Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2024 Sep 1.
Published in final edited form as: Curr Opin Clin Nutr Metab Care. 2023 Jun 9;26(5):409–416. doi: 10.1097/MCO.0000000000000952

How can we assess “thrifty” and “spendthrift” phenotypes?

Tim Hollstein 1, Paolo Piaggi 2,3
PMCID: PMC10531526  NIHMSID: NIHMS1903461  PMID: 37294042

Abstract

Purpose of review

There is a large inter-individual variability in the magnitude of body weight change that cannot be fully explained by differences in daily energy intake and physical activity levels and that can be attributed to differences in energy metabolism. Measuring the short-term metabolic response to acute changes in energy intake can better uncover this inter-individual variability and quantify the degree of metabolic thriftiness that characterizes an individual’s susceptibility to weight gain and resistance to weight loss. This review summarizes the methods used to identify the individual-specific metabolic phenotype (thrifty vs. spendthrift) in research and clinical settings.

Recent findings

The metabolic responses to short-term fasting, protein-imbalanced overfeeding, and mild cold exposure constitute quantitative factors that characterize metabolic thriftiness.

Summary

The energy expenditure response to prolonged fasting is considered the most accurate and reproducible measure of metabolic thriftiness, likely because the largest energy deficit best captures interindividual differences in the extent of metabolic slowing. However, all the other dietary/environmental challenges can be used to quantify the degree of thriftiness using whole-room indirect calorimetry. Efforts are underway to identify alternative methods to assess metabolic phenotypes in clinical and outpatient settings such as the hormonal response to low-protein meals.

Keywords: thrifty, spendthrift, fasting, energy expenditure, whole-room calorimeter

Introduction

The human thrifty phenotype concept refers to the ability to efficiently conserve and use energy in order to ensure survival and reproductivity in environments with limited food resources. This involves tuning the utilization rate of metabolic energy obtained from food to support life processes while avoiding the waste of excess energy. The concept has evolved from an original theory ascribing the high prevalence of type 2 diabetes in today’s modern societies to thrifty genes promoting hyperinsulinemia and to acquired thriftiness through fetal malnutrition to a concept of energy conservation during famine and overnutrition.

The relevance of the human thrifty phenotype to energy balance and body weight change has been consistently demonstrated in various clinical trials including both inpatient and outpatient studies, in which individuals with a more thrifty metabolism have been shown to have: i) reduced rates of weight and fat mass loss and greater energy deficit during highly controlled, prolonged underfeeding conditions(1), ii) higher rates of weight and fat mass regain following clinically meaningful diet-induced weight loss(2), iii) higher rates of weight and fat mass gain and greater energy surplus during sustained overfeeding(3), and iv) higher rates of spontaneous weight gain over time under free-living conditions(4). In addition, a more thrifty metabolism has been associated with a higher percentage body fat and a lower core body temperature (5).

Several methods have been proposed to identify thrifty vs. spendthrift phenotypes via measurements of whole-body energy expenditure (EE): these include the EE responses to: 1) short-term (24–36h) fasting, 2) 24-h low- and high-protein overfeeding, 3) mild cold exposure, and 4) metabolic efficiency in the energy balance state. In this review, we summarize these different methods currently used to assess metabolic thriftiness/spendthriftness.

Methods to identify thrifty and spendthrift phenotypes

(1). Short-term fasting

To date, the most reliable way to assess metabolic thriftiness is the measurement of metabolic rate during short-term (24-h) fasting, thereby mimicking starvation and stimulating whole-body energy conservation mechanisms. In previous studies, a greater fasting-induced decrease in 24-h EE (24hEE, i.e., greater metabolic slowing upon fasting) was associated with reduced rate of diet-induced weight loss during sustained underfeeding(1). It also predicted greater rate of weight gain, more fat storage and more calories stored during sustained low-protein overfeeding(3) and greater spontaneous weight gain over time in free-living conditions(4).

To accurately measure energy metabolism during fasting condition with the goal of obtaining quantitative measures of thriftiness, we assessed the metabolic response during short-term fasting (i.e., adaptive thermogenesis) by measuring the change in EE in a whole-room calorimeter (also known as respiratory or metabolic chamber) during both eucaloric (or energy balance) and fasting conditions in separate 24-h sessions. Adaptive thermogenesis generally refers to the physiological processes that contribute to changes in metabolic rate in response to environmental and behavioral challenges. This physiological response reflects the efficiency of metabolic processes that contribute to changes in resting metabolic rate and non-resting energy expenditure (such as diet-induced thermogenesis and physical activity).

