Half-Life and Clearance of Cardiac Troponin I and Troponin T in Humans

Abstract

BACKGROUND:

Cardiac troponin (cTn) is key in diagnosing myocardial infarction (MI). After MI, the clinically observed half-life of cTn has been reported to be 7 to 20 hours, but this estimate reflects the combined elimination and simultaneous release of cTn from cardiomyocytes. More precise timing of myocardial injuries necessitates separation of these 2 components. We used a novel method for determination of isolated cTn elimination kinetics in humans.

METHODS:

Patients with MI were included within 24 hours after revascularization and underwent plasmapheresis to obtain plasma with a high cTn concentration. After at least 3 weeks, patients returned for an autologous plasma retransfusion followed by blood sampling for 8 hours. cTn was measured with 5 different high-sensitivity cTn assays.

RESULTS:

Of 25 included patients, 20 participants (mean age, 64.5 years; SD, 8.2 years; 4 women [20%]) received a retransfusion after a median of 5.8 weeks (interquartile range, 5.0–6.9 weeks) after MI. After retransfusion of a median of 620 mL (range, 180–679 mL) autologous plasma, the concentration of cTn in participants’ blood increased 4 to 445 times above the upper reference level of the 5 high-sensitivity cTn assays. The median elimination half-life ranged from 134.1 minutes (95% CI, 117.8–168.0) for the Elecsys high-sensitivity cTnT assay to 239.7 minutes (95% CI, 153.7–295.1) for the Vitros high-sensitivity cTnI assay. The median clearance of cTnI ranged from 40.3 mL/min (95% CI, 32.0–44.9) to 52.7 mL/min (95% CI, 42.2–57.8). The clearance of cTnT was 77.0 mL/min (95% CI, 45.2–95.0).

CONCLUSIONS:

This novel method showed that the elimination half-life of cTnI and cTnT was 5 to 16 hours shorter than previously reported. This indicates a considerably longer duration of cardiomyocyte cTn release after MI than previously thought. Improved knowledge of timing of myocardial injury may call for changes in the management of MI and other disorders with myocardial injury.

Clinical Perspective.

What Is New?

• Examination of isolated biomarker elimination kinetics by autologous retransfusion of plasma collected by plasmapheresis several weeks before, here during an ST-elevation myocardial infarction, is feasible.

• The elimination half-lives of cardiac troponin (cTn) I and T were 2 to 10 times shorter than previously observed in patients with an ongoing myocardial infarct (ie, 2 to 4 hours).

• Clearance of cTnI and cTnT was 40 to 77 mL/min.

What Are the Clinical Implications?

• The much shorter half-lives of cTn observed indicate longer release or slower washout of cTn from the myocardium during a myocardial infarction than previously believed, and these findings may change the clinical perception of the dynamics of cTn concentration during myocardial injury.

• The results may also inform us of timing of cTn release during other myocardial injuries than myocardial infarcts, for example, myocarditis, traumas, or episodes of tachycardia.

• Improved knowledge of elimination kinetics may influence the evaluation of timing of myocardial injury and potentially call for changes in the management of myocardial infarction and other myocardial injuries.

In the 1990s, assessment of patients with chest pain was revolutionized by measurement of cardiac troponin (cTn) in blood.¹ cTn analysis has subsequently been refined, leading to the development of the high-sensitivity (hs) assays that are currently recommended.² Today, cTn represents a cornerstone in diagnosing myocardial infarction (MI) and myocardial injury.

Despite the crucial role that cTn plays in the diagnosis of MI, much is still to be learned about the elimination kinetics of these proteins. Elimination kinetics are described by elimination and distribution half-life (T½), clearance, and distribution volume. These parameters are essential to understand how elimination of cTnI and cTnT affects the concentration measured by blood sampling in the clinic and to evaluate ongoing myocardial injury.

Studies in rats have shown that receptor-mediated endocytosis by cells in the liver and kidney are the primary drivers of elimination of cTn in the bloodstream.^3,4 Controversially, some studies have shown that renal excretion is important at lower concentrations but not at higher concentrations.^5,6 However, more studies are needed because this topic is not yet fully explored.

Several previous studies have explored the clinical course of cTn concentration in plasma.^7–9 Although this information is relevant in the clinic, it does not reflect the genuine elimination kinetics of cTn. Because cTn is being degraded in the bloodstream in the aftermath of MI, the same protein may still pour into the bloodstream from the myocardium for an unknown period of time. With 2 unknown variables, influx from the myocardium and simultaneous elimination in blood, it is impossible to determine the elimination kinetics of cTn. In this study, we have found a new method to remove the effect of ongoing release of cTn from cardiomyocytes on determination of the elimination kinetics of cTn.

