The aim of our study was to assess the effect of short-term cold exposure, typical in subarctic climate, on cardiac electrical function among untreated middle-aged hypertensive men.
We conducted a population-based recruitment of 51 hypertensive men and a control group of 32 men without hypertension (age 55–65 years) who underwent whole-body cold exposure (15 min exposure to temperature −10°C, wind 3 m/s, winter clothes). Conduction times and amplitudes, vectorcardiography, arrhythmias, and heart rate variability (autonomic nervous function) were assessed.
Short-term cold exposure increased T-peak to T-end interval from 67 to 72 ms (p<0.001) and 71 to 75 ms (p<0.001) and T-wave amplitude from 0.12 to 0.14 mV (p<0.001) and from 0.17 to 0.21 mV (p<0.001), while QTc interval was shortened from 408 to 398 ms (p<0.001) and from 410 to 401 ms (p<0.001) among hypertensive men and controls, respectively. Cold exposure increased both low (from 390 to 630 ms2 (p<0.001) and 380 to 700 ms2 (p<0.001), respectively) and high frequency heart rate variability (from 90 to 190 ms2 (p<0.001) and 150 to 300 ms2 (p<0.001), respectively), while low-to-high frequency-ratio was reduced. In addition, the frequency of ventricular ectopic beats increased slightly during cold exposure. The cold induced changes were similar between untreated hypertensive men and controls.
Short-term cold exposure with moderate facial and mild whole body cooling resulted in prolongation of T-peak to T-end interval and higher T-wave amplitude while QTc interval was shortened. These changes of ventricular repolarization may have resulted from altered cardiac autonomic regulation and were unaffected by untreated hypertension.
Citation: Hintsala H, Kenttä TV, Tulppo M, Kiviniemi A, Huikuri HV, Mäntysaari M, et al. (2014) Cardiac Repolarization and Autonomic Regulation during Short-Term Cold Exposure in Hypertensive Men: An Experimental Study. PLoS ONE 9(7): e99973. https://doi.org/10.1371/journal.pone.0099973
Editor: Larisa G. Tereshchenko, Johns Hopkins University SOM, United States of America
Received: December 11, 2013; Accepted: May 19, 2014; Published: July 1, 2014
Copyright: © 2014 Hintsala et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: The study was funded through grants from the National Institute for Health and Welfare as well as Paavo Nurmi, Emil Aaltonen and Ida Montin Foundations. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Wintertime is associated with increased morbidity and mortality and a majority of this is related to cardiovascular causes , , such as myocardial infarctions , ruptures/dissection of aortic aneurysms , heart failures , as well as strokes . Also ventricular arrhythmias  and sudden cardiac death  exhibit seasonal changes among cardiac patients, with higher occurrence during cold season.
Whole body exposure to cold and the resulting superficial cooling activates the sympathetic nervous system  that increases heart rate (HR) and constricts peripheral vasculature reducing heat loss from human to the environment. Sympathetic activation in general may produce altered cardiac function involving a higher risk of arrhythmias and cardiac events . Facial cooling, on the other hand, is known to increase the vagal tone through the trigeminal nerve stimulation with a decrease in HR . This may serve as a protective cardiovascular effect. However, for instance during whole body cold water immersion  or facial cooling  a co-activation of sympathetic and parasympathetic nervous system is observed. This causes conflicting inotropic and chronotropic drives to the heart (autonomic conflict) which may have an additional arrhythmogenic effect over the sympathetic activation alone , .
At present, the effects of short-term cold exposure with only superficial cooling on cardiac electrical function are not well known. Previous studies using pronounced whole-body cooling and hypothermia have detected electrocardiographic (ECG) manifestations such as the J (Osborne) waves, interval prolongation, T-wave abnormalities, and atrial and ventricular arrhythmias , , . Furthermore, exercise in a cold environment induces pronounced cardiac symptoms in patients with cardiac disease ,  and impairs post exercise cardiovagal regulation among healthy subjects .
Cold-induced arrhythmias might be pronounced in patients with hypertension in which sympathetic nervous system is already overactive . Hypertension itself is a risk factor for arrhythmias, such as atrial fibrillation . In addition, both short ,  and long term  exposure to cold elevate blood pressure (BP) mainly as a result of the increased sympathetic activity. Hence, this augmented sympathetic drive could further aggravate the course of hypertension and increase the risk for arrhythmias and associated adverse health events. However, no previous studies explored the effect of cold exposure to cardiac electrical function among hypertensive subjects.
