There are various non-invasive methods to predict the bilirubin levels, with each having its own advantage and disadvantages. Original Kramer scale was further improvised by the identification of five zones and its corresponding range of serum bilirubin [5]. Although this was considered a low-cost tool in screening the neonatal jaundice, it has limitations of observer variation and difficulty in visually assessing the jaundice in dark skinned babies [6].

We previously demonstrated that the stool color in addition to clinical staging expressed as STrIP score can accurately predict SBR [11]. This is based on physio-logical plausibility of relation between entero-hepatic circulation, stool color and serum bilirubin. STrIP score is the sum of Kramer score and the matched stool color score, which predicts serum bilirubin. Unlike TcB, stool color is neither influenced by ambient light or skin color, and it does not need a sophisticated optical device. In this study, our objective was to compare the agreement of STrIP score and transcutaneous bilirubinometer values with measured serum bilirubin in neonatal hyper-bilirubinemia.

This prospective study was conducted between January and June, 2019 in the postnatal ward of JSS Hospital. All babies more than 35 weeks of gestation, with clinical jaundice, and on exclusively breastfeed were included in the study. Babies who were clinically unstable or who had received phototherapy based on clinical assessment were excluded. Babies who did not pass stools during the observation time (10 AM to 1 PM) were not included in the study. Ethical clearance was obtained by the institutional ethics committee. Informed consent was obtained from the mother.

Blood sampling for serum bilirubin was done on day 3 of life as per unit protocol or when clinically indicated. Serum bilirubin estimation was done using the principle of diazotization method. Clinical assessment of jaundice was done using Kramer scale on the same day. STrIP weightage was calculated with the stool sample. Stool sample was collected within 3 hours of detection of clinical jaundice, and the color of the stool was compared by the same investigator with the stool color card (stool strip) and stool color weightage was determined. When there was a gradient of color, the darkest portion of the color was taken for determining the stool color weightage as per the STrIP card which ranged from 1 to 5. Matching of stool color with the stool color strip was done in day light between 10 AM and 1 PM beside the window. STrIP score was then calculated by adding the stool color weightage to clinical assessment by Kramer scale. At the same time, average of three bilirubin readings was taken at the forehead by transcutaneous bilirubinometer (MBJ 20, SAAG Medicare system) in a quiet child soon after assessing the STrIP score.

To estimate the agreement between two non-invasive measurements with the gold standard (serum bilirubin measurement), we defined the precision of 95% confidence of limit of agreement to be 40% of the standard deviation of the difference of two measurements based on our pilot study for sample size. A sample size of 84 subjects was calculated.

Statistical analysis: Statistical analysis was done using Microsoft Excel 2016 and Analyse-it version 4.3 for excel. Both the interventional measurements were compared with the measured serum bilirubin using Bland - Altman analysis for mean and the 95% limits of agreement along with their confidence interval.

During the study period, a total of 332 newborns were admitted to the postnatal ward, of which 252 were eligible. However, 94 were excluded as they met exclusion criteria or got transferred to special wards and another 75 due to non-passage of stools between the pre-decided time window. Finally, a total of 83 (53% males) eligible neonates were included in the study. Of these, 12% were 40 weeks and above, 15% had jaundice by 48 hours of age, and 51% between 48-96 hours. Among the neonates who had dehydration, the median (IQR) dehydration was 3% (2%-4%) above the expected for age (Table I). The mean difference between Kramer scale and SBR was –3.2 mg/dL, between TcB and SBR was 1.6 mg/dL, and that between STrIP score and serum bilirubin was 0.9 mg/dL.

Bland-Altman analysis was done for agreement for all the three non-invasive methods of prediction of bilirubin against SBR. Kramer scale underestimates serum bilirubin and 95% of the values assessed by Kramer scale lie between +0.1 to –6.4 mg/dL of serum bilirubin. 95% limits of agreement between TcB, STrIP and SBR is shown in Fig.1. TcB had a 95% limit of agreement between +3.9 to –4 mg/dL. In contrast, the limit of agreement between STrIP and SBR was between +2.2 and –2 mg/dL (Fig. 1).

Fig. 1 Bland-Altman agreement between STrIP score vs SBR and TcB vs SBR.

Median (IQR) TcB value and STrIP were 11.9 (9.6- 12.7) vs 11 (9-13) mg/dL (P= 0.82). The median (IQR) SBR was 11.2 (9.1-12.6) mg/dL.

The present study demonstrates that the STrIP score can predict serum bilirubin more accurately than the other non-invasive method like transcutaneous bilirubino-meter, at three days of age. The mean difference between STrIP score and serum bilirubin was 0.9 mg/dL, much lesser as compared to other non-invasive methods.

The visual assessment using the modified Kramer score is the most widely practiced method of jaundice screening among neonates. Joan, et al. [12] demonstrated Kramer scale to be ineffective in screening jaundice with the sensitivity and specificity being 89% and 54%, respec-tively. Factors like skin color, birth weight, observer difference (nurse, treating physician) are factors contributing for its poor accuracy [13-15]. This further reinforces that the clinical estimate by Kramer scale alone is not sufficient to detect jaundice needing treatment. Transcutaneous bilirubinometer is another non-invasive method used in bilirubin assessment of newborn. Studies have concluded TcB to have better correlation with serum bilirubin levels than visual assessment [6,15]. Although transcutaneous bilirubin assessment was found to be better than the visual assessment of jaundice, bilirubin values were found to be inconsistent in different sites.

Our study has shown that the values of both STrIP score and TcB are very close to measured serum bilirubin and there was no statistically significant difference between them. Hence the comparison of agreement between STrIP score and TcB values against SBR serves as clinically useful information. The STrIP score values hover around +/- 2 mg/dL of SBR compared to TcB which hovers around +/- 4 mg/dL. The clinical implication of this wide variation with TcB is significant, as the margin of error in making management decision in neonatal hyper-bilirubinemia is very small. Among the three non-invasive clinical estimation of bilirubin, the close match with SBR is the STrIP score, throughout the range of clinical utility i.e., between serum bilirubin values of 5-20 mg/dL.

Modified Kramer scale and transcutaneous bilirubin method was developed with the intention of having a test which is simple, reliable, accurate and to avoid repeated blood sampling. STrIP score has all the benefits as above along with added advantage of promising predictive accuracy.

The limitation in this study is the possible bias of the observer as the same person had performed TcB measure-ment and STRiP score assessment.

To conclude, STrIP score, a simple, bedside, easy to use, reliable non-invasive method has the best agreement with serum bilirubin in neonates compared to other non-invasive techniques - transcutaneous bilirubinometry and clinical assessment using Kramer scale. Larger studies might further help in ascertaining its utilization at the community level in early detection of hyperbilirubinemia requiring phototherapy. More studies are needed to conclude the same for reliable use in larger population.

Ethics clearance: Institutional ethics committee, JSS Medical College; No. JSSMC/IEC/ 140120/19NCT/2020-21, dated 30 January, 2020.

Contributors: SK: study design, analyzing the data and preparing the manuscript; BJT: collected the data, did the literature search and contributed for preparation of the manuscript; DT: has helped in literature search, and manuscript preparation; SM: analysis of data, conceived the research question contributed to the study design, and helped in preparing the manuscript. All the authors have approved the manuscript in the present form.

Funding: None; Competing interest: None stated.

1. Bhutani VK, Zipursky A, Blencowe H, et al. Neonatal hyperbilirubinemia and rhesus disease of the newborn: Incidence and impairment estimates for 2010 at regional and global levels. Pediatr Res. 2013;74:86-100.

2. Ullah S, Rahman K, Hedayati M. Hyperbilirubinemia in neonates: Types, causes, clinical examinations, preventive measures and treatments: A narrative review article. Iran J Public Health. 2016;45:558-68.

3. Das S, van Landeghem FKH. Clinicopathological spectrum of bilirubin encephalopathy/kernicterus. Diagnostics (Basel). 2019;9:24.

4. Shapiro SM. Chronic bilirubin encephalopathy: Diagnosis and outcome. Semin Fetal Neonatal Med. 2010;15:157-63.

5. Wan A, Mat Daud S, Teh SH, et al. Management of neonatal jaundice in primary care. Malaysian Fam Physician. 2016;11:16-9.

6. Varughese, P. Kramer’s scale or transcutaneous bilirubinometry: The ideal choice of a paediatrician? Can we trust our eyes? International Journal of Contemporary Pediatrics. 2019;6:1794-801.

7. Cheng NY, Lin YL, Fang MC. Noninvasive transcutaneous bilirubin assessment of neonates with hyperbilirubinemia using a photon diffusion theory-based method. Biomed Opt Express. 2019;10:2969-84.

8. Tan KL, Dong F. Transcutaneous bilirubinometry during and after phototherapy. Acta Paediatr. 2003;92:327-31.

9. Hulzebos CV, Vader-van Imhoff DE, Bos AF, et al. Should transcutaneous bilirubin be measured in preterm infants receiving phototherapy? The relationship between trans-cutaneous and total serum bilirubin in preterm infants with and without phototherapy. PLoS One. 2019;14: e0218131

10. Taylor JA, Stout JW, de Greef L, et al. Use of a smartphone app to assess neonatal jaundice. Pediatrics. 2017;140: e20170312.

11. Doreswamy SM, Vasudev PH, Thandaveshwara D. Stool color test can help to decide which infants with neonatal hyperbilirubinaemia need phototherapy. Acta Paediatr. 2018;108:371-72.

12. Webster J. An appraisal of the use of the Kramer’s scale in predicting hyperbilirubinaemia in healthy full-term infants. Birth Issues. 2006;14:83-89.

13. Olusanya BO, Ogunlesi TA, Kumar P, et al. Management of late-preterm and term infants with hyperbilirubinaemia in resource-constrained settings. BMC Pediatr. 2015;15:39.

14. Slusher TM, Angyo IA, Bode-Thomas F, et al. Transcu-taneous bilirubin measurements and serum total bilirubin levels in indigenous African infants. Pediatrics. 2004; 113:1636-41.

15. Gupta B, Chaudhary N, Bhatia B, et al. Non-invasive trans-cutaneous bilirubin as a screening test to identify the need for serum bilirubin assessment in healthy term neonates. Journal of Universal College of Medical Sciences. 2014;1:17-21.