Heart Rate Variations During Two Historic Geomagnetic Storms: October and November 2003

Abstract

The investigation of the two phenomena of Space Weather, i.e., Forbush decreases in the cosmic ray intensity and geomagnetic storms, is a highly developing field of modern scientific research, since these two phenomena can affect not only technological activities, e.g., electronics, telecommunications, navigations, etc., but also, as evidenced by recent studies, human life as well. This study analyses data of heart rate of volunteers of the Polyclinico Tor Vergata, Rome, Italy, with regard to geomagnetic field’s variations (i.e., geomagnetic storms) and cosmic ray intensity’s fluctuations (i.e., Forbush decreases). Data concerning geomagnetic (Dst- and Ap-index values) and cosmic ray activity derived from the Rome Cosmic Ray Station (Studio Variazioni Intensità Raggi Cosmici: S.V.I.R.CO.) were analyzed. The analysis expands from 24 April 2003 to 12 May 2004 and includes October–November 2003, which was a period of severe activity, when extreme events were recorded (i.e., the Great Halloween Solar Storms and the super storm on November 2003). The variations in heart rate were studied using the ANalysis Of Variance—ANOVA (for various levels of activity of the geophysical environment) and the superimposed epochs methods (during an event’s temporal evolution). Results revealed that high geomagnetic (defined by Dst-index values) and cosmic rays activity is related to heart rate increase. Moreover, the most significant heart rate variations are observed two days before until two days after the development of an event (either geomagnetic storm or a variation in the cosmic ray intensity). The results are in agreement with conclusions presented in the international scientific literature.

1. Introduction

In the early 1920s, the term heliobiology was introduced in order to describe the solar activity’s impact on human health [1,2,3,4,5,6,7]. Nowadays, heliobiological studies are considered to be an important field of contemporary science that is centered around the impact of the geophysical environment’s variations on the human physiological state and particularly on the mortality/morbidity from various diseases, the nervous system’s functional activity, the appearance and evolution of epidemics, and the recording of other psycho-physiological complaints [8,9,10,11,12,13,14,15,16,17,18,19,20,21,22].

For example, the possible effect of the helio-geophysical factors (such as geomagnetic storms, solar proton events, interplanetary coronal mass ejections, etc.) on the number of acute myocardial infarction incidents and the number of ischemic heart disease deaths for the years 2000–2015 in Kaunas, Lithuania was studied in [23]. Amongst others, results revealed that three days before and a day after a geomagnetic storm the possibility of acute myocardial infarction and ischemic heart disease death was higher.

Geronikolou [24] presented an investigation for the time period from 1985 to 1989 on the probable influence of solar activity on vascular stroke mortality. The number of stroke deaths for the city of Piraeus was analyzed in relation to the total solar irradiance by using principal components analysis, regressions and neural networks.

In a different approach, skin conductance, electromyography and the share of abdominal and diaphragmatic breathing in overall ventilation are important psychophysiological parameters which were evaluated in regard to solar activity in [25]. As it was concluded, the solar wind (i.e., electrically charged particles originating from the Sun) which reaches the Earth affect the psychophysiological parameters of the body.

Furthermore, recent studies have examined the relationship between the variations in the geomagnetic activity (GMA) and the heart rate (HR) dynamics [7] and the autonomous nervous system’s self-regulation [26] through HR variability. More specifically, Zenchenko [7] have analyzed the HR data of two women volunteers from 2012 to 2023 and the contemporaneous variations in the geomagnetic field (observed in the horizontal components). Firstly, the correlation between geomagnetic field’s variations and the minutely HR data and secondly, the co-occurrence of the main periods of oscillations (ranging from 3 to 50 min) are evidence of the biogeophysical synchronization effect. This effect was also studied in [26] using the electrocardiogram recordings of 100 min from two female volunteers in good health. Therein, the geomagnetic field’s variations (of the ULF frequency range, i.e., 1–5 mHz) effect of the human HR synchronization was studied for the time periods 24–27 September 2023, 10–13 May 2024 and 10–13 October 2024. In total, 69% of cases showed signs of the biogeosynchronization effect.

The effect of the different phases of a geomagnetic storm on the autonomous nervous system for a group of healthy males was investigated in [27]. Therein, HR data of 61 volunteers from Tbilisi, Georgia for the year 2022 were analyzed during the initial, the main and the restoration phase of geomagnetic storms. It was concluded that the autonomous nervous system presented stress reactions during the main and the restoration phases of geomagnetic storms, making geomagnetic activity an important physical factor for healthy males.

