Water intoxication associated with oxytocin administration during saline-induced abortion NIELS
H.
STANLEY New
York,
LAUERSEN, J.
New
M.D.
BIRNBAUM,
M.D.
York
Four cases of water intoxication in connection with oxytocin administration during salineinduced abortions are described. The mechanism of water intoxication is discussed in regard to these cases. Oxytocin administration during midtrimester-induced abortions is advocated only if it can be carried out under careful observations of an alert nursing staff, aware of the symptoms of water intoxication and instructed to watch the diuresis and report such early signs of the syndrome as asthenia, muscular irritability, or headaches. The oxytocin should be given only in Ringer’s lactate or, alternately, in Ringer’s lactate and a 5 per cent dextrose and water solutions. The urinary output should be monitored and the oxytocin administration discontinued and the serum electrolytes checked if the urinary output decreases. should not be administered in excess of 36 hours. Zf the patient has not aborted oxytocin should be discontinued for 10 to 12 hours in order to perform electrolyte tions and correct any electrolyte imbalance.
K~WNTREE,'*~ in 1923, seems to have been the first to describe the phenomenon of water intoxication in man. He found this complication in a number of patients suffering from diabetes insipidus to whom he had given repeated doses of a solution of pituitary extracts. Abdul-Karim and Assali,Y in 1961, were able to show that oxytocin could exert
its antidiuretic properties clinically. The following year, Liggins* reported a case of water intoxication in a patient who had received large doses of oxytotin for the treatment of missed abortion. Since then. several investigatorP7 have reported on this complication in connection with administration of large doses of oxytocin. The most commonly reported symptom of this syndrome has been grand ma1 seizures.‘-‘” Lilienl” and Gupta and CohenI have reported cases of maternal death. The number of reported cases of water intoxication has recently increased and all cases have occurred in connection with oxytocin administration during saline abortion. Four cases have occurred at The New York Hospital-Cornell Medical Center during the past three years,15* lG including one resulting in severe grand ma1 seizures and brain damage. This has prompted us to review our cases and outline a policy for oxytocin administration.
From the Department of Obstetrics and Gynecology, The New York Hospital-Cornell University Medical College. Supported in #art by Grants 73049 from The Rockefeller Received
for publication
Accepted
May
31,
April
No. 67032 Foundation.
The oxytocin by then the determina-
and
23, 1974.
1974.
Reprint reque.rts: Dr. Niels H. Lauersen, Department of Obstetrics and Gynecology, The New York Hospital-Cornell University Medical College, 1300 York Ave., New York, New York 10021. 2
Volume Number
121 1
Case
Water
reports
intoxication
with oxytocin
3
the saline instillation, intravenous oxytocin was started at a rate of 2 I.U. in 100 ml. of 5 per cent dextrose in water per hour. After 24 hours, the dosage was increased to 10 I.U. of oxytocin per 100 ml. per hour. No contrac-
seizure activity, and the patient was discharged in good condition. Case 3. This patient was a 21-year-old woman, gravida 4, para 2, admitted for elective abortion in her eighteenth week of gestation. She had several unsuccessful attempts at amniocentesis, but since only bloody fluid was obtained, no saline was instilled and it was decided to try to induce the abortion with a high dose of oxytocin administered intravenously. The patient was given 10 U. of oxytocin per hour in 100 ml. of Ringer’s lactate and 5
tions were observed after 20 hours on this dose oxytocin dose was therefore increased to 20 I.U. ml. per hour. This dosage was continued for
and the per 100 7 hours,
per cent dextrose in water solution, alternatingly. The fetus was aborted after 26 hours and 25 minutes and the placenta was passed spontaneously 2 hours later. After
when the patient was found to have edema of the legs and had gained 3.5 Kg. By then, she had received 380 U. of oxytocin and 5,100 ml. of 5 per cent dextrose in water in the course of 51 hours. Serum sodium was 124 mEq. per liter (normal, 130 to 150) and potassium 3.0 mEq. per liter (normal range, 4.0 to 5.5). The oxytocin was immediately discontinued and no further treatment was
the passage of the placenta, the patient experienced a seizure-like episode immediately preceded by a loud scream, characterized by focal motor activity of the upper extremities. She was given 100 mg. of sodium amytol intravenously, with relief of the neurological symptoms. The serum sodium at the time of the seizure was 130 mEq.
