- Email: [email protected]
Effects of maternal glucose infusion on fetal acid-base status in human pregnancy
Effects of maternal glucose infusion on fetal acid-base status in human pregnancy Elliot H. Philipson, M.D., Satish C. Kathan, M.B., Margo M. Riha, R.N., and Renato Pimentel, M.D. Cleveland, Ohio The maternal and fetal metabolic effects of three commonly used intravenous fluids administered before regional anesthesia were studied in 32 gravid women undergoing elective cesarean section at term. Patients were randomized into one of three groups to receive 1 L of either 5% dextrose (50 gm of glucose) or Ringer's lactate or isotonic saline solution before epidural anesthesia. Acute glucose infusion resulted in maternal hyperglycemia, hyperinsulinemia, and an increase in the blood lactate level. Cord blood glucose, insulin, and lactate levels were also increased in this group. The key finding of this study was the significant lowering of pH in the umbilical cord vein (7.31 ± 0.04) and artery (7.21 ± 0.06) in the glucose-infused group when compared with the non-glucose infusion groups (p < 0.05). Confounding perinatal factors such as maternal position, maternal hypotension, and prolonged time of surgery did not influence the fetal acid-base status. Thus acute maternal glucose infusion in normal patients can cause fetal hyperglycemia, metabolic acidosis, and neonatal hypoglycemia. These findings may be of particular clinical importance when fetal distress or fetal hypoxemia is due to other perinatal events. Under these circumstances, acute maternal glucose infusion may further contribute to fetal metabolic acidosis. (AM J 0BSTET GYNECOL 1987;157:866-73.)
Key words: Pregnancy, intravenous fluids, glucose, fetal acidosis, hyperglycemia
Acute intravenous fluid administration has been recommended to prevent maternal hypotension by expanding volume during regional anesthesia.t. 2 Although this "prophylactic" hydration can be accomplished with various intravenous solutions, acute administration of fluids containing glucose can result in maternal and fetal hyperglycemia and hyperinsulinemia.3,., In well-oxygenated fetal lambs, sustained fetal hyperglycemia resulted in increased fetal lactic acid and decreased blood pH. More marked hyperglycemia caused severe metabolic acidosis.' Philipps et al. 6 ·7 demonstrated that sustained fetal hyperglycemia decreased fetal arterial oxygen concentration and induced a state of accelerated fetal oxidative metabolism. Direct insulin infusion to the fetus increased umbilical glucose uptake and decreased fetal arterial oxygen concentration." In contrast to these chronic animal experiments, acute infusion of insulin increased fetal glucose uptake and From the Departments of Obstetrics and Gynecology and Pediatrics, Division of Metabolism; and the Department of Anesthesiology, Case Western Reserve University School of Medicine at Cleveland Metropolitan General Hospital. Supported in part by United States Public Health Service Grants HD 11089 and HD 17336, and Perinatal Clinical Research Center Grant 5M01-RR0021 0. Received for publication january 8, 1987; revised April23, 1987; accepted May 15, 1987. Reprint requests: Elliot H. Philipson, M.D., Department of Obstetrics and Gynecology, Cleveland Metropolitan General Hospital, 3395 Scranton Road, Cleveland, OH 44109.
866
utilization but did not alter fetal oxygen consumption. 9 Therefore it appears that chronic hyperglycemia and hyperinsulinemia influence glucose uptake, arterial oxygen concentration, and pH, whereas the effects of acute hyperglycemia may not be similar. In human pregnancy, Kenepp et a!.'" reported that rapid infusion of large amounts of dextrose may be harmful to the fetus and of no benefit to the mother. In their study of women undergoing cesarean section, acute maternal glucose infusion (57.5 gm) resulted not only in maternal and fetal hyperglycemia and hyperinsulinemia but also in a significant decrease in fetal pH. However, intrapartum and perinatal confounding factors other than maternal glucose could have influenced fetal acid-base status. Placental hypoperfusion, as a result of maternal hypotension or maternal supine position, can cause lactic acidosis and influence neonatal acid-base balance at birth!·'' In addition, prolonged surgical intervals, that is, the time between induction of labor and delivery of the fetus or the time from myometrial incision to delivery, can influence the incidence of neonatal acidosis and low Apgar scores.' 2 · 13 Therefore the purpose of this prospective randomized study was: (1) to evaluate the maternal and fetal metabolic effects of three intravenous fluids commonly used to prevent maternal hypotension during regional anesthesia and (2) to examine intrapartum perinatal factors that could influence fetal acid-base status.
Maternal hyperglycemia and fetal acidosis
Volume 157 Number 4, Part 1
867
Table I. Maternal and neonatal characteristics of the three groups Group 1: Glucose
Group 2: Ringer's lactate
(N = 12)
Group 3: Normal saline
(N = 11)
(N = 9)
Maternal characteristics Age (yr) Gravidity Parity Height (em) Weight (kg) Hematocrit(%)
25.3 2.8 2.0 28.6 73.6 34.1
Neonatal characteristics Birth weight (gm) Dubowitz results (wk)
3245 ± 398 39.1 ± l.3
± 5.6 ± l.3 ± l.2 ± l.O ± 25.9 ± 4.0
25.8 3.1 2.1 28.6 79.8 35.6
± 4.7 ± l.O ± l.O ± 1.0 ± 13.3 ± 3.3
3242 ± 322 38.8 ± l.3
25.3 3.6 2.1 28.4 85.8 34.8
± ± ± ± ± ±
5.1 l.O l.O 0.9 22.8 2.7
3383 ± 454 39.1 ± l.7
Values are mean ± SD.
Methods and material
All patients had normal, singleton, term gestations and were scheduled for elective cesarean section at Cleveland Metropolitan General Hospital. Patients with liver or kidney disease or any medical complications were excluded from the study. All patients had a negative screening test for diabetes mellitus. After written informed consent was obtained, patients were assigned by a random table of numbers into one of three groups. Group 1 patients received 1 L of 5% dextrose in water (glucose = 50 gm); group 2 patients received 1 L of Ringer's lactate; group 3 (control) patients received 1 L of normal saline solution. The intravenous fluids were administered rapidly during 20 to 30 minutes. Immediately after the hydration, all patients were maintained on a regimen of intravenous normal saline solution at a rate of 125 ml/hr until surgery and delivery of the baby. Epidural anesthesia was then induced with 0.5% or 0.75% bupivacaine without epinephrine. All patients were placed on the operating room table in the supine position with left lateral uterine displacement and received oxygen at 4 Llmin by nasal cannula. Maternal blood pressure was measured every minute with an Arteriosonde (Kontron Inc., Everett, Mass.) from the time of epidural anesthesia until delivery of the infant and every 5 minutes thereafter. Maternal hypotension (a decrease in the systolic blood pressure of >30 torr from baseline or an absolute level < 100 torr) was corrected by incremental doses of intravenous ephedrine (5 to 10 mg) and by increasing the rate of infusion of normal saline solution. Maternal blood samples (3 ml each) were obtained from an indwelling cannula in a peripheral hand vein. The sampling site was kept patent by a heparin lock. In order to obtain arterialized samples, the sampling site was kept warm (55° C) by a heating blanket. Maternal blood samples were obtained at the start of and
after acute intravenous hydration at 10, 20, and 30 minutes and continued every 10 minutes until delivery. Fetal arterial and venous blood samples were obtained at birth from a double-clamped umbilical cord. Neonatal blood samples were also obtained every 30 minutes for 2 hours by venipuncture or heel stick. Since the amount of blood drawn from the infant was limited, the samples at 90 and 120 minutes were analyzed for glucose only. All blood samples were obtained in heparinized syringes and immediately placed on ice. Plasma was separated and stored at - 10° C for later analysis. An aliquot of whole blood was precipitated with chilled 10% perchloric acid for lactate measurement. The data collected during the study also included the skin-to-delivery time interval and the uterine incisionto-delivery interval. At birth, 1- and 5-minute Apgar scores were recorded. Each infant was weighed and a complete physical examination was performed. The gestational age was determined by a modification of the Dubowitz examination by Ballard et a!. 11 Blood samples were analyzed for glucose, lactate, insulin, ketones ((3-hydroxybutyrate), and triglycerides. Maternal samples were analyzed for electrolytes, carbon dioxide, and blood urea nitrogen. The baseline maternal samples and umbilical cord samples were also analyzed for acid-base status (pH, Pco2 , Po 2 , and base deficit). Blood gas tensions and pH were measured with a microelectrode system (Model 13031, Instrumentation Laboratory, Lexington, Massachusetts). Plasma glucose was measured by a coupled enzymatic assay with the use of hexokinase and glucose-6-phosphate dehydrogenase on an automated centrifugal analyzer (Baker Instruments, Pleasantville, New York). Blood lactate, (3-hydroxybutyrate, and triglycerides were measured by enzymatic methods and insulin by a doubleantibody radioimmunoassay. All the assay methods have been previously reported from this laboratory. 15
868
October 198 7 Am J Obstet Gynecol
Philipson et al.
