Effects of short-term thermal stress on plasma catecholamine concentrations and plasma renin activity in pregnant and nonpregnant women K. Kalevi Vaha-Eskeli, MD,. Risto U. Erkkola, MD,. Mika Scheinin, MD; and Asko Seppanen, MDc
Turku, Finland OBJECTIVE: The purpose of this study was to find out whether the effects of thermal stress on plasma catecholamines and plasma renin activity are altered by pregnancy. STUDY DESIGN: Nonpregnant (n = 15) and pregnant (13 to 14 weeks, n = 23; 36 to 37 weeks, n = 23) healthy women we~e examined. Samples were taken before the exposure, in the heat chamber (700 G), and during the recovery. The BMDP statistical software was used. RESULTS: The concentration of epinephrine in plasma increased significantly only in the nonpregnant group. The epinephrine concentrations in the pregnancy groups increased most in the subjects who felt discomfort during or after the exposure. Plasma norepinephrine concentrations increased without significant differences between the groups. Plasma renin activity increased in each group, but the increases were significantly lower in the pregnancy groups as compared with the nonpregnant group. CONCLUSIONS: Pregnancy does not alter the thermally induced increase in norepinephrine release. The adrenaline response is mostly dependent on the mental discomfort. The plasma renin activity response to thermal stress is blunted during pregnancy. (AM J OBSTET GVNECOL 1992;167:785-9.)
Key words: Thermal stress, catecholamines, plasma renin activity, pregnancy The responses of plasma epinephrine, norepinephrine, and renin activity to thermal stress have been studied in nonpregnant subjects, mostly male, and sauna and other types of exposure to elevated temperatures are known to cause sympathoadrenal activation and increased release of renin.' Information on the responses of these hormones to thermal stress during pregnancy is lacking. In Finland, traditionally up to 90% of pregnant women regularly visit the sauna until the expected time of delivery/ and sauna bathing has gained popularity in many other countries. The hemodynamic situation and its hormonal control are altered by pregnancy; plasma renin activity, for example, is increased.' The catecholamines and the renin-aldosterone-angiotensin system are of special interest during pregnancy because they are crucial in the control of uterine blood flow and uterine contractions. The purpose of this study was to find out whether the reFrom the Departments of Obstetrics and Gynecologya and Pharmacology,' University of Turku, and the Rehabilitation Research Center of the Social Security Institution.' Supported by the Foundation of the University of Turku, the Paulo Foundation, and the Yrjo Jahnsson Foundation. Received for publication December 4, 1991; revised December 6, 1991; accepted February 13, 1992. Reprint requests: Kalevi Viihii-Eskeli, MD, University Central Hospital of Turku, Department of Obstetrics and Gynecology, Kiinamyllynkatu 2-4, SF-20 520 Turku, Finland. 611 /37261
sponsiveness to thermal stress equivalent to the exposure during ordinary sauna bathing is altered by pregnancy as compared with the nonpregnant state.
Subjects and methods The study population consisted of 47 volunteer subjects who gave written informed consent. The study was approved by the ethical committees of the hospital and the university. The subjects are characterized in Table I. In the nonpregnant control group seven of the subjects were in the preovulatory phase, four in the postovulatory phase, and four in the menstrual phase of the menstrual cycle. Fourteen of the subjects in group 2 were the same as in group 3 and were thus examined twice, first in early and then again in late pregnancy. All participants were in good health, they were nonsmokers, and none received medication, except oral iron supplementation in the pregnancy groups. All pregnancies were uneventful. Ultrasonography had been performed at least once during the first trimester and was repeated 1 hour before the thermal stress. All participants were accustomed to sauna bathing in their normal lives. All tests were performed at the thermal laboratory of the Rehabilitation Research Center of the Social Security Institution of Turku. All participants had a light breakfast in the morning and had been fasting for 2 hours before the exposure. All participants had similar
785
786 Vaha-Eskeli et al.
September 1992 Am J Obstet Gynecol
Table I. Physical characteristics of the subjects Gestational length Study [!;Youp Group 1 Group 2 Group 3
15 23 23
(wk)
Age (yr)
Height (em)
Weight (kg)
13.6 (13-14) 36.7 (36-37)
26 (22-30) 27 (21-35) 28 (21-35)
166 (150-178) 167 (157-170) 165 (157-176)
60 (49-74) 60 (48-80) 68 (56-100)
Values are means and ranges.
diet restrictions. They were requested not to eat salty food, fruit, cheese, peanuts, or sweets or to drink coffee, tea, or chocolate on the morning of the experiment day. Ingestion of fluids was not allowed during the experiment. The exposures were carried out at about noon. The experiment was started with a 20-minute rest at room temperature (21° to 23° C), followed by 20 minutes of thermal stress at 70° C and 15% relative humidity and ended by a 45-minute recovery period at room temperature (21° to 23° C). The temperature of the thermal chamber (Iltek OY, Finland, Landis-Gyr, Switzerland) could be adjusted with an accuracy of ± 1° C and the humidity with an accuracy of ± 3%. The temperature was constant throughout the chamber, because hot air was introduced by laminar flow from beneath the floor; the velocity of the air flow was 0.1 m/sec. During the experiment the subjects, who were dressed in bikinis, sat in wheelchairs and were not allowed to move around. Ten minutes before the experiment a polyethylene cannula was inserted into a forearm vein. Blood samples were obtained by careful aspiration through the cannula without stasis to a plastic tube with dipotassium ethylenediametetraacetic acid, chilled in ice, and immediately centrifuged at + 4° C and stored at - 70° C until analyzed. Samples were taken at the end of the resting period 5 minutes before the exposure (the 15minute sample), 10 minutes from the beginning of the thermal stress (the 30-minute sample), at the end of the stress (the 40-minute sample), 20 minutes after the stress (the 60-minute sample), and finally 45 minutes after the stress (the 85-minute sample). Catecholamine determinations were carried out with high-performance liquid chromatography with electrochemical detection, with intraassay coefficients of variation of about 10% at 0.1 nmollL and about 2% at and above 0.5 nmollL, and interassay coefficients of variation of 3% for norepinephrine (1.6 nmollL, n = 10) and 10% for epinephrine (0.2 nmollL, n = 10).4 The plasma renin activity determinations were made by radioimmunoassay (Pharmacia Diagnostics AB, Sweden) with an intraassay coefficient of variation 8.9% and an interassay coefficient of variation 5.4% at 2.0 to 3.3 angiotensin I ng/(ml x hr).
