Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Is propensity to obesity associated with the diurnal pattern of core body temperature?



Obesity affects more than half a billion people worldwide, but the underlying causes remain unresolved. It has been proposed that propensity to obesity may be associated with differences between individuals in metabolic efficiency and in the energy used for homeothermy. It has also been suggested that obese-prone individuals differ in their responsiveness to circadian rhythms. We investigated both these hypotheses by measuring the core body temperature at regular and frequent intervals over a diurnal cycle, using indigestible temperature loggers in two breeds of canines known to differ in propensity to obesity, but prior to divergence in fatness.


Greyhounds (obesity-resistant) and Labradors (obesity-prone) were fed indigestible temperature loggers. Gastrointestinal temperature was recorded at 10-min intervals for the period of transit of the logger. Diet, body condition score, activity level and environment were similar for both groups. Energy digestibility was also measured.


The mean core body temperature in obesity-resistant dogs (38.27 °C) was slightly higher (P<0.001) than in obesity-prone dogs (38.18 °C) and the former had a greater variation (P<0.001) in 24h circadian core temperature. There were no differences in diet digestibility.


Canines differing in propensity to obesity, but prior to its onset, differed little in mean core temperature, supporting similar findings in already-obese and lean humans. Obese-prone dogs were less variable in daily core temperature fluctuations, suggestive of a degree of circadian decoupling.


The epidemic of obesity, and its debilitating sequelae of metabolic syndromes in humans, is estimated to continue to afflict more than half a billion people globally1 and yet an understanding of the pathogenesis of obesity remains elusive. While at its simplest level obesity must arise from energy intake exceeding energy utilisation (the so-called ‘gluttony and sloth’ syndrome), a more detailed analysis suggests that obesity involves a complex interaction of genetics, sex, epigenetics, diet selection, digestion/absorption, activity, ambient temperature and thermogenesis (see reviews by Pijl2 and Landsberg3). The ‘thrifty gene’ hypothesis of Neel4 proposed a genetic basis to propensity to obesity. In essence, it speculated that genes associated with the efficient storage and retention of excess energy accumulated in evolution, to allow individuals to survive during periods of food insecurity. The development of agriculture and global trade in food removed restrictions on seasonal and year-to-year food availability. Individuals whose energy-retention mechanisms were previously an advantage were thus now prone to being overweight or obese. The nature of these ‘efficiency’ mechanisms is yet to be identified, but may involve a combination of lower basal metabolic rate and insulin resistance.5 Almost two-thirds of the resting metabolic rate is expended to achieve homeothermy, that is, the maintenance of a relatively ‘constant’ core body temperature.6, 7 In sedentary humans this equates to approximately 50% of the total energy expenditure, and even in active humans homeothermy still accounts for approximately 37% of the total energy expenditure.8 Small changes in core body temperature therefore may significantly influence total energy expenditure; for example, it has been estimated that an increase in core body temperature of as little as 1 °C is associated with an increase in metabolic rate of 10–13%.9 Several lines of evidence support an association between obesity and core temperature. Obese (ob/ob) mice are hypothermic compared with lean controls10 and Australian aboriginals have both lower night body temperatures11 and a predisposition to obesity. There are, however, conflicting reports of this association, with some studies supporting a relationship12, 13, 14, 15 and others showing no association.16, 17 A recent report concludes that there are only small differences in core temperature of obese and lean humans, with obese individuals having slightly higher core body temperatures.18

There is currently increasing interest in the notion that metabolic syndromes and obesity are associated with desynchronisation or decoupling of circadian rhythms.19, 20 In other words, obese individuals appear to be less responsive to environmental cues. Diurnal variations in body temperature, for example, are diminished in obese humans21 and obese dogs.15

Most, if not all, studies of obesity and its causes compare obese with lean individuals, thereby confounding cause and effect. It is also difficult to match cohorts of human subjects for diet, environment and level of exercise in such studies. Dogs are a valuable model for studies of obesity in that the genetic variation (between breeds) in propensity to obesity is large. One can also control experimental conditions such as diet, exercise and environment to a much greater extent in dogs than in human subjects. The advent of accurate, large-storage, indigestible temperature loggers (Star-Oddi, Reykjavik, Iceland) allows body temperature to be logged over the period of transit of the logger through the gastrointestinal tract with no interference with the animal’s normal behaviour. We measured the circadian variation in gastrointestinal transit temperature (GITT) in dog breeds differing in propensity to obesity (Greyhounds versus Labradors) of similar bodyweight, body fatness, environmental temperature, diet composition, dietary intake and activity levels. Our hypothesis was that obese-prone dogs (Labradors) would have a diminished diurnal pattern of core body temperature and a lower core temperature than dogs resistant to obesity (Greyhounds).

