Transcutaneous bilirubinometry versus total serum bilirubin measurement for newborns

Clinical evaluation of the installation conditions and accuracy of wearable neonatal jaundice meter

Neonatal jaundice is a physiological phenomenon that occurs in newborns. However, because excess bilirubin can cause cytotoxicity, abnormal hyperbilirubinemia can lead to bilirubin encephalopathy. In the evaluation of neonatal jaundice, serum bilirubin concentrations are indirectly assessed via a transcutaneous bilirubinometer. However, commercially available transcutaneous bilirubinometers are expensive, especially in developing countries. Therefore, many newborns still suffer from bilirubin encephalopathy worldwide. We previously developed a prototype wearable jaundice meter. In this study, we aimed to investigate further improvements in this device and evaluate its measurement accuracy at different measurement sites and at different pressures. The bilirubin values of forty-eight newborns were measured at different sites (middle of the forehead and upper eyebrows) and with different pressures (1 and 2 newtons). The transcutaneous bilirubin values measured by the wearable bilirubin meter were strongly correlated with the transcutaneous bilirubin values measured by a commercially available bilirubin meter (r > 0.9) and the actual serum bilirubin values (> 0.84), even under different conditions. This wearable jaundice meter is much cheaper and smaller than commercially available jaundice meters and is a novel medical device that could be useful for evaluating neonatal jaundice.

Feasibility study of texture-based machine learning approach for early detection of neonatal jaundice

Untreated neonatal jaundice can have severe consequences. Effective screening for neonatal jaundice can prevent long-term complications in infants. Non-invasive approaches may be beneficial in settings with limited resources. This feasibility study explores a texture-based machine learning approach for early detection of neonatal jaundice. Clinical data and skin images of 200 infants were captured from four body locations using the Neonatal Jaundice Screening and Assessment Plate. Data were split into training/validating (n = 160) and blind testing (n = 40) datasets. Ninety-two features (three clinical, 89 texture-based) were extracted after image processing. Eight machine learning models were compared for bilirubin level prediction. The best performing model, Support Vector Machine (SVM), was implemented in a web-based application (AmberSNAP) and tested using blind testing dataset. SVM paired with RRelief-F feature selection achieved optimal performance for head and sternum measurements, while SVM with Univariate Regression performed best for abdomen and lower leg measurements. Blind testing demonstrated good performance in bilirubin level prediction (mean absolute error: 1.675 mg/dL; root mean square error: 2.192 mg/dL), with moderate correlation between predicted and measured values (r = 0.644, p < 0.001). These findings suggest that texture-based machine learning is a feasible approach for neonatal jaundice screening in low-resource settings.

Assessing accuracy of BiliPredics algorithm in predicting individual bilirubin progression in neonates-results from a prospective multi-center study

Background

Neonatal jaundice affects more than half of neonates. As bilirubin values usually peak few days after hospital discharge, jaundice remains a leading cause of rehospitalization. The recently developed BiliPredics algorithm, integrated in the first CE-approved bilirubin prediction tool, predicts individual bilirubin progression for up to 60 h into the future. Goal of the prospective study was to assess accuracy of this algorithm in predicting individual bilirubin prior to hospital discharge in neonates.

Methods

A prospective multi-center study was conducted in 2021 at the University Children's Hospitals in Tübingen and Regensburg, Germany. Various scenarios differing in type and number of bilirubin measurements and in prediction horizon were tested. Primary objective was prediction accuracy of the BiliPredics algorithm based on total serum bilirubin (TSB) measurements or based on transcutaneous bilirubin (TcB) measurements alone. Secondary objective was prediction accuracy based on combinations of TSB and TcB measurements. For assessment of accuracy, two validation metrics, absolute prediction error ( a P E ) and relative prediction error ( r P E ) , and two clinical acceptance conditions, margin of error of the 95%-confidence interval (95%-CI) and percentage of clinically relevant mis-predictions defined as a P E > 85 μ mol / L , were investigated.