The objective of the present study was to assess the effects of short-term cold exposure similar to everyday winter circumstances on cardiac electrical function, incident arrhythmias, and the autonomic nervous system among subjects with untreated hypertension. We hypothesized that, compared to a warm environment, a cold exposure would increase sympathetic drive and induce electrophysiological changes known to be associated with arrhythmias. We also expected these changes to be pronounced among subjects with hypertension. The primary outcome measure of the study was ECG and vectorcardiographic features and secondary measure arrhythmias. The information of how cold exposure affects cardiac electrical function is relevant for medical professionals involved with treating cardiovascular diseases.
Materials and Methods
We conducted a population-based recruitment (Figure 1) from April to November in 2011 which has been previously described in detail . Briefly, men aged 55–65 years were randomly chosen from the population register in Oulu, Finland (65°N, 25°E), interviewed for eligibility and performed home BP measurements according to the recommendations of the European Society of Hypertension . According to the sample size estimation and power analysis (G-Power 3.1.0) we estimated to detect statistically significant differences in brachial BP between a warm and cold environment [Power (1-β err prob), 0.9, Cohen's effect size 0.8, α err prob 0.05] in 23 participants. We aimed at a 2∶1 ratio of hypertensive men and control subjects for possible subgroup analyses among hypertensive, i.e. 46 hypertensive and 23 men in control group in final data-analyses. Based on the home BP measurements 83 subjects were enrolled into the study: 51 hypertensive men (the mean of 28 systolic BP measurements ≥135 and/or the mean of 28 diastolic measurements BP ≥85 mmHg) and 32 controls (the mean of 28 systolic/diastolic BP measurements <135/85 mmHg) (Table 1). The exclusion criteria of the study were presence of coronary heart disease or respiratory disease, use of antihypertensive drugs for any indication, an average home BP ≥175/105 mm Hg, no home BP measurements, inadequate data quality in the laboratory measurements (n = 1), and having a respiratory infection a week before exposure measurements. The other clinical characteristics did not differ among the groups except of 4 subjects had type 2 diabetes in the hypertensive group, and the hypertensive men had higher weight, body-mass index (BMI), and fat percentage. The study was approved by the ethics committee of Oulu University Hospital and all participants of the study gave written informed consent. The protocol for this trial and supporting Trend checklist are available as supporting information; see Checklist S1, Protocol S1, and Protocol S2. The trial was not registered prospectively because this observational controlled study does not any involve any clinical intervention. The authors confirm that all ongoing and related trials for this intervention are registered.
BP, blood pressure. a Four subjects excluded for the usage of BP medication and one for inadequate data quality.b One subject excluded for the usage of BP medication and two did not perform cold exposure for safety reasons.
The experimental measurements were performed from August to November 2011 before the start of the cold season by trained professionals. The experiments began with a short introduction of the measurement protocol and visit to the climatic chamber. Height was measured. Body composition was assessed by bioelectrical impedance analysis (InBody 720 Biospace Ltd, Korea) and physical fitness was measured while resting with the Polar Own Index Fit Test (Polar S610, Polar, Finland). After the measurements the study subjects were equipped with thermistors (skin temperature), ECG electrodes, and a cuff for measuring brachial BP and dressed with winter clothing (ca. 2 clo during cold exposure and 1.6 clo during baseline and recovery measurements ). The exposure protocol consisted of three consecutive 15 minutes phases: baseline, cold exposure, and recovery period in baseline conditions. The subjects were exposed to cold in a wind tunnel (air temperature Ta = −10°C, air velocity Va = 3 m/s, and relative air humidity Ha = 50%) adjacent to a climatic chamber in which baseline and recovery measurements were performed (Ta = 18°C, Va<0.2 m/s, Ha = 30%). All measurements were performed during the office hours.
Digital ECG Recordings
Derived 12-lead ECG recordings (Medilog AR12, Huntleigh Healthcare, Austria) were performed in order to measure the electrical function of the heart throughout the experiment. Recordings were made using 16 bit amplitude resolution and 256 Hz sampling rate.
