 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
LETTER TO THE EDITOR |
Charles E. Ahlfors, MD
LW Ligand, LLC
Vashon, WA 98070-3609
To the Editor.—
The commentary by McDonagh and Maisels1 in response to our recent review2 opens with an assertion that "changes in plasma concentrations of free drug are thought to have little clinical relevance except in unusual situations," implying that measurement of unbound (free) unconjugated bilirubin (Bf) is unlikely to be useful in evaluating risk for bilirubin encephalopathy. We suggest that this reflects a misinterpretation of the cited articles,3–5 which describe pharmacokinetic mechanisms of drug (or, in our case, bilirubin) distribution. In fact, Benet and Hoener3 clearly state that "pharmacologic effect is related to unbound drug concentration." Evidence that brain uptake of bilirubin is primarily determined by Bf was recently presented by Ahlfors and Parker6 on the basis of measured Bf and dissociation rates of protein-bilirubin complexes in jaundiced infants’ sera.
We agree with the cited articles that the quality of serum binding normally has little influence on the pharmacologic response to a given drug dosage. At a given "dose" of bilirubin the Bf will be similar in patients, although there may be marked differences in total serum bilirubin (TSB) resulting from variations in protein binding. Thus, infusion of human serum albumin (HSA) in rats with disrupted blood-brain barriers "appeared" to protect the central nervous system from bilirubin toxicity. Marked electroencephalographic changes in HSA-primed rats occurred at higher concentrations of bilirubin in both serum (TSB) and brain (mostly albumin-bilirubin) than in control rats. However, the toxic threshold for Bf was identical in control and HSA-infused rats, and the dose required for effect was nearly the same.7 However, in a jaundiced newborn, the "dose" of bilirubin is unknown and can only be inferred from either a measured Bf or observed clinical consequence.
We also agree that the presence of binding competitors has little effect on drug distribution or drug effect except when the drug is tightly bound, has a high tissue-extraction ratio, and/or has a narrow "therapeutic (or toxic) index." Bilirubin meets all these criteria: >99.9% of serum bilirubin (TSB) is bound to serum proteins, and the fractional uptake of Bf on single-pass central nervous system perfusion is 
25%. Moreover, in contrast to drug binding, the bilirubin/albumin molar ratio is very high in newborns with hyperbilirubinemia, making them especially vulnerable to the administration of competing drugs. Thus, intravenous administration of a displacing sulfonamide to premature primates with hyperbilirubinemia produced a sudden transient surge in Bf, a persistent decrease in TSB, and almost immediate depression in electrocortical activity.8 Clinically, the sulfonamide9 and benzyl alcohol10 experiences bear tragic witness to this phenomenon.
McDonagh and Maisels also claim that "it is uncertain what particular fraction of bilirubin [is measured]" using the peroxidase method and suggest that Bf might include bilirubin bound to weak binding loci. In our review, we emphasized that plasma bilirubin is bound to multiple receptors, dominated by the high-affinity albumin-binding site. Bf is the tiny fraction of TSB solvated in water and in equilibrium with all of these sites. By using 2 enzyme concentrations, the peroxidase method can measure both Bf and the aggregate dissociation rate of bilirubin bound to all plasma constituents. The advantage of the peroxidase reaction is that it requires binding of bilirubin to the enzyme (in competition with plasma receptors). Therefore, the possibility of oxidizing bilirubin actually bound to a plasma-binding site is extremely remote. It should also be noted that bilirubin-albumin–binding constants and Bf levels as measured in plasma using the peroxidase method are similar to those in bilirubin/purified albumin preparations.11
We are also puzzled by the concern about errors in Bf produced by bilirubin isomers. Although there is evidence that photoisomers bind less avidly to albumin, they do not increase the oxidation rate of bilirubin12,13 and, thus, have a minimal effect on Bf estimations. It is true that the method does not discriminate photoisomers from native bilirubin IX
z,z, but neither do methods commonly used to measure TSB. Current recommendations for intervention on the basis of TSB, and applauded by the commentators, are not modified when the infant is receiving phototherapy.
Finally, although the peroxidase assay is probably the best understood and clinically studied method to measure Bf, we agree with McDonagh and Maisels that alternative methods (eg, front-face fluorometry, etc) warrant development and evaluation. We agree that NICUs provide important venues to relate serial Bf with outcome, but the low prevalence of kernicterus in term and near-term infants with extreme hyperbilirubinemia should not deter our efforts to improve predictors of risk for these infants as well. In doing so, we must accept that the simplistic "one size fits all" approach that has guided jaundice management for many decades needs to be replaced by a multidimensional clinical assessment. As Lucey14 pointed out years ago, there is no specific level of either TSB or Bf that applies to all infants.
The clinical importance of binding and pharmacokinetics has long been difficult for many bilirubin experts to endorse (déjà vu). We hope that our review of clinical and laboratory data2 supporting an important role of Bf in bilirubin toxicity will change that view. Given this evidence and recent improvements in Bf methodology,11 we feel it is time for concerted efforts to automate and apply Bf technologies in well-planned clinical studies relating Bf and TSB to outcome in various clinical settings.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
||
|
||
Home | My Pediatrics | Journal Information | Current Issue | Past Issues | Subscriptions & Services | Contact Us Click here for faster international access © 2006 American Academy of Pediatrics. All rights reserved. |
||
|
||
