Aim The overall objective of this thesis was to increase the knowledge and validate methods for assessment of total daily energy expenditure (TEE), energy requirement and energy intake (EI) in women with COPD. We also investigated the relationship of BMI with clinical characteristics of COPD in a population-based study. Due to the substantial and increasing morbidity in women with COPD, special attention was given to this group. Methods Resting metabolic rate (RMR) was measured by indirect calorimetry (IC). TEE was measured by criterion method, doubly labeled water (DLW) (Paper I-III) during a 14-day period. TEE was simultaneously assessed by SenseWear Armband, software version 5.1 and 6.1 (SWA5 respectively SWA6), and ActiHeart (paper I). EI was assessed by diet history interview and 7-day food diary (paper II), and energy requirement was predicted using pedometer-determined physical activity level (paper III). Energy requirement data was also acquired from studies concerning TEE measured by DLW in patients with COPD (paper IV). BMI and other characteristics in subjects with COPD were compared with non-COPD subjects in a population-based study, Obstructive Lung disease in Norrbotten study (OLIN) (paper V). Results There was a large variation in RMR and TEE measured by DLW in this group of women with COPD. The results of energy expenditure study showed that the SWA5 assessed TEE with good accuracy over a 14-day period in free-living women with COPD. However, the SWA6 and ActiHeart methods tend to underestimate TEE. A higher proportion of women were within ± 5% of the TEE individually measured with the DLW method using SWA5 than SWA6 and AH (63%, 47%, 37% respectively). The agreement between the TEE measured by DLW and SWA5 was strong, and with SWA6 and ActiHeart it was lesser. Bland-Altman plots revealed no systematic bias for TEE. The reported EI was underestimated by 28% respectively 20 % when assessed by diet, and the 7-day food diary compared with the criterion method, DLW. More women were identified as valid-reporters based on their 7- day food diaries than on their diet histories (63% vs 32%). The accuracy of reported EI was only related to BMI. The agreement between the DLW and the EI methods was weak. The Bland-Altman plots revealed a slight systematic bias for both methods. The energy requirement predicted by pedometer-determined PAL multiplied by six different RMR equations was within a reasonable accuracy (±10%) of the measured TEE for all equations except one. The agreement between the DLW and four of six predicted TEE methods was strong. The Bland– Altman plots revealed no systematic bias for predicted energy requirement except for one. Estimated PAL from the pedometer was lower by 14 % than the measured criterion PAL. The energy requirement calculation based on available TEE data measured by DLW varied by BW and FFM. Compared to men, women had a lower RMR and TEE/kg BW/day, and higher RMR and TEE/kg FFM/day. The correlates of RMR/kg BW were gender and forced expiratory volume in 1st second (FEV1) % of predicted value, of TEE/ kg BW the correlates were age and gender, and of TEE/kg FFM were age and FEV1 % predicted. BMI decreased significantly with increase in disease severity and correlated significantly to forced expiratory volume in 1st second % predicted. In the population-based study (OLIN), subjects with COPD had lower BMI and a higher prevalence of underweight than in non-COPD, and its sub-groups namely, normal lung function and restrictive spirometry pattern subjects. There was an independent association between COPD and low BMI. Fewer COPD subjects were obese than in the non-COPD, normal lung function and restrictive spirometry pattern groups. Among the subjects with COPD, women had a lower mean BMI and a higher proportion were under-weight than men. In COPD women with under-weight, FEV1 % predicted values increased with an increase in BMI. Conclusion Compared with the gold standard DLW method, the total daily energy expenditure can be assessed reliably by SenseWear Armband 5 in women with COPD, while other devices underestimated TEE. The energy intake was underestimated by diet history and 7-day food diary methods, and energy requirement was predicted with reasonable accuracy using pedometer-determined PAL and common RMR equations, compared with DLW. Furthermore, the energy requirement was determined per kg BW/day and per kg FFM/day, using DLW based TEE data in patients with COPD. In the population-based study (OLIN), subjects with COPD had