- Email: [email protected]
Implementation of a highly sensitive cardiac troponin I assay: Test volumes, positivity rates and interpretation of results
Clinica Chimica Acta 395 (2008) 57–61
Contents lists available at ScienceDirect
Clinica Chimica Acta j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / c l i n c h i m
Implementation of a highly sensitive cardiac troponin I assay: Test volumes, positivity rates and interpretation of results Stacy E.F. Melanson, Michael J. Conrad, Nima Mosammaparast, Petr Jarolim ⁎ Department of Pathology, Clinical Laboratories Division, Brigham and Women's Hospital, Harvard Medical School, Boston, MA 02115, United States
A R T I C L E
I N F O
Article history: Received 31 December 2007 Received in revised form 30 April 2008 Accepted 1 May 2008 Available online 15 May 2008 Keywords: Cardiac troponin Implementation Sensitive assay Acute coronary syndrome Myocardial injury Renal dysfunction
A B S T R A C T Background: While more sensitive cardiac troponin (cTn) assays may overcome current limitations in diagnosing acute coronary syndrome (ACS), clinicians and laboratorians are concerned about the clinical impact of higher rates of cTn positivity. Methods: We collected statistics on test volume and positivity rates before and after the implementation of a more sensitive assay. We divided patients into 6 groups based on their cardiac troponin I (cTnI) results and utilized additional laboratory test results to determine the impact of the new assay. In addition, we assessed the clinical significance of low-positive cTnI results (0.05–0.10 µg/l). Results: Lowering the diagnostic cutoff from 0.10 to 0.04 µg/l increased the number of positive test results by 44.2% hospital-wide and 114.4% in the emergency department without significantly changing test volume. In some patients, low-positive results were part of the typical rise and fall of cTnI consistent with myocardial damage. Diagnosis was more challenging in patients with consistently low-positive results. Patients in this group were significantly more likely to have impaired renal function than those with cTnI elevations N 0.10 µg/l. Conclusions: More sensitive cTn assays will increase the number of positive results and allow for earlier detection of cardiac injury while also detecting non-ACS related pathologies. © 2008 Elsevier B.V. All rights reserved.
1. Introduction The ability to rapidly determine the presence of myocardial ischemia is crucial for patients presenting with chest pain or other signs and symptoms suggestive of acute coronary syndrome (ACS). Cardiac troponin has become the gold standard among biochemical markers used for both the diagnosis and prognosis of such patients [1–3]. However, until recently, cardiac troponin testing has had various limitations including suboptimal analytical sensitivities and low-end imprecision. As a result of assay performance and the kinetics of cardiac troponin release from damaged cardiac myocytes, patients presenting hours after the onset of symptoms may still have undetectable cardiac troponin concentrations. Typically, patients will have measurable cardiac troponin within 4 to 6 h of symptoms [4], but the opportunity for rapid triage and early intervention may be lost by that time. Thus, improvements in the performance of cardiac troponin assays can
Abbreviations: Acute coronary syndrome, (ACS); American College of Cardiology, (ACC); European Society of Cardiology, (ESC); estimated glomerular filtration rate, (eGFR); Modification of Diet in Renal Disease, (MDRD); chronic kidney disease, (CKD); congestive heart failure, (CHF); laboratory information system, (LIS). ⁎ Corresponding author. Brigham and Women's Hospital, 75 Francis Street, Amory 2, Boston, MA 02115, United States. Tel.: +1 617 732 6672; fax: +1 617 731 4872. E-mail address: [email protected] (P. Jarolim). 0009-8981/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.cca.2008.05.007
potentially overcome the current limitations and benefit patients presenting early in the course of myocardial ischemia. The American College of Cardiology (ACC) and the European Society of Cardiology (ESC) both recommend using a diagnostic cutpoint at the 99th percentile of the reference population with a coefficient of variation (CV) of ≤10% for the diagnosis of ACS [1–3,5–8]. Some of the currently available cardiac troponin assays do not meet these requirements [9]. Siemens Healthcare Diagnostics recently replaced their older cTnI assay with an improved assay, TnI-Ultra, which offers both enhanced sensitivity and precision. We have previously compared the characteristics of cTnI to TnI-Ultra and confirmed that TnI-Ultra offers improved analytical and clinical performance, consistent with the ACC/ESC guidelines [10]. TnI-Ultra achieved a 10% CV at the manufacturer's 99th percentile resulting in a diagnostic decision point of 0.04 µg/l. Because of the improved analytical performance of the TnI-Ultra assay and the decision by the manufacturer to discontinue their cTnI assay, we implemented TnI-Ultra testing in our clinical laboratory. We expected and previously confirmed that the enhanced performance characteristics of the TnI-Ultra assay would have anticipated benefits such as the earlier diagnosis of ACS in a subset of patients [10]. At the same time, both the clinicians and laboratorians were concerned about the clinical utility of cardiac troponin assays with increased sensitivity in other patients. It was clear that the number of positive cardiac troponin I results would increase due to the
58
S.E.F. Melanson et al. / Clinica Chimica Acta 395 (2008) 57–61
new low-positive range of 0.05–0.10 µg/l. However, we did not know whether or not this would translate into an increase in the volume and cost of cardiac troponin I testing for the clinical laboratory or how these low-positive test results would affect diagnostic decisions of our clinicians and subsequent treatment of our patients. Data on the clinical significance of low-positive cardiac troponins is limited. It is known that cardiac troponin can be increased in conditions other than ACS [4,11] such as renal disease [12–14] and congestive heart failure [4,15]. Sommerer et al. [14] have shown that plasma concentrations of cTnT are increased in approximately 40% of asymptomatic chronic haemodialysis patients. Latini et al. [15] reported detectable cTnT in patients with chronic heart failure. Several groups have also studied the predictive value of small cTn elevations for adverse long-term cardiovascular outcomes [16–21]. Using a sensitive cTnI assay, Eggers et al. [17] found that increased cTnI concentrations N0.01 μg/l after an episode of non-ST-elevation ACS predicted mortality during long-term follow-up. However, how these findings will affect the interpretation of low-positive cTnI results and the utility of cardiac troponin assays with increased sensitivity for the diagnosis of ACS is not clear. To clarify the impact of implementation of a more sensitive assay, we collected and analyzed data from 2 comparable, 5-month periods preceding and following implementation of the TnI-Ultra assay at the Brigham and Women's Hospital. 2. Materials and methods 2.1. Cardiac troponin I assay On February 20, 2007, Brigham and Women's Hospital Clinical Chemistry Laboratory replaced the cTnI assay (Siemens Healthcare Diagnostics, Inc., Deerfield, IL, initially introduced by Bayer HealthCare LLC, Diagnostics Division, Tarrytown, NY), with the TnI-Ultra assay (Siemens). Both assays are performed on the ADVIA Centaur analyzer (Siemens). The analytical performance of the assays was as follows: cTnI assay range 0.10–50 ng/ml, analytical sensitivity 0.07 ng/ml, 10% CV in our laboratory at 0.10– 0.13 ng/ml in repeat experiments, 99th percentile of the reference population 0.07 ng/ml. TnI-ultra assay range is 0.006–50 ng/ml, analytical sensitivity 0.006 ng/ml, 10% CV in our laboratory at 0.03 ng/ml, 99th percentile of the reference population 0.04 ng/ml. Due to the enhanced analytical performance of the assay, the diagnostic cutpoint was lowered from N 0.10 µg/l to N0.04 µg/l. 2.2. Estimated glomerular filtration rate (eGFR) Creatinine was measured using the Jaffe method on the Olympus AU2700 analyzer (Olympus America Inc., Center Valley, PA). The Olympus creatinine procedure is linear from 0.2 to 25.0 mg/dl in heparinized plasma, has CV of 1.5% for our lowest quality control material, mean of 0.85 mg/dl, and employs IDMS-traceable calibrators. eGFR was calculated from creatinine results using the Modification of Diet in Renal Disease (MDRD) Study formula for creatinine methods using IDMS-traceable calibrators and was utilized to classify the stage of chronic kidney disease (CKD). eGFRs ≥60 ml/min/1.73 m2 indicated normal renal function, eGFRs between 30 and 59 ml/min/1.73 m2 indicated moderate CKD, eGFRs between 15 and 29 ml/min/1.73 m2 indicated severe CKD and eGFRs b 15 ml/min/1.73 m2 indicated renal failure [22].
