Error simulation modeling to assess the effects of bias and precision on bilirubin measurements used to screen for neonatal hyperbilirubinemia.

OBJECTIVES: Error simulation models have been used to understand the relationship between analytical performance and clinical outcomes. We developed an error simulation model to understand the effects of method bias and precision on misclassification rate for neonatal hyperbilirubinemia using an age-adjusted risk assessment tool. METHODS: For each of 176 measured total bilirubin (TSBM) values, 10,000 simulated total bilirubin (TBS) values were generated at each combination of bias and precision conditions for coefficient of variation (CV) between 1 and 15%, and for biases between -51.3 µmol/L and 51.3 µmol/L (-3 and 3 mg/dL) fixed bias. TBS values were analyzed to determine if they were in the same risk zone as the TSBM value. We then calculated sensitivity and specificity for prediction of ≥75th percentile for postnatal age values as a function of assay bias and precision, and determined the rate of critical errors (≥95th percentile for age TSBM with <75th percentile TBS). RESULTS: A sensitivity >95% for predicting ≥75th percentile bilirubin values was observed when there is a positive fixed bias of greater than 17.1 µmol/L (1.0 mg/dL) and CV is maintained ≤10%. A specificity >70% for predicting <75th percentile bilirubin values was observed when positive systematic bias was 17.1 µmol/L (1 mg/dL) or less at CV ≤ 10%. Critical errors did not occur with a frequency >0.2% until negative bias was -17.1 µmol/L (-1 mg/dL) or lower. CONCLUSIONS: A positive systematic bias of 17.1 µmol/L (1 mg/dL) may be optimal for balancing sensitivity and specificity for predicting ≥75th percentile TSB values. Negative systematic bias should be avoided to allow detection of high risk infants and avoid critical classification errors.

Implementation of a reformulated Roche® bilirubin Gen.3 reagent did not affect the relationship between BiliChek transcutaneous and Roche total serum bilirubin.

OBJECTIVE: Determine whether introduction of a reformulated bilirubin reagent, the Roche bilirubin Gen.3 assay, changed the relationship between BiliChek transcutaneous bilirubin (TcB) and total serum bilirubin (TSB). DESIGN AND METHODS: TcB results from term infants in the level 1 nursery obtained within one hour of a TSB were reviewed over two periods, six months before and after the conversion from the previous generation Roche bilirubin reagent to the new Roche Gen.3 bilirubin assay. TcB measurements were performed using BiliChek transcutaneous devices (Respironics, Marietta GA). Distribution of TSB results, and TcB minus TSB bias, were compared before and after introduction of the reformulated Roche bilirubin Gen.3 assay. Median and interquartile range (IQR) TSB values and bias were calculated. A statistical difference between median TSB values and bias were assessed using Man-Whitney test. RESULTS: A total of 301 paired TcB and TSB results were obtained, 172 before and 129 after implementation of the reformulated Roche bilirubin Gen.3 reagent. Median (IQR) TSB was 7.8 (6.8-8.7)mg/dL (133.3 (116.3-148.8) µmol/L) before and 7.6 (6.7-8.4)mg/dL (130 (114.6-143.6)µmol/L) after implementation of the reformulated reagent (p = .1373). Median (IQR) bias between TcB and TSB was 2.9 (2.2-3.7) mg/dL (49.6 (37.6-63.3)µmol/L) before the reformulated reagent was implemented; and did not change at 2.9 (2.1-3.9) mg/dL (49.6 (35.9-66.7)µmol/L) after implementation (p = .8242). CONCLUSION: Implementation of the reformulated Roche bilirubin Gen.3 reagent did not affect the relationship between BiliChek transcutaneous and total serum bilirubin; thus no changes were needed to the neonatal TcB screening protocol as a result of the new bilirubin reagent.

Evaluation of hemoglobin interference in capillary heel-stick samples collected for determination of neonatal bilirubin.

