Helio-geomagnetic influence in cardiological cases

Available online at www.sciencedirect.com Advances in Space Research 51 (2013) 96–106 www.elsevier.com/locate/asr Helio-geomagnetic influence in cardiological cases Ch. Katsavrias a,⇑, P. Preka-Papadema a, X. Moussas a, Th. Apostolou b, A. Theodoropoulou a, Th. Papadima c a Department of Astrophysics, Astronomy and Mechanism, Faculty of Physics, University of Athens, Panepistimiopolis Zografos 157 71, Greece b St. Panteleimon General Hospital of Nicaea (Piraeus), 3 Mantouvalou, str. 184 54, Greece c Evagelismos’ General Hospital of Athens, 45-47 Ipsilantou str. 106 76, Greece Received 27 July 2011; received in revised form 20 July 2012; accepted 27 July 2012 Available online 6 August 2012 Abstract The effects of the energetic phenomena of the Sun, flares and coronal mass ejections (CMEs) on the Earth’s ionosphere–magnetosphere, through the solar wind, are the sources of the geomagnetic disturbances and storms collectively known as Space Weather. The research on the influence of Space Weather on biological and physiological systems is open. In this work we study the Space Weather impact on Acute Coronary Syndromes (ACS) distinguishing between ST-segment elevation acute coronary syndromes (STE–ACS) and non-ST-segment elevation acute coronary syndromes (NSTE–ACS) cases. We compare detailed patient records from the 2nd Cardiologic Department of the General Hospital of Nicaea (Piraeus, Greece) with characteristics of geomagnetic storms (DST ), solar wind speed and statistics of flares and CMEs which cover the entire solar cycle 23 (1997–2007). Our results indicate a relationship of ACS to helio-geomagnetic activity as the maximum of the ACS cases follows closely the maximum of the solar cycle. Furthermore, within very active periods, the ratio NSTE–ACS to STE–ACS, which is almost constant during periods of low to medium activity, changes favouring the NSTE–ACS. Most of the ACS cases exhibit a high degree of association with the recovery phase of the geomagnetic storms; a smaller, yet significant, part was found associated with periods of fast solar wind without a storm. Ó 2012 COSPAR. Published by Elsevier Ltd. All rights reserved. Keywords: Space weather; Magnetic storms; Substorms; Sunspots 1. Introduction The energetic events of the Sun (flares and coronal mass ejections (CMEs)) release radiation, energetic particles and plasma in the interplanetary space; when they are earthbound, they affect the terrestrial environment. The energetic events induce rapid changes of the solar wind parameters; the Solar Wind speed can increase from its 400 Km/s quiet-period near-Earth mean value, to 800 Km/s which is typical of perturbed periods. This high-speed solar wind triggers magnetic and ionospheric storms; major triggering, however, requires also a southward oriented interplanetary magnetic field, which facilitates reconnection of the solar wind magnetic field lines with the geomagnetic field. These magnetospheric disturbances are known as Space Weather (see Kivelson et al., 1995 for a review). The NASA definition of Space Weather1 is: “ The conditions and processes occurring in space which have the potential to affect the near Earth environment. Space Weather processes can include changes in the interplanetary magnetic field, coronal mass ejections from the Sun and disturbances in Earth’s magnetic field. The effects of space weather can range from damage to satellites to disruption of power grids on Earth”. The magnetospheric response to Space Weather is quantified by geomagnetic indices. In this work, we use the DST index, which is an estimate of the magnitude of the geomagnetic disturbance based on the change of earth’s ring current (e.g. Kivelson et al., 1995). For values of DST less than 50 nT we have a geomagnetic storm or substorm. ⇑ Corresponding author. Tel.: +30 2107276855. E-mail addresses: ckatsavrias@phys.uoa.gr (Ch. Katsavrias), ppreka@ phys.uoa.gr (P. Preka-Papadema), xmoussas@phys.uoa.gr (X. Moussas). 1 Space Weather Centre/NASA. 0273-1177/$36.00 Ó 2012 COSPAR. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.asr.2012.07.030 Ch. Katsavrias et al. / Advances in Space Research 51 (2013) 96–106 97 Fig. 1. Top to bottom: annual distribution of ACS (ST and NST), CMEs and partial-halo CMEs, flares and intense flares (X–M flares) as well as B flares (three upper panels). Annual number of storms, their total annual duration, the annual DST maximum value and their product (four lower panels). The influence of Space Weather on biological and physiological systems is important. The geomagnetic field variations (from helio-geomagnetic disturbances) seem to affect, directly or indirectly, the human physiology and health (Palmer et al., 2006); this has been an open research objective for the last three decades. Possible mechanisms linking solar and geophysical parameters to human health have been proposed (Cherry, 2002). The study of the effects 98 Ch. Katsavrias et al. / Advances in Space Research 51 (2013) 96–106 Fig. 2. Monthly number of ACS–NST (a) and ST (b), sunspots (c), flares (d), X–M flares (e), B flares (f), CMEs (g), P–H CMEs (h) and B–C flares (j). The grey curves represent Savitzky–Golay smoothing with a window of 12 points. Table 1 Pearson’s product moment correlation coefficients. Parameters Coefficients Comments NST–CME’s NST–XM flares ST–B flares NST–CME’s NST–XM flares ST–B flares 0:79  0:00 0:66  21016 0:66  41016 0:86  0:00 0:83  0:00 0:77  0:00 Monthly Correlation coefficients (12 months’ time lag) Monthly Correlation coefficients (18 months’ time lag) of helio-geomagnetic disturbances on health includes the statistical analysis of a plethora of data sets; the results corroborate the association between them:  Palmer et al. (2006) report that 75% of geomagnetic storms are followed by an increase by 50% of hospital cardiological and neurological cases.  Breus et al. (1989) point to a correlation between heart attacks in Moscow and helio-geomagnetic activity.  Cornélissen et al. (2002) indicate that death-rate due to heart attacks increases by 5% in Minnesota USA at the maximum of the solar cycle.  Stoupel et al. (2005) draw attention to the relationship between the death-rate (especially acute myocardial infractions (AMI)) and space weather. Along the same line (Stoupel et al., 2007) show that the monthly rates of AMI (1983–1999 and 2003–2005) are correlated with cosmic ray activity; the latter is anti-correlated to solar sunspot activity. These results are corroborated by Chernouss et al. (2001) and Belov et al., 1998 which present the influence of space weather on the neurological system and brain disruptions.  Specific studies in Israel (Stoupel et al., 1995), Italy (Gavryuseva and Kroussanova, 2002), Bulgaria (Dimitrova, 2006), Mexico and Cuba (Mendoza and Diaz-Sandoval, 2004) provide evidence in support of the Space Weather–Health relation.  Stoupel et al. (1995) as well as Dorman et al. (2001) report increased accident rate due to helio-geomagnetic activity. Ch. Katsavrias et al. / Advances in Space Research 51 (2013) 96–106 99 Fig. 3. Correlation coefficients with 12 months’ time lag between smoothed monthly values of: NST cases and CMEs (a), NST cases and X–M flares (b), ST cases and B flares (c). Correlation coefficients with 18 months’ time lag between smoothed monthly values of: NST cases and CMEs (d), NST cases and X–M flares (e), ST cases and B flares (f). The dash ellipse represents the confidence level which is above 99%. This work examines the helio-geomagnetic activity–health condition relationship focusing on an extended record of Coronary Artery Disease (CAD) cases in Greece covering the solar Cycle 23 (1997–2007); those CAD cases are associated with high mortality and morbidity. The clinical presentation of CAD includes silent ischemia, stable angina pectoris, unstable angina, myocardial infarction (MI), heart failure and sudden death. The MI, unstable angina and silent ischemia consist Acute Coronary Syndromes (ACS) (Kremastinos, 2005; Hamm et al., 2011) and they share a widely common pathophysiological substrate. The pathophysiological mechanism is formed by the under perfusion of myocardial, due to important minimization of the blood flow in the coronary arteries, resulting