- Email: [email protected]
Hemodynamic, autonomic and baroreflex changes after one night sleep deprivation in healthy volunteers
Autonomic Neuroscience: Basic and Clinical 145 (2009) 76–80
Contents lists available at ScienceDirect
Autonomic Neuroscience: Basic and Clinical j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / a u t n e u
Hemodynamic, autonomic and baroreflex changes after one night sleep deprivation in healthy volunteers Massimo Pagani a,⁎, Paolo Pizzinelli a, Anne Pavy-Le Traon b, Cinzia Ferreri a, Silvia Beltrami a, Marie-Pierre Bareille b, Marie-Claude Costes-Salon b, Stéphane Béroud b, Olivier Blin c, Daniela Lucini a, Pierre Philip d a
Centro di Ricerca sulla Terapia Neurovegetativa, Dipartimento Scienze Cliniche “L. Sacco”, University of Milano, Milano, Italy Medes, Clinique Spatiale, Hôpital Rangueil-Larrey, Toulouse, France c Institut des Neurosciences Cognitives de la Mediterranee, Faculte de Medecine, Universite de la Mediterranee, Marseille, France d Sleep Clinic, Bordeaux University School of Medicine, Bordeaux, France b
a r t i c l e
i n f o
Article history: Received 10 June 2008 Received in revised form 30 September 2008 Accepted 2 October 2008 Keywords: Hypertension Sleep deprivation Autonomic nervous system Baroreflex Stress Cardiovascular risk
a b s t r a c t Background: Sleep disorders are associated to a number of cardiovascular disturbances that might increase cardiovascular risk. Sleep deprivation, in particular, might, by inducing autonomic dysregulation, raise arterial pressure and hypertensive risk. Available evidence however is contradictory. Methods: We tested the main hypothesis that one night sleep deprivation in 24 volunteers might alter hemodynamics (heart rate and Arterial Pressure — AP), autonomic regulation (mono and bivariate spectral analysis of RR and non invasive AP variability) and baroreflex control (spectral index alpha and spontaneous baroreflex slope), performance indices (reaction time) and subjective stress (questionnaires and salivary cortisol). Volunteers were studied in normal living conditions and while kept in isolation and confinement, to test the presence of possible bias related to environmental stress. Results: Results indicate that there were no differences between normal living conditions and isolation and confinement (Intraclass Correlation Coefficient N 0.75 for most variables). Conversely, after one night sleep deprivation subjects felt tired (p b 0.05), and performance deteriorated (p b 0.05), while cortisol profile was substantially maintained, hemodynamic parameters did not change and HRV and index alpha increased slightly. Conclusions: Findings support the contention that one night sleep deprivation, in absence of significant additional stress or disturbances, does not lead to increased arterial pressure values or to changes in autonomic or baroreflex profiles that could conceivably favor hypertension development, but induces the expected increase in tiredness and reduction in performance. © 2008 Elsevier B.V. All rights reserved.
1. Introduction Sleep disorders are associated to increased cardiovascular risk and in particular to elevated hypertension risk (Gangwisch et al., 2006). Among the various factors that might be implicated several investigators on the basis of indirect evidence suspected an important role of increased sympathetic activity (Lusardi et al., 1996; Zhong et al., 2005), possibly emphasized by poor sleep quality, such as might occur with the Ambulatory Blood Pressure Monitoring (ABPM) technique (Tropeano et al., 2006), or because of the facilitating effects of strenuous work (Tochikubo et al., 1996) or alcohol addiction (Irwin and Ziegler, 2005). Direct recordings of Muscle Sympathetic efferent
⁎ Corresponding author. Telematica per la Medicina e la Formazione, Ospedale “L. Sacco”, Università di Milano, Via G.B. Grassi, 74, 20157 Milano, Italy. Tel.: +39 02 39042802; fax: +39 02 50319823. E-mail address: [email protected] (M. Pagani). 1566-0702/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.autneu.2008.10.009
Nerve Activity (MSNA), with the exception of extreme conditions such as Obstructive Sleep Apnea (OSA), where sympathetic overactivity is a well known key player (Narkiewicz et al., 1998), do not clarify the issue: in fact, while Ogawa et al. (2003) suggested that sleep deprivation might increase sympathetic activity and, as a consequence, raise arterial pressure, Kato et al. (2000) reported that if anything MSNA was reduced after one night sleep deprivation. Findings might also be contingent upon the specific experimental model that was employed, such as a one night sleep deprivation (Lusardi et al., 1996; Ogawa et al., 2003; Zhong et al., 2005) in normotensive individuals, or a prolonged period of short sleeping hours (Gangwisch et al., 2006) in hypertensive subjects (Verdecchia et al., 2007). In addition, an unfavorable psychosocial environment (Rozanski and Blumenthal, 1999) might lead to a state of sympathetic arousal and thus to higher arterial pressure levels, particularly in patients complaining of chronic stress (Lucini et al., 2005). This complexity, although alluded to, was not usually explicitly addressed in most prior studies. Thus sleep
M. Pagani et al. / Autonomic Neuroscience: Basic and Clinical 145 (2009) 76–80
deprivation and difficult living or working conditions might conjure in determining a deleterious mix of elevated subjective stress and autonomic dysfunction, leading to increased arterial pressure. Of potential clinical interest, some of these alterations can be reverted towards normal with behavioral techniques (Lucini et al., 2007). In order to elucidate the effects of sleep deprivation on individual hemodynamic and stress measures, it might be therefore critical to shield experimental subjects from uncontrolled external and environmental stressors. These requirements can be met in conditions of isolation and confinement, where light and sleep activity schedule can be strictly determined. This setting, although it represents a change from everyday life, permits to maintain environmental stress to a preset level, and it is likely to confer greater stability to baseline autonomic regulation (Pagani et al., 1995). Accordingly we first tested the effects of isolation and confinement, per se, on stability of hemodynamics, autonomic regulation, performance indices and subjective stress in a group of healthy volunteers. Secondly we assessed, on the same variables, the effects of one night sleep deprivation, while controlling for environmental stress and bias. Direct disturbances to subjects well being were further minimized by using only rigorously non invasive recording equipments, and having experimental subjects perform measures themselves, to prevent a possible “white coat effect” (Mancia, 2000).
