- Email: [email protected]
The relationship between reproductive performance and blood pressure in the spontaneously hypertensive rat
The relationship between reproductive performance and blood pressure in the spontaneously hypertensive rat Robert P. Lorenz, M.D., Luciano P. Picchio, M.D., Judith Weisz, M.B., and Tom Lloyd, Ph.D. Hershey, Pennsylvania Blood pressure was measured serially throughout pregnancy in spontaneously hypertensive rats and in normotensive control rats of the Wistar-Kyoto strain. Changes in blood pressure were also correlated with the outcome of pregnancy. In control rats there was a small but significant decrease in blood pressure between days 20 and 22 of pregnancy (day of birth = day 22.5). In the spontaneously hypertensive rat this decrease occurred earlier, between days 18 and 19 of pregnancy, and was of greater magnitude. There was a positive correlation in the spontaneously hypertensive rat between the number of live-born pups and the magnitude of the decrease in blood pressure. Perinatal mortality, but not litter size, was greater in the spontaneously hypertensive rat than in control rats. Thus the physiologic mechanisms responsible for the decrease in blood pressure in the normal rat are preserved in the spontaneously hypertenSive rat, and the successful completion and outcome of pregnancy depend on the capacity of the hypertensive rat to amplify these processes. Consequently, the spontaneously hypertensive rat is not an appropriate experimental model for hypertension of human pregnancy, in particular for preeclampsia, in which the presence of the conceptus characteristically causes blood pressure to rise , especially during the last trimester. However, an investigation of the factors responsible for the profound antihypertensive effect of pregnancy in rats could provide new insights into the mechanisms by which blood pressure is regulated during pregnancy and suggest new therapeutic approaches. (AM J Os STET GVNECOL 1984;150:519-23.)
Hypertensive disorders of pregnancy are responsible in the United States for about 20% of maternal and about 12% of perinatal deaths.! Preexisting (essential) hypertension is a predisposing factor for preeclampsia, the specific hypertensive disorder associated with pregnancy. The factors responsible for the adverse effect of pregnancy on blood pressure regulation in humans are not known. Their identification could be facilitated by the availability of an animal model. However, a similar disease has not yet been identified in laboratory animals , and a satisfactory equivalent has not been induced experimentally . The congenitally (spontaneously) hypertensive rat is generally accepted as a useful model for studies in human essential h ypertension .2 - s However, from the limited number of studies carried out on the influence of pregnancy on this strain of rats it would appear that their response to pregnancy differs markedly from that of the human with essential hypertension . Blood pressure in the spontaneously hypertensive rat has been reported either not to change during the course of pregnancy6. 7 or even to decr~ ase as term approaches. B• 9 To evaluate more fully the conFrom the Department of Obstetrics and Gynecology, The Milton S. Hershey Medical Center, The Pennsylvania State University. Supported in part by Public H ealth Service Grant GM 23134. Sponsored by the Society for Gynecologic Investigation. Reprint requests: Robert P. Lorenz, M.D ., Chief; Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, William Beaumont Hospital, 3601 West 13 Mile R oad, Royal Oak, M1
48072 .
sequences of pregnancy on essential h ypertension in this experimental model, we correlated blood pressure changes with both the o utcome of pregnancy and total reproductive performance. In additio n, we obtained data on the age at vaginal opening, an index of the onset of puberty , and the age at which regular estrous cycles are established. NormotensiveWistar-Kyoto rats , derived from' the same strain, served as controls. Methodology Fourteen animals (Charles River, Massachusetts) from each strain were housed in open bottom cages in two groups of seven. Animals were fed ad libitum a speciall y prepared diet in pellets (7% mixed salts, 24.1 % protein , 55.5% carbohydrate, 5% fiber , 5.5% fat with vitamins [Vitamin Mix 740]; Bio-Serve, Inc., Frenchtown, New Jersey). All animals were maintained in a controlled environment (23 C, 50% humidity, lights on at 0600 hours, off at 1800 hours) and given free access to tap water. The following parameters were evaluated : age at vaginal ()pening (a correlate of puberty) , the estrous cycle as indicated by vaginal smear, blood pressure and weight before and every second day during pregnancy, number of matings required to achieve pregnancy, and outcome of pregnancy including resorptions, litter size, number of stillbirths, and neonatal survival. Vaginal smears were examined d aily beginning at the time of vaginal opening and until day 19 after mating. Females were first mated at 70 to 80 0
519
520
Lorenz et al.