a
Infusion \
~ ~I
Neonate
300 • Glucose o Ringer's Lactate • Saline
~
~
Cl
!
~
200
Cl)
"' 0
0
::>
(!j
100
T
I---H=J 1T 10
~0//
~
4'o'1 t
20
lJ t
cv
D
t
CA
I'
~ 60
120
Minutes
Fig. I. Changes in plasma glucose concentration in the mother and infant after intravenous fluid administration. D = At delivery; CV = cord vein; CA = cord artery. Data are presented as mean± SD.
\ Infusion \
35
EJ
rco;dl
~
Cord Artery
Neonate
• Glucose a Ringer's Lactate • Saline
~
~0
25
!
Q)
a; 0
"'
15
..J
5 '----1,_0_..__3..__~-t/rl-'---'---2-'-o---'-'-4--''o--1/rf-l'f-t.----•t----1/1--'---6--'-0-L -1-'-2-0D CV
CA
Minutes
Fig. 2. Changes in blood lactate concentration in the mother and infant after intravenous administration. For abbreviations see legend to Fig. 1.
All data are presented as mean ± SD. Group comparisons were done with Student's t test. A p value of <0.05 was considered statistically significant.
Results All patients were at term and did not have any underlying medical problems. As shown in Table I, there were no significant differences among the three groups in their antenatal characteristics. The neonatal birth weight and gestational age at birth were also similar between the groups. The neonates were all term sized, appropriate for gestational age, and had 5-minute Apgar scores >7. One neonate in group I had meconium-
stained amniotic fluid below the vocal cords and had a !-minute Apgar score of 6. The 5-minute Apgar score in this neonate was 8. Plasma glucose. The changes in plasma glucose concentration in the three groups are displayed in Fig. I. The initial maternal plasma glucose concentration was similar in the three groups. Acute intravenous glucose (50 gm during 30 minutes) administration in group I caused a rapid increase in plasma glucose concentration, reaching a maximum of 225 ± 66 mg/dl at 20 minutes followed by a continuous gradual decline. At delivery, the maternal plasma glucose level in group I was still significantly elevated (134 ± 37
Maternal hyperglycemia and fetal acidosis 869
Volume 157 Number 4, Part 1
El
Infusion \
Neonate / _ lcOrdl L__ ~ Artery _ ___,
~
100 • Glucose o Ringer's Lactate • Saline
......
e-... ::l
..3
60
.5 3(/) .5
20
~ 10
30
20
~
40 I t
D
t
cv
t
CA
/f'I
~ 60 120
Minutes
Fig. 3. Changes in plasma insulin concentration in the mother and infant after intravenous fluid administration. For abbreviations see legend to Fig. l.
mg/dl). In contrast, acute maternal hydration with Ringer's lactate or with normal saline solution caused a small but insignificant decrease in plasma glucose concentration. The glucose concentration remained stable thereafter, so that at delivery the maternal plasrp.a glucose concentration was 83 ± 16 mg/dl in group 2 and 76 ± 9 mg/dl in group 3. In all three groups, the umbilical cord venous and arterial glucose concentrations were lower than the simultaneously obtained maternal peripheral venous gh.,cose levels at delivery. However, in the group administered glucose, the umbilical cord venous and arterial glucose concentrations were significantly elevated wP,en compared with tpose of groups 2 and 3 (p < 0.001). Even though the glucose concentration in the umbilical cord vein and artery was higher in group 1 than in groups 2 and 3, the plasma glucose concentration in the newborn infants at 30 minutes was similar in all three groups. The plasma glucose concentration in g-roup 1 decreased d1ning the next 30 minutes before stabilizing during the next hour. By 2 hours after birth, the neonatal plasma glucose concentration remained significantly lower in group 1 than in groups 2 and 3 (p < 0.01). In contrast, the plasma glucose concentration in the newborn infants in grqups 2 and 3 stabilized between 44 and 62 mg/dl for the next 90 minutes. Four of 12 neonates in group 1 and two of 11 neonates in group 2 developed asymptomatic hypoglycemia (blood glucose value of <25 mg/dl, Dextrostix, Ames Co., Elkhart, Indiana). No infants in group 3 developed hypoglycemia. One infant in group 1 de~ veloped tachypnea. At this time, the plasma glucose concentration was 18 mg/dL
BlQOd lactate. Changes in maternal, umbilical cord, and neonatal blood lactate concentrations are shown in Fig. 2. Initial maternal blood lactate concentrations in groups 1, 2, and 3 were 12 ± 3, 11 ± 3, and 10 ± 2 mg/dl, respectively. In groups 1 and 2, the maternal blood lactate concentration was significantly higher at the end of the intravenous infusion when compared with the respective initial blood lactate concentrations and also when compared with the value in the group receiving normal saline solution (group 3) (p < 0.01). However, by the time the epidural anesthesia was induced and surgery was qegun, the blood lactate concentratio!l in group 2 was similar to that of group 3. The maternal blood lactate concentration in the glucose infusion group remained significantly elevated throughout surgery and delivery when compare~ with the initial blood lactate concentration and also when compared with values in groups 2 and 3. The umbilical cord venous and arterial blqod lactate concentrations were higher than the maternal venous blood lactate concentrations in all three groups. The glucose infusion in group 1 caused a greater increase in the blood lact
870
Philipson et al.
October 1987 Am J Obstet Gynecol
Table II. Acid-base status at delivery Group 1: Glucose (N = 12)
Maternal vein at delivery pH
Group 2: Ringer's lactate (N = 11)
Group 3: Normal saline (N = 9)
Pco2 Po,
7.36 35.1 61.7 -4.9
± ± ± ±
0.04 5.3 41.1 2.0
7.36 33.7 68.7 -4.7
± ± ± ±
0.12 4.7 43.6 6.3
7.37 34.8 65.1 -3.7
± ± ± ±
0.05 6.9 31.0 2.3
Pco2 Po 2
7.31 41.8 31.4 -4.7
± ± ± ±
0.04* 4.8 8.9 2.5
7.35 40.6 30.5 -2.7
± ± ± ±
0.07 7.0 14.9 1.9
7.34 39.8 31.1 -3.7
± ± ± ±
0.04 4.8 4.6 3.6
7.21 57.7 14.0 -5.1 11.8
± ± ± ± ±
0.06* 5.6* 3.3 3.0 7.0*
7.27 49.2 17.1 -3.1 22.7
± ± ± ± ±
0.12 8.4 5.4 2.7 12.3
7.27 50.8 15.0 -4.3 17.6
± ± ± ± ±
0.06 6.0 3.6 3.4 6.7
Base deficit Cord vein pH Base deficit Cord artery pH
Pco, Po,
Base deficit Oxygen saturation
Values are mean ± SD. *p < 0.05 (group I compared with groups 2 and 3).
