- Email: [email protected]
Blood anions and electrolytes during ritodrine infusion in preterm labor
Blood anions and electrolytes during ritodrine infusion in preterm labor CHRISTINE KIRKPATRICK, M.D . ~~A RIA!'"~ I"~
E QUE t~ 0
I"~,
• • n. lVI.U.
DANIEL DESIR, M.D. Rrussds, Belgium Seven patients with threatened premature labor and six nonpregnant female control subjects were studied for 12 to 24 hours during ritodrine infusion in order to re-evaluate the effects of this beta-adrenergic tocolytic agent on blood anions, electrolytes, and acid-base balance. Dexamethasone was given intramuscularly in all patients and three control subjects at the beginning of ritodrine infusion. Lactate, pyruvate, betahydroxybutyrate, sodium, potassium, chloride, bicarbonate, pH, glucose, and insulin levels were measured in blood samples obtained at intervals of 2 to 6 hours. While pH remained in the normal range, increases in lactate and pyruvate levels and decreases in bicarbonate values resulted in a significant increase in anion gap 4 to 6 hours after the beginning of treatment. The blood potassium level fell simultaneously. These effects were transient, lasting less than 10 hours, despite continuation of ritodrine infusion and maintenance of tocolysis and tachycardia. Although larger increases in the blood glucose level were observed with combined therapy in control subjects, dexamethasone had no additional influence on potassium concentrations. Monitoring of patients treated by intravenous ritodrine should include routine measurements of blood glucose and electrolyte levels in order to detect its potentiaHy harmful metabolic side effects for mother and fetus. (AM. J. OBSTET. GYNECOL. 138:523, 1980.)
EFFECTIVENESS and adverse effects of betasympathomimetics such as terbutaline, salbutamol, fenoterol, and ritodrine. now widely used for inhibition of premature labor. have been the subject of extensive reviews in recent years. 1 In most patients, only cardiovascular side effects of these drugs are reported to cause mild discomfort.~ However, life-threatening hemodynamic3 and metabolic 4 disturbances have been occasionally induced by these agents in pregnant women. Metabolic alterations caused by beta-adrenergic stimulation have been described mainly in obstetric conditions. 5 However. amplitude and duration of these alterations have not been fully characterized, so that From the Departments of Obstetrics and Gynecology, Medical Chemistry and Medicine, Erasme and Saint-Pierre Hospitals, Free University of Brussels. This study was partially supported by the Ministere Beige de la Politique Scientifique (Actions de Recherche Concertees) and the Belgium F.N.R.S. Received for publication February 1, 1980. Revisedjuly 15, 1980. Accepted July 21, 1980. Reprint requests: Dr. C. Kirkpatrick, Department of Obstetrics and Gynecology, HOpital Erasme, Route de Lennik, 808, B-1070 Brussels, Belgium. 0002-9378/80/2!0523+05$00.50/0
©
!980 The C. V. Mosby Co.
guidelines for detection of potentially harmful complications of these drugs, such as hypokalemia or metabolic acidosis, are not available. In the current study, anions, electrolytes, and glucose and insulin levels were measured for periods up to 24 hours in pregnant patients and female nonpregnant control subjects who received a prolonged intravenous infusion of ritodrine, in order to delineate the plasma profiles of these blood constituents in routine clinical conditions.
Subjects and methods Pregnant patients and normal nonpregnant control subjects gave their informed consent for participation in this study after full explanation of its aims and potential risks. Seven women admitted during the third trimester of gestation because of preterm labor and treated by intravenous infusion of ritodrine were investigated during the first 24-hour period after tocolytic therapy was begun. None of these patients had any other abnormality. Blood samples were obtained through an indwelling catheter inserted in a forearm vein for measurements of plasma levels of lactate, pyruvate, sodium, potassium, chloride, bicarbonate, betahydroxybutyrate, and pH. Blood glucose, insulin, and pyruvate levels in patients were not considered,
523
524
Kirkpatrick, Quenon, and Desir Am.
BLOOD GLUCOSE
Table I. Blood values in patients during ritodrine infusion (n = 7) B
A
200
/
/
I
;"\;m·ember l. l9HO Obstet. (;yne< ol
J.