Case 1. The patient was a 33-year-old woman, para 3, with a past history of toxemia of pregnancy. She was admitted for elective abortion in the eighteenth week of pregnancy; 250 ml. of amniotic fluid were aspirated and 250 ml. of saline were instilled. Twenty-four hours after
initiated. The patient responded with good diuresis, which was not quantitated. The electrolytes normalized after 24 hours. The patient did not abort spontaneously, and the pregnancy was finally terminated through surgical evacuation. The patient had no neurological symptoms and wa? discharged in good condition. Case 2. The patient was a 17-year-old girl, gravida 1, who was admitted for elective abortion in the seventeenth week of pregnancy. An oxytocin drip was begun 30 minutes after an uncomplicated amniocentesis with aspiration of 50 ml. of amniotic fluid and instillation of 180 ml. of saline. Ten units of oxytocin were administered hourly in 100 ml. of a 5 per rent dextrose in water solution. The fetus was aborted 31 hours after the saline instillation but the placenta was retained and could not be removed digitally. Oxytocin administration was therefore continued at the same rate for 7 additional hours, after which time it was decided to perform a completion. A total of 380 U. of oxytocin had been administered in 38 hours in 3,800 ml. of a 5 per cent dextrose in water solution. Ten minutes following sodium pentobarbital and atropine medication in preparation for the curettage, the patient had a grand ma1 seizure of 1 to 2 minutes’ duration which responded to sodium amytol administration. Serum electrolyte determination revealed sodium, 122 mEq. per liter, and potassium, 3.1 mEq. per liter; the serum chloride level was not analyzed. She was thereafter maintained on Dilantin, 100 mg. three times a day. The oxytocin was imtnediately discontinued and normal saline and potassium were administered intravenously in small amounts. She responded with good diuresis; the amount was not quantitated and no diuretics were necessary. The electrolytes normalized within the following 10 hours. The serum sodium increased to 134 mEq. per liter and potassium to 3.8 mEq. per liter. The postoperative course was uncomplicated, there was no reoccurrence of the
per liter and potassium was 3.4 mEq. per liter. The patient had then received a total of 290 U. of oxytocin and 2,900 ml. of fluid in the course of 28 hours and 40 minutes. The oxytocin was immediately discontinued and the seizure did not recur. Case 4. The patient was a 29-year-old, single, white woman, gravida 1, para 0, who entered the hospital in her fourteenth week of gestation for a saline abortion. Amniocentesis was performed with difficulty; 20 ml. of clear amniotic fluid were removed and 200 ml. of a 20 per cent saline solution were instilled. An oxytocin infusion of 10 U. in 100 ml. of a 5 per cent dextrose in water solution per hour was started 2 hours after the saline instillation. Due to a combination of circumstances, the intravenous oxytocin administration was continued at the same dose level until the time of abortion on the third day. Forty-eight hours after the saline instillation the urine output had decreased and the diuresis was only 350 ml. in the following 24 hours. About 64 hours after the initiation of the oxytocin administration the patient had developed good uterine contractions and at the same time she had become very tense and nervous. She was therefore given 25 mg. of meperidine and 25 mg. of promethazine intramuscularly. She aborted 45 minutes after this medication and the placenta was passed a few minutes later. Immediately after the passage of the placenta, the patient developed a grand ma1 seizure. It is uncertain how long the seizure lasted but possibly 15 minutes passed before she was attended by a physician and the seizure could be controlled .by administration of 10 mg. of diazepam intravenously. The patient had been given a total of 800 U. of oxytocin in 8,000 ml. of 5 percent dextrose in water solution. The serum sodidm level shortly after the seizure was 111 mEq. per liter and the chloride was 76 mEq. per liter. on fluid restriction and continued three times a day, and diazepam,
The patient was placed on Dilantin, 100 mg. 5 mg. three times a day,
4
Lauersen
and
Birnbaum
and there was no recurrent seizure activity. She responded with an enormous diuresis of 5.280 ml. in the following 24 hours without any administration of diuretics. The patient was stuporous for 24 hours following the incident; then she showed some response to deep pain. The electrolytes normalized within 24 hours. A surgical completion was necessary on the second postabortion day. She developed endometritis, which was treated with high doses of antibiotics before any improvement occurred. Repeated spinal taps were negative for red and white blood cells. Electroencephalogram showed bilateral cerebral dysfunction with some left-sided focal features. The patient was transferred to the neurological service and had to remain in the hospital for five weeks. She remained delirious for three weeks after the seizure, during which time she was aphasic, producing only highpitched screams. By the fourth week she was able to produce a few simple words and thereafter received speech therapy. At the time of discharge she was able to repeat a few words and phrases but was unable to read or write. She was discharged with the following diagnoses: diffuse cerebral damage with left posttemporal dysfunction secondary to hypoxia; grand ma1 seizure secondary to water intoxication with hyponatremia. Comment As mentioned, the syndrome of water intoxication was first described in 1923 by Rowntree’l ’ who observed the phenomenon in a number of experiments in man and animals. He was able to produce the syndrome in patients with diabetes insipidus by giving them repeated doses of pituitary extract and in animals by administration of large amounts of distilled water. The syndrome of water intoxication in all experiments was characterized by asthenia, followed by muscular irritability, convulsions, and finally death. These symptoms could be relieved and the animals saved from death by administration of a salt solution. Rowntree’! 2 was also able to prevent water intoxication in animals by giving a 10 per cent solution of sodium chloride immediately prior to the water administration. He concluded, therefore, that the symptoms were due to a salt-water imbalance in the central nervous system. Pathologic examination of the animals that had suffered from water intoxication consistently revealed cerebral edema, which Rowntree took as the cause of convulsive seizures and subsequent death. It is now believed that it is the hyponatremia that causes the water intoxication symptoms, Plum and Posner17 have theorized that the pathogenesis is related to altered excitability of the neural membranes. The membrane potential is influenced by the ratio of
January 1, 197:~ Am. J. Obstet. Gynecol
