| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Letters |
Department of Laboratory Medicine, Yale University School of Medicine, CB 506 Yale-New Haven Hospital, 20 York St., New Haven, CT 06504, Fax 203-688-3725, E-mail petrie.rainey{at}yale.edu
To the Editor:
In their article investigating digoxin measurements in the presence of Digibind®, Ocal and Green (1) conclude that "useful free digoxin concentrations may be obtained for Digibind-treated patients using either the AxSYM or Stratus immunoassays... ". Yet they cite several references that report that digoxin concentrations measured with the Stratus® instrument were significantly higher than were free digoxin concentrations measured after ultrafiltration (2)(3)(4)(5). The AxSYM® assay has also been shown to yield higher results than ultrafiltration (6). These studies contradict the assertion that the AxSYM and Stratus assays are measuring free digoxin concentrations. Nonetheless, one can agree that results from the Stratus II and AxSYM may be suitable for monitoring the course of Digibind treatment. To use these results most effectively, one must understand their limitations and the probable sources of the discrepancy with ultrafiltration results.
The results of previous investigations, as well as binding kinetics theory, suggest that these immunoassays will overestimate the free digoxin concentration. The extent of overestimation will vary with the extent of digoxin binding to albumin, with the assay measurement strategy, and with the absolute and relative amounts of digoxin and Digibind in the specimen. In most instances, the extent of the overestimation will be small enough that the result can be quite informative. In particular, these results set an upper limit on the concentration of pharmacologically active digoxin and can be used to rule out recurrent toxic concentrations.
One reason that the Stratus and AxSYM assay results exceed free digoxin concentration is that they are designed to measure total digoxin in routine specimens, including digoxin bound to serum proteins, principally albumin. Immunoassay antibodies have digoxin binding affinities significantly greater than that of albumin. Assay incubation times are sufficient for albumin-bound digoxin to dissociate fully, to bind to the assay antibodies, and to be measured. However, albumin-bound digoxin is not free digoxin and will not be measured after ultrafiltration. (It should be noted that any digoxin that is not bound to Digibind will exhibit the usual distribution between free and albumin-bound fractions.) If albumin-bound digoxin were the only source of discrepancy between the two types of measurement, it could be handled simply by using a reference interval for total digoxin [e.g., 1.0–2.6 nmol/L (0.8–2.0 ng/mL)] for the Stratus and AxSYM results and a reference interval for free digoxin [e.g., 0.8–2.0 nmol/L (0.6–1.6 ng/mL)] for ultrafiltration assays. However, the differences that have been observed are >25% and cannot be explained solely by albumin binding [normal range, 20% ± 5% (7)].
Another probable contributor to the observed differences is measurement by the immunoassays of some of the digoxin that is bound to Digibind. There are theoretical reasons to expect that this will occur. According to the manufacturer, Digibind is a polyclonal mixture of antibody Fab fragments having a range of affinities, with dissociation constants (Kd) ranging from 10-9 mol/L to 10-10 mol/L (1–0.1 nmol/L) (8). Immunoassay antibodies may be expected to have comparable or higher affinities because it would be difficult to accurately measure digoxin concentrations in the therapeutic range (1–2.6 nmol/L) if the affinity was lower. When a specimen containing digoxin-Digibind complexes is assayed, some of the bound digoxin will dissociate and be captured by the assay antibodies. The extent of transfer will be limited by the incubation time and the dissociation rates for the complexes. Studies of the antibody binding kinetics of small antigens indicate typical association rates of 107 to 108 L/mol-s (9). This implies a dissociation rate of 1–10% per second for digoxin bound to Digibind fragments with a Kd of 10-9 mol/L, and 0.1–1% per second from fragments with a Kd of 10-10 mol/L. Because bound digoxin concentrations are often in the range of 50–100 nmol/L (40–80 ng/mL) after Digibind treatment (5)(10)(11), only a small percentage needs be transferred to the assay antibodies to increase the apparent unbound digoxin concentration by a significant amount.
