Understanding the high-sensitivity Troponin can be confusing. The literature, for the most part, was done outside of the United States using assays that are not approved by the FDA at this time. The first barrier to understanding the hs-cTn is that, as clinicians, we have to get used to seeing some level of measurable Troponin in a healthy population. To make it even more confusing, the upper reference limit for what is considered a normal value is different for every high-sensitivity assay on the market. This summary is meant to bridge the gap of understanding.
High-Sensitivity Troponin Overview
Troponins (I, C and T) are components of a complex of proteins that modulate the calcium‐mediated interaction between actin and myosin within cardiac myocytes. Troponin T anchors the troponin complex to tropomyosin of the thin filament, Troponin I inhibits actomyosin ATPase and Troponin C is a calcium‐binding subunit. There are three isoforms of troponin I (TnI). Two are unique to skeletal muscle, and one is unique to cardiac muscle (cTnI). This allows us to use cTnI as a highly sensitive marker for myocardial injury.
hs-cTn assays measure conventional Troponin (cTn) concentrations 5-fold to 100-fold lower than conventional assays do. These assays are so sensitive that weare now able to detect troponin levels in normal populations and many of patients will now likely have some measurable level of hs-cTn.
Timing of Elevation
cTnI levels begin to rise in the first few hours after cardiac injury and peaks around 12-16 hours. These levels can remain elevated for 4-9 days following a Myocardial Infarction. One meta-analysis suggests that the duration from symptom onset must be at least 2 hours to achieve a 99% negative predictive value in ruling out an acute MI (Chapman, 2017).
Definitions to help understand the literature...
- Conversion to remember: 1 pg/ml = 1 ng/L…they are the same and you will see both used in the relevant literature
- Limit of Blank (LoB): The background noise present in the analytical measurement system when no Tn is present.
- Limit of Detection (LoD): The lowest detectable concentration of detectable in 95% of measurements
- Old “standard” assays = 100 – 1000 ng/L (0.1 – 1.0 μg/L)
- Sensitive assays = 30 – 50 ng/L (0.03 – 0.05 μg/L)
- High-Sensitivity assays = < 20 ng/L, and some as low as 0.5 ng/L
- Coefficient of Variation (CoV): A measure of assay imprecision at any given concentration. The coefficient of variation value should be 10% or less at the 99th percentile upper reference limit (URL) for hs-cTn assays.
- 99th percentile clinical decision value (99Th% CoV): This is the 99th percentile for normal, healthy individuals of measured troponin and determines the URL of hs-cTn assays.
Troponin results now fall into 3 categories
- Negative = < LoD (undetectable levels)
- A single value below 2 pg/ml (2 ng/L) is considered undetectable
- Normal Range = LoD to the 99th percentile (≠ negative)
- Male: < 20 pg/ml (ng/L)
- Female: < 12 pg/ml (ng/L)
- Abnormal = any number > 99th percentile CV
- An abnormal level is now any number > 99th percentile of normal population URL
(numbers specific to the Beckman Coulter hs-cTn I assay)
hs-cTn > URL does NOT always equal ACS/Type I MI
It is important to point out that the evidence leading to the implementation of protocols, for the most part, set out to either determine 30-day Major Adverse Cardiac Events (MACE) or to rule in and rule out acute MI in the clinical setting of suspected Acute Coronary Syndrome (ACS). Further, the upper reference limits are derived from HEALTHY individuals with an average age of 36 +12 years, most of whom have some degree of measurable troponin. Levels of high-sensitivity cardiac troponin should be interpreted as quantitative markers of cardiomyocyte as it is common to detect circulating levels of cardiac troponin in healthy individual.
What levels are truly significant (highly specific) for an acute MI?
- Elevations > 5-fold the URL have a > 90% PPV for acute type I MI
- Elevations up to 3-fold the URL have only a 50-60% PPV for acute MI and may be associated with a broad spectrum of conditions.
