What is the safety and efficacy data for tranexamic acid injection via the arterial line or intraosseous route?

Is there any safety or efficacy data supporting administration of tranexamic acid injection via the arterial line or intraosseous route?

Comment by InpharmD Researcher

There is limited data on the use of tranexamic acid via intra-arterial or intraosseous routes. Only one small case-control study (Table 1) appears to discuss intra-arterial TXA administration in patients with vessel rupture; TXA was associated with lower mortality and no adverse events. Intraosseous regional TXA has been evaluated in two RCTs (Tables 2 and 3) that found comparable results to IV and topical administration; no obvious safety effects were noted. An expanded literature search did not find any additional data on intra-arterial or intraosseous TXA administration.

Literature Review

A search of the published medical literature revealed 5 studies investigating the researchable question:

Is there any safety or efficacy data supporting administration of tranexamic acid injection via the arterial line or intraosseous route?

Level of evidence

C - Multiple studies with limitations or conflicting results Read more→

Please see Tables 1-5 for your response.

Tranexamic acid as a novel adjunct in the management of vessel perforation complicating Endovascular Clot Retrieval
Design	Retrospective, case-control study N= 1378
Objective	To assess the safety and utility of tranexamic acid (TXA) as an adjunct salvage therapy in iatrogenic vessel perforation complicating endovascular clot retrieval (ECR)
Study Groups	TXA administered (n= 11) Standard care (n= 25)
Inclusion Criteria	Cases where vessel perforation occurred during endovascular clot retrieval between July 2016 and March 2022
Exclusion Criteria	Not specified
Methods	This was a retrospective review of endovascular clot retrieval cases with vessel perforation at a single center in Australia. TXA was administered intra-arterially in some cases. Baseline patient details, imaging, procedural reports, and outcomes at 3 months were recorded. Functional outcomes were assessed using the modified Rankin Scale (mRS).
Duration	July 2016 to March 2022
Outcome Measures	Primary: Good functional outcome at 3 months (mRS 0-2) Secondary: Mortality at 3 months
Baseline Characteristics		TXA administered (n= 11)
	Age, years	72 ± 13
	Female	4 (33%)
	NIHSS, Median (IQR)	18 (7-22)
	IV tPA	3 (27%)
	General anaesthesia	5 (45%)
	TXA was the only hemostasis method in 3 (27%) of the cases.
Results		TXA administered (n= 11)	Standard care (n= 25)	p-value
	Good functional outcome (mRS 0-2)	4 (36%)	3 (12%)	0.09
	Mortality at 3 months	4 (41.7%)	16 (64%)	0.13
Case Series (only three select cases reported)	A patient presenting with a left M1-M2 occlusion underwent ECR after receiving intravenous thrombolysis with tenecteplase. Angiographic evaluation confirmed a distal M1-M2 occlusion, where a single pass using a stent-retriever successfully removed a moderate thrombus burden, resulting in immediate symptom improvement. Despite this, a distal clot located in the angular branch was identified. Aspiration was attempted in the distal cortical middle cerebral artery (MCA) branch, extracting a small thrombus volume. However, subsequent angiography revealed slow contrast extravasation indicative of intracranial vessel perforation in the distal cortical MCA branch. An on-table CT scan corroborated the contrast leak with evidence of a small subarachnoid hemorrhage (SAH). To manage the hemorrhage, the microcatheter and a 5Fr intermediate catheter were retained in situ to exert tamponade on the bleeding site. Hemostasis was achieved following intra-arterial administration of 150 mg of TXA directly through the microcatheter. The patient maintained functional independence (mRS 2) and showed progressive improvement in dysphasia at the 3-month follow-up.