Materials and methods


This trial was conducted with the approval of the University of Adelaide Animal Ethics Committee (approval number S-2012-012). Nine Greyhounds and ten Labradors were used in the trial. The animals were sourced from individual trainers (Greyhounds) and the Royal Society for the Blind Pty Ltd (Labradors). As far as was possible the dogs were matched for sex, age, bodyweight and body condition score, and were housed in individual pens indoors overnight (Table 1). Body condition score was assessed using a 1–5 scoring system (with half scores allocated), with 1 being extremely lean and 5 being grossly obese, based on visual fat inspection and palpation of the ribs, lumbar vertebrae and pelvic bones.22 It was not possible to match groups perfectly (particularly for condition score, because Greyhounds are so lean). The small differences that did exist were tested for effects on gastrointestinal transit temperature using a statistical method and found to be non-significant (see below). All animals had a brief period of light exercise (walk on a lead) at the same time of day (0900–1100 hours). All dogs were fed the same diet (Royal Canin Maxi Adult, Royal Canin Australia, Mount Waverley, VIC, Australia; Royal Canin containing 3989 kcal kg−1 and CP 26%) during the trial period, at the recommended daily intake for each dog based on the manufacturer’s recommendation on a bodyweight basis. The small differences in bodyweight meant total intake varied slightly between dogs. Food was supplied twice daily, once at 0700–0900 hours and again at 1700–1900 hours, and intake was recorded. Water was available ad libitum. Ambient temperature in the pens was logged automatically with a temperature and humidity logger (Thermocron Humidity logger buttons, OnSolution Pty Ltd, Baulkham Hills, NSW, Australia) placed close to the animals.

Table 1 Characteristics of dogs used in the trial

Gastrointestinal transit temperature

The indigestible Star-Oddi DST Micro-T (Star-Oddi) temperature loggers (8.3 mm × 25.4 mm) were validated against a traceable thermometer (CHY 805 RTD thermometer, CHY Firemate Co. Ltd, Tainan City, Taiwan, ROC) in a water bath with varying water temperatures. The water bath temperature was changed at regular intervals (10 min) over a range of temperatures (approximately 35–40 °C). The loggers were set to record temperature every 10 min to coincide with the reading of the traceable thermometer. The temperature readings from both the indigestible logger and the traceable thermometer produced a scatterplot of the two variables. The regression relationship between thermometer and logger was derived for each logger and used to correct subsequent data from each individual logger (Figure 1).

Figure 1

Relationship between logger temperature and temperature recorded by a traceable RTD thermometer recorded in a water bath set at temperatures from 35 to 40 °C. Temperatures were recorded at 10-min intervals coinciding with the times set for the loggers. Values are means±s.e.m. for the 10 loggers. The equation relating logger temperature to thermometer temperature was y=0.977x+1.899. The r2 values for the relationship for individual loggers ranged from 0.913 to 0.986 and none of the relationships differed significantly from the line Y=X.

The dogs were ‘fed’ the loggers at approximately 0900 hours. This was successfully achieved by feeding two boluses of mince (40 g each) prior to a third bolus (40 g) containing the logger. The dogs swallowed the loggers without chewing or damage to the loggers. The loggers were pre-programmed to start recording at 0900 hours at 10-min intervals. All faeces were collected for digestion studies, so recovery of all loggers was readily achieved. Once retrieved, the loggers were washed with water and ethanol, and then dried to retrieve data using the Sea-Star computer programme (Star-Oddi).