Results

Out of 455 enrolled neonates, 276 neonates met bilirubin inclusion criteria and were included in the analyses. Irrespective from tested prediction horizons, median r P E was small (8.5% to 9.5%) utilizing TSB measurements for up to 30 and 60 h and slightly higher (13.8%) utilizing TcB measurements for up to 48 h. The same applied for median a P E . Both clinical acceptance conditions were fulfilled across tested scenarios. Results for combined TSB-TcB scenarios up to a prediction horizon of 48 h without adjustment for type of measurement were comparable to TSB and TcB scenarios fulfilling both clinical acceptance conditions.

Conclusion

Results from this prospective study in neonates confirm that the BiliPredics algorithm accurately predicts bilirubin progression up to 60 h with TSB measurements and up to 48 h with TcB or combined TSB-TcB measurements. As such, prediction tools utilizing this algorithm are expected to facilitate and safely optimize jaundice risk assessment at hospital discharge with the potential to reduce jaundice-related rehospitalizations.

Optimal site for applying transcutaneous bilirubinometer as an outpatient screening tool for neonatal jaundice: a comparison between the sternum and forehead

Background

The gold standard for assessing neonatal jaundice (NJ) is the serum total serum bilirubin (TSB) level by the diazo method. A transcutaneous bilirubinometer (TCB) provides a convenient, noninvasive readout within minutes. The reliability of TCB as the diagnostic tool and the proper site for TCB measurement remains unsettled.

Objectives

This study aimed to (1) evaluate the reliability of TCB in the NJ outpatient management and (2) identify a better site to obtain TCB readings.

Methods

This retrospective study examines data collected prospectively over 15 months at a level III facility. Parents were advised to bring their neonates back to our nursery if neonates were judged to be at risk for NJ or poor weight gain, and a follow-up with the primary practitioner was not available. Those who had received phototherapy or sustained forehead bruising were excluded from the analysis. Blood was collected immediately after TCB readings for TSB measurement using the di-azo method. The primary endpoint was admission for treatment according to the AAP 2004 guidelines. A mixed-effects model was used to assess the correlation of forehead TCB (TCB-f) or sternal TCB (TCB-s) with TSB by adjusting for age at measurement (hours), gestational age (GA), sex, and race. Repeated Measure Receiver Operator Characteristic (ROC) curves were constructed for TCB readings against the hospital admission, and the cutoffs for each method were selected to balance the sensitivity and specificity.

Results

There were 500 visits for 350 neonates, including 136 females, 114 white, 134 black, 71 Hispanic, and 30 Asian. The mean GA was 38.5 weeks [standard deviation (SD) = 1.6], and the mean body weight (BW) was 3,238 g (SD = 506). Forty-five (12.9%) neonates were admitted for phototherapy or blood exchange transfusion according to the TSB levels. Only 43 admitted neonates had all three measurements. Assuming TCB has the same reading as TSB, 30 out of 43 (69.8%) and 20 out of 43 (46.5%) neonates would be sent home if only TCB-f and TCB-s were used, respectively. TCBf has a trend of underestimating the necessity of hospitalization compared to TCBs (p = 0.092 by McNemar test). After adjusting for age of measurement, GA, sex, and race, both TCB-f and TCB-s readings positively correlated with TSB (p < 0.0001). Using repeated measure ROC, with hospital admission for treatment as the primary outcome, the area under the curve (AUC) for TCB-f was 0.79 (95% CI: 0.71-0.86), and AUC for TCB-s was 0.86 (95% CI: 0.81-0.92). A cutoff of 14.3 for TCB-s gave a sensitivity of 81% and a specificity of 78%. A cutoff of 12.6 for TCB-f gave a sensitivity of 80% and a specificity of 65%.

Conclusions

TCB measurements can discriminate well in predicting admission for NJ treatment in our nursery but tend to underestimate the severity. The sternum is a better site for TCB measurements. We must point out that more than 40% of neonates who should be admitted for NJ management would be sent home if TSB were not obtained simultaneously. We recommend adjusting TCB readings according to unit-based calibration before clinical implementation.

