CLINICAL NUTRITION WEEK 2015 ABSTRACTS February 14 17, … · 2016. 8. 24. · Results: PN caused...

Clinical Nutrition Week 2015 1

CLINICAL NUTRITION WEEK 2015 ABSTRACTS

February 14–17, 2015, Long Beach, California

Long Beach Convention Center

NUTRITION AND METABOLISM PAPER SESSIONS

Session Title Page

Premier Paper Session and Live Vars Award Competition 1

Parenteral Nutrition 8

Enteral Nutrition 14

Malnutrition, Obesity, Practice Concepts 22

Critical Care & Other Critical Health Issues 29

GI & Other Metabolic Issues 38

Pediatric and Neonatal 48

Premier Paper Session and Live Vars Award Competition

The Harry M. Vars award is given annually to the person presenting the highest-scoring qualified abstract for Clinical Nutrition Week (CNW).

The candidates are also evaluated on a manuscript based on their abstract, as well as on their expertise and knowledge of the science as

demonstrated during their oral presentation at the Premier Paper Session. CNW15 is February 14–17 in Long Beach, California.

Vars Candidate 2079212 - Oleanolic Acid Improves Gut Atrophy Induced by Parenteral Nutrition Ajay K. Jain, Assistant Professor

1; Joy X. Wen, MD

1; Sumit Arora, Student

1; Keith S. Blomenkamp,

Research Assistant1; Timothy A. Blaufuss, MD

1; John P. Long, DVM

2; Jeffrey H. Teckman, Professor

1

1Pediatrics, Saint Louis University, Saint Louis, MO;

2Comparative Medicine, Saint Louis University, Saint

Louis, MO

Purpose: Parenteral Nutrition (PN) involves delivering all nutritional requirements via the intravenous route. Though PN therapy has grown enormously over the last few decades; it is associated with significant clinical complications including gut atrophy and liver injury. Amelioration of side effects remains a major research focus. We have previously published improvement in PN–associated pathologies with enteral chenodeoxycholic acid a dual Farnesoid X Receptor (FXR) and TGR5 agonist. We have hypothesized that gut growth is regulated via TGR5 activation. We describe role of the pure TGR5 specific agonist, Oleanolic Acid (OA) in animals receiving PN (Figures 1, 2, and 3). Methods: Neonatal pigs were implanted jugular vein (JV) and duodenal catheters (DC). Isocaloric swine milk (n=34) was provided via the DC. Continuous PN solution was infused via the JV. A subset of PN animals (n=34) received the enteral TGR5 agonist, oleanolic acid at 50mg/kg/day via the DC (Figures 1, 2, and 3). Results: PN caused marked gut atrophy in comparison to EN fed control animals. Oleanolic acid treatment led to a robust preservation of gut mass. This was evident grossly and on histology. The mean gut weight as a percent of body weight and (±SD) was 4.30±0.26 for EN, 1.92±0.06 for PN (p


1.69±0.1 for PN (p


Figure 3. Macroscopic and microscopic images of small bowel in Enteral (EN), Parenteral Nutrition (PN), and Oleanolic Acid (OA) treated animals. Note thin appearance and villous atrophy in PN.

Vars Candidate 2078820 - Nutritional Modulation of Cardiotoxicity and Anti-Cancer Efficacy Related to Doxorubicin Chemotherapy by Glutamine and n-3 Polyunsaturated Fatty Acids Hongyu Xue, MD, PhD; Wenhua Ren, PhD; Paul Wischmeyer, MD Anesthesiology, University of Colorado, School of Medicine, Aurora, CO

Purpose: Doxorubicin (DOX) has been one of the most effective antitumor agents against a broad spectrum of malignancies. However, DOX-induced cardiotoxicity (DIC) forms the major factor limiting the dose escalation of the drug and is of great clinical concern with the high morbidity and mortality. Glutamine and n-3 polyunsaturated fatty acids (PUFA) are putatively cardioprotective during various stresses and/or have potential chemo-sensitizing effects during cancer chemotherapy settings. The aim of this study was to investigate whether single and/or combined glutamine and n-3 PUFAs could provide therapeutic benefits in preventing DIC and/or enhancing its anti-tumor efficacy. Methods: To closely recapitulate the clinical scenario, the anti-tumor activity and cardiotoxicity of anthracycline treatment was evaluated simultaneously in the same rat model bearing a transplantable mammary adenocarcinoma treated with doxorubicin (cumulative dose 12mg/kg). Effects of glutamine and n-3 PUFAs, when parenterally provided individually and in combination, were compared in this animal drug-tumor model system. Single or combined treatment of parenteral glutamine (0.35g/kg via Dipeptiven, Fresenius Kabi) and n-3 PUFAs (0.19 g/kg eicosapentaenoic acid and 0.18g/kg docosahexaenoic acid, via Omegaven, Fresenius Kabi) was given to the rats every other day, starting 6 days before initiation of chemotherapy till the end of study (Day 50). Multileveled DIC-associated endpoints, i.e., organ congestion (wet-to-dry weight ratio of liver and lung), cardiac function determined by echocardiography, cardiac histopathology, Plasma cardiac troponin I (cTnI) level, cardiac heat shock protein 70 (Hsp70) expression, and cardiac redox homeostasis as measured by cardiac peroxidation, antioxidant enzymatic defense (i.e., superoxide dismutase, catalase, glutathione peroxidase) and glutathione (GSH) stores in the heart were determined. Tumor-related endpoints include tumor inhibition kinetics related to dox chemotherapy, tumoral GSH stores in tumor tissue and tumoral DOX accumulation. Results: Single glutamine remarkably attenuated lung and liver congestion, prevented cardiac function deterioration, reduced serum cardiac troponin I levels, diminished cardiac lipid peroxidation and preserved anti-oxidant enzymatic reserve in the cardiac tissue, while not affecting tumor inhibition kinetics. Single n-3 PUFA treatment significantly enhanced anti-tumor activity of DOX associated with intensified tumoral oxidative stress and enhanced tumoral DOX concentration while not accentuating the


