Deficiencies in Centralized Jaundice Testing

Deepakshyam Krishnaraju
Sivakumar Palaniswamy


Rapid and early identification of neonates with increasing bilirubin levels coupled with timely intervention can reduce hospital readmissions for hyperbilirubinemia, which accounts for half of all newborn readmissions [1]. Although readmission rates vary widely, preterm and late preterm infants are at greater risk, as up to 80% are susceptible to hyperbilirubinemia [2,3]. The burden of monitoring bilirubin levels is expected to increase in coming years, as the number of premature births increases [4].

Hospital discharge of infants within 72 hours of birth also significantly increases the risk for readmission for hyperbilirubinemia, as bilirubin levels peak at three to four days of age. Reduced lengths of stay – a cost- cutting strategy now used by many hospitals [5] – and increasing rates of pre-term births together pose significant challenges to accessibility and timeliness of bilirubin screening. This is complicated by the need for laboratories to respond quickly with accurate results and for clinicians to receive decision support.

Accurate and timely reporting of bilirubin levels is critical for implementation of appropriate therapy for newborns who require treatment for hyperbilirubinemia.

Inaccurate measurement of bilirubin can adversely impact patient care as well as result in significant financial impact. In a retrospective study conducted across 21 hospitals in the Kaiser Permanente network, serum bilirubin levels were overestimated, on average, by 1.25 mg/dL (21.4 μmol/L). This inaccuracy in measurement resulted in $8 million dollars of unnecessary follow- up testing [9]. Alarmingly, the assay method used by the hospitals in this study is also employed in 380 other labs across the U.S. Therefore, the true cost nationally is likely much, much higher.

The number of newborn readmissions for hyperbilirubinemia and the costs of unnecessary follow- up testing could both be reduced by improving the availability and access to testing, the accuracy of bilirubin measurements, and the speed of clinical diagnosis. The neonatal hyperbilirubinemia healthcare sector is very sensitive to errors in clinical lab methods, as such errors adversely impact patient management and healthcare resources.

Inconsistency of Bilirubin Assay Methods

There are 3 types of bilirubin assays used in the U.S. As shown in Figure 1, approximately three-quarters of all laboratories use the diazo method, while the remainder use spectrophotometry, and oxidation methods14.

Figure 1: Distribution of various bilirubin assay methods reported in the CAP Neonatal Bilirubin NB-C 2017 survey [14]

Accurate and precise bilirubin measurement is required for clinicians to assess the change in levels over time, in order to make appropriate treatment decisions. Hospital laboratories typically require 100-500 μL of blood to measure total and conjugated bilirubin. Unfortunately, different methods for measuring total and conjugated bilirubin in that volume of blood show wide variation and suffer from several significant shortcomings [6-8].

Figure 2 shows the results of neonatal bilirubin measured with 4 different analyzers. All three assay methods were represented. The difference between the instrument reporting the highest measured bilirubin and that reporting the lowest measured value was greater than 3.31 mg/dL. This represents a clinically significant difference which could result in mismanagement and inappropriate treatment of a newborn with jaundice.

Figure 2: Summary data from the College of American Pathologists Neonatal Bilirubin Survey (NB-C 2017) showing mean values for neonatal bilirubin concentrations measured using different methods and instruments [14]

One major source of inconsistency is the susceptibility of the individual methods to in vivo hemolysis interference. In the case of conjugated bilirubin assays, there is an observed cross-reactivity, up to 5%13, between conjugated bilirubin assay and unconjugated bilirubin, due to the routine calibration procedure used. Due to the unavailability of a commercial source of conjugated bilirubin, conjugated bilirubin assays are generally calibrated using synthetic, unconjugated bilirubin at a controlled pH.

A problem for many diazo dye-based bilirubin assays is hemoglobin interference. Interference can be caused by direct spectral interference from oxyhemoglobin, as well as from oxyhemoglobin-mediated breakdown of bilirubin during and after diazotization. Hemoglobin interference is a function of the concentration of plasma hemoglobin, serum bilirubin, and the analyte. The net interference can be positive or negative depending upon the sample dilution.