Potassium (mmoles/L) 3.5 ± 0.1 2.8±0.1 <0.005 Bicarbonate (mmoles/L) 19.0 ± 0.9 13.7 ± 0.8 <0.01 Lactate (mmoles/L) 0.8 ± 0.1 4.9 ± 0.6 <0.001 5.0 ± 1.4 Anion gap (mmoles/L) NS 8.6 ± 0.9 :-IS 0.22 ± 0.05 0.29 ± 0.09 Betahydroxybutyrate (mmoles/L) A: Basal levels (mean± SEM); B: maximal (minimal) values observed 2 to 12 hours after ritodrine infusion was begun (mean ± SEM). NS = Not significant.
I Table II. Blood values in control subjects during ritodrine infusion D
1.>2
(n = 6)
50
0
t
4
8
12
HOURS
Potassium 3.7 ± O.i (mmoles/L) Bicarbonate 20.8 ± 0.8 (mmoles/L) Lactate (mmoles/L) l.O ± 0.1 Pyruvate (mmoles/L) 0.11±0.01 Lactate/ pyruvate 8.9 ± 0.9 Anion gap 5.0 ± 1.1 (mmoles/L) 7.38 ± 0.01 pH Glucose (mg/100 ml) 86 ± 6 Insulin (~-tU/ml) 12 ± 3
(n = 3)
2.7 ± O.i
2.6 ± O.i
19.8 ± 0.7
15.3 ± 0.7
4.6 0.27 17.0 7.5
± l.l 6.8 ± 0.4 ± 0.03 0.53 ± 0.01 ± 2.3 12.9 ± 0.8 ± 1.4 10.6 ± 1.9
7.36 ± 0.03 7.34 ± 0.02 168 ± 30* 223 ± 20* 67 ± 32 78 ± 28
Fig. 1. Mean (±SEM) blood glucose values during treatment with intravenous ritodrine (six control subjects) with (--) or without (----)intramuscular dexamethasone. The arrow indicates the beginning of the treatment.
A: Basal levels (mean± SEMl: B,: maximal (minimall levels observed during ritodrine alo~e (mean ± SEM); B2 : m'aximal (minimal) levels observed during ritodrine and dexamethasone (mean ± SE~f). *B, versus B2 : p < 0.001.
except for the basal blood glucose value, which was in the normal range for all subjects. After a basal blood sample had been collected, ritodrine diluted in a 5% dextrose solution was infused by a pump at a rate increasing from 50 up to 150 /Jog/min until the tocolytic effect was obtained; blood samples were drawn at 2, 4, 6, 12, 18, and 24 hours. At 0 and 12 hours 12 mg of dexamethasone was injected in= tramuscularly in order to accelerate fetal lung matura-
emission flame photometry; plasma glucose was measured on a Beeckman glucose analyzer (glucose oxidase method). Insulin was assayed by radioimmunoassay; pH was measured on a radiometer M 27 pH meter. and betahydroxybutyrate was measured fluorimetrically. The anion gap was estimated as (Na + K + Ca)- (Cl + HC0 3 + protein). 7 Significance was determined by means of Student's t test for paired observations; p values of <0.05 were regarded as significant.
500 ml of dextrose solution) were added when the plasma potassium value \vas lo\ver than 3.0 mmoles/L. In six nonpregnant members of the medical staff, aged 22 to 29, ritodrine diluted in saline solution was infused at a constant rate ( 150 /Jog/min). Three subjects received 12 mg of dexamethasone intramuscularly at the beginning of the ritodrine infusion. Blood was drawn at 2-hour intervals for 12 hours for determination of plasma lactate, pyruvate, sodium, potassium, chloride, bicarbonate, pH, glucose, and insulin values. Lactate and pyruvate were assayed by the method of Henry and associates. 6 Electrolytes were determined by
Results Serum potassium fell significantly in patients (Table I) as well as in normal control subjects (Table II), with minimal values reached in both groups 4 hours after the ritodrine infusion was begun. Control subjects not receiving a supplement exhibited a return to normokalemia 10 to 12 hours after the treatment was started. All other metabolic effects in the control group were clearly observed with ritodrine alone but appeared more marked with the combined administration of ritodrine and dexamethasone (Table II). However,
Volume 138 :-lumber 5
201
Blood anions and electrolytes during ritodrine infusion
T
I
.
