Hyperlactacidemia associated with acute ritodrine infusion

American Journal of

Obstetrics and Gynecology l'o!ume 146 number I

OBSTETRICS

Hyperlactacidemia associated with acute ritodrine infusion STEPHEN R. RICHARDS, M.D. FRANK E. CHANG, M.D. LAURENCE E. STEMPEL, M.D. Columbus, Ohio The use of {1-mimetic tocolytics has been associated with various metabolic derangements which have been the topic of investigational reports. Lactic acidosis has been reported to follow concomitant steroid administration. The purpose of this investigation was to delineate more clearly hyperlactacidemia in patients receiving various tocolytics with and without concomitant use of steroids in a prospective manner. Twenty patients in premature labor who received ritodrine had markedly increased blood levels of lactate (baseline, 1.0 ± 0.1 mmoi/L [SEM] versus 3.5 ± 0.3 mmoi/L [SEM] after 6 hours' intravenous infusion). In the other treatment groups, ritodrine plus hydrocortisone, hydrocortisone alone, and magnesium sulfate alone, lactate levels failed to change significantly. Clinical relevance and implications for metabolic alterations associated with J3-mimetic tocolysis are discussed. (AM. J. OBSTET. GYNECOL. 146:1, 1983.)

THE uSE OF ,8-mimetic tocolytic agents for premature laboring patients is known to cause increased gluconeogenesis, glycogenolysis, and lipolysis. 1 These changes in metabolic activity have been associated with hyperglycemia, 2 hyperinsulinemia, 3 hypokalemia, 4 and hyperlactacidemia. 2 • 3 Two studies involving blood anions, electrolytes, and hyperlactacidemia suffer from small sample size and, in one, lactated Ringer's solution From the Department of Obstetrics and Gynecology, The Ohio State University College of Medicine. Received for publication September 14, 1982. Revised November 5, 1982. Accepted December I, 1982. Reprint requests: Stephen R. Richards, M.D., Department of Obstetrics and Gynecology, The Ohio State University Hospital, Columbus, Ohio 43210.

had been infused. 2 • 5 The importance of delineating lactic acid response to {3-mimetics is highlighted by the report of Desir and associates6 of a case of ritodrineinduced acidosis. This case also involved the administration of hydrocortisone to prevent respiratory distress syndrome. In order to underscore this potential problem, we now report our findings with the use of tocolytics with and without steroids. Patients and methods

A total of 35 patients was studied. Gestational age ranged from 27 to 34 weeks. Premature labor was diagnosed by documented cervical change. The choice of tocolytic agent (ritodrine versus magnesium sulfate) was made by the attending physician. Ritodrine was chosen (n = 20) unless a contraindication was present. Magnesium sulfate was the tocolytic agent in live pa-

2 Richards, Chang, and Stempel

May!. 19X3 Am. J. Obstet. Gvne< nl.

Table I. Patients who initially received subcutaneouslv administered terbutaline

8

:::i

'~

6

0

::!: ::!:

~

w

4

Patient

Dosage (f,Lg)

2

P. H. D. B.

250 250

~

1-' 0

< ..J

6

12 TIME ( hoiJrs}

2.0

1.5

18


0 Tl ME (hours l

:!.5 3.2

Six hours after intravenous ritodrine

Fig. 1. Individual lactate levels for those patients receiving ritodrine alone (n 20).

0

Six houn after intravenous ritodrine

Table II. Patients receiving proionged intravenous ritodrine and subsequently transferred to our facility

OJ

0

Lactate baseline (terbutaline)

6

Fig. 2. Comparison of lactate response for all treatment groups: • - - , ritodrine alone; o-o, ritodrine plus steroid; o--o, steroid alone; • - · magnesium sulfate. Mean± SEM. Asterisk: p < 0.001.