lower BMI and higher prevalence of under-weight than subjects without COPD. There was a gender difference, which was particularly significant in COPD, for women to have lower mean BMI and a higher prevalence of under-weight. The present findings indicate that low BMI is common in COPD and needs to be intervened. For a successful nutritional treatment, it is imperative to assess the patient’s ener / Bakgrund/Syfte Kroniskt obstruktiv lungsjukdom (KOL) är en av de stora folksjukdomarna där dödligheten ökar mest globalt sett. I Sverige är ökningen av KOL mest uttalad för kvinnor. Viktminskning är vanligt vid KOL. Låg kroppsvikt, låg andel fettfrimassa och lågt kroppsmasseindex (BMI) är dessutom riskfaktorer för ökad sjuklighet och dödlighet. Det finns därför behov av att ge anpassad kostbehandling for att motverka viktnedgång eller för att uppnå viktuppgång. För framgångsrik kostbehandlingen behövs metoder som på ett tillförlitligt sätt kan bedöma den totala energiförbrukningen, energiintag, och energibehov., Dessa metoder har ännu inte utvärderats väl hos KOL patienter. Det övergripande syftet därför med denna avhandling var att validera och öka kunskaperna om metoder som är tillförlitliga för att kunna bedöma energiförbrukning, energiintag och energibehov hos patienter med KOL. Vidare, jämfördes BMI och kliniska egenskaper mellan personer med och utan KOL in en populations-baserad studie. Metoder Viloenergiförbrukning mättes med indirekt kalorimeter hos kvinnor med KOL. Den totala energiförbrukningen mättes hos dessa kvinnor med dubbelmärktvatten metoden (DLW) (Arbete 1-3) under en 14-dagars period. Samtidigt bedömdes energiförbrukningen med två aktivitetesmätare, SenseWear Armband, mjukvaruversion 5.1 och 6.1 (SWA5 respektive SWA6), och ActiHeart (arbete 1). Energiintaget bedömdes med kostanamnes och 7-dagars matdagbok (arbete 2); energibehovet beräknades med olika viloenergiförbruknings formler och fysisk aktivitetsnivå uppskattad med en stegräknare (arbete 3). En sammanställning och en analys genomfördes av energibehovet baserad på data från flera studier som har mätt TEE med DLW hos KOL patienter (Arbete 4). BMI och kliniska egenskaper jämfördes mellan personer med och utan KOL i ett stort populationsbaserat material från Obstruktiv Lungsjukdom i Norrbotten (OLIN) studien (Arbete 5). Resultat Det fanns en stor variation in viloenergiförbrukningen och energiförbrukning mätt med DLW hos kvinnor med KOL. Energiförbrukningen mätt med aktivitetsmätare SWA5 visade hög tillförlitlighet både på grupp- och individnivå, medan SWA6 och ActiHeart metoderna underskattade energiförbrukningen. Energiintaget bedömd med kostanamnes och 7-dagars matdagbok visade en underskattning med 28% respektive 20% jämfört med DLW metoden. Det rapporterade energiintaget stämde för flertalet kvinnor bättre mot DLW metoden när energiintaget bedömdes med 7-dagars matdagbok jämfört med kostanamnes (63% mot 32%). Det beräknade energibehovet från stegräknare och från fem av sex olika RMR ekvationer låg inom en rimlig marginal (± 10%) jämfört med den uppmätta energiförbrukningen med DLW. Beräkning av energibehov baserat på tillgänglig data om energiförbrukning mätt med DLW varierade utifrån kroppsvikt och fettfrimassa hos patienter med KOL. Kvinnor hade en lägre viloenergiförbrukning och energiförbrukning per kg kroppsvikt och dag, och högre daglig viloenergiförbrukning och energiförbrukning per kg fettfrimassa och dag jämfört med män. I den populationsbaserade studien hade personer med KOL signifikant lägre BMI och högre prevalens av undervikt än personer utan KOL. Det fanns ett oberoende samband mellan KOL och lågt BMI. Färre personer med KOL var överviktiga än personer som inte hade KOL. Kvinnor med KOL hade både lägre BMI och högre andel med undervikt än män med KOL. Slutsats SenseWear Armband med programvaran 5.1 visar sig kunna beräkna den totala energiförbrukningen på ett tillförlitligt sätt hos kvinnor med KOL. Energiintaget beräknat utifrån kostanamnes och 7-dagars matdagbok underskattades. Energibehovet kan beräknas med en rimlig precision med hjälp av stegräknare och viloenergiförbrukningsformler hos dessa kvinnor. De sammanställda data om energiförbrukning mätt med DLW kan användas för att beräkna energibehov per kg- kroppsvikt och fettfri massa hos patienter med KOL i kliniskt syfte. I det populationsbaserade materialet från OLIN-studien noterades ett lägre BMI och högre prevalens av undervikt hos personer med KOL, och detsamma resultat förekom hos kvinnor med KOL jämfört med män med KOL.