Whole-room calorimeters represent the gold standard method for accurate measurement of 24hEE in research settings (Figure 1A)(6). The measured 24hEE is composed of sleeping metabolic rate, energy cost of being awake, diet-induced thermogenesis (DIT), and the physical activity EE (PAEE) (Figure 1B). Although all components of daily EE can be quantified from 24hEE data, their measurements have different degrees of accuracy. Specifically, sleeping EE is more accurate than DIT and PAEE measurement(7, 8). However, whole-room calorimeters have limitations such as high cost, limited availability, and the need for subjects to remain confined in the chamber for a prolonged period. In addition, standardization of diet, physical activity, and environmental conditions, as well as operating procedures for the calorimeter (9), is required to minimize inter-study variability across research sites. Despite these challenges, the advantage of a whole-room calorimeter compared to other methods of indirect calorimetry, especially those used to obtain short-term measurements of resting metabolic rate (RMR) in the post-absorptive state, is the accurate EE measurement over an extended period of 24 hours in different physiologic conditions including both feeding and fasting conditions as well as daytime (awake) and nighttime (sleeping) settings. Moreover, the reproducibility of EE measurements (especially, 24hEE and sleeping EE) has been shown to be high across different whole-room calorimeters, making this technique suitable for conducting large multi-center studies(10).

Figure 1 – Example of a minute-by-minute EE time course in a whole-room calorimeter and breakdown of the 24hEE components.

Figure 1 –

Open in a new tab

(A) Time course of EE measured minute by minute EE during a 24-h stay in a whole-room calorimeter. Meals are delivered to the participant inside the calorimeter through a small airlock at the indicated times. (B) Components of 24hEE: sleeping metabolic rate, awake metabolic rate, diet-induced thermogenesis, and physical activity EE.

24hEE, 24h energy expenditure; EE, energy expenditure.

To accurately determine 24hEE during the eucaloric state, 24hEE was measured in two separate 24-h sessions in a whole-room calorimeter. After 3 days on a standardized, isocaloric, balanced diet, the first eucaloric EE assessment was obtained while the subjects resided for 24-h in the calorimeter and total energy intake was calculated by using unit-specific formulas to achieve 24-h energy balance in the confined environment of the calorimeter. After a wash-out day outside the calorimeter on an isocaloric balanced diet, the second eucaloric EE assessment was performed in the calorimeter when the prescribed total energy intake was set to be equal to the measured 24hEE value from the first assessment. This allowed to best achieve 24-h energy balance in the confined setting of a calorimeter. The 24hEE obtained during this second assessment was considered the daily EE during eucaloric conditions. Subsequently, subjects entered the whole-room calorimeter again and fasted for 24-h when they were only allowed to drink non-caloric non-caffeinated beverages (e.g., water). Of note, adaptive thermogenesis during fasting is mediated by the rate of depletion of hepatic glycogen stores, which influences metabolic rate in the early phase of fasting along with a decrease in insulin secretion (11). Therefore, to account for inter-individual differences in hepatic glycogen stores, it is essential to standardize food intake prior to 24-hour fasting assessment.

To quantify the individual EE adaptation during acute fasting (i.e., the extent of metabolic slowing in absence of intake), the measured 24hEE value during fasting conditions was subtracted from the 24hEE value during eucaloric conditions obtained during the second eucaloric assessment. To account for interindividual differences in demographics and anthropometrics, each 24hEE value was adjusted for its known determinants, e.g., body composition, age, sex, ethnicity, physical activity and ambient temperature.

By definition, a relatively greater decrease in 24hEE during fasting from isocaloric condition indicates a thriftier metabolism as it reflects a greater degree of metabolic slowing. Of note, to date there is no predefined threshold for the decrease in 24hEE during fasting to clearly separate thrifty from spendthrift individuals, but rather a continuous spectrum identifying individuals with varying degree of thriftiness(12). Therefore, a certain degree of thriftiness as quantified by the change in 24hEE during fasting can be only interpreted in relation to individuals of the same study cohort. To better illustrate the characteristics of two extreme phenotypes, we performed a group-wise comparison of thrifty vs. spendthrift subjects as arbitrarily defined by the median value for the decrease in 24hEE in the whole cohort. Specifically, we classified all individuals who had lower-than-median decrease in 24hEE during fasting as thrifty, and those who exceeded it as spendthrift.