The aim of this study was to determine the elimination kinetics of cTnI and cTnT in humans as measured by 5 different hs-cTn assays and to explore differences between assays and the proteins cTnI and cTnT.

METHODS

In this experimental cohort study, patients who were treated for ST-segment–elevation MI (STEMI) underwent plasmapheresis within 24 hours after acute revascularization. The plasma was stored while patients recovered at home, and the concentration of cTn in the patients’ bloodstream normalized. Patients waited for at least 3 weeks before returning to the hospital for a retransfusion of their own plasma. After retransfusion, blood sampling was performed at fixed intervals for 8 hours to measure the change in concentration of cTn. This was used to estimate elimination kinetics (Figure 1).

Figure 1.

Study course steps. Step 1: Patients were admitted and treated by percutaneous coronary intervention resulting from ST-segment–elevation myocardial infarction. Step 2: Hospital charts were checked for patients eligible for participation. Step 3: Consent was obtained. Step 4: A dialysis catheter was inserted into one of the femoral veins. Step 5: Plasmapheresis was performed, and plasma was immediately frozen. Step 6: Participants finished hospitalized treatment and were discharged. Step 7: Three to 17 weeks later, participants were readmitted to perform autologous plasma transfusion. Step 8: Blood samples were collected at fixed time points from retransfusion to 8 hours later. Created with BioRender.com.

Patients

Admitted patients were screened for eligibility at the Department of Cardiology, Rigshospitalet, Denmark. Eligibility for participation was assessed on the following criteria: admitted because of STEMI with a luminal occlusion of ≥90%, age ≥18 years, a cTnT concentration >500 ng/L on admission or >1000 ng/L within the last 12 hours, time passed since acute revascularization <24 hours, left ventricular ejection fraction of at least 45%, and hemodynamic stability. The criterion for a high cTnT concentration was applied to ensure that plasma collected at plasmapheresis contained a high concentration of cTn for later retransfusion. A full list of patient eligibility criteria is provided in the Supplemental Material.

Plasmapheresis

Included patients were transferred to the intensive care unit at Rigshospitalet, Denmark, where they underwent plasmapheresis with the Prismax membrane-based plasmapheresis system (Baxter, IL). The plasmapheresis protocol used for the study was developed in collaboration with intensive care specialists, nephrologists, and representatives from Baxter. The extracorporeal circuit was primed with a solution of 3 L of isotonic of saline and 15 000 units of heparin. The Thermax blood warmer module was applied and set to 36.5 °C to 37 °C. The blood flow rate was set to 100 to 200 mL/min, and the replacement rate was set to 400 to 800 mL/h. A GamCath triple lumen catheter (Gambro, Hechingen, Germany) was inserted into one of the femoral veins and connected to the Prismax system. No systemic anticoagulation was administered during the procedure. A maximum of 800 mL of plasma was extracted in total. A solution of 800 mL of Ringer’s lactate and 200 mL of 20% albumin replaced the collected plasma 1:1. Patients’ ECG and heart rate were continuously monitored according to the hospital guideline for patients with STEMI. Blood pressure was monitored frequently before, during, and after plasmapheresis. Patients were observed for 2 hours after plasmapheresis before they were returned to the cardiac acute care unit. The plasma was collected in transfusion bags that were connected to the plasma outlet of the Prismax system. A bolus of 1000 to 1500 units of heparin was added to each transfusion bag after plasmapheresis to ensure anticoagulation. The plasma bags were then frozen at −20 °C within 3 hours of plasmapheresis initiation.

Retransfusion

To defrost plasma bags, we used the sous vide system, iVide plus thermal circulator, from SousVideTools to imitate the warm water basins used at blood banks for routine defrosting. Each plasma bag was defrosted for 15 minutes in the sous vide basin, which was set to 36.5 °C. The plasma bags were manipulated manually during defrosting. Defrosted plasma was aspired in 20- to 50-mL syringes and visually checked for signs of cryoprecipitates. A syringe was disposed of if cryoprecipitates had been aspired. Autologous plasma was then injected into a peripheral venous catheter in the cubital vein by hand force. Because the amount of autologous plasma varied between participants, the duration of retransfusion varied from 7 to 69 minutes (Supplemental Material provides individual information on retransfusion).

