Poor agreement between a portable armband and indirect calorimetry in the assessment of resting energy expenditure

A robust permutation test for the concordance correlation coefficient

In this work, we developed a robust permutation test for the concordance correlation coefficient (ρ_c) for testing the general hypothesis H₀ : ρ_c = ρ_c(0). The proposed test is based on an appropriately studentized statistic. Theoretically, the test is proven to be asymptotically valid in the general setting when two paired variables are uncorrelated but dependent. This desired property was demonstrated across a range of distributional assumptions and sample sizes in simulation studies, where the test exhibits robust type I error control in all settings tested, even when the sample size is small. We demonstrated the application of this test in two real world examples across cardiac output measurements and endocardiographic imaging.

Validity of Accelerometers for the Evaluation of Energy Expenditure in Obese and Overweight Individuals: A Systematic Review

OBJECTIVE

Even though the validity of accelerometers for the measurement of energy expenditure (EE) has been demonstrated for normal-weight individuals, the applicability of this instrument in obese individuals remains controversial. This review aims to summarize the level of agreement between accelerometers and the gold standards (indirect calorimetry and doubly labelled water) for the measurement of energy expenditure (EE) in obese or overweight individuals.

METHODS

The literature search was limited to comparison studies assessing agreement in EE determination between accelerometers and indirect calorimetry (IC) or doubly labelled water (DLW). We searched in PubMed and in Scopus until March 1, 2019. The analysis was restricted to obese or overweight adult individuals. The following descriptive information was extracted for each study: sample size, characteristics of participants (sex, age, BMI, fat mass percentage, any pathological conditions, modality of recruitment in the study, and exclusion criteria), accelerometer description (model, type and body position), and type of gold standard and validity protocol (duration, conditions, and requirements during and before the experiment). Three review authors independently screened the obtained results, and the quality of the selected articles was assessed by the QUADAS-2 tool.

RESULTS

We obtained seventeen eligible articles, thirteen of which showed concerns for the applicability section, due to the patient selection. Regarding the accelerometers, nine devices were validated in the included studies with the BodyMedia SenseWear® (SWA) being the most frequently validated. Although correlations between accelerometers and the gold standard were high in some studies, agreement between the two methods was low, as shown by the Bland-Altman plots.

CONCLUSIONS

Most accelerometer estimations of EE were inaccurate for obese/overweight subjects, and authors advise to improve the accuracy of algorithms for SWA software, or the predicted equations for estimating EE from other accelerometers.

Evaluation of a non-invasive multisensor accelerometer for calculating energy expenditure in ventilated intensive care patients compared to indirect calorimetry and predictive equations

Continuous measurement of resting energy expenditure (REE) in critically ill patients remains challenging but is required to prevent malnutrition. SenseWear Pro 3 Armband (SWA) is a research grade accelerometer for assessment of REE with the advantage of easy handling. In a prospective study we compared SWA with indirect calorimetry (IC) and predictive equations in critically ill, ventilated patients. REE was measured by SWA, IC and calculated by predictive formulas. Potential confounding factors that influence REE were also recorded. Results of SenseWear Armband and indirect calorimetry were compared using the Bland-Altman method. 34 ICU patients were investigated. SWA underestimated resting energy expenditure compared to IC with a mean bias of ΔREE = -253.6 ± 333.2 kcal, equivalent to -11.7 % (p = 0.025). This underestimation was seen in both, medical (-14.9 %) and surgical (-12.9 %) patients and the bias was greater in patients with fever (-19.0 %), tachycardia (-18.7 %) or tachypnea (-26.2 %). Differences were also noted when SWA was compared to predictive formulas. At present, SWA cannot be regarded as an alternative to indirect calorimetry. Individual measurements are often inaccurate and should be used with caution until improved algorithms, based on the results of this study, have been implemented.

Irregular meal-pattern effects on energy expenditure, metabolism, and appetite regulation: a randomized controlled trial in healthy normal-weight women

BACKGROUND

Obesity is increasing in parallel with greater all-day food availability. The latter may promote meal irregularity, dysregulation of the energy balance, and poor metabolic health.