decline of cardiac function or aggravating the cardiac oxidative stress. Intriguingly, providing glutamine and n-3 PUFAs together did not confer a greater benefit; conversely, individual benefits on cardiotoxicity associated with single treatments were instead partially lost when the nutrients were combined. In terms of tumor’s response to chemotherapy, a clear antagonism was observed when they were combined. Conclusions: Our data demonstrate an interesting differentiality or even dichotomy in the response of tumor and host, to single parenteral glutamine and n-3 PUFA treatments. These two individual nutrients show therapeutic promise to be used as adjunct to dox chemotherapy as cardiotoxicity antagonist or chemosensitizer, which can ultimately modulate the balance between tumor and host in a manner that favors the host. Intriguingly, an absence of additive benefits or even antagonism in their effects was observed when glutamine and n-3 PUFAs were combined. This draws into question the common assumption that there are additive benefits of combinations of nutrients, which are beneficial on an individual basis.

Financial support received from Fresenius Kabi and the A.S.P.E.N. Rhoads Research Foundation.

Vars Candidate, A.S.P.E.N. Trainee Award 2077285 - Exogenous Glucagon-Like Peptide-2 Therapy Improves Parenteral Nutrition–Associated Liver Disease by Altering Bile Acid Dysmetabolism David W. Lim, MD, CM

1; Paul W. Wales, BSc, MD, MSc, FRCSC, FACS

2,1; Si Mi, MSc

3; Diana Mager,

PhD, RD3,4

; Jonathan Curtis, PhD3; Vera Mazurak, PhD

3; Jason Yap, MBChB, FRACP

4; Pamela R.

Wizzard, BSc, RAHT4; David L. Sigalet, MD, PhD, FRCSC, FACS

5; Justine Turner, MBBS, PhD,

FRACP4,3

1Surgery, University of Alberta, Edmonton, Canada;

2Surgery, University of Toronto, Toronto, Canada;

3Agricultural, Food & Nutritional Science, University of Alberta, Edmonton, Canada;

4Pediatrics, University

of Alberta, Edmonton, Canada; 5Surgery, University of Calgary, Calgary, Canada

Purpose: Parenteral nutrition–associated liver disease (PNALD) occurs in 40-60% of infants with intestinal failure and is the leading cause of morbidity and mortality. The peptide hormone, glucagon-like peptide-2 (GLP-2), is being advanced as therapy for intestinal failure because of its intestinotrophic properties. The effect of GLP-2 on PNALD was previously unknown. Our group recently determined the novel finding that exogenous GLP-2 therapy in a translational piglet model of PNALD improves bile flow and serum markers of cholestasis. We now aim to study the mechanisms underlying the improvement of PNALD with systemically administered GLP-2 in our piglet model. Given recent insights that the pathogenesis of PNALD implicates an alteration in bile acid homeostasis, we hypothesize that bile acid composition and bile acid synthesis pathways are altered in PNALD and ameliorated with GLP-2 therapy. Methods: Neonatal piglets (3 to 5 days old) underwent jugular venous catheterization, followed by 17 days of iso-caloric, iso-nitrogenous parenteral nutrition (PN) therapy. Piglets were randomized to either saline treatment (Group 1, n=8) or GLP-2 treatment at 11 nmol/kg/day (Group 2, n=7). A sow-reared control group (Group 3, n=8) was used for statistical comparison as a group representing normal piglet physiology. After 17 days, a terminal laparotomy was performed, the common bile duct cannulated and bile was collected. Liver specimens were flash frozen in liquid nitrogen. Reverse transcription of total mRNA in liver was performed, followed by quantitative PCR to determine the relative gene expression of enzymes involved in bile acid synthesis (CYP7A1, CYP27A1, CYP8B1), regulation (Farnesoid X receptor, FXR) and transport (multidrug resistance protein 2, MRP2). The composition of bile acids in bile was determined using tandem mass spectrometry. Data are analyzed via Kruskal-Wallis analysis of variance, with post-hoc analysis performed by Mann-Whitney U tests. Results: There was a significant difference in hepatic FXR expression (p


(p 4 and > 12 day samples are shown in Table 1. The mean age was 61 years, 61% were male, and 64% were treated in medical ICUs. APACHE II score was 22, SOFA score 9, and NUTRIC score 5. In the > 4 day sample achieving at least 80% of prescribed protein intake was associated an OR of 0.61 (95% CI 0.47, 0.81) compared to those achieving less than 80%. (Table 2) In the > 12 day sample, achieving at least 80% of prescribed protein intake was also associated with reduced mortality, OR 0.550 (95% CI 0.37, 0.82) and a shorter time to discharge alive, HR 1.25 (95% CI 1.04, 1.49) relative to those achieving less than 80%. (Table 2) In all these cases, the effect of protein administration on outcome persisted when adjusting for energy intake. When adjusting for protein intake, we could not demonstrate any significant relationship with energy intake and outcome. Conclusions: Achieving at least 80% of protein goal may be important to survival and may shorten time in the hospital in critically ill patients. Efforts to achieve goal protein intake should be maximized as a key goal of feeding in such patients and may be more important than optimizing energy intake.

Financial Support received from the C. Richard Fleming Small Research Project Grant and the A.S.P.E.N. Rhoads Research Foundation.


Table 1. Sample Demographics.