	A patient underwent ECR for a left MCA stroke characterized by a large penumbra without an established infarct core on CT. The procedure was performed under general anesthesia due to patient agitation. Angiographic evaluation revealed an occluded distal MCA accompanied by a large, partially occlusive thrombus involving the anterior cerebral artery (ACA), pericallosal, and calloso-marginal branches. Initial attempts at MCA recanalization utilized suction thrombectomy combined with a 4 mm stent-retriever, successfully extracting a moderate thrombus burden. However, this was complicated by distal branch extravasation. Administration of 1000 mg TXA was attempted with minimal hemostatic effect. Subsequent recanalization of the ACA was achieved using the stent-retriever, yielding substantial thrombus extraction. Despite these efforts, final angiographic runs detected ongoing slow extravasation from the left M3/4 branch. On-table CT confirmed extensive SAH and accumulation of contrast material within the Sylvian fissure. Post-procedure, the patient remained intubated and was transferred to the intensive care unit (ICU). Unfortunately, the clinical course deteriorated with ongoing, uncontrollable intracranial hemorrhage leading to death shortly after extubation the following day.
	A 66-year-old patient underwent ECR for an acute proximal left superior M2 division occlusion, associated with a right frontal penumbra and no established infarct core. Angiographic evaluation revealed a focal occlusion in the proximal superior left M2 branch. A single pass with a 4×20 mm stent-retriever successfully removed a small-volume clot. During this procedure, a minor arterial rupture occurred at the stent-retriever instrumentation site, identified as a small volume arterial perforation. To manage this complication, intra-arterial TXA was administered in two 300 mg doses, which led to cessation of bleeding each time without the need for balloon tamponade or surgical intervention. Further endovascular clot retrieval attempts beyond the initial pass were halted due to concerns over exacerbating hemorrhage risk. Imaging confirmed contrast extravasation post-perforation with bleeding into the subarachnoid space, but following TXA administration, bleeding resolved effectively. The patient experienced no neurological deficits post-procedure, was discharged to a rehabilitation center, and had a favorable clinical outcome with a mRS score of 0 at 3 months.
Adverse Events	Seizures were reported in 3 cases (27%) in those who received TXA and in 3 cases (12%) in those who did not receive TXA (p=0.27)
Study Author Conclusions	Tranexamic acid administration in iatrogenic vessel rupture was associated with a lower mortality rate and a larger proportion of patients achieving a good functional outcome at 3 months. This effect trended towards but was not statistically significant. TXA administration was not associated with adverse effects.
Critique	The study's retrospective design and small sample size limit the ability to draw definitive conclusions. The lack of statistical significance in the results suggests that larger, prospective studies are needed to confirm the efficacy of TXA in this context. Additionally, the study did not specify exclusion criteria, which may affect the generalizability of the findings.

