Nocturnal variations in human lower leg subcutaneous blood flow related to sleep stages J. H. SINDRUP, J. KASTRUP, P. L. MADSEN, H. CHRISTENSEN, B. JORGENSEN, AND G. WILDSCHI0DTZ Departments of Clinical PhysiologylNuclear Medicine and Internal Medicine P, Bispebjerg Hospital, DK-2400 Copenhagen NV; Department of Clinical Neurophysiology, Gentofte Sygehus, DK-2900 Hellerup; and Department of Cardiology B, Rigshospitalet and University of Copenhagen, DK-2100 Copenhagen, Denmark SINDRUP, J. H., J. KASTRUP, P. L. MADSEN, H. CHRISTENSEN, B. JBRGENSEN, AND G. WILDSCHIBDTZ.NOC~U~~CZZ uaricztions in human lower leg subcutaneous blood flow related to sleep stages. J. Appl. Physiol. 73(4): 1246-1252, 1992.-Nocturnal subcutaneous adipose tissue blood flow rate was measured in the lower legs of 10 normal human subjects together with systemic arterial blood pressure, heart rate, and registration of sleep stages under ambulatory conditions. The 133Xe washout technique, portable CdTe(C1) detectors, and a portable data storage unit were used for measurement of blood flow rates. The sleep recordings were performed with a portable computerized sleep analysis system. In accordance with the results of previous studies, a hyperemic blood flow rate phase (mean increase 140%) for 100 min was observed -60 min after the subjects went to bed. The moment of onset of the hyperemic phase was closely related to the moment of onset of the first episode of deep sleep (stages 3 and 4). There was a significant (P < 0.01) overrepresentation of deep sleep in the hyperemic phase compared with adjacent phases, and rapid-eye-movement sleep predominantly occurred in the latter part of the night, when the subcutaneous blood flow rate was stable. The results of the present study are in accordance with current theories of the interrelationship between the thermoregulatory and the arousal state control systems and, thus, might suggest that the nightly subcutaneous hyperemia represents a thermoregulatory effector mechanism. adipose tissue blood flow rate; arousal state control system; heart rate; isotope washout technique; lower leg; microcirculation; nocturnal fluctuations; subcutaneous blood flow rate; systemic blood pressure; thermoregulation; xenon-133 washout technique
rate returned to the level measured at the beginning of the night period. Studies of regional variations in subcutaneous blood flow rate, by means of simultaneous measurements on different locations in the right and left lower legs, pointed to a central nervous or humoral elicitation of the nightly subcutaneous hyperemia, although local metabolic factors might participate as well (29). A significant fall in systemic blood pressure and subcutaneous vascular resistance was found in synchronism with the hyperemic blood flow phase (28). It has been shown that during the transitions in sleep stages, changes in peripheral vasomotor tone are influenced by both sympathetic outflow and changes in blood pressure (4). The purpose of the present study was to determine whether the nocturnal fluctuations in subcutaneous adipose tissue blood flow rate could be correlated to the different sleep stages. MATERIALS
AND METHODS
Ten lean normal human subjects were included in the study [age 28.2 t 3.5 (SD) yr]. Four were females (age 26.3 t 2.2 yr) and six were males (age 29.9 t 3.6 yr). Eight of these 10 subjects had participated in four previous studies of the nocturnal blood flow variations on the lower leg (28-31). None of the subjects had tachycardia or bradycardia. The study was approved by the local ethical committee, and informed consent was obtained from all participants before the investigation. Experimental Setup
IN RECENT STUDIES, measurements of subcutaneous blood flow rate over 12-20 h of ambulatory conditions were performed in the lower legs of normal human subjects (28-31). With the change from upright to supine position from day 1 at the beginning of the night period, an instantaneous blood flow rate increment of 3O-40% was observed in accordance with a decrease in central and local postural sympathetic vasoconstrictor activity (10, 17, 23). Approximately 90 min after the subjects went to bed, an additional blood flow rate increment of considerable magnitude was observed. The mean increase was 84%, but in several cases an increment of >200% was measured (30). The duration of this hyperemit phase was -100 min, after which the blood flow 1246
0161-7567/92 $2.00 Copyright
The 133Xewashout measurements. The blood flow rate measurements were performed on the medial aspect of the right lower leg, 10 cm proximal to the malleolar level. The depots of the tracer 133Xewere applied by the atraumatic epicutaneous labeling technique as described previously (28). Portable CdTe( Cl) semiconductor detectors were mounted over the central area of the depots. The detectors were fixed directly to the skin surface with a single layer of Mylar membrane interposed between the detector and the skin surface. Counts were accumulated at l-, 2-, or 4-min intervals and stored in a portable data storage unit (Memolog system 600, B. Simonsen Medical, Randers, Denmark). The measurements started 90 min after the labeling
0 1992 the American
Physiological
Society
Downloaded from www.physiology.org/journal/jappl at Karolinska Institutet University Library (130.237.122.245) on February 12, 2019.
