Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Sep 15.
Published in final edited form as: Sci Total Environ. 2022 May 26;839:156235. doi: 10.1016/j.scitotenv.2022.156235

Geomagnetic disturbances reduce heart rate variability in the Normative Aging Study

Carolina L Zilli Vieira a,*, Kelly Chen a, Eric Garshick b,c,d, Man Liu a, Pantel Vokonas e, Petter Ljungman f,g, Joel Schwartz a,d, Petros Koutrakis a
PMCID: PMC9233046  NIHMSID: NIHMS1812569  PMID: 35644403

Abstract

Background:

Solar and geomagnetic activity (GA) have been linked to increased cardiovascular (CVD) events. We hypothesize that heart rate variability (HRV) may be the biological mechanism between increased CVD risk and intense geomagnetic disturbances (GMD).

Methods:

To evaluate the impact of GA and intense GMD on HRV in 809 elderly men [age mean 74.5 (SD = 6.8)] enrolled in the Normative Aging Study (Greater Boston Area), we performed repeated-measures using mixed-effects regression models. We evaluated two HRV outcomes: the square root of the mean squared differences of successive normal-to-normal intervals (r-MSSD) and the standard deviation of normal-to-normal heartbeat intervals (SDNN) in milliseconds (ms). We also compared the associations between Kp and HRV in patients with and without comorbidities such as diabetes and coronary heart diseases (CHD). We used data on global planetary K-Index (Kp) from middle latitudes as a GA and GMD (>75th Kp) parameters from the National Oceanic and Atmospheric Agency’s Space Weather Prediction Center.

Results:

We found a near immediate effect of continuous and higher Kp on reduced HRV for exposures up to 24 h prior to electrocardiogram recording. A 75th percentile increase in 15-hour Kp prior the examination was associated with a −14.7 ms change in r-MSSD (95 CI: −23.1, −6.3, p-value = 0.0007) and a −8.2 ms change in SDNN (95 CI: −13.9, −2.5, p-value = 0.006). The associations remained similar after adjusting the models for air pollutants over the exposure window prior to the event. In periods of intense GMD, the associations were stronger in patients with CHD and non-diabetes.

Conclusions:

This is the first study to demonstrate the potential adverse effects of geomagnetic activity on reduced heart rate variability in a large epidemiologic cohort over an extended period, which may have important clinical implications among different populations.

Keywords: Solar and geomagnetic activity, Circadian rhythm dysregulation, Reduced heart rate, Cardiovascular risk

GRAPHICAL ABSTRACT

graphic file with name nihms-1812569-f0001.jpg

1. Introduction

Reduced heart rate variability (HRV) is considered an independent risk factor for adverse cardiovascular events, including myocardial infarction and other ischemic events, paroxysmal atrial fibrillation, and sudden cardiac death (Hisako et al., 1996; Kop et al., 2001a; Bettoni and Zimmermann, 2002). Cardiovascular diseases (CVD), including myocardial infarction (MI) and other coronary heart diseases (CHD) and stroke, remain the leading cause of mortality worldwide (Sidney et al., 2016). An emerging body of solar and geomagnetic activity (GA) research has focused on CVD outcomes in the general population, finding that CVD-related mortality and morbidity are associated with greater disturbances of the Earth’s magnetic field (EMF) (McCraty et al., 2017; Palmer et al., 2006; Cornélissen et al., 2002; Vencloviene et al., 2014a; Jaruševičius et al., 2018; Vieira et al., 2019; Vieira et al., n.d.). Earth’s magnetosphere, encircling a region of space around our planet, is subject to complex interactions between the EMF, solar system, and interplanetary space. Solar activity exerts a dominant influence in the short-term EMF disturbances [geomagnetic disturbances (GMD)] through the injection of high-energy plasma and magnetized emissions, such as solar flares and coronal mass ejections, into the Earth’s space environment. These phenomena range in intensity largely according to quasi-periodic 11-year solar cycles (Hasegawa et al., August 2004).

Research focusing on human health has linked short-term increases in GMD to a range of adverse health effects, including CVDs, neurological diseases, behavioral diseases, and increased total mortality (McCraty et al., 2017; Palmer et al., 2006). Many of these studies have been completed among small to moderate-sized cohorts, including that by Cornelissen et al., observing a 5 increase in risk of MI mortality during years of high solar activity compared to years of low activity in Minnesota, USA (Cornélissen et al., March 2002). Vencloviene et al. investigated the association between GMD and the survival of patients hospitalized for acute coronary syndromes in Lithuania, finding a hazard ratio of 1.6 associated with active GMD compared to geomagnetically quiet days during the second day after admission (Vencloviene et al., 2014a). Recently, several large epidemiological studies have sought to determine the impact of short-term exposure to GMD on a greater spatial and temporal scale (Jaruševičius et al., 2018). In our previous study (Vieira et al., 2019), we reported near immediate effects of GMD on cardiovascular outcomes, measured as total cardiovascular, CVD, and MI mortality risk (Vieira et al., 2019). Researchers have further investigated the effect of GMD on biological life forms in both observational and experimental studies, finding links to evolutionary, behavioral, and biological processes in organisms across various taxa, particularly those processes sensitive to or dependent on perception of magnetic fields (Krylov et al., 2014).

