High sensitivity cardiac troponin I (hs-cTnI) assays routinely used to help exclude or diagnose acute myocardial infarction (MI) can be misleading, researchers caution in a paper online in The BMJ March 13, 2019.

In a study of 20,000 consecutive patients at University Hospital Southampton NHS Foundation Trust, about one in 20 had a hs-cTnI concentration above the recommended upper limit of normal (ULN), despite most of them having no clinical suspicion of acute MI.

“This paper should raise questions about the routine use of this test in hospitals,” study chief Dr. Nick Curzen, a consultant cardiologist at University Hospital Southampton, told Reuters Health by email. “This study shows the need for medical staff to interpret troponin levels carefully in order to avoid misdiagnosis of a heart attack and inappropriate treatment.”

Manufacturers of troponin assays provide a recommended “99th centile” based on values from healthy individuals. This level is often used as the ULN when applied to patients in hospitals.

A troponin value above the 99th centile is considered abnormal and indicative of acute MI in appropriate circumstances. But the distribution of troponin levels across the whole hospital population (inpatients, outpatients, patients undergoing surgery, in intensive care, etc.) is largely unknown.

Dr. Curzen’s team measured hs-cTnI in 20,000 inpatients and outpatients undergoing blood tests for any reason at their hospital. Their average age was 61 and 53 percent were women.

The 99th centile of troponin for the whole study population was 296 ng/L, compared with the manufacturer’s recommended ULN of 40 ng/L. Troponin levels were above 40 ng/L in 1,080 individuals, or 5.4 percent, of the total population. After excluding 122 individuals diagnosed with acute MI and 1,707 for whom hs-cTnI was requested for clinical reasons, the 99th centile was 189 ng/L for the remaining 18,171 individuals.

Overall, 39 percent of critical care patients, 14 percent of medical inpatients, and 6 percent of emergency department patients had a troponin concentration greater than the recommended ULN.

This study suggests that the ULN, “which is derived from a group of relatively healthy people, may not be suitable for a hospital population in general,” Dr. Curzen told Reuters Health. “This is of significant concern because if it is measured in patients without a classic heart attack presentation, the level may appear raised and such patients may end up being incorrectly diagnosed as having had a heart attack. They may then receive inappropriate treatment which could be potentially harmful outside of this context.”