OBJECTIVE

We investigated the effect of meal irregularity on the thermic effect of food (TEF), lipid concentrations, carbohydrate metabolism, subjective appetite, and gut hormones in healthy women.

DESIGN

Eleven normal-weight women (18-40 y of age) were recruited in a randomized crossover trial with two 14-d isoenergetic diet periods (identical foods provided and free living) that were separated by a 14-d habitual diet washout period. In period 1, participants followed a regular meal pattern (6 meals/d) or an irregular meal pattern (3-9 meals/d), and in period 2, the alternative meal pattern was followed. Before and after each period, when participants were fasting and for 3 h after intake of a test drink, measurements were taken of energy expenditure, circulating glucose, lipids (fasting only), insulin, glucagon-like peptide 1 (GLP-1), peptide YY (PYY), and ghrelin. An ad libitum test meal was offered. Subjective appetite ratings were assessed while fasting, after the test drink, after the ad libitum meal, and during the intervention. Continuous interstitial glucose monitoring was undertaken for 3 consecutive days during each intervention, and the ambulatory activity pattern was recorded (ambulatory energy expenditure estimation).

RESULTS

Regularity was associated with a greater TEF (P < 0.05) and a lower incremental area under the curve (iAUC) for glucose after intake of the test drink (over 3 h) and, for some identical meals, during the 2 interventions (over 90 min) (day 7: after breakfast; day 9: after lunch and dinner). There was no difference between treatments for the test-drink gut hormone response. A time effect was noted for fasting GLP-1, fasting PYY, PYY responses, and hunger-rating responses to the test drink (P < 0.05). Lower hunger and higher fullness ratings were seen premeal and postmeal during the regular period while subjects were free living.

CONCLUSION

Meal regularity appears to be associated with greater TEF and lower glucose responses, which may favor weight management and metabolic health. This trial was registered at clinicaltrials.gov as NCT02052076.

Measurement of body composition as a surrogate evaluation of energy balance in obese patients

In clinical practice obesity is primarily diagnosed through the body mass index. In order to characterize patients affected by obesity the use of traditional anthropometric measures appears misleading. Beyond the body mass index, there are overwhelming evidences towards the relevance of a more detailed description of the individual phenotype by characterizing the main body components as free-fat mass, muscle mass, and fat mass. Among the numerous techniques actually available, bioelectrical impedance analysis seems to be the most suitable in a clinical setting because it is simple, inexpensive, noninvasive, and highly reproducible. To date, there is no consensus concerning the use of one preferred equation for the resting energy expenditure in overweight and/or obese population. Energy restriction alone is an effective strategy to achieve an early and significant weight loss, however it results in a reduction of both fat and lean mass therefore promoting or aggravating an unfavourable body composition (as sarcobesity) in terms of mortality and comorbidities. Therefore the implementation of daily levels of physical activity should be simultaneously promoted. The major role of muscle mass in the energy balance has been recently established by the rising prevalence of the combination of two condition as sarcopenia and obesity. Physical exercise stimulates energy expenditure, thereby directly improving energy balance, and also promotes adaptations such as fiber type, mitochondrial biogenesis, improvement of insulin resistance, and release of myokines, which may influence different tissues, including muscle.

Validation of the SenseWear Armband in different ambient temperatures

This study examines the validity of the SenseWear Armband in different temperatures using the old (SenseWear v2.2) and newest version of the algorithm (SenseWear v5.2) against indirect calorimetry (IC). Thirty-nine male and female students (21.1 ± 1.41 years) completed an exercise trial in 19°C, 26°C and 33°C consisting of 5 min standing followed by alternating walking/running at 35% and 65% of their maximal oxygen uptake. The accuracy of the algorithms was evaluated by comparing estimated energy expenditure (EE) to IC using a mixed-model design. No difference was reported in EE between the different temperatures for IC. Both algorithms estimated EE significantly higher when exercising at high intensity in 33°C compared to 19°C. Compared to IC, SenseWear v2.2 accurately estimated EE during standing and light intensity exercise but underestimated EE when exercising in a hot environment and at high intensity. SenseWear v5.2 showed a difference when exercising at high intensity in thermoneutral and warm conditions. The new algorithm improved EE estimation in hot environments and at high intensity compared to the old version. However, given the inherent inaccuracy of the EE estimates of SenseWear, greater weight should be given to direct monitor outputs rather than the ability of a monitor to estimate EE precisely.

