Early Human Development 52 (1998) 49–66

The effect of thermal stimulation on the heart-rate variability in neonates a b b a, O. Shefi , S. Davidson , A. Maayan , S. Akselrod * a

Department of Medical Physics, School of Physics and Astronomy, Tel Aviv University, Tel Aviv, Israel b Department of Neonatology, Rabin Medical Center, Beilinson Campus, Tel Aviv University, Tel Aviv, Israel Received 5 December 1997; accepted 23 December 1997

Abstract Thermoregulation in humans can be divided into three broad mechanisms of control, namely: shivering, sweating and vasomotor activity. Previous investigations suggested the presence of an autonomic rhythm, originating in the central nervous system, possibly related to thermal vasomotor control and directly affecting heart rate by reflex changes in cardiac sympathetic and parasympathetic activity. The objective of the present work was to study the maturation process of the thermoregulatory system in newborns. We used peripheral thermal entrainment and focused on the reflections of vasomotor control in the heart-rate (HR) power spectrum (PS). The study included three groups of neonates at three different ages: 10 premature infants, 6 full-terms and 7 older infants (4 to 6 months). Thermal stimulation was achieved by placing a hot and cold surface on the subject’s right palm alternately at three different rates: replacing the touching surface every 4 s (0.25 Hz), 7 s (0.14 Hz) and 10 s (0.1 Hz). ‘Double period’ stimulation was defined as the total duration of each period of hot and cold stimulation at the three rates, namely 8 s (0.125 Hz), 14 s (0.07 Hz), 20 s (0.05 Hz). The ECG of every infant was measured and recorded during the various stages of the experiment. The HR power spectrum from 0.02 Hz up to 2.00 Hz was considered, focusing on narrow ranges around the thermal stimulation frequencies. We found that in most subjects, clear peaks arise in the HR PS at the thermal entrainment frequency and its corresponding half frequency (‘double period’). In premature infants, the reaction is best in response to the longest (10 s) stimulus (9 out of 10 prematures reacted positively), in group B (full-term infants) the reaction responds best to the 7-s stimulus (6 out of 6 reacted) and in older infants the reaction is slightly better at the 4-s stimulus. Since sympathetic control is slower, this ability to entrain the control system at increasing frequencies, might be related to the gradual maturation of parasympathetic control *Corresponding author. Fax: 00 972 3 6429306. 0378-3782 / 98 / $19.00  1998 Elsevier Science Ireland Ltd. All rights reserved. PII: S0378-3782( 98 )00005-X

50

O. Shefi et al. / Early Human Development 52 (1998) 49 – 66

after birth. The different reaction of the three groups may help to understand the maturation process of the thermoregulation system.  1998 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Thermal stimulation; Heart-rate; Neonates

1. Introduction Thermoregulation in humans is achieved mainly by three control mechanisms: shivering, sweating and vasomotor activity [10]. In the adult human, shivering is the most significant involuntary mechanism of thermogenesis; nonshivering thermogenesis occurs in the newborns [16,18]. At a temperature above the thermoneutral zone, thermal balance is achieved by vasomotor adjustment and evaporation of sweat. As ambient temperature falls below the lower limit of thermoneutrality, heat production by shivering progressively increases. The control of peripheral blood flow is the principal thermoregulatory effector process in a thermally neutral environment. The peripheral blood vessels of a normal unanesthetized human subject are in a state of continuous fluctuation, reflecting a rhythmic fluctuation in the nervous outflow from the vasomotor center [17,8]. Monitoring the changes in the digit blood flow has shown that spontaneous frequency components in the range below 0.1 Hz can be attributed to thermal vasomotor control [10]. Analysis applied to the thermal vasomotor control systems has led to the conclusion that the spontaneous fluctuations in digit blood flow are the consequence of a nonlinearity of the system [10]. In his investigation of nonlinear oscillatory systems Minorsky (1962) described an important phenomenon, the ‘entrainment frequency’ that occurs in such systems [10]. When two linear oscillatory systems are coupled and operate at different frequencies, beating occurs. When the oscillators are nonlinear, as in thermal vasomotor activity, and display a small difference in frequency, the coupling may cause the two oscillators to lock into a common frequency [10]. Previous findings have suggested the presence of an autonomic rhythm in vasomotor control, originating in the central nervous system, which may be related to temperature regulation. This rhythm regulates blood flow through skin arteriovenous anastomoses according to the thermal balance of the organism. Such changes affect heart rate by means of reflex changes in cardiac sympathetic and parasympathetic nervous activity. According to this approach, vasomotor changes can be reflected in HR variability and can be displayed and quantified by the Power Spectrum of heart-rate fluctuations [11,12]. In order to investigate thermal effects on HR control, Kitney et al. (1974) [9] exposed adult volunteers to thermal stimuli of 20-s periodicity. The power spectrum of beat to beat intervals calculated from the finger blood volume signal obtained during the 20-s thermal stimulus, showed a large single component at this frequency. The effect of thermal entrainment on HRV in premature infants, small for gestation neonates and full-term infants was investigated by Lindqvist et al. [14]: in 14 out of 17 patients the power in the stimulated frequency band doubled, at least, in size,