Table 1 References:
[1] Morkos K, Maingard J, Barras C, et al. Tranexamic acid as a novel adjunct in the management of vessel perforation complicating Endovascular Clot Retrieval. J Stroke Cerebrovasc Dis. 2023;32(6):107067. doi:10.1016/j.jstrokecerebrovasdis.2023.107067

Effectiveness and safety of intraosseous regional administration versus intravenous tranexamic acid in primary total knee arthroplasty: a randomized, triple-blind, placebo-controlled trial
Design	Prospective, randomized, triple-blind controlled trial N= 97
Objective	To evaluate whether intraosseous regional administration of tranexamic acid (IORA-TXA), alone or combined with intravenous TXA (IV-TXA), reduces perioperative total blood loss (TBL) without increasing complications compared to IV-TXA alone
Study Groups	IORA-TXA (n= 32) IV-TXA (n= 33) IORA+IV-TXA (n= 32)
Inclusion Criteria	Patients aged 20–80 years undergoing primary unilateral conventional TKA with cruciate retaining implants for osteoarthritis
Exclusion Criteria	Allergy to TXA; anemia (preoperative hemoglobin < 11 g/L); high-risk factors for thrombosis; use of anticoagulants or antiplatelet agents within 1 week before procedure; severe organ diseases
Methods	Patients were randomized into three groups: Group A received 1.0 g IORA-TXA, Group B received 15 mg/kg IV-TXA, and Group C received both IORA-TXA and IV-TXA. All patients received an additional 1.0 g IV-TXA at 3 h post-surgery.
Duration	1 December 2022 to 31 December 2024
Outcome Measures	Primary: Total blood loss (TBL), transfusion rate Secondary: Coagulation indicators, inflammatory markers, knee function, patient-reported outcome measures, complications
Baseline Characteristics		IORA-TXA (n= 32)	IV-TXA (n= 33)	IORA+IV-TXA (n= 32)	p-value
	Age, years	67.6 ± 5.0	66.8 ± 7.2	68.1 ± 6.4	0.720
	Female	30	27	28	0.649
	Height, cm	159.3 ± 7.3	161.0 ± 6.7	160.5 ± 7.9	0.685
	Weight, kg	69.5 ± 10.3	69.4 ± 10.2	69.8 ± 14.2	0.992
	BMI, kg/m²	27.3 ± 3.4	26.7 ± 2.7	27.0 ± 4.5	0.809
	Preoperative values Hemoglobin, g/L Platelet count, ×1000/mm³ INR Fibrinogen, g/L	123.8 ± 24.7 230.1 ± 44.4 0.98 ± 0.06 2.8 ± 0.6	134.1 ± 10.1 215.8 ± 57.3 0.97 ± 0.06 2.6 ± 0.5	130.1 ± 10.4 221.6 ± 50.9 0.97 ± 0.07 2.7 ± 0.6	0.071 0.578 0.949 0.718
	Operative time, minutes	88.0 ± 18.2	88.5 ± 16.0	86.6 ± 20.9	0.924
Results		IORA-TXA (n= 32)	IV-TXA (n= 33)	IORA+IV-TXA (n= 32)	p-value
	Total blood loss, mL	430.7 ± 243.4	503.6 ± 277.5	314.4 ± 222.6	0.024
	C-reactive protein at 72h, mg/L	62.1 ± 70.0	72.9 ± 66.0	34.1 ± 26.7	0.047
	Visual analogue scale at POD 14	Lower than IV-TXA	Higher than Groups A and C	Lower than IV-TXA	-
	Flexion contracture at POD 14	Higher than IORA+IV-TXA	Higher than IORA+IV-TXA	Lower than IORA-TXA and IV-TXA	-
Adverse Events	No significant differences in complication rates were observed among the groups. No cases of transfusion were found. No clinically significant DVT or pulmonary embolism was observed within 90 days postoperatively.
Study Author Conclusions	IORA-TXA showed hemostatic efficacy and safety comparable to IV-TXA and reduced pain at POD14. The combination of IORA- and IV-TXA further reduced TBL, and showed a lower CRP levels at POD 3 as well as reductions in pain scores and flexion contracture at POD14, without increasing complications. These findings suggest IORA-TXA as an effective and safe strategy in primary manual TKA.
Critique	The study is well-designed with a triple-blind, randomized controlled trial format, providing robust evidence for the efficacy and safety of IORA-TXA. However, the single-center design and modest sample size may limit the generalizability of the findings. The study did not include a control group without TXA, which could have provided additional insights into the efficacy of TXA administration methods. Further research with larger, multicenter trials is needed to confirm these findings and explore the pharmacokinetics of IORA-TXA.

Table 2 References:
[2] Yu M, Hu M, Xu Y, et al. Effectiveness and safety of intraosseous regional administration versus intravenous tranexamic acid in primary total knee arthroplasty: a randomized, triple-blind, placebo-controlled trial. Int J Surg. Published online December 10, 2025. doi:10.1097/JS9.0000000000004489