Despite the assessment of 24hEE represents the gold standard method to uncover the underlying metabolic phenotype over 24 hours including both sleeping and awake conditions, the change in sleeping metabolic rate (SMR) during fasting may reflect the extent of individual thriftiness. In a previous study, we reported that thrifty individuals on average decreased SMR after short-term fasting whereas spendthrift individuals tended to increase it(12). Likewise, other researchers reported that a lower-than-predicted SMR was associated with greater fat mass retention after an overfeeding period, indicating a thrifty phenotype more susceptible to gain weight over time(13).

This dichotomy in fasting-induced SMR change between thrifty and spendthrift individuals might be useful in clinical settings – where 24-h sessions inside a whole-room calorimeter cannot be conducted – as SMR constitutes a surrogate marker for RMR. The latter is easier to measure due to less complex and expensive equipment (ventilated hood system vs. metabolic chamber) and a shorter duration (20 min vs. 24 hours). However, it is unclear if the assessment of fasting-induced changes in RMR can reliably capture thriftiness derived from SMR as there are considerable differences between both measurements including the post-absorptive state (SMR: few hours after last meal; RMR: >8 hours after last meal), the level of consciousness (SMR: sleep-state; RMR: awake-state) and the duration of measurement (SMR: few hours overnight; RMR: <1 hour).

Other groups reported that RMR measurement might be sufficient to quantify the extent of fasting-related metabolic thriftiness. For instance, Spranger et al. recently reported that a thrifty phenotype defined by a relatively lower RMR (adjusted for fat-free mass) following a 2-month weight loss intervention (800 kcal/d liquid diet) was indicative for greater fat mass regain after 2 years in 80 post-menopausal obese women(14).

(2). Low- and high-protein overfeeding diets

Acute increases in energy intake, such as short-term overfeeding a macronutrient imbalanced diet, elicit metabolic adaptation that typically consists in an acute increase in EE i.e., increased diet-induced thermogenesis (DIT), that, in part, dissipate excess energy ingested. However, there is a large inter-individual variability in the extent of metabolic adaptation to the same overfeeding diet. For instance, some individuals do not substantially change their EE in response to 24-h overfeeding diet accounting for twice their daily caloric needs, while others can increase their metabolic rate up to 15%, and this is independent of changes in physical activity levels(15).

Interestingly, the extent of increase in 24hEE during acute overfeeding correlated with the extent of decrease in EE during acute fasting in the same individual(15). In other words, thriftiness (as quantified by a greater decrease in 24hEE during acute fasting) is associated with reduced metabolic adaptation (i.e., less increase in 24hEE) during acute overfeeding(4, 15). Hence, thrifty individuals have less capacity of energy dissipation in acute states of positive energy balance, thereby conferring greater weight gain susceptibility during long-term overeating. This was confirmed in outpatient studies, where less increase in 24hEE during low-protein overfeeding at baseline was associated with greater free-living weight gain after 6 months (4, 16).

In contrast to fasting, which constitutes a unique and non-modifiable dietary condition, it is possible to employ different overfeeding paradigms to perturb whole-body energy metabolism with the goal of identifying the underlying metabolic phenotype. That is, quantity (e.g., extra calories consumed), quality (e.g., macronutrients consumed), and eating window (e.g., one or more meals throughout the day) can be modified to elicit a metabolic response that quantifies the extent of thriftiness in one individual. We mainly used overfeeding diets accounting for 200% the daily energy requirements, e.g., twice the eucaloric needs as quantified by the 24hEE value measured during two consecutive energy balance assessments inside the calorimeter. With regard to dietary macronutrient content, we found that protein-imbalanced (e.g., 3% low-protein or 30% high-protein) overfeeding diets are better suited to identify thrifty and spendthrift phenotypes compared to high-carbohydrate, high-fat or balanced overfeeding diets(15). These findings support earlier seminal research by Stock M.J., Dulloo A.G. and Jacquet J. (1999, International Journal of Obesity), who suggested that the inter-individual differences in metabolic adaptation (and the associated energy cost of weight gain) during prolonged overfeeding are only modest for normal-protein overfeeding diets (dietary protein content: 15–20%), whereas they are “magnified” during protein-imbalanced overfeeding diets, such as those with low- (<5%) or high- (>20%) protein content.