O. Shefi et al. / Early Human Development 52 (1998) 49 – 66

51

independently of maturation. Halthorn et al., studied HR changes in 25 term infants during periodic thermal stimulation (0.05 to 0.15 Hz). They reported an increase in spectral power at the frequency of stimulation [5]. Jahnukainen et al. (1996) observed in preterm, term and small-for-age term neonates, redistributed oscillations of Skin Blood Flow due to rhythmic thermal stimulation in addition to the synchronized periodic HR variability response in part of the groups [8]. The objective of the present work was to study the maturation process of the thermoregulatory system, focusing on the response to peripheral thermal stimulation. We used the power spectrum of heart-rate signal during thermal entrainment as a measure of maturation in three specific groups of neonates. We consider not only the stimulation frequency related to the frequency of alternating the hot and cold temperature (T ), but also the frequency corresponding to what we call ‘double period’, or full duration of ‘hot and cold’ stimulus. Our working hypothesis was, that the changes in the HR power spectrum as a function of age, in response to peripheral thermal stimulation at various frequencies, should allow us to better understand the functioning and maturation of the thermal regulation, related to vasomotor control.

2. Methods

2.1. Subjects Three groups of neonates participated in the study: Group A: 10 healthy premature infants (6 females and 4 males), born at the Beilinson Medical Center between January 1993 and June 1993. The group had mean gestational age of 31 weeks (range 29–33 weeks) and mean birth weight of 1430 g (range 1270–1940 g). At the time of the study, the mean weight was 1970 g (range 1800–2390 g) and the post-natal age was 35.5 weeks (ranging from 34 to 37 weeks). Group B: 6 healthy full-term neonates (3 females and 3 males, born at Beilinson Medical Center around the same period), with mean gestational age of 39.1 weeks (range 38–40 weeks), and mean birth weight of 3650 g (range 3210–4550 g). The subjects in this group were 24–48-h old. Group C: 7 healthy infants, with mean age of 5 months (range of 4.5–6 months), mean birth weight of 3010 g (range of 2550–3500 g) and mean weight at time of the study of 6990 g (range 5480–8700 g. The infants were fed one hour before the test, dressed and kept in supine position in their own bed. The test was performed at room temperature.

2.2. Temperature entrainment Thermal stimulation was achieved by placing a hot and cold surface on the subject’s right palm alternately. The hot stimulus was achieved with a bag (a product of ‘Flexxum compress’), that was kept in warm water at 458C. To keep the bag’s

52

O. Shefi et al. / Early Human Development 52 (1998) 49 – 66

temperature constant, we replaced it every 1 min. Water temperature was measured with a thermometer (618C). The cold stimulation surface (a similar bag) was maintained at 108C. We used tested thermistor probes (Yellow Springs Instrument, Ohio, Series 400) to monitor the temperature; the probe agreed within 0.18C with the requested temperatures.

2.3. Experimental protocol The ECG of every subject was monitored with a Mennen Medical 740 neonatal monitor and recorded on Teac-R71 analog magnetic tape throughout the various stages of the experiment. The duration of every stage was 7 min, according to considerations of trace length (two slightly overlapping 256 s traces) and spectral analysis (lowest frequency of interest 0.02 Hz). The experiment included the following stages: 1. Baseline, without stimulation. 2. Alternately placing a hot (458C) and cold (108C) stimulation surface on the right palm of the subject at a rate of 0.25 Hz (switching the temperature every 4 s). 3. Same thermal stimulation but at a rate of 0.14 Hz (switching the temperature every 7 s). 4. Same thermal stimulation at a rate of 0.1 Hz (switching the temperature every 10 s). 5. Recording without thermal stimulation (baseline). The recordings were made while the subjects were in relaxed condition. In general, the prematures and full-term neonates were asleep (quiet sleep) during the experiment. The subjects in group C were awake most of the time, but the recordings were taken while relaxed and quiet. The subjects in all groups maintained mean body temperature within the normal range (60.58C) during the experiment. The study was approved by the Human Subjects Review Committee of the Beilinson Medical Center and informed parental consent was obtained for each infant.

2.4. Heart-rate data analysis Spectral analysis of fluctuations in heart rate was performed off-line on 256-s long segments of the ECG. This allowed to analyze two slightly overlapping traces within the 7-min trace recorded during the experiment. A / D conversion of these ECG segments was performed at a frequency of 300 Hz. The peaks of the QRS complex were detected and a smooth HR time series was created and sampled at equal time intervals. Mean HR was subtracted. Next, the heart-rate power spectrum (see Fig. 1a) was calculated by means of the Fast Fourier transform of the heart-rate time series, squaring the absolute values and dividing by (mean HR)2 .

O. Shefi et al. / Early Human Development 52 (1998) 49 – 66

53

Fig. 1. (a) A typical heart-rate time series and the corresponding power spectrum of heart-rate fluctuations for one premature (subject 8 group A) at the baseline stage. (b) A typical heart-rate time series and heart-rate power spectrum for one full-term neonate (subject 5 group B) at the baseline stage. (c) A typical heart-rate time series for one older infant and the corresponding power spectrum for one older infant (subject 3 group C) displaying regions of power both in the high-frequency band and in the low-frequency band.

