Category: Cardiology

Highly Sensitive Troponins – False Positive Bonanza

The “highly sensitive” troponin has received a great deal of publicity, hyped ad nauseum, see: “Simple test could help rule out heart attacks in the ER.”

But, as sensitivity increases – invariably, specificity decreases.  However, that is not the fault of the test – it is a failure of clinicians to ask the correct question of the test.  When asking “does this patient have an acute myocardial infarction?”(most commonly Type 1 MI in the ED), our training and education has been outpaced by assay technology – the test no longer provides a dichotomous “yes” or “no”.

This publication provides a lovely window into precisely the added value of the hsTnI compared with conventional TnI, both assays by Abbott Laboratories.  In this study, the authors simultaneously drew research samples of blood any time a cTnI was ordered.  The sample was frozen, and then analyzed at least 1 month following presentation.  Authors performed hospital records review, telephone follow-up, and vital records search to evaluate adverse events in patients with hsTnI or cTnI elevation.

Overall, they enrolled 808 patients, 40 of which received an adjucated diagnosis of “acute coronary syndrome” – 26 with AMI and 14 with unstable angina.  61 patients had acute heart failure, 7 had volume overload, 7 had pulmonary emboli, and 41 had other non-ACS cardiac diagnoses.

All told, there were 105 elevated cTnI samples – and 164 elevated hsTnI samples.  This means, essentially – in the acute setting, asking our question of interest – there were 50% greater false positives associated with hsTnI.  No patients would have been reclassified as nSTEMI based on the hsTnI result.  The authors sum this up nicely in their discussion:

“The preponderance of novel elevations (roughly 10% in this study) will be observed mainly in subjects with non-ACS conditions.”

The authors go on to note the value in detecting these novel or detectable troponin levels – essentially, non-ACS, subclinical disease – with a much poorer long-term prognosis.  This is almost certainly the case, although it will require further investigation to reliably demonstrate cost-effective management strategies based on these results.

“Troponin Elevations Only Detected With a High-sensitivity Assay: Clinical Correlations and Prognostic Significance”
http://www.ncbi.nlm.nih.gov/pubmed/25112512

Addendum:  As Stephen Smith points out, it may be possible to use the greater precision of hsTnI at the low end of the assay to more accurately adjudicate some MI.  Great insight!

Should the 48-hour Cardioversion Window Be Revised?

It has become generally accepted practice to treat new-onset atrial fibrillation and atrial flutter with electrical cardioversion in the acute setting – provided the known onset of atrial fibrillation is less than 48 hours.  Beyond that, caution tends to be advised – whether through use of transesophageal echocardiography to rule out left atrial thrombus, or through pre- and post-procedural anticoagulation.

However, this data from a research letter in JAMA suggests – possibly we ought to be even more cautious regarding time-of-onset.

This is a re-analysis of FinCV, a 7 year trial registry of cardioversion for atrial fibrillation from Finland.  The study cohort is comprised of 2,481 patients undergoing 5,116 electrical cardioversions, all without peri-procedural anticoagulation for symptom onset <48 hours.  Outcomes were gathered from vital records review, evaluating for cerebrovascular thrombotic complications within 30 days.

Of these patients undergoing cardioversion, there were 38 definite thrombotic complications.  30 of these 38 occurred in patients whose symptom onset was >12 hours.  There were few apparent pro-thrombotic differences between groups, and thus, the authors very reasonably conclude – we should be cautious regarding cardioversion after 12 hours.  Other predisposing factors in their multivariate analysis include female sex, heart failure, and diabetes – but increasing length of time showed the strongest association.

The 12-48 hour window in this study still only represented a 1.1% risk for 30-day thromboembolism, compared to the ~2% risk after 48 hours.  However, it still exceeds the ~0.3% risk of thromboembolism with peri-procedural anticoagulation.  There are other risks associated with anticoagulation, but it is reasonable to suggest the management strategy is no longer as clear-cut around 48 hours.

“Time to Cardioversion for Acute Atrial Fibrillation and Thromboembolic Complications”
http://www.ncbi.nlm.nih.gov/pubmed/25117135

The Glucose-Insulin-Potassium in ACS (IMMEDIATE) Revisited

There are a few folks persistently enamored with the use of glucose-insulin-potassium cocktails in the setting of acute coronary syndrome.  Two years ago, I reported on the initial results of the IMMEDIATE trial, published in JAMA.  At the time, I boggled that such a small, underpowered study – and one that changed their enrollment target and primary endpoint during the trial – would ultimately be accepted into JAMA.  It was also subsequently picked up and promoted by ACEPNow as a “promising therapy” that “slashes death risk” (Tweet still visible, site content since deleted).

