Clinical Chemistry Link to Randox Laboratories Web Site
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Clinical Chemistry 48: 507-516, 2002;
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit an electronic Letter to
the Editor about this paper
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (27)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Steimer, W.
Right arrow Articles by Eber, B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Steimer, W.
Right arrow Articles by Eber, B.
Related Collections
Right arrow Drug Monitoring and Toxicology
(Clinical Chemistry. 2002;48:507-516.)
© 2002 American Association for Clinical Chemistry, Inc.

Digoxin Assays: Frequent, Substantial, and Potentially Dangerous Interference by Spironolactone, Canrenone, and Other Steroids

Werner Steimer1a, Christine Müller1 and Barbara Eber1

1 Institute for Clinical Chemistry and Pathobiochemistry, Munich University of Technology, Klinikum rechts der Isar, D-81675 Munich, Germany.

aAddress correspondence to this author at: Institut für Klinische Chemie und Pathobiochemie, Klinikum rechts der Isar, Technische Universität München, Ismaningerstrasse 22, D-81675 München, Germany. Fax 49-89-4140-4875; e-mail Steimer{at}KlinChem.med.TU-Muenchen.de.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Background: A case of digoxin toxicity resulted from falsely low values with the MEIA II assay for digoxin (AxSYM®; Abbott). The low results were caused by negative interference from canrenone and spironolactone, the latter of which has recently been advocated for the treatment of severe heart failure. Analytical interference from spironolactone has been reported, but little information is available for this effect with newer digoxin assays.

Methods: We examined nine assays (AxSYM, IMx®, TDx®, Emit®, Dimension®, aca®, TinaQuant®, Elecsys®, and Vitros®) for interference by spironolactone, canrenone, and three metabolites. Additionally, all routine digoxin measurements (AxSYM) over a period of 16.5 months (n = 3089) were monitored for interference.

Results: Suppression of the expected values by canrenone (3125 µg/L) was observed for the AxSYM (42% of expected value), IMx (51%), and Dimension (78%) assays. A positive bias was observed for the aca (0.7 µg/L), the TDx (0.62 µg/L), and the Elecsys (>0.58 µg/L). Twenty-five of 669 routinely monitored patients had falsely low results. Nineteen of these had potentially toxic concentrations of digoxin (Emit; >2.0 µg/L), although the AxSYM assay indicated therapeutic or less severe toxic concentrations ({Delta}max = 7.1 µg/L). Except for two unresolved cases, this was attributable to spironolactone, canrenone, hydrocortisone, or prednisolone. Standard doses of spironolactone (up to 50 mg/day) in patients with heart failure displayed inhibition <11%.

Conclusions: The frequency and magnitude of the false-negative results particularly compromise the use of both microparticle enzyme immunoassays. Not only may toxic concentrations remain unidentified, but intoxication could occur should dosage be increased because of falsely low results. With 11 million digoxin tests/year ordered in the US, conceivably many patients could be adversely affected.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Despite declining use in the last few years, digoxin is still one of the most frequently prescribed drugs; it was listed twice among the top 200 prescriptions in the US in 2000 (ranking 36 and 194) (1). In 1995, it was the drug most often monitored therapeutically, with ~11 million tests in the US (2) because of its narrow therapeutic window and potentially serious side effects. Recently, a case of intoxication was reported after the administration of high doses of digoxin (3). Increased dosing had been prompted by continual falsely low serum digoxin concentrations with the AxSYM® (Abbott) microparticle enzyme immunoassay (MEIA II). 1 Spironolactone and canrenone were identified as the major interfering substances. The patient had received digoxin to normalize his heart rate, as well as potassium canrenoate (K-canrenoate), a prodrug of canrenone, for the treatment of ascites.

K-canrenoate is available throughout Europe and in the United Kingdom but not in the US because of its alleged carcinogenic potential (4). According to the package insert for Aldactone-Ampullen® (Roche), K-canrenoate is recommended in doses of up to 800 mg/day for primary and secondary hyperaldosteronism and for the treatment of ascites if a rapid response to therapy is needed or if oral therapy is not possible because of disturbed consciousness or absorption (e.g., portal hypertension). Oral spironolactone is available worldwide and is recommended in doses of up to 200 mg/day for the treatment of edema in congestive heart failure, hepaticcirrhosis, or nephrotic syndrome. Doses up to 100 mg/day are recommended for the treatment of essential hypertension and hypokalemia, whereas doses up to 400 mg/day are suggested for functional testing and diagnosis of primary hyperaldosteronism (package insert for Aldactone®; Searle).

As this case of digoxin toxicity was being reported, a large clinical study was terminated early because of a 30% drop in the mortality rate of patients with severe heart failure who received spironolactone (5). Seventy-five percent of these patients had also received digoxin. On the basis of the findings of this clinical study, there could be a substantial increase in the number of potentially affected patients because both spironolactone and digoxin could be given for the same condition.

