Background: Sepsis-associated brain dysfunction is a frequent disorder in septic patients and has a multifactorial pathophysiology. Cholinergic pathways and brainstem dysfunction may result in pupillary alterations. The aim of this study was to evaluate whether early assessment of the Neurological Pupil Index (NPiTM) derived from an automated pupillometry could predict mortality in critically ill septic patients.
Methods: Retrospective cohort study of adult critically ill septic patients admitted to the intensive care unit of a University Hospital; patients with acute or known brain damage were excluded. Patients' severity was assessed by the daily Sequential Organ Failure Assessment Score and the SOFA<inf>max</inf> (i.e., highest SOFA Score during the first five days) was computed. The worst NPi (i.e., lowest value from one eye) was collected daily and then computed over the first five days of assessment. Mortality was assessed at hospital discharge.
Results: A total of 75 patients were included over the study period (median age 67 [53-75] years and median SOFA Score at admission 10 [8-12]); 64 (85%) presented septic shock; 48 (64%) died at hospital discharge. The worst NPi during the first five days of sepsis was significantly lower in non-survivors compared to survivors (4.4 [3.6-4.6] vs. 4.5 [4.2-4.7]; P=0.042). The worst NPi was also significantly lower in high severity group (i.e., SOFA<inf>max</inf>≥12) when compared to others (4.4 [3.2- 4.5] vs. 4.5 [4.0-4.7] P=0.01). However, in the multivariate analyses, the NPi value was not independently associated with in-hospital mortality or high SOFA<inf>max</inf>.
Conclusions: In this study, no independent prognostic role of NPi was observed in septic patients. Further larger prospective studies are needed to better evaluate the role of automated pupillometry in this setting.