The venous admixture component of the total alveolar-arterial gradient for oxygen (AaD-O2) has been measured in 26 normal infants less than 4 days of age and in 12 others with the respiratory distress syndrome (hyaline membrane disease). The AaD-O2 in normal air-breathing infants (average 28 mm Hg) is nearly three times that seen in adults. Analysis of mixing data from N2-washout curves in these infants suggests such excellent distribution of ventilation that the distribution component of the AaD-O2 must be quite small. By contrast, estimates of the shunt component during oxygen breathing reveal a shunt flow in normal infants of nearly one-fourth cardiac output (AaD-O2 = 311) which is further increased in distressed babies (up to two-thirds cardiac output) and which can completely account for the large AaD-O2's found in both groups of infants. Submitted on July 9, 1962

"Trapped gas" has been defined by Bedell, Marshall, DuBois, and Comroe (1) as "that component of the thoracic gas volume which, having entered the thorax, is present in the lungs or pleural space but is unable to leave owing to intrapulmonary mechanical factors which prevent gaseous outflow." By measuring the ventilated portion of the thoracic gas volume (i.e., the functional residual capacity, FRC) with conventional dilutional methods (2) and comparing it to the total (ventilated and nonventilated) thoracic gas volume (TGV) as determined by their compression-decompression plethysmographic technique (3), these investigators demonstrated trapped gas in several adult patients with blebs, cysts, pneumothorax, intrapulmonary tumors, and especially, emphysema. They conclude that gas which is trapped in alveoli (as opposed to intrapleural or cystic sites) must exist in regions that are in intermittent communication with the airway.Recent measurements indicate a discrepancy between FRC (4) and TGV (5-7) in normal newborn infants. Thus, the data of Geubelle and associates (4) for infants over 1 hour of age yield an average FRC of 26.5 + 5.7 ml per kg as determined by the closed-circuit, helium-dilution technique (8), while Klaus, Tooley, Weaver, and Clements (5) reported a mean TGV of 29 + 7.0 ml per kg, and Auld and colleagues (7) reported 33.8 ± 7.3 ml per kg, as determined by *

The oxygen-dissociation curve of human fetal blood has been the subject of several previous investigations (1)(2)(3)(4)(5)(6). At saturations below 80% the curve has been described with considerable accuracy and, as in other mammalian young, is known to be displaced to the left of that for adult blood although both curves share a similar shape. Most studies have been carried out in vitro, i.e., cord blood specimens have been tonometrically equilibrated at various oxygen tensions, and the oxygen saturation has been measured.Beer, Bartels, and Raczkowski (4) and Rooth, Sj6stedt, and Caligara (5) have investigated the in vivo oxygen-dissociation curve by determining the actual oxygen and saturation at the time cord blood is sampled. Unfortunately, the curves so described do not extend into a range of oxygen content physiologic for the normal newborn infant. It is in just this range along the horizontal upper limb of the curve that precise information is necessary for calculation of venous admixture and the diffusion capacity of the lung for oxygen.One in vitro curve for fetal blood has been published, which extends well into the physiologic range (6). The present investigation represents an attempt at accurate delineation of a comparable in vivo curve. MethodsUmbilical arterial blood was obtained by catheter from 13 newborn infants. The infants comprised a crosssection of a typical newborn population (full-term and * Submitted for publication September 11, 1963; accepted November 29, 1963 (4,6). Darling, Smith, Asmussen, and Cohen (3) had found no difference between maternal and fetal blood with respect to the effect of pH change on the placement of the oxygendissociation curve, but a more recent investigation (11) indicates that there may be a significant difference in response to changes in pH.1 5 = percentage of oxygen saturation; CaO2= arterial oxygen content; PaO2 = arterial oxygen tension; Cap = oxygen capacity. 606

ExtractThe thermal balance of 22 full term, 9 small for date, and 8 large premature infants was studied by gradient layer direct calorimetry on the 1st day of life. In addition, the influence of muscular activity on the ability to sweat of 8 full term infants was investigated. The same measurements were repeated in 20 infants on the 3rd day of life.All i nfants were subjected to the same experimental conditions: ambient temperature 37°, restive humidity 50%, and air flow 30 liters/min.It was demonstrated that the evaporation of water was the primary way to dissipate heat in a heat-gaining environment. Active thermoregulatory sweating was elicited at deep body temperatures of 37.45°, 37.8°, and 37.76° in the full term, small for date, and premature infants, respectively.The set-point temperature of sweating tended to be lower with increasing gestational and postnatal age, and with vigorous muscular activity. The physiologic weight loss of about 3%, and the intrauterine growth retardation resulted in significant increase in this threshold temperature.The maximal sweat rate of the premature infants (1.43 W/kg) was significantly lower than that of the full term (1.81 W/kg) and small for date infants (2.32 W/kg): P < 0.05 and 0.01, respectively.Heat loss by radiation and convection was directly related to the mean skin temperature; it was only when the latter reached the value of 37° that radiative plus convective heat loss became positive.There was a close inverse relation between the total heat loss and heat storage. After the beginning of sweating, the heat storage of full term and small for date infants sharply decreased and stabilized around 0, which showed that a new thermal steady state was reached. The premature infant was not able to reach this new equilibrium and the heat storage remained positive.The full term and small for date infants could keep the body temperature at a steady level of 37.66° and 37.98°, respectively, whereas the esophageal temperature of the premature infants kept on increasing.The cutaneous thermal conductance, an indicator of the heat-transferring blood flow, was at first negative, which showed that heat was transferred from the surface to the core. Later, before the evaporative heat loss, it increased and reached the maximal value at a body temperature of 37.6° in the normal, and of 37.9° in both the small for date and premature groups.The metabolic rate for the small for date and premature infants was significantly higher than that for the full term infants, and it directly related to the deep body temperature in each group. Neonatal thermal balance 889In the new thermal equilibrium, the metabolic rate of the full term infants was about 20% lower than that measured under neutral ambient conditions. We suggest that the term "neutral thermal environment" must be reconsidered. SpeculationA new method, gradient layer direct calorimetry, gave us the opportunity to measure directly the different heat losses of the neonate under well controlled experimental conditions, and to compare thes...