Direct observation of human microcirculation during decompressive craniectomy after stroke

Objectives: Most knowledge related to the pathophysiology of microcirculation in ischemic stroke comes from experimental research. Unfortunately, data on microcirculation in the human brain are limited, partially as a result of the lack of appropriate investigational techniques. The objective of our study was to test the hypothesis that cortical microcirculatory alterations in the brain, in terms of blood flow and vessel density, occur in patients with stroke who require surgical decompression compared with a control group. Design: Prospective and observational study. Setting: Third-level university hospital. Patients: Six patients who had undergone decompressive surgery as a result of a space-occupying hemispheric infarction. These patients were compared with five patients who had undergone craniotomy for a disease not affecting the cortex. Interventions: Cortical microcirculation in the brain was directly observed using sidestream dark-field imaging. All images were analyzed offline. Measurements and Main Results: In patients with stroke with a space-occupying hemispheric infarction, 18 good-quality movie images were compared with 25 control group images. In the control group, cortical vessels showed a continuous flow in small, medium, and large vessels compared with patients with stroke who presented intermittent or no flow in all vessels. The proportion of perfused vessels was near 100% in control subjects and 63.44% in patients with stroke. The perfused vessel density index was also higher in control subjects (6.16 1/mm; interquartile range, 5.65-7.56) than in patients with stroke (2.77 1/mm; interquartile range, 1.75-3.86). Conclusion: Sidestream dark-field imaging allowed direct visualization of cerebral microcirculatory alterations in the operating room. This technique allowed the documentation of a significant blood flow reduction in the cortical microvascular and a decreased vascular density in patients with stroke compared with control subjects. (Crit Care Med 2011; 39:1126-1129)