Blood Sampling

Blood samples were drawn from the peripheral venous catheter. A baseline blood sample was drawn before retransfusion. The next blood sample was drawn after retransfusion. Blood samples were drawn every 10 minutes in the first 1.5 hours and then every half-hour until 6 hours after retransfusion. Blood samples were drawn every hour until 8 hours after retransfusion. Before each sample, at least 3.5 mL of blood was aspired and disposed of. After each sample, 10 mL of isotonic saline was injected.

Biochemical Analysis

Blood samples were continuously sent for direct analysis at the local clinical biochemistry department. cTnI was analyzed in hospital on the Atellica IM Analyzer hs-cTnI assay (Siemens, Munich, Germany). Aliquots of plasma were stored at −80 °C and later sent to other clinical biochemistry departments for analysis on the Dimension Vista hs-cTnI assay (Siemens), the Vitros hs-cTnI assay (Ortho Clinical Diagnostics, Raritan, NJ), the Alinity i STAT hs-cTnI assay (Abbott Laboratories, Chicago, IL), and the Elecsys hs-cTnT assay (Roche Diagnostics, Basel, Germany; assays are described in the Supplemental Material).¹⁰

The Atellica hs-cTnI assay was used to measure the concentration of cTnI in plasma bags for retransfusion before and after freezing and defrosting to examine whether the process affected the concentration of cTnI (Table S1).

Ethics

Written informed consent was obtained from participants. This study was approved by the regional scientific ethics committee of the Capital Region (H-19065459) and registered with the Danish data protection authorities (VD-2019-172).

Statistical Analysis

Descriptive statistics are presented as mean with SD for normally distributed data, median with interquartile range for nonnormally distributed quantitative data, and number with percentage for categorical data. To assess whether the elimination of cTn followed zero- or first-order kinetics, the concentrations of the individual patients and assays were log-transformed and plotted against the time of sampling. T½ of the distribution and elimination phase was estimated directly for each participant with an exponential 2-phase model in which the plateau in the bloodstream was constrained to the assay-specific concentration before retransfusion.¹¹ When assay-specific concentrations before retransfusion were missing, the limit of detection of the assay in question was used in place. Model assumptions were assessed by visual inspection, R², runs test, and residual diagnostics.

Distribution volume, area under the curve, and clearance for the individual study participants were derived from the 2-phase model using the formulas in the Supplemental Material. To compare between assays, the individual distribution and elimination T½, clearances, and distribution volumes were summarized with medians and 95% CIs and visualized in forest plots. Subgroup analyses were conducted excluding participants with an estimated glomerular filtration rate <90 mL·min⁻¹·1.73 m⁻² and stratifying participants by ≤25 and >25 hours from symptom onset to plasmapheresis. The effects of age, body mass index, weight, concentration before and after retransfusion on distribution T½, elimination T½, distribution volume, and clearance were explored with simple linear regression. Sex differences were explored by comparing medians with 95% CIs between men and women in forest plots. Agreement between assays was assessed by Bland-Altman plots and limits of agreement (95% normal ranges).¹²

Because participant 15 exhibited a different pattern of decay compared with the other participants, regression analyses and Bland-Altman plots were reported both excluding and including this participant. P values were adjusted for multiple testing with the method of Benjamini and Hochberg,¹³ which controls the false discovery rate.

Data were managed using Research Electronic Data Capture, a secure, web-based, electronic data capture tool, hosted at the Capital Region server.^14,15 Nonlinear regression analyses were performed with GraphPad Prism (version 9.4.1). Postprocessing of results and descriptive statistics were made with R statistical software (version 4.2.2). The asht package (version 1.0.0) was used to compute 95% CIs for the medians (equation described in the Supplemental Material).¹⁶ Data will be shared on reasonable request and within Danish law.

RESULTS

From December 28, 2021, to September 28, 2022, 263 patients were screened, of whom 25 patients were included (Figure 2). Some eligible patients were not asked to participate because the capacity for plasmapheresis was limited to a maximum of one participant per day. Of the 25 participants included, 20 had a transfusion of autologous plasma and thereby completed participation in the study. Only 5 participants did not want the retransfusion and dropped out. All 20 participants who met for retransfusion completed follow-up blood sampling. One participant experienced venous hematoma after plasmapheresis. No other adverse events were registered.

Figure 2.