In digoxin assays, positive interference by spironolactone and canrenone, an active metabolite of spironolactone and K-canrenoate (for intravenous application), has been described previously (6)(7)(8), but little information is available for newer automated assays (9)(10). To our knowledge, there are no published data concerning the frequency of concomitant therapy of digoxin and spironolactone or K-canrenoate and the possible frequency and significance of this interaction in digoxin monitoring. We therefore tested nine available digoxin assays for interference with spironolactone, canrenone, and three major metabolites and assessed whether the recommended low doses for patients with severe heart failure produce noticeable interference. To determine the significance and frequency of this interaction in a clinical setting, all routine digoxin measurements with the AxSYM MEIA II assay were monitored for interference, and the cause of interference was verified where possible.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
The following digoxin methods were performed as suggested by the manufacturers: AxSYM MEIA II; IMx® MEIA II (singleton analyses; Abbott); TDx® fluorescence polarization immunoassay (FPIA; singleton analyses; Abbott); Emit 2000® (Dade Behring) on the Cobas Mira S system; Dimension® (Dade Behring); aca® (Dade Behring) on the aca SX (singleton analyses); TinaQuant® (Roche) on the Hitachi 912; Elecsys® (Roche) on the Elecsys 2020; and Vitros® slides (Ortho Clinical Diagnostics) on the Vitros 950. Presented results are the means of duplicate measurements unless otherwise specified.

For in vitro experiments, a drug-free serum pool was supplemented with various concentrations of digoxin (98.6% pure; Sigma) and spironolactone (Sigma), canrenone, 7-thiospirolactone, 7-thiomethylspirolactone, or 6-hydroxy-7-thiomethylspironolactone at concentrations of 98, 391, 1563, 6250, and 25 000 µg/L (canrenone and its metabolites from Searle, Skokie, IL). In a separate experiment, the serum pool was supplemented with digoxin and dexamethasone, prednisolone (both from Merck), methylprednisolone (Hoechst), or hydrocortisone (Pharmacia & Upjohn) at concentrations of 50, 250, 500, 2000, 10 000 and 40 000 µg/L.

Ultrafiltration experiments were performed with Centrifree® 30-kDa filters from Amicon. Ultrafiltration was carried out by centrifugation at 3500g for 20 min.

For serum samples from patients undergoing therapy (25–400 mg/day) with spironolactone (oral) or K-canrenoate (intravenous) but not digoxin, the recovery of added digoxin in vitro was determined with the AxSYM, Emit, and TDx assays. After receiving approval from the local institutional committee on human research and obtaining informed consent from all patients, we collected extra serum samples for in vitro experiments.

For some samples, HPLC analysis of spironolactone, canrenone, and 7-{alpha}-thiomethylspironolactone was carried out by BioKinet GmbH. All routine digoxin measurements with the AxSYM MEIA II assay were performed both undiluted and in a 1:1 dilution (Abbott line diluent) over a period of 16.5 months (October 1, 1999, to February 15, 2001). Discrepant results (>0.2 µg/L difference between native sample and corrected dilution) were verified by measurement with an alternative method confirmed to be free of interference from spironolactone and canrenone at therapeutic concentrations (Emit). Suspected treatment with spironolactone, K-canrenoate, or other possible interferents was confirmed by subsequent inquiry on the ward and evaluation of the patients files.

External quality control was ensured through participation in the German proficiency testing scheme (INSTAND, Düsseldorf). During the study, the maximum deviation from target concentrations was 9%, with an average of 4.3% (mean bias, -2.8%). The results of internal quality-control assessments are shown in Table 1 .


View this table:
[in this window]
[in a new window]
  		Table 1. Total imprecision of our routine (AxSYM) and comparison (Emit) methods for digoxin, calculated from control samples.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
in vitro experiments
Minimal or no interference was observed for the Vitros slides, the TinaQuant assay, and the Emit assay when various concentrations of spironolactone, canrenone, and the three metabolites had been added to drug-free serum or drug-free serum supplemented with digoxin to a final concentration of 3.86 µg/L. By contrast, variable positive or negative interference was observed in the other assays, particularly when canrenone had been added. This was also observed to a lower degree with spironolactone, 7-thiospirolactone, and 7-thiomethylspirolactone (Fig. 1, A–D , and Fig. 2, A–D ).



View larger version (42K):
[in this window]
[in a new window]
  		Figure 1. Apparent digoxin concentrations in the presence of spironolactone (A), canrenone (B), and three major metabolites (C–E), all added in vitro to drug-free serum.

No digoxin was present.