The aim of this study was to determine energy expenditure and to find relations between resting energy expenditure (REE) and selected parameters in 14 polytraumatic patients in the ICU of University Hospital in Hradec Kralove, which were easily measurable and usable for REE prediction. In this study 7 men (age 36 ± 18 years) and 7 women (age 58 ± 28 years) with polytrauma were examined. The assessment of REE was measured via indirect calorimetry (IC) method. The examination also included bioimpedance analysis (BIA). BIA was useful especially for obtaining values of overhydration (OH), lean tissue mass (LTM) and metabolically active body cell mass (BCM). Average REE-IC (measured by IC) was 2116 ± 516 kcal·day-1 in men and 1450 ± 407 kcal·day-1 in women (P = 0.018). Statistically significant difference between men's and women's population was also found in these relations: calculation of basal energy expenditure according to Harris-Bennedict equation without (P = 0.001) and with deduction of OH from body weight (P = 0.001), at "breathing energy expenditure" (REE related to respiratory rate) (P = 0.018) and at (REE related to heart rate) "heart rate energy expenditure" (P = 0.038). REE-IC related to kilogram of BCM with and without deduction of overhydration was shown as statistically significant...

An exercise-induced energy deficit may affect post-exercise energy intake, physical activity energy expenditure (PAEE) and energy balance hormones. Therefore, the objective of this thesis was to investigate the impact of a single bout of exercise either of moderate (40% O2max) or high (70% O2max) intensity on post-exercise energy intake, physical activity energy expenditure and energy balance hormones in both sedentary and active males. Physical activity energy expenditure increased between 38 and 62 hours following moderate intensity exercise in sedentary males (Chapters 3 and 4). This was due to increased light intensity energy expenditure (2.4-4.79 METs) such as standing and walking activities (Chapter 4). The Change in PAEE was not associated with circulating leptin and adiponectin concentrations. There was no impact of a single bout of exercise on post-exercise energy intake in sedentary males during a buffet meal (Chapter 4). Chapter 5 aimed to determine whether changes in PAEE, energy intake, and energy balance hormones were related to physical activity status. Interestingly, plasma acylated ghrelin concentration was suppressed while total peptide YY (PYY) concentration tended to be elevated after high intensity exercise in active males (Chapter 5). However, there was no impact of either moderate or high intensity exercise on PAEE and post-exercise energy intake in active males. The final study (Chapter 6) determined whether high intensity exercise in the fed state after a few days of food restriction had an impact on circulating energy balance hormones. Circulating postprandial total PYY and pancreatic polypeptide (PP) were increased for one hour after high intensity exercise in active men. There was no change in PAEE and post-exercise energy intake after exercise.

This thesis investigates the influence of nutrition support on a set of critically ill patients after the administration of parenteral nutrition by measuring their energy expenditure and substrate oxidation. The study was performed on 8 spontaneously breathing patients during the 2nd-5th day of polytrauma and the variables were obtained by Indirect calorimetry. The mean age of the patients was 42 ± 16 years. Four patients were obese, three were overweight and one had normal weight in the examined group. The measurement was conducted on 7 patients after hypocaloric nutrition administration (1323 ± 281 kcal/day) and 1 patient after hypercaloric nutrition administration (3400 kcal/day). Statistical analysis revealed that the administered nutritional support had no significant effect on energy expenditure and the substrate oxidation preference. However, correlations between the values of energy expenditure, nutrients oxidation, and values of some of the laboratory parameters were observed. The energy expenditure of polytraumatic patients measured at least 4 hours after nutritional administration was 2139 ± 518 kcal/day and 24.6 ± 3.5 kcal/kg/day. This energy expenditure was covered by oxidation of proteins (31%), by lipids (46%) and by carbohydrates (23%). Large variability was found among the results...

Purpose: To examine the relationship between hand rim propulsion power and energy expenditure during wheelchair locomotion. Methods: Fourteen individuals who used manual wheelchairs were included in this study. Each participant performed five different locomotion activities in a wheelchair with a PowerTap hub built into the rear wheel. The activities included wheeling on a level surface that elicited a low rolling resistance at three different speeds (4.5, 5.5, and 6.5 km∙hr-1), wheeling on a rubberized 400m track that elicited a higher rolling resistance at one speed (5.5 km∙hr-1), and wheeling on a sidewalk course that included uphill and downhill segments at their self-selected speed. Energy expenditure was measured using a portable indirect calorimetry system. In addition, each subject wore an Actical and a SenseWear activity monitor on the right wrist and upper arm, respectively. Stepwise, linear regression was performed to predict energy expenditure from power output variables. A repeated measures ANOVA was used to compare the measured energy expenditure to the estimates from the power models, the Actical, and the SenseWear. Bland-Altman plots were used to assess the agreement between the criterion values and the predicted values. Results: The relationship between energy expenditure and power was significantly correlated (r = 0.694, p < 0.001). Stepwise, linear regression analysis yielded three significant prediction models utilizing measured power; measured power and speed; and measured power, speed, and heart rate. A repeated measures ANOVA demonstrated a significant main effect between measured energy expenditure and estimated energy expenditure (p < 0.01). There were no significant differences between the criterion method and the power models or the Actical. The SenseWear significantly overestimated energy expenditure when wheeling at 4.5 km·hr-1, 5.5 km·hr-1, 6.5 km·hr-1, and during self-paced sidewalk wheeling (p < 0.05). Conclusion: Energy expenditure can be accurately and precisely estimated based on wheelchair propulsion power. These results indicate that wheelchair power could be used as a method to assess physical activity in people who use wheelchairs.

The main purpose of this study was to examine the constructs of the theory of planned behaviour (TPB) in two different exercise settings structured versus unstructured. Owing to the assumption that individuals may perceive less volitional control in a structured setting versus an unstructured setting, it was hypothesized that perceived behavioural control would be a stronger predictor of exercise behaviour in the structured setting. A secondary purpose of the current study was to assess the utility of using two different exercise outcomes energy expenditure and exercise frequency to assess exercise behaviour. Participants (N = 207) were recruited from a first-year kinesiology university class. Data collection occurred over two time periods, nine days apart, and was conducted during class periods. During the first testing session, participants were provided with a questionnaire that assessed TPB constructs and physical activity level in the two settings (structured versus unstructured) using the Modifiable Activity Questionnaire (MAQ). Half of the participants were requested to complete the TPB constructs using energy expenditure as the outcome measure and half were requested to complete the constructs using exercise frequency as the outcome measure. In the second testing session, all participations were asked to report their exercise levels over the previous seven days using the MAQ. In terms of predicting intention, hierarchical regression analyses revealed that, contrary to the hypothesis, perceived behavioural control was more predictive in the unstructured setting versus the structured setting. Although not predicted, it also was found that subjective norms were a significant predictor of activity intention in the structured but not the unstructured setting. The results using the two different outcome measures (e.g., energy expenditure versus frequency) also revealed differences; however, no consistent pattern emerged. One relationship that did emerge was the finding that perceived behavioural control was found to be a stronger predictor of intention in the unstructured setting using energy expenditure as an outcome versus exercise frequency. <p> Finally, the results revealed little support for the TPB constructs predicting self-reported physical activity behaviour. Practical implications and future directions are discussed.

