dynamics of vasomotor thermoregulation of the skin in term and preterm neonates

Early Human Development, 33 (1993) 133-143 @ 1993 Elsevier Scientific Publishers Ireland Ltd. All rights reserved 037%3782/93/$06.00

133

EHD 01410

Dynamics of vasomotor thermoregulation of the skin in term and preterm neonates

Timo Jahnukainen”, Conny van Ravenswaaij-Artsb, Jarmo Jalonena and Ilkka V%limBkia

‘Cardiorespiratory Research Unit, University of Turku (Finland) and bDepartment of Paediatrics, University Hospital Nijmegen (Netherlancis)

(Accepted 24 March 1993)

Summary

Eighteen fullterm infants and 17 preterm infants were studied on their 3rd day of life to investigate the reactivity of skin blood flow to thermal stimulation. The infants were studied during quiet sleep. After a lo-min control period a constant air current was used to synchronise the external cutaneous stimulus to the distal lower extremity of each infant: the heating element of an air blower was automatically switched on and off to generate successive warm and cool periods of equal duration (5 cycles/min). Heart rate (HR), skin blood flow (SBF) and respiratory waveform signals were recorded and their variability was analysed using the fast Fourier trans- form and spectral analysis. Fullterm infants showed a clear response to external thermal stimulation: both HR and SBF were synchronised to the stimulation frequency. A response of preterm infants was present but it was markedly attenuated in comparison to term infants. The effect of stimulation did not seem to be dependent on postnatal age. The results suggest that the vasomotor control is immature in preterm infants.

Key words: heart rate variability; skin blood flow; newborn infant; autonomic control

Introduction

The cutaneous vasculature is morphologically immature at birth: there is a disorderly capillary network which gradually develops into an organised sub-

Correspondence IO: T. Jahnukainen, Cardiorespiratory Research Unit, University of Turku, Kiinamyllynkatu 10, SF-20520 Turku, Finland.

134

papillary plexus and papillary loops during the first weeks of life [20]. Still the thermoregulation of skin blood flow is considered to be mature and functional in the newborn [4].

Like all biological control systems with a negative feedback mechanism, the cutaneous thermoreflex oscillates [8,12]. Since the time-delay in this feedback loop is fairly long, the oscillation frequency of vasomotor thermoregulation is low. The peripheral vascular oscillations have natural frequency components over a frequency range of 0.01-o. 1 Hz under spontaneous conditions [ 13,221. External thermal stimulation by repeated rhythmic warming and cooling of the skin causes entrainment of the multiple spontaneous blood flow oscillations as long as this periodic stimulation recurs at a stable frequency of 0.01-0.1 Hz. This entrainment causes contractions and relaxations of the skin blood vessels synchronous with the stimulus frequency, while spontaneous oscillations close to the stimulus frequency are attenuated t&12,13]. This thermal entrainment can be studied by measuring the variability of the signals of skin blood flow, blood pressure or heart rate [ 14,191. Using this kind of cutaneous stimulation Lindqvist et al. [ 171 showed that entrainment of the heart rate variability (HRV) in both term and preterm infants takes place. The response to stimulation tended to increase with maturity. The entrainment of the HRV at the thermal stimulation frequency as found by Lindqvist et al. is apparently secondary to the entrainment of the oscillations of the peripheral vascular resistance (i.e. skin blood flow). Quantitative monitoring of the skin circulation has recently become possible by direct non-invasive laser Doppler flowmetry [ 1,241. This technique together with modern computerised signal analysis methods have opened new ways of studying quantitatively and non-invasively cutaneous vascular dynamics. In the present report the influence of periodic thermal stimulation on skin blood flow (SBF) and heart rate was studied in term and preterm infants on their third day of life. The aim was to obtain insight in the maturation of the dynamics of human cutaneous vasomotor thermoregulation.