View larger version (45K):
[in this window]
[in a new window]
  		Figure 2. Apparent digoxin concentrations in the presence of spironolactone (A), canrenone (B), and three major metabolites (C–E), all added in vitro to drug-free serum.

All samples were supplemented with 3.86 µg/L digoxin. Elecsys assay results reported as >5 µg/L are depicted in the plots as 5 µg/L (A–D).

Falsely increased digoxin concentrations could be detected with the aca and TDx, mainly in the absence of digoxin, and with the Elecsys, mainly when digoxin was present. The strongest positive bias was caused by canrenone, starting at concentrations of ~400 µg/L (Figs. 1BUp and 2BUp ). At 3125 µg/L, the approximate serum canrenone concentration for the patient reported (3), the bias could be estimated as 0.7 µg/L (aca), 0.62 µg/L (TDx), or >0.58 µg/L (Elecsys in the presence of digoxin; Figs. 1BUp and 2BUp ).

Falsely low readings were observed for the Dimension (78% recovery at 3125 µg/L canrenone) and both MEIA assays (recovery, 42% with the AxSYM and 51% with the IMx assay; Fig. 2, A–DUp ). These false-negative results were noticeable starting at canrenone concentrations of 100 µg/L (Fig. 2BUp ).

The extent of inhibition of the AxSYM MEIA assay increased not only with the concentration of canrenone but also with the concentration of digoxin (Fig. 3 ).



View larger version (20K):
[in this window]
[in a new window]
  		Figure 3. Drug-free serum supplemented with various concentrations of digoxin and canrenone.

Samples were measured with the AxSYM MEIA II assay. The graph shows that the extent of inhibition depends on both canrenone and digoxin concentrations.

patients receiving spironolactone or k-canrenoate, but not digoxin
The concentrations of spironolactone, canrenone, and 7-{alpha}-thiomethylspironolactone were measured (BioKinet) in 24 serum samples from eight patients undergoing therapy with spironolactone or K-canrenoate. Five serum samples originated from two patients who were receiving digoxin as well. Spironolactone was below the detection limit of 15 µg/L in 19 of the 24 samples (31–104 µg/L in the other 5 samples). 7-{alpha}-Thiomethylspironolactone concentrations ranged from 46 to 371 µg/L in 17 samples and were below the limit of detection (15 µg/L) in 7 samples. The highest concentrations were found for canrenone (36–3949 µg/L). Fig. 4 shows the correlation between the log of the measured canrenone concentration and the recovery of the AxSYM digoxin assay.



View larger version (23K):
[in this window]
[in a new window]
  		Figure 4. Canrenone concentrations measured in 24 samples from eight patients undergoing therapy with spironolactone or K-canrenoate.

The correlation between the log of the measured canrenone concentration and the recovery of the AxSYM digoxin assay was calculated relative to in vitro-added digoxin (3.86 µg/L) in those patients not receiving digoxin and relative to the results of the Emit digoxin assay in those samples from patients receiving digoxin.

Sera from 10 patients [7 from the gastroenterology department and 3 from the anesthesiology intensive care unit (ICU)] receiving 100–600 mg of K-canrenoate per day, mostly for the treatment of ascites as a consequence of cirrhosis of the liver, were tested for interference in the AxSYM assay. Digoxin-like immunoreactive factor (DLIF) is often present in such patients and has been reported to suppress MEIA results (11)(12) and cause a positive interference in the Emit assay (13). Several other studies, however, have not been able to reproduce the latter findings (14)(15)(16). Before digoxin was added, there was no Emit result in excess of 0.5 µg/L, indicating that the strong inhibition of the AxSYM assay observed after the addition of digoxin was primarily attributable to the presence of canrenone and not DLIF (recovery, 35–81%; total mean, 60%). The observed inhibition of the AxSYM assay in these 10 patients increased with higher doses (100, 200, 400, and 600 mg) of K-canrenoate per day (recoveries, 63.3% for 100 mg/day; 61.4% for 200 mg/day; 59.8% for 400 mg/day; and 50.8% for 600 mg/day; single results for 100 and 600 mg/day; means for 200 and 400 mg/day).

In one patient who received 100 mg of K-canrenoate intravenously twice a day, blood samples were drawn before dosing as well as 2 and 15 min after dosing and were supplemented with 3.38 µg/L digoxin. The AxSYM assay showed a significant inhibition before dosing (recovery, 56%) that became more pronounced both 2 (43%) and 15 min (47%) after intravenous dosing.