Several small cohort studies have observed associations between the short-term effect of solar activity and altered HRV. Ranging from prior day to 3-day exposure windows, solar activity has been associated with reduced HRV during relatively short-term follow-up periods that range from two months to one year (Alabdulgader et al., 2018; McCraty et al., n.d.). In addition, a gap remains in accounting for potential confounding by exposure to ambient particulate pollution, as the links between particulate exposures and both reduced HRV and adverse cardiac outcomes are well established (Mordukhovich et al., 2015; Baja et al., 2013; Park et al., 2005; Zanobetti et al., 2010). Both particulate pollution and solar and geomagnetic activity play a role in the dysregulation of the autonomic nervous system (ANS), which may be the link between them and increased risk of CVD (Vieira et al., n.d.). Intense solar and geomagnetic activity can disrupt the 24 h-circadian rhythm, promoting the imbalance of the ANS, and reducing melatonin secretion (Vieira et al., n.d.). A predominance of sympathetic activity and/or diminished parasympathetic activity is believed to drive the association between HRV and adverse cardiovascular outcomes (Bell et al., 1972).

While the aforementioned studies observed associations on hospital admissions and mortality, gaps remain in elucidating specific mechanisms. In this study, we evaluated the Veteran Affairs (VA) Normative Aging Study (NAS) cohort to investigate the impact of geomagnetic activity (GA) and periods of intense GMD on heart rate variability in a well-defined geographical area. We performed sensitivity analyses by adjusting the models for air pollutants, and comparing patients with and without different comorbidities such as diabetes and CHD. To the best of our knowledge, this study will be the first to evaluate this association in a large population over an extended period, allowing for repeated measurements over 17-year follow ups.

2. Methods

2.1. Study population

The subjects included in this analysis were enrolled in the NAS, an ongoing longitudinal study of aging established by the U.S. Department of Veterans Affairs in 1963 (Zanobetti et al., n.d.; Bell, Rose, and Damon, n.d.). The NAS cohort consists of men from the Greater Boston area aged 21–80 at baseline who were enrolled after an initial health screening determined they were free of known chronic medical conditions. Since 1963, every 3 to 5 years, participants have returned onsite to undergo routine physical examinations, laboratory tests, collection of medical history information, and completion of questionnaires on smoking history, education level, food intake, and other factors that may affect health. Heart-rate variability measurements were conducted on 809 participants with 2540 visits from November 14, 2000, to December 16, 2017. Participants were examined under conditions of abstaining from smoking and an overnight fast. Diabetes was defined as a physician diagnosis of diabetes, use of any diabetes medication, or fasting glucose >126 mg/dL. Body mass index (BMI) was computed as the weight in kilograms divided by the square of the height in meters and recalculated based on height and weight measurements at each visit. Self-reported data on diabetes status and statin use were updated at each study visit and confirmed by a physician interview. The study was approved by the Human Research Committees of the Harvard School of Public Health and the Veterans Affairs Boston Healthcare System, and written informed consent was obtained from participants prior to participation (Shaffer and Ginsberg, 2017).

2.2. Electrocardiogram (ECG) measurement and analysis

Participants were examined in the morning between 7 am and 9 am during each participant’s regularly scheduled visit. Following a 5-minute rest, participants were seated and a 5-to 10-minute ECG was taken using a two-channel, five-lead ECG monitor (Trillium 3000; Forest Medical, Inc., East Syracuse, NY) using a sampling rate of 256 Hz per channel. Heart rate and HRV measures were calculated using PC-based software (Trillium 3000 PC Companion Software for MS Windows; Forest Medical, East Syracuse, NY) and assessed by an experienced reviewer to correct for any errors or artifacts. The best 7 min were used for computation. The following HRV measures were obtained from these calculations: the square root of the mean squared differences of successive normal-to-normal intervals (r-MSSD) and standard deviation of normal-to-normal heart beat intervals (SDNN) (Park et al., n.d.). A detailed description of the protocol has been previously described (Baja et al., n.d.; Mordukhovich et al., n.d.).

2.3. Solar activity-induced GMD assessment

The parameter planetary Kp-index (or Kp) is an excellent indicator of disturbances in Earth’s magnetic field. The Kp index (“Kennziffer” or characteristic digit) is derived for each of worldwide observatories by first obtaining the amplitude of the largest variation observed in either of the horizontal components of Earth’s magnetic field in units of nanotesla (nT) during a 3-hour interval after the regular daily variation is removed. Kp index ranges from 0 (a calm day of low GMD activity) to 9 (extreme geomagnetic disturbance). Since 1994, the National Oceanic and Atmospheric Agency (NOAA)’s Space Weather Prediction Center (SWPC) (NOAA, n.d.) have derived this measure from the following mid-to high-latitude ground-based magnetometers: Sitka, Alaska; Meanook, Canada; Ottawa, Canada; Fredericksburg, Virginia; Hartland, UK; Wingst, Germany; Niemegk, Germany; and Canberra, Australia.

As the participants in our study resided in a mid-latitude region, we used mid-latitude Kp index from the NOAA Space Weather Prediction Center. We calculated 8 exposure windows using the 3-hour Kp data. The exposure windows ranged from the time of ECG examination (typically around 7 to 9 am) to 24 h prior to the visit (0–3, 0–6, 0–9, 0–12, 0–15, 0–18, 0–21, and 0–24 h moving averages), using continuous Kp, and Kp dichotomized at the 75th percentile (periods of intense GMD).

2.4. Air pollution exposure assessment

We performed sensitivity analysis to investigate the confounding impact of ambient air pollution exposures in the GA models. Ambient pollution was measured at the Harvard Supersite located in downtown Boston, MA, located approximately 1 km from the medical center where the study participants were examined. Hourly measurements were made for fine particle mass (PM2.5) using a tapered element oscillation microbalance (TEOM) (model 1400A; Rupprecht and Pastashnick, East Greenbush, NY). Particle Number (PN) concentrations were measured using a condensation particle counter (model 3022A; TSI Inc., Shoreview, MN). Black Carbon (BC) was measured using an Aethalometer (Magee Scientific Company, model AE-16, Berkeley, CA). A detailed description of the supersite has been previously published (Kang et al., 2010; Nyhan et al., 2018). To evaluate lagged effects of air pollutants, we used the same 8 exposure windows as we had considered for Kp, ranging from the current hour to the 24-hour moving average of each of the air pollutants.

