Table 2 Measured energy expenditure

Resting Energy Expenditure (REE)

Restricting

Melchior 1989

831

1017

P < 0.05

22.38%

When expressed per kg of LBM, REE was unchanged after dietary repletion and weight gain. After weight gain, REE was correlated with LBM (r = 0.69, P < 0.02). When expressed as % basal values, the increase of REE was positively correlated with LBM (r = 0.64, P < 0.05).

Scalfi 1993

836.28

1181.17

P < 0.01

41.24%

EE remained significantly lower even after adjustment for body composition (FFM and FM) or body weight.

Moukaddem 1997

1031.55

1104.69

P < 0.021

7.09%

When the increase in body weight and LBM was taken into account, one week of refeeding did not modify REE.

Van Vymelbeke 2004

918.74

1397.80

P < 0.001

52.14%

The ratio of REE to FFM was significantly related to energy intake (P < 0.01), anxiety (P < 0.01), abdominal pain (P < 0.05), and depressive mood (P < 0.05). The ratio also increased significantly with physical activity (P < 0.01) and cigarette smoking (P < 0.02). This rise in REE leveled off after recovery from AN.

Yoshida 2006

761.9

934.0

P = 0.000

paired t-test

22.59%

EE remained significantly lower at baseline even after adjustment for LBM.

Cuerda 2007

1106

1241

P = 0.011

Wilcoxon

12.21%

REE (kcal/day) significantly increased during hospitalization. However, the increase in REE/kg FFM did not reach statistical significance.

Unspecified

Dempsey 1984

766

1240

Not reported

61.88%

Vaisman 1991

831.98

1108.99

P < 0.025

33.30%

EE remained significantly lower at admission even when expressed per kg of body weight or unit of LBM. No significant correlation was found between total energy or protein intake and total REE or REE per kg of LBM.

Krahn 1993

1166

1769

Not reported

51.72%

Schebendach 1997

895.2

1123.8

P < 0.001

25.54%

Svobodovå 1999

1188.58

1257.89

n.s.

5.83%

Pagliato 2000

857

1100

P = 0.0028

28.35%

A positive correlation was found at T1 (r = 0.62) between lean body mass and REE, whereas at T0, the correlation was not significant.

Pauly 2000

1086

1193

P = 0.0037

9.85%

Rigaud 2000

918.74

1233.27

P < 0.01

34.23%

Russell 2001

1038

1211

P < 0.001

16.67%

Satoh 2003

830

1230

P < 0.01

48.19%

Vaisman 2004

882.4

1006.7

P < 0.01

14.09%

Cuerda Compés 2005

1165.2

1225.5

P < 0.05

5.18%

Haas 2005

989.48

1123.33

P < 0.05

13.53%

No significant correlation between serum leptin and REE (tested within AN patients only).

Onur 2005

1027.72

1123.33

P < 0.01

Wilcoxon

9.30%

There was a highly significant association between weight gain-induced changes in Triiodothyronine and changes in adjusted REE (r = 0.78, P < 0.001, based on Pearson’s correlation). An increase in plasma Triiodothyronine concentrations of 1.8 pmol/l could explain an increase in REE of 0.6 MJ/day (that is, a 32% increase in Triiodothyronine was associated with a 13% increase in REE).

Winter 2005

1058

1133

n.s.

7.09%

Dragani 2006

882.74

992.38

P < 0.03

12.42%

Pre-/post-rehabilitation difference: P < 0.005. REE/day shows significant positive correlation with BMI, weight, arm muscle area, and arm fat area (P < 0.001). Change in REE/day significantly correlates with change in BMI (P < 0.01).

Haas 2010

995

1158

P < 0.001

16.38%

Positive association between leptin and REE_LTM in AN (r² = 0.14, P = 0.001). After weight gain, the slope of the regression line was less steep when compared with AN before weight gain.

Dostálová 2010

1060.0

1176.3

P < 0.05

10.97%

Sum 2011

1087

1378

P < 0.05

26.77%

Soto-Célix 2013

1023

1128

P = 0.08

10.26%

Kochavi 2020

972.60

1255.40

Not reported

29.08%

Admission measured REE was positively correlated with admission weight and BMI, as well as with longer duration of illness. Measured discharge REE was positively correlated with duration of illness. Greater changes in measured REE were correlated with lower admission weight, BMI, and weight/height ratio, and with greater increase in weight from admission to discharge. No significant correlations were found between any of the REE measurements and thyroid hormone and cortisol levels at admission and discharge or the change in their levels from admission to discharge.

Total Energy Expenditure (TEE)

Restricting

Platte 1994

1946

2602

n.s.

33.71%

Unspecified

Pettersson 2016

1568

2034

P < 0.0001

29.72%

Resting Metabolic Rate (RMR)

Restricting

Platte 1994

1171

1330

n.s.

13.58%

Significant correlations between RMR and LBM were found in the weight-regained group (r = 0.84; P < 0.05), but not in the acute group (r = 0.39; ns).

Unspecified

Obarzanek 1994

742

1208

P < 0.05

62.80%

There were no significant correlations among plasma norepinephrine, thyroid hormones and RMR.

Konrad 2007

1015.33

1137.33

P < 0.05

12.02%

A trend toward smaller increases in RMR being associated with higher admission BMI (r =-0.49, P = 0.08), but not with highest lifetime BMI.

Refeeding is associated with an increase in RMR that is not accounted for by the increase in FFM.

DosReis 2022

864.00

1325.00

P = 0.037

Pairwise Kruskal-Wallis test adjusted by bonferroni

53.36%

Basal Metabolic Rate (BMR)

Restricting

Moukaddem 1997

993.07

1099.67

P < 0.021

10.73%

Unspecified

Forman-Hoffman 2006

1000

1220

Yes

22.00%

Median change: 182 kcal/day, P = 0.02

Pichard 1996

969

1360

Not reported

40.35%

Polito 2000

938.93

1110.08

P < 0.002

18.23%

BMR was significantly correlated with FFM (r² = 0.48, P < 0.0000) and body weight (r² = 0.62, P < 0.0000). The regression equation showed that 62% of the variance in BMR was attributed to differences in body weight. When BMR was regressed against FFM, 48% of the variance was explained by FFM. A positive correlation was found also between the logarithm of leptin concentration and BMR (r² = 0.28, P < 0.00006) even after adjustment for FFM (r² = 0.21, P < 0.0003).