in the partial or fully necrosis of myocardium. Such a state can be caused when an atherosclerotic plaque ruptures or erodes, leading to different degrees of superimposed thrombosis and distal embolization. The latter may lead to electrical instability which may, in turn, cause ventricular tachycardia or fibrillation and may end to sudden death. The leading symptom that initiates the diagnostic and the therapeutic cascade is chest pain, but the classification of patients is based on the electrocardiogram (ECG). Based on the morphology of ECG there are two categories of patients:  Patients with acute chest pain and persistent (P20 min) ST-segment elevation: The ST-segment elevation is identified by the presence of P1 mm ST elevation in at least two adjacent limb leads, P2 mm ST elevation in at least two contiguous pecordial leads or new onset bundle branch block. This is termed ST elevation ACS (STE– ACS) and generally reflects an acute total coronary occlusion which, mostly, develop a ST elevation MI (STEMI).  Patients with acute chest pain but without persistent STsegment elevation: They exhibit a persistent or transient ST-segment depression or T-wave inversion, flat T waves, pseudo-normalization of T waves, or no ECG changes are presented. This is dubbed non-ST-elevation ACS (NSTE–ACS) and the diagnosis is based on the measurement of troponine enzyme. Abnormal troponine level qualifies the situation as a non-ST- elevation MI (NSTEMI), otherwise is managed as an unstable angina. Also, a NSTEMI event can be caused by dynamic obstruction (eg.Vasospasm) and extrinsic factors leading 100 Ch. Katsavrias et al. / Advances in Space Research 51 (2013) 96–106 Table 2 Dates of increased number of NST (P5) and ST (P4) cases (per day) during storm recovery phase for the three periods examined in Section 3. Increased NST–ST are marked by . Date NST 6/8/2002  ST Multiple storm 3–29/9 5/9/2002  11/9/2002  Date NST 24/8/2002  9/9/2002 19/9/2002 Multiple storm 3–17/10 3/10/2002  14/10/2002  5/10/2002 17/10/2002 Double storm 23/10–17/11 29/10/2002 16/11/2002   10/11/2002 Double storm 17/11–18/12 22/11/2002  10/12/2002   4/12/2002 Multiple storm 18/12–9/1 22/12/2002  9/1/2003  26/2/2003 2/11/2003 8/11/2003  4/12/2003  10/3/2005  7/4/2005  14/4/2005  Multiple storm 6/5–8/7 10/5/2005  26/5/2005  16/6/2005  Multiple storm 21/8–25/9 26/8/2005          3/1/2003   ST  16/10/2003 6/11/2003 2/12/2003 21/2/2005 11/3/2005 10/4/2005 28/4/2005      19/5/2005 9/6/2005 21/6/2005   8/9/2005       to poor coronary perfusion (eg.hypotension, hypervolaemia or hypoxia). NSTEMI and unstable angina represent the majority of ACS. In this study, we examine the relationship between heliogeomagnetic activity and NSTE–ACS and STE–ACS in Greece for the whole solar cycle 23. We use the 1997– 2007 patients records from the St. Panteleimon General Hospital of Nikaea (Piraeus, Greece). 2. Data selection and analysis In this work we used data from the following databases:  OMNIWEB2: For the number of sunspots (Rz), the solar wind speed (VSW ) and the geomagnetic index DST  NGDC3 GOES: For the daily number of solar flares by intensity class (intense solar flares of X and M class, henceforward X–M and weak flares of B and C class, henceforward B–C). 2 3 http://omniweb.gsfc.nasa.gov/. http://ftp.ngdc.noaa.gov/stp/. Table 3 Dates of increased number of NST (P5) and ST (P4) Cases (identified by ) Outside storm recovery phase (includes pre-phase, Main-phase and intervals of increased VSW without storm and some unclassified). The results cover the three periods examined in Section 3. Date NST Main phase 24/10/2002 13/9/2005  Pre–phase 18/8/2002 27/1/2003 15/2/2005 Unclassified 13/1/2003 3/2/2005 21/7/2005 18/11/2005 18/12/2005 Solar wind speed 18/12/2003 26/12/2003 8/8/2005 13/10/2005 30/11/2005 ST Date NST 20/11/2003  21/1/2003 27/10/2003 5/5/2005    ST        14/11/2003 17/3/2005 25/10/2005 24/11/2005 19/12/2005 (P600 Km/s)      20/12/2003 29/3/2005 20/8/2005 6/11/2005 1/12/2005                  SOHO/LASCO4: For the daily number of CMEs (including the highly geoeffective Full and Partial Halo (width P200 degrees) hence P-H CMEs). There are two data gaps in this database in 25/6/1998–14/10/ 1998 and 1/1/1999–4/2/1999.  