ECG and Vectorcardiographic Features
The signal processing and analysis were carried out with custom-made software written in Matlab (MathWorks, inc., Natick, MA, USA). Baseline wander was removed with cubic spline interpolation and high-frequency noise was suppressed with a bidirectional 40 Hz low-pass filter. Ectopic and abnormally shaped beats were removed from the analysis. Beats were considered ectopic if the preceding RR interval differed by more than 20% from the last valid RR interval. Beats with deviating morphology were identified with template matching. The template was updated with each accepted beat and the threshold for acceptance was 95% match. Subsequently, the signal was resampled from 256 Hz to 512 Hz in order to improve the morphological alignment of the beats. All beats were aligned in each of the leads based on cross-correlation (time alignment between the leads was not changed). The aligned beats were then filtered with a ten-beat wide sliding window in order to produce representative median beats for each cardiac cycle and to suppress the noise within the beats.
Three representative median beats were taken five minutes apart from each other, starting from the beginning of each phase, i.e. baseline, cold and recovery phases. In order to assure the accuracy of the interval measurements, these samples were analyzed manually by a trained operator, who identified the P-wave onset, QRS boundaries, T-wave peak and T-wave offset from each of the 12 leads for each sample. Subsequently, PR, QRS, QT and T-peak to T-end intervals were calculated in leads II and V5. QT interval was corrected for HR with nomogram method (QTc) . Left ventricular hypertrophy (LVH) was evaluated by Sokolow-Lyon voltage criteria. LVH was defined as a sum of S amplitude in lead V1 and R amplitude in lead V5 or V6 (lead with higher amplitude was chosen) ≥3.5 mV.
Mean spatial QRS-T angle and maximum T-wave amplitude were calculated in a beat-to-beat manner. For this purpose, the QRS boundaries, T-wave peak and T-wave offset were located automatically for each beat. The QRS onset and offset detection was based on the method described by Daskalov and Cristov , utilizing the morphology of the ECG in the determination of the QRS boundaries. T-wave offset was defined as the intersection of the isoelectric line and a line fitted by least squares to the maximum slope of the T-wave. Isoelectric line was defined as the mean amplitude of the preceding and following TP intervals. The maximum amplitude between QRS offset and T-wave offset was taken as the T-wave peak (minimum T-wave amplitude if inverted T-wave was observed).
Mean QRS and T axes were calculated from the orthogonal X, Y and Z leads, which were reconstructed from the 12-lead ECG samples with inverse-Dower matrix . The mean spatial axes were based on the position vectors defined by the areas of the QRS and T-wave loops in the X, Y and Z leads and the mean spatial QRS-T angle was defined as the angle between these axes. Additionally, the maximum T-wave amplitude was calculated from leads V5 and II. However, instead of lead V5, lead V4 or V6 (which ever had the highest amplitude of T-wave) was used if the amplitude in V5 was below 0.15 mV.
We also analyzed the frequency of single ventricular and atrial ectopic beats. Automatic detection of ectopic beats was performed from baseline, cold exposure, and recovery ECG recordings with Medilog Darwin Holter Analyses software version V1.13.4 (TOM Medical Handels Gmbh, Austria). Automatic detection of ectopic beats was based on Medilog ADAPT algorithm . Prematurity threshold was set to 15% compared to three normal preceding beats. Detected ectopic beats were confirmed with visual inspection by trained health care professionals.
Heart Rate Variability
Heart rate variability (HRV) analyses were performed in time and frequency domain as described previously  (Hearts 8, Heart Signal Co., Oulu, Finland) to assess activity of the autonomic nervous system. Artifacts and ectopic beats were removed and replaced with the local average. Linear de-trending was applied in measured RR interval data. Fast Fourier transform was applied to estimate the power spectral densities of the RR interval variability. Low (LF, 0.04–0.15 Hz) and high frequency power (HF, 0.15–0.40 Hz) of RR interval oscillations were quantified. The ratio between LF and HF powers (LF/HF-ratio) was also analyzed. The power spectral components of HRV were computed in 5 min periods and then averaged over each 15 min phase, i.e. baseline, cold, and recovery. To analyze the overall HRV, we computed the standard deviation of all normal RR intervals (SDNN) in time domain for each 15 minutes phase.
Brachial BP was measured with the use of oscillometric sphygmomanometry (Schiller BP 200 Plus, Schiller, Switzerland) at 3-minute intervals during baseline, cold exposure, and recovery measurements. The arm of the subject was supported and the position of the cuff was at the level of the heart.
