Effect of food withdrawal on arterial blood glucose and plasma 13, 14-dihydro-15-keto-prostaglandin F2« concentrations and nocturnal myometrial electromyographic activity in the pregnant rhesus monkey in the last third of gestation: A model for preterm labor?

Z. Binienda, A. Massmann, M. D. Mitchell, R. D. Gleed,J. P. Figueroa, and P. W. Nathanielsz

Ithaca, New York, and San Diego, California

Pregnant rhesus monkeys were studied between 109 and 149 days of gestation. Food withdrawal for 48 hours (with free access to water) was accompanied by a decrease in maternal whole blood glucose concentration and an increased maternal arterial plasma 13,14-dihydro-15-keto-prostaglandin F2n

concentration. On s\Jccessive nights of the 48-hour period of food withdrawal, there was an increase in the frequency of myometrial contractions as recorded by uterine electromyogram. In the period after food was returned, blood glucose, arterial 13,14-dihydro-15-keto-prostaglandin F2n concentration, and contraction frequency returned to baseline. Because food withdrawal results in the appearance of the nocturnal contraction pattern seen at term, we suggest that this experimental paradigm may be used as a model for preterm labor. (AM J OBSTET GVNECOL 1989;160:746-50.)

Key words: Preterm labor, myometrial activity, maternal nutrition

In the pregnant sheep, withdrawal of food after 120 days of gestational age for a period of 48 hours results in a fall in maternal and fetal plasma glucose concen­tration, a rise in fetal plasma cortisol concentration, and increased maternal and fetal plasma prostaglandin and maternal plasma estrone sulfate concentrations. I

.3 In

addition, myometrial activity as recorded by uterine electromyogram' and intrauterine pressure increases. 1

Similar observations were originally made in the preg­nant mare. I·" If the food withdrawal occurs after 137

days of gestational age, preterm labor may be precip­itated in the sheep. 1

Myometnal e1ectromyographlC patterns in the preg­nant sheep and pregnant monkey have certain simi­larities and differences.7.11 In both species two distinct types of myometrial events have been described. Throughout the major part of pregnancy the com­monest form of activity is epochs of myometrial elec-

From the Laboratory for Pregnancy and Newborn Research, New York State College of Veterinary MedIcine, Cornell University, Ithaca, and Department of Reproductive MedIcine, University of California San Diego MedIcal Center, San Dzego.

Supported by National Institute of Health Grants No. HD-18870 and No. HD-20779 (M. D. M.).

ReceIVed for publication February 23,1988; revised July 20,1988; accepted October 9, 1988.

Reprint requests: Dr. Peter W. Nathanzelsz, MD, PhD, Laboratory for Pregnancy and Newborn Research, NYS College of Veterinary Medicine, Cornell University, Ithaca, NY 14853.

746

tromyographic activity lasting longer than 3 minutes: contractu res. In the pregnant monkey we have noted that the shorter electromyographic events characteristic of labor and delivery (contractions) occur at three dis­tinct times: (1) in the period immediately after most surgeries involving laparotomy and hysterotomy, (2)

occasionally spontaneously at nighttime, and (3) in the period immediately before delivery.7 In our studies of uterine electromyographic activity concomitant with in­trauterine pressure measurement, intrauterine pres­sure changes accompany the electromyographic bursts, although the relationship of their amplitudes may change.7 Electromyographic recordings may be seen as indicating the drive to the myometrium better than intrauterine pressure changes, which are affected at least in part by the viscoelastic properties of the uterus and contact between the fetus and myometrium that may lessen the effect of myometrial activity on intra­uterine pressure. Finally, the incision required for placement of catheters to increase intrauterine pressure may cause irritation of the uterus and disruption of other activity patterns. In the pregnant sheep there is also an increase in the shorter electro myographic events in the days immediately after surgery and before delivery.9 In this species there is also evidence of a noc­turnal increase in these shorter events throughout the latter part of pregnancy.12

We examined the effect of 48 hours of food with-

Volume 160 Number 3

Table I. Outcome of four pregnant rhesus monkeys

47

Gestational age at sur- 99 gery (days)

Gestational age intervals at which studies were undertaken (days)

109-120 + 130-137 + 145-149 +

Gestational age at de- 165 livery (days)

drawal on uterine electro myographic activIty in the pregnant rhesus monkey. We focused on the incidence of myometrial contractions during the hours of dark­ness, because we have previously shown7 that contrac­tion activity when it occurs in late pregnancy generally ocurs at nighttime. We have related the changes that occur when food is withdrawn to the occurrence of maternal hypoglycemia and changes in plasma cortisol and 13,14-dihydro-15-keto-prostaglandin F2• (PGFM) levels.