Similarly to the fasting assessment described above, we accurately assessed metabolic adaptation to short-term overfeeding inside a whole-room calorimeter with repeated 24-h measurements: two 24-h sessions in eucaloric conditions to precisely estimate the individual-specific baseline 24hEE level, and one 24-h session during low-protein (3% protein, 46% fat, and 51% carbohydrate) or high-protein (30% protein, 44% fat, and 26% carbohydrate) overfeeding, with at least one “wash-out” day on an isocaloric, balanced diet in-between each assessment. After the overfeeding session, food leftovers were weighed by the metabolic kitchen staff and the actual energy intake consumed during the 24h in the calorimeter was calculated. Individuals who consumed less than 95% of the total kcal given were excluded to minimize confounding effects of varying caloric intake.

To calculate the extent of metabolic adaptation during low- and high-protein overfeeding diets as compared to the eucaloric assessment, the measured 24hEE value during overfeeding was subtracted from the 24hEE value during energy balance. A relatively smaller increase in 24hEE during overfeeding was associated with greater decrease in 24hEE during short-term fasting(15), thus indicating a thriftier metabolism.

(3). Mild cold exposure

Similar to fasting and overfeeding dietary interventions, mild cold exposure also leads to short-term metabolic adaptation to maintain temperature homeostasis, primarily by inducing non-shivering “cold-induced thermogenesis” (CIT). We previously reported that a greater extent of thriftiness (as defined by a larger decrease in 24hEE during short-term fasting compared to eucaloric conditions) is strongly associated with reduced CIT during mild cold exposure(17), presumably due to lower brown adipose tissue activity which is a main determinant of non-shivering thermogenesis(18). Therefore, the measurement of CIT during mild cold exposure constitutes another way to assess metabolic thriftiness using a different stressor compared to fasting or overfeeding diets.

We quantified the extent of CIT over 24 hours in a metabolic chamber with two subsequent 24-h measurements, both obtained during eucaloric feeding: one session at thermoneutrality (24°C ambient temperature) and one session at mild cold temperature (19°C ambient temperature, lower ambient temperatures are not used in this paradigm to avoid shivering-related thermogenesis). Standardized clothing (hospital gown, pants, ankle-length socks) was provided to minimize confounding effects. For sleep, three blankets were provided, such that participants could create their own microenvironment to achieve a level of comfort overnight.

To calculate the individual metabolic adaptation during mild cold exposure, the measured 24hEE value during mild cold exposure was subtracted from the 24hEE value during energy balance at thermoneutrality. A decrease or a relatively smaller increase in 24hEE during mild cold exposure was associated with a greater decrease in 24hEE during short-term fasting at thermoneutrality(17), indicating a thriftier metabolism.

(4). Metabolic efficiency in sedentary, eucaloric conditions

The most reproducible variable to quantify metabolic thriftiness in terms of its longitudinal association with future weight change was the 24hEE response to short-term fasting, during which thriftier individuals showed a relatively greater decrease in 24hEE compared to energy balance conditions. Because metabolic thriftiness was calculated as the difference in the 24hEE values during fasting and eucaloric conditions, we investigated which component was responsible for the larger decrease in 24hEE characterizing thrifty individuals (e.g., lower fasting 24hEE, higher eucaloric 24hEE, or both). Contrary to what could have been expected, we found that thrifty individuals had relatively higher 24hEE during eucaloric conditions, but similar 24hEE during fasting, compared to spendthrift individuals after adjustment for differences in body composition and other know EE determinants(12). In other words, a greater metabolic slowing during fasting was the result of a relatively higher 24hEE level during energy balance conditions rather than a lower 24hEE level during fasting conditions. This finding might appear paradoxical as, per definition, thriftier individuals are more “energy efficient” and, as such, should have a lower (instead of higher) rate of EE in any dietary condition.

