Current Evidence for Estimating Energy Requirements Clare Soulsby, Research Dietitian.

Current Evidence for Estimating Energy Requirements

Clare Soulsby, Research Dietitian

Main components of energy expenditure:

– basal metabolic rate (BMR)– alteration in BMR due to disease process

(stress factors)– activity– diet induced thermogenesis (DIT)

Estimating BMR: controversies

basal metabolic rate (BMR) vs. resting energy expenditure (REE)

prediction equations vs. measured energy expenditure (MEE)

Conditions essential for measuring BMR

post-absorptive (12 hour fast) lying still at physical and mental rest thermo-neutral environment (27 – 29oC) no tea/coffee/nicotine in previous 12 hours no heavy physical activity previous day gases must be calibrated establish steady-state (~ 30 minutes)

* if any of the above conditions are not met = REE

Estimating BMR: controversies

basal metabolic rate (BMR) vs. resting energy expenditure (REE)

prediction equations vs. measured energy expenditure (MEE)

Estimating BMR: prediction equations

may over or under-estimate (compared with MEE)

inadequately validated poor predictive value for individuals open to misinterpretation

(Cortes & Nelson, 1989; Malone, 2002; Reeves & Capra, 2003)

Estimating BMR:which equation?

• Harris-Benedict

• Schofield Equations

• disease specific eg Ireton Jones

• Kcal/kg

Estimating BMR: Harris Benedict Equations

• Developed in 1919• From data collected between 1909 and 1917

(Harris Benedict 1919)

• Study population: – 136 men; mean age 27 ± 9 yrs, mean BMI

21.4 ± 2.8– 103 women; mean age 31 ± 14 yrs, mean BMI

21.5 ± 4.1• Tends to overestimate in healthy individuals

(Daly 1985, Owen 1986, Owen 1987)

Estimating BMR: Schofield Equations

• developed in 1985 (Schofield 1985)• meta analysis of 100 studies of 3500men and

1200 women• studies conducted between 1914 and 1980

(including Harris Benedict data)• 2200 (46%) subjects were military Italian adults • 88 (1.2%) subjects were >60 years • SE 153-164kcal/d (women) 108 -119kcal/d

(men) (Schofield 1985)

Estimating BMR: disease specific equations

• developed for specific patient groups (Ireton Jones 1992, Ireton Jones 2002)

• advantage over Schofield/ HB equations:– Schofield /HB estimate BMR of a healthy

individual then necessary to adjust for disease using a stress factors

– disease specific equations include patients in their database so aim to more accurately reflect BMR of hospitalised patients

Estimating BMR: Ireton-Jones energy equations

• ventilated and breathing ICU patients

• 3 x 1 minute measurements 200 patients

• unclear whether measurements took place during feed infusion/ after treatment etc

• 52% burns, 31% trauma

• validation studies, IJEE had a better agreement with MEE: – HBx1.2, HBx1.3, 21kcal/kg

Estimating BMR• Schofield equation derived using meta analysis:

– greater power than small/ local studies

• compiled from unstructured data set obtained for diverse reasons:– problems with sampling assumptions

• accuracy approx ±15%• disease specific equations useful in some

circumstances

Estimating BMR

• what about:– the elderly?– the obese?

Estimating BMR: the elderly

• Original Schofield equations:– only 88 (1.2%) of subjects >60 years– particularly unsuitable for >75yr– included data on subjects from the tropics

• Revised equations for the elderly:– published in the 1991 COMA (DH 1991)– include additional data from 2 studies; 101

Glaswegian men (60-70yr) 170 Italian men and 180 Italian women

– excluded data collected in the tropics

Estimating BMR: Obesity

• equations (such as Schofield) are linear• weight increases linearly with estimated BMR• may overestimate in obese

weight

BMR

Estimating BMR: obesity

BMI % of Schofield database

% of UK population (DOH 1999)

> 25 14.6% 40.8%

> 30 4.5% 9.7%

Estimating BMR: Obesity

• obese data primarily obtained from 2 groups:– Burmese hill dwellers– retired Italian military

• there were significant differences in weight/ BMR association between groups, Italian group showed greatest difference

• obese subjects in Schofield data may not be a statistically representative sample of the population is general

Estimating BMR: Obesity

• recent (Horgan 2003) reassessed validity of the Schofield data to predict BMR in obese

• conclusions:– BMR increases more slowly at heavier weights– to ignore this is to over predict energy requirements– any general equation for predicting BMR may be

biased for some groups or populations.

