Fig. 1
Position of probes on brain surface and near infrared optical topogram of weight coefficients from a healthy volunteer
Patients breathed through an OX-135 oxygen face mask (Atom Medical Corporation, Tokyo, Japan) and peripheral oxygen saturation (SpO2) was measured at the tip of the index finger using an OLV 3100 SpO2 monitor (Nihon Koden Corporation, Tokyo, Japan). Air was delivered through the oxygen face mask at a flow rate of 8 L/min, and OT measurements were started. After the OT values were stabilized, the patients inhaled air for 1 min, followed by oxygen at the same flow rate for 2 min, and then air once again at the same flow rate for 3 min. The SpO2 changed in a trapezoidal manner under these conditions. Changes in HbOxy associated with oxygen inhalation in the cerebral tissues were measured using OT [1].
Analysis of OT Data
The time courses of HbOxy data were analyzed as follows. Data from OT were processed using MATLAB R14 (Math Works Inc., Natick, MA, USA) and principal component analysis (PCA) was applied to the data. Signal interference has been identified and removed from extracerebral tissues using this technique [11]. The weights are the coefficients of each channel for the synthesis of the principal component, and they reflect the degree to which the data in each channel contribute to the principal component data. Cross-correlations between each component and SpO2 were calculated. The principal component with the highest correlation coefficient included changes in HbOxy associated with oxygen inhalation, and this was analyzed in detail. To evaluate the transmission of systemic SpO2 changes to cerebral tissues, the weight coefficients of the principal components were calculated for each channel. Contour maps (topograms) were drawn using weight coefficients positioned at each measurement point (Fig. 1).
Evaluation of Cerebral Ischemia
The laterality of OT was determined by comparing the channel inspections in the symmetrical positions on the topogram. The side with a lower weight coefficient was defined as the ischemic side. Statistical significance was determined using the nonparametric Mann-Whitney U test. Differences with P < 0.05 were considered significant. Two neurosurgeons and two radiologists determined the ischemic side on SPECT images. The ischemic sides in OT and SPECT were compared, and the possibility of identifying cerebral ischemia was investigated. The ischemic side of OT was compared with those of SPECT, and two neurosurgeons investigated the possibility of identifying ischemic findings and ischemic sides.
Results
Table 1 shows the results of the 29 patients who were assessed by OT and 123I-IMP SPECT on the same day. DIND developed in 7 (24 %) of the patients who had concurrent ischemic findings on both images. However, ischemia was detected on OT before DIND. Early ischemic findings were not evident in OT images of 10 patients, and none of them developed DIND. In 22 patients who did not develop DIND, OT and 123I-IMP SPECT image showed ischemic findings in 11 and 13 of them, respectively. Among the 13 patients with ischemic findings on 123I-IMP SPECT images, 11(84.6 %) were concurrent with those of OT and all 9 patients without ischemic findings on 123I-IMP SPECT images also had no ischemic findings on OT images. Among the 22 patients who did not develop DIND, the findings on OT and 123I-IMP SPECT agreed in 20 (90.9 %) of them. Among all 29 patients, 123I-IMP SPECT and OT images showed ischemic findings in 20 and 18 of them, respectively, and the findings were concurrent in 18 (90 %) of the 20 patients with such findings on 123I-IMP SPECT images. Among all 29 patients, 9 patients had no ischemic findings on either 123I-IMP SPECT or OT images. In total, the presence or absence of ischemia on OT and 123I-IMP SPECT images concurred in 27 (93.1 %) of the 29 patients.
Table 1
Patient characteristics