Skin Temperatures and Thermal Sensations
Skin temperature was measured continuously with the use of thermistors (NTC DC95, Digi-Key, Thief River Falls, MN, USA) and data was recorded at 12 s intervals with an eight channel temperature data logger (SmartReaderPlus, Acr Systems Inc., BC, Canada). The thermistors were placed on the middle finger, back of the hand, shoulder blade, and cheek. Thermal perception for the whole body and face was assessed using subjective judgement scales .
Subject characteristics between study groups were compared with independent t–test for continuous variables and chi-square test for categorical variables. Sensitivity analyses were conducted separately for BMI, weight, fat percentage and chronic diseases. Due to lack of effect to the results these were ignored in further analyses. Means between baseline, cold, and recovery as well as study groups were compared by 2-way repeated measures ANOVA and contrast tests. Parameters with non-Gaussian distribution were transformed into natural logarithm. In addition, we executed linear regression analyses to evaluate the rate-dependency (HR changes) of T-wave amplitude change. The results are expressed as means and their standard deviations (SD) or 95% confidence intervals (CI). Statistical analyses were performed with IBM SPSS for Windows version 19 (IBM Corp. Released 2010, Armonk, NY, USA) and significance was set at p<0.05.
The employed cold exposure involved facial cooling which could be demonstrated as a rapid (30°C to 21°C in 2 min) and robust (30°C to 15°C in 10 min) reduction in skin temperature of the cheek. In general, the winter clothing slowed down or even prevented superficial cooling of most areas of the body, as demonstrated by the only slightly lowered shoulder blade temperature (baseline 34°C vs. cold 31°C). Finger temperatures decreased by 3–4°C in the cold (26°C to 22°C). Skin temperatures on cheek and finger remained 4–7°C lower 10–15 minutes after the exposure compared to baseline measurements both in hypertensive men and controls. Skin temperature responses during baseline, cold exposure and the recovery measurements did not differ between hypertensive men and control group. Thermal perceptions (median) of the whole body and face were neutral during baseline and recovery measurements and cool during cold exposure in both study groups.
Cold exposure, ECG, HRV, BP, and arrhythmias
The detected changes in T-wave amplitude, T-peak to T-end interval, QTc, and HR are depicted in Figure 2. Cold exposure resulted in higher maximum T-wave amplitude compared to baseline in both hypertensive and control subjects in leads II and V5 (Table 2, Figure 2). We also examined the HR dependency of this parameter and estimated the following regression equation: T-wave amplitude change = 48.370+0.426 * RR change (adjusted R2 = 0.296, p<0.001). QRS-T angle and QTc interval decreased during cold exposure in both groups (Table 2). T-peak to T-end interval increased in lead II (Figure 2), but remained constant in lead V5. An example ECG presenting changes in cardiac repolarization during cold exposure is depicted in Figure 3. PR and QRS intervals were mainly unaltered. Cold exposure increased systolic (26–27 mmHg) and diastolic BP (12 mmHg) both among hypertensive and control subjects. Cold induced ECG changes did not return to the baseline level during 15 minutes recovery follow-up after the exposure (Table 2). Cold exposure increased the frequency of ventricular, but not atrial ectopic beats (Table 2). In all phases, frequency of ectopic beats was low in both study groups.
T-peak to T-end interval in lead II (A), T-wave amplitude in lead V5 (B), QTc in lead V5 (C), and HR (D) of hypertensive men (N = 51) and controls (N = 32) before (baseline), during, and after (recovery) cold exposure. Values represent means and standard error of the means.
Figure presents a median ECG from 10 consecutive beats as an example of changes in T-wave in lead II before (baseline), during, and after (recovery) cold exposure. Cold exposure increased T-peak to T-end interval and T-wave amplitude.
When examining the HRV parameters, cold exposure increased both LF (60–80%) and HF (100–110%) components of HRV, as well as SDNN (25%) in both hypertensive and control subjects (Table 2). LF/HF-ratio and HR decreased during cold exposure in both groups. HRV parameters did not return to the baseline level during 15 minutes recovery follow-up after cold exposure.