sodium outside to sodium inside the cell and, more important, by the ratio of potassium outside to POtassium inside the cell. During hyponatremic episodes there may be shifts of sodium and potassium in neurons, accentuating the excitability and producing the neurological symptoms. Moss18 and Brockbank’” have reported extreme muscular cramps in workmen who perspire freely and consume large quantities of water. These symptoms could be relieved by administration of salt water. The need for extra intake of salt in stokers and others who work under very hot conditions is well known. Helwig and associates20 reported a death due to water intoxication in a woman who absorbed 9,000 ml. of water in 26 hours. Van Dyke and co-worker? have made extensive studies of the oxytocic and antidiuretic properties of the two neurohypophysial hormones oxytocin and vasopressin (or antidiuretic hormone) . In pharmacological assays, the antidiuretic activity of oxytocin is only about 0.5 per cent of the activity of vasopressin; likewise, the oxytocic activity of vasopressin is only a few per cent of that of oxytocin. Not only in animals, but also in the human subject, oxytocin is found to have antidiuretic properties and vasopressin oxytocic properties, but the ratios based on animal assays do not apply to the human subject at all. Thus vasopressin has, in equimolar amounts, a stronger oxytocic effect on the human nonpregnant uterus than does oxytocin.“, 23 The response of the target organ depends on other hormones, in case of the uterus upon ovarian or placental steroids.24 The effect of oxytocin on the urinary excretion of water and electrolytes in the human subject was first studied by ThomsorP in Great Britain and Abdul-Karim and Assali3 in the United States; both studies clearly demonstrated an antidiuretic effect of oxytocin. To exclude the possibility that oxytocin antidiuresis could be mediated through stimulation of antidiuretic hormone secretion, Abdul-Karim and Assali administered oxytocin to two patients with diabetes insipidus and three normal subjects whose endogenous antidiuretic hormone release had been blocked by alcohol. The results demonstrated that oxytocin has a direct action on the kidney. In pregnant patients who were given oxytocin for induction of labor, Abdul-Karim and Assali found the antidiuretic action to become evident before the oxytocic effect. Abdul-Karim and Assali demonstrated that the
Volume Number
121 1
antidiuretic effect during an infusion of oxytocin began at an infusion rate of about 15 mu. per minute and is maximal at 45 mu. per minute. The amount of oxytocin did not seem to be a controlling factor in the production of water intoxication when it exceeded 4.5 mu. per minute. This explains why several of the patients reported to have water intoxication have developed symptoms on relatively low doses of oxytocin. to According to Pickford, 26 the renal response vasopressin is modified by a variety of hormones, adrenomedullary hormones, corticosteroids, parathyroid hormones, and ovarian steroids. Much less is known about the influence of other hormones on the renal response to oxytocin, but Pickford did find an effect of estrogens in dogs, and Abdul-Karim and Assali3 found a difference in the human renal response between term pregnant and nonpregnant subjects. This suggests that the placental steroids influence the renal sensitivity to oxytocin but whether the kidney in the midtrimester is particularly sensitive to oxytocin is not known. The doses of oxytocin given in cases of saline-induced abortion are usually of such a magnitude that even with a relatively low renal sensitivity, an antidiuretic effect is to be expected. In all four cases of water intoxication described here there was a decrease in the sodium levels in the serum. This hyponatremia is believed by some investigators17 to be the primary cause of the syndrome. In the majority of the reported case&l6 water intoxication occurred when large amounts of oxytocin were given in (1) saline-induced abortion, (2) missed abortion, and (3) postpartum hemorrhage. In all cases, oxytocin was administered intravenously in electrolyte-free solutions; this undoubtedly was an additional factor, since Rowntree could prevent the symptoms in animals by treatment with salt solutions. In three of our four cases (nos. 1, 2, and 4)) the patients were given large amounts of oxytocin in 5 per cent dextrose in water without any electrolyte administration. All three patients responded with good diuresis and relief of symptoms after discontinuation of the oxytocin infusion and administration of electrolyte solutions. In ‘case 3, water intoxication developed in spite of the fact that oxytocin was administered in Ringer’s lactate solution, but this patient was not given an intraamniotic instillation of hypertonic saline. Several investigators 28, 29 have shown that there is a slight increase in sodium chloride concentra-
Water
intoxication
with
oxytocin
5
tion in the serum in the first hours after an intraamniotic instillation of a 20 per cent saline solution. This might have some beneficial effect in preventing water intoxication during subsequent oxytocin administration, but the most important precaution appears to be to administer the oxytocin in an electrolyte solution. The hypertonic saline blocks vasopressin (ADH) release ; this tends to facilitate excretion of the salt but may lead to dehydration. Oxytotin will slow down the excretion through the kidneys and the carrier solution will tend to hydrate the patient. Up to a point, the two processes will tend to balance each other but, if oxytocin is given too long without electrolytes, water intoxication will occur. We have now treated more than 800 cases of oxytocin administration during midtrimester saline abortions where the oxytocin was administered at the rate of 167 mu. per minute in alternating solutions of Ringer’s lactate and 5 per cent dextrose and water without any cases of water intoxication; oxytocin has not been administered for more than 36 hours. All the cases described occurred when large amounts of oxytocin in nonelectrolyte solutions were used over a prolonged period. As shown by Lauersen and Schulman,‘5 the abortion time can be shortened to about 20 hours and the complication rate greatly reduced when 10 U. of oxytocin are given hourly ( 167 mu. per minute), starting immediately after saline instillation. The oxytocin should be given in Ringer’s lactate or, alternately, in Ringer’s lactate and 5 per cent dextrose and water solutions. These authors stressed the importance of careful monitoring of the urinary output and of stopping the oxytocin administration to check the serum electrolytes if the urinary output decreased. In addition, they recommended discontinuation of the oxytocin administration for 10 to 12 hours if the patient had not aborted within 36 hours, in order to perform electrolyte determination and correct any electrolyte imbalance. These recommendations are fully supported by the experience gained from our four cases of water intoxication. In three of our cases the patients were attended by a physician immediately after the convulsive seizure occurred and the hypoxic state was apparently so short that the patient could be discharged without cerebral damage. In case 4, the seizure lasted several minutes and the patient developed diffuse cerebral damage; at the time of discharge, six weeks after the incident, she was still
6
Lauersen
unable
to
important the
and
Birnbaum
speak, to
read,
have
symptoms
of
an
or alert
water
write. nursing
intoxication,
It
is
therefore
staff,
aware
of
and
instructed
to
watch
the
of
the
syndrome
or
hradache.
