OUP user menu

Early diagnosis of acute myocardial infarction in the elderly using more sensitive cardiac troponin assays

Miriam Reiter, Raphael Twerenbold, Tobias Reichlin, Philip Haaf, Frederico Peter, Julia Meissner, Willibald Hochholzer, Claudia Stelzig, Michael Freese, Corinna Heinisch, Tobias Breidthardt, Heike Freidank, Katrin Winkler, Isabel Campodarve, Joaquim Gea, Christian Mueller
DOI: http://dx.doi.org/10.1093/eurheartj/ehr033 1379-1389 First published online: 1 March 2011

Abstract

Aims To examine the diagnostic accuracy of sensitive cardiac troponin (cTn) assays in elderly patients, since elevated levels with sensitive cTn assays were reported in 20% of elderly patients without acute myocardial infarction (AMI).

Methods and results In this multi-centre study, we included 1098 consecutive patients presenting with symptoms suggestive of AMI, 406 (37%) were >70 years old. Measurement of three investigational sensitive cTn assays [Roche high-sensitive cTnT (hs-cTnT), Siemens cTnI-Ultra, and Abbott-Architect cTnI) and the standard assay (Roche cTnT) was performed in a blinded fashion. The final diagnosis was adjudicated by two independent cardiologists. Acute myocardial infarction was the adjudicated final diagnosis in 24% of elderly patients. Among elderly patients without AMI, baseline cTn levels were elevated above the 99th percentile in 51% with Roche hs-cTnT, in 17% with Siemens TnI-Ultra, and 13% with Abbott-Architect cTnI. The diagnostic accuracy as quantified by the area under the receiver operating characteristic (ROC) curve (AUC) was significantly greater for the sensitive cTn assays compared with the standard assay (AUC for Roche hs-cTnT, 0.94; Siemens cTnI-Ultra, 0.95; and Abbott-Architect cTnI, 0.95 vs. AUC for the standard assay, 0.90; P < 0.05 for comparisons). The best cut-offs for the sensitive cTn-assays determined by the ROC-curve in elderly patients differed clearly from those in younger patients. Furthermore, the prognostic value regarding 90-day mortality varied among the sensitive cTn assays.

Conclusion Sensitive cTn assays have high diagnostic accuracy also in the elderly. Mild elevations are common in elderly non-AMI patients, therefore the optimal cut-off levels are substantially higher in elderly as compared with younger patients. Furthermore, sensitive cTn assays yielded different prognostic value (ClinicalTrials.gov number, NCT00470587).

  • Acute coronary syndromes
  • Myocardial infarction
  • New biomarkers
  • Troponin
  • Coronary
  • Artery disease
  • Elderly
  • Prognosis

Introduction

Acute myocardial infarction (AMI) is the leading cause of death and morbidity in the elderly population of Europe and the USA.1 Its rapid and accurate diagnosis is an unmet clinical need. Delayed ‘rule in’ increases morbidity and mortality.2,3 Delayed ‘rule out’ prolongs the time spent in the emergency department (ED), increasing patients’ anxiety and causes enormous costs for the health-care system.4

New sensitive cardiac troponin (cTn) assays, with a lower limit of detection (LoD), below the 99th percentile of a normal reference population, have recently become available in clinical practice.5,6 These assays improved the early diagnosis of an AMI in unselected patients with acute chest pain.7,8 However, their diagnostic accuracy in the elderly is uncertain, as elevated levels of cTn were found in up to 22% of persons living in the community who were 70 years of age or older.911

Therefore, and for other reasons, elderly patients merit particular attention. First, pre-existing heart diseases, risk factors, as well as diseases that potentially cause AMI-like chest pain such as gastritis or pneumonia are more common in the elderly.1214 Thus, they are at increased risk for both: AMI, as well as anxiety related to non-cardiac causes of chest pain. Secondly, the interpretation of the 12-lead electrocardiography (ECG) is particularly challenging in elderly patients due to common pre-existing ECG alterations such as a bundle branch block or a pacemaker. In contrast, new ST-segment-elevation is less common in the elderly compared with younger patients.15 Thirdly, the diagnosis of an AMI is even more challenging in the elderly since they differ from younger patients in perception and presentation of the symptoms.16,17 Fourthly, due to frequently pre-existing heart diseases and renal failure, cTn levels may also be elevated in elderly patients without an AMI.18,19 Fifthly, the elderly are more prone to adverse events related to cardiovascular medication, e.g. anticoagulation, as well as to cardiovascular procedures, e.g. coronary angiography, and percutaneous coronary intervention. In this context, an accurate diagnosis of an AMI is mandatory for therapeutic indications.2022

We therefore examined the diagnostic performance of novel sensitive cTn assays for the early diagnosis of an AMI in patients >70 years of age, presenting with acute chest pain to the ED.