Table 1 Classification of patients with positive cardiac troponin results Group 1
Low-positive group Highest result in the low-positive range of 0.05–0.10 µg/l (n = 648)
1a. Either only one low-positive or fluctuating between low-positive and negative (n = 410)
1b. All test results in the low-positive range (n = 238) Consistently low-positive group Group 2
High-positive group
2a. Initial positive cardiac troponin in the low-positive range, followed by one or more cardiac troponin N0.10 µg/l (n = 238)
At least one cardiac troponin N 0.10 µg/l (n = 1409)
2b. Initial cardiac troponin(s) N 0.10 µg/l, subsequent result(s) in the low-positive range (n = 185) 2c. At least one result N0.10 µg/l, no results in the low-positive range (n = 986)
Patients are divided into two groups depending upon their cardiac troponin results and their subsequent trend over time. to as the negative group. Patients with positive cardiac troponins were divided into two groups with additional subsets in each group (Table 1): Group 1. Patients whose highest result was in the low-positive range, i.e., with highest cardiac troponin between 0.05 and 0.10 µg/l, further referred to as the lowpositive group (n = 648). Group 1a. Patients with either only one low-positive result and therefore no trend data or with results that fluctuate between low-positive and negative (n = 410). Group 1b. Patients in whom all results fell into the low-positive range, further referred to as the consistently low-positive group (n = 238). Group 2. Patients who had at least one cardiac troponin result N 0.10 µg/l, further referred to as high-positive group (n = 1409). Group 2a. Patients in whom the initial positive cardiac troponin result was lowpositive, followed by one or more cardiac troponin results N 0.10 µg/l (n=238). This group was described in our previous publication and confirmed that the increased sensitivity of the TnI-Ultra assay would lead to earlier detection of ACS in a subset of patients [10]. Group 2b. Patients whose initial positive cardiac troponin result(s) were N0.10 µg/l, followed by a low-positive result (n = 185). This group included patients with ACS in whom the low-positive result confirms the downward slope but otherwise offers limited additional diagnostic information. Group 2c. Patients with at least one cardiac troponin N 0.10 µg/l and no results in the low-positive range (n = 986).
We retrieved the eGFR results for all patients in groups 1b and 2, if results were within 24 h of a cardiac troponin result. If more than one creatinine was ordered during a patient event, we used the first eGFR result for statistical evaluation of renal function.
2.4. Association of eGFR and cardiac troponin I concentrations To support our hypothesis that patients with impaired renal function may have higher cardiac troponin concentrations, we retrieved specimens from 138 outpatients and 106 inpatients with a range of eGFR values and measured TnI-Ultra. Inpatients with the diagnosis of ACS and those who had undergone surgery or a transplant during hospital stay were excluded from the analysis.
2.3. Patients
2.5. Statistical analysis
We acquired an approval from our Institutional Human Research Committee for this study. Using our laboratory information system (LIS), we retrieved all TnI-Ultra test results for the 5-month period following TnI-Ultra implementation (March 1, 2007 to July 31, 2007) and all cTnI results from the same 5-month period prior to TnI-Ultra implementation (March 1, 2006 to July 31, 2006). These 2 periods have identical numbers of weekdays, weekends, and holidays. We compared the number of cardiac troponin tests performed during both time periods, the corresponding cardiac troponin results and the rates of positivity for cTnI and TnI-Ultra both hospital-wide and in the Emergency Department (ED). Our previous diagnostic cutpoint of N 0.1 µg/l was used to determine positivity for cTnI and our current cutpoint N0.04 µg/l was used for TnI-Ultra. We have previously established that TnI-Ultra and cTnI test results correlate very well with comparable results obtained by both assays. Therefore, when comparing positivity rates between cTnI and TnI-Ultra, it was assumed that test results N 0.10 µg/l by TnIUltra would also be positive by cTnI. TnI-ultra results between 0.05 µg/l and 0.10 µg/l were designated low-positive, as these results were positive by TnI-Ultra and negative by cTnI.
Data was analyzed using Analyse-It (Analyse-It Software Ltd, Yorkshire, UK). A 1way ANOVA was utilized to compare population means. A χ2 analysis was performed to determine differences in observed proportions of CKD in each cardiac troponin group. A Pearson correlation was utilized to determine the relationship between eGFR and TnIUltra. A p b 0.05 was considered significant.
The number of patients who had cardiac troponin testing during their hospital visit was also determined. Patients who never had a positive cardiac troponin were referred
3. Results 3.1. Tests performed and positivity rates We evaluated a total of 17,332 tests performed before and 17,086 tests performed after implementation of the TnI-Ultra assay. This corresponds to 7173 hospital visits by 6028 unique patients (16% repeat visits) pre- and 7003 hospital visits by 5914 unique patients (16% repeat visits) post-implementation who received cardiac troponin testing. Similar results were seen in the ED with 4251 visits by 3867 unique
S.E.F. Melanson et al. / Clinica Chimica Acta 395 (2008) 57–61
patients (9% repeat visits) pre- and 4248 visits by 3823 unique patients (10% repeat visits) post-implementation. The total number of tests ordered hospital-wide (Fig. 1a) and in the ED (Fig. 1b), did not change significantly after TnI-Ultra was implemented. Specifically, the number of cardiac troponin tests performed daily hospital-wide averaged 113.3 ± 16.0 before implementation, and 111.7± 20.7 after TnI-Ultra implementation (p = 0.45) and the average number of tests per day in the ED was 31.5 ± 6.7 before implementation and 32.2 ± 6.6 after implementation (p = 0.36). Similar results were obtained when we compared the number of patients who received cardiac troponin testing, hospital-wide and in the ED, prior to and after TnI-Ultra implementation, with a 2.4% decrease and a 0.1% decrease, respectively. There was a 44.2% relative increase in the number of positive test results hospital-wide (Fig. 1a). The absolute percentage of positive tests hospital-wide increased from 27.6% to 40.3%. The percentage of cardiac troponin results N0.10 µg/l remained constant after TnI-Ultra implementation, at 27.6%, so the additional 12.7% of positive results are indeed all in the new, low-positive range (i.e., 0.05–0.10 µg/l). We saw a 114.4% relative increase in the number of positive tests in the ED (Fig. 1b), from 8.4% positivity rate pre-implementation to 17.5% positivity with TnIUltra. Again, the percentage of cardiac troponin results N0.10 µg/l remained essentially unchanged with 8.4% before and 8.7% after implementation of TnI-Ultra, i.e., practically all additional positive results fell into the low-positive range. Similar to the number of positive tests, there was a 50.7% and 115.9% increase in the number of patients with positive results hospital-wide and in the ED, respectively. 3.2 Low-positive cardiac troponin results and renal function Since patients with CKD may have slightly increased cardiac troponin concentrations [12,13], we hypothesized that, as compared to patients with negative cardiac troponin results or patients with cardiac troponin elevations N0.10 µg/l, more patients in the consistently low-positive range (i.e., group 1b) would have abnormal renal function [23,24]. We therefore compared differences in eGFR between groups 1b and 2. In addition, we compared group 1b with patients that never had a positive cardiac troponin (i.e., negative group).
Fig. 1. Change in cardiac troponin test volume and the number of positive cardiac troponin tests both hospital-wide and in the emergency department. a. Hospital-wide. The test volume and number of positive cardiac troponin tests hospital-wide is plotted prior to TnI-Ultra implementation (white) and after TnI-Ultra implementation (diagonal stripes). The corresponding percentage increase or decrease is noted above the bar graph. b. Emergency department. The test volume and number of positive cardiac troponin tests in the emergency department is plotted prior to TnI-Ultra implementation (white) and after TnI-Ultra implementation (diagonal stripes). The corresponding percentage increase or decrease is noted above the bar graph.