6.2, 10.5 ± 5.5, and 12.6 ± 9.0 f.LU/ml, respectively (Fig. 3). The glucose infusion in group 1 caused a shortterm rise in the insulin concentration so that at 20 minutes the plasma insulin concentration was 80.7 ± 44.8 f.LU/ml (p < 0.0001). At delivery, the maternal plasma insulin concentration in group 1 remained significantly elevated at 58.9 ± 38.4 f.LU/ml (p < 0.002) with the plasma insulin concentrations in groups 2 and 3 remaining unchanged. The insulin concentration in the umbilical cord vein (65.9 ± 67.9 f.LU/ml) and artery (55.6 ± 66.5 f.LU/ml) at delivery in group 1 was also significantly elevated when compared with the other two groups (p < 0.02). The insulin concentrations could not be measured in all the neonates at 30, 60, 90, and 120 minutes because of limited blood sample size. However, the neonatal insulin concentrations in groups 2 and 3 remained <3.5 f.LU/ml while in the neonates of group 1 higher insulin concentrations were maintained at 30 and 90 minutes after birth. Acid-base status. The maternal venous pH before the infusion was similar in the three groups. By the end of the infusion, the maternal pH in group 1 (7.36 ± 0.03) was significantly lower than the pH in group 2 (7.38 ± 0.2) and group 3 (7.39 ± 0.02) (p < 0.05). As shown in Table II, the maternal pH at delivery was similar between the groups. However, the umbilical arterial pH in group 1 (7.21 ± 0.06) was significantly lower than the umbilical arterial pH in group 2 (7.27 ± 0.12) and group 3 (7.27 ± 0.06) (p < 0.05). The umbilical venous blood pH in group 1 (7 .31 ± 0.04) was also significantly lower than the umbilical cord vein pH in group 2 (7.35 ± 0.07) and group 3 (7.34 ± 0.04) (p < 0.05). In the neonatal period, there was a tendency for lower venous pH values
in the neonates from group 1. However, no conclusions can be made as complete studies were done in only eight neonates. Additional acid-base results in the maternal vein at delivery and the umbilical cord vein and artery of the three groups are shown in Table II. There was a significantly higher umbilical cord artery Pco 2 in the glucose-infused group. The umbilical arterial hemoglobin oxygen saturation in the glucose group (11.8 ± 7.0) was significantly lower than that of the Ringer's lactate (22.7 ± 12.3) or normal saline solution (17.6 ± 6.7) groups (p < 0.05). f3-Hydroxybutyrate and triglycerides. As shown in Table Ill, acute glucose infusion in group 1 was associated with a significant decrease in the concentration of [3-hydroxybutyrate in the maternal blood at delivery (p < 0.001). In addition, the umbilical venous (p < 0.003) and arterial (p < 0.04) concentrations of [3-hydroxybutyrate were lower in group 1 than in groups 2 and 3. The concentration of [3-hydroxybutyrate in the maternal plasma tended to rise in the Ringer's lactate and normal saline solution groups. The triglyceride levels in the three groups demonstrated wide variations with large standard deviations within each group. No significant differences in the mean triglyceride levels between the groups were identified. Serum electrolytes, carbon dioxide, and blood urea nitrogen. The serum electrolytes, carbon dioxide, and blood urea nitrogen were similar in the three groups. Acute glucose infusion significantly decreased the serum sodium level from 137 ± 3 to 131 ± 3 mEq/L (p < 0.01). No significant change in other electrolytes, carbon dioxide, or blood urea nitrogen was seen in any of the groups.
Maternal hyperglycemia and fetal acidosis
Volume 157 Number 4, Part I
871
Table III. Changes in 13-hydroxybutyrate and triglycerides in the three groups {3-Hydroxybutyrate (mmol!L)
Mother Preinfusion Delivery Umbilical cord Vein Artery Neonate 30 min 60 min
Triglycerides (mgldl)
Group 1: Glucose
Group 2: Ringer's lactate
Group 3: Normal saline
Group 1: Glucose
Group 2: Ringer's lactate
Group 3: Normal saline
0.23 ± 0.26 0.11 ± 0.07*
0.21 ± 0.09 0.26 ± 0.15
0.24 ± 0.15 0.33 ± 0.16
209 ± 112 183 ± 95
221 ± 91 214 ± 83
233 ± 112 229 ± 91
0.09 ± 0.04t 0.07 ± 0.03:j:
0.12 ± 0.04 0.12 ± 0.06
0.17 ± 0.07 0.10 ± 0.02
15 ± 9 12 ± 5
18 ± 12 16 ± 9
13 ± 9 12 ± 5
0.09 ± 0.01 0.08 ± 0.02
0.07 ± 0.01 0.07 ± 0.03
0.09 ± 0.01 0.06 ± 0.03
23 ± 5 24 ± 8
16 ± II 22 ± 8
20 ± 5 20 ± 8
Values are mean ± SD. *p < 0.001, compared with groups 2 and 3. tp < 0.003, compared with groups 2 and 3. :j:p < 0.04, compared with groups 2 and 3.
Table IV. Influence of maternal hypotension on pH, glucose, and lactate at delivery
Maternal vein at delivery pH Glucose Lactate Umbilical cord vein pH Glucose Lactate Umbilical cord artery pH Glucose Lactate
Hypotension (N = 9)
No hypotension (N = 23)
Significance
7.31 ± 0.12 126 ± 49 19.1 ± 5.3
7.38 ± 0.04 90 ± 24 12.6 ± 5.0
p < 0.05 p< 0.01 p < 0.01
7.30 ± 0.04 Ill± 45 23.9 ± 9.8
7.34 ± 0.05 78 ± 22 15.9 ± 6.0
p < 0.05 p< 0.01 p < 0.01
7.19 ± 0.11 95 ± 42 26.3 ± 9.3
7.27 ± 0.06 62 ± 20 18.8 ± 7.1
p < 0.02 p < 0.01 p < 0.05
Values are mean ± SD.
Maternal hypotension. The influence of maternal hypotension on pH, plasma glucose, and blood lactate are shown in Table IV. Nine patients (28%) developed hypotension: six in group 1, two in group 2, and one in group 3. In comparison with the group that remained normotensive, the hypotensive group of patients had a significantly lower maternal vein pH at delivery as well as a lower pH in the umbilical cord vein and artery. Because the hypotensive patients included a greater proportion from the glucose group, this group also had elevated plasma glucose and blood lactate levels in the maternal vein at delivery and in the umbilical cord. In addition, all of these patients had received ephedrine for the correction of hypotension. Because of these differences in plasma glucose and blood lactate in the hypotensive group, the influence of maternal hypotension was examined within group 1 alone. When the six hypotensive patients were compared with six normotensive patients in group 1, there
were no significant differences in the pH, plasma glucose level, or blood lactate level. The glucose concentration in the maternal venous blood at delivery tended to be higher, although not significantly so, in the hypotensive group, possibly because of ephedrine administration. Uterine incision-to-delivery interval. In order to examine the effects of a prolonged uterine incisionto-delivery interval, patients with a uterine incisiondelivery interval > 180 seconds were compared with patients with a uterine incision-delivery interval of < 180 seconds. A prolonged uterine incision-todelivery interval did not significantly alter any of the parameters measured. Comment
The purpose of this study was to evaluate the maternal and fetal metabolic effects of three commonly used intravenous fluids during delivery. The results
872
Philipson et al.