NOCTURNAL
PERIPHERAL
BLOOD
FLOW
1247
REGULATION
procedure, between 7:00 and 9:00 P.M., and ended at approximately 9:00 A.M. the following day. During this period, which took place under ambulatory conditions, the subjects recorded their physical activities, time of meals, and sleeping periods. The subjects performed their normal activities during daytime (standing, walking, and sitting), and they went to bed at approximately 1 l:oo P.M., slept in supine position during the night, and got up at approximately 7:00 A.M. The subjects were not allowed to drink alcohol.
and only rate constants (h) from curve sections with a correlation coefficient (r) > 0.80 were taken into consideration. A few minor curve sections were rejected on the basis of this criterion; however, in no case were all data from a subject omitted. Irregularities on the curves were typical ly a res ult of either geometric distu .rbances between the detector and the isotope depot or temporary detector circuit noise. The curves were analyzed for goodness of fit and linearity by linear regression analysis, and the distributions Registration of onset of sleep and different sleep stages. of the residuals were inspected (28). According to the The sleep recordings were performed with a portable former, the majority of the curves could be divided into five phases as follows: phase 1 is day 1; phases 2,3, and 4 computerized sleep analysis system (Somnolog 3, Ventee, Hellerup, Denmark). The system measured the elec- are night; and phase 5 is day 2. Phase 2 was the initial troe lncephalogram (EEG), eye movements (with a p iezo- supine period in bed after standing. This period was electric transducer), and electromyogram (EMG) to without any central and local sympathetic or humoral the comply with the stan dard techniques (22). In addition, orthostatic vasoconstrictive activity. Therefore, the sound pressure level was measured to detect disturbblood flow level during this phase was chosen as reference level (index = l.OO), and the changes in blood flow ing noise, and hand movem .ents were measured to improve the arous al detection. From the raw data a primary rate durin .g the measurem .ent were calculated relative to feature extracti on was performed from the rele vant polyphase 2. graphic sleep parameters in 6-s epochs [for RechtschafDuring the night period, three distinct phases were defen and Kales (22) scoring]. These features were dis- tected, with significantly different isotope washout rates played and used in a visual scoring procedure with a (phases 2, 3, and 4) (28). The two break points of the 83.5% correlation to paper Rechtschaffen and Kales vi- three nocturnal phases were not associated with any sual scoring (33, 34). known events. The determination of these two break The EEG was recorded bipolar from F3 to A2 (the in- points was not solely based on visual inspection. First, ternational lo-20 system), the EEG amplifier frequency the logarithmically transformed washout curve of the response was 0.5-30 Hz (-3 dB) with a noise level
rate returned to the level measured at the beginning of the night period. Studies of regional variations in subcutaneous blood flow rate, by means of simultaneous measurements on different locations in the right and left lower legs, pointed to a central nervous or humoral elicitation of the nightly subcutaneous hyperemia, although local metabolic factors might participate as well (29). A significant fall in systemic blood pressure and subcutaneous vascular resistance was found in synchronism with the hyperemic blood flow phase (28). It has been shown that during the transitions in sleep stages, changes in peripheral vasomotor tone are influenced by both sympathetic outflow and changes in blood pressure (4). The purpose of the present study was to determine whether the nocturnal fluctuations in subcutaneous adipose tissue blood flow rate could be correlated to the different sleep stages. MATERIALS
AND METHODS
Ten lean normal human subjects were included in the study [age 28.2 t 3.5 (SD) yr]. Four were females (age 26.3 t 2.2 yr) and six were males (age 29.9 t 3.6 yr). Eight of these 10 subjects had participated in four previous studies of the nocturnal blood flow variations on the lower leg (28-31). None of the subjects had tachycardia or bradycardia. The study was approved by the local ethical committee, and informed consent was obtained from all participants before the investigation. Experimental Setup