Hypertension, ECG, HRV, BP, and arrhythmias
There were no differences in ECG parameters or arrhythmias between the study groups, except for T-wave amplitude which was ca. 0.1 mV higher in controls compared with hypertensive men during all phases (p<0.001) and for unadjusted QT (p<0.05). HR was lower (6–7 beats) and LF/HF-ratio higher among hypertensive compared to control subjects. Otherwise, there were no differences in the HRV parameters between the study groups. Hypertensive subject had a higher SBP and DBP compared to controls during all phases (p<0.001), as expected.
The present study is the first to assess cardiac electrical activity, cardiovascular autonomic regulation, and arrhythmias in hypertensive middle-aged men during cold exposure similar to everyday winter circumstances in a subarctic climate. We observed that cold exposure slightly increased T-peak to T-end interval, T-wave amplitude and decreased QTc interval reflecting changes in ventricular repolarization. Simultaneously an autonomic co-activation with increased sympathetic and parasympathetic activity during the cold exposure was observed. The frequency of ventricular ectopic beats increased slightly in the cold. These cold-related responses were similar in hypertensive and control subjects.
According to our results, short-term cold exposure not involving substantial whole body cooling increased T-peak to T-end interval in lead II, reflecting prolongation of cardiac repolarization. The observed increase in T-wave amplitude, on the other hand, may indicate dispersion in duration of action potentials . In addition, cold exposure had no effect on QT interval despite a reduction in HR. The frequency of ventricular ectopic beats was slightly increased during cold exposure, which may reflect the observed cardiac electrical changes.
Previous experimental or clinical studies examining the effects of cold on cardiac electrical function have involved intense whole-body exposures, such as hypothermia , ,  and immersion to cold water  or employed exercise in cold –. These studies have shown several different effects on ECG, such as atrial , , ,  and ventricular ,  arrhythmias, interval prolongation , , , T-wave abnormalities , , Osborne waves , , , and pronounced ST-depression during exercise among cardiac patients , depending on cold exposure duration and its intensity. Only a few studies have examined cardiac electrical activity under superficial or local cold exposure, such as facial cooling with  or without breath holding .
The observed changes in our study in cardiac electrophysiology during cold exposure are based on current knowledge caused by an altered autonomic nervous system activity , , , , . This altered autonomic regulation persisted throughout the 15 minutes recovery period. One plausible reason for this could be lowered skin temperatures at recovery, which could have provided a continuing parasympathetic and sympathetic stimulus. Also, irrespective of skin temperature, the autonomic nervous system regulation may not recover immediately after the cessation of stressor. The applied cold exposure increased sympathetic activity , as demonstrated by the increased peripheral BP and cardiac work . Furthermore, the spectral analysis of HRV showed that both LF and HF increased while HR decreased as a sign of parasympathetic activation during cold exposure. Sympathetic stimulation increases T-peak to T-end interval in limb leads . Sympathetic stimulation results in either inverted or positive T-wave depending on the stimulated area , . Sympathetic activation also increases risk of arrhythmias . The observed shortened cold-related QTc interval with reduced HR but unaltered QT interval in the present study probably reflects the simultaneous effects of increased vagal and sympathetic activity , .
Sympathovagal co-activation of autonomic nervous system is a regulatory process but could cause conflicting pressures on the heart and may partially explain the observed increases in T-peak to T-end interval, T-wave amplitude, and ventricular arrhythmias. Such co-activation has been previously described in association with stimuli, like facial cooling , during cold water immersion involving breath holding , as well as during the recovery phase of exercise . The co-activation of the sympathetic and vagal outflow changes HR dynamics from more fractal to random HR organization  and could predispose to arrhythmias , .
Hypertension is related to increased sympathetic activity ,  and is a risk factor of cardiac events such as arrhythmia  and ischemic changes , . In our study, untreated hypertensive men had higher HR and LF/HF- ratio and lower T-wave amplitude compared to controls in both warm and cold conditions, consistent with previous non-cold related studies ,  and indicating elevated sympathetic activity in hypertension. Sympathetic hyperactivity combined with the cold stressor could potentate adverse ECG changes during cold exposure. Contrary to our hypothesis, we did not observe major differences in cardiac responses to cold exposure between hypertensive men and controls. The lack of major ECG changes may be due to a short disease history among the untreated hypertensive subjects in our study. It is possible that in persons whose cardiovascular disease is at a more advanced stage, disturbed circulatory adjustments or compromised cardiac oxygen supply could lead to aggravated cardiovascular responses and impaired cardiac function.