Digestive efficiency

All feed provided was consumed by all dogs. Samples of the feed were taken regularly and all faeces were collected in the trial period (4–6 days). Faeces were stored at 2 °C until all samples were obtained for an individual. All faecal samples from each individual dog were placed into a container and mixed well. The dry matter (%) content of two subsamples (40–75 g) of faeces for each dog and of feed subsamples were determined by drying at 103 °C to constant weight. The remaining samples were bagged and stored at −20 °C. Dry matter digestibility was estimated using the following formula:

DMD (%)=[(Dry matter intake−Dry matter faeces)/Dry matter intake] × 100

The subsamples were then ground, pelleted and placed into a bomb calorimeter (Parr Oxygen Bomb Calorimeter 1281, Parr Instrument Co., Moline, IL, USA), where the energy value of each sample was determined (MJ kg−1). Gross energy intake (GEI) and gross faecal energy output (GFE) were estimated and the energy digestibility was calculated as follows:

Energy digestibility (%)=((GEI−GFE)/GEI) × 100

Data handling and statistical analysis

The first 80 min of temperature data after logger ingestion for each dog were discarded on the grounds of large variations, presumably associated with the warming of the loggers from room temperature to body temperature, and in response to food and water entering the stomach (Figure 2).

Figure 2

Variation in temperature (temperature at time n minus temperature at time n−10 min) for the Greyhounds (diamonds) and Labradors (squares) with time after pill ingestion. Values are means±s.e.m.

For dogs that retained the logger for more than 24 h, the additional data beyond 24 h were incorporated into a mean value for each 10-min interval over a ‘standard’ 24-h period. In other words, all the data available for each dog were used to construct its 24-h temperature cycle. This means that for some dogs the 10-min values represent just 1 datum, while for others there was >1 data value used to construct the mean 10-min values. GenStat statistical programme version 14 (VSN International, Hemel Hempstead, UK) was used to analyse the results. A two-way ANOVA for GITT with breed and time as variables was used to identify any differences between breeds in GITT over time. A general linear model was used to identify any effects of the small differences in variables such as condition score, ambient temperature in the different pens, body mass, age and sex. An independent t-test was used to identify breed differences in energy digestibility, dry matter digestibility, logger passage time and variables associated with variation in circadian rhythmicity. The timing of peaks in GITT was assessed by averaging the time of day corresponding to the highest four GITT values occurring in the morning (am) and afternoon (pm) periods for each dog. Data are presented as means±s.e.m.


The temperatures logged by the indigestible loggers were highly correlated with concomitant temperatures observed on a reference thermometer over the range 35–40 °C (r2 values ranged from 0.913 to 0.986 and none of the regression lines differed significantly from Y=X, where Y is the logger temperature and X is the reference temperature).

All loggers were recovered successfully from the faeces and all loggers recorded a temperature value at each of the 10-min intervals as programmed. Transit times of the loggers differed widely between dogs (Table 2). On average the transit time of the loggers was approximately 30 h (Table 2), which provides sufficient data to provide a comprehensive description of one 24-h cycle.

Table 2 Transit time for indigestible loggers in Labradors and Greyhounds

Figure 3 shows the changes in GITT for the two breeds over a 24-h cycle. There was a highly significant (P<0.001) effect of dog breed, time of day, and interaction between breed and time of day. The mean GITT of the Greyhounds was 38.27 °C and the Labradors 38.18 °C (P<0.001). The circadian pattern of GITT change was similar for the two breeds but the amplitude of GITT change was greater in the Greyhounds than the Labradors. Both peaked at approximately 1400 hours and reached a nadir between 0300 and 0500 hours. Cubic polynomials explained 70% (P<0.001) of the daily variance in GITT in Greyhounds, but only 49% (P<0.01) of the variance in Labrador GITT (Figure 3). In general the daytime GITT in Greyhounds was higher than that of Labradors but there was little difference in night time GITT. In addition to the diurnal wave of GITT, the dogs displayed two secondary peaks in GITT; the morning peak for greyhounds occurred at 0955 hours and for Labradors at 0849 hours. This difference in peak timing differed significantly (P<0.03). Similarly, in the afternoon, the Labrador peak (1612 hours) occurred earlier than the greyhound peak (1634 hours), but this difference was not statistically significant (P=0.60).