Efficacy of Intraosseous Tranexamic Acid in Primary Total Hip Arthroplasty: A Prospective Randomized Controlled Trial
Design	Prospective randomized controlled trial N= 126
Objective	To evaluate whether intraosseous (IO) tranexamic acid (TXA) administration is noninferior to intravenous (IV) or topical administration in reducing blood loss in primary total hip arthroplasty (THA)
Study Groups	IO (n= 42) IV (n= 42) Topical (n= 42)
Inclusion Criteria	Patients diagnosed with osteoarthritis or osteonecrosis of the femoral head scheduled for cementless primary unilateral THA
Exclusion Criteria	Bilateral arthroplasty, femoral neck fracture, developmental dysplasia of hip, history of DVT or PE, clotting disorders, ongoing anticoagulant treatment, preoperative hepatic or renal dysfunction, serious cardiac or cerebrovascular comorbidities, allergy to TXA, refusal to participate
Methods	Patients were randomized to receive tranexamic acid (TXA) 20 mg/kg via intraosseous (IO), intravenous (IV), or topical administration. The IV group received 20 mg/kg TXA diluted in 100 mL of normal saline as an IV bolus 5 minutes before skin incision. The topical group received 20 mg/kg TXA diluted in saline and applied directly to the surgical site before suture placement. In the IO group, the 20 mg/kg TXA dose was divided equally between two sites: the cancellous bone of the femur and the acetabulum. Hemoglobin levels were measured preoperatively and on postoperative days 1 through 3, and blood loss and transfusion rates were recorded.
Duration	October 2024 to April 2025
Outcome Measures	Primary: Hemoglobin reduction on the day of surgery and postoperative days one to three Secondary: Blood loss, transfusion rate, adverse events
Baseline Characteristics		IV (n= 42)	Topical (n= 42)	IO (n= 42)	p-value
	Age, years	54 (38-78)	55 (38-79)	53 (40-76)	0.668
	Male	25 (59.5%)	20 (51.2%)	22 (56.4%)	0.868
	Height, cm	1.67 ± 0.08	1.65 ± 0.06	1.67 ± 0.09	0.448
	Weight, kg	74.7 ± 13.2	73.4 ± 13.0	71.5 ± 10.7	0.394
	BMI, kg/m²	26.6 (19.6-37.1)	26.7 (18.3-40.1)	25.6 (20.0-36.3)	0.272
	Predicted blood volume, L	4.5 ± 0.7	4.4 ± 0.6	4.4 ± 0.7	0.542
	Operated side Right Left	20 (47.6%) 22 (52.4%)	23 (54.7%) 19 (45.3%)	22 (52.4%) 20 (47.6%)	0.621 - -
	Diagnosis ONFH OA RA TA	29 (69%) 11 (26.2%) 0 1 (1.8%)	35 (83.3%) 6 (14.3%) 1 (2.4%) 0	34 (80.1%) 6 (14.3%) 1 (1.8%) 1 (1.8%)	- - - -
	Preoperative laboratory values Hb, g/L Hct, %	132.8 ± 18.3 41.3 ± 4.9	134.9 ± 15.0 42.1 ± 4.2	131.30 ± 15.8 40.7 ± 4.4	0.505 0.286
	Duration of surgery, hours	1.93 ± 0.56	1.91 ± 0.62	1.83 ± 0.53	0.672
	Abbreviations: BMI, body mass index; ONFH, osteonecrosis of the femoral head; OA, osteoarthritis; RA, rheumatoid arthritis; TA, traumatic arthritis; Hb, hemoglobin; Hct, hematocrit; IV, intravenous; IO, intraosseous
Results		IV (n= 42)	Topical (n= 42)	IO (n= 42)	p-value
	Postoperative Hb, g/dL DOS POD 1 POD 2 POD 3	123.1 ± 2.4 120.5 ± 2.4 108.6 ± 2.4 103.5 ± 2.5	128.5 ± 2.3 125.4 ± 2.6 114.8 ± 2.3 109.9 ± 2.2	128.3 ± 2.2 125.0 ± 2.4 114.2 ± 2.2 108.8 ± 2.1	0.162 0.297 0.115 0.108
	Reduction of Hb, g/dL DOS POD 1 POD 2 POD 3	8.8 ± 1.5 11.4 ± 1.4 23.2 ± 1.6 28.3 ± 1.8	8.6 ± 1.3 11.8 ± 1.7 22.3 ± 1.7 27.3 ± 1.6	6.6 ± 1.6 10.6 ± 1.3 20.7 ± 1.6 26.1 ± 1.8	0.427 0.528 0.477 0.443
	Abbreviations: Hb, hemoglobin; IV, intravenous; IO, intraosseous; SD, standard deviation; CI, confidence interval; DOS, day of surgery; POD, postoperative day
Adverse Events	One case of deep vein thrombosis in the topical group; no pulmonary embolism or infections reported
Study Author Conclusions	The blood-sparing efficacy of IO TXA administration is noninferior to that of IV and topical administration, with potential benefit in early postoperative blood loss control. Further high-quality studies are needed to confirm its superiority and establish its clinical value.
Critique	The study's strengths include its randomized controlled design and comprehensive safety monitoring. However, the single-center setting and modest sample size limit generalizability. The absence of pharmacokinetic data and the lack of significant differences in transfusion rates or other patient-centered outcomes suggest that further research is needed to establish the clinical relevance of IO TXA administration.

Table 3 References:
[3] Chen Y, Coa L, Ji B, Li G, Zhang X. Efficacy of Intraosseous Tranexamic Acid in Primary Total Hip Arthroplasty: A Prospective Randomized Controlled Trial. J Arthoplasty. 2026. doi.org/10.1016/j.arth.2026.05.065.