Material and methods

Surgical procedures. Four pregnant rhesus monkeys 5 to 7 years old with known gestational age were ob­tained from the California Regional Primate Center, Davis, Calif., and acclimated to the laboratory condi­tions as previously described.9 The animals were housed in rooms with controlled light cycles (12 hours light, 12 hours dark) with lights on at 8 AM. Food pro­vided was Purina 5045 Monkey Chow and fresh fruits. Water was provided ad libitum during the experimental period. Animals were instrumented by the use of tech­niques described previously.5 Briefly, with the use of halothane anesthesia, catheters were placed in the ma­ternal femoral artery and vein. Two pairs of myome­trial electromyographic leads were placed on the uterus. Gestational age at surgery and outcome of the four operations are given in Table I. Heparinized ar­terial blood samples were drawn at lOAM and at 3 and 9 PM and plasma was processed as described previously, with the return of red cells to the monkey. I \ Each plasma sample was measured for PGFM and glucose concentration. Data from the three plasma samples taken on each day ot the study were averaged to give a single data point for that day. Experiments began at least 5 days after surgery. On days that animals were fed, fresh food was provided at 8 AM and was left for the monkey to consume ad libitum. On the day of food withdrawal, food was removed at 3 PM and the next 24-hour period was called the first day of food withdrawal. Food was offered ad libitum after exactly 48 hours.

Effect of food withdrawal in last third of gestation 747

Animal no.

48 49 53

101 105 107

+ + + + +

159 159 167

Measurement of PGFM. PGFM was measured by a previously described and validated specific radioim­munoassay."

Measurement of maternal blood glucose. Glucose was measured in whole blood on a Yellow Springs In­strument Co. glucose oxidase-based glucose analyzer (model YSI 27; Yellow Springs, Ohio).

Recording of electromyogram, data storage, and calculation of myometrial contractions. The uterine electromyogram was recorded from multistranded bi­polar stainless steel wires (Cooner AS 632, Cooner Sales Co., Inc., Chatsworth, Calif.) implanted 3 to 5 mm apart in the myometrium. These were placed on at least two sites of the myometrium as previously described in the pregnant monkey and sheep." 7 " Visual observation of electromyographic records showed that the most char­acteristic and remarkable change in pattern that oc­curred during food withdrawal was the appearance of contractions at nighttime. We therefore analyzed the electromyographic patterns for the numbers of con­tractions occurring throughout the entire night (8 PM

to 8 AM). Recording and quantification of myometrial electro myographic activity were performed with tech­niques similar to those undertaken in sheep7 by use of a computer-based acquisition system. Myometrial con­tractions were defined as electro myographic epochs less than I minute in duration terminated by at least I min­ute of quiescence.

Statistical analysis. Because our hypothesis was that food withdrawal would produce changes in myometrial electro myographic activity and in maternal plasma glu­cose and PGFM concentrations, e1ectromyographic ac­tivity (expressed as total number of contractions at night) and mean maternal plasma glucose and PGFM concentrations on the day before food withdrawal and on each day of food withdrawal were calculated. A pre­liminary analysis of variance was performed to show significant differences between means. The value on the day before food withdrawal was compared with each day of food withdrawal by the Student paired t test. Because the Bonferroni correction for two com-

748 Binienda et al.

~ ..... 60 I :a Qo 50 ! ~ 40 '" 0 0 ::s

30 t;, "d 0

..9 20

.0

01 I:l 10 ... ~ ... .,

::II 0 C1 C2 "1 FW2 R1 R2

EXPERIMENTAL TIME (Daya)