ICU ≥ 4 days

(N=2828)

ICU ≥ 12 days

(N=1584)

Variable Mean (SD) or N (%) Mean (SD) or N (%)

Age 61.4 (17.3) 59.8 (17.4)

BMI (kg/m2) 26.9 (7.4) 27.4 (7.5)

Gender

Male (%)

Female (%)

1731 (61.2%)

1097 (38.8%)

991 (62.6%)

593 (37.4%)

Admission Type (%)

Medical

Surgical Elective

Surgical Emergent

1833 (64.8%)

238 (8.4%)

757 (26.8%)

1048 (66.2%)

119 (7.5%)

417 (26.3%)

APACHE II Score 22.5 (8.5) 22.2 (7.9)

SOFA Score 8.95 (3.7) 9.1 (3.7)

NUTRIC Score 4.84 (2.03) 4.8 (2)

Table 2-2. Mortality and Time to Discharge Alive Outcomes Relative to Protein and Caloric Intake Delivered.

Patients in ICU ≥ 4 d Patients in ICU ≥ 12 d

Variable 60-Day Mortality, Odds Ratio (95% CI) 60-Day Mortality, Odds Ratio (95% CI)

Unadjusted Adjusteda Adjusted

b Unadjusted Adjusted

a Adjusted

b

Protein intake (Delivery ≥ 80% of prescribed vs. < 80%)

0.650

(0.506, 0.84)

0.613

(0.465, 0.808)

0.662

(0.498, 0.88)

0.620

(0.436, 0.88)

0.551

(0.371, 0.82)

0.587

(0.393, 0.88)

Energy intake (Delivery ≥ 80% vs. < 80% of Prescribed)

0.738

(0.598, 0.91)

0.705

(0.560, 0.889)

0.882

(0.703, 1.11)

0.736

(0.541, 1.000)

0.679

(0.476, 0.97)

0.896

(0.629, 1.28)

Variable Time to Discharge Alive, Hazard Ratio

(95% CI) Time to Discharge Alive, Hazard Ratio

(95% CI)

Unadjusted Adjusteda Adjusted

b Unadjusted Adjusted

a Adjusted

b

Protein intake (Delivery ≥ 80% of prescribed vs. < 80%)

0.898

(0.788, 1.02)

0.913

(0.786, 1.06)

1.022

(0.874, 1.195)

1.075

(0.900, 1.29)

1.248

(1.044, 1.49)

1.260

(1.030, 1.54)

Energy intake (Delivery ≥ 80% vs. < 80% of Prescribed)

0.817

(0.714, 0.94)

0.826

(0.713, 0.96)

0.817

(0.697, 0.96)

0.973

(0.818, 1.16)

1.006

(0.977, 1.04)

0.983

(0.803, 1.20)

a Adjusted for BMI, Sister Site, Gender, Admission Type, Age, Evaluable Days in the ICU, APACHE II Score, SOFA

Score. b Adjusted for BMI, Sister Site, Gender, Admission Type, Age, Evaluable Days in the ICU, and APACHE II Score,

SOFA Score, Calories and Protein.


Vars Candidate 2078946 - Adequate Nutrition May Get You Home: The Effect of Caloric/Protein Deficits on Critically Ill Surgical Patients Daniel D. Yeh, MD

1; Eva Fuentes, MD

1; Sadeq A. Quraishi, MD, MHA, MMSc

1; Catrina Cropano, MSc

1,2;

Haytham Kaafarani, MD, MPH1; Jarone Lee, MD, MPH

1; David R. King, MD

1; Marc de Moya, MD

1; Peter

Fagenholz, MD1; Yuchiao Chang, PhD

1; George Velmahos, MD, PhD

1

1Massachusetts General Hospital, Boston, MA;

2Quinnipiac University, Frank H. Netter MD School of

Medicine, North Haven, CT

Purpose: Nutrition is a fundamental component in the care of critically ill patients. Adequate nutrition is known to influence wound healing, protein catabolism, risk of infection, length of stay (LOS), and mortality in the intensive care unit (ICU). Despite its importance, nutritional delivery is generally poor, with between 50-66% of prescribed calories/protein actually being delivered in clinical practice. Previous reports have shown that increasing macronutrient deficit in the surgical ICU is associated with worse in-hospital outcomes. We hypothesized that increased caloric and protein deficit may also be associated with more intermediate outcomes, such as likelihood of being discharged to home versus transfer to a rehabilitation or skilled nursing facility. Methods: We prospectively enrolled adult surgical ICU patients who received >72 h of enteral nutrition (EN) from 03/2012-05/2014. Patients with absolute contraindications to EN, short surgical ICU stay (


Table 1. Demographics, Nutrition Delivery, and Outcomes. Caloric deficit Protein deficit


wished to run another marathon. Gastroparesis with significant nausea and vomiting precluded standard enteral hydration; therefore the use of intravenous (IV) hydration was needed. For short runs, pre-run and/or post-run hydration were adequate, but a marathon (26.2 miles), would be too long to run without IV hydration. Based on recommendations from the International Marathon Medical Director's Association guidelines for fluid replacement during endurance running, a sweat rate was used to determine a hydration strategy. During training, we instructed our consumer to record weights (pre/post run), ambient temperature, running distance, and duration of time. These data were used to calculate her sweat rate and estimate hydration volume during the marathon. An additional dilemma was determining how to administer the needed hydration while running. A traditional battery operated pump with hydration would be too cumbersome for running; therefore we elected to administer fluids via an elastomeric infusion device (EID) that would provide 500 mL of D5 0.45% NaCl over 2 hours. Results: Ambient temperature was a significant factor influencing sweat rate (R2=0.6878). To predict hydration needs during the marathon, sweat rate (x) plotted against temperature (y) which resulted in a slope of y=0.0482x + 30.223. The estimate temperature for the marathon was 65 degrees Fahrenheit; therefore our consumer would have an estimated sweat rate of approximately 720 mL/hr. This exceeded the amount of fluid that could be infused via the EID, therefore we advised our consumer to over hydrate with 1000 mL 0.45% NaCl in addition to her 2000 mL her standard HPN prior to the race. She would completed 14 miles in just over 2 hours, therefore, at this point the initial EID was complete and she received a bolus of 240 mL a second EID was attached. At 23.5 miles, a third EID was attached making the total IV fluids infused during the marathon 1740 mL of D5 0.45% NaCl (295 kcal). When combined with the pre-race 1L 0.45% NaCl, the average volume infused was approximately 520 mL/h. Initial post-race urine output (UO) was low and concentrated, but returned to baseline after post-race hydration. Our consumer did not experience any symptoms of dehydration and only had minor muscle soreness. She completed this marathon 46 minutes faster than her previous marathon prior to illness. Conclusions: Our consumer was able to complete a marathon with IV hydration administered via a combination pre-race hydration, continuous infusion via an elastomeric infusion device, and fluid boluses. Reliance on IV hydration and nutrition does increase the complexity of maintaining adequate hydration and nutrition during endurance sports, but it does not prevent participation. We have shown that with carefully preparation, calculation, and planning, our HPN consumer was able to adequately maintain her state of hydration and accomplish her goal of completing a marathon.