2.5. Weather parameter assessment

Models were adjusted a priori for room temperature (°C), and daily ambient temperature (°C) and relative humidity (RH) (Schwartz et al., 2004; Wilmshurst, 1994). Daily local weather data were obtained from the NOAA.

2.6. Statistical analyses

In separate models, we examined whether GA and intense GMD were associated with two HRV outcomes (r-MSSD and SDNN). We considered eight exposure windows, from current hour (ECG exam time) to 24-hour moving average prior to the exam, using continuous Kp and Kp dichotomized at the 75th percentile. We applied linear mixed effects regression models to account for correlation among measurements on the same subject across different medical visits. In the models, we included a random intercept for each subject, either SDNN and r-MSSD as the dependent variable, and fixed effects for covariates. The primary models included a single exposure variable (Kp, PM2.5, BC, or PN). As previous studies have found associations with HRV and ambient particulate pollution, we controlled for several pollutants in our models to evaluate their confounding impact on our findings. We performed sensitivity analyses to evaluate the confounding impact of pollutants in the models, and also to compare patients with and without comorbidities (diabetes and CHD). To do so, we first included each of the pollutants (PM2.5, BC, and PN) as the exposure variable to determine the hourly exposure window with the maximum moving day average effect on each of the HRV outcomes. Then we ran second set of models including the full set of covariates and two exposure variables: Kp and PM2.5; Kp and BC; Kp and PN, including one of the eight exposure windows for Kp and the exposure window for the pollutant with its corresponding maximum effect. We presented the effect estimates as unit change (ms) in HRV. In the continuous Kp analysis, the estimates are described per interquartile range (IQR) increase in Kp.

Based on previous NAS studies, we selected the following covariates a priori as clinically important predictors to be included in the models: age, race (white or non-white), BMI, smoking status, cumulative smoking in pack-years, use of cardiac medications, supine pulse, cardiac medication use (alpha blockers, beta blockers, or ACE inhibitors), statin-drugs, diabetes medication, room temperature, ambient temperature, relative humidity, and to capture short term and seasonal changes, the day of the week and sine and cosine of time (Luttmann-Gibson et al., 2010). We assessed the correlation among the outcomes and covariates (Supplementary material) by calculating Pearson correlation coefficients. We performed all analyses using R software 4.0.2. (http://cran-r-project.org).

3. Results

3.1. Descriptive results

Tables 1 and 2 show the demographic and clinical characteristics of the participants at the first visit and all visits from 2000 to 2017. In the first visit, participants were mostly all elderly white male, having a mean age of 74.5 years (SD = 6.8 years). Furthermore, 14.2% had diabetes and the majority was former smokers. CHD, stroke, and hypertension prevalence were respectively 29.3, 7.1, and 72.1%. In addition, 34.8% reported having two or more drinks per day. Participants on average had 3.1 visits. Participants with more than one visit were on average older, had a higher prevalence of CHD, stroke, diabetes, and hypertension, and more likely to be taking at least one cardiac medication and to be on statin-drugs (Table 1).

Table 1.

Characteristics of NAS subjects 2000–2017.

First visit All visits
N = 809 Nobservations = 2540
Mean (SD) Mean (SD)
Age (y) 74.5 (6.8) 76.2 (7.2)
BMI (kg/m3) 28.3 (4.1) 27.9 (4.1)
Cumulative cigarette (packyears) 21.6 (26.8) 19.5 (23.7)
Pulse supine (bpm) 69.8 (8.4) 67.9 (9.9)
N (%) N (%)
Race
White 789 (98.3) 2467 (97.9)
 Non-White 20 (1.7) 73 (2.1)
Alcohol intake (2 drinks/day) 282 (34.8) 465 (18.3)
Smoking status
 Current 247 (30.6) 803 (31.7)
 Former 38 (4.7) 103 (4.1)
 Never 522 (64.7) 1630 (64.3)
History of CVD
CHD 237 (29.3) 889 (35)
Stroke 57 (7.1) 208 (8.2)
Diabetes 115 (14.2) 411 (16.2)
Hypertension 583 (72.1) 1953 (76.9)
Diabetes medication 76 (9.4) 263 (10.4)
Cardiac medications
 Any 430 (53.3) 1599 (65.8)
 Alpha 87 (10.8) 432 (17)
 ACE inhibitors 205 (25.4) 774 (34.2)
 Beta 263 (32.5) 1019 (40.1)
Statin drugs 282 (34.9) 1298 (51.1)
Open in a new tab

Table 2.

Distribution of HRV measures r-MSSD and SDNN during 2000–2017.

Mean ± SD Min Max IQR
r-MSSD (ms) 68 ± 87 0 597 65.0
SDNN (ms) 57 ± 58 0 673 44.0
Open in a new tab

Exposure levels for GA, particulate pollution, and meteorological parameters across all the study visits are reported in Table 3. For daily Kp, we observed a mean exposure level of 8.5 (SD = 11.3; IQR = 7). The mean concentrations of PM2.5 and BC were below the USEPA’s National Ambient Air Quality Standard (NAAQS) levels. For ambient air pollution exposure, corresponding values for daily exposures were: for PM2.5, 9.4 μg/m3 (SD = 6.4; IQR = 6.6); for BC, 0.8 μg/m3 (SD = 0.4; IQR = 0.5); and for PN, 20,963 particles/cm3 (SD = 11,852; IQR = 15,798). We reported exposure levels across all study visits for relative humidity, room temperature, and ambient temperature. For relative humidity, we observed a mean relative humidity of 68 (SD = 17.4; IQR = 27.2). There was a significant correlation between r-MSSD and SDNN and middle-latitude Kp from 6 h to 12 h prior to the events (Supplementary material). The distribution of annual middle latitude Kp and HRV (ms) parameters are showed in the Supplementary material (SF1).

