Article Text

Download PDFPDF

Effects of sleeping position on development of infant cardiovascular control
  1. S R Yiallourou,
  2. A M Walker,
  3. R S C Horne
  1. Ritchie Centre for Baby Health Research, Monash Institute of Medical Research, Monash University, Melbourne, Victoria, Australia
  1. Associate Professor Rosemary S C Horne, Ritchie Centre for Baby Health Research, Level 5, Monash Medical Centre, 246 Clayton Rd, Clayton, Victoria, Australia 3168; rosemary.horne{at}med.monash.edu.au

Footnotes

  • Competing interests: None.

  • Ethics approval: The protocol for this study was approved by the Southern Health and Monash University Human Research Ethics Committees.

  • Patient consent: Written parental consent was obtained prior to commencement of the study and no monetary incentive was provided for participation.

View Full Text

Statistics from Altmetric.com

Sudden Infant Death Syndrome (SIDS) is presumed to occur during sleep, and prone sleeping has been identified as a major risk factor for SIDS.13 Epidemiological studies have also shown that SIDS has a peak incidence occurring between 2 and 3 months postnatal age (PNA).13 Although the underlying mechanism remains elusive, circulatory dysfunction leading to uncompensated hypotension may be a crucial element of the fatal event.4 5 Preliminary studies indicate that vasomotor tone is reduced in the prone position,6 7 suggesting the potential for hypotension to develop in this sleep position, but this requires verification for infants at the age of highest risk for SIDS.

Assessments of the development of cardiovascular control during sleep in infancy to date have focused on heart rate (HR), and blood pressure (BP) control has not been examined. Exploiting a new method for the continuous recording of BP during sleep in infants,810 we aimed to provide normative data on the development of BP and HR over the first 6 months of PNA, the period when most SIDS deaths occur. Our study focused on the effects of sleeping position and sleep state because of their potential importance to understanding the pathogenesis of SIDS.

METHODS

Subjects

Twenty full-term infants (12 female/8 male) born at 38–42 weeks gestational age with normal birth weights ranging from 2970 g to 4250 g (mean (SEM) 3590 g (100 g)) and Apgar scores of 9–10 (median 9) at 5 minutes were studied. Each infant was of appropriate birth weight for gestation and had no congenital abnormalities. All infants were born to non-smoking mothers and routinely slept supine at home.

Each infant was studied at three ages within the first 6 months of life: at 2–4 weeks PNA (mean (SEM) age 3 (0.1 weeks)), at 2–3 months PNA (mean (SEM) age 10 (0.1 weeks)) and at 5–6 months PNA (mean (SEM) age 22 (0.3 weeks)).

Polysomnography

Infants were studied using daytime polysomnography under dim lighting and constant room temperature (22–23°C). Electrodes for recording physiological variables were attached during the infants’ routine morning feed, and they were allowed to sleep naturally. Using a polygraph (Grass Instrument Co, Quincy, Massachusetts, USA) electroencephalogram (EEG), electrooculogram (EOG), submental electromyogram (EMG), electrocardiogram (ECG), thoracic and abdominal breathing movements (Resp-ez Piezo-electric sensor, EPM Systems, Midlothian VA, USA), arterial blood oxygen saturation (SpO2) (Biox 3700e Pulse Oximeter, Ohmeda, Louisville, Colorado, USA) and abdominal skin temperature (YSI 400 series thermistor, Yellow Springs Instruments, Yellow Springs, Ohio, USA), were recorded. Sleep state was assessed as either quiet sleep (QS) or active sleep (AS) using EEG, behavioural, HR and breathing pattern criteria.11 Infants slept in both the prone and supine sleeping positions, with the initial position randomised. Infant sleep position was changed after the mid-day feed.

BP measurement

BP was measured using a non-invasive cuff placed around the infant’s wrist (Finometer, FMS, Finapres Medical Systems, Amsterdam, The Netherlands) using a method previously validated by our group.10 BP measurements were recorded in both AS and QS in both the supine and prone positions in 1–2-minute epochs, with 5–8 minutes being recorded in each infant for each condition. Care was taken to continuously observe infants during BP recordings and there was no evidence that inflation of the BP cuff induced any changes in behaviour, sleep or HR.

