The haemodynamic, renal excretory and hormonal changes induced by resting in the left lateral position in normal pregnant women during late gestation

Abstract

Objective  To characterise the haemodynamic, renal-electrolyte and hormonal parameters in normal near-term pregnancy.

Design  Observational prospective case-series study.

Setting and population  Eleven women with normal pregnancies at 35–39 weeks gestation.

Methods  Following baseline laboratory assessments and placement of a right-atrial catheter, serial measurements were obtained for 2 hours in the supine position (SP) followed by a change to the (LLP) and subsequent observations for 2 hours.

Main outcome measures  Blood pressure (BP), central venous pressure (CVP), atrial natriuretic peptide (ANP), plasma renin activity (PRA), plasma aldosterone (ALDO), diuresis, creatinine clearance, sodium and potassium excretion.

Results  In the SP, the subjects’ BP remained stable while their CVP decreased. In the LLP, the subjects’ systolic and diastolic BP consistently decreased by about 15 mmHg and their CVP increased within the first 60 minutes. ANP levels doubled in the subjects while they rested in the LLP, whereas the subjects’ PRA and ALDO levels decreased by half compared with when they rested in the SP. In the LLP, the subjects’ creatinine clearance significantly increased by 12% and their sodium excretion and diuresis increased by 38% and 59% respectively.

Conclusion  Rest in the LLP induces systemic and intra-renal haemodynamic and hormonal changes that may play a central physiological role in the renal excretory response to restore excessive sodium/water retention in late pregnancy.

Introduction

Pregnancy is a physiological state in which primary peripheral vasodilatation occurs, and, as a consequence, sodium and water are retained in excess.^1,2 Many hormonal adaptations may be responsible for the extracellular volume increase, particularly, the activation of the renin–angiotensin–aldosterone system (RAAS) and the initial suppression of atrial natriuretic peptides (ANP).³ During late pregnancy, it is very common for women to experience edaema as a clinical expression of this physiological phenomenon. For this condition, as well as for gestational hypertension and other conditions, many obstetricians recommend bed rest with the intent of reducing the patient’s edaema and blood pressure (BP).⁴ This therapeutic/preventive principle continues to be a matter of discussion and has recently been questioned in clinical practice.^4–6 Although there are few studies on this topic, there is an agreement among physicians that water and sodium excretion is greater in the lateral recumbent position.^4–6 The rationale is that, in the lateral recumbent position, the gravid uterus releases the peripheral venous return to the central circulation, which improves cardiac output and consequently renal perfusion.^7,8

The aims of the present study were to characterise the renal-electrolyte physiological adaptations when late-pregnant women move from bed rest in the SP to the left lateral position (LLP) and to identify the RAAS and ANP changes that accompany this physiological phenomenon. We hypothesised that the haemodynamically induced RAAS and ANP adaptations are responsible for the alterations in the renal-electrolyte physiological parameters.

Methods

Eleven healthy pregnant women, 18–32 years of age and at 35–39 weeks gestation, were included in the study. None of them had a previous history of hypertension or renal disease nor were any of them taking any medications, except for oral vitamins and iron supplements. All pregnancies had a normal outcome for mothers and children. After fasting overnight, the pregnant women were admitted to a clinic research centre and submitted to the study procedures.

The study protocol and informed consent were approved by a Central Ethical Institutional Research Board. The gravid participants signed the informed consent form after the entire study protocol had been explained and before any procedure of the study had been performed.

Experimental design

A 2-hour pre-study period was necessary to set all the methodological points. During this period, the pregnant women ingested 1 litre of fresh drinking water and, from that point on, we offered 300 ml of water every hour to ensure a continuous urine flow for collection. After admission, a blood sample was collected to determine haematocrit, creatinine, urea, sodium, potassium, glucose, uric acid and albumin. A spot urine sample was collected for analysis, including a microscopic examination of the urine sediment. After that, a catheter was placed into the right atrium via a right internal jugular vein puncture to monitor the central venous pressure (CVP), which was determined at a sternal angle (normal values range from −2 to +2 cmH₂O).

A schematic of the study design is shown in Figure 1. Two study periods were performed. In the first 2 hours, the pregnant women laid horizontally in a SP, which was followed by a second 2-hour period in which they laid in a LLP.

Figure 1.  Protocol schematic. BP, blood pressure; HR, heart rate; CVP, central venous pressure; PRA, plasma renin activity; Aldo, plasma concentration of aldosterone and ANP, plasma concentration of atrial natriuretic peptide.

