Clinical Chemistry 43: 1635-1640, 1997;
(Clinical Chemistry. 1997;43:1635-1640.)
© 1997 American Association for Clinical Chemistry, Inc.
Multicenter study of Abbott AxSYM® Digoxin II assay and comparison with 6 methods for susceptibility to digoxin-like immunoreactive factors
Hassan M. E. Azzazy1,2,a,
Show-Hong Duh1,
Andrew Maturen3,
Elisabeth Schaller3,
Leslie Shaw4,
Robert Grimaldi5,
Gloria Shock6 and
Robert H. Christenson1,2
1
Departments of Pathology
2
Medical & Research Technology, University of
Maryland School of Medicine, Baltimore, MD 21201.
3
Rush–Presbyterian–St. Lukes Medical Center,
Chicago, IL 60612.
4
Hospital of the University of Pennsylvania,
Philadelphia, PA 19104.
5
New England Medical Center, Boston, MA 02111.
6
Abbott Laboratories, Abbott North Park, IL 60064.
a Address correspondence to this author, at: University of Maryland Medical Center, Clinical Pathology, 22 South Greene St., Baltimore, MD 21201; Fax (410) 328-5880; e-mail hazzazy{at}umabnet.ab.umd.edu
 |
Abstract
|
|---|
Performance characteristics of the Abbott nonpretreatment
AxSYM® Digoxin II assay were evaluated for quantification
of digoxin at four independent sites. Correlation of digoxin
measurements with the Abbott pretreatment AxSYM®, Baxter
Stratus® II, Abbott TDx/TDxFLx II®, Abbott
IMx®, Emit® 2000, and Beckman Synchron
CX® digoxin assays showed acceptable agreement, as
indicated by: slope values >0.84, r >0.90,
y-intercepts for all comparisons at or below the assay
detection limit, and
Sy|x ranging between
7.5% and 15.4% of the average digoxin value. Susceptibility to
interference from digoxin-like immunoreactive factors (DLIFs) was
examined in 233 samples from renal patients, liver disease patients,
cord blood, and third-trimester pregnancies; the AxSYM Digoxin II assay
demonstrated the least DLIFs interference. DLIF susceptibility for four
of the methods was significantly greater (P <0.05) than in
the AxSYM Digoxin II assay; susceptibilities of the Stratus II and Emit
2000 methods were similar to the AxSYM Digoxin II assay.
|
Introduction
|
|---|
Digoxin, a glycosylated steroid-like drug isolated from
foxglove, is a potent cardiac glycoside and is administered to patients
with cardiac arrhythmia and congestive heart failure (1).
Digoxin acts as a reversible inhibitor of the
Na+/K+-ATPase pump, causing inhibition of
cardiac contractions (1). The elimination half-life of
digoxin in healthy test individuals ranges between 26 and 45 h,
but increases dramatically in patients with renal disease
(1). About 25% of digoxin is bound to plasma proteins
(1).
Marked patient variability in response to the same dosage of digoxin,
resulting in unpredictable concentrations in serum, as well as the low
therapeutic index of this drug, both potentiate toxicity. For these
reasons digoxin is considered one of the most frequent causes of drug
toxicity, particularly in the elderly (2), and regular
therapeutic measurements are required to strike the balance between
effective serum concentrations and either toxic or subtherapeutic
values. For life-threatening digoxin intoxication, digoxin-specific Fab
antibody fragments are the treatment of choice (3);
activated charcoal is used for treatment of less-severe cases
(4).
Digoxin-like immunoreactive factors
(DLIFs)1
are reactive substances that can cause false-positive results
and affect the accuracy of digoxin monitoring tests
(5)(6). DLIFs have been reported in neonates
(7), cord blood (8), pregnant women
(8)(9), and patients with renal
(10)(11) or hepatic
(12)(13) disease. Various commercial digoxin
assays show different susceptibilities to DLIFs; in fact, with one
solid-phase RIA, both heparinized blood and urine samples of healthy
volunteers were reported to contain DLIFs (14). Strategies
to reduce DLIF activity include changing the incubation temperature and
time for the immunoassay kits (15).
Assays that require acid pretreatment have also been documented to show
substantial DLIF interference (16). Although previous
digoxin assays from Abbott Labs. have included acid pretreatment, a new
assay requiring no pretreatment, the AxSYM Digoxin II assay, has
recently been developed.