ECG parameters are commonly rate-dependent , and some of the observed changes in our study could be explained by slightly decreased HR in the cold. According to the linear regression model less than one third of the observed change in T-wave amplitude could be explained by change in HR in our study, i.e. most of the change is related to other factors. T-peak to T-end interval has been presented to have minor or no connection with HR . In contrast, QT interval is known to be strongly dependent of HR  and rate dependency of it varies with altered autonomic function , . QRS-T angle decreased with decreasing HR in our study, consistent with a previous study assessing rate-dependency of QRS-T-angle during exercise test .
Strengths and limitations
The strength of our study is that all subjects were drawn from the general population and thus the results reflect typical cardiovascular response to cold among untreated middle aged hypertensive men and men without hypertension. Furthermore, we were able to produce a strictly controlled and equal cold exposure to all subjects. Hence, we consider that our results have public health implications due to the population-based sample and the utilized cold exposure which was similar to everyday winter circumstances in a cold climate.
Performing a randomized controlled trial would have reduced possible bias related to anticipation caused by measurements. However, we emphasized familiarization and reduction of stress of all participants caused by the experimental conditions. Another limitation of our study was that some of the study participants had other chronic diseases in addition to hypertension. However, we conducted sensitivity analyses by excluding the participants with these chronic diseases, but did not detect any effects on the main results. Also, despite the population based recruitment, some selection may still occur through non-participation. We do not have information of the characteristics of the subjects declining to participate in the study.
In conclusion, our results demonstrate that moderate short-term whole body cold exposure commonly occurring in everyday life  results in altered cardiac repolarization among middle-aged men. Rather than being of clinical importance, the detected changes may be physiological responses to short-term cold exposure. Furthermore, as compared to the control group, the untreated hypertensive men are not more susceptible to cold induced changes in cardiac electrical function. However, the detected cold effects could be more substantial with more intense cold exposure or among patients with cardiac disease. Future studies are warranted to examine cardiac electrical function during cold exposure among persons with advanced cardiovascular diseases.
The study protocol approved by ethics committee – original version in Finnish.
The authors wish to thank Saana Rautakoski, Henna Hyrkäs, Jukka Jauhiainen, Heta Helakari, Arno Kandelberg, and Jaakko Takkunen for their great help in the collection of the data.
Conceived and designed the experiments: HH TVK MT AK HVH MM SKK KHH RA HR JJKJ TMI. Performed the experiments: HH MT AK HR JJKJ TMI. Analyzed the data: HH TVK MT AK HVH RB HR JJKJ TMI. Wrote the paper: HH TVK MT AK HVH MM SKK RB KHH RA HR JJKJ TMI.
- 1. The Eurowinter Group (1997) Cold exposure and winter mortality from ischaemic heart disease, cerebrovascular disease, respiratory disease, and all causes in warm and cold regions of Europe. The Eurowinter Group. Lancet 349: 1341–1346.
- 2. Fares A (2013) Winter cardiovascular diseases phenomenon. N Am J Med Sci 5: 266–279.
- 3. Manfredini R, Manfredini F, Boari B, Bergami E, Mari E, et al. (2009) Seasonal and weekly patterns of hospital admissions for nonfatal and fatal myocardial infarction. Am J Emerg Med 27: 1097–1103.
- 4. Mehta RH, Manfredini R, Hassan F, Sechtem U, Bossone E, et al. (2002) International Registry of Acute Aortic Dissection (IRAD) Investigators (2002) Chronobiological patterns of acute aortic dissection. Circulation 106: 1110–1115.
- 5. Gallerani M, Boari B, Manfredini F, Manfredini R (2011) Seasonal variation in heart failure hospitalization. Clin Cardiol 34: 389–394.
- 6. Palm F, Dos Santos M, Urbanek C, Greulich M, Zimmer K, et al. (2013) Stroke seasonality associations with subtype, etiology and laboratory results in the ludwigshafen stroke study (LuSSt). Eur J Epidemiol 28: 373–381.
- 7. Anand K, Aryana A, Cloutier D, Hee T, Esterbrooks D, et al. (2007) Circadian, daily, and seasonal distributions of ventricular tachyarrhythmias in patients with implantable cardioverter-defibrillators. Am J Cardiol 100: 1134–1138.