Physical activity after surgically obtained weight loss: study with a SenseWear armband in subjects undergoing biliopancreatic diversion

BACKGROUND

This study aimed to understand the role of the extra load of body mass in limiting physical activity and in preventing an active lifestyle in severely obese patients.

MATERIALS AND METHODS

The study was carried out in a University Hospital setting, and investigates severely obese patients, having undergone biliopancreatic diversion (BPD) for obesity and control subjects with a body weight closely similar to that of the BPD subjects; energy intake was evaluated by alimentary interview and energy expenditure was assessed with the Body Media SenseWear® Pro armband (SWA).

RESULTS

SWA metabolic efficiency (MET) was negatively associated with body mass index values (ρ = -0.464, p < 0.01), and the SWA overall energy expenditure was very similar to the energy alimentary intake in the obese patients and in the control subjects. Among the controls, the MET mean value was higher (p < 0.05) than that observed both in obese and in BPD subjects (1.392 ± 0.211 vs. 1.149 ± 0.181 and 1.252 ± 0.284, respectively); furthermore, in comparison with the obese and the BPD groups, among the control individuals a greater number of active persons (27% vs. 0 and 7%, respectively) and a lower number of sedentary persons (27% vs. 70 and 43%, respectively) was found (p < 0.01).

CONCLUSIONS

These preliminary results suggest the reliability of SWA data in assessing energy expenditure and tend to rule out the hypothesis that in severely obese patients the extra load of body mass by itself is a main factor limiting physical activity and leading to a sedentary lifestyle.

Accuracy of SenseWear Pro2 Armband to predict resting energy expenditure in childhood obesity

OBJECTIVE

The accuracy of the SenseWear Pro2 Armband (SWA) in estimating resting energy expenditure (REE) in children and adolescents with obesity, using indirect calorimetry (IC) as a reference was evaluated.

DESIGN AND METHODS

REE was assessed using both the SWA and IC in 40 obese subjects (26 M/14 F, age 11.5 ± 2.57 years, z-score BMI 3.14 ± 0.53). The agreement between methods was assessed by the Bland-Altman procedure. The relationship between REE assessments and patients' characteristics was also analyzed.

RESULTS

SWA- and IC-derived estimates of REE showed a significant correlation (r = 0.614; P < 0.001), but the SWA overestimated mean REE by 13% (P < 0.001). Age and kilogram of fat-free mass (kg-FFM) were significantly correlated with both REE estimation by SWA (r = 0.434 and r = 0.564, respectively) and IC (r = 0.401 and r = 0.518, respectively). Only kg-FFM was demonstrated to be the main predictor factor of REE variability (r(2) 79% SWA; 75% IC).

CONCLUSIONS

The SWA overestimated mean REE in childhood obesity, suggesting that the SWA and IC are not yet interchangeable methods. This would require improving the SWA by developing better algorithms for predicting REE and, probably, bias in each individual REE could be reduced by an adjustment for subjects' kg-FFM.

Energy expenditure by multisensor armband in overweight and obese lactating women validated by doubly labeled water

OBJECTIVE

To validate total energy expenditure (TEE) and activity energy expenditure (AEE) from the portable SenseWear armband (SWA) Pro 2 (TEESWA and AEESWA ; InnerView software versions SWA 5.1 and SWA 6.1) against TEE from doubly labeled water (DLW) and AEE from DLW and indirect calorimetry (TEEDLW and AEEDLW ) in overweight/obese lactating women at 10 weeks postpartum.

DESIGN AND METHODS

TEE was measured simultaneously with DLW (14 days) and SWA (first 7 days). Lactating women (n = 62), non-smoking, with a BMI > 25 kg/m(2) and wearing time SWA ≥ 90% were included.