In a more recent study, He [28] analyzed more than 500,000 measurements of arterial blood pressure obtained from two different hospitals located in the cities Qingdao and Weihai in China in regard to GMA (expressed by the geomagnetic index Ap). This analysis, which covered the time period from January 2015 until December 2020, concluded that there is a correlation between arterial blood pressure fluctuations and the GMA and more precisely similar patterns and periodicities are evident in both data series.

In this work HR data obtained during the medical assessment of volunteers in the Polyclinico Tor Vergata, Rome, Italy were evaluated during periods of GMA variations (i.e., geomagnetic storms) and CRI fluctuations (i.e., Forbush decreases). The analysis expands from 24 April 2003 until 12 May 2004, when severe physical events occurred. The HR data were acquired with the method of Holter Electrocardiogram, while geomagnetic indices (Dst-index and Ap-index) and CRI (Rome Cosmic Ray Station/Studio Variazioni Intensità Raggi Cosmici: S.V.I.R.CO.) data were processed as well. The HR variations for various levels of activity of the geophysical environment but also during an event’s temporal evolution were studied by applying the ANalysis Of Variance (ANOVA) and the superimposed epochs methods.

2. Data and Method

2.1. Medical Data

A number of 2047 daily mean values of HR (beats per minute, bpm), recorded during the medical assessment of volunteers (men and women aged from 5 to 96 years old) in the Polyclinico Tor Vergata, Rome, Italy was studied. The analysis covers the time period from 24 April 2003 until 12 May 2004. The Polyclinico Tor Vergata provided the mean HR for every different volunteer, i.e., for the same day the clinic may have treated more than one person. The data were obtained using a Holter Electrocardiogram [29].

For the 786 women in the dataset the average HR was 73.95 ± 11.07 bpm and for the 758 men the average HR was 70.4 ± 10.9 bpm (for 503 cases information about the gender was not provided).

It should be noted that all volunteers provided information about their general medical and psycho-physiological state before obtaining any data. Therefore, a complete database was created, including not only information on physiological parameters (HR, RR intervals, arterial blood pressure, etc.) but also about medication (beta-blockers or others medicine), disturbances (arrhythmias, vertigo, tachycardia, etc.) and other information about any type of discomfort. These data are of great importance as they will allow a future multi-factor analysis so as to determine the reaction of each of the above mentioned parameters to the variations in the physical environment.

2.2. Cosmic Ray Intensity Data

The Rome Cosmic Ray Station (Studio Variazioni Intensità Raggi Cosmici: S.V.I.R.CO.) is one of the fundamental members of the European High-resolution Neutron Monitor Database (NMBD; http://www.nmdb.eu, accessed on 5 April 2025) [30]. Its 20 BF3 proportional counters (BP-28 type) qualifies SVIRCO (Effective vertical cutoff rigidity 6.27 GV) to study primary cosmic rays and their modulation in the heliosphere. Since 1997, the SVIRCO Observatory (INAF/IFSI-UNIRomaTre collaboration) was relocated in the Department of Physics ‘E. Amaldi’ of ‘Roma Tre’ University (41.86° N, 12.47° E, about sea level) and has been continually operating.

Daily CRI data (counts/sec) corrected for pressure and efficiency were acquired from the Rome Cosmic Ray Station (http://www.nmdb.eu, accessed on 5 April 2025). CRI was normalized according to the $\mathrm{CRI}={\displaystyle \frac{{CRI}_{obs}-CR{I}_{aver}}{CR{I}_{aver}}}$, where $CR{I}_{obs}$ is the observed CRI value and $CR{I}_{aver}$ is the average value that is calculated for a quiet period during the time span of the analysis. $CR{I}_{aver}$ represents the quiet-time background CRI and when decreases are included in the time period under study, $CR{I}_{aver}$ is calculated after excluding data corresponding to the FD onset, minimum, and recovery phase (e.g., [31,32,33]).

CRI variations span from −17% decreases up to +3% increases. The CRI was organized into five levels of intensity (

Table 1. CRI levels and the corresponding number of measurements for all the volunteers.

Intensity	CRI Levels	CRI, %	Number of Measurements
Severe decreases	−3	−17 ≤ CRI ≤ −11	33
Major decreases	−2	−11 < CRI ≤ −6	114
Moderate decreases	−1	−6 < CRI ≤ −1	1274
Quiet	0	CRI = 0	233
CRI increases	1	1 ≤ CRI ≤ 3	393

), according to the CRI % values.

2.3. Geomagnetic Activity Data

The World Data Centre for Geomagnetism, Kyoto (https://wdc.kugi.kyoto-u.ac.jp/dstae/index.html, accessed on 5 April 2025) and the German Research Center for Geosciences, GFZ (https://www.gfzpotsdam.de/en/section/geomagnetism/dataproductsservices/geomagnetic-kp-index, accessed on 5 April 2025) provided the Dst- and Ap- geomagnetic indices values that were used in this analysis.

GMA was classified into 5 different levels, as shown in

Table 2. Dst-index levels and the corresponding number of measurements for all the volunteers.

Activity	Dst/Ap Levels	Dst-index Values (nT)	Number of Measurements	Ap-Index Values	Number of Measurements
Quiet	I0	Dst ≥ 0	212	Ap < 8	375
Minor	I	−20 < Dst < 0	779	8 ≤ Ap < 15	459
Moderate	II	−50 < Dst ≤ −20	891	15 ≤ Ap < 30	695
Strong	III	−100 < Dst ≤ −50	105	30 ≤ Ap < 50	367
Severe	IV	Dst ≤ −100	60	Ap ≥ 50	151

, based on the Dst- and Ap-index daily values.

2.4. Statistical Methods

The methods of Analysis of Variance (ANOVA) [34] and the Chree analysis, i.e., the superimposed epochs method [35] were applied in order to analyze the physical activity in regard to HR data. The one way ANOVA method defines the dependent variable, i.e., HR, and the factors or independent variables, i.e., Dst, Ap and CRI levels and evaluates the impact of the different levels of physical activity on HR.

In order to do that, the ANOVA method assumes that the dependent variable HR’s variation is not affected significantly by the factors Dst, Ap and CRI levels. In other words, the mean values of HR at every geophysical level have no difference between them that is statistically significant, i.e., the null hypothesis. In case such disagreements do exist, the null hypothesis is dismissed and the alternative hypothesis is valid, i.e., the physiological parameter HR is affected by the Dst, Ap and CRI levels.

Whether the null or the alternative hypothesis is valid will be determined by the statistical significance level, which is set by the software system at p < 0.05. As a result, for p-level values < 0.05 the alternative hypothesis is valid and the null hypothesis is rejected and vice versa for p-level values greater than 0.05.

For examining whether, for the sample under investigation, the HR, defined as the dependent variable, is normally distributed and therefore the ANOVA method can be performed, the Shapiro–Wilk test of normality was applied. It is proven that the ANOVA method can be employed since the sample does not show any signs of non-normality (W = 0.993620, p = 0.000000).

In order to study the geophysical activity’s effect on HR from three days before until three days after an event besides the ANOVA method, the superimposed epoch method or the Chree analysis was also applied.

This statistical analysis defines the day that HR data were obtained as Day 0. Geophysical activity data (Dst- and Ap-index values and CRI data) are also obtained for day 0 and consequently the level of activity for this day is determined (see Table 1, Table 2 and

Table 3. Dates with the highest GMA (minimum Dst-index and maximum Ap-index values) and the highest cosmic ray’s activity (minimum CRI values, Forbush decreases).

Date	Dst-Index Values (nT)	CRI, %	Ap-Index Values
20 November 2003	−156	−3	150
21 November 2003	−140	−4	42
29 October 2003	−128	−11	204
30 October 2003	−221	−16	191
31 October 2003	−117	−17	116

). Days −3, −2 and −1 and +1, +2 and +3 before and after the event, respectively, designate the initial and the restoration phase, because geophysical events fully develop over a period of several days [36,37], even though their main phase lasts for some hours. In this way the HR’s temporal distribution during the development of a geomagnetic storm or a CRI decrease is examined, i.e., preparation of the physical environment conditions, main event and restoration of the physical environment conditions [38].

3. Results

The time interval under investigation is part of the declining phase of solar cycle 23, one of longest cycles since 1847, which lasted from May 1996 to December 2008 [39]. The maximum of Solar cycle 23 was reached between 2000 and 2002 [40]. Even though during the Space Age (1957–today) the strongest geomagnetic storm was registered on March 13–14, 1989 (daily Dst = −225 nT and Ap = 246), the events of October (daily Dst = −221 nT and Ap = 204) and November 2003 (daily Dst = −156 nT and Ap = 150) are also included in the list of most severe decreases in the CRI (level −3) and geomagnetic storms [41]. The storms from late October to early November are known as the Great Halloween Solar Storms, while the geomagnetic storm in the mid-November was characterized as a super storm.

The Great Halloween Solar Storms (level G5 according to the NOAA Space Weather Scales) were associated with three active regions of particular size and complexity, i.e., AR484, AR486, and AR488. These ARs were responsible for the great solar flares associated with high-speed CMEs and intense energetic particle events registered from October 19 to November 5. In total, 17 major (>R2) solar flares, six radiation storms (>S1) and four severe (>G2) geomagnetic storms occurred during a 20-day period.

In particular, 12 (out of 17) great solar flares that were registered during this time period were associated with AR486 and the X17 on 28 October, the X10 on 29 October and the X28 on 4 November are the ones that stand out.

The technological impact of these events was enormous. Damages and malfunctions were reported in over half of the spacecrafts orbiting around the Earth, while communications and navigation systems were severely affected. Airline flights connecting North America to Asia through the North Pole were disrupted and in order to protect the astronauts aboard the ISS from the increased radiation specific measures, which had only been taken twice in the past, were enforced. Moreover for the geomagnetic storms under consideration, Lotz [42] and Švanda [43] studied the strong effects on power lines and transformers at middle latitudes (in South Africa and in the Czech Republic, respectively) one day to several days after the beginning of geomagnetic storms due to geomagnetically induced currents during and after geomagnetic storms. Despite all the above, the 2003 Great Halloween Solar Storms events also rewarded the public with a widespread worldwide view of auroras. On 29–30 October, regions of middle or even low latitude enjoyed this unique manifestation of Space Weather. California to Texas and Florida and Australia to central Europe and even as far south in countries of the Mediterranean Sea, had all the opportunity to enjoy the marvelous displays of aurora (https://www.ncei.noaa.gov/news/great-halloween-solar-storm-2003, accessed on 8 April 2025).

Regarding the G5 geomagnetic storm on 20 November 2003, Ebihara [44] list it as the second since 1957 (geomagnetic storm on 13–14 March 1989), in relation to the values of the Dst-index (Dst- and Kp-index were −472 nT and 9⁻, respectively). Sunspot groups 501 and 508 were highly active on 18 November 2003. A M4 solar flare, from group 508, registered at the eastern limb was associated with a powerful CME, while two long-lasting M solar flares (i.e., M3.2 and M3.9), from group 501, registered at the center of the disk were also associated with CMEs. The two central flares and maybe the third are responsible for the super storm on 20 November 2003.

Table 3 includes these days along with the corresponding Dst-index, Ap-index (daily mean values) and the CRI values, while Figure 1 shows the daily variations in the Dst-index (red dashed line), the Ap-index (green continuous line) and the CRI (blue punctuated line) from 1 January 2003 until 31 December 2004. The time period 29–31 October 2003 is marked in the circle.

As already mentioned, ANOVA method was used to estimate the significance levels (p) of the impact of the various GMA and CRI levels on HR during geomagnetic storms and CRI variations, but also for the days preceding and following these events.

Table 4. Potential effect of the GMA and CRI on HR on the days before (−), during (0) and after (+) GMA and CRI variations, as described by the significance levels (p-values).

Days	Dst	Ap	CRI
−3	0.36972	0.56033	0.01596 *
−2	0.43717	0.42622	0.14030
−1	0.74530	0.23454	0.05954
0	0.83491	0.79509	0.01172 *
+1	0.97541	0.01415 *	0.02898 *
+2	0.90324	0.64736	0.53414
+3	0.81379	0.58817	0.06004

* Statistical significant results.

includes these results for the three factors under investigation for all volunteers. As it can be seen, results that are statistically significant were obtained for all volunteers for Ap-index on day +1st and CRI for days −3rd, 0 and +1st. More specifically, for the aforementioned days both for Ap-index and CRI, the alternative hypothesis, i.e., the physiological parameter is affected by the physical activity levels, is valid. For the geomagnetic Dst-index no statistically significant results were acquired for none of the days.

The HR trend for the various intensities of GMA, i.e., Dst- and Ap-index and different variations in CRI for Day 0 are shown in Figure 2, Figure 3 and Figure 4, respectively. The 0.95 confidence interval is denoted by vertical bars in these figures. It is interesting to comment that even though for Dst-index results are expected and in accordance with previous conclusions [38,45,46] results regarding Ap-index and CRI are quite divergent. Moreover, it should be mentioned that the HR behavior presented in Figure 2, Figure 3 and Figure 4 is not a definite behavior but merely a trend. Although these HR trends are in agreement with findings already introduced in the international literature, it is recommended to evaluate them conservatively, mainly because statistically significant results for GMA and CRI variations were obtained only for certain days for Ap and CRI (see Table 4).

It is stated that the highest GMA levels, for example, levels III and IV of the Dst- and Ap-index categorization are related to greater HR values. In addition, the lowest GMA levels, i.e., levels I0, I and II of the Dst- and Ap-index categorization are related to small variations or even no variations at all [29,46,47,48]. This trend is noticed in Figure 2, where for severe geomagnetic storms (Dst-index level IV) the HR takes its maximum value.

On the other hand, for Ap-index (Figure 3) HR takes its maximum value for moderate geomagnetic storms (level II) and not during severe GMA as expected. Also, a decrease is recorded from level II to IV, when HR minimum is recorded.

Moreover, results for CRI levels are presented in Figure 4. As already seen in Table 4, HR is affected by the CRI levels for Day 0 (statistically significant results). For severe and major CRI decreases (levels −3 and −2 of the CRI, % classification) high HR values are observed. In general HR’s behavior from level −3 to level 0 is consistent with previous results [29,46,47]. However, there is a notable HR increase for level 1 of the CRI, % classification (i.e., CRI increases) to values comparable to the ones of severe and major CRI decreases.

Figure 5, Figure 6 and Figure 7 depict the HR dynamics during the development of a geomagnetic storm or CRI variations. In these diagrams days −3rd, −2nd and −1st refer to the initial phase of the geophysical event, day 0 defines the day of the main event and days +1st, +2nd and +3rd after refer to the restoration phase of the geophysical event. The various colored lines in these diagrams represent the different levels of geophysical activity.

Figure 5 shows the variations in HR in relation to GMA, expressed by the Dst-index. For levels I0, I and II (blue, red and green line, respectively) of GMA (i.e., low GMA) no significant variation was noticed. On the contrary, peak values of HR before and after the event are observed for the two highest levels of GMA, which is in accordance with prior findings [38,45,46]. HR increased on the days before the event and decreased during the event and until 2 days after the event and increased again on day +3rd. More precisely, HR was maximized on day −2nd before the event and decreased from there on until day +2nd after the event (for level III—pink line). Similar behavior is also noticed for level IV (black line). The only difference in behavior between the two higher levels is that HR maximum value has one day delay for level IV (maximum on day −1st).

On the other hand, Ap-index results are not consistent with Dst-index findings or previous results. As seen in Figure 6, all levels of GMA present HR variations and peak values before or after a geomagnetic storm. For levels I (red line) and III (pink line) a similar behavior is noticed from day –1st before the event until day +2nd after the event (decrease–increase–decrease). The opposite trend (increase from day −1st to day 0, decrease till day +1st and increase from there on until day +3rd) is observed for level II (green line). For this level HR had its maximum value on Day 0 as was also commented in Figure 4. HR decreased from day −3rd until day +1st after the geomagnetic storm and increased from there on to values similar to that of day’s −3rd (for level IV). Particularly on day +1st, the HR decrease for Ap level IV is significant compared to level I0.

Variations HR for the various levels of cosmic rays activity are seen in Figure 7. For level −1 (moderate decreases, green line) and +1 (increases in CRI, black line) no significant variations were registered. For the highest level −3 of the cosmic rays activity (blue line) a decrease in HR is noticed from a maximum to a minimum value from day −3rd to day +3rd. For level −2 (red line), HR increased from day –1st to day 0 and decreased from there on until day +2nd. HR’s behavior for level 0 (no CRI variations) is also worth mentioning, since an obvious decrease from day −2nd until minimum value on day 0 and increase from there on until day +2nd is registered. This behavior is in agreement with results already mentioned in the literature that state that minor or even no variations in the physical environment can have an impact on human physiological parameters [49,50]. In general, it can be argued that HR’s significant correlation to CRI is apparently very weak.

In general it can be argued that HR variations are not only more evident for high levels of physical activity but are also more pronounced in the time period from two days before until two days after the events under consideration. For days −3rd and +3rd the physiological parameter of HR behaved differently according to the level of activity. Moreover, for low levels of physical activity (Dst-index and CRI) HR on day +3rd returned near its value registered on day −3rd, but for the highest levels of activity this was not always the case.

4. Conclusions and Discussion

Over the years many studies have been conducted addressing the issue of the possible effect of Space Weather phenomena (e.g., geomagnetic storms and Forbush decreases) on human physiological parameters variations as well as the frequency of myocardial infractions, strokes, traffic accidents, etc. [51,52,53,54,55,56]. Results reported in the international literature show human physiology is sensitive to variations in the geomagnetic field and the CRI. It should be mentioned that [51] in an attempt to summarize his results, extended the National Oceanic and Atmospheric Administration (NOAA) scale for geomagnetic storms. Thus, for strong geomagnetic storms, this scale includes not only their effects on the operation and reliability of various technological systems, but also their corresponding biological effects.

This work examined the relationship between HR variations and GMA and CRI variations. In total 2047 daily mean values of HR of healthy volunteers from the Polyclinico Tor Vergata, Rome, Italy, were accessed in regard to indices Dst and Ap and variations in CRI. The analysis expanded from 24 April 2003 to 12 May 2004. The most interesting results are summarized in the following.

From the analysis of the physiological parameter HR variations in regard to different levels of geomagnetic and cosmic rays’ activity it was concluded that:

(1)

Concerning the p-values, results statistically significant were obtained for geomagnetic Ap-index one day after and for CRI on days before, during and after the development of physical events, showing that physical activity levels affect the physiological parameter HR. For the geomagnetic Dst-index no statistically significant results were obtained for the days under study.

(2)

For the geomagnetic index Dst, strong and severe geomagnetic storms (levels III and IV, respectively) were related to HR increase.

(3)

For the geomagnetic index Ap, strong and severe geomagnetic storms (levels III and IV) were connected to low HR values.

(4)

For the CRI levels, severe and major CRI decreases (levels −3 and −2) are related to high HR values.

(5)

For level 1 of the CRI, % classification (i.e., CRI increases) there is a notable HR increase to values comparable to the ones of severe and major CRI decreases.

Heart rate variations were examined during the development of a physical event, i.e., three days before (days −3rd, −2nd and −1st), the day during (day 0) and three days after (days +1st, +2nd and +3rd) the event and it was concluded that:

(6)

Regarding the geomagnetic index Dst, for levels I0, I and II (i.e., low GMA) no notable variations in HR were recorded. HR presented peak values before and after the event for levels III and IV.

(7)

Regarding the geomagnetic index Ap, all GMA levels present variations in HR and peak values before or after a geomagnetic storm. Moreover, for day +1st, the HR decrease for Ap level IV is significant compared to level I0.

(8)

Regarding CRI, for levels −1 (moderate decreases) and +1 (increases in CRI) no significant variations were registered. For the highest levels −3 and −2 of the cosmic rays activity, on the days before and after the event HR presented peak values. Generally, it can be stated that HR is only marginally modified for CRI levels −3 to +1, showing that the significant correlation of HR with CRI is apparently very weak.

(9)

Concerning CRI, for level 0 (no CRI variations) HR’s behavior is also noticeable.

(10)

The most significant HR variations for high levels of physical activity were noticed mainly for the time period from two days before until 2 days after the events under consideration.

Different investigations support the last statement, i.e., the physical activity can affect the human physiological state in a time period from almost 2 days before the physical event’s onset and 2 days after the manifestation of the main event [57,58,59,60,61,62]. This is possibly due to the fact that most geophysical events usually follow precursory signs. These precursors may be responsible for provoking regulating adjustments of the human physiological parameters a few days before the main phase of the geophysical event.

For example, the increase in the arterial blood pressure from day −2nd before until day +2nd after moderate, major and severe geomagnetic storms was discussed in [45]. In another study by [63] the most significant variations in arterial blood pressure during geomagnetic storms were reported in women and were observed 2 days before and 1 day after the storm (mainly for systolic pressure). Furthermore, Vaičiulis [23] showed that the probability of acute myocardial infarction and death due to ischemic heart disease increases 3 days before and 1 day after a geomagnetic storm. Moreover, Vencloviene [64] showed that geomagnetic storms occurring within 2 days of hospital admission increased the risk of cardiovascular death by 1.58 times. For women, geomagnetic storms occurring within 1 day of hospital admission increased the risk of cardiovascular death by almost 4 times.

Even though the majority of findings of this work agree with results from previous studies, there are some points that diverge from the expected and need further research. Since this specific topic remains open for the scientific community, thorough studies are needed to arrive at clearer and more reliable conclusions.