It is well-established that neutrophils can act as important mediators of immunopathology and it has been previously demonstrated that ROS production is elevated in patients with COVID-19 [9]. Here, we sought to assess the prognostic impact of ROS measurement in relation to mortality in a cohort of patients admitted to ICU with COVID-19.

Our study confirmed previous work that COVID-19 values at entry to ICU are markedly elevated compared to healthy donors and reflect the high levels of peripheral ROS production in patients with pulmonary complications of COVID-19 [9]. The reasons for increased ROS production in COVID patients are likely to be multi-factorial and may relate to the immature or dysfunctional mature neutrophil phenotype reported in severe COVID-19 or direct activation of neutrophils through ACE2 engagement. It is uncertain if ROS is a direct mediator of tissue damage but neutrophil infiltration is thought to contribute to pulmonary complications and ROS production initiates a range of downstream activities, including generation of neutrophil extracellular traps (NETs) and stimulation of production of pro-inflammatory cytokines [18, 23,24,25].

As expected, LIT values were strongly correlated with the peripheral neutrophil count, as this population is the dominant source of ROS and this relationship has been reported previously [9]. However, we also observe a strong association with CRP values, an important systemic marker of inflammation which is synthesized in the liver. As such, the degree of ROS production is directly related to the overall inflammatory profile but whether ROS acts as a causative factor, or a simple correlate of the inflammatory state, is currently unknown.

The major aim of our study was to assess the utility of the near-patient LIT assay for ROS production as a predictive biomarker for clinical outcome in patients with COVID-19. In particular, within the study cohort of 44 patients, 24 (54.5%) survived the ICU admission and were discharged into ward care, whilst 20 patients (45.5%) died whilst on ICU (mortality timepoint cutoff: 42 days).

We initially determined the prospective neutrophil counts over time within these two groups. As expected from prior literature [18], neutrophil counts were somewhat higher within the mortality cohort, but the confidence interval overlapped, so that this was not an effective predictive marker. Elevated CRP values are also seen in patients with severe COVID-19 and a trend towards increased values within the mortality cohort was seen between days 1 and 10 of ICU admission, although the predictive value was not sustained.

LIT values had better prognostic value and diverged consistently, and with increasing magnitude, between the two study groups (survivors and non-survivors) beyond day 7 of infection. As such, the measurement of LIT values may represent an important prognostic marker for this disorder. An even more noticeable effect was seen in relation to LIT/N values which diverged markedly after day 5 of ICU admission, falling substantially within the group that survived, and with an effect that increased over time. These findings indicate that LIT/N measurement may represent a novel and highly predictive marker of clinical outcome for patients undergoing ICU care for COVID-19. Importantly, the LIT test has several technical advantages over current tests as it is a near-patient assay, delivers a result within 10 min from blood droplets from venepuncture samples or finger prick assays, and is inexpensive.

The statistical validity of this approach would need to be assessed in larger clinical studies, although this will be challenging to deliver at the current time due to the low rate of ICU admission for COVID-19 patients in the current era of vaccination and high levels of prior infection. As such, the utility of LIT/N should also be assessed in other clinical settings on ICU, such as patients with sepsis, where our initial data have already shown high levels of LIT/N at the time of ICU admission. As the LIT test is portable and inexpensive it can be used in hospitals, outreach clinics and in the field, and raises the additional possibility of widening access of information available to clinicians in isolated regions and in low- and middle-income countries. In particular, if predictive value is confirmed and defined, it could potentially be used to triage patients, thus freeing up vital resources.

Our study has a number of limitations. The cohort size is modest, although the denominator of 44 patients with severe disease is larger than many other prognostic assessments and was undertaken when the incidence of severe disease was falling. The demographic details of the study group are also somewhat limited, and we do not have long-term clinical follow-up data. Finally, ROS production is only one measure of neutrophil activation and additional measurements may serve to increase the predictive power of this phenotype. For example, while this study did not directly investigate the role of neutrophil extracellular traps (NETs), future research may benefit from studying the correlation between ROS generation and NETs.

For decades, the clinical assessment of neutrophils within disease has relied on the quantification of cells within blood, aided in some situations with morphological analysis of a blood smear. Crucially, these fail to assess neutrophil function, and here we use the LIT test and corrected LIT/N values to assess this rapidly at the patient bedside. We show that these values appear highly promising as predictive markers of mortality in a patient group on ICU. Given that neutrophils are known to orchestrate cellular damage in various disease conditions, such as sepsis [15, 26], the LIT/N biomarker has the potential to be a powerful prognostic tool in assessing disease severity in those conditions as well. As such we propose that larger studies are undertaken in conditions, such as sepsis to assess the potential importance of these values to aid prognosis, and potentially direct treatment pathways, in disorders, where neutrophil activation may act as a determinant of disease course.