diuresis
and as
asthrnia,
to
report
such
muscular
early
si,qns
irritability.
REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16.
Rowntree, I,. G.: Arch. Intern. Med. 32: 157, 1923. Rowntree, L. G.: J. Pharmacol. Exp. Ther. 29: 135, 1926. Abdul-Karim, R., and Assali, N. S.: J. Lab. Clin. Med. 57: 522, 1961. Liggins, G. C.: J. Obstet. Gynaecol. Br. Commonw. 69: 277, 1972. Pittman, J. G.: N. Engl. J. Med. 268: 481, 1963. Potter, R. R.: Obstet. Gynecol. 23: 699, 1964. Josey, W. E., Pinto, A. P., and Plant, R. A.: AM. J. OBSTET. GYNECOL. 104: 926, 1969. Silva, P., and Allan, M. S.: Obstet. Gynecol. 27: 517, 1966. Self, J.: Am. J. Med. Sci. 252: 573, 1966. Leventhal, J. M., and Reid, D. E.: AM. J. OBSTET. GYNECOL. 102: 310, 1968. Burt, R. L., Oliver, K. L., and Whitener, D. I,.: Obstet. Gynecol. 34: 212, 1969. Bilek, W., and Dorr, P.: Can. Med. Assoc. J. 103: 379, 1970. Lilien, A. A.: Obstet. Gynecol. 32: 171, 1968. Gupta, D. R., and Cohen, N. H.: J. A. M. A. 220: 681, 1972. Lauersen, N. H., and Schulman, J. D.: AM. J. OBSTET. GYNECOL. 115: 420, 1973. Mann, L. I., Duchin, S., Newman, M., and Weiss, R. R.: Obstet. Gynecol. 41: 748, 1973.
Attention The
Editors
to
Authors: Publisher
OF OBSTETRICS
have
in
AND
agreed
Reference to add
18. 19. 20. 21. 22. 23. 24. 25. 26.
27. 28. 29.
Plum, F., and Posner, J. B.: In Davis, F. A., editor: Diagnosis of Stupor and Coma, ed. 2, Philadelphia, 1972, F. A. Davis Company. Moss, K. N.: Proc. R. Sot. Land. Ser. B 95: 181, 1923-1924. Brockbank, E. M.: Br. Med. J. 1: 65, 1929. Helwig, F. C., Schutz, C. B., and Curry, D. E.: J. A. M. A. 104: 1569, 193.5. Van Dyke, H. B., Adamsons, K. Jr., and Engcl, S. L.: Recent Progr. Hormone Res. 11: 1, 1955. Dahle, T.: Acta Obstet. Gynecol. Stand. 30: (Suppl. 1): 1950. Coutinho, E. M., and Lopes, A. C. V.: Am. J. Obstet. Gynecol. 102: 479, 1968. Berde, B.: In Handbook of Experimental Pharmacology, Vol. XXIII, Heidelberg, 1968, Springer Verlag. Thomson, W. B.: J. Physiol. 150: 284, 1960. Pickford, M.: In Harris, G. W., and Donovan, B. D., editors: The Pituitary Gland, London, 1966, Butterworth & Company. Abdul-Karim, R. W., and Rizk, P. T.: Obstet. Gynecol. Surv. 25: 80.5, 1970. Brewer, I,. I,., Johnson, W. I,., and Hunter, C. A.: Asr. J. OBSTET. GYNECOI,. 101: 1078, 1968. Berkowitz, R. L. T.: Clin. Obstet. Gynecol. 14: 107, 1971.
Style
the article title to references in the AMERICAN References will now conform to the style of the Cumulated Index Medicus, viz., name of author, title of article, name of periodical, volume, page, and year. Authors are always encouraged to limit references to sixteen for the following JOURNAL sections: Obstetrics, Gynecology, and Fetus, Placenta, and Newborn.
JOURNAL
and
Change
17.