Table 3.

Distribution of daily solar and particulate pollution exposures, and weather parameters during 2000–2017.

Mean ± SD Min Max IQR
Mid-latitude geomagnetic disturbance (Kp-index) 8.5 ± 11.3 0.0 199 7.0
PM2.5 (μg/m3) 9.4 ± 6.4 0.2 58.4 6.6
Black carbon (μg/m3) 0.8 ± 0.4 0.4 2.5 0.5
Particle number (number/cm3) 20,963 ± 11,852 3447 92,400 15,798
Relative humidity (%) 68 ± 17.4 7.2 99.7 27.2
Room temperature (°C) 23.9 ± 1.6 12.0 31.0 2.0
Ambient temperature (°C) 13.2 ± 8.8 −14.5 32.3 13.8
Open in a new tab

3.2. GMD and HRV

We calculated the effect estimates of GA and intense GMD beginning at ECG time to up to 24 h prior to the examination. Overall there were significant associations between continuous hourly Kp and r-MSSD up to 9 h, and SDNN up to 6 h prior to the examination (Fig. 1). The magnitude of the associations in the models adjusted for pollutants was similar to the non-adjusted models (Fig. 2). Fig. 3 reports change in r-MSSD and SDNN (ms) during intense GMD (Kp > 75th percentile) compared to calmer days (Kp < 75th percentile) without adjustment for pollutants. We observed the greatest effects during the 0–15-hour moving average, with a change of −14.7 ms change in r-MSSD (95 CI: −23.1, −6.3, p-value = 0.0007) and a −8.2 ms change in SDNN (95 CI: −13.9, −2.5, p-value = 0.006). Furthermore, all exposure windows between the 3- and 24-hour exposure windows were significantly associated with reductions in r-MSSD, and nearly all for SDNN. The reductions in HRV tended to diminish as the exposure moving average increased beyond the 15-hour moving average.

Fig. 1.

Fig. 1.

Open in a new tab

Reduction in r-MSSD and SDNN associated with hourly geomagnetic activity adjusted for all covariates. Change in r-MSSD and SDNN (ms) per IQR increase in continuous Kp (95%CI). Models were adjusted for age, race (white or non-white), BMI, smoking status, cumulative smoking in pack-years, use of cardiac medications, supine pulse, cardiac medication use (alpha blockers, beta blockers, or ACE inhibitors), statin-drugs, diabetes medication, room temperature, ambient temperature, relative humidity, and to capture short term and seasonal changes, the day of the week and sine and cosine of time.

Fig. 2.

Fig. 2.

Open in a new tab

Reduction in r-MSSD and SDNN associated with hourly geomagnetic activity adjusted for all covariates and pollutants. Change in r-MSSD and SDNN (ms) per IQR increase in continuous Kp (95%CI). Models were adjusted for age, race (white or non-white), BMI, smoking status, cumulative smoking in pack-years, use of cardiac medications, supine pulse, cardiac medication use (alpha blockers, beta blockers, or ACE inhibitors), statin-drugs, diabetes medication, room temperature, ambient temperature, relative humidity, and to capture short term and seasonal changes, the day of the week and sine and cosine of time.

Fig. 3.

Fig. 3.

Open in a new tab

Reduction in r-MSSD and SDNN associated with intense hourly GMD (75th in Kp index) adjusted for all covariates. Change in r-MSSD and SDNN (ms) in 75th percentile increase in Kp over 25th percentile (95%CI). Models were adjusted for age, race (white or non-white), BMI, smoking status, cumulative smoking in pack-years, use of cardiac medications, supine pulse, cardiac medication use (alpha blockers, beta blockers, or ACE inhibitors), statin-drugs, diabetes medication, room temperature, ambient temperature, relative humidity, and to capture short term and seasonal changes, the day of the week and sine and cosine of time.

In the sensitivity analyses, in periods of intense GMD, there were significant reductions of HRV in patients with non-diabetes and CHD (Table 4). There were significant associations between HRV and continuous Kp in patients with CHD (Supplementary material and Table 4), and insignificant associations in patients with no history of CHD. There was no meaningful difference in the associations between continuous Kp and r-MSSD and SDNN among patients with and without diabetes (Supplementary material). The associations between GMD and reduced HRV remained similar after adjusting for pollutants (Figs. 2 and 4, and Supplementary material).

Table 4.

Association between SDNN and r-RMSSD and intense GMD (>75th percentile) in patients with/without comorbidities.

