Fetal-maternal during chronic MARY

fluid and electrolyte relations fetal urine loss in sheep

E. WLODEK,

RICHARD

HARDING,

Fetal and Neonatal Unit, Department Melbourne, Victoria 3168, Australia

Wlodek, Mary E., Richard Harding, and Geoffrey D. Thorburn. Fetal-maternal fluid and electrolyte relations during chronic fetal urine loss in sheep. Am. J. Physiol. 263 (Renal Fluid Electrolyte Physiol. 32): F671-F679, 1992.-Our aim was to determine the effects of prolonged removal of fetal urine during late gestation on fetal-maternal fluid and electrolyte relationships. We measured the volume and composition of fetal urine and amniotic and allantoic fluids and the composition of fetal and maternal plasma in sheep before and during continuous urine drainage, which began at 130 days of gestation and continued until the onset of labor; a control group was also studied. The response to fetal urine drainage occurred in two phases. In the “acute” phase (l-3 days), amniotic and allantoic fluid volumes decreased significantly, presumably due to their reabsorption into the fetal chorionic circulation or swallowing of amniotic fluid by the fetus. During the “chronic” phase, starting 3-5 days after urine drainage, a significant reversal in the transplacental osmotic gradient occurred due to a decrease in maternal plasma osmolality. During the entire drainage period (14.1 t 1.1 days, mean t SE, n = 5) at least 542 ml/day of water and 24 mmol/day of electrolytes passed from the mother into the fetal circulation and fetal plasma osmolality was unchanged. We conclude that, despite the loss of substantial water and electrolytes, the fetus is able to maintain its growth and fluid and electrolyte homeostasis by obtaining water and electrolytes initially from the amniotic and allantoic fluids and subsequently from its mother. The movement of water and electrolytes to the fetus would have been facilitated by the reversed transplacental osmotic gradient. fetus; placenta; amniotic fluid; allantoic fluid; osmolality; placental transfer PREGNANCY, maternal and fetal fluid compartments are in a state of balance. The placenta is permeable to water and electrolytes, and there is normally a small net movement of these to the fetus and placenta for their growth (3). The fetus develops within a pool of amniotic fluid, which is subject to constant turnover, being added to by fetal lung liquid and urine and being removed by fetal swallowing. Furthermore, fluxes of water and electrolytes may take place across the surrounding membranes into the fetal (chorionic) circulation. In those species, such as sheep, in which an allantoic sac persists, allantoic fluid is formed from fetal urine (40). The fate of this fluid is unknown, but it is probably reabsorbed into the fetal circulation. The role of amniotic fluid is commonly thought to be largely protective, allowing fetal movements yet preventing mechanical damage. However, this fluid, together with allantoic fluid, may also serve as a reservoir of water and electrolytes readily available to the fetus should maternal or fetal fluid status become compromised. It is well established that fetal urine is a major component of amniotic and allantoic fluids (9,27,40). Little is known, however, about the role of amniotic and allantoic fluids in the maintenance of fetal fluid homeoDURING

0363-6127/92

$2.00

AND GEOFFREY

D. THORBURN

of Physiology, Monash University,

Copyright

stasis under normal conditions and under conditions of perturbed fluid and electrolyte balance. Removal of fetal urine experimentally can potentially provide information on the role of fetal urine and amniotic and allantoic fluids, on water and electrolyte turnover in the fetus and its sacs, and on fluid and electrolyte transfer across the placenta. Both Gresham et al. (15) and Wintour et al. (39) continuously drained urine from a single sheep fetus to the exterior for 18 days until 148 days and death at 120 days of gestation, respectively. Despite the loss of large amounts of water and electrolytes via fetal urine and a significant reduction in amniotic and allantoic fluid volumes, the composition of fetal plasma remained constant and fetal growth was normal (15). Drainage of fetal urine had no effect on the osmolalities of fetal plasma, urine, and amniotic fluid (39). In another study (30), the drainage of fetal urine for an average of 2.6 days on one to four occasions in different animals increased transplacental fluxes of water and electrolytes. Taken together, the results of these three studies indicate that, despite the loss of urinary water and electrolytes, the sheep fetus can readily obtain, via the placenta, the fluids and electrolytes necessary for maintenance of blood composition and growth. We considered that more comprehensive studies were necessary to examine in detail the relationship between fetal and maternal fluid status when large volumes of water and electrolytes are lost via removal of fetal urine. In particular, we were interested in the ability of the fetus to obtain water and electrolytes 1) from its fluid sacs and 2) from the mother via the placenta. To accomplish these objectives we determined, over a prolonged study period, the effects of continuous removal of fetal urine on the volume and composition of urine and amniotic and allantoic fluids and on the composition of fetal and maternal plasma. We have also investigated possible effects on fetal growth and fetal endocrine status. METHODS General surgical procedure. Thirteen Merino cross-bred ewes underwent surgery (U-2.0% halothane in O,-N,O) at 116-120 days of pregnancy (full term 147 days). These were divided into a control group (n = 6) and an experimental group (n = 7) from which fetal urine was drained from 130 days until term. The uterus was incised after a midline laparotomy, and the fetal hindlimbs and trunk were removed to the level of the umbilicus. A catheter was inserted into the femoral artery so that its tip lay in the abdominal aorta. A catheter was inserted into the fetal bladder to monitor fetal bladder pressure as an indicator of the presence or absence of fetal voiding (40, 41). A catheter was implanted to measure fetal intraperitoneal pressure as a reference pressure. Silicone rubber catheters (n = 3-4) were sutured to the skin of the fetal leg and ventral and dorsal abdomen and over the spine, and two catheters were sutured into the allantoic

0 1992 the American

Physiological

Society

F671

Downloaded from www.physiology.org/journal/ajprenal by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on January 16, 2019.

