Recently, tissue oxygenation in pediatric heart surgery is measured by using near-infrared spectroscopy. Monitoring of cerebral oxygen saturation (ScO2) is most common but that of somatic tissue oxygen saturation (SrO2) is also gradually becoming widespread. However, the value of their monitoring is not well established. One of the reasons for this may be that the physiological factors affecting ScO2 and SrO2 have not been sufficiently clarified. Accordingly, we prospectively observed the changes in ScO2 and SrO2 simultaneously throughout cardiac surgery with cardiopulmonary bypass (CPB) in children weighing under 10 kg and evaluated their relationships with physiological parameters by using the random-effects model. ScO2 and SrO2 were measured with an INVOS 5100C (Somanetics, Troy, MI, USA). The random-effects analysis was applied for ScO2 and SrO2, as dependent variables, and seven physiological parameters (mean blood pressure, central venous pressure, rectal temperature, SaO2, hematocrit PaCO2, and pH) were entered as independent covariates. The analysis was performed during the pre-CPB, CPB, and post-CPB periods. Next, the same analysis was performed by dividing the patients into univentricular and biventricular physiological types. Forty-one children were evaluated. Through the whole surgical period, ScO2 correlated strongly with mean blood pressure regardless of the physiological type. On the other hand, the contribution of mean blood pressure to SrO2 was weak and various other parameters were related to SrO2 changes. Thus, the physiological parameters affecting ScO2 and SrO2 were rather different. Accordingly, the significance of monitoring of cerebral and somatic tissue oxygen saturation in pediatric cardiac surgery should be further evaluated.