Buuuut…. according to the fourth universal definition of MI…
”it is not clinically possible to distinguish which increases of cTn levels are due to which mechanisms. However, regardless of the mechanism, acute myocardial injury, when associated with a rising and/or falling pattern of cTn values with at least one value above the 99th percentile URL and caused by myocardial ischemia, is designated as an acute MI.”
What do you do with an incidental hs-cTn elevation?
Bottom line….it may not be benign.
An elevated high sensitivity troponin has been shown to predict overall badness. In one study, an elevated hs-cTn level above the upper reference limit has been described to predict up to a 31% 12-month mortality, and even higher if due to a non-ACS related elevation (Petrie, et al., 2014). At present, experts are uncertain on what to do with this information for the most part, but something to keep in mind as evidence continues to build on the topic of incidental hs-cTn elevations. As always, if you are ordering a test, make sure you have a good reason to order it in the first place.
Higher Sensitivity = Lower Specificity
Clinical Implications of high-sensitivity (low specificity) Assays
- hs-cTn has a higher negative predictive value for acute MI
- Reduces the “troponin-blind” interval leading to earlier detection of acute MI
- This results in a 4% absolute and 20% relative increase in detection of type I MI and a corresponding decrease in the diagnosis of unstable angina as this is more likely to be classified as a NSTEMI now.
- Associated with a 2-fold increase in the detection of type 2 MI
Does lower specificity mean higher cost and more work-ups?
- Compared to conventional diagnostic pathways, the 0/1-hour accelerated diagnosis algorithm resulted in fewer overnight ED stays (60.3% vs. 24.1%), fewer cardiology referrals (32.4% vs. 20.9%), and lower healthcare human resources ($11,236.80 vs. $8,528.20 USD). Use of hs-cTnT was associated with an ICER of $14,082.40 per QALY gained. The majority of incremental cost-QALY pairs fall in the dominant quadrant of the CE plane, indicating that using hs-cTnT is cost-saving with better outcomes (Hsu, 2018).
- Hs-cTn testing is more cost-effective in younger age, pre-existing coronary artery disease, and symptom onset < 3 hrs ago (Ramaekers, 2014).
- In one multicenter study of 2544 patients, there was not only no change in the rate of angiography, but a significant decrease in the need for stress testing from 29% to 19% and median time to discharge was decreased by 79 minutes (P < 0.001).They concluded that introducing a high-sensitivity troponin assay “…was associated with an improved rule-out process and thereby helped to substantially reduce the need for cardiac stress testing and time to discharge.” Total costs were also reduced (Twerenbold, 2016).
- In a recent prospective observational multicenter study performed at 6 centers in Sweden, combining a 0/1hour hs-cTn with the HEART score decreased admission rates from 59% to 33% and decreased median health-care costs from $1,748 to $1,079 with no significant difference in clinical outcomes. However, it is important to note that this study was not randomized and was underpowered to detect differences in clinical outcome (Ljung, 2019).
What is a significant delta?
Some numbers to keep in mind (specific to the Beckman Coulter assay):
- ~ 30 % of patients will have a level below the LoD (Apple, 2016).
- Variation in a single individual is ~ 6% (Apple, 2016).
- Variation based on analytical error is ~ 14% (Apple, 2016).
- In one study, 54% of patients with renal dysfunction will have baseline > 99th % URL (Twerenbold, 2015).
Setting the Delta:
- A Delta of 20% has the highest sensitivity and was the initial recommended delta in literature (Keller, et al., 2011).
- More recently, a delta of 50% has been advocated by some, including the American College of Cardiology expert analysis. However, it seems apparent that this recommendation is based off of early research done with hs-cTn T in the Europe and the Pacific-Asian islands.
- Troponin I literature shows loss of sensitivity with a 50% cutoff, but this doesn’t appear to change overall mortality (Saajeev, 2015).
- The ACC also recommends setting an absolute number when defining a delta below the 99th percentile CV, and not a percentage.
- Study Title: “Application of High-Sensitivity Troponin in Suspected Myocardial Infarction.”