59
Table 2 eGFR in patients with low-positive cardiac troponins Cardiac troponin concentration
% eGFR b 60 (p valuea)
% eGFR b 30 (p valuea)
% eGFR b 15 (p valuea)
Consistently low-positive group Cardiac troponin 0.05–0.10 µg/l n = 237b High-positive group Cardiac troponin N 0.1 µg/l n = 1404b Negative group Cardiac troponin b 0.05 µg/l n = 2492b
61.2%
26.6%
12.2%
50.7% (0.0007)
19.8% (b 0.0001)
7.7% (b0.0001)
25.7% (b0.0001)
5.9% (b0.0001)
1.6% (b0.0001)
a p value was calculated using the χ2. The high-positive and negative group were compared to the low-positive group. b n represents the number of patients who had an eGFR within 24 h of their cardiac troponin result.
Percentages of patients in the consistently low-positive (group 1b), high-positive (group 2) and negative groups with stage 3, stage 4 or stage 5 CKD were compared (Table 2). A total of 61.2% patients in the consistently low-positive cardiac troponin group had at least moderate CKD as opposed to 50.7% (p=0.0007) in the high-positive cardiac troponin group and 25.7% (pb 0.0001) in the negative group. 26.6% patients in the consistently low-positive cardiac troponin group had severe CKD while there were only 19.8% in the high-positive cardiac troponin group (p=0.0001) and 5.9% in the negative group (pb 0.0001). Finally, 12.2% of the patients in the consistently low-positive cardiac troponin group met the criteria for renal failure as opposed to 7.7% in the high-positive cardiac troponin group (pb 0.0001) and 1.6% in the negative group (pb 0.0001). 3.3. Association of eGFR and cardiac troponin I concentrations Patients with CKD have been reported to have slightly increased cardiac troponin concentrations [14,25]. We hypothesized that a significant subset of the newly positive patients would have an abnormal renal function. We therefore measured cardiac troponin in patients with various degrees of renal disease. We found a negative correlation between eGFR and TnI-Ultra in both outpatients (Fig. 2a) and inpatients (Fig. 2b). The correlation for outpatients (n=138) was −0.29 (p=0.0005). The
Fig. 2. Correlation between eGFR and cardiac troponin I in outpatients and inpatients. The concentration of TnI-Ultra (µg/l) is plotted as a function eGFR. a. Outpatients (n = 138). Negative correlation between eGFR and TnI-Ultra is illustrated (r = −0.29, p = 0.0005). b. Inpatients (n = 106). Similar negative correlation between eGFR and TnIUltra was detected (r = −0.37, p b 0.0001).
60
S.E.F. Melanson et al. / Clinica Chimica Acta 395 (2008) 57–61
correlation for inpatients (n=106) was −0.37 (pb 0.0001) after excluding ACS, surgeries and significant co-morbidities. 4. Discussion Implementation of TnI-Ultra significantly increased the rate of cardiac troponin positivity hospital-wide. This is not surprising given the higher sensitivity of the assay. At the same time, the total number of ordered tests did not increase. The higher number of positive cardiac troponins was solely due to results that fall in the low-positive range (0.05–0.10 µg/l). It can therefore be attributed to the increased sensitivity and precision of the assay and not to a change in the patient population and an ensuing increase in the number of results that would have also been positive previously using the less sensitive assay. These results suggest that while the laboratory budget will not increase, the hospital may have to allocate additional resources to triaging positive cardiac troponins. In particular, the cardiology service may receive considerably more requests for consultation for additional, low-positive patients. While we did not collect the number of cardiology consults, this conclusion was supported by discussions with multiple cardiologists who all felt that they have been receiving many more consult requests since the TnI-ultra was implemented. Interestingly, while the absolute percentage increase in cardiac troponin positivity was similar both hospital-wide (9.1%) and in the ED (12.7%), the relative percentage increase in cardiac troponin positivity was far greater in the ED, where it more than doubled. We believe that this difference is due to the nature of testing in each location. The ED more frequently evaluates new onset chest pain increasing the likelihood of low-positive results, while the rest of the hospital more frequently follows patients with known ACS, increasing the likelihood of higher cardiac troponin results. In fact, 136 out of the 238 patients (57%) in group 2a had their low-positive specimen obtained in the ED. Therefore, clinical laboratories implementing TnIUltra or another high-sensitivity cardiac troponin assay should address the differential impact on the ED and focus on discussing the impact of high-sensitivity cardiac troponin assays not only with cardiologists, but also with emergency room physicians. Correct interpretation of the low-positive cardiac troponin results is important for development of algorithms that would allow triage of patients with low-positive results without overwhelming cardiology service with unnecessary consults. While the marked increase in the number of positive cardiac troponin results may initially appear overwhelming to many clinicians, we believe that the low-positive results can be interpreted relatively easily in the context of a set of serial measurements. Patients with low-positive cardiac troponins fall into several groups (i.e., groups 1a, 1b, 2a and 2b). Patients from group 1a, (i.e., those with only one low-positive test or those that fluctuate between positive and negative), do not need urgent intervention although they may require follow-up at a later time. Low-positive cardiac troponins in patients from group 2b (i.e., patients with preceding high-positive cardiac troponin results) merely confirm a downward trend in patients with ACS. Low-positive cardiac troponins in patients from group 2a confirm that TnI-Ultra, a more sensitive cardiac troponin assay, can detect myocardial injury earlier than cTnI, as we have shown previously [10]. Group 2a accounts for 11.6% of patients with increased cardiac troponin. As compared to groups 1a, 2a and 2b, group 1b poses a greater diagnostic challenge. We asked if results in the consistently lowpositive group were suggestive of an additional underlying disease and, if so, what was its pathology and clinical significance. Our results show that patients in group 1b who have consistently low-positive cardiac troponin results often have impaired renal function when compared to patients with negative cardiac troponins. This suggests that some of the low-positive cardiac troponin results are indicative of renal pathology and that more sensitive assays can more clearly
delineate patients with and without the disease. Irrespective of pathophysiology, it has been shown by several groups that even lowpositive cTn predict poor prognostic outcomes such as subsequent MI and mortality [16,19–21]. Patients in group 1b have no rise and fall in cardiac troponin and remain in the low-positive range. Therefore they do not meet the classic diagnostic criterion of a rise and fall in plasma cardiac troponin concentrations and most likely do not have ACS. At the same time, their persistently increased cardiac troponin results are suggestive of an underlying, associated pathological condition. Since diseases other than ACS can cause elevations of cardiac troponin, we postulated that renal dysfunction may explain the consistently low-positive results seen in many patients. While multiple studies have demonstrated detectable concentrations of cardiac troponin T (TnT) in patients with CKD [12,13], it has been a general belief that cardiac troponin I (TnI) is much less frequently increased in CKD. However, using more sensitive TnI methods, two groups have recently found increases not only in TnT but also in TnI [26,27]. We confirmed that our patients with consistently low-positive concentrations of TnI-Ultra (i.e., group 1b) were significantly more likely to have renal dysfunction as judged by eGFR than patients with major cardiac troponin elevations (i.e., the high-positive group), again demonstrating the association between renal dysfunction and lowpositive cardiac troponins in some patients. Although the correlation in inpatients is confounded by acute illness and co-morbidities, we found a similar negative correlation in the non-acutely ill outpatients. As a result, low-positive cardiac troponin results should be considered in the context of other diagnostic tests such as serum creatinine and/or eGFR. A single timepoint can have a completely different clinical significance depending to which of the four groups, 1a, 1b, 2a, 2b, a patient belongs. All patients in groups 1a, 1b, and 2a present with an initial positive troponin result in the 0.05–0.10 μg/l range. Only additional cardiac troponin testing allows categorizing them further, with approximately 25% of these patients remaining in the lowpositive range and about 25% of patients with subsequent rise in troponin concentrations. Approximately 50% of patients are categorized in group 1a. For this reason, the importance of serial testing cannot be overemphasized. Furthermore, more sensitive cardiac troponin assays may allow the time window for serial measurements to be narrowed as reported previously [28]. Due to the improved precision and sensitivity of TnIUltra, significant differences in cardiac troponin may be detected in considerably less than the recommended 6 h. Narrowing the interval between serial cardiac troponin measurements will allow for more rapid interpretation and triage of patients with suspect ACS by either indicating significant rise in cardiac troponin much sooner or by detecting steady, low-concentration cardiac troponin elevations that should then be interpreted in the context of other diagnostic and clinical findings. This study has several limitations. It is a retrospective study with a focus on the impact of new cardiac troponin assays on the volume of laboratory testing and the role of the laboratory in interpretation of results. Although all patients with positive cardiac troponin(s) have myocardial injury by definition, we do not report the exact etiology of this injury, due to logistical problems and our desire to focus on the role of the laboratory. Patient outcomes before and after implementation of the more sensitive assay, such as admission rates, referrals for cardiac imaging studies and numbers of cardiology consults, were not determined. Lastly, information about the race of patients was not available in the hospital information system and, consequently, the eGFR results could not be modified for African American patients. Based on our experience, implementation of a new, highly sensitive cardiac troponin assay does not significantly change the volume of cardiac troponin testing. If this result can be generalized to other hospitals, such implementation should not increase staffing requirements and reagent budget for clinical chemistry laboratories.
