Circadian Blood Pressure Variability as a Function of Parity in Normotensive Pregnant Women

Abstract

Studies based on office blood pressure measurements concluded that parity has significant effects on blood pressure during pregnancy. The authors evaluated possible differences in the circadian pattern of blood pressure as a function of parity in normotensive women systematically studied by ambulatory blood pressure monitoring. They analyzed 1400 blood pressure series sampled for 48 consecutive hours every 4 weeks from the first obstetric visit (usually within the first trimester of pregnancy) until delivery in 234 women. The circadian pattern of blood pressure variation for each group (nulliparous vs. multiparous) and trimester of gestation was established by population multiple‐component analysis. A highly statistically significant circadian pattern, described by a model that includes components with periods of 24 and 12 hours, was demonstrated for systolic and diastolic blood pressure for both groups of pregnant women in all trimesters (p<0.001). There was no significant difference in 24‐hour mean among groups divided by parity at any stage of pregnancy (p>0.315). Data obtained from systematic ambulatory monitoring in normotensive pregnant women indicate the lack of differences in blood pressure according to parity. Reference thresholds for blood pressure in pregnancy could thus be developed as a function of rest‐activity cycle and gestational age only, independent of parity.

Previous studies have shown the prognostic significance of parity in the differential diagnosis of various hypertensive diseases of pregnancy. ¹ , ² , ³ , ⁴ , ⁵ Multiparous women with preeclampsia or even eclampsia seem to be different on long‐term follow‐up from nulliparous women with the same complications in pregnancy. ¹ , ⁶ , ⁷ Moreover, clear differences have been demonstrated between nulliparous and multiparous women in both maternal presentation and impact of maternal disease on fetal growth and development. ² , ⁸ , ⁹ , ¹⁰ It has been thus concluded from these observations that multiparous and nulliparous patient groups should be analyzed separately whenever hypertensive diseases of pregnancy are evaluated. Therefore, it is of interest to examine the underlying variations of blood pressure (BP) during pregnancy as a function of parity.

In a cohort study of more than 6500 pregnant women, Christianson ¹¹ showed that both maternal age and parity have highly significant effects on BP measurements obtained at the clinic during pregnancy. Although other studies based on office BP determinations have also provided similar conclusions, ¹² , ¹³ results are still controversial due to the lack of correlation between parity and BP shown in several other trials. ¹⁴ , ¹⁵ , ¹⁶ , ¹⁷ The controversy could come from, among other factors, the inclusion in most studies of both healthy and complicated pregnancies, as well as from the shortcomings of office BP values. These provide a measurement that represents only a fraction of the 24‐hour BP profile, usually obtained under circumstances that may have a pressor effect; and the technique is fraught with potential errors, including instrument defects and examiner technique. ¹⁸ Despite the extremely poor sensitivity and specificity of clinic values, the diagnosis of gestational hypertension still relies on office BP measurements and the use of arbitrary constant critical thresholds, that is, 140/90 mm Hg after 20 weeks of gestation in a previously normotensive woman. ¹⁹ , ²⁰ Recent studies have tried to overcome many of the limitations of isolated BP measurements by relying on ambulatory BP monitoring (ABPM). ¹⁷ , ²¹

By using ABPM, a predictable pattern of BP variability throughout the 24 hours of the day (circadian variability) has been shown to characterize clinically healthy pregnant women as well as women who develop gestational hypertension or preeclampsia. ²² , ²³ , ²⁴ This circadian variation in BP represents, on one hand, the influence of internal factors such as ethnicity, gender, autonomic nervous system tone, vasoactive hormones, and hematologic and renal variables. ²⁵ BP is also affected by a variety of external factors, including ambient temperature/humidity, physical activity, emotional state, alcohol or caffeine consumption, meal composition, and sleep/wake routine. ²⁶ , ²⁷ During gestation, another source of variability comes from the predictable pattern of BP changes. ²⁸ In clinically healthy pregnant women, BP steadily decreases up to the middle of gestation and then increases up to the day of delivery, with final BP values similar to those found early in pregnancy in the same women. For women who develop gestational hypertension or preeclampsia, BP is stable during the first half of pregnancy and then continuously increases until delivery. ²⁸

Moreover, differences between healthy and complicated pregnancies in the circadian pattern of BP, previously documented for the second trimester of pregnancy, ²⁹ can be observed by ABPM as early as in the first trimester of pregnancy, well before the actual clinical diagnosis of gestational hypertension or preeclampsia takes place for the women investigated. ²³ , ²⁴ Therefore, healthy and complicated pregnancies should be studied separately when investigating other possible factors influencing BP during gestation. Accordingly, we studied the possible influence of parity on the circadian pattern of BP in clinically healthy normotensive pregnant women who were systematically sampled by 48‐hour ABPM from the first obstetric visit to the hospital until delivery.

Subjects

We studied 234 (125 nulliparous) untreated white pregnant women with uncomplicated pregnancies who fulfilled all required criteria for this trial. All women received obstetric care at the Obstetric Physiopathology Unit, Hospital Clínico Universitario, Santiago de Compostela, Spain. The demographic characteristics of the women investigated are included in the Table. Gestational age and fetal growth were determined by monthly echography assessments. Conventional office BP measurements (three to six at each obstetric visit) were always obtained by the same midwife to avoid examiner bias. Women were seated during BP determination, and Korotkoff phase V was used for diastolic BP (DBP) assessment. Inclusion criteria were absence of any condition requiring the use of antihypertensive medication; maternal age (18–40 years); gestational age (<16 weeks at the time of inclusion), clinic BP measurements below 140/90 mm Hg for the duration of pregnancy; and hyperbaric index consistently below the previously established threshold for diagnosing hypertension in pregnancy ³⁰ , ³¹ as an added measure to corroborate normotension in all women investigated. The hyperbaric index, as a determinant of BP excess, has been defined as the total area of any given patient's BP above a reference threshold, provided by the upper limit of a tolerance interval specified as a function of gestational age and rest‐activity cycle. ³⁰ , ³¹ , ³² Exclusion criteria were gestational hypertension, preeclampsia, multiple pregnancy, chronic hypertension, chronic liver disease, any disease requiring the use of anti‐inflammatory medication, diabetes or any other endocrine disease such as hyperthyroidism, and intolerance to the use of an ABPM device. Apart from the 234 women providing all required information, 19 subjects who provided <4 profiles of ABPM (two spontaneous abortions and 17 who withdrew from the trial) were eliminated from the study. The State Ethics Committee of Clinical Research approved the study. All women signed consent forms before entering the study.

ABPM Assessment

In this trial, the systolic BP (SBP) and DBP of each woman were scheduled to be measured by ABPM every 20 minutes during the day (7 a.m. to 11 p.m.) and every 30 minutes during the night for 48 consecutive hours with a validated SpaceLabs 90207 device (SpaceLabs Inc., Issaquah, WA) at the time of recruitment (usually within the first trimester of pregnancy) and then every 4 weeks until delivery. Women were assessed while adhering to their usual diurnal (9 a.m. to midnight for most) activity‐nocturnal sleep routine (average duration of nocturnal rest 9.0±1.2 hours). Subjects were instructed to go about their usual activities with minimal restrictions but to follow a similar schedule during the 2 consecutive days of ABPM and to avoid the use of over‐the‐counter and other medication for the duration of the trial. BP series were eliminated from analysis (a total of 52) when the women showed an irregular rest‐activity schedule during the 2 days of sampling, an odd sampling with spans of >3 hours without BP measurement, or a night resting span of <6 hours or >12 hours. The total number of BP series provided by the 234 women under investigation fulfilling all mentioned requirements was 1400. ABPM was performed in addition to the women's routine antenatal care, and no person was hospitalized during monitoring. The BP cuff was worn on the nondominant arm with cuff size determined by upper arm circumference at each study visit. The monitor was always set to the so‐called "blind function." Accordingly, the display never shows the actual BP readings after measurement, keeping the information blind to the patient. ABPM always started between 10 a.m. and 1 p.m. During monitoring, each woman maintained a diary listing the time of going to bed at night and awakening in the morning and of meals, exercise, and unusual physical activity, plus events and mood/emotional states that might affect BP.

Statistical Methods

Each individual' clock hour BP values were first rereferenced from clock time to hours before and after awakening from nocturnal sleep. This transformation avoided the introduction of bias in the shape and phasing of the 24‐hour BP pattern due to differences among subjects in their sleep‐activity routine. ³³ BP values were then edited according to commonly used criteria for the removal of outliers and measurement errors. ³⁴ The circadian rhythm of BP for each group of pregnant women (nulliparous vs. multiparous) in each trimester of gestation was established by population multiple‐component analysis, ³⁵ a method designed for analysis of nonsinusoidal hybrid data (time series of data collected from a group of subjects) consisting of values distributed at equal or unequal intervals.

The method produces estimates of the rhythm‐adjusted mean or midline estimating statistic of rhythm, average value of the rhythmic function fitted to the data (MESOR), as well as the amplitude (one half the extent of change explainable by the rhythmic fitted curve) and acrophase (crest time expressed as a lag from a designated reference) for every fitted component. When all fitted components are harmonics from a fundamental period, the method of multiple components also provides three additional parameters: the overall amplitude (one half the difference between the maximum and the minimum of the best fitted curve); the orthophase; and the bathyphase (peak and trough times, respectively, here expressed as a lag from the time of awakening from nocturnal sleep). ³⁵ Circadian parameters were subsequently compared between groups of women in each trimester of pregnancy with a nonparametric test developed to compare parameters obtained from population multiple components analysis. ³⁵ Hourly means of BP were compared between groups of women by t test, corrected for multiple testing using Holm' procedure. ³⁶ Additionally, the demographic and perinatal characteristics included in the Table were compared between groups of pregnant women by analysis of variance (quantitative variables) or nonparametric χ ² testing (incidence of complications).

RESULTS

The baseline characteristics of the two groups of pregnant women investigated (nulliparous and multiparous, Table) differed in age (p=0.002) but not in maternal weight (p=0.105) or height (p=0.798). The comparison of the average office BP measurements obtained at the time of the first visit to the hospital indicates the lack of statistically significant differences between nulliparous and multiparous pregnancies (p>0.296 for both SBP and DBP) and also similar office BP measurements taken at the last obstetric visit (shortly before delivery) (p>0.365 for both SBP and DBP). Gestational age at delivery, newborn weight, and Apgar scores at 1, 5, and 10 minutes after birth were also similar among both groups of pregnant women. The Table further illustrates the lack of statistically significant differences between groups in preterm delivery (before 37 weeks gestation), intrauterine growth retardation, and incidence of delivery by cesarean section.

The circadian rhythms (obtained by population multiple component analysis) for SBP and DBP in each trimester of pregnancy for nulliparous and multiparous clinically healthy women are represented in Figures 1‐3. The graphs show the lack of differences in 24‐hour mean (MESOR), amplitude, and orthophase of BP as a function of parity in normotensive women systematically sampled throughout gestation by 48‐hour ABPM. The lower horizontal axis represents circadian time in hours after awakening; the resting span is indicated by the dark bar in the lower horizontal axis. The nonsinusoidal curve represented for each group corresponds to the best fitted waveform model obtained by population multiple components analysis applied to all original BP values (not just to the hourly means). The arrow from the upper horizontal axis indicates the circadian orthophase for each group. Differences or similarities in rhythm characteristics, as well as the general waveform of circadian variability in BP can be readily seen from this graphic representation.

The comparison of SBP and DBP between nulliparous and multiparous pregnant women sampled in the first trimester of gestation (Figure 1) indicates not only the similarity in circadian MESOR of BP, but also that all 24‐hourly averages are practically overlapped for both groups without statistically significant differences among groups detected by t test adjusted for multiple testing. In the second (Figure 2) and third (Figure 3) trimesters of pregnancy, the differences in the 24‐hour mean of BP between nullipara and multipara are below 0.1 mm Hg for both SBP and DBP.

Figure 1.

Circadian variation of systolic blood pressure (left) and diastolic blood pressure (right) in clinically healthy pregnant women who were assessed by 48‐hour ambulatory monitoring in the first trimester of pregnancy. Each graph shows hourly means and standard errors (thin lines) of data collected from nulliparous (continuous line) and multiparous pregnant women (dashed line). Average resting span is indicated by the dark bar in the lower horizontal axis. Nonsinusoidal shaped curves (thick lines) correspond to the best‐fit waveform model determined by population multiple‐component analysis. Arrows descending from the upper horizontal axis point to the peak time (orthophase) of the circadian rhythm of blood pressure determined by the waveform approximation.

Figure 2.

Circadian variation of systolic blood pressure (left) and diastolic blood pressure (right) in clinically healthy pregnant women who were assessed by 48‐hour ambulatory monitoring in the second trimester of pregnancy. Each graph shows hourly means and standard errors (thin lines) of data collected from nulliparous (continuous line) and multiparous pregnant women (dashed line). Average resting span is indicated by the dark bar in the lower horizontal axis. Nonsinusoidal shaped curves (thick lines) correspond to the best‐fit waveform model determined by population multiple‐component analysis. Arrows descending from the upper horizontal axis point to the peak time (orthophase) of the circadian rhythm of BP determined by the waveform approximation.

Figure 3.

Circadian variation of systolic blood pressure (left) and diastolic blood pressure (right) in clinically healthy pregnant women who were assessed by 48‐hour ambulatory monitoring in the third trimester of pregnancy. Each graph shows hourly means and standard errors (thin lines) of data collected from nulliparous (continuous line) and multiparous pregnant women (dashed line). Average resting span is indicated by the dark bar in the lower horizontal axis. Nonsinusoidal shaped curves (thick lines) correspond to the best‐fit waveform model determined by population multiple‐component analysis. Arrows descending from the upper horizontal axis point to the peak time (orthophase) of the circadian rhythm of blood pressure determined by the waveform approximation.

DISCUSSION

Results from this study on normotensive women systematically measured by 48‐hour ABPM during different stages of their pregnancies indicate the lack of differences in BP according to parity, in contrast to conclusions from earlier reports based on routine prenatal visit BP measurements. ¹¹ , ¹² , ¹³ These reports indicated that, for a given maternal age, nulliparous women had a higher BP than multiparous women. This observation was usually related to the prognostic significance of parity in the differential diagnosis of gestational hypertension, preeclampsia, and eclampsia. ² , ³ , ⁴ , ⁵ Not all previous studies have shown a significant relation between parity and office BP measurements. Moutquin et al. ¹⁵ conducted a prospective study on 366 pregnant women whose BP was measured at each antenatal visit using an automatic random‐zero sphygmomanometer; they found no difference in BP during pregnancy between nulliparous and multiparous women who remained normotensive. Lee‐Feldstein et al. ¹⁴ analyzed the BP values measured in 755 women in relation to parity, race, and residential stress; none of the regression relationships between BP and parity were found to be significant in the race‐stress groups included in their study. In a more recent trial, Okonofua et al. ¹⁶ monitored the BP of 189 women from early pregnancy and up to term, during labor, and 24 hours after delivery; they also found no significant correlation of BP with parity. The results from our trial, corroborating previous findings, ¹⁷ indicate the lack of statistically significant differences in BP as a function of parity not just on ABPM at all stages of gestation (Figures 1–3), but also in office BP values (Table). Differences in office BP as a function of parity in some previous studies could be directly related to the inclusion of both normotensive and hypertensive pregnant women. ¹¹ , ¹² , ¹³

Results from this study are based on data systematically measured by 48‐hour ABPM. Although most studies assessing the circadian BP profile have used 24‐hour ABPM, as a compromise with practicability, monitoring over at least 48 hours has been shown to present advantages in the analysis of BP variability, ²⁷ , ³³ , ³⁷ , ³⁸ diagnosis of disease, ³⁰ , ³¹ , ³² and evaluation of a patient's response to treatment. ³⁸ The individualized estimation of rhythm characteristics becomes more reliable; new end points are obtained, like the circadian period, that cannot usually be estimated from 24‐hour records. ³⁹ Moreover, there may be relatively large day‐to‐day changes in BP, due in part to differences in day‐to‐day schedule, that are at least partly accounted for by sampling over two or more days. ³³ , ³⁷ Finally, ABPM for 48 consecutive hours revealed a statistically significant pressor response that could mostly reflect a novelty effect in the use of the monitoring device for the first time. ⁴⁰ This ABPM effect, which significantly increases BP for about the first 10 hours of monitoring, is diminished in extent and duration for successive sessions of ABPM on the same patients. This pressor effect, which has been shown to be independent of any change or modification in physical activity between consecutive days of monitoring, ⁴⁰ supports the need for monitoring over spans of time longer than 24 hours.

The circadian pattern with large amplitude that characterizes BP in healthy pregnancies at all gestational ages ²³ , ²⁴ suggests that the constant threshold currently used for diagnosing hypertension in pregnancy ¹⁹ , ²⁰ should be replaced by a time‐specified reference limit reflecting the mostly predictable BP variability. ³³ , ⁴¹ The ideal reference interval for a variable of clinical interest would be specific for all deterministic factors affecting that variable. Results from Figures 1–3 corroborate previous reports showing a predictable pattern of BP variation with gestational age in normotensive women independent of parity. ²⁸ Moreover, Figures 1–3 also show the expected highly significant circadian variation in BP for all groups of women and stages of gestation. This circadian pattern, here expressed in hours after awakening from nocturnal sleep, was demonstrated as statistically significant in 97% of all pregnant women studied by 48‐hour ABPM who participated in this trial.

This study of women systematically sampled by 48‐hour ABPM throughout gestation confirms the predictable pregnancy‐associated variability in BP, shows the lack of any significant influence of parity on BP, and provides proper information for the establishment of reference limits for BP to be used in the early diagnosis of hypertensive complications in pregnancy. ³⁰ , ³¹ , ⁴¹ Those limits could thus be developed as a function of rest‐activity cycle and gestational age, independent of parity.

Acknowledgment: This research was supported in part by grants from Xunta de Galicia (PGIDIT03‐PXIB‐32201PR) and Vicerrectorado de Investigación, University of Vigo.