1.2 1.3

to 34 weeks), in whom steroids were given by random assignment of an existing protocol, were entered into the study as a control group. Two patients were given 250 p.g of subcutaneously administered terbutaline before transfer to our facility and subsequently received ritodrine tocolysis. Informed consent was obtained from all patients. Each patient served as her own control with the baseline laboratory swdies being performed after the receipt of l L of 0.45% normal saline in a 5% dextrose solution over 30 minutes prior to tocolytic or steroid agent. Blood collections were then made every 6 hours while the patient was receiving intravenous drug and every 12 hours while receiving oral tocolytic. Maternal vital signs were monitored every 15 minutes dur-

ing intravenous drug administration and fetal heart tients. Ritodrine was administered intravenously according to manufacturers instructions, beginning at 0.1 mg/min and increasing by 0.05 mg/min increments to a maximum of 0.35 mg/min as needed until the patient experienced cessation of premature labor for 12 hours

rate was monitored continuously by a standard fetal monitor. Blood analyses included serum lactate, sodium, potassium, chloride, glucose, and venous blood gases. Anion gap was derived from the formula sodium plus

or patient symptomatology required cessation of ther=

potassium minus chloride plus bicarbonate. Signifi-

apy or reduced dosage. Transition to oral ritodrine was made at 11.5 hours to 10 mg every 2 hours for the first 24 hours. Magnesium sulfate was administered intravenously as a 4 gm loading dose over 30 minutes followed by a 1 to 3 gm/hour infusion. The dose at which labor stopped was then maintained for 12 hours and the patient was placed on a regimen of an oral ,8-mimetic, with dosage adjusted to the maternal pulse. 7- 9 Steroids (hydrocortisone succinate, 500 mg every 8 hours for four doses by intravenous infusion) were given to those patients (n = 4) who continued to have uterine contractions after 4 to 6 hours of ritodrine therapy. In addition, a group of patients (n = 4) with premature rupture of membranes (gestational age, 27

cance was determined by means of Student's t test with p values <0.05 being regarded as significant.

Results Individual blood lactate levels for those patients receiving ritodrine are plotted in Fig. 1. Maximal levels (range, 1.5 to 6.2 mmol/L; mean, 3.5 ± 0.3 [SEM)) were recorded from 6 to 12 hours during ritodrine infusion (p < 0.001 ). At 18 hours, the lactate levels had fallen to or were returning to baseline levels, despite continued intravenous ritodrine in most instances. Lactate levels failed to significantly increase for the other treatment groups at 6 hours (Fig. 2): Magnesium sulfate group, 1.2 ± 0.3 mmol/L (SEM) baseline versus 1.3 ± 0.2 mmol/L (SEM) at 6 hours; steroid alone

Hyperlactacidemia with ritodrine infusion

Volume 14() Number I

3

Table III. Patients admitted in premature labor and receiving an oral (:3-mimetic

Patient

S. H. CH. D. M. K. B. C. F. R. C. R. W.

*R

=

Ritodrine; T

Oral agent

Lactate baseline

Tocolytic agent

R R R

0.6 1.4

T

0.7 1.0 1.0 0.8

Ritodrine Ritodrine Ritodrine Ritodrine Ritodrine Magnesium sulfate Magnesium sulfate

R R

T =

0.9

Six hours after intravenous tocolytic

'X Increase

1.5

l:iO

2.2 2.8 1.2 3.7 1.0 0.6

.")0

211 71 270 ()

terbutaiine.