2. Materials and methods 2.1. Study subjects 24 young healthy subjects (12 male, 12 female; age 27–45) who had previously passed a formal selection at MEDES Clinic (Toulouse) were studied. Briefly, the selection included a complete physical and psychological (including morning–night sleep type, no sleep disorders, no psychopathology, no alcohol, no smoking, no shift-work) assessment, and a battery of clinical measures (normal blood tests: glucose, urea, creatinine, electrolytes (Na+, K+, CI−, P, HCO3−, Ca++), transaminases (ALT and AST), γGT, alkaline phosphatases, hepatitis and HIV serology. The mean height of the crew was 172 ± 8 cm, the mean weight was 66.6 ± 13.3 kg and the mean Body Mass Index (BMI) was 22.4 ± 2.9 kg/m2. 2.2. Protocol All subjects were randomly exposed to 2 different environmental conditions: Medes (M) and Home (H), in a Latin square design. Condition M: lasted 8 days from M1 (first day) to M8 (eighth day), and was designed in order to obtain the controlled conditions of a monotonous working environment in terms of low physical activity, social isolation, confinement and light exposure. To this aim subjects were housed in the MEDES “Clinical Research Facility” with specific light conditions: 300 lux in living area and 50 lux in passageways during the artificial day (16 h); and 50 lux in living areas and 10 to 50 lux in passageways during night periods (8 h: low light). Notice that because subjects chose voluntarily to participate to the experiment and could exit from it in any moment, if they decided to, these conditions, according to the Karasek model of stress-reward should be considered as low chronic work stress (Karasek et al., 1981). At the end of day 8 (M8) subjects were kept awake for 24 h under low light exposure thus undergoing a one-night sleep deprivation under continued TV surveillance. To further ensure that subjects did not inadvertently fall asleep, they had to perform a simple computerized task every 2 h. At the end of this sleep deprivation period volunteers were subjected to final tests the morning of day 9 (M9). Subjects were isolated in groups of six (3 men and 3 women). Condition H: subjects were also studied at the beginning (H1) and at the end (H8) of an 8-day period during which they followed a
77
normal life routine at home, with however a strict sleep-wake schedule (23 p.m.–7 a.m.), and avoiding heavy activities. Based on previous experience (Pagani et al., 1995), subjects were individually trained once for about 4 h by two investigators (CF and PP) to collect data by themselves, with a high level of efficiency. Overall 85% of data were fully analyzable, and only 7% of data were lost. This approach would most likely avoid any “white coat effect” (Mancia, 2000). All recordings (both for condition M and H) were performed in MEDES Clinic in the morning, between 9 and 11 a.m., at least 2 h after a light breakfast not containing caffeine. In all subjects hemodynamic, endocrine, psychological, autonomic and performance measures were obtained, as specified below. Recorded variables: using standard Ag–AgCl electrodes, both the ECG (CM5) and the respiratory signal (transthoracic impedance) were monitored with a two-way radiotelemetry system (Marazza, Monza, Milan, Italy). The arterial pressure waveform was continuously estimated non invasively with a plethysmographic device (Portapres, TNO Institute of Applied Physics Biomedical Instrumentation, Amsterdam). Data were acquired on a PC, using an acquisition rate of 300 samples/channel/s. All subjects were studied in resting conditions (lying down with 15° back support): after a 10 minute period, allowed for stabilization, a 10 minute rest period was obtained. 2.2.1. Assessment of overall stress level A battery of non-invasive tests gauged the overall stress level: – Endocrine involvement (Lucini et al., 2002) was assessed by measuring by radioimmunoassay (DPC, Los Angeles, California, USA) free salivary cortisol levels which reflect the concentration of free hormone in plasma. Samples were in all instances obtained at approximately the same hours (10:30 a.m. ±1 and 6 p.m. ±1) to take circadian variation into account. – Psychological involvement was gauged by a battery of questionnaires (Lucini et al., 2002), providing self-rated scales (higher values indicate higher degrees of symptoms) that, in line with previous and more recent studies, focuses on appraisal, coping and health. In brief: – The appraisal of stress and tiredness was assessed by a global scoring index (0–10) for each measure. – A symptom list was used to score 18 somatic complaints (score 0–180) (4SQ). – This questionnaire has been validated against Cognitive Behavioral Assessment, Italian version (Lucini et al., in press). 2.2.2. Study of autonomic regulation Indirect autonomic parameters were assessed with a non-invasive approach based on spectral analysis of RR, Systolic Arterial Pressure (SAP) and respiration variability (Pagani et al., 1986; Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology, 1996). Analysis was conducted using an ad hoc software designed to perform spectral analysis of cardiovascular signals (HeartScope) (Badilini et al., 2005): from the ECG derived tachogram an autoregressive approach provides both absolute (i.e. [ms2]) and normalized units (i.e. [nu]) spectral powers of Low (LF) and High Frequency (HF) components. The normalized power of these two components provides gross estimates of the sympathetic and vagal oscillatory modulations of the sinoatrial (SA) node, and of their balance (from LF/ HF ratio) (Pagani et al., 1986). Spectral analysis was also performed on the systolic arterial pressure, in order to estimate sympathetic oscillatory vasomotor modulation from the absolute power of the LF component (HF power of systolic arterial pressure variability is a measure of mechanical effects of respiration on the peripheral circulation) (Pagani et al., 1997).
78
M. Pagani et al. / Autonomic Neuroscience: Basic and Clinical 145 (2009) 76–80
Table 1 Summary statistics for baseline autonomic, humoral, psychological and performance indices REST
Measure units
Average std dev
RR RR σ2 LFRR LFRR HFRR HFRR LF/HFRR SAP SAP σ2 LFSAP HFSAP BRS Alpha Cort_M Cort_A Stress Tired S4Q SRTT SRTT 10%L
ms ms2 ms2 nu ms2 nu – mm Hg ms2 mm Hg2 mm Hg2 ms/mm Hg ms/mm Hg ng/ml ng/ml A.U. A.U. A.U. ms ms
914.84 ± 135.44 5278.60 ± 5415.66 2190.28 ± 3012.52 55.65 ± 20.47 1186.28 ± 1536.88 39.68 ± 19.25 2.28 ± 2.26 113.94 ± 13.69 41.19 ± 39.32 8.68 ± 11.26 2.26 ± 2.39 21.08 ± 9.95 21.25 ± 11.67 8.03 ± 3.29 3.26 ± 1.35 0.80 ± 1.32 1.33 ± 1.59 7.43 ± 12.82 383.64 ± 131.95 682.90 ± 351.766
working conditions at MEDES (M), or comparing Day 1 or Day 8 (data not shown for simplicity). Consequently average data and standard deviations are reported from combined experiments of days H1, H8, M1, M8 in Table 1 that accordingly provides summary baseline evaluation of all hemodynamic and autonomic variables, in conditions of normal sleep routine. Intraclass Correlation Coefficients (ICC) computed while a regular sleep schedule was maintained (H1, H8, M1, M8) further confirms the strong stability of baseline data across days and contexts. It is apparent that all variables in spite of a well known high variability between subjects possess a high degree of stability (Fig. 1), with a few exceptions: SAP variance and LFSAP. Notably ICCs computed for time domain measures (RR, SAP, BRS), for several frequency domain indices
Average ± standard deviation values of indices from 24 healthy volunteers at rest, combining results obtained in 4 different instances (at Home and at MEDES, day 1 and day 8). Abbreviations: σ2 = variance, LF = Low Frequency, nu = Normalized Unit, HF = High Frequency, SAP = Systolic Arterial Pressure, BRS = Baroreceptor Reflex Sensitivity, Alpha = frequency domain index of baroreflex sensitivity, Cort_M = salivary cortisol (morning evaluation), Cort_A = salivary cortisol (afternoon evaluation), S4Q = Subjective Stress-Related Somatic Symptoms Questionnaire, SRTT = Simple Reaction Time Test, SRTT10%L = mean of the ten percent slowest RT trials.
The center frequency of the HF component of the respiratory signals, obtained with a similar procedure, assessed the main respiratory frequency. This latter usually should be higher than 0.14 Hz, in order to avoid entrainment (Pagani et al., 1986). From the simultaneous analysis of arterial pressure and RR interval variability a frequency domain index (Alpha) (Pagani et al., 1988) can be derived at both LF and HF frequencies which provides a measure of the overall gain of the arterial pressure–heart period relationship. The following formula providing an average value (Lucini et al., 2002) was employed: 1=2 Alpha = ðPRR =PSAP Þ1=2 LF + ðPRR =PSAP ÞHF =2 The program also provides values of the time domain spontaneous Baroreflex slope, as originally described by Bertinieri et al. (1985). Performance was estimated with a Simple Reaction Time Test (SRTT), which lasted 10 min and was based on a PSION personal organizer (Gilberg et al., 1994). 2.2.3. Statistical analysis Data are presented as Mean ± Standard Deviation. Significance of differences between repeated measures and contexts was estimated with a mixed model or GLM analysis as appropriate, followed by individual contrasts. Intraclass Correlation analysis (van de Borne et al., 1997) was used to assess the stability of measures across days M1, M8, H1, H8, treating data as independent measures. Significance was set at p b 0.05. Computations were performed with a commercial statistical package (SPSS13). 3. Results 3.1. Baseline measures No significant differences were found between baseline data obtained during the free living conditions at home (H), or during the controlled
Fig. 1. Intraclass Correlation Coefficients for variables related to Heart Rate Variability (A), arterial pressure, respiration and BRS (B), stress and performance (C), considering the effects of time and context: ICC was computed across days H1, H8, M1 and M8 of each condition (M and H). A dotted line at 0.75 highlights those variables above the value demonstrating high stability. Abbreviations: LF = Low Frequency, σ2 = variance, HF = High Frequency, SAP = Systolic Arterial Pressure, Alpha = frequency domain index of baroreflex sensitivity, BRS = Baroreceptor Reflex Sensitivity, S4Q = Subjective Stress-Related Somatic Symptoms Questionnaire, SRTT = Simple Reaction Time Test, SRTT10%L = mean of the ten percent slowest RT trials, Cort(N) = Night Cortisol, Cort(M) = Morning Cortisol.
