| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Articles |
1
Hennepin County Medical Center, Department of Laboratory Medicine and Pathology, Minneapolis, MN 55415.
2
Medical College of Virginia, Departments of Pathology
and Medicine, Richmond, VA 23225.
3
Mayday University Hospital, Surrey CR7 7YE, United
Kingdom.
4
Alameda County Medical Center–Highland Campus,
Department of Emergency Medicine, Oakland, CA 94602.
a Address correspondence to this author at: Hennepin County Medical Center, Clinical Laboratories 812, 701 Park Ave., Minneapolis, MN 55415. Fax 612-904-4229; e-mail fred.apple{at}co.hennepin.mn.us
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Methods: This study evaluated the clinical diagnostic sensitivity and specificity of the First Medical Cardiac Test device operated by nursing and laboratory personnel that simultaneously measures cardiac troponin I (cTnI), creatine kinase (CK) MB, myoglobin, and total CK on the Alpha Dx analyzer in whole blood for detection of MI. Over a 6-month period, 369 patients initially presenting to the emergency department with chest pain were evaluated for MI using modified WHO criteria. Eighty-nine patients (24%) were diagnosed with MI.
Results: In whole blood samples collected at admission and at 3- to 6-h intervals over 24 h, ROC curve-determined MI decision limits were as follows: cTnI, 0.4 µg/L; CKMB, 7.0 µg/L; myoglobin, 180 µg/L; total CK, 190 µg/L. Based on peak concentrations within 24 h after presentation, the following sensitivities (± 95% confidence intervals) were found: cTnI, 93% ± 5.5%; myoglobin, 81% ± 9.7%; CKMB, 90% ± 6.3%; total CK, 86% ± 7.5%. Sensitivities were maximal at >90% for both cTnI and CKMB at >12 h in MI patients, without differences between ST-segment elevation and non-ST-segment elevation MI patients.
Conclusions: The First Medical point-of-care device provides cardiac marker assays that can be used by laboratories and clinicians in a variety of hospital settings for ruling in and ruling out MI.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Several acute coronary syndrome patient groups with varying clinical needs are targets for implementation of POC testing of markers of myocardial injury, including patients with acute MI (ST-segment elevation and depression, Q-wave and non-Q-wave or non-ST-segment elevation) and patients with noncardiac chest pain. At least one report has documented that a fast-track POC marker testing protocol was successful for a rapid rule-out protocol with cost reductions (10). The purpose of this study was to clinically evaluate the diagnostic sensitivity and specificity of the First Medical (FM) Cardiac Panel test device (11) for the quantitative, simultaneous measurement of myoglobin, creatine kinase MB (CKMB) mass, cTnI, and total CK mass in whole blood in patients presenting with ischemic chest pain to rule in or rule out MI.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
patient samples
Specimens from 166 apparently healthy individuals (68 males and 98
females) were used to estimate the reference interval for all four
markers evaluated on the FM Alpha Dx system, including myoglobin, CKMB
mass, cTnI, and total CK mass. None of the healthy subjects had
evidence of cardiac, renal, or skeletal muscle disease. The median ages
for males and females were 28 years (range, 20–51 years) and 37 years
(range, 20–62 years), respectively. Serial whole blood (EDTA)
specimens (n = 1550) from 369 acute coronary syndrome patients
(211 males and 158 females) presenting with chest pain suggestive of
myocardial ischemia were also included, of which 89 had acute MI. The
median ages for males and females were 58 years (range, 25–93 years)
and 60 years (range, 30–92 years), respectively. The diagnosis of
acute MI was made according to modified WHO criteria in which two of
three of the following were present: chest pain duration >20 min,
ST-segment elevation on the electrocardiogram, or increased cardiac
markers (12)(13)(14). The WHO criteria were considered modified
because at least one site used cTnI as their routine biomarker
(13)(14). An increased cardiac marker was
defined as any one of the serial specimens with an increased
concentration above the diagnostic decision cutoff noted below.