- Study Design: Meta-analysis of 15 international cohorts of 22,651 patients presenting to the ED with symptoms suggestive of myocardial infarction.
- Study objective: To determine the concentrations of hs-cTn I or hs-cTn T at presentation and after early or late serial sampling.
- Study Methods: Diagnostic and prognostic performance of multiple hs-cTn cutoff combinations were assessed with the use of a derivation-validation design.
- Results:
- Prevalence of MI was 15.3%
- hs-cTn I concentrations of < 6 ng/l with a delta of < 4 ng/l after 45-120 minutes (early serial sampling) had a NPV of 99.5% for MI, with an associated 30-day risk of subsequent MI or death of 0.2%
- 56.5% of patients would be classified as low risk.
- These findings were confirmed in an external validation data set
- Study Conclusions: This study proposes a risk-assessment tool based on initial sample concentrations and time of repeat sample to provide a NPV and 30-day risk or MI or death based on either hs-cTn I or hs-cTn T assays. This tool can be accessed here.
My Recommendation for interpretation of a significant delta based on review of the evidence:
- If the initial value is BELOW the URL for age with symptoms > 3 hrs: a significant 1-hour delta is a > 50% change
- If the initial value is BELOW the URL for age with symptoms < 3 hrs: a significant 2-hour delta is a > 50% change
- If the initial value is ABOVE the URL for age with symptoms < 3 hrs: a significant 2-hour delta is a > 20% change
Are there false negatives?
- Yes, but they are rare and usually due to anti-cTn antibodies.
- Hs-cTn assay values are lowered by hemolysis.
- Supplemental biotin may also interfere with the assay and falsely lower values.
- Chronic diseases such as CKD or HF can be associated with elevated background concentrations of hs-cTn. Thus, assessment of acute symptoms in affected patients can be challenging. Although detection of chronic myocardial injury identifies a patient at higher risk, a standard approach for evaluation of such patients remains undefined.
Special Clinical Circumstances
hs-cTn after Non-Cardiac Surgery
- Deveraux, 2017:
- In contrast to conventional Troponin assays, high-sensitivity assays reveal that up to 35% of post-operative patients have cardiac injury with measured hs-cTnT concentrations above the 99th percentile value of 14 ng/l.
- Elevations can be used to predict 30-day mortality:
- < 14 ng/l: mortality = 0.01% to .05%
- 14 to 20 ng/l: mortality = 1.1%
- 21 to 64 ng/l: mortality = 3.0%
- 65 to 999 ng/l: mortality = 9.1%
- ≥1,000 ng/l: mortality = 29.6%
- The investigators also found the larger the change from baseline, the worse the prognosis.
- Besides AMI, it is reasonable to assume multiple causes of troponin elevation (such as PE, HF, sepsis) were present
- Only 7% of patients in this study had ischemic symptoms, and the ECGs were often not informative.
- Among those thought to have AMI, most were assumed to be type 2 Mis.
- 50% of the events that lead to mortality are due to plaque rupture (Gualando, 2012).
- Interpretation of hs-cTn after non-cardiac surgery must incorporate the entire clinical picture without underestimating the importance of a large rise in hs-cTn.
- Some authors recommended a baseline, pre-operative sample in patients deemed at risk to facilitate interpretation (Thygesen, 2018).
hs-cTn after Percutaneous Coronary Intervention (PCI)
- Thygesen, 2018:
- A rising delta is expected for those with rising concentration before the procedure.
- On the basis of the Universal Definition of MI, criteria for a type 4A MI include elevation of cardiac troponin values > 5 times the 99th percentile URL in patients with normal baseline values (≤ 99th percentile URL) or a rise of cardiac troponin values > 20% if the baseline values were elevated and are stable or falling.
- However, abnormal hs-cTn alone is not diagnostic of a type 4A MI, because the definition requires:
- 1) new ischemic ECG changes
- 2) angiographic findings consistent with a procedural flow-limiting complication
- 3) imaging demonstration of new loss of viable myocardium or new regional wall motion abnormality.
