584001

research-article2015

PENXXX10.1177/0148607115584001Journal of Parenteral and Enteral NutritionMogensen et al

Brief Communication

Validation of the Society of Critical Care Medicine and American Society for Parenteral and Enteral Nutrition Recommendations for Caloric Provision to Critically Ill Obese Patients: A Pilot Study

Journal of Parenteral and Enteral Nutrition Volume XX Number X Month 201X 1­–9 © 2015 American Society for Parenteral and Enteral Nutrition DOI: 10.1177/0148607115584001 jpen.sagepub.com hosted at online.sagepub.com

Kris M. Mogensen, MS, RD, LDN, CNSC1; Benjamin Y. Andrew, RD1,4; Jasmine C. Corona, BS2; and Malcolm K. Robinson, MD2,3

Abstract Background:The Society of Critical Care Medicine and A.S.P.E.N. recommend that obese, critically ill patients receive 11–14 kcal/ kg/d using actual body weight (ABW) or 22–25 kcal/kg/d using ideal body weight (IBW), because feeding these patients 50%-70% maintenance needs while administering high protein may improve outcomes. It is unknown whether these equations achieve this target when validated against indirect calorimetry, perform equally across all degrees of obesity, or compare well with other equations. Methods: Measured resting energy expenditure (MREE) was determined in obese (body mass index [BMI] ≥30 kg/m2), critically ill patients. Resting energy expenditure was predicted (PREE) using several equations: 12.5 kcal/kg ABW (A.S.P.E.N.-Actual BW), 23.5 kcal/kg IBW (A.S.P.E.N.-Ideal BW), Harris-Benedict (adjusted-weight and 1.5 stress-factor), and Ireton-Jones for obesity. Correlation of PREE to 65% MREE, predictive accuracy, precision, bias, and large error incidence were calculated. Results: All equations were significantly correlated with 65% MREE but had poor predictive accuracy, had excessive large error incidence, were imprecise, and were biased in the entire cohort (N = 31). In the obesity cohort (n = 20, BMI 30–50 kg/m2), A.S.P.E.N.-Actual BW had acceptable predictive accuracy and large error incidence, was unbiased, and was nearly precise. In super obesity (n = 11, BMI >50 kg/m2), A.S.P.E.N.-Ideal BW had acceptable predictive accuracy and large error incidence and was precise and unbiased. Conclusions: SCCM/A.S.P.E.N-recommended body weight equations are reasonable predictors of 65% MREE depending on the equation and degree of obesity. Assuming that feeding 65% MREE is appropriate, this study suggests that patients with a BMI 30–50 kg/m2 should receive 11–14 kcal/kg/d using ABW and those with a BMI >50 kg/m2 should receive 22–25 kcal/kg/d using IBW. (JPEN J Parenter Enteral Nutr. XXXX;xx:xx-xx)

Keywords critical care; energy expenditure; indirect calorimetry; obesity; underfeeding; predictive equations

Clinical Relevancy Statement This study validates the Society of Critical Care Medicine/ A.S.P.E.N predictive equations for estimating caloric needs based on actual body weight or ideal body weight for obese, critically ill patients by comparing them with metabolic cart studies. These findings are clinically relevant for clinicians who need to estimate energy needs in this population when a metabolic cart is not available or in patients for whom accurate metabolic cart studies are not feasible.

dosing requirements, and nutrition support.2 Of note, a recent study indicates that obese individuals, like their nonobese counterparts, have worse outcomes in the ICU when they are From the 1Department of Nutrition, Brigham and Women’s Hospital, Boston, Massachusetts; 2Department of Surgery, Brigham and Women’s Hospital, Boston, Massachusetts; 3Department of Surgery, Harvard Medical School, Boston, Massachusetts; and 4Duke University School of Medicine, Durham, North Carolina. Financial disclosure: None declared. Presented in part as an Abstract of Distinction at Clinical Nutrition Week 2011, Vancouver, BC, Canada, January 2011.