Flowchart of study inclusion and completion. Patients who had been examined by coronary angiography within 24 hours were screened for eligibility. Some eligible patients were not asked to participate because of capacity limitations, as only one participant could be included each day. cTnT indicates cardiac troponin T.

Baseline characteristics for all participants included (n=25) are presented in the Table S2 and for participants completing the study (n=20) in the Table. All participants were of European ancestry. The mean age of participants completing the study was 64.5 years (SD, 8.2 years), and 4 were women (20%). Mean left ventricular ejection fraction was 50.8% (SD, 4.7%), and 19 had a luminal stenosis of at least one coronary artery of ≥99% (95%). Mean body mass index was 29.2 kg/m² (SD, 4.4 kg/m²), and one participant had chronic kidney disease (5%).

Table.

Baseline Characteristics of Participants Completing the Study

A median of 18.4 hours passed between acute revascularization and plasmapheresis (interquartile range, 16.5–20.3 hours; range, 3.9–26.0 hours). From plasmapheresis to retransfusion, a median of 5.8 weeks passed (interquartile range, 5–6.9 weeks; range, 3.0–17.5 weeks; Table S3). The concentration of cTnI, as measured on the 4 included hs-cTnI assays, and cTnT in plasma gained by plasmapheresis varied considerably, but the amount of cTn harvested was high for all participants (Table S3). Hence, the concentration of cTn in participants increased to concentrations 4 to 445 times the upper reference limit of all hs-cTn assays after autologous retransfusion (Table S3).

Of the 440 samples planned (22 per participant at retransfusion), 17 samples (3.9%) were not carried out because of a lack of time to perform sampling repeatedly every 10th minute at the start of follow-up after retransfusion. A total of 2115 biochemical analyses on hs-cTn assays were performed, of which 65 (3.1%) failed, primarily because of hemolysis. Four results of biochemical analysis were removed for the Alinity hs-cTnI assay during data quality control (further information is provided in the Supplemental Material).

The concentration of cTn was plotted on a logarithmic scale from retransfusion to 8 hours later, resulting in a curvilinear course (Figure S1). An exponential 2-phase model was found to be suitable to describe the first-order decay. In Figure 3 the concentrations of hs-cTnT and hs-cTnI are plotted as a scatterplot from the end of retransfusion to 8 hours later, and the exponential 2-phase model is plotted as curves for each participant. The elimination-phase median T½ ranged from 134.1 minutes (95% CI, 117.8–168.0) for the Elecsys hs-cTnT assay to 239.7 minutes (95% CI, 153.7–295.1) for the Vitros hs-cTnI assay (Figure 4A and 4B). The median elimination of hs-cTnI by the Atellica assay and median elimination of hs-cTnT by the Elecsys assay are illustrated in Figure S2, and the median elimination from a standardized starting concentration is illustrated for all assays in Figure S3.

Figure 3.

Decay of cTnI and cTnT over time. Decay of cardiac troponin (cTn) visualized by concentration from the end of retransfusion to 8 hours after measured by 5 high-sensitivity (hs) cTn assay. Scatterplots represent individual measurements for each participant. The exponential 2-phase model is plotted as curves for each participant. Participant 10 is marked by dark red; ◊ indicates the only participant with known chronic kidney disease. cTnI indicates cardiac troponin I; and Conc., concentration.

Figure 4.

Comparison of elimination kinetics across cTn assays. Comparison of elimination kinetics across cardiac troponin (cTn) assays. A, Median distribution half-life of cTn according to high-sensitivity (hs) cardiac troponin I (cTnI) and hs–cardiac troponin I (cTnT) assays with 95% CIs. B, Median elimination half-life of cTns according to hs-cTnI and hs-cTnT assays with 95% CIs. C, Median clearance of cTns by hs-cTnI and hs-cTnT assays with 95% CI. D, Median distribution volume of cTns according to hs-cTnI and hs-cTnT assays with 95% CIs.

The median clearance of cTnI ranged from 40.3 mL/min (95% CI, 32.0–44.9) for the Vista hs-cTnI assay to 52.7 mL/min (95% CI, 42.2–57.8) for the Atellica hs-cTnI assay. The clearance of cTnT was 77.0 mL/min (95% CI, 45.2–95.0), as measured on the Elecsys hs-cTnT assay (Figure 4C). Median distribution volume ranged from 5.1 L (95% CI, 4.7–5.7) to 6.2 L (95% CI, 5.2–6.9; Figure 4D). The dose of cTn retransfused was not sampled for the first 3 participants. As a result, clearance and distribution volume could not be estimated for them.

