 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
To the Editor. —
In their recent article describing a benchmarking model for the management of hyperbilirubinemia, Chou et al1 are to be commended for creating within their health care organization a systematic approach to the prevention of hyperbilirubinemia. At the same time, the prominent inclusion of a full-color nomogram in the article is a potential disservice for the reader.
To their credit, the authors indicate that their model for a laminated pocket card was developed prior to the 1994 publication of the American Academy of Pediatrics (AAP) practice guidelines,2 but readers are encouraged to examine carefully the very significant differences. Unlike the AAP guidelines, the pocket card has no adjustment for consideration of lower birth weights, nor does it reflect the recommendation for more aggressive management when hemolysis is present. The area shaded in blue suggests that an infant with a bilirubin level as high as 20 mg/dL at 3 days of age need not start phototherapy if repeat measurement is performed and implies that phototherapy is optional for a level as high as 24 mg/dL at 4 days of age, 2 examples inconsistent with the AAP practice guidelines. The pocket card has the words "Photo Rx" in one area, whereas the reverse side notes that this means at least double phototherapy; with some confusion, the words "intensive phototherapy" are said to refer to "double or triple phototherapy." The pocket card suggests that an umbilical cord bilirubin level of 8 mg/dL could be managed simply with repeat measurement, a practice not well addressed by the AAP practice guideline and not consistent with the current approach of most neonatologists to start phototherapy immediately for such an infant in hopes of avoiding exchange transfusion.
The reader therefore is encouraged to compare and contrast a similar nomogram, developed for use at Boulder Community Hospital (Boulder, CO) in 2000 and shared with Maisels1 at that time (Fig 1). The similarities are striking in the color portrayal, but the differences are critically important to the appropriate management of hyperbilirubinemia.
|
In translating the intention of that panel, it is quickly apparent that the format of their recommendations using age spans (expressed in hours) proves to be quite problematic. For example, the panel describes 1 measure of hyperbilirubinemia as a total serum bilirubin measurement of 
12 mg/dL at 25 to 48 hours, but an infant with a bilirubin of 12 mg/dL at 25 hours of age is clearly of much greater worry than an infant with the same measurement at 48 hours of age.2 Because of this very serious limitation of the 1994 recommendations, the midpoint of each recommended range was used in creating the graph shown (in this example, the level of 12 mg/dL would be plotted graphically at 36 hours of age). Using specific total bilirubin levels provided by the panel (with each level plotted at the midpoint of the consensus time range) allows for the creation of a series of 3 parallel smooth curves, characterized by an initial rising threshold for intervention. Extrapolation of the resulting curves to the left provides some suggestion of appropriate management during the first few hours of life, such as with the interpretation of umbilical cord bilirubin measurements.
The graph is designed to portray the panel’s recommendations for beginning phototherapy and does not pretend to assist the practitioner in deciding when to discontinue phototherapy after improvement has been measured. For measurements of total bilirubin plotted against age in hours, each color zone can be followed without interruption to the right side of the page, at which a succinct summary of the panel’s recommendation is found. (For simplicity, the use of more intensive phototherapy for higher bilirubin levels is implied rather than stated specifically.) Using this format, the nuances of management for lower birth weights or in the presence of hemolysis can be included according to the panel’s consensus.
At Boulder Community Hospital, the graph has been used for hundreds of jaundiced infants as a full-page worksheet, with the inclusion of areas for recording the birth weight, blood types of the infant and mother, Coombs test results, and each numeric value of sequential bilirubin measurements. Interestingly, the graph has been reduced to a laminated pocket card in our hospital as well.
The idea of benchmarking presented by Chou et al is laudable. Nevertheless, it is suggested that physicians use a tool that more accurately reflects the expert opinions of the AAP panel of experts. With such a tool, investigators may even eventually exceed the results of the Henry Ford Health Systems in the prevention of hyperbilirubinemia. Additionally, rather than waiting for necessity to once again become the mother of invention, the AAP panel, if and when it reconvenes, would be well advised to present their consensus in a graphic format for more rapid acceptance and ease of use.
REFERENCES
This article has been cited by other articles:
 |
|
 |
|
  R. H. Palmer, S. Ezhuthachan, C. Newman, M. J. Maisels, and M. A. Testa Management of Hyperbilirubinemia in Newborns: Measuring Performance Using a Benchmarking Model Pediatrics, September 1, 2004; 114(3): 902 - 902. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
||
|
||
Home | My Pediatrics | Journal Information | Current Issue | Past Issues | Subscriptions & Services | Contact Us Click here for faster international access © 2004 American Academy of Pediatrics. All rights reserved. |
||
|
||
