Percentage HScore confirms low incidence of secondary haemophagocytic lymphohistiocytosis in hospitalised COVID-19 patients

Ardern-Jones, M. R.; Stammers, M.; Phan, H. T. T.; Borca, F.; Koutalopoulou, A.; Teo, Y.; Batchelor, J.; Smith, T.; Duncombe, A.

Objective: It has been assumed that a significant proportion of COVID-19 patients show evidence of hyperinflammation of which secondary haemophagocytic lymphohistiocytosis (sHLH) is the most severe manifestation. To facilitate diagnosis of sHLH the HScore has been developed and validated. We set out to examine the prevalence of sHLH-like hyperinflammation in COVID-19. Methods: We retrospectively examined HScore parameters in 626 COVID-19 cases admitted to our institute of which 567 were suitable for analysis and compared these to a cohort of confirmed infection associated sHLH cases. To account for missing data, we calculated the maximum possible HScore of the recorded parameters (%HScore). Results: Early measurement of HScore parameters (day -1 to 4 from diagnosis) strongly predicted the %HScore over the course of the admission (p <0.0001). The retrospective cohort of sHLH showed significantly higher %HScores as compared to COVID-19 (median 73.47 vs 18.13 respectively, p <0.0001). The overall prevalence of individuals with an 80% probability of sHLH in our COVID-19 cohort was 1.59% on admission and only rose to 4.05% during the whole disease course. In the small cohort with scores suggestive of sHLH, there was no excess mortality compared with the whole cohort. %HScores were higher in younger patients (p<0.0001) and did not reliably predict outcome at any cut-off value (AUROC 0.533, p=0.211; OR 0.99). Conclusion: Surprisingly, these findings show that sHLH-type hyperinflammation is not prevalent in COVID-19, and %HScores do not predict outcome. Therefore, new algorithms are required to optimise case selection for clinical trials of targeted anti-inflammatory interventions.