In the subgroup analysis excluding participants with an estimated glomerular filtration rate <90 mL·min⁻¹·1.73 m⁻² (n=11), the estimated medians of distribution and elimination T½ and clearances were similar to those of the main analyses (Table S4). This was also the case when participants were stratified by ≤25 hours (n=10) and >25 hours (n=10) from symptom onset to collection of plasma by plasmapheresis (Table S5).

Simple linear regression did not reveal any systematic associations of distribution T½, elimination T½, clearance, and distribution volume with age, weight, body mass index, or concentration before or after retransfusion when participant 15 was excluded (Tables S19 through S28). Moreover, differences in distribution T½, elimination T½, clearance, and distribution volume between men and women could not be assessed because CIs could not be calculated for the women, as the sample size was too small (Table S29).

Agreements between assays for distribution T½, elimination T½, clearance, and distribution volume are presented in Figures S4 through S11; Tables S31 through S38. Limits of agreement (LOAs) were calculated with and without participant 15, who exhibited substantially different kinetics compared with the other participants and poor agreement between assays. After the exclusion of participant 15, the best agreement for the distribution T½ was found between Atellica hs-cTnI and Elecsys hs-cTnT (LOA, −12.5 to 10.1 minutes) and the least between Alinity hs-cTnI and Elecsys hs-cTnT (LOA, −21.5 to 26.6 minutes). The best agreement for the elimination T½ was found between Vista hs-cTnI and Atellica hs-cTnI (LOA, −9.2 to 111.5 minutes) and the least between Alinity hs-cTnI and Elecsys hs-cTnT (LOA, −401.4 to 553.6 minutes).

Individual clinical information and kinetic profiles are given in Tables S6 through S18.

DISCUSSION

The main findings in this study were: (1) when cTn entered the bloodstream, it followed an exponential 2-phase model with a distribution and elimination phase; (2) the elimination T½ of cTnI and cTnT was 5 to 16 hours shorter than what has previously been observed clinically in patients with ongoing myocardial injury; (3) clearance of cTnI and cTnT in humans, examined for the first time, ranged between 40.3 and 77.0 mL/min; (4) the results were consistent across cTnI and cTnT and across the different assays; and (5) in humans, it is often not feasible to inject exogenous biomarkers because of safety and ethical concerns. This study showed that it is possible to examine biomarker elimination kinetics in humans by autologous retransfusion.

In this study, myocardial cTn release was simulated by injecting cTn intravenously. The decay of both cTnI and cTnT exhibited a nearly perfect exponential 2-phase model with a distribution and an elimination phase. A study of the decay of cTn in rats has shown that after infusion, cTn is distributed to cells of the liver and kidney and, to a smaller extent, to the spleen and small intestines.⁴ A similar distribution in humans fits well with the proposed model. If a T½ measured in a patient in the clinic is higher than the 95% CIs of the elimination T½ of this study, then it can be assumed that myocardial cTn release is ongoing in that patient.

When the distribution has reached a steady state, the elimination of cTn can be measured as the T½ of the elimination phase. Our study revealed notably shorter elimination T½ compared with previously reported elimination T½ of patients with ongoing MI. One study found the elimination T½ of cTnI in patients with MI (n=22) to be 20.4 hours (SD, 10.7 hours) and 6.8 hours (SD, 5.6 hours), according to whether Q waves were present on the ECG.⁹ Another study by our group examined the elimination T½ as estimated by 4 hs-cTnI assays and 1 hs-cTnT assay in another cohort of 36 patients with STEMI.¹⁷

The median elimination T½ was found to be 12.4 hours (95% CI, 11.0–14.1) to 14.7 hours (95% CI, 12.4–18.2) for cTnI and 17.3 hours (95% CI, 14.9–20.8) for cTnT as measured on hs-cTn assays. These studies did not take the distribution phase into account because the time period used to estimate elimination T½ started from the peak concentration of cTn in the bloodstream, and thus it must be assumed that the distribution of cTn had reached steady state.

