The randomized clinical trial, SafeBoosC III, evaluates the effect of treatment guided by cerebral tissue oximetry monitoring in extremely preterm infants. Treatment should be considered, when cerebral oxygen saturation (StO₂) drops below a predefined hypoxic threshold. This threshold value differs between different brands of instruments. To achieve high external validity, in this pragmatic trial all commercially available cerebral tissue oximeters have been accepted, provided their specific hypoxic threshold value has been determined in phantom studies. Since most companies produce sensors with an adhesive surface on the patient-contacting side, in the phantom studies these sensors were applied according to the specifications, i.e., the glossy cover was removed from the sensor. However, since the skin of preterm infants is particularly fragile, some neonatologists keep this cover on the adhesive sensors, to avoid the risk of skin injury when removing the sensor. Therefore, the aim of this study was to determine whether keeping this cover on leads to different StO₂ values. To evaluate the effect of the cover, we performed multiple deoxygenations in a blood-lipid phantom and compared an INVOS neonatal sensor (Medtronic), with and without the cover, to a reference oximeter (OxiplexTS, ISS). As expected, the relationship of the StO₂ between the INVOS neonatal sensor and OxiplexTS was linear (r² = 0.999) with and without cover, but the cover influenced the linear equation: StO_{2_INVOS_cover} = 1.133*StO_{2_ISS} + 7.1 as opposed to StO_{2_INVOS_nocover} = 1.103*StO_{2_ISS} + 12.0. Furthermore, the hypoxic SafeBoosC III threshold differed as well: 60.3% with cover and 63.8% without cover. In conclusion, keeping the adhesive cover on an INVOS neonatal sensor results in lower measured values. At the hypoxic threshold, this is more than 3% (from 60.3% to 63.8%), and therefore, if clinicians keep the cover on the sensor, they need to be aware of this difference.