Development and Validation of a Smartphone Application for Neonatal Jaundice Screening

Importance

This diagnostic study describes the merger of domain knowledge (Kramer principle of dermal advancement of icterus) with current machine learning (ML) techniques to create a novel tool for screening of neonatal jaundice (NNJ), which affects 60% of term and 80% of preterm infants.

Objective

This study aimed to develop and validate a smartphone-based ML app to predict bilirubin (SpB) levels in multiethnic neonates using skin color analysis.

Design, Setting, and Participants

This diagnostic study was conducted between June 2022 and June 2024 at a tertiary hospital and 4 primary-care clinics in Singapore with a consecutive sample of neonates born at 35 or more weeks' gestation and within 21 days of birth.

Exposure

The smartphone-based ML app captured skin images via the central aperture of a standardized color calibration sticker card from multiple regions of interest arranged in a cephalocaudal fashion, following the Kramer principle of dermal advancement of icterus. The ML model underwent iterative development and k-folds cross-validation, with performance assessed based on root mean squared error, Pearson correlation, and agreement with total serum bilirubin (TSB). The final ML model underwent temporal validation.

Main Outcomes and Measures

Linear correlation and statistical agreement between paired SpB and TSB; sensitivity and specificity for detection of TSB equal to or greater than 17mg/dL with SpB equal to or greater than 13 mg/dL were assessed.

Results

The smartphone-based ML app was validated on 546 neonates (median [IQR] gestational age, 38.0 [35.0-41.0] weeks; 286 [52.4%] male; 315 [57.7%] Chinese, 35 [6.4%] Indian, 169 [31.0%] Malay, and 27 [4.9%] other ethnicities). Iterative development and cross-validation was performed on 352 neonates. The final ML model (ensembled gradient boosted trees) incorporated yellowness indicators from the forehead, sternum, and abdomen. Temporal validation on 194 neonates yielded a Pearson r of 0.84 (95% CI, 0.79-0.88; P < .001), 82% of data pairs within clinically acceptable limits of 3 mg/dL, sensitivity of 100%, specificity of 70%, positive predictive value of 10%, negative predictive value of 100%, positive likelihood ratio of 3.3, negative likelihood ratio of 0, and area under the receiver operating characteristic curve of 0.89 (95% CI, 0.82-0.96).

Conclusions and Relevance

In this diagnostic study of a new smartphone-based ML app, there was good correlation and statistical agreement with TSB with sensitivity of 100%. The screening tool has the potential to be an NNJ screening tool, with treatment decisions based on TSB (reference standard). Further prospective studies are needed to establish the generalizability and cost-effectiveness of the screening tool in the clinical setting.

Validity of transcutaneous bilirubin measurements during and after phototherapy in term and late preterm infants

The purpose of the study was to investigate correlation and concordance between total serum bilirubin (TSB) and transcutaneous bilirubin measured at covered (TcBC) and uncovered (TcBU) skin during and after discontinuation of phototherapy. A cross-sectional study included ≥ 34 weeks gestation infants requiring phototherapy for neonatal hyperbilirubinemia. In-house, photo-opaque patches were placed on infants' sternums before phototherapy initiation. Simultaneous blood sampling for TSB, TcBC, and TcBU measurements were performed. Among 103 infants included in the final analysis, 70% were full-term. Covering skin during phototherapy resulted in strong TcBC-TSB correlation (r = 0.91, 95% CI 0.87-0.94, P < 0.001) compared to TcBU (r = 0.53, 95% CI 0.37-0.65, P < 0.001), persisting post-phototherapy (r = 0.88, 95% CI 0.82-0.91, P < 0.001). Bland-Altman analysis showed a higher mean difference and wider 95% limits of agreement for TcBU-TSB during phototherapy (-6.3 mg/dL and -11.1 to -1.6) vs TcBC-TSB (0.9 mg/dL and -1.2 to 2.9). Passing-Bablok regression analysis confirmed good agreement between TcBC and TSB.