F672

FETAL-MATERNAL

RELATIONS

DURING

fluid sac via the pregnant and nonpregnant uterine horns (12). The ewe’s carotid artery and jugular vein were catheterized. Sterile isotonic saline (250 ml) was infused into the amniotic sac to replace fluid lost during surgery. Procaine penicillin (500 mg) and dihydrostreptomycin (500 mg) were injected intramuscularly into the fetus before replacing the fetus and closing the uterine incision. Catheters were exteriorized through the flank of the ewe and the abdomen was closed in layers. After surgery the sheep were housed in individual cages in a room with 12-h light (0700-1900 h) and dark cycles. Experimental protocol. A postsurgical recovery period of at least 5 days was allowed before any measurementswere made. Data were obtained from both groups of animals during a pretreatment period (124-130 daysof gestation) and during a treatment period (131 days of gestation to term). During the treatment period fetal urine was drained continuously from the experimental animalsbut not from control animals.Fetal urine production and amniotic and allantoic fluid volumeswere measured once during the pretreatment period and once every 5 days during the treatment period in all animals. Samplesof fetal and maternal plasma, fetal urine, and amniotic and allantoic fluids were obtained during the pretreatment and treatment periodsin both groupsand were analyzed for osmolalities,Na+, K+, and Cl- concentrations and pH. Fetal and maternal plasma sampleswere also analyzed for concentrations of glucose,total protein, albumin, urea, creatinine, cortisol, arginine vasopressin (AVP), and prolactin. Fetal and maternal blood sampleswere also analyzed for pH, PO,, Pco~, bicarbonate concentration, percent oxygen saturation, hemoglobin content, and hematocrit. Fetal urine production was measuredover 24 h from 1000 h once every 5 days in control fetusesfrom 125 days and in experimental fetusesat 125days by drainagefrom the bladder into an external sterile bag from which it was transferred every 4-6 h to the amniotic sac(60%) and allantoic sac(40%) (40). During the treatment period in the experimental animals (from 130 days of pregnancy until the onset of labor), urine was continuously drained from the bladder via a flowmeter (17) into a sterile l-liter bagbut wasnot returned to the fluid sacs.During bladder drainage there were no increments in fetal bladder pressure, indicating that micturition episodeswere absent (41). In both groups,samplesof fetal (10 ml) and maternal (10 ml) arterial blood, fetal urine (IO ml), and amniotic (10 ml) and allantoic (10 ml) fluids were collected between0900 and 1000h before the ewereceived food and water. In both groupssamples werecollected 6 and 3 days before the drainagecommenced,and on the 1st and 3rd days of eachof the subsequent5-day periods. An additional set of sampleswascollected on the 2nd day of the treatment period. All sampleswere collected into tubes that werecappedand stored on ice until centrifuged at 2,500rpm for 10 min at 4°C. The supernatant was stored at -20°C until analyzed. Maternal food and water intake and urine output were monitored daily. A measuredamount of food (1.2 kg of lucernechaff and 100 g of oats) and water (8 liters) was given to the eweeach day (1000h, before sampleswere taken), and the amount of food (to nearest5 g) and water (to nearest5 ml) left from the previousday and urine output (to nearest 5 ml) wascalculated. The ewesand fetuses were killed with an intravenous overdoseof pentobarbital sodium immediately after the ewesentered spontaneouslabor. Fetal body and organ weights were measured. Bloodgas analysis. Arterial PO,, Pco~, and pH were analyzed with a blood gas analyzer (ABL30, Radiometer), and percent oxygen saturation and hemoglobincontent were measuredusing a hemoximeter (OSM2, Radiometer). The pH of fetal urine and amniotic and allantoic fluids wasmeasuredusing the blood gas analyzer on samplesthat werecappedand stored on ice. Arterial hematocrit wasanalyzed in ouadruplicate and read on a micro-

CHRONIC

FETAL

URINE

LOSS

hematocrit reader using a x40 microscope.Fetal and maternal arterial blood bicarbonate concentrations were calculated according to the formula (3) bicarbonate = 0.0294X arterial PCO,

+ 10(-4.gg1 + o6576 x PH2)

