Neonatal bilirubin levels, phototherapy decline after laboratory recalibration

Last Updated: 2016-04-12

By Will Boggs MD

NEW YORK (Reuters Health) - Neonatal bilirubin levels and phototherapy use declined after recalibration of the most widely used method to quantify total serum bilirubin (TSB), researchers have found.

"As physicians, we often take lab values at face value," Dr. Michael W. Kuzniewicz, from Kaiser Permanente Northern California, Oakland, told Reuters Health by email. "Therefore, it was very surprising to see that changes in lab calibration had such a dramatic effect on the percentage of infants over the American Academy of Pediatrics (AAP) phototherapy curve thresholds. We saw a 50% reduction in phototherapy use with no change in our clinical practice."

In 2012, Ortho Clinical Diagnostics adjusted the calibrator values for the Vitros BuBc Neonatal Bilirubin Assay slides and predicted that the adjustment would reduce reported unconjugated bilirubin values by 0.1 to 2.16 mg/dL.

Dr. Kuzniewicz's team investigated the clinical effect of this recalibration on maximum TSB levels and phototherapy use in their study of 61,677 neonates born before the recalibration and 42,571 born after the recalibration.

The mean maximum TSB level declined from 10.1 mg/dL in the pre-recalibration period to 8.8 mg/dL in the post-recalibration period (p<0.001), according to the April 11 JAMA Pediatrics online report.

Phototherapy use during the birth hospitalization and readmissions for phototherapy dropped by more than 50% after the recalibration.

In moving average models, recalibration was associated with an 8.0% absolute reduction in infants with a TSB level at or above 15 mg/dL, a 4.8% absolute reduction in infants with a TSB level at or above the AAP phototherapy threshold, a 5.5% absolute reduction in infants receiving phototherapy during the birth hospitalization, and a 2.0% absolute reduction in phototherapy readmissions.

The average number of TSB tests sent per patient declined slightly, from 2.65 to 2.20.

"When using laboratory measurements to make clinic decisions, it is important for physicians to be aware of the accuracy and precision of those measurements," Dr. Kuzniewicz said. "This is especially important when comparing those measurements to a nomogram or threshold values. Differences in measurements made using different assays or laboratories could alter clinical decision making."

Dr. Stanley F. Lo, from Medical College of Wisconsin, Milwaukee, who wrote an accompanying editorial, told Reuters Health by email, "Laboratory test results for neonatal bilirubin (and other tests) from different labs are not always interchangeable. Rarely does a significant change within the same lab occur. The AAP guidelines regarding hyperbilirubinemia do not address the potential inaccuracies of different methods with respect to interpreting bilirubin results."

"That's why the 'art' of medicine remains most prominent in patient care. I would encourage physicians and laboratorians to establish open communication lines so issues such as recalibration or concerns of suspected changes can be efficiently communicated," he said.

"From the laboratory perspective, I think manufacturers need to move toward using calibrators comprised of human serum and unconjugated bilirubin," Dr. Lo said. "This will help in the standardization process and hopefully minimize the effects of changes to calibration. Unfortunately, this is a pretty significant undertaking for companies and I don't expect this change to happen quickly."

"The original study by Bhutani to create the neonatal bilirubin nomogram in the AAP guidelines inherently has some bias or inaccuracy," Dr. Lo added. "Unfortunately, we do not know the extent of this inaccuracy; in fact, we may never know. I've no hard data, but it seems that in spite of this, the guideline remains helpful in determining how to diagnose and treat hyperbilirubinemic patients. Hopefully we have fewer incidences of kernicterus and better care for those in need of it."

The Agency for Healthcare Research and Quality supported this research. The authors made no disclosures.

SOURCE: http://bit.ly/1S7jI7B and http://bit.ly/1SddXbN

JAMA Pediatr 2016.