RESULTS

Mean TEESWA5.1 was overestimated with 85 kcal/day compared to TEEDLW (P = 0.040), while mean TEESWA6.1 was underestimated with 241 kcal/day compared to TEEDLW (P < 0.001). Mean AEESWA5.1 was similar to mean AEEDLW (P = 0.818), while mean AEESWA6.1 was underestimated with 581 kcal/day compared to AEEDLW (P < 0.001). TEESWA6.1 and AEESWA6.1 were systematically underestimated at higher levels of energy expenditure and BMI while only AEESWA5.1 was systematically overestimated at higher levels of energy expenditure.

CONCLUSIONS

TEESWA5.1 and AEESWA5.1 were fairly estimated on a group level while TEESWA6.1 and AEESWA6.1 were significantly and systematically underestimated. Both SWA software versions showed large individual variation in agreement with TEEDLW and AEEDLW , limiting the validity on individual level.

Energy expenditure evaluation in humans and non-human primates by SenseWear Armband. Validation of energy expenditure evaluation by SenseWear Armband by direct comparison with indirect calorimetry

Introduction

The purpose of this study was to compare and validate the use of SenseWear Armband (SWA) placed on the arm (SWA ARM) and on the back (SWA BACK) in healthy humans during resting and a cycle-ergometer exercise and to evaluate the SWA to estimate Resting Energy Expenditure (REE) and Total Energy Expenditure (TEE) in healthy baboons.

Methods

We studied 26 (15F/11M) human subjects wearing SWA in two different anatomical sites (arm and back) during resting and a cycle-ergometer test and directly compared these results with indirect calorimetry evaluation (IC), performed at the same time. We then inserted the SWA in a metabolic jacket for baboons and evaluated the TEE and REE in free living condition for 6 days in 21 (8F/13M) non-human primates.

Results

In humans we found a good correlation between SWA place on the ARM and on the BACK with IC during the resting experiment (1.1±0.3 SWAs, 1±0.2 IC kcal/min) and a slight underestimation in the SWAs data compared with IC during the cycle-ergometer exercise (5±1.9 SWA ARM, 4.5±1.5 SWA BACK and 5.4±2.1 IC kcal/min). In the non-human primate (baboons) experiment SWA estimated a TEE of 0.54±0.009 kcal/min during free living and a REE of 0.82±0.06 kcal/min.

Conclusion

SWA, an extremely simple and inexpensive apparatus, provides quite accurate measurements of energy expenditure in humans and in baboons. Energy expenditure data obtained with SWA are highly correlated with the data obtained with “gold standard”, IC, in humans.

[Postoperative assessment of daily energy expenditure. Comparison of two methods]

BACKGROUND

The reference method for determining resting energy expenditure (REE) in clinical nutrition practice is measurement by indirect calorimetry; however, indirect calorimetry has some limitations, is expensive and not widely available. Therefore, the most used methods to estimate the caloric requirements in intensive care patients are predictive equations. The Harris-Benedict equations (HBE) are the most common formulae in the clinical setting. The SenseWear(®) armlet (SWA) is a noninvasive device that monitors skin temperature, heat flux, galvanic skin response and movement. These data as well as anthropometric characteristics are used to calculate REE. The aim of this study was to evaluate the levels of agreement and interchangeability of REE estimated by HBE (EEHBE) and measured by SWA (EESWA) in normometabolic patients after elective bowel resection with laparotomy. Furthermore, postsurgical pain therapy by continuous thoracic epidural anaesthesia (t-PDA) was compared with continuous intravenous pain therapy regarding EESWA in these patients.

METHODS

After obtaining approval by the ethics committee and written informed consent 57 patients participated in the study procedures. A total of 50 patients (23 male, 27 female) were finally included in the data analysis because 7 patients did not meet the criterion of > 80% on-body time of the SWA. Additional (a priori) exclusion criteria were metabolic or cardiopulmonary decompensation or postoperative mechanical ventilation. Before induction of general anesthesia 26 patients received a thoracic epidural catheter. Immediately after surgery the SWA was placed on the right upper arm of each patient for 24 h. A continuous pain therapy was started either an epidural application of ropivacain 0.2% and sufentanil or in the other 24 patients an intravenous infusion of metamizol and tramadol.