Financial support: none.

Abstract of Distinction 2079715 - Mental Health Support of Consumers and Caregivers on Home Parenteral and Enteral (HPEN) Therapy Abigail E. Brogan; Ann Weaver; Deborah Pfister, RD, CNSC, MS Nutrition, ThriveRx, Cincinnati, OH

Purpose: Consumers and caregivers of those on Home Parenteral and Enteral Nutrition (HPEN) frequently experience an emotional toll from both their diagnosis as well as medical therapy; including experiencing depression, anxiety, and post-traumatic stress, among other symptoms. Many also expressed the lack of support within the medical community. Prior research reported incidence of anxiety and depression among those with chronic medical conditions as well those on HPEN. Methods: To better understand the emotional and mental health challenges HPEN consumers/caregivers experience and the support they receive, an online survey was created. The link to the survey was distributed to HPEN consumers, and their caregivers, via social media. Respondents were asked to report their experience with symptoms such as, grief, anxiety, fatigue, appetite changes, sleep disruptions and inability to cope, as well as duration of symptoms. Respondents were additionally asked if they were provided mental health support and resources by their medical team, and if they felt that such support should be an integral part of coordinated care. Results: We received 184 completed Surveys (65% caregivers and 35% consumers). 95% of respondents felt that mental health support should be provided as part of comprehensive care for individuals on HPEN. 55% of all respondents indicated that they experienced 6 or more symptoms that could be indicative of a mental health disorder. 29% of respondents indicated that their symptoms began


at diagnosis and 50% indicated that their symptoms have been ongoing throughout their illness. 76% of caregivers and 72% of consumers did not have any type of mental health support offered by their practitioner or clinical team. 53% sought professional help for their symptoms and of those, 61% were self-referred while 39% were referred by their medical team. Only 33% of respondents reported having a consumer advocate provided by their home infusion company and of those respondents, 55% received resources from their consumer advocate to inform them of the possibility of mental/emotional issues. Conclusions: Our research shows that symptoms of metal/emotional disorders may be common among consumers on nutrition support and their caregivers, as suggested by prior research. This paired with the fact that 76% of caregivers and 72% of consumers and caregivers did not have any type of mental health support offered to them, indicates that there is a need that is not being met by medical teams supporting individuals on HPEN. Consumers and caregivers overwhelmingly feel that mental health support should be offered as part of their or their loved one’s comprehensive care. The data highlights the importance of integrating mental health support into comprehensive care for individuals and their caregivers on Home Parenteral and Enteral therapy. It also indicates the importance of offering mental health support and resources throughout treatment and therapy.


Abstract of Distinction 2079408 - Maintaining Patient Safety and the Patient–Physician Relationship: Management of the Noncompliant Home Parenteral Nutrition Patient Sandra I. Austhof, MS, RD, LD, CNSC

1; Karen M. Giaco, MS, RD, LD, CNSC

1; Ronelle Mitchell, MA, RD,

LD, CNSC1; Denise Konrad, RD, LD, CNSC

1; NaSheema Anderson

2; Donald F. Kirby, MD, FACP,FACN,

AGAF,CNSC,CPNS3

1Home Nutrition Support Service, Cleveland Clinic, Cleveland, OH;

2Ombudsman, Cleveland Clinic,

Cleveland, OH; 3Gastroenterology & Hepatology, Cleveland Clinic, Cleveland, OH

Purpose: The Home Nutrition Support Service (HNSS) at the Cleveland Clinic manages nearly 300 patients on home parenteral nutrition (HPN) each month. The goal is to provide high quality intravenous nutrition therapy with the patient/caregiver as a key team member. Patient noncompliance has become challenging over the past several years. As a result, a process was developed to help ensure patient safety, quality of care and provider liability protection. This process involves the physician, clinician, and ombudsman who help guide the patient to improve compliance or to pursue termination of the physician-patient relationship when necessary. Methods: Noncompliance may be suspected by unexplainable electrolyte abnormalities, dehydration, recurrent catheter-related blood stream infections, weight loss or lack of weight gain. Patients may routinely miss lab draws, clinic visits, or may not provide intake/output records. Additionally, these patients may not answer or return phone calls nor contact our service as required. In short, when adverse outcomes cannot be explained by changes to medical condition, medications, or gastrointestinal output, noncompliance is suspected. An ad hoc team of physicians, clinicians, an attorney, and ombudsman developed a patient agreement document and intervention process. The agreement educates on expectations for service and is documented in the medical record. While it creates a context of mutual understanding and collaboration between the providers and patient/caregiver, it is not a legally binding contract. The patient agreement is presented as a collaboration between patient/caregiver and providers. The HNSS team and ombudsman provide clarification and support to the patient as needed. Patients/caregivers must sign the agreement to participate in the HNSS but may elect for another suitable provider if preferred. For the existing patient, when non-compliance is suspected, the HNSS notifies the ombudsman who creates a reference file. The team notifies the ombudsman with continued noncompliance and actions are documented. The clinician works closely with the patient to develop a plan and target date for improvement compliance. If the patient cannot be reached by phone, the physician mails a letter to ensure communication, clarification and encouragement. If improvement is not demonstrated, a warning letter is sent by certified mail from the physician, outlining safety concerns, requesting specific behavior changes and defined timelines. Continued noncompliance results in a termination letter notifying that the patient will be terminated from service within 30 days. At the end of the 30 days, the patient is withdrawn from the service.