As I noted last time, the authors reported 30-day outcomes.  Now, they’re back, as promised, with 1-year outcomes – hoping to build on the “trends” previously observed.

Did it work?  Did a longer follow-up period help?  I’ll let the authors answer in their own words:

“The 1-year outcomes (Table 2) generally demonstrate point estimates favoring GIK for individual events and for composites, but CIs overlap 1.0 in most analyses.”

Which is to say, no.  Trends stayed trends, for the most part.  And, considering the neutral pre-study evidence from such studies as CREATE-ECLA, which enrolled 20,201 patients, even such hopeful interpretations of these data should not inspire a change in practice.

It is still reasonable, however, as these authors suggest, to continue studying this treatment.  There are small differences in important patient-oriented outcomes (10.9% vs. 13.0% in 1-year mortality) favoring the intervention.  These effects were even more pronounced in the subgroup of patients for whom ST-elevation was present on the pre-hospital ECG.  Considering there’s not much room left in acute medical management of STEMI offering a survival advantage – if there is, in fact, a 4.3% absolute 1-year survival attributable to GIK, it is worth continuing to investigate and tailor trials to the subgroup most likely to benefit.

“One-Year Outcomes of Out-of-Hospital Administration of Intravenous Glucose, Insulin, and Potassium (GIK) in Patients With Suspected Acute Coronary Syndromes”
http://www.ncbi.nlm.nih.gov/pubmed/24792735

The Return of Metoprolol – for Anterior STEMI

Beta-blockade early in the course of myocardial infarction was once fashionable – until COMMIT demonstrated an excess of early cardiogenic shock detracting from subsequent late, favorable effects.  This led to beta-blockade initiation being deferred until after hemodynamic stability established.

This study, METOCARD-CNIC, is a trial of early intravenous metoprolol prior to primary PCI in patients with anterior STEMI.  270 patients were randomized over two years to receive IV metoprolol pre-reperfusion versus standard initiation following PCI.  These two publications describe a surrogate outcome based on infarct size seen in follow-up MRI, and patient-oriented outcomes of 2-year MACE and heart failure progression.  And, overall, it’s a good thing – infarct size at 1 week was reduced from 32.0 to 25.6 grams, and long-term MACE at median 2-year follow up was reduced from 18.3% to 10.8%.

However, long-term MRI follow-up at twelve months showed infarcted myocardium measured at 15.7 grams in the intervention group versus 18.2 grams in the control – no longer statistically significant.  And, the patient-oriented outcome of MACE is a combined endpoint of death, heart failure admission, reinfarction, and malignant arrhythmias – with most of the separation in groups coming from heart failure admissions and malignant arrhythmias, as opposed to hard endpoints.

But, at the minimum, this is worth continuing to investigate.  There are likely patients, such as this anterior STEMI cohort, with Killip Class II or lower at presentation, that reasonably have a greater chance of benefit than harm from early metoprolol.  This is also quite small study – but taken in the context of the prior evidence, an argument could be made to cautiously re-introduce this treatment strategy, ideally as part of prospective investigation.

“Effect of Early Metoprolol on Infarct Size in ST-Segment−Elevation Myocardial Infarction Patients Undergoing Primary Percutaneous Coronary Intervention”
http://www.ncbi.nlm.nih.gov/pubmed/24002794

“Long-Term Benefit of Early Pre-Reperfusion Metoprolol Administration in Patients With Acute Myocardial Infarction”
http://www.ncbi.nlm.nih.gov/pubmed/24694530

Guideline Recommendations Are Written in Dry Erase Marker, Not Stone.

A guest post by Anand Swaminathan (@EMSwami) of EM Lyceum and Essentials of EM fame.

