Received: 21 July 2016 Accepted: 10 February 2017 DOI: 10.1111/jpn.12709 ORIGINAL ARTICLE Estimation of maintenance energy requirements in German shepherd and Labrador retriever dogs in Bangalore, India H. S. Madhusudhan 1 K. Chandrapal Singh 1 U. Krishnamoorthy 1 K. G. Umesh 2 R. Butterwick 3 D. Wrigglesworth 3 1 Department of Animal Nutrition, Veterinary College, Bangalore, India 2 Mars India International, Hyderabad, India 3 WALTHAM Centre for Pet Nutrition, Leicestershire, UK Correspondence Dr. H. S. Madhusudhan, Department of Animal Nutrition, Veterinary College, Hassan, Karnataka, India. Email: hsmsgowda@gmail.com Funding information WALTHAM Centre for Pet Nutrition, England Summary Maintenance energy requirements (MERs) were calculated for 17 German shepherd and 20 Labrador retriever adult dogs using an in- home prospective dietary trial. The dogs were fed the same dry pet food and body weight, food intake, body condition score and physical activity were monitored for 10 weeks. Labrador retrievers were significantly heavier and had higher body condition scores than German shepherd dogs, but there was no difference between males and females within each group. Body weights remained stable over the study period, with an average daily gain of 9.1 g. Mean (SD) MER was 103.4 (16.3) kcal/kg BW 0.75, which was some 20% lower than that currently suggested for moderately active young adult dogs. Individual MER ranged from 66.8 to 141 kcal/kg BW 0.75. There were no significant differences in MER between the two breeds, or between males and females within and between the two breeds. There was a significant inverse relationship between MER and body condition score, reflecting the lower energy expenditure of adipose tissue. The lower MER of dogs in this study, relative to previous observations, may reflect climatic and environmental differences and highlight the necessity for accurate estimates of MER in relation to the production and feeding of pet foods. KEYWORDS maintenance energy requirement, Labrador retriever, German shepherd 1 INTRODUCTION The issue of canine energy requirements has occupied physiologists and nutritionists for well over 100 years, ever since Rubner published the first predictive equations for the metabolisable energy (ME) content of foods for dogs in 1901 (Rubner, 1901). Understanding energy intakes is important since it provides the foundation for the formulation of nutritionally complete and balanced pet foods, as well as the calculation of feeding guides. Metabolisable energy defines the amount of food energy that is available for utilisation by the body following digestion and absorption, plus losses in urine and faeces, and equates to energy expenditure. It is the value used to calculate and express energy requirements for growth, maintenance of adult body weight, physical activity, gestation and lactation in dogs as well as cats. Estimation of canine ME requirements is complicated by the great diversity in size, shape and body weight, which may range from 1 to 90 kgs or more, between dog breeds. Studies in animals varying in size from mice to elephants have shown that energy requirements are not related directly to body weight, but are more closely related to body weight raised to some power (i.e., BW x, also referred to as metabolic body weight; Brody, Proctor, & Ashworth, 1934). There has been considerable debate as to which power value best describes the relationship between body mass and energy requirement in dogs. Figures of between 0.64 and 0.88 have been proposed (Burger & Johnson, 1991), with current consensus being that 0.75 is the most appropriate estimate (National Research Council (NRC), 2006). Mean estimates of maintenance energy requirements (MERs) for adult dogs, which reflect an individual s utilisation of energy for thermoregulation, metabolic processes and spontaneous activity, have e106 2017 Blackwell Verlag GmbH wileyonlinelibrary.com/journal/jpn J Anim Physiol Anim Nutr. 2018;102:e106 e110.