Figure 3

Gastrointestinal transit temperature (GITT) of Greyhounds (n=9/data point; black markers) and Labradors (n=10/data point; grey markers) over a 24-h period. Values are means at 10-min intervals during the gastrointestinal transit of an indigestible temperature logger. The fitted polynomials accounted for 0.70 and 0.49 of the variance in GITT for the Greyhounds and Labradors, respectively. The average standard error of difference is indicated as sed. The data have been transposed to correspond to the daylight hours. The time of logger ingestion is indicated by the arrow (0900 hours).

The maximum GITT and the variability of GITT as measured by standard error of the GITT throughout the day were greater in Greyhounds than in Labradors (Table 3). There was no significant difference in minimum GITT or diurnal range of GITT between dog breeds.

Table 3 Maximum, minimum, range and standard error of GITT for Labradors and Greyhounds

Digestive efficiency

There were no significant differences between the breeds in digestion of dry matter or energy. Dry matter digestibility was 0.834±0.018 and 0.824±0.047 (P>0.05) for the Greyhounds and Labradors, respectively. Digestibility of energy was 0.877±0.013 and 0.868±0.041 (P>0.05) for the Greyhounds and Labradors, respectively.


The use of canines differing in propensity to obesity, coupled with the use of indigestible temperature loggers, provided a powerful model for investigations of association between diurnal core temperature and obesity disposition. There is strong evidence that canine genotypes differ in their incidence of obesity.23, 24 Labradors and Labrador Retrievers are overrepresented in the overweight and obese categories,23, 25, 26, 27 whereas Greyhounds are underrepresented in the overweight and obese cohorts.27, 28 These cohort associations, of course, are epidemiological only and will remain so until the genes associated with obesity are identified and mapped to the canine breeds. Nevertheless, the diversity of obesity among dog breeds, and the ability to control environmental factors such as exercise, diet, age, sex, body condition and temperature/humidity, makes canines a good model of the underlying biological causes of obesity. To date all the studies of obesity, to the best of our knowledge, have compared obese with non-obese individuals, thereby confounding cause and effect (for example, Piccione et al.15 and Corbalan-Tutau et al.21). The current study is, to the best of our knowledge, the first to use canines differing in propensity to obesity, at similar levels of fatness, and with diet and other variables matched as closely as possible. Our hypothesis was that these animals would differ in the diurnal pattern of core body temperature, reflecting both a difference in homeothermy3, 5 and differences in circadian responsiveness.21, 29

The indigestible temperature loggers proved to be precise and accurate in comparison to a traceable, reference thermometer. They are a useful tool for studying homeothermy in animals, provided the data for the first hour or so are discarded (Figure 2). Much of this initial variation reflects the warming of the logger from 15 °C (room temperature) to body temperature, and could be reduced by warming the logger to body temperature before ingestion. Some of the initial variations, however, also reflected the effects of food and water intake, which generated large (>2 °C) changes over short periods of time, while the logger remained in the gastric stomach. After approximately 1 h these adjacent variations in temperature became minimal, presumably due to the logger moving to the small intestines. That the temperatures logged thereafter reflect the true core body temperature (defined as temperature in the pulmonary artery) remains equivocal. The loggers did, however, provide very similar values to a reference thermometer in a water bath test. Relative changes between dogs with time after logger ingestion can be confidently taken to represent real body temperature changes.

There was a wide range of gastrointestinal transit times of the loggers between dogs, ranging from 23 to 107 hours for the Labradors and from 23 to 53 hours for the Greyhounds. Given the uniformity between dogs in diet digestibility, there seems to be no reason to suggest differences in peristaltic activity, and it may be that the loggers become lodged for various times in various parts of the gastrointestinal tract of some animals. Radiographic imaging of the passage of the loggers or the use of pH and pressure changes to identify the location of the loggers30 might usefully identify the cause of variation in logger passage and may help to ensure that data are captured from passage through the same segments of the gut. However, there appears to be little effect of gastrointestinal site on temperature,30 so this is unlikely to be a significant source of error. All loggers provided a complete set of data with no apparent drop-outs. They produced data likely to be more accurate than rectal, aural or skin surface probes, and without the need for surgical intervention. More frequent data logging (as frequent as 1 s−1) is possible, but 10-min intervals proved adequate for the purpose of studying circadian temperature changes in the present study.