The patient records of the 2nd Cardiologic Department of the General Hospital of (Piraeus) ‘St. Panteleimon’, Greece: For the daily number of ACS cases, STE– ACS and NSTE–ACS (hence ST and NST respectively) Our data sets cover the Solar Cycle 23 from January 1st, 1997 to December 31st, 2007. We calculated the Pearson’s Product Moment Correlation Coefficient (Benesty et al., 2009) which gives an indication on the strength of the linear relationship between two random variables X and Y. X r ¼ 1=ðn  1Þ ðððX i  X Þ=sX ÞððY i  Y Þ=sY ÞÞ ð1Þ i Where ðX i  X Þ=sX , X and sX are the standard score, sample mean, and sample standard deviation respectively. If r(X,Y)=0, then X and Y are said to be uncorrelated (or linearly independent as a special case). The closer the value of r(X, Y) is to 1, the stronger the correlation between the two variables. For a non-linear dependency of X, Y, however, the coefficient may, also, be equal to zero. The solar activity cycle is quantified by the number of sunspots which is correlated with the flare and CME rate. The solar cycle 23 started in 1996–1997 and ended in 2006–2007 with maximum within 2000–2002. Surprisingly there were two very active periods, with strong flares and fast CMEs, during the decay phase of the solar cycle; these 4 http://cdaw.gsfc.nasa.gov/. Ch. Katsavrias et al. / Advances in Space Research 51 (2013) 96–106 101 Fig. 4. Daily number of DST index, solar wind speed (VSW ) and ACS (NST and ST) cases during the period August 2002–February 2003. The red curve represents the smoothed values (Savitzky–Golay smoothing with a window of 7 days). The ACS cases come from the database of General Hospital of the town of Nicaea (Piraeus) ‘St. Panteleimon’. Fig. 5. Daily distribution of DST and date labels of increased of NST (P5) and ST (P4) cases in 1/8/2002–8/2/2003). Time t is in days after 1/8/2002 for all panels. 102 Ch. Katsavrias et al. / Advances in Space Research 51 (2013) 96–106 Fig. 6. Same as Fig. 4 for the period October–December 2003. Fig. 7. Daily distribution of DST and date labels of increased of NST (P5) and ST (P4) cases in 1/10/2003 to 31/12/2003; t is in days after 1/10/2003. are the extremely active intervals of October-November 2003 (Alissandrakis et al., 2005) and of January 2005 (Bouratzis et al., 2006). Flares and CMEs were recorded throughout the 11 years cycle triggering disturbances in the terrestrial magnetosphere. In Fig. 1 (Three upper panels) we present the annual distribution of the number of flares and CMEs during the years 1997–2007, including the distribution of X–M type flares and P-H CMEs. In Fig. 1 (four lower panels), on the other hand, the magnetospheric disturbances are depicted. These include the annual number of geomagnetic storms and substorms (DST 6 50 nT), total storm duration per year, the DST peak of each year, and the product of DST peak times total storm duration multiplied by the annual number of storms (bottom panel); the product represents a phenomenological quantification of geomagnetic activity for each year and is used in comparisons between years. The SOHO/LASCO data included 12278 CMEs of which 6215 were of the Partial-Halo or Halo (P–H in this work) group (51%). There were 12019 GOES SXR flares of which 1173 X–M class (9%). The CME and flare rate reached annual peak in 2000–2002 which roughly coincides with the solar cycle maximum. However, when examined by category (or class) there was some differentiation regarding the peak times. For the flares the peak was in 2000–2001 (1400 flares); the X–M flares, however reached their maximum rate (250) in 2002. In the same year the number of CMEs and P–H CMEs peaked at 1740 and 850 respectively. Surprisingly, in the decay phase, the annual number of flares increases to 1950 in year 2004 and 1800 in the year 2005; the strong to medium (X–M) flare do not follow this trait dropping in numbers as expected. Therefore this irregular increase is due mainly to a large number of weak flares. The CME rate, also increases a little in 2005 (1247) and more in 