Subjects and Methods

Subjects Eighteen fullterm (gestational age 39 weeks) and 17 preterm newborn babies

(gestational age 33.5 weeks) with an uncomplicated neonatal course were studied on their 3rd day of life at the Turku University Hospital. To exclude the effects of postnatal adaptation, 10 of these term infants were studied also on their postnatal days 1 and 2 and eight of the total number of 18 preterm infants were studied once again at the age of 1 week. Only healthy infants appropriate for gestational age (AGA) were included in the study. The mean birth weight was 3735 g for the term infants and 2300 g for the preterm infants. Babies had not received any vasoactive or sedative drugs or had not needed any resuscitation procedures at birth. The moth- ers had not received vasoactive or sedative drugs at any time of pregnancy or parturi- tion. Parental permission was obtained for the study of each infant. The study was also approved by the ethical committee of the hospital.

135

Signal acquisition Since baseline SBF decreases [24] and HRV increases [27] with age, all infants

were studied on the third day of life. Measurements were performed 1 h after feeding, while the infant was in quiet sleep. The infant was observed continuously by one of the investigators and body and eye movements were documented. The babies were studied in open cribs at room temperature. They were wrapped in a blanket except for a lower extremity onto which the thermal stimulation was applied. Electrodes for recording the ECG and respiratory waveform (transthoracic impedance, Corometrics 512 Neonatal Monitor, Corometrics Medical System Inc., Wallingford, CT) were placed in such a way that optimal R-peaks were obtained (usually on the left and right side of the trunk and on the lower abdomen or back). The laser Doppler transducer (Periflux Pf 2B, Perimed KB, Stockholm, Sweden) was attached to the forehead [24]. The laser Doppler technique is based on the measurement of the Doppler shift of light, which is the result of scattering by moving par- ticles [ 11. Laser Doppler penetrates several millimeters into the tissue. Consequently, the blood flow signal is derived from capillaries, mainly from thermoregulatory microvessels [ 151.

All signals were recorded on magnetic tape by a four-channel FM-recorder (Racal Thermionic Store 4 DS, Racal Recorders Ltd, Hythe, Southampton, UK) under visual control on an oscilloscope. Recordings were made (1) before stimulation (10 min), (2) during thermal stimulation (10 min) and (3) again for 10 min of rest. Rhythmic thermal stimulation was produced by stimulating the skin of the baby’s foot with two air blowers which changed the temperature of the air; the blowers were controlled with a programmable automatic control unit. The duration of the cold and warm periods could be selected with an accuracy of 1 s. The stimulation frequency in this study was 5 cycles/min (0.08 Hz), and thus the alternating warm and cold periods had a duration of 6 s each, which has previously been found to be appropriate for thermal cycling [ 171. The amplitude of the fluctuation of the air temperature was about 7°C. The air blowers were allowed to run during the whole experiment (total time 30 min) in order to allow the infants to get accustomed to the hum of the air blowers.

Signal analysis The signals were divided into segments of 2 min preprocessed for further analysis.

The recording was replayed at eight times the real speed, while the ECG was subjected to R-wave detection after band-pass filtering by a voltage threshold trig- ger, under visual quality control (Gould DSO 1604, Roebuck Road, Ilford, Essex, UK). The time-intervals between successive R-waves were measured with an interval counter operating at a clock frequency 1 kHz which gives an accuracy of about 1 ms. The reciprocal of each R-R interval was calculated to obtain the instantaneous heart rate. After low-pass filtering to avoid aliasing, the respiration and blood flow signals were digitised with a sample-frequency of 33 Hz. Subsequently, the signals were fed into an Eclipse MV4000 laboratory computer (Data General Corp., Southboro, MA). Before spectral analysis the instantaneous heart rate, skin blood flow and respiration signal were plotted for inspection of data quality (Fig. 1). Soli-

136

A HRIBPMI

6 6‘0 li0 SBF

205-

SBFV[AUl.lO’ 70

FREOUEKY lHz1

I3 HI? fBPt.41

11L

6 $0 li0 SBF

710

170

b $0 1;o TIME Is1

I-W I AU1 SBFV MUI ~10’

FFXWENCY IHz 1

TERM INFANT

HR I ePMl 155-

205-

TIME is.1

HRV[AUI SEFVIAUIxlO’

FREOUENCY 1 Hz1

HR lf3PMl 155-

1254 I 0 60 120

SBF

365-

20%

b $0 120 TIME Is1

HRVIAUI SBFV htJI.10’

FREQUENCY (Hz1

PRETERM INFANT

Fig. 1. Signals and spectra of one fullterm infant (left) and preterm infant (right) during control (A) situ- ation and thermal stimulation (B). The arrows show the stimulation frequency (AU, arbitrary units; BPM, beats/min).