Medicine is filled with guidelines, professional recommendations and expert consensus statements. These documents guide clinical practice. In Emergency Medicine, we often rely on non-EM specialty guidelines. For instance, we often state that a patient in whom you consider ACS should get evocative testing (i.e. stress test) within 72 hours of presentation according to the American College of Cardiology/American Heart Association (ACC/AHA) guidelines. As more guidelines and subsequent revisions are released a number of questions arise. Should we adopt the guidelines immediately? If so, which pieces are ready for immediate incorporation into clinical care? At the heart of these questions is the strength and durability of the recommendations.
This article is unique in the question it asked: what is the durability of class I recommendations from the ACC/AHA? The looked at 11 guidelines published between 1998 and 2007 along with revisions to these guidelines from 2006 to 2013. What they found was surprising. Out of 619 original class 1 recommendations, about 80% were retained in subsequent revisions. About 9% were downgraded or reversed and about 11% were omitted.  Not surprisingly, recommendations with multiple randomized studies backing them up tended to stick around (90.5%) but those recommendations supported by opinion only did not (73.7%).
What can we take away from this? First, we shouldn’t adopt recommendations (even level 1) that don’t have good evidence backing them up. Secondly, guidelines should be updated frequently (these authors suggest every 3 to 5 years) to incorporate new evidence that may up or downgrade recommendations. Guideline adherence shouldn’t be used as a performance measure since the recommendations are anything but written in stone. Lastly, this is a further call for our specialty to take the reigns and start writing our own, high-quality guidelines from which we can base clinical practice.
“Durability of Class 1 American College of Cardiology/American Heart Association Clinical Practice Guideline Recommendations”

Conclusively Settling Azithromycin’s Cardiac Toxicity Forever

We’ve been obsessed with azithromycin’s cardiovascular effects for quite some time – with some studies showing an increase in events, and other studies using azithromycin as the active agent to decrease coronary events.  Why is it such an issue?  Mostly because azithromycin has become the nonsensical cure-all of eager-to-please primary care physicians for self-limited or viral conditions, let alone the mainstay of treatment for pneumonia.

This latest study comes from a retrospective cohort of Veterans Affairs patients admitted and receiving IDSA guideline-compliant treatment for community-acquired pneumonia.  These authors compared a cohort of patients receiving ß-lactam + azithromycin with any other guideline-compliant therapy, typically fluoroquinolone monotherapy.  They created two propensity-matched cohorts based on known confounders, resulting in comparison groups of 31,863 patients each, with a treatment period spanning 2001 to 2012.

Of these, 1,948 patients exposed to azithromycin had a myocardial infarction recorded within 90 days, compared with 1,523 in the non-azithromycin cohort, for an OR of 1.11 (95% CI 1.03-1.20).  No other cardiovascular disease was increased, and no specific subgroup conferred a higher or lower risk of MI after azithromycin use.  Most of the difference in MI occurred within 30 days of exposure.

However, interestingly, the overall 90-day mortality was 6,582 in the azithromycin cohort, compared with 8,152 in the non-azithromycin cohort, for an OR of 0.73 (95% CI 0.70-0.76).  And, the authors happily run with this mortality advantage – concluding “azithromycin compared with other antibiotics was associated with a lower risk of 90-day mortality (number needed to treat of 21)”.  But, how does a short course of a macrolide antibiotic generate such a profound survival curve that progressively widens months after exposure?  The authors do not provide data on causes of death, nor do they provide much explanation for the observed survival advantage.  Either short-course azithromycin provides a powerful, anti-inflammatory effect with long-term advantage – as implied by the authors – or there’s a problem with the data and the matching.

My vote is for problems with the data.  Propensity matching is only as good as the prognostic importance of variables included in the algorithm, and suffers tremendously when performed on retrospective data not gathered specifically to support such analyses.  Sadly, this study is probably best served to be assigned to the scrap heap of unreliable retrospective observations.

“Association of Azithromycin With Mortality and Cardiovascular Events Among Older Patients Hospitalized With Pneumonia”

The Unusable Manchester Chest Pain Instrument

With probably underpowered derivation and validation, a model that seems to overfit the data, and incorporating an impractical and questionable cardiac biomarker – despite a lovely continuous predictive function – this instrument is doomed in its current form.

This is the Manchester Acute Coronary Syndromes (MACS), a prospectively derived and validated risk-stratification instrument.  These authors identify an 8 variable decision instrument based on 698 patients at Manchester Infirmary – including hsTnT, heart-type fatty-acid binding protein, ECG changes, diaphoresis, vomiting, radiation to right shoulder, worsening angina, and hypotension – and then validate it on 463 patients from Stepping Hill Hospital.  In the validation, 27.0% of patients were ultimately classified as “very low risk” with 98% sensitivity (95% CI 93.0% to 99.8%) for 30-day MACE, and the authors feel this tool could reduce unnecessary admissions.