group, 0.8 ± 0.2 mmol/L (SEM) baseline versus 1.2 ± 0.15 mmol/L (SEM) at 6 hours. When steroids were given to patients already receiving ritodrine lactate levels still did not increase above the preexisting baseline, 2.2 ± 0.5 mmol/L (SEM) versus 2.0 ± l.l mmol/L (SEM) at 6 hours after steroids. Two patients received subcutaneous terbutaline (250 ~-tg) before transfer to out facility and subsequently received intravenous ritodrine (Table 1). The baseline lactate was above laboratory normal range (0.5 to I .6 mmol!L) and increased approximately 60% at 6 hours in both instances. There were two patients who were transferred from their primary hospital because of failure to respond to intravenous ritodrine tocolysis (Table II). Baseline and 6=hour lactate levels \vere normal after resumption of intravenous ritodrine at our institution. A group of patients who had received oral (:3mimetics as outpatients and then admitted for intravenous tocolysis are identified in Table III. Five of these patients received ritodrine. Corresponding 6-hour lactate levels ranged from 50% to 270'/f above baseline. Two patients received magnesium sulfate and lactate levels did not increase. i1~ll baseline lactates v:ere ·within normal range. Venous pH and anion gap (Fig. 3) failed to reveal any significant trend in the ritodrine-treated group. Baseline pH ranged from 7.34 to 7.47 with 6-hour values ranging from 7.30 to 7.45. Similarly, other treatment groups showed no change in these parameters (data not shown). A significant decrease in serum potassium was noi:ed in the ritodrine-treated group: mean baseline, 3.7 ± 0.1 mEq/L (SEM), versus mean 6-hour posttreatment, 2.9 ± 0.1 mEq/L (SEM), p < 0.001. Values beyond 6 hours are not plotted in Fig. 4 as potassium supplementation was administered to most patients after this time. Potassium levels did not significantly change in the other treatment groups (data not shown). The response of glucose to various drug regimens is depicted by Fig. 5. Significant increases occurred in the

ritodrine group (baseline, 87 ± 6 mg/dl [SEM], versus 6-hour, 124 ± 7 mg/dl, p < 0.001) and steroid group (baseline, 84 ± 3 mg/dl, versus 6-hom, Hi ± 11 mg/dl, p < 0.01). The ritodrine plus steroid group exhibited an increasing trend that failed to achieve statistical significance (baseline, 106 ± 13 mg/dl, versus 6hour, 147 ± 15 mg/dl). This was probably a reflection of sample size. No cases of ketoacidosis were detected. Comment

The use of (:3-mimetic tocolytic agents has profound metabolic consequences 1 in addition to the well-known physiologic responses involving the respiratory and cardiovascular systems as well as uterine musculature. 7 Previous investigators have reported increased blood lactate levels fo!!O\·ving ,8-mimetic administration.l· 2 • :-). 10 In these reports various (:3-mimetic agents were studied, but a small population sample was used and there was no alternate tocolytic control group. With the use of separate control groups, magnesium sulfate toco!ysis, and steroids, we have compared ritodrine-treated patients with and without concomitant stt:roids for their metabolic consequences, specifically changes in blood lactate, venous pH and anion gap, glucose. and potassium. Hendricks 11 demonstrated a 64% increase in lactate levels during term labor. Our patients admitted in premature labor all demonstrated normal baseline lactate levels, though prelabor levels were not available for comparison. Our patients receiving intravenous ritodrine had significantly increased blood lactate levels in contrast to the magnesium sulfate and steroid alone groups. This obviates tocolysis per se as a cause of hyperlactacidemia. However, the preterm patients receiving only steroids were not in labor, therefore a true comparison cannot be made. The mechanism of (:3mimetic-induced hyperlactacidemia relates to {3-receptor activation of adenyl cyclase '\Vith the production of cyclic adenosine monophosphate. 12 This cyclic adenosine monophosphate production via kinase activity then stimulates glycogenolysis and lipolysis. 7 As

4 Richards, Chang, and Stempel

Mav 1. JYitl

Am.

7.5

~

7.4

18

180

14

160

Q.

~

7.3

.J 0

10

'E

1

140

tl'

7.2

0

12 6 TIME(hoursJ

18

6

0

6 12 TIME (hours)

18

Fig. 3. Individual venous pH and anion gap data for ritodrine

LIJ

~ u

Obstet.Gynecol

I

z Q

J:

J.

1

120

::::> .J

(!)

