Decreased cardiac parasympathetic nerve activity of pregnant women during foot baths

Abstract

Aim:  This study aimed to examine the effect of foot baths on the cardiac parasympathetic outflow of five pregnant (31–32 weeks; mid-pregnancy) women and 16 non-pregnant women.

Methods:  The cardiac parasympathetic outflow was assessed by using the respiratory variability of the R-R interval under controlled breathing according to three different methods: (i) respiratory sinus arrhythmia; (ii) a respiratory-synchronized component of the power spectrum of R-R interval variability with a fast Fourier transform; and (iii) a high-frequency component (HF at 0.15–0.40 Hz) of the power spectrum of R-R interval variability with a maximum entropy method.

Results:  The rate and amplitude of spontaneous respiration and arterial blood pressure at rest were the same between the non-pregnant and pregnant women. However, the baseline R-R interval was shorter and the respiratory-synchronized and HF components of the power spectrum of R-R interval variability were smaller in the pregnant women, indicating a decreased baseline cardiac parasympathetic outflow with mid-pregnancy. Although a foot bath for 15 min did not significantly affect the systemic hemodynamics and body temperature of both groups, the foot bath reduced the respiratory arrhythmia and the respiratory-synchronized and HF components of the power spectrum of R-R interval variability in both groups. The reduction in the respiratory R-R interval became more prominent in the pregnant group.

Conclusion:  A foot bath, used as part of midwives' daily nursing care, is able to decrease the cardiac parasympathetic outflow of pregnant women.

INTRODUCTION

Foot baths have been used in the daily nursing care of midwives during the early stage of labor to make a laboring woman relax and facilitate a labor process. However, the mechanism responsible for the beneficial role of foot baths in labor remains to be identified. The uterus is divided into the corpus and cervix, which can contract independently. The uterus is one of the organs that receive abundant autonomic innervation. Sympathetic adrenergic fibers, via the hypogastric nerve, widely terminate in the smooth muscles and blood vessels of the upper and lower regions of the uterus and the cervix in rats, guinea pigs, pigs, and horses (Bae et al., 2001; Houdeau et al., 1998; Kulkarni et al., 1976; Thilander & Rodriguez-Martinez, 1989; Thorbert, Alm, Omen & Sjoberg, 1977). Parasympathetic cholinergic fibers, via the pelvic nerve, non-uniformly distribute in the myometrium and vasculature of the uterus in mice, rats, guinea pigs, and pigs (Moscarini, Cantagalli, Cavallotti, De Luca & Amenta, 1982; Papka et al., 1999; Thilander & Rodriguez-Martinez; Thorbert et al.). The number and density of the parasympathetic cholinergic fibers increase as they proceed toward the cervix, which shows the richest cholinergic innervation (Moscarini et al.; Papka et al.; Thilander & Rodriguez-Martinez; Thorbert et al.). Therefore, the uterine sympathetic nerve projects to the entire uterus, whereas the uterine parasympathetic nerve chiefly projects to the cervix.

The physiological function of the uterine autonomic nerve has been studied by using field stimulation of an isolated uterine tissue or by stimulation of the pelvic or hypogastric or uterine nerve in association with autonomic blockades. The stimulation of uterine sympathetic adrenergic fibers induced the vasoconstriction and contraction of uterine smooth muscle, while the stimulation of the cholinergic nerve induced the vasodilatation and contraction of uterine smooth muscle in rats, dogs, and humans (Hotta et al., 1999; Morizaki et al., 1989; Ryan, Clark & Brody, 1974; Sato et al., 1996; Stjernquist & Owman, 1985). Accordingly, during the early stage of labor, an increase in sympathetic nerve activity to the uterine corpus might cause the contraction of smooth muscle and a decrease in parasympathetic nerve activity to the uterine cervix might cause the dilatation of smooth muscle, which in turn could accelerate labor. Uterine parasympathetic nerve activity might play a more important role in the neural regulation of uterine function as pregnancy progresses because the disappearance of adrenergic terminals to the myometrium has been noticed and only sparse adrenergic fibers have been found around uterine vessels during pregnancy (Ryan et al.; Sjoberg, 1968). We hypothesized that foot baths as empirical nursing care would affect the autonomic nervous system, promoting the smooth progress of labor, especially a reduction in the amount of parasympathetic nerve activity to the uterine cervix, which in turn might enhance cervical dilatation. Unfortunately, we are not able to directly observe the parasympathetic nerve activity to the uterine cervix in humans.

To test the hypothesis, we examined the parasympathetic nerve activity to the heart, assuming that the change in cardiac parasympathetic nerve activity would correspond to the change in the parasympathetic nerve activity to the cervical uterus. The cardiac parasympathetic nerve activity was assessed in terms of the heart rate variability, assuming that the size of the respiratory sinus arrhythmia of the R-R interval or the respiratory-synchronized frequency component of the power spectrum of R-R interval variability was proportional to the size of the cardiac parasympathetic nerve activity (Fouad et al., 1984; Grossman, Stemmler & Meinhardt, 1990; Katona & Felix, 1975; Pomeranz et al., 1985; Takahashi et al., 2007). We identified the effect of foot baths on the respiratory variability of the R-R interval as an index of cardiac parasympathetic nerve activity under controlled breathing in non-pregnant and pregnant women. Such uninvasive analysis of respiratory variability of the R-R interval was suitable because it produced the least stress on the mothers and fetuses during the experiments.

METHODS

Participants

Sixteen non-pregnant women (aged 22 ± 0.4 years) and five pregnant women (aged 30 ± 2.1 years) at 31–32 weeks (mid-pregnancy) participated in this study. All the participants were healthy and had no cardiovascular or autonomic disease and did not take any medication. This study was carried out in accordance with the Declaration of Helsinki (World Medical Association, 2008) and was approved by the Institutional Ethics Committee, Seoul, Korea. The experimental protocols and procedures were well explained to all the participants and written informed consent was obtained from them before starting the experiments.

Measurements

An electrocardiogram (ECG) was carried out with a pair of electrodes (Magnerode TE-18M-3; Fukuda-Denshi, Tokyo, Japan) attached to the chest. The ECG signal was monitored with a telemeter system (Dynascope DS-3140; Fukuda-Denshi). The thoracic respiratory movement was monitored with an impedance belt (manufactured in our laboratory) that was wound around the chest. The heart rate (HR) and R-R interval were obtained from the R-wave of the ECG. The arterial blood pressure (AP) was measured by using a sphygmomanometer with a cuff wrapped around the upper arm (STBT-780B; Colin, Tokyo, Japan). The axillary body temperature was recorded with a digital thermometer. The surface and deep skin temperatures were continuously measured with thermosensors affixed to the forehead, back, and knee (Core-temp CM-210; Terumo, Tokyo, Japan) of the non-pregnant women alone; the skin temperature of the pregnant women was not measured to avoid uncomfortable feelings arising from putting the thermosensors on the forehead and other places.

Experimental procedures

After completing instrumentation, each participant sat on a comfortable recumbent chair in a sound-proof room with an ambient temperature of 25–26°C and a relative humidity of 50–60%. A period of >20–30 min was allowed to stabilize the cardiovascular and temperature variables before starting the experiments. The frequency of spontaneous respiration during resting was determined so that all the participants could become comfortable. The frequency of a metronome was adjusted so that it was similar to the arbitrary frequency of respiration. The participants were asked to match the rate and amplitude of respiration for a certain period with their predetermined values (controlled breathing) because the rate and amplitude of respiratory movement are known to strongly influence the size of respiratory arrhythmia (Grossman et al., 1990; Kollai & Mizsei, 1990). The respiratory movement was displayed on an oscillograph (AD Instruments, Sydney, Australia) to help achieve controlled breathing.

During the foot bath, the participants' legs were immersed for 15 min in hot water just below knee level. The depth of the water was 15–20 cm and the water temperature was automatically maintained at 41–42°C. The ECG, thoracic respiratory movement, and surface and deep skin temperature results, which were measured simultaneously for 15 min before, during, and after the foot bath, were stored in a computer via an analogue and digital converter at a sampling frequency of 1 kHz (MacLab; AD Instruments, Sydney, Australia). Each period before, during, and after the foot bath involved three sessions of controlled breathing for 2 min for the non-pregnant women, while one session of controlled breathing for 1 min was conducted in each period for the pregnant women in order to minimize stress and discomfort. In each session, the AP and axillary temperature were recorded.

Data treatment and statistical analysis

The respiratory variability of the R-R interval, which is known to represent cardiac parasympathetic nerve activity, was assessed according to three different methods: (i) respiratory sinus arrhythmia; (ii) the power spectrum of R-R interval variability using a fast Fourier transform (FFT); and (iii) the power spectrum of R-R interval variability using a maximum entropy method. The respiratory sinus arrhythmia was calculated from a change in the R-R interval synchronized with respiratory movement, as shown in Figure 1. A respiratory-synchronized component of the power spectrum of R-R interval variability was obtained by using a conventional FFT software system (AcqKnowledge version 3.7.3; BIOPACK Systems, Santa Barbara, CA, USA). A high-frequency component (HF at 0.15–0.40 Hz) of the power spectrum of R-R interval variability was determined by using the maximum entropy method (TARAWA/WIN; SUWA Trust, Tokyo, Japan). The respiratory variability data of the R-R interval and the cardiovascular and temperature data were compared before, during, and after the foot bath. Their relative changes were statistically analyzed by using a one-way analysis of variance with repeated measures. If a significant F-value in the main effect was present, a Dunnett Post-hoc test was carried out to detect a significant difference between the mean values. The baseline values of the cardiovascular parameters, respiratory rate, and respiratory variability of the R-R interval were compared by using an unpaired t-test between the non-pregnant and pregnant groups. The level of statistical significance was defined as P < 0.05. The data are expressed as the means ± standard error.

Figure 1. Changes in the amplitude and rate of thoracic respiratory movement under controlled breathing before, during, and after foot baths in (a) non-pregnant women and (b) pregnant women. The horizontal dotted lines show the control levels before the foot baths.

RESULTS

Baseline values of the heart rate, arterial blood pressure, and body temperature, and their changes during the foot bath

The baseline values of the HR, AP, and axillary temperature, as well as their changes before, during, and after the foot bath, are summarized in Table 1. The baseline HR of 80 ± 4.5 beats/min in the pregnant women was significantly (P < 0.05) higher than the baseline HR of 69 ± 2.3 beats/min in the non-pregnant women. However, the baseline values of the systolic, mean, and diastolic AP and the axillary temperature were not different (P > 0.05) between the two groups. Furthermore, the foot bath did not significantly (P > 0.05) alter any parameters of the HR, AP, and axillary and skin temperatures in both the non-pregnant and pregnant groups, although the skin-surface temperature of the back and the deep skin temperature of the knee tended to increase during and after the foot bath in the non-pregnant women.

Table 1.  Changes in heart rate, arterial blood pressure (AP), and body temperature during foot baths for non-pregnant and pregnant women

Variable	N	Before the foot bath	During the foot bath	After the foot bath
* Significant difference (P < 0.05) between the non-pregnant and pregnant women.
Non-pregnant women
Heart rate (beats/min)	16	69 ± 2.3	68 ± 2.2	69 ± 2.3
Systolic AP (mmHg)	16	108 ± 1.4	108 ± 1.8	108 ± 1.4
Diastolic AP (mmHg)	16	64 ± 2.3	65 ± 2.4	65 ± 2.3
Mean AP (mmHg)	16	79 ± 1.8	79 ± 2.0	79 ± 1.8
Axillary temperature (°C)	7	36.3 ± 0.1	36.4 ± 0.1	36.4 ± 0.1
Skin surface temperature (°C)
Back temperature (°C)	4	34.5 ± 0.7	35.1 ± 0.6	35.3 ± 0.1
Knee temperature (°C)	4	33.8 ± 1.0	34.0 ± 1.0	34.2 ± 0.9
Deep skin temperature (°C)
Forehead temperature (°C)	4	36.3 ± 0.2	36.5 ± 0.7	36.5 ± 0.1
Knee temperature (°C)	4	34.9 ± 0.8	35.5 ± 0.7	35.6 ± 0.7
Pregnant women
Heart rate (beats/min)	5	80 ± 4.5*	84 ± 4.9*	85 ± 6.7*
Systolic AP (mmHg)	5	103 ± 1.0	105 ± 2.0	104 ± 5.0
Diastolic AP (mmHg)	5	62 ± 1.0	64 ± 1.0	66 ± 1.0
Mean AP (mmHg)	5	76 ± 1.0	77 ± 1.1	78 ± 1.2
Axillary temperature (°C)	5	36.4 ± 0.3	36.5 ± 0.2	36.6 ± 0.3

Controlled respiratory movement during the foot bath

Figure 2 demonstrates the rate and amplitude of thoracic respiratory movement during controlled breathing carried out before, during, and after the foot bath, which were actually identical among the three periods in the non-pregnant and pregnant groups (P > 0.05). This result was further supported by the fact that the power spectrum of thoracic respiratory movement was essentially the same before, during, and after the foot bath (Fig. 3).

Figure 2. Effects on the R-R interval, electrocardiogram (ECG), and thoracic respiratory movement under controlled breathing before, during, and after foot baths in an (a) non-pregnant woman and (b) pregnant woman.

Figure 3. Fast Fourier transform power spectrums of thoracic respiratory movement and R-R interval variability under controlled breathing are compared before, during, and after foot baths in an (a) non-pregnant woman and (b) pregnant woman.

Baseline respiratory variability of the R-R interval between the non-pregnant and pregnant women

Table 2 compares the baseline values of the respiratory rate, R-R interval, and respiratory variability of the R-R interval (respiratory sinus arrhythmia and the respiratory-synchronized or HF component of the power spectrum of R-R interval variability) during controlled respiration in the non-pregnant and pregnant women. The respiratory rate before the foot bath was not significantly different (P > 0.05) between the non-pregnant and pregnant groups (0.22 ± 0.089 Hz vs 0.24 ± 0.025 Hz, respectively). However, the baseline R-R interval was shorter in the pregnant group than in the non-pregnant group. Furthermore, both the respiratory-synchronized and HF component of R-R interval variability showed a clear reduction in the pregnant women, as compared to the non-pregnant women, although the size of the respiratory sinus arrhythmia was not different between the two groups (Table 2).

Table 2.  Changes in the respiratory rate and respiratory variability of the R-R interval during foot baths for non-pregnant and pregnant women

Effect of the foot bath on the respiratory variability of the R-R interval in the non-pregnant women

Figure 1 shows an example of the changes in respiratory sinus arrhythmia of the R-R interval, ECG, and thoracic respiratory movement during controlled breathing, which were taken before, during, and after the foot bath in a non-pregnant woman. The amplitude and rate of respiratory movement under controlled breathing were maintained throughout the experiments. The amplitude of the respiratory sinus arrhythmia slightly decreased after the foot bath in the non-pregnant woman, whereas it decreased during the foot bath for the pregnant woman. The FFT power spectrums of thoracic respiratory movement and R-R interval variability are exemplified in Figure 3. An apparent change in the mean R-R interval was not induced by the foot bath. The power spectrum of respiratory movement was essentially the same throughout the experiment for each participant. In contrast, the respiratory-synchronized component of the power spectrum of R-R interval variability decreased during and after the foot bath for the non-pregnant woman and during the foot bath for the pregnant woman. The size of the respiratory sinus arrhythmia did not change during the foot bath but tended to decrease following the foot bath (Fig. 2a). In fact, the respiratory-synchronized component of the FFT power spectrum of R-R interval variability was attenuated during and after the foot bath (Fig. 3a), despite the identical power spectrum of thoracic respiratory movement during controlled breathing.

The effect of the foot bath on the respiratory variability of the R-R interval was assessed according to the sizes of respiratory sinus arrhythmia and the respiratory-synchronized or HF components of the power spectrum of R-R interval variability in the non-pregnant group (Table 2 and Fig. 4a). The amplitude of the respiratory sinus arrhythmia considerably decreased (P < 0.05) during the foot bath in the pregnant women and also decreased after the foot bath in the non-pregnant women. The respiratory-synchronized component of the FFT power spectrum of R-R interval variability significantly decreased (P < 0.05) during and after the foot bath in the non-pregnant women and tended to reduce during the foot bath in the pregnant women. Furthermore, the HF component of the power spectrum of R-R interval variability decreased (P < 0.05) during the foot bath in the pregnant women and after the foot bath in the non-pregnant women.

Figure 4. Effects of foot baths on the respiratory variability of the R-R interval under controlled breathing in (a) non-pregnant and (b) pregnant women. *Significant changes (P < 0.05) from the control before the foot baths. HF, a higher-frequency component (at 0.15–0.40 Hz) of the power spectrum of R-R interval variability.

Effect of the foot bath on the respiratory variability of the R-R interval in the pregnant women

The blunted respiratory variability of the R-R interval during the foot bath was more evident in the pregnant women than the non-pregnant women. The size of the respiratory sinus arrhythmia and the respiratory-synchronized component of the power spectrum of R-R interval variability obviously decreased during the foot bath for the pregnant women (Figs 2,3). Based on the data that are summarized in Table 2 and Figure 4b, the size of the respiratory sinus arrhythmia significantly (P < 0.05) reduced to 67% of the control value during the foot bath. The respiratory-synchronized component of the power spectrum of R-R interval variability tended to decrease to 71% of the control, although the decrease was not significant, which was probably related to the small sample size. Moreover, the HF component of the power spectrum of R-R interval variability was reduced (P < 0.05) during the foot bath to 30% from the control value.

DISCUSSION

A precise mechanism for the regulation of spontaneous labor is not fully understood in midwifery and obstetrics, although many factors, like autonomic nerve activity, various hormones, and cytokines, are likely to play a role. The foot bath has been used during the early stage of labor as Japanese empirical nursing care by midwives. The present study was conducted to understand the beneficial role of the foot bath in labor. We hypothesized that the foot bath would reduce parasympathetic nerve activity to the uterine cervix, which in turn might evoke neurogenic cervical dilatation and thereby promote the smooth progress of labor. As the outflow from the parasympathetic nerve innervating the cervix is inaccessible, we attempted to measure the changes in cardiac parasympathetic outflow from the respiratory variability of the R-R interval in response to the foot bath in non-pregnant and pregnant women. The major new findings of this study are: (i) that the baseline respiratory variability of the R-R interval was reduced in the pregnant women, as compared to the non-pregnant women; (ii) that the respiratory variability of the R-R interval with controlled breathing was decreased by the foot bath in both the non-pregnant and pregnant women; and (iii) that the reduction in the respiratory variability of the R-R interval was much greater in the pregnant group. Thus, the pregnant women not only had a decreased baseline cardiac parasympathetic outflow but also had an enhanced reduction in the cardiac parasympathetic outflow in response to the foot bath. In concert with the response in cardiac parasympathetic outflow, it is considered that the amount of parasympathetic nerve activity to the uterine cervix might be decreased during the foot bath, which could lead to cervical dilatation and explain the beneficial role of the foot bath as empirical nursing care.

Assessment of cardiac parasympathetic nerve activity

The cardiac parasympathetic outflow has been assessed from the respiratory variability of the R-R interval (Fouad, Tarazi, Ferrario, Fighaly & Alicandri, 1984; Grossman et al., 1990; Katona & Felix, 1975; Pomeranz et al., 1985; Takahashi et al., 2007). In order to raise the reliability in this analysis, three different methods (respiratory sinus arrhythmia, and the respiratory-synchronized and HF component of the power spectrum of R-R interval variability) were employed in this study. If all the methods provided an identical result in relation to the changes in respiratory variability of the R-R interval, the result would be considered to be highly reliable. As a matter of fact, the different procedures led to the same conclusion: the foot bath decreased the respiratory variability of the R-R interval in the non-pregnant and pregnant women (Table 2 and Fig. 4). As another way to raise the reliability in the cardiac parasympathetic analysis, breathing was intermittently controlled for 1–2 min throughout the experiments, thus avoiding the possibility that a change in respiration during the foot bath might influence the respiratory variability of the R-R interval (Grossman et al.; Kollai & Mizsei, 1990). As the rate, amplitude, and power spectrum of thoracic respiratory movement were actually the same before, during, and after the foot bath (Figs 1,3), we were able to analyze the effect of the foot bath on the respiratory variability of the R-R interval under the identical breathing pattern.

Modulation of autonomic nerve activity with pregnancy

Remarkable adjustments of the cardiovascular and autonomic nervous systems occur as pregnancy progresses (Capeless & Fry, 2001). For example, the blood volume and cardiac output increase during pregnancy. Regarding the cardiac autonomic outflow, we found that not only did the baseline respiratory sinus arrhythmia become much smaller, but the respiratory-synchronized and HF component of the power spectrum of R-R interval variability became much smaller in the pregnant women, compared to the non-pregnant women (Table 2), suggesting a decreased level of cardiac parasympathetic outflow in association with pregnancy. This finding is supported by previous studies reporting a reduction in the HF component of HR variability with pregnancy (Avery et al., 2001; Ekholm et al., 1992; Lucini, Strappazzon, Vecchia, Maggoini & Pagani, 1999; Yang, Chao, Kuo, Yin & Chen, 2000). The decrease in cardiac parasympathetic outflow might correspond to a higher baseline HR in pregnant women, as compared to non-pregnant women. As the tidal volume usually increases during pregnancy (Capeless & Fry), an increase in tidal volume might enhance the size of the respiratory arrhythmia, suggesting that the decrease in cardiac parasympathetic outflow with pregnancy might be underestimated. However, the cardiac sympathetic outflow increased during pregnancy, according to the power spectral analysis of HR variability (Ekholm et al.; Lucini et al.; Yang et al.). Taken together, a decrease in cardiac parasympathetic outflow and an increase in cardiac sympathetic outflow might appear as pregnancy progresses.

To our knowledge, the direct measurement of sympathetic nerve activity has been conducted by only a few studies using pregnant humans and animals. Greenwood et al. (1998, 2001) assessed the muscle sympathetic nerve activity with microneurography in pregnant women and found that the muscle sympathetic nerve activity was increased in women with normal pregnancy and was more enhanced in hypertensive pregnant women. O'Hagan and Alberts (2003) reported that an increase in renal sympathetic nerve activity during dynamic exercise did not differ between pregnant and non-pregnant conscious rabbits. They also reported that the uterine blood flow was very high in the pregnant rabbits and that the decrease in uterine blood flow during dynamic exercise, which occurred in the non-pregnant rabbits, was markedly attenuated. This finding suggests a decreased baseline sympathetic outflow to the uterus and a blunted sympathetic response during exercise in association with pregnancy. Contrary to this idea, Bower (1966) reported that sympathetic nerve activity to the uterus was more intense in anesthetized pregnant rabbits, compared to non-pregnant rabbits. Even though the available data have been limited so far, pregnancy-induced adjustment of sympathetic outflows to the organs or tissues might be heterogeneous and the direct measurement of uterine sympathetic nerve activity using pregnant conscious animals will be needed.

Decreased cardiac parasympathetic nerve activity related to the foot bath

In this study, there were no significant changes in the hemodynamics and body and skin temperatures during the foot bath in both the non-pregnant and pregnant women. The insignificant responses were probably related to the short duration (15 min) of the foot bath because a foot bath for a longer duration of 60 min elicited cardiovascular and thermoregulatory responses (Inoue et al., 1998). In other words, the decrease in cardiac parasympathetic outflow during the foot bath was not induced secondarily by the changes in the systemic cardiovascular hemodynamics and body temperature. The decrease in cardiac parasympathetic outflow might have been evoked by a reflex originating from cutaneous thermoreceptors that were stimulated during the foot bath. We do not have a clear answer as to why the reduction in cardiac parasympathetic outflow as a result of the foot bath is enhanced with pregnancy. If the uterine parasympathetic nerve activity is reduced by the foot bath in concert with the cardiac parasympathetic outflow, the decrease in uterine parasympathetic nerve activity might promote cervical dilatation, leading to laboring women feeling less labor pain and less anxiety.

Limitations of the study

There are several substantial limitations. First, we extrapolated the effect of the foot bath on uterine parasympathetic nerve activity from the effect on cardiac parasympathetic nerve activity because it was not possible to observe the parasympathetic nerve activity to the uterine cervix in humans. This fundamental assumption must be confirmed by the direct examination of uterine parasympathetic nerve activity by using experimental animals. Second, it is not evident entirely whether the activity of the autonomic nervous system at the early stage of labor is the same as the activity in mid-pregnancy because there are no studies about the effect of the foot bath on the parasympathetic nervous system at both periods. In a future study, we would like to examine the effect of the foot bath on cardiac parasympathetic nerve activity at the early stage of labor. Third, although the augmenting effect of the foot bath on the cardiac parasympathetic outflow in pregnant women was quite impressive, the sample size of pregnant women was relatively small, as compared to the sample size of non-pregnant women. Thus, a definitive conclusion about the augmenting effect of the foot bath on the cardiac parasympathetic outflow with pregnancy should be verified by increasing the sample size. However, as the present research was not an epidemiological study but a physiological and interventional study, a quite large number of participants would not be needed, so long as statistical significance about the main effect was obtained. In fact, we obtained the significant effect of the foot bath on the respiratory-synchronized and HF components of the power spectrum of R-R interval variability in the pregnant women, as well as the non-pregnant women. Fourth, we have to mention that there was a significant difference in age between the non-pregnant and pregnant groups. Previous studies reported that the respiratory sinus arrhythmia and the variability of the R-R interval decreased with age (Shannon et al., 1987; Umetani et al., 1998). However, the reduction in respiratory sinus arrhythmia and the variability of the R-R interval related to the age difference in this study seems to be only 4–5%, based on the previous studies. Thus, the age difference cannot explain the greater reduction (35–92%) in cardiac parasympathetic outflow at rest and during the foot bath in association with pregnancy in the present study. Fifth, the menstrual phase was not determined in the individual non-pregnant women. Although the present effect of the foot bath on cardiac parasympathetic outflow might reflect its average response over the menstrual cycle, it cannot be neglected that the effect of the foot bath on the cardiac parasympathetic outflow in non-pregnant women might be dependent on the menstrual phase and/or sex hormones. Finally, it might be possible that the foot bath could impose emotional stress and thereby affect cardiac autonomic nerve activity, especially in pregnant women. However, this possibility is unlikely because all of the participants felt comfortable during the foot bath.

CONCLUSION

The assessment of the respiratory variability of the R-R interval revealed a reduction in cardiac parasympathetic nerve activity during and after the foot bath in the pregnant women more than in the non-pregnant women. It is likely that the foot bath, as empirical nursing care used by Japanese midwives at the early stage of labor, reduces the cardiac parasympathetic outflow more intensely in pregnant women than in non-pregnant women. The uterine parasympathetic nerve activity might be reduced by the foot bath in concert with the cardiac parasympathetic outflow, which in turn might promote cervical dilatation and the smooth progress of labor.

ACKNOWLEDGMENTS

This study was supported by a Grant-in-Aid for Scientific Research (C) from the Japan Society for the Promotion of Science, Japan.