2007 (1441). The irregular increases in flares and CMEs correspond to an increase of geomagnetic activity as recorded in OMNIWEB. We had 19 GMS with total duration of 63 days in 2002 (coincides with the CME peak in the same year) Ch. Katsavrias et al. / Advances in Space Research 51 (2013) 96–106 103 Fig. 8. The same as Fig. 4 for the year 2005. and 21 GMS with total duration of 51 days in 2005 (coincides with the CME peak of 2005); the peaks of DST , on the other hand, appear at 2001, 2003 and 2004. The full record of ACS cases includes 5160 NSTE–ACS (NST) and 2638 STE–ACS (ST); the NST cases were 62– 64% of the total. In the years 2001–2004, however, the NST reached 69–74% (74% in 2003); these are depicted in the three top panels of Fig. 1. In Fig. 2 we present monthly distributions of the number of: sunspots (Rz), flares, X–M flares, B flares, B–C flares, CMEs and P–H CMEs and ACS (ST and NST) cases; we have smoothed our data using a Savitzky–Golay filter with a window of 12 points. The maximum number of flares per month (185) was recorded in August 2001; the strong–medium (X–M) flare number, however, peaked in July 2000 (37), September 2001 and August 2002 (36). The maximum number of CMEs per month was also recorded in August 2002 (177) coinciding with one of the X–M peaks. The halo-partial halo (P–H) CMEs exhibited maxima in September 2002 (89), June 2000 (91) and March 2001 (92) near the X–M peaks. There was an important secondary maximum of flare number in January 2005 (217 flares), during the unusual active period mentioned at the beginning of this section; there was another unexpected secondary peak of CMEs and P–H CMEs within the solar minimum in April 2007 (161 and 90 respectively). We calculated the Pearson’s Product Moment Correlation Coefficient between the cardiological cases (NST, ST separately) and solar and geomagnetic activity parameters in order to specify their relationship. The results are summarized in Table 1 and Fig. 3. We restrict our presentation to correlations coefficients exceeding 50% with significance level above 99%. Best results were obtained with 12 months’ time lag for the cardiological cases. The smoothed values of CMEs and NST are fairly well corre- lated (0:79  0:00, see Fig. 3(a)), same holds for NST cases and X–M flares (0:66  21016 , see Fig. 3(a)), and number of B flares and ST cases(0:66  41016 , Fig. 3(c) with 12 months lag for the ST and NST). We also calculated the same correlation coefficients using a 18 months’ time lag with similar results (see Table 1 and 3(d)–(f)). 3. A detailed study of the relationship between cardiological cases and helio-geomagnetic activity in three characteristic periods of Solar Cycle 23 Based on the discussion of the previous section, we focus on characteristics periods of the solar cycle 23 as marked by peaks in active phenomena. We proceed therefore to analyze in more detail three interval of interest as follows:  The August–September 2002 peak: In this period appear the number of CMEs maximum, the number of P–H CMEs and the number of X–M flare peaks. Shortly after, in October 2002 the global maximum of NST cases (72) is recorded; a little later in November 2002 an ST monthly peak (32) is recorded. This was deemed significant as only 6 times after 2002 ST exceeds 32.  The Active Period October–December 2003 (Halloween Events): The most intense GMS of cycle 23 (220 nT) was recorded in October 30, 2003, within a period of extremely high helio-geomagnetic activity5 at the beginning of the decline phase of the solar cycle.  The year 2005: Although it was near the solar minimum, the quiet interval was interrupted by a period of extreme helio-geomagnetic activity in January 2005 with the 5 There were Solar Energetic Particle events and three Ground Level Enhancements of Cosmic Rays in October 28 and 29 and in November 2, 2003. 104 Ch. Katsavrias et al. / Advances in Space Research 51 (2013) 96–106 Fig. 9. Daily distribution of DST and date labels of increased of NST (P5) and ST (P4) cases in 1/1/2005–31/12/2005; t is in days after 1/1/2005 for all panels. maximum GLEs (Ground Level Enhancements of Cosmic Rays) of the past decades and a period of increased geomagnetic activity in April–September. This succession of intense, medium and small activity presents a unique opportunity of comparative study of the impact of this activity on ACS. For each period of interest we compare the variations of DST and VSW to the daily number of NST and ST cardiological cases from our records; Savitzky–Golay 7-day smoothing is used in order to eliminate the 7 days periodicity of the hospital duty cycle. The NST–ST daily number is characterized as increased when daily numbers are greater than 5 for NST and greater than 4 for ST. These increased cases are in excess of the average for each period. The days with increased NST-ST rate are summarized in Table 2 for the recovery phase of geomagnetic storms (single or multiple) and in Table 3 for increased rate outside recovery phase. This includes dates within the pre-phase or the main-phase of storms as well as dates of increased VSW without storm and some unclassified recordings. 3.1. The August–September 2002 peak The variations of DST and VSW and the daily numbers of NST and ST are presented in Fig. 4 for comparison. In Fig. 5 we show details of the DST variation annotated by the days with increased NST-ST rate. We note increased Ch. Katsavrias et al. / Advances in Space Research 51 (2013) 96–106 NST during September–October 2002 and ST during October–November 2002. This is near the peak of cycle 23 heliogeomagnetic activity as indicated by the DST and VSW . As second period of increased NST-ST rate appears in January 2003 (mostly NST); the origin of this peak is somewhat unclear as increased cases during and after Christmas and New Year’s Eve are always expected. It coincides, however, with increased helio-geomagnetic activity within the past months and with very high VSW (exceeding 700 Km/s). At this point we note that the fast solar wind is not always accompanied by storm (in cases when magnetospheric field lines do not open and reconnect usually when Bz is negative); there is however increased ST–NST rate in this case too. From Fig. 5 the 80% of increased NST-ST rates appear during the recovery phase of storms. 3.2. The active period October–December 2003 In Fig. 6 we present DST , VSW and numbers of NST and ST per day. In Fig. 7 we mark the days with increased NSTST rate on the DST time series. There are four intense storms in mid October, at the end of October (the most powerful GMS of solar cycle 23), at the beginning of November and after mid November. This period is associated with increased NST cases from mid October until the end of December. There are also increased ST cases although there are fewer. In this period the 50% of increased NST-ST rates appears during the recovery phase of storms (Fig. 7, Table 2). A number of increased ACS (NST–ST) cases (18, 20 and 26/12/2003) do not coincide with storm but follow a period of high speed solar wind (9–15/12/2003). 3.3. The year 2005 The DST , VSW and ACS (NST–ST) relationship are in Fig. 8 with details of increased NST–ST rate compared to DST in Fig. 9. Extreme helio-geomagnetic activity was recorded in January 2005, accompanied by the largest GLEs of the past decades and was followed by a period of increased activity in April–September; there were no GMS after September. Within January 2005, a number of GMS with duration of 10 days, a Solar Energetic Particle event (SEP) and two GLEs (17 and 20) occured. During the following period high geomagnetic activity including multiple powerful GMS was recorded; in May 2005, the DST peak of 2005 was observed. Increased ACS rate follows this activity; the number of ST cases increases in January and February and NST increase in March. Within April, ACS (ST–NST) numbers drop increasing again after the end of the month (NST in particular), following increasing GMS activity and culminating at the peak of May. After a drop in June–July, a new increased rate of ACS numbers is recorded in October 2005; this is, probably, related to the increased GMS within September. Within the quiet October–December interval, however, a surprising enhancement of the ST rate appears in November 2005; it included 40 cases within this month which is 105 twice the mean number of ST per month and the peak of cycle 23. Despite the absence of storms, high-speed solar wind was recorded. We note, at the same period, an unusually fast drop of the monthly number of flares. In 118 flares were recorded in August, 40 in September, only 1 in October but 148 flares in November. From Fig. 9 the 45% of increased NST–ST rates appear during the recovery phase of storms and 21% within the high-speed solar wind interval. 4. Discussion and conclusions In this report we have performed a long–term analysis of the relationship of helio-geomagnetic activity to Acute Coronary Syndromes (ACS) including both STE–ACS (ST) and NSTE–ACS (NST) cases separately. This study covered the 23d solar cycle (1997–2007); it was focused, mostly, on certain periods of the solar cycle 23 where the activity or its variation were quite pronounced. The results are summarized as follows:  The STE–ACS (ST) and NSTE–ACS (NST) cases are related to the helio-geomagnetic activity but, in detail, this relationship may be somewhat different. The calculated correlation coefficient using smoothed monthly values when the ACS time-series was lagging by 12 months behind the helio-geomagnetic activity are: 0:79  0:00 between CMS and NST, 0:66  21016 between NST–intense flares (X and M) and 0:66  41016 between ST–faint flares (B type).  Using annual values of the NST/ST cases we find an almost constant ratio of 62–64% which increases towards periods on increased solar activity (2001– 2004) reaching a peak of 74% in 2003.  The maximum of NST cases (72) were recorded in October and January 2003. The second, in magnitude, peak of ST cases (32), on the other hand, appeared in November 2002. This supports time coincidence between increased ACS cases and strong helio-geomagnetic activity as the peak of this activity took place in August–September 2002 as presented in 3.1. The maximum number of ST cases in November 2005 (40) coincides, in time, with the unusual fluctuation of the number of flares discussed in 3.3.  In the period October–November 2003 (subSection 3.2) the strongest recorded GMS had a duration of 5 days where the DST peaked at 220 nT. A less strong (120 nT) but of longer duration (18 days) GMS occured October 2002 (subSection 3.1). A comparison of the effects of these two storms indicates that the ACS reach maximum in October–November 2002 as mentioned above. This suggests that the health effects of geomagnetic activity are cumulative as they appear to depend on duration.  For the three periods of interest examined in Section 3 we have studied the phase (pre-phase, main, recovery) of the storm that coincides with increased ACS cases.: 106 Ch. Katsavrias et al. / Advances in Space Research 51 (2013) 96–106 – 60% of increased ACS cases coincide with the storm recovery phase; the probability rises to 75% for the NST cases. (see Table 2) – Only 3 dates (4.5%) with increased ACS cases took place at the main phase (see Table 3). – 6 dates (9% of the increased cases took place at the prephase of the storms (see Table 3). – From Table 3, we note a significant number of increased ACS cases (15%) in periods of high solar wind speed (P600 Km/s) without geomagnetic storms. These results corroborate the geomagnetic effects on cardiological cases (Palmer et al., 2006 and references therein); the availability of accurate hospital records covering decades, supports a detailed analysis, in the future, in order to improve the statistics and secure the conclusions. Acknowledgements This work was supported in part by the University of Athens research center (ELKE/EKPA). The authors appreciate discussions with Alexander Hillaris and critical reading of the manuscript by George Xystouris. They also acknowledge the suggestions and recommendations of the two anonymous referrees which lead to improvements in the quality of this report. References Alissandrakis, C.E., Nindos, A., Hilaris, A., Caroubalos, C., The Artemis Team, Complex solar events observed with the Artemis-Iv RadioSpectrograph in October/November 2003. In: The Dynamic Sun: Challenges for Theory and Observations, vol. 600 of ESA Special Publication. p. 106.1, December 2005. Belov, D., Kanunikov, I., Kiselev, B. Dependence of human EEG spatial synchronization on the geomagnetic activity on the day of experiment. Ross Fiziol Zh Im I M Sechenova 84, 761–774 (in Russian), 1998. Benesty, J., Chen, J., Huang, Y., Cohen, I. Noise Reduction in Speech Processing Topics in Signal Processing. Springer, Berlin, Germany. April, 2009. Bouratzis, K., Preka-Papadema, P., Hillaris, A., Moussas, X., Caroubalos, C., Petoussis, V., Tsitsipis, P., Kontogeorgos, A. Radio bursts in the active period January 2005. In: N. Solomos (Ed.), Recent advances in astronomy and astrophysics, 7th International Conference of the Hellenic Astronomical Society, vol. 848 of American Institute of Physics Conference Series, Kefalinia, Greece, pp. 213–217, August 2006. Breus, T., Komarov, F., Musin, M., Naborov, I., Rapaport, S. Heliophysical factors and their influence on cyclical processes in biosphere. Itogi Nauki I Techniki: Medicinskaya Geografica 18, 172– 174, 142, 145, 147, 148, (in Russian), 1989. Chernouss, S., Vinogradov, A., Vlassova, E. Geophysical hazard for human health in the circumpolar auroral belt: evidence of a relationship between heart rate variation and electromagnetic disturbances. Nat Hazards 23, 121–135, 2001. Cherry, N. Schumann resonances, a plausible biophysical mechanism for the human health effects of solar. Nat Hazards 26, 279–331, http:// dx.doi.org/10.1023/A:1015637127504, 2002. Cornélissen, G., Halberg, F., Breus, T., Syutkina, E.V., Baevsky, R., Weydahl, A., Watanabe, Y., Otsuka, K., Siegelova, J., Fiser, B., Bakken, E.E. Non-photic solar associations of heart rate variability and myocardial infarction. J. Atmos. Sol-Terr. Phys. 64, 707–720, March, 2002. Dimitrova, S. Relationship between human physiological parameters and geomagnetic variations of solar origin. Adv. Space Res. 37, 1251–1257, 2006. Dorman, L.I., Iucci, N., Ptitsyna, N.G., Villoresi, G. Cosmic rays as indicator of space weather influence on frequency of infarct myocardial, brain strokes, car and train accidents. in: International Cosmic Ray Conference of International Cosmic Ray Conference, vol. 9, p. 3511–3514, August 2001. Gavryuseva, E., Kroussanova, N. The relationship between the human state and external perturbations of atmospheric, geomagnetic and solar origin. in: A. Wilson (Ed.), Solar Variability: From Core to Outer Frontiers, ESA Special Publication, vol. 506, pp. 1053–1056, December 2002. Hamm, C., Bassand, J.P., Agewall, S., et al. ESC Guidelines for the management of acute coronary syndromes in patients presenting without persistent ST – segment elevation. Euro. Heart J. 32, 2999– 3054, 2011. Kivelson, M.G., Russell, C.T. Introduction to Space Physics. Cambridge University Press, New York, USA. April, 1995. Kremastinos, D. Cardiology Clinical Cardiology. Medical Publications Paschalides, Athens, Greece, 2005. Mendoza, B., Diaz-Sandoval, R. Effects of solar activity on myocardial infarction deaths in low geomagnetic latitude regions. Nat. Hazards 32, 25–36, http://dx.doi.org/10.1023/B:NHAZ.0000026789.71030.31, 2004. Palmer, S.J., Rycroft, M.J., Cermack, M. Solar and geomagnetic activity, extremely low frequency magnetic and electric fields and human health at the Earth’s surface. Surveys Geophys. 27, 557–595, September, 2006. Stoupel, E., Abramson, E., Sulkes, J., Martfel, J., Stein, N., Handelman, M., Shimshoni, M., Zadka, P., Gabbay, U. Relationship between suicide and myocardial infarction with regard to changing physical environmental conditions. Int. J. Biometeorol. 38, 199–203, December, 1995. Stoupel, E., Domarkiene, S., Radisauskas, R., Bernotiene, G., Abramson, E., Israelevich, P., Sulkes, J., Breus, T. Variants of acute myocardial infarction in relation to cosmophysical parameters. Sem. Cardiology 11, 51–55, 2005. Stoupel, E., Kalediene, R., Petrauskiene, J., Starkuviene, S., Abramson, E., Israelevich, P., Sulkes, J. Monthly deaths number and concomitant environmental physical activity: 192 months observation (1990–2005). Sun Geosphere 2, 78–83, December, 2007.