- Without one of these major criteria, a PCI-related MI cannot be reliably diagnosed.
Chronic Kidney Disease
- Cardiovascular diseaseaccounts for 43% of all-cause mortality in ESRD.
- Interpretation of elevations of hs-cTn can be difficult in these patients as concentrations are frequently above the 99th percentile in the absence of AMI
- hs-cTn I is elevated in 38% of CKD patients
- hs-cTn T is elevated in 68% of CKD patients
- As CKD progresses, baseline levels of both hs-cTnT and hs-cTnI gradually rise (Parikh, 2015).
- This is possibly due to underlying structural heart disease and decreased clearance (deFilippi, 2012).
- Recent dialysis has been shown to decrease hs-cTn T by up to 12% (Chen, 2017).
- Despite these issues, accurate use of hs-cTn to identify or exclude AMI in those with CKD is still possible.
- While differences in optimal cutoffs might exist, the ACCdoes not endorse specific cutoffs in those with CKD because of concerns for false negatives.
- The American College of Cardiology does emphasize using the absolute changes in hs-cTn I or T concentrations during serial sampling as these do not differ between MI patients with and without CKD (Twerenbold, 2015).
- A recent prospective, European multicenter study demonstrated that when using hs-cTnT in a 0/1 hour triage algorithm for patients with CKD, there was overall similar sensitivity of rule out but lower specificity of rule in and lower overall efficacy (Twerenbold, 2018).
Heart Failure
- hs-cTn above the 99th percentile are common among patients with HF, and a rise or fall can often be seen in those with acute decompensation of HF.
- Similar to patients with CKD, serial testing of hs-cTn concentrations can help differentiate MI from chronic hs-cTn elevations due to HF
- That being said, an elevated hs-Tn does have some prognostic implications in the setting of chronic heart failure. In a meta-analysis of 10 studies, reporting data on 11 cohorts and 9289 patients, a hs-Tn T above the URL was independently associated with all-cause mortality (hazard ratio, 1.48; 95% CI, 1.41-1.55), cardiovascular mortality (hazard ratio, 1.40; 95% CI, 1.33-1.48), and cardiovascular hospitalization (hazard ratio, 1.42; 95% CI, 1.36-1.49), over a median 2.4-year follow-up (all P<0.001) (Aimo, 2018).
Society Specific Recommendations
- Acceptable miss rate: “Based on limitations in diagnostic technology and the need to avoid the harms associated with false-positive test results, the committee based its recommendations on the assumption that the majority of patients and providers would agree that a missed diagnosis rate of 1%-2% for 30-day MACE in NSTE ACS is acceptable.”
- A single high-sensitivity troponin result below the level of detection on arrival to the ED, or negative serial high-sensitivity troponin result at 0 and 2 hours is predictive of a low rate of MACE. (Level C Recommendation)
- In adult patients with suspected acute NSTE ACS who are determined to be low risk based on validated ADPs that include a non-ischemic ECG result and negative serial high-sensitivity troponin testing results both at presentation and at 2 hours can predict a low rate of 30-day MACE allowing for an accelerated discharge pathway from the ED. (Level C Recommendation)
ACEP Clinical Policy Summary: “In summary, although the studies varied a great deal in the type of troponin test used and whether a repeated test was performed, a few reasonable conclusions are possible. At least in the Asia-Pacific region, a 2-hour ADP applied to a select group of low-risk ACS patients that uses a high-sensitivity troponin test can identify those with a low 30- day rate of MACE (ASPECT Trial). A single high-sensitivity troponin test result below the level of detection, a single high-sensitivity troponin test result applied to a low-risk cohort, or serial high-sensitivity troponin test results within 3 hours have all been demonstrated to reduce MACE. “
- Due to the higher sensitivity and diagnostic accuracy for the detection of acute MI at presentation, the time interval to the second cardiac troponin assessment can be shortened with the use of high-sensitivity assays.