54

O. Shefi et al. / Early Human Development 52 (1998) 49 – 66

2.5. Frequency ranges of interest We investigated the HR power spectrum from 0.02 Hz up to 2.00 Hz. The examined bandwidth (0.02–2.00 Hz) was divided into several relevant frequency bands. The low (0.02–0.2 Hz) and the high (0.2–2.00 Hz) frequency bands were considered, as well as the total power (over 0.02–2.00 Hz). The main ranges of interest were narrow bands around the frequencies of thermal stimulation. The corresponding band for the entrainment at 0.25 Hz was 0.23–0.27 Hz, for 0.14 Hz the band was 0.12–0.16 Hz and for the entrainment at 0.1 Hz, the band was 0.08–0.12 Hz. Accordingly, the integrals of the power of HR fluctuations in the frequency ranges around the 3 ‘double period’ frequencies were investigated as well (0.125, 0.07 and 0.05 Hz). The integrals over the above relevant 6 frequency ranges were computed and compared between the different experimental stages (1 through 5).

2.6. Statistical analysis The total integrals of the HR PS and the integral of every relevant spectral band (around the 6 frequencies of interest were calculated for every subject at every stage of the study. The integral of every spectral band was normalized by dividing it by the total integral of the PS. We averaged these values separately for every group and graphically plotted the mean6SD. The results were compared (after averaging over the subjects) within each group for the various entrainment conditions (1 to 5) by applying a paired t-test (between the stage with the maximal integral and the other stages). For all spectral integrals, the tests were performed after log transformation of the results to obtain a more Gaussian distribution. Results with P , 0.05 were considered to be statistically significant.

3. Results

3.1. Baseline A typical baseline HR time series and HR power spectrum (PS) for one premature, one full-term neonate and one older infant, (during baseline conditions), are shown in Fig. 1a–c. From the HR traces and their PS for each subject, we computed for each group the mean6SD of HR, total power and the ratio between the integrals of low to high-frequency bands under initial baseline conditions (Fig. 2a–c). As shown in Fig. 2a, the mean HR of the prematures is the highest (group A, 14966 bpm) and the full-term heart rate is the lowest (group B, 126610 bpm), lower than that of the older

O. Shefi et al. / Early Human Development 52 (1998) 49 – 66

55

Fig. 2. (a) Mean heart rate of the three age groups: prematures, full-term neonates and older infants during the baseline stage. (b) Mean total power of the power spectrum (0.02–2.00 Hz) during the baseline stage of three groups of age. (c) The low / high ratio (0.02–0.2 / 0.2–2.0 Hz) of the three groups at the baseline stage.

babies (group C, 134614 bpm). This slower HR in group B is typical for neonates immediately after birth (24 h) [4]. The total power spectrum increased with age (Fig. 2b), and the LF / HF ratio decreased with age (Fig. 2c). The high-frequency power increased and the lowfrequency power remained unchanged. The respiratory peak at the high-frequency band, around 0.5 Hz, is single, clear and well defined in the older infants (group C) and, in contrast, is poorly defined and almost unobservable in a linear scale in the younger infants, groups A and B (Fig. 1).

56

O. Shefi et al. / Early Human Development 52 (1998) 49 – 66

3.2. Thermal entrainment The same subjects were monitored and recorded under various thermal entrainment conditions as described in Section 2. An example of the HR power spectra for one premature neonate (the same baby as shown in Fig. 1a under baseline conditions) during the three thermal entrainment stages is shown in Fig. 3. During the first thermal entrainment (Fig. 3a), switching from hot to cold stimulus is performed every 4 s, causing two clear peaks to arise in the HR PS, one at 0.25 Hz (4-s period) and the other at 0.125 Hz (8-s double period) according to the thermal entrainment frequency. No peaks arise at 0.14 Hz, 0.07 Hz, nor at 0.1 Hz and 0.05 Hz, which are the frequencies corresponding to the other entrainment frequencies applied later). In this case, we consider the reaction to the 4-s stimulus as a positive response. The criterion to consider a peak as a positive response was a peak with a power that is at least twice as high as the level in its close vicinity. During the second thermal stimulation in the same neonate, changing the hot to cold stimulus every 7 s causes the HR PS to display peaks at 0.14 Hz and 0.07 Hz (Fig. 3b), thus indicating a positive reaction to that stimulus. In the third session, changing the hot to cold stimulus every 10 s causes peaks to arise in the HR PS at 0.1 Hz and 0.05 Hz (Fig. 3c), again indicating a positive reaction to the thermal stimulus. Such a fully positive reaction to all 3 stimuli was not always observed in every subject in the different groups. Some of the subjects responded to the thermal entrainment only at the higher frequencies (0.25 Hz, 0.14 Hz and 0.1 Hz) and did not respond at the frequencies that correspond to the ‘double period’ (0.125 Hz, 0.07 Hz and 0.05 Hz). Such a phenomenon is displayed in Fig. 4, showing the HR power spectrum of a premature neonate during a 7-s long thermal stimulus: the only thermal peak that was entrained in that session, is at 0.14 Hz and not at 0.07 Hz. The results for the three groups in the different sessions, are summarized in Tables 1–3. The first row corresponds to thermal entrainment session with the 4-s long stimulus. Each column thereafter matches the corresponding frequency bands in the HR PS. In the column of 0.25 Hz, we count the number of subjects in the group for which a peak at 0.25 Hz has been elicited in their HR PS during a specific session (row) of the study. It should be emphasized that these peaks are visually quantitative, according to the peak criterion, by two blind observers. Of course, the ideal case is a reaction at 0.25 Hz and 0.125 Hz during the first stimulation session, a reaction at 0.14 Hz and 0.07 Hz during the second session and a reaction at 0.1 Hz and 0.05 Hz during the third stage. This would correspond to a reaction only along the diagonal of the table. The reality is far from being so simple and we can sometimes see peaks arising at frequencies which are not related to the thermal entrainment rate (not necessarily on the diagonal of the tables). Table 1 summarizes the responses of the premature neonates (group A) to the three thermal entrainment sessions: 5 out of 10 neonates react to the 4-s long stimulus (at 0.25 Hz), 6 out of 10 react to the 7-s stimulus (at 0.14 Hz) and 9 out of 10 react to the 10-s stimulus (at 0.1 Hz). Thus a strong tendency to diagonal response is