The discrepancy between the elimination T½ observed in the clinic and that in our study may be attributed to several factors. Observational clinical studies are prone to estimate a longer elimination T½ if ongoing release of cTn from the myocardium occurs while blood sampling is performed. Our results show that the elimination T½ of cTn in the bloodstream in fact is much shorter than previously anticipated. This strengthens the assumption that the previous clinically estimated elimination T½ does not represent the genuine elimination T½ of cTn in the bloodstream; instead, it may be due to ongoing release from the myocytes days after MI. Because cTn can be measured for an extended time period after MI, we question whether the term myocardial injury, which is central to the field of cardiology and a diagnosis of MI, is the proper term. Release of cTn from the myocardium may not necessarily imply ongoing injury, as shown in several studies.^18–20 The ongoing release from myocytes may also explain the discordance between the elimination T½ as estimated in animal models and in patients with MI. Studies involving animal models in which exogenous cTnI was injected have demonstrated a considerably shorter elimination T½ in dogs (1.9 hours), rats (0.8 hours), and horses (0.5 hours), which better resemble the results of this study (2–4 hours).^21,22 However, some differences remain. The elimination T½ was examined up to 240 minutes after injection in 4 ponies by Kraus et al.²¹ The calculations of the elimination T½ in ponies did not account for a distribution phase of cTn or the possibility of cTn plateau; therefore, the reported elimination T½ may have been overestimated (faster).

The T½ has also been examined in 30 rats and 9 beagle dogs from injection of exogenous cTn to 24 hours after by Dunn et al.²² The study accounted for the distribution phase and baseline concentration of cTn; however, those investigators did not describe the statistical method used to examine the elimination T½ and clearance. Clearance was reported per kilogram in rats and dogs. In humans, we did not find evidence that weight influences clearance.

Seven samples were drawn in dogs and rats, and serum was analyzed on the Singulex Erenna Ultrasensitive Immunoassay in the study by Dunn et al.²² Six samples were drawn, and whole blood was analyzed on a point-of-care ELISA in the study by Kraus et al.²¹ The difference in samples drawn and assay used may also explain the differences compared with the present study.

It is worth noting that the use of exogenous, recombinant cTn may result in autoantibodies against the exogenous cTn, which may influence elimination kinetics.^23,24 The elimination kinetics may also vary when an infusion of cTn generated exogenously is administered compared with autologous plasma, which includes different forms of the cTn complex and plasma components. Thus, species-specific differences and the use of exogenous cTn in the animal models pose challenges in the extrapolation of findings to humans. Furthermore, differences in methodology, blood product, and assays used should be taken into account when the results obtained with animal models are compared with the current study.

In studies of the elimination kinetics during exercise-induced cTn release, the elimination T½ from peak concentration has not been reported.^25,26 A recent study by Legaz-Arrese et al²⁶ showed that the median human cTnT concentration peaked at 3 hours after exercise and had reverted to baseline within 24 hours.²⁵ The peak concentration varied considerably according to age group, and 43% to 77% of participants exceeded the upper reference limit at median peak concentration. The rapid decline from peak concentration to baseline concentration after exercise-induced cTn release may feature elimination kinetics closer to the results of this study than the elimination kinetics measured after STEMI in a clinical setting.

Interindividual variation may also lead to differences in elimination T½ and clearance. This may be attributable to differences in individual characteristics such as age, sex, weight, body mass index, and liver and kidney function. Differences in liver or kidney function may be attributable to disease or genetic variation.^27,28 We have not found any previous studies reporting interindividual variation in relation to elimination kinetics of cTn. In this study, we found no evidence of a significant association between elimination kinetics and examined characteristics.

Participant 15 exhibited a different kinetic profile compared with all the other participants, with notably longer distribution and elimination T½ of cTnI measured on the Vista hs-cTnI assay and the Atellica hs-cTnI assay compared with the other participants. The intraindividual estimation of distribution and elimination T½ on the Vista hs-cTnI and Atellica hs-cTnI assays in participant 15 was also notably longer than estimations on the other assays. Before retransfusion, the concentrations of cTnI and cTnT in this participant were some of the highest of all participants, even though 4.4 weeks had passed since acute revascularization. The Vista and Atellica hs-cTnI assays featured the same epitopes for capture and detection but in reverse (Table S30). This may explain why these assays exhibited different distribution and elimination T½ for this participant compared with the estimations by the other assays. The difference in the distribution and elimination T½ in participant 15 compared with the other participants may be due to circulating macrocomplexes of immunoglobin-bound fragments of cTnI.²⁹ This has previously been found to cause increased concentration of cTn in patients without myocardial ischema.³⁰

The interindividual elimination kinetics may vary, depending on a difference in circulating cTn fragments and circulating cTn complex formations.³¹ cTn is degraded primarily by receptor-mediated endocytosis, and the uptake may vary for different fragments and cTn complex formations because of a difference of receptor affinity or size of the complex formation.^3,4 Likewise, smaller fragments are more easily excreted into urine.^5,6 No evidence exists suggesting that there is a difference in cTn fragmentation or complex formation in infused autologous plasma by retransfusion compared with the real-world scenario.