M. Pagani et al. / Autonomic Neuroscience: Basic and Clinical 145 (2009) 76–80 Table 2 Mean ± standard deviation of stress and performance parameters assessed before and after moderate sleep deprivation
RR RR σ2 LFRR LFRR HFRR HFRR LF/HFRR SAP SAP σ2 LFSAP HFSAP BRS Alpha Cort_M Cort_A STRESS TIRED S4Q SRTT SRTT 10%L
Measure units
MEDES M8
M9
ms ms2 ms2 nu ms2 nu – mm Hg ms2 mm Hg2 mm Hg2 ms/mm Hg ms/mm Hg ng/ml ng/ml A.U. A.U. A.U. ms ms
895.17 ± 161.64 4610.35 ± 5143.10 2315.15 ± 3528.00 59.12 ± 19.34 988.03 ± 1071.31 36.56 ± 17.75 2.21 ± 1.50 113.44 ± 10.43 35.04 ± 19.43 11.02 ± 18.49 2.58 ± 3.05 18.61 ± 7.57 20.03 ± 10.51 7.87 ± 3.91 3.61 ± 0.98b 1.13 ± 1.69 1.43 ± 1.85 6.57 ± 11.72 379.63 ± 131.03 726.42 ± 429.89
990.20 ± 147.14a 7448.26 ± 8803.18 2218.48 ± 2829.37 57.62 ± 17.46 1490.79 ± 2182.78 37.51 ± 16.68 2.32 ± 2.20 113.13 ± 11.40 42.01 ± 18.38 7.33 ± 6.08 1.66 ± 1.12 25.87 ± 20.25a 27.82 ± 16.10a 7.06 ± 3.22 3.48 ± 1.05b 1.09 ± 1.90 5.35 ± 2.42a 6.96 ± 11.04 508.44 ± 182.80a 1107.52 ± 492.83a
M8 = day 8 at Medes, M9 = day 9 at Medes, i.e. after sleep deprivation. Abbreviations: σ2 = variance, LF = Low Frequency, nu = Normalized Unit, HF = High Frequency, SAP = Systolic Arterial Pressure, BRS = Baroreceptor Reflex Sensitivity, Alpha = frequency domain index of baroreflex sensitivity, Cort_M = salivary cortisol (morning evaluation), Cort_A = salivary cortisol (afternoon evaluation), S4Q = Subjective Stress-Related Somatic Symptoms Questionnaire, SRTT = Simple Reaction Time Test, SRTT10%L = mean of the ten percent slowest RT trials. a Significant p b 0.05 M8 vs M9. b Significant p b0.05 morning vs night.
(RR σ2, LFRR(nu), HFRR(nu), LF/HFRR; HFRESP, Alpha index) and for stress and performance (SRTT) indices were all above the 0.75 value, which is usually taken to indicate a high degree of consistency. 3.2. Sleep deprivation Table 2 focuses on the effect of sleep deprivation: with the exception of prolonged RR and increased BRS and α index, there were no changes in spectral parameters. Salivary cortisol was not affected and maintained its circadian variations throughout the experiment. Tiredness and SRTT scores increased significantly (i.e. performance was impaired).
79
namic and cardiac autonomic indices have a high stability, but we show in addition that hormonal and psychological stress indicators are also highly stable in the setting of the present experiment. In particular, the hypothetical stressful condition of isolation and confinement, as in MEDES environment, was not reflected in any measurable change in the multidimensional parameters under scruting. Our data moreover extend the notion of elevated stability (van de Borne et al., 1997) over time also for bivariate indices of baroreflex regulation, inclusive of phase and coherence between arterial pressure and RR variability oscillations in the LF and HF regions, as well as for indices of motor performance, as assessed by a simple reaction time task. Similarity of findings was maintained also comparing data obtained during normal life at home, or during a controlled work routine, in MEDES environment. It should be added that, because subjects were in control of their work relationship, MEDES environment provided the additional advantage of a stable situation, as far as work stress is concerned (Karasek et al., 1981). This contention is supported by the physiologically low cortisol and stress score values. Baseline values of hemodynamic, autonomic and baroreflex regulation were all within the normal range of our laboratory (Lucini et al., 2002). Accordingly, isolation and confinement per se do not seem to introduce autonomic or subjective signs of stress. Different results may be obtained in extreme conditions (Palinkas and Houseal, 2000) or considering other targets, such as the immune system or more demanding environments (Shimamiya et al., 2004). 4.2. Autonomic assessment In our investigation, based on multiple recordings several days apart, only non invasive non intrusive recordings were employed, as per routine in our clinical laboratory (Lucini et al., 2005). Obviously, direct electroneurographic recordings, as in the case of MSNA (Pagani et al., 1997), might provide more robust physiology, although at the risk of introducing an unknown stress related bias. The multiparameter approach, although not providing direct data about sympathetic efferent traffic is likely to furnish more abundant information (Pagani and Malliani, 2000) on the complexity of autonomic cardiovascular regulatory processes (Ahn et al., 2006), and their changes, particularly in the oscillatory domain (Massimini et al., 2000). These changes have been shown to track faithfully prehypertensive risk (Lucini et al., 2002), its link with stress symptoms (Lucini et al., 2005) and possible behavioral treatment (Lucini et al., 2007).
4. Discussion
4.3. Sleep deprivation
This study provides evidence against the hypothesis that one night sleep deprivation might increase arterial pressure or its variability and sympathetic drive in healthy individuals. Unintended emotional bias, such as that produced by unfavorable or uncontrolled work environment, or by invasive examinations (e.g., needles as in MSNA or in plasma catecholamine determinations), was virtually excluded by the strict use of controlled non invasive, non intrusive experimental conditions. Moreover, the potential weakness of indirect autonomic assessment was counterbalanced by the enhanced information (Pagani and Malliani, 2000) provided by the use of several concurrent indices (Haken,1983, Ahn et al., 2006): hemodynamic, autonomic, hormonal, psychological and of performance. This approach, could furnish functional elements that could be included in the design of personalized preventive clinical approaches (Turner et al., 2007).
After one night sleep deprivation we observed no signs of increases in arterial pressure or its variability. RR interval, BRS variance and the index alpha increased slightly, suggesting an increase in cardiac vagal regulation. This finding is in line with previous reports of a mirror reduction of MSNA activity after one night sleep deprivation (Ogawa et al., 2003). Subjective stress scores and salivary cortisol were unaffected by sleep deprivation, while tiredness and performance selectively deteriorated. Taken together these findings suggest that previous observations of increased arterial pressure variability levels after one night sleep deprivation (Zhong et al., 2005) might rather reflect the disturbing effect of psychological stress on autonomic arousal, particularly in cases where arterial pressure was assessed with ABPM (Lusardi et al., 1996). A role of psychological factors is also likely in those instances where prolonged short sleeping hours are considered (Gangwisch et al., 2006; Gottlieb et al., 2006; Tochikubo et al., 1996). In susceptible individuals therefore, short term sleep deprivation might induce a condition of psychological stress, mediating sympathetic arousal and an increase in arterial pressure or its variability (Lucini et al., 2005; Mancia, 2000); conversely in normal healthy humans one night sleep deprivation does
4.1. Baseline stability In line with previous findings on repeated resting hemodynamic and MSNA variability data (van de Borne et al., 1997), not only we confirm that over relatively short periods of time baseline hemody-
80
M. Pagani et al. / Autonomic Neuroscience: Basic and Clinical 145 (2009) 76–80
not lead to signs of sympathetic arousal and increased arterial pressure or variability, but dissociates performance, that is reduced, from autonomic regulation, where signs of vagal activation seem present. A different condition is provided by individuals with disturbed sleep because of OSA, where sympathetic arousal is well documented (Grassi et al., 2005; Narkiewicz et al., 1998). 4.4. Limitations Important limitations should be indicated, such as the use of indirect, non invasive measures that were suggested by methodological considerations, and were also imposed by the multiple recording sessions. Also sleep quality was not directly assessed but was judged from self reports of subjects. These subjects moreover may be considered somewhat “better than usual”, because they were enrolled after a severe clinical selection routine, inclusive of a specific psychological evaluation. In addition isolation and confinement was obtained in easy, non-demanding, housing conditions. This allowed, however, to obtain data, at variance with other studies (Zhong et al., 2005), in a unique environment, that shielded subjects from uncontrolled, external stressors. 4.5. Perspectives Although based on a small highly selected group of healthy volunteers, we believe that our study strikes a balance between completeness of findings and minimal experimental intrusiveness. We conclude that one night sleep deprivation dissociates performance from autonomic regulation. In fact the observed increase in tiredness and reduction in performance is not accompanied by a rise in arterial pressure or its variability, and does not produce signs of sympathetic overactivity but, if anything, suggests an augmented cardiac vagal drive. We hope that these findings might further support the use of indirect non-invasive techniques to address the “autonomic” component in the emerging field of personalized medicine (Turner et al., 2007). Acknowledgments This study was partially supported by the following grants: ASI (Project on Disorders of Motor and Cardiorespiratory Control, DCMC), ESA, FIRST. Disclosure statement This was not an industry supported study. The authors have reported no financial conflicts of interest. References Ahn, A.C., Tewari, M., Poon, C., Phillips, R.S., 2006. The clinical applications of a system approach. PLoS MEDICINE 3, 0956–0960. Badilini, F., Pagani, M., Porta, A., 2005. Heartscope: a software tool addressing autonomic nervous system regulation. Computers in Cardiology 32, 259–262. Bertinieri, G., Di Rienzo, M., Cavallazzi, A., Ferrari, A.U., Pedotti, A., Mancia, G., 1985. A new approach to analysis of the arterial baroreflex. Journal of Hypertension 3, S79–S81. Gangwisch, J.E., Heymsfield, S.B., Boden-Albala, B., Buijs, R.M., Kreier, F., Pickering, T.G., Rundle, A.G., Zammit, G.K., Malaspina, D., 2006. Short sleep duration as a risk factor for hypertension: analyses of the first National Health and Nutrition Examination Survey. Hypertension 47, 833–839. Gilberg, M., Kecklund, G., Åkerstedt, T., 1994. Relation between performance and subjective ratings of sleepiness during a night awake. Sleep 17, 236–241. Gottlieb, D.J., Redline, S., Nieto, F.J., Baldwin, C.M., Newman, A.B., Resnick, H.E., Punjabi, N.M., 2006. Association of usual sleep duration with hypertension: the Sleep Heart Health Study. Sleep 29, 1009–1014. Grassi, G., Facchini, A., Trevano, F.Q., Dell'Oro, R., Arenare, F., Tana, F., Bolla, G., Monzani, A., Robuschi, M., Mancia, G., 2005. Obstructive sleep apnea-dependent and independent adrenergic activation in obesity. Hypertension 46, 321–325. Haken, H., 1983. Synergetics — an introduction: Non equilibrium Phase Transitions and Self-Organization in Physics, Chemistry and Biology. Third edition and Enlarged Edition.