Other Sources of Variation

Preanalytical Errors: There are 3 phases in laboratory testing, namely preanalytical, analytical, and post- analytical testing. Of these, the preanalytical errors – prior to the assaying of the specimen – account for approximately 70% of all errors that occur in the testing process [19]. Preanalytical errors can occur at numerous points in the process, including patient preparation, blood collection, sample transport, and sample storage.

Due to concerns about the stability of bilirubin following collection and interference from a variety of domestic and foreign substances, results of bilirubin testing may need to be interpreted with a degree of caution.

Bilirubin is stable following collection if the sample is shielded from exposure to light. Exposure to light changes the structure of the bilirubin molecule into several different water-soluble products. After such exposure, the measured bilirubin value is not representative of the true bilirubin value. Some studies have shown that in vitro bilirubin may be degraded as much as 50% following exposure to light [10].

The site where the blood sample is collected may also be a cause for variation in bilirubin measurement. While there is no agreement in the literature regarding the clinically significant difference in bilirubin concentrations in samples obtained from different body sites, capillary blood samples are preferred. Capillary blood draw ensures that only small sample volumes are obtained in order to avoid iatrogenic anemia, especially in premature infants. However, a collection of capillary blood is also associated with greater rates of in vitro hemolysis and clot formation [11], both of which lead to inaccurate bilirubin measurements.

Lack of Reference Intervals: Currently, there is a very limited knowledge regarding the laboratory practices to validate reference intervals. A U.S.-based survey, conducted under the auspices of the Q-probe study program of the College of American Pathologists (CAP) [15], reported that of 163 participating labs, only 25% had established appropriate pediatric reference intervals [16]. Most laboratories adopted reference intervals from external sources, such as medical literature or manufacturer published reference ranges.

Overcentralized Healthcare Resources:

The centralization of healthcare resources, meant to address patient demand, the paucity of resources, the need for specialty care, or simply contain costs also create delays throughout the healthcare system. While these delays may be administrative or clinical in nature, they may directly delay a lab request, blood draw, or sample delivery, thereby affecting time to result and time to treatment. Once a sample is received in a laboratory, the general turnaround times (TAT) can range from 45 minutes to 3 hours. While the lab TAT is easily quantifiable, the other factors may be responsible for delays of several hours. These wait times raise concerns worsening clinical status, delays in treatment, and the need for additional procedures and longer length of stay.

Many neonatal intensive care units have improved response times by adopting the point of care testing for select lab tests. However, bilirubin assays via hand-held, the point of care devices have not yet been adopted. Dependence on central lab assays and their respective TATs continues to produce long wait times. These wait times directly impact patient discharge times and increase the length of stay, as infants are often kept unnecessarily for 2 hours or more until test results are reported and interpreted.

Large Sample Volumes:

Iatrogenic blood loss resulting from the intensive clinical monitoring in the weeks immediately following birth remains a primary cause of neonatal anemia and the need for blood transfusion. This risk is amplified in premature and extreme low birth weight (ELBW) infants. In one study of 63 ELBW infants, the number of laboratory blood tests performed during the first week of life averaged 9.8 ± 4.3 per infant per day [17]. With a total blood volume in a premature infant of 90-100 mL per kg of body weight, a 1,500 g infant has a total of 125mL intravascular blood volume. Blood transfusion may be required when 10% or more of a neonate’s blood volume is drawn over a 2 to 3-day period [18,21].

Central laboratories are designed to facilitate high volume throughput, rather than to minimize blood loss in a neonate. Testing for bilirubin typically requires a minimum of 0.4 mL of blood, and many automated analyzers cannot accommodate pediatric sample tubes. Blood samples with insufficient volume may be analyzed after dilution: However, manual dilution predisposes the sample to preanalytical errors, causes delays, and exposes the specimen to external elements, thereby leading to inaccurate results. Most automated chemistry assays, which utilize plasma or serum, require relatively large volumes of whole blood due to the poor yield of high-hematocrit specimens from neonates. Some analyzers have been modified to allow for smaller sample volumes to be processed, but the additional handling of specimens remains a problem.


Bilirubin is generally considered one of the most problematic analytes in clinical chemistry [12], and no standardized reference method for total bilirubin determination is currently available.