The same investigator determined each subject’s BP throughout the entire period of study, measuring on the right arm every 30 minutes using a mercury sphygmomanometer, according to recommended procedure.⁹ BP values for each point are the mean of three consecutive measurements. Urine was collected through spontaneous voiding and programmed bladder emptying at the beginning and the end of each study period. A catheter was inserted into an antecubital vein for blood sampling, which was collected at the midpoint of each experimental period, for sodium, potassium and creatinine. To determine the circulating levels of ANP, plasma renin activity (PRA) and plasma aldosterone (ALDO), peripheral blood samples were collected at the end of the second hour in the SP and at the 10-minute and 2-hour timepoints in the LLP.

Creatinine clearance

Two-hour creatinine clearance was calculated using the formula Ucr.Vmin/Pcr, where ‘Ucr’ is the value of an urinary creatinine sample collected from the total 2-hour urine volume, ‘Vmin’ is the minute urinary volume (total 2-hour urinary volume divided by 120 minutes) and ‘Pcr’ is the serum creatinine collected at midpoint of each experimental period. The result was normalised to 1.73 m² of body surface area (BSA), calculated using the DuBois formula BSA (m²) = 0.20247 × Height(m)^0.725 × Weight(kg)^0.425 considering pregestational weight. The results are expressed in ml/min/1.73 m².

Blood and urine parameters

Sodium and potassium levels in the urine and blood were determined by flame spectrophotometer. Serum creatinine, uric acid, albumin and plasma glucose were determined using an automated chemistry analyser Selectra II (Vital Scientific NV, 6950 AC Dieren, the Netherlands).

Blood samples for ANP, PRA and ALDO were collected into chilled tubes that were kept in ice until being centrifuged at 0°C and stored at −20°C until assaying. Plasma alpha-ANP levels were determined using a RIA Kit (Peninsula Laboratories, Belmont, CA, USA) after pre-extraction with 0.1% trifluoroacetic acid and 60% acetonitrile gradient in C₁₈ SEP-PAK^® columns as previously described.¹⁰ PRA and ALDO levels were determined using a RIA (Fleury Medicina Diagnóstica, São Paulo, SP, Brazil—normal values in healthy women for PRA are 0.4–0.7 ng/ml/h and for ALDO 1.0–16.0 ng/dl).

Statistical analysis

A one-way analysis of variance (anova) for repeated measures was used to compare different means. When differences were significant (P < 0.05), a post hoc Tukey’s test that incorporates a method for controlling for type I errors was performed to identify the differences between specific means.

Results

The mean values for the demographic, gestational, clinical and biochemical parameters are shown in Table 1. As expected, all of these data are characteristic of normal pregnant women in late gestation.

Table 1.   Demographic, gestational, hemodynamic and biochemical parameters of the normal pregnant women

Parameter	Mean (n = 11)	SD	Range
Age (y)	24.5	4.3	18–32
Body weight (kg)	69.6	8.6	55–84
Height (cm)	158	6	152–168
Number of pregnancies	3.0	1.3	1–5
Gestational age (weeks)	37.3	1.1	35–39
Systolic blood pressure (mmHg)	116	16	96–138
Diastolic blood pressure (mmHg)	72	11	50–88
Heart rate (bpm)	77	8	64–96
Haematocrit (%)	35	3	30–40
Urea (mg/dl)	14	3	8–16
Creatinine (mg/dl)	0.7	0.1	0.5–0.9
Sodium (mmol/l)	136	6	125–140
Potassium (mmol/l)	3.3	0.3	2.6–3.6
Glycemia (mg/dl)	74	15	60–108
Uric acid (mg/dl)	3.5	0.6	2.3–4.3
Albumin (g/dl)	3.3	0.4	2.8–4.0

The haemodynamic physiological parameters were determined every 30 minutes in the supine and LLP and are shown in Figure 2. In the SP, there is a trend of increasing systolic and diastolic BP (not significant), whereas after moving to the LLP, a consistent decrease in both systolic and diastolic BP was detected in the subjects without any change in the heart rate. This decrease in BP occurs immediately after the decumbency change, and it could be registered as soon as 10 minutes after the subjects changed to the LLP. The BP decrease persisted through the course of this experimental period. Also, the subjects’ CVP slightly but significantly decreased during the SP at the 90- and 120-minute timepoints, then increased in the first 60 minutes after moving to the LLP. This increase in CVP seems to be the stimulus that leads to the release of ANP (see below).

Figure 2.  Haemodynamic parameters in late gestation normal gravidas resting in the supine and left lateral positions. Values are expressed as the means ± SD, n = 11. Left Y-axis legend refers to systolic blood pressure, diastolic blood pressure and heart rate. Right Y-axis refers to central venous pressure. *P < 0.01 as compared to all values recorded in the supine position; †P < 0.05 as compared to minute ‘Zero’; ‡P < 0.05 as compared to minute ‘120’.

The renal excretory parameters measured during the two experimental periods are depicted in Table 2. The urine volume increased by 59% after the subjects moved to the LLP and sodium excretion increased by 38%. Also, creatinine clearance significantly increased by 12% when subjects were resting in the LLP compared with the SP. In addition, the fractional sodium excretion tended to increase in subjects resting in the LLP, whereas the absolute and fractional potassium excretions tended to decrease in LLP resting subjects (NS).

Table 2.   Renal excretory parameters in late gestation normal gravidas resting in the supine and left lateral positions

Parameter	Supine position	Left lateral position	% Change
Values are expressed as the means ± SD, n = 11. *P < 0.01 as compared to the supine position.
Urine volume (ml/min)	3.7 ± 1.5	5.9 ± 2.0*	59
Sodium excretion (μmol/min)	180 ± 102	248 ± 79*	38
Fractional sodium excretion (%)	1.2 ± 0.8	1.5 ± 0.6	25
Potassium excretion (μmol/min)	38 ± 13	33 ± 11	−14
Fractional potassium excretion (%)	10.8 ± 4.8	7.9 ± 2.7	−27
Creatinine clearance (ml/min/1.73 m^2)	120 ± 40	134 ± 35*	12

The hormonal parameters were determined in the subjects at 120 minutes in the SP, and at the 10- and 120-minute timepoints after the subjects changed to the LLP and the results are shown in Table 3 and Figure 3. Observe that the ANP experienced a rapid increase, to double the level recorded during SP, only 10 minutes after changing to the LLP and the elevated ANP level persisted until the 120 minute timepoint. Inversely, both the PRA and ALDO levels progressively decreased at the 10- and 120-minute timepoints after the subjects changed to the LLP. At 120 minutes, the values for both parameters were approximately half of those recorded when the patients rested in the SP.

Table 3.   Hormonal parameters in late gestation normal gravidas resting in the supine and left lateral positions

Parameter	Supine position (after 120  minutes)	Left lateral position (after 10  minutes)	Left lateral position (after 120  minutes)
Values are expressed as the means ± SE, n = 11. ANP, atrial natriuretic peptide plasma concentration; PRA, plasma renal activity and Aldo, aldosterone plasma concentration. *P < 0.05 as compared to the supine position. **P < 0.01 as compared to the supine position.
ANP (pmol/l)	7.4 ± 2.3	12.4 ± 4.2	15.0 ± 4.5*
PRA (ng/ml/h)	14.7 ± 4.5	12.9 ± 4.6	8.9 ± 3.4**
Aldo (ng/dl)	29.7 ± 6.0	28.8 ± 6.1	14.5 ± 2.3*

Figure 3.  Hormonal parameters in late gestation normal gravidas resting in the supine and left lateral positions. Values are expressed as the means ± SE, n = 11. *P < 0.05 as compared to the supine position at 120 minutes; †P < 0.01 as compared to the supine position at 120 minutes. ANP, atrial natriuretic peptide plasma concentration; PRA, plasma renin activity and Aldo, aldosterone plasma concentration.

Discussion

Pregnancy is a physiological state, and it is often accompanied by vasodilatation, activation of the RAAS and renal sodium and water retention.^1,2 ANP, which possess diuretic, natriuretic and vasodilatatory effects have a close relationship with the RAAS, and in many physiological states their activities are reciprocal, i.e. when ANP levels are high, RAAS levels are depressed and vice-versa.¹¹ This reciprocal phenomenon does not occur during pregnancy, since the RAAS and ANP are both stimulated as early as week 6 of gestation for the RAAS and from week 12 to delivery for the ANP. This increase in ANP activity may, at least in part, be responsible for the vasodilatation and resistance to angiotensin and aldosterone that is typical for this physiological state.² Despite these physiological adaptations, women often experience edaema in late pregnancy, which is in part due to compression of the vena cava by the gravid uterus.^7,8 A common recommendation for reduction of the edaema associated with late gestational pregnancy is to rest in the LLP.^4–6 However, few data are available that explore the physiological phenomenon underlying the water and electrolyte excretion induced by resting in the LLP in late pregnancy.

In the present study, we evaluated normal pregnant women in late gestation by quantifying their systemic haemodynamic parameters, renal excretion of water and electrolytes, as well as hormonal changes in the RAAS and ANP when resting in the SP and then moving to the LLP. We observed that a change from the SP to the LLP was associated with a rapid and significant reduction in systolic and diastolic BP that persisted throughout the 2-hour experimental period, while the heart rate did not change and the CVP increased transiently in the LLP. These haemodynamic alterations were accompanied by a progressive increase in the plasma levels of ANP and a decrease in the concentrations of ALDO and PRA.

We hypothesised that all of these alterations are sequentially interrelated. When pregnant women move from the SP to the LLP, the gravid uterus decompresses the inferior vena cava, which in turn enhances the venous return to the central circulation.^7,8 Consequently, the CVP increases releasing ANP from the atrial wall into the circulation, and also increasing the cardiac output and renal perfusion, which in turn suppresses the RAAS.^7,8,11,12 Elevation of ANP may also directly impair renin and aldosterone release.¹¹ The documented renal consequences of all of theses changes are an increase in the glomerular filtration rate (12% as estimated by creatinine clearance), diuresis (59%) and natriuresis (38%). All of these intra-renal haemodynamic and excretory changes occur despite a decrease in the systemic BP, suggesting that RAAS depression and ANP activation overcome these systemic haemodynamic modifications. Interestingly, previous studies have demonstrated that a doubling in the ANP plasma levels, which is similar to the values observed in our gravidas, by either decreasing the clearance of endogenous ANP or by exogenous administration of ANP, produces renal haemodynamic and excretory changes very close to those observed in this study.^10,13,14 Although these findings have a very consistent relationship, a direct distal tubular effect of ANP cannot be ruled out.¹⁵

Other studies have documented an increase in ANP plasma levels and/or reciprocal inhibition of RAAS in late normal pregnant women who assume LLP as compared to sitting position.^16–18 Indeed, Steegers et al.¹⁶ observed ANP values in the sitting position that are very close to those of our subjects lying in the SP. In addition, the magnitude of plasma ANP increments stimulated by moving to LLP or by sodium overload in the Steegers study are very similar to the increase in ANP plasma levels of our pregnant women in the LLP.¹⁶ These findings suggest that the observed increase in plasma ANP in LLP is due at least in part to an increase in central volume, particularly in the right atrium.

We had many ethical concerns before deciding to place catheters into the right atria of our subjects. CVP may be estimated by non-invasive methods, but this procedure has greater variability and is more imprecise than direct measurement. We anticipated that minimal physiological changes in CVP could be crucial for our study to identify the sequence of events that induces diuresis when women in late gestation assume the LLP. To minimise the risks, all catheters were inserted by the same investigator (FAA) and we did not observe any relevant adverse event related to the procedure.

In summary, this study describes the physiological phenomenon induced by changing decumbency from the SP to the LLP in late gestation gravidas. In the LLP, we observed a marked reduction in the subjects’ systolic and diastolic BPs and a transitory increase in the CVP. This was accompanied by ANP release and RAAS depression and concurrent increase in glomerular filtration rate plus sodium and water excretion. Thus, resting in the LLP induces systemic and intra-renal haemodynamic and hormonal changes in the RAAS and ANP that may play a central physiological role in the renal excretory response to restore excessive sodium/water retention in late pregnancy. However, documented benefits with regards to both the mother and fetal outcome following bed rest in the LLP prescribed for pregnant women with edaema and/or hypertension remains to be established.

Disclosure of interests

There are no conflicts of interests that the authors should disclose.

Contribution to authorship

Fernando A. Almeida and Maria V. Pavan designed and conducted the experimental protocol as well as analysed the data and wrote the manuscript. Cibele I. S. Rodrigues participated in the study design, data analysis and revision of the manuscript.

Details of ethics approval

The study protocol and informed consent were approved by a Central Ethical Institutional Research Board. The gravid participants signed the informed consent after the entire study protocol had been explained and before any procedure of the study had been performed.

Funding

The entire study has been supported by a regular grant from our own institution (Pontifícia Universidade Católica de São Paulo).

Acknowledgements

We would like to thank João Gilberto Vieira, MD, PhD at Laboratório Fleury, São Paulo, Brazil for his kind help in determining the plasma renin activities and ALDO levels. We would also like to thank Mirian de Carvalho Vilela, MD for her kind participation in the study.