Irwin, M.R., Ziegler, M., 2005. Sleep deprivation potentiates activation of cardiovascular and catecholamine responses in abstinent alcoholics. Hypertension 45, 252–257. Karasek, R., Baker, D., Marxer, D., Ahlbom, A., Theorell, T., 1981. Job decision latitude, job demands, and cardiovascular disease: a prospective study of Swedish men. American Journal of Public Health 71, 694–705. Kato, M., Phillips, B.G., Sigurdsson, G., Narkiewicz, K., Pesek, C.A., Somers, V.K., 2000. Effects of sleep deprivation on neural circulatory control. Hypertension 35, 1173–1175. Lucini, D., Norbiato, G., Clerici, M., Pagani, M., 2002. Hemodynamic and autonomic adjustments to real life stress conditions in humans. Hypertension 39, 184–188. Lucini, D., Di Fede, G., Parati, G., Pagani, M., 2005. Impact of chronic psychosocial stress on autonomic cardiovascular regulation in otherwise healthy subjects. Hypertension 46, 1201–1206. Lucini, D., Riva, S., Pizzinelli, P., Pagani, M., 2007. Stress management at the worksite: reversal of symptoms profile and cardiovascular dysregulation. Hypertension 49, 1–2. Lucini, D., Cannone, V., Malacarne, M., Bruno, D., Beltrami, S., Pizzinelli, P., Piazza, E., Di Fede, G., Pagani, M., in press. Evidence of autonomic dysregulation in otherwise healthy cancer caregivers: a possible link with health hazard. European Journal of Cancer. Available on line: http://dx.doi.org/10.1016/j.ejca.2008.08.006. Lusardi, P., Mugellini, A., Preti, P., Zoppi, A., Derosa, G., Fogari, R., 1996. Effects of a restricted sleep regimen on ambulatory blood pressure monitoring in normotensive subjects. American Journal of Hypertension 9, 503–505. Mancia, G., 2000. White coat effect. Innocuous or adverse phenomenon? American Journal of Hypertension 21, 1647–1648. Massimini, M., Porta, A., Mariotti, M., Malliani, A., Montano, N., 2000. Heart rate variability is encoded in the spontaneous discharge of thalamic somatosensory neurones in cat. Journal of Physiology 526 (Pt 2), 387–396. Narkiewicz, K., Pesek, C.A., Kato, M., Phillips, B.G., Davison, D.E., Somers, V.K., 1998. Baroreflex control of sympathetic nerve activity and heart rate in obstructive sleep apnea. Hypertension 32, 1039–1043. Ogawa, Y., Kanbayashi, T., Saito, Y., Takahashi, Y., Kitajima, T., Takahashi, K., Hishikawa, Y., Shimizu, T., 2003. Total sleep deprivation elevates blood pressure through arterial baroreflex resetting: a study with microneurographic technique. Sleep 26, 986–989. Pagani, M., Malliani, A., 2000. Interpreting oscillations of muscle sympathetic nerve activity and of heart rate variability. Journal of Hypertension 18, 1709–1719. Pagani, M., Lombardi, F., Guzzetti, S., Rimoldi, O., Furlan, R., Pizzinelli, P., Sandrone, G., Malfatto, G., Dell'Orto, S., Piccaluga, E., Turiel, M., Baselli, G., Cerutti, S., Malliani, A., 1986. Power spectral analysis of heart rate and arterial pressure variabilities as a marker of sympathovagal interaction in man and conscious dog. Circulation research 58, 178–193. Pagani, M., Somers, V.K., Furlan, R., Dell'Orto, S., Conway, J., Baselli, G., Cerutti, S., Sleight, P., Malliani, A., 1988. Changes in autonomic regulation induced by physical training in mild hypertension. Hypertension 12, 600–610. Pagani, M., Iellamo, F., Lucini, D., Pizzinelli, P., Castrucci, F., Peruzzi, G., Malliani, A., 1995. Adaptational changes in the neural control of cardiorespiratoy function in a confined environment: the CNEC#3 experiment. Acta Astronautica 36, 449–461. Pagani, M., Montano, N., Porta, A., Malliani, A., Abboud, F.M., Birkett, C.L., Somers, V.K., 1997. Relationship between spectral components of cardiovascular variabilities and direct measures of muscle sympathetic nerve activity in humans. Circulation 95, 1441–1448. Palinkas, L.A., Houseal, M., 2000. Stages of change in mood and behavior during a winter in Antarctica. Environment and Behavior 32, 128–141. Rozanski, A., Blumenthal, J.A., 1999. Impact of psychological factors on the pathogenesis of cardiovascular disease and implications for therapy. Circulation 99, 2192–2217. Shimamiya, T., Terada, N., Hiejima, Y., Wakabayashi, S., Kasai, H., Mohri, M., 2004. Effects of 10-day confinement on the immune system and psychological aspects in humans. Journal of Applied Physiology 97, 920–924. Task Force of the European Society of Cardiology, the North American Society of Pacing and Electrophysiology, 1996. Heart rate variability. Standards of measurement, physiological interpretation, and clinical use. Circulation 93, 1043–1065. Tochikubo, O., Ikeda, A., Miyajima, E., Ishii, M., 1996. Effects of insufficient sleep on blood pressure monitored by a new multibiomedical recorder. Hypertension 27, 1318–1324. Tropeano, A.I., Roudot-Thoraval, F., Badoual, T., Goldenberg, F., Dolbeau, G., Gosse, P., quin-Mavier, I., 2006. Different effects of ambulatory blood pressure monitoring on subjective and objective sleep quality. Blood Pressure Monitoring 11, 315–320. Turner, S.T., Schwartz, G.L., Boerwinkle, E., 2007. Personalized medicine for high blood pressure. Hypertension 50, 1–5. van de Borne, P., Montano, N., Zimmerman, B.G., Pagani, M., Somers, V.K., 1997. Relationship between repeated measures of hemodynamics, muscle sympathetic nerve activity and the spectral oscillations. Circulation 96, 4326–4332. Verdecchia, P., Angeli, F., Borgioni, C., Gattobigio, R., Reboldi, G., 2007. Ambulatory blood pressure and cardiovascular outcome in relation to perceived sleep deprivation. Hypertension 49, 777–783. Zhong, X., Hilton, H.J., Gates, G.J., Jelic, S., Stern, Y., Bartels, M.N., Demeersman, R.E., Basner, R.C., 2005. Increased sympathetic and decreased parasympathetic cardiovascular modulation in normal humans with acute sleep deprivation. Journal of applied physiology 98, 2024–2032.
