Resting and recruitable endothelial function – Evidence of two distinct circadian patterns

Abstract

BACKGROUND:

Evidence of a circadian rhythm in endothelium-dependent vasomotor function, with a nadir in Flow-Mediated Dilation (FMD) in the early morning hours, has been previously reported. These changes have been proposed to be one of the mechanisms explaining the circadian pattern in the incidence of cardiovascular events. We set out to investigate the circadian rhythm of FMD, low-flow mediated dilation (L-FMC) and sympathetic vascular tone.

METHODS AND RESULTS:

10 young healthy male volunteers (mean age, 28.9±3.7 years) underwent measurements of radial artery endothelium-dependent FMD and L-FMC at 8AM, 2PM and 8PM on the same day. Sympathetic vascular tone was assessed with laser Doppler and Fourier transform analysis. Compared with 2PM and 8PM, FMD decreased markedly in the early morning (2.9±3.4%; 6.2±2.9%; 6.0±4.0%; P = 0.007). In contrast, L-FMC was maximal at 8AM, decreased significantly at 2PM, and returned to higher values at 8PM (–5.1±1.3%; –2.7±2.0%; –4.6±2.2%; P = 0.030), such that the composite endpoint of endothelial function (sum of FMD+L-FMC) was not significantly different among timepoints. Vascular sympathetic tone was maximal early in the morning and lowest in the evening (P = 0.014) without a correlation with the changes in FMD or L-FMC.

CONCLUSIONS:

Endothelial responsiveness (FMD) and basal tone (L-FMC) appear to follow different circadian rhythms, with an impaired responsiveness in the early morning and a nadir in baseline tone in the early afternoon.

Keywords

Endothelial function circadian rhythm vascular physiology

1 Introduction

The study of endothelial function provides important information on cardiovascular homeostasis with both prognostic and pathophysiological implications [1 –4]. For its simplicity and non-invasiveness, flow-mediated dilation (FMD) remains the most commonly used method for the assessment of endothelial function. FMD consists of the measurement of the changes in radial or brachial artery diameter in response to sudden increases in shear stress; as such, it is a marker of endothelial responsiveness, ie it quantifies the reactivity of this tissue to a specific stimulus for the production of endothelial mediators.

In analogy with several other homeostatic mechanisms such as blood pressure, heart rate and sympathetic tone, endothelial function displays a circadian rhythm. In the study by Otto et al., it was shown that FMD is lowest early in the morning, an observation which the authors proposed as possible co-causal mechanism for the higher incidence of coronary events and higher cardiovascular mortality observed during morning hours [5, 6]. A number of other studies have addressed the existence of this circadian variation in FMD with similar results [7 –9]. As such, given the prognostic importance of endothelial function [10], it has been hypothesized that an impairment in endothelial function may occur in the early morning hours, an observation that fits well with the increased incidence of coronary events at this time of the day [11].

The endothelium is however a complex tissue, whose status can be hardly described by one single parameter. One limitation to the concept of FMD, for instance, is that this method provides information on the capacity of the endothelium to cause vasodilation in response to a specific stimulus, but it does not take into account nor quantifies resting endothelial activity (i.e. the endothelial contribution to vascular tone in resting conditions). In the present study, we set out to investigate whether resting endothelial function, as measured by low-flow mediated vasoconstriction (L-FMC), like FMD, has a circadian rhythm, and the relationship between this rhythm and that of FMD and that of vascular sympathetic tone.

2 Methods

2.1 Study population

Ten healthy non-smoking male volunteers (age range 18–35 years) with no cardiac risk factors, blood pressure <130/70mmHg and normal blood counts, normal kidney and liver function and no history of disease were enrolled in this study. The protocol was approved by the local ethics committee, and is registered in www.clinicaltrials.gov (NCT01713374). All subjects gave informed consent.

