA, White CC, Kunkle BF, Eichinger JK, & Friedman RJ. Effects of the obesity epidemic on total hip and knee arthroplasty demographics. Journal of Arthroplasty 2021 36 3097–3100. (https://doi.org/10.1016/j.arth.2021.04.017)
- 3↑
Sizer SC, Cherian JJ, Elmallah RD, Pierce TP, Beaver WB, & Mont MA. Predicting blood loss in total knee and hip arthroplasty. Orthopedic Clinics of North America 2015 46 445–459. (https://doi.org/10.1016/j.ocl.2015.06.002)
- 4↑
Prudovsky I, Kacer D, Zucco VV, Palmeri M, Falank C, Kramer R, Carter D, & Rappold J, Tranexamic acid: beyond antifibrinolysis. Transfusion 2022 62(Suppl 1) S301–S312. (https://doi.org/10.1111/trf.16976)
- 5↑
Colomina MJ, Contreras L, Guilabert P, Koo M, M Ndez E, & Sabate A. Clinical use of tranexamic acid: evidences and controversies. Brazilian Journal of Anesthesiology 2022 72 795–812. (https://doi.org/10.1016/j.bjane.2021.08.022)
- 6↑
Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, Shamseer L, Tetzlaff JM, Akl EA, Brennan SE, et al.The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ 2021 372 n71. (https://doi.org/10.1136/bmj.n71)
- 7↑
Sterne JAC, Savović J, Page MJ, Elbers RG, Blencowe NS, Boutron I, Cates CJ, Cheng HY, Corbett MS, Eldridge SM, et al.RoB 2: a revised tool for assessing risk of bias in randomised trials. BMJ 2019 366 l4898. (https://doi.org/10.1136/bmj.l4898)
- 8↑
Slim K, Nini E, Forestier D, Kwiatkowski F, Panis Y, & Chipponi J. Methodological index for non-randomized studies (minors): development and validation of a new instrument. ANZ Journal of Surgery 2003 73 712–716. (https://doi.org/10.1046/j.1445-2197.2003.02748.x)
- 9↑
Guyatt G, Oxman AD, Akl EA, Kunz R, Vist G, Brozek J, Norris S, Falck-Ytter Y, Glasziou P, DeBeer H, et al.GRADE guidelines: 1. Introduction-GRADE evidence profiles and summary of findings tables. Journal of Clinical Epidemiology 2011 64 383–394. (https://doi.org/10.1016/j.jclinepi.2010.04.026)
- 10↑
Qahri-Saremi H, & Sharma S. Introduction to Meta-analysis 2017. (https://doi.org/10.13140/RG.2.2.23047.91046)
- 11↑
Goh GS, D'Amore T, Lonner JH, & Fillingham YA. Tranexamic acid is associated with decreasing the risk of complications in patients undergoing arthroplasty with preoperative coagulopathy. Journal of Arthroplasty 2021 36 3864–3869.e1. (https://doi.org/10.1016/j.arth.2021.08.009)
- 12↑
Heller S, Secrist E, Shahi A, Chen AF, & Parvizi J. Tranexamic acid can be administered to arthroplasty patients who receive aspirin for venous thromboembolic prophylaxis. Journal of Arthroplasty 2016 31 1437–1441. (https://doi.org/10.1016/j.arth.2015.12.042)
- 13↑
Poeran J, Chan JJ, Zubizarreta N, Mazumdar M, Galatz LM, & Moucha CS. Safety of tranexamic acid in hip and knee arthroplasty in high-risk patients. Anesthesiology 2021 135 57–68. (https://doi.org/10.1097/ALN.0000000000003772)
- 14↑
Porter SB, Spaulding AC, Duncan CM, Wilke BK, Pagnano MW, & Abdel MP. Tranexamic acid was not associated with increased complications in high-risk patients with hip fracture undergoing arthroplasty. Journal of Bone and Joint Surgery 2021 103 1880–1889. (https://doi.org/10.2106/JBJS.21.00172)
- 15↑
