, Germaine Cornelissen2 and Franz Halberg2
(1)
Department of Chronomics & Gerontology, Tokyo Women’s Medical University Medical Center East, Arakawa-ku, Tokyo, Japan
(2)
Halberg Chronobiology Center, University of Minnesota, Minneapolis, MN, USA
Abstract
Franz Halberg (1919–2013) developed chronobiology and founded the field of chronomedicine including chronomics, chronoastrobiology, and chronobioethics. The progress of chronomedicine is one and the same as his history of research. This chapter introduces his foundation of chronobioethics.
Natural geophysical near-weeks, half-years, the solar wind’s near- and far-transyears, etc., have evolved in living matter into a transdisciplinary system of coperiodisms in and around us. These ubiquitous cycles, some in the archaeon and the human, constitute congruent signatures not only of the photic day and year but also of the nonphotic, including gravitational and magnetic cyclic influences upon our mental functions in particular. They leave their signatures in terrorism, military-political affairs, crime, suicide, church memberships, and religious proselytism.
Keywords
Glocal methods in time and spaceAeolian original dataChronousphereChronobioethics3.1 Encounters
Prof. Dr. (med.) Dr. (theol.) Mag. (pharmacist) Gustav Sauser, whom I (FH) met while he practiced as a pharmacist in Wels, Upper Austria, after he lost his professorship during the occupation of Austria in World War II, granted me an opportunity for an academic connection. After the war, he was appointed as the head of the anatomy department at the University of Innsbruck and in succession became its dean and rector. Sauser offered me a room in his bombed-out anatomy department building; as his assistant, I participated in international academic weeks [1, 2] where I met the physicist Arthur March (a friend of Erwin Schrödinger), with whom I developed an interest in physics. There, I also read Galileo and the statement attributed to him that I (and others including physicists and historians I consulted had cited) could never identify in his writings: “Measure what is measurable and render measurable what as yet is not,” which became the guiding principle of my endeavors and to which I added qualifications like “in time and hence meaningfully”: “Measure, in a partial system, everything pertinent that is measurable and render measurable what is pertinent but as yet is not measurable, as simply as possible, but not simpler, in time (tempestive) and hence meaningfully, taking into account chronomic maps, a described and quantified complementary system” [Omnia propria ex systemate partiali metire quæcumque licet et propria immensa, quam simplicissime sed non simplicius, ad mensuram tempestive et ergo significative redige, reddens rationem tabulae chronomicae – systematis complementaris, descripti et quantificati].
While in Innsbruck from 1946 to 1948, I also headed the section of venereology and dermatology at the French military hospital, where I encountered my first case in which timing could possibly have played a role. In two identically treated soldiers who contracted gonorrhea from the same prostitute on consecutive nights, the same treatment begun the same day cleared gonococci first from the patient who was infected a day later [3]. I also owe Sauser my World Health Organization fellowship that brought me to Harvard in 1948.