To determine whether the recommended low doses of oral spironolactone for patients with heart failure produce interference, we added digoxin to in vitro serum samples from 12 patients receiving daily spironolactone doses between 25 and 100 mg to achieve a concentration of 3.86 µg/L (9 cardiology and 3 gastroenterology patients). The recovery in the AxSYM assay was lower compared with the control in all but one sample tested. The average recoveries before and 3 h after spironolactone was administered were 90% and 87% for the nine cardiology patients for both samples drawn (difference not significant). For the eight patients receiving 25 or 50 mg/day, the respective numbers were 92% and 89%. The recovery for the four patients receiving 100 mg/day was 64%.

patients receiving spironolactone, k-canrenoate, or other interferents and digoxin
Ultrafiltration is the standard procedure used to eliminate high-molecular weight interfering substances and to verify unbound digoxin concentrations (12)(17)(18)(19)(20)(21). Despite canrenone being 89.9–98.3% protein bound (22)(23), the interference in the AxSYM assay could not be eliminated by ultrafiltration because it appeared to be a logarithmic function of canrenone concentrations (Fig. 4Up ). Fig. 5 shows serial dilution and ultrafiltration of serum from the previously reported patient (3). Ultrafiltration removed only a fraction of the interference, as can be seen by the increased recovery upon further dilution. Interpretation of the ultrafiltration results was complicated by the fact that it also removes protein-bound digoxin, which accounts for ~23% of total digoxin (20).



View larger version (23K):
[in this window]
[in a new window]
  		Figure 5. Serum sample from a patient (3) treated with K-canrenoate and digoxin diluted 1+1 (1 mL of sample diluted with 1 mL of drug-free serum) with drug-free serum and then ultrafiltered.

Both the ultrafiltrate and the unfiltered serum were diluted again to achieve a final dilution of 1+3 (1 mL of 1+1 dilutions diluted with 1 mL of drug-free serum). All specimens were measured with the AxSYM assay and the Emit assay as reference (Emit result of native sample, >5.0 µg/L). Only a small portion of the interferent was removed by ultrafiltration.

In this study, the identification of canrenone interference in the AxSYM assay was accomplished by diluting the sample, which yielded higher concentrations after correction for dilution (Fig. 5Up ). The procedure was limited by the limit of quantification of the assay and the digoxin concentration in the sample. Minor interference could not be detected this way.

In a period from October 1, 1999, to February 15, 2001, we monitored all routine digoxin measurements with the AxSYM MEIA II assay (3089 samples from 669 patients) for interference with spironolactone or canrenone, using the dilution procedure outlined above. In 25 of 669 monitored patients (3.74%), significant inhibition of the AxSYM assay compared with the Emit assay was observed (Fig. 6 ). Nineteen of these patients (2.84%) had potentially toxic concentrations of digoxin (Emit result >2.0 µg/L) despite therapeutic (n = 17; 2.54%) or less toxic (n = 2) concentrations being reported by the AxSYM assay. The maximum difference observed was 11.1 µg/L with the Emit compared with 3.99 µg/L with the AxSYM. All but two of these patients were in intensive care units (three different units run by the internal, surgical, and anesthesiology departments in our tertiary care university hospital). This indicates that 8.15% of all ICU patients being monitored for their digoxin concentrations had falsely low results reported by the AxSYM assay. Most of the patients had received intravenous K-canrenoate doses between 200 and 600 mg/day, but inhibition was also observed at 25 mg/day K-canrenoate (intravenous), 50 mg/day spironolactone (oral), and 200 mg/day spironolactone (oral). In most patients, inhibition could be observed in several samples obtained during therapy.



View larger version (40K):
[in this window]
[in a new window]
  		Figure 6. Inhibition of AxSYM results in 25 patients.

Over a period of 16.5 months, all routine samples (669 patients) were monitored for interference. The maximum observed difference between the Emit and AxSYM results is shown for each patient. C, K-canrenoate; S, spironolactone; H, hydrocortisone; P, prednisolone. The number after each letter denotes the daily dose in mg.

During the screening for discrepant results, serum samples from three patients showed significant inhibition in the AxSYM assay, despite the fact that the patients were not receiving spironolactone or K-canrenoate. Two of the patients had received hydrocortisone and one had received prednisolone at doses of 96–200 mg/day. Cortisol was measured in both patients receiving high-dose cortisol (hydrocortisone) therapy (Table 2 ), and the interference of these steroids then verified in vitro. Recovery in the AxSYM assay at 13 000 nmol/L was 72% for hydrocortisone and 81% for prednisolone. No significant interference was observed in the Emit and TDx assays for these interferents, or for methylprednisolone or dexamethasone in any of the three assays.


View this table:
[in this window]
[in a new window]
  		Table 2. Important clinical data from 25 patients with falsely low digoxin results reported by the AxSYM assay.1

One patient was reported via telephone as having received K-canrenoate, but this could not be verified in the file (patient 9 in Table 2Up ). The patient died on the day of the discrepant sample. DLIF remains a possible explanation for the suppression of the AxSYM results (11)(12).