Porter SB, White LJ, Osagiede O, Robards CB, & Spaulding AC. Tranexamic acid administration is not associated with an increase in complications in high-risk patients undergoing primary total knee or total hip arthroplasty: a retrospective case-control study of 38,220 patients. Journal of Arthroplasty 2020 35 45–51.e31. (https://doi.org/10.1016/j.arth.2019.08.015)
- 16↑
Qiu J, Sun X, Zhang W, Ke X, Yang G, & Zhang L. Effect of topical tranexamic acid in total hip arthroplasty patients who receive continuous aspirin for prevention of cardiovascular or cerebrovascular events: a prospective randomized study. Orthopaedics and Traumatology, Surgery and Research 2019 105 1327–1332. (https://doi.org/10.1016/j.otsr.2019.06.018)
- 17↑
Sabbag OD, Abdel MP, Amundson AW, Larson DR, & Pagnano MW. Tranexamic acid was safe in arthroplasty patients with a history of venous thromboembolism: a matched outcome study. Journal of Arthroplasty 2017 32 S246–S250. (https://doi.org/10.1016/j.arth.2017.02.008)
- 18↑
Tang S, Zhou Z, Yang J, Kang P, Shen B, Pei F, & Shi X. Effects of tranexamic acid on vascular occlusive events and perioperative resuscitation in patients with atrial fibrillation undergoing total joint arthroplasty. Chinese Medical Journal 2022 135 2354–2356. (https://doi.org/10.1097/CM9.0000000000002114)
- 19↑
Whiting DR, Gillette BP, Duncan C, Smith H, Pagnano MW, & Sierra RJ. Preliminary results suggest tranexamic acid is safe and effective in arthroplasty patients with severe comorbidities. Clinical Orthopaedics and Related Research 2014 472 66–72. (https://doi.org/10.1007/s11999-013-3134-0)
- 20↑
Yen SH, Lin PC, Wu CT, & Wang JW. Comparison of effects of a thrombin-based hemostatic agent and topical tranexamic acid on blood loss in patients with preexisting thromboembolic risk undergoing a minimally invasive total knee arthroplasty. A prospective randomized controlled trial. BioMed Research International 2021 2021 2549521. (https://doi.org/10.1155/2021/2549521)
- 21↑
Joo YB, Kim YM, An BK, Lee CW, Kwon ST, & Song JH. Topical tranexamic acid can be used safely even in high risk patients: deep vein thrombosis examination using routine ultrasonography of 510 patients. Medicina 2022 58. (https://doi.org/10.3390/medicina58121750)
- 22↑
Banach L, Janczewski L, Kajdanek J, Milowska K, Kolodziejczyk-Czepas J, Galita G, Rozpedek-Kaminska W, Kucharska E, Majsterek I, & Kolesinska B. Synthesis and hemostatic activity of new amide derivatives. Molecules 2022 27. (https://doi.org/10.3390/molecules27072271)
- 23↑
Jiang J, Xing F, Zhe M, Luo R, Xu J, Duan X, & Xiang Z. Efficacy and safety of tranexamic acid for patients with intertrochanteric fractures treated with intramedullary fixation: a systematic review and meta-analysis of current evidence in randomized controlled trials. Frontiers in Pharmacology 2022 13 945971. (https://doi.org/10.3389/fphar.2022.945971)
- 24↑
Lecker I, Wang DS, Whissell PD, Avramescu S, Mazer CD, & Orser BA. Tranexamic acid-associated seizures: causes and treatment. Annals of Neurology 2016 79 18–26. (https://doi.org/10.1002/ana.24558)
- 25↑
Li GL, & Li YM. Oral tranexamic acid can reduce blood loss after total knee and hip arthroplasty: a meta-analysis. International Journal of Surgery 2017 46 27–36. (https://doi.org/10.1016/j.ijsu.2017.08.009)
- 26↑
Poeran J, Rasul R, Suzuki S, Danninger T, Mazumdar M, Opperer M, Boettner F, & Memtsoudis SG. Tranexamic acid use and postoperative outcomes in patients undergoing total hip or knee arthroplasty in the United States: retrospective analysis of effectiveness and safety. BMJ 2014 349 g4829. (https://doi.org/10.1136/bmj.g4829)