3.2 Opportunities Enjoyed and Yet to Be Followed Up On
In Minnesota in 1950, I met Prof. Alan Treloar, who renewed my interest in the inferential statistical validation of all I learned. In 1961, I began receiving a lifetime career award from the US National Institutes of Health (NIH); at over 92 years of age, I am still receiving it. I may be one of only two still-active recipients of these awards. A contract to study the endocrine associations of breast cancer, also from the NIH data, led to multifrequency models of infradian risk assessment. We found modulations by both about-30-day and about-yearly rhythms of the circadian MESOR and amplitude demonstrating them stepwise as partial spectral structures based on parameters obtained by the least-squares fit of cosine curves of 24 hours and 28 and 365 days to data on plasma prolactin, Figs. 3.1, 3.2, 3.3, and 3.4 [4].1 These are obtained every 20 min, and data on other hormones are obtained at 100-min intervals for 24 hours, in 4 seasons on 3 age groups of Japanese and Minnesotan women, Fig. 3.1a–f, to derive tentative gauges of breast cancer and other disease risks. We then wrote something still pertinent today:
Fig. 3.1
(a) Individual sampling schemes (spans) for assessment of multifrequency rhythms in human blood and urine. Experimental design of Minnesota-Japan chronoepidemiologic study on women of 3 age groups: schedule of hospital admission and at-home self-measurements of subjects classified according to age and geographic location, with specification of menstrual cycle stage on day of admission. Limitation: ignorance, at the time, of quinmensals and transyears discovered later in some other hormones that remain to be examined (© Halberg Chronobiology Center). (b) Comparison of statistical significance of circadian rhythm, assessed by least-squares fit of a 24-hour cosine curve for 13 hormones in plasma of healthy women sampled for individualized rhythm assessment by cosinor. P from F-test of zero-amplitude hypothesis. Since the hierarchical statistical significance represented by P-values is sampling dependent, prolactin and cortisol, measured every 20 min, were also analyzed at 100-min intervals as for the other hormones. Total number of determinations: 40,765.(© Halberg Chronobiology Center). (c) Comparison of plasma prolactin in clinically healthy Japanese and Minnesotan women in four seasons. Samples taken at 20-min intervals over a 24-hour span in Fukuoka City, Japan, and Minneapolis, Minnesota, USA (© Halberg Chronobiology Center). (d) Geographic difference in circannual rhythm of human plasma prolactin. Contradictory comparisons of circadian MESOR could be obtained by sampling in different seasons. Moreover, controlling only time of year (or only time of day) may not avoid misleading results due to circannual (or circadian, as the case may be) variations, whereas assessment of circannual (or circadian) rhythm characteristics may reveal important chronoepidemiological or chronoprotopathological results (as would most likely extended focus on rhythms discovered in the interim). Conclusions from single samples obtained in different seasons (or times of day) can be resolved as differences of circannual (circadian) prolactin rhythm between Japanese and American women. Each subject provided 72 blood samples (8640 determinations, each in duplicate). W winter, Sp spring, S summer, F fall (© Halberg Chronobiology Center). (e) Circadian rhythm of plasma prolactin in Japanese and North American women in winter. Dense sampling on plasma prolactin suggests both a larger circadian amplitude and a higher rhythm-adjusted mean or MESOR in Japanese as compared to North American women (P < 0.002 from Hotelling T2 test) (© Halberg Chronobiology Center). (f) Difference of plasma prolactin between clinically healthy Japanese and American women in March 1978 when circannual rhythm is near its maximum, confirmed by complementary study. Fifteen whites of mixed ethnic background (18–24 years old) in Minnesota and 20 Japanese (~20 years old) in Kyushu, Japan (© Halberg Chronobiology Center)
Fig. 3.2
Stepwise theoretical reconstruction of partial spectral structure of human plasma prolactin for group of Japanese women (based on parameter estimates obtained from separate least-squares fits of cosine curves with periods of 24 hours [circadian], 28 days [circatrigintan], and 365 days [circannual] to data on plasma prolactin obtained every 20 min for 24 hours, ~4 times a year on a few women). Didactic example modeling the interaction of human plasma prolactin rhythms with different frequencies, including modulation of circadian MESOR and amplitude by circatrigintan and circannual rhythms (© Halberg Chronobiology Center)
Fig. 3.3
