Abstract
The article by Tsai et al. in Annals 1 describes a significant inverse association between total serum bilirubin and total white (blood) cell count (WCC) in a cohort of healthy Taiwanese adults. This relationship was detected in cross-sectional data for cases with total serum bilirubin within the reference interval and total WCC less than 15 × 109/L. We have investigated similar relationships in two distinct cohorts of Australian patients who had routine pathology conducted via Sullivan Nicolaides Pathology (Taringa, Queensland), including assessments of inflammatory markers.
The first cohort consisted of community patients (n = 980), and included cases tested for hepatitis B surface antigen (HBsAg) (n = 510). The second cohort consisted of intensive care unit (ICU) cases from Brisbane Hospital (n = 955). Ages ranged between 17 and 94 years for both cohorts, but with a higher mean age for the ICU cohort (66 years versus 40 years). Fifty-seven per cent (57%) of the ICU cohort was male, while 42% of the community cohort was male. Bivariate correlation (Pearson) was conducted to assess the association in each cohort between total serum bilirubin and total WCC (SPSS, version 22).
In order to reflect the Tsai et al. study, cases with elevated serum bilirubin (>20 µmol/L) and WCC (>12 × 109/L) were removed from the community cohort prior to analysis. A significant inverse correlation was detected between bilirubin and WCC (r = −0.195, P < 0.001, n = 833). Significant inverse correlations were also found separately for HBsAg positive cases (n = 340, r = −0.223, P < 0.001), and cases not tested for HBsAg (n = 402, r = −0.155, P < 0.005), but not the HBsAg negative cases (n = 91, r = −0.190, P > 0.07). The same analyses were performed on the total cohort (including cases with elevated bilirubin and/or WCC); no significant correlations were observed.
By contrast, in the ICU cohort, with all cases included, a significant positive correlation was found between total serum bilirubin and WCC (n = 955, r = 0.152, P < 0.001), whereas after the removal of elevated bilirubin and WCC cases, a significant association was not found (n = 498, r = 0.062, P > 0.16).
Our findings in the community cohort agree with the findings of the Tsai study. The association was also observed in a subcohort of HBsAg positive cases where increased acute inflammation might reasonably have been expected. We speculate that the inverse association between bilirubin and WCC seen in apparently healthy subjects reflects homeostasis in terms of normal variations of oxidative stress required to maintain health.
The opposite was found for high-dependency patients requiring intensive care. In this cohort, acute inflammation and multi-system disease processes resulted in a significant positive correlation between bilirubin and WCC. Consistent with this explanation, elevated inflammatory markers were seen in the ICU cohort, with ICU median CRP, ESR and WCC significantly higher compared with the community cohort (30 µg/mL versus 2 µg/mL; 18.5 mm h versus 6 mm h; and 11.1 × 109/L versus 6.8 × 109/L, respectively: all P < 0.001, Mann--Whitney U test).
Footnotes
Acknowledgements
We would like to thank Sullivan Nicolaides Pathology Brisbane for access to de-identified patient results.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
This research was funded by the Quality Use of Pathology Programme (QUPP), the Department of Health, Commonwealth of Australia.
Ethics approval
Approved by the Human Ethics Committees of Bond University (Robina, Queensland) and The Australian National University (Canberra, ACT).
Guarantor
TB.
Contributorship
TB researched literature; TB and BAL conceived the study. AR guided the statistical analysis, TB and BAL gained ethics approvals, TB patient recruitment and BAL performed the data analysis. TB wrote the first draft of the manuscript. All authors reviewed and edited the manuscript and approved the final version of the manuscript.