Methods

Study design and population

The Advantageous Predictors of Acute Coronary Syndrome Evaluation (APACE) Study is an ongoing prospective international multi-centre study designed and co-ordinated by the University Hospital Basel. From April 2006 to June 2009, a total of 1247 consecutive patients presenting to the ED with symptoms suggestive of an AMI with an onset or peak within the last 12 h were recruited. Elderly patients were considered those >70 years of age.

The study was carried out according to the principles of the Declaration of Helsinki and approved by the local Ethics Committees. Written informed consent was obtained from all patients. The authors designed the study, gathered, and analysed the data, vouch for the data and analysis, wrote the paper, and made the decision to submit it for publication. The assays were donated by the manufacturers, who had no role in the design of the study, the analysis of the data, the preparation of the manuscript, or the decision to submit for publication.

Routine clinical assessment

All patients underwent an initial clinical assessment that included history taking, a physical examination, 12-lead ECG, continuous ECG monitoring, pulse oximetry, standard blood tests, and chest radiography. Cardiac troponin I or cTnT, CK-MB, and myoglobin were measured at presentation and 6–9 h after presentation or as long as clinically indicated. The precise timing of clinical post-baseline measurements and the treatment of patients were left to the discretion of the attending physician.

Adjudicated final diagnosis

To determine the final diagnosis for each patient, two independent cardiologists reviewed all available medical records—the clinical history, findings on physical examination, and results of laboratory tests (including cTn values obtained at the participating hospitals but not those being assessed as part of this study), radiologic testing, ECG, echocardiography, cardiac exercise test, coronary angiography—from the time of the patient's arrival in the ED to the end of the 90-day follow-up period. When there was disagreement about the diagnosis, cases were reviewed and adjudicated in conjunction with a third cardiologist.

An AMI was defined in accordance with current guidelines.23 In brief, an AMI was diagnosed when there was evidence of myocardial necrosis in association with clinical signs of myocardial ischaemia. Necrosis was diagnosed by a rising and/or falling pattern of the local cTn level, with at least one value above the 99th percentile, at a level of imprecision of <10%.5 The following cTn assays were used for the adjudication of the final diagnosis at the participating hospitals: Abbott Axsym cTnI ADV, Beckmann Coulter Accu cTnI, and Roche cTnT. All three are well-validated current cTn assays with comparable performance in the diagnosis of an AMI.5 Unstable angina was diagnosed when a patient had normal cTn levels but new onset or a deterioration of stable angina, positive cardiac exercise testing, coronary artery stenosis of 70% in subsequent cardiac catheterization. Unstable angina was also diagnosed when the diagnosis during hospitalization was uncertain but AMI or sudden cardiac death within 90 days occurred, that was evidently associated with the symptoms described at prior presentation. Further pre-defined diagnostic categories were cardiac but not coronary symptoms (e.g. perimyocarditis or tachyarrhythmias), and non-cardiac causes. If AMI was ruled out at presentation, but no sufficient diagnostic procedures were performed during follow up to establish a definite final diagnosis, symptoms were classified as chest pain of unknown origin.

Investigational cardiac troponin analysis

Blood samples for determination of four investigational cTn assays (three sensitive: Roche High-Sensitive-cTnT, Siemens cTnI-Ultra,6,24 Abbott cTnI-Architect, and one standard: Roche cTnT25) were collected at the time of the patient's presentation to the ED. Additional samples were collected at 1, 2, 3, and 6 h. Serial sampling was discontinued when the diagnosis of an AMI was certain and treatment required transferring the patient to the catheter laboratory or coronary care unit. After centrifugation, samples were frozen at −80°C until they were assayed in a blinded fashion in two batches in a dedicated core laboratory.

The two Roche assays were performed with the use of the Elecsys 2010 system (Roche Diagnostics): for cTnT (fourth generation), the LoD is 0.01 μg/L, the 99th percentile <0.01 μg/L, and the coefficient of variation of <10% (10% CV) is 0.035 μg/L; for hs-cTnT, the LoD is 0.005 μg/L, the 99th percentile 0.014 μg/L, and the 10% CV is 0.013 μg/L.26 For the Siemens cTnI-Ultra assay, performed with the ADVIA Centaur immunoassay system (Siemens), the LoD is 0.006 μg/L, the 99th percentile 0.04 μg/L, and the 10% CV is 0.03 μg/L, as specified by the manufacturer.6,7,24 For the Abbott-Architect cTnI assay, performed with the Architect system (Abbott Diagnostics), the LoD is 0.01 μg/L, the 99th percentile 0.028 μg/L, and the 10% CV is 0.032 μg/L, as specified by the manufacturer.