Results: In the last 15 months, 70 patients have been identified as noncompliant. Improved compliance was seen in the majority after a call from the ombudsman; however, 10 certified warning letters were mailed leading to improved compliance and one termination letter. Conclusions: Non-compliance has become a challenge in safely managing HPN patients resulting in increased hospital readmissions and healthcare costs. A multi-disciplinary process was developed including physicians, clinicians, and ombudsman to ensure patient safety, quality of care and provider liability protection. Alerting patients of the danger of their noncompliance has resulted in improved compliance.


2079395 - Predictors of Insulin Requirement for Patients on Parenteral Nutrition Kristen Lach, MS, RD, LDN

2,1; Kelly A. Roehl, MS, RD, LDN, CNSC

1; Diane Sowa, MBA, RD, LDN

1;

Sarah Peterson, MS, RD, LDN, CNSC1

2Nutrition and Operational Support Services, Aramark, Downers Grove, IL;

1Clinical Nutrition, Rush

University Medical Center, Chicago, IL

Purpose: Research has shown that hyperglycemia can lead to serious complications including morbidity and mortality. Patients receiving parenteral nutrition (PN) are at higher risk for developing hyperglycemia due to the administration of intravenous dextrose. Taking a proactive approach to dosing insulin may be crucial in limiting these complications associated with hyperglycemia. The objectives of this quality improvement project were to determine factors predictive of insulin requirement and insulin dosing among adult patients receiving PN. Methods: Patients 18 years of age and older, admitted to an adult patient care unit and started on PN between January 2009 to June 2013 were included. The following variables were collected: age, gender, body mass index (BMI), subjective global assessment (SGA, categorized as well nourished (A) versus malnourished (B/C)), diagnosis of DM (categorized as yes versus no), glycosylated hemoglobin (Hgb A1c) value within 3 months prior to PN start, ICU versus general medical/surgical floor admission, steroid administration, infusion of dextrose containing IV fluid with PN, reason for PN, total days PN, average PN dextrose (defined as grams of dextrose administered divided by days on PN), average dextrose infusion rate (DIR, defined as the amount of dextrose delivered per minute per kg body mass [mg/kg/min]) and average total insulin (defined as total insulin in the PN solution, plus supplemental insulin [subcutaneous and/or continuous infusion] divided by the days the patient received insulin). Descriptive statistics, count with percentage and mean with standard deviation (SD), were utilized to describe the sample. Stepwise forward linear regression was utilized determine the best predictors of average total insulin. Results: A total of 513 patients were started on PN during the time period, 57% of the sample were female with a mean age (SD) of 61.2 (14.1) years, and BMI of 27.4 (8.0) kg/m2. Approximately 49% of patients were malnourished. The average (SD) Hgb A1C was 6.2 (1.4, range 3.3-15.6) and 44% had a diagnosis of DM. Patients received PN for an average of 9.7 days (8.6) and ileus, obstruction and post-operative bowel rest were the most common indications for PN, representing 61% of all PN. The strongest predictors of average total insulin included Hgb A1c (P


2079048 - Ethanol Lock Therapy in Prevention of Central Line-Associated Bloodstream Infections in Pediatric Patients on Home Parenteral Nutrition Alexia Janes, PharmD Candidate; Maxime Thibault, PharmB, MSc, CNSC Pharmacy, CHU Sainte-Justine, Montreal, Canada

Purpose: Central venous catheter-associated bloodstream infections (CABSI) are a frequent and severe complication of home parenteral nutrition (HPN). Ethanol lock therapy (ELT) has been used in pediatric patients to decrease the frequency of infections and prevent associated complications. There is uncertainty regarding the frequency of adverse events related to this technique, such as line breakage or thrombosis. Polyurethane-based catheters have been purported to be susceptible to damage with ELT. The purpose of this study was to evaluate the efficacy and safety of ethanol locks in children receiving HPN. Methods: We performed a retrospective cohort study. Patients were included if they were followed at our center, received HPN and were on ELT after 2009. Efficacy was evaluated as the incidence of CABSI, total and CABSI-related central line changes, total and CABSI-related hospitalizations and total and CABSI-related emergency department (ED) visits, per 1000 catheter-days. Safety was evaluated as the incidence of line breakage, thrombosis, and line changes related to these events per 1000 catheter-days. Analysis was performed as intent-to-treat. Parameters were compared before and after initiation of ELT, each patient serving as his own control. Comparative analysis was performed using the Wilcoxon signed-rank test. P-values


Table 1 - Patient characteristics

Sex, n (%) Male 2 (29)

Female 5 (71)

Age (years) Median [Min-Max] 5.4 [1.4-12.1]

Duration of HPN (days) Median [Min-Max] 1089 [158 – 2124]

Indication for HPN, n (%)

Short bowel syndrome secondary to gastroschisis 3 (44)

Short bowel syndrome secondary to omphalocele 1 (14)

Tricho-hepato-enteric syndrome 1 (14)

Intestinal atresias 1 (14)

Liver failure 1 (14)

Table 2 - Outcomes

Outcome (per 1000 catheter-days) Pre-ELT Median

[interquartile range]

Post-ELT Median [interquartile

range]