Complex formations featuring fragmented cTnI with the specific epitopes of the Atellica and Vista hs-cTnI assays may explain why participant 15 exhibited slower distribution and elimination T½ for these assays; these complex formations may recirculate without being eliminated.

Interassay variation of elimination kinetics may be attributable to the different capture and detector antibodies used in the assays because some fragments of cTn may be more prevalent than others. Interassay variation may also be attributable to a difference in the reuptake of fragments during receptor-mediated endocytosis.⁴

The cTn concentrations in response to a controlled significant myocardial injury (eg, after transcoronary septal alcohol ablation in patients with obstructive hypertrophic cardiomyopathy) have been reported. The controlled myocardial injury in this setting may represent better circumstances for the determination of the elimination kinetics of cTn. However, in this situation, we will also expect ongoing cell death and therefore an ongoing cTn release for hours or even days after the injury.³² It is also not possible to estimate the clearance and distribution volume because it is impossible to know the amount of cTn released from cardiomyocytes.

Limitations

To enhance participant comparability and to provide a comprehensive view of kinetics, this study used strict inclusion criteria and frequent sampling. In addition, the simultaneous sampling of all biomarkers allowed direct comparison between the biomarkers.

However, there are some limitations. Because of the nature of the study, it was not possible to administer the same amount of cTn to patients. In addition, it must be assumed that the decay of cTn had begun before retransfusion. However, this may be of minor importance because the breakdown of cTn is thought to begin intracellularly before it has even been released into the bloodstream, and the ongoing breakdown is also a feature present in the aftermath of MI as seen in the clinic.¹⁹ However, we do know that the size of cTn fragments changes over time after MI, and this could affect the kinetics.³¹ A sensitivity analysis of clearance stratifying participants according to time from symptom onset to collection of plasma did not show a difference, but this may be because of the small sample size or misregistered timing of symptom onset by patients. The fragmentation of cTn may differ for the early release of cTn found in the cytosol compared with the later release from disintegrating myofibrils as shown by Katus et al³³ and Remppis et al³⁴ for cTnT.

We reported that elimination kinetics may be applied directly for each of 5 assays in clinical use. However, it would also be of interest to examine the elimination kinetics of the full cTnI and cTnT peptides without cTn fragments by mass spectrometry.

Plasma for autologous retransfusion was obtained by membrane-based plasmapheresis, which filters molecules according to size (up to 3×10⁶ Da). The plasma was fresh-frozen to quickly stop the process of decay. We believe the extraction process to be well suited for the purpose of this experimental study, but interference from unknown factors cannot be ruled out. Studies have previously examined the stability of cTn in stored frozen plasma, which is stable for at least 1 year.^35,36 Plasma bags were stored at −20 °C to mitigate cryoprecipitate formation during the process of defrosting before retransfusion. Before the study, we examined whether defrosting was affected by the temperature of the plasma bags before defrosting by defrosting 5 plasma bags stored at −20 °C compared with 5 plasma bags stored at −80 °C . Fewer to no cryoprecipitates formed in the plasma bags stored at −20 °C by visual inspection. To the best of our knowledge, no study has examined whether changes occur in the structure of cTnI or cTnT during storage as frozen plasma.

A bolus of 1000 to 1500 units of heparin was added to each transfusion bag with plasma after plasmapheresis to ensure anticoagulation. All participants had also received heparin intravenously during primary percutaneous coronary intervention (5000–10 000 units). A previous study has shown that the specific cleavage of cTnT into the 29-kDa fragment of cTnT occurs mainly as a result of the activation of thrombin in vials of serum in vitro.³⁷ All sampling of cTnT in the clinic was done in vials containing heparin, which was also the case in our study. We do not expect the addition of heparin to influence the elimination kinetics in plasma, but there may be differences between the elimination kinetics in plasma and serum resulting from the inactivation of thrombin in plasma. To the best of our knowledge, no study has examined whether heparin influences cleavage of cTnI.