Iii
BICARBONATE
15
525
LACTATE
5
ANiON GAP
5
2
0
0
t
8
16
24
HOURS
Fig. 2. Mean (±SEM) plasma bicarbonate and anion gap values during treatment with intravenous ritodrine (seven patients). The arrow indicates the beginning of the treatment.
since only three subjects were investigated in each subgroup, these differences were not significant, except for blood glucose levels, which were markedly higher (p < 0.00 1) in subjects receiving dexamethasone (Fig. I). Plasma sodium and chloride concentrations (ranging from 130 to 141 mmoles/L and from 98 to Ill mmoles/L, respectively) were not affected significantly during ritodrine infusion. Plasma bicarbonate concentration decreased in patients (Table I) and control subjects (Table II), with minimal values being observed 12 and 8 hours. respectively, after the beginning of treatment (Fig. 2). .A. significant rise of plasma lactate \vas detected after 2 hours of infusion in patients (0.8 ± 0.1 versus 2.0 ± 0. 2 mmoles/L, p < 0.02) and control subjects ( 1.0 ± O.l versus 2.5 ± 0.2 mmoles/L, p < 0.001). while maximal concentrations of lactate (Table I and !I) occurred after 6 and 8 hours, respectively (Fig. 3). Pyruvate concentrations (Table II), measured only in normal control subjects, increased during ritodrine, with maximal values observed 8 hours after treatment was begun (Fig. 3). The lactate/pyruvate ratio (Table II) showed an ab-
PYRUVATE
1 t:::":
oj
0
t
4
8
12
HOURS
Fig. 3. Mean (±SEM) plasma lactate and pyruvate values during treatment with intravenous ritodrine (six control subjects). The arrow indicates the beginning of the treatment.
rupt increase 2 hours after the ritodrine infusion was begun; this increase was maintained subsequently above 12 up to 12 hours of infusion. In some individuals, an extreme rise in lactate and a fall in bicarbonate levels \vere associated ·with an anion gap above lO mmoles/L. However, anion gap increase reached significance only in the group of control subjects as a whole (Fig. 2). A systematic control of pH was performed only in control su~jects; a significant decrease was observed, but the values remained in the normal range. Measurements of betahydroxybutyrate levels in five patients did not disclose any effect of ritodrine infusion (Table I).
Comment The most striking metabolic effects in this series (hyperlactacidemia and hypokalemia) were observed in
526
Kirkpatrick, Quenon, and Desir Am.
pregnant subjects and nonpregnant control subjects, suggesting that pregnancy itself does not predispose to these effects. Hyperiactacidemia has been described in previous reports, 8 but the course of lactate variations in plasma during a sustained infusion had not yet been fuiiy characterized. The increased production of lactate caused by beta-adrenergic stimuiation is thought to be due to the combination of increased glycogenolysis and lipolysis 9 ; beta-adrenergic stimulation causes enhanced glycogenolysis and glycolysis and results in the exaggerated production of lactate and pyruvate. On the other hand, accelerated lipolysis provides large amounts of free fatty acids, 10 which have been shown to inhibit pyruvate oxidation and to favor anaerobic glycolysis and lactate accumulation and to increase the lactate/pyruvate ratio. 11 We observed such an elevation of the lactate/pyruvate ratio (Fig. 3) but we did not measure free fatty acid levels; lipolysis, if present, did not result in a significant elevation of production of ketone bodies since betahydroxybutyrate levels remained in the normal range in all subjects (Table I) and were not affected by prolonged infusion. The variations of these anions (lactate, pyruvate, and betahydroxybutyrate) in plasma during ritodrine infusion suggest that lactate was the major metabolic anion responsible for anion gap increase. We were not able to detect deviations of the blood pH from the normal range in our series. However, metabolic acidosis has been encountered occasionally in otherwise normal pregnant women undergoing acute tocolysis; in one single case, this acidosis has been clearly ascribed to an exaggeration of the hyperlactacidemia. 12 Hypokalemia resulting fr~m administration of beta= adrenergic agents has been frequently reported in dinical series. 13 This effect has been regarded as a by~ product of the combined glucose and insulin elevations which induce accelerated transfer of potassium into the intracellular compartmentY We did not observe significant differences in the amplitude of the plasma potassium drop in normal control subjects, who had mild hyperglycemia ( 168 ± 30 mg/dl), or in control subjects receiving dexamethasone, who exhibited much higher blood glucose levels (223 ± 20 mg/dl). Another mechanism for intracellular transfer of potassium could be found in the ultimate action of beta-
November 1. 1V80 Obstet. (;ynecol.
