NAHCO~EFFECTS ON BRAIN BLOOD FLOW 003 1-3998/85/1908-08 15$02.00/0 PEDIATRIC RESEARCH Vol. 19, No. 8, 1985 Copyright O 1985 International Pediatric Research Foundation, lnc Printed in U.S.A. The Effects of Sodium Bicarbonate on Brain Blood Flow and Oz Delivery during Hypoxemia and Acidemia in the Piglet ABBOT R. LAPTOOK Urziversifyof Texus Heultll Sciencr Center at Dullus, Department of Pediatrics, Dalla.~,Texas 75235 ABSTRACT. Metabolic acidosis in the neonate is often infants metabolic acidemia is commonly secondary to cardio8- secondary to hypoxemia and cardiopulmonary disturb- pulmonary disorders (e.g. hyaline membrane disease, pneu- ances. Sodium bicarbonate, an agent used to treat metabolic monia, asphyxia neonatorum, sepsis, etc.) and often hypoxemia acidemia in newborns, is often administered during hypox- accompanies metabolic acidemia. In the absence of acid-base emia. In the absence of acid-base alterations, during hy- alterations hypoxemia results in increases of CBF (2), the incre- poxemia a reciprocal relationship exists between arterial ments of which are reciprocally related to reductions in CaOz O2 content (Ca02)and brain blood flow (BBF). However, (3). Ca02represents the total Q2 carried by arterial blood and is when hypoxemia is compounded by acidemia it is unclear a function of Pa02,hemoglobin concentration, and hemoglobin- whether the increase in arterial pH achieved by infusions Oz saturation. Studies of hypoxia in newborn lambs, produced of sodium bicarbonate alters BBF. To investigate this, BBF by either hypoxemia or acute normovolemic anemia, demon- (microsphere technique), arterial blood gases, and CaOl strate that Ca02 appears to be a good predictor and possibly ,a were measured in 14 ventilated piglets. Variables were regulator of cerebral vascular responses (4). During hypoxemi,a assesed during a control period, a period of hypoxemia (50 the fall in CaOz is exacerbated when there is an associated min) associated with metabolic acidemia (hypoxemia + metabolic acidemia because of a pH mediated shift (down and acidemia), and after infusions of either saline (n = 6) or rightward) in the hemoglobin-O2 saturation curve (5). The role NaHC03 (n = 8, 2 mEq/kg) during continued hypoxemia. of hydrogen ion per se in the control of CBF remains controver- Arterial pH was similar in both groups at control, and sial. Conflicting results have been obtained from both newborn hypoxemia + acidemia resulted in comparable reductions (6) and adult animal models (7), and human adults (8) with of pH in both saline- and NaHC03-treated piglets (7.21 2 respect to correlations between CBF and changes in arterial pH 0.02 versus 7.21 L 0.03, respectively). NaHC03 infusions of metabolic origin. In contrast, it has been demonstrated usin:g produced a significant rise in pH, 7.30 2 0.03 versus 7.15 the cranial window technique in cats (9) and ventriculocisternr;d 2 0.03, p < 0.05. In each group Ca02 paralleled changes perfusion in dogs (lo), that CBF varies directly with changes in in pH but did not differ between groups. In all animals cerebral extraceliular fluid acidity. BBF increased more than 2-fold during hypoxemia + aci- Liberal use of sodium bicarbonate in neonates has been tem- dernia and was unaltered by infusions of either saline or porized by associated alterations of serum osmolality (1 l)?a role NaHC03. Brain O2 delivery decreased in both groups in the pathogenesis of intracranial hemorrhage (12), and effects during hypoxemia + acidemia and was unchanged by in- on the BBB (13). Despite prominent effects within the CNS it fusions of saline or NaHC03. During hypoxemia + acide- remains unclear whether sodium bicarbonate alters CBF. Sodiun~ mia the change in arterial p1-I induced by NaHC03 (2 bicarbonate could potentially affect CBF via osmolar effects (14:1, mEq/kg) does not alter BBF or brain O2delivery. (Pediatr transient elevations of arterial C02 tension (2, 15). and BB13 Res 19: 815-819,1985) disruption with secondary change in cerebral interstial fluid pH (9). Since sodium bicarbonate is often administered during hy- Abbreviations poxemia and acidemia, another potential mechanism is that thle rise in arterial pH induced by sodium bicarbonate may improvl: Ca02, arterial oxygen content hemoglobin-lo2 saturation sufficiently to increase Ga02and thus, BBF, brain blood flow CBF may decrease. Reductions in CBF during hypoxemia anti CBF, cerebral blood flow acidemia would appear to be detrimental considering the cerebral total carbon dioxide content vascular response to hypoxemia. Howcver, evaluation of tissue ECF, extracellular fluid O2 delivery, which is dependent on both blood flow and CaOz, BBB, blood-brain barrier may be important to assess whether changes in CRF are appro- MAP, mean arterial pressure priate. Therefore, to determine if sodium bicarbonate alters brain blood flow and 02 delivery, piglets with metabolic acidemia secondary to hypoxemia were studied before and after alkali treatment. Sodium bicarbonate has been recommended for the treatment of metabolic acidemia in neonates (I). In both preterm and term METHODS Received July 5. 1984; accepted March 14, 1985. Fourteen miniature swine were the subjects of this investiga- Reprint requests to Abbot R. Laptook M.D., Ilniversiry of Texas Health Science tion. Pregnant sows were housed and farrowed in the Animal Center at Dallas. Department of Pediatrics, 5323 Harry Hines Boulevard, Dallas, TX 75235. Resources Center of The University of Texas Neahh Science Supported by American Heart Association, National Chapter and American Center at Dallas. Newborns were kept with the sow until the Heart Assoc~ation.'Texas Affiliate, Grant 8373 1. morning of the study at which time one piglet was removed from 816 LAPTOOK the litter. Premedication consisted of a single dose of intramus- Wet tissue weight was used to express blood flow as ml/min. 100 cular ketamine (5 mg/kg) after which a tracheostomy was per- g. Regional brain blood flow (cerebrum, cerebellum, brainstem) formed with placement of a 3.0-mni endotracheal tube. The was calculated by the above formula using the respective regional piglet was ventilated (Harvard Apparatus Rodent Respirator, cpm in lieu of brain tissue cpm. Brain O2 delivery was derived model 680) with 70% nitrous oxide and 30% oxygen, using a by the product of BBF and GO2. By the Fick principle, cerebral tidal volume of 12 ml/kg and rates of 40-50 breathslmin. O2 uptake was equivalent to the product of cerebral blood flow Polyethylene catheters were placed in the left ventricle (via the and the cerebal arteriovenous difference of O2 content. left common carotid artery), abdominal aorta (via a femoral Statistical analysis was performed by analysis of variance with artery), inferior vena cava (via a femoral vein), and the left repeated measures (SAS Statistical Package) to compare the axillary artery. A 22-gauge Teflon catheter was inserted into the measured variables between groups. Significant interactions were sagittal sinus through a burr hole in the midline of the exposed localized by a Newman-Keuls multiple comparison procedure. calvarium. Statistical significance was designated at p < 0.05. Values re- All procedures were performed while maintaining rectal tem- ported are the mean and 1 SE. In one saline-infused piglet the perature between 38.5-39.5" C. Following catheter placement, reference sample withdrawal apparatus malfunctioned and blood the inspired gas was changed to 70% nitrogen and 30% 02, 0.3 flow analysis was limited to the 13 animals with complete data. mg/kg D-tubocurarine was given intravenously, and a 1-h stabi- lization period commenced. After stabilization a control measure RESULTS of BBF was obtained during normoxemia. EIypoxemia was then induced by decreasing the inspired O2 to 13%. Hypoxemia was Age and weight were comparable between groups at the time continued until arterial pH was less than 7.25 (assessed by serial of study (8 c 2 versus 8 + 1 days and 11.54 + 0.12 versus 1.77 + blood gases), at which time a second BBF determination was 0.13 kg for the saline- and bicarbonate-treated piglets, respec- performed. Once completed, the piglets were randomly divided tively). The changes in Pa02, arterial pH, and C02TOT are shown into two groups; eight piglets received an infusion of 0.5 M in Figure 1. Values for Pa02 were comparable between groups NaHC03 (2 mEq/kg) over 3 rnin via the femoral vein catheter, throughout the study. For all animals a significant reduction in whereas the remaining six piglets received a comparable volume Pa02 was observed at 10 min after initiation of 13% oxygen, of 0.9% NaCl (4 mllkg) over 3 min. Hypoxemia was continued falling to 37.7 + 1.3 mm Hg and was unchanged thereafter. The during the respective infusions. The third blood flow determi- time necessary for hypoxemia to result in metabolic acidemia nation was performed 10 min following the completion of either varied, ranging from 41 to 62 min; however, there were no NaHC03 or saline infusion. The ventilator rate was adjusted to differences between groups (48 +. 7 versus 51 +. 9 rnin for the maintain PaC02 constant (-35-40 mm Hg) throughout the saline- and bicarbonate-treated piglets, respectively). Therefore, experiment. Immediately prior to each microsphere injection, values for hypoxemia with acidemia are plotted as the mean blood was obtained for measurement of hematocrit, plasma value for all animals. at 50 rnin after the onset of hypoxemia. osmolality and sodium concentration, C:02TOT and cerebral ar- Arterial pH was comparable in both groups at control and fell teriovenous differences of O2 content, blood gases, and pH. At during hypoxemia to similar values at 60 min; however, during each blood flow measurement 4.0 ml of blood was removed for continued hypoxemia the respective infusions resulted in differ- blood analyses and the microsphere reference sample. After each microsphere injection packed red blood cells previously obtained CONTROL HYPOXEMIA + ACIDEMIA INFUSION from a donor piglet of the same litter and stored with ACD were slowly administered to replace blood loss.