2.2 Endothelial function assessments

Each subject returned to the laboratory for three visits (8AM, 2PM and 8PM) on the same day. Studies were conducted in a quiet, temperature- and humidity-controlled environment. Briefly, subjects were required to abstain from food for 6 hours before measurements and from performing sport, and drinking caffeine-containing beverages from the day before. Radial artery diameter, blood flow, L-FMC and FMD and their composite end point were measured using a Vivid 7 (General electrics, München, Germany) ultrasound machine with a 14 MHz matrix transducer and automatic analysis software as previously published [12, 13]. Briefly, the forearm was immobilized and radial artery diameter was measured at rest, during suprasystolic inflation of a pneumatic cuff placed at the wrist (4.5 min) and for the 5 min following deflation. The inflation and deflation of the cuff cause, respectively, a decrease and an increase in blood flow which can be measured proximal to the cuff. These changes modulate the endothelial production of vasoactive mediators resulting in, respectively, a constriction during the low-flow state (L-FMC, measured as the % constriction observed during cuff inflation) and a dilation during hyperemia (FMD, measured as % dilation following deflation of the cuff). The implications and mechanisms of these parameters have been recently reviewed [14 –16]. Finally, the time to peak vasodilation (Tmax) was calculated at the time between release of the wrist cuff and maximal dilation [17].

2.3 Vascular sympathetic function

Blood flow was measured at the level of the forearm using laser Doppler flowmetry (LDF) [18]. LDF measures skin blood flow by measuring the frequency shift in the backscattering of a laser beam which penetrates the skin and is partially reflected by moving blood cells [19]. The single point LDF apparatus used in the present study (Periflux PF4001, Perimed, Sweden) had the following settings: wavelength 780 nm, bandwidth 10 Hz-19 kHz, time constant 0.1 s, sampling frequency 32 Hz. LDF measurements reflect the blood flow in capillaries, arterioles, venules and dermal vascular plexus. The LDF signal was continuously recorded for 10 min prior to endothelial function assessment with Perisoft-dedicated software. Blood flow oscillations were recorded and examined off-line by means of spectral Fourier analysis. Based on previous studies, the frequency range 0.02–0.06 Hz identified sympathetic-dependent vasomotion [20]. Since measurements of blood flow oscillations are influenced by changes in blood flow, PSD values were normalized for the blood flow values measured during the same time-window.

2.4 Statistical analysis

All source data were coded after acquisition. To limit reader's bias, the analysis of endothelial function and laser-Doppler data was performed in a randomized fashion by staff blinded to the code. Data are presented as mean±standard deviation unless otherwise noted. Normality was assessed with the Kolmogorov-Smirnov test. A repeated measured ANOVA was used for comparison between different visits with Bonferroni/Dunn correction for post-hoc comparisons. The sample size was based on our previous observation of a mean FMD of 5.0±2.1% [12]. Assuming a variance explained by effect of 0.7%, a power of 90% and an alpha of 5%, with three repeated measures, 8 subjects would be necessary. All statistical analyses were performed with Statview (SAS). The primary endpoint was the change in FMD and L-FMC among visits, and a P < 0.05 was considered statistically significant.

3 Results

The mean age of the participants was 28.9±3.7 years. The principal blood values are presented in Table 1.

Table 1
Clinical laboratory characteristics of the subjects, with reference ranges