Statistical analysis

Continuous variables are presented as means (±SD) or medians [with the inter-quartile range (IQR)], and categorical variables as numbers and percentages. Continuous variables were compared with the use of the Mann–Whitney test and categorical variables with the use of the Pearson χ2 test. Receiver operating characteristic (ROC) curves were constructed to assess the sensitivity and specificity of cTn measurements obtained at specific times with the four assays and to compare their ability to diagnose an AMI. Logistic regression was used to combine cTn levels at presentation with early changes in cTn levels. The comparison of areas under the ROC curves (AUC) was performed as recommended by DeLong et al.27 The best cut-off values were determined by the point farthest from the bisector of the ROC curve. For the analysis of the prognostic value of the sensitive cTn assays, we calculated Kaplan–Meier curves, using the log-rank test for comparisons and Cox regression. All hypothesis testing was two-tailed, and P-values of <0.05 were considered to indicate statistical significance. All statistical analyses were performed with the use of SPSS for Windows, version 15.0 (SPSS), and MedCalc software, version 10.3.0 (MedCalc).

Results

Patient characteristics

Of the 1247 consecutively enrolled patients, measurement of all four investigational cTn assays were obtained at presentation from 1098 patients, of whom 406 (37%) were older than 70 years. Elderly patients differed from those younger than 70 years of age in several of the baseline characteristics (Table 1). The 90-day follow-up was completed in 100% of patients.

View this table:
Table 1

Baseline characteristics

All patientsPatients >70 yearsPatients ≤ 70 yearsP-value*Patients >70 years
(n = 1098)(n = 406)(n = 681)Acute myocardial infarctionP-value
Yes (n = 98)No (n = 308)
Male gender, n (%)756 (67)218 (54)511 (75)<0.00163 (64)155 (50)0.016
Age (years)
 Median64 (51–75)78 (74–82)54 (46–63)<0.00179 (75–85)78 (74–85)0.093
Risk factors, n (%)
 Hypertension693 (63)330 (81)360 (53)<0.00183 (85)247 (80)0.320
 Dyslipidaemia492 (45)210 (52)282 (41)0.00152 (53)158 (51)0.761
 Diabetes217 (20)110 (27)107 (16)<0.00131 (32)79 (26)0.244
 Current smoking265 (24)36 (9)224 (33)<0.00112 (12)24 (8)0.179
 Prior smoking391 (36)154 (38)236 (35)0.23539 (40)115 (37)0.675
History, n (%)
 Prior coronary artery disease401 (37)212 (52)189 (27)<0.00156 (57)156 (51)0.262
 Prior myocardial infarction271 (25)146 (36)125 (18)<0.00142 (43)104 (34)0.102
 Prior revascularization296 (27)140 (35)155 (23)<0.00133 (34)107 (35)0.847
 PCI254 (23)116 (29)137 (20)0.00124 (25)92 (30)0.304
 Surgery111 (10)68 (17)43 (6)<0.00118 (18)50 (16)0.622
 Peripheral artery disease76 (7)53 (13)23 (3)<0.00113 (13)40 (13)0.943
 Impaired kidney function115 (11)90 (22)25 (4)<0.00128 (29)62 (20)0.080
Prior stroke64 (6)51 (13)13 (2)<0.00117 (17)34 (11)0.101
All patientsPatients >70 yearsPatients ≤70 yearsP-value*Patients >70 years
(n = 1098)(n = 406)(n = 692)Acute Myocardial InfarctionP-value
Yes (n = 98)No (n = 308)
Vital status, median (IQR)
 Heart rate (b.p.m.)75 (66–89)75 (64–90)76 (66–89)0.08681 (71–100)73 (62–88)<0.001
 Systolic blood pressure (mmHg)142 (127–160)145 (127–164)142 (127–158)0.097136 (119–161)146 (130–164)0.011
Diastolic blood pressure (mmHg)84 (74–93)79 (70–88)86 (78–95)<0.00178 (70–90)79 (69–88)0.501
 Body mass index26 (24–30)26 (24–29)27 (24–30)0.1426 (24–29)26 (24–29)0.448
ECG, n (%)
 Potential ischaemic ECG changes247 (23)118 (29)129 (19)0.00354 (55)64 (21)<0.001
 ST segment depression131 (12)66 (16)65 (9)0.00134 (35)32 (10)<0.001
 Abnormal Q-wave112 (10)59 (15)53 (8)<0.00114 (14)45 (15)0.937
 Left bundle branch block42 (4)27 (7)15 (2)<0.00114 (14)13 (4)<0.001
 T-wave inversion147 (13)72 (18)75 (11)0.00129 (30)43 (14)<0.001
  • *Comparison of patients >70 years and ≤70 years.