100

only group. 80

of

5

r--0

Tl ME ( hours )

6

Fig. 5. Comparison of glucose response for treatment groups: e--e, ritodrine alone; o--o, ritodrine plus steroid;

o----o, steroid alone. Mean :t SEM. Double asterisk: p < 0.001; asterisk: p < 0.01: NS not significant.

0

0

6 TIME (hours)

Fig. 4. Decrease in serum potassium exhibited by ritodrine only group.

muscle lacks glucose-6-phosphatase, lactate is the end product of muscle glycogenolysis. Increased free fatty acids from lipolysis 1 inhibits pyruvate oxidation which also results in lactate production. Importantly, none of our patients experienced the adverse effects of hyperlactacidemia. Indeed, consistent with the study of Schreyer and associates, 4 our patients displayed no trend in venous pH or anion gap despite increased lactate levels. Cotton and co-workers2 reported a fall in pH which nadired at one hour, remaining below baseline (but within normal range). Similarly, Kirkpatrick and colleagues5 reported a fall in pH, but within normal range. This group also reported an increase in anion gap and fall in bicarbonate. These disparate results are not readily explainable unless the small sample sizes are considered (n 6 and n 7, respectively). Desir and associates 6 reported a case of ritodrine-induced acidosis with concomitant steroid usage. The patient was found to have normal glucose tolerance after pregnancy; however, gestational diabetes was not excluded. Except for this case, all reports of ketoacidosis have involved diabetic patients, some receiving a /3-mimetic alone, 13• 14 others receiving a /3-mimetic plus steroid, 15• 16 and others receiving steroids alone. 9 When given in the manner we describe

steroids do not appear to synergistically increase ritodrine-induced hyperlactacidemia. /3-Mimetic therapy is thought to induce hyperglycemia via two different pathways; first, increased glycogenolysis and, second. increased gluconeogenesis. Lunnell and co-workers 1 demonstrated decreased plasma alanine with a concomitant increase in glucose levels in patients receiving salbutamol. The isolated gluconeogenic effect of glucocorticoid induction of hyperglycemia explains the lack of hyperlactacidemia with steroid administration alone or with /3-mimetics. Since glycogenolysis is not stimulated by glucocorticoids, the lactate formation by muscle does not take place as it does with /3-receptor stimulation. Therefore, steroid administration alone or when given with ritodrine does not induce or heighten existing hyperlactacidemia. Concurrent administration of these two agents is therefore not contraindicated when this single metabolic effect is considered. Hyperinsulinemia is a consequence of the hyperglycemia, however /3 2-mediated insulin release of the pancreatic /3-cells has been demonstrated. 11 Relevance to the pregnant diabetic patient is the increased insulin requirement to forestall ketoacidotic consequences_I7-HJ Therefore, the pregnant diabetic patient deserves intensive metabolic surveillance when /3-mimetic and/or steroid treatment is instituted. Hyperinsulinemia is felt to contribute greatly to the associated hypokalemia of /3-mimetic therapy by facilitating cell uptake of potassium. 4 • 20 Schreyer and colleagues4 have demonstrated that erythrocyte uptake of potassium can account for potassium loss from the plasma space. Total body potassium is maintained since urinary excretion of potassium is unaffected by /3-mi-