O. Shefi et al. / Early Human Development 52 (1998) 49 – 66

57

Fig. 3. (a) A typical heart-rate power spectrum for one preterm neonate (subject 8 in group A) during thermal entrainment – 4-s stimulation stage. Clear peaks occur at the entrainment frequencies, 0.25 Hz and 0.125 Hz, in addition to the energy in the low-frequency band that exists at the baseline stage. (b) Power spectrum of heart-rate fluctuations for the same subject (subject 8 group A) during the 7-s thermal stimulation stage. Clear peaks have risen at the corresponding 0.14 Hz and 0.07 Hz frequencies (in addition to the low-frequency power already existing at baseline). (c) Heart-rate power spectrum for the same preterm neonate during thermal entrainment with clear response to the 10-s stimulation at 0.1 Hz and 0.05 Hz.

58

O. Shefi et al. / Early Human Development 52 (1998) 49 – 66

Fig. 4. Heart-rate power spectrum during thermal stimulation (7-s stage) where only one entrained peak (the higher) has elicited, and not the peak at the ‘double period’.

observed. A fairly similar, strong diagonal, response occurs in the premature group at the frequencies corresponding to the double stimulus period (5 / 10, 8 / 10, 8 / 10 at 0.125, 0.07 and 0.05 Hz respectively).

Table 1 Premature infants Stimulus

0.25 Hz

0.14 Hz

0.1 Hz

0.125 Hz

0.07 Hz

0.05 Hz

4s 7s 10 s

5 / 10 0 / 10 0 / 10

4 / 10 6 / 10 2 / 10

3 / 10 5 / 10 9 / 10

5 / 10 0 / 10 0 / 10

3 / 10 8 / 10 1 / 10

2 / 10 3 / 10 8 / 10

Summary for group A, prematures, of the reaction to thermal entrainment during the three thermalstimulation stages at the 6 relevant frequencies. (If a peak has elicited in the power spectrum of the subject at the relevant frequency, this subject is added to the count) The ratios m /n indicates that m newborns out of the n members of the group showed a peak at the specific and frequency (columns) as a function of the 3 different stimuli (rows). Very few peaks have occurred off diagonal.

O. Shefi et al. / Early Human Development 52 (1998) 49 – 66

59

Table 2 Full-term neonates Stimulus

0.25 Hz

0.14 Hz

0.1 Hz

0.125 Hz

0.07 Hz

0.05 Hz

4s 7s 10 s

4/6 0/6 0/6

0/6 6/6 0/6

2/6 2/6 3/6

2/6 0/6 0/6

1/6 3/6 0/6

2/6 0/6 4/6

Summary as described in Table 1 for group B, the full-term neonates, of the reaction to thermal entrainment. All subjects had a peak at 0.14 Hz that corresponds to the 7-s stimulation stage.

Table 2 shows for the full-term neonates (group B) that 4 out of 6 react to the 4-s stimulus, 6 out of 6 (the entire group) react to the 7-s long stimulus and 3 out of 6 react to the 10-s stimulus. A smaller number of full-term neonates displayed in those sessions, a peak at the frequencies that correspond to the ‘double period’ of stimulus (2 / 6 at 0.125 Hz, 3 / 6 at 0.07 Hz and 4 / 6 at 0.05 Hz). Table 3 summarizes the response of the older infants (group C) to the different thermal entrainment frequencies. The maximal reaction occurs again at the diagonals of the table. However more subjects in this group had a positive reaction at the second group of frequencies (double period) – 0.125 Hz, 0.07 Hz and 0.05 Hz, than at the first group of frequencies (0.25 Hz, 0.14 Hz, 0.1 Hz). In this ‘double period’ part of the table, 5 / 7 reacted positively to the 4-s stimulus (0.125 Hz), 4 / 7 reacted to the 7-s stimulus (0.07 Hz) and 4 / 7 reacted to the 10-s stimulation (0.05 Hz). In groups A and B (prematures and full-terms) a larger number of subjects reacted to the thermal entrainment at the alternation frequencies (0.25, 0.14 and 0.1 Hz), than at the lower frequencies which correspond to the ‘double period’ (0.125, 0.07 and 0.05 Hz). This is opposite to the pattern observed in group C (older infants). As can be seen, the tables display a clear diagonal preponderance. In the first group the reaction is best to the 10-s long stimulus, in group B the reaction is best to the 7-s stimulus, and the reaction in group C is slightly better at 4 s. A more quantitative analysis of the results is presented in Fig. 5a–c Fig. 6a–c Fig. 7a–c for the different stages in the study, by computing integrals of the HR PS over the specific frequency bands. Fig. 5a presents the mean normalized integrals (6SD) for group A (prematures) for the spectral band 0.23–0.27 Hz (divided by the total integral), at the 5 different stages of the study (pre- for baseline, 4-s stimulus, 7-s stimulus, 10-s stimulus, and post- for baseline again). As expected for this frequency band, the mean integral is Table 3 Older infants Stimulus