137

tary R-R intervals that diverged markedly from the adjacent intervals were corrected. The number of corrected intervals never exceeded 5% of the total number of intervals per 2-min segment.

Data reduction was performed according to the Nyquist criterion by taking every 8th point of the SBF signal. This resulted in a final sampling rate of 4.125 Hz instead of 2 Hz, which was the upper limit of interest. The heart rate signal was also inter- polated and sampled at 4.125 Hz. Trend correction was performed by computing the 4th order polynomial least-squares approximation and subtracting the polynomial trend from the original source data. The auto-covariance functions of heart rate and skin blood flow were computed by Fast Fourier transformation. The cross-spectra between HR and SBF were computed to obtain the cross-covariance functions and to assess simultaneous frequency-specific variations of these signals. To reduce noise a spectral (Tukey) window was applied to smoothen the spectral estimator [lo].

To compare the results before, during and after thermal stimulation one artifact- free 2-min period was selected and the power spectral density was integrated over frequency in the following six spectral bands: (1) 0.025-0.060 Hz, (2) 0.060-0.100 Hz, (3) 0.100-0.200 Hz, (4) 0.200-0.400 Hz, (5) 0.400-0.700 Hz and (6) 0.700-1.0 Hz. The spectral contents of the frequency bands of the heart rate and SBF power spectra were calculated in terms of absolute and relative (percentage of the total variability power) values. The slowest oscillation that can be appropriately analyzed in a 120-s period has a periodicity of 40 s (0.025 Hz) [2S]. Band 2 was around the stimulation frequency of 0.08 Hz. Also the baroreflex-related HRV of newborn infants occur around this frequency [7]. Beside baroreflex and sympathetically mediated vasomotor thermoregulation, also respiration is known to cause HRV. Respira- tion induces heart rate fluctuations at a frequency equal to the breathing rate (respiratory sinus arrhythmia) and thus will be reflected in bands 4-6. Periodic breathing with a frequency of 0.05-o. 15 Hz is often found in (preterm) newborns and this is accompanied by synchronous heart rate fluctuations (bands 2 and 3) [7].

Statistical methods Wilcoxon’s test was used to compare control versus stimulation or post-

stimulation periods. The spectral contents of the stimulation and post-stimulation periods were always expressed as compared to the spectral content during the control period. The dependency on gestational age of the difference between stimulation and control periods was analysed with the Kruskal-Wallis test, while analysis of covariance for repeated measurements was used to study the age dependency in 10 term and eight preterm infants.

Results

Procedure and behavioural pattern The ambient temperature and humidity were recorded at the beginning and end

of each experiment. The humidity did not change (median difference O%, range -2 to +l.S%), but the room temperature tended to increase (median difference 0.6”C, range -0.7 to +1.6”C).

138

TABLE I

Mean heart rate, skin blood flow and respiratory rate (standard deviation) during control, stimulation and post-stimulation periods (n = 20). AU, arbitrary units.

Thermal control Stimulation Post-stimulation

Heart rate (mm-‘) 118 (19) 119 (17) 120 (17) Skin blood flow (AU) 467 (135) 506 (191) 547’ (266) Respiratory rate (mm-‘) 43 (12) 44 (10) 42 (8)

In general the infants responded to the thermal stimulation by moving less and increasing the occurrence of quiet sleep. Term infants tended to wake especially when the thermal stimulation was stopped. Data from the post-stimulation periods were available in only 10 of the 18 term and in 10 of the 17 preterm infants. The mean heart rate, respiration and skin blood flow before, during and after stimulation