IN RECENT STUDIES, measurements of subcutaneous blood flow rate over 12-20 h of ambulatory conditions were performed in the lower legs of normal human subjects (28-31). With the change from upright to supine position from day 1 at the beginning of the night period, an instantaneous blood flow rate increment of 3O-40% was observed in accordance with a decrease in central and local postural sympathetic vasoconstrictor activity (10, 17, 23). Approximately 90 min after the subjects went to bed, an additional blood flow rate increment of considerable magnitude was observed. The mean increase was 84%, but in several cases an increment of >200% was measured (30). The duration of this hyperemit phase was -100 min, after which the blood flow 1246
0161-7567/92 $2.00 Copyright
The 133Xewashout measurements. The blood flow rate measurements were performed on the medial aspect of the right lower leg, 10 cm proximal to the malleolar level. The depots of the tracer 133Xewere applied by the atraumatic epicutaneous labeling technique as described previously (28). Portable CdTe( Cl) semiconductor detectors were mounted over the central area of the depots. The detectors were fixed directly to the skin surface with a single layer of Mylar membrane interposed between the detector and the skin surface. Counts were accumulated at l-, 2-, or 4-min intervals and stored in a portable data storage unit (Memolog system 600, B. Simonsen Medical, Randers, Denmark). The measurements started 90 min after the labeling
0 1992 the American
Physiological
Society
Downloaded from www.physiology.org/journal/jappl at Karolinska Institutet University Library (130.237.122.245) on February 12, 2019.
NOCTURNAL
PERIPHERAL
BLOOD
FLOW
1247
REGULATION
procedure, between 7:00 and 9:00 P.M., and ended at approximately 9:00 A.M. the following day. During this period, which took place under ambulatory conditions, the subjects recorded their physical activities, time of meals, and sleeping periods. The subjects performed their normal activities during daytime (standing, walking, and sitting), and they went to bed at approximately 1 l:oo P.M., slept in supine position during the night, and got up at approximately 7:00 A.M. The subjects were not allowed to drink alcohol.
and only rate constants (h) from curve sections with a correlation coefficient (r) > 0.80 were taken into consideration. A few minor curve sections were rejected on the basis of this criterion; however, in no case were all data from a subject omitted. Irregularities on the curves were typical ly a res ult of either geometric distu .rbances between the detector and the isotope depot or temporary detector circuit noise. The curves were analyzed for goodness of fit and linearity by linear regression analysis, and the distributions Registration of onset of sleep and different sleep stages. of the residuals were inspected (28). According to the The sleep recordings were performed with a portable former, the majority of the curves could be divided into five phases as follows: phase 1 is day 1; phases 2,3, and 4 computerized sleep analysis system (Somnolog 3, Ventee, Hellerup, Denmark). The system measured the elec- are night; and phase 5 is day 2. Phase 2 was the initial troe lncephalogram (EEG), eye movements (with a p iezo- supine period in bed after standing. This period was electric transducer), and electromyogram (EMG) to without any central and local sympathetic or humoral the comply with the stan dard techniques (22). In addition, orthostatic vasoconstrictive activity. Therefore, the sound pressure level was measured to detect disturbblood flow level during this phase was chosen as reference level (index = l.OO), and the changes in blood flow ing noise, and hand movem .ents were measured to improve the arous al detection. From the raw data a primary rate durin .g the measurem .ent were calculated relative to feature extracti on was performed from the rele vant polyphase 2. graphic sleep parameters in 6-s epochs [for RechtschafDuring the night period, three distinct phases were defen and Kales (22) scoring]. These features were dis- tected, with significantly different isotope washout rates played and used in a visual scoring procedure with a (phases 2, 3, and 4) (28). The two break points of the 83.5% correlation to paper Rechtschaffen and Kales vi- three nocturnal phases were not associated with any sual scoring (33, 34). known events. The determination of these two break The EEG was recorded bipolar from F3 to A2 (the in- points was not solely based on visual inspection. First, ternational lo-20 system), the EEG amplifier frequency the logarithmically transformed washout curve of the response was 0.5-30 Hz (-3 dB) with a noise level
