–Cardiologists fear the new high sensitivity troponin test will lead to lots of unnecessary testing.
Along with many other cardiologists, Sek Kathiresan (Massachusetts General Hospital and Broad Institute) is worried that a new and improved diagnostic test will have disastrous unintended effects.
Kathiresan expressed his concern on Twitter yesterday in response to my summary of a new paper in Annals. The study showed that the newly approved high-sensitivity cardiac troponin T (hs-TnT or hs-cTnT) blood tests may allow the rapid dismissal from the Emergency Department of 30% of chest pain patients. But, in addition to this clear benefit, Kathiresan is worried that the test will have other, far less desirable consequences. Kathiresan tweeted:
- Will we see a large increase in additional testing and consults for ‘leaklets’ (i.e. small increases in hs-troponin)?
- Will we resist the urge to perform coronary angiograms on all these leaklets?
These questions sparked a wide-ranging discussion among cardiologists and other physicians on Twitter (and offline as well).
Ethan Weiss (UCSF) seconded Kathiresan’s concerns:
- Almost certainly. Negative predictive power will increase, but false positives will as well… hope the hospital can make more beds
Weiss went on to predict a huge increase in CT angiograms. Maulik Majmudar (Massachusetts General Hospital) had a similar prediction:
- I bet we will see a lot more cardiology consults and follow on stress testing. Will need study on cost-effectiveness once intro in US
Scott Wright pointed out what should be obvious, but often is not, that reflex testing is bad medicine:
- When did an elevated troponin justify a cath? ACS justifies a cath if the patient is intermediate or high risk and the patient concurs
- My concern w/ the “troponin advocates” is that they fail to ask physicians to be clinicians first and foremost. Treat the patient not the lab
- MI is a clinical diagnosis, not a biochemical one. Treat the patient not the lab value.
John Pickering (Christchurch, New Zealand) is the first author of the Annals paper. He observed that the new test, which has been commercially available outside the US for several years, has not resulted in these feared consequences so far:
- interesting discussion – NZ, AU, & I think European experience of introduction of hsTns has not been huge increase in MIs diagnosed.
- We CONSERVATIVELY estimate 1.7 million false positives (likely far far far greater) using hs-#troponin
Weiss went on to speculate about additional related problems that may be caused by the new tests. Emphasizing the likely absence of specificity of the new test, and the absurdity of the situation, Weiss asked:
- Serious question: If you had gone out to measure hsTnI in all of the people who ran the Boston Marathon today, what would you have seen?
- How will we deal with people who come in after a big workout, or huge emotional stress? I do not think it will be easy…
Weiss then raised yet another provocative question:
- Who will be the first person to propose measuring hsTroponin in people with severe constipation?
Sanjay Kaul is the co-author, with John Brush and Harlan Krumholz, of a sober JACC paper exploring the important issues raised by the new troponin test. Kaul is not an active Twitter participant, but he sent the following comment about the discussion:
The high-sensitivity assay is clearly a step forward for the ED physicians whose primary focus is to rule-out MI and triage patients out of the hospital. The false-negative rate of 0.5% is quite reassuring for the emergency physicians because they can confidently discharge patients from the ED without the fear of medicolegal liability (the generally acceptable threshold of medicolegal risk being <1% false-negative rate). But this low false-negative rate only applies to low risk patients (defined as those without ECG changes) who have a low pretest probability of having acute coronary syndromes (ACS). However, the high-sensitivity of the assay comes at the expense of lower specificity which could potentially be a liability for cardiology consultants whose major focus is to rule-in MI and appropriately triage high-risk patients to more aggressive treatment interventions. It is important to keep in mind that high values do not establish a diagnosis of ACS with confidence if the pretest probability is low (as in the setting of renal failure, sepsis, heart failure, etc.). Conversely, even very low values of troponin do not reliably exclude the diagnosis of ACS if the pretest probability is high. Although new high-sensitivity or more accurately the ‘next generation’ troponin assays offer potential advantages over the conventional assays, the major problem with them, as with any other laboratory test, is often an inappropriate request and improper interpretation of the results, not the marker itself. Just as a tool is only as good as its operator, a diagnostic test can be only as good as its interpretation. Expecting the test to provide all the answers without including the proper clinical context can lead to erroneous diagnoses.
John Brush, first author of the JACC paper, sent the following comment:
Much of the push back regarding the new hs troponin assay comes from a misunderstanding of the meaning of “high sensitivity.” In this case, this term means analytical sensitivity, not clinical sensitivity. Clinicians are used to thinking about sensitivity as a trade-off with specificity and mistakenly think that this assay lowers the cut point, which would increase the rate of false positive test results. This is not the case, as discussed in our recent JACC review on troponin. The cut point (or positivity criterion) for the newer assay is about the same as for the existing assay for troponin T (with more precision, 0.014 ng/ml vs. 0.01 ng/ml). With the newer assay, the clinician must get used to having troponin test results which are now measurable and are in a “normal range.” This is no different than most laboratory tests. Formerly, any detectable troponin was abnormal because the prior generations of the assay were not as analytically sensitive and precise. Any detectable troponin using the old assays was abnormal. The newer assay, if used properly, will be an advance for clinical medicine.