The purpose of this study was to determine the method of endotracheal suctioning (ETS) that resulted in the least compromise to the cerebrovascular status of adult patients with severe head injuries. A two-group (two vs three ETS), two-protocol (100% tidal volume [VT] vs 135% VT) design was used. The dependent variables were mean intracranial pressure (MICP), mean arterial pressure (MAP), cerebral perfusion pressure (CPP), heart rate (HR), and oxygen saturation (SaO2). By random assignment, 14 subjects were in the two-ETS group and 16 subjects were in the three-ETS group. Intracranial pressure response to ETS in these patients with head injury can be characterized as falling into three patterns: (1) a rise in baseline beginning with ETS and continuing throughout the ETS sequences; (2) intracranial pressure spiking during the suctioning component of the protocol; (3) a combination of both a rising baseline and spiking. There was a significant (p less than or equal to 0.001) increase from baseline for both two- and three-ETS groups with both hyperoxygenation protocols (100% VT vs 135% VT) for MICP, MAP, HR, and CPP. No significant difference was found for SaO2 for either of the protocols regardless of number of suction passes. No significant differences were found between two- and three-ETS groups for any of the dependent variables. All groups, however, regardless of number of suction passes, demonstrated a cumulative increase in MICP, MAP, and CPP with each consecutive suction sequence.(ABSTRACT TRUNCATED AT 250 WORDS)