Author + information
- ↵*Reprint requests and correspondence:
Dr. Allan S. Jaffe, Professor of Medicine, Mayo Clinic and Graduate Medical School, Rochester, Minnesota 55905, USA.
The European Society of Cardiology/American College of Cardiology (ESC/ACC) redefinition of acute myocardial infarction (MI), with its move to a troponin standard (1,2)and particularly the idea that even minor elevations of troponin with contemporary assays are indicative of risk, has been slow to be embraced. This is related in part to a lack of data, in part to a lack of understanding, and, in some instances, to difficulties with the assays themselves. This investigator has spoken out about several of these issues in other venues (3,4). A substantial amount of re-education of the cardiovascular population about how to appropriately use troponin is necessary and has, in my estimation, been lacking. However, large clinical trials in patients with acute coronary syndromes, despite the reluctance of clinical trial groups to accept the recommendations (3), have made it clear that any elevation of troponin is indicative of increased risk (5,6). This has led to a variety of studies defining new therapies and the efficacy of such therapies predicated on troponin markers (7,8).
In other areas, however, the idea that minor elevations are of importance has been slower to evolve. This is the case with perioperative MI. However, in this issue of the Journal,Landesberg et al. (9)from Israel break new ground in this important area. For the first time, the investigators examine with contemporary assays very low levels of troponin and demonstrate that there is a substantial increase in risk when even minor elevations occur in patients who have undergone vascular surgery. The Landesberg et al. (9)study builds on the initial studies utilizing troponin, which used very low levels.
The first study, published in the New England Journal of Medicine(10), used a value of 3.1 μg/ml, which at the time for that research assay was at the upper limit of the reference range for hospitalized controls. This value was extrapolated to a subsequent value of 0.6 μg/ml when the assay was commercialized (11), just about the 10% coefficient of variation (CV) level utilized in the study by Landesberg et al. (9). The studies by Lee and Goldman (12,13)utilized 0.2 μg/ml or 0.1 μg/ml depending upon the assay for cardiac troponin T (cTnT), which also, at that time, was the lowest level that could be measured accurately and presaged the results reported by Landesberg et al. (9). These initial data, however, were not pursued with these low cutoff values. Part of this may be related to the fact that many individuals did not understand that the 3.1 μg/ml value had changed to 0.6 μg/ml when the assay was commercialized. Most studies, as in the earlier one by Landesberg et al. (14), utilized this artificially high cutoff value (five times the upper limit of the reference range) or a value of 1.5 μg/ml, which is known as the receiver operator curve for MI cutoff and which promulgates a criteria close to that which would be the equivalent for elevations in creatine kinase-MB fraction (CK-MB). Not surprisingly, these studies demonstrated that patients with such marked elevations often met criteria for ischemia utilizing continuous electrocardiographic (ECG) monitoring and had an adverse short-term prognosis. Patients with substantial-sized MIs should be expected to have an adverse prognosis. Indeed, the present dataset corroborates this finding as well (see subsequent text).
Given the data that any increase in troponin has prognostic importance and the recommendations of the ESC and ACC committee that any value over the 99th percentile of the normal range should be considered abnormal (1,2), Landesberg and colleagues, when asked, tested what is known as the 10% CV level. This is a level close to the 99th percentile but a level where the variability of the assays are sufficiently good to preclude analytic false positives. Several of us have advocated the use of this value until the assays improve (3). These cutoff values are close to those used in studies done in patients with acute coronary syndromes.
The data presented by Landesberg et al. (9)are impressive. They first demonstrate that only 3.6% of patients present with symptoms that might alert the clinician to a possible MI, and only 5.6% would have met the prior World Health Organization criteria for acute MI. Despite these facts, patients with low-level elevations of troponin (107 of the 447 patients or 23.9%) had an adverse prognosis with a subsequent cardiovascular event over a mean follow-up period of 32.3 months. Whether one calls these individuals “patients with acute myocardial infarction” or simply “at risk,” the improvement in the ability to detect risk was increased five-fold. Second, and importantly, a relationship existed between the finding of ischemia with continuous ECG monitoring and elevations of troponin. Those with greater elevations tended to have more prolonged ischemia, but the relationship between troponin elevations of any degree and ECG ischemia was substantial, suggesting that the elevations found were indicative of ischemic heart disease in this population. Finally, Landesberg et al. showed that the larger the increment in troponin, the more likely an event was to occur, in keeping with the established concept that the larger the elevation, the worse the prognosis. Thus, the data suggest that, in this group of patients, elevations of troponin should be treated from the prognostic standpoint similar to elevations of troponin in patients who present with acute coronary syndromes. Even low levels of troponin elevations identify a cohort at markedly increased risk for subsequent events.
As impressive as the data are, several caveats are appropriate.
1. The risk of mortality even in the lowest group, which was 2.15-fold greater, was evaluated over a mean of 32.3 months. It appears from the curves that individuals had a substantial acute hazard, but these data are not provided. This is an important consideration because it would substantially impact on how and when one would evaluate and perhaps intervene in these patients. These outcome data should be amplified in additional studies.
2. A second important implication is that the assays utilized for cTnT and, in this instance, the first-generation Dade cTnI assay are reasonable contemporary assays. However, the Dade assay has been replaced by a more sensitive one, and several newer assays are more sensitive than this particular one. It must be appreciated that all assays do not measure the same type or amount of cTnI. Although it is likely that the principle of using the 10% CV as a cut-point for analysis for all assays will work reasonably well (3), one needs to exercise caution and rigor when attempting to extrapolate the significance of these data to risks faced by patients in your hospital, which probably utilizes a different assay for troponin.
3. Finally, this group was composed of 447 vascular surgery patients. This population of patients reflects a group with a very high a priori incidence of coronary artery disease (CAD). Thus, and not unexpectedly, elevations were correlated with ischemic-looking ST-T-wave changes. It is very likely that the etiology of most of these elevations were indicative of MI. Therefore, the most appropriate evaluation for most of these patients would be an evaluation for underlying CAD. However, this may not be the case in other surgical populations. For example, if this were a population of patients with orthopedic surgery, many of the more modest elevations might be due to pulmonary embolism (15). Patients with acute central nervous system disease can also have cardiac injury, and there is a strong experimental literature suggesting that CAD need not be the mechanism (16). Thus, extrapolation from vascular surgery patients to other patient groups must be done thoughtfully. It is likely that elevations associated with ischemic heart disease found in these other populations will have similar prognostic importance, but a larger percentage may be due to other disease entities requiring a different type of evaluation and imparting different prognostic implications.
As with any important research advance, this dataset raises new and important questions. Most of these patients likely had some preoperative evaluation. One wonders if, in retrospect, one could have predicted which of these patients was at risk. If so, perhaps one ought to intervene prior to surgery in those patients whose surgery was elective. It may also be, if one could identify these patients preoperatively, that even if one chooses not to intervene invasively, that strategies such as the use of beta-blockers might be appropriate in reducing the incidence of these small infarctions (17)and improving prognosis. Finally, the questions of when the events occurred in each of these groups and whether all patients should undergo immediate, somewhat delayed, or more markedly delayed evaluations to assist in their long-term management need to be delineated. Part of this delineation will include datasets where, in contrast to this study (9), there are data concerning the etiology of the long-term outcomes so that strategies specific to this circumstance can be crafted.
New standards always require new knowledge. These data, which break new ground, help the field progress. If those who work in other areas would take the same approach of questioning and evaluating the ESC/ACC criteria in large enough datasets so that longer-term prognosis can be assessed, equivalent progress might be made in clarifying the importance of similar troponin elevations in other areas as well. Nonetheless, fields progress slowly. This is a large baby step in the right direction.
☆ Dr. Jaffe is a consultant and receives research support from Dade-Behring, Roche, and Beckman-Coulter. He is also a consultant to Ortho Diagnostics. All these companies make troponin assays.
↵* Editorials published in the Journal of the American College of Cardiologyreflect the views of the authors and do not necessarily represent the views of JACCor the American College of Cardiology.
- American College of Cardiology Foundation
- ↵Myocardial infarction redefined—a consensus document of the Joint European Society of Cardiology/American College of Cardiology Committee for the redefinition of myocardial infarction. J Am Coll Cardiol 2000;36:959–69
- Jaffe A.S.,
- Ravkilde J.,
- Roberts R.,
- et al.
- Jaffe AS. Caveat emptor. Am J Med 2003;115:241–4
- ↵James S, Armstrong P, Califf R, et al. A lower cut-off level of troponin-T is more appropriate for prediction of low risk in acute coronary syndrome patients—a prospective verification in a large cohort of patients from the GUSTO-IV trial. Am J Med 2003;115:178–84
- TACTICS-TIMI 18 Investigators,
- Morrow D.A.,
- Cannon C.P.,
- Rifai N.,
- et al.
- Morrow D.A.,
- Antman E.M.,
- Tanasijevic M.,
- et al.
- ↵Landesberg G, Shatz V, Akopnik I, et al. Association of cardiac troponin, CK-MB, and postoperative myocardial ischemia with long-term survival after major vascular surgery. J Am Coll Cardiol 2003;42:1547–54
- Jaffe A.S.,
- Landt Y.,
- Parvin C.A.,
- Abendschein D.R.,
- Geltman E.M.,
- Ladenson J.H.
- Lopez-Jimenez F.,
- Goldman L.,
- Sacks D.B.,
- et al.
- Landesberg G.,
- Mosseri M.,
- Zahger D.,
- et al.
- Kostantinides S.,
- Geibel A.,
- Olshewski M.,
- et al.