- For rapid rule-out, two alternative approaches to the 0 h/1 h or 0 h/3 h algorithms have been adequately validated and may be considered.
- High-sensitivity cardiac troponin assays also maintain high diagnostic accuracy in patients with renal dysfunction. To ensure the best possible clinical use, assay-specific optimal cut-off levels, which are higher in patients with renal dysfunction, should be used.
- For conventional troponin, it is recommended to use the 0 h/3 h algorithm (figure 2).
- 0 h/1 h assessments are recommended when high-sensitivity cardiac troponin assays with a validated algorithm are available (figure 3).
- The cut-off levels within the 0 h/1 h algorithm are assay specific. Those algorithms should always be integrated with a detailed clinical assessment and 12-lead ECG and repeat blood sampling is mandatory in case of ongoing or recurrent chest pain.
figure 3 (accessed here)
AHA/ACC Recommendations
The AHA/ACC guidelines were last updated in 2014 (can be accessed here) and do not specifically address high-sensitivity Troponin. A 2019 publication from the AHA/ACC suggests they are attempting to fill this gap and address a revised guideline in the future, but this has yet to be released. The American College of Cardiology does not have a specific guideline for the use of high-sensitivity Tn. Below is a summary of the AAC expert analysis of the use of hs-cTn T. Keep in mind, the Beckman Coulter assay is a hscTn I.
Journal of American College of Cardiology Expert Panel Recommendations (Januzzi, 2019)
hs-cTn concentration exceeding the 99th percentile of a normal population will be referred to as elevated or abnormal. Although changes below the 99th percentile may reveal acute coronary events, use of such lower concentrations is not yet endorsed by the Fourth Universal Definition of MI.
- Delta Troponin
- The cutoff quantity for defining a rise or fall must be determined for each individual troponin assay. One approach for interpretation at lower values suggests that a change threshold be set at 50% to 80% of the baseline concentration (which comports to the sum of analytic and biological variation) (Apple, 2017).
- For example, for hs-cTnT, a change of 7 ng/l from a baseline of 14 ng/l would be significant (Mueller, 2011 and Keller, 2011).
- Present data and the fourth universal definition of MI suggest that the use of absolute values rather than percentages provides better diagnostic information (Mueller, 2011).
- Chronic Comorbid Conditions
- Serial testing becomes even more important in patients with chronic comorbid conditions, such as the elderly and those with chronic kidney disease (CKD) or HF (Twerenbold, 2015).
- However, no hs-cTnI or T change criteria have perfectsensitivity and specificity for AMI, and thus, clinician judgment remains essential to confirm or refute the diagnosis.
- In general, with lower change criteria, sensitivity is higher and specificity is lower.
- Accelerated Diagnostic Pathways (ADPs)
- Although relevant studies provide a body of evidence supporting rapid rule-out ADPs, some have questioned their generalizability (Andruchow, 2018).
- Another limitation is that guidelines recommend using these ADPs in conjunction with all available clinical data, such as the patient’s ECG and historical data, but no specific guidance is given regarding how to incorporate these data into the algorithm.
- For ED-based testing:
- Current evidence supports excluding MI by use of a rapid algorithm (e.g., baseline plus 1-, 2-, or 3-h second sample) for hs-cTn alongside a validated risk score such as EDACS or HEART to identify patients suitable for early discharge.
- Compared with other rule-out strategies, theserialtesting approach for hs-cTn is less likely to miss MI among early presenters and uses a change value that is less susceptible to assay imprecision.
- Patients presenting > 3 h from symptom onset: 0/1h algorithm (particularly when paired with an ADP) can provide acceptable sensitivity and NPV.
- Patients presenting > 3 h from symptom onset with a low-risk presentation and hs-cTn < LoD: The single-test approach is reasonable if initial value is below the limit of detection.