A negative correlation between the total relative breast cancer risk evaluated from epidemiological criteria and the circannual prolactin amplitude (left) corroborates the finding that an evaluation of breast cancer risk is associated with a decrease in circannual amplitude (based on least-squares fit of 365.25-day cosine curve to circadian MESORs assessed in each of the four seasons). Note further the positive correlation between epidemiologically assessed breast cancer risk and the circannual amplitude of thyroid-stimulating hormone (TSH) (right). Clinical thyroid malfunction has empirically been associated with breast cancer risk. If, then, this topic is still controversial, this may perhaps be accounted for by the circannual stage dependence of the correlation. Note also that in prostatic cancer (a condition characterized by geographic differences in morbidity and mortality similar to those of breast cancer), the extent of circannual variation also changes as a function of risk and/or cancer. In blood sampled with serial independence in the morning at different times of the year, a prominent circannual rhythm in TSH of healthy subjects is lost in prostatic cancer (and perhaps even in men at high risk of prostatic cancer). For prolactin, a circannual rhythm becomes demonstrable in the case of prostatic cancer, while it is not demonstrable with serially independent sampling in healthy men of low or high prostatic cancer risk. Thus, TSH and prolactin show opposite behavior along the 1-year scale in cancers of both breast and prostate (rather than responding in the same way, as is the case along the scale of minutes to hours, following the application of stimuli such as thyrotropin-releasing hormone). The two correlations shown in the figure are just part of a larger correlation matrix. The circumstance is noted that correlations emerged as statistically significant for the very hormones which clinicians have long considered have some relation to breast cancer, yet thus far could not rigorously establish such a relation as biologically significant, perhaps because of too-limited sampling. While these conclusions rest on large samples, they describe only a small number of subjects. Moreover, all conditions required to apply a linear regression between two variables are not satisfied. A test for lack of fit indicates that the model is not adequate for TSH; the error term is not normally distributed. In addition, the assumption of homogeneity of variance is not verified for prolactin as well as for TSH. Finally, in the case of prolactin, there seems to be an age effect on both this circannual amplitude (decreases with age) and the breast cancer risk (increases with age). This may account for the negative correlation illustrated in the figure. Hence, the correlations in the figure are of ordering rather than documenting value. They are intended to emphasize that circannual rhythmicity deserves further study in relation to carcinogenesis. If such correlations can be confirmed and if the circannual rhythms involved should prove to be determinants of carcinogenesis in the human breast, these same correlations will point to the possibility of a chemoprevention of breast cancer (© Halberg Chronobiology Center)
Fig. 3.4
Top: negative correlation between the circannual aldosterone amplitude (based on least-squares fit of 365.25-day cosine curve to circadian MESORs assessed in each of the four seasons on each of the women investigated in Minnesota and Japan) and the circadian diastolic blood pressure MESOR (in winter, the season when blood pressure profiles at about 10-min intervals were obtained in both locations) observed in women in Minnesota (N = 13, left), Kyushu, Japan (N = 10, middle), and both (N = 23, right). Bottom: negative correlation of the circannual aldosterone amplitude and both the circadian diastolic blood pressure MESOR and the individual risk of diseases associated with high blood pressure. Results from 23 presumably healthy women, 13 in the USA (∎) and 10 in Japan (●); estimate of circadian diastolic blood pressure MESOR based on appropriate data obtained only in winter; circannual aldosterone amplitude determined by fit of 365.25-day cosine curve to circadian MESORs assessed in each of the four seasons; factors determining risk value included familial morbidity or mortality, obesity, “high” heart rate, and a history of blood pressure spiking. The same restrictions as outlined for Fig. 3.1g applies to Fig. 3.2 and this figure. For qualification, see legend of Fig. 3.3 (© Halberg Chronobiology Center)
Once we realize that a spectrum of rhythms exists [Figures 3.1, 3.2, 3.3, and 3.4] and what its components are, we have a chance to specify how, with very few samples if not with a single sample, at the right time for each pertinent variable, one may [try to] achieve multiple aims. One may quantify certain aspects of health. One may recognize risk — before disease sets in, as a basis for prevention. One may also learn when one should look for what and where in the body — in a study of physiological mechanisms, safeguarding our performance — the neuroendocrine and cellular feedsidewards leading to a yet to be tested hemopsy [4–7]
The extension of this model of 1981 [4] to include the periods mapped in the interim of 30 years as coperiodisms in and around us, up to paratridecadals, remains a major challenge for marker rhythmometry.
Whenever possible, for disease risk assessment by infradian modulation, we should prefer the noninvasively collected saliva or urine to plasma. A study of 11,702 samples on six hormones in saliva collected every 4 hours around the clock for many months reveals new facts, including not only multiple distinct frequencies in a very broad spectrum but also, i.e., multiple circadians. Multiple circaseptan frequencies, Fig. 3.5a, b, were also found in cases where, under life in contemporary society, we expected to find only a single circadian or circaseptan frequency, which is or is not environmentally synchronized. As noted earlier, there are multiple coexisting circadian frequencies, Table 3.1. Thus, we arrive at new diagnoses; once we rely on time series, we may quantify even the effects of the earth tides on humans [8]. Serial data in turn require glocal methods in time and space, notably if focus is extended beyond the circadian system.