Introduction The rise in obesity in the general American population has led to a concomitant rise in the number of obese patients seen in the intensive care unit (ICU), with a prevalence most recently estimated at 31.1%.1 The obese patient poses a number of unique challenges in the provision of care in the ICU, including airway management, circulatory support, pharmaceutical

Received for publication June 18, 2014; accepted for publication March 24, 2015. Corresponding Author: Malcolm K. Robinson, MD, Department of Surgery, Brigham and Women’s Hospital, 75 Francis Street, Boston, MA 02115, USA. Email: [email protected]

Downloaded from pen.sagepub.com at BROWN UNIVERSITY on April 23, 2015

2

Journal of Parenteral and Enteral Nutrition XX(X)

malnourished.3 This suggests that nutrition assessment and, when indicated, intervention are of paramount importance for obese, critically ill patients. A cornerstone of nutrition intervention is estimating energy needs to reduce the risk of overfeeding or inappropriate underfeeding, both of which have known risks that may affect outcome.4-6 Indirect calorimetry or metabolic carts are considered the gold standard for determining energy needs for obese individuals in the ICU, as these methods overcome many of the problems associated with estimating needs using predictive formulas. Unfortunately, metabolic carts are not uniformly available for use in many ICUs. In addition, some patients have conditions that preclude accurate measurement of energy expenditure. Under such circumstances, one or more predictive formulas must be used to estimate energy needs for the obese, critically ill individual. In 2009 the Society of Critical Care Medicine (SCCM) and the American Society for Parenteral and Enteral Nutrition (A.S.P.E.N.) jointly published guidelines for the nutrition support of critically ill patients, which included specific guidelines for obese patients.7 Guidelines specifically for hospitalized critically ill and non–critically ill obese patients were published by A.S.P.E.N. in 2013.8 It is recommended that obese, critically ill patients without severe renal or hepatic dysfunction receive a trial of hypocaloric feeding with high protein administration.8 The expert panels making these recommendations specifically suggest that somewhere between 50% and 70% of energy requirements be provided for hospitalized obese individuals. The authors further state that in the absence of a metabolic cart, energy requirement may be estimated by the use of basic body weight predictive equations: 22–25 kcal/kg/d using ideal body weight (IBW) or 11–14 kcal/kg/d using actual body weight (ABW).7,8 Although A.S.P.E.N. has published forms of these guidelines twice,7,8 it is unknown whether the recommended predictive equations reliably estimate energy needs at 50%–70% of maintenance needs as measured by indirect calorimetry. The purposes of this study were (1) to validate current body weight predictive equations recommended by A.S.P.E.N. in obese, critically ill patients compared with indirect calorimetry; (2) to compare the A.S.P.E.N. equations to other predictive equations in this population; and (3) to identify any differences in the performance of these equations across multiple levels of obesity.

body composition (eg, obesity, amputation), or in other clinical situations where there was concern for inadequate or excessive energy delivery. Measured resting energy expenditure (MREE) was determined with indirect calorimetry using a metabolic cart (VMax 29N; SensorMedics, San Diego, CA) following standard manufacturer instructions and our institutional protocol. The metabolic cart was calibrated prior to each indirect calorimetry study. Patients were studied after 30 minutes of rest in a thermoneutral environment; changes in ventilator settings or nebulizer treatments were completed no later than 2 hours prior to the study. Exclusion criteria for indirect calorimetry included oxygen requirements >60%, presence of chest tubes with air leaks, continuous renal replacement therapy, or after a hemodialysis session. Spontaneously breathing patients were required to be stable on room air for at least 2 hours prior to the study. Studies were conducted for 30 minutes and were deemed valid if steady state was achieved. Steady state was defined as a 5-minute period of variation of 30 kg/m2) and 2 subgroups: subjects with obesity (BMI 30–50 kg/m2) and subjects with super obesity (BMI >50 kg/m2). The categorization of patients with BMI >50 kg/m2 as super obese has been used by several other investigators.15,16 While the use of the National Institutes of Health (NIH) categorization of obesity allows stratification of patients with varying degrees of obesity, the class III obesity threshold (BMI >40 kg/m2) does not accurately capture the differences in pathophysiology, body composition, or surgical outcomes in these patients. Furthermore, there is a documented difference in metabolically active fat-free mass between patients with super obesity (BMI >50 kg/m2) and the general obese population (BMI 65% MREE