P-value

Incidence of CABSI (n=7) 10.30 [4.27-12.90] 2.60 [0-4.12] 0.046

Total hospitalizations (n=7) 12.87 [8.55-27.35] 14.52 [4.55-19.39] 1.000

CABSI-related hospitalizations (n=7) 4.57 [4.12-7.08] 1.95 [0-8.06] 0.463

Total ED visits (n=6)* 4.71 [2.57-9.54] 8.24 [0-29.39] 0.500

CABSI-related ED visits (n=6)* 3.64 [2.42-5.77] 0.97 [0-6.09] 0.225

Line breakages (n=7) 5.13 [0-6.87] 9.10 [0-13.85] 0.345

Thrombosis (n=7) 3.22 [0-3.42] 0 [0-0.65] 0.043

Total central line changes (n=7) 12.36 [4.27-16.13] 5.97 [0-9.10] 0.075

CABSI-related central line changes (n=7) 3.22 [2.29-6.45] 0 [0-1.61] 0.028

Line breakage or thrombosis-related central line changes (n=7)

2.29 [0-4.12] 2.99 [0-7.15] 0.345

* One patient was excluded from analysis for these outcomes because ED visits occurred at another center and detailed information could not be obtained.

Abstract of Distinction 2079481 - Central Line–Associated Bloodstream Infection and Blood Drawing Practices in an Adult Home Parenteral Nutrition Population With Tunneled Catheters Marianne Opilla, RN, CNSC; Brenda Gray, PharmD, CNSC, BCNSP; Thomas Diamantidis, PharmD; Rodney Okamoto, RPh Nutrishare, Elk Grove, CA

Purpose: Long-term home parenteral nutrition (HPN) is often administered through a tunneled central venous catheter (CVC). One significant complication of HPN is central line associated bloodstream infection (CLABSI). Periodic blood draws for laboratory monitoring are required for safe management of HPN, and may be collected from the CVC or by venipuncture. The CVC is usually chosen for blood draws when there is poor venous access, needle phobia, pain with venipuncture, or for patient convenience. The aim of this study is to compare CLABSI rates between CVC and peripheral vein (PV) blood sampling, and to correlate infection rates associated with the person performing the CVC blood draw. Methods: Retrospective data was collected by medical record review from one home infusion pharmacy on adult HPN patients with tunneled CVCs. The collection period for tunneled CVC days and


corresponding CLABSI was January 2005 to December 2013. The minimum number of tunneled CVC days in the study was 730. CLABSI incidents were defined as presence of infection symptoms and positive blood cultures by peripheral and/or CVC sample. Positive blood cultures for the same organism within 30 days were counted as the same infection episode. There were 108 total patients studied, with 68 patients having blood drawn by PV and 40 patients with blood drawn from the CVC. The CVC group was further divided into four categories based on the person performing the blood collection: self, clinic/MD staff, home health nurse, or family caregiver. Groups were compared for CLABSI per 1000 CVC days. Demographic data was collected for age, sex, diagnosis, and infusion days per week. Results: The average age of the study group was 58 years, 69% were female, 70% had a diagnosis of short bowel syndrome, and 80% infused HPN 6-7 days per week. There were a total of 278,635 tunneled CVC days with 210 infections at an overall CLABSI rate of 0.75/1000 CVC days. The PV group had 118 infections in 181,650 CVC days with a rate of 0.65/1000 days compared to the CVC group with 92 infections in 96,985 CVC days at 0.95/1000 days. It should be noted that in both groups some patients experienced multiple infections while others remained infection free. See Table 1 for a summary of CLABSI data. Conclusions: There was a clinically significant increase in CLABSI when blood was drawn from the CVC compared to PV. Of the subset of patients with CVC draws, the group drawing their own blood had the lowest incidence of infection and family caregiver blood draws had the highest. Of interest, the clinic/MD office clinicians had a lower infection incidence than home health nurses. The HPN patient is the guardian of their CVC and appear to be the most meticulous with their own care. Patients should closely scrutinize catheter care techniques when anyone manipulates their CVC, and voice concerns when uncomfortable with technique. Ongoing education and review of aseptic technique is vital for HPN patients, health care providers, and family caregivers who work with CVCs. It is recognized that CLABSI is a complex complication and other risk factors may have contributed to this cohort's infection rate, but this data indicates that blood drawing practices should be part of a total CLABSI incidence assessment for HPN patients. More investigation is needed to identify strategies to reduce CLABSI in the HPN population.


Table 1. Blood Collection Data Summary.

Draw N patients N CLABSI CVC Days CLABSI/1000 Days

PV 68 118 181650 0.65

CVC 40 92 96985 0.95

Self 13 13 35499 0.37

Clinic/MD Office 11 17 27246 0.62

Home Health RN 10 30 21792 1.37

Family Caregiver 6 32 12448 2.57

Nutrition and Metabolism Research Paper Session: Enteral Nutrition A.S.P.E.N. Trainee Award 2078842 - A Volume-Based Enteral Nutrition Support Regimen Improves Caloric Delivery but May Not Impact Clinical Outcomes Ivy Haskins, MD; Mary E. Baginsky, MS, RD, CNSC; Alison Campbell, RD, CNSC; Nathaniel Gamsky, BS; Kia Sedghi, BS; Sojung Yi, BS; Richard Amdur, PhD; Mary Gergely, RD, CNSC; Babak Sarani, MD, FACS General Surgery, George Washington University Hospital, Washington, D.C.

Purpose: Underfeeding in the critically ill patient is associated with increased morbidity and mortality. Nevertheless, meeting enteral nutrition goals is an ongoing challenge in the intensive care unit. Most hospitals utilize an hourly rate based protocol (RB) for nutrient delivery. This method of feeding is predicated upon patients being fed around the clock without interruption, which is impractical. Previous studies have found that a volume based protocol (VB) allows for improved delivery of prescribed calories.