The current study investigated two hypotheses related to the aetiology of obesity. The first is that the metabolic cost of homeothermy differs between obese-prone and obese-resistant individuals. Small differences in core body temperature are associated with large differences in the metabolic cost to the animal. DuBois9 estimated that every 1 °C difference in core temperature is equivalent to a difference of 10–13% in metabolic rate, and by inference, the energy used to generate that temperature difference. Of all the studies on obesity and temperature, ours is the first, to the best of our knowledge, to compare animals before differences in fatness were apparent. This is important because differences in fatness are likely to significantly influence thermoregulation by insulation and through effects on activity levels, although the core temperature should be controlled by a homeostatic set point. We hypothesised that the obese-prone dogs (Labradors) would have a lower core body temperature than obese-resistant dogs (Greyhounds), and further, that it would be core temperature at night that would generate these differences. This hypothesis was based on evidence that so-called ‘thrifty’ phenotypes enter a state of mini-torpor during sleep31 thereby conserving more energy than lean phenotypes. Indeed the Labradors did have a lower core temperature than Greyhounds when averaged over a 24-h period (P<0.001), but the difference was small (0.09°C) and was generated largely during the daylight hours (Figure 3). Such a small difference is in accord with previous studies of obese versus lean humans18 and supports the conclusion that the small differences in metabolic rate likely to accompany such temperature differences (that is, approximately 1–1.3% of the metabolic rate)9 are unlikely to be a causative factor in obesity. This is an important finding because previous studies have compared already-obese versus lean humans. The fact that similar small differences in core temperature are apparent prior to the development of obesity suggests that differences in temperature set point and metabolic rate are likely to be negligible relative to total energy intake and possibly energy utilisation rates.

Reduced amplitude of circadian rhythms is increasingly being recognised as a feature of obese individuals,19, 20 a phenomenon known as circadian decoupling. In this regard, there were two major peaks of temperature in the Greyhounds, one at about 1000 hours and the other at about 1630 hours. The Labradors similarly showed secondary peaks of core temperature superimposed on the larger diurnal rhythm, but these occurred earlier than in the Greyhounds; the morning peak in the Labradors was significantly earlier (1 h) than that of the Greyhounds. This association between core body temperature variation throughout the day and propensity to obesity requires further analysis. Our results support the notion that circadian decoupling exists prior to the development of obesity. In other words, a reduction in circadian responsiveness, in our case as indicated by a reduction in diurnal amplitude of core body temperature, appears to be associated with propensity to obesity. Corbalan-Tutau et al.21 similarly found a lower mean wrist temperature and a flatter diurnal temperature in obese women compared with normal-weight women. That such disruptions to circadian rhythms are responsible at least in part for obesity and metabolic syndromes is indicated by targeted disruptions to the clock genes.32 Piccione et al.15 reported lower rectal temperatures in obese dogs than lean dogs of the same bodyweight, although the difference was of the order of 0.2 °C in their study, twice the difference we report in our study. These authors also reported a much more-defined diurnal rhythm in rectal temperatures of Labradors than we report. The indigestible temperature loggers we used recorded at frequent intervals (10 min) and are more likely to accurately reflect core temperature than rectal probes, so it is perhaps not surprising that our data are more sensitive to short-term fluctuations in core temperature. Indeed our data demonstrate interesting short-term wavelets of temperature change that would be difficult to detect by other means (Figure 3). The cause of the spike in temperature of the Greyhounds at 1000 hours is not clear but it did coincide with the mild exercise period and may reflect a strong sympathetically mediated anticipation response that produced an elevated heart rate and metabolism. No such spike was apparent in the Labradors. Again this may reflect differences between the breeds in sympathovagal regulation of the body systems, another indicator of environmental responsiveness. Indeed there is evidence of a reduction in heart rate variability and poorer autonomic nervous system function in obese individuals.33 Further studies of autonomic nervous system functioning and circadian decoupling in canines differing in propensity to obesity are warranted on the basis of our results.

In conclusion, we demonstrate, for the first time, that animals differing in propensity to obesity display different patterns of core body temperature throughout the day. It is unlikely that the small differences in mean temperature and associated energy utilisation for homeothermy would be sufficient to produce differences in energy retention between the breeds. On the other hand, the different patterns of diurnal temperature between the breeds are of interest and may reflect the underlying physiological mechanisms associated with the development of obesity.


  1. 1

    WHO. 2006.

  2. 2

    Pijl H . Obesity: evolution of a symptom of affluence. J Med 2011; 69: 159–166.

    CAS  Google Scholar 

  3. 3

    Landsberg L . Core temperature: a forgotten variable in energy expenditure and obesity? Obesity Rev 2012; 13 (Suppl 2): 97–104.

    Article  Google Scholar 

  4. 4

    Neel JV . Diabetes mellitus: a ‘thrifty’ genotype rendered detrimental by ‘progress’? Am J Hum Genet 1962; 14: 353–362.

    CAS  PubMed  PubMed Central  Google Scholar 

  5. 5

    Landsberg L, Young JB, Leonard WR, Linsenmeier RA, Turek FW . Do the obese have lower body temperatures? A new look at a forgotten variable in energy balance. Trans Am Clin Climatol Assoc 2009; 120: 287–295.

    PubMed  PubMed Central  Google Scholar 

  6. 6

    Landsberg L, Young JB . Autonomic regulation of thermogenesis. In Giradier L, Stock MJ, editors. Mammalian Thermogenesis 1st edn. Chapman and Hall: London, UK, 1983. pp 99–140.

    Chapter  Google Scholar 

  7. 7

    Girardier L, Stock MJ . Mammalian thermogenesis: an introduction. In Giradier L, Stock MJ, editors. Mammalian Thermogenesis 1st edn. Chapman and Hall: London, UK, 1983. pp 1–8.

    Chapter  Google Scholar 

  8. 8

    Westerterp KR, Speakman JR . Physical activity energy expenditure has not declined since the 1980s and matches energy expenditure of wild mammals. Int J Obes 2008; 32: 1256–1263.

    CAS  Article  Google Scholar 

  9. 9

    Dubois EF . The basal metabolism in fever. JAMA 1921; 77: 352–355.

    CAS  Article  Google Scholar 

  10. 10

    Davis TRA, Mayer J . Imperfect homeothermia in the hereditary obese-hyperglycemic syndrome of mice. Am J Physiol 1954; 177: 222–226.

    CAS  Article  PubMed  Google Scholar 

  11. 11

    Hammel HT, Elsner RW, LeMessurier DH, Andersen HT, Milan FA . Thermal and metabolic responses of the Australian aborigine exposed to moderate cold in summer. J Appl Physiol 1959; 14: 605–615.

    Article  Google Scholar 

  12. 12

    Rising R, Keys A, Ravussin E, Bogardus C . Concomitant interindividual variation in body temperature and metabolic rate. Am J Physiol 1992; 263: E730–E734.

    CAS  Google Scholar 

  13. 13

    Adam K . Human body temperature is inversely correlated with body mass. Eur J Appl Physiol 1989; 58: 471–475.

    CAS  Article  Google Scholar 

  14. 14

    Kim H, Richardson C, Roberts J, Gren L, Lyon JL . Cold hands, warm heart. Lancet 1998; 351: 1492.

    CAS  Article  Google Scholar 

  15. 15

    Piccione G, Giudice E, Fazio F, Refinetti R . Association between obesity and reduced body temperature in dogs. Int J Obesity 2011; 35: 1011–1018.

    CAS  Article  Google Scholar 

  16. 16

    Rising R, Fontvieille AM, Larson DE, Spraul M, Bogardus C, Ravussin E . Racial difference in body core temperature between Pima Indian and Caucasian men. Int J Obes 1995; 19: 1–5.

    CAS  Google Scholar 

  17. 17

    Ericksson H, Svardsudd K, Larsson B, Welin L, Ohlson LO, Wilhelmsen L . Body temperature in general population samples. The study of men born in 1913 and 1923. Acta Med Scand 1985; 217: 347–352.

    Article  Google Scholar 

  18. 18

    Hoffman ME, Rodriguez SM, Zeiss DM, Wachsberg KN, Kushnerr RF, Landsberg L, Linsenmeier RA . 24-hour core body temperature in obese and lean men and women. Obesity (Silver Spring) 2012; 20: 1585–1590.

    Article  Google Scholar 

  19. 19

    Marcheva B, Ramsey KM, Affinati A, Bass JJ . Clock genes and metabolic disease. Appl Physiol 2009; 107: 1638–1646.