MADHUSUDHAN et al. e107 ranged from 94 to 183 kcal/kg BW 0.75 (Patil & Bisby, 2001). Factors that influence MER, and thereby account for this variation, include breed, gender, age, lean body mass, activity and environmental temperature. As a guide, the NRC s Ad Hoc Committee on Dog and Cat Nutrition currently suggests a MER of 130 kcal/kg BW 0.75 for moderately active pet dogs (NRC, 2006). Several methods have been employed to calculate MERs in dogs, including direct and indirect calorimetry, double- labelled water and other stable isotope washout techniques, as well as digestibility studies and feeding trials where ME intake is calculated from a diet history or weighed record (Hill, 2006). None of these methods are universally ideal. Calorimetry and isotope washout studies are accurate but are expensive in terms of equipment and analytical costs, thus restricting their use to laboratory or experimental situations. Feeding trials arguably provide the best means of assessing energy requirements of free- living pets, but rely upon accuracy in recording food intakes and determining the ME content of the foods consumed, which has limited their contribution to assessing canine MER. The present study was designed to determine the MER of freeliving adult pet German shepherd and Labrador retriever dogs using a prospective in- home feeding trial and was conducted in the Indian city of Bangalore. The study design also allowed for the assessment of how MER was influenced by breed, gender, body condition score and activity level. 2 MATERIALS AND METHODS Dogs were recruited from clients of a private veterinary practice in Bangalore. A total of 46 healthy adult dogs, aged 14 60 months, were entered into the study; these comprised 22 German Shepherds and 24 Labrador retrievers. The majority of animals were sexually intact, the exceptions being one neutered German shepherd bitch, one castrated Labrador retriever dog and two neutered Labrador retriever bitches. All of the animals lived in single dog households. Dogs were excluded from the study if they became pregnant or the owners failed to comply with the trial protocol. TABLE 1 Nutrient composition a of the diet Nutrient Organic matter 94.3 Crude protein 23.5 Ether extract 8.2 Crude fibre 4.4 Total Ash 5.7 Gross energy (kcal/100 g) 451 Metabolisable energy (kcal/100 g) b 371 Percentage a On percent dry matter basis. The dry matter content of the diet as fed was 97.2%. b Calculated in separate digestibility feeding trials, following the equation of the American Association of Feed Control Officials (AAFCO, 2010). The feeding trial was conducted in Bangalore city (India), which is situated at 13 north latitude and 77 37 east longitude, at 899 m above m.s.l. The trial was conducted in two phases during the period between July and October. During this period, daily atmospheric temperature ranged from 20.7 C to 26.1 C, mean (SD) daily relative humidity was 73.6% (4.0), and mean (SD) monthly rainfall was 1.6 mm (0.6). The first phase was a baseline period of 30 45 days in which the dogs were introduced to the trial diet (Table 1, a commercial dry pet food Pedigree (Mars International India Private Limited, Hyderabad, Telangana State, India) Adult Maintenance), and the feeding of snacks, treats and other foods discontinued. This was followed by the second experimental period of 10 weeks when the trial diet was fed solus and data collected. Initial daily food allowances were calculated to provide adequate or slightly in excess of ME requirements according to NRC (2006) guidelines. Owners were provided with a measuring cup and the daily ration offered in two to six meals. Any food left over the next day was sealed in polythene bags, and collected and weighed by the study investigators. But, there was no difference observed between breeds regarding food left over. Compliance with feeding procedures was assessed during visits by the investigators to each home every 3 4 days. Every day, the dog s daily activity was assessed based on a composite measure from owner s estimates of the time that the dog spent walking, playing, resting and sleeping each day. Body weight and body condition score were recorded at the practice by the same veterinarian every 2 weeks, following the 7- point WALTHAM S.H.A.P.E Guide for Dogs (German et al., 2006). Assessments of general health, coat condition and any ailments were also made at these times. Daily ME intakes were calculated from the owners feeding records. Where dogs had gained or lost the body weight during the trial phase, the ME contribution to the body weight change was calculated assuming a value of 7.92 kcal per gram of change in body weight (Laflamme & Kuhlman, 1995). This was then subtracted from the average daily ME intake to determine the ME intake for maintenance, which was expressed per kg BW 0.75 (Table 3). MER (KCal/Kg BW 0.75 ) 160.0 150.0 140.0 130.0 120.0 110.0 100.0 90.0 80.0 70.0 60.0 2.5 3 3.5 4 4.5 5 5.5 e German shepherd Labrador retriever Neutered FIGURE 1 Correlation between maintenance energy requirements (MERs) and body condition score in 37 dogs. (R square =.249, p < 0.05)
