Objectives. To compare, using cerebral near-infrared spectroscopy (NIRS), the effects of blood withdrawal (W) and infusion (I) via, respectively, UAC with tip in the thoracic aorta and UVC. Method Twelve infants of 30 weeks (25-33) gestational age, 1175 g (760-1690) birth weight, mechanically ventilated for RDS were studied at 19 hours (2-25) postnatal age. The protocol involved W, through UAC of 3 ml/kg blood in 30 seconds, followed by I at the same rate, after recovery of the NIRS trace. The procedure was repeated 3 times for UAC and 3 times for UVC. NIRS (NIRO 500, Hamamatsu Photonics), a technique based on the differential absorption of NIR light by hemoglobin and cytochrome aa3 (CytO2), depending on their oxidation state, was used to assess changes in cerebral blood volume (DCBV), calculated from total hemoglobin, oxy- (DHbO2), deoxy-(DHb) hemoglobin and CytO2, time of max variation and recovery (from start of W and I). Vital parameters, including mean arterial blood pressure (MABP) were continuously measured. Results No significant change in Hb or CytO2, heart rate, pCO2 or pO2 was seen. Concordant changes in HbO2 and CBV indicate modifications of cerebral blood flow. MABP increased significantly during I-UVC and decreased during W-UVC. A positive correlation between MABP and CBV changes during W-UVC(r2 =0.41, p = 0.036) was observed. Conclusion Blood withdrawal and infusion via UVC cause significant MABP changes and related CBV oscillation which could potentially precipitate germinal matrix hemorrhage in preterm infants; utilizing the UAC this effect may be reduced or altogether avoided. Table

Table 1 No caption available.
Full size table