Fig. 3.5
(a) Transdisciplinary congruence, defined by overlapping 95 % confidence intervals of the periods in the circaseptan spectral region (© Halberg Chronobiology Center). (b) Combination of gliding spectral window (top) with special focus on the behavior of two selected periods (serial sections, middle and bottom), with the time course of the phase validating the 6.77-day period by a more or less horizontal trajectory of phases in only part of the record, but invalidating a precise 7-day periodicity since no dots bracket any horizontal time course of phases, and the small initial section with dots shows a gradual advance (© Halberg Chronobiology Center). (c) Fourier power spectral density of rest activity of a clinically healthy man (JFC) on a self-selected sleep-wake schedule (© Halberg Chronobiology Center). (d) Sleep-wakefulness on a self-selected schedule analyzed by the fit of cosine curves of 24.00 hours to intervals of 168 hours (top 3 rows) or of 24.84 hours to intervals of 173.88 hours displaced with increments of 12.00 or 12.42 hours, respectively. Nonlinear analyses (not shown here) reveal coexisting circadian periods including a 24.84-hour period (24.828; 24.859) during an initial more or less 24 hour synchronized span seen in row 2. Note, in the first 18 lunar months, in row 2, the dominating, albeit undulating and slightly delaying, not quite horizontal time course of doubly plotted acrophases indicating perhaps the interaction of a 24.00-hour schedule with a competing free-running and/or other mechanism; periods shorter than 24.8 hours are seen in the first 15 lunar months in row 4. This row also shows spans of 24.8-hour (earth tidal, also greatly undulating) synchronization (Original data from John F. Costella © Halberg Chronobiology Center)
Table 3.1
On a self-selected schedule (JFC), a double tidal pull competes with 24-hour environment while initially coexisting free-running circadian fades (global analysis)
Calendar date | Period | Amplitude | DT-A/S-A (%) | |||
|---|---|---|---|---|---|---|
All (3-year) data | Double tidal (DT) | 24.836 (24.833, 24.838) | 0.10606 (0.0854, 0.1268) | Global | ||
1990/02/19 to 1993/03/11 | Compromise | 24.432 (24.427, 24.436) | 0.05643 (0.0357, 0.0772) | |||
Free-running | 24.260 (24.252, 24.268) | 0.03253 (0.0118, 0.0533) | ||||
Societal (S) | 24.001 (24.000, 24.001) | 0.25443 (0.2337, 0.2751) | 0.41685 | |||
Tidal | 12.414 (12.410, 12.418) | 0.01728 (−0.0034, 0.038) | ||||
Semidian | 11.999 (11.997, 12.001) | 0.03204 (0.0113, 0.0527) | ||||
Local | 1: 1990/02/19 to 1991/04/28 | Double tidal | 24.872 (24.830, 24.914) | 0.02513 (−0.0070, 0.0573) | ||
Compromise | 24.409 (24.384, 24.435) | 0.04117 (0.0089, 0.0734) | ||||
Free-running | 24.222 (24.210, 24.235) | 0.07921 (0.0468, 0.1116) | ||||
Societal | 24.008 (24.004, 24.011) | 0.31605 (0.2838, 0.3483) | 0.07951 | |||
Tidal | 12.421 (12.411, 12.431) | 0.02615 (−0.0059, 0.0582) | ||||
Semidian | 12.007 (12.001, 12.013) | 0.04105 (0.0090, 0.0731) | ||||
2: 1991/04/28 to 1991/09/23 | Double tidal | 24.898 (24.832, 24.964) | 0.12744 (0.0707, 0.1842) | |||
Compromise | 24.707 (24.671, 24.743) | 0.25208 (0.1973, 0.3068) | ||||
Compromise | 24.488 (24.434, 24.541) | 0.10274 (0.0552, 0.1502) | ||||
Societal | 24.011 (23.992, 24.030) | 0.22758 (0.1815, 0.2736) | 0.55998 | |||
Tidal | Did not converge | – | ||||
Semidian | – | – | ||||
3: 1991/09/23 to 1992/03/08 | Double tidal | 24.862 (24.852, 24.872) | 0.34858 (0.3089, 0.3882) | |||
Compromise | 24.523 (24.468, 24.579) | 0.06108 (0.0212, 0.1009) | ||||
Free-running | Did not converge | – | ||||
Societal | 23.997 (23.984, 24.011) | 0.22362 (0.1841, 0.2631) | 1.55881 | |||
Tidal | 12.432 (12.416, 12.448) | 0.04969 (0.0106, 0.0888) | ||||
Semidian | Did not converge | – | ||||
4: 1992/03/08 to 1992/07/14 | Double tidal | 24.942 (24.896, 24.989) | 0.15133 (0.0960, 0.2067) | |||
Compromise | 24.451 (24.394, 24.508) | 0.11847 (0.0630, 0.1739) | ||||
Free-running | – | – | ||||
Societal | 23.987 (23.967, 24.006) | 0.32099 (0.2660, 0.3760) | 0.47145 | |||