Electrolyte concentrations and osmolalities. Samplesof fetal and maternal plasma, urine, and amniotic and allantoic fluids were cappedand stored on ice for analysisof Na+, K+, and Clconcentrations (using ion-selective electrodes, Ciba-Corning model 644) and were frozen for subsequentanalysis of total protein, albumin, urea and creatinine concentrations in fetal and maternal plasma (Kone). Measurement of osmolality was made by freezing-point depression (Advanced Instruments model 302) on samplesthat were capped and stored on ice to prevent bicarbonate loss(3). The sum of the measuredplasmaelectrolyte concentrations was calculatedto determine whether they could account for the measuredosmolalitiesof fetal and maternal plasma. The concentration of albumin was converted to millimoles per liter by dividing by its molecular weight of 60,000. The sum of the following constituents (in mM) was calculated: Na+, K+, Cl-, albumin, creatinine, urea, glucose,and bicarbonate. In the calculation of the difference between the electrolyte sumsof maternal and fetal plasmaconstituents no correction wasmadefor changesin total protein concentrations, which were unaltered in both groups. We used only data for which values for all constituents in both fetal and maternal plasmawere present, so that there were 40 measurementsfor control animals (n = 5) and 25 measurementsfor experimental animals (n = 4). Hormone and glucose concentrations. The concentrations of cortisol and immunoreactive AVP in fetal and maternal plasma and of prolactin in fetal and maternal plasmaand amniotic and allantoic fluids weremeasuredby radioimmunoassaysusingpreviously describedand validated techniques (8, 23, 35). Glucose concentrations in fetal and maternal plasmawas measuredusing the calorimetric glucoseoxidasemethod (33). Measurement of fluid volumes. Amniotic (127, 133, 138, and 143 days) and allantoic (128, 132, 137, and 142 days) fluid volumes were measuredover 6 h on different days using the indicator-dilution technique with 1251-labeled bovine serum albumin (BSA) (32). Amberlite resin wasusedto remove any free 1251from the 1251-BSAsolution before its use. We have confirmed that there was minimal free iodine in the 1251-BSAby separating 1251-BSAchromatographically on a G-25 column (Pharmacia) and comparingits elution volume with that of free iodine. During the measurementsof fluid volumes, amniotic (or allantoic) fluid catheters were connected to a sterile l-liter bag. Six amniotic (or allantoic) fluid samples(1.0 ml) were obtained before injection of 1251-BSAto calculatethe basalactivity of the 1251-BSAin thesefluids by draining amniotic (or allantoic) fluid into the bag by gravity. Three sampleswere taken from the catheters and three from the bag. At 1000 h, -lo6 cpm of 1251-BSAwere injected into the bag that contained amniotic (or allantoic) fluid which had beendrained by gravity. The fluid wasthen returned to the fluid sac.During the next 3 h amniotic (or allantoic) fluid was drained into the bag and returned to the amniotic sac at least every 30 min to ensurethorough mixing. Six fluid samples(1.0 ml) wereobtained at 1300,1400,1500, and 1600h by draining amniotic (or allantoic) fluid into the bag, and three weretaken from the catheters and three from the bag. After sampling, fluids were returned to the appropriate sac. Fluid sampleswere weighedand counted for 10 min (Packard model5400) for 1251activity. Immediately after making the injection, the amount of 1251in the fluid was calculated using standard least-squaresregressionmethodsfrom the six samples taken at eachof the four times. The volume of the compartment

Downloaded from www.physiology.org/journal/ajprenal by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on January 16, 2019.

FETAL-MATERNAL

RELATIONS

DURING

wasobtained by dividing the number of injected counts, lessthe counts per minute per milliliter removed in previous samples,by the extrapolated value and then adding the volume removed in previous samples(32). The rate of disappearanceof the 1251BSA was calculated by dividing the slopeof the regressionline by the intercept and expressedas a percentageper hour. The rate of disappearanceof 1251or “fetal swallowing”wascalculated asthe product of disappearancerate of 1251-BSAfrom the amniotic fluid sacand amniotic fluid volume (32). The product of the apparent disappearancerate of 1251-BSAfrom the allantoic fluid sac and allantoic fluid volume representsthe amount of fluid reabsorbedacross the membranes. This assumption is basedon the fact that there is no route for 1251-BSAto leavethe compartment and that fluid is thoroughly mixed during the measurement. Samplesof fetal and maternal blood and plasma,fetal urine, and fluid from the other fluid sac (i.e., allantoic sac) were obtained before injection of 1251-BSAinto the fluid sac (amniotic sac) and at the end of the measurement(1600h). There was no significant 1251-BSAdetected in fetal or maternal blood or in plasma,urine, or the other fluid sac(i.e., allantoic) at the end of the measurementperiod after correction for basallevelsof 1251BSA obtained before the start of the measurementperiod. At the end of the measurementperiod 87.7 t 3.6% (mean t SE, n = 30) of the 1251 remained bound to precipitated proteins as determined by trichloroacetic acid precipitation. Statistics and data presentation. All results are presentedas meanst SE for n animals. Data were analyzed by three-way analysis of variance (ANOVA) with repeated measureswith treatment, animal, and time as the factors. To reduce heterogeneity of variance, values of cortisol and prolactin concentrations were logarithmically transformed before their analysis. The Newman-Keuls test for multiple comparisonswasusedto test for significant differences between means, within groups detected by the ANOVA (P < 0.05). When there was no significant difference in electrolyte concentrations in fetal and maternal plasma, fetal urine, and amniotic and allantoic fluids between control and experimental animals and no significant effects of the treatment, asdetected by ANOVA, the valueswere aggregatedfor the two groups over the study period (124-142 days; seeTables I and 2). However, if a significant effect of gestational age for both groups was detected by ANOVA, a pooled mean value for both groups during the pretreatment period is presented. If a significant difference between control and experimental animalswas detected by ANOVA, meanvaluesfor control and experimental animalspooledover the entire gestational age range (124-142 days) is presented.

CHRONIC

FETAL

URINE

F673

LOSS

feet of treatment. Plasma composition. In control animals plasma osmotic pressure was consistently greater in the ewe than in the fetus by 2.7 t 0.4 mosmol/kgHzO (P < 0.05) both before and during the treatment period (124-142 days of gestation). Fetal plasma osmolality was significantly greater in the experimental animals than in control animals by 3.8 mosmol/kgH,O (P = 0.027) both before and during the treatment period. However, the difference between maternal and fetal plasma osmolality before the treatment period (124-130 days) was not significantly different between control and experimental animals. In the experimental animals, urine drainage had no significant effect on fetal plasma osmolality at any gestational age between 131 and 142 days. Maternal plasma osmolality was not significantly different between the control and experimental animals before the onset of urine drainage or for the first 5 days of urine drainage. Before and during the first 5 days of urine drainage there was no significant change in the difference in the maternal-to-fetal plasma osmolality gradient between control and experimental animals. However, at 137, 140, and 142 days in experimental animals (i.e., over the last 6 days of urine drainage), there was a significant reversal in the maternalto-fetal osmolality gradient, so that maternal plasma osmolality became significantly lower than fetal plasma osmolality by 2.7 t 1.1 (n = 6) mosmol/kgH,O (P c 0.05; Fig. 1, Table 1). Table 1 summarizes the mean values of electrolyte concentrations in fetal and maternal plasma and the maternal-to-fetal differences pooled for both control and experimental animals averaged over all gestational ages (124-142 days). Concentrations of K+, Cl-, total protein, albumin and urea in fetal plasma and Na+, Cl-, total protein, albumin, creatinine, and bicarbonate concentrations in maternal plasma were not significantly altered by urine drainage or gestational age. There was a significant and progressive increase in fetal arterial plasma creatinine concentration after 135 days in control fetuses (Fig.

cl-

0 co NTROL n

N=5

UR INE DRAIN

N=6

RESULTS

Fetal well being. Both control (n = 6) and experimental (n = 7) ewes entered spontaneous labor at term (145.5 t 0.8 and 144.7 t 1.3 days, respectively). There was no

significant difference between the weights of control and experimental fetuses (4.2 t 0.4 and 3.9 t 0.4 kg, respectively). There were one set of twins and five singleton pregnancies in the control group and one set of triplets, two sets on twins, and four singleton pregnancies in the experimental group. The mean weight of the unoperated fetuses in the control and experimental groups was 3.4 (n = 1) and 3.2 t 0.4 kg (n = 4), respectively. Fetal organ weights, expressed in grams or as a percentage of fetal body weight, were not significantly different between the two groups. Fetal and maternal arterial Pco~, PO,, pH, oxygen saturation, hemoglobin content, and hematocrit were within the normal ranges in our laboratory during the study period (124-142 days), with no significant ef-

-a

I 120

I 125

1 130

1 135

GESTATIONAL

AGE (days)

1 140

I 145

Fig. 1. Difference between maternal and fetal plasma osmolality (mosmol/kgHzO) for control and experimental animals. Solid horizontal bar represents period of fetal urine drainage from 130 days (after samples were taken) until term. There was a significant reduction in difference between plasma osmolalities in ewe and fetus 5 days after onset of fetal urine drainage. * Experimental values significantly different to control values, P = 0.002.

Downloaded from www.physiology.org/journal/ajprenal by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on January 16, 2019.

F674

FETAL-MATERNAL

RELATIONS

DURING

CHRONIC

FETAL

URINE

LOSS

Table 1. Fetal and maternal plasma electrolyte concentrations Fetal Plasma (n = 11)

Na+, mM K+, mM

135.6&1.2* 3.99kO.04

Cl-, mM Urea, mM

108.lt0.6 9.0t0.5

Bicarbonate, mM Glucose, mM Creatinine, PM Total protein, g/l Albumin, g/l Osmolality, mosmol/kgH20

28.3t0.4* 0.92t0.07 117.6k7.23 36.8t0.6 22.3t0.4 291.1t0.6 293.921.4

Maternal Plasma (n = 10)

141.01co.7 4.01t0.03 4.11t0.05 112.3t0.5 ll.lt0.9 7.5kO.6 25.6t0.2 2.78kO.10 71.1t2.4 67.4tl.l 32.521.0 292.7+1.8$

(C)T

Maternal-to-Fetal Gradient (n = 9)

5.1t0.6 0.11t0.04

(E)t (C)T

4.5t0.8 0.4t0.5

(E)t -1.8t0.3* 1.79t0.11 -35.2kl5.91 31.3tl.6 10.6t0.9 1.3fl.Q

(C)t (E)t Values are means & SE for control and experimental animals; n, no. of animals. * When ANOVA detected a significant effect of gestational age for both groups (see text for details), mean value during pretreatment period is presented. “f When ANOVA detected a significant difference between control and experimental animals (see text for details), mean value for control (C) and experimental (E) animals is presented. $ When ANOVA detected a significant effect of gestational age for experimental group (see text for details), mean value during pretreatment period is presented. All other values were aggregated throughout gestation for each animal, and ANOVA demonstrated no significant difference between control and experimental animals nor gestational age for data presented.

2; P < 0.05). Before urine drainage, fetal plasma creatinine concentrations were similar in both groups. After the onset of urine drainage, creatinine concentrations in experimental fetuses remained low and did not change significantly during late gestation (Fig. 2). There was a signi .ficant rise in fetal plasma Na+ concentrations in both groups, with concen trations at 124 days (134.6 t 0.8 mM) being significantly lower than those on 135 (136.4 t 0.8 mM) and 140 days (136.1 t 0.9 mM; P < 0.05). There was no significant difference in arterial bicarbonate concentration between control and experimental fetuses, but there was a significant effect of gestational age in both fetal groups (P < 0.05), with values at 124 days (28.1 t 0.5 mM) being significantly higher than at 142 days (25.2 t 0.6 mM). Control ewes had higher plasma urea concentrations (by 3.4 mM) than experimental ewes at all ages both before and during the treatment period (P = 0.03). At all ages maternal plasma K+ concentrations were sig-

300

o-0

CONTROL

m-m

URINE

-5 200

N=5

1 cd

7

N=6

DRAIN PA LU

bc

ab

5

ab

,6----b,T

01 120

, 125

ab

6’

&/a

cd

j,

I/ I/

ab / -1 IfI

-1 130 GESTATIONAL

r 135 AGE (days)

I 140

1 145

Fig. 2. Plasma creatinine concentrations (PM) for control and experimental fetuses. Solid horizontal bar represents period of fetal urine drainage from 130 days (after samples were taken) until term. Creatinine concentrations in control fetuses increased significantly and progressively, whereas concentrations in experimental fetuses were not significantly different in late gestation. Different letters indicate significantly different values for control fetuses, P c 0.05.

nificantly and consistently higher by 0.16 mM in experimental animals than in the control animals (P = 0.021). There were no significant differences between groups both before and during the treatment period for both fetal and maternal glucose concentrations. Maternal plasma glucose concentrations were higher than in the fetus by 1.9 + 0.2 and 1.7 t 0.1 mM in control and experimental animals, respectively (P < 0.05). The electrolyte sum (i.e., sum of osmotically active substances) was similar to measured osmolality in maternal plasma (100.4 t 0.2%) but significantly lower in fetal plasma (97.9 t 0.2%) for both groups. The major osmotically active substances that were not measured in this study (Ca2+, Mg2+, phosphate, fructose, lactate, and a-amino acids) account for 7.4 and 13.1 mM in maternal and fetal plasma, respectively, in normal animals (3). When these values were added to the electrolyte sum it reached slightly >lOO% of the osmolality (102.9 t 0.2% for maternal plasma and 102.4 t 0.2% for fetal plasma in both control and experimental animals) and there were no significant differences between maternal and fetal plasma. Thus, for both ewes and fetuses the sum of the plasma electrolytes closely reflected the osmolalities measured although no correction was made for the activity coefficients of the solutes (3). To determine the presence of transplacental gradients for the constituents of plasma, we determined whether the concentrations of constituents were significantly different between fetal and maternal plasma. There were no significant differences between the groups, and no trends existed in relation to gestational age both before and during the treatment period for the difference between maternal and fetal plasma concentrations of Na+, K+, Cl-, total protein, albumin, urea, and glucose (Table 1). Concentrations of Na+, K+, Cl-, albumin, total protein, and glucose and osmolality and the sum of the electrolytes were significantly greater in maternal plasma than in fetal plasma (P c 0.05). Concentrations of creatinine and bicarbonate were significantly greater in fetal plasma

Downloaded from www.physiology.org/journal/ajprenal by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on January 16, 2019.

FETAL-MATERNAL

RELATIONS

DURING

than in ewe’s plasma (P < 0.05). There was no significant gradient in the concentration of urea. The difference between maternal and fetal plasma creatinine concentrations decreased significantly with advancing gestation in control animals but not in experimental animals (see Fig. 2; P < 0.05). The difference between maternal and fetal blood bicarbonate concentrations and the sum of electrolyte concentrations decreased significantly and progressively in both groups with advancing gestation (P < 0.05). Urine production and fluid volumes. At 125 days (before the onset of urine drainage) urine production was 668 t 173 (n = 6) and 671 t 95 (n = 7) ml/day for control and experimental fetuses, respectively. After the onset of urine drainage at 130 days in the experimental fetuses there was a mean loss of 542 t 49 ml urine/day over 14.1 t 1.1 days (n = 7). The total volume of fetal urine drained ranged from 5,353 ml over 11 days to 11,961 ml over 17 days. There was no significant difference in urine production measured every 5 days between the groups, and there were no significant age-related trends (Fig. 3). In the experimental fetuses (n = 5) amniotic fluid volume decreased significantly from 803 t 195 ml at 127 days (before urine drainage) to 114 t 43 and 165 t 54 ml at 133 and 138 days, respectively (P < 0.05; Fig. 4). In control fetuses (n = 5), amniotic fluid volume increased from 641 t 97 ml at 127 days to 1,024 t 178 ml on 143 days. There was no significant difference between the groups, and there were no trends before and during the treatment period in disappearance rates of 1251-BSA from amniotic fluid (fetal swallowing; 127-143 days). The mean values of disappearance rates of 1251-BSA from amniotic fluid in control and experimental fetuses over the study period were 859 t 123 (n = 18) and 519 t 79 ml/day (n = 13)) respectively. Allantoic fluid volume was 447 t 229 ml at 128 days (before urine drainage) in the experimental fetuses (n = 4). By at most the 2nd day of urine drainage it could not be sampled or drained from either catheter (P < 0.05; Fig. 4). In control fetuses (n = 3) allantoic fluid volume was variable (96-935 ml). Before the treatment period there was no significant difference in allantoic fluid 1251-BSA ^x 0 2000

CONTROL N=6 URINE DRAIN N=7

-

P L g-

1500

-

E? Q.

1000

-

W z E 3 2

500

-

G 2

k d i

0

120

:IIi+.L 125

130

GESTATIONAL AGE (days)

Fig. 3. Fetal urine production (ml/day) for control and experimental fetuses. Solid horizontal bar represents period of fetal urine drainage from 130 days until term. There was no significant difference between control and experimental fetuses, and there were no trends in relation to gestational age.

CHRONIC

1500

-

1000

-

500

-

FETAL

URINE

F675

LOSS 0

CONTROL

m

URINE

N=5

DRAIN

N=5

z W

f G 0 3 I2 0 5 z 5

b

b

T

750

,

I

Ot

0

CONTROL

m

URINE

N=3

DRAIN

N=4

-

GESTATIONAL

AGE (days)

Fig. 4. Amniotic fluid volume (ml) (top) and allantoic fluid volume (ml) (bottom) for control and experimental fetuses. Solid horizontal bars represent the period of fetal urine drainage from 130 days until term. There was a significant decrease in amniotic fluid volume after onset of fetal urine drainage and allantoic fluid volume became immeasurably small by at most the 2nd day after urine drainage, P < 0.05. Different letters indicate significantly different values, P c 0.05.

disappearance rate between control (35 t 19 ml/day; n = 3) and experimental (13 t 38 ml/day; n = 4) animals. There were no age-related trends (128-142 days) in the control animals, with a mean value of 1 t 32 ml/day (n = 10). Negative values represent disappearance of 1251 or fluid reabsorption and positive values represent an increase in allantoic fluid volume. Urine and fluid compositions. In both groups of fetuses urine osmolalities remained below 180 mosmol/kgH20. Na+ and K+ concentrations, the Na+-to-K+ ratio, and osmolalities in fetal urine and amniotic fluid and Clconcentrations in fetal urine were not significantly different between groups and no age-related trends existed (Table 2). There was a significant increase in amniotic fluid Cl- concentration during urine drainage (P < 0.05; Fig. 5). The pH of fetal urine and amniotic and allantoic fluid was not significantly different between the groups, and there were no age-related trends. Allantoic fluid Na+, K+, and Cl- concentrations, the Na+-to-K+ ratio and osmolality were not significantly different between the groups before and for 2 days after the onset of urine drainage (124, 127, 130, 131, and 132 days; Table 2). There was no significant change in the daily loss of Na+ and K+ in drained urine at all gestational ages in the

Downloaded from www.physiology.org/journal/ajprenal by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on January 16, 2019.

F676

FETAL-MATERNAL

RELATIONS

DURING

CHRONIC

FETAL

URINE

LOSS

Table 2. Electrolyte concentrations and osmolality in fetal urine and amniotic and allantoic fluids Fetal Urine (n = 11)

Amniotic Fluid (n = 11)

o-0

CONTROL

m---m

URINE DRAIN N=5

N=6

Allantoic Fluid (n = 10)

Na+, mM 16.2t1.8 111.9t6.0 110.2t3.2 K+, mM 5.95t0.74 4.31k0.22 5.36t0.65 Cl-, mM 12.2t0.7 97.0*3.5* 86.0t5.2 Na+-to-K+ ratio 3.8t0.7 28.3t1.3 24.6t3.0 Osmolality, 119.6t7.5 271.6t3.4 278.4~14.2 mosmol/kgH20 Values are means t SE for control and experimental animals; n, no. of animals. * When ANOVA detected a significant effect of gestational age in experimental animals (see text for details), mean value during pretreatment period is presented. All other values were aggregated throughout gestation for each animal and ANOVA demonstrated no significant difference between control and experimental animals nor gestational age for data presented.

experimental fetuses. In contrast there was an increase in Cl- loss in urine with advancing gestation (P < 0.05). The average urinary loss of Na+, K+ and Cl- over the entire drainage period (13.0 t 1.1 days; n = 5) was 12.0 t 3.3,4.2 t 0.6, and 7.6 t 1.1 mmol/day, respectively. Overall, there was an electrolyte (Na+, K+ and Cl-) loss of 23.8 t 4.3 mmol/day (n = 5) in experimental fetuses. Maternal food and water intake and urine output. There was no significant difference in food intake between control (1.1 t 0.1 kg, n = 6) and experimental (1.0 t 0.1 kg, n = 5) ewes both before and during the treatment period; all ewes consumed their entire daily food ration. The water intake of experimental ewes (n = 7) was significantly greater than that of control ewes (n = 6) by 1,594 ml/day at all gestational ages both before and during the treatment period (P = 0.01). In both groups there was a significant progressive increase in maternal water intake from before the onset of urine drainage [124-129 days; 2,650 t 229 (n = 6) and 4,133 t 450 ml/day (n = 7) for control and experimental animals, respectively] to 14O145 days (3,387 t 160 and 4,790 t 588 ml/day for control and experimental animals, respectively). Although maternal urine output was not significantly different between groups both before and during the treatment period, the experimental animals (2.2 t 0.2 1, n = 4) had a higher mean urine output than control animals (1.4 t 0.11, n = 6). Hormone concentrations. Fetal

and maternal plasma AVP concentrations were all below the assay sensitivity of 3.4 t 0.5 pg/ml between 124 days and the onset of labor. There were no significant effects of fetal urine drainage on fetal and maternal plasma cortisol concentrations and concentrations of prolactin in maternal and fetal plasma and amniotic and allantoic fluid, nor were there any significant gestational age trends before and during the treatment period for maternal plasma cortisol and prolactin concentrations and allantoic fluid prolactin concentrations. Fetal plasma cortisol and prolactin concentrations and amniotic fluid prolactin concentrations in control and experimental ewes rose significantly and progressively after 135-13’7 days (P < 0.05).

DISCUSSION

We have shown that, despite the continuous loss of large quantities of fluid and electrolytes via drained urine,

50 8

0

r

b

b

130

f yj

110

i W 0 or s

90

6

cl 70 L 120

0

125

I

r

130

135

I

140

145

GESTATIONAL AGE (doys)

Fig. 5. Na+ (top), K+ (middle), and Cl- (bottom) concentrations (mM) in amniotic fluid for control and experimental fetuses. Solid horizontal bars represent period of fetal urine drainage from 130 days (after samples were taken) until term. There was no significant difference between control and experimental fetuses, and there were no trends in relation to gestational age for Na+ and K+ concentrations. Cl- concentration increased significantly in experimental fetuses, P < 0.05, with no changes in control fetuses. Different letters indicate significantly different values for experimental fetuses, P < 0.05.

the fetus is able to maintain its fluid and electrolyte homeostasis and a normal rate of growth. Our data indicate that the fetal-maternal response to this challenge occurs in two phases. During the first l-3 days of urine drainage, the “acute” phase, the volumes of the amniotic and allantoic sacsare markedly decreased. These changes can be explained in part by the lack of urine entering the sacs and by the fetus continuing to swallow amniotic fluid. In the case of the allantoic sac, and possibly the amniotic sac, it is probable that fluid is reabsorbed across the fetal membranes. Although this study does not provide definitive evidence, it seems likely that fluid reabsorbed from the fetal sacs is returned to the fetal circulation. Because amniotic and allantoic fluids are separated from the fetal

Downloaded from www.physiology.org/journal/ajprenal by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on January 16, 2019.

FETAL-MATERNAL

RELATIONS

DURING

circulation by one layer of epithelial cells, these extrafetal fluids may be regarded as part of the fetal extracellular fluid (24). A transmembranous route for fluid movement from the fluid sacs to the fetus (but not to the mother) has been demonstrated in normal fetal sheep; it has been estimated that 209 ml/day of amniotic fluid is returned to the fetus via this route (14). The reduction in amniotic and allantoic fluid volumes during this acute period would have provided -380 ml of fluid to the fetus over 3 days, replacing some of the fetal urine (530 t 80 ml/day) lost over the initial phase of the drainage period. Our data indicate that during the “chronic” phase of the fetal-maternal response (i.e., after 3-5 days of urine drainage), the fetus obtains substantial amounts of water and solutes from the maternal circulation via the placenta to maintain fluid and electrolyte homeostasis. Throughout the chronic phase, 522 t 78 ml of water, 12.0 t 3.3 mmol Na+, 7.6 t 1.1 mmol Cl-, and 4.2 t 0.6 mmol K+ were lost via the fetal urine per day. It has been estimated that a water and solute (Na+, K+, Cl-) flux from ewe to fetus of 24.5 ml day-l kg body wt-l and 7.0 mmol day-l. kg body wt-l, respectively, are necessary for normal fetal growth (10). Because fetal plasma osmolality did not change and fetal growth was maintained, the quantities of water (542 ml/day) and electrolytes (24 mmol/day) lost in drained urine provide an underestimate for the maximum potential rate of transfer of water and electrolytes across the ovine placenta in late gestation. Similar values for water and electrolyte fluxes have been obtained during short-term fetal urine drainage (30) and from one fetus during 18 days of urine drainage (15). The absence of a change in fetal plasma osmolality during prolonged drainage of amniotic and allantoic fluids also indicates that large amounts (-385 ml/day) of water and electrolytes can be acquired by the fetus via the placenta (12). During the chronic phase of urine drainage, the composition of the dilute fetal urine may be indicative of flux of water and electrolytes crossing the placenta. An unexpected and novel finding was that after the acute phase of our study (i.e., after the reabsorption of amniotic and allantoic fluids) the plasma osmolality of the ewes decreased and became significantly lower than that of the fetuses. That is, the normal osmotic gradient between ewe and fetus, which has been previously documented (3) and confirmed in this study, became reversed over the last 6 days of urine drainage (i.e., at 137,140, and 142 days). The reversal of this osmotic gradient starting 5 days after the onset of urine drainage in the absence of changes in reflection coefficients of the solutes would be expected to favor water and electrolyte movement from ewe to fetus, thereby replacing water and solutes lost in fetal urine. Because data were obtained close to term at 142-147 days and samples were obtained only at 2- to 3-day intervals, we were unable to confirm the maintenance of the reversal in osmotic gradient across the placenta. Further studies beginning earlier in gestation with more frequent sampling would be necessary to confirm this observation. We were unable to determine which particular constituents were responsible for the reduction in maternal plasma osmolality, and it is most likely that there was a l

l

CHRONIC

FETAL

URINE

LOSS

F677

small undetectable change in the concentration of many, or all, constituents. The only measurable change in plasma composition during fetal urine drainage was the absence of the normal gestational increase in the concentration of creatinine in fetal plasma (2), but the reason for this is unknown. We confirmed that there are measurable gradients in the concentrations of most plasma constituents across the placenta. The plasma concentrations of Na+, K+, Cl-, albumin, glucose, and total protein were significantly greater and the plasma concentrations of creatinine and bicarbonate were significantly lower, in ewes than in fetuses (2, 3). However, we did not find a significant difference in the fetal and maternal plasma concentrations of urea (3). The mechanism by which maternal plasma osmolality decreased 5 days after the onset of urine drainage is not readily apparent. In confirmation of a study in which amniotic and allantoic fluids were drained for 20 days during late gestation in sheep (12), we did not detect a change in the water intake or urine production of the ewes. Although experimental ewes consumed significantly more water and produced more urine than control ewes, these changes did not occur in response to fetal urine drainage but were evident both before and during the treatment period. Despite fetal plasma osmolality being significantly greater in the experimental than in control animals at all ages, the maternal-to-fetal osmotic gradient was not significantly different between the two groups before and during the first 5 days of drainage. This indicates that although the ewes may have had different rates of fluid turnover, fluid transfer across the placenta to the fetus was unlikely to have been different between the two groups. The urine composition of the mother was not examined because of the difficulty of obtaining uncontaminated samples. It is of interest that fetal and maternal plasma AVP concentrations remained immeasurably low during fetal urine drainage. In humans (11), but not sheep (7), maternal plasma osmolality is reduced during pregnancy in the absence of changes in plasma AVP concentrations, suggesting a lowering in the set point of the osmotic threshold for AVP secretion. It is possible that such a change in the set point for AVP secretion occurred in response to the prolonged loss of water and electrolytes from the maternal circulation, thereby allowing the plasma osmolality of the mother to fall. This implies that during the chronic phase the mother retains fluid, thereby lowering her plasma osmotic pressure. More accurate methods for measuring maternal fluid balance would be needed to confirm this suggestion by reducing the variability of maternal water intake and urine output measurements. A reversal in the osmotic gradient between maternal and fetal plasma would be expected to favor water and electrolyte transfer from the ewe to fetus. However, it is also possible that changes in the placental reflection coefficients for specific electrolytes or in the permeability of the placenta to major electrolytes could have contributed to this transfer (13). Under normal conditions, Na+ and Cl- exert significant osmotic pressures opposing the

Downloaded from www.physiology.org/journal/ajprenal by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on January 16, 2019.

F678

FETAL-MATERNAL

RELATIONS

DURING

transfer of water to the fetus (3). In the absence of significant changes in the transplacental concentrations of Na+ and Cl- it is possible that the increase in water transfer across the placenta to the fetus could have occurred as a result of alterations in the reflection coefficients for Na+ and Cl- in the placental vessels or in the permeability of the placenta to Na+ and Cl-. Placental reflection coefficients in sheep have been found to be inversely related to fetal urine osmolality (30). Because fetal urine osmolality is considered to be under the control of fetal plasma AVP during late gestation (37, 38) it is likely that fetal AVP also modulates permeability of the placenta (30), thus providing a mechanism whereby the fetal acquisition of water may be regulated. Because we were unable to detect any change in fetal plasma AVP concentrations and urine osmolality was consistently

Fetal-maternal fluid and electrolyte relations during chronic fetal urine loss in sheep.

Our aim was to determine the effects of prolonged removal of fetal urine during late gestation on fetal-maternal fluid and electrolyte relationships. ...
3MB Sizes 0 Downloads 0 Views