How do guideline recommended energy targets compare with measured energy expenditure in critically ill adults with obesity: A systematic literature review

BACKGROUND

Critically ill patients with obesity have unique and complex nutritional needs, with clinical practice guidelines conflicting regarding recommended energy targets. The aim of this systematic review was to 1) describe measured resting energy expenditure (mREE) reported in the literature and; 2) compare mREE to predicted energy targets using the European (ESPEN) and American (ASPEN) guideline recommendations when indirect calorimetry is not available in critically ill patients with obesity.

METHODS

The protocol was registered apriori and literature was searched until 17th March, 2022. Original studies were included if they reported mREE using indirect calorimetry in critically ill patients with obesity (BMI≥30 kg/m²). Group-level mREE data was reported as per the primary publication using mean ± standard deviation or median [interquartile range]. Where individual patient data was available, Bland-Altman analysis was used to assess mean bias (95% limits of agreement) between guideline recommendations and mREE targets (i.e. ASPEN for BMI 30-50, 11-14 kcal/kg actual weight compared to 70% mREE and ESPEN 20-25 kcal/kg adjusted weight compared to 100% mREE). Accuracy was assessed by the percentage (%) of estimates within ±10% of mREE targets.

RESULTS

After searching 8019 articles, 24 studies were included. mREE ranged from 1607 ± 385 to 2919 [2318-3362]kcal and 12-32kcal/actual body weight. For the ASPEN recommendations of 11-14 kcal/kg, a mean bias of -18% (-50% to +13%) and 4% (-36% to +44%) was observed, respectively (n = 104). For the ESPEN recommendations 20-25 kcal/kg, a bias of -22% (-51% to +7%) and -4% (-43% to +34%), was observed, respectively (n = 114). The guideline recommendations were able to accurately predict mREE targets on 30%-39% occasions (11-14 kcal/kg actual) and 15%-45% occasions (20-25 kcal/kg adjusted), for ASPEN and ESPEN recommendations, respectively.

CONCLUSIONS

Measured energy expenditure in critically ill patients with obesity is variable. Energy targets generated using predictive equations recommended in both the ASPEN and ESPEN clinical guidelines have poor agreement with mREE and are frequently not able to accurately predict within ±10% of mREE, most commonly underestimating energy needs.

Resting Energy Expenditure in the Critically Ill and Healthy Elderly-A Retrospective Matched Cohort Study

The use of indirect calorimetry to measure resting energy expenditure (mREE) is widely recommended as opposed to calculating REE (cREE) by predictive equations (PE). The aim of this study was to compare mREE with cREE in critically ill, mechanically ventilated patients aged ≥ 75 years and a healthy control group matched by age, gender and body mass index. The primary outcome was the PE accuracy rate of mREE/cREE, derived using Bland Altman plots. Secondary analyses included linear regression analyses for determinants of intraindividual mREE/cREE differences in the critically ill and interindividual mREE differences in the matched healthy cohort. In this retrospective study, 90 critically ill patients (median age 80 years) and 58 matched healthy persons were included. Median mREE was significantly higher in the critically ill (1457 kcal/d) versus the healthy cohort (1351 kcal/d), with low PE accuracy rates (21% to 49%). Independent predictors of mREE/cREE differences in the critically ill were body temperature, heart rate, FiO₂, hematocrit, serum sodium and urea. Body temperature, respiratory rate, and FiO₂ were independent predictors of interindividual mREE differences (critically ill versus healthy control). In conclusion, the commonly used PE in the elderly critically ill are inaccurate. Respiratory, metabolic and energy homeostasis variables may explain intraindividual mREE/cREE as well as interindividual mREE differences.

Use of Predictive Equations for Energy Prescription Results in Inaccurate Estimation in Trauma Patients

BACKGROUND

Overfeeding and underfeeding are associated with poor clinical outcomes. In the absence of indirect calorimetry (IC), the Society of Critical Care Medicine/ASPEN recommend prescribing 25-30 kcal/kg. The Harris-Benedict equation (HBE) multiplied by a stress factor is commonly applied in critically ill patients. We describe the difference between estimated and actual energy needs in critically injured patients.