GYNECOLOGY.
H.
STANLEY New
York,
LAUERSEN, J.
New
M.D.
BIRNBAUM,
M.D.
York
Four cases of water intoxication in connection with oxytocin administration during salineinduced abortions are described. The mechanism of water intoxication is discussed in regard to these cases. Oxytocin administration during midtrimester-induced abortions is advocated only if it can be carried out under careful observations of an alert nursing staff, aware of the symptoms of water intoxication and instructed to watch the diuresis and report such early signs of the syndrome as asthenia, muscular irritability, or headaches. The oxytocin should be given only in Ringer’s lactate or, alternately, in Ringer’s lactate and a 5 per cent dextrose and water solutions. The urinary output should be monitored and the oxytocin administration discontinued and the serum electrolytes checked if the urinary output decreases. should not be administered in excess of 36 hours. Zf the patient has not aborted oxytocin should be discontinued for 10 to 12 hours in order to perform electrolyte tions and correct any electrolyte imbalance.
K~WNTREE,'*~ in 1923, seems to have been the first to describe the phenomenon of water intoxication in man. He found this complication in a number of patients suffering from diabetes insipidus to whom he had given repeated doses of a solution of pituitary extracts. Abdul-Karim and Assali,Y in 1961, were able to show that oxytocin could exert
its antidiuretic properties clinically. The following year, Liggins* reported a case of water intoxication in a patient who had received large doses of oxytotin for the treatment of missed abortion. Since then. several investigatorP7 have reported on this complication in connection with administration of large doses of oxytocin. The most commonly reported symptom of this syndrome has been grand ma1 seizures.‘-‘” Lilienl” and Gupta and CohenI have reported cases of maternal death. The number of reported cases of water intoxication has recently increased and all cases have occurred in connection with oxytocin administration during saline abortion. Four cases have occurred at The New York Hospital-Cornell Medical Center during the past three years,15* lG including one resulting in severe grand ma1 seizures and brain damage. This has prompted us to review our cases and outline a policy for oxytocin administration.
From the Department of Obstetrics and Gynecology, The New York Hospital-Cornell University Medical College. Supported in #art by Grants 73049 from The Rockefeller Received
for publication
Accepted
May
31,
April
No. 67032 Foundation.
The oxytocin by then the determina-
and
23, 1974.
1974.
Reprint reque.rts: Dr. Niels H. Lauersen, Department of Obstetrics and Gynecology, The New York Hospital-Cornell University Medical College, 1300 York Ave., New York, New York 10021. 2
Volume Number
121 1
Case
Water
reports
intoxication
with oxytocin
3
the saline instillation, intravenous oxytocin was started at a rate of 2 I.U. in 100 ml. of 5 per cent dextrose in water per hour. After 24 hours, the dosage was increased to 10 I.U. of oxytocin per 100 ml. per hour. No contrac-
seizure activity, and the patient was discharged in good condition. Case 3. This patient was a 21-year-old woman, gravida 4, para 2, admitted for elective abortion in her eighteenth week of gestation. She had several unsuccessful attempts at amniocentesis, but since only bloody fluid was obtained, no saline was instilled and it was decided to try to induce the abortion with a high dose of oxytocin administered intravenously. The patient was given 10 U. of oxytocin per hour in 100 ml. of Ringer’s lactate and 5
tions were observed after 20 hours on this dose oxytocin dose was therefore increased to 20 I.U. ml. per hour. This dosage was continued for
and the per 100 7 hours,
per cent dextrose in water solution, alternatingly. The fetus was aborted after 26 hours and 25 minutes and the placenta was passed spontaneously 2 hours later. After
when the patient was found to have edema of the legs and had gained 3.5 Kg. By then, she had received 380 U. of oxytocin and 5,100 ml. of 5 per cent dextrose in water in the course of 51 hours. Serum sodium was 124 mEq. per liter (normal, 130 to 150) and potassium 3.0 mEq. per liter (normal range, 4.0 to 5.5). The oxytocin was immediately discontinued and no further treatment was
the passage of the placenta, the patient experienced a seizure-like episode immediately preceded by a loud scream, characterized by focal motor activity of the upper extremities. She was given 100 mg. of sodium amytol intravenously, with relief of the neurological symptoms. The serum sodium at the time of the seizure was 130 mEq.