Contents lists available at ScienceDirect
Autonomic Neuroscience: Basic and Clinical j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / a u t n e u
Hemodynamic, autonomic and baroreflex changes after one night sleep deprivation in healthy volunteers Massimo Pagani a,⁎, Paolo Pizzinelli a, Anne Pavy-Le Traon b, Cinzia Ferreri a, Silvia Beltrami a, Marie-Pierre Bareille b, Marie-Claude Costes-Salon b, Stéphane Béroud b, Olivier Blin c, Daniela Lucini a, Pierre Philip d a
Centro di Ricerca sulla Terapia Neurovegetativa, Dipartimento Scienze Cliniche “L. Sacco”, University of Milano, Milano, Italy Medes, Clinique Spatiale, Hôpital Rangueil-Larrey, Toulouse, France c Institut des Neurosciences Cognitives de la Mediterranee, Faculte de Medecine, Universite de la Mediterranee, Marseille, France d Sleep Clinic, Bordeaux University School of Medicine, Bordeaux, France b
a r t i c l e
i n f o
Article history: Received 10 June 2008 Received in revised form 30 September 2008 Accepted 2 October 2008 Keywords: Hypertension Sleep deprivation Autonomic nervous system Baroreflex Stress Cardiovascular risk
a b s t r a c t Background: Sleep disorders are associated to a number of cardiovascular disturbances that might increase cardiovascular risk. Sleep deprivation, in particular, might, by inducing autonomic dysregulation, raise arterial pressure and hypertensive risk. Available evidence however is contradictory. Methods: We tested the main hypothesis that one night sleep deprivation in 24 volunteers might alter hemodynamics (heart rate and Arterial Pressure — AP), autonomic regulation (mono and bivariate spectral analysis of RR and non invasive AP variability) and baroreflex control (spectral index alpha and spontaneous baroreflex slope), performance indices (reaction time) and subjective stress (questionnaires and salivary cortisol). Volunteers were studied in normal living conditions and while kept in isolation and confinement, to test the presence of possible bias related to environmental stress. Results: Results indicate that there were no differences between normal living conditions and isolation and confinement (Intraclass Correlation Coefficient N 0.75 for most variables). Conversely, after one night sleep deprivation subjects felt tired (p b 0.05), and performance deteriorated (p b 0.05), while cortisol profile was substantially maintained, hemodynamic parameters did not change and HRV and index alpha increased slightly. Conclusions: Findings support the contention that one night sleep deprivation, in absence of significant additional stress or disturbances, does not lead to increased arterial pressure values or to changes in autonomic or baroreflex profiles that could conceivably favor hypertension development, but induces the expected increase in tiredness and reduction in performance. © 2008 Elsevier B.V. All rights reserved.
1. Introduction Sleep disorders are associated to increased cardiovascular risk and in particular to elevated hypertension risk (Gangwisch et al., 2006). Among the various factors that might be implicated several investigators on the basis of indirect evidence suspected an important role of increased sympathetic activity (Lusardi et al., 1996; Zhong et al., 2005), possibly emphasized by poor sleep quality, such as might occur with the Ambulatory Blood Pressure Monitoring (ABPM) technique (Tropeano et al., 2006), or because of the facilitating effects of strenuous work (Tochikubo et al., 1996) or alcohol addiction (Irwin and Ziegler, 2005). Direct recordings of Muscle Sympathetic efferent
⁎ Corresponding author. Telematica per la Medicina e la Formazione, Ospedale “L. Sacco”, Università di Milano, Via G.B. Grassi, 74, 20157 Milano, Italy. Tel.: +39 02 39042802; fax: +39 02 50319823. E-mail address: [email protected] (M. Pagani). 1566-0702/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.autneu.2008.10.009
Nerve Activity (MSNA), with the exception of extreme conditions such as Obstructive Sleep Apnea (OSA), where sympathetic overactivity is a well known key player (Narkiewicz et al., 1998), do not clarify the issue: in fact, while Ogawa et al. (2003) suggested that sleep deprivation might increase sympathetic activity and, as a consequence, raise arterial pressure, Kato et al. (2000) reported that if anything MSNA was reduced after one night sleep deprivation. Findings might also be contingent upon the specific experimental model that was employed, such as a one night sleep deprivation (Lusardi et al., 1996; Ogawa et al., 2003; Zhong et al., 2005) in normotensive individuals, or a prolonged period of short sleeping hours (Gangwisch et al., 2006) in hypertensive subjects (Verdecchia et al., 2007). In addition, an unfavorable psychosocial environment (Rozanski and Blumenthal, 1999) might lead to a state of sympathetic arousal and thus to higher arterial pressure levels, particularly in patients complaining of chronic stress (Lucini et al., 2005). This complexity, although alluded to, was not usually explicitly addressed in most prior studies. Thus sleep
M. Pagani et al. / Autonomic Neuroscience: Basic and Clinical 145 (2009) 76–80
deprivation and difficult living or working conditions might conjure in determining a deleterious mix of elevated subjective stress and autonomic dysfunction, leading to increased arterial pressure. Of potential clinical interest, some of these alterations can be reverted towards normal with behavioral techniques (Lucini et al., 2007). In order to elucidate the effects of sleep deprivation on individual hemodynamic and stress measures, it might be therefore critical to shield experimental subjects from uncontrolled external and environmental stressors. These requirements can be met in conditions of isolation and confinement, where light and sleep activity schedule can be strictly determined. This setting, although it represents a change from everyday life, permits to maintain environmental stress to a preset level, and it is likely to confer greater stability to baseline autonomic regulation (Pagani et al., 1995). Accordingly we first tested the effects of isolation and confinement, per se, on stability of hemodynamics, autonomic regulation, performance indices and subjective stress in a group of healthy volunteers. Secondly we assessed, on the same variables, the effects of one night sleep deprivation, while controlling for environmental stress and bias. Direct disturbances to subjects well being were further minimized by using only rigorously non invasive recording equipments, and having experimental subjects perform measures themselves, to prevent a possible “white coat effect” (Mancia, 2000).