Missingness of data in this study was very low, and because the data exhibited a nearly perfect exponential 2-phase model, this study could have been carried out with less frequent sampling, especially for the first 1.5 hours.

Most patients had concentrations of cTn below the 99th sex- and assay-specific upper reference limit at retransfusion. Measurable cTn concentrations may be attributable to macrotroponin complexes recirculating or slight myocardial release.¹⁹ Given that several weeks had passed between STEMI and retransfusion, we can assume that the concentration of cTn in the bloodstream was close to steady state. The concentration before retransfusion was accounted for in all calculations by subtracting the plateau. The individual baseline cTn concentration by each assay was missing for 3 of a total of 100 samples (Table S11). For the 2 participants for whom these 3 concentrations were missing, none of the measurements by the other assays were above the 99th percentile; therefore, we used the limit of detection as baseline concentration for assays for which the baseline was missing. This study was based on cTn harvested from patients with STEMI. Elimination kinetics may potentially have been different if cTn had been harvested from patients with non-STEMI.

Conclusions

The elimination T½ of cTn is 2 to 10 times shorter than previous observations of the combined myocardial cTn release and elimination T½. We also determined the clearance of cTn in the human bloodstream for the first time, which ranged between 40.3 and77.0 mL/min. These results are of importance when evaluating the turnover of cTn in the bloodstream of patients presenting with myocardial injury. Because elevated concentrations of cTn can be measured several days after MI, this study supports the assumption that myocardial cTn release or washout from the infarct area is prolonged for days in the aftermath of MI.

These results will enable us to better adjudicate timing of not only myocardial infarcts but also myocarditis and other disorders with myocardial injury. This study showed that it is possible to examine biomarker elimination kinetics in humans by autologous retransfusion.

ARTICLE INFORMATION

Sources of Funding

This study was supported by funding from Candy’s Foundation, Murermester Lauritz Peter Christensen og hustru Kirsten Sigrid’s grant, Research Council of Herlev and Gentofte Hospital, and Mauritzen la Fountaine foundation. The benefactors had no influence on study design or execution.

Disclosures

Dr Bundgaard received payment or honoraria for lectures, presentations, speakers’ bureaus, manuscript writing, or educational events from Amgen, Sanofi, BMS, and MSD. Dr Bundgaard owns stock or stock options in Novo Nordic. Dr Kamstrup reported a grant from Gangsted Fonden outside the present work. P.R.K. reported consulting fees from Novartis and Silence Therapeutics. Dr Kamstrup reported payment or honoraria for lectures, presentations, speakers’ bureaus, manuscript writing, or educational events from Physicians’ Academy for Cardiovascular Education, Novartis, and PCSK9 Forum. Dr Frikke-Schmidt reported grants or contracts outside the present work from Lundbeck Foundation, The Danish Heart Foundation, and Sygeforsikringen Denmark Research Fund. Dr Kjærgaard reports a grant or contract from Novo Nordisk Foundation outside the present work. Dr Holmvang reported personal payment or honoraria for lectures from Boehringer Ingelheim and payment or honoraria to her institution for lectures from Bayer. Dr Holmvang reported receiving support for travel from Abbott to her institution. Dr Bor reported receiving honoraria for a lecture from Bristol-Myers. Dr Thygesen reported participation on the Data Safety Monitoring Board of DANBLOCK (Danish Trial of Beta Blocker Treatment After Myocardial Infarction Without Reduced Ejection Fraction) and REDUCE (Randomized Evaluation of Decreased Usage of Betablockers After Myocardial Infarction in the SWEDEHEART Registry). Dr Jaffe reported royalties or licenses and stock or stock options to RCE Technologies; consulting fees from Abbott, Roche, Beckman-Coulter, Radiometer, Siemens, Ortho Diagnostics, Spinship, and LuminaRx; and support for attending meetings and/or travel from the American Association for Clinical Chemistry. Dr Dahl reporting receiving payment or honoraria for lectures, presentations, speakers’ bureaus, manuscript writing, or educational events from Grifols and Ciesi to himself and support for attending ERS Congress 2022 from Grifols.

Supplemental Material

List of inclusion and exclusion criteria

Assays used for biochemical analysis

Formulas for calculation of area under the curve, clearance, and distribution volume

Method of calculating CIs for medians

Data quality control

Tables S1–S40

Figures S1–S11

Supplementary Material

cir-150-1187-s001.pdf ^{(2MB, pdf)}