Object: Proinflammatory adhesion molecule expression has been demonstrated to be elevated in patients with aneurysmal subarachnoid hemorrhage (SAH). Recent studies have shown that elevations in soluble intercellular adhesion molecule-1 (ICAM-1) may be predictive of poor outcome in patients with good grade (Hunt and Hess Grades 1-2) aneurysmal SAH at delayed time points that correspond with the risk period for cerebral vasospasm. In addition, ICAM-1 is upregulated in experimental models of vasospasm. Unfortunately, the relationship of adhesion molecule expression to human vasospasm remains unclear. The authors hypothesized that the delayed elevation of soluble ICAM-1 in patients with aneurysmal SAH is associated with the development of cerebral vasospasm.
Methods: Eighty-nine patients with aneurysmal SAH were prospectively enrolled in a study and stratified according to the presence or absence of vasospasm, as evidenced by daily monitoring of transcranial Doppler (TCD) velocities (presence, > 200 cm/second; absence, < 120 cm/second). Levels of soluble ICAM-1 were determined using enzyme-linked immunosorbent assay every other day for 12 days post-SAH. An analysis of covariance model was used to evaluate trends in soluble ICAM-1 levels from 2 days prior to 6 days after the occurrence of documented vasospasm. Two groups of patients, matched for admission admission Hunt and Hess grade, were compared: nine patients with TCD velocities greater than 200 cm/second and nine patients with TCD velocities less than 120 cm/second. From among the patients with TCD velocities greater than 200 cm/second six patients with angiographically documented vasospasm were selected. Patients with TCD velocities less than 120 cm/second and matched admission Hunt and Hess grades but without angiographically documented vasospasm were selected. Patients with TCD-demonstrated vasospasm showed a significant mean rate of rise (p < 0.01) in soluble ICAM-1 levels during the perivasospasm period, but admission Hunt and Hess grade-matched control patients did not (p = not significant [NS]). There was a significant difference between these groups' rates of soluble ICAM increase (p < 0.01). Patients with both TCD- and angiographically demonstrated vasospasm likewise showed a highly significant mean rate of increase in soluble ICAM-1 levels during the perivasospasm period (p < 0.01), whereas admission Hunt and Hess grade-matched controls did not (p = NS). The difference beween these groups' rates of increase was highly significant (p < 0.001).
Conclusions: These data suggest a role for ICAM-1 in the pathophysiology of cerebral vasospasm or its ischemic sequelae. As this relationship is further elucidated, adhesion molecules such as ICAM-1 may provide novel therapeutic targets in the prevention of vasospasm or its ischemic consequences.