However, these studies did not assess clinical outcomes. The purpose of this study is to investigate the difference between a RB and VB strategy on delivery of prescribed calories and clinical outcomes. We hypothesize that VB strategy leads to improved delivery of prescribed nutritional goals. Methods: In 2013, our institution changed the method of enteral nutrition delivery for critically ill patients from a RB to a VB protocol. We carried out a single-institution, before and after study. Inclusion criteria included ICU patients age 18 years or older who received continuous enteral nutrition support for at least 24 hours. Patients less than 18 years old and those not receiving enteral nutrition were excluded. Consecutive medical and surgical patients from April-July 2013 comprise the RB group and patients from December 2013 through February 2014 comprise the VB arm. Admission APACHE II score was calculated for all patients. The effect of RB and VB on percentage of goal calories received, ICU length of stay (LOS), hospital LOS, and days on the ventilator were investigated using the Kruskal-Wallis test of differences. Multivariate regression was used to identify independent predictors of outcome. Significance was defined as p


Abstract of Distinction 2072741 - Use of Blenderized Tube Feeding in Patients on Home Enteral Nutrition Adele Pattinson, RDN, LD, CNSC

2; Lisa Lammert, RDN, LD, CNSC

2; Lisa Epp, RDN, LD, CNSC

2;

Jennifer Lintz, RDN, LD2; Jithinraj Edakkanambeth Varayil, MD

1; Manpreet Mundi, MD

2; Ryan Hurt, MD,

PhD1

1Division of General Internal Medicine, Mayo Clinic Rochester, Rochester, MN;

2Division of

Endocrinology, Mayo Clinic Rochester, Rochester, MN;

Purpose: The use of long-term enteral tube feeding has increased dramatically in the United States and worldwide with the number of patients receiving enteral nutrition at home increasing from 34,000 in 1989 to 344,000 in 2014. While most patients are initiated on commercially available enteral formulas, it has been our experience that a majority of home enteral nutrition (HEN) patients also use blenderized tube feeding (BTF) to partially or completely deliver daily nutritional requirements. They have anecdotally reported better tolerance and decreased gastrointestinal symptoms. Despite widespread use, limited resources are available for patients who are interested in BTF and studies evaluating the safety, tolerability, and clinical effectiveness of BTF are limited. Methods: We conducted a prospective cross-sectional survey with 54 patients to understand BTF prevalence and use. BTF was defined as any food (liquid or food mixture) administered via feeding tube that is not water, medication or commercial formula. Participants were provided the survey at HEN follow-up appointments with those fed post-pylorically being excluded. Survey questions included gender, whether or not they have used BTF, rationale for using BTF, number of days per week of BTF use, percentage of daily food that is comprised of BTF, preparation and storage of BTF, weight changes and gastrointestinal symptoms while using either BTF or a commercial formula. Results: The mean age of our cohort was 58.9+/-13.7 years and 23 (42.5%) patients were female. 30 (55.5%) patients had used BTF in the past and a total of 16 (30%) patients were on HEN greater than 1 year, 21 (21%) were on HEN for more than 6 months, 3 (6%) were on HEN for less than a month. 38 (70.3%) were not working/retired, 12 (22.2%) were working part-time and 4 (7.4%) were working full time. Patients in our cohort used blenderized tube feeding for a median of 4 days per week ranging from 1-7 days. Patients provided various reasons for using BTF including 13 (43%) citing it as being more natural, 10 (33%) citing they like eating what their family eats, and 9 (30%) citing they can tolerate BTF better (Table 1). In patients who use BTF, 17 (56.6%) patients reported that it comprised up to 50% of their daily enteral feedings. 79.3% of patients who used BTF reported that they were successful in maintaining their body weight. 67% reported that they did not experience any symptoms such as nausea, vomiting, fever, diarrhea while using BTF. A comparison of patients using BTF and commercial tube feeding showed less reported nausea, vomiting, gas/bloating, diarrhea, constipation, and pain with BTF (Table 2). Conclusions: From our survey, it is clear that more than 50% of patients surveyed use BTF and close to 90% expressed a desire to use BTF if provided with adequate information about the appropriate preparation and storage practices. However, limited resources are available regarding appropriate preparation and storage of BTF. Additionally, studies to evaluate the safety, tolerability and clinical effectiveness of BTF are lacking. In the future, we are planning to conduct a study assessing the safety and efficacy of BTF. Outcomes will compare the efficacy (calories, micro, macronutrients delivered), safety (side effects including symptoms and possible foodborne illness), quality of life and traditional enteral nutrition clinical outcomes (tube occlusion, infection, hospitalization). It is our hope that results from this study could help to direct the development of better resources for HEN patients choosing to use BTF.



Table 1. Reasons for Using Blenderized Tube Feeding.

Reasons Number of Patients (%)

It makes me feel “normal” 9 (24%)

I can tolerate it better 11 (30%)

I like eating what my family eats 12 (32%)

It is more natural 17 (46%)

I don’t like the ingredients of commercial formulas 9 (24%)

I have food allergies 2 (5%)

Other reasons 6 (16%)

Table 2. Comparison of Self-Reported Symptoms on Blenderized Tube Feedings and Commercial Tube

Feedings.