S.E.F. Melanson et al. / Clinica Chimica Acta 395 (2008) 57–61
However, the number of positive results increases, in particular in the ED, so the laboratory should help both the cardiologists and ED physicians to develop algorithms for interpretation of results. With increasing sensitivity of the cardiac troponin assays, serial testing in conjunction with other diagnostic markers is critical for interpretation of more sensitive cardiac troponin assays. In summary, we illustrate the clinical significance of a cardiac troponin assay with increased sensitivity by demonstrating that lowpositive results will lead to an earlier detection of cardiac injury and will help to identify patients with cardiac troponin elevations associated with kidney disease.
[12] [13]
[14] [15] [16]
[17]
References [1] Alpert JS, Thygesen K, Antman E, Bassand JP. 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. [2] Morrow DA, Cannon CP, Jesse RL, et al. National Academy of Clinical Biochemistry Laboratory Medicine Practice Guidelines: clinical characteristics and utilization of biochemical markers in acute coronary syndromes. Clin Chem 2007;53:552–74. [3] Thygesen K, Alpert JS, White HD, et al. Universal definition of myocardial infarction: Kristian Thygesen, Joseph S. Alpert and Harvey D. White on behalf of the Joint ESC/ACCF/AHA/WHF Task Force for the redefinition of myocardial infarction. Eur Heart J 2007;28:2525–38. [4] Babuin L, Jaffe AS. Troponin: the biomarker of choice for the detection of cardiac injury. Can Med Assoc J 2005;173:1191–202. [5] Apple FS, Quist HE, Doyle PJ, Otto AP, Murakami MM. Plasma 99th percentile reference limits for cardiac troponin and creatine kinase MB mass for use with European Society of Cardiology/American College of Cardiology consensus recommendations. Clin Chem 2003;49:1331–6. [6] Bassand JP, Hamm CW, Ardissino D, et al. Guidelines for the diagnosis and treatment of non-ST-segment elevation acute coronary syndromes. The Task Force for the Diagnosis and Treatment of Non-ST-Segment Elevation Acute Coronary Syndromes of the European Society of Cardiology. Eur Heart J 2007;28:1598–660. [7] Braunwald E, Antman EM, Beasley JW, et al. ACC/AHA 2002 guideline update for the management of patients with unstable angina and non-ST-segment elevation myocardial infarction—summary article: a report of the American College of Cardiology/American Heart Association task force on practice guidelines (Committee on the Management of Patients With Unstable Angina). J Am Coll Cardiol 2002;40:1366–74. [8] Newby LK, Alpert JS, Ohman EM, Thygesen K, Califf RM. Changing the diagnosis of acute myocardial infarction: implications for practice and clinical investigations. Am Heart J 2002;144:957–80. [9] Melanson SE, Tanasijevic MJ, Jarolim P. Cardiac troponin assays: a view from the clinical chemistry laboratory. Circulation 2007;116:e501–4. [10] Melanson SE, Morrow DA, Jarolim P. Earlier detection of myocardial injury in a preliminary evaluation using a new troponin I assay with improved sensitivity. Am J Clin Pathol 2007;128:282–6. [11] Wu AH, Jaffe AS, Apple FS, et al. National Academy of Clinical Biochemistry Laboratory Medicine Practice Guidelines: use of cardiac troponin and B-type
[18]
[19]
[20] [21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
61
natriuretic peptide or N-terminal proB-type natriuretic peptide for etiologies other than acute coronary syndromes and heart failure. Clin Chem 2007. Abbas NA, John RI, Webb MC, et al. Cardiac troponins and renal function in nondialysis patients with chronic kidney disease. Clin Chem 2005;51:2059–66. Fahie-Wilson MN, Carmichael DJ, Delaney MP, Stevens PE, Hall EM, Lamb EJ. Cardiac troponin T circulates in the free, intact form in patients with kidney failure. Clin Chem 2006;52:414–20. Sommerer C, Beimler J, Schwenger V, et al. Cardiac biomarkers and survival in haemodialysis patients. Eur J Clin Invest 2007;37:350–6. Latini R, Masson S, Anand IS, et al. Prognostic value of very low plasma concentrations of troponin T in patients with stable chronic heart failure. Circulation 2007;116:1242–9. Apple FS, Smith SW, Pearce LA, Ler R, Murakami MM. Use of the Centaur TnI-Ultra assay for detection of myocardial infarction and adverse events in patients presenting with symptoms suggestive of acute coronary syndrome. Clin Chem 2008;54:723–8. Eggers KM, Lagerqvist B, Venge P, Wallentin L, Lindahl B. Persistent cardiac troponin I elevation in stabilized patients after an episode of acute coronary syndrome predicts long-term mortality. Circulation 2007;116:1907–14. James S, Armstrong P, Califf R, et al. Troponin T concentrations and risk of 30-day outcomes in patients with the acute coronary syndrome: prospective verification in the GUSTO-IV trial. Am J Med 2003;115:178–84. Kavsak PA, Newman AM, Ko DT, et al. Is a pattern of increasing biomarker concentrations important for long-term risk stratification in acute coronary syndrome patients presenting early after the onset of symptoms? Clin Chem 2008;54:747–51. Kavsak PA, Newman AM, Lustig V, et al. Long-term health outcomes associated with detectable troponin I concentrations. Clin Chem 2007;53(2):220–7. Kontos MC, Shah R, Fritz LM, et al. Implication of different cardiac troponin I concentrations for clinical outcomes and prognosis of acute chest pain patients. J Am Coll Cardiol 2004;43:958–65. Levey AS, Coresh J, Balk E, et al. National Kidney Foundation practice guidelines for chronic kidney disease: evaluation, classification, and stratification. Ann Intern Med 2003;139:137–47. Ix JH, Shlipak MG, Liu HH, Schiller NB, Whooley MA. Association between renal insufficiency and inducible ischemia in patients with coronary artery disease: the heart and soul study. J Am Soc Nephrol 2003;14:3233–8. Sarnak MJ, Levey AS, Schoolwerth AC, et al. Kidney disease as a risk factor for development of cardiovascular disease: a statement from the American Heart Association Councils on Kidney in Cardiovascular Disease, High Blood Pressure Research, Clinical Cardiology, and Epidemiology and Prevention. Circulation 2003;108:2154–69. Satyan S, Light RP, Agarwal R. Relationships of N-terminal pro-B-natriuretic peptide and cardiac troponin T to left ventricular mass and function and mortality in asymptomatic hemodialysis patients. Am J Kidney Dis 2007;50:1009–19. Hickman PE, Koerbin G, Southcott E, et al. Newer cardiac troponin I assays have similar performance to troponin T in patients with end-stage renal disease. Ann Clin Biochem 2007;44:285–9. Lamb EJ, Kenny C, Abbas NA, et al. Cardiac troponin I concentration is commonly increased in nondialysis patients with CKD: experience with a sensitive assay. Am J Kidney Dis 2007;49:507–16. Macrae AR, Kavsak PA, Lustig V, et al. Assessing the requirement for the 6-hour interval between specimens in the American Heart Association Classification of Myocardial Infarction in Epidemiology and Clinical Research Studies. Clin Chem 2006;52:812–8.