Tidal | 12.432 (12.398, 12.465) | 0.04960 (−0.0051, 0.1042) | ||||
Semidian | 11.982 (11.951, 12.012) | 0.05028 (−0.0044, 0.1049) | ||||
5: 1992/07/14 to 1993/03/11 | Double tidal | 24.789 (24.782, 24.797) | 0.31836 (0.2798, 0.3569) | |||
Compromise | 24.474 (24.438, 24.510) | 0.06218 (0.0236, 0.1008) | ||||
Free-running | – | – | ||||
Societal | 24.009 (23.997, 24.022) | 0.17892 (0.1405, 0.2173) | 1.77934 | |||
Tidal | 12.352 (12.328, 12.376) | 0.02358 (−0.0147, 0.0619) | ||||
Semidian | 12.001 (11.988, 12.013) | 0.04128 (0.0030, 0.0796) |
In the early 1960s, representatives from NASA were the first to knock on my door and ask what they could do for Minnesotan chronobiology. NASA, the National Science Foundation (NSF), and NIH enabled the completion of a laboratory for periodicity analysis, with standardization, if not full control, of environmental temperature and lighting, and as best we could of sound and smell, since they were the candidate factors accounting for the synchronization of the eosinophil count’s circadian rhythm in commonly but not in singly housed mice without eyes (deprived of them by surgery or genetics, ZRD). NASA also supported the introduction of temperature and other telemetry in preparation for a biosatellite study that remains a challenge, as does that of a lunar laboratory [9]. In extraterrestrial space, away from hospitals, it will be especially important to detect undue strain and to alleviate it, a task also critical on earth.
It cannot be overemphasized that cycles, when not considered, can be confounders, or when two cycles being compared differ in phase, or in frequency or for those very many investigators sampling at a single clock hour with an interest in aging. A human mental function, such as the duration of an estimation of one minute, may decrease with age at one time of day, while at another clock hour in the same person in the same longitudinal time series, an increase with age is found, Fig. 3.6. This is a necessary a priori consequence of a decrease in circadian amplitude (or of a change in phase) with age. If other characteristics do not change (they actually do in the case considered), a decrease with age of the circadian amplitude results in an increase at the trough, and a decrease at the peak of a daily rhythm, as readily visualized by a 24-hour cosine fit at the start and at the end (not shown).
Fig. 3.6
The data of the first and last years of 1-min estimation by RBS were stacked along the 24-hour scale. Whereas parameter tests did not detect any change in the characteristics of the 24-hour synchronized rhythm, a paired t-test comparing eight 3-hourly mean values shows a difference (P < 0.05). Moreover, Student’s t-test on data in each of eight equidistant bins reveal that at certain circadian stages, 1 min passed faster in the last year than in the first year (year 35 vs. year 1). This difference was statistically significant at all test times between 06:00 and 18:00 (in some of them with P < 0.001), but was not found between 18:00 and 06:00, suggesting that the change in 1-min time estimation with age is circadian stage dependent, with highly significant differences during part of the daily active phase, but not at other circadian times. In RBS, chronomics demonstrate interaction between the circadian rhythm’s stage and age (© Halberg Chronobiology Center)
Fortunately, the puzzles were solved and nonsensical partial results were not published. Interest in full- and overtime study of the cosmos resulted from our inability to validate a test developed and published by ourselves in Minnesota [10], confirmed abroad for 2 years, yet subject to a modulation by an initially unknown circadecadal rhythm, which could eventually be found in the circadian blood pressure amplitude of newborns. The challenge remains to test other-than-circadian, e.g., circaseptan amplitudes that may be less confounded by decadal changes for diagnosing the neonatal risk of developing cardiovascular disease later in life, even though an effect of geomagnetism upon them has been demonstrated [11].
