The Journal of Pediatrics
Volume 156, Issue 3 , Pages 508-509, March 2010

Nomogram can help estimate risk of serious hyperbilirubinemia in healthy infants

Tufts Evidence-Based Practice Center, Boston, Massachussets

Article Outline

 

Varvarigou A, Fouzas S, Skylogianni E, Mantagou L, Bougioukou D, Mantagos S. Transcutaneous bilirubin nomogram for prediction of significant neonatal hyperbilirubinemia. Pediatrics 2009;124:1052-9.

Back to Article Outline

Question 

In healthy term and near-term neonates, can transcutaneous bilirubin (TcB) measurements be used to develop a predictive nomogram to assess the risk of significant hyperbilirubinemia?

Back to Article Outline

Design 

Prospective cohort.

Back to Article Outline

Setting 

Well-infant nursery in a university hospital in Greece, between September 2005 and December 2007.

Back to Article Outline

Participants 

2039 healthy neonates (gestational age ≥35 weeks and birth weight ≥2000 g).

Back to Article Outline

Intervention 

10 382 TcB measurements were performed with a BiliCheck bilirubinometer (SpectRx, Norcross, GA) at designated time points between 12 and 120 hours of life.

Back to Article Outline

Outcomes 

Significant hyperbilirubinemia was defined on the basis of the hour-specific threshold values for phototherapy proposed by the American Academy of Pediatrics.

Back to Article Outline

Main Results 

Using likelihood ratios (LRs), the high- and low-risk demarcators for each designated time were calculated and presented on an hour-specific nomogram. Significant hyperbilirubinemia was documented for 122 neonates (6%). At 24 hours of life, the high-risk zone of the nomogram had 73.9% sensitivity and a positive LR of 12.1 in predicting significant hyperbilirubinemia, whereas the low-risk zone had 97.7% sensitivity and a negative LR of 0.04. At 48 hours, the high-risk zone had 90% sensitivity and a positive LR of 12.1, whereas the low-risk zone had 98.8% sensitivity and a negative LR of 0.02. In this study population, the probability of significant hyperbilirubinemia would be >35% for values in the high-risk zone and <0.5% for values in the low-risk zone of the nomogram.

Back to Article Outline

Conclusions 

The authors provide a predictive TcB tool that could allow for a noninvasive, risk-based approach to neonatal hyperbilirubinemia.

Back to Article Outline

Commentary 

This paper presents nomograms for interpreting hour-specific transcutaneous bilirubin (TcB) measurements with the Bilicheck device, by predicting whether infants should receive phototherapy according to the American Academy of Pediatrics (AAP) guidelines.1 The authors' approach appears to be a pragmatic one—pediatricians will have to act on the basis of TcB measurements, and, presumably, they intend to follow the AAP guidelines. Briefly, the AAP guidelines interpret total serum bilirubin (TSB) measurements differently for infants in “low,” “intermediate,” and “high” baseline risk strata that are defined by gestational ages and risk factor profiles. Yet, the Varvarigou nomogram explicitly averages over the baseline risk strata. I do not argue that this is a bad decision—I merely point out the oddity of predicting a combination of TSB and clinical factors from a surrogate of TSB alone. So, to which baseline risk stratum is the Varvarigou nomogram “high-risk zone” cutoff most applicable? Judging from the distribution of gestational ages (infants with risk factors were practically excluded), 20% of the studied infants are of “intermediate” baseline risk with the rest being “low” risk. Therefore, the Varvarigou “high-risk zone” boundary is a bit of an underestimate for “low risk,” a bit of an overestimate for “intermediate risk,” and an overestimate for (the excluded) “high-risk” infants in the AAP guidelines. It is my own belief that we can only fully understand the implications of variations in screening through careful decision modeling.2 The authors have a wonderful data set to model the TcB trajectories of individual infants over time and the opportunity to pursue more detailed models than those published to date. It may be worth a try.

Back to Article Outline

References 

  1. American Academy of Pediatrics Subcommittee on Hyperbilirubinemia . Management of hyperbilirubinemia in the newborn infant 35 or more weeks of gestation. Pediatrics. 2004;114:297–316
  2. Trikalinos TA, Siebert U, Lau J. Decision-analytic modeling to evaluate benefits and harms of medical tests: uses and limitations. Med Decis Making. 2009 Sep 4;[Epub ahead of print]

PII: S0022-3476(09)01203-7

doi:10.1016/j.jpeds.2009.11.068

The Journal of Pediatrics
Volume 156, Issue 3 , Pages 508-509, March 2010