- 27↑
Gillette BP, DeSimone LJ, Trousdale RT, Pagnano MW, & Sierra RJ. Low risk of thromboembolic complications with tranexamic acid after primary total hip and knee arthroplasty. Clinical Orthopaedics and Related Research 2013 471 150–154. (https://doi.org/10.1007/s11999-012-2488-z)
- 28↑
American Society of Anesthesiologists Task Force on Perioperative Blood Management. Practice guidelines for perioperative blood management: an updated report by the American Society of Anesthesiologists Task Force on Perioperative Blood Management. Anesthesiology 2015 122 241–275. (https://doi.org/10.1097/ALN.0000000000000463)
- 29↑
Wang F, Zhao KC, Zhao MM, & Zhao DX. The efficacy of oral versus intravenous tranexamic acid in reducing blood loss after primary total knee and hip arthroplasty: a meta-analysis. Medicine (Baltimore) 2018 97 e12270. (https://doi.org/10.1097/MD.0000000000012270)
- 30↑
Reale D, Andriolo L, Gursoy S, Bozkurt M, Filardo G, & Zaffagnini S. Complications of tranexamic acid in orthopedic lower limb surgery: a meta-analysis of randomized controlled trials. BioMed Research International 2021 2021 6961540. (https://doi.org/10.1155/2021/6961540)
- 31↑
Richardson MK, Liu KC, Mayfield CK, Kistler NM, Lieberman JR, & Heckmann ND. Tranexamic acid is safe in patients with a history of venous thromboembolism undergoing total joint arthroplasty. Journal of Bone and Joint Surgery 2024 106 30–38. (https://doi.org/10.2106/JBJS.23.00254)
- 32↑
Fillingham YA, Ramkumar DB, Jevsevar DS, Yates AJ, Bini SA, Clarke HD, Schemitsch E, Johnson RL, Memtsoudis SG, Sayeed SA, et al.Tranexamic acid in total joint arthroplasty: the endorsed clinical practice guides of the American Association of Hip and Knee Surgeons, American Society of Regional Anesthesia and Pain Medicine, American Academy of Orthopaedic Surgeons, Hip Society, and Knee Society. Regional Anesthesia and Pain Medicine 2019 44 7–11. (https://doi.org/10.1136/rapm-2018-000024)
- 33↑
Bourne RB, Chesworth B, Davis A, Mahomed N, & Charron K. Comparing patient outcomes after THA and TKA: is there a difference? Clinical Orthopaedics and Related Research 2010 468 542–546. (https://doi.org/10.1007/s11999-009-1046-9)
- 34↑
Fillingham YA, Ramkumar DB, Jevsevar DS, Yates AJ, Shores P, Mullen K, Bini SA, Clarke HD, Schemitsch E, Johnson RL, et al.The safety of tranexamic acid in total joint arthroplasty: a direct meta-analysis. Journal of Arthroplasty 2018 33 3070–3082.e1. (https://doi.org/10.1016/j.arth.2018.03.031)
- 35↑
Duncan CM, Gillette BP, Jacob AK, Sierra RJ, Sanchez-Sotelo J, & Smith HM. Venous thromboembolism and mortality associated with tranexamic acid use during total hip and knee arthroplasty. Journal of Arthroplasty 2015 30 272–276. (https://doi.org/10.1016/j.arth.2014.08.022)
- 36↑
Shakur H, Roberts I, Bautista R, Caballero J, Coats T, Dewan Y, El-Sayed H, Gogichaishvili T, Gupta S, Herrera J, et al.Effects of tranexamic acid on death, vascular occlusive events, and blood transfusion in trauma patients with significant haemorrhage (CRASH-2): a randomised, placebo-controlled trial. Lancet 2010 376 23–32. (https://doi.org/10.1016/S0140-6736(1060835-5)