Reference range Mean SD

Creatinine, mg/dl 0.7–1.3 0.9 0.1

Sodium, mmol/L 136–145 141 2

Potassium, mmol/L 3.5–5.1 3.6 0.4

Calcium, mmol/L 2.18–2.5 2.3 0.1

Uric acid, mg/dl 3.4–7.0 6.0 1.3

Azotemia, mg/dl 8–26 14.0 2.5

Glucose, mg/dl 70–100 77.9 18

C-reactive protein, mg/l <5 1.8 1.4

International normalized ratio 1.0 0.1

gamma-Glutamyl transferase, U/l 12–46 21 10

Aspartate aminotransferase, U/l 5–35 30 9

Alanine aminotransferase, U/l 5–35 27 7

Total Cholesterol, mg/dl <200 190 30

HDL-Cholesterol, mg/dl >40 50 8

LDL-Cholesterol, mg/dl <160 122 28

LDL/HDL-Ratio 2.5 0.7

Triglycerides, mg/dl <150 94 42

Hematocrit, % 39–49 45 3

Erythrocytes, n/pl 4.2–5.6 5.1 0.4

Haemoglobin, g/dl 13.5–17.5 15.1 1.2

Leukocytes, n/nl 3.5–10 5.7 1.1

Thrombocytes, n/nl 150–360 206 38

	Reference range	Mean	SD
Creatinine, mg/dl	0.7–1.3	0.9	0.1
Sodium, mmol/L	136–145	141	2
Potassium, mmol/L	3.5–5.1	3.6	0.4
Calcium, mmol/L	2.18–2.5	2.3	0.1
Uric acid, mg/dl	3.4–7.0	6.0	1.3
Azotemia, mg/dl	8–26	14.0	2.5
Glucose, mg/dl	70–100	77.9	18
C-reactive protein, mg/l	<5	1.8	1.4
International normalized ratio	1.0	0.1
gamma-Glutamyl transferase, U/l	12–46	21	10
Aspartate aminotransferase, U/l	5–35	30	9
Alanine aminotransferase, U/l	5–35	27	7
Total Cholesterol, mg/dl	<200	190	30
HDL-Cholesterol, mg/dl	>40	50	8
LDL-Cholesterol, mg/dl	<160	122	28
LDL/HDL-Ratio	2.5	0.7
Triglycerides, mg/dl	<150	94	42
Hematocrit, %	39–49	45	3
Erythrocytes, n/pl	4.2–5.6	5.1	0.4
Haemoglobin, g/dl	13.5–17.5	15.1	1.2
Leukocytes, n/nl	3.5–10	5.7	1.1
Thrombocytes, n/nl	150–360	206	38

All values, for each individual subject, were within the limits of normality. HDL: high-density lipoprotein; LDL: low-density lipoprotein.

3.1 Endothelial function

Radial artery diameter, blood flow and endothelial function data are presented in Table 2 and Figs. 1 and 2.

Fig.1

Changes in FMD and L-FMC at the three timepoints: FMD showed a nadir at 8AM, L-FMD at 2PM. ^*: P < 0.05 compared to the other two timepoints.

Fig.2

Changes in vascular endothelial, myogenic and sympathetic tone.

Table 2

Endothelial function parameters

Parameter	8AM	2PM	8PM	P (RM ANOVA)
Baseline diameter, mm	2.71±0.40	2.72±0.54	2.91±0.78	0.632
Delta during cuff inflation, mm	0.14±0.04	0.07±0.05^*	0.13±0.05	0.014
Delta after cuff deflation, mm	0.08±0.10^**	0.17±0.08	0.17±0.11	0.035
L-FMC, %	–5.1±1.3	–2.7±2.0^*	–4.6±2.2	0.030
FMD, %	2.9±3.4^**	6.2±2.9	6.0±4.0	0.007
Composite endpoint, %	8.05±4.1	8.92±3.5	10.59±5.3	0.230
Time to Peak FMD, sec.	51±31	103±65	78±24	0.057
% Change in blood flow during cuff occlusion	21±9	27±18	19±9	0.070
% Hyperemia	401±256	466±218	411±326	0.933

Resting radial artery diameter was not different among visits (P = 0.632). The change in diameter after cuff deflation was lowest at 8AM (P = 0.035, P < 0.01 for both post-hoc comparisons). As a result of these differences, and reproducing previous results, FMD was significantly lower at 8AM as compared to 2PM and 8PM (P = 0.007, Fig. 1). In contrast, the change in diameter observed during cuff occlusion was significantly smaller at 2PM as compared to the other two timepoints (P = 0.014, P < 0.05 for both post-hoc analyses). In line with this, L-FMC also showed a circadian rhythm, with the lowest values at 2 PM and similar, higher, values at 8AM and 8PM (P = 0.030, Fig. 1). There was no significant difference in the composite endpoint of endothelial function among timepoints (P = 0.230). Time to peak was also lowest at 8PM, although this difference formally did not reach statistical significance (P = 0.057). Blood flow responses to cuff inflation and deflation (Table 2) were not different among timepoints, however a non-significant trend towards a smaller change in blood flow at 8PM was seen.

3.2 Laser Doppler

Laser Doppler data are presented in Table 3 and Fig. 2. There was no difference between timepoints in mean perfusion units or total spectral power. The Fourier analysis of laser Doppler signals of the subcutaneous blood flow showed significant differences for all three components, with consistently lower values at 8PM. By virtue of correction for repeated testing, a decrease in the endothelial component was evident at 8PM compared to 8AM (P = 0.011). Further, a decreased sympathetic activity was seen at 8PM as compared to both 8AM and 2PM (both P < 0.0125, Bonferroni-Dunn correction). There was no difference among timepoints in the other components of the laser Doppler spectrum.