2. Materials and methods 2.1. Study subjects 24 young healthy subjects (12 male, 12 female; age 27–45) who had previously passed a formal selection at MEDES Clinic (Toulouse) were studied. Briefly, the selection included a complete physical and psychological (including morning–night sleep type, no sleep disorders, no psychopathology, no alcohol, no smoking, no shift-work) assessment, and a battery of clinical measures (normal blood tests: glucose, urea, creatinine, electrolytes (Na+, K+, CI−, P, HCO3−, Ca++), transaminases (ALT and AST), γGT, alkaline phosphatases, hepatitis and HIV serology. The mean height of the crew was 172 ± 8 cm, the mean weight was 66.6 ± 13.3 kg and the mean Body Mass Index (BMI) was 22.4 ± 2.9 kg/m2. 2.2. Protocol All subjects were randomly exposed to 2 different environmental conditions: Medes (M) and Home (H), in a Latin square design. Condition M: lasted 8 days from M1 (first day) to M8 (eighth day), and was designed in order to obtain the controlled conditions of a monotonous working environment in terms of low physical activity, social isolation, confinement and light exposure. To this aim subjects were housed in the MEDES “Clinical Research Facility” with specific light conditions: 300 lux in living area and 50 lux in passageways during the artificial day (16 h); and 50 lux in living areas and 10 to 50 lux in passageways during night periods (8 h: low light). Notice that because subjects chose voluntarily to participate to the experiment and could exit from it in any moment, if they decided to, these conditions, according to the Karasek model of stress-reward should be considered as low chronic work stress (Karasek et al., 1981). At the end of day 8 (M8) subjects were kept awake for 24 h under low light exposure thus undergoing a one-night sleep deprivation under continued TV surveillance. To further ensure that subjects did not inadvertently fall asleep, they had to perform a simple computerized task every 2 h. At the end of this sleep deprivation period volunteers were subjected to final tests the morning of day 9 (M9). Subjects were isolated in groups of six (3 men and 3 women). Condition H: subjects were also studied at the beginning (H1) and at the end (H8) of an 8-day period during which they followed a
77
normal life routine at home, with however a strict sleep-wake schedule (23 p.m.–7 a.m.), and avoiding heavy activities. Based on previous experience (Pagani et al., 1995), subjects were individually trained once for about 4 h by two investigators (CF and PP) to collect data by themselves, with a high level of efficiency. Overall 85% of data were fully analyzable, and only 7% of data were lost. This approach would most likely avoid any “white coat effect” (Mancia, 2000). All recordings (both for condition M and H) were performed in MEDES Clinic in the morning, between 9 and 11 a.m., at least 2 h after a light breakfast not containing caffeine. In all subjects hemodynamic, endocrine, psychological, autonomic and performance measures were obtained, as specified below. Recorded variables: using standard Ag–AgCl electrodes, both the ECG (CM5) and the respiratory signal (transthoracic impedance) were monitored with a two-way radiotelemetry system (Marazza, Monza, Milan, Italy). The arterial pressure waveform was continuously estimated non invasively with a plethysmographic device (Portapres, TNO Institute of Applied Physics Biomedical Instrumentation, Amsterdam). Data were acquired on a PC, using an acquisition rate of 300 samples/channel/s. All subjects were studied in resting conditions (lying down with 15° back support): after a 10 minute period, allowed for stabilization, a 10 minute rest period was obtained. 2.2.1. Assessment of overall stress level A battery of non-invasive tests gauged the overall stress level: – Endocrine involvement (Lucini et al., 2002) was assessed by measuring by radioimmunoassay (DPC, Los Angeles, California, USA) free salivary cortisol levels which reflect the concentration of free hormone in plasma. Samples were in all instances obtained at approximately the same hours (10:30 a.m. ±1 and 6 p.m. ±1) to take circadian variation into account. – Psychological involvement was gauged by a battery of questionnaires (Lucini et al., 2002), providing self-rated scales (higher values indicate higher degrees of symptoms) that, in line with previous and more recent studies, focuses on appraisal, coping and health. In brief: – The appraisal of stress and tiredness was assessed by a global scoring index (0–10) for each measure. – A symptom list was used to score 18 somatic complaints (score 0–180) (4SQ). – This questionnaire has been validated against Cognitive Behavioral Assessment, Italian version (Lucini et al., in press). 2.2.2. Study of autonomic regulation Indirect autonomic parameters were assessed with a non-invasive approach based on spectral analysis of RR, Systolic Arterial Pressure (SAP) and respiration variability (Pagani et al., 1986; Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology, 1996). Analysis was conducted using an ad hoc software designed to perform spectral analysis of cardiovascular signals (HeartScope) (Badilini et al., 2005): from the ECG derived tachogram an autoregressive approach provides both absolute (i.e. [ms2]) and normalized units (i.e. [nu]) spectral powers of Low (LF) and High Frequency (HF) components. The normalized power of these two components provides gross estimates of the sympathetic and vagal oscillatory modulations of the sinoatrial (SA) node, and of their balance (from LF/ HF ratio) (Pagani et al., 1986). Spectral analysis was also performed on the systolic arterial pressure, in order to estimate sympathetic oscillatory vasomotor modulation from the absolute power of the LF component (HF power of systolic arterial pressure variability is a measure of mechanical effects of respiration on the peripheral circulation) (Pagani et al., 1997).
78
M. Pagani et al. / Autonomic Neuroscience: Basic and Clinical 145 (2009) 76–80
Table 1 Summary statistics for baseline autonomic, humoral, psychological and performance indices REST
Measure units
Average std dev
RR RR σ2 LFRR LFRR HFRR HFRR LF/HFRR SAP SAP σ2 LFSAP HFSAP BRS Alpha Cort_M Cort_A Stress Tired S4Q SRTT SRTT 10%L
ms ms2 ms2 nu ms2 nu – mm Hg ms2 mm Hg2 mm Hg2 ms/mm Hg ms/mm Hg ng/ml ng/ml A.U. A.U. A.U. ms ms
914.84 ± 135.44 5278.60 ± 5415.66 2190.28 ± 3012.52 55.65 ± 20.47 1186.28 ± 1536.88 39.68 ± 19.25 2.28 ± 2.26 113.94 ± 13.69 41.19 ± 39.32 8.68 ± 11.26 2.26 ± 2.39 21.08 ± 9.95 21.25 ± 11.67 8.03 ± 3.29 3.26 ± 1.35 0.80 ± 1.32 1.33 ± 1.59 7.43 ± 12.82 383.64 ± 131.95 682.90 ± 351.766
working conditions at MEDES (M), or comparing Day 1 or Day 8 (data not shown for simplicity). Consequently average data and standard deviations are reported from combined experiments of days H1, H8, M1, M8 in Table 1 that accordingly provides summary baseline evaluation of all hemodynamic and autonomic variables, in conditions of normal sleep routine. Intraclass Correlation Coefficients (ICC) computed while a regular sleep schedule was maintained (H1, H8, M1, M8) further confirms the strong stability of baseline data across days and contexts. It is apparent that all variables in spite of a well known high variability between subjects possess a high degree of stability (Fig. 1), with a few exceptions: SAP variance and LFSAP. Notably ICCs computed for time domain measures (RR, SAP, BRS), for several frequency domain indices
Average ± standard deviation values of indices from 24 healthy volunteers at rest, combining results obtained in 4 different instances (at Home and at MEDES, day 1 and day 8). Abbreviations: σ2 = variance, LF = Low Frequency, nu = Normalized Unit, HF = High Frequency, SAP = Systolic Arterial Pressure, BRS = Baroreceptor Reflex Sensitivity, Alpha = frequency domain index of baroreflex sensitivity, Cort_M = salivary cortisol (morning evaluation), Cort_A = salivary cortisol (afternoon evaluation), S4Q = Subjective Stress-Related Somatic Symptoms Questionnaire, SRTT = Simple Reaction Time Test, SRTT10%L = mean of the ten percent slowest RT trials.
The center frequency of the HF component of the respiratory signals, obtained with a similar procedure, assessed the main respiratory frequency. This latter usually should be higher than 0.14 Hz, in order to avoid entrainment (Pagani et al., 1986). From the simultaneous analysis of arterial pressure and RR interval variability a frequency domain index (Alpha) (Pagani et al., 1988) can be derived at both LF and HF frequencies which provides a measure of the overall gain of the arterial pressure–heart period relationship. The following formula providing an average value (Lucini et al., 2002) was employed: 1=2 Alpha = ðPRR =PSAP Þ1=2 LF + ðPRR =PSAP ÞHF =2 The program also provides values of the time domain spontaneous Baroreflex slope, as originally described by Bertinieri et al. (1985). Performance was estimated with a Simple Reaction Time Test (SRTT), which lasted 10 min and was based on a PSION personal organizer (Gilberg et al., 1994). 2.2.3. Statistical analysis Data are presented as Mean ± Standard Deviation. Significance of differences between repeated measures and contexts was estimated with a mixed model or GLM analysis as appropriate, followed by individual contrasts. Intraclass Correlation analysis (van de Borne et al., 1997) was used to assess the stability of measures across days M1, M8, H1, H8, treating data as independent measures. Significance was set at p b 0.05. Computations were performed with a commercial statistical package (SPSS13). 3. Results 3.1. Baseline measures No significant differences were found between baseline data obtained during the free living conditions at home (H), or during the controlled
Fig. 1. Intraclass Correlation Coefficients for variables related to Heart Rate Variability (A), arterial pressure, respiration and BRS (B), stress and performance (C), considering the effects of time and context: ICC was computed across days H1, H8, M1 and M8 of each condition (M and H). A dotted line at 0.75 highlights those variables above the value demonstrating high stability. Abbreviations: LF = Low Frequency, σ2 = variance, HF = High Frequency, SAP = Systolic Arterial Pressure, Alpha = frequency domain index of baroreflex sensitivity, BRS = Baroreceptor Reflex Sensitivity, S4Q = Subjective Stress-Related Somatic Symptoms Questionnaire, SRTT = Simple Reaction Time Test, SRTT10%L = mean of the ten percent slowest RT trials, Cort(N) = Night Cortisol, Cort(M) = Morning Cortisol.
