003 I -399S19 1/3004-0368$03.00/U PEDIATRIC RESEARCH Vol. 30. No. 4. 199 1 Copyrig111ic) 1991 Inrcrnat~onalPed~atric Research Foundation. Inc. Pri~iie~IIn U.S'.,l. Effects of Acute Hypercapnia on Maternal and Fetal Vasopressin and Catecholamine Release1 DANIEL J. FAUCHER,? ABBOT R. LAPTOOK. JOHN C. PORTER, AND CHARLES R. ROSENFELD Deparrrnenrs cfPedialrics, Obs~erricsand Gynecology, and Physio1og.v. Tlre Univer.si/ji of Texas So~ltl~western Medical Center at Dallas, Dallas, Texas 75235 ABSTRACT. Although fetal asphyxia, i.e. hypoxemia, Alterations in the homeostatic milieu trigger a complex series acidosis, and hypercapnia, increases plasma arginine va- of events that serve to protect the individual organism against sopressin (AVP) >40-fold, hypoxemia and metabolic aci- environmental insults. Asphyxia, i.e. the simultaneous occur- dosis occurring independently cause only 5-fold and 2-fold rence of hypoxemia, acidosis, and hypercapnia, is a striking increases, respectively. To determine the effects of hyper- illustration of an intense "stress," and subsequent adaptive re- capnia on AVP release, we examined the effects of acute sponses reflect a complex interaction of factors related to the hypercapnia on AVP secretion in six pregnant sheep and metabolic, neuroendocrine, and cardiovascular systems. Al- their fetuses at 135 f 4 d (x f SD), exposing the ewe though these responses have been examined in adult animals of successively to room air, 30% 02,30% O2 plus 10% C02, several species (1-6), the ontogeny of these responses during fetal 30% 02,and room air, and monitoring uterine blood flow, development is not fully understood. as well as maternal and fetal mean arterial pressure, heart In fetal sheep the response to stress has been characterized by rate, arterial blood gases, and plasma AVP and catechol- increased secretion of AVP and adrenal catecholamines (7). By amines. Oxygen exposure had no effect on the ewe or fetus. using the fetal sheep as a model, we and others (5, 8-12) have During O2 plus COz exposure, the ewes and fetuses devel- observed that fetal asphyxia, regardless of method of induction, oped hypercapnia in the absence of hypoxia, arterial C02 results in substantial increases in plasma AVP concentrations, tension increasing to 8.38 f 0.87 kPa (62.9 f 6.5 mm Hg) i.e. >40-fold rises over baseline levels. In contrast, we have shown and 10.0 + 0.61 kPa (75.2 + 4.6 mm Hg) (p < 0.001), (I I) that during moderate fetal hypoxemia, i.e. a 45% decrease respectively, at 30 min of exposure. Although fetal heart in Pa02without alterations in either arterial pH or PcoZrplasma rate and mean arterial pressure were unchanged, maternal levels of AVP rose only modestly (5-fold over baseline values). values rose 61 and 30% (p < 0.001), respectively. At 30 Furthermore, induction of marked metabolic acidemia had little min of O2 + C02 exposure, maternal norepinephrine in- effect on fetal AVP secretion, i.e. only a 2-fold rise over baseline creased from 2.23 + 0.74 to 8.52 f 3.97 nmol/L (p = 0.15) levels (I 3). Because neither hypoxemia alone nor pure metabolic and fetal epinephrine increased from 0.27 + 0.10 to 2.271 acidosis resulted in the substantial increase in plasma AVP seen 2 0.90 nmol/L (p = 0.01); plasma AVP was not signifi- during fetal asphyxia, the role of hypercapnia in fetal secretion cantly increased in the ewe or fetus, although levels rose of AVP, as well as the interaction of these variables, remained to from -45 to 127 + 48 and 137 2 64 pmol/L (p = 0.10), be elucidated. respectively. Hypercapnia alone is not a major determinant Catecholamines are also released during asphyxia. Rose ef a/. of AVP secretion in the mother or fetus; thus, the marked (14) demonstrated that acute hypercapnic acidosis resulted in rise in fetal AVP secretion observed during asphyxia1 increased circulating levels of both E and NE in adult dogs. insults may reflect interaction between the effects of hy- However, NE was the predominant catecholamine released, es- percapnia and hypoxemia. (Pediatr Res 30: 368-374,1991) pecially when hypercapnia was combined with hypoxemia. They suggested that the major source of NE was the sympathetic nerve terminals. They also observed an increase in MAP and heart rate Abbreviations in response to hypercapnia. No one, to our knowledge. has addressed the effects of acute hypercapnia on circulating cate- MAP, mean arterial pressure cholamines in a fetal animal. Moreover, the relationship between E, epinephrine plasma levels of these hormones and cardiovascular responses NE, norepinephrine during hypercapnia has not been examined in the fetus. PaOz, arterial O2tension Therefore, the purpose of the study presented here was to PaCOZ,arterial COz tension determine the effects of acute hypercapnia on fetal secretion of ANOVA, analysis of variance AVP, dopanline, NE. and E, to examine the concomitant fetal pH,, arterial pH hemodynamic responses, and to compare these alterations with those occurring simultaneously in the pregnant ewe. MATERIALS AND METHODS Received October 26, 1990; accepted June 17, 1991. .4nimu/prepu~ulion.Pregnant sheep (n= 6) of mixed Western Correspondence: Charles R. Rosenfeld, M.D., Department of Pediatrics, Uni- breed bearing singleton fetuses were used in these versity of Texas Southwestern Medical Center. 5323 Hany Hines Blvd., Dallas, TX 75235-9063. Studies (n = 8) were performed at 135 t 4 (x k SD) d of gestation Supported by NIH Grants HD-08783, HD-20720, JD-I I 149, and DK-01237. (term, - 145 d). The surgical preparation has been described ' Presented in part at the annual meeting of the Endocrine Society, Baltimore, previou~ly(1 1, 15). In brief, animals were fasted 24 h before MD, June 1985. surgery, which was performed at 124 to 126 d of gestation. After Dr. Faucher was a postdoctoral Fellow In the Division of Neonatal-Pennatal Medicine from July 1, 1983 to June 30, 1986 and is presently at McGill University, spinal anesthesia with hyperbaric lidocaine ( 12 mg), i.~,pento- Montreal, Canada. barbital (- 15 mg/kg) was administered as needed. While em- FETAL VASOPRESSIN DURING HYPERCAPNIA 369 ploying sterile procedures, the maternal abdomen was opened, 7.5 min after the end of C02 exposure. Each sample of fetal the pregnant uterus was delivered, and a fetal hindlimb was blood collected for AVP and catecholamine measurements was brought out through a uterine incision. Polyvinyl catheters were centrifuged immediately, the plasma was removed, and the eryth- implanted into a fetal femoral artery (7.5 cm) and vein (I 5 cm). rocytes were resuspended in a volume of sterile isotonic saline The fetal hindlimb was returned to the amniotic sac, an amniotic equal to the plasma volume and reinfused into the fetus in the catheter was inserted, and the uterus was closed with a double manner previously reported (1 5). This procedure has been shown layer of sutures. The uterus was then returned to the maternal to have no effect on AVP secretion (13). Maternal blood was abdomen, and electromagnetic flow probes (Micron Instruments handled similarly, but red cells were not reinfused. Fetal (0.5 Inc., Los Angeles, CA) were implanted around both main uterine mL) and maternal (0.5 mL) plasma osmolalities were measured arteries proximal to their bifurcation. The fetal catheters and at each of these times in the first four experiments. Fetal plasma flow probe leads were brought out of the maternal abdomen lactate levels (0.4 mL) were determined in two initial experi- through a stab wound in the fascia, and the abdominal incision ments. was closed. Polyvinyl catheters were implanted in both maternal Samples of amniotic fluid (2 mL) were collected at 15 min in femoral arteries (I5 cm) and veins (20 to 25 cm). All catheters room air, at 20 rnin in 02,at 10-min intervals in COz plus 02, were filled with heparinized saline, closed with sterile pins, and at 15 and 30 min in O2 after ending COz, and at 30 min, 60 min. brought out to the maternal flank with the flow probe leads via and 20 h in room air for measurement of AVP. a S.C. tunnel, where they were maintained in a canvas pouch Assav.~.Arterial blood samples for blood gases, pH, and hem- attached to the skin with steel pins. Pregnant animals received atocrit were collected in 1.0-mL heparinized syringes and were intramuscular mezlocillin (0.5 g) and penicillin (600 000 U) on kept on ice until analyzed. The blood gases and pH were meas- the day of surgery and the 2 following days. Mezlocillin (50 mg) ured with a pH/gas analyzer (model 113; Instrumentation Lab- was also instilled in the amniotic sac at surgery and every other oratory, Lexington, MA). The hematocrit was measured on these day thereafter. After surgery, all animals were maintained in samples by the micro-method of Wintrobe. Plasma osmolality individual stalls in the laboratory and had free access to water was measured by freezing-point depression with an automatic and food (Purina Commercial Creep Chow 11, G:Ralston-Purina osmometer. Lactic acid concentrations were determined by using Co., St. Louis, MO). Studies were performed at least 5 d post- a lactate dehydrogenase enzymatic assay on plasma (200 pL) operatively. Fetal sheep with abnormal blood gases or pH were after deproteinization with 8% perchloric acid. not studied. Blood samples for measurement of AVP were collected in Experimental protocol. Acute hypercapnia was induced in chilled heparinized syringes and were immediately transferred to pregnant ewes and their fetuses by sequentially exposing the ewe sterile plastic centrifuge tubes. After centrifugation at 10 000 x to the following gas mixtures: room air for 30 min, 30% O-, for g for 60 s in a Beckman microfuge, the plasma was removed.