HEART RATE POWER SPECTRUM HEART RATE POWER SPECTRUM

Arbitrary units * _. *

--I * I___(

18 term infants

0 t 0.2 0.4 0.6 0.8 1

Frequency I Hz1

17 Preterm infants

t I +-____-_- I 0 t 0.2 0.L 0.6 0.8 1

Frequency I Hz]

SKIN BLOOD FLOW POWER SPECTRUM SKIN BLOOD FLOW POWER SPECTRUM

Arbitrary units I I

18 term infants

- 0.L 06 0.8 1

Frequency I Hz 1

Arbitrary units

17 Preterm infants

0.6 0.8 1 I

Frequency I Hz1

-Control ----Stimulation

Fig. 2. Absolute variation in heart rate and skin blood flow during control (solid lines) and stimulation (dotted lines). The arrow shows frequency band which included stimulation frequency. Wilcoxon’s test: *P < 0.05, **p < 0.01.

139

are shown in Table I. No significant differences in HRV and SBF oscillations with a frequency below 0.2 Hz were found between control and post-stimulation periods (N = 20).

Response to thermal stimulation In some of the recordings spontaneous SBF oscillations were visible during the

control periods. These spontaneous oscillations had a frequency between 3 and 4/min in the youngest pretext infants (32-33 weeks), between 4 and 6/min in the less preterm infants (34-36 weeks) and between 5 and 7/min in the term infants. The frequency of these spontaneous oscillations changed towards the stimulation frequency (6/min) and the signal amplitude tended to increase when thermal stimulation was applied. Also seven of the nine preterm infants and nine of the 11 term infants who did not show spontaneous SBF oscillations generated oscillations around 0.08 Hz during the thermal stimulation.

The effect of stimulation on the absolute spectral density of the frequency bands of the heart rate and SBF are shown in Fig. 2 in infants 3 days old. The absolute power of the heart rate and SBF oscillations around the stimulus frequency increased significantly in the term infants only. However, also in preterm infants a significant redistribution of the heart rate and SBF power spectra occurred with more power located around 0.08 Hz during the thermal stimulation. The absolute frequency-specific heart rate response to the stimulation was significantly greater in the term than in the preterm babies (P < 0.05).

The effect of thermal stimulation on HRV and SBF oscillations below 0.2 Hz showed no age-dependency in either term (postnatal days 1, 2 and 3; n = 10) or preterm (days 3 and 7; n = 8) infants. An effect of thermal stimulation on heart rate was already present on day 1; it was significant on days 2 and 3 in the 10 term infants that were measured on days I, 2 and 3.

Discussion

Methodological considerations The study group was directed to AGA infants, because in our former study SGA

infants showed less HRV response to thermal stimulation than AGA infants [17]. Further SGA infants are known to have a delayed increase of HRV with postnatal age [27].

Several studies that have compared laser Doppler flowmetry with other direct and indirect methods for the measurement of skin blood flow have demonstrated reason- able correlations [ 11. In the present study, all infants were studied under stable conditions with regular respiration and no body movements. This is important because the Periflux Doppler flowmeter is rather sensitive to movements. Further Suichies et al. [24] found a significant increase in the mean level and variability of skin blood flow during active sleep compared to quiet sleep in term newborns.

In general, the condition of the infants remained stable during the thermal stimulation. However, many infants were aroused when the stimulation was discontinued. Cessation of the stimulation was achieved by turning the air jet away from the baby; the device was not switched off. Therefore, the noise level did not change. Mean

140

heart rate, skin blood flow and respiration remained constant during the experi- ments. Thus these variables cannot explain the effect of thermal stimulation on HRV and skin blood flow oscillations.

The spectral analysis of signal variability provides repeatable detection of frequency-specific periodic components of variation in cardiorespiratory signals. It is important that the studied signals are stationary i.e. there are no global trends in the signal or its overall variation and signal-to-noise ratio is good. After the signals have been A/D converted the fast Fourier transformation is done in order to com- pute autospectra which display quantitatively the components of variation in the source signals. Interactions between two source signals may be assessed by cross- spectral analysis which displays the magnitude of joint-variability in the two signals. Synchronization of the periodicities may be quantified by computing the respective coherence spectrum i.e. the coherence is a measure of time-linkage between the signals. The spectra can be finally divided into frequency bands. The area under the spectral curve of these frequency bands is comparable to the variation at that frequency range. We found the method of spectral band integration convenient for the inter-group study of frequency-selective responses to the thermal stimulation.