Another patient was admitted to the hospital after an attempted suicide with high doses of digoxin, diltiazem, phenprocoumon, and oxazepam, which was verified by quantitative or qualitative analysis. The ingestion of enalapril, trospium chloride, and a ginkgo biloba preparation (Tebonin intens®; Dr. Wilmar Schwabe GmbH & Co) was also plausible after partly or completely empty packages were found. When digoxin was measured with the AxSYM assay in three samples taken within ~4 h, the AxSYM reported the following results: 3.99, 3.43 and 3.36 µg/L. Diluting the first sample with 2 parts of diluent produced a concentration of 9.5 µg/L. The corresponding results from the first sample for the Emit, TDx, and aca assays were 11.1, 10.5, and 12.3 µg/L digoxin. In contrast to the AxSYM assay, results from the undiluted sample were reported as being above the upper limits for the assays. After the increased digoxin results were confirmed with various assays and in various dilutions, the patient received anti-digoxin antibody therapy.

According to all available information, this patient had not taken spironolactone, K-canrenoate, hydrocortisone, or prednisolone. Canrenone could also not be detected by HPLC in a pool of the three samples before antibody therapy. Cortisol was increased both in a pool of the three samples before initiation of digoxin antibody therapy (1148 nmol/L) and after treatment (1093 nmol/L), possibly because of stress in connection with the attempted suicide. However, the concentrations found do not explain the extent of inhibition observed. Because the cortisol assay used (Elecsys®; Roche) shows a >100% cross-reactivity for prednisolone, both hydrocortisone and prednisolone could be ruled out as causing the observed inhibition. In vitro testing for interference from any of the medications the patient had taken in the suicide attempt failed to explain the strong inhibition of the AxSYM assay. Other medications that had been previously ruled out as inhibitors when the first case of canrenone inhibition was investigated (3) were metamizol, distigmine bromide, norfenefrin, ciprofloxacin, fluconazole, imipenem, dopamine, adrenaline, heparin, ethacrynic acid, omeprazole, ambroxole, and intravenous preparations of amino acids, lipids, and several vitamins.

DLIF also does not serve as an explanation of the interference in this latter case. The TDx assay, despite being sensitive for positive DLIF interference (24), reported no results in excess of the other methods, nor was a significant portion of the interference removed by ultrafiltration (21). Nevertheless, this patient displayed the greatest analytical inhibition with the potentially most serious consequences observed during the 16.5-month observation period of our study.

Important information about the 25 patients found to have significant inhibition of their digoxin results in the AxSYM assay is summarized in Table 2Up .


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Except for the Vitros slides, the TinaQuant, and the Emit assays, both falsely increased [TDx (9)(10)(25), aca, and Elecsys] and inhibited results (AxSYM, IMx, and Dimension) have to be expected when spironolactone or K-canrenoate and digoxin therapy are combined. The clinically most significant interaction found in this study was the massive inhibition in the MEIA II assays (AxSYM and IMx) followed by a weaker inhibition of the Dimension assay. The overall inhibition observed in vivo is a compound effect of spironolactone, canrenone, and at least two metabolites. All of these showed interference in this study and are present at high concentrations in vivo (26)(27).

The amount of positive interference observed in some of the assays will most likely not significantly influence the clinical course of patients affected. Toxic concentrations resulting from positive interference alert both pathologists and clinicians and lead to further investigation, during which interference should be detected. At worst, interference leads to the administration of insufficient digoxin dosages.

Negative interference, particularly in the magnitude observed in both Abbott MEIA II assays (AxSYM and IMx), is much more dangerous. Toxic concentrations may remain unidentified or toxicity could occur should dosing be based on such misleading low values (3). Supposedly "normal" therapeutic concentrations are measured when the true concentration is much higher. Unless all samples are diluted or cross-checked with another method that is insensitive to the interference, there is no indication for the laboratory that the results are incorrect.

In cases where simultaneous treatment with aldosterone inhibitors is reported or known, ultrafiltration of the sample can only partly remove the inhibition and does not lead to a correct result.

A hypothetical mechanism for the inhibition of the MEIA II assay has been described based on the assumption that the cross-reactant dissociates from the antibody during the wash step and allows greater binding of the tracer to the available antibody sites (11). Considering this mechanism, it could be that MEIA assays other than the assay for digoxin also suffer negative interference from cross-reactants known to interfere positively in other assay designs. Similar assay formats should, therefore, be evaluated for this phenomenon. Testing interference in the presence of the primary ligand is mandatory (11).

Which doses are needed to produce a significant suppression of the AxSYM digoxin assay? This study shows that canrenone concentrations reached during conventional ascites therapy (100–600 mg/day K-canrenoate or spironolactone) will most likely cause significant inhibition (recovery, 64% at 100 mg/day). DLIF adds to the problem because it has also been reported to inhibit MEIA assays (11) and to be present in end stage liver disease (28)(29)(30)(31)(32), a condition prompting the use of high-dose spironolactone for ascites therapy.