Figure 1.  Regression analysis of the 4 predictive equations vs 65% MREE. Solid black line represents the regression model. Solid gray line represents the 95% confidence interval of observed values. Dashed gray line represents the 95% confidence interval of mean values. (A) A.S.P.E.N.-Actual BW equation (12.5 kcal/kg actual body weight). (B) A.S.P.E.N.-Ideal BW equation (23.5 kcal/kg ideal body weight). (C) Harris-Benedict (HB) equation. (D) Ireton-Jones equation for obesity (IJ-Ob). MREE, measured resting energy expenditure; PREE, predicted resting energy expenditure. Table 3.  Correlation of PREE to 65% MREE.a Equation A.S.P.E.N.-Actual A.S.P.E.N.-Ideal Harris-Benedict Ireton-Jones

Total Cohort b

0.46 0.70b 0.77b 0.67b

Obesity b

0.49 0.72b 0.74b 0.69b

Table 4.  Precision Analysis.a Super Obesity b

0.46 0.75b 0.75b 0.63b

Equation A.S.P.E.N.-Actual A.S.P.E.N.-Ideal Harris-Benedict Ireton-Jones

Total Cohort

Obesity

Super Obesity

15.2–34.3 10.1–16.4 10.5–16.4 14.7–26.3

7.9–16.6 11.0–18.3 8.3–17.8 9.7–19.4

27.5–67.3 6.3–15.0b 9.9–18.5 19.7–42.7

MREE, measured resting energy expenditure; PREE, predicted resting energy expenditure. a Values are represented as r for correlation between PREE and 65% MREE. b P < .05 for correlation.

a Values are presented as 95% confidence intervals of the root mean square error of predicted resting energy expenditure as a percentage of 65% measured resting energy expenditure. The upper bound of the confidence interval must be ≤15% to be considered precise. b Precise.

enteral nutrition versus eucaloric feeding and found that hypocalorically fed patients had a significantly reduced ICU length of stay,21 duration of antibiotic therapy,21 hospital mortality,22 and 180-day mortality.22 The hypocaloric group also showed a

trend toward reduced number of days receiving mechanical ventilation.21,22 Of note, there were no differences in nitrogen balance or changes in serum prealbumin and albumin.21 It should be noted that hypocaloric feeding in the obese patient is

Downloaded from pen.sagepub.com at BROWN UNIVERSITY on April 23, 2015

6

Journal of Parenteral and Enteral Nutrition XX(X)

100 90

Considered acceptable if

70% (dashed line)

Predictive Accuracy (%)

80 70 60 50 40 30 20 10 0

A.S.P.E.N.-Actual

A.S.P.E.N.-Ideal

Harris- Benedict

Ireton-Jones

All

45

58

61

55

Obese Obese Simple

70

50

55

70

Super Obese

0

73

73

27

Figure 2.  Predictive accuracy of tested equations. Bars represent the percentage of subjects with predicted resting energy expenditure within the range of 55%-75% measured resting energy expenditure. An equation is considered to have acceptable predictive accuracy if this value is ≥70% (dashed line).

100

Large Error Incidence (%)

90 80 70 60 50

Considered excessive if > 15% (dashed line)

40 30 20 10 0

Harris- Benedict

Ireton-Jones

All

A.S.P.E.N.-Actual 39

A.S.P.E.N.-Ideal 19

26

23

Obese Obese Simple

15

25

25

20

Super Obese

82

9

27

27

Figure 3.  Large error incidence of tested equations. Bars represent the percentage of subjects with predicted resting energy expenditure outside the range of 50%-80% measured resting energy expenditure. An equation is considered to have an excessive large error incidence if this value is >15% (dashed line).

also theorized to be clinically appropriate due to the high availability of endogenous fat for energy production. However, data on the metabolic response to critical illness in the obese population show reduced lipid mobilization in obese patients compared with their nonobese counterparts.23 Based on these and other data, the 2009 SCCM/A.S.P.E.N. guidelines recommend that critically ill, obese patients receive 60%–70% of estimated needs through the provision of 11–14

kcal/kg/d ABW or 22–25 kcal/kg/d IBW.7 The more recent 2013 A.S.P.E.N. guidelines suggest that hospitalized obese patients receive a trial of feeding at 50%–70% of eucaloric needs through the provision of 800 kcal compared with the average 65% MREE. This finding supports the fact that while correlation may be a reasonable approach to assess the performance of a predictive equation at the group level, it may be completely inadequate in informing decision making at the individual patient level. A caloric estimate that is off by >800 kcal from the desired goal may impose significant risk to patients in the critical care setting, leading either to rapid accumulation of caloric deficit or to uncontrolled overfeeding. We also determined predictive equation performance through precision and bias analyses and an assessment of predictive accuracy and large error incidence. Information gleaned from these analyses shows how well each predictive equation estimates the measured target goal (ie, 65% MREE) on an individual patient basis and therefore provides more practical information for individual patient care decisions. We found that neither of the A.S.P.E.N. formulas, based on ABW or IBW, performed well using such analyses when validated against the 65% MREE target for the entire cohort. However, we found that these formulas performed better when a priori subgroup analyses were done on those with obesity (BMI = 30–50 kg/ m2) and super obesity (BMI >50 kg/m2). Providing calories at 12.5 kcal/kg of ABW was a good estimate of 65% MREE for patients with obesity, while 22.5 kcal/kg of IBW was a good estimate of 65% MREE for super obese patients. We also arbitrarily selected 2 other well-known predictive equations and tested their precision, bias, large error incidence, and predictive accuracy relative to the 65% MREE goal. Neither the IJ-Ob nor HB equation performed well when tested in the entire cohort or in the obesity and super obesity subgroups. Of note, these 2 formulas were designed to accurately estimate eucaloric needs—that is, those calories needed to match 100% of the daily energy expenditure under specified conditions—and not hypocaloric needs. Hence, these formulas were adjusted by a factor of 0.65 in our analyses. Such adjustment likely weakened their utility and validity, which is not surprising because neither formula was developed, nor ever intended to be used, for estimating hypocaloric needs. The 2013 A.S.P.E.N. guidelines state that the Penn State University (PSU) formulas should be used to determine caloric

Downloaded from pen.sagepub.com at BROWN UNIVERSITY on April 23, 2015

8

Journal of Parenteral and Enteral Nutrition XX(X)

needs of the hospitalized obese patient if a metabolic cart is not available.8 This is in contrast to a statement in the same document that recommends hypocalorically feeding at 50%–70% of maintenance needs using 50 kg/m2. It should be noted that the number of Americans with super obesity (BMI >50 kg/m2) is increasing at a higher rate than any other class of obesity.28 This class of obese individuals may be better served by using a predictive equation specifically developed for their group, rather than a one-size-fits-all approach.

Conclusions The results from the present study suggest that current A.S.P.E.N. body weight predictive equations designed to hypocalorically feed critically ill, obese patients perform reliably, but only when used based on a patient’s degree of obesity. Our preliminary analysis suggests that if indirect calorimetry is unavailable and a clinician requires a simple method for determining nutrition support hypocaloric targets, 11–14 kcal/kg/d ABW can be used for critically ill patients with BMI 30–50 kg/ m2, and 22–25 kcal/kg/d IBW can be used for critically ill patients with BMI >50 kg/m2. Adjusting predictive formulas originally designed to estimate maintenance needs to instead estimate a hypocaloric goal may be unreliable. Such adjustment should not be done unless the adjusted formula is validated against the hypocaloric target as measured by indirect calorimetry. While preliminary results show promising outcomes with this particular approach to nutrition support, further research is needed to determine whether hypocaloric feeding is appropriate at all and, if so, whether this should be the standard of care for all critically ill, obese patients across the wide range of medical conditions and throughout the entire spectrum of obesity. Additional studies with a larger patient sample and with comparison to the PSU equations will help to further evaluate the relative accuracy and usefulness of these predictive methods in the critically ill obese population.

Statement of Authorship Kris M. Mogensen, Jasmine C. Corona, and Malcolm K. Robinson equally contributed to the conception/design of the research; Benjamin Y. Andrew contributed to the design of the research; Kris M. Mogensen contributed to the acquisition and analysis of the data; Benjamin Y. Andrew and Jasmine C. Corona contributed

Downloaded from pen.sagepub.com at BROWN UNIVERSITY on April 23, 2015

Mogensen et al

9

to the analysis of the data; Malcolm K. Robinson contributed to the acquisition, analysis, and interpretation of the data. All authors drafted and critically revised the manuscript, agree to be fully accountable for ensuring the integrity and accuracy of the work, and read and approved the final manuscript.

References 1. Finkielman J, Gajic O, Bekele A. Underweight is independently associated with mortality in post-operative and non-operative patients admitted to the intensive care unit: a retrospective study. BMC Emerg Med. 2004;4(1):3. 2. Kiraly L, Hurt RT, Van Way CW III. The outcomes of obese patients in critical care. JPEN J Parenter Enteral Nutr. 2011;35(5 suppl):29S-35S. 3. Robinson MK, Mogensen KM, Casey JD, et al. The relationship between obesity, nutritional status, and mortality in the critically ill. Crit Care Med. 2015;43:87-100. 4. Alberda C, Gramlich L, Jones N, et al. The relationship between nutritional intake and clinical outcomes in critically ill patients: results of an international multicenter observational study. Intensive Care Med. 2009;35(10):1728-1737. 5. Villet S, Chiolero RL, Bollmann MD, et al. Negative impact of hypocaloric feeding and energy balance on clinical outcome in ICU patients. Clin Nutr. 2005;24(4):502-509. 6. Dissanaike S, Shelton M, Warner K, O’Keefe GE. The risk for bloodstream infections is associated with increased parenteral caloric intake in patients receiving parenteral nutrition. Crit Care. 2007;11(5):R114. 7. McClave SA, Martindale RG, Vanek VW, et al. Guidelines for the provision and assessment of nutrition support therapy in the adult critically ill patient: Society of Critical Care Medicine (SCCM) and American Society for Parenteral and Enteral Nutrition (A.S.P.E.N.). JPEN J Parenter Enteral Nutr. 2009;33(3):277-316. 8. Choban P, Dickerson R, Malone A, et al. A.S.P.E.N. clinical guidelines: nutrition support of hospitalized adult patients with obesity. JPEN J Parenter Enteral Nutr. 2013;37(6):714-744. 9. Harris JA, Benedict FG. A biometric study of human basal metabolism. Proc Natl Acad Sci U S A. 1918;4(12):370-373. 10. Ireton-Jones CS, Turner WW Jr, Liepa GU, Baxter CR. Equations for the estimation of energy expenditures in patients with burns with special reference to ventilatory status. J Burn Care Rehabil. 1992;13(3):330-333. 11. Hamwi GJ. Therapy: changing dietary concepts. In: Danowski T, ed. Diabetes Mellitus: Diagnosis and Treatment. New York, NY: American Diabetes Association; 1964:73-78. 12. Anderegg BA, Worrall C, Barbour E, Simpson KN, Delegge M. Comparison of resting energy expenditure prediction methods with measured resting energy expenditure in obese, hospitalized adults. JPEN J Parenter Enteral Nutr. 2009;33(2):168-175. 13. The Academy of Nutrition and Dietetics. Academy of Nutrition and Dietetics Evidence Analysis Library. In critically ill patients, what is

the relationship between resting metabolic rate (RMR) and RMR predicted by the Ireton-Jones 1992 equations? http://www.andeal.org/topic. cfm?cat=4311&conclusion_statement_id=251235&highlight=equations &home=1. Accessed June16, 2014. 14. Frankenfield DC, Coleman A, Alam S, Cooney RN. Analysis of estimation methods for resting metabolic rate in critically ill adults. JPEN J Parenter Enteral Nutr. 2009;33(1):27-36. 15. Salihoglu T, Salihoglu Z, Zengin AK, Taskin M, Colakoglu N, Babazade R. The impacts of super obesity versus morbid obesity on respiratory mechanics and simple hemodynamic parameters during bariatric surgery. Obes Surg. 2013;23(3):379-383. 16. Strain GW, Gagner M, Pomp A, Dakin G, Inabnet WB, Saif T. Comparison of fat-free mass in super obesity (BMI >/ = 50 kg/m2) and morbid obesity (BMI

Validation of the Society of Critical Care Medicine and American Society for Parenteral and Enteral Nutrition Recommendations for Caloric Provision to Critically Ill Obese Patients: A Pilot Study.

The Society of Critical Care Medicine (SCCM) and American Society for Parenteral and Enteral Nutrition (ASPEN) recommend that obese, critically ill pa...
526KB Sizes 0 Downloads 11 Views