Physiological variables were digitised via a 16-channel Powerlab (ADInstruments, Sydney, Australia) onto a computer program for data storage, analysis and visualisation (Chart 5.2, ADInstruments, Sydney, Australia). All data were recorded at a frequency of 400 Hz.

Data analysis

Beat to beat values of mean (MAP), systolic (SAP) and diastolic (DAP) arterial pressure, and HR were obtained using peak detection. Movement artefacts that disrupted the physiological signals and BP data that lay 1.5 × the interquartile range outside of the first and third quartiles were removed from subsequent analyses. Data were averaged for each sleep state and sleeping position and mean BP and HR values calculated for each infant, then pooled at each study for comparison. Missing BP values (n = 3/80 values at 2–3 weeks, 3/80 values at 2–3 months and 14/80 values at 5–6 months) were calculated.12

Statistical analysis

Statistical analysis was performed using SigmaStat (Systat Software, Richmond, California, USA). The effects of sleep state and sleeping position were compared using a two-way repeated measures ANOVA, and the effects of PNA using three-way ANOVA with Student-Newman Kuels post hoc analysis. Results are presented as mean (SEM), with significance taken at the p<0.05 level.

RESULTS

A typical polygraphic record of the effects of sleeping position and sleep state on BP and HR is presented in fig 1.

Figure 1 Blood pressure (BP, mmHg) recorded with standard polysomnography variables in an infant aged 2–3 months, illustrating differences of BP (top channel) between quiet sleep (QS) and active sleep (AS) in the supine and prone sleeping positions. Note the lower level of BP and HR in QS versus AS, in both supine and prone sleeping positions. Also note the further reduction in BP, along with a relative HR increase, in QS in the prone position compared to the supine position. ECG, electrocardiogram; EEG, electroencephalogram; EMG, submental electromyogram; EOG, right and left electro-oculogram; HR, heart rate; Resp, abdominal and thoracic breathing movements; SpO2, oxygen saturation; Temp, abdominal skin temperature.

Sleep characteristics

There was no effect of sleep position or PNA on sleep epoch length in either QS or AS at any of the three ages studied (table 1). At 2–4 weeks there was no effect of sleep state on epoch length in either sleeping position. However, at 2–3 months and 5–6 months, QS epoch lengths were longer compared to AS in both the supine and prone sleeping positions (p<0.05).

Table 1 Average epoch length in each sleep sate and position at 2–4 weeks, 2–3 months and 5–6 months postnatal age for 20 infants

Effects of sleep position

The effects of sleeping position on MAP, SAP, DAP and HR are presented in table 2. During QS at 2–3 months, SAP was significantly lower in the prone sleeping position compared to supine (p<0.05). In AS there was no effect of position on MAP, SAP or DAP at this age.

Table 2 Effects of sleep position and sleep state on MAP, SAP, DAP (mmHg) and HR (bpm) during QS supine, AS supine, QS prone and AS prone at 2–4 weeks, 2–3 months and 5–6 months PNA

Sleeping position also had effects on HR. At 2–3 months, HR was significantly higher in the prone position during both QS (p<0.01) and AS (p<0.01). At 5–6 months in QS, HR was higher in the prone position (p<0.05), however there was no effect of position during AS at 5–6 months.

Effects of sleep state

Sleep state had a significant effect on BP at all three ages studied in both sleeping positions (table 2). At 2–4 weeks, 2–3 months, and 5–6 months MAP, SAP and DAP were lower in QS compared to AS in both the supine (p<0.05) and prone positions (p<0.05).

Sleep state had no effect on HR at 2–4 weeks or 2–3 months in either sleeping position. In contrast, at 5–6 months HR was significantly lower in QS compared to AS in both the supine (p<0.05) and prone (p<0.05) positions.

Effects of PNA

The effects of PNA on BP are presented in fig 2. An overall effect of PNA on MAP (fig 2a) and DAP (fig 2c) was identified (p<0.05) with MAP and DAP consistently averaging less at 2–3 months in both sleep states and in both sleeping positions. However the differences were too marginal to isolate the conditions which differed using a post hoc multiple-comparison procedure.