The BMDP statistical software was used. The significance of differences identified in the analyses of variance for repeated measures was further tested with the unpaired or paired t test with the Bonferroni correction. Linear regression analysis was used to generate Pearson's product-moment correlation coefficients. Logarithmic transformations were used if the distributions were not normal. Because 14 of the subjects in group 2 were the same as in group 3, the results of these subjects are included in the statistical analyses only where they serve as their own controls in comparisons between the two pregnancy groups. The results of all subjects are included in mean values and comparisons between the different time points within the group. Results
In group 2 six subjects and in group 3 five subjects felt discomfort during the recovery period. Of these subjects four in group 2 and three in group 3 wanted to interrupt the experiment 15 minutes before the end of the recovery period but spent the planned time, 20 minutes, in the heat chamber. Their results until the point of interruption did not differ significantly from the others in the respective groups and are included in the total results. The number of subjects carrying out the entire study, i.e., up until 85 minutes, was thus 19 in group 2 and 20 in group 3. Plasma catecholamines. The results are shown in Table II. The preexposure values of neither of the catecholamines differed significantly between the groups. Nor were there differences within group 1 between the different phases of the menstrual cycle. The subjects who felt discomfort (groups 2 and 3) had higher plasma epinephrine levels than those who tolerated the stress well; this increase was, however, statistically significant only in group 3 at the end of the experiment at 85 minutes (p < 0.05). The plasma concentration of epinephrine increased significantly only in group 1 from 15 to 30 minutes by 68% (p < 0.001) and from 15 to 60 minutes by 28% (p < 0.05). Plasma concentration of norepinephrine increased significantly by 36% (p < 0.05) in group 1 from 15 to 40 minutes. In group 2 plasma norepinephrine increased from 15 to 40 minutes by 24% (p < 0.05) and decreased thereafter from 40 to 60 minutes
Thermal stress and pregnancy 787
Volume 167 Number 3
Table II. Plasma epinephrine and norepinephrine concentrations (nanomoles per liter) before thermal stress (15 minutes), 10 minutes after beginning of heat administration (30 minutes), at end of stress (40 minutes), during recovery (60 minutes), and at end of recovery (85 minutes) Heat stress Group
2
3
Discomfort
15 min prestress
I
30 min
Recovery 40 min
60 min
1
85 min
0.25 ± 0.11 (15) 0.42 ± 0.23 (15)* 0.36 ± 0.21 (15) 0.32 ± 0.16 (15)t 0.28 ± 0.13 (15) 2.35 ± 1.04 (15) 2.60 ± 0.96 (15) 3.19 ± 1.31 (15)t 2.40 ± 1.04 (15) 2.17 ± 0.52 (15)
Epinephrine Norepinephrine Epinephrine Epinephrine Norepinephrine
+
0.34 ± 0.19 (17) 0.42 ± 0.20 (17) 0.62 ± 0.82 (6) 0.70 ± 0.56 (6) 2.75 ± 0.81 (23) 2.87 ± 0.86 (23)
Epinephrine Epinephrine Norepinephrine
+
0.20 ± 0.07 (18) 0.19 ± 0.09 (18) 0.21 ± 0.09 (18) 0.25 ± 0.16 (18) 0.26 ± 0.12 (18) 0.76 ± 0.83 (5) 0.54 ± 0.51 (5) 0.64 ± 0.28 (2) 0.17 ± 0.10 (5) 0.22 ± 0.16 (5) 1.74 ± 0.54 (23) 2.38 ± 0.71 (23)* 2.81 ± 1.04 (23)* 2.18 ± 0.71 (23)* 2.08 ± 0.59 (20)*
0.45 ± 0.21 (17) 0.47 ± 0.28 (17) 0.54 ± 0.31 (6) 0.43 ± 0.16 (6) 3.40 ± 1.02 (23)t 2.53 ± 0.62 (23)
0.42 ± 0.28 (17) 0.37 ± 0.68 (2) 2.56 ± 0.63 (19)
Number of subjects is in parentheses. -, No discomfort during or after thermal stress; +, discomfort after thermal stress. Values are means ± SD. *p < 0.001, increase in comparisons within group from prestress level (15 minutes). tp < 0.05, increase in comparisons within group from prestress level (15 minutes). *p < 0.005, increase in comparisons within group from prestress level (15 minutes).
(P < 0.001). In group 3 norepinephrine increased more rapidly: already by 30 minutes was there an increase of 37% and by 40 minutes of 61 % (P < 0.001); norepinephrine remained elevated until the end of the experiment. There were no differences between the groups in the proportional changes of either of the catecholamines. Plasma renin activity. Plasma renin aCtiVIty mcreased in all groups (Table III). The starting values were highest in the pregnancy groups as compared with the nonpregnant group (p < 0.001), but the difference between the pregnancy groups was not significant. The peak levels were reached at the end of the thermal stress in group 1 (184%, P < 0.001) and in group 2 (61%, P< 0.001), whereas in group 3 the peak was reached 20 minutes after the end of the stress (26%, p < 0.001). The values remained elevated in each group until the end of the recovery period but were significantly raised only in groups 1 and 2 (P < 0.05). The lower the prestress values were, the higher the percentage increase was at 40 and 60 minutes when all subjects were pooled together (r = - 0.26, P < 0.001). The percentage increase was significantly more pronounced in group 1 as compared with groups 2 and 3 at 40 (P < 0.001) and 60 minutes (p < 0.005). In group 2 there was also a significantly higher increase in plasma renin activity by 40 minutes as compared with group 3 (p < 0.005). The absolute prestress norepinephrine and plasma renin activity values correlated positively in group 1 (r = 0.68, P< 0.005) but not in the other groups nor during the thermal stress. There was also a positive correlation between the percentage changes of plasma renin activity and norepinephrine from 15 to 40 min-
utes in group 1 (r = 0.26, not significant) and a negative correlation in groups 2 (r = - 0.48, P < 0.05) and 3 (r = - 0.13, not significant).