OBJECTIVE: In this study, we determined the assay performance criteria necessary to produce acceptable results for >or=98% of neonate bilirubin samples collected by capillary heel-stick. STUDY DESIGN AND METHODS: We determined serum free hemoglobin levels in 151 heel-stick serum samples to determine the hemolysis level. We then tested the effect of hemolysis on total bilirubin levels determined by four commercially available assays. RESULTS: The mean level of serum free hemoglobin was 1.62 g/L. Of the serum total bilirubin assays tested, the Total Bilirubin Special (Roche Diagnostics) and the TBILI (Roche Diagnostics) reagents did not show significant interference at the concentrations of free hemoglobin observed in >or=98% of heel-stick samples. The Vitros Bu/Bc slide (Ortho-Clinical Diagnostics) showed significant interference only at normal bilirubin concentrations; while the Bilirubin DPD reagent (Amresco Inc.) showed significant interference starting at hemoglobin concentrations of 1.0 g/L. CONCLUSIONS: Bilirubin assays that are not sensitive to approximately 6 g/L free hemoglobin should provide accurate results for most samples obtained via capillary heel-stick. Of the four assays tested, the Bilirubin DPD reagent (Amresco Inc.) was the most susceptible to the presence of free hemoglobin and will result in a higher rejection rate of neonate capillary heel-stick samples.

Accuracy of neonatal transcutaneous bilirubin measurement in the outpatient setting.

OBJECTIVE. To evaluate the effectiveness of transcutaneous bilirubin (TcB) measurement in predicting risk for neonatal hyperbilirubinemia in outpatients. DESIGN. Subjects were infants ≤8 days old seen in an outpatient clinic. Infants discharged with high-risk (HR) or high-intermediate risk (HIR) total serum bilirubin (TSB) values and jaundiced infants were recruited. TSB and TcB (BiliChek) levels were plotted on an hour-specific nomogram to determine risk for hyperbilirubinemia. RESULTS. A total of 79 infants provided 87 sets of TcB and TsB values. Mean bias and standard deviation between TcB and TsB was 1.5 ± 2.1 mg/dL for outpatients, compared with 2.7 ± 1.3 mg/dL for inpatients. The sensitivity and specificity of HR or HIR TcB for predicting an HR or HIR TSB were 87% and 58%, respectively. Of 9 infants readmitted for phototherapy, 1 had a low-risk TcB and high-risk TSB. CONCLUSIONS. TcB screening in the outpatient environment may not be safe and efficient.

BiliChek transcutaneous bilirubin meter overestimates serum bilirubin as measured by the Doumas reference method.

OBJECTIVES: To determine the relationship between BiliChek TcB (Respironics, Marietta GA) and Doumas reference serum or plasma total bilirubin (TSB). DESIGN AND METHODS: Pooled samples with values assigned by the Doumas reference method were used to establish the relationship between a local laboratory and reference Doumas TSB. We then established the relationship between TcB and TSB in the 3 months before and after reassignment of calibrator setpoints undertaken to match the local laboratory to Doumas reference bilirubin values. RESULTS: Before calibrator setpoint reassignment TSB as measured in our laboratory overestimated Doumas reference bilirubin. After calibrator adjustment laboratory TSB was within 1.7-6.8 micromol/L (0.1-0.4 mg/dL) of Doumas reference values. Mean bias between BiliChek TcB and TSB was 42.8+/-22.2 micromol/L (2.5+/-1.3mg/dL) (n=94) before and 49.6+/-22.2 micromol/L (2.9+/-1.3mg/dL) (n=115) after calibration adjustment. CONCLUSIONS: BiliChek TcB significantly overestimates TSB as measured by the Doumas reference method.

Evaluation of the BiliChek noninvasive bilirubin analyzer for prediction of serum bilirubin and risk of hyperbilirubinemia.

We identified clinical and laboratory variables affecting the relationship between transcutaneous and serum bilirubin levels and determined whether transcutaneous bilirubin values could be used to predict the risk of hyperbilirubinemia. Median bias between transcutaneous and diazo serum bilirubin was 2.0 mg/dL (34.2 micromol/L), while median bias between transcutaneous and the Vitros (Ortho Clinical Diagnostics, Rochester, NY) serum bilirubin values was 1.3 mg/dL (22.2 micromol/L). The mother's ethnicity, the gestational age, and postnatal age did not impact the relationship between transcutaneous and serum bilirubin values. In contrast, the serum bilirubin method (diazo vs Vitros) and collection container (clear vs amber tube) significantly impacted the relationship between transcutaneous and serum bilirubin values. Transcutaneous bilirubin was a sensitive but not specific predictor of the risk of hyperbilirubinemia using a conventional risk nomogram. Because systematic differences between serum bilirubin methods and local laboratory practices impact the relationship between transcutaneous and serum bilirubin values, the effectiveness of transcutaneous prediction of the serum bilirubin risk zone will vary by institution.