M. Pagani et al. / Autonomic Neuroscience: Basic and Clinical 145 (2009) 76–80 Table 2 Mean ± standard deviation of stress and performance parameters assessed before and after moderate sleep deprivation
RR RR σ2 LFRR LFRR HFRR HFRR LF/HFRR SAP SAP σ2 LFSAP HFSAP BRS Alpha Cort_M Cort_A STRESS TIRED S4Q SRTT SRTT 10%L
Measure units
MEDES M8
M9
ms ms2 ms2 nu ms2 nu – mm Hg ms2 mm Hg2 mm Hg2 ms/mm Hg ms/mm Hg ng/ml ng/ml A.U. A.U. A.U. ms ms
895.17 ± 161.64 4610.35 ± 5143.10 2315.15 ± 3528.00 59.12 ± 19.34 988.03 ± 1071.31 36.56 ± 17.75 2.21 ± 1.50 113.44 ± 10.43 35.04 ± 19.43 11.02 ± 18.49 2.58 ± 3.05 18.61 ± 7.57 20.03 ± 10.51 7.87 ± 3.91 3.61 ± 0.98b 1.13 ± 1.69 1.43 ± 1.85 6.57 ± 11.72 379.63 ± 131.03 726.42 ± 429.89
990.20 ± 147.14a 7448.26 ± 8803.18 2218.48 ± 2829.37 57.62 ± 17.46 1490.79 ± 2182.78 37.51 ± 16.68 2.32 ± 2.20 113.13 ± 11.40 42.01 ± 18.38 7.33 ± 6.08 1.66 ± 1.12 25.87 ± 20.25a 27.82 ± 16.10a 7.06 ± 3.22 3.48 ± 1.05b 1.09 ± 1.90 5.35 ± 2.42a 6.96 ± 11.04 508.44 ± 182.80a 1107.52 ± 492.83a
M8 = day 8 at Medes, M9 = day 9 at Medes, i.e. after sleep deprivation. Abbreviations: σ2 = variance, LF = Low Frequency, nu = Normalized Unit, HF = High Frequency, SAP = Systolic Arterial Pressure, BRS = Baroreceptor Reflex Sensitivity, Alpha = frequency domain index of baroreflex sensitivity, Cort_M = salivary cortisol (morning evaluation), Cort_A = salivary cortisol (afternoon evaluation), S4Q = Subjective Stress-Related Somatic Symptoms Questionnaire, SRTT = Simple Reaction Time Test, SRTT10%L = mean of the ten percent slowest RT trials. a Significant p b 0.05 M8 vs M9. b Significant p b0.05 morning vs night.
(RR σ2, LFRR(nu), HFRR(nu), LF/HFRR; HFRESP, Alpha index) and for stress and performance (SRTT) indices were all above the 0.75 value, which is usually taken to indicate a high degree of consistency. 3.2. Sleep deprivation Table 2 focuses on the effect of sleep deprivation: with the exception of prolonged RR and increased BRS and α index, there were no changes in spectral parameters. Salivary cortisol was not affected and maintained its circadian variations throughout the experiment. Tiredness and SRTT scores increased significantly (i.e. performance was impaired).
79
namic and cardiac autonomic indices have a high stability, but we show in addition that hormonal and psychological stress indicators are also highly stable in the setting of the present experiment. In particular, the hypothetical stressful condition of isolation and confinement, as in MEDES environment, was not reflected in any measurable change in the multidimensional parameters under scruting. Our data moreover extend the notion of elevated stability (van de Borne et al., 1997) over time also for bivariate indices of baroreflex regulation, inclusive of phase and coherence between arterial pressure and RR variability oscillations in the LF and HF regions, as well as for indices of motor performance, as assessed by a simple reaction time task. Similarity of findings was maintained also comparing data obtained during normal life at home, or during a controlled work routine, in MEDES environment. It should be added that, because subjects were in control of their work relationship, MEDES environment provided the additional advantage of a stable situation, as far as work stress is concerned (Karasek et al., 1981). This contention is supported by the physiologically low cortisol and stress score values. Baseline values of hemodynamic, autonomic and baroreflex regulation were all within the normal range of our laboratory (Lucini et al., 2002). Accordingly, isolation and confinement per se do not seem to introduce autonomic or subjective signs of stress. Different results may be obtained in extreme conditions (Palinkas and Houseal, 2000) or considering other targets, such as the immune system or more demanding environments (Shimamiya et al., 2004). 4.2. Autonomic assessment In our investigation, based on multiple recordings several days apart, only non invasive non intrusive recordings were employed, as per routine in our clinical laboratory (Lucini et al., 2005). Obviously, direct electroneurographic recordings, as in the case of MSNA (Pagani et al., 1997), might provide more robust physiology, although at the risk of introducing an unknown stress related bias. The multiparameter approach, although not providing direct data about sympathetic efferent traffic is likely to furnish more abundant information (Pagani and Malliani, 2000) on the complexity of autonomic cardiovascular regulatory processes (Ahn et al., 2006), and their changes, particularly in the oscillatory domain (Massimini et al., 2000). These changes have been shown to track faithfully prehypertensive risk (Lucini et al., 2002), its link with stress symptoms (Lucini et al., 2005) and possible behavioral treatment (Lucini et al., 2007).
4. Discussion
4.3. Sleep deprivation
This study provides evidence against the hypothesis that one night sleep deprivation might increase arterial pressure or its variability and sympathetic drive in healthy individuals. Unintended emotional bias, such as that produced by unfavorable or uncontrolled work environment, or by invasive examinations (e.g., needles as in MSNA or in plasma catecholamine determinations), was virtually excluded by the strict use of controlled non invasive, non intrusive experimental conditions. Moreover, the potential weakness of indirect autonomic assessment was counterbalanced by the enhanced information (Pagani and Malliani, 2000) provided by the use of several concurrent indices (Haken,1983, Ahn et al., 2006): hemodynamic, autonomic, hormonal, psychological and of performance. This approach, could furnish functional elements that could be included in the design of personalized preventive clinical approaches (Turner et al., 2007).
After one night sleep deprivation we observed no signs of increases in arterial pressure or its variability. RR interval, BRS variance and the index alpha increased slightly, suggesting an increase in cardiac vagal regulation. This finding is in line with previous reports of a mirror reduction of MSNA activity after one night sleep deprivation (Ogawa et al., 2003). Subjective stress scores and salivary cortisol were unaffected by sleep deprivation, while tiredness and performance selectively deteriorated. Taken together these findings suggest that previous observations of increased arterial pressure variability levels after one night sleep deprivation (Zhong et al., 2005) might rather reflect the disturbing effect of psychological stress on autonomic arousal, particularly in cases where arterial pressure was assessed with ABPM (Lusardi et al., 1996). A role of psychological factors is also likely in those instances where prolonged short sleeping hours are considered (Gangwisch et al., 2006; Gottlieb et al., 2006; Tochikubo et al., 1996). In susceptible individuals therefore, short term sleep deprivation might induce a condition of psychological stress, mediating sympathetic arousal and an increase in arterial pressure or its variability (Lucini et al., 2005; Mancia, 2000); conversely in normal healthy humans one night sleep deprivation does
4.1. Baseline stability In line with previous findings on repeated resting hemodynamic and MSNA variability data (van de Borne et al., 1997), not only we confirm that over relatively short periods of time baseline hemody-
80
M. Pagani et al. / Autonomic Neuroscience: Basic and Clinical 145 (2009) 76–80
not lead to signs of sympathetic arousal and increased arterial pressure or variability, but dissociates performance, that is reduced, from autonomic regulation, where signs of vagal activation seem present. A different condition is provided by individuals with disturbed sleep because of OSA, where sympathetic arousal is well documented (Grassi et al., 2005; Narkiewicz et al., 1998). 4.4. Limitations Important limitations should be indicated, such as the use of indirect, non invasive measures that were suggested by methodological considerations, and were also imposed by the multiple recording sessions. Also sleep quality was not directly assessed but was judged from self reports of subjects. These subjects moreover may be considered somewhat “better than usual”, because they were enrolled after a severe clinical selection routine, inclusive of a specific psychological evaluation. In addition isolation and confinement was obtained in easy, non-demanding, housing conditions. This allowed, however, to obtain data, at variance with other studies (Zhong et al., 2005), in a unique environment, that shielded subjects from uncontrolled, external stressors. 4.5. Perspectives Although based on a small highly selected group of healthy volunteers, we believe that our study strikes a balance between completeness of findings and minimal experimental intrusiveness. We conclude that one night sleep deprivation dissociates performance from autonomic regulation. In fact the observed increase in tiredness and reduction in performance is not accompanied by a rise in arterial pressure or its variability, and does not produce signs of sympathetic overactivity but, if anything, suggests an augmented cardiac vagal drive. We hope that these findings might further support the use of indirect non-invasive techniques to address the “autonomic” component in the emerging field of personalized medicine (Turner et al., 2007). Acknowledgments This study was partially supported by the following grants: ASI (Project on Disorders of Motor and Cardiorespiratory Control, DCMC), ESA, FIRST. Disclosure statement This was not an industry supported study. The authors have reported no financial conflicts of interest. References Ahn, A.C., Tewari, M., Poon, C., Phillips, R.S., 2006. The clinical applications of a system approach. PLoS MEDICINE 3, 0956–0960. Badilini, F., Pagani, M., Porta, A., 2005. Heartscope: a software tool addressing autonomic nervous system regulation. Computers in Cardiology 32, 259–262. Bertinieri, G., Di Rienzo, M., Cavallazzi, A., Ferrari, A.U., Pedotti, A., Mancia, G., 1985. A new approach to analysis of the arterial baroreflex. Journal of Hypertension 3, S79–S81. Gangwisch, J.E., Heymsfield, S.B., Boden-Albala, B., Buijs, R.M., Kreier, F., Pickering, T.G., Rundle, A.G., Zammit, G.K., Malaspina, D., 2006. Short sleep duration as a risk factor for hypertension: analyses of the first National Health and Nutrition Examination Survey. Hypertension 47, 833–839. Gilberg, M., Kecklund, G., Åkerstedt, T., 1994. Relation between performance and subjective ratings of sleepiness during a night awake. Sleep 17, 236–241. Gottlieb, D.J., Redline, S., Nieto, F.J., Baldwin, C.M., Newman, A.B., Resnick, H.E., Punjabi, N.M., 2006. Association of usual sleep duration with hypertension: the Sleep Heart Health Study. Sleep 29, 1009–1014. Grassi, G., Facchini, A., Trevano, F.Q., Dell'Oro, R., Arenare, F., Tana, F., Bolla, G., Monzani, A., Robuschi, M., Mancia, G., 2005. Obstructive sleep apnea-dependent and independent adrenergic activation in obesity. Hypertension 46, 321–325. Haken, H., 1983. Synergetics — an introduction: Non equilibrium Phase Transitions and Self-Organization in Physics, Chemistry and Biology. Third edition and Enlarged Edition.