One potential explanation is that the relatively higher rate of EE in constrained, highly controlled, fed conditions observed in thrifty individuals could represent a driver for overeating in free-living conditions where energy intake is not constrained. In support of this hypothesis, numerous independent studies have consistently demonstrated that relatively higher EE (both resting and non-resting components) after accounting for body size and other confounders is associated with higher habitual and ad libitum food consumption(19). This putative, causal link between EE and energy intake has been termed “energy sensing”(20, 21). According to this mechanism whereby higher EE may drive greater energy intake, it could be speculated that thrifty individuals may have a greater tendency to overeat in environments with unrestricted access to food due to their elevated energy requirements that might increase hunger and promote overeating. In other words, the orexigenic effect of relatively higher EE might exceed the energy-consuming effect of increased metabolic rate, thereby promoting positive energy balance and consequently leading to weight gain over time(22). The effect of increased EE on energy intake might be stronger in settings of low energy-turnover (e.g., inside a metabolic chamber when physical activity level is reduced) where appetite control is weaker and tendency to overeating is increased(23) compared to high energy-turnover settings such as those characterized by higher levels of physical activity. This “energy sensing” hypothesis underlying the greater propensity for weight gain in thrifty individuals with relatively higher 24hEE during sedentary conditions was confirmed by an independent study showing that higher eucaloric 24hEE predicted greater weight and fat mass regain in overweight individuals after 6-week diet-induced weight loss(2).

Future directions

The accurate assessment of metabolic phenotype based on 24hEE measurements requires specific instrumentation and clinical protocols. The gold standard method for measuring 24hEE involves the operation of a whole-room indirect calorimeter, which is an expensive and complex apparatus that requires extensive maintenance by research technicians and monitoring of volunteers by nursing and medical staff, especially for overnight studies. Therefore, a simplified procedure or method for assessing metabolic phenotypes in everyday clinical practice would be desirable for large-scale population studies.

To address this issue, the search of a cost-effective “thrifty” biomarker would be helpful. Several hormonal signatures, such as epinephrine, leptin, and ghrelin have been associated with the extent of metabolic thriftiness observed during acute fasting conditions(12, 24). However, the strength of these hormonal associations with the concomitant changes in EE is not sufficiently strong to render these hormones as viable biomarkers that can replace the actual EE measurement via indirect calorimetry. Therefore, future studies are warranted to identify novel biomarker(s) that mediate the EE changes during prolonged fasting condition, such as using plasma metabolomics to identify proteins or protein patterns robustly associated with metabolic thriftiness.

Because the metabolic differences between thrifty vs. spendthrift phenotypes mostly arise from the fed state rather than the fasted state(12), measurement of DIT during meal tests, preferably using low- or high-protein meals, along with blood draws for hormonal measurements could constitute another method to quantify metabolic thriftiness. Specifically, the fibroblast growth factor 21 (FGF21) response to low-protein meals has been advocated as the most promising biomarker for thriftiness as plasma FGF21 concentration markedly increases up to 3-fold during 24-h low-protein overfeeding, with smaller increase in FGF21 concentration being associated with smaller DIT and greater extent of future weight gain(16).

Identifying thrifty and spendthrift phenotypes in overweight individuals or individuals with obesity can contribute to personalizing therapies in clinical practice. For example, normal-weight or overweight individuals with a thrifty phenotype could be identified before gaining weight and informed about their greater susceptibility to future weight gain. Further, individuals with obesity displaying a thrifty phenotype, who would not fully benefit from dietary restriction interventions due to their greater energy-saving capabilities in such dietary conditions, could be directed more quickly to alternative weight loss interventions, such as pharmaceutical therapies or bariatric surgery. By gaining more insight into the phenotypic differences among different metabolic phenotypes in the general population, obesity therapies could be personalized for thrifty individuals to mitigate their inherent predisposition for energy saving and greater susceptibility to future weight gain. However, future larger clinical trials are warranted to determine the prevalence of thrifty vs. spendthrift metabolic phenotypes in the general population, as well as the efficacy of new anti-obesity drugs in subjects with different metabolic profiles.

Conclusion

Metabolic thriftiness can be assessed by EE measurements during short-term fasting, overfeeding, mild cold exposure, and sedentary conditions (Figure 2). The metabolic response to 24-h fasting has proven to be the most reliable method for quantifying thriftiness, likely due to the pronounced effects of fasting on energy homeostasis, which best activates individual-specific energy conservation mechanisms. Because whole-room calorimetry is required to quantify the extent of metabolic thriftiness in research settings, future studies should be conducted to identify alternative methods (e.g., hormonal biomarkers during meal tolerance trsting) for clinical and outpatient settings that allow the assessment of metabolic phenotypes in large-scale population studies.

Figure 2 – Current methods to assess metabolic thriftiness.