A minimum of three serial specimens were collected, at admission and at approximately 3- to 6-h intervals to 24 h, depending on each hospital’s protocol. Whole blood specimens were analyzed within 30 min of collection on the Alpha Dx system. Paired serum specimens (frozen at below -20 °C) were analyzed for each cardiac marker for comparison in the central laboratory at each of the four participating institutions on the following analyzers (diagnostic decision cutoffs): Dade Stratus II myoglobin (110 µg/L); Dade Stratus II CKMB (7.0 µg/L); Dade Stratus II cTnI (1.5 µg/L) or Beckman Access cTnI (0.15 µg/L); Ortho Clinical Diagnostics Vitros total CK (men, 300 U/L; women, 200 U/L).
evaluation of cutoff concentrations
The 97.5th percentiles for the apparently healthy individuals and
the non-MI patients were calculated. To determine the diagnostic
cutoffs for acute MI patients, we performed ROC curve analysis for the
Alpha Dx assays. The relationship between each concentration measured
by the Alpha Dx vs the comparative device was determined using
specimens whose Alpha Dx concentrations fell within the dynamic range
of each assay.
analytical performance
The minimum detectable concentration for each FM assay was defined
as the analyte concentration corresponding to the mean + 2 SD from the
response of 20 replicates of the zero calibrator. Imprecision was
studied across sites, using frozen serum pools at two concentrations,
over 20 days and with three lots of FM test disks and a single
calibration per Alpha Dx instrument.
Alpha Dx system
The Alpha Dx System, a fluorescence immunoassay platform that
integrates automated solid-phase sandwich immunoassay capabilities with
fluorescence detection, provides quantitative measurement of four
cardiac markers: myoglobin, total CK mass, CKMB mass, and cTnI
(11). The system consists of test disks, safe-T-coupler
units, a fluid cassette, and a calibration diskette. The test disk
contains all required test-specific reagents, along with bilevel
quality controls for each test in stabilized dry form. The
safe-T-coupler unit is used to load a primary blood collection tube
into the analyzer, where 80 µL of sample is automatically metered
from the sample tube into the sample chamber, through the
safe-T-coupler unit, where it is mixed with the fluorescein-labeled
antibodies. A minimum of 1.0 mL of blood is needed for analysis. The
fluid cassette contains a stabilized, buffered detergent solution that
is used for both the rehydration of dried reagents and protocol washes.
The calibration diskette contains lot-specific calibration information,
limits for system quality control, and expiration dating for each lot
of test disks. The system is designed for STAT operation, with assay
results available in <20 min. Biohazard waste and assay fluids are
safely contained within the test disc and safe-T-coupler unit. The test
disc built-in bilevel controls are processed in parallel with the
patient sample. External quality-control procedures are not required.
In a typical sandwich fluorescent immunoassay in the test device, antigens are detected by binding to two distinct antibodies, one immobilized on a solid phase and the other coupled to a fluorescent label. The test disc uses three antibodies: a solid-phase antibody, a hapten (fluorescein)-labeled antibody, and a fluorescently labeled anti-hapten (fluorescein) antibody. The three-antibody system provides enhanced signal amplification and improved assay sensitivity. The test disc consists of three zones, one for the test sample and two for the quality controls. In addition, each disc contains a hematocrit chamber and a waste ring. All zones have four spatially distinct but fluid-connected features that function as analyte capture sites within a shared reaction chamber. In all four assays, fluorescence intensity is proportional to the amount of bound label and, therefore, to the concentration of analyte in the sample, measured after excitation by a 640 nm diode laser in the analyzer fluorometer. The packed cell volume, obtained form the hematocrit procedure, is used to convert the measured whole blood sample result to a matched serum sample. All reactions are temperature controlled at 37 °C.
statistical comparison
To compare the Alpha Dx System with each laboratory’s established
devices, Passing-Bablock linear regression analysis was performed on
single tests from each sample over the dynamic range of both assays.
Values below the minimum detectable concentration of either assay were
excluded from the data analysis. ROC curves were analyzed for each
assay (15). Diagnostic sensitivity and specificity data are
presented with 95% confidence intervals (95% CIs) and compared
between assays by the McNemar test. All statistical tests were
two-tailed, with significance set at P <0.05. SAS Ver. 6.09
software was used for all statistical analyses. The data presented here
are the basis of materials submitted to the Food and Drug
Administration for 510K approval of the four cardiac markers on the FM
Alpha Dx instrument.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, the total imprecision (CV) of the Alpha Dx assays was
4.0–8.8% for the two frozen serum controls. The lower limits of
detection were as follows: myoglobin, 5 µg/L; CK, 10 µg/L; CKMB,
0.5 µg/L; cTnI, 0.09 µg/L.