Fig. 1. Mean (+ SEM) maternal whole blood arterial glucose levels in pregnant rhesus monkeys in eight experiments in which food, but not water, was removed for 48 hours. C, Con­trol day; FW, food withdrawal day; R, recovery day. *P < 0.025 compared with the second control day. Food was removed at 3 PM and maternal whole blood glucose concentration in the 9 PM, 10 AM, and 3 PM samples were averaged each day.

parisons was used, the significance level was set at p < 0.025. Eight experiments were conducted on four animals over a range of 30 days of gestation. Each food withdrawal experiment was treated independently for the purpose of analysis (n = 8) because of differences in individual resting plasma concentrations and peak responses.

Results

Outcome of the four animal preparations. The outcome of these four rhesus monkeys is shown in Ta­ble I.

Maternal blood glucose concentrations. Fig. 1 shows the mean maternal blood glucose concentration on the 2 days before food withdrawal, during food with­drawal, and for the two days after the return to normal feeding. The maternal plasma glucose level fell in the first 24 hours but was not significantly depressed (p < 0.028) over the entire day until the second day of food withdrawal (p < 0.004).

Maternal PGFM concentrations. The mean maternal arterial PGFM concentration rose during the first day of food withdrawal but was significantly elevated only during the second day of food withdrawal (p < 0.007) (Fig. 2).

Myometrial electromyographic activity. The inci­dence of short-term epochs of myometrial activity in the 12 hours of daytime was between 2 and 11 for the entire 6-day study. The remarkable increase in electro­myographic activity was clearly visible at nighttime. Be­cause the animals were not disturbed at night by such activities as cage cleaning, we analyzed only the night­time recordings. In each animal there was an increase

0 .., " .>: I ~ I 0 ... '0,", :., .... .0 I :a 8 I .jo t.o .... Il. M~ ..... IS ., CII e r... '" I:l ., ....

- '0 Il. <: 'iii

., E ~ ... ~ '" ..,

0

" ... " Il.

120 110 100 90 80 70 80 50 40 30 20 10 0

C1 C2 "1 FW2

March 1989 Am J Obstet Gynecol

R1 R2

EXPERIMENTAL TIME (Daya)

Fig. 2. Mean (+ SEM) maternal arterial plasma PGFM con­centration in pregnant rhesus monkeys in eight experiments in which food, but not water, was removed for 48 hours. C, Control day; FW, food withdrawal day; R, recovery day. *P < 0.025 compared with the second control day. Food was removed at 3 PM and maternal arterial PGFM concentration in the samples were averaged each day.

in contraction activity on the first evening after food withdrawal. This increase always occurred by midnight, within 9 hours of food removal. The frequency of con­tractions was also significantly elevated on the second evening of food withdrawal (Fig. 3). The data have been displayed simply as the total number of contrac­tions occurring during the entire night period, al­though on each occasion the contractions occurred for only I or 2 hours early in the period of darkness. However, because the exact hour at which contractions increased during the nighttime differed slightly for each experiment, contraction frequency has been ex­pressed based on contractions over the entire dark period.

Comment

Several investigators have demonstrated a rhythm of myometrial activity in the pregnant rhesus monkey. The short-lived contraction type epochs of electromyo­graphic activity tend to occur at nighttime.7 A similar increase in nocturnal myometrial activity has been shown by other workers. 15

.17 This is true both for the

contraction activity that occurs immediately before la­bor and delivery, as well as for the spontaneous episodes of contraction activity that we have seen and occasional examples after maternal stress, such as hemorrhage." In the present study we observed an increased inci­dence of myometrial contractions during the nights of food withdrawal that was significant within 17 hours and occurred each evening during the food withdrawal period. Because the increased myometrial activity we observed is of the short-activity-burst, contraction type, we suggest that food withdrawal may be a useful ex-

Volume 160 Number 3

'0 0 80 cD 0 I 70 0

0 a 60 ~ III 50 ~ 0

:;:l 40 " til ... .... 30 c

0

" '0 20 ...