Case 1. The patient was a 33-year-old woman, para 3, with a past history of toxemia of pregnancy. She was admitted for elective abortion in the eighteenth week of pregnancy; 250 ml. of amniotic fluid were aspirated and 250 ml. of saline were instilled. Twenty-four hours after
initiated. The patient responded with good diuresis, which was not quantitated. The electrolytes normalized after 24 hours. The patient did not abort spontaneously, and the pregnancy was finally terminated through surgical evacuation. The patient had no neurological symptoms and wa? discharged in good condition. Case 2. The patient was a 17-year-old girl, gravida 1, who was admitted for elective abortion in the seventeenth week of pregnancy. An oxytocin drip was begun 30 minutes after an uncomplicated amniocentesis with aspiration of 50 ml. of amniotic fluid and instillation of 180 ml. of saline. Ten units of oxytocin were administered hourly in 100 ml. of a 5 per rent dextrose in water solution. The fetus was aborted 31 hours after the saline instillation but the placenta was retained and could not be removed digitally. Oxytocin administration was therefore continued at the same rate for 7 additional hours, after which time it was decided to perform a completion. A total of 380 U. of oxytocin had been administered in 38 hours in 3,800 ml. of a 5 per cent dextrose in water solution. Ten minutes following sodium pentobarbital and atropine medication in preparation for the curettage, the patient had a grand ma1 seizure of 1 to 2 minutes’ duration which responded to sodium amytol administration. Serum electrolyte determination revealed sodium, 122 mEq. per liter, and potassium, 3.1 mEq. per liter; the serum chloride level was not analyzed. She was thereafter maintained on Dilantin, 100 mg. three times a day. The oxytocin was imtnediately discontinued and normal saline and potassium were administered intravenously in small amounts. She responded with good diuresis; the amount was not quantitated and no diuretics were necessary. The electrolytes normalized within the following 10 hours. The serum sodium increased to 134 mEq. per liter and potassium to 3.8 mEq. per liter. The postoperative course was uncomplicated, there was no reoccurrence of the
per liter and potassium was 3.4 mEq. per liter. The patient had then received a total of 290 U. of oxytocin and 2,900 ml. of fluid in the course of 28 hours and 40 minutes. The oxytocin was immediately discontinued and the seizure did not recur. Case 4. The patient was a 29-year-old, single, white woman, gravida 1, para 0, who entered the hospital in her fourteenth week of gestation for a saline abortion. Amniocentesis was performed with difficulty; 20 ml. of clear amniotic fluid were removed and 200 ml. of a 20 per cent saline solution were instilled. An oxytocin infusion of 10 U. in 100 ml. of a 5 per cent dextrose in water solution per hour was started 2 hours after the saline instillation. Due to a combination of circumstances, the intravenous oxytocin administration was continued at the same dose level until the time of abortion on the third day. Forty-eight hours after the saline instillation the urine output had decreased and the diuresis was only 350 ml. in the following 24 hours. About 64 hours after the initiation of the oxytocin administration the patient had developed good uterine contractions and at the same time she had become very tense and nervous. She was therefore given 25 mg. of meperidine and 25 mg. of promethazine intramuscularly. She aborted 45 minutes after this medication and the placenta was passed a few minutes later. Immediately after the passage of the placenta, the patient developed a grand ma1 seizure. It is uncertain how long the seizure lasted but possibly 15 minutes passed before she was attended by a physician and the seizure could be controlled .by administration of 10 mg. of diazepam intravenously. The patient had been given a total of 800 U. of oxytocin in 8,000 ml. of 5 percent dextrose in water solution. The serum sodidm level shortly after the seizure was 111 mEq. per liter and the chloride was 76 mEq. per liter. on fluid restriction and continued three times a day, and diazepam,
The patient was placed on Dilantin, 100 mg. 5 mg. three times a day,
4
Lauersen
and
Birnbaum
and there was no recurrent seizure activity. She responded with an enormous diuresis of 5.280 ml. in the following 24 hours without any administration of diuretics. The patient was stuporous for 24 hours following the incident; then she showed some response to deep pain. The electrolytes normalized within 24 hours. A surgical completion was necessary on the second postabortion day. She developed endometritis, which was treated with high doses of antibiotics before any improvement occurred. Repeated spinal taps were negative for red and white blood cells. Electroencephalogram showed bilateral cerebral dysfunction with some left-sided focal features. The patient was transferred to the neurological service and had to remain in the hospital for five weeks. She remained delirious for three weeks after the seizure, during which time she was aphasic, producing only highpitched screams. By the fourth week she was able to produce a few simple words and thereafter received speech therapy. At the time of discharge she was able to repeat a few words and phrases but was unable to read or write. She was discharged with the following diagnoses: diffuse cerebral damage with left posttemporal dysfunction secondary to hypoxia; grand ma1 seizure secondary to water intoxication with hyponatremia. Comment As mentioned, the syndrome of water intoxication was first described in 1923 by Rowntree’l ’ who observed the phenomenon in a number of experiments in man and animals. He was able to produce the syndrome in patients with diabetes insipidus by giving them repeated doses of pituitary extract and in animals by administration of large amounts of distilled water. The syndrome of water intoxication in all experiments was characterized by asthenia, followed by muscular irritability, convulsions, and finally death. These symptoms could be relieved and the animals saved from death by administration of a salt solution. Rowntree’! 2 was also able to prevent water intoxication in animals by giving a 10 per cent solution of sodium chloride immediately prior to the water administration. He concluded, therefore, that the symptoms were due to a salt-water imbalance in the central nervous system. Pathologic examination of the animals that had suffered from water intoxication consistently revealed cerebral edema, which Rowntree took as the cause of convulsive seizures and subsequent death. It is now believed that it is the hyponatremia that causes the water intoxication symptoms, Plum and Posner17 have theorized that the pathogenesis is related to altered excitability of the neural membranes. The membrane potential is influenced by the ratio of
January 1, 197:~ Am. J. Obstet. Gynecol