Table 3
Sympathetic activity

Parameter 8AM 2PM 8PM P (RM ANOVA)

Perfusion units 55±64 46±27 63±23 0.716

Total spectral power, units 56±74 39±18 33±10 0.510

Normalized sympathetic component 0.53±0.17 0.52±0.21 0.27±0.14^† 0.005

Normalized endothelial component 0.22±0.11 0.18±0.08 0.12±0.05^ 0.035

Normalized myogenic component 0.35±0.20 0.34±0.22 0.18±0.04 0.029

Parameter	8AM	2PM	8PM	P (RM ANOVA)
Perfusion units	55±64	46±27	63±23	0.716
Total spectral power, units	56±74	39±18	33±10	0.510
Normalized sympathetic component	0.53±0.17	0.52±0.21	0.27±0.14^*†	0.005
Normalized endothelial component	0.22±0.11	0.18±0.08	0.12±0.05^*	0.035
Normalized myogenic component	0.35±0.20	0.34±0.22	0.18±0.04	0.029

Post-hoc tests (Bonferroni/Dunn): ^*P < 0.0125 for the comparison with 8AM; ^†P < 0.0125 for the comparison with 2PM.

4 Discussion

For its importance in the regulation of cardiovascular homeostasis, parameters of systemic vascular endothelial function have often been used as a surrogate marker of coronary atherosclerosis and in the stratification of cardiovascular risk [10]. Conduit artery endothelial function follows a circadian rhythm, with an attenuation of FMD in the morning to levels that, in healthy subjects, are very similar to those recorded in patients with cardiovascular disease [5 , 22]. Beyond their importance for the design of studies, these findings have been suggested to provide a background for the increased incidence of cardiac and vascular events observed in the early morning hours [6].

Complicating the interpretation of FMD, however, while this method quantifies endothelial responses to a specific stimulus, it does not provide information on resting endothelial activity (ie, the endothelial production of vasomotor mediators in resting conditions). The relationship between resting function and endothelial reactivity is complex, and it has been previously demonstrated that measures of resting endothelial tone and endothelial responses are, to a certain extent, inversely proportional. For instance, during physical activity or immediately following administration of a systemic inflammatory stimulus, the endothelium is activated, resulting in increased L-FMC and decreased FMD [16 , 24]. In order to assess the true meaning of a decreased FMD, it would be therefore advisable to also provide an index of resting endothelial tone.

In the present paper, we report on the circadian pattern of L-FMC. Like FMD, we show that L-FMC also has a circadian rhythm, which, in contrast to that of FMD, has a bimodal peak in the early morning and evening hours, and a nadir in the early afternoon. This L-FMC peak compensates for the nadir in FMD in the morning hours, such that “overall” endothelial function remains relatively constant during the day. Although we did not address the mechanisms of the variations in FMD and L-FMC, the changes observed are in line with previous reports in the circadian rhythms of ET-1 (which is involved in determining L-FMC [15]) and nitric oxide (the principal mediator of FMD [1]). In the paper by Elherik et al. [25], indeed, ET-1 showed a bimodal pattern, with a peak early in the morning and late in the evening, compatible with our nadir in L-FMC at 2PM. In contrast, nitric oxide levels appear to show a nadir early in the morning, which is also compatible with our reported blunting in FMD. Further, we provide direct evidence of a circadian rhythm in the sympathetic regulation of vascular tone, which is also in line with the patterns of L-FMC and FMD. While sympathetic withdrawal in the evening hours is compatible with the peak of FMD [26], an increased bioavailability of vasoconstrictors in the morning hours (and/or an increased endothelial stimulation via beta2-adrenergic receptors) is compatible with the morning peak of L-FMC. Finally, we show a pattern of the endothelial and myogenic components of cutaneous tone, which were both blunted at 8PM in repeated measures ANOVA. The mechanisms of these observations, and the discrepancy between conduit (showing a peak at 8PM) and cutaneous endothelial pattern (with a nadir at the same timepoint) remain to be explained.