10 " .0 8 0 :;j z

C1 C2 FW1 FW2 R1 R2 R3 R4

EXPERIMENTAL TIME (Days)

Fig. 3. Mean ( + SEM) number of contractions occurring dur­ing the entire nighttime period in pregnant rhesus monkeys in eight experiments in which food, but not water, was re­moved for 48 hours. C, Control day; FW, food withdrawal day; R, recovery day. *P < 0.025 compared with the second control day. Food was removed at 3 PM and an uterine elec­tromyogram was analyzed in 24-hour blocks around this time point.

perimental paradigm of preterm labor. In these ani­mals there was no increased incidence of preterm de­livery; however, it should be noted that all preterm labor does not lead to preterm delivery.

Hypoglycemia has several metabolic effects. Mobili­zation of free fatty acids, including arachidonic acid, may result in generation of the prostaglandins that stimulate uterine contractility in those fetal and mater­nal tissues that are responsible for the regulation of myometrial contractions. It is of interest that although the hypoglycemia was maintained over the entire 48-hour period of food withdrawal, the increased myo­metrial activity occurred only at night. It may be that the sensitivity of the myometrial cell to some nocturnal input was increased by the food withdrawal. Catechol­amines released either into the vasculature or at sym­pathetic nerve terminals may play a role in the in­creased myometrial activity that occurs during the"hy­poglycemia induced by food withdrawal.

Further studies are required to elucidate the mech­anisms that underlie the increased myometrial activity we observed after food withdrawal. Increased uterine activity almost certainly involves prostaglandin gener­ation, because exogenous prostaglandins will increase myometrial activity in the pregnant monkei8 and depression of endogenous prostaglandin production will decrease myometrial activity in pregnant sheep'9

and block the normal onset of parturition in primates.2o

Food withdrawal in the pregnant sheep and mare leads to increased circulating maternal plasma PGFM con­centrations and is associated with increased uterine ac­tivity. 1. 2. 4 Our finding of increased production of PGFM

Effect of food withdrawal in last third of gestation 749

concomitant with maternal hypoglycemia in the preg­nant rhesus monkey is consistent with these observa­tions in other species.

Our observation may also have importance for the preparation of the pregnant nonhuman primate for experimental surgery. We suggest that future experi­mental surgeries for instrumentation of the pregnant rhesus monkey be conducted under euglycemic con­ditions, with careful monitoring of maternal blood glu­cose concentration and administration of glucose to the mother if necessary during the period of preoperative food withdrawal and in the early postsurgical days, when food intake may be decreased. This precaution may lead to a higher success rate of instrumented prep­arations. Preterm labor in the human parturient has a higher incidence after fasting during Yom Kippur2' ;

thus our observation may have relevance to human pregnancy. The psychic stress of food withdrawal may play an important role in the observed increased myo­metrial activity. Possible mechanisms would be central arginine vasopressin and corticotropin-releasing factor release with subsequent activation of the hypothalamo­hypophyseal-adrenal axis. Further studies are required to define any relationship to more chronic undernu­trition in women.

REFERENCES 1. Fowden AL, Silver M. The effects of food withdrawal on

uterine contractile activity and on plasma cortisol concen­trations in ewes and their fetuses during late gestation. In: Jones CT, and Nathanielsz PW, eds. The physiological development of the fetus and newborn. London: Aca­demic Press, 1985: 157-61.

2. Fowden AL, Silver M. The effect of the nutritional state on uterine prostaglandin F metabolite concentrations in the pregnant ewe during late gestation. Q J Exp Physiol 1983;68:337-9.

3. Milvae R, Mitchell MD, Nathanielsz PW, Pimentel G, Rosen ED. The effect of food withdrawal on myometrial electro myographic (EMG) activity and maternal plasma concentrations of 13,14-dihydro-15-keto prostaglandin F20 (PGFM) and oestrone sulphate in the pregnant ewe at 122-127 days gestation (d G.A.). J Physiol 1986;381:4IP.

4. Barnes RJ, Comline RS, Jeffcott LB, Mitchell MD, Ross­dale PD, Silver M. Foetal and maternal plasma concen­trations of 13,14-dihydro-15-oxo-prostaglandin F in the mare during late pregnancy and at parturition. J Endo­crinol 1978;78:201-15.