My favorite feature from this study is the derivation of a continuous function for prediction of 30-day outcomes.  The authors state an AUC of 0.92 for the function predicting MACE, which suggests potential as a useful tool for discussing individualized risks with patients.  Rather than simply dichotomize a “very low risk” cohort, the predictive function could help aid shared decision-making conversations with patients.

However, the utilization of fatty-acid binding protein is questionable.  These same authors presented work favoring H-FABP with an AUC for diagnosis of AMI of 0.86, but compared it against a troponin assay with an AUC of 0.70.  A response to that same article notes the authors probably made inappropriate comparisons, and modern conventional troponin assays and/or high-sensitivity troponin assays have AUCs >0.90.  It’s not clear what, or how much, additional value this biomarker adds to this study – and its inclusion essentially obviates the generalizability of the tool.  No rapid, automated H-FABP assay is available suitable for use in an ED context.  It is also unfortunate the corresponding author declares conflict-of-interest with the manufacturers of the assays used.

Interestingly, as well, the authors focus only on the “very low risk” group when an odd thing happens in their validation population – the “low risk” group actually had fewer MACE than the “very low risk” group (1.2% vs. 1.6%).  This is a substantial reversal of the derivation population (5.8% vs. 0.4%), suggesting the attempted validation reveals their model may be overfitting the data.  The authors state the second site validation is a strength regarding combating overfitting, but do not mention this inconsistency in the outcomes.

And, finally, I’m not entirely certain what question this study was designed to answer.  The focus on a “very low risk” cohort in the discussion doesn’t entirely match the study design – it seems other studies focused on outcomes in low risk chest pain have specifically excluded patients whose presentation is clearly AMI on initial presentation.  The inclusion of the entire spectrum of disease, while valuable for their general model, dilutes the strength of their “very low risk” conclusions, as evidenced by wide confidence intervals around the sensitivity for MACE.

As the authors note in their discussion, additional work needs to be done to compare their model with other risk-stratification tools.  And, for anyone to use this tool other than the authors, the H-FABP needs to be dropped.  At the least, however, it fits in nicely with another recent critique – that many “low risk” and “very low risk” patients do not require observation and immediate provocative testing, where the false-positives and resource expenditures are simply preposterous.

“The Manchester Acute Coronary Syndromes (MACS) decision rule for suspected cardiac chest pain: derivation and external validation”
http://heart.bmj.com/content/early/2014/04/29/heartjnl-2014-305564

Varying Degrees of Diagnostic Discomfort

A guest post by Rory Spiegel (@EMNerd_) who blogs on nihilism and the art of doing nothing at emnerd.com.

I think we can all agree, there is a fair degree of variability in our (dis)comfort with uncertainty.  As illustrated by an article published in this month’s JAMA Internal Medicine, our discomfort in the unknown and its effect on practice variability has never been more obvious than in the management of acute coronary syndrome (ACS). Kyan et al examined just how this variation affected patient care on a hospital based level. Using a prospectively gathered national database of 2,700 hospitals, the authors extracted those patients in whom cardiac ischemia was considered, and examined the number of patients from each hospital surveyed who received non-invasive cardiac imaging. The authors divided the hospitals into quartiles based on the proportion of their patients who underwent non-invasive testing and examined how this variation in testing affected downstream care.

Of the 224 hospitals included in the data, variation of number of patients who underwent non-invasive cardiac imaging ranged from an impressively nihilistic 0.2% to extraordinarily high 55.7%.  The majority of these tests consisted of myocardial perfusion studies (80.4%) and to a lesser extent stress echocardiography (16.6%). Despite its recent hype CTCA was utilized sparingly as an imaging modality (1.2%). Not surprisingly the hospitals with higher rates of non-invasive testing had higher number of hospital admissions, angiographic studies and revascularization procedures. The increased testing and subsequent interventions failed to demonstrate a noticeable effect on patient outcomes. Both the hospitals in the lowest quartile and those in the highest quartile of non-invasive cardiac testing had equivalent readmission rates for acute myocardial infarction within the next 2-months (0.3%).