The lower doses recommended for patients with heart failure (12.5–50 mg spironolactone/day) usually translate into inhibition <11% (recovery >89%). The number of patients possibly affected by this limited inhibition is significant because spironolactone therapy has been advocated for the therapy of severe heart failure (5) and 75% of patients in that study had also received digoxin. The observed inhibition could be greater in individual cases because outpatients are often told to delay their digoxin dose when therapeutic drug monitoring is planned, creating a situation where samples for digoxin concentration monitoring are drawn during the peak concentrations of spironolactone.

Another clinically important finding of this study is the negative interference from compounds other than aldosterone inhibitors. Despite the fact that most immunoassays routinely used to measure digoxin have reasonably low cross-reactivities with the known physiologic hormones (<1%) (2), positive interference has been reported with therapeutically given steroids such as hydrocortisone (33), prednisolone, and other synthetic steroids (34). It could, therefore, be possible that the MEIA II assays are also sensitive to steroids or steroid-like compounds other than hydrocortisone and prednisolone.

Apart from steroids, weaker negative interference for the AxSYM assay has been reported for digitoxin, bufalin (a cardioactive component of the Chinese medicines Chan Su and Lu-Shen-Wan), and oleandrin, a compound not in clinical use (11)(12)(21)(35)(36).

More concerns were raised by the unresolved case of suicidal intoxication. This case illustrates the possible consequences of falsely low digoxin measurements for the adequate therapy of patients with true digoxin toxicity. Antibody therapy would have not been initiated based on digoxin results <4 µg/L as opposed to the reported concentration of 11.1 µg/L.

How many patients are clinically affected? In this study, digoxin concentrations measured with the Abbott AxSYM assay were falsely low in 3.74% of patients (8.15% of ICU patients) in a tertiary care setting mainly because of canrenone (37), hydrocortisone, and prednisolone. Three of four of these patients had potentially toxic concentrations when the AxSYM assay actually reported therapeutic concentrations or greatly underestimated the extent of toxic concentrations. Subsequent action and reporting of the correct higher results ensured that the current digoxin dose was stopped, paused, or reduced. In a routine setting where this interference is not recognized, one must assume that the affected patients would develop even higher digoxin concentrations than reported here. This could be the result of continuation of standard doses despite unrecognized toxic concentrations, or even increased dosing based on falsely low, subtarget concentrations.

Because of our limited screening strategy, probably more patients than were discovered here, in particular those treated with lower doses, were affected by less pronounced falsely low results with the AxSYM assay. It is important to note that, despite repeated instruction, we could not rely on concurrent spironolactone or K-canrenoate therapy being reported when digoxin monitoring was requested in our hospital.

In conclusion, this study shows beyond the previous case report on negative canrenone interference (3) that:


   Acknowledgments
 
We thank Searle (Skokie, IL) for providing canrenone and its metabolites, Heumann Pharma for supporting the measurement of spironolactone and its metabolites by BioKinet GmbH (Vienna, Austria), and all of our colleagues who helped with the collection of samples.