Comment The present study is the first to examine the responses of plasma catecholamines and plasma renin activity to thermal stress during pregnancy as compared with the nonpregnant state. The thermal stress that was used may appear rather moderate. Longer duration of the exposure and higher temperatures might have caused more pronounced changes in the hormones that were studied. However, most of the subjects informed us that they would not have tolerated the stress much longer. Catecholamines. The starting values of the catecholamines did not differ among the groups. Previous reports on the effect of pregnancy on the concentration of catecholamines in plasma are controversial. Venous plasma levels of norepinephrine and epinephrine have been reported to be unchanged during pregnancy.'· 6 According to another study, the venous plasma norepinephrine level declines toward term,' whereas arterial plasma norepinephrine is unchanged. 8 In the same study epinephrine concentration sank as compared with the situation post partum. In a longitudinal study the levels of both catecholamines were reported to decrease with advancing gestation. 9 It has been clearly documented that the concentration of norepinephrine in blood and urine increases during thermal stress, reflecting increased sympathetic nervous activity. The magnitude of the increase depends on the duration and quality of the exposure; in healthy young men staying in sauna until exhaustion
788
Vaha-Eskeli et al.
September 1992 Am J Obstet Gyneco1
Table III. Plasma renin activity (nanograms angiotensin I per milliliter x hour) before thermal stress (IS minutes), 10 minutes after beginning of heat administration (30 minutes), at end of stress (40 minutes), during recovery (60 minutes), and at end of recovery (85 minutes) Heat stress Group
I
2
3
15 min Prestress
30 min
1.37 ± 1.09 (IS) 4.70 ± 1.69 (23) 6.72 ± 3.S1 (23)
1.69 ± I.S4 (8) S.93 ± 2.36 (23)t 6.8S ± 2.86 (23)
I
Recovery
40 min
60 min
3.89 ± 3.04 (IS)* 7.S8 ± 3.81 (23)t 7.96 ± 3.43 (23)t
2.82 ± 2.20 (IS)* 7.20 ± 3.91 (23)+ 8.S0 ± 3.34 (23)*
I
85 min
2.29 ± 1.82 (lS)t 6.70 ± 3.92 (19)§ 7.96 ± 3.17 (20)
Number of subjects is in parentheses. Values are means ± SD.
*p < 0.001, increase in comparisons within group from prestress level (IS minutes). tp < O.OS, increase in comparisons within group from prestress level (IS minutes). :f:P < O.OOS, increase in comparisons within group from prestress level (IS minutes). §p < 0.01, increase in comparisons within group from prestress level (IS minutes). the plasma norepinephrine concentration increased threefold. lo The level of plasma norepinephrine remained elevated for >45 minutes after sauna bath and cooling in a swimming pool. 11 The reports concerning the changes in blood concentrations of epinephrine are conflicting. In most studies the levels in peripheral venous plasma have remained unchanged or increased slightly during thermal stress. 1 One explanation for the increase in epinephrine levels in some sauna studies may be that either the subjects have not been familiar to sauna bathing I2 or that some other conditions of the experiment have raised epinephrine because of nonspecific stress factors. The subjects in the present study were regular sauna bathers. In spite of this, the epinephrine concentration increased significantly in nonpregnant subjects who all tolerated the stress well. Fear of study procedures such as blood sampling may have affected the reaction in the nonpregnant and the early pregnancy group. All subjects in groups 1 and 2 participated in the test for the first time, whereas 14 of the subjects in group 3 were already familiar with the experimental setting and procedure, which may have influenced the results. The starting values of epinephrine in the subjects in group 2 who later felt discomfort were almost twofold as compared with the rest of the group. This phenomenon was not, however, seen in group 3, where the values were almost identical to the end of the stress. The subjects sat in a wheelchair during the whole experiment to eliminate the effect of change in posture on the results, whereas in most previous studies the subjects have walked to and from the thermal chamber. Plasma norepinephrine increases immediately after standing up, although the effect of standing on plasma epinephrine is minimal. Is Thus the rather small increases in plasma norepinephrine concentration in the present study as compared with sauna studies may be merely due to differences in study design; our purpose was to examine the effects of a pure thermal stress.
The proportional changes of either norepinephrine or epinephrine among the groups were not significantly different, in spite of increases in plasma norepinephrine of 61 % in group 3 as compared with 24% in group 1. According to a previous report the arterial plasma epinephrine levels are lower during pregnancy as compared with the nonpregnant state after a cold pressor test and tilting, whereas the concentration of plasma norepinephrine after isometric hand gripping, cold pressor test, and mental stress are unchanged. The vascular responses to norepinephrine and epinephrine infusions are blunted during pregnancy.s The hemodynamic reactions to these provocation tests are, however, different from thermal stress and comparisons may be irrelevant. Norepinephrine is known to stimulate and epinephrine to depress uterine activity in pregnant women.I4 As previously reported of the same subjects as in group 3, only five of 23 of the subjects had mild uterine contractions. I5 Infusions of both norepinephrine and epinephrine I6 decrease uterine and placental blood flow in sheep. There is some evidence that the response of the uterine vascular bed to adrenergic stimuli is blunted during pregnancy. 16. 17 During or after thermal stress there are no major disturbances in fetal heart rate reactivity,l5 which could indicate adverse changes in uteroplacental blood flow. No significant changes in either uterine or umbilical artery blood flow, measured by Doppler ultrasonography, have been recorded during or after thermal stress in conditions similar to the present study. IS Plasma renin activity. Plasma renin activity increases early in pregnancy.3 The rise is twofold as compared with the nonpregnant level by 12 weeks of pregnancy, and elevated levels until termS have been reported. In another study I3-fold levels at 10 weeks of pregnancy have been found. 19 Our results on the prestress plasma renin activity concentration, a threefold increase in
Volume 167 Number 3
early pregnancy and a fivefold increase in late pregnancy, falls between these reported changes. Many stressful situations increase renin secretion, e.g., thermal stress.' The reason for the increased plasma renin activity may be the increased sympathetic activity.20 This was the case in this study only in group I (not significant) but not in groups 2 or 3. Loss of sodium also increases plasma renin activity levels during prolonged thermal stress. 2' The heat-induced decrement in renal and perhaps also hepatic blood flow may contribute to the rise in plasma renin activity. Plasma renin activity may increase 50% in response to an upright posture!2 but this source of error was eliminated in the present study. The rise of plasma renin activity was lowest and most markedly delayed in the late pregnancy group (Table III). We can only speculate the explanation for this observation. Sympathetic activation as measured by norepinephrine concentration did not differ significantly between the groups. Renal blood flow, which affects renin release, was not measured in the present study. Thus proportional changes affected by thermal stress in glomerular filtration may be different during pregnancy. The catecholamine response to thermal stress is characterized by activation of the sympathetic nervous system and a slight increase in epinephrine concentration. This reflects, in part, the effect of mental stress in the experimental situation and not pure thermal stress. The response of norepinephrine and epinephrine is not altered by pregnancy. Plasma renin activity increases in thermal stress, but perhaps because change in posture was restricted in the present study the rise was less marked than reported previously. The plasma renin activity response is blunted and delayed in late pregnancy, but the reason for this remains to be resolved. We thank Robert Paul, MD, for revising the manuscript. REFERENCES I. Kukkonen-Harjula K, Kauppinen K. How the sauna af-
fects the endocrine system. Ann Clin Res 1988;20:262-6. 2. Vaha-Eskeli K, Erkkola R. The sauna and pregnancy. Ann Clin Res 1988;20:279-82. 3. Tulchinsky D, Ryan Kj. In: Maternal-fetal endocrinology. Philadelphia: WB Saunders, 1980. 4. Schein in M, Koulu M, Laurikainen E, Allonen H. Hypokalaemia and other non-bronchial effects of inhaled fenoterol and salbutamol: a placebo-controlled dose-response study in healthy volunteers. Br j Clin Pharmacol 1987;24:645-53. 5. Lederman RP, McCann DS, Work B, Huber MJ. Endogenous plasma epinephrine and norepinephrine in lasttrimester pregnancy and labor. AM j OBSTET GYNECOL 1977;129:5-8.
Thermal stress and pregnancy
789
6. Pedersen EB, Rasmussen AB, Christensen Nj, et al. Plasma noradrenaline and adrenaline in pre-eclampsia, essential hypertension in pregnancy and normotensive pregnant control subjects. Acta Endocrinol 1982;99:594600. 7. Tunbridge RDG, Donnai P. Plasma noradrenaline in normal pregnancy and in hypertension of late pregnancy. Br j Obstet Gynaecol 1981 ;88: 105-8. 8. Nisell H. Studies of cardiovascular and sympatho-adrenal function in normal pregnancy and pregnancy-induced hypertension [Dissertation]. Stockholm: Karolinska Institute, 1985. 9. Natrajan PG, McGarrigle HHG, Lawrence DM, Lachelin GCL. Plasma noradrenaline and adrenaline levels in normal pregnancy and in pregnancy-induced hypertension. Br j Obstet Gynaecol 1982;89: 1041-5. 10. Kukkonen-Harjula K, Oja P, Laustiola K, et al. Hemodynamic and hormonal responses to three heat exposures in Finnish sauna bathing. Eur j Appl PhysioI1989;58:54350. 11. Hussi E, Sonck T, Poso H, Remes j, Eisalo A, janne J. Plasma catecholamines in Finnish sauna. Ann Clin Res 1977;9:301-4. 12. Tatar P, Vigas M, jurovicova j, jezova D, Strec V, Palat M. Impaired glucose utilization in man during acute exposure to environmental heat. Endocrinol Exp 1985;19:227-81. 13. Robertson D, johnson GA, Robertson RM, Nies AS, Schand DG, Oates jA. Comparative assessment of stimuli that release neuronal and adrenomedullary catecholamines in man. Circulation 1979;59:637-43. 14. Zuspan FP, Cibils LA, Pose SV. Myometrial and cardiovascular responses to alterations in plasma epinephrine and norepinephrine. AMj OBSTET GYNECOL 1962;84:84151. 15. Vaha-Eskeli K, Erkkola R. The effect of short-term heat stress on uterine contractility, fetal heart rate and fetal movements at late pregnancy. Eur j Obstet Gynecol Reprod Bioi 1990;38:9-14. 16. Greiss FC. Differential reactivity of the myoendometrial and placental vasculatures: adrenergic responses. AM j OBSTET GYNECOL 1972;112:20-30. 17. Erkkola R, Tabsh K, Ushioda E, Nuwayhid B, Brinkman CR, Assali NS. Responses of the pelvic vascular bed to intra-arterial stimulation of [3-adrenergic and cholinergic receptors in pregnant and non-pregnant sheep. AMj OBSTET GYNECOL 1981; 141 :599-607. 18. Vaha-Eskeli K, Pirhonen j, Seppanen A, Erkkola R. Doppler How measurement of uterine and umbilical arteries in heat stress during late pregnancy. Amj Perinatol 1991 ;8:385-9. 19. Lammintausta R, Erkkola R. Renin-angiotensin-aldosterone system and sodium in normal pregnancy: a longitudinal study. Acta Obstet Gynecol Scand 1977;56:221-5. 20. Berlyne GM, Finberg jPM, Yoran C. The effect of [3adrenoreceptor blockade in body temperature and plasma renin activity in heat-exposed man. Br j Clin Pharmacol 1974; I :307-12. 21. Lammintausta R, Syvalahti E, Pekkarinen A. Change in hormones reHecting sympathetic activity in the Finnish sauna. Ann Clin Res 1976;8:266-71. 22. Ylikorkala 0, Haapalahti J. Plasma renin activity (PRA) and some Huid equilibrium values in recumbent and upright positions in non-pregnant women and in normal and complicated term pregnancy. Ann Chir Gyn Fenn 1974;63:39.
Turku, Finland OBJECTIVE: The purpose of this study was to find out whether the effects of thermal stress on plasma catecholamines and plasma renin activity are altered by pregnancy. STUDY DESIGN: Nonpregnant (n = 15) and pregnant (13 to 14 weeks, n = 23; 36 to 37 weeks, n = 23) healthy women we~e examined. Samples were taken before the exposure, in the heat chamber (700 G), and during the recovery. The BMDP statistical software was used. RESULTS: The concentration of epinephrine in plasma increased significantly only in the nonpregnant group. The epinephrine concentrations in the pregnancy groups increased most in the subjects who felt discomfort during or after the exposure. Plasma norepinephrine concentrations increased without significant differences between the groups. Plasma renin activity increased in each group, but the increases were significantly lower in the pregnancy groups as compared with the nonpregnant group. CONCLUSIONS: Pregnancy does not alter the thermally induced increase in norepinephrine release. The adrenaline response is mostly dependent on the mental discomfort. The plasma renin activity response to thermal stress is blunted during pregnancy. (AM J OBSTET GVNECOL 1992;167:785-9.)
Key words: Thermal stress, catecholamines, plasma renin activity, pregnancy The responses of plasma epinephrine, norepinephrine, and renin activity to thermal stress have been studied in nonpregnant subjects, mostly male, and sauna and other types of exposure to elevated temperatures are known to cause sympathoadrenal activation and increased release of renin.' Information on the responses of these hormones to thermal stress during pregnancy is lacking. In Finland, traditionally up to 90% of pregnant women regularly visit the sauna until the expected time of delivery/ and sauna bathing has gained popularity in many other countries. The hemodynamic situation and its hormonal control are altered by pregnancy; plasma renin activity, for example, is increased.' The catecholamines and the renin-aldosterone-angiotensin system are of special interest during pregnancy because they are crucial in the control of uterine blood flow and uterine contractions. The purpose of this study was to find out whether the reFrom the Departments of Obstetrics and Gynecologya and Pharmacology,' University of Turku, and the Rehabilitation Research Center of the Social Security Institution.' Supported by the Foundation of the University of Turku, the Paulo Foundation, and the Yrjo Jahnsson Foundation. Received for publication December 4, 1991; revised December 6, 1991; accepted February 13, 1992. Reprint requests: Kalevi Viihii-Eskeli, MD, University Central Hospital of Turku, Department of Obstetrics and Gynecology, Kiinamyllynkatu 2-4, SF-20 520 Turku, Finland. 611 /37261
sponsiveness to thermal stress equivalent to the exposure during ordinary sauna bathing is altered by pregnancy as compared with the nonpregnant state.
Subjects and methods The study population consisted of 47 volunteer subjects who gave written informed consent. The study was approved by the ethical committees of the hospital and the university. The subjects are characterized in Table I. In the nonpregnant control group seven of the subjects were in the preovulatory phase, four in the postovulatory phase, and four in the menstrual phase of the menstrual cycle. Fourteen of the subjects in group 2 were the same as in group 3 and were thus examined twice, first in early and then again in late pregnancy. All participants were in good health, they were nonsmokers, and none received medication, except oral iron supplementation in the pregnancy groups. All pregnancies were uneventful. Ultrasonography had been performed at least once during the first trimester and was repeated 1 hour before the thermal stress. All participants were accustomed to sauna bathing in their normal lives. All tests were performed at the thermal laboratory of the Rehabilitation Research Center of the Social Security Institution of Turku. All participants had a light breakfast in the morning and had been fasting for 2 hours before the exposure. All participants had similar
785
786 Vaha-Eskeli et al.
September 1992 Am J Obstet Gynecol
Table I. Physical characteristics of the subjects Gestational length Study [!;Youp Group 1 Group 2 Group 3
15 23 23
(wk)
Age (yr)
Height (em)
Weight (kg)
13.6 (13-14) 36.7 (36-37)
26 (22-30) 27 (21-35) 28 (21-35)
166 (150-178) 167 (157-170) 165 (157-176)
60 (49-74) 60 (48-80) 68 (56-100)
Values are means and ranges.
diet restrictions. They were requested not to eat salty food, fruit, cheese, peanuts, or sweets or to drink coffee, tea, or chocolate on the morning of the experiment day. Ingestion of fluids was not allowed during the experiment. The exposures were carried out at about noon. The experiment was started with a 20-minute rest at room temperature (21° to 23° C), followed by 20 minutes of thermal stress at 70° C and 15% relative humidity and ended by a 45-minute recovery period at room temperature (21° to 23° C). The temperature of the thermal chamber (Iltek OY, Finland, Landis-Gyr, Switzerland) could be adjusted with an accuracy of ± 1° C and the humidity with an accuracy of ± 3%. The temperature was constant throughout the chamber, because hot air was introduced by laminar flow from beneath the floor; the velocity of the air flow was 0.1 m/sec. During the experiment the subjects, who were dressed in bikinis, sat in wheelchairs and were not allowed to move around. Ten minutes before the experiment a polyethylene cannula was inserted into a forearm vein. Blood samples were obtained by careful aspiration through the cannula without stasis to a plastic tube with dipotassium ethylenediametetraacetic acid, chilled in ice, and immediately centrifuged at + 4° C and stored at - 70° C until analyzed. Samples were taken at the end of the resting period 5 minutes before the exposure (the 15minute sample), 10 minutes from the beginning of the thermal stress (the 30-minute sample), at the end of the stress (the 40-minute sample), 20 minutes after the stress (the 60-minute sample), and finally 45 minutes after the stress (the 85-minute sample). Catecholamine determinations were carried out with high-performance liquid chromatography with electrochemical detection, with intraassay coefficients of variation of about 10% at 0.1 nmollL and about 2% at and above 0.5 nmollL, and interassay coefficients of variation of 3% for norepinephrine (1.6 nmollL, n = 10) and 10% for epinephrine (0.2 nmollL, n = 10).4 The plasma renin activity determinations were made by radioimmunoassay (Pharmacia Diagnostics AB, Sweden) with an intraassay coefficient of variation 8.9% and an interassay coefficient of variation 5.4% at 2.0 to 3.3 angiotensin I ng/(ml x hr).
The BMDP statistical software was used. The significance of differences identified in the analyses of variance for repeated measures was further tested with the unpaired or paired t test with the Bonferroni correction. Linear regression analysis was used to generate Pearson's product-moment correlation coefficients. Logarithmic transformations were used if the distributions were not normal. Because 14 of the subjects in group 2 were the same as in group 3, the results of these subjects are included in the statistical analyses only where they serve as their own controls in comparisons between the two pregnancy groups. The results of all subjects are included in mean values and comparisons between the different time points within the group. Results
In group 2 six subjects and in group 3 five subjects felt discomfort during the recovery period. Of these subjects four in group 2 and three in group 3 wanted to interrupt the experiment 15 minutes before the end of the recovery period but spent the planned time, 20 minutes, in the heat chamber. Their results until the point of interruption did not differ significantly from the others in the respective groups and are included in the total results. The number of subjects carrying out the entire study, i.e., up until 85 minutes, was thus 19 in group 2 and 20 in group 3. Plasma catecholamines. The results are shown in Table II. The preexposure values of neither of the catecholamines differed significantly between the groups. Nor were there differences within group 1 between the different phases of the menstrual cycle. The subjects who felt discomfort (groups 2 and 3) had higher plasma epinephrine levels than those who tolerated the stress well; this increase was, however, statistically significant only in group 3 at the end of the experiment at 85 minutes (p < 0.05). The plasma concentration of epinephrine increased significantly only in group 1 from 15 to 30 minutes by 68% (p < 0.001) and from 15 to 60 minutes by 28% (p < 0.05). Plasma concentration of norepinephrine increased significantly by 36% (p < 0.05) in group 1 from 15 to 40 minutes. In group 2 plasma norepinephrine increased from 15 to 40 minutes by 24% (p < 0.05) and decreased thereafter from 40 to 60 minutes
Thermal stress and pregnancy 787
Volume 167 Number 3
Table II. Plasma epinephrine and norepinephrine concentrations (nanomoles per liter) before thermal stress (15 minutes), 10 minutes after beginning of heat administration (30 minutes), at end of stress (40 minutes), during recovery (60 minutes), and at end of recovery (85 minutes) Heat stress Group
2
3
Discomfort
15 min prestress
I
30 min
Recovery 40 min
60 min
1
85 min
0.25 ± 0.11 (15) 0.42 ± 0.23 (15)* 0.36 ± 0.21 (15) 0.32 ± 0.16 (15)t 0.28 ± 0.13 (15) 2.35 ± 1.04 (15) 2.60 ± 0.96 (15) 3.19 ± 1.31 (15)t 2.40 ± 1.04 (15) 2.17 ± 0.52 (15)
Epinephrine Norepinephrine Epinephrine Epinephrine Norepinephrine
+
0.34 ± 0.19 (17) 0.42 ± 0.20 (17) 0.62 ± 0.82 (6) 0.70 ± 0.56 (6) 2.75 ± 0.81 (23) 2.87 ± 0.86 (23)
Epinephrine Epinephrine Norepinephrine
+
0.20 ± 0.07 (18) 0.19 ± 0.09 (18) 0.21 ± 0.09 (18) 0.25 ± 0.16 (18) 0.26 ± 0.12 (18) 0.76 ± 0.83 (5) 0.54 ± 0.51 (5) 0.64 ± 0.28 (2) 0.17 ± 0.10 (5) 0.22 ± 0.16 (5) 1.74 ± 0.54 (23) 2.38 ± 0.71 (23)* 2.81 ± 1.04 (23)* 2.18 ± 0.71 (23)* 2.08 ± 0.59 (20)*
0.45 ± 0.21 (17) 0.47 ± 0.28 (17) 0.54 ± 0.31 (6) 0.43 ± 0.16 (6) 3.40 ± 1.02 (23)t 2.53 ± 0.62 (23)
0.42 ± 0.28 (17) 0.37 ± 0.68 (2) 2.56 ± 0.63 (19)
Number of subjects is in parentheses. -, No discomfort during or after thermal stress; +, discomfort after thermal stress. Values are means ± SD. *p < 0.001, increase in comparisons within group from prestress level (15 minutes). tp < 0.05, increase in comparisons within group from prestress level (15 minutes). *p < 0.005, increase in comparisons within group from prestress level (15 minutes).
(P < 0.001). In group 3 norepinephrine increased more rapidly: already by 30 minutes was there an increase of 37% and by 40 minutes of 61 % (P < 0.001); norepinephrine remained elevated until the end of the experiment. There were no differences between the groups in the proportional changes of either of the catecholamines. Plasma renin activity. Plasma renin aCtiVIty mcreased in all groups (Table III). The starting values were highest in the pregnancy groups as compared with the nonpregnant group (p < 0.001), but the difference between the pregnancy groups was not significant. The peak levels were reached at the end of the thermal stress in group 1 (184%, P < 0.001) and in group 2 (61%, P< 0.001), whereas in group 3 the peak was reached 20 minutes after the end of the stress (26%, p < 0.001). The values remained elevated in each group until the end of the recovery period but were significantly raised only in groups 1 and 2 (P < 0.05). The lower the prestress values were, the higher the percentage increase was at 40 and 60 minutes when all subjects were pooled together (r = - 0.26, P < 0.001). The percentage increase was significantly more pronounced in group 1 as compared with groups 2 and 3 at 40 (P < 0.001) and 60 minutes (p < 0.005). In group 2 there was also a significantly higher increase in plasma renin activity by 40 minutes as compared with group 3 (p < 0.005). The absolute prestress norepinephrine and plasma renin activity values correlated positively in group 1 (r = 0.68, P< 0.005) but not in the other groups nor during the thermal stress. There was also a positive correlation between the percentage changes of plasma renin activity and norepinephrine from 15 to 40 min-
utes in group 1 (r = 0.26, not significant) and a negative correlation in groups 2 (r = - 0.48, P < 0.05) and 3 (r = - 0.13, not significant).
Comment The present study is the first to examine the responses of plasma catecholamines and plasma renin activity to thermal stress during pregnancy as compared with the nonpregnant state. The thermal stress that was used may appear rather moderate. Longer duration of the exposure and higher temperatures might have caused more pronounced changes in the hormones that were studied. However, most of the subjects informed us that they would not have tolerated the stress much longer. Catecholamines. The starting values of the catecholamines did not differ among the groups. Previous reports on the effect of pregnancy on the concentration of catecholamines in plasma are controversial. Venous plasma levels of norepinephrine and epinephrine have been reported to be unchanged during pregnancy.'· 6 According to another study, the venous plasma norepinephrine level declines toward term,' whereas arterial plasma norepinephrine is unchanged. 8 In the same study epinephrine concentration sank as compared with the situation post partum. In a longitudinal study the levels of both catecholamines were reported to decrease with advancing gestation. 9 It has been clearly documented that the concentration of norepinephrine in blood and urine increases during thermal stress, reflecting increased sympathetic nervous activity. The magnitude of the increase depends on the duration and quality of the exposure; in healthy young men staying in sauna until exhaustion
788
Vaha-Eskeli et al.
September 1992 Am J Obstet Gyneco1
Table III. Plasma renin activity (nanograms angiotensin I per milliliter x hour) before thermal stress (IS minutes), 10 minutes after beginning of heat administration (30 minutes), at end of stress (40 minutes), during recovery (60 minutes), and at end of recovery (85 minutes) Heat stress Group
I
2
3
15 min Prestress
30 min
1.37 ± 1.09 (IS) 4.70 ± 1.69 (23) 6.72 ± 3.S1 (23)
1.69 ± I.S4 (8) S.93 ± 2.36 (23)t 6.8S ± 2.86 (23)
I
Recovery
40 min
60 min
3.89 ± 3.04 (IS)* 7.S8 ± 3.81 (23)t 7.96 ± 3.43 (23)t
2.82 ± 2.20 (IS)* 7.20 ± 3.91 (23)+ 8.S0 ± 3.34 (23)*
I
85 min
2.29 ± 1.82 (lS)t 6.70 ± 3.92 (19)§ 7.96 ± 3.17 (20)
Number of subjects is in parentheses. Values are means ± SD.
*p < 0.001, increase in comparisons within group from prestress level (IS minutes). tp < O.OS, increase in comparisons within group from prestress level (IS minutes). :f:P < O.OOS, increase in comparisons within group from prestress level (IS minutes). §p < 0.01, increase in comparisons within group from prestress level (IS minutes). the plasma norepinephrine concentration increased threefold. lo The level of plasma norepinephrine remained elevated for >45 minutes after sauna bath and cooling in a swimming pool. 11 The reports concerning the changes in blood concentrations of epinephrine are conflicting. In most studies the levels in peripheral venous plasma have remained unchanged or increased slightly during thermal stress. 1 One explanation for the increase in epinephrine levels in some sauna studies may be that either the subjects have not been familiar to sauna bathing I2 or that some other conditions of the experiment have raised epinephrine because of nonspecific stress factors. The subjects in the present study were regular sauna bathers. In spite of this, the epinephrine concentration increased significantly in nonpregnant subjects who all tolerated the stress well. Fear of study procedures such as blood sampling may have affected the reaction in the nonpregnant and the early pregnancy group. All subjects in groups 1 and 2 participated in the test for the first time, whereas 14 of the subjects in group 3 were already familiar with the experimental setting and procedure, which may have influenced the results. The starting values of epinephrine in the subjects in group 2 who later felt discomfort were almost twofold as compared with the rest of the group. This phenomenon was not, however, seen in group 3, where the values were almost identical to the end of the stress. The subjects sat in a wheelchair during the whole experiment to eliminate the effect of change in posture on the results, whereas in most previous studies the subjects have walked to and from the thermal chamber. Plasma norepinephrine increases immediately after standing up, although the effect of standing on plasma epinephrine is minimal. Is Thus the rather small increases in plasma norepinephrine concentration in the present study as compared with sauna studies may be merely due to differences in study design; our purpose was to examine the effects of a pure thermal stress.
The proportional changes of either norepinephrine or epinephrine among the groups were not significantly different, in spite of increases in plasma norepinephrine of 61 % in group 3 as compared with 24% in group 1. According to a previous report the arterial plasma epinephrine levels are lower during pregnancy as compared with the nonpregnant state after a cold pressor test and tilting, whereas the concentration of plasma norepinephrine after isometric hand gripping, cold pressor test, and mental stress are unchanged. The vascular responses to norepinephrine and epinephrine infusions are blunted during pregnancy.s The hemodynamic reactions to these provocation tests are, however, different from thermal stress and comparisons may be irrelevant. Norepinephrine is known to stimulate and epinephrine to depress uterine activity in pregnant women.I4 As previously reported of the same subjects as in group 3, only five of 23 of the subjects had mild uterine contractions. I5 Infusions of both norepinephrine and epinephrine I6 decrease uterine and placental blood flow in sheep. There is some evidence that the response of the uterine vascular bed to adrenergic stimuli is blunted during pregnancy. 16. 17 During or after thermal stress there are no major disturbances in fetal heart rate reactivity,l5 which could indicate adverse changes in uteroplacental blood flow. No significant changes in either uterine or umbilical artery blood flow, measured by Doppler ultrasonography, have been recorded during or after thermal stress in conditions similar to the present study. IS Plasma renin activity. Plasma renin activity increases early in pregnancy.3 The rise is twofold as compared with the nonpregnant level by 12 weeks of pregnancy, and elevated levels until termS have been reported. In another study I3-fold levels at 10 weeks of pregnancy have been found. 19 Our results on the prestress plasma renin activity concentration, a threefold increase in
Volume 167 Number 3
early pregnancy and a fivefold increase in late pregnancy, falls between these reported changes. Many stressful situations increase renin secretion, e.g., thermal stress.' The reason for the increased plasma renin activity may be the increased sympathetic activity.20 This was the case in this study only in group I (not significant) but not in groups 2 or 3. Loss of sodium also increases plasma renin activity levels during prolonged thermal stress. 2' The heat-induced decrement in renal and perhaps also hepatic blood flow may contribute to the rise in plasma renin activity. Plasma renin activity may increase 50% in response to an upright posture!2 but this source of error was eliminated in the present study. The rise of plasma renin activity was lowest and most markedly delayed in the late pregnancy group (Table III). We can only speculate the explanation for this observation. Sympathetic activation as measured by norepinephrine concentration did not differ significantly between the groups. Renal blood flow, which affects renin release, was not measured in the present study. Thus proportional changes affected by thermal stress in glomerular filtration may be different during pregnancy. The catecholamine response to thermal stress is characterized by activation of the sympathetic nervous system and a slight increase in epinephrine concentration. This reflects, in part, the effect of mental stress in the experimental situation and not pure thermal stress. The response of norepinephrine and epinephrine is not altered by pregnancy. Plasma renin activity increases in thermal stress, but perhaps because change in posture was restricted in the present study the rise was less marked than reported previously. The plasma renin activity response is blunted and delayed in late pregnancy, but the reason for this remains to be resolved. We thank Robert Paul, MD, for revising the manuscript. REFERENCES I. Kukkonen-Harjula K, Kauppinen K. How the sauna af-
fects the endocrine system. Ann Clin Res 1988;20:262-6. 2. Vaha-Eskeli K, Erkkola R. The sauna and pregnancy. Ann Clin Res 1988;20:279-82. 3. Tulchinsky D, Ryan Kj. In: Maternal-fetal endocrinology. Philadelphia: WB Saunders, 1980. 4. Schein in M, Koulu M, Laurikainen E, Allonen H. Hypokalaemia and other non-bronchial effects of inhaled fenoterol and salbutamol: a placebo-controlled dose-response study in healthy volunteers. Br j Clin Pharmacol 1987;24:645-53. 5. Lederman RP, McCann DS, Work B, Huber MJ. Endogenous plasma epinephrine and norepinephrine in lasttrimester pregnancy and labor. AM j OBSTET GYNECOL 1977;129:5-8.
Thermal stress and pregnancy
789
6. Pedersen EB, Rasmussen AB, Christensen Nj, et al. Plasma noradrenaline and adrenaline in pre-eclampsia, essential hypertension in pregnancy and normotensive pregnant control subjects. Acta Endocrinol 1982;99:594600. 7. Tunbridge RDG, Donnai P. Plasma noradrenaline in normal pregnancy and in hypertension of late pregnancy. Br j Obstet Gynaecol 1981 ;88: 105-8. 8. Nisell H. Studies of cardiovascular and sympatho-adrenal function in normal pregnancy and pregnancy-induced hypertension [Dissertation]. Stockholm: Karolinska Institute, 1985. 9. Natrajan PG, McGarrigle HHG, Lawrence DM, Lachelin GCL. Plasma noradrenaline and adrenaline levels in normal pregnancy and in pregnancy-induced hypertension. Br j Obstet Gynaecol 1982;89: 1041-5. 10. Kukkonen-Harjula K, Oja P, Laustiola K, et al. Hemodynamic and hormonal responses to three heat exposures in Finnish sauna bathing. Eur j Appl PhysioI1989;58:54350. 11. Hussi E, Sonck T, Poso H, Remes j, Eisalo A, janne J. Plasma catecholamines in Finnish sauna. Ann Clin Res 1977;9:301-4. 12. Tatar P, Vigas M, jurovicova j, jezova D, Strec V, Palat M. Impaired glucose utilization in man during acute exposure to environmental heat. Endocrinol Exp 1985;19:227-81. 13. Robertson D, johnson GA, Robertson RM, Nies AS, Schand DG, Oates jA. Comparative assessment of stimuli that release neuronal and adrenomedullary catecholamines in man. Circulation 1979;59:637-43. 14. Zuspan FP, Cibils LA, Pose SV. Myometrial and cardiovascular responses to alterations in plasma epinephrine and norepinephrine. AMj OBSTET GYNECOL 1962;84:84151. 15. Vaha-Eskeli K, Erkkola R. The effect of short-term heat stress on uterine contractility, fetal heart rate and fetal movements at late pregnancy. Eur j Obstet Gynecol Reprod Bioi 1990;38:9-14. 16. Greiss FC. Differential reactivity of the myoendometrial and placental vasculatures: adrenergic responses. AM j OBSTET GYNECOL 1972;112:20-30. 17. Erkkola R, Tabsh K, Ushioda E, Nuwayhid B, Brinkman CR, Assali NS. Responses of the pelvic vascular bed to intra-arterial stimulation of [3-adrenergic and cholinergic receptors in pregnant and non-pregnant sheep. AMj OBSTET GYNECOL 1981; 141 :599-607. 18. Vaha-Eskeli K, Pirhonen j, Seppanen A, Erkkola R. Doppler How measurement of uterine and umbilical arteries in heat stress during late pregnancy. Amj Perinatol 1991 ;8:385-9. 19. Lammintausta R, Erkkola R. Renin-angiotensin-aldosterone system and sodium in normal pregnancy: a longitudinal study. Acta Obstet Gynecol Scand 1977;56:221-5. 20. Berlyne GM, Finberg jPM, Yoran C. The effect of [3adrenoreceptor blockade in body temperature and plasma renin activity in heat-exposed man. Br j Clin Pharmacol 1974; I :307-12. 21. Lammintausta R, Syvalahti E, Pekkarinen A. Change in hormones reHecting sympathetic activity in the Finnish sauna. Ann Clin Res 1976;8:266-71. 22. Ylikorkala 0, Haapalahti J. Plasma renin activity (PRA) and some Huid equilibrium values in recumbent and upright positions in non-pregnant women and in normal and complicated term pregnancy. Ann Chir Gyn Fenn 1974;63:39.