A third possible source of overestimation is the presence of empty Digibind fragments. Unless these are removed by pretreatment or an early wash step, they can be expected to bind some of the labeled digoxin tracer in the assay, thereby reducing the amount of tracer available to compete with unbound digoxin for assay antibody. If these Digibind-tracer complexes are subsequently removed, as occurs in the Stratus assay, this will yield less bound tracer and an increase in apparent digoxin. This effect would not be expected in the AxSYM assay because Digibind is removed by a wash step before tracer addition.
Other possible components of the discrepancy between immunoassay and ultrafiltration values could be underestimation of free digoxin by ultrafiltration methods because of differential ultrafiltration rates for digoxin and water molecules (10)(12), nonspecific binding of digoxin to the ultrafiltration device (13), or matrix effects from calibrators not prepared in serum ultrafiltrate (11)(14). As Ocal and Green (1) note, the measurement of free digoxin by ultrafiltration has not been well-validated. Thus, it is not currently possible to estimate to what extent, if any, measurements after ultrafiltration underestimate actual free digoxin concentrations.
Currently, patients treated with Digibind can be monitored reasonably, using either the Stratus or AxSYM immunoassays or an ultrafiltration method free of matrix effects. It must be remembered that the immunoassays will overestimate free digoxin concentrations, whereas ultrafiltration may underestimate it. Occasionally, it might be appropriate to use both approaches to set upper and lower limits on the actual concentration.
Footnotes
1 Present address: Memorial Sloan-Kettering Cancer Center, Department of Pathology, 1275 York Ave., New York, NY 10021. *Author for correspondence. Fax 503-494-8148; e-mail greent{at}ohsu.edu
References
Division of Laboratory Medicine, L471, Department of Pathology, Oregon Health Sciences University, Portland, OR 97201
To the Editor:
Although we agree in general terms with the premise put forth by Dr. Rainey, we believe it is important to not lose sight of the fact that the various reasons proposed to account for the biases between the AxSYM and Stratus direct immunoassays and ultrafiltration, although plausible, are hypotheses, not proven observations. Further experimental work should be done to determine more definitively the root causes of these biases. As stated in our report (1), and reiterated by Dr. Rainey in his letter, some of the biases can be accounted for in terms of albumin-bound digoxin (retained by ultrafiltration but measured by the direct method). This, without doubt, accounts for a significant portion of the bias.
On a more conceptual point, under physiological conditions, receptors to which digoxin binds have access to both free and albumin-bound digoxin because albumin-bound digoxin must, by definition, remain in equilibrium with serum free digoxin in that they share the same "compartment". Therefore, an emphasis on the free form of digoxin at the exclusion of albumin-bound digoxin should be viewed as an underestimation of the true serum pool of digoxin. We view albumin-bound digoxin as another "reservoir" of digoxin that should not be ignored in monitoring digitoxic patients. In this regard, in the absence of Digibind® therapy, patients are routinely monitored for total, not free, digoxin. Why then, should there be an emphasis on measuring only free digoxin following Digibind therapy when the patient's baseline therapeutic values, before digitoxicity, are measured in terms of total, not free, digoxin? Indeed, problems in accounting for variability in albumin concentrations among different patients that can be attributed to renal or liver disease, dehydration, or clearance of Digibind may actually be amplified by the very nature of the ultrafiltration technique, which by definition excludes albumin-bound digoxin from being measured. This issue of what to measure, "available" digoxin not sequestered by Digibind vs free digoxin, in tracking patients on Digibind therapy, seems to us to be one of the central questions that need further consideration in reaching a consensus on acceptable methods of determining serum digoxin measurements in patients treated with Digibind.
This having been said, we agree with Dr. Rainey's analysis that the overall differences are relatively small between the methods cited, and that useful therapeutic information can be obtained using either, providing, as Dr. Rainey aptly noted, that the limitations of each method are fully appreciated.
In response to Dr. Rainey's letter, the Journal is publishing a correction in this issue. We thank Dr. Rainey for calling this error to our attention in a personal communication to the editor.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