J.
adrenergic agents on smooth muscles 15 : Beta-receptor agonists induce relaxation of smooth muscle by stimulating adenylate cyclase which in turn produces cyclic adenosine monophosphate, activation of the Na+fK+ pump, and net exchange of intracellular Na+ for extraceliuiar K '. The increased gradient for Na 1 across the cell membrane accelerates the rate of Na+f Ca++ exchange, resuiting in increased ca++ etnux from the cytoplasm and finally changes in intracellular Ca'' distribution, which decreases the muscle contractibility. A byproduct of these movements of cations across the cell membrane could be a net flux of potassium from the extracellular to the intracellular compartment, with significant hypokalemia. The return to normal of most metabolic disturbances induced by ritodrine could be related to the selective desensitization of beta-adrenergic receptors, which has been characterized in various systems in vitro and in vivo. 16 The decrease in the number of binding sites for beta-adrenergic agonists observed after exposure to isoproterenol probably results in loss of cell responsiveness. The maintenance of tachycardia and tocolysis for days in pregnant women given betamimetics while metabolic effects, even for submaxiinal doses, are transient in terms of hours could reflect differential desensitization processes in various target organs, with liver and skeletal muscles exhibiting a faster and more pronounced loss of beta-adrenergic responsiveness than the heart and uterus. Although the number of subjects receiving ritodrine alone was small, our data suggest that most metabolic disturbances induced by betamimetics, and particularly blood glucose variations, are exaggerated when large doses of corticosteroids are associated \AJith toco!ytics in order to foster fetal lung maturation. From a clinical point of vie'.v, metabolic monitoring of all pregnant patients given intensive beta-adrenergic therapy for tocolysis should include determinations of blood glucose and electrolytes at inter'Cals of 2 to 6 hours during the first day of betamimetic treatment. Moreover, particular caution must be recommended in high-risk patients such as those with diabetes and those receiving diuretics. We thank A. Elslander, R.N., for her valuable help in this study.
REFERENCES 1. Caritis, S. N ., Edelstone, D. 1., and Mueller-Heubach, E.: Pharmacologic inhibition of preterm labor, AM. J. OsSTET. G~JECOL. 133:557, 1979. 2. Lauersen, N.H., Mer katz, I. R., Tejani, N ., Wilson, K. H.,
Roberson, A., Mann, L. l., and Fuchs, F.: Inhibition of premature labor: A multicenter comparison of ritodrine and ethanol, AM.j. OssTET. GYNECOL.l27:837, 1977. 3. Tinga, D. J ., and Aarnoudse,]. G.: Post-partum pulmo-
Volume 1:1H
Blood anions and electrolytes during ritodrine infusion
527
Number'\
4.
5.
6. 7.
8.
9.
nary oedema associated with preventive therapy for premature labor. Lancet I: 1026, 1979. Chapman. M. G.: Salbutamol induced acidosis in pregnant diabetics, Br. Med. J. 1:639, 1977. Lunell, N. 0., Joelsson, I.. Larsson, A., and Persson, B.: The immediate effect of a beta-adrenergic agonist (salbutamol) on carbohydrate and lipid metabolism during the third trimester of pregnancy, Acta Obstet. Gynecol. Seand. 56:475, 1977. Hemy,R.J.,Cannon, D.C .. and Winkelman,]. W.: Clinical Chemistry-Pt·inciples and Technics, Hagerstown, Maryland. 1974, Harper & Row, Publishers. Oh. \1. S., and Caroll, H. J.: The anion gap, N. Engl. J. Med. 297:814, 1977. Goldberg. R .. Joffe, B. I .. Bersohn. 1., Van As, M., Krut, L, and Seftel, M. C.: Metabolic responses to selective beta-adrenergic stimulation in man, Postgrad. Med. J, 51:53, 1975. Conradt, A .. Schlotter. C. M., and Unbehaun, V.: Lvpolysis, ketonemia, estrogen and gestagen concentrations during short time infusion therapy with betamimetics, in Weidinger. H., editor: Labour InhibitionBetamimetic Drugs in Obstetrics, Stuttgart, West C.ermanv, 1977, Gustav Fisher, p. 103.
