Brain Oxygen Levels Stay Low in Hemorrhage Patients Despite Normal Pressure Monitoring

Critically ill patients with spontaneous intracerebral hemorrhage can show persistently low oxygen levels in brain tissue that appears structurally untouched, even when every conventional monitor suggests the brain is being adequately perfused. That is the central observation of a prospective, single-center observational cohort study conducted in a dedicated neurocritical care unit between January 2021 and March 2023, published in the journal Neurocritical Care. The researchers enrolled consecutive adults with spontaneous intracerebral hemorrhage who required invasive multimodal brain monitoring and analyzed their neuromonitoring data during the first 48 hours and between days 3 and 7.

The team placed particular emphasis on brain tissue oxygen pressure, abbreviated PbtO2, a parameter measured by a probe inserted directly into brain parenchyma. While intracranial pressure and cerebral perfusion pressure are standard components of neurocritical monitoring, the authors note that whether PbtO2 adds information beyond these conventional parameters in spontaneous intracerebral hemorrhage has remained unclear. To address that gap, they monitored intracranial pressure, cerebral perfusion pressure, the pressure reactivity index known as PRx, bedside optimal cerebral perfusion pressure, and PbtO2 in the contralateral hemisphere, the side opposite the bleeding.

A total of 31 patients were included in the study, of whom 18 underwent PbtO2 monitoring. The investigators analyzed neuromonitoring data during two windows: the first 48 hours and a later interval between days 3 and 7. Paired comparisons were used to assess temporal changes, and linear regression models were used to explore the relationship between PbtO2, cerebral perfusion pressure and the pressure reactivity index.

The conventional parameters told a reassuring story. Intracranial pressure remained within normal ranges throughout the monitored periods. Cerebral perfusion pressure increased significantly from the first 48 hours to days 3 through 7, and the pressure reactivity index improved over time, a pattern the authors interpret as better cerebrovascular autoregulation as the patients progressed through the acute phase. By these measures, the hemodynamic management of these patients appeared to be succeeding.

The oxygen measurements told a different story. PbtO2 values remained close to hypoxic thresholds and did not differ significantly between the two monitoring periods. During the first 48 hours, median PbtO2 was 21.72 millimeters of mercury, with an interquartile range of 18.67 to 27.04. That figure is notable because it was recorded despite preserved systemic oxygenation in these patients. When the researchers examined the relationship between brain tissue oxygen and the ratio of arterial oxygen partial pressure to inspired oxygen fraction, the standard index of lung gas exchange, they found no significant correlation. In other words, lungs that were delivering oxygen satisfactorily did not translate into brain tissue that was well oxygenated.

The regression analysis added a further layer of nuance. During the first 48 hours, the association between PbtO2 and cerebral perfusion was more evident than in the later window, although visual inspection of the data suggested a nonlinear relationship. Higher PbtO2 values clustered around a relatively narrow range of cerebral perfusion pressure. The authors suggest that PbtO2 may be closely related to cerebral perfusion dynamics and could serve as a complementary tool for individualized neurocritical care management, providing information that intracranial pressure and perfusion pressure alone do not capture.

In these patients, PbtO2 values remained close to hypoxic thresholds while PRx improved over time, a pattern the authors interpret as better cerebrovascular autoregulation.

The authors are careful about the limits of their data. Because the PbtO2 probe was positioned in the contralateral hemisphere, no inferences regarding perihematomal oxygenation, the tissue immediately surrounding the hematoma, can be made from these findings. The single-center design and the modest sample size, with 18 of 31 patients monitored for brain tissue oxygen, also constrain how broadly the results can be generalized. The observational design means the study can describe associations between monitored parameters but cannot establish that manipulating one parameter would change outcomes.

Even with those caveats, the study suggests a potential role for PbtO2 as a complementary tool in neurocritical care. In this cohort, patients whose intracranial pressure remained within normal ranges, whose cerebral perfusion pressure increased over time and whose pulmonary gas exchange was preserved still had brain tissue oxygen values close to hypoxic thresholds. The authors conclude that PbtO2 may be closely related to cerebral perfusion dynamics and may provide information beyond intracranial pressure and cerebral perfusion pressure alone. The authors state that PbtO2 may have a potential role as a complementary tool for individualized neurocritical care management in spontaneous intracerebral hemorrhage, while noting that no inferences regarding perihematomal oxygenation can be made from these data.

Subject of Research: Brain tissue oxygen monitoring in the uninjured hemisphere of patients with spontaneous intracerebral hemorrhage

Article Title: Brain Tissue Oxygenation in the Contralateral Hemisphere of Critically Ill Patients with Spontaneous Intracerebral Hemorrhage: A Prospective Observational Study

Article References: Ferreira, A., Dias, M., Silva, P. A., Monteiro, E., & Dias, C. (2026). Brain Tissue Oxygenation in the Contralateral Hemisphere of Critically Ill Patients with Spontaneous Intracerebral Hemorrhage: A Prospective Observational Study. Neurocritical Care. https://doi.org/10.1007/s12028-026-02705-x

Image Credits: AI Generated

DOI: 10.1007/s12028-026-02705-x

Keywords: intracerebral hemorrhage, brain tissue oxygenation, PbtO2, neurocritical care, intracranial pressure, cerebral perfusion pressure, pressure reactivity, multimodal monitoring, cerebral autoregulation, observational cohort study, critical care, oxygen diffusion