   Footnotes
 
1 Nonstandard abbreviations: MEIA, microparticle enzyme immunoassay; ICU, intensive care unit; and DLIF, digoxin-like immunoreactive factor.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. RxList—the internet drug index. http://www.rxlist.com/top200.htm (Accessed November 2001)..
  2. Jortani SA, Valdes R. Digoxin and its related endogenous factors. Crit Rev Clin Lab Sci 1997;34:225-274.[ISI][Medline] [Order article via Infotrieve]
  3. Steimer W, Müller C, Eber B, Emmanuilidis K. Intoxication due to negative canrenone interference in digoxin drug monitoring. Lancet 1999;354:1176-1177.[ISI][Medline] [Order article via Infotrieve]
  4. Cook CS, Hauswald CL, Schoenhard GL, Piper CE, Patel A, Radzialowski FM, et al. Difference in metabolic profile of potassium canrenoate and spironolactone in the rat: mutagenic metabolites unique to potassium canrenoate. Arch Toxicol 1988;61:201-212.[ISI][Medline] [Order article via Infotrieve]
  5. Pitt B, Zannad F, Remme WJ, Cody R, Castaigne A, Perez A, et al. The effect of spironolactone on morbidity and mortality in patients with severe heart failure. N Engl J Med 1999;341:709-717.[Abstract/Free Full Text]
  6. Huffman DH. The effect of spironolactone and canrenone on the digoxin RIA. Res Commun Chem Pathol Pharmacol 1974;9:787-790.[ISI][Medline] [Order article via Infotrieve]
  7. Lichey J, Rietbrock N, Borner K. The influence of intravenous canrenoate on the determination of digoxin in serum by radio- and enzyme-immunoassay. Int J Clin Pharmacol Biopharm 1979;17:61-63.[ISI][Medline] [Order article via Infotrieve]
  8. Silber B, Sheiner LB, Powers JL, Winter ME, Sadee W. Spironolactone-associated digoxin RIA interference. Clin Chem 1979;25:48-50.[Abstract/Free Full Text]
  9. Foukaridis GN. Influence of spironolactone and its metabolite canrenone on serum digoxin assays. Ther Drug Monit 1990;12:82-84.[ISI][Medline] [Order article via Infotrieve]
  10. Pleasants RA, Williams DM, Porter RS, Gadsden RH, Sr. Reassessment of cross-reactivity of spironolactone metabolites with four digoxin immunoassays. Ther Drug Monit 1989;11:200-204.[ISI][Medline] [Order article via Infotrieve]
  11. Jortani SA, Miller JJ, Helm RA, Johnson NA, Valdes R. Suppression of immunoassay results by cross-reactivity. J Clin Ligand Assay 1997;20:177-179.
  12. Dasgupta A, Trejo O. Suppression of total digoxin concentrations by digoxin-like immunoreactive substances in the MEIA digoxin assay. Elimination of negative interference by monitoring free digoxin concentrations. Am J Clin Pathol 1999;111:406-410.[ISI][Medline] [Order article via Infotrieve]
  13. Lucena MI, Andrade RJ, Fraile JM, Alcantara R, Gonzalez-Correa JA, De La Cuesta FS. Endogenous digoxin-like substance in liver failure. Int J Clin Pharmacol Ther 1994;32:567-569.[ISI][Medline] [Order article via Infotrieve]
  14. Azzazy HM, Duh SH, Maturen A, Schaller E, Shaw L, Grimaldi R, et al. Multicenter study of Abbott AxSYM digoxin II assay and comparison with 6 methods for susceptibility to digoxin-like immunoreactive factors. Clin Chem 1997;43:1635-1640.[Abstract/Free Full Text]
  15. Saccoia NC, Hackett LP, Morris RG, Ilett KF. Enzyme-multiplied immunoassay (EMIT 2000) digoxin assay compared with fluorescence polarization immunoassay and Amerlex 125I-RIA at two Australian centers. Ther Drug Monit 1996;18:672-677.[ISI][Medline] [Order article via Infotrieve]
  16. Sanchez BM, De Cos MA, Peralta FG, Arribas C, Armijo JA. Syva Emit 2000 and Roche "on line" subject to less interference by digoxin-like factors than Abbott TDx FPIA in newborns and pregnant women. Clin Chem 1996;42:974-976.[Free Full Text]
  17. Jortani SA, Miller JJ, Valdes R. Resolving discrepant digoxin results: a case study involving multiple interferences. J Clin Ligand Assay 1996;19:131-137.
  18. Jortani SA, Pinar A, Johnson NA, Valdes R. Validity of unbound digoxin measurements by immunoassays in presence of antidote (Digibind). Clin Chim Acta 1999;283:159-169.[ISI][Medline] [Order article via Infotrieve]
  19. Valdes R, Jortani SA. Monitoring of unbound digoxin in patients treated with anti-digoxin antigen-binding fragments: a model for the future?. Clin Chem 1998;44:1883-1885.[Free Full Text]
  20. Valdes R, Jr, Jortani SA, Gheorghiade M. Standards of laboratory practice: cardiac drug monitoring. National Academy of Clinical Biochemistry. Clin Chem 1998;44:1096-1109.[Abstract/Free Full Text]
  21. Dasgupta A, Scott J. Unexpected suppression of total digoxin concentrations by cross-reactants in the microparticle enzyme immunoassay: elimination of interference by monitoring free digoxin concentration. Am J Clin Pathol 1998;110:78-82.[ISI][Medline] [Order article via Infotrieve]
  22. Karim A, Zagarella J, Hribar J, Dooley M. Spironolactone. I. Disposition and metabolism. Clin Pharmacol Ther 1976;19:158-169.[ISI][Medline] [Order article via Infotrieve]