METHODS

From March to November 2018, we collected demographics and energy needs determined by continuous IC (started within 4 days) in intubated adults. Ideal or adjusted body weight was used for 25-30 kcal/kg, and HBE was multiplied by a 1.3 stress factor (1.3HBE). Daily requirements up to 14 days, extubation, or death were calculated using all 3 methods and compared with IC.

RESULTS

Fifty-five subjects were included. Median age was 38 [27-58] years, 38 (69%) were male, body mass index was 28 [25-33] kg/m² , and Acute Physiology and Chronic Health Evaluation II score was 17 [14-24] Mechanism of injury was blunt (38, 69%), penetrating (9, 16%), and burn (8, 15%). By day 14, compared with measured energy requirements by IC, the other methods could result in a cumulative 1827-kcal (+7%) surplus (1.3HBE), a 1313-kcal (-5%) deficit (25 kcal/kg), or a 3950-kcal (+14%) surplus (30 kcal/kg) per patient over a median 9 days.

CONCLUSION

In critically injured patients, predictive equations for energy needs do not account for dynamic metabolic changes over time and could result in underfeeding or overfeeding. Adjusting daily prescription based on continuous IC may result in better individualized treatment.

Systematic review of factors associated with energy expenditure in the critically ill

BACKGROUND AND AIMS

Indirect calorimetry is the reference standard for energy expenditure measurement. Predictive formulae that replace it are inaccurate. Our aim was to review the patient and clinical factors associated with energy expenditure in critically ill patients.

METHODS

We conducted a systematic review of the literature. Eligible studies were those reporting an evaluation of factors and energy expenditure. Energy expenditure and factor associations with p-values were extracted from each study, and each factor was classified as either significantly, indeterminantly, or not associated with energy expenditure. Regression coefficients were summarized as measures of central tendency and spread. Metanalysis was performed on correlations.

RESULTS

The search strategy yielded 8521 unique articles, 307 underwent full text review, and 103 articles were included. Most studies were in adults. There were 95 factors with 352 evaluations. Minute volume, weight, age, % body surface area burn, sedation, post burn day, and caloric intake were significantly associated with energy expenditure. Heart rate, fraction of inspired oxygen, respiratory rate, respiratory disease diagnosis, positive end expiratory pressure, intensive care unit days, C- reactive protein, and size were not associated with energy expenditure. Multiple factors (n = 37) were identified with an unclear relationship with energy expenditure and require further evaluation.

CONCLUSIONS

An important interval step in the development of accurate formulae for energy expenditure estimation is a better understanding of relationships between patient and clinical factors and energy expenditure. The review highlights the limitations of currently available data, and identifies important factors that are not included in current prediction formulae of the critically ill.

Are Predictive Energy Expenditure Equations Accurate in Cirrhosis?

Malnutrition is associated with significant morbidity and mortality in cirrhosis. An accurate nutrition prescription is an essential component of care, often estimated using time-efficient predictive equations. Our aim was to compare resting energy expenditure (REE) estimated using predictive equations (predicted REE, pREE) versus REE measured using gold-standard, indirect calorimetry (IC) (measured REE, mREE). We included full-text English language studies in adults with cirrhosis comparing pREE versus mREE. The mean differences across studies were pooled with RevMan 5.3 software. A total of 17 studies (1883 patients) were analyzed. The pooled cohort was comprised of 65% men with a mean age of 53 ± 7 years. Only 45% of predictive equations estimated energy requirements to within 90⁻110% of mREE using IC. Eighty-three percent of predictive equations underestimated and 28% overestimated energy needs by ±10%. When pooled, the mean difference between the mREE and pREE was lowest for the Harris⁻Benedict equation, with an underestimation of 54 (95% CI: 30⁻137) kcal/d. The pooled analysis was associated with significant heterogeneity (I² = 94%). In conclusion, predictive equations calculating REE have limited accuracy in patients with cirrhosis, most commonly underestimating energy requirements and are associated with wide variations in individual comparative data.

Indirect calorimetry as point of care testing

Determining energy requirement is a fundamental of nutrition support. Indirect calorimetry (IC) has been long recognized as the gold standard for assessing basal or resting energy expenditure (REE). The measurement of REE is recommended particularly in the situation where adjustment of energy provision is critical. The result of the IC measurement can lead to changes in treatment and since the change can be carried out immediately at the bedside, this may be considered as point-of-care testing. Beyond the nutritional aspects, studies of energy expenditure with IC have brought out more understanding of the metabolic changes during the natural course of diseases or conditions as well as those related to the intervention. The literature in various disease states has shown that changes in energy expenditure may reveal hidden metabolic information that might be translated into clinical information and have the potential of being both prognostic indicators and/or treatment targets.

Reliability of resting energy expenditure in major burns: Comparison between measured and predictive equations

BACKGROUND & AIMS

Poor outcomes can result from inadequate energy intake. We aimed to investigate the reliability of resting energy expenditure (REE) measured by indirect calorimetry (IC) with REE calculated using predictive equations for nutritional support in patients with major burns.

METHODS

REE was measured using IC and compared with predictive equations in 215 adult severe burns patients from Jan 2011 to Jun 2015. Agreement between IC and predictive equations was assessed using Bland-Altman methods.

RESULTS

All predictive equations, including newly developed Hangang equation, were compared with REE measured using IC. The mean measured REE was 1712 kcal/d. Bland-Altman analysis showed that 1.2 times HBE, Thumb 25, and Ireton-Jones equations had higher accuracy and reliability. The concordance correlation coefficient was higher (0.49) in the Ireton-Jones equation, and root mean square error (RMSE) was lowest (471.5) in the Thumb 25 equation. The proportion of patients with predicted REE within ±10% of measured REE was highest with Thumb 25 (52.5%). In the validation set, the Hangang equation showed the highest Lin's concordance correlation coefficient (0.67) and the lowest RMSE (311.4). Other equations for burns patients had higher mean bias and overestimated REE when compared with IC results.

CONCLUSIONS

This study suggests that Thumb 25 can be used as an alternative method for estimating energy requirements of patients with major burns when IC is not available or applicable. However, for these patients with significant variation in metabolism over time, an alternative equation is the new Hangang equation.

Establishment and assessment of new formulas for energy consumption estimation in adult burn patients

OBJECTIVE

An accurate knowledge of energy consumption in burn patients is a prerequisite for rational nutrition therapy. This study sought to create a formula that accounts for the metabolic characteristics of adult burn patients to accurately estimate energy consumption of patients with different areas and extents of burn and at different times after injury.

METHODS

Resting energy expenditure (REE) data on 66 burn patients, with total body surface area (TBSA) of burns ranging from 4% to 96%, were evaluated at different times after injury. REE values were determined in patients using indirect calorimetry at days 1, 2, 3, 7, 14, 21, and 28 after injury. We then constructed a mathematical model of REE changes post-burn. Next, established two new formulas (one non-linear and the other linear) for energy consumption estimation using model-based analytical solution and regression analysis. The new formulas were compared with measured REE and commonly used formulas including those of Carlson, Xie, Curreri, and Milner to determine accuracy and reliability.

RESULTS

Comparative analysis showed that the new formulas offered significantly higher accuracy and reliability than the Milner formula, which is considered the most accurate of commonly used burn energy consumption estimate formulas. The accuracy of the new nonlinear formula (94.29%) and that of the linear formula (91.43%) were significantly higher than that of Milner formula (72.86%) when compared to measured REE (χ2  =  11.706, P  =  0.001; χ2  =  8.230, P  =  0.004, respectively). The reliabilities of the new estimation formulas were both 100% and that of Milner formula was 74.24% (χ2  =  19.513, P  =  0.000).

CONCLUSION

The new formulas constructed in this study provide reliable simulation of the impact of the degree of burn and post-burn days on energy consumption and offer notably higher accuracy and reliability than other formulas. These formulas will help determine nutritional needs of burn patients.

TRIAL REGISTRATION

The study was registered on Chinese Clinical Trial Registry as ChiCTR-TRC-13003806.

Resting energy expenditure in subjects with fibro-calculous pancreatic diabetes

BACKGROUND

Fibro-calculous pancreatic diabetes is an indigenous disorder present in populations largely in tropical regions. Energy expenditure through indirect calorimetry has not been studied in this disorder and may provide important clues as to the pathogenesis of diabetes in these patients.

METHODS

A total of 51 males in three groups comprising fibrocalculous pancreatic diabetes (FCPD) (group 1; n = 24), type 2 diabetes (group 2; n = 15) and healthy controls (group 3; n = 12) were studied. The body composition was measured using Dual Energy X-ray Absorptiometry (DEXA) and the REE was estimated using indirect calorimetry. The predicted energy expenditure (PEE) was calculated using three different equations.

RESULTS

Patients in both groups with diabetes had a higher mean waist-hip ratio than the controls (P = 0.002). However patients with type 2 diabetes alone had a significantly higher mean body mass index (P = 0.012), percentage of fat (P = 0.016) and total fat content (P = 0.031). There was no significant difference in REE among the three groups. After adjustment of body mass index (BMI), the REE was significantly higher in patients with FCPD than in those patients with Type 2 diabetes. PEE correlated poorly with indirect calorimetry.

CONCLUSIONS

Energy expenditure in patients with diabetes varies according to the composition and distribution of body fat and is lower in patients with FCPD. Standard predictive equations were not accurate for the assessment of energy expenditure in patients with FCPD. Further research is required to recommend specific nutritional therapy for this group of patients.

A single-blinded randomised clinical trial of permissive underfeeding in patients requiring parenteral nutrition

BACKGROUND & AIMS

The importance of adequate nutritional support is well established, but characterising what 'adequate nutrition' represents remains contentious. In recent years there has been increasing interest in the concept of 'permissive underfeeding' where patients are intentionally prescribed less nutrition than their calculated requirements. The aim of this study was to evaluate the effect of permissive underfeeding on septic and nutrition related morbidity in patients requiring short term parenteral nutrition (PN).

METHODS

This was a single-blinded randomised clinical trial of 50 consecutive patients requiring parenteral nutritional support. Patients were randomized to receive either normocaloric or hypocaloric feeding (respectively 100% vs. 60% of estimated requirements). The primary end point was septic complications. Secondary end points included the metabolic, physiological and clinical outcomes to the two feeding protocols.

RESULTS

Permissive underfeeding was associated with fewer septic complications (3 vs. 12 patients; p = 0.003), and a lower incidence of the systemic inflammatory response syndrome (9 vs. 16 patients; p = 0.017). Permissively underfed patients had fewer feed related complications (2 vs. 9 patients; p = 0.016).

CONCLUSION

Permissive underfeeding in patients requiring short term PN appears to be safe and may results in reduced septic and feed-related complications.

TRIAL REGISTRATION

NCT01154179 TRIAL REGISTRY: http://clinicaltrials.gov/ct2/show/NCT01154179.

[Guidelines for specialized nutritional and metabolic support in the critically-ill patient. Update. Consensus of the Spanish Society of Intensive Care Medicine and Coronary Units-Spanish Society of Parenteral and Enteral Nutrition (SEMICYUC-SENPE): patient with polytrauma]

Patients with polytrauma can be viewed as paradigmatic of the critically-ill patient. These previously healthy patients undergo a life-threatening aggression leading to an organic response that is no different from that in other types of patients. The profile of trauma patients has changed and currently corresponds to patients who are somewhat older, with a higher body mass index and greater comorbidity. Severe injuries lead to intense metabolic stress, posing a risk of malnutrition. Therefore, early nutritional support, preferentially through the enteral route, with appropriate protein intake and glutamine supplementation, provides advantages over other routes and types of nutritional formula. To avoid overnutrition, reduced daily calorie intake can be considered in obese patients and in those with medullary lesions. However, little information on this topic is available in patients with medullary lesions.

Hypocaloric considerations in patients with potentially hypometabolic disease States

The provision of nutrition has traditionally been driven by the desire to provide adequate calories. However, over the past decade it has become evident that provision of excess calories can be detrimental to patients' outcomes in both critical care and long-term care settings. This review examines patient populations for whom hypocaloric nutrition can be both appropriate and beneficial. In specific situations, critically ill patients, such as those with obesity, stroke, and spinal cord injury, may have decreased energy requirements. In patients with spinal cord injury, the level of injury significantly correlates with the extent of reduced caloric energy expenditure. Ventilator-dependent patients with amyotrophic lateral sclerosis (ALS) have decreased energy needs; energy expenditure for ALS patients is lower than the predictive equation value. Aging patients will have decreased energy needs relative to a reduction in lean body mass. Patients with cerebral palsy (CP) have significantly lower caloric requirements than anticipated using predictive equations. Patients with CP pose a particular challenge in nutrition assessment. Several studies demonstrate that patients with CP have significantly lower caloric requirements than anticipated using predictive equations; thus, patients with CP benefit from indirect calorimetry. Provision of nutrition for obese patients is briefly addressed, as this is an increasingly important consideration for nutrition support in both the critical care and long-term care settings. When indirect calorimetry is not available, clinicians should remember that most patients will have low resting energy expenditure regardless of functional status and require frequent evaluation during institution of nutrition recommendations to guard against overfeeding and obesity.

Review article: permissive underfeeding in short-term nutritional support

BACKGROUND

The importance of adequate nutritional support in selected patient groups is well established. Traditionally, the amounts of macronutrients provided have been based on a perceived need to achieve, if not exceed, energy and protein balance. In recent years, there has been increasing interest in the concept of 'permissive underfeeding'.

AIM

To determine whether or not there is evidence of benefit for permissive underfeeding in selected groups.

METHODS

Studies were identified from MEDLINE, Embase and PubMed databases and the Cochrane collaboration. The search was limited from January 1950 to January 2010. Further searches were made from the references of original articles. The literature search revealed 591 abstracts of relevant studies. All abstracts were initially reviewed by the primary author (AO) and those that did not fulfil the inclusion criteria were discarded. The remaining articles were requested and were reviewed independently by two authors (AO, JM).

RESULTS

Twelve studies were included in the final analysis. Eight of these were randomized interventional trials. Three were prospective cohort studies and one was a retrospective analysis.

CONCLUSION

This review suggests that permissive underfeeding may be associated with improved outcomes and reduced morbidity in patients requiring short-term nutritional support.

Metabolic support of the obese intensive care unit patient: a current perspective

PURPOSE OF REVIEW

Obesity is a widespread condition associated with a variety of mechanical, metabolic, and physiologic changes that affect both health outcomes and delivery of care. Nutrition support is a key element of management during critical illness known to improve outcomes favorably, but is likewise complicated in the presence of obesity. This review serves to discuss the challenges unique to management of critically ill obese patients and an evidence-based approach to nutrition support in this patient population.

RECENT FINDINGS

High-protein, hypocaloric feeding has emerged as a nutrition support strategy capable of reducing hyperglycemia and protein catabolism, while promoting favorable changes in body composition and fluid mobilization. Recent data have shown a protective effect of mild-moderate obesity (BMI 30-39.9 kg/m2), with improved morbidity and mortality outcomes in this subgroup. Therefore, it is unclear whether hypocaloric feeding represents an inferior approach in this subgroup in which weight maintenance may be preferable.

SUMMARY

There are many obstacles that limit provision of nutrition support in the obese ICU patient. Calculating energy needs accurately is extremely problematic due to a lack of reliable prediction equations and a wide variability in body composition among the obese patients. Further research is needed to determine a better approach to estimating energy needs in this population, in addition to validating hypocaloric feeding as the standard approach to nutrition support in the obese patients.