Estimating BMR: adjusted body weight (ADJ)

estimate of how much of the extra body weight is lean and thus metabolically active

2 methods: 25% adjusted weight

= (actual body weight x 0.25) + ideal body weight

adjusted average weight = (actual body weight + ideal body weight) x 0.5

Estimating BMR: adjusted body weight (ADJ)

first reported in newsletter Q&A format not validated studies suggest adjusted average weight

has better predictive value than 25% adjusted weight (Glynn 1998, Barak 2002)

no longer included in ASPEN guidelines (2002)

Estimating BMR: Obesity

predicting BMR is very difficult (without measuring lean body mass)

adequacy of specific equations? (Ireton-Jones et al., 1992; Glynn et al., 1998)

• actual body weight + stress + activity = overestimate

access to indirect calorimetry is limited

Determining energy requirements in obesity

• non stressed patients:– calculate as normal and - 400-1000kcal for decrease

in energy stores

• mild to moderately stress:– calculate as normal – omission of stress and activity avoids the adverse

effects of overfeeding

• severe stress– might be necessary to add a stress factor to BMR

• *monitoring essential eg blood glucose

Estimating energy requirements

• The main components of energy expenditure are estimated:– BMR– Alteration in BMR due to disease process

(stress factors)– Activity– DIT

Levels of evidence

1. a) Meta-analysesb) Systematic reviews of randomised controlled trials (RCTs)c) RCTs

2. a) Systematic reviews of case-control or cohort studiesb) Case-control or cohort studies

3. Non-analytic studies e.g. case studies4. Expert opinion(adapted from: Draft NICE Guidelines for Nutrition Support in Adults, 2005)

Stress factors

timing of measurements over (hyperalimentation) vs. under-feeding changes in therapeutic interventions

e.g. improved wound care, anti-pyretics, sedation, control of ambient room temperature

err towards lower end of the range and monitor

Stress factors

• unable to include a stress factor for every disease or condition

• many measured in far from ideal circumstances• limited by data available• may choose to underfeed in certain

circumstances• necessary to refer back to the literature• included a checklist of factors to look for when

reviewing papers

Adverse effect of over-feeding

• excess carbohydrate:– difficulties controlling blood glucose– increased CO2 production– respiratory problems in vulnerable patients (eg

COPD/ ventilated)• swings in blood glucose increase mortality in

critically ill• aim not to exceed the glucose oxidation rate (4-7

mg glucose/ kg/ min)• long term excess carbohydrate can lead to

steatohepatosis or fatty liver (Elwyn DH, 1987).

Estimating energy requirements

• The main components of energy expenditure are estimated:– BMR– Alteration in BMR due to disease process– Activity– DIT

Total energy expenditure

BMR

Activity+ DIT

Activity+ DIT

Health Disease

BMR

Activity factor

• energy expended during active movement of skeletal muscle

• approximately 20-40% of energy expenditure in free living individuals

• depends on duration and intensity of the exercise

• activity is less than 20% of the energy expenditure in hospitalised or institutionalised

• NB assumes normal muscle function

Activity factor for activity: institutionalised patients combined

with DIT

Activity level Males and females

Bedbound immobile

Bedbound mobile/ sitting

Mobile on ward

+ 10%

+ 15 – 20%

+ 25%

Activity factor:abnormal muscle function

• hospital patients likely to have higher activity levels:– abnormal neuro-muscular function e.g. brain

injury, Parkinson’s, cerebral palsy, motor neurone disease, and Huntington’s chorea

– prolonged active physiotherapy– effort involved in moving injured or painful

limbs

Community patients

• free living individuals have higher energy expenditure due to physical activity

• nursing home and house bound patients ? similar activity levels to hospital patients

• for active patients in the community a PAL should be added

Physical activity level (PAL) of adults

Non-occupational activity

occupational activity

light

M F

occupational activity

moderate

M F

occupational activity

mod/ heavy

M F

non active

m. active

very active

1.4 1.4

1.5 1.5

1.6 1.6

1.6 1.5

1.7 1.6

1.8 1.7

1.7 1.5

1.8 1.6

1.9 1.7

Estimating energy requirements

• The main components of energy expenditure are estimated:– BMR– Alteration in BMR due to disease process– Activity– DIT

Diet-induced thermogenesis

Continuous infusion of enteral feed and parenteral nutrition do not significantly increase REE

Bolus feeding increases REE by ~ 5% Mixed meal increases REE ~ 10 % PALs include DIT (COMA, 1991)

guidelines include combined factor for activity and DIT

Estimating requirements: sources of error

• prediction equation for BMR

• stress factor:– degree of stress inaccurately assessed– poor evidence to support stress factor used

• activity level inaccurately assessed or poorly understood

• DIT varies by 10% depending on feeding method

Sources of error: inaccurate weight

• Inaccurately measured weight – estimated weight– inaccurate scales– patient had their feet on the floor (chair

scales)– patient was fluid overloaded ( 20% of

hospital patients)– amputees

Conclusions

Estimated requirements are only a starting point- set realistic goals of treatment for each patient- monitor and amend as patient’s condition changes

Review and criticise the literature regularly- be aware of gaps in the evidence- understand the limitations of guidelines- check applicability to your patients

Contribute to research and audit projects