10. Pilkington, T. R. E., Love, R. D., J11 of smooth muscle. N ature 277:32, 1979. 16. Wolfe, B. B .. Harden, T. K., and Molinoff P. B.: In vitro studY. ofbeta-ad~~':ergi~,:eceptnrs, Annu Re\·. Ph annacol. J OXICOI. 17:.1/:J, 19,1.
E QUE t~ 0
I"~,
• • n. lVI.U.
DANIEL DESIR, M.D. Rrussds, Belgium Seven patients with threatened premature labor and six nonpregnant female control subjects were studied for 12 to 24 hours during ritodrine infusion in order to re-evaluate the effects of this beta-adrenergic tocolytic agent on blood anions, electrolytes, and acid-base balance. Dexamethasone was given intramuscularly in all patients and three control subjects at the beginning of ritodrine infusion. Lactate, pyruvate, betahydroxybutyrate, sodium, potassium, chloride, bicarbonate, pH, glucose, and insulin levels were measured in blood samples obtained at intervals of 2 to 6 hours. While pH remained in the normal range, increases in lactate and pyruvate levels and decreases in bicarbonate values resulted in a significant increase in anion gap 4 to 6 hours after the beginning of treatment. The blood potassium level fell simultaneously. These effects were transient, lasting less than 10 hours, despite continuation of ritodrine infusion and maintenance of tocolysis and tachycardia. Although larger increases in the blood glucose level were observed with combined therapy in control subjects, dexamethasone had no additional influence on potassium concentrations. Monitoring of patients treated by intravenous ritodrine should include routine measurements of blood glucose and electrolyte levels in order to detect its potentiaHy harmful metabolic side effects for mother and fetus. (AM. J. OBSTET. GYNECOL. 138:523, 1980.)
EFFECTIVENESS and adverse effects of betasympathomimetics such as terbutaline, salbutamol, fenoterol, and ritodrine. now widely used for inhibition of premature labor. have been the subject of extensive reviews in recent years. 1 In most patients, only cardiovascular side effects of these drugs are reported to cause mild discomfort.~ However, life-threatening hemodynamic3 and metabolic 4 disturbances have been occasionally induced by these agents in pregnant women. Metabolic alterations caused by beta-adrenergic stimulation have been described mainly in obstetric conditions. 5 However. amplitude and duration of these alterations have not been fully characterized, so that From the Departments of Obstetrics and Gynecology, Medical Chemistry and Medicine, Erasme and Saint-Pierre Hospitals, Free University of Brussels. This study was partially supported by the Ministere Beige de la Politique Scientifique (Actions de Recherche Concertees) and the Belgium F.N.R.S. Received for publication February 1, 1980. Revisedjuly 15, 1980. Accepted July 21, 1980. Reprint requests: Dr. C. Kirkpatrick, Department of Obstetrics and Gynecology, HOpital Erasme, Route de Lennik, 808, B-1070 Brussels, Belgium. 0002-9378/80/2!0523+05$00.50/0
©
!980 The C. V. Mosby Co.
guidelines for detection of potentially harmful complications of these drugs, such as hypokalemia or metabolic acidosis, are not available. In the current study, anions, electrolytes, and glucose and insulin levels were measured for periods up to 24 hours in pregnant patients and female nonpregnant control subjects who received a prolonged intravenous infusion of ritodrine, in order to delineate the plasma profiles of these blood constituents in routine clinical conditions.
Subjects and methods Pregnant patients and normal nonpregnant control subjects gave their informed consent for participation in this study after full explanation of its aims and potential risks. Seven women admitted during the third trimester of gestation because of preterm labor and treated by intravenous infusion of ritodrine were investigated during the first 24-hour period after tocolytic therapy was begun. None of these patients had any other abnormality. Blood samples were obtained through an indwelling catheter inserted in a forearm vein for measurements of plasma levels of lactate, pyruvate, sodium, potassium, chloride, bicarbonate, betahydroxybutyrate, and pH. Blood glucose, insulin, and pyruvate levels in patients were not considered,
523
524
Kirkpatrick, Quenon, and Desir Am.
BLOOD GLUCOSE
Table I. Blood values in patients during ritodrine infusion (n = 7) B
A
200
/
/
I
;"\;m·ember l. l9HO Obstet. (;yne< ol
J.