5. Silver M, Barnes RJ, Comline RS, Fowden AL, Clover L, Mitchell MD. Prostaglandins in maternal and fetal plasma and in allantoic fluid during the second half of gestation in the mare. J Reprod FertiI1979;27:531-9.

6. Silver M, Fowden AL. Uterine prostaglandin F metabolite production in relation to glucose availability in late preg­nancy and a possible influence of diet on time of delivery in the mare. J Reprod Fertil 1982;32(suppl):511-19.

7. Taylor NF, Martin MC, Nathanielsz PW, Seron-Ferre M. The fetus determines the circadian oscillation of myo­metrial electro myographic activity in the pregnant rhesus monkey. AMJ OSSTET GYNECOL 1983;146:557-67.

8. Germain G, Cabrol D, Visser A, Sureau C. Electrical ac­tivity of the pregnant uterus in the cynomolgus monkey. AM J OSSTET GYNECOL 1982; 142:513-19.

750 Binienda et al.

9. Figueroa ]P, Mahan S, Poore ER, Nathanielsz PW. Char­acteristics and analysis of uterine electro myographic ac­tivity in the pregnant sheep. AM ] OBSTET GVNECOL 1985; 151 :524-31.

10. Toutain PL, Garcia-Villar R, Hanzen C, Ruckebusch Y. Electrical and mechanical activity of the cervix in the ewe during pregnancy and parturition.] Reprod Fertil 1983; 68: 195-204.

11. Nathanielsz PW, Frank D, Gleed R, Dillingham L, Poore ER, Figueroa ]P. Methods of investigation of the chron­ically instrumented pregnant rhesus monkey preparation maintained on a tether and swivel system. In: Nathanielsz PW, ed. Animal models in fetal medicine. vol III. Ithaca, NY: Perinatology Press, 1984:110-60.

12. Figueroa ]P, McDonald T], Nathanielsz PW, Poore ER, Wentworth RA. Circadian variation in myometrial activity in the chronically instrumented pregnant sheep at 120-130 days gestation.] Physiol 1985;369:116P.

13. Nathanielsz PW, Lowe KC, Beck NFG, et at. Circulating plasma protein concentrations in the fetal and neonatal sheep. Bioi Neonate 1980;38:126-33.

14. Strickland DM, Brennecke SP, Mitchell MD. Measure­ment of 13,14-dihydro-15-keto-prostaglandin F2• and 6-keto-prostaglandin Fl. by radioimmunoassay without prior extraction and chromatography. Prostaglandins Leukotrienes Med 1982;9:491-3.

March 1989 Am.J Obstet Gynecol

15. Novy MJ. Endocrine and pharmacological factors which influence the onset of labour in rhesus monkeys. In: Knight], O'Connor M, eds. The fetus and birth. CIBA Foundation symposium 47 (new series). Amsterdam: El­sevier/North Holland, 1977:259-88.

16. Ducsay CA, Cook M], Walsh SW, Novy MJ. Circadian patterns and dexamethasone-induced changes in uterine activity in pregnant rhesus monkeys. AM ] OBSTET Gv­NECOL 1983;145:389-96.

17. Harbert GM. Biorhythms of the pregnant uterus (Macaca mulatta). AM] OBSTET GVNECOL 1977;129:401.

18. Novy M], Thomas CL, Lees MH. Uterine contractility and regional blood flow responses to oxytocin and prosta­glandin E2 in pregnant rhesus monkeys. AM] OBSTET Gv­NECOL 1975;122:419-33.

19. El Badry A, Figueroa ]P, Poore ER, et al. Effect of fetal intravascular 4-aminoantipyrine infusion on myometrial activity (contractures) at 125 to 143 days' gestation in the pregnant sheep. AM] OBSTET GVNECOL 1984; 150:474-9.

20. Novy M], Cook M], Manaugh L. Indomethacin block of normal onset of parturition in primates. AM ] OBSTET GVNECOL 1974;116:412-6.

21. Kaplan M, Eidelman AI, Aboulafia Y. Fasting and the precipitation of labor. The Yom Kippur effect. JAMA 1983;250: 1317-8.