Tranexamic Acid Dosing for Cardiac Surgical Patients With Chronic Renal Dysfunction: A New Dosing Regimen

Angela Jerath, FRCPC, FANZCA, MBBS, BSc; Qi Joy Yang, MSc; K. Sandy Pang, PhD; Nikita Looby, MSc; Nathaly Reyes-Garces, MSc; Tijana Vasiljevic, BSc; Barbara Bojko, PhD; Janusz Pawliszyn, PhD; Duminda Wijeysundera, PhD, FRCPC; W. Scott Beattie, PhD, FRCPC; Terrence M. Yau, PhD, FRCSC, MD, MSc, BA; Marcin Wąsowicz, FRCPC, PhD, MD


Anesth Analg. 2018;127(6):1323-1332. 

In This Article


We recruited 26 low-risk and 22 high-risk patients. Stage 4 and low-risk stage 2 CRD patients were more difficult to recruit; thus, patient numbers were lower in these categories. Perioperative patient characteristics are reported in Table 3 and Supplemental Digital Content, Table 1,

Clinical Outcomes

Four patients (8%) suffered postoperative seizures in CRD stages 5 (3 patients) and 3 (1 patient) in the high-risk group (Supplemental Digital Content, Table 2, Brain computed tomography scans revealed no new focal mass lesion in any of these patients. There were 5 (10%) postoperative deaths in low-risk stage 3 (1 patient), high-risk stage 3 (1 patient), and stage 5 (3 patients) CRD, which included 2 patients who suffered a seizure. There was no significant difference in postoperative bleeding, chest reexploration, transfusion, ischemic-thrombotic complications, or length of stay.

Pharmacokinetic Modeling and Simulations

Pharmacokinetic parameters obtained from fitting are summarized in Table 2 and include data previously obtained on 15 high-risk stage 1 CRD patients.[10] The loading or bolus dose was effective at establishing the initial plasma TXA concentration to attain a therapeutic threshold above 100 mg/L. TXA concentration–time profiles demonstrated that plasma levels were elevated with increasing severity of CRD in both study groups (Figure 1). The total body CL (estimated as dose/AUC) values for TXA remain similar for stages 1 and 2 in the low- (0.094 stage 1 vs 0.073 stage 2, L/h/kg) and high-risk (0.110 stage 1 vs 0.092 stage 2, L/h/kg) groups. However, the total body CL reduces markedly with declining renal function between stages 3 and 5 in both the low- and high-risk treatment groups (Table 1). Collectively speaking, the CL progressively decreased among patients who were more renally compromised. A similar trend was observed for V1, volume of the central compartment. While there was no change in low-risk stages 1–5, V1 progressively reduced for high-risk stages 4 and 5 groups, likely due to accompanying dialysis and other procedures that reduced the plasma volume. The intercompartmental CL, CL12 or CL21, was >2-fold higher for the high-risk group versus the low-risk group for each stage. Due to the combined changes in CL, CL12, and V1, plasma levels fell below the therapeutic threshold in 2–4 hours in low-risk group stages 1 and 2 CRD patients, retaining a short t ½ of 2.6–3.4 hours. However, the t ½ values progressively increased to 6.3, 12, and 38 hours (Table 1) in low-risk CRD stages 3–5, with TXA displaying a slower decay and sustained plasma levels near 100 mg/L for approximately 6 hours. By contrast, the high-risk group showed consistently elevated plasma TXA concentrations above 100 mg/L that were sustained for up to 12 hours in stages 3–5. This reduction in CL was consistent with CRD stages (represented by baseline eGFR), with a significant correlation of r 2 of 0.67 (Figure 2).

Figure 1.

Fitted (lines) and observed (symbols) plasma TXA concentration–time profiles in low- and high-risk study groups for each stage of CRD. The different colors represent different stages of CRD (1–5). Red dashed lines indicate the TXA threshold level of 100 mg/L that allows 100% antifibrinolysis.9,10 The yellow circles (high-risk group, stages 3 and 5) represent TXA concentration data in the 4 seizure patients. Based on the newly recommended reduction in loading dose and maintenance infusion (Table 2) for the high-risk stages 2–5 CRD groups, simulation of TXA profiles was performed: the dashed and solid turquoise lines represent the pharmacokinetic profile when loading dose is 25 or 30 mg/kg, respectively. For the CRD 3–5 groups, the turquoise solid and the dashed lines represent the loading dose of 30 and 25 mg/L, respectively, with the same upper boundary maintenance infusion rate (shown in Table 2); the brown solid and dashed lines represent the loading dose of 30 and 25 mg/kg, respectively, with the same lower boundary maintenance infusion rate recommended (Table 2). CRD indicates chronic renal dysfunction; TXA, tranexamic acid.

Figure 2.

Baseline eGFR versus TXA plasma clearance. Different colors represent different stages of CRD. The yellow closed circles represent the data obtained from seizure patients. The measured TXA concentrations of 1 seizure patient were consistently rising; therefore, the AUC of that patient could not be calculated. The dashed and solid black lines denote the regression line and line of identity (y = x), respectively. AUC indicates area under the curve to time infinity; eGFR, estimated glomerular filtration rate; TXA, tranexamic acid.

Population Pharmacokinetics

With population pharmacokinetic modeling, we investigated which covariates influence CL, V1, and CL12. We identified that eGFR and patient risk group (low or high) were important covariates influencing the CL, V1, V2, CL12 and the unexplained intersubject variability (represented by ETAs) associated with the pharmacokinetic parameters were also reduced (Figure 3; Supplemental Digital Content, Table 3, The final model is robust with reasonably good predictions for characterizing TXA pharmacokinetics among CRD patients.

Figure 3.

Plots of contribution of covariates (including eGFR, CRD stages, and risk groups [low versus high]) to the random effects (ETAs) for CL1, V1, CL12 or CL21, and V2 of tranexamic acid for (A) base and (B) final models. These plots were generated from output of NONMEM using R (R Development Core Team, Vienna, Austria). Risk groups are assigned as 0 (low risk) or 1 (high risk). CL indicates clearance; CRD, chronic renal dysfunction; eGFR, estimated glomerular filtration rate; V1, volume of the central compartment; V2, volume of peripheral compartment.

Seizure Versus Nonseizure Patients

We examined the pharmacokinetic profiles of the 4 seizure patients, which are highlighted in yellow symbols for high-risk group stages 3 and 5 (Figure 1, right panel), reflecting the higher area under the curve (AUC) and lower CL. The seizure group showed higher TXA concentrations and prolonged t ½ (29.6 vs 3.4 hours; P = .014) and the area above the 100 mg/L mark, or AUC>100 mg/L (AUC − AUC≤100 mg/L) compared to the nonseizure group (Supplemental Digital Content, Table 2, This suggests that higher TXA systemic accumulation and reduced CL may have contributed to seizure development.

Dosing Adjustment for CRD

Given observations of reduced V1 and CL but higher CL12 among high-risk patients, we developed a new proposal for the TXA infusion regimen, aiming to maintain a plasma concentration around 100 mg/L (Table 2). We lowered the loading dose in stages 3–5 CRD (due to reduced V1 but higher CL12) slightly to 25 mg/L and the maintenance infusion rate (corresponding to eGFR) for all CRD stages to bring TXA plasma levels into a safe range. Simulations based on the newly recommended regimen demonstrate this expected pattern for high-risk stages 2–5 CRD patients (Figure 1). Given that TXA exhibits minimal protein binding, any change in protein levels is unlikely to influence TXA plasma levels.