Acute myocardial infarction was the adjudicated final diagnosis in 24% of patients over 70 years when compared with 9% in younger patients (P < 0.001). Other adjudicated diagnoses were unstable angina in 16%, cardiac symptoms from causes other than coronary artery disease in 15%, non-cardiac causes in 34%, and symptoms of unknown origin in 11% of the elderly patients (Table 2).

View this table:
Table 2

Final diagnoses, n (%)

All patientsPatients >70 yearsPatients ≤70 yearsP-value*
(n = 1098)(n = 406)(n = 692)
Acute myocardial infarction173 (16)98 (24)75 (11)<0.001
 ST-segment-elevation41 (4)21 (5)20 (3)0.054
 Non-ST-segment-elevation132 (12)77 (19)55 (8)<0.001
Unstable angina152 (14)66 (16)86 (12)0.076
Cardiac cause, but not CAD146 (13)59 (15)88 (13)0.394
Non-cardiac cause518 (48)140 (34)388 (56)<0.001
Unknown98 (9)43 (11)55 (8)0.138
  • *Comparison of patients >70 years and ≤70 years.

Cardiac troponin levels at presentation

Among the patients, whose final diagnosis was not an AMI, patients over 70 years old had significantly higher baseline levels of all three sensitive cTn compared with younger patients: median levels in patients aged ≥70 were 0.014 μg/L (IQR: 0.009–0.024), with hs-cTnT; 0.01 μg/L (IQR: 0.004–0.025), with cTnI-Ultra; and 0.003 μg/L (IQR: 0–0.011), with Abbott-Architect cTnI; compared with 0.005 μg/L (IQR: 0.003–0.009), with hs-cTnT; 0.004 μg/L (IQR: 0.001–0.011), with cTnI-Ultra; and 0.000 μg/L (IQR: 0.000–0.002) with Abbott-Architect cTnI in younger patients (P < 0.001 for comparisons). Of the total, 51% of the elderly patients with a final diagnosis other than AMI, had elevated baseline levels of hs-cTnT, 17% had elevated baseline levels of the Siemens cTnI-Ultra, and 13% had elevated baseline levels of Abbott-Architect cTnI above the 99th percentile. Among younger patients, the percentages were significantly smaller with the sensitive cTn assays (14, 8, and 7%; P < 0.001, P = 0.002, and P = 0.004, respectively; Figure 1).

Figure 1

Baseline levels of sensitive troponin at presentation. Troponin levels at presentation to the emergency department displayed as multiples of the 99th percentile in patients whose adjudicated final diagnosis was other than acute myocardial infarction. Boxes represent inter-quartile ranges, while whiskers display ranges (without outliers further than 1.5 inter-quartile ranges from the end of the box). Left side: patients, >70 years of age. Right side: patients, age 70 years or younger.

In elderly as well as in younger patients, cTn levels at presentation, as assessed by all the assays, were significantly higher in patients whose final diagnosis was AMI when compared with those with other diagnoses (P < 0.001 for comparisons).

Diagnostic accuracy of cardiac troponin at presentation

In elderly patients, the diagnostic accuracy for an AMI, quantified by the AUC, was significantly higher with the three sensitive cTn assays than that with the standard assay [AUC for Roche hs-cTnT: 0.94; 95% confidence interval (CI): 0.91–0.96; for Siemens cTnI-Ultra: 0.95; 95% CI: 0.93–0.97; and for Abbott-Architect cTnI: 0.95; 95% CI: 0.92–0.97; vs. AUC for the standard assay: 0.90; 95% CI: 0.87–0.93; P = 0.015, P = 0.003, P = 0.006, respectively, for comparisons; Table 3 and Figure 2A]. Overall, among all three sensitive assays, the diagnostic accuracy was similar (P > 0.05). In younger patients, the diagnostic accuracy for an AMI was also significantly higher with the sensitive assays than that with the standard assay (AUC for Roche hs-cTnT: 0.94; 95% CI: 0.91–0.96; for Siemens cTnI-Ultra: 0.95; 95% CI: 0.93–0.97; and for Abbott-Architect cTnI: 0.95; 95% CI: 0.92–0.97; vs. AUC for the standard assay: 0.90; 95% CI: 0.87–0.93; P = 0.015, P = 0.003, P = 0.006, respectively, for comparisons; Table 2 and Figure 2C). Overall, comparing the AUC of the three sensitive assays in elderly vs. younger patients, we found no significant differences (P > 0.4 for all tests).