Intravenous Versus Intraosseous Use of Tranexamic Acid in Patients With Traumatic Brain Injury
Design	Retrospective analysis of prospectively collected data from a prehospital TXA for traumatic brain injury trial (NCT01990768) N= 966
Objective	To determine if the total drug exposure of TXA given in the prehospital setting in patients with moderate or severe brain injury differs based on route of administration
Study Groups	2-g TXA bolus IV (n= 233) IO (n= 8) 1-g TXA bolus followed by 1-g TXA infusion IV (n= 152) IO (n= 8)
Inclusion Criteria	Participants recruited prehospital from 20 trauma centers and 39 emergency medical services across the US and Canada between May 2015 and November 2017, with traumatic brain injury, a Glasgow Coma Scale of 12 or less, and a systolic blood pressure of at least 90 mmHg
Exclusion Criteria	Participants with missing plasma TXA concentration data and outlier AUC measurements (defined as the 5% most extreme in each subgroup)
Methods	Participants were enrolled in the Prehospital TXA for a TBI RCT. TXA blood concentrations were drawn at 0, 6, 12, 24, and 48 h with respect to emergency department presentation. Plasma TXA concentrations were measured using liquid chromatography with tandem mass spectrometry. The area under the plasma drug concentration-time curve (AUC) was calculated using the trapezoidal rule.
Duration	May 2015 and November 2017
Outcome Measures	Primary: Total drug exposure measured by area under the plasma drug concentration-time curve (AUC) Secondary: Differences in clot lysis at 30 min on thromboelastography
Baseline Characteristics		1-g TXA bolus 1-g infusion arm	2-g TXA bolus arm
	Median age, years (IQR)	36 (25-55)	39 (26-55)
	Female	39 (24%)	58 (24%)
	Median BMI, kg/m2 (IQR)	26 (23-29)	25 (22-30)
	Serum creatinine, mg/dL (IQR)	1.0 (0.8-1.1)	0.9 (0.8-1.1)
	eGFR, mL/min/1.73 m2 (IQR)	86 (72-103)	84 (71-100)
Results		1-g TXA bolus 1-g infusion arm	2-g TXA bolus arm
	Mean AUC difference (IO vs IV) 95% confidence interval	-13.0 mcg/mL/h -236.2 to 210.3 mcg/mL/h	-2.6 mcg/mL/h -28.4 to 23.3 mcg/mL/h
	No differences in LY30 based on thromboelastography were observed between participants with IO compared to IV access in either the 2-g TXA bolus or the 1-g TXA bolus 1-g TXA infusion treatment groups
Adverse Events	Not reported
Study Author Conclusions	These preliminary data suggest that the administration of TXA via IO and IV routes may result in similar total drug exposure. Further studies incorporating larger numbers with clinical outcomes are needed to confirm this finding.
Critique	The study is limited by the small number of participants receiving TXA via IO administration, which may result in an underpowered analysis to detect differences in TXA exposure. The separation of analysis by TXA dosing further limits the power, with only eight participants with IO access in each subgroup. These findings should be considered exploratory and interpreted with caution.

Table 4 References:
[4] Newman ZC, Ogbeifun VO, Barbosa CE, et al. Intravenous Versus Intraosseous Use of Tranexamic Acid in Patients With Traumatic Brain Injury. J Surg Res. 2024;302:798-804. doi:10.1016/j.jss.2024.08.003