The effect of thermal stimulation In adults, Burton [5] used linger plethysmography to show that thermoregulation

produces slow fluctuations in peripheral blood flow with a periodicity of 15-120 s (0.008-0.07 Hz). Rosenbaum [22] studied the delay-time and overshooting of peripheral vasoconstriction after sympathetic stimulation in the dog and found a mean resonance or intrinsic frequency of peripheral vascular resistance of 0.034 Hz. This was confirmed by Kitney [ 131 who showed by computer simulation that intrinsic peripheral blood flow oscillations have a frequency of 0.038 Hz. The frequency components found by Burton [5] probably arise from the interaction between the fluctuations in peripheral blood temperature and the intrinsic oscillation of the thermoregulatory control system [ 131.

In newborn infants spontaneous oscillations of skin blood flow have been recorded by Piischl et al. [21]. These oscillations are present on the 4th postnatal day. The oscillation frequency was higher in term infants (3/min to 6/min) than in preterm infants (l/min to S/min). We observed similar spontaneous oscillations in term and preterm infants. The frequency of these oscillations increased with gestational age and approached the natural oscillation frequency of the baroreflex in heart rate. In newborns the baroreflex-related heart rate oscillations have a frequency of approx- imately 0.06-0.08 Hz [7].

The slow periodic vasoconstriction is accompanied by an initial increase of heart rate. Thus a discharge of impulses in the cardiac sympathetic nerves occurs simultaneously with sympathetic peripheral vasoconstriction. The vasoconstriction subsequently induces a rise in blood pressure and a baroreflex-mediated decrease of heart rate. Burton [5] demonstrated this biphasic response of heart rate to the cutaneous vasoconstrictions in man already in 1939.

In adults, thermal stimulation at 0.1 Hz entrains the periodic heart rate variability (HRV) by synchronizing the oscillations at 0.1 Hz. This indicates that influences of thermoregulation and baroreceptors are coupled in the heart [ 181, since the latter are known to occur at 0.1 Hz in the adult [8].

141

In analogy with the studies in adults Lindqvist [ 171 found a greater response in newborn babies to a stimulation frequency of 0.08 Hz than to 0.04 or 0.025 Hz. So, also in newborns there appears to be coupling between thermoregulatory and baroreflex-related heart rate fluctuations (around 0.07 Hz). The HRV is influenced by thermal stimulation probably through fluctuations in skin blood flow. Indeed, in our group of 18 term infants, the increase in the spectral density of heart rate variability around the stimulation frequency, was parallelled by an increase in the spectrum of skin blood flow variability. An increase in the joint-variability around 0.08 Hz was also found in the respective cross spectrum (Table II), and the squared coherence spectrum of heart rate and SBF produced a significant synchronisation of the signals.

Maturation Lindqvist [17] found in neonates that the impact of thermal stimulation on

periodic heart rate variability increased with increasing maturity. Our preterm infants showed also a lower response to stimulation in terms of the variability of both absolute heart rate and skin blood flow oscillations. However, the redistribution of the heart rate and SBF variability towards the stimulus frequency was significant in both term and preterm infants. This proves that preterm infants are able to respond to thermal stimulation, although the magnitude of the response as regards both heart rate and SBF is attenuated in comparison to term infants. The heart rate power differed significantly between the term and preterm infants when the stimulation response was studied.

Stimulation in preterm infants was less pronounced; this might be due to immaturity of the autonomic nervous system. The time-delay of the baroreflex is longer in preterm than in term infants. Therefore, the optimal stimulus frequency may be a little lower in preterm than term infants. Preterm infants did respond on stimulation, albeit less markedly than the term infants. Thus thermoregulation might be less mature in preterm babies, leaving them in a poorer position in comparison to term infants. Also immature sympathetic innervation of the cutaneous vessels may cause the attenuated response of the SBF oscillations in preterms [23]. Besides immaturity of the autonomic nervous system in general, also enhanced activity of the renin- angiotensin system and delayed postnatal adaptation may influence the vasomotor

TABLE II

Appearance of oscillation at the thermal stimulation frequency on heart rate and skin blood flow variability on third day of life. (HR x SBF cross spectrum (control/stimulation)).