Figure 2 –

Open in a new tab

24hEE, 24h energy expenditure; RMR, resting metabolic rate.

Key Points.

  • Inter-individual variability in body weight change is not fully explained by differences in energy intake and physical activity levels, suggesting the existence of distinct metabolic phenotypes (e.g., thrifty vs. spendthrift).

  • Short-term metabolic responses to acute fasting, low-protein overfeeding, and mild cold exposure can quantify the degree of metabolic thriftiness in each individual.

  • Currently, whole-room calorimetry is the primary method used to assess thriftiness, but alternative methods such as hormonal biomarkers during meal tolerance testing are being tested for use in large-scale studies.

Acknowledgements

The figure has been designed using images from Flaticon.com, made from the following authors: Dailypm Studio, Freepik, max.icons, Flat icons, Vectors market.

Funding source:

This study was supported by the Intramural Research Program of the National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases.

Financial support and sponsorship

T.H. was supported by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) - Project 413490537. P.P. was supported by the program “Rita Levi Montalcini for young researchers” from the Italian Minister of Education and Research (Ministero dell’Istruzione, dell’Università e della Ricerca).

Abbreviations:

24hEE

24h energy expenditure

EE

energy expenditure

RMR

resting metabolic rate

Footnotes

Conflict of interest Statement: The authors have nothing to disclose.