Potassium (mmoles/L) 3.5 ± 0.1 2.8±0.1 <0.005 Bicarbonate (mmoles/L) 19.0 ± 0.9 13.7 ± 0.8 <0.01 Lactate (mmoles/L) 0.8 ± 0.1 4.9 ± 0.6 <0.001 5.0 ± 1.4 Anion gap (mmoles/L) NS 8.6 ± 0.9 :-IS 0.22 ± 0.05 0.29 ± 0.09 Betahydroxybutyrate (mmoles/L) A: Basal levels (mean± SEM); B: maximal (minimal) values observed 2 to 12 hours after ritodrine infusion was begun (mean ± SEM). NS = Not significant.
I Table II. Blood values in control subjects during ritodrine infusion D
1.>2
(n = 6)
50
0
t
4
8
12
HOURS
Potassium 3.7 ± O.i (mmoles/L) Bicarbonate 20.8 ± 0.8 (mmoles/L) Lactate (mmoles/L) l.O ± 0.1 Pyruvate (mmoles/L) 0.11±0.01 Lactate/ pyruvate 8.9 ± 0.9 Anion gap 5.0 ± 1.1 (mmoles/L) 7.38 ± 0.01 pH Glucose (mg/100 ml) 86 ± 6 Insulin (~-tU/ml) 12 ± 3
(n = 3)
2.7 ± O.i
2.6 ± O.i
19.8 ± 0.7
15.3 ± 0.7
4.6 0.27 17.0 7.5
± l.l 6.8 ± 0.4 ± 0.03 0.53 ± 0.01 ± 2.3 12.9 ± 0.8 ± 1.4 10.6 ± 1.9
7.36 ± 0.03 7.34 ± 0.02 168 ± 30* 223 ± 20* 67 ± 32 78 ± 28
Fig. 1. Mean (±SEM) blood glucose values during treatment with intravenous ritodrine (six control subjects) with (--) or without (----)intramuscular dexamethasone. The arrow indicates the beginning of the treatment.
A: Basal levels (mean± SEMl: B,: maximal (minimall levels observed during ritodrine alo~e (mean ± SEM); B2 : m'aximal (minimal) levels observed during ritodrine and dexamethasone (mean ± SE~f). *B, versus B2 : p < 0.001.
except for the basal blood glucose value, which was in the normal range for all subjects. After a basal blood sample had been collected, ritodrine diluted in a 5% dextrose solution was infused by a pump at a rate increasing from 50 up to 150 /Jog/min until the tocolytic effect was obtained; blood samples were drawn at 2, 4, 6, 12, 18, and 24 hours. At 0 and 12 hours 12 mg of dexamethasone was injected in= tramuscularly in order to accelerate fetal lung matura-
emission flame photometry; plasma glucose was measured on a Beeckman glucose analyzer (glucose oxidase method). Insulin was assayed by radioimmunoassay; pH was measured on a radiometer M 27 pH meter. and betahydroxybutyrate was measured fluorimetrically. The anion gap was estimated as (Na + K + Ca)- (Cl + HC0 3 + protein). 7 Significance was determined by means of Student's t test for paired observations; p values of <0.05 were regarded as significant.
500 ml of dextrose solution) were added when the plasma potassium value \vas lo\ver than 3.0 mmoles/L. In six nonpregnant members of the medical staff, aged 22 to 29, ritodrine diluted in saline solution was infused at a constant rate ( 150 /Jog/min). Three subjects received 12 mg of dexamethasone intramuscularly at the beginning of the ritodrine infusion. Blood was drawn at 2-hour intervals for 12 hours for determination of plasma lactate, pyruvate, sodium, potassium, chloride, bicarbonate, pH, glucose, and insulin values. Lactate and pyruvate were assayed by the method of Henry and associates. 6 Electrolytes were determined by
Results Serum potassium fell significantly in patients (Table I) as well as in normal control subjects (Table II), with minimal values reached in both groups 4 hours after the ritodrine infusion was begun. Control subjects not receiving a supplement exhibited a return to normokalemia 10 to 12 hours after the treatment was started. All other metabolic effects in the control group were clearly observed with ritodrine alone but appeared more marked with the combined administration of ritodrine and dexamethasone (Table II). However,
Volume 138 :-lumber 5
201
Blood anions and electrolytes during ritodrine infusion
T
I
.