Unfortunately this study does not tell us which patients if any, should receive further non-invasive testing following a negative ED work up. What is becoming increasingly clear is if you take a cohort of patients, the vast majority of whom, are not experiencing the pathology in question, no amount of further testing will lead to improved negative predictive values. The only thing gained will be the iatrogenic harms caused by the downstream interventions this increased testing will invariably cause. A high price to pay for a salve for our discomfort…

“Hospital Variation in the Use of Noninvasive Cardiac Imaging and Its Association With Downstream Testing, Interventions, and Outcomes” 
http://www.ncbi.nlm.nih.gov/pubmed/24515551

Pre-Hospital Furosemide – No, No, Also No

In Ottawa, pre-hospital care includes paramedics authorized to treat acute cardiogenic pulmonary edema in the setting of respiratory distress.  Their treatment, sensibly, includes nitroglycerin.  It also, insensibly, includes furosemide.

Decompensated heart failure, resulting in pulmonary edema and dyspnea, is indeed a sort of fluid overload.  However, these patients frequently are not hypervolemic – they may be euvolemic or even hypovolemic, with other underlying etiologies for decompensation than fluid retention.  This pushes the concept of a strategy for the treatment of acute cardiogenic pulmonary edema with furosemide even further down the nonsense pathway.  Yet, there it is.

This study, a retrospective review of presentations with pre-hospital furosemide administration and hospital diagnoses of acute decompensated heart failure, demonstrates essentially nothing.  The primary outcome was designed to detect serious adverse outcomes associated with furosemide administration, but their comparison groups – furosemide given to heart failure, furosemide given to misdiagnosed heart failure, and furosemide not given to heart failure – are clinically heterogeneous and require probably meaningless adjusted comparisons.  The authors find no significant difference, but this is simply a matter of sample size and study design – a treatment given to a group with no chance of benefit obviously suffers only harms.

Most damning, however, is the utter failure of pre-hospital providers to correctly diagnose heart failure.  Of the 272 cases of heart failure diagnosed on arrival to the ED, pre-hospital providers made the diagnosis in only 110 instances.  Then, pre-hospital providers incorrectly diagnosed an additional 58 cases with heart failure and administered furosemide – when the patient was diagnosed with pneumonia, COPD, or another alternative.

Just say no.

“Prehospital use of furosemide for the treatment of heart failure”

The Troponin to End All Troponins

Yet again, the internet has exploded with magical thinking: “Simple test could help rule out heart attacks in the ER: study”

What is this one “weird” trick that builds muscle, saves electricity, gives you a flat belly, and detects heart attacks in the Emergency Department?

It’s just another high-sensitivity troponin publication.

This is a two-year retrospective evaluation of patients presenting to two Emergency Departments in Sweden, culling the electronic health record for patients aged greater than 25 who were evaluated for chest pain and had at least one hs-cTnT level measured.  These patients were then followed through the central Swedish Health Registry for subsequent hospital admissions or death for 30, 180, and 365 days following their Emergency Department Visit.

Of this cohort, 8,883 had an initial hs-TnT <5 ng/L; within 30 days, 15 of these patients received the diagnosis of MI and two patients died.  Thus, a negative initial hs-TnT was associated with a 0.17% absolute risk for MI and a 0.023% risk of death from cardiovascular causes within 30 days.  Therefore, this test is magic.

Except, it isn’t.  We’ve known for almost two decades that negative biomarkers confer an excellent 30-day prognosis.  This is simply a more sensitive version of those prior assays, with the potential to pick up troponin elevations earlier in the time course of cardiac ischemia.  Earlier detection of cardiac ischemia then potentially enables a one-set rule-out, rather than a two-set or three-set traditional evaluation.

However, as is the natural order of things, when the balance of a test shifts to assign greater significance to a negative test result, the significance of a positive test results declines.  Patients admitted to the hospital with hs-TnT levels >14 ng/L received a diagnosis of MI only 30% of the time.  Their data tables are insufficient to estimate the specificity of the test at the cut-offs provided in the paper, but, clearly it is poor.

The authors promote this test as a tool to discharge greater numbers of patients from the Emergency Department.  However, any potential performance improvement will depend on the baseline admit rate at your institution.  Then, ultimately, we’ll need a more sophisticated approach to interpreting this test – moving beyond the dichotomous interpretation of “positive” and “negative” biomarkers, and describe the detectable levels on a continuum in the context of the concomitant disease processes – including, but not limited to, acute coronary syndrome.

“Undetectable High Sensitivity Cardiac Troponin T Level in the Emergency Department and Risk of Myocardial Infarction”
https://content.onlinejacc.org/article.aspx?articleID=1854323