Comparison of Outcomes of Multimodal Intraosseous Femoral Injection and Multimodal Intraosseous Tibial Injection: A Randomized Controlled Trial in Simultaneous Bilateral Total Knee Arthroplasty Patients
Design	Double-blind, randomized controlled trial N= 40 (80 total knees)
Objective	To compare the outcomes of multimodal intraosseous femoral injections with those of tibial injections in patients undergoing bilateral total knee arthoplasty (TKA)
Study Groups	Group A: Femoral injection on right, tibial on left (n= 20) Group B: Tibial injection on right, femoral on left (n= 20)
Inclusion Criteria	Patients who underwent simultaneous bilateral TKA due to osteoarthritis of both knees, aged >55 years, underwent cemented PS design TKA surgery, and were able to give informed consent
Exclusion Criteria	Patients with revision TKA, history of previous knee or hip surgery, allergic reaction to study drugs, chronic kidney or liver disease, pregnancy, coagulopathy, platelet dysfunction, thromboembolism, or use of anticoagulants
Methods	Patients received multimodal intraosseous injections in either the femur or tibia for each TKA, with alternating sites between groups. Injections included 15 mg ketorolac and 500 mg tranexamic acid. Postoperative outcomes assessed included VAS pain score, blood loss, painkiller use, range of motion (ROM), adverse effects, and complications.
Duration	December 2023 to February 2024
Outcome Measures	Primary: Postoperative pain and blood loss Secondary: Amount of painkiller used, ROM, adverse effects, complications
Baseline Characteristics		All	Group A	Group B	p-value
	Female	33 (82.5%)	17 (85%)	16 (60%)	0.687
	Age, years	68.6 ± 66.3	68.4 ± 66.8	68.8 ± 65.1	0.857
	BMI, kg/m²	26.94 ± 4.2	26.07 ± 5.16	27.82 ± 3.95	0.235
	Kellgren-Lawrence grade III/V Femoral side Tibial side	9/31 9/31	3/17 4/16	6/14 5/15	0.267 0.478
	Operative time, min Femoral side Tibial side	57.7 ± 10.0 58.2 ± 9.6	57.0 ± 8.9 58.4 ± 9.1	58.3 ± 11.2 58.0 ± 10.3	0.667 0.911
	Knee Injury and Osteoarthritis Outcome Score Pre-operative Post-operative	41.5 ± 8.6 61.5 ± 4.6	40.5 ± 9.6 60.9 ± 4.4	42.2 ± 7.52 62.1 ± 4.8	0.459 0.438
	Length of stay, days	5.7 ± 1.5	5.8 ± 1.6	5.7 ± 1.5	0.918
Results		Femoral injection Site	Tibial injection Site	p-value	Mean difference	95% CI
	Visual Analog Scale (VAS) Preoperative Postop 12 hr Postop 24 hr Postop 48 hr Postop 2 weeks	7.02 ± 1.18 3.48 ± 1.95 3.75 ± 1.82 3.55 ± 1.61 3.28 ± 1.48	7.08 ± 1.45 4.13 ± 2.04 4.33 ± 2.03 3.83 ± 1.61 3.45 ± 1.55	0.675 0.149 0.186 0.449 0.493	0.06 -0.65 -0.58 -0.28 -0.17	(20.63, 0.51) (21.52, 0.22) (21.42, 0.26) (20.98, 0.42) (20.83, 0.49)
	Extra painkiller requested in first 48 hr, number of times	0.60 ± 0.74	0.75 ± 0.92	0.110	-0.15	(20.51, 0.21)
	Total postoperative blood loss, mL Redivac drain blood loss, mL	303.5 ± 83.5 289.6 ± 76.7	365.4 ± 98.4 350.4 ± 94.6	<0.001 <0.001	-61.9 -60.80	(2101.89, 221.90) (298.54, 223.06)
Adverse Events	No drug adverse effects or complications were reported
Study Author Conclusions	Multimodal intraosseous injection at the femoral site seemed to be as effective for pain management as at the tibial site, but it resulted in markedly reduced postoperative blood loss in patients undergoing TKA. Additional research is required to explore long-term benefits of this technique and confirm its safety profile.
Critique	The study's double-blind, randomized controlled design is a strength, reducing participant heterogeneity. However, the single-surgeon and single-institution design limits generalizability. The study group consisted mostly of Thai women, which may affect the applicability of results to other populations. The small sample size may not fully capture adverse effects of intraosseous TXA/ketorolac, and the influence of periarticular injections on outcomes was not fully explored.

Table 5 References:
[5] Chompoosang T, Wijit T, Ploynumpon P. Comparison of Outcomes of Multimodal Intraosseous Femoral Injection and Multimodal Intraosseous Tibial Injection: A Randomized Controlled Trial in Simultaneous Bilateral Total Knee Arthroplasty Patients. J Am Acad Orthop Surg Glob Res Rev. 2026;10(7):e25.00375. Published 2026 Jul 1. doi:10.5435/JAAOSGlobal-D-25-00375