Contents lists available at ScienceDirect
Clinica Chimica Acta j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / c l i n c h i m
Implementation of a highly sensitive cardiac troponin I assay: Test volumes, positivity rates and interpretation of results Stacy E.F. Melanson, Michael J. Conrad, Nima Mosammaparast, Petr Jarolim ⁎ Department of Pathology, Clinical Laboratories Division, Brigham and Women's Hospital, Harvard Medical School, Boston, MA 02115, United States
A R T I C L E
I N F O
Article history: Received 31 December 2007 Received in revised form 30 April 2008 Accepted 1 May 2008 Available online 15 May 2008 Keywords: Cardiac troponin Implementation Sensitive assay Acute coronary syndrome Myocardial injury Renal dysfunction
A B S T R A C T Background: While more sensitive cardiac troponin (cTn) assays may overcome current limitations in diagnosing acute coronary syndrome (ACS), clinicians and laboratorians are concerned about the clinical impact of higher rates of cTn positivity. Methods: We collected statistics on test volume and positivity rates before and after the implementation of a more sensitive assay. We divided patients into 6 groups based on their cardiac troponin I (cTnI) results and utilized additional laboratory test results to determine the impact of the new assay. In addition, we assessed the clinical significance of low-positive cTnI results (0.05–0.10 µg/l). Results: Lowering the diagnostic cutoff from 0.10 to 0.04 µg/l increased the number of positive test results by 44.2% hospital-wide and 114.4% in the emergency department without significantly changing test volume. In some patients, low-positive results were part of the typical rise and fall of cTnI consistent with myocardial damage. Diagnosis was more challenging in patients with consistently low-positive results. Patients in this group were significantly more likely to have impaired renal function than those with cTnI elevations N 0.10 µg/l. Conclusions: More sensitive cTn assays will increase the number of positive results and allow for earlier detection of cardiac injury while also detecting non-ACS related pathologies. © 2008 Elsevier B.V. All rights reserved.
1. Introduction The ability to rapidly determine the presence of myocardial ischemia is crucial for patients presenting with chest pain or other signs and symptoms suggestive of acute coronary syndrome (ACS). Cardiac troponin has become the gold standard among biochemical markers used for both the diagnosis and prognosis of such patients [1–3]. However, until recently, cardiac troponin testing has had various limitations including suboptimal analytical sensitivities and low-end imprecision. As a result of assay performance and the kinetics of cardiac troponin release from damaged cardiac myocytes, patients presenting hours after the onset of symptoms may still have undetectable cardiac troponin concentrations. Typically, patients will have measurable cardiac troponin within 4 to 6 h of symptoms [4], but the opportunity for rapid triage and early intervention may be lost by that time. Thus, improvements in the performance of cardiac troponin assays can
Abbreviations: Acute coronary syndrome, (ACS); American College of Cardiology, (ACC); European Society of Cardiology, (ESC); estimated glomerular filtration rate, (eGFR); Modification of Diet in Renal Disease, (MDRD); chronic kidney disease, (CKD); congestive heart failure, (CHF); laboratory information system, (LIS). ⁎ Corresponding author. Brigham and Women's Hospital, 75 Francis Street, Amory 2, Boston, MA 02115, United States. Tel.: +1 617 732 6672; fax: +1 617 731 4872. E-mail address: [email protected] (P. Jarolim). 0009-8981/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.cca.2008.05.007
potentially overcome the current limitations and benefit patients presenting early in the course of myocardial ischemia. The American College of Cardiology (ACC) and the European Society of Cardiology (ESC) both recommend using a diagnostic cutpoint at the 99th percentile of the reference population with a coefficient of variation (CV) of ≤10% for the diagnosis of ACS [1–3,5–8]. Some of the currently available cardiac troponin assays do not meet these requirements [9]. Siemens Healthcare Diagnostics recently replaced their older cTnI assay with an improved assay, TnI-Ultra, which offers both enhanced sensitivity and precision. We have previously compared the characteristics of cTnI to TnI-Ultra and confirmed that TnI-Ultra offers improved analytical and clinical performance, consistent with the ACC/ESC guidelines [10]. TnI-Ultra achieved a 10% CV at the manufacturer's 99th percentile resulting in a diagnostic decision point of 0.04 µg/l. Because of the improved analytical performance of the TnI-Ultra assay and the decision by the manufacturer to discontinue their cTnI assay, we implemented TnI-Ultra testing in our clinical laboratory. We expected and previously confirmed that the enhanced performance characteristics of the TnI-Ultra assay would have anticipated benefits such as the earlier diagnosis of ACS in a subset of patients [10]. At the same time, both the clinicians and laboratorians were concerned about the clinical utility of cardiac troponin assays with increased sensitivity in other patients. It was clear that the number of positive cardiac troponin I results would increase due to the
58
S.E.F. Melanson et al. / Clinica Chimica Acta 395 (2008) 57–61
new low-positive range of 0.05–0.10 µg/l. However, we did not know whether or not this would translate into an increase in the volume and cost of cardiac troponin I testing for the clinical laboratory or how these low-positive test results would affect diagnostic decisions of our clinicians and subsequent treatment of our patients. Data on the clinical significance of low-positive cardiac troponins is limited. It is known that cardiac troponin can be increased in conditions other than ACS [4,11] such as renal disease [12–14] and congestive heart failure [4,15]. Sommerer et al. [14] have shown that plasma concentrations of cTnT are increased in approximately 40% of asymptomatic chronic haemodialysis patients. Latini et al. [15] reported detectable cTnT in patients with chronic heart failure. Several groups have also studied the predictive value of small cTn elevations for adverse long-term cardiovascular outcomes [16–21]. Using a sensitive cTnI assay, Eggers et al. [17] found that increased cTnI concentrations N0.01 μg/l after an episode of non-ST-elevation ACS predicted mortality during long-term follow-up. However, how these findings will affect the interpretation of low-positive cTnI results and the utility of cardiac troponin assays with increased sensitivity for the diagnosis of ACS is not clear. To clarify the impact of implementation of a more sensitive assay, we collected and analyzed data from 2 comparable, 5-month periods preceding and following implementation of the TnI-Ultra assay at the Brigham and Women's Hospital. 2. Materials and methods 2.1. Cardiac troponin I assay On February 20, 2007, Brigham and Women's Hospital Clinical Chemistry Laboratory replaced the cTnI assay (Siemens Healthcare Diagnostics, Inc., Deerfield, IL, initially introduced by Bayer HealthCare LLC, Diagnostics Division, Tarrytown, NY), with the TnI-Ultra assay (Siemens). Both assays are performed on the ADVIA Centaur analyzer (Siemens). The analytical performance of the assays was as follows: cTnI assay range 0.10–50 ng/ml, analytical sensitivity 0.07 ng/ml, 10% CV in our laboratory at 0.10– 0.13 ng/ml in repeat experiments, 99th percentile of the reference population 0.07 ng/ml. TnI-ultra assay range is 0.006–50 ng/ml, analytical sensitivity 0.006 ng/ml, 10% CV in our laboratory at 0.03 ng/ml, 99th percentile of the reference population 0.04 ng/ml. Due to the enhanced analytical performance of the assay, the diagnostic cutpoint was lowered from N 0.10 µg/l to N0.04 µg/l. 2.2. Estimated glomerular filtration rate (eGFR) Creatinine was measured using the Jaffe method on the Olympus AU2700 analyzer (Olympus America Inc., Center Valley, PA). The Olympus creatinine procedure is linear from 0.2 to 25.0 mg/dl in heparinized plasma, has CV of 1.5% for our lowest quality control material, mean of 0.85 mg/dl, and employs IDMS-traceable calibrators. eGFR was calculated from creatinine results using the Modification of Diet in Renal Disease (MDRD) Study formula for creatinine methods using IDMS-traceable calibrators and was utilized to classify the stage of chronic kidney disease (CKD). eGFRs ≥60 ml/min/1.73 m2 indicated normal renal function, eGFRs between 30 and 59 ml/min/1.73 m2 indicated moderate CKD, eGFRs between 15 and 29 ml/min/1.73 m2 indicated severe CKD and eGFRs b 15 ml/min/1.73 m2 indicated renal failure [22].
Table 1 Classification of patients with positive cardiac troponin results Group 1
Low-positive group Highest result in the low-positive range of 0.05–0.10 µg/l (n = 648)
1a. Either only one low-positive or fluctuating between low-positive and negative (n = 410)
1b. All test results in the low-positive range (n = 238) Consistently low-positive group Group 2
High-positive group
2a. Initial positive cardiac troponin in the low-positive range, followed by one or more cardiac troponin N0.10 µg/l (n = 238)
At least one cardiac troponin N 0.10 µg/l (n = 1409)
2b. Initial cardiac troponin(s) N 0.10 µg/l, subsequent result(s) in the low-positive range (n = 185) 2c. At least one result N0.10 µg/l, no results in the low-positive range (n = 986)
Patients are divided into two groups depending upon their cardiac troponin results and their subsequent trend over time. to as the negative group. Patients with positive cardiac troponins were divided into two groups with additional subsets in each group (Table 1): Group 1. Patients whose highest result was in the low-positive range, i.e., with highest cardiac troponin between 0.05 and 0.10 µg/l, further referred to as the lowpositive group (n = 648). Group 1a. Patients with either only one low-positive result and therefore no trend data or with results that fluctuate between low-positive and negative (n = 410). Group 1b. Patients in whom all results fell into the low-positive range, further referred to as the consistently low-positive group (n = 238). Group 2. Patients who had at least one cardiac troponin result N 0.10 µg/l, further referred to as high-positive group (n = 1409). Group 2a. Patients in whom the initial positive cardiac troponin result was lowpositive, followed by one or more cardiac troponin results N 0.10 µg/l (n=238). This group was described in our previous publication and confirmed that the increased sensitivity of the TnI-Ultra assay would lead to earlier detection of ACS in a subset of patients [10]. Group 2b. Patients whose initial positive cardiac troponin result(s) were N0.10 µg/l, followed by a low-positive result (n = 185). This group included patients with ACS in whom the low-positive result confirms the downward slope but otherwise offers limited additional diagnostic information. Group 2c. Patients with at least one cardiac troponin N 0.10 µg/l and no results in the low-positive range (n = 986).
We retrieved the eGFR results for all patients in groups 1b and 2, if results were within 24 h of a cardiac troponin result. If more than one creatinine was ordered during a patient event, we used the first eGFR result for statistical evaluation of renal function.
2.4. Association of eGFR and cardiac troponin I concentrations To support our hypothesis that patients with impaired renal function may have higher cardiac troponin concentrations, we retrieved specimens from 138 outpatients and 106 inpatients with a range of eGFR values and measured TnI-Ultra. Inpatients with the diagnosis of ACS and those who had undergone surgery or a transplant during hospital stay were excluded from the analysis.
2.3. Patients
2.5. Statistical analysis
We acquired an approval from our Institutional Human Research Committee for this study. Using our laboratory information system (LIS), we retrieved all TnI-Ultra test results for the 5-month period following TnI-Ultra implementation (March 1, 2007 to July 31, 2007) and all cTnI results from the same 5-month period prior to TnI-Ultra implementation (March 1, 2006 to July 31, 2006). These 2 periods have identical numbers of weekdays, weekends, and holidays. We compared the number of cardiac troponin tests performed during both time periods, the corresponding cardiac troponin results and the rates of positivity for cTnI and TnI-Ultra both hospital-wide and in the Emergency Department (ED). Our previous diagnostic cutpoint of N 0.1 µg/l was used to determine positivity for cTnI and our current cutpoint N0.04 µg/l was used for TnI-Ultra. We have previously established that TnI-Ultra and cTnI test results correlate very well with comparable results obtained by both assays. Therefore, when comparing positivity rates between cTnI and TnI-Ultra, it was assumed that test results N 0.10 µg/l by TnIUltra would also be positive by cTnI. TnI-ultra results between 0.05 µg/l and 0.10 µg/l were designated low-positive, as these results were positive by TnI-Ultra and negative by cTnI.
Data was analyzed using Analyse-It (Analyse-It Software Ltd, Yorkshire, UK). A 1way ANOVA was utilized to compare population means. A χ2 analysis was performed to determine differences in observed proportions of CKD in each cardiac troponin group. A Pearson correlation was utilized to determine the relationship between eGFR and TnIUltra. A p b 0.05 was considered significant.
The number of patients who had cardiac troponin testing during their hospital visit was also determined. Patients who never had a positive cardiac troponin were referred
3. Results 3.1. Tests performed and positivity rates We evaluated a total of 17,332 tests performed before and 17,086 tests performed after implementation of the TnI-Ultra assay. This corresponds to 7173 hospital visits by 6028 unique patients (16% repeat visits) pre- and 7003 hospital visits by 5914 unique patients (16% repeat visits) post-implementation who received cardiac troponin testing. Similar results were seen in the ED with 4251 visits by 3867 unique
S.E.F. Melanson et al. / Clinica Chimica Acta 395 (2008) 57–61
patients (9% repeat visits) pre- and 4248 visits by 3823 unique patients (10% repeat visits) post-implementation. The total number of tests ordered hospital-wide (Fig. 1a) and in the ED (Fig. 1b), did not change significantly after TnI-Ultra was implemented. Specifically, the number of cardiac troponin tests performed daily hospital-wide averaged 113.3 ± 16.0 before implementation, and 111.7± 20.7 after TnI-Ultra implementation (p = 0.45) and the average number of tests per day in the ED was 31.5 ± 6.7 before implementation and 32.2 ± 6.6 after implementation (p = 0.36). Similar results were obtained when we compared the number of patients who received cardiac troponin testing, hospital-wide and in the ED, prior to and after TnI-Ultra implementation, with a 2.4% decrease and a 0.1% decrease, respectively. There was a 44.2% relative increase in the number of positive test results hospital-wide (Fig. 1a). The absolute percentage of positive tests hospital-wide increased from 27.6% to 40.3%. The percentage of cardiac troponin results N0.10 µg/l remained constant after TnI-Ultra implementation, at 27.6%, so the additional 12.7% of positive results are indeed all in the new, low-positive range (i.e., 0.05–0.10 µg/l). We saw a 114.4% relative increase in the number of positive tests in the ED (Fig. 1b), from 8.4% positivity rate pre-implementation to 17.5% positivity with TnIUltra. Again, the percentage of cardiac troponin results N0.10 µg/l remained essentially unchanged with 8.4% before and 8.7% after implementation of TnI-Ultra, i.e., practically all additional positive results fell into the low-positive range. Similar to the number of positive tests, there was a 50.7% and 115.9% increase in the number of patients with positive results hospital-wide and in the ED, respectively. 3.2 Low-positive cardiac troponin results and renal function Since patients with CKD may have slightly increased cardiac troponin concentrations [12,13], we hypothesized that, as compared to patients with negative cardiac troponin results or patients with cardiac troponin elevations N0.10 µg/l, more patients in the consistently low-positive range (i.e., group 1b) would have abnormal renal function [23,24]. We therefore compared differences in eGFR between groups 1b and 2. In addition, we compared group 1b with patients that never had a positive cardiac troponin (i.e., negative group).
Fig. 1. Change in cardiac troponin test volume and the number of positive cardiac troponin tests both hospital-wide and in the emergency department. a. Hospital-wide. The test volume and number of positive cardiac troponin tests hospital-wide is plotted prior to TnI-Ultra implementation (white) and after TnI-Ultra implementation (diagonal stripes). The corresponding percentage increase or decrease is noted above the bar graph. b. Emergency department. The test volume and number of positive cardiac troponin tests in the emergency department is plotted prior to TnI-Ultra implementation (white) and after TnI-Ultra implementation (diagonal stripes). The corresponding percentage increase or decrease is noted above the bar graph.