The main purpose of this study was to examine the constructs of the theory of planned behaviour (TPB) in two different exercise settings structured versus unstructured. Owing to the assumption that individuals may perceive less volitional control in a structured setting versus an unstructured setting, it was hypothesized that perceived behavioural control would be a stronger predictor of exercise behaviour in the structured setting. A secondary purpose of the current study was to assess the utility of using two different exercise outcomes energy expenditure and exercise frequency to assess exercise behaviour. Participants (N = 207) were recruited from a first-year kinesiology university class. Data collection occurred over two time periods, nine days apart, and was conducted during class periods. During the first testing session, participants were provided with a questionnaire that assessed TPB constructs and physical activity level in the two settings (structured versus unstructured) using the Modifiable Activity Questionnaire (MAQ). Half of the participants were requested to complete the TPB constructs using energy expenditure as the outcome measure and half were requested to complete the constructs using exercise frequency as the outcome measure. In the second testing session, all participations were asked to report their exercise levels over the previous seven days using the MAQ. In terms of predicting intention, hierarchical regression analyses revealed that, contrary to the hypothesis, perceived behavioural control was more predictive in the unstructured setting versus the structured setting. Although not predicted, it also was found that subjective norms were a significant predictor of activity intention in the structured but not the unstructured setting. The results using the two different outcome measures (e.g., energy expenditure versus frequency) also revealed differences; however, no consistent pattern emerged. One relationship that did emerge was the finding that perceived behavioural control was found to be a stronger predictor of intention in the unstructured setting using energy expenditure as an outcome versus exercise frequency. <p> Finally, the results revealed little support for the TPB constructs predicting self-reported physical activity behaviour. Practical implications and future directions are discussed.

Cattle classified as having low residual feed intake (RFI) phenotypes are those that consume less feed than expected based on body weight and growth performance. Mechanisms contributing to the variation in RFI are not fully understood. Previous studies have shown that cattle of divergent RFI phenotypes have different levels of energy expenditures, which are associated with heat increment, basal metabolism, thermoregulation responses, and physical activity. The objectives of this experiment were to characterize residual feed intake (RFI) in growing heifers and to determine if variation in whole-animal energy expenditure contributes to differences in RFI. Brangus heifers (n =120) were individually fed a roughage-based diet (1.93 Mcal ME/kg DM) diet twice daily and feed refusals measured weekly. Heifers were weighed once weekly for 70 d and RFI calculated as the difference between actual and expected DMI from linear regression of DMI on ADG and mid-test BW0.75. Immediately following the 70 d study, oxygen pulse rate (mL O2/heart beat) and 48-h heart rates were measured on 8 high and 8 low RFI heifers to estimate energy expenditure. Daily heart rates and oxygen pulse rates were higher (P < 0.05) in heifers with high RFI compared to those with low RFI. As a result, energy expenditure (kcal/BW0.75) was estimated to be 17.4 percent greater (P < 0.05) in high-RFI heifers then low-RFI heifers. Mortality and morbidity losses caused by bovine respiratory disease (BRD) continue to negatively impact the net revenues of the beef cattle industry. Stress can predispose calves arriving at feedlots to BRD by impairing their immune system with calves having more excitable temperaments possibly having a greater risk. The objectives of the second study was to examine the effects of metaphylactic treatment and temperament on performance, feed intake, feed efficiency, and feeding behavior traits in steers. Santa Gertrudis steers (n =119) were weighed and randomly to control (CON; no antimicrobial treatment) or metaphylactic (MET; 1.5 mL/45 kg BW of ceftiofur crystalline free acid) treatments. Steers were weighed at 14-d intervals and individual intakes and feeding behavior traits measured using a GrowSafe sytems while fed a roughage-based diet (2.21 Mcal ME/kg DM). Objective (relative exit velocity; REV) and subjective (chute score; CS) measurements of temperament were measured on arrival and on day 28 of the study. Steers with higher REV weighed less, grew slower, consumed less feed, spent less time consuming feeding, had more feeding bouts per meal, had less backfat, smaller longissimus muscle area, and higher cortisol levels. Steers treated with MET had higher ADG than those receiving CON. Cattle with higher REV that received MET had less of a decrease in ADG, DMI, time spend consuming feed, and less of an increase in feeding bouts compared to high REV steers receiving CON. Results from this study suggest that process-control strategies, which quantify and manage inter-animal variation in calf temperament may facilitate more judicious use of antimicrobial products and provide more consistent and predictable responses to metaphylactic strategies.