e108 MADHUSUDHAN et al. TABLE 2 Age (months, mean) Initial, final & mean body weight (kg, Mean ± SE), average daily gain in bodyweight (ADG, g), mean body condition score of dogs and activity (minutes per day, walking + playing) Breed Sex Mean age Initial BW Final BW Mean ADG BCS Activity German Shepherd Male 32 28.3 ± 1.9 29.1 ± 2.0 28.7 ± 2.0 11.6 ± 4.0 3.9 ± 0.2 138 ± 17 Female* 38 26.8 ± 1.4 27.7 ± 1.3 27.3 ± 1.3 11.9 ± 7.4 4 ± 0.1 103 ± 45 Mean 35 27.5 ± 1.1 a 28.4 ± 1.1 a 27.9 ± 1.1 a 11.8 ± 4.2 3.9 ± 0.1 a 118 ± 27 Labrador retriever Male* 34 33.9 ± 1.3 34.5 ± 1.3 34.3 ± 1.3 8.9 ± 5.8 4.3 ± 0.1 86 ± 11 Female 32 32.5 ± 1.5 32.8 ± 1.5 32.6 ± 1.5 3.6 ± 3.3 4.3 ± 0.2 117 ± 15 Mean 33 33.4 ± 1.0 b 33.8 ± 1.0 b 33.6 ± 1 b 6.8 ± 3.7 4.3 ± 0.1 b 99 ± 9 Overall mean 34 30.7 ± 0.9 31.3 ± 0.9 31.0 ± 0.9 9.1 ± 2.8 4.1 ± 0.1 106 ± 12 Values in a column bearing a different superscript letter differ (p <.05). *Includes one neutered animal. Includes two neutered animals. Relationships between variables were evaluated using Pearson correlation coefficients, and multivariate models tested using general linear model methodology. Differences between breeds and gender within and between breeds were analysed by analysis of variance. The level of statistical significant was taken as p < 0.05, unless stated otherwise. 3 RESULTS Data from nine dogs were excluded from analysis because four dogs became pregnant during the study and the owners of five dogs failed to comply with the study protocol. Relatively stable body weight was achieved in the remaining 37 dogs (17 German shepherd and 20 Labrador retriever), with a mean gain over the 10- week period of just 0.637 kg (range 2 kg loss to 3.5 kg gain). Mean daily times engaged walking or playing ranged from 45 to 371 min/day, and any weight change was unrelated to activity level. The Labrador retrievers were significantly heavier and had higher body condition scores, than their German shepherd counterparts (Table 2). There was no difference in body weight or body condition score between males and females within the breed, but there was a statistical difference in initial, final and average body weight between the breeds. Body weight gain was similar in both breeds, and there was no difference between males and females. Although there were no significant differences in activity levels between the two breeds, or between the two sexes within each breed, male German shepherds were significantly more active than the male Labrador retrievers. Mean (SD) MER was calculated to be 103.4 (±16.3) kcal/kg BW 0.75, with a range of 66.8 141 kcal/kg BW 0.75. There were no significant differences in MER between the two breeds, between male and females within each breed, or between male and females of the two breeds (Table 3). MERs were negatively correlated with average Body Condition Score (BCS) (p < 0.05) (Figure 1). Multivariate analysis, including breed, gender, initial body weight, body condition score and activity level, indicated that none of these factors was an independent predictor of MER. 4 DISCUSSION This is, to the best of our knowledge, the first report of MER in freeliving domestic dogs on the Indian subcontinent. The results bear testament to the contribution that prospective feeding trials can make to understand canine energy requirements and highlights important variations in MER among young adult (1 5 years of age) moderately active pet dogs. The average MER calculated in the present study (103.4 kcal/kg BW 0.75 ) is somewhat lower than that suggested by the NRC for moderately active young adult dogs (130 kcal/kg BW 0.75 ; NRC, 2006), which is supported by data from young (<6 years old) laboratory dogs of various breeds (Taylor, Adams, & Neville, 1995), young- to middleaged (1 7 years old) laboratory Huskies (Finke, 1991), and mediumsized pet dogs living in multiple dog households (Patil & Bisby, 2001). The observed value of 105.9 kcal/kg BW 0.75 is in fact comparable with mean values reported for older Labradors of 7 years of age upwards (Finke, 1991; Rainbird & Kienzle, 1990). Comparison with these studies suggests