View this table:
Table 3

Diagnostic performance of cardiac troponin in elderly patients at presentation (n = 406)

Sensitivity (95% CI)Specificity (95% CI)Negative-predictive value (95% CI)Positive-predictive value (95% CI)
Sensitive troponin assays
 Roche high-sensitive troponin T
  Limit of detection (0.005 μg/L)100 (96–100)1 (0–3)100 (40–100)24 (20–29)
  99th percentile (0.014 μg/L)a98 (93–100)49 (44–55)99 (95–100)38 (32–44)
 Siemens troponin I-Ultra
  Limit of detection (0.006 μg/L)99 (94–100)30 (25–35)99 (94–100)31 (26–36)
  99th percentile (0.040 μg/L)a92 (85–96)83 (79–87)97 (94–99)64 (55–72)
 Abbott-architect troponin I
  Limit of detection (0.010 μg/L)94 (87–98)72 (67–78)97 (94–99)52 (45–60)
  99th percentile (0.028 μg/L)89 (81–94)87 (83–91)96 (93–98)69 (60–77)
  10% CV (0.032 μg/L)88 (80–94)88 (84–92)96 (93–98)70 (62–78)
Standard troponin assay
 Roche troponin T fourth generation
  99th percentile (unknown)
  Limit of detection (0.010 μg/L)83 (74–90)90 (86–93)94 (91–97)72 (62–80)
  10% CV (0.035 μg/L)76 (57–79)96 (93–98)93 (89–95)86 (77–93)
  • aCriterion of 10% variation coefficient (10% CV) fulfilled at the 99th percentile.

Figure 2

Diagnostic performance of cardiac troponin at presentation. Receiver operating characteristic curves describing the diagnostic performance of different cardiac troponin assays at presentation for the diagnosis of acute myocardial infarction in patients >70 years of age (top), and in those, aged 70 years or younger (bottom). Left side: presenting within 12 h from chest pain onset. Right side: presenting within <3 h of chest pain onset. (A and C): the calculated best cut-off values for elderly and for younger patients are defined by the point farthest from the bisector of the receiver operating characteristic curves.

Best cut-off determined by receiver operating characteristic curve

The best cut-off value to separate AMIs from non-AMIs determined by ROC analysis in elderly patients was nearly four times the 99th percentile for hs-cTnT (0.054 μg/L), but close to the 99th percentile for both sensitive cTnI assays (0.032 μg/L for Abbott-Architect cTnI and 0.045 μg/L for Siemens cTnI-Ultra; see Table 4 and Figure 2A).

View this table:
Table 4

Diagnostic performance of sensitive troponin assays in elderly patients (>70 years old) at the best cut-off determined by the receiver operating characteristic curve (95% confidence interval)

Best cut-off ROCBest cut-off ROC99th percentileP-value*
Sensitive troponin assays
 Roche high-sensitive troponin T, 99th percentile (0.014 μg/L)0.054 μg/LSensitivity79 (69–86)98 (92–100)<0.001
Specificity96 (93–98)49 (44–55)<0.001
Negative-predictive value93 (90–96)99 (95–100)0.042
Positive-predictive value86 (77–92)38 (32–44)<0.001
 Siemens troponin I-Ultra, 99th percentile (0.040 μg/L)0.045 μg/LSensitivity92 (85–96)92 (85–96)1
Specificity88 (84–91)83 (79–87)0.458
Negative-predictive value97 (94–99)97 (94–99)0.997
Positive-predictive value71 (62–79)64 (55–72)0.643
 Abbott-Architect Troponin I, 99th percentile (0.028 μg/L)0.032 μg/LSensitivity88 (80–94)89 (81–94)0.976
Specificity88 (84–91)87 (83–91)0.934
Negative-predictive value96 (93–98)96 (93–98)0.984
Positive-predictive value70 (62 –78)69 (60–77)0.970
  • *Comparisons among elderly patients.

In contrast, the best cut-off value to separate AMIs from non-AMIs in younger patients was close to the 99th percentile for hs-cTnT (0.017 μg/L) and for cTnI-Ultra (0.039 μg/L), but less than a third of the 99th percentile for Abbott-Architect cTnI (0.008 μg/L; see Table 5 and Figure 2C).

