Body temperature is one of the vital signs and it is a complex clinical variable, which can be captured accurately and quantitatively analysed 1-2. According to German physician Wunderlich, the normal body temperature is defined as 37 0C and fever as 38 0C 3.
A healthy, resting adult human normal core body temperature is 37 0C. However body temperature is not constant and varies among individuals throughout the day, because of individual’s metabolism rate, which is directly proportional to the normal core body temperature, time of a day or part of the body in which the temperature measured at, in the early morning the body temperature is lower and in late evening it is high due to after muscular activity and food intake. Body temperature also varies at different sites.
In clinical practice the rectal, oral, axillary, forehead and ear are used to measure body temperature. An oral site, which is more convenient to measure temperature is at 37 0C. Axillary site is not accurate to measure the temperature, where temperature fall at least value36.4 0C is noted from this site. Generally rectal temperature is considered to be the gold standard for core body temperature and average temperature is fall at 37.60C.Being an internal core body temperature, it is least time consuming procedure. The temperature is higher than at other sites, due to the low blood flow and high isolation of the area, giving a low heat loss16. Rectal temperature measurement is unhygienic and can pose a risk of injury to the intestinal mucosa, especially in infants and in rectal surgery. It increases physical and psychological stress and can cause embarrassment, anxiety and physical discomfort17. The tympanic is a good site for non-invasive measurement of core body temperature. However care should be exercised with the different modes of operation offered.18
When summarizing studies with able or adequately able affirmation, the ambit for articulate temperature was 33.2-38.2 0C, rectal: 34.4-37.8 0C, tympanic: 35.4-37.80C. The ambit in articulatetemperature for men and women, respectively, was 35.7-37.7and 33.2-38.1 0C, in abdominal 36.7-37.5 in tympanic 35.5-37.5 and35.7-37.5 0C1. Mackowiak et al. in 1992 recorded the body temperatures of 65 men with the average value of36.8 0C (98.2 0F).4
The thermometer is one of the most attempted and trusted clinical instruments, yet surprising surround the information which it yields. The use of thermometer in clinical medicine was started in the middle of 19th century. However its understanding and significance of temperature measurement in health and diseased condition was occurred from past twenty decades 19.
Thermometer is invented in seventeenth-century, it did not reach medicine until the 1870’s, it was already in veterinary use because it provided an early diagnosis of the dreaded cattle plague-as discovered 100 years earlier by a French veterinary student. On the other hand the existence of fever had been recognized since 600 B.C. For most of this time, fever was believed to be beneficial, even to absurd degrees. `If there were a physician skilful1 enough to produce a fever it would be useless to seek any other remedy against disease’-according to Rufus of Ephesus in A.D. 100. By the 17OO’s, however, the ability of willow bark to reduce fever became known and, as ever, once an effective drug was available, excellent use was found for it. The ability to control fever fostered the belief that it ought to be controlled, or at least that it was unhelpful, without much real evidence: `The role of fever in disease remains unexplained. Fever may eventually be shown to confer a greater advantage to the defence mechanism of the host than to the invasive properties of the microorganism19.
In 1861, Carl Wunderlich was the first German physician performed the systemic measurement of human core body temperature in healthy individuals, the average reported value was 37 0C or 98.6 degree Fahrenheit. Because of his work on temperature Wunderlich is generally regarded as the father of clinical thermometry20-21. According to Wunderlich, normal body temperature lies within a range of 97.2 0F/36.2 0C to 99.5 0F/37.5 0C. Wunderlich found that the body temperature is not constant and varies in both healthy and unhealthy individuals. He wrote, “The lowest point is reached in the morning hours between two and eight, and the highest in the afternoon between four. In his investigation the body temperature rises in mental exertion, constipation and urine retention.He observed that women have slightly higher body temperature than men and among age groups; older individuals have significantly lower body temperature compare to younger individuals20.
Body temperature is influenced by several factors, such as diurnal variation and cellular metabolism, due to muscle activity during the day exercise and ambient temperature 22-23.
Daily body temperature is not dependent on site of measurement, which is non-linear, and characterized by moment to moment complex variability 4 .The cosinor analysis of temperature variability data is well established in circadian research of body temperature rhythms, which is described in a simple cosine wave, which is typically characterized in terms of acrophase, amplitude, and mesor, where it filtered out the complex variability data4.
Under natural conditions expected timing of the nadir and acrophase , value of the mesor, and amplitude of temperature rhythm was significantly different in an individual’s temperature rhythm, which will be influenced by many endogenous countenance of the environments as well as health status4.
During menstrual period, there is an increase in body temperature ranging from 0.5-1.0° F/0.25-0.5° C is typically observed at or around ovulation(ref- Circannual and menstrual rhythm characteristics in manic episodes and body temperature. in note.) . Comparison of between follicular phase and the post-ovulation luteal phase, body temperature is elevated, but the amplitude of the temperature rhythm is reduced (ref- Circadian rhythms, sleep, and the menstrual cycle. In word doc.). It has been reported that the temperature in luteal stage is 0.4 0C higher than follicular stage (26 ref- in note ). (Check original ref no. in protocol.) This is mainly because of the progesterone hormone level in luteal stage and some studies proved that rise in body temperature is due to effect of progesterone hormone whereas oestrogen has lowering effect. the temperature remain elevated in luteal phase as long as the progesterone levels are increased. ( ref-29 and 33 in note.) )(check 27-28 original ref )
Physical fitness varies the normal body temperature; Atkinson G et al found that the physically active groups have higher amplitude of temperature than inactive groups. However the oral temperature of physically active group had lower than inactive group at 2.00 and 6.00. Rest of the time physically active group had significantly higher oral temperature than inactive group.
Age has also an important role in variability of temperature. In most of the studies reported that cosinor analysis of temperature; mesor and amplitude decreases with increase in age. Gubin et al reported that normal temperature range is higher in young adults than in elders. Mesor is higher in young adults (97.5° F/36.38° C) than in elderly subjects (97.1° F/36.17° C) and amplitude was also increased in young adults than in elderly subjects. The mean circadian acrophase was similar in both age groups (17:19 versus 16:93); however, inter-individual differences were higher in the older group, with individual values varying between 10:00 and 23:00 hours (ref.-48 in note ). In another study Howell et al recorded the oral temperature using electronic thermometer in 105 females age ranges from 61-105 years and reported a group mean of 36 0C,which is significantly lower than in a younger adults.(29 in original protocol). Touitou et al, found that the daily body temperature amplitude was decreased in the elderly subjects when compared to healthy adult individuals.(30 in original protocol).
Nonlineardynamics and complexity theory appear to offer an alternative approach. Many biological structures can be regarded as natural fractals and much physiological behaviour can be explained by deterministic chaos (heart rate, bacterial population growth, hormonal secretion pulses, and epidemics. Furthermore, in many cases, pathological conditions and ageing are known to be accompanied by losses in complexity. The complexity of the temperature curve could be regularly measured in all cases. Consistent results were obtained using three entirely independent methods that measure different aspects (in two cases the dynamic behaviour, in the other the anfractuosity of the curve), with good correlation between all methods. None of the complexity measures was significantly different for the two sexes, nor were they affected by the BMI. On the other hand, all measures of complexity were inversely correlated with age. This finding is somehow “coded” in the temperature readings, but classical statistical indicators are not capable of bringing it to light. The finding of an inverse correlation between age and the degree of complexity of the temperature curve was likewise not unexpected. Ageing and illness are known to be accompanied by a loss of complexity in certain patterns of chaotic behaviour12. For instance; variability in heart rate decreases with age and in certain conditions is associated with a poor prognosis31. It could be argued that, as has been postulated for heart rate, body temperature is governed by several different regulatory systems (thermogenesis, vasoconstriction- vasodilatation mechanisms, sweating, breathing rate) and at the same time is subject to external factors (ambient temperature, exercise, clothing). Perhaps illness and ageing cause a certain decoupling or isolation of the thermal regulatory system from its surroundings. This in turn could result in less complexity of the temperature plot, leading to lower ApEn and FDc values and higher DFA values12. Varela et al. reported that in healthy subjects, the temperature curve behaves like a natural fractal whose complexity may be analyzed in a consistent manner. In addition, they observed that complexity decreased significantly with age.
The complexity of the temperature curve is tightly inversely correlated with the severity of the patient’s condition. Both mean and minimum ApEn were significantly lower in patients who died than in patients who survived. Consequently one would expect to see a reduced complexity in the temperature readings of critically ill patients, the level of complexity mirroring the patient’s clinical evolution. In this respect, the mean ApEn value for the patient series was significantly lower than the mean ApEn for a series of 21 healthy subjects. The inverse correlation between the ApEn values and the SOFA scores in most of thein their series was likewise consistent with that premise32.
In another study of Varela et al. reported that there was good correlation between complexity results and clinical scores for each patient. Non survivors exhibited lower complexity values than survivors, so low levels of complexity in the temperature curve are poor indicators of prognosis in patients with multiple organ failure. The predictive ability of temperature curve complexity is similar to that of the SOFA score33.
Available evidence suggested that body temperature is a complex, non-linear physiological variable and has an accepted 24 hour rhythm associated with health. Body temperature is also subject to many sources of endogenous and exogenous variation4.
Temperature curve analysis may provide relevant information on the aetiology of fever thus may assist in early diagnosis of disease 12. There are few cases where rectal temperature is measured intermittently using thermometer to generate fever patterns but it is a tedious procedure and has a limited usefulness in diagnosis of certain clinical conditions such as, double quotidian fever curves in diagnosis of mixed malarial infections, visceral leishmaniasis, right-sided gonococcal endocarditis and sustained fever patterns in typhoid fever 13. Musher D M et al found that the fever pattern is not likely to be helpful in diagnosis of sustained fever in Gram-negative pneumonia or in CNS damage with possible exceptions.(new ref in fever pattern).
Papaioannou et al studied temperature curve complexity using wavelet transformation in 22 patients with systemic inflammation found that there is a decrease in complexity of temperature especially more in sepsis condition. They suggested that complexity analysis of temperature signals can help in assessment of inherent thermoregulatory dynamics during systemic inflammation and also can increase discriminating value in patients with infectious versus non-infectious conditions, probably associated with severity of illness.
However monitoring of 24 hour ambulatory core body temperature so far has been limited and still remains obscure. Studies have shown that core body temperature is not constant, and fluctuates in different clinical conditions and in various endogenous and exogenous factors, where variability of core body temperature patterns is noted. However assessment of 24 hour core body has not been established in detail. Thus standardization of core body temperature using spectral analysis might play a significant role in clinical practice, which would potentially help us to predict clinical outcome in the early part of fever in patients and with other associated clinical conditions.