In this multicenter study, we examined the performance of seven
automated digoxin methods, including the new Abbott AxSYM Digoxin II
assay and the susceptibility of each to DLIFs. In addition, we
evaluated the precision, accuracy, and minimum detectable concentration
(MDC) for the AxSYM Digoxin II immunoassay.
|
Materials and Methods
|
|---|
clinical specimens
A total of 798 leftover serum samples submitted for routine
digoxin analysis to the hospital laboratories at the University of
Maryland Medical Center, Rush–Presbyterian–St. Lukes Medical Center,
University of Pennsylvania Hospital, and New England Medical Center
were used. Digoxin concentrations in all these clinical specimens
exceeded 0.3 µg/L.
materials
The following digoxin assays were included: AxSYM Digoxin, AxSYM
Digoxin II, IMx Digoxin, and TDx/TDxFLx Digoxin II, all provided by
Abbott Labs.; Stratus II (Dade Labs.), Synchron CX®
(Beckman Instruments), and Emit® 2000 (Behring) were
purchased for use in this study. All assays were performed according to
the manufacturers' protocols. Controls recommended by each
manufacturer were used to determine the precision of each run.
The manufacturers' reported MDC is 0.3 µg/L for the AxSYM Digoxin,
AxSYM Digoxin II, IMx Digoxin, and Baxter Stratus II digoxin assays.
The TDx/TDxFLx II, Emit 2000, and Synchron CX assays list a MDC of 0.2
µg/L. The digoxin therapeutic range for all methods included was
1.0–2.0 µg/L (1).
AxSYM Digoxin II and IMx Digoxin assays.
Both of these
assays are microparticle enzyme immunoassays, in which the digoxin in
the sample binds to anti-digoxin-coated microparticles; after
separation, digoxin–alkaline phosphatase conjugate binds to the
available sites remaining. The substrate 4-methylumbelliferyl phosphate
is added and the fluorescent product is measured. Neither assay
requires acid pretreatment of samples. Linearity, recovery, and
interferences with the AxSYM Digoxin II assay were determined by the
manufacturer (in the package insert, list no. 5B73).
Stratus II digoxin assay.
In this radial partition
immunoassay, digoxin in the specimen binds to the immobilized
anti-digoxin antibodies; any remaining unbound antibody sites are
occupied by the addition of enzyme-conjugated digoxin. After washing to
remove excess conjugate, added substrate is converted to measurable
fluorescent product.
AxSYM Digoxin and TDx/TDxFLx Digoxin II assays.
Both of
these assays require pretreatment (with sulfosalicylic acid, 45 g/L, in
a solution of 750 mL/L methanol and 250 mL/L water) to deproteinate the
samples. After centrifugation, both systems measure digoxin in the
supernatant by competitive fluorescence polarization immunoassay.
Emit 2000 digoxin assay.
This assay is based on
competition between the digoxin in the sample and that labeled with
recombinant glucose-6-phosphate dehydrogenase (G6PD) for antibody
binding sites. Activity of the G6PD label is decreased upon antibody
binding, and the remaining active G6PD label converts oxidized
NAD+ to NADH, which is measured on a Beckman CX7
instrument.
Beckman Synchron CX digoxin assay.
In this method,
digoxin is measured by a method utilizing ß-galactosidase separated
in two inactive components: an enzyme donor, to which digoxin is
attached, and an enzyme acceptor. Competitive binding of donor/digoxin
to anti-digoxin antibodies blocks donor and acceptor interaction and
inhibits the formation of active enzyme. Quantification is based on the
measurement of enzyme activity. (The Synchron CX digoxin reagent was
reformulated after completion of this study.)
axsym digoxin ii assay evaluation
Imprecision study.
Low, medium, and high controls were
each run in duplicate twice a day for 20 separate days with the AxSYM
Digoxin II assay.
Detection limit.
Ten replicates of the AxSYM
Digoxin II zero calibrator and duplicates of the lowest nonzero
calibrator were measured. A two-point curve was plotted between the
mean analytical signals of the zero calibrator and the lowest nonzero
calibrator. The MDC was determined by extrapolating the concentration
corresponding to 2 SD above the mean zero calibrator signal from this
two-point curve. The MDC reported represents the mean of 4 runs, 1
performed at each of the 4 participating sites.
assessment of digoxin-like immunoreactivity
False-positive digoxin results for each method were determined by
analyzing sera from four different groups of patients susceptible to
DLIF interference (n = 233), none of whom was taking digoxin.