- Patients presenting < 3 hours from symptom onset: Serial testing is still recommended in this group as evidence suggests that the single-test approach might have lower sensitivity in patients presenting < 3 h from symptom onset. The authors agree with the Universal Definition of MI and recommends use of sex-based 99th percentile URLs and has provided recommended values for “significant” changes of hs-cTn to be nested within the strategies used for AMI evaluation.
- For inpatient testing of acutechest pain or other signs of myocardial ischemia:
- It is reasonable to use a 3-h approach, with the recognition that the rapid response for inpatients might run the potential risk of earlyfalse negative testing and that at times, there are difficulties in seeing a delayed falling pattern.
- In contrast to the ED setting, a substantially greater percentage of patients in thehospital setting are likely to have significant, acute comorbid medical conditions that can cause myocardial injury.
Designing the Algorithm
When using any algorithm, some caveats apply:
- Algorithms should only be used in conjunction with all available clinical information, including detailed assessment of chest pain characteristics and ECG
- Algorithms may not encompass every clinical scenario, and requires some level of clinical judgement
- If you are uncertain of what to do, repeat a 2-hour level:
- If the initial level was within normal range and the delta goes up by > 50% = this is a significant change
- If the initial level was above normal range and the delta goes up by > 20% = this is a significant change
- In patients presenting very early (immediately after chest pain onset), the second cardiac troponin level should be obtained at 3 h, due to the time dependency of troponin release
- Late increases in cTn have been described in 1% of patients. So, serial cardiac troponin testing should be pursued if the clinical suspicion remains high or whenever the patient develops recurrent chest pain.
- High-sensitivity cardiac troponin assays also maintain high diagnostic accuracy in patients with renal dysfunction.
Can we use Accelerated Decision Pathways (ADPs) with adequate Sensitivity?
(note: these pathways only flies if ECG is non-ischemic)
- POSSIBLY at presentation (if patient is low risk)
- My take: Using a level below the LoD (< 2 ng/l for the Beckman Coulter hs-cTnI assay) has a sensitivity of 99.0% for MI with 99.5% NPV, although a more liberal cutoff set (~ 5 ng/l for the Beckman Coulter hs-cTnI assay) was shown to have only 94.5% sensitivity. If this is used, it should only be for levels < LoD for the assay used in patients who are truly low risk based on heart score and no ECG changes and pain at presentation that has been going on for > 2-3 hours to maintain sensitivity.
- PROBABLY at 1 hour
- MOST DEFINITELY at 2 hours
- DEFINITELY at 3 hours
- POSSIBLY at presentation (if patient is low risk)
How can we design an algorithm to get the most Sensitivity?
- Use sex specific cutoffs
- 0/1-hour algorithm is reasonable if it has been validated for your assay (Beckman Coulter assay has not)
- Combining 0/2-hour levels with HEART score < 3 will yield a high Sensitivity for all assays
- No Objective Test Rule (NOT rule) = 0, combined with a 0/2 hour ADP is 99-100% Sensitive at all times of presentation
- If the initial cTn is < URL, a 50% delta is reasonable. If the initial cTn is > URL, a 20% delta will offer higher sensitivity.
- Non-ischemic ECG is a must for all accelerated pathways
- If clinical question about being low risk…I suggest a 3 hour level be obtained
So, what is a reasonable (and safe) approach to ADP interpretation timeframes?
Example Evidence Based hs-cTn Pathway
This pathway includes data on Sensitivity and Specificity specific to the Beckman-Coulter Access hs-Tn I assay. However, it can be used as a reasonable starting point for interpretation of any of the FDA approved assays in the United States. It is meant to serve as a starting point when designing a pathway specific to your institution. However, you should always review the specific literature pertaining to your specific assay prior to implementation and design should include a multidisciplinary approach with input from Emergency Physicians, Cardiology, Hospitalists and Lab personnel.
- Monoclonal Antibodies for COVID-19 – An evidence based update - January 18, 2022
- COVID-19 Vaccines in Pregnancy - January 6, 2022
- COVID-19 Boosters….are they necessary? - October 8, 2021