Sooner rather than later, we must deal with a broad spectrum of nonphotic, partly gravitational, and/or magnetic novel paired transdisciplinary cycles in and around us in the sphere of the human mind (noös), Vernadsky’s noösphere. Mikulecky and Pales’ ~500-year cycle [12, 13] in the emergence of prominent poets, historians, and physicians, with counterparts in cave temperature and the incidence pattern of international battles [14], Fig. 3.7a–e, show the extent to which the cosmos modulates human affairs, including science and art, leading to a temporally structured chronousphere, coined by portmanteauing the Attic Gk nous (= noos), into Gk chronos (=time) and Gk sphairos (=globe as well as sphere). There is no choice for a unified art and science, whether in the service of health care, anthropology, or a unified science, but to monitor from birth to death with repeated passes over the accumulating data so as to determine loads before they become undue strain and harbingers of disease, to start with in blood pressure and heart rate. Monitoring has to start during wellness for the prevention of illness as soon as affordable unobtrusive instrumentation yields data analyzed via an international multilingual website, the recommendation of a World Forum on “Natural Cataclysms and Global Problems of the Modern Civilization,” held September 19–21, 2011 in Istanbul, Turkey.
Fig. 3.7
(a) About-500-year cycles in the emergence of famous physicians are visible to the naked eye and seem to be synchronized in three completely different geographical regions among which there was originally little if any communication (© Halberg Chronobiology Center). (b) About-500-year cycles are also apparent in the emergence of famous historians and poets in three completely different geographical regions among which there was originally little if any communication (© Halberg Chronobiology Center). (c) Indirect proxy approximations of solar activity via the effects of climate upon the growth of trees during spans when no other human dynamic indicesFig. 3.7 (continued) exist. A cycle with a period of over 500 years here shown was obtained in the course of studies reported earlier [15]. A similar cycle was also found, among others, in the spectrum of international battles (in log-transformed data) with a period of 499 years and a 95 % CI extending from 459 to 539 years, as also found in human creative cultural growth by Pales and Mikulecky [12, 13] (see Fig. 3.7a, b) (© Halberg Chronobiology Center). (d) Chart of about-500-year cycles in the emergence of great historians, physicians, and poets, compared with similar cycles in the Wheeler index of international battles, and in two series likely related to climatic changes, namely, tree ring widths and stalagmite coloring (© Halberg Chronobiology Center). (e) Acrophase chart of about-500-year cycles shown in Fig. 3.7d, estimated at average period (© Halberg Chronobiology Center)
3.3 So What?
That aeolian sunspots (actually their cycles) affect many aspects of our physiopathology was documented by Vallot et al., whose data are meta-analyzed in Fig. 3.8. The proposition that infectious diseases such as cholera may be influenced by the cycle in Wolf numbers is one of Alexander Leonidovich Chizhevsky’s many lines of evidence, sometimes aptly based on aeolian original data with solar cycles reporting millions of cases near the peak of one cycle of solar activity, while in another, no cases are recorded. On the average, there is a semidecadal (instead of a decadal) cycle [17]. With selected superposed epochs, Chizhevsky certainly made the case that infections in humans are influenced by solar activity, even though he only mentioned the spectra produced by his friend Vladimir Boleslavovich Shostakovich, which in the case of cholera in Moscow, in meta-analyses by Lyazzat Gumarova, shows that overall (globally), the harmonics of the circaundecennian sunspot cycle were more prominent than the fundamental, which was the one Chizhevsky illustrated as of interest, locally in time: the privilege of genius.