Volume Hti Cllumber l

metic treatment. 4 ' 21 The hypokalemia is transient, returning to normal by 24 hours. 20 Hence, in the normokalemic patient not receiving digitalis of diuretics, potassium supplementation is of questionable benefit. Disparities between metabolic consequences of intravenous versus oral or intramuscular administration has been noted by other investigators. 22 - 24 In contrast to intravenously administered ritodrine, neither oraJ2 2 nor intramuscular23 ritodrine affected carbohydrate homeostatis. Similarly, oral ritodrine/salbutamol displayed variable effects of serum potassium. 23 • 24 Intramuscular ritodrine gave a paradoxic increase in potassium.23 A similar lactate response was noted by our patients on oral betamimetic; baseline lactates were normal, however significant increases occurred in these same patients when given intravenous ritodrine. Subcutaneously administered terbutaline elevated baseline lactate levels, but further augmentation was derived from intravenous ritodrine. Extended intravenous rirodrine, however, is accompanied by normalization of lactate levels. This response is similar to that reported by Young and associates 20 who observed a return of potassium and glucose levels to baseline over 24 hours. Also patients with short lapses in intravenous ritodrine treatment do not respond with apparent augmentation of serum lactate. This infers that tachyphylaxis develops to this metabolic alteration of intravenous ritodrine. Kirkpatrick and co-workers 5 suggested a selective desensitization whereby actions of the /1-mimetic are maintained on heart and uterine muscle but lost on liver and skeletal muscle. The differential effect of oral ritodrine may represent subtherapeutic dosage and hence subrnetabolic effect unless titrated to patient pulse. 7 The metabolic disturbances we have outlined are of minimal concern in the usual patient in premature labor~ The patient at risk \vould be the individual \Vith known impaired glucose tolerance or the patient presenting late for prenatal care in premature labor with unknown carbohydrate status. These patients when given ~-mimetic tocolytics and/or steroids should be followed closely with serial blood sugar levels, electrolytes, lactate, and venous pH. Lactated Ringer's solutions should be relatively contraindicated in patients receiving ~-mimetic tocolysis.

REFERENCES I. Lunnel, N. D., Joelsson, I., Harsson, 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. Gynaecol. &and. 56:475, 1977. 2. Cotton, D. B., Strassner, H. T., Lipson, L. G., and Goldstein, D. A.: The effects of terbutaline on acid base, serum electrolytes, and glucose homeostasis during the management of premature labor, AM. J. 0BSTET. GYNECOL, 141:617, 1981.

Hyperlactacidemia with ritodrine infusion

5

3. Lipshitz, J., and Vinik, A. I.: The effects of hexoprenaline, a t3 2-sympathomimetic drug, on maternal glucose, insulin, glucogen, and free fatty acid levels, AM. J. 0BSTET. GYNECOL. 130:761, 1978. 4. Schreyer, P., Cos pi, E., Arieli, S., Maor, J .. and Modai, D.: Metabolic effects of intravenous ritodrine infusion in pregnancy, Acta Obstet. Gynaecol. Scand. 59:197, 1980. 5. Kirkpatrick. C., Quenon, M., and Desir, D.: Blood anions and electrolytes durin~~: ritodrine infu$ion in oreterm labor, AM. j.' 0BSTET. GYNECOL. 138:523, 1980. ' 6. Desir, D., Van Coevorden, A., Kirkpatrick, C., and Caufriez, A.: Ritrodrine-induced acidosis in i)regnancv, Br. Med. J. 2:1194, 1978. 7. Lipshitz, J.: Beta-adrenergic agonists, ~e rnllL Perina to!. 5:252, 1981. 8. TambyRaja, R. L., Atputharajah, \'., and Salmon, Y.: Prevention of prematurity in twins, Aust. NZ J. Obstet. Gynaecol. 18:179, 1978. 9. Baillie, P., Malan, A. F., Saunders, M. C., and Davey, D. A.: The active management of pretenn labour and its effects on fetal outcome, Aust. NZ J. Ohstet. Gynaecol. 16:94, 1976. 10. Goldberg, R., Joffe, B. I., Bersohn, 1., VanAs, M., Krut, L., and Seftel, H. C.: Metabolic responses to selective /3-adrenergic stimulation in man, Postgrad. Med. J. 51:53, 1975. 1 I. Hendricks, C. H.: Studies on lactic acid metabolism in pregnancy and labor. AM. J. 0BSTET. (~Y"SECOL. 73:492. 1957. 12. Kauppila, A., Tuimala, R., Ylikorkala, 0 .. Haapalahti . .J., Karppanen, H., and Viinikka, L.: Effects