Preeclampsia is a common disorder of human pregnancy in which the normal hemodynamic response to pregnancy is compromised.1 It is diagnosed primarily by a rapidly progressive increase in blood pressure (BP) and proteinuria after 20 weeks' gestation.2 As a leading cause of maternal and infant morbidity and mortality, preeclampsia accounts for 22% of maternal deaths and 18% of all premature births and increases maternal risk for future cardiovascular disease.3 Scholars in this field believe that, although preeclampsia is diagnosed in late gestation, gradual pathologic processes leading to the diagnosis begin much earlier, as soon as, if not before, the time of the placentation.4 No preventive treatment is currently recommended, although a number of randomized trials have been conducted.2
Previously, we found that women in a daily prenatal stretching exercise program were found to experience a lower incidence of preeclampsia than those in a walking program. Protective evidence found in the stretching exercise group compared with the walking group included lower resting heart rates and higher antioxidant levels in tissues.5 There is general consensus that gradual decline takes place in vascular compliance and neural modulation among women who eventually develop preeclampsia.6 Therefore, it is conceivable that regular stretching exercise may preserve the integrity of vascular health if stretching enhances autonomic nervous function without aggravating gradual pathologic progress. We also reported that heart rates in the stretching group were lower than those in a walking group in another study.7 Given that heart rate reflects sympathetic activity, this result may suggest that stretching exercise modulates autonomic response in pregnant women and may further influence cardiovascular risk of pregnancy, including preeclampsia.
Heart rate variability (HRV) is a noninvasive measure to assess sympathetic and vagus nerve activity to the heart.8 Heartbeat signals captured by electrocardiography (ECG) provide the standard measures of HRV in time and frequency domains. Simple time domain variables include the standard deviation of the normal-to-normal intervals (SDNN) and the root mean square of the successive differences of normal to normal intervals (RMSSD). The SDNN is a global index of HRV and estimates total variability of the entire cyclic component during the period of recording. The RMSSD reflects parasympathetic tone. The frequency domain measures of HRV include high frequency (HF), low frequency (LF), and LF/HF ratio. Although HF is recognized as a marker of parasympathetic activity, interpretation of LF and LF/HF ratio remains inconclusive.9
Building on our previous findings, we explored the possible beneficial effects of stretching exercise on modulating autonomic response measured by HRV. This pilot study aims to compare HRVs, BPs, and HRs before and after 20 minutes of stretching exercise in healthy pregnant women.
Methods
Subjects
After institutional review board approval was obtained, a total of 15 pregnant women were recruited for the study from the community using flyers posted on bulletin boards and word of mouth. Pregnant women were eligible for the study if they were older than 18 years, carried a singleton pregnancy, were normotensive, and were not taking medication to control hypertension.
Stretching Exercise Intervention
The intervention was developed based on recommendations of the American College of Obstetricians and Gynecologists.10 Although in sports medicine and physical therapy, stretching is viewed as a means of flexibility enhancement and injury prevention,11 the 20 minutes of stretching exercise as prescribed in this study is focused on muscle stretching, not joint flexibility. The 20 minutes of stretching exercise consisted of a sequence of large skeletal muscle stretches from the neck to the upper limbs, to the torso, and the lower limbs. Briefly, the 20 minutes of exercise included 10 minutes of upper body movements and 10 minutes of lower body movements. All stretching movements were completed either in a sitting or knees-and-hands position on the floor. Subjects were instructed to view and follow a video demonstration. To ensure that the participants stretched safely and effectively, they were instructed to breathe normally as they stretch, not to force the stretching but to stretch the muscles up to the point of slight resistance. A research assistant also followed the video demonstration along with the participants, observing and correcting their stretching motions.
Measures Contextual Factors
Age (in years), gestational weeks, and prepregnant weight were self-reported. Height (in meters) was measured with a wall-mounted stadiometer, and weight (in kilograms) was measured with an electronic scale. Body mass index was calculated as kilograms per square meter.
Blood Pressure and Heart Rate
Peripheral brachial BP and heart rate were measured using a Welch Allyn Vital SignsMonitor 300 Series (Skaneateles Falls, New York). Because incorrect cuff size causes errors in BP measurement (cuff too large reads low and cuff too small reads high), the midsection circumference of the nondominant upper arm was obtained with a tape measure, and the proper sized cuff was selected according to the upper arm circumference.12 Using this automated BP monitoring device, BP and HR were measured twice in the Semi-Fowler position, with a 2-minute rest between measures.
Heart Rate Variability
Heart rate variability is a noninvasive measure of the variation in the beat-to-beat interval. Subjects were asked to breathe at their own normal respiration rate. Electrocardiography (ECG) was obtained for 10 minutes in the semi-Fowler position. Of 15 participants, 12 participants' ECG recordings were free of body movements and ectopic beats. Using the Nevrokard HRV analysis program, two 5-minute HRV of the frequency domain (LF power and HF power in absolute [in milliseconds squared] and in normalized values and LF/HF ratio) and the time domain (SDNN and RMSSD in milliseconds) were calculated and averaged for each participant.
Procedure
Data collection was conducted in the Biobehavioral Laboratory at the University of North Carolina at Chapel Hill. The Biobehavioral Laboratory has a sleep laboratory, which provides a sleep room equipped with ECG and BP measures and a living room equipped with a VCR screen. Outside these 2 rooms, there is an observation station for the sleep room and a weight and height measure station. After obtaining informed consent in the sleep room, height and weight were assessed in the station. Next, the subjects walked back to the sleep room and took a 10-minute rest in the Semi-Fowler position. Then, the baseline BP and HR were measured, and HRV were recorded for 10 minutes. Next, the subjects slowly walked cross the station to the living room where a yoga mat was laid out and a VCR screen was ready for showing a 20-minute stretching exercise. Subjects followed the exercise video. After the stretching exercise, subjects stood up slowly and walked across the station back to the sleep room. Subjects lied in a semi-Fowler position again. Blood pressure, HR, and HRV were measured again in the same manner as before the exercise.
Data Analysis
All analyses were performed with SAS (version 9.2 for Windows). The assumption of normally distributed difference scores was examined by the Shapiro-Wilk test. To test the hypothesis that the cardiovascular measures before and after the stretching exercise were equal, paired-samples t tests or Wilcoxon signed-rank test for the nonnormally distributed variables were performed. All reported P values were 2 sided, and P = .05 was considered statistically significant.
Results
The age range of the 15 participants was between 24 and 37 years, and their mean (SD) age was 29.47 (4.07) years. The mean (SD) body mass index before pregnancy was 29.25 (5.96) kg/m2, with 4 participants (36.4%) being normal, 1 participant (9.1%) being overweight, 5 participants (45.5%) being medically obese at stage 1, and 1 participant (9.1%) being medically obese stage 2. The gestational weeks ranged between 15 and 39 weeks, and the mean (SD) gestational weeks was 26.53 (8.35) weeks. The mean (SD) of weight change before pregnancy and the present was 7.98 (7.52) kg.
The Shapiro-Wilk test showed that mean difference scores in absolute values of HF and RMSSD were nonnormally distributed at P = .000 and P = .003, respectively. Results for the paired-samples t test or Wilcoxon signed-rank test (for the nonnormally distributed variables) are summarized in the Table. The results demonstrate that there were significant increases in SDNN and RMSSD levels after the stretching exercise. Compared with before the stretching exercise, SDNN increased by 7.40 milliseconds (P = .042) and RMSSD increased by 11.68 milliseconds (P = .041) after the stretching exercise. Diastolic BP and heart rate also decreased by 2.13 mmHg and 3.31 bpm, but they did not reach statistical significance (P = .074 and P = .053, respectively).
Discussion
In this pilot study, we tested the effects of 20 minutes of stretching exercise on HRV, BP, and HR in pregnant women. We found that in comparison with before the stretching exercise, SDNN, which reflects total variability of heart rate, significantly increased. Furthermore, RMSSD, a surrogate measure of parasympathetic outflow, also significantly increased. It is possible that the time spent completing the stretching exercise provided simply a prolonged resting time, although this is unlikely. Pregnant participants were in a semi-Fowler position for 10 minutes before any cardiovascular data were measured, which should provide sufficient time to reach a resting state of vital signs. Although it was not statistically significant, possibly attributable to the small sample size, heart rate, which is a surrogate measure of sympathetic function, also decreased after the stretching exercise. Of note, sympathetic overactivity usually occurs after aerobic exercise. Our results may suggest that while the stretching exercise enhances parasympathetic function, it does not stimulate sympathetic activity.
A well-balanced interaction between the sympathetic and the parasympathetic activity is essential for the cardiovascular system to adapt to hemodynamic needs. Along with hemodynamic change, change in autonomic nervous function is reported to start during the first trimester of normal pregnancy. This change is characterized by an increase in sympathetic function and a reduction in parasymphathetic tone, demonstrating that RMSSD, SDNN, and HF all are reduced, whereas LF is increased during pregnancy.13 Previous studies showed that these changes are enhanced in preeclamptic pregnancy compared with normotensive pregnancy.14,15 An intervention to promote a decrease in sympathetic function and an increase in parasympathetic function may be beneficial to reduce preeclampsia risk.
An observational study has reported that a reduced risk of preeclampsia is associated with physical activity such as walking or high-intensity activities before pregnancy and during early pregnancy.16 However, the role of low-intensity activities such as stretching during pregnancy remains inconclusive. Pregnancy is accompanied by significant hemodynamic changes, including increase in cardiac output, blood volume expansion, and maternal heart rate17; thus, low-intensity activities that promote parasympathetic function without increasing sympathetic activity may have significant benefit to pregnant women, especially in the latter half of pregnancy, when many pregnant women may find it difficult to conduct high-intensity activity.
Our previous study observed that regular stretching exercise resulted in lowering systolic BP18 and pulse pressure.19 In the present study, which included a 20-minute stretching exercise, diastolic BP mildly decreased (P = .074). As for the HRV measures of the frequency domain, although HF is widely accepted as an index of parasympathetic function, it is somewhat controversial that LF reflects sympathetic activity.9 In the current study, absolute values of HF and LF showed an increasing trend after the stretching exercise, but the normalized values, which show the balanced function of the sympathetic and parasympathetic branches, failed to produce meaningful results.
Further research is needed to examine how the application of low-intensity physical activity interventions such as stretching exercise may impact hemodynamic adaptation processes. Although HRV is a noninvasive measure of autonomic function, it takes significant time for accurate measure in clinical settings, which may create a barrier for translational research. The use of alternative methods to measure autonomic function, including Valsalva maneuver, deep breathing, isometric handgrip test, or orthostatic test,20 may promote studies that investigate this important mechanism leading to the prevention of preeclampsia in pregnant women.
Conclusion
Reduced HRV is manifested as a preeclampsia risk in pregnant women. This pilot study shows that 20 minutes of stretching exercise may be effective in increasing HRV and promoting parasympathetic tone in pregnant women. Future studies are warranted to replicate these results in a large sample of pregnant women and to further investigate the potential of developing stretching exercises as an important preventive measure to decrease sympathetic function and increase parasympathetic function in pregnant women at risk for preeclampsia.
What's New and Important
* Following 1 episode of stretching exercise, SDNN (total variability of heart rate) and RMSSD (a surrogate measure of parasympathetic outflow) increased.
* Regular stretching exercise may be beneficial in reducing preeclampsia risk in pregnant women by increasing parasympathetic tone.
REFERENCES