Cardiac output can be measured using a modified carbon dioxide Fick (mCO2F) method. A validation study was performed comparing mCO2F method-derived cardiac output (Q(mCO2F)) with invasively measured pulmonary blood flow. In seven randomly bred ventilated newborn lambs, cardiac output was manipulated by creating hemorrhagic hypotension. When steady state was reached, Q(mCO2F) was measured. Gas analysis was performed in simultaneously obtained arterial and venous blood samples (right atrium [RA], superior vena cava [SVC], and inferior vena cava [IVC]). Carbon dioxide exchange and pulmonary blood flow was measured continuously using a CO2SMO Plus monitor and a pulmonary ultrasonic flow probe (Q), respectively. Mean bias, defined as Q(mCO2F) - Q(ufp), was small (respectively, -0.082 L.min, -0.085 Lx min(-1) and -0.183 Lxmin(-1) for venous sampling from RA, SVC, and IVC). The limits of agreement were -0.328 to 0.164 Lxmin(-1) (RA), -0.335 to 0.165 Lxmin(-1) (SVC), and 0.415 to 0.049 Lxmin(-1) (IVC). In conclusion, measurement of cardiac output with the mCO2F method is reliable and easily applicable in ventilated newborn lambs. For clinical use, the site of venous blood sampling is of minor importance.