These samples were collected at University of Maryland Medical Center,
University of Pennsylvania Hospital, and New England Medical Center
from women in the third trimester of pregnancy (n = 50), from cord
blood (n = 35), and from patients with liver disease (n = 49)
or renal disease (n = 99). Results exceeding the MDC for each
method were considered positive for DLIFs.
statistical analysis
Within-run, between-run, and total imprecision was determined with
a Nested Analysis of Variance (ANOVA) (17). Linear
regression analysis (least squares method) was used for correlation
studies. The McNemar test (18) was used to compare
DLIF susceptibility of each patient group, for each assay
comparison. The McNemar test was also used for comparing overall
sensitivity to DLIFs of the different assays relative to the AxSYM
Digoxin II; P <0.05 was considered to show significance.
|
Results
|
|---|
The MDC determined for the AxSYM Digoxin II assay was 0.12 µg/L.
Table 1
shows the precision data for this assay. The total CV was <7%
at all concentrations tested.
The comparison data for patients' samples for the AxSYM Digoxin II and
various other methods are shown in Fig. 1
; Table 2
lists the regression parameters for these data. Slopes ranged
from 0.84 to 0.98, all correlation coefficients exceeded 0.90, the
y-intercepts for all comparisons were at or below the
assays' MDC, and variability around the regression curve
(Sy|x) was between 7.5% and 15.4%
of the average digoxin value.
View larger version (34K):
[in this window]
[in a new window]
|
Figure 1. Comparison of AxSYM Digoxin II assay with AxSYM Digoxin,
Beckman Synchron, IMx, Stratus II, Emit 2000, and TDx/TDxFLx digoxin
immunoassays in serum from patients on digoxin therapy; dotted
line = line of identity (x = y).
|
|
Substantial differences were observed among the methods as to
susceptibility to DLIFs in the non-digitalis-treated patients (n =
233) from the four clinical groups, as indicated in Fig. 2
. The highest incidence of DLIFs was observed for the TDx/TDxFLx
(in 75% of cord blood, 12% of renal disease, and 69% of liver
disease samples) and the AxSYM Digoxin assay (in 34% of cord blood,
10% of renal disease, and 36% of liver disease samples). The Synchron
CX assay also showed susceptibility to DLIFs in a substantial
proportion of patients (in 22% of renal disease and 65% of liver
disease samples). Overall, most of the positive results were evident in
liver disease and cord blood samples; only one sample from a woman in
the third trimester of pregnancy had detectable DLIFs. Figs. 2A
and 2F
show that both of the assays requiring acid pretreatment, AxSYM Digoxin
and TDx/TDxFLx Digoxin II, were significantly more susceptible than the
AxSYM Digoxin II assay to DLIFs in cord blood samples. Figs. 2B
and 2C
show that the Synchron and IMx assays were significantly more
susceptible to DLIFs in patients with liver disease than was the AxSYM
Digoxin II assay. Figs. 2D
and 2E
indicate no significant differences
between the AxSYM Digoxin II assay and either the Stratus II or Emit
2000 methods.
View larger version (41K):
[in this window]
[in a new window]
|
Figure 2. Scattergrams of apparent digoxin concentrations obtained
by the seven digoxin immunoassays applied to digoxin-free sera from
cord blood (n = 35), women in third-trimester pregnancy (n =
50), renal disease patients (n = 99), and liver disease patients
(n = 49).
Points enclosed by parentheses indicate digoxin values >4 µg/L.
Horizontal bars indicate the MDC for each assay.
|
|
Table 3
shows the McNemar test analysis of DLIF sensitivity of all
methods for the overall results from the cord blood, pregnant women,
and renal disease and liver disease patients included. There was no
significant difference in DLIF susceptibility for the Stratus II, Emit
2000, and AxSYM Digoxin II methods. The AxSYM digoxin, TDx/TDxFLx, IMx,
and Synchron methods, however, were all significantly more susceptible
to DLIF interference.
|
Discussion
|
|---|
Whereas other evaluation studies have utilized a limited number of
patients' samples from a single population base, this was a
multicenter study and included data from >1000 patients, 800 of whom
were receiving digoxin and 233 who were patients in groups susceptible
to DLIFs. This large, diverse population allowed valid statistical
analysis for detecting differences between the numerous methods and
patients' groups included.