View this table:
Table 5

Diagnostic performance of sensitive troponin assays in younger patients (70 years or younger) at the best cut-off determined by the receiver operating characteristic curve (95% confidence interval)

Best cut-off ROCBest cut-off ROC99th percentileP-value*
Sensitive troponin assays
 Roche high-sensitive troponin T, 99th percentile (0.014 μg/L)0.017 μg/LSensitivity88 (78–94)88 (78–94)1
Specificity90 (87–92)86 (83–89)0.215
Negative-predictive value98 (97–99)98 (97–99)0.997
Positive-predictive value51 (42–60)44 (36–52)0.527
 Siemens troponin I-Ultra, 99th percentile (0.040 μg/L)0.039 μg/LSensitivity87 (77–93)87 (77–93)1
Specificity92 (90–94)92 (89–94)0.979
Negative-predictive value98 (97–99)98 (97–99)1
Positive-predictive value57 (47–66)56 (47–72)0.989
 Abbott-Architect troponin I, 99th percentile (0.028 μg/l)0.008 μg/LSensitivity91 (82–96)79 (68–87)0.125
Specificity85 (82–88)93 (91–95)<0.001
Negative-predictive value99 (97–99)97 (96–98)0.256
Positive-predictive value43 (35–51)57 (47–67)0.065
  • *Comparisons among younger patients.

Cardiac troponin levels in patients with recent onset of chest pain

The accuracy of the sensitive cTn assays was also high among patients with recent onset of chest pain (Figure 2B and D, Figure 3, and Supplementary material online, Table S5A).

Figure 3

Diagnostic accuracy at presentation according to chest pain onset. Area under the receiver operating characteristic curves for the different cardiac troponin assays at presentation in the diagnosis of acute myocardial infarction according to the time since chest pain onset. (A): patients, >70 years of age. (B): patients, age 70 years or younger.

Diagnostic performance at the 99th percentile in elderly and younger patients

Overall, at the 99th percentile all cTn assays showed higher sensitivity in elderly when compared with younger patients. The increase in sensitivity in elderly patients was associated with a decrease in specificity, which was particularly pronounced with hs-cTnT (49% in the elderly vs. 86% in younger patients; P < 0.001; see Table 6).

View this table:
Table 6

Diagnostic performance of sensitive troponin assays at the 99th percentile; at 10% coefficient of variation for the standard assay (95% confidence interval)

>70 years≤70 yearsP-value*
Sensitive troponin assays
 Roche high-sensitive troponin T, 99th percentile (0.014 μg/L)Sensitivity98 (92–100)88 (78–94)0.019
Specificity49 (44–55)86 (83–89)<0.001
Negative-predictive value99 (95–100)98 (97–99)0.984
Positive-predictive value38 (32–44)44 (36–52)0.288
 Siemens troponin I-Ultra, 99th percentile (0.040 μg/L)Sensitivity92 (85–96)87 (77–93)0.394
Specificity83 (79–87)92 (89–94)<0.001
Negative-predictive value97 (94–99)98 (97–99)0.370
Positive-predictive value64 (55–72)56 (47–65)0.253
 Abbott-Architect troponin I, 99th percentile (0.028 μg/L)Sensitivity89 (81–94)79 (68–87)0.109
Specificity87 (83–91)93 (91–95)0.008
Negative-predictive value96 (93–98)97 (96–98)0.481
Positive-predictive value69 (60–77)57 (47–67)0.088
Standard troponin assay
 Roche troponin T fourth generation, 99th percentile (unknown) 10% CV (0.035 μg/L)Sensitivity76 (57–79)59 (47–70)0.028
Specificity96 (93–98)98 (97–99)0.135
Negative-predictive value93 (89–95)95 (93–97)0.152
Positive-predictive value86 (77–93)79 (66–88)0.351
  • *Comparison of patients >70 years and ≤70 years.

Serial cardiac troponin levels

During serial sampling, the AUC for all cTn assays increased in patients over 70 years (Supplementary material online, Table S5B). Absolute values of changes in cTn levels from presentation to 1 and 2 h had similar diagnostic accuracy as the baseline cTn levels. The combination of baseline levels plus early changes did slightly improve the performance of the baseline level for all cTn assays. This increase in accuracy was statistically significant for the hs-cTnT assay and the Siemens cTnI-Ultra assay with the combination of the baseline levels and the change already within the first hour after presentation (P = 0.013 and P = 0.017, respectively). For the Abbott-Architect cTnI assay, the improvement in accuracy was only significant with the combination of the baseline level and the change within the first 2 h after presentation (P = 0.014; Supplementary material online, Table S5C). The diagnostic performance of the standard cTn was significantly improved by the measurement of early changes in cTn levels after 2 h as well as by the combination of baseline levels plus early changes at 1 h and at 2 h (P < 0.05; see Supplementary material online, Table S5C).