0.25 Hz

0.14 Hz

0.1 Hz

0.125 Hz

0.07 Hz

0.05 Hz

4s 7s 10 s

3/7 0/7 0/7

0/7 2/7 0/7

1/7 3/7 2/7

5/7 1/7 0/7

0/7 4/7 0/7

2/7 2/7 4/7

Summary as described in Table 1 for group C, the older infants, of the reaction to thermal entrainment. The response is more evident at the frequencies that correspond to the ‘double period’ stimulation (0.125, 0.07 and 0.05 Hz).

60

O. Shefi et al. / Early Human Development 52 (1998) 49 – 66

Fig. 5. (a) Mean values, for group A, the premature infants, of the power in the spectral band 0.23–0.27 Hz (around 0.25 Hz) divided by the total power, at the 5 different stages. The maximum value occurs during the 4-s stimulation stage and is significantly higher than at all other experimental steps. Significance is indicated by *. (b) Mean values, for group A, of the power in the spectral band 0.12–0.16 Hz (around 0.14 Hz) divided by the total power, at the 5 different stages. The value during the 7-s stage is higher than at the 4- and 10-s stages. (c) Same presentation for group A in the spectral band 0.08–0.12 Hz.

clearly maximal and significantly higher during the 4-s stimulus stage (peaks arise around 0.25 Hz in the 4 s session, so the power in this band grows higher for this stimulus than at the stages with other entrainment frequencies). Fig. 5b–c were obtained for group A by the same technique as in Fig. 5a, but displaying the integrals over 0.12–0.16 Hz (around 0.14 Hz) and 0.08–0.12 Hz (around 0.1 Hz) respectively.

O. Shefi et al. / Early Human Development 52 (1998) 49 – 66

61

Fig. 6. (a) Mean values for full-term infants of the power spectrum between 0.23 to 0.27 Hz during the 5 stages of the study, divided by the total integral of the spectrum. The value during the 4-s stage is significantly higher than during the other 4 stages. (b) Same presentation for the frequency band 0.12–0.16 Hz for group B (full-term neonates). The highest value is during the 7-s stage, this value is significantly higher than during the 10-s stage and during the last baseline stage. (c) Same graph for group B for the frequency band 0.08–0.12 Hz. The maximum value is during the 10-s stage (corresponding to 0.1 Hz) and this maximum is highly significant compared to the values during the 7-s stage and the last baseline stage.

Similarly, they display a clear maximum for the 7-s and for the 10-s (respectively) stimulus stages. Fig. 6a–c and Fig. 7a–c are the equivalent presentation of the quantitative results of groups B and C respectively, for the 3 relevant spectral bands (0.23–0.27 Hz, 0.12–0.16 Hz, 0.08–0.12 Hz) respectively. These quantitative results are independent of the tabular presentation above. We tested statistically, the quantitative results in Figs. 5–7 and checked whether the

62

O. Shefi et al. / Early Human Development 52 (1998) 49 – 66

Fig. 7. (a) Mean values for group C, older infants, of the power spectrum between 0.23 to 0.27 Hz (around 0.25 Hz) during the 5 stages of the study, divided by the total integral of the spectrum. The value during the 4-s stage is not maximal in comparison to the other stages. This value is significantly higher than at baseline stage. (b) Same presentation for group C of the power between 0.12–0.16 Hz. The value of the normalized power of this band is significantly higher than all other stages but the 4-s stimulation stage. (c) Same presentation for group C of the power between 0.08–0.12 Hz. The maximum value is during the 10-s stimulation stage.

maximum value is statistically significantly higher than the values at the other stages (requiring P , 0.05 by paired t-test). For example, in Fig. 5a, the integral value (corresponding to the 0.23–0.27 Hz range) is maximal at the 4-s stimulus and this maximum is highly significant compared to its value at the baseline stages and the stages of 7-s and 10-s stimuli.

O. Shefi et al. / Early Human Development 52 (1998) 49 – 66

63

The maximum values observed were not always significantly different from the other stages. The significance depends very much on the frequency range considered. Indeed, at the higher frequencies of stimulation, the results were more significant, because there is not much background power (or noise) in the corresponding frequency range except for the energy specifically due to the thermal entrainment. At lower frequencies the spectrum has more background power and peaks than normally exist, such as the baroreceptor peak, at baseline without thermal stimulus [3,16]. Therefore in this ‘busy’ frequency range, the power at the thermal entrainment peak is not necessarily significantly higher.