- 8. Page RL, Zipes DP, Powell JL, Luceri RM, Gold MR, et al. (2004) Seasonal variation of mortality in the antiarrhythmics versus implantable defibrillators (AVID) study registry. Heart Rhythm 1: 435–440.
- 9. Leppäluoto J, Pääkkönen T, Korhonen I, Hassi J (2005) Pituitary and autonomic responses to cold exposures in man. Acta Physiol Scand 184: 255–264.
- 10. Vaseghi M, Shivkumar K (2008) The role of the autonomic nervous system in sudden cardiac death. Prog Cardiovasc Dis 50: 404–419.
- 11. Khurana RK, Wu R (2006) The cold face test: A non-baroreflex mediated test of cardiac vagal function. Clin Auton Res 16: 202–207.
- 12. Shattock MJ, Tipton MJ (2012) ‘Autonomic conflict’: A different way to die during cold water immersion? J Physiol (Lond) 590: 3219–3230.
- 13. Tulppo MP, Kiviniemi AM, Hautala AJ, Kallio M, Seppänen T, et al. (2005) Physiological background of the loss of fractal heart rate dynamics. Circulation 112: 314–319.
- 14. Aslam AF, Aslam AK, Vasavada BC, Khan IA (2006) Hypothermia: Evaluation, electrocardiographic manifestations, and management. Am J Med 119: 297–301.
- 15. De Souza D, Riera AR, Bombig MT, Francisco YA, Brollo L, et al. (2007) Electrocardiographic changes by accidental hypothermia in an urban and a tropical region. J Electrocardiol 40: 47–52.
- 16. Backman C, Linderholm H (1991) Reaction of patients with effort angina to cold exposure during exercise. Arctic Med Res Suppl: 526–528.
- 17. Schmid JP, Morger C, Noveanu M, Binder RK, Anderegg M, et al. (2009) Haemodynamic and arrhythmic effects of moderately cold (22 degrees C) water immersion and swimming in patients with stable coronary artery disease and heart failure. Eur J Heart Fail 11: 903–909.
- 18. Sanchez-Gonzalez MA, Figueroa A (2013) Cold exposure attenuates post exercise cardiovagal reactivation and sympathetic withdrawal. Auton Neurosci 176: 95–97.
- 19. Parati G, Esler M (2012) The human sympathetic nervous system: Its relevance in hypertension and heart failure. Eur Heart J 33: 1058–1066.
- 20. Manolis AJ, Rosei EA, Coca A, Cifkova R, Erdine SE, et al. (2012) Hypertension and atrial fibrillation: Diagnostic approach, prevention and treatment. position paper of the working group ‘hypertension arrhythmias and thrombosis’ of the european society of hypertension. J Hypertens 30: 239–252.
- 21. Hintsala H, Kandelberg A, Herzig KH, Rintamäki H, Mäntysaari M, et al. (2014) Central aortic blood pressure of hypertensive men during short-term cold exposure. Am J Hypertens 27 (5) 656–664.
- 22. Komulainen S, Tähtinen T, Rintamäki H, Virokannas H, Keinänen-Kiukaanniemi S (2000) Blood pressure responses to whole-body cold exposure: Effect of carvedilol. Eur J Clin Pharmacol 56: 637–642.
- 23. Barnett AG, Sans S, Salomaa V, Kuulasmaa K, Dobson AJ, et al. (2007) The effect of temperature on systolic blood pressure. Blood Press Monit 12: 195–203.
- 24. Parati G, Stergiou GS, Asmar R, Bilo G, de Leeuw P, et al. (2008) European society of hypertension guidelines for blood pressure monitoring at home: A summary report of the second international consensus conference on home blood pressure monitoring. J Hypertens 26: 1505–1526.
- 25. ISO 9920 (2007) Ergonomics of the thermal environment - estimation of thermal insulation and water vapour resistance of a clothing ensemble.
- 26. Karjalainen J, Viitasalo M, Mänttäri M, Manninen V (1994) Relation between QT intervals and heart rates from 40 to 120 beats/min in rest electrocardiograms of men and a simple method to adjust QT interval values. J Am Coll Cardiol 23: 1547–1553.
- 27. Daskalov IK, Christov II (1999) Electrocardiogram signal preprocessing for automatic detection of QRS boundaries. Med Eng Phys 21: 37–44.