Attempts to measure bilirubin have resulted in widely varying results [13], even in a research setting. Most importantly, clinically significant variation is observed among bilirubin results obtained from different assays of the same specimen. Lack of assay standardization adversely impacts patient care, especially when different instruments and methods are used for measurement of bilirubin in a single individual. The various point-of-care devices and laboratory instruments used within a single healthcare location or system may create incomparable results. If an infant has bilirubin measured in several locations (e.g. maternity unit, pediatric unit, emergency room, or clinic), the ability to interpret change over time may be questioned due to lack standardization of assays.

With total bilirubin levels a major driver for the treatment of neonatal hyperbilirubinemia, hospitals must improve the current diagnostic system through harmonization of assays to reduce the variability of measurements, reduce times to treatment, and improve patient outcomes. An easily accessible, low cost, and accurate assay for jaundice is essential to detect and to prevent the development of severe hyperbilirubinemia, especially in infants discharged before 48 hours of life. A system incorporating efficiency, accuracy and clear quality metrics is the basis for cost-efficient treatment.


  1. Metcalfe, Amy, et al. “Proportion of neonatal readmission attributed to the length of stay for childbirth: a population-based cohort study.” BMJ open 6.9 (2016): e012007.
  2. NICE: Detection and treatment of neonatal jaundice. Lancet 2010, 375(9729):1845.
  3. Rennie J, Burman-Roy S, Murphy MS: Neonatal   jaundice:   summary   of   NICE   guidance. BMJ 2010, 340:c2409.
  4. Reddy UM, Ko CW, Raju TN, Willinger M. Delivery indications at late-preterm gestations and infant mortality rates in the United States. Pediatrics 2009; 124: 234-40.
  5. Marbella, Anne M., Veerappa K. Chetty, and Peter M. Layde. “Neonatal hospital lengths of stay, readmissions, and charges.” Pediatrics 101.1 (1998): 32-36.
  6. Doumas BT, Eckfeldt JH. Errors in the measurement of total bilirubin: A perennial problem. Clin Chem 1996; 42, 6: 845-48.
  7. Newman TB, Maisels   MJ.   Evaluation   and   treatment   of   jaundice   in   the   term   newborn: A kinder, gentler approach. Pediatrics 1992; 89, 5: 809-18.
  8. Vreman et al. Interlaboratory variability of   bilirubin   measurements.   Clin   Chem   1996; 42, 6: 869-73.
  9. Kuzniewicz M, Greene D, Walsh E et al. Association between laboratory calibration of a serum bilirubin assay, neonatal bilirubin levels, and phototherapy use. JAMA Pediatrics 2016
  10. Young DS. Effects of preanalytical variables on clinical laboratory tests. 2nd ed. Washington, DC: AACC Press, 1997.
  11. Young DS. Preanalytical issues in neonatology. Blood Gas News 2002; 11,1: 14-18
  12. Rolinski, Boris, et al. “Total bilirubin measurement by photometry on a blood gas analyzer: potential for use in neonatal testing at the point of care.” Clinical chemistry 47.10 (2001): 1845-1847.
  13. Kirk JM. Neonatal jaundice: a critical review of the role and practice of bilirubin analysis. Ann Clin Biochem 2008; 45: 452-62
  14. College of American Pathologists: Neonatal Bilirubin NB-C 2017 Participant Summary
  15. Howanitz P. Quality assurance measurements in departments of pathology and laboratory medicine. Arch Pathol Lab Med 1990; 114: 1131-35.
  16. Freidberg R, Souers R, Wager E et al. The origin of reference intervals. Arch Pathol Lab Med 2007; 131: 348-57.
  17. Alves-Dunkerson JA, Hilsenrath PE, Cress GA, Widness JA. Cost analysis of a neonatal point- of-care monitor. Am J Clin Pathol 2002;117:809-18
  18. Widness, John A., et al. “Changing patterns of red blood cell transfusion in very low birth weight infants.” The Journal of pediatrics 129.5 (1996): 680-687.
  19. e=pt42.5.493&rgn=div5#se42.5.493_11289 (last accessed 06/26/2018)
  20. Bonini P, Plebani M, Ceriotti F et al. Errors in laboratory medicine. Clin Chem 2002; 48,5. 691-98
  21. (last accessed 06/26/2018)