sodium outside to sodium inside the cell and, more important, by the ratio of potassium outside to POtassium inside the cell. During hyponatremic episodes there may be shifts of sodium and potassium in neurons, accentuating the excitability and producing the neurological symptoms. Moss18 and Brockbank’” have reported extreme muscular cramps in workmen who perspire freely and consume large quantities of water. These symptoms could be relieved by administration of salt water. The need for extra intake of salt in stokers and others who work under very hot conditions is well known. Helwig and associates20 reported a death due to water intoxication in a woman who absorbed 9,000 ml. of water in 26 hours. Van Dyke and co-worker? have made extensive studies of the oxytocic and antidiuretic properties of the two neurohypophysial hormones oxytocin and vasopressin (or antidiuretic hormone) . In pharmacological assays, the antidiuretic activity of oxytocin is only about 0.5 per cent of the activity of vasopressin; likewise, the oxytocic activity of vasopressin is only a few per cent of that of oxytocin. Not only in animals, but also in the human subject, oxytocin is found to have antidiuretic properties and vasopressin oxytocic properties, but the ratios based on animal assays do not apply to the human subject at all. Thus vasopressin has, in equimolar amounts, a stronger oxytocic effect on the human nonpregnant uterus than does oxytocin.“, 23 The response of the target organ depends on other hormones, in case of the uterus upon ovarian or placental steroids.24 The effect of oxytocin on the urinary excretion of water and electrolytes in the human subject was first studied by ThomsorP in Great Britain and Abdul-Karim and Assali3 in the United States; both studies clearly demonstrated an antidiuretic effect of oxytocin. To exclude the possibility that oxytocin antidiuresis could be mediated through stimulation of antidiuretic hormone secretion, Abdul-Karim and Assali administered oxytocin to two patients with diabetes insipidus and three normal subjects whose endogenous antidiuretic hormone release had been blocked by alcohol. The results demonstrated that oxytocin has a direct action on the kidney. In pregnant patients who were given oxytocin for induction of labor, Abdul-Karim and Assali found the antidiuretic action to become evident before the oxytocic effect. Abdul-Karim and Assali demonstrated that the
Volume Number
121 1
antidiuretic effect during an infusion of oxytocin began at an infusion rate of about 15 mu. per minute and is maximal at 45 mu. per minute. The amount of oxytocin did not seem to be a controlling factor in the production of water intoxication when it exceeded 4.5 mu. per minute. This explains why several of the patients reported to have water intoxication have developed symptoms on relatively low doses of oxytocin. to According to Pickford, 26 the renal response vasopressin is modified by a variety of hormones, adrenomedullary hormones, corticosteroids, parathyroid hormones, and ovarian steroids. Much less is known about the influence of other hormones on the renal response to oxytocin, but Pickford did find an effect of estrogens in dogs, and Abdul-Karim and Assali3 found a difference in the human renal response between term pregnant and nonpregnant subjects. This suggests that the placental steroids influence the renal sensitivity to oxytocin but whether the kidney in the midtrimester is particularly sensitive to oxytocin is not known. The doses of oxytocin given in cases of saline-induced abortion are usually of such a magnitude that even with a relatively low renal sensitivity, an antidiuretic effect is to be expected. In all four cases of water intoxication described here there was a decrease in the sodium levels in the serum. This hyponatremia is believed by some investigators17 to be the primary cause of the syndrome. In the majority of the reported case&l6 water intoxication occurred when large amounts of oxytocin were given in (1) saline-induced abortion, (2) missed abortion, and (3) postpartum hemorrhage. In all cases, oxytocin was administered intravenously in electrolyte-free solutions; this undoubtedly was an additional factor, since Rowntree could prevent the symptoms in animals by treatment with salt solutions. In three of our four cases (nos. 1, 2, and 4)) the patients were given large amounts of oxytocin in 5 per cent dextrose in water without any electrolyte administration. All three patients responded with good diuresis and relief of symptoms after discontinuation of the oxytocin infusion and administration of electrolyte solutions. In ‘case 3, water intoxication developed in spite of the fact that oxytocin was administered in Ringer’s lactate solution, but this patient was not given an intraamniotic instillation of hypertonic saline. Several investigators 28, 29 have shown that there is a slight increase in sodium chloride concentra-
Water
intoxication
with
oxytocin
5
tion in the serum in the first hours after an intraamniotic instillation of a 20 per cent saline solution. This might have some beneficial effect in preventing water intoxication during subsequent oxytocin administration, but the most important precaution appears to be to administer the oxytocin in an electrolyte solution. The hypertonic saline blocks vasopressin (ADH) release ; this tends to facilitate excretion of the salt but may lead to dehydration. Oxytotin will slow down the excretion through the kidneys and the carrier solution will tend to hydrate the patient. Up to a point, the two processes will tend to balance each other but, if oxytocin is given too long without electrolytes, water intoxication will occur. We have now treated more than 800 cases of oxytocin administration during midtrimester saline abortions where the oxytocin was administered at the rate of 167 mu. per minute in alternating solutions of Ringer’s lactate and 5 per cent dextrose and water without any cases of water intoxication; oxytocin has not been administered for more than 36 hours. All the cases described occurred when large amounts of oxytocin in nonelectrolyte solutions were used over a prolonged period. As shown by Lauersen and Schulman,‘5 the abortion time can be shortened to about 20 hours and the complication rate greatly reduced when 10 U. of oxytocin are given hourly ( 167 mu. per minute), starting immediately after saline instillation. The oxytocin should be given in Ringer’s lactate or, alternately, in Ringer’s lactate and 5 per cent dextrose and water solutions. These authors stressed the importance of careful monitoring of the urinary output and of stopping the oxytocin administration to check the serum electrolytes if the urinary output decreased. In addition, they recommended discontinuation of the oxytocin administration for 10 to 12 hours if the patient had not aborted within 36 hours, in order to perform electrolyte determination and correct any electrolyte imbalance. These recommendations are fully supported by the experience gained from our four cases of water intoxication. In three of our cases the patients were attended by a physician immediately after the convulsive seizure occurred and the hypoxic state was apparently so short that the patient could be discharged without cerebral damage. In case 4, the seizure lasted several minutes and the patient developed diffuse cerebral damage; at the time of discharge, six weeks after the incident, she was still
6
Lauersen
unable
to
important the
and
Birnbaum
speak, to
read,
have
symptoms
of
an
or alert
water
write. nursing
intoxication,
It
is
therefore
staff,
aware
of
and
instructed
to
watch
the
of
the
syndrome
or
hradache.