- 28. Edenbrandt L, Pahlm O (1988) Vectorcardiogram synthesized from a 12-lead ECG: Superiority of the inverse dower matrix. J Electrocardiol 21: 361–367.
- 29. Pardey J, Jouravleva S (2004) The next-generation holter revolution: From “analyse, edit, print” to “analyse, print”. Computers in cardiology 31: 373–376.
- 30. Huikuri HV, Valkama JO, Airaksinen KE, Seppänen T, Kessler KM, et al. (1993) Frequency domain measures of heart rate variability before the onset of nonsustained and sustained ventricular tachycardia in patients with coronary artery disease. Circulation 87: 1220–1228.
- 31. ISO 10551 (1995) Ergonomics of the thermal environment - assessment of the influence of the thermal environment using subjective judgement scales.
- 32. Bernardo D, Murray A (2000) Explaining the T-wave shape in the ECG. Nature 403: 40.
- 33. Mattu A, Brady WJ, Perron AD (2002) Electrocardiographic manifestations of hypothermia. Am J Emerg Med 20: 314–326.
- 34. Wierzba T, Musial P, Cwikalowska H (2011) Variable profile of individual heart rate responses to cold water immersion apnea in healthy late adolescent men. - Georgian Med News 196–197: 28–38.
- 35. Sundaram S, Carnethon M, Polito K, Kadish AH, Goldberger JJ (2008) Autonomic effects on QT-RR interval dynamics after exercise. Am J Physiol Heart Circ Physiol 294: H490–H497.
- 36. Vaseghi M, Yamakawa K, Sinha A, So E, Zhou W, et al. (2013) Modulation of regional dispersion of repolarization and T-peak to T-end interval by right and left stellate ganglia. - Am J Physiol Heart Circ Physiol 305 (7) H1020–H1030.
- 37. Kralios FA, Martin L, Burgess MJ, Millar K (1975) Local ventricular repolarization changes due to sympathetic nerve-branch stimulation. Am J Physiol 228: 1621–1626.
- 38. Conrath CE, Opthof T (2006) Ventricular repolarization: An overview of (patho)physiology, sympathetic effects and genetic aspects. Prog Biophys Mol Biol 92: 269–307.
- 39. Arai K, Nakagawa Y, Iwata T, Horiguchi H, Murata K (2013) Relationships between QT interval and heart rate variability at rest and the covariates in healthy young adults. Auton Neurosci 173: 53–57.
- 40. Tulppo MP, Kiviniemi AM, Hautala AJ, Kallio M, Seppänen T, et al. (2011) Sympatho-vagal interaction in the recovery phase of exercise. Clin Physiol Funct Imaging 31: 272–281.
- 41. Chen PS, Tan AY (2007) Autonomic nerve activity and atrial fibrillation. Heart Rhythm 4: S61–S64.
- 42. Grassi G (2010) Sympathetic neural activity in hypertension and related diseases. Am J Hypertens 23: 1052–1060.
- 43. Murphy BP, Stanton T, Dunn FG (2009) Hypertension and myocardial ischemia. Med Clin North Am 93: 681–695.
- 44. Wu JS, Lu FH, Yang YC, Lin TS, Chen JJ, et al. (2008) Epidemiological study on the effect of pre-hypertension and family history of hypertension on cardiac autonomic function. J Am Coll Cardiol 51: 1896–1901.
- 45. Dilaveris P, Gialafos E, Poloniecki J, Hnatkova K, Richter D, et al. (2000) Changes of the T-wave amplitude and angle: An early marker of altered ventricular repolarization in hypertension. Clin Cardiol 23: 600–606.
- 46. Andersen MP, Xue JQ, Graff C, Kanters JK, Toft E, et al. (2008) New descriptors of T-wave morphology are independent of heart rate. J Electrocardiol 41: 557–561.
- 47. Kenttä T, Karsikas M, Kiviniemi A, Tulppo M, Seppänen T, et al. (2010) Dynamics and rate-dependence of the spatial angle between ventricular depolarization and repolarization wave fronts during exercise ECG. Ann Noninvasive Electrocardiol 15: 264–275.
- 48. Mäkinen TM, Raatikka VP, Rytkönen M, Jokelainen J, Rintamäki H, et al. (2006) Factors affecting outdoor exposure in winter: Population-based study. Int J Biometeorol 51: 27–36.