  23. Ho PC, Bourne DW, Triggs EJ, Smithurst BA. Comparison of plasma levels of canrenone and metabolites after base hydrolysis in young and elderly subjects following single and multiple doses of spironolactone. Eur J Clin Pharmacol 1984;27:435-439.[ISI][Medline] [Order article via Infotrieve]
  24. Mojaverian P, Green PJ, Jhangiania RK, Chase GD. Digoxin-like immunoreactive substance: monoclonal and polyclonal RIA and FPIA compared. J Pharm Biomed Anal 1989;7:585-592.[ISI][Medline] [Order article via Infotrieve]
  25. Morris RG, Frewin DB, Taylor WB, Glistak ML, Lehmann DR. The effect of renal and hepatic impairment and of spironolactone on digoxin immunoassays. Eur J Clin Pharmacol 1988;34:233-239.[ISI][Medline] [Order article via Infotrieve]
  26. Steimer W, Eber B, Müller C. The interference of spironolactone, canrenone and three metabolites in nine immunoassays for the analysis of digoxin [Abstract]. Clin Chem 2000;46:A192-A193.
  27. Vergin H, Mahr G, Metz R, Eichinger A, Nitsche V, Martens H. Analysis of metabolites—a new approach to bioequivalence studies of spironolactone formulations. Int J Clin Pharmacol Ther 1997;35:334-340.[ISI][Medline] [Order article via Infotrieve]
  28. Rosenkranz B, Frolich JC. Falsely increased digoxin concentrations in patients with liver disease. Ther Drug Monit 1985;7:202-206.[ISI][Medline] [Order article via Infotrieve]
  29. Gervais A, Nanji AA, Greenway DC, McLean W. Digoxin-like immunoreactive substance (DLIS) in liver disease: comparison of clinical and laboratory parameters in patients with and without DLIS. Drug Intell Clin Pharm 1987;21:450-452.[Abstract]
  30. DiPiro JT, Cote JR, DiPiro CR, Bustrack JA. Spironolactone interference with digoxin RIA in cirrhotic patients. Am J Hosp Pharm 1980;37:1518-1521.[Abstract]
  31. Nanji AA, Greenway DC. Correlation between serum albumin and digoxin-like immunoreactive substance in liver disease. J Clin Pharmacol 1986;26:152-153.[ISI][Medline] [Order article via Infotrieve]
  32. Yang SS, Korula J, Sundheimer JE, Keyser AJ. Digoxin-like immunoreactive substances in chronic liver disease. Hepatology 1989;9:363-366.[ISI][Medline] [Order article via Infotrieve]
  33. Scholer A, Boecker J, Engelmayer U, Feldmann K, Hannak D, Kattermann R, et al. Comparability of a new turbidimetric digoxin test with other immunochemical tests and with HPLC—a multicenter evaluation. Clin Chem 1997;43:92-99.[Abstract/Free Full Text]
  34. Yosselson-Superstine S. Drug interferences with plasma assays in therapeutic drug monitoring. Clin Pharmacokinet 1984;9:67-87.[ISI][Medline] [Order article via Infotrieve]
  35. Datta P, Dasgupta A. Bidirectional (positive/negative) interference in a digoxin immunoassay: importance of antibody specificity. Ther Drug Monit 1998;20:352-357.[ISI][Medline] [Order article via Infotrieve]
  36. Dasgupta A, Biddle DA, Wells A, Datta P. Positive and negative interference of the Chinese medicine Chan Su in serum digoxin measurement. Elimination of interference by using a monoclonal chemiluminescent digoxin assay or monitoring free digoxin concentration. Am J Clin Pathol 2000;114:174-179.[ISI][Medline] [Order article via Infotrieve]
  37. Steimer W, Müller C, Kopf S, Eber B. False low digoxin readings in every 40th patient! Canrenone and spironolactone interference in the Abbott AxSYM MEIA II assay [Abstract]. Ther Drug Monit 2001;23:482.
  38. Cardeno EF, Carrigan PJ, Saldana SC, Tetin SY. Assessment of aldosterone inhibitors on digoxin immunoassay performance [Abstract]. Ther Drug Monit 2001;23:480.



The following articles in journals at HighWire Press have cited this article:


Home page
 Clin. Chem.Home page
Z. Cao, T. A. Swift, C. A. West, T. G. Rosano, and R. Rej
Immunoassay of Estradiol: Unanticipated Suppression by Unconjugated Estriol
Clin. Chem., January 1, 2004; 50(1): 160 - 165.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Chem.Home page
G. A. McMillin, W. E. Owen, T. L. Lambert, B. K. De, E. L. Frank, P. R. Bach, T. M. Annesley, and W. L. Roberts
Comparable Effects of DIGIBIND and DigiFab in Thirteen Digoxin Immunoassays
Clin. Chem., September 1, 2002; 48(9): 1580 - 1584.
[Full Text] [PDF]

Home page
 Clin. Chem.Home page
R. Valdes Jr. and S. A. Jortani
Unexpected Suppression of Immunoassay Results by Cross-Reactivity: Now a Demonstrated Cause for Concern
Clin. Chem., March 1, 2002; 48(3): 405 - 406.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit an electronic Letter to
the Editor about this paper
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (27)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Steimer, W.
Right arrow Articles by Eber, B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Steimer, W.
Right arrow Articles by Eber, B.
Related Collections
Right arrow Drug Monitoring and Toxicology

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS