CC BY
22
CONGRUENCE OF MORTALITY FROM MYOCARDIAL INFARCTION IN BULGARIA AND PC1 PULSATIONS
Germaine Cornelissen, Franz Halberg, Tamara Breus, P.K. Ivanova, V.F. Ozheredov,
N. G. Kleimenova, O. V. Kozireva, Agarwal Rajesh K.
Aim. To compare the time structure of mortality from myocardial infarction (MI) in Bulgaria with that of Pel pulsations. Materials and methods. Data on the monthly incidence of mortality from MI in Bulgaria during the span 1970-1995 (26 years) are compared with concurently recorded Pcl pulsations in the 0.2-5 Hz range from Borok's Observatory. The Pcl data analyzed herein are the monthly numbers of 15-min intervals during which Pcl pulsations were present and lasted at least 5 minutes. Both data series were analyzed by the extended cosinor method. Least squares periodograms in the frequency range from one cycle in 26 years to about 5.4 cycles per year were computed for the Pcl data and for the MI incidence data, the latter after removal of a marked increased linear trend. Anticipated components corresponding to spectral peaks were further analyzed by nonlinear least squares to obtain point and 95% confidence interval estimates for the period as well as for the amplitude and acrophase. Nonlinear analyses were performed on the original data, including a linear trend in the fitted model.
Results. Detrended monthly MI incidence in Bulgaria correlates positively with Pcls (r=0.l9l, P<0.00l). As both variables are characterized by a prominent circannual variation, peaking in the winter (Pcl: around mid-December; MI: around mid-February), the extended cosinor (l-3) was used to check for any congruence of long-term variations between these two variables. They are indeed also characterized by an about l0-year variation that is more pronounced for the physical than for the biological data. The nonlinear model applied to the original (raw) data consisted of a linear trend, a fixed l.0-year cosine curve and an adjusted l0.5-year cosine curve. As shown in Table l, both components are resolved nonlinearly with statistical significance, the infra-annual cycle being congruent between the two variables (see Figure).
Table 1
Variable Fixed 1.0-year Trial period = 10.5 years
Amplitude Period Amplitude
[95%CI] [95%CI] [95%CI]
Pcl 75.73 10.60 130.06
[43.82, 107.64] [9.91, 11.29] [97.59, 162.54]
MI 933.1 12.62 169.41
[802.0, 1064.2] [9.98, 15.27] [28.24, 310.58]
Incidence of Myocardial Infarction in Bulgaria and Pattern of Pc1
NL: 12.62
Time (calendar date)
Data received from T Breus
Discussion. It has been reported that variations in MI and Pcl pulsations both exhibit a summer mimimum (4).
In the present analyses, the calendar year is the most prominent for MI and the about l0.5-year component predominates in the Pcl. Hence, the relative prominence of the circannual and circadecadal components is opposite between the two variables. In view of the much smaller amplitude of the about l0.5-year cycle in MI, the 95% CI of its period is much broader than that of the Pcl also recorded during the same 26-year span. An about l0.5-year cycle was also found to characterize mortality from MI in Minnesota, recorded for 29 years from l968 to l996 (5).
1. Halberg F. Chronobiology: methodological problems. Acta med rom l980; l8: 399-440.
2. Cornelissen G, Halberg F. Chronomedicine. In: Armitage P, Colton T (Eds.) Encyclopedia of Biostatistics, 2nd ed. Chichester, UK: John Wiley & Sons Ltd; 2005. p. 796-8l2.
Материалы IX международного конгресса «Здоровье и образование в XXI веке» РУДН, Москва
3. Refinetti R, Comélissen G, Halberg F. Procedures for numerical analysis of circadian rhythms. Biological Rhythm Research 2007; 38: 275-325.
4. Kleimenova NG, Kozyreva OV, Breus TK, Rapoport SI. Pcl geomagnetic pulsations as a potential hazard of the myocardial infarction. JASTP 2007; 69: 1759-1764.
5. Comélissen G, Halberg F, Breus T, Syutkina EV, Baevsky R, Weydahl A, Watanabe Y, Otsuka K, Siegelova J, Fiser B, Bakken EE. Non-photic solar associations of heart rate variability and myocardial infarction. JASTP 2002; 64: 707-720.
Материалы! IX международного конгресса «Здоровье и образование в XXI веке» РУДН, Москва