Iii
BICARBONATE
15
525
LACTATE
5
ANiON GAP
5
2
0
0
t
8
16
24
HOURS
Fig. 2. Mean (±SEM) plasma bicarbonate and anion gap values during treatment with intravenous ritodrine (seven patients). The arrow indicates the beginning of the treatment.
since only three subjects were investigated in each subgroup, these differences were not significant, except for blood glucose levels, which were markedly higher (p < 0.00 1) in subjects receiving dexamethasone (Fig. I). Plasma sodium and chloride concentrations (ranging from 130 to 141 mmoles/L and from 98 to Ill mmoles/L, respectively) were not affected significantly during ritodrine infusion. Plasma bicarbonate concentration decreased in patients (Table I) and control subjects (Table II), with minimal values being observed 12 and 8 hours. respectively, after the beginning of treatment (Fig. 2). .A. significant rise of plasma lactate \vas detected after 2 hours of infusion in patients (0.8 ± 0.1 versus 2.0 ± 0. 2 mmoles/L, p < 0.02) and control subjects ( 1.0 ± O.l versus 2.5 ± 0.2 mmoles/L, p < 0.001). while maximal concentrations of lactate (Table I and !I) occurred after 6 and 8 hours, respectively (Fig. 3). Pyruvate concentrations (Table II), measured only in normal control subjects, increased during ritodrine, with maximal values observed 8 hours after treatment was begun (Fig. 3). The lactate/pyruvate ratio (Table II) showed an ab-
PYRUVATE
1 t:::":
oj
0
t
4
8
12
HOURS
Fig. 3. Mean (±SEM) plasma lactate and pyruvate values during treatment with intravenous ritodrine (six control subjects). The arrow indicates the beginning of the treatment.
rupt increase 2 hours after the ritodrine infusion was begun; this increase was maintained subsequently above 12 up to 12 hours of infusion. In some individuals, an extreme rise in lactate and a fall in bicarbonate levels \vere associated ·with an anion gap above lO mmoles/L. However, anion gap increase reached significance only in the group of control subjects as a whole (Fig. 2). A systematic control of pH was performed only in control su~jects; a significant decrease was observed, but the values remained in the normal range. Measurements of betahydroxybutyrate levels in five patients did not disclose any effect of ritodrine infusion (Table I).
Comment The most striking metabolic effects in this series (hyperlactacidemia and hypokalemia) were observed in
526
Kirkpatrick, Quenon, and Desir Am.
pregnant subjects and nonpregnant control subjects, suggesting that pregnancy itself does not predispose to these effects. Hyperiactacidemia has been described in previous reports, 8 but the course of lactate variations in plasma during a sustained infusion had not yet been fuiiy characterized. The increased production of lactate caused by beta-adrenergic stimuiation is thought to be due to the combination of increased glycogenolysis and lipolysis 9 ; beta-adrenergic stimulation causes enhanced glycogenolysis and glycolysis and results in the exaggerated production of lactate and pyruvate. On the other hand, accelerated lipolysis provides large amounts of free fatty acids, 10 which have been shown to inhibit pyruvate oxidation and to favor anaerobic glycolysis and lactate accumulation and to increase the lactate/pyruvate ratio. 11 We observed such an elevation of the lactate/pyruvate ratio (Fig. 3) but we did not measure free fatty acid levels; lipolysis, if present, did not result in a significant elevation of production of ketone bodies since betahydroxybutyrate levels remained in the normal range in all subjects (Table I) and were not affected by prolonged infusion. The variations of these anions (lactate, pyruvate, and betahydroxybutyrate) in plasma during ritodrine infusion suggest that lactate was the major metabolic anion responsible for anion gap increase. We were not able to detect deviations of the blood pH from the normal range in our series. However, metabolic acidosis has been encountered occasionally in otherwise normal pregnant women undergoing acute tocolysis; in one single case, this acidosis has been clearly ascribed to an exaggeration of the hyperlactacidemia. 12 Hypokalemia resulting fr~m administration of beta= adrenergic agents has been frequently reported in dinical series. 13 This effect has been regarded as a by~ product of the combined glucose and insulin elevations which induce accelerated transfer of potassium into the intracellular compartmentY We did not observe significant differences in the amplitude of the plasma potassium drop in normal control subjects, who had mild hyperglycemia ( 168 ± 30 mg/dl), or in control subjects receiving dexamethasone, who exhibited much higher blood glucose levels (223 ± 20 mg/dl). Another mechanism for intracellular transfer of potassium could be found in the ultimate action of beta-
November 1. 1V80 Obstet. (;ynecol.