demonstrate that acute glucose infusion before delivery resulted in maternal hyperglycemia, hyperinsulinemia, and an increase in blood lactate. These maternal perturbations were associated with fetal hyperglycemia and hyperinsulinemia, higher fetal blood lactate levels, and lower fetal arterial pH. The key finding of this study was the lower fetal arterial pH in the glucose-infused group. This increase in blood hydrogen ion concentration was accompanied by a significant rise in blood lactate, an increase in carbon dioxide, and a decreased fetal hemoglobin oxygen saturation. An explanation for this lactic acidosis may be obtained from animal data. In chronically catheterized ewes, direct glucose infusion to the fetus can induce an acceleration of the fetal metabolic rate and fetal oxygen consumption. 7 The increase in fetal oxygen consumption has been speculated to be responsible for the decrease in fetal arterial oxygen content. Robillard et aJ.5 have shown that chronic fetal hyperglycemia in sheep induces a similar state of fetal arterial hypoxia, metabolic acidosis, and fetal death. Direct infusion of insulin to the fetus caused an increased glucose uptake by the fetus and a decreased fetal arterial oxygen content.8· 9 Crandell et al. 16 showed that in chronically catheterized ewes hyperglycemia resulted not only in fetal hypoxemia and acidemia but also in a redistribution of the fetal combined ventricular output and a decrease in umbilical blood flow to the placenta. Although there was no significant change in oxygen delivery to the fetal heart, brain, or kidneys in this study, there was a marked decrease in fetal arterial oxygen saturation. On the basis of these studies in animals, one may speculate that acute maternal glucose infusion in human pregnancy results in an increased metabolic rate of the fetus. Glucose consumption and lactic acid production are increased. Fetal oxygen consumption is also increased and results in a decrease in fetal oxygen content and subsequent fetal hypoxemia. Fetal hypoxemia leads to a decrease in pH and, if severe, lactic acidosis. Redistribution of fetal cardiac output may be a compensatory adaptation to these metabolic changes. This study also examined other metabolic consequences of acute glucose infusion. Acute maternal hyperglycemia persisted more than 40 minutes after the infusion so that, by delivery, the maternal glucose remained elevated and the concentration of blood lactate had doubled. The umbilical cord vein and artery blood lactate concentrations remained twice as high as those of the control group. In contrast, the group that received the Ringer's lactate (28 mEq/L of lactate) did not demonstrate this altered metabolic lactic acidosis. Other perinatal factors, such as maternal position, maternal hypotension, and surgical time intervals at the time of cesarean section, were examined since these factors are known to influence fetal acid-base status at
October 1987 Am J Obstet Gynecol
delivery. 9· 13 All patients m this study had left lateral uterine displacement to avoid the effects of aortocaval compression. Maternal hypotension was rapidly corrected and did not influence neonatal acid-base status when controlled for glucose concentration. In this study, a prolonged skin-to-delivery interval did not influence the neonatal acid-base balance. Similar results have been reported with spinal anesthesia in the absence of hypotension." The uterine incision-todelivery interval also did not appear to influence the pH although there were only a few cases with a prolonged interval of >3 minutes. Therefore, by controlling for these confounding variables, the alterations in substrates anq lower pH were attributed to the metabolic effects of acute glucose infusion alone. In summary, short-term maternal administration of glucose before delivery is associated with fetal hyperglycemia, hyperinsulinemia, an increase in lactic acid, and a significant decrease in pH. Nevertheless, no obvious detrimental effects were observed on the mother, fetus, or neonate in this study. However, these metabolic changes occurred in normal patients at term without any signs of fetal distress or acidosis during the infusion, epidural anesthesia, or surgery. Since fetal distress and acidosis may occur during labor and delivery, acute glucose infusion in this clinical setting may exaggerate the metabolic acidosis associated with fetal distress. When intravenous fluids are necessary in obstetrics, either for hydration or administering medications, large amounts of glucose should be avoided. We thank Ms. Darlene Spagnoli, Ms. Monica Bakaitis, and Ms. Ellen Brown for their technical assistance, Ms. Sue Fitz for patient recruitment, and Sharon Anthony, R.N., and the Perinatal Clinical Research Center staff for their assistance with patient care. REFERENCES 1. 2.
3.
4. 5.
6. 7.
Wollm~n SB, Marx GF. Acute hydration for prevention of hypotension of spinal anesthesia in parturients. Anesthesiology 1968;29:374-9. Wright RG, Shnider SM. Hypotension and regional anesthesia in obstetrics. In: Shnider SM, Levinson G, eds. Anesthesia for obstetrics. Baltimore: Williams & Wilkins, 1987:293-9. Light IJ, Keenan WJ, Sutherland JM. Maternal intravenous glucose administration as a cause of hypoglycemia in the infant of the diabetic mother. AM J 0BSTET GYNECOL 1972; 113:345-50. . Tobin JD, Roux JF, Soeldner JS. Human fetal insulin response after acute maternal glucose administration during labor. Pediatrics 1969;44:668-71. Robillard JE, Sessions C, Kennedy RL, Smith FG. Metabolic effects of constant hypertonic glucose infusion in well-oxygenated fetuses. AM J 0BSTET GYNECOL 1978; 130:199-203. Philipps AF, Dubin JW, Matty PJ, Raye JR. Arterial hypoxemia and hyperinsulinemia in the chronically hyperglycemic fetal lamb. Pediatr Res 1982; 16:653-8. Philipps AF, Porte PJ, Stabinsky S, Rosenkrantz TS, Raye JR. Effects of chronic fetal hyperglycemia upon oxygen
Maternal hyperglycemia and fetal acidosis 873
Volume 157 Number 4, Part I
8.
9. 10. ll. 12.
consumption in the ovine uterus and conceptus. J Clin Invest 1984;74:279-86. Carson BS, Philipps AF, Simmons MA, Battaglia FC, Meschia G. Effects of a sustained insulin infusion upon glucose uptake and oxygenation of the ovine fetus. Pediatr Res 1980;14:147-52. Simmons MA, Jones MD, Battaglia FC, Meschia G. Insulin effect on fetal glucose utilization. Pediatr Res 1978;12: 90-2. Kenepp NB, Shelley WC, Gabbe SG, et al. Fetal and neonatal hazards of maternal hydration with 5% dextrose before cesarean section. Lancet 1982; l: 1150-2. Crawford JS, Burton M, Davies P. Time and lateral tilt at caesarean section. Br J Anaesth 1972;44:477-84. Fothergill RJ, Robertson A, Bond RA. Neonatal acidaemia
13. 14. 15. 16.
related to procrastination at cesarean section. J Obstet Gynaecol Br Commonw 1971;78:1010-23. Datta S, Ostheimer GW, WeissjB, Brown WU, Alper MH. Neonatal effect of prolonged anesthetic induction forcesarean section. Obstet Gynecol 1981 ;58:331-5. Ballard JL, Novak K, Driver M. A simplified score for assessment of fetal maturation and newly born infant. J Pediatr 1979;95:769-74. Kalhan SC, Tserng KY, Gilfillan CA, Dierker LJ. Metabolism of urea and glucose in normal and diabetic pregnancy. Metabolism 1982;31 :824-33. Crandell SS, Fisher DJ, Morriss FH. Effects of ovine maternal hyperglycemia on fetal regional blood flows and metabolism. Am J Physiol 1985;249:454-9.
Bound volumes available to subscribers Bound volumes of the AMERICAN jOURNAL OF OBSTETRICS AND GYNECOLOGY are available to subscribers (only) for the 1987 issues from the Publisher, at a cost of $51.00 ($74.00 international) for Vol. 156 (January-June) and Vol. 157 (July-December). Shipping charges are included. Each bound volume contains a subject and author index and all advertising is removed. Copies are shipped within 60 days after publication of the last issue in the volume. The binding is durable buckram with the JouRNAL name, volume number, and year stamped in gold on the spine. Payment must accompany all orders. Contact The C. V. Mosby Company, Circulation Department, 11830 Westline Industrial Drive, St. Louis, Missouri 63146, USA; phone (800) 325-4177, ext. 351. Subscriptions must be in force to qualify. Bound volumes are not available in place of a regular JOURNAL subscription.
Key words: Pregnancy, intravenous fluids, glucose, fetal acidosis, hyperglycemia
Acute intravenous fluid administration has been recommended to prevent maternal hypotension by expanding volume during regional anesthesia.t. 2 Although this "prophylactic" hydration can be accomplished with various intravenous solutions, acute administration of fluids containing glucose can result in maternal and fetal hyperglycemia and hyperinsulinemia.3,., In well-oxygenated fetal lambs, sustained fetal hyperglycemia resulted in increased fetal lactic acid and decreased blood pH. More marked hyperglycemia caused severe metabolic acidosis.' Philipps et al. 6 ·7 demonstrated that sustained fetal hyperglycemia decreased fetal arterial oxygen concentration and induced a state of accelerated fetal oxidative metabolism. Direct insulin infusion to the fetus increased umbilical glucose uptake and decreased fetal arterial oxygen concentration." In contrast to these chronic animal experiments, acute infusion of insulin increased fetal glucose uptake and From the Departments of Obstetrics and Gynecology and Pediatrics, Division of Metabolism; and the Department of Anesthesiology, Case Western Reserve University School of Medicine at Cleveland Metropolitan General Hospital. Supported in part by United States Public Health Service Grants HD 11089 and HD 17336, and Perinatal Clinical Research Center Grant 5M01-RR0021 0. Received for publication january 8, 1987; revised April23, 1987; accepted May 15, 1987. Reprint requests: Elliot H. Philipson, M.D., Department of Obstetrics and Gynecology, Cleveland Metropolitan General Hospital, 3395 Scranton Road, Cleveland, OH 44109.
866
utilization but did not alter fetal oxygen consumption. 9 Therefore it appears that chronic hyperglycemia and hyperinsulinemia influence glucose uptake, arterial oxygen concentration, and pH, whereas the effects of acute hyperglycemia may not be similar. In human pregnancy, Kenepp et a!.'" reported that rapid infusion of large amounts of dextrose may be harmful to the fetus and of no benefit to the mother. In their study of women undergoing cesarean section, acute maternal glucose infusion (57.5 gm) resulted not only in maternal and fetal hyperglycemia and hyperinsulinemia but also in a significant decrease in fetal pH. However, intrapartum and perinatal confounding factors other than maternal glucose could have influenced fetal acid-base status. Placental hypoperfusion, as a result of maternal hypotension or maternal supine position, can cause lactic acidosis and influence neonatal acid-base balance at birth!·'' In addition, prolonged surgical intervals, that is, the time between induction of labor and delivery of the fetus or the time from myometrial incision to delivery, can influence the incidence of neonatal acidosis and low Apgar scores.' 2 · 13 Therefore the purpose of this prospective randomized study was: (1) to evaluate the maternal and fetal metabolic effects of three intravenous fluids commonly used to prevent maternal hypotension during regional anesthesia and (2) to examine intrapartum perinatal factors that could influence fetal acid-base status.
Maternal hyperglycemia and fetal acidosis
Volume 157 Number 4, Part 1
867
Table I. Maternal and neonatal characteristics of the three groups Group 1: Glucose
Group 2: Ringer's lactate
(N = 12)
Group 3: Normal saline
(N = 11)
(N = 9)
Maternal characteristics Age (yr) Gravidity Parity Height (em) Weight (kg) Hematocrit(%)
25.3 2.8 2.0 28.6 73.6 34.1
Neonatal characteristics Birth weight (gm) Dubowitz results (wk)
3245 ± 398 39.1 ± l.3
± 5.6 ± l.3 ± l.2 ± l.O ± 25.9 ± 4.0
25.8 3.1 2.1 28.6 79.8 35.6
± 4.7 ± l.O ± l.O ± 1.0 ± 13.3 ± 3.3
3242 ± 322 38.8 ± l.3
25.3 3.6 2.1 28.4 85.8 34.8
± ± ± ± ± ±
5.1 l.O l.O 0.9 22.8 2.7
3383 ± 454 39.1 ± l.7
Values are mean ± SD.
Methods and material
All patients had normal, singleton, term gestations and were scheduled for elective cesarean section at Cleveland Metropolitan General Hospital. Patients with liver or kidney disease or any medical complications were excluded from the study. All patients had a negative screening test for diabetes mellitus. After written informed consent was obtained, patients were assigned by a random table of numbers into one of three groups. Group 1 patients received 1 L of 5% dextrose in water (glucose = 50 gm); group 2 patients received 1 L of Ringer's lactate; group 3 (control) patients received 1 L of normal saline solution. The intravenous fluids were administered rapidly during 20 to 30 minutes. Immediately after the hydration, all patients were maintained on a regimen of intravenous normal saline solution at a rate of 125 ml/hr until surgery and delivery of the baby. Epidural anesthesia was then induced with 0.5% or 0.75% bupivacaine without epinephrine. All patients were placed on the operating room table in the supine position with left lateral uterine displacement and received oxygen at 4 Llmin by nasal cannula. Maternal blood pressure was measured every minute with an Arteriosonde (Kontron Inc., Everett, Mass.) from the time of epidural anesthesia until delivery of the infant and every 5 minutes thereafter. Maternal hypotension (a decrease in the systolic blood pressure of >30 torr from baseline or an absolute level < 100 torr) was corrected by incremental doses of intravenous ephedrine (5 to 10 mg) and by increasing the rate of infusion of normal saline solution. Maternal blood samples (3 ml each) were obtained from an indwelling cannula in a peripheral hand vein. The sampling site was kept patent by a heparin lock. In order to obtain arterialized samples, the sampling site was kept warm (55° C) by a heating blanket. Maternal blood samples were obtained at the start of and
after acute intravenous hydration at 10, 20, and 30 minutes and continued every 10 minutes until delivery. Fetal arterial and venous blood samples were obtained at birth from a double-clamped umbilical cord. Neonatal blood samples were also obtained every 30 minutes for 2 hours by venipuncture or heel stick. Since the amount of blood drawn from the infant was limited, the samples at 90 and 120 minutes were analyzed for glucose only. All blood samples were obtained in heparinized syringes and immediately placed on ice. Plasma was separated and stored at - 10° C for later analysis. An aliquot of whole blood was precipitated with chilled 10% perchloric acid for lactate measurement. The data collected during the study also included the skin-to-delivery time interval and the uterine incisionto-delivery interval. At birth, 1- and 5-minute Apgar scores were recorded. Each infant was weighed and a complete physical examination was performed. The gestational age was determined by a modification of the Dubowitz examination by Ballard et a!. 11 Blood samples were analyzed for glucose, lactate, insulin, ketones ((3-hydroxybutyrate), and triglycerides. Maternal samples were analyzed for electrolytes, carbon dioxide, and blood urea nitrogen. The baseline maternal samples and umbilical cord samples were also analyzed for acid-base status (pH, Pco2 , Po 2 , and base deficit). Blood gas tensions and pH were measured with a microelectrode system (Model 13031, Instrumentation Laboratory, Lexington, Massachusetts). Plasma glucose was measured by a coupled enzymatic assay with the use of hexokinase and glucose-6-phosphate dehydrogenase on an automated centrifugal analyzer (Baker Instruments, Pleasantville, New York). Blood lactate, (3-hydroxybutyrate, and triglycerides were measured by enzymatic methods and insulin by a doubleantibody radioimmunoassay. All the assay methods have been previously reported from this laboratory. 15
868
October 198 7 Am J Obstet Gynecol
Philipson et al.
a
Infusion \
~ ~I
Neonate
300 • Glucose o Ringer's Lactate • Saline
~
~
Cl
!
~
200
Cl)
"' 0
0
::>
(!j
100
T
I---H=J 1T 10
~0//
~
4'o'1 t
20
lJ t
cv
D
t
CA
I'
~ 60
120
Minutes
Fig. I. Changes in plasma glucose concentration in the mother and infant after intravenous fluid administration. D = At delivery; CV = cord vein; CA = cord artery. Data are presented as mean± SD.
\ Infusion \
35
EJ
rco;dl
~
Cord Artery
Neonate
• Glucose a Ringer's Lactate • Saline
~
~0
25
!
Q)
a; 0
"'
15
..J
5 '----1,_0_..__3..__~-t/rl-'---'---2-'-o---'-'-4--''o--1/rf-l'f-t.----•t----1/1--'---6--'-0-L -1-'-2-0D CV
CA
Minutes
Fig. 2. Changes in blood lactate concentration in the mother and infant after intravenous administration. For abbreviations see legend to Fig. 1.
All data are presented as mean ± SD. Group comparisons were done with Student's t test. A p value of <0.05 was considered statistically significant.
Results All patients were at term and did not have any underlying medical problems. As shown in Table I, there were no significant differences among the three groups in their antenatal characteristics. The neonatal birth weight and gestational age at birth were also similar between the groups. The neonates were all term sized, appropriate for gestational age, and had 5-minute Apgar scores >7. One neonate in group I had meconium-
stained amniotic fluid below the vocal cords and had a !-minute Apgar score of 6. The 5-minute Apgar score in this neonate was 8. Plasma glucose. The changes in plasma glucose concentration in the three groups are displayed in Fig. I. The initial maternal plasma glucose concentration was similar in the three groups. Acute intravenous glucose (50 gm during 30 minutes) administration in group I caused a rapid increase in plasma glucose concentration, reaching a maximum of 225 ± 66 mg/dl at 20 minutes followed by a continuous gradual decline. At delivery, the maternal plasma glucose level in group I was still significantly elevated (134 ± 37
Maternal hyperglycemia and fetal acidosis 869
Volume 157 Number 4, Part 1
El
Infusion \
Neonate / _ lcOrdl L__ ~ Artery _ ___,
~
100 • Glucose o Ringer's Lactate • Saline
......
e-... ::l
..3
60
.5 3(/) .5
20
~ 10
30
20
~
40 I t
D
t
cv
t
CA
/f'I
~ 60 120
Minutes
Fig. 3. Changes in plasma insulin concentration in the mother and infant after intravenous fluid administration. For abbreviations see legend to Fig. l.
mg/dl). In contrast, acute maternal hydration with Ringer's lactate or with normal saline solution caused a small but insignificant decrease in plasma glucose concentration. The glucose concentration remained stable thereafter, so that at delivery the maternal plasrp.a glucose concentration was 83 ± 16 mg/dl in group 2 and 76 ± 9 mg/dl in group 3. In all three groups, the umbilical cord venous and arterial glucose concentrations were lower than the simultaneously obtained maternal peripheral venous gh.,cose levels at delivery. However, in the group administered glucose, the umbilical cord venous and arterial glucose concentrations were significantly elevated wP,en compared with tpose of groups 2 and 3 (p < 0.001). Even though the glucose concentration in the umbilical cord vein and artery was higher in group 1 than in groups 2 and 3, the plasma glucose concentration in the newborn infants at 30 minutes was similar in all three groups. The plasma glucose concentration in g-roup 1 decreased d1ning the next 30 minutes before stabilizing during the next hour. By 2 hours after birth, the neonatal plasma glucose concentration remained significantly lower in group 1 than in groups 2 and 3 (p < 0.01). In contrast, the plasma glucose concentration in the newborn infants in grqups 2 and 3 stabilized between 44 and 62 mg/dl for the next 90 minutes. Four of 12 neonates in group 1 and two of 11 neonates in group 2 developed asymptomatic hypoglycemia (blood glucose value of <25 mg/dl, Dextrostix, Ames Co., Elkhart, Indiana). No infants in group 3 developed hypoglycemia. One infant in group 1 de~ veloped tachypnea. At this time, the plasma glucose concentration was 18 mg/dL
BlQOd lactate. Changes in maternal, umbilical cord, and neonatal blood lactate concentrations are shown in Fig. 2. Initial maternal blood lactate concentrations in groups 1, 2, and 3 were 12 ± 3, 11 ± 3, and 10 ± 2 mg/dl, respectively. In groups 1 and 2, the maternal blood lactate concentration was significantly higher at the end of the intravenous infusion when compared with the respective initial blood lactate concentrations and also when compared with the value in the group receiving normal saline solution (group 3) (p < 0.01). However, by the time the epidural anesthesia was induced and surgery was qegun, the blood lactate concentratio!l in group 2 was similar to that of group 3. The maternal blood lactate concentration in the glucose infusion group remained significantly elevated throughout surgery and delivery when compare~ with the initial blood lactate concentration and also when compared with values in groups 2 and 3. The umbilical cord venous and arterial blqod lactate concentrations were higher than the maternal venous blood lactate concentrations in all three groups. The glucose infusion in group 1 caused a greater increase in the blood lact
870
Philipson et al.
October 1987 Am J Obstet Gynecol
Table II. Acid-base status at delivery Group 1: Glucose (N = 12)
Maternal vein at delivery pH
Group 2: Ringer's lactate (N = 11)
Group 3: Normal saline (N = 9)
Pco2 Po,
7.36 35.1 61.7 -4.9
± ± ± ±
0.04 5.3 41.1 2.0
7.36 33.7 68.7 -4.7
± ± ± ±
0.12 4.7 43.6 6.3
7.37 34.8 65.1 -3.7
± ± ± ±
0.05 6.9 31.0 2.3
Pco2 Po 2
7.31 41.8 31.4 -4.7
± ± ± ±
0.04* 4.8 8.9 2.5
7.35 40.6 30.5 -2.7
± ± ± ±
0.07 7.0 14.9 1.9
7.34 39.8 31.1 -3.7
± ± ± ±
0.04 4.8 4.6 3.6
7.21 57.7 14.0 -5.1 11.8
± ± ± ± ±
0.06* 5.6* 3.3 3.0 7.0*
7.27 49.2 17.1 -3.1 22.7
± ± ± ± ±
0.12 8.4 5.4 2.7 12.3
7.27 50.8 15.0 -4.3 17.6
± ± ± ± ±
0.06 6.0 3.6 3.4 6.7
Base deficit Cord vein pH Base deficit Cord artery pH
Pco, Po,
Base deficit Oxygen saturation
Values are mean ± SD. *p < 0.05 (group I compared with groups 2 and 3).
6.2, 10.5 ± 5.5, and 12.6 ± 9.0 f.LU/ml, respectively (Fig. 3). The glucose infusion in group 1 caused a shortterm rise in the insulin concentration so that at 20 minutes the plasma insulin concentration was 80.7 ± 44.8 f.LU/ml (p < 0.0001). At delivery, the maternal plasma insulin concentration in group 1 remained significantly elevated at 58.9 ± 38.4 f.LU/ml (p < 0.002) with the plasma insulin concentrations in groups 2 and 3 remaining unchanged. The insulin concentration in the umbilical cord vein (65.9 ± 67.9 f.LU/ml) and artery (55.6 ± 66.5 f.LU/ml) at delivery in group 1 was also significantly elevated when compared with the other two groups (p < 0.02). The insulin concentrations could not be measured in all the neonates at 30, 60, 90, and 120 minutes because of limited blood sample size. However, the neonatal insulin concentrations in groups 2 and 3 remained <3.5 f.LU/ml while in the neonates of group 1 higher insulin concentrations were maintained at 30 and 90 minutes after birth. Acid-base status. The maternal venous pH before the infusion was similar in the three groups. By the end of the infusion, the maternal pH in group 1 (7.36 ± 0.03) was significantly lower than the pH in group 2 (7.38 ± 0.2) and group 3 (7.39 ± 0.02) (p < 0.05). As shown in Table II, the maternal pH at delivery was similar between the groups. However, the umbilical arterial pH in group 1 (7.21 ± 0.06) was significantly lower than the umbilical arterial pH in group 2 (7.27 ± 0.12) and group 3 (7.27 ± 0.06) (p < 0.05). The umbilical venous blood pH in group 1 (7 .31 ± 0.04) was also significantly lower than the umbilical cord vein pH in group 2 (7.35 ± 0.07) and group 3 (7.34 ± 0.04) (p < 0.05). In the neonatal period, there was a tendency for lower venous pH values
in the neonates from group 1. However, no conclusions can be made as complete studies were done in only eight neonates. Additional acid-base results in the maternal vein at delivery and the umbilical cord vein and artery of the three groups are shown in Table II. There was a significantly higher umbilical cord artery Pco 2 in the glucose-infused group. The umbilical arterial hemoglobin oxygen saturation in the glucose group (11.8 ± 7.0) was significantly lower than that of the Ringer's lactate (22.7 ± 12.3) or normal saline solution (17.6 ± 6.7) groups (p < 0.05). f3-Hydroxybutyrate and triglycerides. As shown in Table Ill, acute glucose infusion in group 1 was associated with a significant decrease in the concentration of [3-hydroxybutyrate in the maternal blood at delivery (p < 0.001). In addition, the umbilical venous (p < 0.003) and arterial (p < 0.04) concentrations of [3-hydroxybutyrate were lower in group 1 than in groups 2 and 3. The concentration of [3-hydroxybutyrate in the maternal plasma tended to rise in the Ringer's lactate and normal saline solution groups. The triglyceride levels in the three groups demonstrated wide variations with large standard deviations within each group. No significant differences in the mean triglyceride levels between the groups were identified. Serum electrolytes, carbon dioxide, and blood urea nitrogen. The serum electrolytes, carbon dioxide, and blood urea nitrogen were similar in the three groups. Acute glucose infusion significantly decreased the serum sodium level from 137 ± 3 to 131 ± 3 mEq/L (p < 0.01). No significant change in other electrolytes, carbon dioxide, or blood urea nitrogen was seen in any of the groups.
Maternal hyperglycemia and fetal acidosis
Volume 157 Number 4, Part I
871
Table III. Changes in 13-hydroxybutyrate and triglycerides in the three groups {3-Hydroxybutyrate (mmol!L)
Mother Preinfusion Delivery Umbilical cord Vein Artery Neonate 30 min 60 min
Triglycerides (mgldl)
Group 1: Glucose
Group 2: Ringer's lactate
Group 3: Normal saline
Group 1: Glucose
Group 2: Ringer's lactate
Group 3: Normal saline
0.23 ± 0.26 0.11 ± 0.07*
0.21 ± 0.09 0.26 ± 0.15
0.24 ± 0.15 0.33 ± 0.16
209 ± 112 183 ± 95
221 ± 91 214 ± 83
233 ± 112 229 ± 91
0.09 ± 0.04t 0.07 ± 0.03:j:
0.12 ± 0.04 0.12 ± 0.06
0.17 ± 0.07 0.10 ± 0.02
15 ± 9 12 ± 5
18 ± 12 16 ± 9
13 ± 9 12 ± 5
0.09 ± 0.01 0.08 ± 0.02
0.07 ± 0.01 0.07 ± 0.03
0.09 ± 0.01 0.06 ± 0.03
23 ± 5 24 ± 8
16 ± II 22 ± 8
20 ± 5 20 ± 8
Values are mean ± SD. *p < 0.001, compared with groups 2 and 3. tp < 0.003, compared with groups 2 and 3. :j:p < 0.04, compared with groups 2 and 3.
Table IV. Influence of maternal hypotension on pH, glucose, and lactate at delivery
Maternal vein at delivery pH Glucose Lactate Umbilical cord vein pH Glucose Lactate Umbilical cord artery pH Glucose Lactate
Hypotension (N = 9)
No hypotension (N = 23)
Significance
7.31 ± 0.12 126 ± 49 19.1 ± 5.3
7.38 ± 0.04 90 ± 24 12.6 ± 5.0
p < 0.05 p< 0.01 p < 0.01
7.30 ± 0.04 Ill± 45 23.9 ± 9.8
7.34 ± 0.05 78 ± 22 15.9 ± 6.0
p < 0.05 p< 0.01 p < 0.01
7.19 ± 0.11 95 ± 42 26.3 ± 9.3
7.27 ± 0.06 62 ± 20 18.8 ± 7.1
p < 0.02 p < 0.01 p < 0.05
Values are mean ± SD.
Maternal hypotension. The influence of maternal hypotension on pH, plasma glucose, and blood lactate are shown in Table IV. Nine patients (28%) developed hypotension: six in group 1, two in group 2, and one in group 3. In comparison with the group that remained normotensive, the hypotensive group of patients had a significantly lower maternal vein pH at delivery as well as a lower pH in the umbilical cord vein and artery. Because the hypotensive patients included a greater proportion from the glucose group, this group also had elevated plasma glucose and blood lactate levels in the maternal vein at delivery and in the umbilical cord. In addition, all of these patients had received ephedrine for the correction of hypotension. Because of these differences in plasma glucose and blood lactate in the hypotensive group, the influence of maternal hypotension was examined within group 1 alone. When the six hypotensive patients were compared with six normotensive patients in group 1, there
were no significant differences in the pH, plasma glucose level, or blood lactate level. The glucose concentration in the maternal venous blood at delivery tended to be higher, although not significantly so, in the hypotensive group, possibly because of ephedrine administration. Uterine incision-to-delivery interval. In order to examine the effects of a prolonged uterine incisionto-delivery interval, patients with a uterine incisiondelivery interval > 180 seconds were compared with patients with a uterine incision-delivery interval of < 180 seconds. A prolonged uterine incision-todelivery interval did not significantly alter any of the parameters measured. Comment
The purpose of this study was to evaluate the maternal and fetal metabolic effects of three commonly used intravenous fluids during delivery. The results
872
Philipson et al.
demonstrate that acute glucose infusion before delivery resulted in maternal hyperglycemia, hyperinsulinemia, and an increase in blood lactate. These maternal perturbations were associated with fetal hyperglycemia and hyperinsulinemia, higher fetal blood lactate levels, and lower fetal arterial pH. The key finding of this study was the lower fetal arterial pH in the glucose-infused group. This increase in blood hydrogen ion concentration was accompanied by a significant rise in blood lactate, an increase in carbon dioxide, and a decreased fetal hemoglobin oxygen saturation. An explanation for this lactic acidosis may be obtained from animal data. In chronically catheterized ewes, direct glucose infusion to the fetus can induce an acceleration of the fetal metabolic rate and fetal oxygen consumption. 7 The increase in fetal oxygen consumption has been speculated to be responsible for the decrease in fetal arterial oxygen content. Robillard et aJ.5 have shown that chronic fetal hyperglycemia in sheep induces a similar state of fetal arterial hypoxia, metabolic acidosis, and fetal death. Direct infusion of insulin to the fetus caused an increased glucose uptake by the fetus and a decreased fetal arterial oxygen content.8· 9 Crandell et al. 16 showed that in chronically catheterized ewes hyperglycemia resulted not only in fetal hypoxemia and acidemia but also in a redistribution of the fetal combined ventricular output and a decrease in umbilical blood flow to the placenta. Although there was no significant change in oxygen delivery to the fetal heart, brain, or kidneys in this study, there was a marked decrease in fetal arterial oxygen saturation. On the basis of these studies in animals, one may speculate that acute maternal glucose infusion in human pregnancy results in an increased metabolic rate of the fetus. Glucose consumption and lactic acid production are increased. Fetal oxygen consumption is also increased and results in a decrease in fetal oxygen content and subsequent fetal hypoxemia. Fetal hypoxemia leads to a decrease in pH and, if severe, lactic acidosis. Redistribution of fetal cardiac output may be a compensatory adaptation to these metabolic changes. This study also examined other metabolic consequences of acute glucose infusion. Acute maternal hyperglycemia persisted more than 40 minutes after the infusion so that, by delivery, the maternal glucose remained elevated and the concentration of blood lactate had doubled. The umbilical cord vein and artery blood lactate concentrations remained twice as high as those of the control group. In contrast, the group that received the Ringer's lactate (28 mEq/L of lactate) did not demonstrate this altered metabolic lactic acidosis. Other perinatal factors, such as maternal position, maternal hypotension, and surgical time intervals at the time of cesarean section, were examined since these factors are known to influence fetal acid-base status at
October 1987 Am J Obstet Gynecol
delivery. 9· 13 All patients m this study had left lateral uterine displacement to avoid the effects of aortocaval compression. Maternal hypotension was rapidly corrected and did not influence neonatal acid-base status when controlled for glucose concentration. In this study, a prolonged skin-to-delivery interval did not influence the neonatal acid-base balance. Similar results have been reported with spinal anesthesia in the absence of hypotension." The uterine incision-todelivery interval also did not appear to influence the pH although there were only a few cases with a prolonged interval of >3 minutes. Therefore, by controlling for these confounding variables, the alterations in substrates anq lower pH were attributed to the metabolic effects of acute glucose infusion alone. In summary, short-term maternal administration of glucose before delivery is associated with fetal hyperglycemia, hyperinsulinemia, an increase in lactic acid, and a significant decrease in pH. Nevertheless, no obvious detrimental effects were observed on the mother, fetus, or neonate in this study. However, these metabolic changes occurred in normal patients at term without any signs of fetal distress or acidosis during the infusion, epidural anesthesia, or surgery. Since fetal distress and acidosis may occur during labor and delivery, acute glucose infusion in this clinical setting may exaggerate the metabolic acidosis associated with fetal distress. When intravenous fluids are necessary in obstetrics, either for hydration or administering medications, large amounts of glucose should be avoided. We thank Ms. Darlene Spagnoli, Ms. Monica Bakaitis, and Ms. Ellen Brown for their technical assistance, Ms. Sue Fitz for patient recruitment, and Sharon Anthony, R.N., and the Perinatal Clinical Research Center staff for their assistance with patient care. REFERENCES 1. 2.
3.
4. 5.
6. 7.
Wollm~n SB, Marx GF. Acute hydration for prevention of hypotension of spinal anesthesia in parturients. Anesthesiology 1968;29:374-9. Wright RG, Shnider SM. Hypotension and regional anesthesia in obstetrics. In: Shnider SM, Levinson G, eds. Anesthesia for obstetrics. Baltimore: Williams & Wilkins, 1987:293-9. Light IJ, Keenan WJ, Sutherland JM. Maternal intravenous glucose administration as a cause of hypoglycemia in the infant of the diabetic mother. AM J 0BSTET GYNECOL 1972; 113:345-50. . Tobin JD, Roux JF, Soeldner JS. Human fetal insulin response after acute maternal glucose administration during labor. Pediatrics 1969;44:668-71. Robillard JE, Sessions C, Kennedy RL, Smith FG. Metabolic effects of constant hypertonic glucose infusion in well-oxygenated fetuses. AM J 0BSTET GYNECOL 1978; 130:199-203. Philipps AF, Dubin JW, Matty PJ, Raye JR. Arterial hypoxemia and hyperinsulinemia in the chronically hyperglycemic fetal lamb. Pediatr Res 1982; 16:653-8. Philipps AF, Porte PJ, Stabinsky S, Rosenkrantz TS, Raye JR. Effects of chronic fetal hyperglycemia upon oxygen
Maternal hyperglycemia and fetal acidosis 873
Volume 157 Number 4, Part I
8.
9. 10. ll. 12.
consumption in the ovine uterus and conceptus. J Clin Invest 1984;74:279-86. Carson BS, Philipps AF, Simmons MA, Battaglia FC, Meschia G. Effects of a sustained insulin infusion upon glucose uptake and oxygenation of the ovine fetus. Pediatr Res 1980;14:147-52. Simmons MA, Jones MD, Battaglia FC, Meschia G. Insulin effect on fetal glucose utilization. Pediatr Res 1978;12: 90-2. Kenepp NB, Shelley WC, Gabbe SG, et al. Fetal and neonatal hazards of maternal hydration with 5% dextrose before cesarean section. Lancet 1982; l: 1150-2. Crawford JS, Burton M, Davies P. Time and lateral tilt at caesarean section. Br J Anaesth 1972;44:477-84. Fothergill RJ, Robertson A, Bond RA. Neonatal acidaemia
13. 14. 15. 16.
related to procrastination at cesarean section. J Obstet Gynaecol Br Commonw 1971;78:1010-23. Datta S, Ostheimer GW, WeissjB, Brown WU, Alper MH. Neonatal effect of prolonged anesthetic induction forcesarean section. Obstet Gynecol 1981 ;58:331-5. Ballard JL, Novak K, Driver M. A simplified score for assessment of fetal maturation and newly born infant. J Pediatr 1979;95:769-74. Kalhan SC, Tserng KY, Gilfillan CA, Dierker LJ. Metabolism of urea and glucose in normal and diabetic pregnancy. Metabolism 1982;31 :824-33. Crandell SS, Fisher DJ, Morriss FH. Effects of ovine maternal hyperglycemia on fetal regional blood flows and metabolism. Am J Physiol 1985;249:454-9.
Bound volumes available to subscribers Bound volumes of the AMERICAN jOURNAL OF OBSTETRICS AND GYNECOLOGY are available to subscribers (only) for the 1987 issues from the Publisher, at a cost of $51.00 ($74.00 international) for Vol. 156 (January-June) and Vol. 157 (July-December). Shipping charges are included. Each bound volume contains a subject and author index and all advertising is removed. Copies are shipped within 60 days after publication of the last issue in the volume. The binding is durable buckram with the JouRNAL name, volume number, and year stamped in gold on the spine. Payment must accompany all orders. Contact The C. V. Mosby Company, Circulation Department, 11830 Westline Industrial Drive, St. Louis, Missouri 63146, USA; phone (800) 325-4177, ext. 351. Subscriptions must be in force to qualify. Bound volumes are not available in place of a regular JOURNAL subscription.