RESULTS AND DISCUSSION

The data showed good agreement between EESWA and EEHBE. The mean on-body time was found to be 22.94±4.77 h. There were no significant differences between EESWA and EEHBE (p>0.05) corresponding to a high Pearson's coefficient of correlation of 0.985. The mean bias (EESWA-EEHBE) was -0.569±0.378 kcal/kgBW/24 h reflecting a minimal systematic underestimation of REE by SWA of -2.9% compared to EEHBE. The Bland-Altman plot shows interchangeability of EESWA and EEHBE. It was noted that 94% of the data points (47 out of 50 patients) were within ±2 SD and the remaining 3 data points were lying close to the 95% interval. The same results (no significant differences between EESWA and EEHBE) were obtained after differentiation of EEHBE into low (<18 kcal/kgBW/24 h, n=9), medium (18-21 kcal/kgBW/24 h, n=30) and high (>21 kcal/kgBW/24 h, n=11) energy ranges. There were no significant differences in EESWA regarding postsurgical pain therapy regimens.

CONCLUSIONS

The SWA showed reliable concordance with daily REE estimated by HBE in normometabolic postsurgery patients. This noninvasive, convenient and easy to handle device may be helpful in determining energy requirements as part of metabolic monitoring. Further research is needed to validate the method in patients with severe metabolic disturbances. The energetic requirements of patients with postoperative t-PDA were not different from those with intravenous pain therapy.

Estimating resting energy expenditure with a portable armband device in an ambulatory setting

BACKGROUND

The measurement of resting energy expenditure (REE) in an ambulatory setting raises methodological problems. Therefore, the use of predictive equations for the estimation of REE is common. Alternatively, the measurement of sleeping energy expenditure (SEE) has been proposed. The authors retrospectively analyzed data on SEE assessed with a portable armband (PA) device in an ambulatory setting and evaluated this approach against predictive equations and REE measured by indirect calorimetry (IC).

METHODS

REE was measured with IC, and SEE was assessed with the PA using standardized conditions in 81 participants (aged 46 ± 13 years) over a wide range of body weight (mean body mass index [BMI] 36.4 ± 9.3 kg/m(2); range, 21.6-55.7).

RESULTS

SEE (1756 ± 393 kcal/d) was 7.6% higher than REE (1632 ± 346 kcal/d) (P < .001). This difference (123 ± 214 kcal/d) was smaller than that using the predictive equation for REE by Harris and Benedict (207 ± 217 kcal/d) and the BMI group-specific equations according to Müller et al (209 ± 190 kcal/d). Linear regression analysis was significant (r (2) = 0.705; P < .001). SEE showed similar 95% confidence intervals compared with both of the predictive equations.

CONCLUSIONS

The described standardized assessment of SEE by a PA device appears to be a promising approach to estimate REE in an ambulatory setting. SEE reflects REE at least as precisely as the predictive equations.

Patterns of physical activity and sedentary behavior in normal-weight, overweight and obese adults, as measured with a portable armband device and an electronic diary

BACKGROUND & AIMS

Accurate data on domain-specific physical activity and sedentary behavior among normal-weight, overweight and obese adults are scarce. This study described a comprehensive physical (in)activity profile and examined variations in activity across the week.

METHODS

Physical activity was measured in 442 Flemish adults (41.4 ± 9.8 years) for 7 days using the SenseWear Armband and an electronic diary. Differences in (in)activity between BMI subgroups were examined using one-way analyses of variance.

RESULTS

Physical activity level decreased with increasing BMI in men (1.77-1.46 MET) and women (1.67-1.31 MET). Sedentary time was higher in obese than normal-weight men (+1.09 h/day) and overweight and obese than normal-weight women (+1.04 and +1.88 h/day). Total hours of moderate-to-vigorous activity and bouts of moderate-to-vigorous activity were lower in overweight and obese than normal-weight subjects. The average duration of a sedentary bout and the number of breaks in sedentary time were only different between female BMI groups. The intensity of domain-specific activities decreased with increasing BMI. Activity patterns across the week differed between subgroups, with the difference most pronounced on Saturday for the male BMI groups.

CONCLUSIONS

This study described activity patterns among normal-weight, overweight and obese adults. The results should be used to design obesity prevention strategies.