J.
adrenergic agents on smooth muscles 15 : Beta-receptor agonists induce relaxation of smooth muscle by stimulating adenylate cyclase which in turn produces cyclic adenosine monophosphate, activation of the Na+fK+ pump, and net exchange of intracellular Na+ for extraceliuiar K '. The increased gradient for Na 1 across the cell membrane accelerates the rate of Na+f Ca++ exchange, resuiting in increased ca++ etnux from the cytoplasm and finally changes in intracellular Ca'' distribution, which decreases the muscle contractibility. A byproduct of these movements of cations across the cell membrane could be a net flux of potassium from the extracellular to the intracellular compartment, with significant hypokalemia. The return to normal of most metabolic disturbances induced by ritodrine could be related to the selective desensitization of beta-adrenergic receptors, which has been characterized in various systems in vitro and in vivo. 16 The decrease in the number of binding sites for beta-adrenergic agonists observed after exposure to isoproterenol probably results in loss of cell responsiveness. The maintenance of tachycardia and tocolysis for days in pregnant women given betamimetics while metabolic effects, even for submaxiinal doses, are transient in terms of hours could reflect differential desensitization processes in various target organs, with liver and skeletal muscles exhibiting a faster and more pronounced loss of beta-adrenergic responsiveness than the heart and uterus. Although the number of subjects receiving ritodrine alone was small, our data suggest that most metabolic disturbances induced by betamimetics, and particularly blood glucose variations, are exaggerated when large doses of corticosteroids are associated \AJith toco!ytics in order to foster fetal lung maturation. From a clinical point of vie'.v, metabolic monitoring of all pregnant patients given intensive beta-adrenergic therapy for tocolysis should include determinations of blood glucose and electrolytes at inter'Cals of 2 to 6 hours during the first day of betamimetic treatment. Moreover, particular caution must be recommended in high-risk patients such as those with diabetes and those receiving diuretics. We thank A. Elslander, R.N., for her valuable help in this study.
REFERENCES 1. Caritis, S. N ., Edelstone, D. 1., and Mueller-Heubach, E.: Pharmacologic inhibition of preterm labor, AM. J. OsSTET. G~JECOL. 133:557, 1979. 2. Lauersen, N.H., Mer katz, I. R., Tejani, N ., Wilson, K. H.,
Roberson, A., Mann, L. l., and Fuchs, F.: Inhibition of premature labor: A multicenter comparison of ritodrine and ethanol, AM.j. OssTET. GYNECOL.l27:837, 1977. 3. Tinga, D. J ., and Aarnoudse,]. G.: Post-partum pulmo-
Volume 1:1H
Blood anions and electrolytes during ritodrine infusion
527
Number'\
4.
5.
6. 7.
8.
9.
nary oedema associated with preventive therapy for premature labor. Lancet I: 1026, 1979. Chapman. M. G.: Salbutamol induced acidosis in pregnant diabetics, Br. Med. J. 1:639, 1977. Lunell, N. 0., Joelsson, I.. Larsson, A., and Persson, B.: The immediate effect of a beta-adrenergic agonist (salbutamol) on carbohydrate and lipid metabolism during the third trimester of pregnancy, Acta Obstet. Gynecol. Seand. 56:475, 1977. Hemy,R.J.,Cannon, D.C .. and Winkelman,]. W.: Clinical Chemistry-Pt·inciples and Technics, Hagerstown, Maryland. 1974, Harper & Row, Publishers. Oh. \1. S., and Caroll, H. J.: The anion gap, N. Engl. J. Med. 297:814, 1977. Goldberg. R .. Joffe, B. I .. Bersohn. 1., Van As, M., Krut, L, and Seftel, M. C.: Metabolic responses to selective beta-adrenergic stimulation in man, Postgrad. Med. J, 51:53, 1975. Conradt, A .. Schlotter. C. M., and Unbehaun, V.: Lvpolysis, ketonemia, estrogen and gestagen concentrations during short time infusion therapy with betamimetics, in Weidinger. H., editor: Labour InhibitionBetamimetic Drugs in Obstetrics, Stuttgart, West C.ermanv, 1977, Gustav Fisher, p. 103.
10. Pilkington, T. R. E., Love, R. D., J