References

  • 1.Reinhardt M, Thearle MS, Ibrahim M, Hohenadel MG, Bogardus C, Krakoff J, et al. A Human Thrifty Phenotype Associated With Less Weight Loss During Caloric Restriction. Diabetes. 2015;64(8):2859–67. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Hollstein T, Heinitz S, Basolo A, Krakoff J, Votruba SB, Piaggi P. Reduced metabolic efficiency in sedentary eucaloric conditions predicts greater weight regain in adults with obesity following sustained weight loss. Int J Obes (Lond). 2021;45(4):840–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Hollstein T, Ando T, Basolo A, Krakoff J, Votruba SB, Piaggi P. Metabolic response to fasting predicts weight gain during low-protein overfeeding in lean men: further evidence for spendthrift and thrifty metabolic phenotypes. Am J Clin Nutr. 2019;110(3):593–604. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Schlogl M, Piaggi P, Pannacciuli N, Bonfiglio SM, Krakoff J, Thearle MS. Energy Expenditure Responses to Fasting and Overfeeding Identify Phenotypes Associated With Weight Change. Diabetes. 2015;64(11):3680–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Reinhardt M, Schlogl M, Bonfiglio S, Votruba SB, Krakoff J, Thearle MS. Lower core body temperature and greater body fat are components of a human thrifty phenotype. International journal of obesity (2005). 2016;40(5):754–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Chen KY, Smith S, Ravussin E, Krakoff J, Plasqui G, Tanaka S, et al. Room Indirect Calorimetry Operating and Reporting Standards (RICORS 1.0): A Guide to Conducting and Reporting Human Whole-Room Calorimeter Studies. Obesity (Silver Spring). 2020;28(9):1613–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.*.Dörner R, Hägele FA, Koop J, Rising R, Foerster T, Olsen T, et al. Validation of energy expenditure and macronutrient oxidation measured by two new whole‐room indirect calorimeters. Obesity. 2022;30(9):1796–805. [DOI] [PubMed] [Google Scholar]; Annotation: Measurements of 24-hour energy expenditure are reproducible between different whole-room indirect calorimeters.
  • 8.Henriksen HB, Henriksen C, De Sousa ARS, Alavi DT, Augestad EMS, Rising R, et al. Validity and reproducibility of a whole‐room indirect calorimeter for the estimation of VO2, VCO2. Physiological Reports. 2023;11(7). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Piaggi P, Rodzevik TL, Wohlers E, Ruud K, Moon J, Krakoff J, et al. Closed-loop control of air supply to whole-room indirect calorimeters to improve accuracy and standardize measurements during 24-hour dynamic metabolic studies. Obesity. 2023;31(3):780–8. [DOI] [PubMed] [Google Scholar]
  • 10.Stinson EJ, Rodzevik T, Krakoff J, Piaggi P, Chang DC. Energy expenditure measurements are reproducible in different whole‐room indirect calorimeters in humans. Obesity. 2022;30(9):1766–77. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Müller MJ, Heymsfield SB, Bosy‐Westphal A. Changes in body composition and homeostatic control of resting energy expenditure during dietary weight loss. Obesity. 2023;31(4):892–5. [DOI] [PubMed] [Google Scholar]
  • 12.**.Hollstein T, Basolo A, Ando T, Votruba SB, Walter M, Krakoff J, et al. Recharacterizing The Metabolic State of Energy Balance in Thrifty and Spendthrift Phenotypes. J Clin Endocrinol Metab. 2020;105(5):1375–92. [DOI] [PMC free article] [PubMed] [Google Scholar]; Annotation: Thrifty individuals expend more energy relative to their body size in the sedentary, eucaloric state.
  • 13.*.Johannsen DL, Marlatt KL, Conley KE, Smith SR, Ravussin E. Metabolic adaptation is not observed after 8 weeks of overfeeding but energy expenditure variability is associated with weight recovery. Am J Clin Nutr. 2019;110(4):805–13. [DOI] [PMC free article] [PubMed] [Google Scholar]; Annotation: Low sleeping metabolic rate (as indicative of a more thrifty phenotype) was associated with greater fat mass retention after sustained overfeeding.
  • 14.Spranger L, Bredow J, Zeitz U, Grittner U, Boschmann M, Dickmann S, et al. Thrifty energy phenotype predicts weight regain - results of a randomized controlled trial. 2021. [DOI] [PubMed]
  • 15.*.Hollstein T, Basolo A, Takafumi A, Krakoff J, Piaggi P. Reduced adaptive thermogenesis during acute protein-imbalanced overfeeding is a metabolic hallmark of the human thrifty phenotype. Am J Clin Nutr. 2021;114(4):1396–407. [DOI] [PMC free article] [PubMed] [Google Scholar]; Annotation: Thrifty individuals have less ability to increase their metabolic rate during protein-imbalanced overfeeding.
  • 16.Vinales KL, Begaye B, Bogardus C, Walter M, Krakoff J, Piaggi P. FGF21 is a Hormonal Mediator of the Human “Thrifty” Metabolic Phenotype. Diabetes. 2018. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Hollstein T, Heinitz S, Ando T, Rodzevik TL, Basolo A, Walter M, et al. Metabolic Responses to 24-Hour Fasting and Mild Cold Exposure in Overweight Individuals Are Correlated and Accompanied by Changes in FGF21 Concentration. Diabetes 2020;69(7):1382–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.*.Hollstein T, Vinales K, Chen KY, Cypess AM, Basolo A, Schlögl M, et al. Reduced brown adipose tissue activity during cold exposure is a metabolic feature of the human thrifty phenotype. Metabolism. 2021;117:154709. [DOI] [PMC free article] [PubMed] [Google Scholar]; Annotation: Thrifty individuals show lower activity of cold-induced brown adipose tissue.
  • 19.Blundell JE, Gibbons C, Beaulieu K, Casanova N, Duarte C, Finlayson G, et al. The drive to eat in homo sapiens: Energy expenditure drives energy intake. Physiology & Behavior. 2020;219:112846. [DOI] [PubMed] [Google Scholar]
  • 20.Piaggi P, Vinales KL, Basolo A, Santini F, Krakoff J. Energy expenditure in the etiology of human obesity: spendthrift and thrifty metabolic phenotypes and energy-sensing mechanisms. J Endocrinol Invest. 2018;41(1):83–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Piaggi P Metabolic Determinants of Weight Gain in Humans. Obesity. 2019;27(5):691–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Piaggi P, Basolo A, Martin CK, Redman LM, Votruba SB, Krakoff J. The counterbalancing effects of energy expenditure on body weight regulation: Orexigenic versus energy‐consuming mechanisms. Obesity. 2022;30(3):639–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Hägele FA, Büsing F, Nas A, Hasler M, Müller MJ, Blundell JE, et al. Appetite control is improved by acute increases in energy turnover at different levels of energy balance. The Journal of Clinical Endocrinology & Metabolism. 2019;104(10):4481–91. [DOI] [PubMed] [Google Scholar]
  • 24.Hollstein T, Basolo A, Unlu Y, Ando T, Walter M, Krakoff J, et al. Effects of Short-term Fasting on Ghrelin/GH/IGF-1 Axis in Healthy Humans: The Role of Ghrelin in the Thrifty Phenotype. The Journal of Clinical Endocrinology & Metabolism. 2022;107(9):e3769–e80. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES