Exploring bilirubin/albumin ratio as predictor of significant hyperbilirubinemia among neonate in Southwest Nigeria

INTRODUCTION

Jaundice is the most common reason for readmission in most health facilities after early discharge and normal spontaneous vaginal delivery of a healthy-term newborn.¹

Clinically significant neonatal hyperbilirubinemia refers to jaundice in which immediate treatment with phototherapy or exchange blood transfusion is indicated to prevent complications.² When serum bilirubin reaches a dangerously high level, the albumin-binding capacity of bilirubin could be exceeded leading to free unbound bilirubin exiting the circulation and staining the extravascular tissues. Bilirubin staining of the tissues including the brain could result in irreversible damage with fatal consequences.³ Bilirubin measurement is a frequently performed test in the newborn and total serum bilirubin (TSB) is the gold standard for the assessment of the risk of hyperbilirubinemia.^4-7

Early discharge of healthy term newborn after spontaneous vaginal delivery has advantages such as promotion of maternal-infant bonding, reduction of risk of nosocomial infection and reduction in cost of hospital stay.⁸ This could make early detection and immediate intervention in significant neonatal hyperbilirubinemia increasingly difficult. Therefore, the need for accurate prediction of babies who may be at risk of developing significant hyperbilirubinemia before discharge from the hospital becomes pertinent. This offers hope of preventing problems associated with delayed recognition of the significant jaundice in the babies discharged early.^9,10

TSB measures both conjugated and unconjugated bilirubin together and therefore, might not reflect the exact risk of bilirubin toxicity. The absolute measurement of the unconjugated bilirubin which is unbound bilirubin that freely exits the intravascular space is a better reflection of bilirubin toxicity. The magnitude of free bilirubin is more predictive of bilirubin neurotoxicity compared with TSB.¹¹ Therefore, measurement of TSB alone is not sufficient for risk assessment of bilirubin toxicity because the amount of unconjugated bilirubin in the central nervous system depends on the amount of free unbound bilirubin that crosses the blood-brain barrier to cause neurotoxicity.^12,13

The bilirubin/albumin ratio is better predictor of bilirubin toxicity in patient with severe hyperbilirubinemia than the TSB estimation alone.¹⁴ while other studies have reported to the contrary.¹⁵

This study therefore will serve as a tie-breaker on the bilirubin/albumin ratio as a surrogate for free bilirubin and subsequently a predictor of significant hyperbilirubinemia in the newborn.

METHODS

Study design

The study was a hospital-based longitudinal, cross-sectional study conducted in selected hospitals in Ogbomoso Oyo State Nigeria over a period of nine months between July 2022 to April 2023.

Selection criteria

The study included apparently healthy term newborns with birth weight ≥2500 gm, APGAR score of ≥7 at one minute and five minutes of life and delivered by spontaneous vaginal delivery or elective cesarean section while those with life-threatening congenital anomaly, maternal diseases in pregnancy such as diabetes, hepatitis or those with instrumental delivery-forceps and vacuum delivery were excluded from the study.

Sample size determination

The sample size was determined with Leslie-kish’s formula using a prevalence of 55% from a previous study and the level of statistical significance chosen as 1.96 at a 95% confidence interval.^16,17

The estimated sample size was 380.4 and adding 20% attrition rate a total of 456 were calculated for the study.

Sampling technique

A multistage sampling technique was employed in the selection of subjects for this study. Two out of five local government areas were randomly selected by balloting. A list of all the registered health facilities (Government and private) in the selected LGAs were obtained from Oyo State ministry of health. A total of 4 health facilities were chosen. The required sample size was attained by allocating subjects proportionally among the four hospitals. This was achieved by dividing the calculated sample size by the total yearly deliveries across the selected centres and multiplying it by the annual delivery in each of the selected centres. Neonatal recruitment proceeded consecutively until the desired sample size was attained in each of the selected centres.

Procedure

The four selected hospitals were visited before the commencement of sample collection. Consents were obtained from mothers, in the first phase of labor, upon admission into the labor ward.

Samples from cord blood were collected into lithium heparin bottles immediately after birth, wrapped with foil paper and transported to the laboratory for analysis. Follow-up samples were collected within 72 hours and 96 hours post-delivery. The TSB for both cord and follow-up samples was measured with the colorimetric method.

Data analysis

The data were analyzed using statistical package for the social sciences (SPSS) version 23.0 for Windows. Data were represented using tables and/or charts. Cord bilirubin and bilirubin/albumin ratio were represented by mean and standard deviation.

Ethical clearance

Ethical clearance was obtained from the ethical review committee of LAUTECH teaching hospital and Oyo State ministry of health ethical committee

RESULTS

A total of 471 were analyzed, among which 243 (51.6%) were males and 228 (48.4%) were females.

Table 1 show the mean and standard deviation of anthropometric parameters of the subjects, with birth weight, length and Pondera index of 3.19±0.42, 47.50±3.48 and 3.07±0.82 respectively.

Table 2 showed the cord parameters and follow-up bilirubin in significant neonatal hyperbilirubinemia (≥12 mg/dl) was recorded in 20 (4.2%) of neonates between 72 and 96 hours of life.

Table 3 shows the association between the cord bilirubin, albumin, bilirubin/albumin ratio and follow-up bilirubin values at 72-96 hours of life. There is a statistically significant relationship between cord bilirubin/ albumin ratio and follow-up bilirubin (p=0.025, χ²=0.104).

Comparison of the mean cord blood bilirubin, cord blood albumin and bilirubin/albumin ratio between subjects with and without significant neonatal hyperbilirubinemia shows a higher mean cord bilirubin, albumin and bilirubin/albumin ratio in subjects with significant neonatal hyperbilirubinemia. The association between cord bilirubin/albumin ratio and significant neonatal hyperbilirubinemia shows no statistically significant relationship (p=0.117, χ²=0.072). However, there is a significant relationship between cord bilirubin and significant neonatal hyperbilirubinemia.

The association between cord bilirubin/ albumin ratio cutoff of 0.61 and follow-up bilirubin shows no statistically significant relationship (χ²=0.016, p=0.733) (Table 4).

For the prediction of significant neonatal hyperbilirubinemia, a cut-off of 1.58 mg/dl was set for cord blood total bilirubin based on the ROC curve analysis. At this cut-off point, the sensitivity, specificity, PPV and NPV are 70%, 48.34%, 5.7% and 97.3% respectively.

For the prediction of significant hyperbilirubinemia, a cut-off value of 4.05 was obtained for albumin with sensitivity, specificity, PPV and NPV of 65%, 39.47%, 4.5% and 96.2% respectively.

A cut-off value of 0.38 was obtained for cord bilirubin-albumin ratio with sensitivity, specificity, PPV and NPV of 60%, 46.78%, 4.6 % and 96.3% respectively, and AUC of 0.533 for bilirubin/ albumin ratio.

The low AUC value (0.51-≤0.60) recorded shows a poor prediction of significant neonatal hyperbilirubinemia for each of the parameters (cord bilirubin, albumin and bilirubin /albumin ratio).

BAR=Bilirubin/albumin ratio.

DISCUSSION

Mean cord albumin in this study was significantly higher compared to a similar study but in resonant with another study by Venkatraman et al.^14,19 This observation could result from racial and genetic difference in study population.

This study shows that whenever there is an increase in cord bilirubin, cord albumin and cord bilirubin/albumin ratio, there is a corresponding increase in follow-up bilirubin as opposed to a similar study by Bhat et al, which demonstrated a strong correlation between cord blood bilirubin and development of pathological jaundice in the term neonate.²⁰ The difference in the two studies could be attributed to the fact that all respondents in Bhat’s study had either physiological or pathological jaundice compared to the current study in which a small percentage developed jaundice at follow-up.

There is no statistically significant relationship between cord bilirubin/albumin ratio and the development of significant hyperbilirubinemia at 72-96 hours follow-up. This is in contrast to the findings of Deghady et al in which bilirubin/albumin ratio was overall best of 4 cord parameters in predicting significant hyperbilirubinemia.²¹ Similarly, Ramteke et al demonstrated the ability of cord bilirubin/albumin to predict significant neonatal hyperbilirubinemia, though cord bilirubin was found to be more predictive of significant hyperbilirubinemia.²²

This study observed the ability to predict significant hyperbilirubinemia by the cord bilirubin and the cord bilirubin/ albumin ratio was further tested using ROC curve analysis. In the current study, the cut-off value of cord bilirubin for predicting significant hyperbilirubinemia was comparable to that of a similar study of 418 newborns by Aktas et al in which the cut-off value of cord bilirubin for the prediction of significant hyperbilirubinemia requiring phototherapy was 1.67 mg/dl, this is comparable with the cut-off of 1.58 mg/dl in the current study however the sensitivity and specificity were comparatively higher in Aktas et al study.²³ Knupfer et al in a similar study concluded that cord bilirubin of greater than 1.75 mg/dl is predictive of significant hyperbilirubinemia with a sensitivity of 90% and negative predictive value of 99.1%.²⁴ Therefore neonates with cord bilirubin value below 1.75 mg/dl are considered not likely to develop significant hyperbilirubinemia. These findings are consistent with the findings of the current study especially with the strong negative predictive value observed, newborns with cord bilirubin levels of less than 1.58 mg/dl are at a low risk for the development of significant hyperbilirubinemia at follow-up although with poor predicting significant hyperbilirubinemia.

The cut-off value of cord bilirubin/albumin ratio for the prediction of significant hyperbilirubinemia obtained in the present study is low when compared with most similar studies, In Khairy et al study a higher cord bilirubin/albumin ratio of 0.61 was shown to have a good predictive value for the development of significant hyperbilirubinemia with a sensitivity and negative predictive value of 100% respectively.¹⁴ Similarly Ramteke et al reported that 87.5% of neonates who developed significant neonatal hyperbilirubinemia had cord bilirubin/albumin ratio above the cut-off point of 0.89.²² The lower cut-off value for cord bilirubin/albumin ratio observed in the current study could be attributed to the difference with mean value of cord albumin in the study population, 4.78 gm/dl in the current study versus 2.8 g/dl in Khairy’s study.¹⁴

The current study also shows that a low cord bilirubin/albumin ratio is protective against the development of significant hyperbilirubinemia. This is consistent with the findings of Ramteke and Khairy et al in which cord bilirubin/albumin ratio was significantly higher in newborns who developed neonatal hyperbilirubinemia compared with those who did not.^14,22 Rhena et al also concluded that the higher the bilirubin/albumin ratio, the higher the risk of hyperbilirubinemia.²⁵

This present study demonstrated that both cord bilirubin and cord bilirubin/albumin ratios were weak in predicting significant hyperbilirubinemia. However, cord bilirubin is better predictor of significant neonatal hyperbilirubinemia than cord bilirubin/albumin ratio. Rhena et al concluded that cord bilirubin is better than cord bilirubin/albumin ratio and that both parameters were strong in predicting significant jaundice.²⁵ Ramteke et al reported that cord bilirubin has a better diagnostic accuracy than cord bilirubin/ albumin ratio and this was corroborated by finding by Sun et al on usefulness of cord bilirubin for subsequent hyperbilirubinemia prediction.^22,26

There are several studies into the predictive ability of cord blood parameters on significant hyperbilirubinemia with varying conclusions. Rostami and Mehrabi, concluded that cord bilirubin is not useful in determining significant hyperbilirubinemia, this contrasts the findings of the current study as stated above.²⁷ Ahlfors et al maintained that the bilirubin/albumin ratio could provide a good estimate of free bilirubin,²⁸ while Sharma et al in a study of healthy term neonates found that cord bilirubin/albumin ratio was a better predictor compared to cord albumin and cord bilirubin.²⁹ Bhat et al concluded that cord bilirubin, cord albumin and bilirubin/albumin ratio are excellent at predicting neonatal hyperbilirubinemia with cord albumin being the best of the parameters.²⁰

Limitations

The low number of subjects with significant hyperbilirubinemia at follow-up under-powers this study and thus a major draw-back. Moreso, the follow-up period for this study was 72-96 hours. Longer period of follow up would have enhanced detection of babies who could developed significant hyperbilirubinemia beyond the 96 hours of life. The non-inclusion of preterm and low birth weight babies in the recruitment for this study is a limit generalization of our findings. Future research should include a broader population of newborn such as low birth weight and preterm babies, so that the result can be more representative. A multi-centre study involving all the geopolitical zones in the country would also give a more encompassing view of the predictive ability of cord bilirubin/albumin ratio for significant hyperbilirubinemia.

CONCLUSION

The cord bilirubin/albumin ratio and cord bilirubin are weak in predicting significant hyperbilirubinemia at 72-96 hours follow-up, they are of limited usefulness in the early identification of babies at risk of significant hyperbilirubinemia. However, cord bilirubin is better than cord bilirubin/albumin ratio in predicting significant hyperbilirubinemia in term neonates. Cord albumin is also of limited usefulness in early identification of babies at risk of significant hyperbilirubinemia.

Funding: No funding sources

Conflict of interest: None declared

Ethical approval: The study was approved by the Institutional Ethics Committee

REFERENCES

1. 1. Habib HS. Impact of discharge timings of healthy newborns on the rates and etiology of neonatal hospital readmissions. J Coll Physicians Surg Pak. 2013;23(10):715-9.
2. 2. Ipek IO, Bozaykut A. Clinically significant neonatal Hyperbilirubinemia: an analysis of 646 cases in Istanbul. J Trop Pediatr. 2008;54(3):211-3.
3. 3. Watchko JF. Bilirubin-Induced Neurotoxicity in the Preterm Neonate. Clin Perinatol. 2016;43(2):297-311.
4. 4. Keren R, Bhutani VK, Luan X, Nihtianova S, Cnaan A, Schwartz JS. Identifying newborns at risk of significant hyperbilirubinemia: a comparison of two recommended approaches. Arch Dis Child. 2005;90(4):415-21.
5. 5. Hansen TWR. Prevention of neurodevelopmental sequelae of jaundice in the newborn. Dev Med Child Neurol. 2011;53(4):24-8.
6. 6. Puppalwar P, Goswami K, Dhok A. Review on ‘Evolution of Methods of Bilirubin Estimation’. J Dental Med Sci. 2012;1(3):2278-84.
7. 7. Adiga DUS. Icteric index and its significance. Int J Med Heal Res. 2016;2(4):32-4.
8. 8. Benitz WE, Committee on Fetus and Newborn, American Academy of Pediatrics. Hospital stay for healthy term newborn infants. Pediatrics. 2015;135(5):948-53.
9. 9. Oyapero O, Njokanma F, Disu E. Relevance of the jaundice meter in determining significant bilirubin levels in term neonates at a tertiary hospital in Lagos State. Arch Med Heal Sci. 2018;6(2):232.
10. 10. Ogunlesi TA, Ogunfowora OB. Predictors of Acute Bilirubin Encephalopathy Among Nigerian Term Babies with Moderate-to-severe Hyperbilirubinemia. J Trop Pediat. 2011;57(2):80-6.
11. 11. Mansor MN, Yaacob S, Hariharan M, Basah SN, Jamil SHFSA, Khidir MLM, et al. Jaundice in Newborn Monitoring using Color Detection Method. Procedia Engineering. 2012;29:1631-5.
12. 12. Odutolu Y, Emmerson AJ. Low bilirubin kernicterus with sepsis and hypoalbuminemia. BMJ Case Rep. 2013;2013:bcr2012008042.
13. 13. Xie B, da Silva O, Zaric G. Cost-effectiveness analysis of a system-based approach for managing neonatal jaundice and preventing kernicterus in Ontario. Paediatr Child Health. 2012;17(1):11-6.
14. 14. Khairy M, Abuelhamd W, Elhawary I, Mahmoud-Nabayel A. Early predictors of neonatal Hyperbilirubinemia in full term newborn. Pediatr Neonatol. 2019;60(3):285-90
15. 15. Iskander I, Gamaleldin R, El Houchi S, El Shenawy A, Seoud I, El Gharbawi N, et al Serum Bilirubin and Bilirubin/Albumin Ratio as Predictors of Bilirubin Encephalopathy. Pediatrics. 2014;134(5):e1330-9.
16. 16. Charan J, Biswas T. How to Calculate Sample Size for Different Study Designs in Medical Research. Indian J Psychological Med. 2013;35(2):121-6.
17. 17. Brits H, Adendorff J, Huisamen D, Beukes D, Botha K, Herbst H, et al The prevalence of neonatal jaundice and risk factors in healthy term neonates at National District Hospital in Bloemfontein. Afr J Prim Health Care Fam Med. 2018;10(1):1582.
18. 18. Akbar M, Tjipta GDS. Cord Blood Albumin correlation as a predictor of neonatal hyperbilirubinemia in Healthy neonate. Azebaijan Med J. 2020;62(10)5769-75.
19. 19. L. Venkatraman, V. Anand, Jethifa M.P. A Prospective Cohort Study On Predictive Markers Of Neonatal Hyperbilirubinemia In Cord Blood. Int J Acad Med Pharm. 2023;5(1):546-52.
20. 20. Bhat JA, Sheikh SA, Ara R. Cord blood bilirubin, albumin, and bilirubin /albumin ratio for predicting subsequent neonatal Hyperbilirubinemia. PI. 2019;59(5):244-51.
21. 21. Deghady SAR, Rowisha MA, Ibrahim AM, Kishk WA. Cord Blood Bilirubin, Albumin and Bilirubin/Albumin Ratio versus Hydrogen Peroxide as Early Predictors of Significant Neonatal Hyperbilirubinemia. J Adv Med Med Res. 2020;38(18):34-42.
22. 22. Ramteke S, Shrivastav J, Agrawal A, Mishra NR, Saravanan AT, Tikkas R. Comparison of cord bilirubin and bilirubin albumin ratio to predict significant Hyperbilirubinemia in healthy full-term neonates. IJCH. 2018;05(02):108-11.
23. 23. Aktas S, Dogan C, Okmen ZH, Gulec SG. Is Cord Blood Bilirubin Level a Reliable Predictor for Developing Significant Hyperbilirubinemia. Am J Perinatol. 2019;36(3):317-21.
24. 24. Knüpfer M, Pulzer F, Gebauer C, Robel-Tillig E, Vogtmann C. Predictive value of umbilical cord blood bilirubin for postnatal hyperbilirubinemia. Acta Paediatrica. 2007;94(5):581-7.
25. 25. Rehna T, Shiyas K. Comparison of Umbilical Cord blood Bilirubin (UCB) and Bilirubin Albumin Ratio (BAR) in Predicting Neonatal Hyperbilirubinemia: A Prospective Observational Study. IJNMR. 2021;10(4):PC05-9.
26. 26. Sun G, Wang YL, Liang JF, Du LZ. Predictive value of umbilical cord blood bilirubin level for subsequent neonatal jaundice. Zhonghua Er Ke ZaZhi. 2007;45(11):848-52.
27. 27. Rostami N, Mehrabi Y. Identifying the newborns at risk for developing significant Hyperbilirubinemia by measuring cord bilirubin levels. J Arab Neonatal Forum. 2022;2(6):81-5.
28. 28. Ahlfors CE. Predicting bilirubin neurotoxicity in jaundiced newborns. Curr Opinion Pediatr. 2010;22(2):129-33.
29. 29. Sharma IK, Kumar D, Singh A, Mahmood T. Ratio of cord blood bilirubin and albumin as predictors of neonatal hyperbilirubinemia. Clin Exp Hepatol. 2020;6(4):384-8.