Background and objective:
Radiographic cerebral vasospasm (CVS) after subarachnoid haemorrhage (SAH) may occur in the absence of clinical deficit and vice-versa. A practical method for measuring cerebral blood flow (CBF) at the bedside in the ICU is still lacking. New techniques combining near infra-red spectroscopy (NIRS) and indocyanine green (ICG) dye dilution to estimate cerebral haemodynamics have recently been developed. The aims were: (1) to establish the new method in the clinical setting and (2) to estimate its value in detection and treatment of CVS.
Methods and materials:
Four NIRS optodes were placed bilaterally on the forehead (extended prototype, Oxymon system, Artinis, Njimegen). After daily central venous injections of 0.5 mg kg−1 ICG dye dilution curves were recorded and regional values for the mean transit time (mttlCG), cerebral blood volume (CBV) and CBF were quantified. Data are compared with U-test.
No side-effects occurred. In all, 135 ICG injections in 26 patients were performed. In three patients no values of cerebral haemodynamics could be calculated because of artifacts. Mean values obtained over both hemispheres were for mttlCG 9.4+2.5s (n = 182), for CBV 3.4+2.1mL per 100g (n = 177) and for CBF 22.6+14.5 mL per 100 g min−1 (n = 177). In all, 9 of 23 patients developed symptomatic CVS confirmed with angiography. Mean values for CBV and CBF were significantly lower with the presence of CVS (for CBV 3.1+2.1 vs. 4.5+2.4 mL per 100 g; P = 0.0145 and for CBF 19.6+14.3 vs. 29.9+15.5mL per 100 g min−1; P = 0.0029). There were no significant differences between mttlCGmean with and without CVS.
The new technique is a promising method for serial bedside CBF measurements in the ICU. It has the advantage of being non-invasive and does not require the patient to be transported. The technique could be a powerful tool in detection and treatment of CVS. Investigations in a larger set of patients are needed to evaluate its diagnostic accuracy in detection of CVS.