BACKGROUND Sepsis-associated brain dysfunction (SABD) is frequent and is associated with poor outcome. Changes in brain hemodynamics remain poorly described in this setting. The aim of this study was to investigate the alterations of cerebral perfusion pressure and intracranial pressure in a cohort of septic patients. METHODS We conducted a retrospective analysis of prospectively collected data in septic adults admitted to our intensive care unit (ICU). We included patients in whom transcranial Doppler recording performed within 48 h from diagnosis of sepsis was available. Exclusion criteria were intracranial disease, known vascular stenosis, cardiac arrhythmias, pacemaker, mechanical cardiac support, severe hypotension, and severe hypocapnia or hypercapnia. SABD was clinically diagnosed by the attending physician, anytime during the ICU stay. Estimated cerebral perfusion pressure (eCPP) and estimated intracranial pressure (eICP) were calculated from the blood flow velocity of the middle cerebral artery and invasive arterial pressure using a previously validated formula. Normal eCPP was defined as eCPP ≥ 60 mm Hg, low eCPP was defined as eCPP < 60 mm Hg; normal eICP was defined as eICP ≤ 20 mm Hg, and high eICP was defined as eICP > 20 mm Hg. RESULTS A total of 132 patients were included in the final analysis (71% male, median [interquartile range (IQR)] age was 64 [52-71] years, median [IQR] Acute Physiology and Chronic Health Evaluation II score on admission was 21 [15-28]). Sixty-nine (49%) patients developed SABD during the ICU stay, and 38 (29%) were dead at hospital discharge. Transcranial Doppler recording lasted 9 (IQR 7-12) min. Median (IQR) eCPP was 63 (58-71) mm Hg in the cohort; 44 of 132 (33%) patients had low eCPP. Median (IQR) eICP was 8 (4-13) mm Hg; five (4%) patients had high eICP. SABD occurrence and in-hospital mortality did not differ between patients with normal eCPP and patients with low eCPP or between patients with normal eICP and patients with high eICP. Eighty-six (65%) patients had normal eCPP and normal eICP, 41 (31%) patients had low eCPP and normal eICP, three (2%) patients had low eCPP and high eICP, and two (2%) patients had normal eCPP and high eICP; however, SABD occurrence and in-hospital mortality were not significantly different among these subgroups. CONCLUSIONS Brain hemodynamics, in particular CPP, were altered in one third of critically ill septic patients at a steady state of monitoring performed early during the course of sepsis. However, these alterations were equally common in patients who developed or did not develop SABD during the ICU stay and in patients with favorable or unfavorable outcome.