59
Table 2 eGFR in patients with low-positive cardiac troponins Cardiac troponin concentration
% eGFR b 60 (p valuea)
% eGFR b 30 (p valuea)
% eGFR b 15 (p valuea)
Consistently low-positive group Cardiac troponin 0.05–0.10 µg/l n = 237b High-positive group Cardiac troponin N 0.1 µg/l n = 1404b Negative group Cardiac troponin b 0.05 µg/l n = 2492b
61.2%
26.6%
12.2%
50.7% (0.0007)
19.8% (b 0.0001)
7.7% (b0.0001)
25.7% (b0.0001)
5.9% (b0.0001)
1.6% (b0.0001)
a p value was calculated using the χ2. The high-positive and negative group were compared to the low-positive group. b n represents the number of patients who had an eGFR within 24 h of their cardiac troponin result.
Percentages of patients in the consistently low-positive (group 1b), high-positive (group 2) and negative groups with stage 3, stage 4 or stage 5 CKD were compared (Table 2). A total of 61.2% patients in the consistently low-positive cardiac troponin group had at least moderate CKD as opposed to 50.7% (p=0.0007) in the high-positive cardiac troponin group and 25.7% (pb 0.0001) in the negative group. 26.6% patients in the consistently low-positive cardiac troponin group had severe CKD while there were only 19.8% in the high-positive cardiac troponin group (p=0.0001) and 5.9% in the negative group (pb 0.0001). Finally, 12.2% of the patients in the consistently low-positive cardiac troponin group met the criteria for renal failure as opposed to 7.7% in the high-positive cardiac troponin group (pb 0.0001) and 1.6% in the negative group (pb 0.0001). 3.3. Association of eGFR and cardiac troponin I concentrations Patients with CKD have been reported to have slightly increased cardiac troponin concentrations [14,25]. We hypothesized that a significant subset of the newly positive patients would have an abnormal renal function. We therefore measured cardiac troponin in patients with various degrees of renal disease. We found a negative correlation between eGFR and TnI-Ultra in both outpatients (Fig. 2a) and inpatients (Fig. 2b). The correlation for outpatients (n=138) was −0.29 (p=0.0005). The
Fig. 2. Correlation between eGFR and cardiac troponin I in outpatients and inpatients. The concentration of TnI-Ultra (µg/l) is plotted as a function eGFR. a. Outpatients (n = 138). Negative correlation between eGFR and TnI-Ultra is illustrated (r = −0.29, p = 0.0005). b. Inpatients (n = 106). Similar negative correlation between eGFR and TnIUltra was detected (r = −0.37, p b 0.0001).
60
S.E.F. Melanson et al. / Clinica Chimica Acta 395 (2008) 57–61
correlation for inpatients (n=106) was −0.37 (pb 0.0001) after excluding ACS, surgeries and significant co-morbidities. 4. Discussion Implementation of TnI-Ultra significantly increased the rate of cardiac troponin positivity hospital-wide. This is not surprising given the higher sensitivity of the assay. At the same time, the total number of ordered tests did not increase. The higher number of positive cardiac troponins was solely due to results that fall in the low-positive range (0.05–0.10 µg/l). It can therefore be attributed to the increased sensitivity and precision of the assay and not to a change in the patient population and an ensuing increase in the number of results that would have also been positive previously using the less sensitive assay. These results suggest that while the laboratory budget will not increase, the hospital may have to allocate additional resources to triaging positive cardiac troponins. In particular, the cardiology service may receive considerably more requests for consultation for additional, low-positive patients. While we did not collect the number of cardiology consults, this conclusion was supported by discussions with multiple cardiologists who all felt that they have been receiving many more consult requests since the TnI-ultra was implemented. Interestingly, while the absolute percentage increase in cardiac troponin positivity was similar both hospital-wide (9.1%) and in the ED (12.7%), the relative percentage increase in cardiac troponin positivity was far greater in the ED, where it more than doubled. We believe that this difference is due to the nature of testing in each location. The ED more frequently evaluates new onset chest pain increasing the likelihood of low-positive results, while the rest of the hospital more frequently follows patients with known ACS, increasing the likelihood of higher cardiac troponin results. In fact, 136 out of the 238 patients (57%) in group 2a had their low-positive specimen obtained in the ED. Therefore, clinical laboratories implementing TnIUltra or another high-sensitivity cardiac troponin assay should address the differential impact on the ED and focus on discussing the impact of high-sensitivity cardiac troponin assays not only with cardiologists, but also with emergency room physicians. Correct interpretation of the low-positive cardiac troponin results is important for development of algorithms that would allow triage of patients with low-positive results without overwhelming cardiology service with unnecessary consults. While the marked increase in the number of positive cardiac troponin results may initially appear overwhelming to many clinicians, we believe that the low-positive results can be interpreted relatively easily in the context of a set of serial measurements. Patients with low-positive cardiac troponins fall into several groups (i.e., groups 1a, 1b, 2a and 2b). Patients from group 1a, (i.e., those with only one low-positive test or those that fluctuate between positive and negative), do not need urgent intervention although they may require follow-up at a later time. Low-positive cardiac troponins in patients from group 2b (i.e., patients with preceding high-positive cardiac troponin results) merely confirm a downward trend in patients with ACS. Low-positive cardiac troponins in patients from group 2a confirm that TnI-Ultra, a more sensitive cardiac troponin assay, can detect myocardial injury earlier than cTnI, as we have shown previously [10]. Group 2a accounts for 11.6% of patients with increased cardiac troponin. As compared to groups 1a, 2a and 2b, group 1b poses a greater diagnostic challenge. We asked if results in the consistently lowpositive group were suggestive of an additional underlying disease and, if so, what was its pathology and clinical significance. Our results show that patients in group 1b who have consistently low-positive cardiac troponin results often have impaired renal function when compared to patients with negative cardiac troponins. This suggests that some of the low-positive cardiac troponin results are indicative of renal pathology and that more sensitive assays can more clearly
delineate patients with and without the disease. Irrespective of pathophysiology, it has been shown by several groups that even lowpositive cTn predict poor prognostic outcomes such as subsequent MI and mortality [16,19–21]. Patients in group 1b have no rise and fall in cardiac troponin and remain in the low-positive range. Therefore they do not meet the classic diagnostic criterion of a rise and fall in plasma cardiac troponin concentrations and most likely do not have ACS. At the same time, their persistently increased cardiac troponin results are suggestive of an underlying, associated pathological condition. Since diseases other than ACS can cause elevations of cardiac troponin, we postulated that renal dysfunction may explain the consistently low-positive results seen in many patients. While multiple studies have demonstrated detectable concentrations of cardiac troponin T (TnT) in patients with CKD [12,13], it has been a general belief that cardiac troponin I (TnI) is much less frequently increased in CKD. However, using more sensitive TnI methods, two groups have recently found increases not only in TnT but also in TnI [26,27]. We confirmed that our patients with consistently low-positive concentrations of TnI-Ultra (i.e., group 1b) were significantly more likely to have renal dysfunction as judged by eGFR than patients with major cardiac troponin elevations (i.e., the high-positive group), again demonstrating the association between renal dysfunction and lowpositive cardiac troponins in some patients. Although the correlation in inpatients is confounded by acute illness and co-morbidities, we found a similar negative correlation in the non-acutely ill outpatients. As a result, low-positive cardiac troponin results should be considered in the context of other diagnostic tests such as serum creatinine and/or eGFR. A single timepoint can have a completely different clinical significance depending to which of the four groups, 1a, 1b, 2a, 2b, a patient belongs. All patients in groups 1a, 1b, and 2a present with an initial positive troponin result in the 0.05–0.10 μg/l range. Only additional cardiac troponin testing allows categorizing them further, with approximately 25% of these patients remaining in the lowpositive range and about 25% of patients with subsequent rise in troponin concentrations. Approximately 50% of patients are categorized in group 1a. For this reason, the importance of serial testing cannot be overemphasized. Furthermore, more sensitive cardiac troponin assays may allow the time window for serial measurements to be narrowed as reported previously [28]. Due to the improved precision and sensitivity of TnIUltra, significant differences in cardiac troponin may be detected in considerably less than the recommended 6 h. Narrowing the interval between serial cardiac troponin measurements will allow for more rapid interpretation and triage of patients with suspect ACS by either indicating significant rise in cardiac troponin much sooner or by detecting steady, low-concentration cardiac troponin elevations that should then be interpreted in the context of other diagnostic and clinical findings. This study has several limitations. It is a retrospective study with a focus on the impact of new cardiac troponin assays on the volume of laboratory testing and the role of the laboratory in interpretation of results. Although all patients with positive cardiac troponin(s) have myocardial injury by definition, we do not report the exact etiology of this injury, due to logistical problems and our desire to focus on the role of the laboratory. Patient outcomes before and after implementation of the more sensitive assay, such as admission rates, referrals for cardiac imaging studies and numbers of cardiology consults, were not determined. Lastly, information about the race of patients was not available in the hospital information system and, consequently, the eGFR results could not be modified for African American patients. Based on our experience, implementation of a new, highly sensitive cardiac troponin assay does not significantly change the volume of cardiac troponin testing. If this result can be generalized to other hospitals, such implementation should not increase staffing requirements and reagent budget for clinical chemistry laboratories.
S.E.F. Melanson et al. / Clinica Chimica Acta 395 (2008) 57–61
However, the number of positive results increases, in particular in the ED, so the laboratory should help both the cardiologists and ED physicians to develop algorithms for interpretation of results. With increasing sensitivity of the cardiac troponin assays, serial testing in conjunction with other diagnostic markers is critical for interpretation of more sensitive cardiac troponin assays. In summary, we illustrate the clinical significance of a cardiac troponin assay with increased sensitivity by demonstrating that lowpositive results will lead to an earlier detection of cardiac injury and will help to identify patients with cardiac troponin elevations associated with kidney disease.
[12] [13]
[14] [15] [16]
[17]
References [1] Alpert JS, Thygesen K, Antman E, Bassand JP. 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. [2] Morrow DA, Cannon CP, Jesse RL, et al. National Academy of Clinical Biochemistry Laboratory Medicine Practice Guidelines: clinical characteristics and utilization of biochemical markers in acute coronary syndromes. Clin Chem 2007;53:552–74. [3] Thygesen K, Alpert JS, White HD, et al. Universal definition of myocardial infarction: Kristian Thygesen, Joseph S. Alpert and Harvey D. White on behalf of the Joint ESC/ACCF/AHA/WHF Task Force for the redefinition of myocardial infarction. Eur Heart J 2007;28:2525–38. [4] Babuin L, Jaffe AS. Troponin: the biomarker of choice for the detection of cardiac injury. Can Med Assoc J 2005;173:1191–202. [5] Apple FS, Quist HE, Doyle PJ, Otto AP, Murakami MM. Plasma 99th percentile reference limits for cardiac troponin and creatine kinase MB mass for use with European Society of Cardiology/American College of Cardiology consensus recommendations. Clin Chem 2003;49:1331–6. [6] Bassand JP, Hamm CW, Ardissino D, et al. Guidelines for the diagnosis and treatment of non-ST-segment elevation acute coronary syndromes. The Task Force for the Diagnosis and Treatment of Non-ST-Segment Elevation Acute Coronary Syndromes of the European Society of Cardiology. Eur Heart J 2007;28:1598–660. [7] Braunwald E, Antman EM, Beasley JW, et al. ACC/AHA 2002 guideline update for the management of patients with unstable angina and non-ST-segment elevation myocardial infarction—summary article: a report of the American College of Cardiology/American Heart Association task force on practice guidelines (Committee on the Management of Patients With Unstable Angina). J Am Coll Cardiol 2002;40:1366–74. [8] Newby LK, Alpert JS, Ohman EM, Thygesen K, Califf RM. Changing the diagnosis of acute myocardial infarction: implications for practice and clinical investigations. Am Heart J 2002;144:957–80. [9] Melanson SE, Tanasijevic MJ, Jarolim P. Cardiac troponin assays: a view from the clinical chemistry laboratory. Circulation 2007;116:e501–4. [10] Melanson SE, Morrow DA, Jarolim P. Earlier detection of myocardial injury in a preliminary evaluation using a new troponin I assay with improved sensitivity. Am J Clin Pathol 2007;128:282–6. [11] Wu AH, Jaffe AS, Apple FS, et al. National Academy of Clinical Biochemistry Laboratory Medicine Practice Guidelines: use of cardiac troponin and B-type
[18]
[19]
[20] [21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
61
natriuretic peptide or N-terminal proB-type natriuretic peptide for etiologies other than acute coronary syndromes and heart failure. Clin Chem 2007. Abbas NA, John RI, Webb MC, et al. Cardiac troponins and renal function in nondialysis patients with chronic kidney disease. Clin Chem 2005;51:2059–66. Fahie-Wilson MN, Carmichael DJ, Delaney MP, Stevens PE, Hall EM, Lamb EJ. Cardiac troponin T circulates in the free, intact form in patients with kidney failure. Clin Chem 2006;52:414–20. Sommerer C, Beimler J, Schwenger V, et al. Cardiac biomarkers and survival in haemodialysis patients. Eur J Clin Invest 2007;37:350–6. Latini R, Masson S, Anand IS, et al. Prognostic value of very low plasma concentrations of troponin T in patients with stable chronic heart failure. Circulation 2007;116:1242–9. Apple FS, Smith SW, Pearce LA, Ler R, Murakami MM. Use of the Centaur TnI-Ultra assay for detection of myocardial infarction and adverse events in patients presenting with symptoms suggestive of acute coronary syndrome. Clin Chem 2008;54:723–8. Eggers KM, Lagerqvist B, Venge P, Wallentin L, Lindahl B. Persistent cardiac troponin I elevation in stabilized patients after an episode of acute coronary syndrome predicts long-term mortality. Circulation 2007;116:1907–14. James S, Armstrong P, Califf R, et al. Troponin T concentrations and risk of 30-day outcomes in patients with the acute coronary syndrome: prospective verification in the GUSTO-IV trial. Am J Med 2003;115:178–84. Kavsak PA, Newman AM, Ko DT, et al. Is a pattern of increasing biomarker concentrations important for long-term risk stratification in acute coronary syndrome patients presenting early after the onset of symptoms? Clin Chem 2008;54:747–51. Kavsak PA, Newman AM, Lustig V, et al. Long-term health outcomes associated with detectable troponin I concentrations. Clin Chem 2007;53(2):220–7. Kontos MC, Shah R, Fritz LM, et al. Implication of different cardiac troponin I concentrations for clinical outcomes and prognosis of acute chest pain patients. J Am Coll Cardiol 2004;43:958–65. Levey AS, Coresh J, Balk E, et al. National Kidney Foundation practice guidelines for chronic kidney disease: evaluation, classification, and stratification. Ann Intern Med 2003;139:137–47. Ix JH, Shlipak MG, Liu HH, Schiller NB, Whooley MA. Association between renal insufficiency and inducible ischemia in patients with coronary artery disease: the heart and soul study. J Am Soc Nephrol 2003;14:3233–8. Sarnak MJ, Levey AS, Schoolwerth AC, et al. Kidney disease as a risk factor for development of cardiovascular disease: a statement from the American Heart Association Councils on Kidney in Cardiovascular Disease, High Blood Pressure Research, Clinical Cardiology, and Epidemiology and Prevention. Circulation 2003;108:2154–69. Satyan S, Light RP, Agarwal R. Relationships of N-terminal pro-B-natriuretic peptide and cardiac troponin T to left ventricular mass and function and mortality in asymptomatic hemodialysis patients. Am J Kidney Dis 2007;50:1009–19. Hickman PE, Koerbin G, Southcott E, et al. Newer cardiac troponin I assays have similar performance to troponin T in patients with end-stage renal disease. Ann Clin Biochem 2007;44:285–9. Lamb EJ, Kenny C, Abbas NA, et al. Cardiac troponin I concentration is commonly increased in nondialysis patients with CKD: experience with a sensitive assay. Am J Kidney Dis 2007;49:507–16. Macrae AR, Kavsak PA, Lustig V, et al. Assessing the requirement for the 6-hour interval between specimens in the American Heart Association Classification of Myocardial Infarction in Epidemiology and Clinical Research Studies. Clin Chem 2006;52:812–8.