CONCLUSIONS

The application of in-house, photo-opaque patches enhanced the correlation and agreement between TcBC and TSB during and after discontinuation of phototherapy. This may prove particularly useful in resource-limited settings where commercial devices are unavailable.

WHAT IS KNOWN

• Transcutaneous bilirubin measurement has been widely used as a screening method for neonatal hyperbilirubinemia. • The accuracy of transcutaneous bilirubin measurements during and after phototherapy in infants with hyperbilirubinemia has been debated.

WHAT IS NEW

• Our study demonstrated that utilizing carefully designed photo-opaque patches enhanced the accuracy of transcutaneous bilirubin measurement during and after phototherapy. • Effective in-house alternatives are crucial in resource-limited settings where commercial opaque patches are not always accessible or affordable.

Prevalence and associated factors of mother-reported jaundice in newborns

OBJECTIVE

Newborns' jaundice is the result of bilirubin accumulation as fetal hemoglobin is metabolized by the immature liver. This study aimed to estimate the prevalence of mother-reported newborn jaundice and associated factors.

METHODS

A cross-sectional study was carried out using data from a longitudinal study involving 914 children. The data were reviewed using Poisson regression with a robust estimator in a hierarchical model in which the sociodemographic variables constituted the first level, those related to the conditions of the pregnancy constituted the second level, and those related to the report of jaundice, the third level. Prevalence ratios and their relevant confidence intervals were estimated.

RESULTS

The prevalence of reported jaundice in newborns was 17.9%. The variables late pregnancy, urinary infection during pregnancy, and preterm and post-term birth were independently statistically associated with a higher prevalence of newborn jaundice reports.

CONCLUSION

We can conclude that mother-reported newborn jaundice was 17.9% associated with maternal and birth aspects.

Transcutaneous bilirubinometry for detecting jaundice in term or late preterm neonates

BACKGROUND

The American Academy of Pediatrics and the Canadian Paediatric Society both advise that all newborns should undergo bilirubin screening before leaving the hospital, and this has become the standard practice in both countries. However, the US Preventive Task Force has found no strong evidence to suggest that this practice of universal screening for bilirubin reduces the occurrence of significant outcomes such as bilirubin-induced neurologic dysfunction or kernicterus.

OBJECTIVES

To evaluate the effectiveness of transcutaneous screening compared to visual inspection for hyperbilirubinemia to prevent the readmission of newborns (infants greater than 35 weeks' gestation) for phototherapy.

SEARCH METHODS

We searched CENTRAL, MEDLINE, Embase, CINAHL, ClinicalTrials.gov, ICTRP, and ISRCTN in June 2023. We also searched conference proceedings, and the reference lists of included studies.

SELECTION CRITERIA

We included randomized controlled trials (RCTs), quasi-randomized, cluster-randomized, or prospective cohort studies with control arm that evaluated the use of transcutaneous bilirubin (TcB) screening for hyperbilirubinemia in newborns before hospital discharge.

DATA COLLECTION AND ANALYSIS

We used standard methodologic procedures expected by Cochrane. We evaluated treatment effects using a fixed-effect model with risk ratio (RR) and 95% confidence intervals (CI) for categorical data and mean, standard deviation (SD), and mean difference (MD) for continuous data. We used the GRADE approach to evaluate the certainty of evidence.

MAIN RESULTS

We identified one RCT that met our inclusion criteria. The study included 1858 African newborns at 35 weeks' gestation or greater who were receiving routine care at a well-baby nursery, and were randomly recruited prior to discharge to undergo TcB screening. The study had good methodologic quality. TcB screening versus visual assessment of hyperbilirubinemia in newborns: - probably reduces readmission to the hospital for hyperbilirubinemia (RR 0.25, 95% CI 0.14 to 0.46; P < 0.0001; moderate-certainty evidence); - may have little or no effect on the rate of exchange transfusion (RR 0.20, 95% CI 0.01 to 14.16; low-certainty evidence); - probably increases the number of newborns who require phototherapy prior to discharge (RR 2.67, 95% CI 1.56 to 4.55; moderate-certainty evidence). - may have little or no effect on the rate of acute bilirubin encephalopathy (RR 0.33, 95% CI 0.01 to 8.18; low-certainty evidence). The study did not evaluate or report cost of care.

AUTHORS' CONCLUSIONS

Moderate-certainty evidence suggests that TcB screening probably reduces hospital readmission for hyperbilirubinemia compared to visual inspection. Low-certainty evidence also suggests that TcB screening may have little or no effect on the rate of exchange transfusion compared to visual inspection. However, moderate-certainty evidence suggests that TcB screening probably increases the number of newborns that require phototherapy before discharge compared to visual inspection. Low-certainty evidence suggests that TcB screening may have little or no effect on the rate of acute bilirubin encephalopathy compared to visual inspection. Given that we have only identified one RCT, further studies are necessary to determine whether TcB screening can help to reduce readmission and complications related to neonatal hyperbilirubinemia. In settings with limited newborn follow-up after hospital discharge, identifying newborns at risk of severe hyperbilirubinemia before hospital discharge will be important to plan targeted follow-up of these infants.

Innovative approaches to neonatal jaundice diagnosis and management in low-resourced settings

Persistent challenges in addressing severe neonatal hyperbilirubinaemia in resource-constrained settings have led to ongoing and often unacceptable rates of morbidity, disability and mortality. These challenges stem from limitations such as inadequate, inefficient or financially inaccessible diagnostic and therapeutic options. However, over the past decade, noteworthy innovations have emerged to address some of these hurdles, and these innovations are increasingly poised for broader implementation. This review provides a concise summary of these novel, economically viable diagnostic solutions, encompassing point-of-care assays and smartphone applications, as well as treatment modalities, notably more effective phototherapy and filtered sunlight. These advancements hold promise and have the potential to meaningfully reduce the burden of neonatal hyperbilirubinaemia, signifying a promising shift in the landscape of neonatal healthcare.

Skin color influences transcutaneous bilirubin measurements: a systematic in vitro evaluation

OBJECTIVE

Concerns have been raised about the effect of skin color on the accuracy of transcutaneous bilirubin (TcB) measurements, a widely used method for hyperbilirubinemia diagnosis in newborns. Literature is inconclusive, with both reported under- and overestimations of the TcB with increasing skin pigmentation. Therefore, the influence of skin color on TcB measurements was systematically evaluated in a controlled, in vitro setting.

METHODS

A bilirubin meter (JM-105) was evaluated on layered phantoms that mimic neonatal skin with varying dermal bilirubin concentrations (0-250 µmol/L) and varying epidermal melanosome volume fractions (0-40%; light-dark skin color).

RESULTS

TcB measurements were influenced by skin pigmentation. Larger mimicked melanosome volume fractions and higher bilirubin levels led to larger underestimations of the measured TcB, compared to an unpigmented epidermis. In the in vitro setting of this study, these underestimations amounted to 26-132 µmol/L at a TcB level of 250 µmol/L.

CONCLUSION

This in vitro study provides insight into the effect of skin color on TcB measurements: the TcB is underestimated as skin pigmentation increases and this effect becomes more pronounced at higher bilirubin levels. Our results highlight the need for improved TcB meter design and cautious interpretation of TcB readings on newborns with dark skin.

IMPACT

Key message: Skin color influences transcutaneous bilirubin measurements: the darker the skin, the larger the underestimation. What this study adds to existing literature: Existing literature is inconclusive regarding the influence of skin color on transcutaneous bilirubin measurements. This study systematically evaluates and clarifies the influence of skin color on transcutaneous bilirubin measurements in a controlled, in vitro setting.

IMPACT

This study aids to better interpret the measured TcB level in patients with varying skin colors, and is particularly important when using TcB meters on patients with dark skin colors.