4. Discussion Previous studies investigating HRV power in neonates revealed fluctuations of three main origins, arising from: respiration, blood pressure control (baroreceptor reflex) and vasomotor tone activity associated with temperature regulation [4,6]. The option of synchronizing self-sustained oscillations by an active, forced function has been known since the early experiments by Appleton and Van der Pol back in 1922 [10]. In 1974, Kitney et al. performed an experiment with adult volunteers: thermal stimuli of 20-s periodicity were achieved by placing one hand alternately in two buckets containing water at different temperatures (188C and 468C). Changes in blood volume in one finger were monitored as a representation of the state of the thermoregulatory system. The power spectrum of beat to beat intervals, calculated from the finger blood volume signal measured during the 20-s thermal stimulus, showed a large, single component at the 0.05 Hz frequency [9]. In 1990 and 1983, Lindqvist et al. investigated the effect of thermal entrainment on heart-rate variability in 20 neonates (their foot was placed in the jet of an air blower), classified according to maturity [14,13]. They reported that the response of the periodic HRV was selective at the spectral band corresponding to the frequency of stimulation. That report did not mention a reaction at the high frequencies bands, they reported only about the response in the frequency band that corresponds to the ‘double period’ (as called in our study). Jahnukaien-T. et al. observed in a similar study that full-term infants showed a clear response while the response of preterm infants and small for age neonates was markedly attenuated [7,8]. Lindqvist et al. assessed the physiological ability of an autonomic thermal stimulation method to induce synchronized oscillations in the neural cardiovascular control system in 18 adults [15]. The responses to the periodic thermal stimulation were measured from skin blood flow and heart rate. The rhythmic thermal stimulation synchronized the oscillations of the forearm skin blood flow but no stable synchronization of the periodic heart-rate variability was found (in supine rest). In the present study, we investigated the influence of thermal entrainment on 3 groups of infants. Thermal stimulation was achieved at frequencies that correspond to 4 s, 7 s and 10 s long stimuli, applied to one region of thermal sensors, the palm of the neonates. Our goal was to test the heart-rate response to changes in peripheral

64

O. Shefi et al. / Early Human Development 52 (1998) 49 – 66

blood flow induced by thermal stimulation of the skin, and to look for differences in reaction between the three groups. A positive response to the thermal entrainment was observed in all groups: each subject reacted at least at one stage of the study (at least in one stimulation frequency band). The differences between the three groups, as reacted in Tables 1–3, are: (a) the number of subjects displaying a positive response at each stage, (b) the appearance of peaks at frequencies that correspond to the thermal frequency and its ‘double period’ frequency, (c) the appearance of only one peak at the time. Most premature neonates reacted to the 10-s long stimulus, all full-term neonates reacted to the 7-s stimulus (at 0.14 Hz) and the older infants reacted with slight priority to the 4-s stimulation, the shorter one (see Tables 1–3). This fact may indicate that premature and full-term newborns require thermal stimuli of longer duration than older infants in order to affect their vasomotor activity and react by a change in HR control so that their reaction can be detected by spectral analysis. This may point towards a developmental process of maturation with age of the ability to respond to changes in skin temperature. Most premature neonates, who responded to the thermal entrainment, reacted by raising two peaks, at the entrainment: frequency and at the frequency corresponding to the ‘double period’. Half of the full-term neonates had a reaction at both frequencies and the remaining full-term neonates reacted at the entrainment frequency (and not at the one that corresponds to the double period) as can be seen in Fig. 4. Most subjects in the older infants group reacted to the thermal entrainment by raising a peak at the frequency that corresponded to the ‘double period’ stimulus and not at the entrainment frequency itself. It seems that the younger infants (preterms and full-terms) react to every change of the skin temperature while the older ones preferably synchronize on the ‘double period’ (hot and cold stimulus). The peak that occurs at the entrainment frequency cannot be considered as a harmony of the ‘double period’ peak, as can be concluded from Fig. 4 which only displays a peak at the stimulation frequency without raising peak at the double period. Previous studies on adults revealed that different branches of autonomic control contribute to the HR power spectrum in different, well defined, frequency bands. Slow HR fluctuations in the frequency range of 0.02 to 0.15 Hz, are presumably of vasomotor and baroreceptor origin, and mediated by both sympathetic and parasympathetic tone. Higher frequency fluctuations, represented by a peak centred around the respiratory frequency are of vagal origin [1]. The spectral analysis of HR fluctuations in neonates provides a tool for estimating the maturation of infants [4,2,3]. The maturation of the parasympathetic control is reflected by an increase in HF values expressed in a clear and well defined single respiratory peak in the older infants in contrary to the younger infants (groups A and B). A similar lack of maturation of thermal responsiveness can explain the preferred reaction of the younger infants to the slowest thermal stimulus (0.1 Hz). On the other hand, the preference of the older infants to raising their ‘double period’ peak, may reflect the fact that the sympathetic system (inherently slower) has matured and is now able to assume its vasocontrol role.

O. Shefi et al. / Early Human Development 52 (1998) 49 – 66

65

5. Conclusions The present study attempts to provide some insight in the maturation of thermal control. We observe that: 1. The reaction to thermal entrainment occurs already in preterm infants. 2. Thermal entrainment has a marked effect at the three different frequencies of stimulation (0.25, 0.14 and 0.1 Hz) in the three groups of subjects: a clear ‘entrainment’ peak usually occurs either at the frequency of stimulation (alternations) or at the frequency corresponding to the ‘double period’ or at both. 3. When comparing the frequency of thermal entrainment, we observe that premature infants seem to react preferably to the slowest stimulus (every 10 s alternation), full-term neonates react to the intermediate frequency (0.14 Hz, every 7-s stimulation), while the older infants have a slight tendency to respond to the quickest stimulus (4 s alternation). 4. When comparing the preponderance of entrainment at the, basic period of stimulus, with the ‘double period’, we observe that preterm infants and full-term infants react preferably to every change of skin temperature (single period), while the older infants react rather to the ‘double period’ (‘hot and cold’ stimulus) than to every change of skin temperature. This could represent the maturation of thermoregulatory response, the ability of sympathetic control to mediate this response and its preponderance in the lower frequency range (below 0.125 Hz), as usually quoted in adults.

6. Notation Keywords: HR: Heart Rate PS: Power Spectrum ECG: Electrocardiogram A / D: Analog to Digital

References [1] Akselrod S. Spectral analysis of fluctuations in heart rate and other cardiovascular parameters as a tool for the assessment of autonomic control. In: Korczyn AD, editor. Handbook of autonomic nervous system dysfunction. New York: Marcel Dekker Inc., 1985:469–93. [2] Akselrod S, Gordon D, Ubel FA, Shannon DC, Barger AC, Cohen RJ. Power spectrum analysis of heart-rate fluctuation: a quantitative probe of beat-to-beat cardiovascular control. Science 1981;213:220–2. [3] Akselrod S, Gordon D, Madwed JB, Snidman NC, Shannon DC, Cohen RJ. Hemodynamic regulation: investigation by spectral analysis. Am J Physiol 1985;249:H867–75. [4] Chatow U, Davidson S, Reichman BL, Akselrod S. Development and maturation of the autonomic nervous system in premature and full-term infants using spectral analysis of heart-rate fluctuations. Pediatr Res 1995;37(3):294–302.

66

O. Shefi et al. / Early Human Development 52 (1998) 49 – 66

[5] Hathorn MKS, Mannix PA, Costeloe K. Effect of periodic thermal stimulation on heart rate in term newborn infants. J Appl Physiol 1995;79(4):1093–9. [6] Hirsch M, Karin J, Shechter B, Segal O, Merlov P, Kaplan B, Akselrod S. In: Cosmi EV, Renzo, GCDI, editors. Detection of fetal breathing activity (BA) in real-time by means of spectral analysis of fetal heart-rate (HR) fluctuations. 2nd World Congress of Perinatal Medicine. Rome, Italy, 1993. [7] Jahnukainen T, van-Ravenswaaij AC, Jalonen J, Valimaki I. Dymamics of vasomotor thermoregulation of the skin in term and preterm neonates. Early Hum Dev 1993;33(2):133–43. [8] Jahnukainen T, Lindqvist A, Jalonen J, Kero P, Valimaki I. Reactivity of skin blood flow and heart-rate to thermal stimulation in infants during the first postnatal days and after a two-month follow-up. Stockholm ISSN 0803-5253. Acta Paediatr 1996;85:733–8. [9] Kitney RI. The analysis and simulation of the human thermoregulatory control system. Med Biol Eng 1974;12(1):57–65. [10] Kitney RI. An analysis of the nonlinear behavior of the human thermal vasomotor control system. J Theoret Biol 1975;52:231–48. [11] Kitney RI, Rompelman O, editors. The study of heart-rate variability. Oxford: Clarendon Press, 1980. [12] Lossius K, Eriksen M, Walloe L. Thermoxegulatory fluctuations in heart-rate and blood pressure in humans: Effect of cooling and parasympathetic blockade. J Autonomic Nerv Syst 1994;47:245–54. [13] Lindqvist A, Oja R, Helena O, Valimaki I. Impact of thermal vasomotor control on the heart-rate variability of newborn infants. Early Hum Dev 1983;8(1):37–47. [14] Lindqvist A. Contribution of thermal stimulation to the periodic heart-rate variability in neonates (V). Acta Physiol Scand Suppl 1990;588:1–107. [15] Lindqvist A, Parviainen P, Jalonen J, Tuominen J, Valimaki I, Laitinen LA. Clinical testing of thermally stimulated cardiovascular oscillations in man. Cardiovasc Res 1991;25(8):666–75. [16] Schmidt RF, Thews G. Human physiology. Berlin: Springer-Verlag, 1989:624–44. [17] Shitzer A, Eberhat RC, editors. Heat transfer in medicine and biology. New York: Plenum Press, 1985. [18] Tokuo O, Low PA. Autonomic regulation of temperature and sweating. In: Low PA, editor. Clinical autonomic disorders. Rochester: Little Brown and Company, 1993:79–92.

The effect of thermal stimulation on the heart-rate ... - Semantic Scholar

Analysis applied to the thermal vasomotor control systems has led to the conclusion .... Heart-rate data analysis. Spectral analysis of .... 0.125 Hz, in addition to the energy in the low-frequency band that exists at the baseline stage. (b) Power.

1MB Sizes 0 Downloads 216 Views

Recommend Documents

The effect of thermal stimulation on the heart-rate ... - Semantic Scholar
bDepartment of Neonatology, Rabin Medical Center, Beilinson Campus, Tel Aviv University, Tel Aviv,. Israel .... Same thermal stimulation but at a rate of 0.14 Hz (switching the temperature every. 7 s). 4. ... Heart-rate data analysis. Spectral ...

Effect of substrate modes on thermal transport in ... - Semantic Scholar
Aug 12, 2011 - thus creates an average net thermal current from the center to the edge. ... steady-state heat current and that the time-averaged ε of the middle atoms is ... 2(a). This profile is nonlinear, with temperature “slips” near the ther

Thermal lens technique to study the effect of pH on ... - Semantic Scholar
b Centre for Laser Spectroscopy, MAHE Life Science Centre, MAHE, Manipal, Karnataka 576 ... resonance energy transfer does not require physical contact of.

The effect of disfluency on mind wandering during ... - Semantic Scholar
United States b University of Notre Dame, Department of Computer Science .... findings and those for tasks like list learning suggests that disfluency only affects surface level learning (e.g. ... The experiment utilized a between-subjects design.

The Kuleshov Effect - Semantic Scholar
Statistical parametric maps (from right hemisphere to left hemisphere) and parameter estimate plots illustrating the main effect of faces presented in the (c) negative-neutral contexts and (d) positive-neutral contexts. Faces presented in both negati

The effect of disfluency on mind wandering during ... - Semantic Scholar
United States b University of Notre Dame, Department of Computer Science .... findings and those for tasks like list learning suggests that disfluency only affects surface level learning (e.g. ... The experiment utilized a between-subjects design.

Subthalamic nucleus stimulation influences ... - Semantic Scholar
Sep 22, 2010 - 1 Neurology Department, University of Virginia Health Systems, Virginia, ... analytic methods to separate the strength of the initial response ...

25 Effect of the Brazilian thermal modification process on the ...
25 Effect of the Brazilian thermal modification process ... Part 1: Cell wall polymers and extractives contents.pdf. 25 Effect of the Brazilian thermal modification ...

An Experimental Study on the Capture Effect in ... - Semantic Scholar
A recent measurement work on the cap- ture effect in 802.11 networks [12] argues that the stronger frame can be successfully decoded only in two cases: (1) The.

Subthalamic nucleus stimulation influences ... - Semantic Scholar
Sep 22, 2010 - 3 Cognitive Science Centre Amsterdam, University of Amsterdam, ... Psychological Sciences, University of Northern Colorado, Colorado, USA.

Effect of Global Context on Homophone Ambiguity ... - Semantic Scholar
provide converging evidence for this interpretation. Eye-tracking data analysis. .... to the evolution of perceptual and cognitive systems. Cognitive Science: A ...

on the difficulty of computations - Semantic Scholar
any calculation that a modern electronic digital computer or a human computer can ... programs and data are represented in the machine's circuits in binary form. Thus, we .... ory concerning the efficient transmission of information. An informa-.

Effect of Melatonin on Sleep, Behavior, and ... - Semantic Scholar
Article Plus (online only) materials for this article appear on the Journal's Web ..... c Mann-Whitney U test. VAN DER HEIJDEN ET ..... Neuropsychol Dev Cogn C.

Dual Ligand Stimulation of RAW 264.7 Cells ... - Semantic Scholar
lecular Science and Technology, Ajou University, Suwon, Korea. Received for ..... than additive) response increased with time and by 4 h, the ma- jority of nonadditive ..... It is worth noting that of the features that were rapidly induced by. IFN- a

Effect of vitamin AD3E utilization at the pregnancy ... - Semantic Scholar
system such as vitamin A,E and D (Priyantha et al .,. 2012). Some studies refer to that the deficiency in the .... AD3E in G2 and low level of iron content in feedstuff of G1These results agree with results of Hafez ... 1- At the last of 3 months pre

The Effect of Unemployment Benefits and ... - Semantic Scholar
key policy parameters: the effect of UI on reservation wages and the effect of nonemployment durations on wage offers. ... we call the causal effect of unemployment duration on wages – is important for understanding the costs and ...... of UI exten

Kondo Effect in the Presence of Itinerant-Electron ... - Semantic Scholar
Dec 10, 2003 - system's ground state can be tuned to have a fully com- pensated local spin, in which case the QD conductance is found to be the same for each spin channel, G" И G#. The model.—For ferromagnetic leads, electron- electron interaction

The Effect of Unemployment Benefits and ... - Semantic Scholar
in the part of the wage offer distribution relevant for workers' employment decisions. If the condition on reemployment wages is satisfied in the data, the only effect of nonem- ployment durations on wages must arise from a change in the wage-offer d

The Impact of the Lyapunov Number on the ... - Semantic Scholar
results can be used to choose the chaotic generator more suitable for applications on chaotic digital communica- .... faster is its split from other neighboring orbits [10]. This intuitively suggests that the larger the Lyapunov number, the easier sh

Field-Effect Tunneling Transistor Based on Vertical ... - Semantic Scholar
Feb 2, 2012 - Page 1. www.sciencemag.org/cgi/content/full/science.1218461/DC1. Supporting Online Material for. Field-Effect Tunneling Transistor ...

Exploiting the Unicast Functionality of the On ... - Semantic Scholar
any wired infrastructure support. In mobile ad hoc networks, unicast and multicast routing ... wireless network with no fixed wired infrastructure. Each host is a router and moves in an arbitrary manner. ..... channel reservation control frames for u

the impact of young workers on the aggregate ... - Semantic Scholar
years, from various years of the Statistical Abstract of the United States. The following observations are missing: 16–19 year olds in 1995 and 1996; 45–54 year old men in Utah in 1994; and women 65 and over in. Delaware, Idaho, Mississippi, Tenn

Exploiting the Unicast Functionality of the On ... - Semantic Scholar
Ad hoc networks are deployed in applications such as disaster recovery and dis- tributed collaborative computing, where routes are mostly mul- tihop and network ... the source must wait until a route is established before trans- mitting data. To elim