Irwin, M.R., Ziegler, M., 2005. Sleep deprivation potentiates activation of cardiovascular and catecholamine responses in abstinent alcoholics. Hypertension 45, 252–257. Karasek, R., Baker, D., Marxer, D., Ahlbom, A., Theorell, T., 1981. Job decision latitude, job demands, and cardiovascular disease: a prospective study of Swedish men. American Journal of Public Health 71, 694–705. Kato, M., Phillips, B.G., Sigurdsson, G., Narkiewicz, K., Pesek, C.A., Somers, V.K., 2000. Effects of sleep deprivation on neural circulatory control. Hypertension 35, 1173–1175. Lucini, D., Norbiato, G., Clerici, M., Pagani, M., 2002. Hemodynamic and autonomic adjustments to real life stress conditions in humans. Hypertension 39, 184–188. Lucini, D., Di Fede, G., Parati, G., Pagani, M., 2005. Impact of chronic psychosocial stress on autonomic cardiovascular regulation in otherwise healthy subjects. Hypertension 46, 1201–1206. Lucini, D., Riva, S., Pizzinelli, P., Pagani, M., 2007. Stress management at the worksite: reversal of symptoms profile and cardiovascular dysregulation. Hypertension 49, 1–2. Lucini, D., Cannone, V., Malacarne, M., Bruno, D., Beltrami, S., Pizzinelli, P., Piazza, E., Di Fede, G., Pagani, M., in press. Evidence of autonomic dysregulation in otherwise healthy cancer caregivers: a possible link with health hazard. European Journal of Cancer. Available on line: http://dx.doi.org/10.1016/j.ejca.2008.08.006. Lusardi, P., Mugellini, A., Preti, P., Zoppi, A., Derosa, G., Fogari, R., 1996. Effects of a restricted sleep regimen on ambulatory blood pressure monitoring in normotensive subjects. American Journal of Hypertension 9, 503–505. Mancia, G., 2000. White coat effect. Innocuous or adverse phenomenon? American Journal of Hypertension 21, 1647–1648. Massimini, M., Porta, A., Mariotti, M., Malliani, A., Montano, N., 2000. Heart rate variability is encoded in the spontaneous discharge of thalamic somatosensory neurones in cat. Journal of Physiology 526 (Pt 2), 387–396. Narkiewicz, K., Pesek, C.A., Kato, M., Phillips, B.G., Davison, D.E., Somers, V.K., 1998. Baroreflex control of sympathetic nerve activity and heart rate in obstructive sleep apnea. Hypertension 32, 1039–1043. Ogawa, Y., Kanbayashi, T., Saito, Y., Takahashi, Y., Kitajima, T., Takahashi, K., Hishikawa, Y., Shimizu, T., 2003. Total sleep deprivation elevates blood pressure through arterial baroreflex resetting: a study with microneurographic technique. Sleep 26, 986–989. Pagani, M., Malliani, A., 2000. Interpreting oscillations of muscle sympathetic nerve activity and of heart rate variability. Journal of Hypertension 18, 1709–1719. Pagani, M., Lombardi, F., Guzzetti, S., Rimoldi, O., Furlan, R., Pizzinelli, P., Sandrone, G., Malfatto, G., Dell'Orto, S., Piccaluga, E., Turiel, M., Baselli, G., Cerutti, S., Malliani, A., 1986. Power spectral analysis of heart rate and arterial pressure variabilities as a marker of sympathovagal interaction in man and conscious dog. Circulation research 58, 178–193. Pagani, M., Somers, V.K., Furlan, R., Dell'Orto, S., Conway, J., Baselli, G., Cerutti, S., Sleight, P., Malliani, A., 1988. Changes in autonomic regulation induced by physical training in mild hypertension. Hypertension 12, 600–610. Pagani, M., Iellamo, F., Lucini, D., Pizzinelli, P., Castrucci, F., Peruzzi, G., Malliani, A., 1995. Adaptational changes in the neural control of cardiorespiratoy function in a confined environment: the CNEC#3 experiment. Acta Astronautica 36, 449–461. Pagani, M., Montano, N., Porta, A., Malliani, A., Abboud, F.M., Birkett, C.L., Somers, V.K., 1997. Relationship between spectral components of cardiovascular variabilities and direct measures of muscle sympathetic nerve activity in humans. Circulation 95, 1441–1448. Palinkas, L.A., Houseal, M., 2000. Stages of change in mood and behavior during a winter in Antarctica. Environment and Behavior 32, 128–141. Rozanski, A., Blumenthal, J.A., 1999. Impact of psychological factors on the pathogenesis of cardiovascular disease and implications for therapy. Circulation 99, 2192–2217. Shimamiya, T., Terada, N., Hiejima, Y., Wakabayashi, S., Kasai, H., Mohri, M., 2004. Effects of 10-day confinement on the immune system and psychological aspects in humans. Journal of Applied Physiology 97, 920–924. Task Force of the European Society of Cardiology, the North American Society of Pacing and Electrophysiology, 1996. Heart rate variability. Standards of measurement, physiological interpretation, and clinical use. Circulation 93, 1043–1065. Tochikubo, O., Ikeda, A., Miyajima, E., Ishii, M., 1996. Effects of insufficient sleep on blood pressure monitored by a new multibiomedical recorder. Hypertension 27, 1318–1324. Tropeano, A.I., Roudot-Thoraval, F., Badoual, T., Goldenberg, F., Dolbeau, G., Gosse, P., quin-Mavier, I., 2006. Different effects of ambulatory blood pressure monitoring on subjective and objective sleep quality. Blood Pressure Monitoring 11, 315–320. Turner, S.T., Schwartz, G.L., Boerwinkle, E., 2007. Personalized medicine for high blood pressure. Hypertension 50, 1–5. van de Borne, P., Montano, N., Zimmerman, B.G., Pagani, M., Somers, V.K., 1997. Relationship between repeated measures of hemodynamics, muscle sympathetic nerve activity and the spectral oscillations. Circulation 96, 4326–4332. Verdecchia, P., Angeli, F., Borgioni, C., Gattobigio, R., Reboldi, G., 2007. Ambulatory blood pressure and cardiovascular outcome in relation to perceived sleep deprivation. Hypertension 49, 777–783. Zhong, X., Hilton, H.J., Gates, G.J., Jelic, S., Stern, Y., Bartels, M.N., Demeersman, R.E., Basner, R.C., 2005. Increased sympathetic and decreased parasympathetic cardiovascular modulation in normal humans with acute sleep deprivation. Journal of applied physiology 98, 2024–2032.