Figure 2 Effects of PNA on MAP (mmHg) at 2–4 weeks, 2–3 months and 5–6 months PNA in the supine and prone sleeping positions during QS and AS. Values are mean (SEM). AS, active sleep; DAP, diastolic arterial pressure; MAP, mean arterial pressure; P, prone; PNA, postnatal age; QS, quiet sleep; S, supine; SAP, systolic arterial pressure.

The effects of PNA on HR are presented in fig 3. HR significantly decreased with PNA during QS and AS while infants slept both supine (p<0.05) and prone (p<0.05).

Figure 3 Effects of PNA on HR (bpm) at 2–4 weeks, 2–3 months and 5–6 months PNA in the supine and prone sleeping positions during QS and AS. Values are mean (SEM). *p<0.05, **p<0.01, ***p<0.001. AS, active sleep; HR, heart rate; PNA, postnatal age; QS, quiet sleep.

Physiological variables

Physiological variables recorded in AS and QS in the prone and supine positions are presented in table 3.

Table 3 Temperature (°C), SpO2 and respiratory rate (breath/min) measured during QS and AS in the prone and supine sleeping positions at 2–4 weeks, 2–3 months and 5–6 months postnatal age

Sleep state had an effect on respiratory rate, which was increased in AS compared to QS in the supine position at 2–4 weeks (p<0.05), 2–3 months (p<0.05) and 5–6 months (p<0.05). In the prone sleeping position, respiratory rates were only increased in AS compared to QS at 2–3 months (p<0.05). Position had no effect on respiratory rates at 2–4 weeks or 5–6 months PNA. However, at 2–3 months respiratory rates were significantly lower in the prone position compared to supine position during AS (p<0.05). Respiratory rate was also significantly affected by PNA with respiratory rate falling significantly across the three ages in both sleep states and sleeping positions.

What is already known on this topic

  • Infant heart rate is altered by sleep state, sleep position and postnatal age, being elevated in active sleep, higher in the prone position, and declining with increasing postnatal age after 1 month.

  • Systolic arterial pressure is decreased in the prone sleeping position in quiet sleep in infants aged 6–9 weeks.

What this study adds

  • Normative data on the effects of sleep state, sleep position and postnatal age on blood pressure and heart rate in healthy-term infants over the first 6 months of life.

  • Systolic arterial pressure is lower in the prone sleeping position during quiet sleep at 2–3 months, coincident with the time of increased risk for SIDS.

Abdominal skin temperature was significantly higher in the prone sleeping position in both QS and AS at 2–4 weeks (p<0.05) and 2–3 months (p<0.05) PNA. However at 5–6 months abdominal skin temperature was higher in the prone position only during QS (p<0.05). There were no effects of PNA on abdominal skin temperature in any sleeping position or sleep state.

SpO2 was not different between sleep states or sleeping positions at any of the ages studied. However, PNA had a significant effect on SpO2 with SpO2 being significantly lower at 2–4 weeks than at both 2–3 months and 5–6 months in both positions in AS and in the supine position in QS.

DISCUSSION

To our knowledge, this was the first study to provide normative data on the development of autonomic BP control, including both cardiac and circulatory parameters, over the first 6 months of life in healthy-term infants during sleep. We found that sleeping position and sleep state altered cardiovascular control and that the affects were age dependent.

Sleeping position

The prone sleeping position altered autonomic cardiovascular control, with BP tending to be lower and HR higher when infants slept prone in both sleep states and at all three ages studied. This effect of position was most marked at 2–3 months PNA when QS prone sleeping induced a fall in SAP by 6 mmHg. This finding is in agreement with data previously reported by Chong et al using intermittent BP measurements in QS in infants aged 6–9 weeks.6 Extending the Chong et al study, we also recorded data in AS, and found that in contrast to QS, prone sleeping in AS did not alter BP. In our study at 2–3 months PNA, MAP and DAP were not altered in the prone position during QS (table 1). The maintenance of MAP and DAP could be explained by the increase in HR in this position at this age; as heart period is shortened the beat to beat fall in DAP would be decreased.