|
Regression analysis statistics are demonstrated in Fig. 1
. Samples were from paired cardiac marker measurements between
the Alpha Dx (whole blood) and the comparative devices (serum) in
subsets from the total specimens (n = 1550) collected from 276
suspected acute MI patients and 134 healthy individuals over the
dynamic range of each assay. Comparison of the Alpha Dx cTnI with the
Stratus cTnI assay (y = 0.18x + 0.56;
r = 0.92; n = 226) and Access cTnI assay
(y = 1.44x + 0.96; r = 0.88;
n = 244) showed substantially different slopes: 0.18 and 1.44,
respectively. Bias plots revealed a proportional (38%) increasing
positive bias across the range of concentrations tested for by the
Stratus II cTnI compared with the Alpha Dx cTnI. Bias plots for total
CK, myoglobin, and CKMB demonstrated random scatter at the upper end of
linearity for each marker between assays (data not shown). A
representative time-vs-cTnI concentration profile for one MI patient,
shown in Fig. 2
, demonstrates the different absolute and relative
concentrations that are observed among the different cTnI assays.
|
|
clinical performance
Fig. 3
shows cTnI, CKMB, myoglobin, and total CK ROC curves determined
using peak concentrations over the 24 h after presentation from
the 369 ischemic chest pain patients. The sensitivities and
specificities (95% CIs), respectively, at decision cutpoints were as
follows: 0.4 µg/L cTnI, 93% (87.5–98.5%) and 94% (91.3–96.7%);
7.0 µg/L CKMB, 90% (82.7–96.3%) and 90% (86.6–93.4%); 100
µg/L myoglobin, 81% (71.5–90.5%) and 81% (77.3–84.7%); 190
µg/L total CK, 86% (78.5–93.5%) and 86% (81.9–90.1%). There was
a 92% concordance between paired results for diagnosis between the FM
cTnI and FM CKMB compared with the Dade cTnI and Dade CKMB assays,
respectively. The area under the cTnI ROC curve (0.919) was
significantly larger (P <0.05) compared with CKMB (0.890),
myoglobin (0.541), and total CK (0.856). There were no statistical
differences for the areas under the ROC curves for the FM cTnI assay
compared with the Dade cTnI assay (0.925), whose cutpoint of 1.5 µg/L
gave a 93% (95% CI, 87.4–98.6%) sensitivity and 93% (90.1–95.9%)
specificity.
|
For sensitivity and specificity calculations, McNemar
2 analysis showed no statistical differences
in concordances between the FM CKMB and Stratus CKMB assays, the FM
cTnI and Stratus cTnI assays, or the FM myoglobin and Stratus myoglobin
assays (data not shown). Based on peak values over the 24 h after
onset of symptoms, the sensitivities (± SD) of all four markers on the
FM device were not statistically different in MI patients with
ST-segment elevation (n = 60) and non-ST-segment elevation (n
= 29) infarctions, respectively: myoglobin, 76% ± 10% and 58% ±
18%; CK, 87% ± 8% and 80% ± 15%; cTnI, 93% ± 5% and 90% ±
9%; CKMB, 90% ± 7% and 83% ± 14%. cTnI demonstrated >90%
sensitivity compared with 77–90% sensitivity for myoglobin, total CK,
and CKMB. Fig. 4
shows the clinical sensitivity of the FM cTnI assay for
ST-segment elevation MI patients (n = 60) compared with
non-ST-segment elevation MI patients (n = 29) based on time
intervals from onset of symptoms. Myoglobin was the most sensitive in
the early 0–4 h time frame (61%), compared with total CK (43%), CKMB
(50%), and cTnI (29%). At 5–11 h, the sensitivity was 90% for CKMB
and 75% for cTnI. At 12–23 h, the sensitivities for both CKMB and
cTnI were >90%. Also using peak values over the 24-h period after
onset of chest pain, specificities between ST-segment elevation and
non-ST-segment elevation MIs were not statistically different from the
overall specificities noted above.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
), with maximum
sensitivity 12–23 h after presentation at >90% in both MI groups
when cTnI was used. Although not statistically significant,
sensitivities for CKMB in both groups were lower. It should be noted
that MI patients who may have received thrombolytic therapy from the
ST-segment elevation MI group were not separated for analysis.
cTnI, a cardiac-specific protein that is challenging CKMB as a new
marker for MI detection
(1)(13)(14)(16),
demonstrated concentration differences when the FM cTnI assay was
compared with the Dade Stratus II and Beckman Access assays, as shown
in Figs. 1
and 2
. The two- to sevenfold concentration differences in
assays substantiate the lack of appropriate standardization between
assay systems (16)(17)(18). This underlines the need for
individual laboratories and hospitals to establish their own decision
cutpoints for each cTnI assay used for medical decisions if appropriate
peer-reviewed studies are not available in the literature.
In addition to CKMB mass and cTnI, myoglobin and total CK mass
concentrations are also obtained on the Alpha Dx panel. Although not
cardiac specific, myoglobin was shown to be the most sensitive marker
early after presentation in chest pain, confirming what is established
in the literature (19). This study did not address the role
of myoglobin as a predictor for ruling out MI (20). Unique
to this assay system is the measurement of total CK mass. Similar to
the myoglobin, CKMB, and cTnI assays, the total CK mass assay
demonstrated acceptable concordance with the total CK activity assays
(Fig. 1
). The range in values between assays found in the regression
analysis was likely because the comparison assay measures activity
(Vitros), whereas the Alpha Dx is a mass immunoassay. However, it is
not clear what clinical utility the addition of this marker will add,
if any, to the rule-in/rule-out process for MI. Furthermore, the range
in CKMB values found in the regression analysis (Fig. 1
) was likely
attributable to the lack of standardization between the two mass
assays. All four assays on the Alpha Dx panel were found to be
analytically acceptable for use in clinical laboratory testing.
Although cardiac markers are often not necessary for the diagnosis of
acute MI in patients presenting with an ST-segment elevation MI,
markers are essential criteria for the diagnosis of non-ST-segment
elevation MI (21). The findings in our study showed no
statistical differences in sensitivity for any of the four markers
between patients with ST-segment elevation and non-ST-segment elevation
infarctions based on peak concentrations within 24 h after
presentation or in serial determinations for 0–4, 5–11, or 12–23 h
groupings after presentation. However, there was a trend toward a lower
clinical sensitivity in the non-ST-segment elevation infarction
patients, as shown in Fig. 4
. This may be influenced by the central
laboratory assay cutpoints used for each assay. We purposely separated
out the clinical sensitivities for the cardiac markers by ST-segment
elevation and non-ST-segment elevation MI to emphasize the point that
markers such as cTnI and CKMB, which are heavily relied on in the
diagnostic criteria in the non-ST-segment elevation MI diagnosis, are
diagnostically similar to ST-segment elevation MI in clinical
utilization. Finally, no long- or short-term prognostic information was
obtained in this study for either the MI patients or the patients with
increased cTnI concentrations (minor myocardial damage) who ruled out
for MI because it was not part of our study design.
In conclusion, the FM Alpha DX Cardiac Panel provides a platform for the simultaneous, quantitative measurement of total CK mass, CKMB mass, myoglobin, and cTnI in either whole blood or serum in <20 min. This is relevant in medical centers that are unable to support a rapid (<1 h) turnaround time from the central laboratory (1). The diagnostic sensitivity and specificity of each test panel are equivalent to other commercially available assay systems. The FM system provides a POC testing system suitable for use by laboratory or non-laboratory personnel in a coronary care unit, emergency department, or central laboratory setting in both rural hospitals and major medical centers.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
1 Nonstandard abbreviations: MI, myocardial infarction; cTnI, cardiac troponin I; POC, point of care; FM, First Medical; CKMB, creatine kinase MB; and CI, confidence interval.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  F. S. Apple, R. Ler, A. Y. Chung, M. J. Berger, and M. M. Murakami Point-of-Care i-STAT Cardiac Troponin I for Assessment of Patients with Symptoms Suggestive of Acute Coronary Syndrome, Clin. Chem., February 1, 2006; 52(2): 322 - 325. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Panteghini Acute Coronary Syndrome: Biochemical Strategies in the Troponin Era Chest, October 1, 2002; 122(4): 1428 - 1435. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. S. Apple, M. M. Murakami, R. L. Jesse, M. A. Levitt, A. K. Berger, L. A. Pearce, and P. Collinson Near-Bedside Whole-Blood Cardiac Troponin I Assay for Risk Assessment of Patients with Acute Coronary Syndromes Clin. Chem., October 1, 2002; 48(10): 1784 - 1787. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) vs Stratus cTnI (
) and Access
cTnI (
) measured in serial specimens from a representative acute MI
patient collected from admission.