Group N 0.25-0.06 0.06-O. 1 0.1-0.2 Hz

Preterm 17 18.9/10.7 6.219.3 10.7/13.6 Term 18 12.0112.7 6.7/15.1** 19.U25.8’

Spectral density at first three frequency bands. The influence of thermal stimulation has been calculated as the absolute spectral content during control and stimulation period. The cross-spectrum (HR*SBF) displays the magnitude of similar periodicities in heart rate and skin blood flow signals against frequency. Significance of the difference between control and stimulation: *P < 0.05, **P < 0.01. (Wilcoxons test).

142

response in preterm infants. The renin-angiotensin system is known to attenuate the fluctuations in peripheral vascular resistance in the adult dog [2]. In the newborn, especially in the preterm infants, the renin-angiotensin system appears to be hyper- active [3]. The influence of the neonatal renin-angiotensin system on HRV or SBF fluctuations is not known.

Postnatal adaptation The effect of thermal stimulation on heart rate and SBF oscillations appeared to

be independent of the immediate postnatal age in of a subgroup of the infants studied. Although Briick [4] suggested that the skin blood flow responds less well to temperature on the first postnatal day than a week later, an effect of thermal stimulation was visible already on the very first day of life in our term infants. In con- trast, this effect was not present on day 7 in the preterm infants. Therefore we believe that the attenuated response to thermal stimulation is not likely to be due to delayed postnatal adaptation during the first week of life.

We describe a convenient quantitative method to assess the vasomotor capacity of the newborn. A thermoregulatory response is present in both term and preterm neonates, but it is more pronounced in the term infants. The ability to regulate skin blood flow by vasomotion is important for the prevention of hyper- or hypothermia. Entrainment of the spontaneous skin blood flow fluctuations by external thermal stimulation, resulting in a redistribution of periodic variability of heart rate and skin blood flow, occurs in both term and preterm infants but is different in magnitude. The lower response of preterm babies to thermal stimulation may be due to a less mature skin blood flow regulation.

Acknowledgement

This study has been supported by the Varsinais-Suomi Regional Fund of the Finnish Cultural Foundation. The participation of CvR has been possible by a De IBM Frye award of the University of Nijmegen.

References

1 Aarnoudse, J.G. (1991): Use of Doppler flowmetry for perinatal assessment. In: Fetal and Neonatal Physiological Measurement, pp. 137-144. Editor: H.N. Lafeber. Elsevier, Amsterdam.

2 Akselrod, S., Gordon D., Ubel F.A., Shannon D.C., Barger A.C. and Cohen R.J. (1981): Power spectrum analysis of heart rate fluctuations: a quantitative probe of beat-to-beat cardiovascular control. Science, 213, 220-222.

3 Broughton-Pipkin, F. and Smales, O.R.C. (1977): A study of factors affecting blood pressure and angiotensin II in newborn infants. J. Pediatr., 91, 113-l 19.

4 Brilck, K., Brilck, M. and Lemtis, H. (1961): Temperature regulation in the newborn infant. Biol. Neonate, 3, 65-l 19.

5 Burton, A.C. (1939): The range and variability of the blood flow in the human fingers and the vasomotor regulation of body temperature. Am. J. Physiol., 127, 437-453.

143

6

8

9

14

15

16

17

18

19

20

21

Gadzinowski, J., Moczko, J., Breborowicz, C., Twanowska, 1. and Slomko, Z. (1988): Heart rate variability in neonates with respiratory distress syndrome: an analysis in frequency domain. In: XIth European Congress on Perinatal Medicine, Rome, pp. 299. Giddens, D.P. and Kitney, RI. (1985): Neoantal heart rate variability and its relation to respiration. J. Theor. Biol., 113, 759-780. Hyndman, B.W., Kitney, RI. and Sayers, B.Mc.A. (1971): Spontaneous rhythms in biological control systems. Nature, 233, 339-341. Jenkins, J.G., Reid, M.Mc.C. and McClure, B.G. (1980): Study of heart rate variability in sick newborn infants. Acta Paediatr. Stand., 69, 393-396. Jenkins, G.M. and Watts D.G. (1969): Spectral Analysis and its Applications. Holden-Day, San Francisco. Kero, P. (1974): Heart rate variation in infants with respiratory distress syndrome. Acta Paediatr. Suppl., 250, l-70. Kitney, R.I. (1975): An analysis of the nonlinear behaviour of the human thermal vasomotor control system. J. Theor. Biol., 52, 231-248. Kitney, R.I. (1974): The analysis and simulation of the human thermoregulatory control system. Med. Biol. Eng., 12, 57-65. Kitney, R.I. (1975): Entrainment of the human RR interval by thermal stimuli. J. Physiol. Lond., 252, 37-38P. Kolari, P.J. (1985): Penetration of unfocussed laser light into the skin. Arch. Dermatol. Res., 277, 342-344. Korvenranta, H., Kero, P. and ViilimPki, I. (1983): Cardiovascular monitoring in infants with respiratory distress syndrome. Biol. Neonate, 44, 138-145. Lindqvist, A., Oja, R., Hellman, 0. and Vllimiiki, I. (1983): Impact of thermal vasomotor control on the heart rate variability of newborn infants. Early Hum. Dev., 8, 37-47. Lindqvist, A., Parviainen, P., Kolari, P., Tuominen, J., VHlimiiki, I., Antila, K. and Laitinen, L. (1989) non-invasive method for testing neural circulatory control in man. Cardiovasc. Res., 23, 262-272. Lindqvist, A., Jalonen, J., Parvainen. P., Antila. K. and Laitinen. L.A. (1990): Effect of posture on spontaneous and thermally stimulated cardiovascular oscillations. Cardiovasc. Res., 24, 373-380. Perera, P., Kurban, A.K. and Ryan, T.J. (1970): The development of the cutaneous microvascular system in the newborn. Br. J. Dermatol., 82 Suppl. 5, 86-91. Piischl, J., Weisch, T., Diehm, C. and Linderkamp, 0. (1991): Periodic variations in skin perfusion in full term and pretetm neonates using laser Doppler technique. Acta Paediatr. Stand., 80, 999-1007.

22 Rosenbaum, M. and Race, D. (1968): Frequency-response characteristics of vascular resistance vessels. Am. J. Physiol., 215, 1397-1402.

23 Rowe& L.B. (1977): Reflex control of the cutaneous vasculature. J. Invest. Dermatol., 69, 154-166. 24 Suichies, H.E., Brouwer, C., Aarnoudse, J.G., Jentink, H.W., de Mul, F.F.M. and Greve, J. (1990):

Skin blood flow changes, measured by laser Doppler flowmetry, in the first week after birth. Early Hum. Dev., 23, l-8.

25 VBlimHki, I., Aarimaa, T., Metsala, T., Parviainen, P. and Lindqvist A. (1990): Variability of cardiovascular signals duing the perinatal period-acts and traps in spectral analysis. In: Progress in Peri- natal Medicine, pp. 101-l 13. Editors: D.F. Hawkins, G.C. DiRenzo, and E.V. Cosmi. Harwood Academic Publishers, Singapore.

26 Van Ravenswaaij-Arts, C., Hopman, J.C. W., Kollee, L.A.A., Van Amen, J.P.L., Stoelinga, G.B.A. and Van Geijn H.P. (1991): The influence of respiratory distress syndrome on heart rate variability in very preterm infants. Early Hum. Dev., 27, 207-221.

27 Van Ravenswaaij-Arts, C., Hopman, J.C.W., Kollte, L.A.A., Van Amen, J.P.L., Stoehnga, G.B.A. and Van Geijn H.P. (1991): Influences on heart rate variability in spontaneously breathing preterm infants. Early Hum. Dev., 27, 187-205.

dynamics of vasomotor thermoregulation of the skin in term and preterm neonates

Documents