As suggested previously,6 the lower BP in the prone sleeping position may be due to lower vasomotor tone leading to peripheral vasodilatation, in turn reflected in an increase in skin temperature.6 Consistent with previously reported data,1315 we also found that abdominal skin temperature was higher in the prone position in both sleep states at all three ages studied. Adding to the evidence that peripheral vasodilatation occurs in prone sleeping, reduced autonomic vasoconstriction during postural tilting in 1-month-old infants in the prone position has been demonstrated.7 Therefore, the fall in BP may be due to a peripheral vasodilatation to increase heat loss ability. As a consequence, we suggest that the increase in HR in prone sleeping represents a baroreflex compensation to the fall in BP. Notably, in this study the HR increase appeared effective in maintaining BP in AS, but not in QS.

Alternatively, it has also been proposed that the fall in BP in the prone position may be associated with infants simply being in a “deeper sleep”.6 16 Previously, we have shown that at both 2–3 weeks and 2–3 months infants have decreased induced and spontaneous arousal responses in both AS and QS when sleeping prone and these differences were most marked at 2–3 months.14 Thus, infants at this age may have a less disturbed sleep when prone.

Sleep state

In adults, rapid eye movement (REM) sleep is believed to be sympathetically dominated, in contrast to non-rapid eye movement (NREM) which is dominated by the parasympathetic nervous system.17 We also found that sleep state altered autonomic cardiovascular control, with BP being higher in AS compared to QS by 5–9 mmHg in both positions at all three ages studied. Our BP data are consistent with the only other previously reported values recorded during both AS and QS18 19 and in QS6 in infants.

Our studies also showed HR was higher in AS compared to QS in both sleeping positions at 5–6 months. Previous studies have also demonstrated sleep state differences for HR in early infancy.1921 As the differences reported in these studies were all small (3–6 beats/min) and similar to those found in our study, we suggest that the variability between infants may have prevented us obtaining statistical significance at the earlier ages studied.

Effects of PNA

Studies during quiet wakefulness have shown that BP increases from birth to 6 months PNA in full-term infants.22 Although we identified a significant overall PNA effect on BP during sleep, the differences were too marginal to isolate the conditions which differed. However, it was evident that mean BP was consistently lower at 2–3 months (fig 2), and this may have reached statistical significance with a larger sample size. Previous large-scale cross-sectional studies have reported a fall in DAP and constant SAP between 1 and 4 months of age22; consistent with a coincident fall in BP, as we have found. At this age, when SIDS risk is greatest there is a nadir in physiological anaemia.23 Thus, this age could represent a critical time period when the effects of low BP could accentuate decrements in oxygen-carrying capacity and delivery to critical organs.

We identified a consistent pattern of HR maturation, in that HR fell significantly from 2–4 weeks to 2–3 months to 5–6 months PNA. Previous studies have also shown that during sleep, HR falls with increasing PNA between 1 month and 6 months,24 25 suggesting an increase in parasympathetic dominance with increasing PNA.26

In summary, the prone sleeping position alters cardiovascular control in healthy-term infants, inducing a fall in BP which is most prominent in QS at 2–3 months PNA. We suggest that this positional fall in BP may be due to a combination of thermoregulatory stress inducing peripheral vasodilatation, and decreased arousal mechanisms promoting deeper sleep. Previous studies have shown that infants who subsequently succumbed to SIDS had evidence of a pre-existing autonomic dysfunction,27 reduced spontaneous arousals from sleep28 29 and altered medullary 5-HT signalling pathways associated with cardiofvascular and respiratory control, thermoregulation and arousal.30 With these deficits, infants at risk for SIDS may be less able to compensate for the falls in BP associated with the prone position.

In conclusion, our studies have provided normative data on the changes in BP over the first 6 months of life during sleep and highlight that sleeping in the prone position can significantly alter autonomic cardiovascular control in otherwise healthy-term infants, with these alterations being most marked at 2–3 months, coincident with the age of greatest SIDS risk.

Acknowledgments

The authors thank the parents and infants who participated in this study and the staff of the maternity wards at Monash Medical Centre and Jessie McPherson Private Hospital. This project was supported by the National Health and Medical Research Council of Australia Project Number 284357.

REFERENCES

View Abstract

Footnotes

  • Competing interests: None.

  • Ethics approval: The protocol for this study was approved by the Southern Health and Monash University Human Research Ethics Committees.

  • Patient consent: Written parental consent was obtained prior to commencement of the study and no monetary incentive was provided for participation.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.