Cardiac troponin in the diagnosis of unstable angina and acute coronary syndromes

In elderly patients, the sensitive cTn assays had very poor accuracy for differentiating unstable angina from non-cardiac causes of chest pain (AUC for Roche hs-cTnT: 0.62; 95% CI: 0.55–0.69; for Siemens cTnI-Ultra: 0.58; 95% CI: 0.51–0.65; for Abbott-Architect cTnI: 0.59; 95% CI: 0.52–0.66). The diagnostic accuracy for differentiating unstable angina from non-cardiac causes of chest pain was significantly higher in younger patients than in elderly with Roche hs-cTnT (AUC: 0.74; 95% CI: 0.69–0.80; P = 0.019 for comparison). Within the group of younger patients and within the group of elderly patients the diagnostic accuracy of all sensitive cTn assays was similarly low (P > 0.2 for comparisons). However, the diagnostic performance among sensitive assays was still significantly greater than that of the standard assay (AUC for the standard assay 0.52 in elderly as well as in younger patients, P < 0.001 for comparisons of the standard assay with the sensitive assays among elderly and younger patients).

In elderly patients, for the diagnosis of acute coronary syndromes (AMI or unstable angina), the negative-predictive value of a negative assay result, defined as a value below the 99th percentile with a level of imprecision of <10%, was 64% (57–71%) for Roche hs-TnT, 65% (60–71%) for Siemens cTnI-Ultra, 64% (58–70%) for Abbott-Architect cTnI, and 61% (56–66%) for the standard assay. In younger patients, the negative-predictive value was 86% (83–89%), 86% (82–88%), 84% (81–87%), and 82% (78–85%), respectively.

Prognostic value of sensitive cardiac troponin assays

The 90-day follow-up was completed in 100% of patients. Elevated levels of hs-cTnT and Siemens cTnI-Ultra above the 99th percentile at presentation in patients who had a final diagnosis other than AMI strongly predicted death within 90 days, while elevated levels of Abbott-Architect cTnI had no predictive value (Figure 4). Accordingly, in a Cox regression model, adjusted for age >70 years, hs-cTnT, and Siemens TnI-Ultra levels above the 99th percentile represented independent predictors for death within 90 days among patients whose final diagnosis was other than AMI (HR: 4.7; 95% CI: 1.2–18.6; P = 0.027 for Roche hs-cTnT; and HR: 3.6; 95% CI: 1.2–11.0; P = 0.027 for Siemens cTnI-Ultra). Regarding AMI within 90 days, none of the sensitive cTn assays had predictive value (all P > 0.05).

Figure 4

Prognostic impact of elevations in cardiac troponin in patients with final diagnosis other than acute myocardial infarction. Kaplan–Meier curves showing mortality within 90 days in patients whose final diagnosis was other than acute myocardial infarction, comparing patients with levels of sensitive troponins above (green lines) and below (blue lines) the 99th percentile.

Discussion

This prospective multi-centre study involving consecutive patients >70 years of age examined the diagnostic performance of new sensitive cTn assays for the early diagnosis of AMI. We report seven major findings that have important clinical implications for ‘rule in’ and ‘rule out’ AMI:

First, among elderly patients with a final diagnosis other than AMI, the percentage of patients with a baseline level of sensitive cTnI and hs-cTnT above the 99th percentile was high and differed largely among the three sensitive cTn assays, ranging from 13 to 51%. Clinically, the high incidence of elevated cTn levels in the elderly challenges the application of the 99th percentile as the decision limit for the diagnosis of AMI as suggested in current guidelines. Careful clinical assessment is essential to separate AMIs from a variety of acute and chronic disorders also associated with low-level myocardial necrosis, especially in elderly patients, since rhythm disorders, heart failure, and impaired renal function are common in them.19

In addition, the substantial difference between the sensitive cTnI assays and the hs-TnT assay, regarding the incidence of cTn levels above the 99th percentile in patients without AMI, might indicate a variation of release of cTnI and cTnT in these non-AMI settings. The hypothesis of a difference in the release between cTnI and cTnT is supported by recent data indicating a higher prevalence of hs-cTnT levels above the 99th percentile than of cTnI levels above the 99th percentile in elderly patients without symptoms or signs of AMI.9,10

Secondly, despite the above findings, the diagnostic accuracy was similarly high already at presentation for all three sensitive cTn assays in elderly (AUC: 0.94–0.95) as well as in younger patients (AUC: 0.94–0.96) and significantly higher than that of the standard assay.

Thirdly, although the AUC of sensitive cTn assays did not differ significantly comparing elderly and younger patients, the best cut-off values to separate AMIs from non-AMIs varied substantially. Among elderly patients, the best cut-off values to separate AMIs from non-AMIs were around the 99th percentile with the sensitive cTnI assays, but nearly four times the 99th percentile with hs-cTnT. In contrast, the best cut-off values in younger patients were close to the 99th percentile for hs-cTnT and for cTnI-Ultra, but less than a third of the 99th percentile for Abbott-Architect cTnI. These findings highlight the clinical need to develop test-specific algorithms that fine tune the application of these novel tests for the diagnosis of patients with acute chest pain.9,11

Fourthly, the accuracy of sensitive cTn assays was also very high among elderly patients with a recent onset of chest pain. Improvement in the early diagnosis of AMI offers the opportunity to minimize myocardial damage by extending early treatment options to AMI patients without ST-segment-elevation.1,2 Our findings highlight that the ‘rule in’ benefits of sensitive cTn assays can be confirmed in the high-risk group of elderly patients.1,2,23 The cost savings associated with increased sensitivity already at presentation might be substantial.4 On the other hand, the lower specificity increases the incidence of patients with elevated cTn levels not having an AMI, which may trigger additional costly investigations to identify the cause of myocardial necrosis. Detailed cost analyses need to be performed to define the net economic effect of the clinical application of the sensitive cTn assays.28

Furthermore, the distinction between an AMI and other medical conditions associated with cTn elevations including myocarditis, tako-tsubo cardiomyopathy, and heart failure will still require full clinical evaluation. Also, the ECG remains an indispensable tool to immediately identify patients with ST-segment-elevation AMI.26

Fifthly, the overall diagnostic accuracy of the sensitive assays at the 99th percentile, quantified by the AUC for the sensitive assays did not differ significantly in elderly and younger patients. However, regarding the diagnostic parameters in detail, the sensitivity and specificity of the sensitive cTn assays diverged in the elderly compared with younger patients, especially with hs-TnT. All cTn assays showed higher sensitivity but lower specificity in elderly when compared with younger patients, reflecting the higher incidence of baseline levels above the 99th percentile in elderly patients with a final diagnosis other than AMI.

Sixthly, sensitive cTn assays seemed not helpful in the diagnosis of unstable angina in elderly patients and had only little value in younger patients. Further research is necessary to identify biomarkers that reliably detect myocardial ischaemia without necrosis.8,29

Seventhly, elevated levels of Roche hs-cTnT and Siemens TnI Ultra in patients with a final diagnosis other than AMI were strongly related to all-cause mortality within 90 days. This association was not seen for the Abbott-Architect cTnI. Further investigations are required to specify the prognostic capabilities of sensitive cTn assays for short- and long-term prognosis and to illuminate potential explanations for differences among sensitive cTn assays and outcomes. Our findings extend the results of previous studies, investigating the long-term mortality of apparently healthy subjects with elevated levels measured with sensitive cTnI assays.11,30 Further, we found that none of the sensitive cTnI assays predicted an AMI within 90 days.11,30

The following limitations of the current study merit consideration. First, we evaluated three sensitive cTn assays. We hypothesize that our findings can be generalized to other cTn assays with similar sensitivity and precision. However, additional studies need to confirm this hypothesis. Secondly, in this ongoing prospective study, the subgroup analysis of patients aged >70 years was not pre-defined at the time of the initial protocol written in 2005. It was added while we were still blinded to the results in 2008,8 with regard to recent investigations, showing that up to 21% of elderly, presumably healthy patients had elevated levels of sensitive cTn.911 Thirdly, this observational study cannot quantify exactly the clinical benefit associated with the increase in early diagnostic accuracy. To add this important information, interventional studies seem warranted. Fourthly, some of the patients with positive sensitive cTn values classified as non-AMIs might have had small AMIs below the decision value of conventional cTn. Presumably, this contributed to the reduced specificity of the sensitive assays. Fifthly, in this study, we found discrepant prognostic performance among the sensitive cTnI assays. Further analyses are necessary to evaluate possible explanations for these differences.

In conclusion, the excellent diagnostic performance of sensitive cTn assays in the early diagnosis of AMI can be confirmed in elderly as well as in younger patients. However, elevated cTn levels are common in elderly patients with diagnoses other than AMI and challenge differential diagnoses and currently recommended cut-off levels in the elderly, particularly with hs-cTnT, highlighting the clinical need to develop test-specific algorithms for patients with acute chest pain.

Funding

The study was supported by research grants from the Swiss National Science Foundation (PP00B-102853), the Swiss Heart Foundation, Abbott, Roche, Siemens, and the Department of Internal Medicine, University Hospital Basel.

We disclose that C.M. has received research support from the Swiss National Science Foundation (PP00B-102853), the Swiss Heart Foundation, the Novartis Foundation, the Krokus Foundation, Abbott, Astra Zeneca, Biosite, Brahms, Nanosphere, Roche, Siemens, and the Department of Internal Medicine, University Hospital Basel, as well as speaker honoraria from Abbott, Biosite, Brahms, Roche, and Siemens.

Conflict of interest: none declared.

Acknowledgements

We are indebted to the patients who participated in the study and to the ED staff as well as the laboratory technicians for their most valuable efforts, and we thank Dr. C. Schindler for expert statistical advice.

References

View Abstract