The aim of this thesis was to describe the relationships between energy expenditure and body composition in healthy adults and children and in patients with metabolic disorders. In a healthy population resting energy expenditure (REE) is highly influenced by body composition, specifically lean mass (LM). Prediction equations can therefore accurately predict REE from body composition in healthy individuals. However, application of these equations to clinical patients, in whom metabolism is disordered, risks miscalculation of energy metabolism due to their dissociation between body composition and energy expenditure. Therefore new prediction equations were derived based on precise body composition measurements in healthy adults and children. Then, in patients with metabolic disorders, differences between their measured and equation-predicted REE and LM were presented as standardised Z scores. REE in healthy adults was predicted by the coefficients: age, fat mass and fat-free mass. LM in healthy adults was predicted by the coefficients: bone mineral content and height2 in men; and by fat and height2 in women. In healthy children, REE was predicted using gender specific models: by fat and LM in boys; and by solely LM in girls. REE and LM were then measured in adult and paediatric patients with metabolic disorders (Lipodystrophy, Thyrotoxicosis and Resistance to Thyroid Hormone β or α), and Z scores were calculated to highlight their deviations from the healthy populations. In adults, thyrotoxicosis patients displayed the highest REE Z scores (5.8), followed by lipodystrophy (2.9) and RTHβ cases (1.8), with RTHα demonstrating the lowest REE Z scores (-2.3). For LM, lipodystrophy patients exhibited with the highest Z scores (4.2), followed by RTHα patients (2.1), with RTHβ patients showing normal LM Z scores (-0.2) and thyrotoxicosis patients presenting with the lowest LM Z scores (-1.2). In the paediatric patients, RTHβ patients demonstrated REE Z scores similar to healthy controls (males; -0.15, females; 0.15), but RTHα patients displayed lower REE Z scores (male; -0.82, female; -2.2) compared to RTHβ patients and healthy controls. These studies highlight the disassociation between REE and body composition in patients with metabolic disorders. The application of a prediction equation for REE to calculate Z scores between measured and predicted values allows quantification of the differences between patients with metabolic disorders and healthy populations, and is a new and important concept.

There is a need to examine the difference in the relationship between oxygen uptake (VO2) and output from hip- and wrist-worn accelerometers in adults with Down syndrome (DS). The purpose of this study is to identify if that relationship is different between adults with and without DS. Hip- and wrist-worn accelerometer accuracy was also assessed. The sample included 16 adults with DS (10 men; age 31±15 years) and 19 adults without DS (10 men; age 24±6 years). We measured VO2 with a portable spirometer and accelerometer output (Vector Magnitude [VM]) with a hip- and a wrist-worn accelerometer. VM and group were significant predictors of VO2 (p?0.021). BMI became a significant predictor in the second model and DS was no longer significant for both accelerometer models. The Bland-Altman plots indicated nearly zero mean error for both groups. Hip-worn accelerometers showed greater accuracy, and showed less error based on 95% confidence intervals.