November I, 1984 Am J Obstet Gynecol
Table I. Reproductive outcome of spontaneously hypertensive rats compared to Wistar-Kyoto strain Spontaneously hypertensive rats
No. of matings, n Total pregnancies, n Resorptions, n (%) Live births, n Stillbirths, n Litter size, mean ± SD Neonatal deaths (0-10 days), n Perinatal loss, n (%) Length of pregnancy, days Prepregnancy weight, gm Weight gain, gm Weight gain, %
31 23 9 (39)
102 4
7.3 ± 2.6 71 75 (71)
22 ± 0.5
168.5 ±16.2 65.2 ± 10.4 39
Wistar-Kyoto strain
p
13 12 1(8.5) 66
4 6.4 ± 2.8
34 38 21 126 63.8
(54)
p P
x2
=
± 1.0
± 17 ± 18 52
t = t=
t=
= =
0.17* 0.06*
10.6, P < 0.01 6.14, P < 0.Ql 0.23, p> 0.5 2.21, P < 0.05t
*Fisher's exact test. t Arcsine transformation. 11 days of life. Blood pressure was measured using a tail cuff, sensor, and dynograph. lO , 11 Each blood pressure value reported is the mean of at least three determinations that were taken during the same session and varied by less than 5%. Mating was accomplished by housing overnight with a male of the same strain up to four females that showed vaginal proestrus. If vaginal smears did not show a typical estrous pattern and sperm, the process was repeated the following day. Fertility was assessed by determining the number of matings at the appropriate stage of the estrous cycle per pregnancy. Pregnancy was confirmed if the smear remained diestrus from day 2 until day 14 after mating. Pregnant females were housed individually in closed-bottom cages with bedding material from day 19 until weaning of the offspring. If no delivery occurred in spite of evidence of pregnancy, it was assumed resorption had occurred. Perinatal loss included all stillborn pups and those born live that did not survive to 10 days of age. Data were analyzed statistically by means of Student's t test, Newman-Keuls test, and X2 • Arcsine transformation of data expressed as percent (relative weight gain, relative decrease in blood pressure) was used. 12 Blood pressure was analyzed by a two-way analysis of variance (AN OVA) with one independent measure (strain) and one repeated measure (gestational age). Blood pressures were grouped by gestational age for statistical analysis; p values less than 0.05 were considered significant. Results
Vaginal opening occurred somewhat later in the spontaneous hypertensive rat (42 ± 3 days) than in the control rat (39 ± 3 days) (p < 0.02). All animals achieved regular 4-day estrous cycles by 50 days of age. Fertility was not significantly different in the two strains. The spontaneously hypertensive rat had a higher prepregnancy weight, but a similar absolute total weight gain during pregnancy as the control rat
(Table I). There was no significant difference in absolute weight gain between groups. Weight gain as a percent of prepregnancy weight did, however, differ between strains, with the spontaneously hypertensive rat gaining less weight as a percent of prepregnancy weight (39.1% versus 52%, t = 2.21, P < 0.05, arcsine transformation l1 ). Duration of pregnancy also did not differ between the strains. Mean blood pressure values for both the spontaneously hypertensive rat and control rat throughout gestation are shown in Fig. 1. Blood pressure in the spontaneously hypertensive rat was significantly higher than in the control rat (F = 139.8; df = 1, 23; P < 0.01). Blood pressure also varied significantly with gestational age (F = 37.7; df = 5, 15; P < 0.01). At every gestational age, the spontaneously hypertensive rat had a significantly higher blood pressure than the control rat (e.g., days 20-22; t = 3.92; df = 34; P < 0.01). The timing of the decrease in blood pressure differed between the strains. In the control rat, blood pressure was stable until a decrease occurred on days 20 to 22 of gestation (Newman-Keuls, p < 0.01). In the spontaneously hypertensive rat, blood pressure decreased earlier, on days 18 to 19 (Newman-Keuls, p < 0.01). Furthermore, t tests indicated that the absolute decrease in blood pressure from days 12 to 17 to the final blood pressure readingjust before parturition was significantly greater in spontaneously hypertensive rats than in control rats (t = 3.00; df = 23; P < 0.005). Relative decrease in blood pressure over this time period, however, was not significantly different between the strains. Table II and Fig. 2 illustrate the relationship between the decrease in blood pressure over this time period and the number of live-born pups. In the spontaneously hypertensive rat, the number of live-born pups was positively correlated with the absolute decrease in blood pressure (r=0.78; df= 13; p
Reproduction and blood pressure in hypertensive rat
Volume 150 Number 5, Part I
(r = 0.76; df= 13; P < 0.002, arcsine transformation) but not with the absolute blood pressure readings on days 12 to 17 in the spontaneously hypertensive rat. Also, the number of live-born pups did not correlate with any blood pressure measures in the control rat. Details of the outcome of pregnancy are listed in Table I. There was no difference in litter size between the strains, but perinatal mortality was higher in the spontaneously hypertensive rat, due to a higher neonatal mortality rate. The rate of resorption of pregnancy was higher in the spontaneously hypertensive rat but did not attain statistical significance. Three control rats and one spontaneously hypertensive rat adult female rats died during the study. All had clinical evidence of respiratory infection, which was confirmed by autopsy. These animals were isolated from the survivors at the time of onset of symptoms. Two of the control rats died 7 and 10 days after a resorbed pregnancy was diagnosed. These two pregnancies were not included in the final data because symptoms of pneumonia were noted during the pregnancy. The remaining two animals were never pregnant. Surviving animals were examined and showed no clinical evidence of infection. Comment
Pregnancy in humans, as in other mammals, is associated with marked changes in electrolyte and fluid balance . Blood and extracellular fluid volume increase as a result of salt retention. These changes are associated with increased cardiac output, heart rate, and decreased peripheral vascular resistance. These hemodynamic adjustments, combined with the presence of the fetoplacental unit acting as an arteriovenous shunt, ensure that blood pressure remains essentially stable in face of the altered salt and water balance. Some of the humoral agents and mechanisms responsible for establishing and maintaining homeostasis in the cardiovascular system during pregnancy have been identified, e.g., progesterone and increased resistance to angiotensin . These homeostatic mechanisms not only prevent any rise in blood pressure but may actually cause it to fall at some stage of pregnancy. In the rat such a decline occurs, as shown previously by othersB and confirmed in this study, during the last days of pregnancy. In humans also, blood pressure may fall transiently during the first two trimesters in the normal patient as well as some patients with essential hypertension . However, the two species differ profoundly in the timing of blood pressure changes in normal pregnancy and in their ability to maintain circulatory homeostasis in the presence of hypertension. In women with essential hypertension, blood pressure commonly escapes from control, in particular during the third trimester when they may become suscep-
Gestational Age vs. Blood Pressure
leo C.
i
SHRWKY--
ISO
oS ~
140
Ii:
130
...,"
"8 ffi
521
120
110
1"
f---
----t------+------ r
1"~ \WKY
\i
,
prepregnancy
l'
'"
'" I
~ 'r ....
.,
II)
l'
II) II)
Gestational Age (Days)
Fig. 1. Relationship of gestational age and blood pressure. SHR, Spontaneously hypertensive rats; WKY, Wistar-Kyoto strain of rats.
tible to preeclampsia. In contrast, in the rat, the ability to lower blood pressure before delivery is preserved as shown in the present study. This phenomenon, as well as the fact that the magnitude of the decline in blood pressure is greater than normal, has been reported previously by Aoi et al. B In addition, in the present study we have shown that the decline in blood pressure in the spontaneously hypertensive rat occurs significantly earlier than in the control rat and that the magnitude of the decline in blood pressure correlates positively with the number of live pups born. Most other aspects of reproductive function, such as age when regular estrous cycles become established, fertility rate, and the number of fetuses, were not significantly affected by the hypertensive state of the mothers. Thus the major adverse consequences were noted on the number of fetuses born alive and the number of surviving after birth. There have been two previous reports that pregnancy had no influence on established hypertension in the spontaneously hypertensive rat. In one of these, by McCarty and Kopin,6 the rats were studied only at one time point, on day 20 of gestation, and under halothane anesthesia. In the second study Yamada et aU measured blood pressure serially but pooled the data across days 15 to 20 of pregnancy, which could account for their failure to identify a decline in blood pressure in either the normal or the hypertensive rats. A single study has been carried out to evaluate the influence of hypertension on hemodynamic parameters. 9 Two days before the expected day of delivery, the only time studied, the sole abnormality identified in the spontaneously hypertensive rat was a marked decrease in myometrial and placental blood flow. The percent
522
Lorenz et al.
November I, 1984 Am J Obstet Gynecol
10
r=.76
p<.002
~
~
Zw 30 -a: w::J
t ~~
Zw 30 -a:
w::Jt
CI)CI)
CI) CI) c(CI)
o a..
we eo 0
,
..J
~ ffi a: a: 20 o a..
20
we eO
10
0
..J
III
10
III
o
o Number of liveborn in individual rats vs % decrease in blood pressure in late pregnancy.
Fig. 2. Number of live births of individual rats in relation to the percent decrease in blood pressure in late pregnancy.
Table II. Number of live births of individual spontaneously hypertensive rats and control rats in relation to the decrease in blood pressure during pregnancy Day 12-17 mean blood pressure (mm Hg) (A)
Spontaneously hypertensive rats I
2 3 4 5 6 7 8 9 10
II
12 13 14 Wistar-Kyoto strain
1 2 3 4 5 6 7 8 9 10 11 *C = (BfA) x 100. t Arcsine transformation."
Final blood pressure (mm Hg)
101 104 120
Difference (mm Hg) (8)
157.6 151.8 173.9 161 142.3 167.8 165.5 159.6 155.9 149.1 167.9 162.3 139.7 168.5
101 122.7 120 11 9 127 126 142.8 140 128 167
56.6 47 .8 53.9 50 41.3 45.1 45.5 40.6 28.9 23.1 25.1 22.3 11.7 1.5
117.3 121.8 129.5 106.6 116.8 105.7 130.3 113.2 122.3 104.8 124.3
77 91 102.7 86.3 95 87 121 106 115.3 101 120
40.3 30.8 26.8 20.3 21.8 18.7 9.3 7.2 7 3.8 4.3
III
% decrease
Live births (n)
(C) *
=
8 10 10 9 8 9 7 8 8 9 4 5 6 1 13, P < 0.002t
34 25 21 19 19 18 7 6 6 4 3 r = 0.464, df
9 8 8 7 5 2 7 9 7 2 2 10, P > O.lt
36 31 31 31 29 27 27 25 19 15 15 14 8 1 r = 0.760, df
=
Volume 150 Number 5, Part 1
increase in mean cardiac output and heart rate and the percent decrease in total peripheral resistance associated with pregnancy were similar in the spontaneously hypertensive rat and the Wistar-Kyoto rats. It should be noted , however, that these studies were carried out with the rats under barbiturate anesthesia. These investigators also noted a decrease during pregnancy in mean blood pressure in awake spontaneously hypertensive rats and control rats. B, 9 The antihypertensive effect of pregnancy observed in the spontaneously h ypertensive rat by us as well as by Aoi and coworkers, however, is not limited to the congenital form of hypertension in the rat. Goldblatt et al. 13 demonstrated this phenomenon as early as 1939 in dogs with experimentally induced renal hypertension. Their observations were subsequently confirmed on rats and rabbits by Pa ge et al. 14 These investigators were also the first to point out the importance of trying to identify the basis for the differences between these species and the human. Using the too ls then available, the y eliminated a number of factors and agents that might be implicated, including estrogen, progesterone, "chorionic hormone ," "lactogenic hormone," and pregnant mare serum. IS They made the interesting observations that the presence of experimentally induced deciduoma, but not pseudopregnancy, protected rats from the hypertensive effects of partial ligation of one renal artery. In addition , they demonstrated that pregnancy could not protect rats from hypertension induced by the administration of desox ycorticosterone acetate a nd salt. 16 From this they concluded that the simple physical presence of the placenta, acting as a vascular shunt, could not be responsible for the protection offered by pregnancy against renal hypertension. Their interesting observations and perceptive conclusions have not been followed up and have remained largely ignored since the 1940s. The questions concerning mechanisms of blood pressure regulations originally raised by Page and coworkers deserve to be reinvestigated with use of the tools now available. Although the spontaneously hypertensive rat is a good model for essential hypertension in the nonpregnant human, it does not parallel hypertensive disorders of human pregnancy. The value of studying animals with diseases tha t resemble those that occur in human is well recognized. Potentially important information could, however, be obtained also by investigating species that appear to possess mechanisms that protect them from diseases to which humans are prone.
Reproduction and blood pressure in hypertensive rat
523
Identifying the humoral agents and the physiologic mechanisms responsible for the hypotensive effect of pregnancy in the normal rat and the ability of hypertensive rats to amplify such mechanisms could provide clues to new approaches to therapy. We gratefully acknowledge the assistance of Dr. Byron Ward, Villanova University, and Catherine Hansen and Mary Ann Ryan in the preparation of this manuscript. REFERENCES
1. Chesley LC, ed. Hypertensive disorders in pregnancy. New York: Appleton-Century-Crofts, 1978:3. 2. Okamoto K, Aoki K. Development of a strain of spontaneousl y hypertensive rats. jpn Circ j 1963;27:282. 3. Folkow BUG, Hallback M. Pathophysiology of spontaneous hypertension in rats. In: Genestj, Kiow E, Kuchel 0 , eds. Hypertension . New York: McGraw-Hill, 1977:507. 4. Brody Mj, HaywoodjR, Toure KB. Neural mechanisms in hypertension. Ann Rev Physiol 1980;42:441. 5. Trippodo NC, Frohlich ED. Similarities of genetic (spontaneous) hypertension-man and rat. Cir Res 1981 ; 48(3) :309. 6. McCarty R, Kopin IJ. Pregnancy: its effects on blood pressure, heart rate and sympatho-adrenal activity in spontaneously hypertensive rats. Proc Soc Exp Bioi Med 1978; 158:242. 7. Yamada N, et al. Hypertensive effects on pregnancy in spontaneousl y hypertensive rat and spontaneously hypertensive rat-stroke prone. Int j Bioi Res Preg 1981 ;2(2) :80. 8. Aoi W, et al. Antihypertensive effect of pregnancy in spontaneously hypertensive rats. Proc Soc Exp Bioi Med 1976;153(1):13. 9. Lundgren Y, Karlsson K, Ljungblad U. Circulatory changes during pregnancy in spontaneously and renal hypertensive rats. Clin Sci 1979;57:337s. 10. Bunag RD, Riley E. Simultaneous measurements in awake rats of drug-induced changes in carotid and tailcuff systolic pressures. j Appl Physiol 1974;36:621. 11. Frohlich ED, Pfeffer jM, Pfeffer MA. Validity of an indirect tail-cuff method for determining systolic arterial pressure in unanesthetized normotensive and spontaneously hypertensive rats. j Lab Clin Med 1971;78(6):957. 12. Zar jH, ed. Biostatistical analysis. Prentice-Hall: Englewood Cliffs, New jersey, 1974:261. 13. Goldblatt H, KahnjR, Honzal RF. Studies on experimental hypertension. IX. The effect on blood pressure of constriction of the abdominal aorta above and below the site of origin of both main renal arteries. j Exp Med 1939;69:649. 14. Page EW, Patton HS, Ogden E. The effect of pregnancy on experimental hypertension. AM j OBSTET GYNECOL 1941 ;41:53. 15. Page EW, Ogden E. Endocrine influences upon the blood pressure of normal and hypertensive rats. AM j OBSTET GYNECOL 1947;47:150. 16. Page EW, Glendening MG. Influence of pregnancy upon hypertension induced in rats by sodium chloride and desoxycorticosterone. Proc Soc Exp Bioi Med 1953;82: 466.
Hypertensive disorders of pregnancy are responsible in the United States for about 20% of maternal and about 12% of perinatal deaths.! Preexisting (essential) hypertension is a predisposing factor for preeclampsia, the specific hypertensive disorder associated with pregnancy. The factors responsible for the adverse effect of pregnancy on blood pressure regulation in humans are not known. Their identification could be facilitated by the availability of an animal model. However, a similar disease has not yet been identified in laboratory animals , and a satisfactory equivalent has not been induced experimentally . The congenitally (spontaneously) hypertensive rat is generally accepted as a useful model for studies in human essential h ypertension .2 - s However, from the limited number of studies carried out on the influence of pregnancy on this strain of rats it would appear that their response to pregnancy differs markedly from that of the human with essential hypertension . Blood pressure in the spontaneously hypertensive rat has been reported either not to change during the course of pregnancy6. 7 or even to decr~ ase as term approaches. B• 9 To evaluate more fully the conFrom the Department of Obstetrics and Gynecology, The Milton S. Hershey Medical Center, The Pennsylvania State University. Supported in part by Public H ealth Service Grant GM 23134. Sponsored by the Society for Gynecologic Investigation. Reprint requests: Robert P. Lorenz, M.D ., Chief; Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, William Beaumont Hospital, 3601 West 13 Mile R oad, Royal Oak, M1
48072 .
sequences of pregnancy on essential h ypertension in this experimental model, we correlated blood pressure changes with both the o utcome of pregnancy and total reproductive performance. In additio n, we obtained data on the age at vaginal opening, an index of the onset of puberty , and the age at which regular estrous cycles are established. NormotensiveWistar-Kyoto rats , derived from' the same strain, served as controls. Methodology Fourteen animals (Charles River, Massachusetts) from each strain were housed in open bottom cages in two groups of seven. Animals were fed ad libitum a speciall y prepared diet in pellets (7% mixed salts, 24.1 % protein , 55.5% carbohydrate, 5% fiber , 5.5% fat with vitamins [Vitamin Mix 740]; Bio-Serve, Inc., Frenchtown, New Jersey). All animals were maintained in a controlled environment (23 C, 50% humidity, lights on at 0600 hours, off at 1800 hours) and given free access to tap water. The following parameters were evaluated : age at vaginal ()pening (a correlate of puberty) , the estrous cycle as indicated by vaginal smear, blood pressure and weight before and every second day during pregnancy, number of matings required to achieve pregnancy, and outcome of pregnancy including resorptions, litter size, number of stillbirths, and neonatal survival. Vaginal smears were examined d aily beginning at the time of vaginal opening and until day 19 after mating. Females were first mated at 70 to 80 0
519
520
Lorenz et al.
November I, 1984 Am J Obstet Gynecol
Table I. Reproductive outcome of spontaneously hypertensive rats compared to Wistar-Kyoto strain Spontaneously hypertensive rats
No. of matings, n Total pregnancies, n Resorptions, n (%) Live births, n Stillbirths, n Litter size, mean ± SD Neonatal deaths (0-10 days), n Perinatal loss, n (%) Length of pregnancy, days Prepregnancy weight, gm Weight gain, gm Weight gain, %
31 23 9 (39)
102 4
7.3 ± 2.6 71 75 (71)
22 ± 0.5
168.5 ±16.2 65.2 ± 10.4 39
Wistar-Kyoto strain
p
13 12 1(8.5) 66
4 6.4 ± 2.8
34 38 21 126 63.8
(54)
p P
x2
=
± 1.0
± 17 ± 18 52
t = t=
t=
= =
0.17* 0.06*
10.6, P < 0.01 6.14, P < 0.Ql 0.23, p> 0.5 2.21, P < 0.05t
*Fisher's exact test. t Arcsine transformation. 11 days of life. Blood pressure was measured using a tail cuff, sensor, and dynograph. lO , 11 Each blood pressure value reported is the mean of at least three determinations that were taken during the same session and varied by less than 5%. Mating was accomplished by housing overnight with a male of the same strain up to four females that showed vaginal proestrus. If vaginal smears did not show a typical estrous pattern and sperm, the process was repeated the following day. Fertility was assessed by determining the number of matings at the appropriate stage of the estrous cycle per pregnancy. Pregnancy was confirmed if the smear remained diestrus from day 2 until day 14 after mating. Pregnant females were housed individually in closed-bottom cages with bedding material from day 19 until weaning of the offspring. If no delivery occurred in spite of evidence of pregnancy, it was assumed resorption had occurred. Perinatal loss included all stillborn pups and those born live that did not survive to 10 days of age. Data were analyzed statistically by means of Student's t test, Newman-Keuls test, and X2 • Arcsine transformation of data expressed as percent (relative weight gain, relative decrease in blood pressure) was used. 12 Blood pressure was analyzed by a two-way analysis of variance (AN OVA) with one independent measure (strain) and one repeated measure (gestational age). Blood pressures were grouped by gestational age for statistical analysis; p values less than 0.05 were considered significant. Results
Vaginal opening occurred somewhat later in the spontaneous hypertensive rat (42 ± 3 days) than in the control rat (39 ± 3 days) (p < 0.02). All animals achieved regular 4-day estrous cycles by 50 days of age. Fertility was not significantly different in the two strains. The spontaneously hypertensive rat had a higher prepregnancy weight, but a similar absolute total weight gain during pregnancy as the control rat
(Table I). There was no significant difference in absolute weight gain between groups. Weight gain as a percent of prepregnancy weight did, however, differ between strains, with the spontaneously hypertensive rat gaining less weight as a percent of prepregnancy weight (39.1% versus 52%, t = 2.21, P < 0.05, arcsine transformation l1 ). Duration of pregnancy also did not differ between the strains. Mean blood pressure values for both the spontaneously hypertensive rat and control rat throughout gestation are shown in Fig. 1. Blood pressure in the spontaneously hypertensive rat was significantly higher than in the control rat (F = 139.8; df = 1, 23; P < 0.01). Blood pressure also varied significantly with gestational age (F = 37.7; df = 5, 15; P < 0.01). At every gestational age, the spontaneously hypertensive rat had a significantly higher blood pressure than the control rat (e.g., days 20-22; t = 3.92; df = 34; P < 0.01). The timing of the decrease in blood pressure differed between the strains. In the control rat, blood pressure was stable until a decrease occurred on days 20 to 22 of gestation (Newman-Keuls, p < 0.01). In the spontaneously hypertensive rat, blood pressure decreased earlier, on days 18 to 19 (Newman-Keuls, p < 0.01). Furthermore, t tests indicated that the absolute decrease in blood pressure from days 12 to 17 to the final blood pressure readingjust before parturition was significantly greater in spontaneously hypertensive rats than in control rats (t = 3.00; df = 23; P < 0.005). Relative decrease in blood pressure over this time period, however, was not significantly different between the strains. Table II and Fig. 2 illustrate the relationship between the decrease in blood pressure over this time period and the number of live-born pups. In the spontaneously hypertensive rat, the number of live-born pups was positively correlated with the absolute decrease in blood pressure (r=0.78; df= 13; p
Reproduction and blood pressure in hypertensive rat
Volume 150 Number 5, Part I
(r = 0.76; df= 13; P < 0.002, arcsine transformation) but not with the absolute blood pressure readings on days 12 to 17 in the spontaneously hypertensive rat. Also, the number of live-born pups did not correlate with any blood pressure measures in the control rat. Details of the outcome of pregnancy are listed in Table I. There was no difference in litter size between the strains, but perinatal mortality was higher in the spontaneously hypertensive rat, due to a higher neonatal mortality rate. The rate of resorption of pregnancy was higher in the spontaneously hypertensive rat but did not attain statistical significance. Three control rats and one spontaneously hypertensive rat adult female rats died during the study. All had clinical evidence of respiratory infection, which was confirmed by autopsy. These animals were isolated from the survivors at the time of onset of symptoms. Two of the control rats died 7 and 10 days after a resorbed pregnancy was diagnosed. These two pregnancies were not included in the final data because symptoms of pneumonia were noted during the pregnancy. The remaining two animals were never pregnant. Surviving animals were examined and showed no clinical evidence of infection. Comment
Pregnancy in humans, as in other mammals, is associated with marked changes in electrolyte and fluid balance . Blood and extracellular fluid volume increase as a result of salt retention. These changes are associated with increased cardiac output, heart rate, and decreased peripheral vascular resistance. These hemodynamic adjustments, combined with the presence of the fetoplacental unit acting as an arteriovenous shunt, ensure that blood pressure remains essentially stable in face of the altered salt and water balance. Some of the humoral agents and mechanisms responsible for establishing and maintaining homeostasis in the cardiovascular system during pregnancy have been identified, e.g., progesterone and increased resistance to angiotensin . These homeostatic mechanisms not only prevent any rise in blood pressure but may actually cause it to fall at some stage of pregnancy. In the rat such a decline occurs, as shown previously by othersB and confirmed in this study, during the last days of pregnancy. In humans also, blood pressure may fall transiently during the first two trimesters in the normal patient as well as some patients with essential hypertension . However, the two species differ profoundly in the timing of blood pressure changes in normal pregnancy and in their ability to maintain circulatory homeostasis in the presence of hypertension. In women with essential hypertension, blood pressure commonly escapes from control, in particular during the third trimester when they may become suscep-
Gestational Age vs. Blood Pressure
leo C.
i
SHRWKY--
ISO
oS ~
140
Ii:
130
...,"
"8 ffi
521
120
110
1"
f---
----t------+------ r
1"~ \WKY
\i
,
prepregnancy
l'
'"
'" I
~ 'r ....
.,
II)
l'
II) II)
Gestational Age (Days)
Fig. 1. Relationship of gestational age and blood pressure. SHR, Spontaneously hypertensive rats; WKY, Wistar-Kyoto strain of rats.
tible to preeclampsia. In contrast, in the rat, the ability to lower blood pressure before delivery is preserved as shown in the present study. This phenomenon, as well as the fact that the magnitude of the decline in blood pressure is greater than normal, has been reported previously by Aoi et al. B In addition, in the present study we have shown that the decline in blood pressure in the spontaneously hypertensive rat occurs significantly earlier than in the control rat and that the magnitude of the decline in blood pressure correlates positively with the number of live pups born. Most other aspects of reproductive function, such as age when regular estrous cycles become established, fertility rate, and the number of fetuses, were not significantly affected by the hypertensive state of the mothers. Thus the major adverse consequences were noted on the number of fetuses born alive and the number of surviving after birth. There have been two previous reports that pregnancy had no influence on established hypertension in the spontaneously hypertensive rat. In one of these, by McCarty and Kopin,6 the rats were studied only at one time point, on day 20 of gestation, and under halothane anesthesia. In the second study Yamada et aU measured blood pressure serially but pooled the data across days 15 to 20 of pregnancy, which could account for their failure to identify a decline in blood pressure in either the normal or the hypertensive rats. A single study has been carried out to evaluate the influence of hypertension on hemodynamic parameters. 9 Two days before the expected day of delivery, the only time studied, the sole abnormality identified in the spontaneously hypertensive rat was a marked decrease in myometrial and placental blood flow. The percent
522
Lorenz et al.
November I, 1984 Am J Obstet Gynecol
10
r=.76
p<.002
~
~
Zw 30 -a: w::J
t ~~
Zw 30 -a:
w::Jt
CI)CI)
CI) CI) c(CI)
o a..
we eo 0
,
..J
~ ffi a: a: 20 o a..
20
we eO
10
0
..J
III
10
III
o
o Number of liveborn in individual rats vs % decrease in blood pressure in late pregnancy.
Fig. 2. Number of live births of individual rats in relation to the percent decrease in blood pressure in late pregnancy.
Table II. Number of live births of individual spontaneously hypertensive rats and control rats in relation to the decrease in blood pressure during pregnancy Day 12-17 mean blood pressure (mm Hg) (A)
Spontaneously hypertensive rats I
2 3 4 5 6 7 8 9 10
II
12 13 14 Wistar-Kyoto strain
1 2 3 4 5 6 7 8 9 10 11 *C = (BfA) x 100. t Arcsine transformation."
Final blood pressure (mm Hg)
101 104 120
Difference (mm Hg) (8)
157.6 151.8 173.9 161 142.3 167.8 165.5 159.6 155.9 149.1 167.9 162.3 139.7 168.5
101 122.7 120 11 9 127 126 142.8 140 128 167
56.6 47 .8 53.9 50 41.3 45.1 45.5 40.6 28.9 23.1 25.1 22.3 11.7 1.5
117.3 121.8 129.5 106.6 116.8 105.7 130.3 113.2 122.3 104.8 124.3
77 91 102.7 86.3 95 87 121 106 115.3 101 120
40.3 30.8 26.8 20.3 21.8 18.7 9.3 7.2 7 3.8 4.3
III
% decrease
Live births (n)
(C) *
=
8 10 10 9 8 9 7 8 8 9 4 5 6 1 13, P < 0.002t
34 25 21 19 19 18 7 6 6 4 3 r = 0.464, df
9 8 8 7 5 2 7 9 7 2 2 10, P > O.lt
36 31 31 31 29 27 27 25 19 15 15 14 8 1 r = 0.760, df
=
Volume 150 Number 5, Part 1
increase in mean cardiac output and heart rate and the percent decrease in total peripheral resistance associated with pregnancy were similar in the spontaneously hypertensive rat and the Wistar-Kyoto rats. It should be noted , however, that these studies were carried out with the rats under barbiturate anesthesia. These investigators also noted a decrease during pregnancy in mean blood pressure in awake spontaneously hypertensive rats and control rats. B, 9 The antihypertensive effect of pregnancy observed in the spontaneously h ypertensive rat by us as well as by Aoi and coworkers, however, is not limited to the congenital form of hypertension in the rat. Goldblatt et al. 13 demonstrated this phenomenon as early as 1939 in dogs with experimentally induced renal hypertension. Their observations were subsequently confirmed on rats and rabbits by Pa ge et al. 14 These investigators were also the first to point out the importance of trying to identify the basis for the differences between these species and the human. Using the too ls then available, the y eliminated a number of factors and agents that might be implicated, including estrogen, progesterone, "chorionic hormone ," "lactogenic hormone," and pregnant mare serum. IS They made the interesting observations that the presence of experimentally induced deciduoma, but not pseudopregnancy, protected rats from the hypertensive effects of partial ligation of one renal artery. In addition , they demonstrated that pregnancy could not protect rats from hypertension induced by the administration of desox ycorticosterone acetate a nd salt. 16 From this they concluded that the simple physical presence of the placenta, acting as a vascular shunt, could not be responsible for the protection offered by pregnancy against renal hypertension. Their interesting observations and perceptive conclusions have not been followed up and have remained largely ignored since the 1940s. The questions concerning mechanisms of blood pressure regulations originally raised by Page and coworkers deserve to be reinvestigated with use of the tools now available. Although the spontaneously hypertensive rat is a good model for essential hypertension in the nonpregnant human, it does not parallel hypertensive disorders of human pregnancy. The value of studying animals with diseases tha t resemble those that occur in human is well recognized. Potentially important information could, however, be obtained also by investigating species that appear to possess mechanisms that protect them from diseases to which humans are prone.
Reproduction and blood pressure in hypertensive rat
523
Identifying the humoral agents and the physiologic mechanisms responsible for the hypotensive effect of pregnancy in the normal rat and the ability of hypertensive rats to amplify such mechanisms could provide clues to new approaches to therapy. We gratefully acknowledge the assistance of Dr. Byron Ward, Villanova University, and Catherine Hansen and Mary Ann Ryan in the preparation of this manuscript. REFERENCES
1. Chesley LC, ed. Hypertensive disorders in pregnancy. New York: Appleton-Century-Crofts, 1978:3. 2. Okamoto K, Aoki K. Development of a strain of spontaneousl y hypertensive rats. jpn Circ j 1963;27:282. 3. Folkow BUG, Hallback M. Pathophysiology of spontaneous hypertension in rats. In: Genestj, Kiow E, Kuchel 0 , eds. Hypertension . New York: McGraw-Hill, 1977:507. 4. Brody Mj, HaywoodjR, Toure KB. Neural mechanisms in hypertension. Ann Rev Physiol 1980;42:441. 5. Trippodo NC, Frohlich ED. Similarities of genetic (spontaneous) hypertension-man and rat. Cir Res 1981 ; 48(3) :309. 6. McCarty R, Kopin IJ. Pregnancy: its effects on blood pressure, heart rate and sympatho-adrenal activity in spontaneously hypertensive rats. Proc Soc Exp Bioi Med 1978; 158:242. 7. Yamada N, et al. Hypertensive effects on pregnancy in spontaneousl y hypertensive rat and spontaneously hypertensive rat-stroke prone. Int j Bioi Res Preg 1981 ;2(2) :80. 8. Aoi W, et al. Antihypertensive effect of pregnancy in spontaneously hypertensive rats. Proc Soc Exp Bioi Med 1976;153(1):13. 9. Lundgren Y, Karlsson K, Ljungblad U. Circulatory changes during pregnancy in spontaneously and renal hypertensive rats. Clin Sci 1979;57:337s. 10. Bunag RD, Riley E. Simultaneous measurements in awake rats of drug-induced changes in carotid and tailcuff systolic pressures. j Appl Physiol 1974;36:621. 11. Frohlich ED, Pfeffer jM, Pfeffer MA. Validity of an indirect tail-cuff method for determining systolic arterial pressure in unanesthetized normotensive and spontaneously hypertensive rats. j Lab Clin Med 1971;78(6):957. 12. Zar jH, ed. Biostatistical analysis. Prentice-Hall: Englewood Cliffs, New jersey, 1974:261. 13. Goldblatt H, KahnjR, Honzal RF. Studies on experimental hypertension. IX. The effect on blood pressure of constriction of the abdominal aorta above and below the site of origin of both main renal arteries. j Exp Med 1939;69:649. 14. Page EW, Patton HS, Ogden E. The effect of pregnancy on experimental hypertension. AM j OBSTET GYNECOL 1941 ;41:53. 15. Page EW, Ogden E. Endocrine influences upon the blood pressure of normal and hypertensive rats. AM j OBSTET GYNECOL 1947;47:150. 16. Page EW, Glendening MG. Influence of pregnancy upon hypertension induced in rats by sodium chloride and desoxycorticosterone. Proc Soc Exp Bioi Med 1953;82: 466.