Resting energy expenditure assessment in anorexia nervosa: comparison of indirect calorimetry, a multisensor monitor and the Müller equation

The aim of this study was to compare the estimations provided by three different means of measuring the resting energy expenditure (REE) in anorexia nervosa (AN) patients. REE was measured, after 24 h of refeeding, using a portable multisensor body monitor [SenseWear Pro2 Armband (SWA)], FitMate™ method and the Müller equation for individuals with body mass index < 18.5, the latter being based on dual-energy X-ray absorptiometry assessment of body composition. The mean differences between REE values estimated by SWA and those provided by the Müller equation and the FitMate™ method were significantly different from zero in both cases. In contrast, the mean differences between FitMate™ method and Müller equation were weakly significantly different from zero, and a significant correlation was noted between these two methods. In conclusion, the SWA does not appear to be an alternative to FitMate™ and Müller equation methods for assessing REE in AN patients.

Accuracy of a portable multisensor body monitor for predicting resting energy expenditure in older people: a comparison with indirect calorimetry

BACKGROUND

Accurate and comfortable methods are needed to determine resting energy expenditure (REE) in older people who are characterized by a lowered metabolic rate. The portable SenseWear® armband (SWA) body monitor, worn on the right upper arm, can easily be used by this age group in an ambulatory manner.

OBJECTIVE

The purpose of this study was to evaluate the reliability and accuracy of the SWA armband in determining REE in healthy, normal-weight older people.

METHODS

Participants were 49 older, community-dwelling volunteers aged 60-87 years. Reliability was tested in all subjects. Accuracy was analyzed in 32 subjects by comparing REE estimated by SWA against REE measured by indirect calorimetry as a criterion method. Data were simultaneously and continuously recorded for 20 min in the morning. In the same subjects, REE determined by SWA from night recording was evaluated for accuracy against indirect calorimetry. The agreement between methods was assessed by the Bland-Altman procedure.

RESULTS

(1) REE measured by SWA was reliable when comparing 2 consecutive measurements (typical error: 1.9%). (2) REE calculated from morning recording (1,543 ± 181 kcal/24 h) was higher than REE from night recording (1,564 ± 192 kcal/24 h; p = 0.018). (3) Compared with REE by indirect calorimetry (1,377 ± 228 kcal/24 h) for accuracy, the SWA overestimated mean REE in the older people by 12 and 14%, respectively, for morning and night recording (each p < 0.001).

CONCLUSIONS

The SWA provides a reliable estimate of REE in healthy older subjects and has the advantage of easy handling. The 20-min recording time, which was recommended by the manufacturer, can be applied. However, the SWA overestimates REE in this group, possibly due to age-related changes in skin conductance and thermoregulation, both being measured by sensors of the SWA armband. This requires improving the SWA by developing better fitting algorithms for predicting REE in older people.

Differences of energy expenditure and physical activity patterns in subjects with various degrees of obesity

BACKGROUND & AIMS

An imbalance of energy intake and energy expenditure leads to obesity. However, little detailed information of energy expenditure and physical activity patterns in obese subjects is available. Therefore, we assessed total energy expenditure (TEE) with its components resting energy expenditure (REE) and activity thermogenesis (AT) and the patterns of physical activity in non-obese and in subjects with different degrees of obesity.

METHODS

TEE and activity pattern were assessed with the SenseWear™ armband in 78 subjects (46 ± 12 years; 28 with normal weight/overweight, 13 each with obesity I° and II°, and 24 with obesity III°). In addition, REE was measured by indirect calorimetry and AT was calculated.

RESULTS

Although TEE (and REE) increased with increasing weight category from 2567 (1437) kcal/d in non-obese subjects to 3033 (1931) kcal/d in subjects with obesity III° (p=0.016, p<0.001, respectively) body weight adjusted TEE decreased from 33.1 to 22.1 kcal/kg/d (p<0.001). This was mainly due to decreased body weight adjusted AT (11.3-5.8 kcal/kg/d, p<0.001). AT consisted almost completely of non-exercise AT. In particular, for obese subjects exercise-related AT was negligible.

CONCLUSIONS

Higher degrees of obesity are associated with decreased body weight adjusted AT. These differences have to be considered for therapeutic strategies.