In the present pilot study, only male healthy young volunteers were enrolled in order to limit the influence of possible confounders. The role of hormones, age-related changes, and disease is acknowledged [27, 28]. A fast Fourier method was used for the analysis of the components of cutaneous flow. Although this method is commonly used, it is less sensitive than other ones, e.g. the Wavelet analysis [29]. Finally, an important limitation is that endothelium-independent vasodilation was not tested. Previous papers have however addressed the relationship between blunting in FMD and ischemic episodes in patients with vasospastic angina [9], and the absence of a circadian rhythm in endothelium-independent vasodilation has also been already shown [5].

5 Conclusions

“Endothelial function” is a complex homeostatic phenomenon with important clinical implications. We show that both FMD and L-FMC, i.e. endothelial responsiveness and resting endothelial tone, undergo similar but asynchronous circadian patterns.

Author contributions

Conceptualization, MV and TG; methodology, TG and MV; formal analysis, MES and TG; investigation, all authors; resources, TG; data curation, TG MV and MES; writing— original draft preparation, TG; writing— review and editing, all authors; project administration, MV; funding acquisition, TG.

Conflicts of interest

The authors declare no conflict of interest.

Funding

This research received no external funding.

Footnotes

Acknowledgments

TG is DZHK professor.

References

Corretti

, Anderson

, Benjamin

, Celermajer

, Charbonneau

, Creager

, et al. Guidelines for the ultrasound assessment of endothelial-dependent flow-mediated vasodilation of the brachial artery: A report of the International Brachial Artery Reactivity Task Force. Journal of the American College of Cardiology. 2002;39(2):257–65.

Ghiadoni

, Versari

, Giannarelli

, Faita

, Taddei

. Non-invasive diagnostic tools for investigating endothelial dysfunction. Current Pharmaceutical Design. 2008;14(35):3715–22.

Vita

, Treasure

, Ganz

, Cox

, Fish

, Selwyn

. Control of shear stress in the epicardial coronary arteries of humans: Impairment by atherosclerosis. Journal of the American College of Cardiology. 1989;14(5):1193–9.

Silva

, Teixeira

, Oliveira

, Almeida

, Vasconcelos

. Endothelial dysfunction, microvascular damage and ischemic peripheral vasculopathy in systemic sclerosis. Clin Hemorheol Microcirc. 2017;66(2):117–30.

Otto

, Svatikova

, Barretto

, Santos

, Hoffmann

, Khandheria

, et al. Early morning attenuation of endothelial function in healthy humans. Circulation. 2004;109(21):2507–10.

Ridker

, Manson

, Buring

, Muller

, Hennekens

. Circadian variation of acute myocardial infarction and the effect of low-dose aspirin in a randomized trial of physicians. Circulation. 1990;82(3):897–902.

Bau

, Bau

, Naujorks

, Rosito

, Fuchs

. Diurnal variation of vascular diameter and reactivity in healthy young men. Braz J Med Biol Res. 2008;41(6):500–3.

Etsuda

, Takase

, Uehata

, Kusano

, Hamabe

, Kuhara

, et al.

Morning attenuation of endothelium-dependent, flow-mediated dilation in healthy young men: Possible connection to morning peak of cardiac events?

Clin Cardiol. 1999;22(6):417–21.

Kawano

, Motoyama

, Yasue

, Hirai

, Waly

, Kugiyama

, et al. Endothelial function fluctuates with diurnal variation in the frequency of ischemic episodes in patients with variant angina. Journal of the American College of Cardiology. 2002;40(2):266–70.

10.

Munzel

, Sinning

, Post

, Warnholtz

, Schulz

. Pathophysiology, diagnosis and prognostic implications of endothelial dysfunction. Ann Med. 2008;40(3):180–96.

11.

Nagarajan

, Fonarow

, Ju

, Pencina

, Laskey

, Maddox

, et al. Seasonal and circadian variations of acute myocardial infarction: Findings from the Get With The Guidelines-Coronary Artery Disease (GWTG-CAD) program. Am Heart J. 2017;189:85–93.

12.