The total CV for the AxSYM Digoxin II assay was <7% at both the
therapeutic and supratherapeutic concentrations. Although this study
showed that the AxSYM Digoxin II method demonstrated generally
acceptable correlation with all other assays examined, potential users
of the AxSYM Digoxin II must be aware that a bias may be expressed in
the results, as indicated by the slopes listed in Table 2
.
The incidence of DLIFs was determined by analyzing sera from cord
blood, pregnant women in the third trimester, and patients with liver
or renal disease—populations reported to be particularly susceptible
to DLIF interference (7)(8)(9)(10)(11)(12)(13). The molecular structure of
digoxin and DLIFs are reported to change after incubation of patient
samples with sulfosalicylic acid (16). The acid
pretreatment step, which could cause removal of sugar moieties and
formation of less-polar compounds, is used in both the AxSYM Digoxin
assay and TDx/TDxFLx Digoxin II assays. In this study both assays
showed high susceptibility to DLIFs, particularly in assays of cord
blood and liver disease samples. Acid-pretreatment-induced structural
changes of digoxin and DLIFs could contribute to the discrepancies
observed between measured digoxin values. The higher susceptibility of
DLIFs by TDx/TDxFLx digoxin II assay is in accordance with results
previously obtained by Gault et al. (19) and Dodds et al.
(20). On the other hand, the AxSYM Digoxin II assay showed
satisfactory correlation with the Stratus II digoxin assay that has
been reported to have minimal DLIF interference (9). The
strategy of removing the acid pretreatment step in the AxSYM Digoxin II
method greatly reduced the susceptibility of this technology to DLIFs.
The AxSYM Digoxin II and Emit 2000 assays also showed less DLIF
susceptibility than the AxSYM Digoxin, TDx/TDxFLx, IMx, and Synchron
assays.
Most recently, in a preliminary study, Jortani et al. (21)
reported that in the presence of DLIFs, the IMx Digoxin assay
artifactually reduced digoxin measurements in serum samples containing
digoxin. Although addressing this specific issue was beyond the scope
of our study, we observed that all slopes for the digoxin comparisons
were <1.0 for the AxSYM Digoxin II comparisons, which may indicate a
slight bias or suppression in some patients. This important issue
should be considered in future digoxin studies if similar results are
reported by others.
We conclude that AxSYM Digoxin II assay showed acceptable
precision, agreed well with other assays, and demonstrated DLIF
interference that is equivalent to state-of-the art assays. It is
important to emphasize, however, that rare occurrences of DLIF
interference may be encountered by users of this assay.
|
Acknowledgments
|
|---|
We thank Abbott Laboratories for funding this study and for
donations of reagents and supplies.
|
Footnotes
|
|---|
1 Nonstandard abbreviations: DLIF(s), digoxin-like immunoreactive factor(s); G6PD, glucose-6-phosphate dehydrogenase; and MDC, minimum detectable concentration. 
|
References
|
|---|
-
Lisalo E. Clinical pharmacokinetics of digoxin. Clin Pharmacokinet 1977;2:1-16.
[ISI][Medline]
[Order article via Infotrieve]
-
Mahdyoon H, Battilana G, Rosman H, Goldstein S, Gheorghiade M. The evolving pattern of digoxin intoxication: observations at a large urban hospital from 1980 to 1988. Am Heart J 1990;120:1189-1194.
[ISI][Medline]
[Order article via Infotrieve]
-
Antman EM, Wenger TL, Butler VP, Haber E, Smith TW. Treatment of 150 cases of life-threatening digitalis intoxication with digoxin-specific Fab antibody fragments. Final report of a multicenter study. Circulation 1990;81:1744-1752.
[Abstract/Free Full Text]
-
Lalonde RL, Deshpande R, Hamilton PP, McLean WM, Greenway DC. Acceleration of digoxin clearance by activated charcoal. Clin Pharmacol Ther 1985;37:367-371.