diuresis
and as
asthrnia,
to
report
such
muscular
early
si,qns
irritability.
REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16.
Rowntree, I,. G.: Arch. Intern. Med. 32: 157, 1923. Rowntree, L. G.: J. Pharmacol. Exp. Ther. 29: 135, 1926. Abdul-Karim, R., and Assali, N. S.: J. Lab. Clin. Med. 57: 522, 1961. Liggins, G. C.: J. Obstet. Gynaecol. Br. Commonw. 69: 277, 1972. Pittman, J. G.: N. Engl. J. Med. 268: 481, 1963. Potter, R. R.: Obstet. Gynecol. 23: 699, 1964. Josey, W. E., Pinto, A. P., and Plant, R. A.: AM. J. OBSTET. GYNECOL. 104: 926, 1969. Silva, P., and Allan, M. S.: Obstet. Gynecol. 27: 517, 1966. Self, J.: Am. J. Med. Sci. 252: 573, 1966. Leventhal, J. M., and Reid, D. E.: AM. J. OBSTET. GYNECOL. 102: 310, 1968. Burt, R. L., Oliver, K. L., and Whitener, D. I,.: Obstet. Gynecol. 34: 212, 1969. Bilek, W., and Dorr, P.: Can. Med. Assoc. J. 103: 379, 1970. Lilien, A. A.: Obstet. Gynecol. 32: 171, 1968. Gupta, D. R., and Cohen, N. H.: J. A. M. A. 220: 681, 1972. Lauersen, N. H., and Schulman, J. D.: AM. J. OBSTET. GYNECOL. 115: 420, 1973. Mann, L. I., Duchin, S., Newman, M., and Weiss, R. R.: Obstet. Gynecol. 41: 748, 1973.
Attention The
Editors
to
Authors: Publisher
OF OBSTETRICS
have
in
AND
agreed
Reference to add
18. 19. 20. 21. 22. 23. 24. 25. 26.
27. 28. 29.
Plum, F., and Posner, J. B.: In Davis, F. A., editor: Diagnosis of Stupor and Coma, ed. 2, Philadelphia, 1972, F. A. Davis Company. Moss, K. N.: Proc. R. Sot. Land. Ser. B 95: 181, 1923-1924. Brockbank, E. M.: Br. Med. J. 1: 65, 1929. Helwig, F. C., Schutz, C. B., and Curry, D. E.: J. A. M. A. 104: 1569, 193.5. Van Dyke, H. B., Adamsons, K. Jr., and Engcl, S. L.: Recent Progr. Hormone Res. 11: 1, 1955. Dahle, T.: Acta Obstet. Gynecol. Stand. 30: (Suppl. 1): 1950. Coutinho, E. M., and Lopes, A. C. V.: Am. J. Obstet. Gynecol. 102: 479, 1968. Berde, B.: In Handbook of Experimental Pharmacology, Vol. XXIII, Heidelberg, 1968, Springer Verlag. Thomson, W. B.: J. Physiol. 150: 284, 1960. Pickford, M.: In Harris, G. W., and Donovan, B. D., editors: The Pituitary Gland, London, 1966, Butterworth & Company. Abdul-Karim, R. W., and Rizk, P. T.: Obstet. Gynecol. Surv. 25: 80.5, 1970. Brewer, I,. I,., Johnson, W. I,., and Hunter, C. A.: Asr. J. OBSTET. GYNECOI,. 101: 1078, 1968. Berkowitz, R. L. T.: Clin. Obstet. Gynecol. 14: 107, 1971.
Style
the article title to references in the AMERICAN References will now conform to the style of the Cumulated Index Medicus, viz., name of author, title of article, name of periodical, volume, page, and year. Authors are always encouraged to limit references to sixteen for the following JOURNAL sections: Obstetrics, Gynecology, and Fetus, Placenta, and Newborn.
JOURNAL
and
Change
17.
GYNECOLOGY.