Sensitive analysis Exposure window SDNN RMSSD
Middle latitude Kp index at 75th CHD Without CHD 0–3 −3.42 (95% CI: −13.16, 6.31) −6.1 (95% CI: −21.22, 9.02)
0–6 −5.08 (95% CI: −11.82, 1.67) −9.21 (95% CI: −19.7, 1.27)
0–9 −2.21 (95% CI: −8.79, 4.37) −5.14 (95% CI: −15.37, 5.09)
0–12 −3.52 (95% CI: −10.25, 3.21) −7.57 (95% CI: −18.03, 2.89)
0–15 −5.99 (95% CI: −12.78, 0.79) −11.03 (95% CI: −21.56, −0.49)*
0–18 −5.15 (95% CI: −11.49, 1.2) −11.87 (95% CI: −21.72, −2.01)*
0–21 −2.75 (95% CI: −9.12, 3.61) −8.81 (95% CI: −18.69, 1.08)
0–24 −2.3 (95% CI: −8.79, 4.2) −8.76 (95% CI: −18.84, 1.32)
With CHD 0–3 −11.49 (95% CI: −23.61, 0.63) −16.21 (95% CI: −32.89, 0.47)
0–6 −13.82 (95% CI: −27.42, −0.21)* −22.12 (95% CI: −40.81, −3.44)*
0–9 −9.74 (95% CI: −22.05, 2.57) −16.69 (95% CI: −33.61, 0.23)
0–12 −12.09 (95% CI: −24.28, 0.1) −20.28 (95% CI: −37.03, −3.52)*
0–15 −11.85 (95% CI: −23.64, −0.06)* −19.13 (95% CI: −35.24, −3.03)*
0–18 −9.76 (95% CI: −21.74, 2.23) −15.07 (95% CI: −31.48, 1.33)
0–21 −10.32 (95% CI: −22.36, 1.72) −16.03 (95% CI: −32.48, 0.42)
Diabetes Without diabetes 0–24 −10.92 (95% CI: −23.02, 1.17) −16.35 (95% CI: −32.88, 0.19)
0–3 −6.37 (95% CI: −16.29, 3.55) −7.26 (95% CI: −21.73, 7.21)
0–6 −7.7 (95% CI: −14.56, −0.84)* −12.52 (95% CI: −22.5, −2.54)*
0–9 −6.11 (95% CI: −12.73, 0.52) −9.84 (95% CI: −19.48, −0.2)*
0–12 −8.36 (95% CI: −14.77, −1.95)* −13.5 (95% CI: −22.81, −4.18)*
0–15 −8.49 (95% CI: −14.74, −2.25)* −14.87 (95% CI: −23.92, −5.82)*
0–18 −7.01 (95% CI: −13.43, −0.6)* −13.26 (95% CI: −22.55, −3.97)*
0–21 −5.93 (95% CI: −12.33, 0.48) −12.34 (95% CI: −21.62, −3.07)*
With diabetes 0–24 −4.81 (95% CI: −11.12, 1.5) −10.97 (95% CI: −20.1, −1.83)*
0–3 −2.97 (95% CI: −22.34, 16.39) −8.65 (95% CI: −37.6, 20.29)
0–6 −4.24 (95% CI: −19.73, 11.25) −4.62 (95% CI: −27.77, 18.53)
0–9 −6.41 (95% CI: −20.82, 8) −11.17 (95% CI: −32.71, 10.37)
0–12 −5.68 (95% CI: −20.05, 8.7) −10.87 (95% CI: −32.34, 10.6)
0–15 −8.9 (95% CI: −23.95, 6.15) −15.06 (95% CI: −37.57, 7.44)
0–18 −11.12 (95% CI: −26.22, 3.97) −18.72 (95% CI: −41.26, 3.82)
0–21 −11.32 (95% CI: −26.59, 3.96) −18.83 (95% CI: −41.69, 4.03)
0–24 −11.91 (95% CI: −26.52, 2.7) −20.26 (95% CI: −42.07, 1.55)
Open in a new tab

[Bold results(*): p-value<0.05].

Fig. 4.

Fig. 4.

Open in a new tab

Reduction in r-MSSD and SDNN associated with intense hourly GMD adjusted for covariates and air pollution. Change in r-MSSD and SDNN (ms) in 75th percentile increase in Kp over 25th percentile (95%CI). Models were adjusted for age, race (white or non-white), BMI, smoking status, cumulative smoking in pack-years, use of cardiac medications, supine pulse, cardiac medication use (alpha blockers, beta blockers, or ACE inhibitors), statin-drugs, diabetes medication, room temperature, ambient temperature, relative humidity, and to capture short term and seasonal changes, the day of the week and sine and cosine of time. Models were also adjusted for each of the pollutants, using the exposure window of each pollutant determined to have the maximum effect.

4. Discussion

To our knowledge, this study is the first to examine the impact of GMD on heart rate variability in a large cohort of elderly men using repeated measurements for each subject over 16 years. We observed consistent and statistically significant reductions in both HRV parameters (r-MSSD and SDNN) associated with geomagnetic activity, and stronger association with periods of greater GMD intensity. The associations were greater in patients with CHD and no diabetes. Notably, we found almost an immediate effect, with the strongest effects observed during the zero (exam time–between 7 am and 9 am) to 24-hour exposure window prior to the exam. These HRV-GMD associations remained even after adjusting for air pollution exposures.

The morning period between 6 am and noon has been described as the highest risk period of adverse cardiovascular events, including myocardial infarction and sudden cardiac death, when compared to the rest of the day, signaling a link to the circadian pattern (Cohen et al., 1997), which may be linked to oscillations of GA. Dysregulation of the 24 h-circadian rhythm, ANS, and melatonin secretion related to intense solar and geomagnetic activity, can increase the risk of abnormal heart rhythm, and also potentiate the air pollution-related CVD risk (Vieira et al., 2019; Vieira et al., n.d.). A growing body of literature suggests that short-term GMD may modulate ANS functions and subsequent downstream pathways regulated by the ANS, including those that affect HRV (McCraty et al., 2017; Palmer et al., 2006; Cornélissen et al., 2002; Vencloviene et al., 2014a; Jaruševičius et al., 2018; Vieira et al., 2019; Vieira et al., n.d.). Studies show that magneto-reception systems composed by photosensitive retinal proteins called cryptochromes detect GMD-related electromagnetic field variations from the environment (Vieira et al., 2019). The overstimulation of these systems over-activates the functions of the central nervous system, unbalancing standard circadian rhythm processes and the ANS activity (Vieira et al., 2019). It is well known that sympathetic and parasympathetic nervous system activities of the ANS regulate core functions such as HRV, breathing, and metabolic processes in the body (Odemuyiwa et al., 1991; Kop et al., 2001b; Hisako, 1996). Intracellularly, these pathways may raise intracellular Ca2+ levels via voltage-gated ion channels leading to increased oxidative stress and to irregular cardiac action potentials in cardiac cells to reduce HRV (Vieira et al., 2019). Our results substantiate prior studies conducted within smaller cohorts, in which solar activity parameters ranging from prior day to 3-day windows were associated with reductions in heart rate variability, as observed over two-week to 1 year follow-up periods (Baja et al., 2013; Park et al., 2005). Furthermore, the findings reported here corroborate results of large epidemiological studies reporting near immediate effects of solar activity on cardiovascular outcomes, measured as total cardiovascular, CVD, and MI mortality risk (Vieira et al., 2019; Mendoza and Diaz-Sandoval, 2004; Vencloviene et al., 2014b). HRV is an indicative of ANS dysregulation, and has been associated with increased risk for CHD and worse CHD prognosis (Greiser et al., 2005). Although our findings showed patients with CHD or non-diabetic presented lower HRV during intense GMD, the biological mechanisms remain unclear. Reduced HRV during intense GMD may increase the risk of adverse events in patients with CHD, including heart failure and mortality risk.

Overall, this study reported an almost immediate reduction in both HRV parameters, r-MSSD and SDNN, associated with GA, especially during greater GMD, which remained significant even after adjustment for air pollutants. This exposure nexus could have important implications for cardiac health and preventative strategies, as HRV is a predictor of CVD morbidity and mortality (Hillebrand et al., 2013; Sessa et al., 2018; Boudreau et al., 2012). Previous studies have reported a 32–45% increased risk of a first cardiovascular event associated with low HRV in populations without known CVD, and have also attributed low HRV as a predictor for sudden cardiac death, which is responsible for about 25 of deaths in clinical cardiology (Hillebrand et al., 2013). Hillebrand et al. concluded that an increase of 1 s in SDNN can reduce 1 of the risk of fatal or non-fatal CVD (Hillebrand et al., 2013).

Our study has limitations. Mostly of our patients were healthy elderly white men (97.2), living in a high latitudinal area at sea level (Boston, MA, area). Our findings may not be applicable in women, different racial groups, and in populations living in other locations. However, our study presents several strengths. To begin with, we studied this association in a larger cohort than those previously studied and over a longer, 17-year follow up. The NAS cohort is a well-validated cohort that allowed for use of repeated measurements and adjustment of numerous covariates that may confound the association between mid-latitude GMD and HRV. Furthermore, particulate pollution, which may be an important confounder of the association between GMD and HRV, was measured with three ambient particulate measurements and adjusted on the same hourly scale as the solar activity exposure. Lastly, while electromagnetic fields driven by the solar activity are known to vary across a large area, there is likely minimal exposure misclassification given the relatively small geographic area of the cohort within the greater Boston area.

5. Conclusions

Our findings showed a significant impact of GA and intense GMD on the reduction in HRV up to 24 h prior to the electrocardiogram in elderly individuals. The associations remained even after adjusting for air pollutants. In periods of intense GMD, the reduction of HRV was even stronger in patients with CHD and no diabetes. This exposure nexus could have important clinical implications for cardiac health and preventative strategies, as HRV is a predictor of CVD morbidity and mortality caused by the dysregulation of sympathetic and/or parasympathetic activity. Future studies may investigate differential susceptibility and related biological mechanisms based on location, age, and other pre-existing comorbidities in different populations.

Supplementary Material

Supplement

HIGHLIGHTS.

  • Intense geomagnetic activity up to 24 h reduced heart rate variability (HRV).

  • Patients with coronary heart diseases presented higher risk.

  • The associations remained similar after the adjustments for air pollutant exposures.

  • Geomagnetic activity may account for temporal HRV-related cardiovascular outcomes.

Acknowledgements

We offer our special thanks to the NAS participants. The authors would like to thank the study participants for their dedicated participation. The authors would like to thank J.M. Wolfson and C.M. Kang for their assistance in collecting and managing data for this study.

Funding sources

The VA Normative Aging Study is supported by the Cooperative Studies Program/Epidemiology Research and Information Center of the U.S. Department of Veterans Affairs and is a component of the Massachusetts Veterans Epidemiology Research and Information Center (MAVERIC), Boston, Massachusetts. This study was also supported by NIH National Institute of Environmental Health Sciences R21-ES029637 and resources and use of facilities at the Veterans Affairs Boston Healthcare System. The views expressed in this article are those of the authors and do not reflect the position or policy of the Department of Veterans Affairs or the United States Government. This publication was made possible by U.S. EPA grant RD-835872. Its contents are solely the responsibility of the grantee and do not necessarily represent the official views of the U.S. EPA. Further, U.S. EPA does not endorse the purchase of any commercial products or services mentioned in the publication.

Abbreviations:

HRV

heart rate variability

r-MSSD

mean squared differences of successive normal-to-normal intervals

SDNN

standard deviation of normal-to-normal heart beat intervals

ECG

Electrocardiogram

CVD

cardiovascular diseases

MI

myocardial infarction

CHD

coronary heart diseases

VA

Veterans Affairs

NAS

Normative Aging Study

BMI

body mass index

PM2.5

ambient particulate matter <2.5 μm (PM2.5)

PN

ambient ultrafine particles

BC

ambient black carbon

NOAA

National Oceanic and Atmospheric Administration

SPWC

Space Weather Prediction Center

EMF

earth magnetic field

GMD

geomagnetic disturbances

Kp index

geomagnetic disturbance parameter

USEPA

United States Environmental Protection Agency

NAAQS

National Ambient Air Quality Standard (NAAQS)

IQR

interquartile range

Temp

ambient temperature

RH

relative humidity

SD

standard deviation

CI

Confident Interval

UTC

Coordinated Universal Time

ppb

parts per billion

ANS

autonomic nervous system

SCN

suprachiasmatic nucleus

#

number

GA

geomagnetic activity

Footnotes

CRediT authorship contribution statement

C.L.Z.V. and P.K. designed the study; C.L.Z.V. and K.C. conducted the statistical analysis and wrote the draft manuscript. M.L. helped edit the figures. P.V. is the NAS project PI, and has conducted study funding, data collection and data management. The manuscript was edited by C.L.Z.V., K.C., P.K., J.S. and E.G. All authors revised and approved the final version.

Declaration of competing interest

The authors declare no competing financial interests.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.scitotenv.2022.156235.

References

  1. Alabdulgader Abdullah, et al. , 2018. Long-term study of heart rate variability responses to changes in the solar and geomagnetic environment. Sci. Rep 8 (1), 2663. 10.1038/s41598-018-20932-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Baja Emmanuel S., et al. , 2013. Structural Equation Modeling of Parasympathetic and Sympathetic Response to Traffic Air Pollution in a Repeated Measures Study. 13. [DOI] [PMC free article] [PubMed]
  3. Baja et al. (n.d.), “Structural Equation Modeling of Parasympathetic and Sympathetic Response to Traffic Air Pollution in a Repeated Measures Study.”. [DOI] [PMC free article] [PubMed]
  4. Bell Rose, and Damon (n.d.), “The Normative Aging Study.”.
  5. Bell Benjamin, Rose Charles L., Damon Albert, 1972. The normative aging study: an inter-disciplinary and longitudinal study of health and aging. Int. J. Aging Hum. Dev 3 (1), 5–17. 10.2190/GGVP-XLB5-PC3N-EF0G. [DOI] [Google Scholar]
  6. Bettoni Marco, Zimmermann Marc, 2002. Autonomic tone variations before the onset of paroxysmal atrial fibrillation. Circulation 105 (23), 2753–2759. 10.1161/01.CIR.0000018443.44005.D8. [DOI] [PubMed] [Google Scholar]
  7. Boudreau Philippe, Yeh Wei Hsien, Dumont Guy A., Boivin Diane B., 2012. A circadian rhythm in heart rate variability contributes to the increased cardiac sympathovagal response to awakening in the morning. Chronobiol. Int 29 (6), 757–768. 10.3109/07420528.2012.674592. [DOI] [PubMed] [Google Scholar]
  8. Cohen MC, Rohtla KM, Lavery CE, Muller JE, Mittleman MA, 1997. Meta-analysis of the morning excess of acute myocardial infarction and sudden cardiac death. Am. J. Cardiol 79 (11), 1512–1516. 10.1016/s0002-9149(97)00181-1 Erratum in: Am J Cardiol 1998 Jan 15;81(2):260. [DOI] [PubMed] [Google Scholar]
  9. Cornélissen Germaine, et al. , March 2002. Non-photic solar associations of heart rate variability and myocardial infarction. J. Atmos. Sol. Terr. Phys 64 (5–6), 707–720. 10.1016/S1364-6826(02)00032-9. [DOI] [Google Scholar]
  10. Greiser KH, Kluttig A, Schumann B, et al. , 2005. Cardiovascular disease, risk factors and heart rate variability in the elderly general population: design and objectives of the CARdiovascular disease, living and ageing in Halle (CARLA) study. BMC Cardiovasc. Disord 5, 33. 10.1186/1471-2261-5-33. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Hasegawa H, et al. , August 2004. Transport of solar wind into Earth’s magnetosphere through rolled-up kelvin-helmholtz vortices. Nature 430 (7001), 755–758. 10.1038/nature02799. [DOI] [PubMed] [Google Scholar]
  12. Hillebrand Stefanie, et al. , 2013. Heart rate variability and first cardiovascular event in populations without known cardiovascular disease: meta-analysis and dose-response meta-regression. Europace 15 (5), 742–749. 10.1093/europace/eus341. [DOI] [PubMed] [Google Scholar]
  13. Hisako Tsuji, 1996. Impact of reduced heart rate variability on risk for cardiac events. Circulation 94 (11), 2850–2855. 10.1161/01.CIR.94.11.2850. [DOI] [PubMed] [Google Scholar]
  14. Hisako Tsuji, et al. , 1996. Impact of reduced heart rate variability on risk for cardiac events. Circulation 94 (11), 2850–2855. 10.1161/01.CIR.94.11.2850. [DOI] [PubMed] [Google Scholar]
  15. Jaruševičius Gediminas, et al. , 2018. Correlation between changes in local earth’s magnetic field and cases of acute myocardial infarction. Int. J. Environ. Res. Public Health 15 (3), 399. 10.3390/ijerph15030399. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Kang Choong-Min, Koutrakis Petros, Suh Helen H., 2010. Hourly measurements of fine particulate sulfate and carbon aerosols at the Harvard–U.S. Environmental Protection Agency supersite in Boston. J. Air Waste Manage. Assoc 60 (11), 1327–1334. 10.3155/1047-3289.60.11.1327. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Kop Willem J., et al. , 2001. Changes in heart rate and heart rate variability before ambulatory ischemic events11 the opinions and assertions expressed herein are those of the authors and are not to be construed as reflecting the views of the USUHS or the US Department of Defense. J. Am. Coll. Cardiol 38 (3), 742–749. 10.1016/S0735-1097(01)01451-6. [DOI] [PubMed] [Google Scholar]
  18. Kop Willem J., et al. , 2001. Changes in heart rate and heart rate variability before ambulatory ischemic events11 the opinions and assertions expressed herein are those of the authors and are not to be construed as reflecting the views of the USUHS or the US Department of Defense. J. Am. Coll. Cardiol 38 (3), 742–749. 10.1016/S0735-1097(01)01451-6. [DOI] [PubMed] [Google Scholar]
  19. Krylov Viacheslav V., et al. , 2014. An experimental study of the biological effects of geomagnetic disturbances: the impact of a typical geomagnetic storm and its constituents on plants and animals. J. Atmos. Sol. Terr. Phys 110–111, 28–36. 10.1016/j.jastp.2014.01.020. [DOI] [Google Scholar]
  20. Luttmann-Gibson H, et al. , 2010. Systemic inflammation, heart rate variability and air pollution in a cohort of senior adults. J. Occup. Environ. Med 67 (9), 625–630. 10.1136/oem.2009.050625. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. McCraty Rollin, et al. , 2017. Synchronization of human autonomic nervous system rhythms with geomagnetic activity in human subjects. Int. J. Environ. Res. Public Health 14 (7), 770. 10.3390/ijerph14070770. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. McCraty et al. (n.d.), “Synchronization of Human Autonomic Nervous System Rhythms with Geomagnetic Activity in Human Subjects.”. [DOI] [PMC free article] [PubMed]
  23. Mendoza Blanca, Diaz-Sandoval Rosa, 2004. Effects of solar activity on myocardial infarction deaths in low geomagnetic latitude regions. Nat. Hazards (Dordrecht) 32 (1), 25–36. 10.1023/B:NHAZ.0000026789.71030.31. [DOI] [Google Scholar]
  24. Mordukhovich Irina, et al. , 2015. Exposure to sub-chronic and long-term particulate air pollution and heart rate variability in an elderly cohort: the normative aging study. Environ. Health 14 (1), 87. 10.1186/s12940-015-0074-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Mordukhovich et al. (n.d.), “Exposure to Sub-Chronic and Long-Term Particulate Air Pollution and Heart Rate Variability in an Elderly Cohort.”. [DOI] [PMC free article] [PubMed]
  26. NOAA (n.d.). Kp index. Site: https://www.swpc.noaa.gov/products/planetary-k-index.
  27. Nyhan Marguerite M., et al. , 2018. Associations between ambient particle radioactivity and blood pressure: the NAS (Normative Aging Study). J. Am. Heart Assoc 7 (6). 10.1161/JAHA.117.008245. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Odemuyiwa Olusola, et al. , 1991. Comparison of the predictive characteristics of heart rate variability index and left ventricular ejection fraction for all-cause mortality, arrhythmic events and sudden death after acute myocardial infarction. Am. J. Cardiol 68 (5), 434–439. 10.1016/0002-9149(91)90774-F. [DOI] [PubMed] [Google Scholar]
  29. Palmer SJ, Rycroft MJ, Cermack M, 2006. Solar and geomagnetic activity, extremely low frequency magnetic and electric fields and human health at the earth’s surface. Surv. Geophys 27 (5), 557–595. 10.1007/s10712-006-9010-7. [DOI] [Google Scholar]
  30. Park Sung Kyun, et al. , 2005. Effects of air pollution on heart rate variability: the VA normative aging study. Environ. Health Perspect 113 (3), 304–309. 10.1289/ehp.7447. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Park et al. (n.d.), “Effects of Air Pollution on Heart Rate Variability.”.
  32. Schwartz Joel, Samet Jonathan M., Patz Jonathan A., 2004. Hospital admissions for heart disease: the effects of temperature and humidity. Epidemiology 15 (6), 755–761. 10.1097/01.ede.0000134875.15919.0f. [DOI] [PubMed] [Google Scholar]
  33. Sessa Francesco, et al. , 2018. Heart rate variability as predictive factor for sudden cardiac death. Aging (Albany NY) 10 (2), 166–177. 10.18632/aging.101386. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Shaffer Fred, Ginsberg JP, 2017. An overview of heart rate variability metrics and norms. Front. Public Health 5, 258. 10.3389/fpubh.2017.00258. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Sidney Stephen, et al. , 2016. Recent trends in cardiovascular mortality in the United States and public health goals. JAMA Cardiol 1 (5), 594. 10.1001/jamacardio.2016.1326. [DOI] [PubMed] [Google Scholar]
  36. Vencloviene Jone, et al. , 2014. The effect of solar-geomagnetic activity during and after admission on survival in patients with acute coronary syndromes. Int. J. Biometeorol 58 (6), 1295–1303. 10.1007/s00484-013-0725-0. [DOI] [PubMed] [Google Scholar]
  37. Vencloviene Jone, et al. , 2014. The effect of solar-geomagnetic activity during and after admission on survival in patients with acute coronary syndromes. Int. J. Biometeorol 58 (6), 1295–1303. 10.1007/s00484-013-0725-0. [DOI] [PubMed] [Google Scholar]
  38. Vieira CLZ, Link M, Garshick E, Mark S. Link, et al. Solar and Geomagnetic Activity Enhance the Effects of Air Pollutants on Atrial Fibrillation. EUROPACE. Just accepted(n.d.). [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Vieira CLZ, et al. , 2019. Geomagnetic disturbances driven by solar activity enhance total and cardiovascular mortality risk in 263 U.S. cities. Environ. Health 18 (1), 83. 10.1186/s12940-019-0516-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Wilmshurst Peter, 1994. Temperature and cardiovascular mortality: excess deaths from heart disease and stroke in northern Europe are due in part to the Cold. Br. Med. J 309 (6961), 1029–1030. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Zanobetti Antonella, et al. , 2010. Reduction in heart rate variability with traffic and air pollution in patients with coronary artery disease. Environ. Health Perspect 118 (3), 324–330. 10.1289/ehp.0901003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Zanobetti et al. (n.d.), “Reduction in Heart Rate Variability with Traffic and Air Pollution in Patients with Coronary Artery Disease.”. [DOI] [PMC free article] [PubMed]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplement
NIHMS1812569-supplement-Supplement.docx (392.4KB, docx)

RESOURCES