[ISI][Medline]
[Order article via Infotrieve]
-
Valdes R, Jr. Endogenous digoxin-like immunoreactive factors: impact on digoxin measurements and potential physiological implications. Clin Chem 1985;31:1525-1532.
[Abstract/Free Full Text]
-
Soldin SJ. Digoxin—issues and controversies [Review]. Clin Chem 1986;32:5-12.
[Abstract/Free Full Text]
-
Beyers AD, Spruyt LL, Seifart HI, Kriegler A, Parkin DP, Van Jaarsueld PP. Endogenous immunoreactive digitalis-like substance in neonatal serum and placental extracts. S Afr Med J 1984;65:878-882.
[ISI][Medline]
[Order article via Infotrieve]
-
Diamandis EP, Papanastasiou-Diamandi A, Soldin SJ. Digoxin immunoreactivity in cord and maternal serum and placental extracts. Partial characterization of immunoreactive substances by high performance liquid chromatography and inhibition of Na+, K+-ATPase. Clin Biochem 1985;18:48-55.
[ISI][Medline]
[Order article via Infotrieve]
-
Cook JD, Koch TR, Cook MS, Knoblock EC. Inaccuracies in digoxin measurement. Clin Biochem 1988;21:353-357.
[ISI][Medline]
[Order article via Infotrieve]
-
Graves SW, Brown B, Valdes R. An endogenous digoxin-like substance in patients with renal impairment. Ann Intern Med 1983;99:604-608.
-
Pleasants RA, Gadsden RH, McCormack JP, Piveral K, Sawyer WT. Interference of digoxin-like immunoreactive substances with three digoxin immunoassays in patients with various degrees of renal function. Clin Pharm 1986;5:810-816.
[ISI][Medline]
[Order article via Infotrieve]
-
Nanji AA, Greenway DC. Falsely raised plasma concentrations in liver disease. Br J Med 1985;290:432-433.
-
Pudek MR, Seccombe DW, Humphries KG. Digoxin-like immuno-reactive substance and bile acids in the serum of patients with liver disease [Letter]. Clin Chem 1986;32:2005-2006.
[Free Full Text]
-
Balzan S, Clerico A, del Chicca MG, Montali U, Ghione S. Digoxin-like immunoreactivity in normal human plasma and urine, as detected by a solid-phase radioimmunoassay. Clin Chem 1984;30:450-451.
[Abstract]
-
Yannakou L, Diamandis EP, Souvatzoglou A. Effect of incubation time and temperature on the interference of digoxin-like immunoreactive substances in digoxin immunoassays. Ther Drug Monit 1987;9:461-466.
[ISI][Medline]
[Order article via Infotrieve]
-
Qazzaz HMAM, Goudy S, Miller JJ, Valdes R, Jr. Treatment of human serum with sulfosalicylic acid structurally alters digoxin and endogenous digoxin-like immunoreactive factors. Ther Drug Monit 1995;17:53-59.
[ISI][Medline]
[Order article via Infotrieve]
-
Krinkle RO, Johnson GF. Estimating variance components by nested ANOVA. Burtis CA Ashwood ER eds. Tietz textbook of clinical chemistry 2nd ed. 1986:324-332 WB Saunders Philadelphia. .
-
Woolson RF. Statistical methods for the analysis of biomedical data 1987:205-213 Wiley and Sons New York. .
-
Gault MH, Vasdev S, Longerich L. Higher values for digitalis-like factors with TDx Digoxin II [Letter]. Clin Chem 1986;32:2000-2001.
[Free Full Text]
-
Dodds HM, Norris RLG, Johnson AG, Pond SM. Evaluation and comparison of the TDx, II Stratus, and OPUS digoxin assays. Ther Drug Monit 1995;17:68-74.
[ISI][Medline]
[Order article via Infotrieve]
-
Jortani SA, Miller JJ, Helm RA, Valdes R, Jr. Unexpected suppression of digoxin values caused by DLIFs [Abstract]. Clin Chem 1996;42:S124.
The following articles in journals at HighWire Press have cited this article:
|
|
|
|
 
W. Steimer, C. Muller, and B. Eber
Digoxin Assays: Frequent, Substantial, and Potentially Dangerous Interference by Spironolactone, Canrenone, and Other Steroids
Clin. Chem.,
March 1, 2002;
48(3):
507 - 516.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