Gori

, Dragoni

, Lisi

, Di Stolfo

, Sonnati

, Fineschi

, et al. Conduit artery constriction mediated by low flow a novel noninvasive method for the assessment of vascular function. Journal of the American College of Cardiology. 2008;51(20):1953–8.

13.

Bartoli

, Menegaz

, Lisi

, Di Stolfo

, Dragoni

, Gori

. Model-based analysis of flow-mediated dilation and intima-media thickness. Int J Biomed Imaging. 2008;2008:738545.

14.

Ghiadoni

, Salvetti

, Muiesan

, Taddei

. Evaluation of endothelial function by flow mediated dilation: Methodological issues and clinical importance. High Blood Press Cardiovasc Prev. 2014.

15.

Gori

, Parker

, Munzel

. Flow-mediated constriction: Further insight into a new measure of vascular function. European Heart Journal. 2011;32(7):784–7.

16.

Humphreys

, Green

, Cable

, Thijssen

, Dawson

. Low-flow mediated constriction: The yin to FMD’s yang? Expert Rev Cardiovasc Ther. 2014;12(5):557–64.

17.

Liuni

, Luca

, Lisi

, Dragoni

, di Stolfo

, Mariani

, et al. Observations of time-based measures of flow-mediated dilation of forearm conduit arteries: Implications for the accurate assessment of endothelial function. Am J Physiol Heart Circ Physiol. 2010;299(3):H939–45.

18.

Rossi

, Carpi

, Di Maria

, Galetta

, Santoro

. Spectral analysis of laser Doppler skin blood flow oscillations in human essential arterial hypertension. Microvasc Res. 2006;72(1-2):34–41.

19.

Nilsson

, Tenland

, Oberg

. Evaluation of a laser Doppler flowmeter for measurement of tissue blood flow. IEEE Trans Biomed Eng. 1980;27(10):597–604.

20.

Landsverk

, Kvandal

, Kjelstrup

, Benko

, Bernjak

, Stefanovska

, et al. Human skin microcirculation after brachial plexus block evaluated by wavelet transform of the laser Doppler flowmetry signal. Anesthesiology. 2006;105(3):478–84.

21.

Khan

, Ahmad

. Circadian variation–increased morning incidence of acute myocardial infarction in patients with coronary artery disease. JPMA The Journal of the Pakistan Medical Association. 2003;53(2):84–7.

22.

Gaenzer

, Sturm

, Kirchmair

, Neumayr

, Ritsch

, Patsch

. Circadian variation of endothelium-dependent vasodilatation of the brachial artery as a confounding factor in the evaluation of endothelial function. Atherosclerosis. 2000;149(1):227–8.

23.

Gori

, Grotti

, Dragoni

, Lisi

, Di Stolfo

, Sonnati

, et al. Assessment of vascular function: Flow-mediated constriction complements the information of flow-mediated dilatation. Heart. 2010;96(2):141–7.

24.

Spiro

, Digby

, Ghimire

, Mason

, Mitchell

, Ilsley

, et al. Brachial artery low-flow-mediated constriction is increased early after coronary intervention and reduces during recovery after acute coronary syndrome: Characterization of a recently described index of vascular function. European Heart Journal. 2011;32(7):856–66.

25.

Elherik

, Khan

, McLaren

, Kennedy

, Belch

. Circadian variation in vascular tone and endothelial cell function in normal males. Clin Sci (Lond). 2002;102(5):547–52.

26.

Panza

, Epstein

, Quyyumi

. Circadian variation in vascular tone and its relation to alpha-sympathetic vasoconstrictor activity. N Engl J Med. 1991;325(14):986–90.

27.

Shiogai

, Stefanovska

, McClintock

. Nonlinear dynamics of cardiovascular ageing. Phys Rep. 2010;488(2-3):51–110.

28.

Stupin

, Stupin

, Baric

, Matic

, Kolar

, Drenjancevic

. Sex-related differences in forearm skin microvascular reactivity of young healthy subjects. Clin Hemorheol Microcirc. 2019.

29.

Martini

, Bagno

. The wavelet analysis for the assessment of microvascular function with the laser Doppler fluxmetry over the last 20 years. Looking for hidden informations. Clin Hemorheol Microcirc. 2018;70(2):213–29.