    CAS  Article  Google Scholar 

  20. 20

    Wyse CA . Does human evolution in different latitudes influence susceptibility to obesity via the circadian pacemaker? Bioessays 2012; 34: 921–924.

    Article  PubMed  PubMed Central  Google Scholar 

  21. 21

    Corbalan-Tutau MD, Madrid JA, Ordovas JM, Smith CE, Nicolas F, Garaulet M . Differences in daily rhythms of wrist temperature between obese and normal weight women: associations with metabolic syndrome features. Chronobiol Int 2011; 28: 425–443.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  22. 22

    Bland IM, Guthrie-Jones A, Taylor RD, Jill J . Dog obesity: owner attitudes and behaviour. Prev Vet Med 2008; 92: 333–340.

    Article  Google Scholar 

  23. 23

    Lund EM, Armstrong PJ, Kirk CA, Klauser JS . Prevalence and risk factors for obesity in adult dogs from private practices. Int J Appl Res Vet Med 2006; 4: 177–186.

    Google Scholar 

  24. 24

    McGreevy PD, Thomson PC, Pride C, Fawcett A, Grassi T, Jones B . Prevalence of obesity in dogs examined by Australian veterinary practices and the risk factors involved. Vet Rec 2012; 156: 695–702.

    Article  Google Scholar 

  25. 25

    Edney ATB, Smith PM . Study of obesity in dogs visiting veterinary practices in the United Kingdom. Vet Rec 1986; 118: 391–396.

    CAS  Article  PubMed  Google Scholar 

  26. 26

    Kronfeld DS, Donoghue S, Glickman LT . Body condition and energy intake of dogs in a referral teaching hospital. J Nutr 1991; 121 (Suppl S): 157–158.

    Article  Google Scholar 

  27. 27

    Zoran DL . Obesity in dogs and cats: a metabolic and endocrine disorder. Vet Clin Small Anim 2010; 40: 221–239.

    Article  Google Scholar 

  28. 28

    Diez M, Nguyen P . The epidemiology of canine and feline obesity. Waltham Focus 2006; 16: 2–8.

    Google Scholar 

  29. 29

    Wyse CA, Selman C, Page MM, Coogan AN, Haelrigg DG . Circadian desynchrony and metabolic dysfunction; did light pollution make us fat? Med Hypoth 2011; 77: 1139–1144.

    CAS  Article  Google Scholar 

  30. 30

    Saad RJ, Hasler WL . A technical review and clinical assessment of the wireless motility capsule. Gastroenterol Hepatol 2011; 7: 795–804.

    Google Scholar 

  31. 31

    Kreider MB, Buskirk ER, Bass DE . Oxygen consumption and body temperatures during the night. J Appl Physiol 1958; 12: 361–366.

    CAS  Article  PubMed  Google Scholar 

  32. 32

    Turek FW, Joshu C, Kohsaka A, Lin E, Ivanova G, McDearmon E et al. Obesity and metabolic syndrome in circadian clock mutant mice. Science 2005; 308: 1043–1045.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  33. 33

    Windham BG, Fumagalli S, Ble A, Sollers JJ, Thayer JF, Naijar SS, Griswold ME, Ferrucci L . The relationship between heart rate variability and adiposity differs for central and overall adiposity. J Obes 2012; 2012: 149516.

    Article  PubMed  PubMed Central  Google Scholar 

Download references


We wish to thank the staff at the Royal Society for the Blind Pty Ltd South Australia for providing access to the Labrador Retriever guide dogs. In particular, we thank Dr Chris Muldoon, Ms Celeste Osmond, Ms Daisy Piccoli and the foster-carers of the guide dogs. Dr Jane McNicholl is also gratefully acknowledged for assistance in accessing Greyhounds for the study.

Author information



Corresponding author

Correspondence to P I Hynd.

Ethics declarations

Competing interests

The authors declare no conflict of interest.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Hynd, P., Czerwinski, V. & McWhorter, T. Is propensity to obesity associated with the diurnal pattern of core body temperature?. Int J Obes 38, 231–235 (2014).

Download citation


  • Canis familiaris
  • diurnal rhythm
  • body temperature

Further reading


Quick links