that the lower calculated MER in the present study is not readily explained by disparity in breed or age of dogs. This leaves the possibilities that differences in methodology, gender, lean body mass or activity of the dogs, or environmental temperature may be responsible. Using long- term feeding studies in which ME intakes were estimated from measurement of digestible energy intake, Finke (1991) found the average MER of six Labrador retrievers aged over 8.6 years to be 101 kcal/kg BW 0.75 and a value of 104 kcal/kg BW 0.75 was reported in 14 Labradors of greater than 7 years of age using similar methodology by Rainbird and Kienzle (1990). Data indicating that MERs are greater for younger dogs, similar to the age of the dogs in the present study, have been acquired from both digestibility (mean 132 kcal/kg BW 0.75 ; Finke, 1991) and calorimetry (mean 129 kcal/kg BW 0.75 ; Taylor et al., 1995) studies. Unfortunately, since no other MER data have been published from in- home feeding studies, the extent to which this technique might underestimate MER cannot be fully gauged. The reliability of prospective in- home feeding studies is entirely dependent upon rigorous compliance on the part of the dog owner
MADHUSUDHAN et al. e109 TABLE 3 Daily food intake (g, dry matter basis), metabolisable energy (ME) intake, ME for daily gain in body weight, ME for maintenance and ME per kg BW 0.75 of dogs Total ME intake ME for a daily gain ME for maintenance ME per kg BW 0.75b Breed Sex Total food intake kcal German shepherd Male 388.8 ± 18.4 1440 ± 68 91.9 ± 32 1348 ± 75 109.5 ± 5.3 Female 356.4 ± 16.2 1320 ± 60 94.3 ± 59 1226 ± 105 102.7 ± 8.2 Mean 371.7 ± 12.4 1376 ± 46 93.2 ± 34 1283 ± 66 105.9 ± 4.9* Labrador retriever Male 396.9 ± 8.1 1470 ± 30 70.7 ± 46 1399 ± 55 99.2 ± 3.6 Female 388.8 ± 0.0 1440 ± 0.0 28.3 ± 26 1412 ± 26 104.5 ± 4.3 Mean 393.7 ± 4.9 1458 ± 18 53.7 ± 29 1404 ± 34 101.3 ± 2.7* Overall mean 383.6 ± 6.5 1421 ± 24 71.9 ± 22 1349 ± 36 103.4 ± 2.7* Differences in the mean values between male and female within a breed and between the breeds for all parameters are non- significant (p >.05). a Assuming 7.92 kcal of energy is required per g gain in body weight. b All the values are significantly different from the NRC (2006) recommendation of 130 kcal/kg BW 0.75. *Values significantly different from 130.0 (p <.05). and his or her family, as well as frequent monitoring by the trial originators. Great care was taken in this study to ensure that dogs did not receive treats or obtain access to other foods, since this could have resulted in underestimation of MER from measurement of the trial food consumed. One other potential source of error is in determining the ME density of the food (Hill, 2006). This was not a concern in the present study because the ME content of the trial food was directly measured in digestibility trials and not predicted from its nutrient profile. With regard to the physiological determinants of MER, it is widely assumed that variations associated with breed, gender, age and activity are largely due to interaction between these factors and their influence of these factors on lean body mass (NRC, 2006). For instance, declining MER with age may well reflect reduced activity and subsequent loss of lean body mass. Differences in body composition may also explain why breeds of similar size, such as Great Danes and Newfoundlands, can differ greatly in their MER (Kienzle and Rainbird, 1991). Adipose tissue is metabolically much less active than lean body mass, and weight gain in dogs is associated with declining MER on a per kg BW 0.75 basis; increasing the body fat content of beagles from 25% to 38% was associated with a decline in energy expenditure from 130 to 107 kcal/kg BW 0.75 (Pouteau et al., 2000). There was a significant correlation between MER and body condition score in the present study, such that MER declined with increasing adiposity. The average body condition score for the German shepherds and Labrador retrievers were 3.9 and 4.3, respectively, on a scale where 4 represents an ideal body condition with a normal amount of body fat and 5 represents mildly overweight with a small excess of body fat. Given that none of the dogs had mean body condition scores of greater than 5, it is observed that the lower MERs in the present study were related to moderate or severe obesity. The higher BCS in Labrador retriever might have resulted in a feed allowance being higher than actual MER. The dogs in the present study were reasonably active, with mean daily physical activity totalling between 45 min and over 6 hr, and so the relatively low