Blood pressure variability, nocturnal heart rate variability and endothelial function predict recurrent cerebro-cardiovascular events following ischemic stroke

Introduction Cardiovascular parameters characterizing blood pressure (BP), heart rate (HR), endothelial function and arterial stiffness predict cerebro-cardiovascular events (CCVE) in the general population. Considering the paucity of data in stroke patients, we assessed these parameters as potential predictors of recurrent CCVE at acute stroke stroke. Patients and methods This is a secondary outcome analysis of a prospective observational longitudinal Sleep Deficiency & Stroke Outcome Study (ClinicalTrials.gov Identifier: NCT02559739). The study consecutively recruited acute ischemic stroke patients. Cardiovascular parameters (blood pressure variability [BPV], heart rate variability [HRV], endothelial function, and arterial stiffness) were assessed within the first week post-stroke. Future CCVE were recorded over a 3-year follow-up. Multivariate Cox regression analysis was used to investigate the prognostic value of 48 cardiovascular parameters regarding CCVE risk. Results Out of 447 recruited patients, 359 were included in this analysis. 20% of patients developed a future CCVE. A high variability of systolic BP (n = 333) and nocturnal HR (non-linear parameters; n = 187) at acute stroke predicted CCVE risk after adjustment for demographic parameters, cardiovascular risk factors and mean BP or HR, respectively. Endothelial dysfunction (n = 105) at acute stroke predicted CCVE risk after adjustment for age and sex, but not after adjustment for cardiovascular risk factors. Diurnal HR and arterial stiffness at acute stroke were not associated with CCVE risk. Conclusion High blood pressure variability, high nocturnal HRV and endothelial function contribute to the risk for future CCVE after stroke.

consecutive measurements in the sitting position (back and arm supported) (1) at 2-minute intervals 3 times per day: morning, day and evening; after 3-5 minutes of rest in a quiet room.
Only the patients with at least 12 available BP measurements were included in the analysis.

EndoPAT: diurnal heart rate, endothelial function and arterial stiffness
Approximately 20% of the patients were randomly selected for additional assessments of diurnal HRV, endothelial function and arterial stiffness with digital plethysmography (EndoPAT2000-device; Itamar Medical Ltd., Caesarea, Israel).The assessment was performed within the first week after hospital admission according to the instructions from the manufacturer.
The participants were instructed to fast and to refrain from smoking, alcohol-and xanthinecontaining beverages for 12 hours prior to the assessment of cardiovascular parameters.The measurement of cardiovascular characteristics was performed in the morning (between 08:00 and 11:00).The participants were in supine position for a minimum of 20 minutes before measurements in a quiet room (temperature 21-24°C) and stayed motionless during the entire measurement period.

Supplementary material 2.
The assessment of cerebro-cardiovascular events.
The main outcome of the study was a composite of fatal and non-fatal recurrent CCVE that included ischemic or hemorrhagic stroke, TIA, myocardial infarction, unplanned  (12).Night-time HR seems to have better predictive ability regarding CCVE risk compared to daytime and 24-h HR (13).
HRV HRV reflects the fluctuations in HR during the investigated period of time and can be characterized by 3 groups of parameters as specified below (14).
Time domain of HRV These parameters (for example, SDNN, SDSD, RMSSD, pNN50, TI, TINN) quantify the amount of variability in measurements of the interbeat interval, which is the time period between successive heartbeats (14).High values of these parameters indicate high HRV.
These parameters could reflect both sympathetic and parasympathetic autonomic functioning (14).Low diurnal HRV with high levels of sympathetic components and lower parasympathetic as well as unpredictable nocturnal HRV was described to be associated with CCVE risk and CCVE risk factors (15)(16)(17)(18).
Frequency domain of HRV These parameters (for example, nuLF, nuHF, LF, HF, VLF, LF/HF) quantify signal energy within a specific frequency band (14).The physiological regulation of these parameters is complex and involves circadian, endocrine and autonomic components (14,19).
There is no consensus regarding the mechanisms that define these signals, however, current literature overview points to the association of LF power with sympathetic activation, HF power with parasympathetic activation and VLF power with circadian rhythm and sympathetic activation (14,20,21).In a simplified way, LF/HF ratio is supposed to reflect sympatho-vagal balance, however, considering the non-linear relationship between sympathetic and parasympathetic regulation, this paradigm is largely debated (22).
Non-linear HRV measurements These parameters (for example, DFAa1, DFAa2, ApEn, SD1, SD2, SD1/SD2) quantify the unpredictability of a time series: high values of these parameters indicate high HRV (14,23).Little is known about the physiological regulation of these parameters.According to few available studies, higher values of DFA, ApEn and SD2 reflect sympathetic modulation, while SD1 could reflect parasympathetic modulation (7,24,25).Abbreviations: ApEn -approximate entropy, AI -augmentation index, AI75 -augmentation index normalized for 75 bpm, ARV -average real variability, bpm -beats per minute, CV -coefficient of variation, DBP -diastolic blood pressure, DFA1 -detrended fluctuation analysis component α1, DFA2 -detrended fluctuation analysis component α1, FRHI -Framingham Reactive Hyperemia Index, HFnu -normalized high frequency power, HFhigh frequency power, HR -heart rate, HRV -heart rate variability, LFnu -normalized low frequency power, LFlow frequency power, MAD -mean absolute deviation, pNN50 -percentage of adjacent NN intervals that differ from each other by more than 50 ms, RHI -reactive hyperemia index, RMSSD -root mean square of successive differences of successive RR intervals, SBP -systolic blood pressure, SD -standard deviation, SD1 -SD1 from Poincaré plot, SD2 -SD2 from Poincaré plot, SDNN -standard deviation of normal-to-normal beats, SDSDstandard deviation of successive RR interval differences, SV -successive variation, TI -triangular index, TINNtriangular interpolation of NN interval histogram, VLF -very low frequency power.Supplementary table 4. Significant (p<0.05)associations between cardiovascular parameters at acute stroke with demographic characteristics, stroke characteristics and comorbidities according to correlation analysis followed by multiple linear regression analysis (both approaches lead to significance).Multiple linear regression was adjusted for age and sex.
the magnitude of limb reperfusion following a brief period of ischemia induced by arterial occlusion (the ratio of hyperemic pressure to baseline pressure in the measurement and control arms)(26).Compared to RHI, FRHI, uses the natural logarithmic transformation of the RHI ratio, does not include the baseline correction factor and utilizes only the readings from 90 to 120 seconds for post occlusion(27

table 2 .
Characteristics of participants with specific cardiovascular assessments at acute stroke (overall n=359).There were no significant differences between the populations with specific cardiovascular assessments.Continuous data is presented as median [interquartile range].Categorical data is presented as count (% of total).Kruskall-Wallis test was used for continuous variables, Chi-square test was used for categorical variables. 1BMI≥30 kg/m 2 . 2 SDB is defined as AHI≥20/h.Abbreviations: AHI -apnea-hypopnea index, BDI -Beck Depression Inventory, BMI -body mass index, CADcoronary artery disease, CCVE -cerebro-cardiovascular event, CE -cardioembolism, DBP -diastolic blood pressure, HRV -heart rate variability, LAAS -large-artery atherosclerosis, mRS -modified Rankin Scale, NIHSS -National Institute of Health Stroke Scale, PAD -peripheral artery disease, SDB -sleep-disordered breathing, SVO -small vessel occlusion, TIA -transient ischemic attack.

table 3 .
Description of cardiovascular characteristics at acute stroke.