MER cannot be simply explained by very sedentary lifestyles. The two groups of dogs were also well balanced with respect to males and females, making it unlikely that a predominance of one or another gender may have unduly influenced the MER calculations. There remains the possibility that it is environmental and climatic factors that largely explain why the MER of these dogs are lower than those in other studies. Differences in housing between pet and laboratory animals may be important in determining energy expenditure and hence MER. The climate in the region where this study was performed could also be significant since annual mean maximum and minimum temperatures are 39 C and 8 C, respectively, with humidity ranging between 57% and 91%. These conditions contrast with those of northern Europe and north- west United States where much of the previous work on canine MER has been undertaken. While this study indicated that, for young adult pet dogs on the Indian subcontinent, mean MER are some 20% lower than standard recommendations (NRC, 2006), it should also be borne in mind that there was substantial variation around this mean with individual values ranging from 66.8 to 141 kcal/kg BW 0.75. Such interindividual variation is not unique to this study, with others reporting SD typically equivalent to 10% of the mean value with ranges of between 4% and 42% cited (NRC, 2006). Substantial over- or underestimations of energy requirements can result in under- or overfeeding with consequent loss of body weight or risk of obesity. This study has highlighted the fact there can be substantial variation between individuals, as well as between geographic regions, that should to be considered in the formulation of pet foods and their feeding guides. REFERENCES AAFCO (2010). Official Publication. The Association of American Feed Control Officials. Illinois. Retrieved from http://www.aafco.org/publications/ PublicationListing
e110 MADHUSUDHAN et al. Brody, S., Proctor, R. C., & Ashworth, U. S. (1934). Growth and Development with Special Reference to Domestic Animals. XXXIV. Basal Metabolism, Endogenous Nitrogen, Creatinine and Neutral Sulphur Excretions as Functions of Bodyweight. Agricultural Experiment Station Research Bulletin No. 220. Columbia: University of Missouri. Burger, I. H., & Johnson, J. V. (1991). Dogs large and small: The allometry of energy requirements within a single species. Journal of Nutrition., 121, S18 S21. Finke, M. D. (1991). Evaluation of energy requirements of adult kennel dogs. Journal of Nutrition., 121, S22 S28. German, A. J., Holden, S. L., Moxham, G. L., Holmes, K. L., Hackett, R. M., & Rawlings, J. M. (2006). A simple, reliable tool for owners to assess the body condition of their dog or cat. Journal of Nutrition., 136, 2031 2033. Hill, R. C. (2006). Challenges in measuring energy expenditure in companion animals: A clinician s perspective. Journal of Nutrition., 136, 1967S 1972S. Kienzle, E., & Rainbird, A. (1991). Maintenance energy requirement of dogs: what is the correct value for the calculation of metabolic body weight in dogs?. Journal of Nutrition, 121(11S), S39 S40. Laflamme, D. P., & Kuhlman, G. (1995). Effect of weight loss regimen on subsequent weight maintenance in dogs. Nutrition Research, 15(7), 1019 1028. National Research Council (NRC) (2006). Nutrient requirements of dogs and cats. Washington, DC, USA: National Academy Press. Patil, A. R., & Bisby, T. M. (2001). Comparison of maintenance energy requirement of client-owned dogs and kennelled dogs. St Louis, MO: Purina Nutrition Forum. Pouteau, E., Mariot, S., Martin, L., Dumon, H., Robins, R., Darmaun, D., Ngyuen, P. (2000). Effect of weight variation (fattening and slimming) on energy expenditure in dogs. Journal of Veterinary Internal Medicine, 14, 390. Rainbird, A., & Kienzle, E. (1990). Untersuchungen zum Energiebedarf des Hundes in Abhaengigkeit von Rassezugehoerigkeit und Alter [Investigations on energy requirement of dogs in relation to breed and age]. Kleintierpraxis, 35, 145 158. Rubner, M. (1901). Der Energiewert der Kost der Menschen [The energy value of human food]. Zeitschrift fur Biologie, 42, 261 308. Taylor, E. J., Adams, C., & Neville, R. (1995). Some nutritional aspects of ageing in dogs and cats. Proceedings of the Nutrition Society, 54, 645 656. How to cite this article: Madhusudhan HS, Chandrapal Singh K, Krishnamoorthy U, Umesh KG, Butterwick R, Wrigglesworth D. Estimation of maintenance energy requirements in German shepherd and Labrador retriever dogs in Bangalore, India. J Anim Physiol Anim Nutr. 2018;102:e106 e110. https://doi.org/10.1111/jpn.12709