Symptom While Using BTF While Not Using BTF P Value

No symptoms 25 (68%) 16 (43%) .002

Weight Loss 6 (17%) 10 (33%) .10

Gas/ Bloating 6 (16%) 9 (24%) .08

Nausea/ Vomiting 4 (11%) 3 (8%) .31

Fever 1 (3%) 4 (11%) .18

Diarrhea 2 (5%) 5 (13%) .17

Constipation 1 (3%) 4 (11%) .08

Pain 1 (3%) 3 (8%) .31

Abstract of Distinction 2079668 - An Enteral Access Team Utilizing Electromagnetic Placement Device and Clinical Nutrition Practice Management Guidelines Improves Quality of Care Nancy J. Park, MS, RD, CNSC

1; Thomas E. Knuth, MD, MPH, CNSC

2

1Food and Nutrition-Touchpoint, St. John Hospital and Medical Center, Detroit, MI;

2Department of

Surgery, Trauma and Critical Care, St. John Hospital and Medical Center, Detroit, MI

Purpose: In 2011 our tertiary care, level II trauma center adopted the use of an Electromagnetic Placement Device (EMPD) for insertion of small-bore feeding tubes. The major impetus for this change was overuse of Parenteral Nutrition (PN), and inappropriate use of PN in patients with functional gastrointestinal tracts. PN was primarily used for patients who exhibited gastric intolerance of feedings. Determining gastric "intolerance" was often interpreted anecdotally by nurses and physicians. Existing policies and procedures for blind post-pyloric placement of small-bore feeding tubes were not observed, nor were endoscopic or fluoroscopic placements utilized. We therefore trained an expert group of nurses, known as the Enteral Access Team (EAT), to be adept at placing tubes post-pylorically via EMPD. Three team "captains" are responsible for recruiting team members, training and certification on EMPD, bridling of tubes, documentation, and quality assurance. Competency includes interpretation of tip location on the EMPD screen. Tip placement is confirmed by the nutrition support physician. Utilization of EMPD has resulted in near elimination of confirmatory radiograph; (


Management Guidelines is $1,033,147.00.With the success of our EMPD process and EAT, we continue to seek out further improvement. Physician and nurse education, along with improvement in quality data tracking of enterally fed patients are goals to which we are striving. Even with a post-pyloric feeding program, we still experience mechanical and administrative complications including clogging, dislodgement, and withholding of nutrition for procedures and therapies. Follow up of patients with small-bore feeding tubes who are discharged to skilled nursing facilities, long-term acute care, and home is sporadic. We want to institutionalize our Nutrition Support Practice Management Guidelines. Other goals include developing a "PEP uP" enteral protocol, improving communication with our EAT and dietitians for enteral access selection and nutrition care plan development, and tracking outcomes of enterally fed patients. Financial support: none.

Abstract of Distinction 2078065 - Optimizing Enteral Nutrition Delivery in a Level I Trauma Intensive Care Unit by Using a Physician-Driven Catch-Up Feeding Protocol Jenny C. Lee, MS, RD, LD, CNSC

1; Rosemary A. Kozar, MD, PhD

2; Charles C. Miller, PhD

3; Laura J.

Moore, MD, FACS2

1Clinical Nutrition, Memorial Hermann Hospital - Texas Medical Center, Houston, TX;

2Department of

Surgery, University of Texas Health Science Center at Houston, Houston, TX; 3Department of

Cardiothoracic and Vascular Surgery, University of Texas Health Science Center at Houston, Houston, TX

Purpose: Prior studies indicate patients admitted to intensive care unit (ICU) after trauma receive approximately 60% of their estimated calorie requirements. Frequent operations and procedures limit escalation and maintenance of feeding goals. The purpose of this quality improvement project was to compare nutritional adequacy before and after initiation of a physician-driven catch-up feeding protocol designed to optimize enteral nutrition (EN) delivery to trauma patients in the ICU. Methods: All trauma patients aged ≥18 admitted to the Shock Trauma ICU and receiving enteral nutrition (EN) for >48 hours were included. Patients who suffered non-trauma injuries, sustained non-survivable injuries, received any parenteral nutrition, or re-admitted to the ICU were excluded. Data collected were: admission diagnosis, age, gender, prescribed calories and protein, and daily total EN volume received. EN data collection started on the day of ICU admission and continued until the day patients received an oral diet, were discharged from ICU, or reached ICU Day 12. Starting in April 2014, fellows, residents, and medical students were trained on the Physician-Driven Catch-Up Feeding Protocol. Early initiation of EN (within 24-48 hours of ICU admission) was also emphasized. Residents were required to report EN volume infused in the past 24 hours and percent of EN volume during ICU rounds. A registered dietitian recorded EN volume in a calorie tracker. If patients received


Conclusions: The use of a Physician-Driven Catch-Up Feeding Protocol resulted in a significant increase in nutritional adequacy in a group of injured patients at high risk for underfeeding due to frequent operations and procedures. The keys to the success of this protocol were nutrition education and compliance by our physicians, simplicity of the protocol, and a dietitian dedicated to the ICU. The effect of this protocol on clinical outcomes is currently being investigated.


Table 1. Demographic and Nutritional Data for Baseline and Intervention Groups.

Characteristics Baseline (n=86) Intervention (n=70) P Value

Age, year, mean (range) 46 (18-92) 43 (19-98) 0.269

Male, sex, % 86% 70% 0.015

Boday Mass Index (BMI) 27.8±6.4 28.2±6.4 0.648

Calories prescribed, kcal/day 1861±230 1965±348 0.034

Calories received, kcal/day 1224±713 1783±817


Figure 2. Cumulative caloric deficit during the first 12 ICU days.

2078352 - Impact of Timing and Dose of Energy Exposure in Intensive Care Unit Patients With Acute Lung Injury Sarah Peterson, MS, RD, CNSC

1,3; Omar Lateef, DO

2; Miyoung Kim, BS

3; Liam McKeever, MS, RD

3;

Sally Freels, PhD4; Carol Braunschweig, PhD, RD

3

1Department of Food and Nutrition, Rush University Medical Center, Chicago, IL;

3Department of

Kinesiology and Nutrition, University of Illinois at Chicago, Chicago, IL; 2Pulmonary and Critical Care

Medicine, Rush University Medical Center, Chicago, IL; 4School of Public Health Division of Epidemiology

and Biostatistics, University of Illinois at Chicago, Chicago, IL

Purpose: High quality prospective randomized control trials (PRCT) have reported no difference or worse outcomes in intensive care unit (ICU) patients randomized to receive greater percentage of daily calorie requirements. We recently conducted a PRCT to assess the impact of intensive medical nutrition therapy from diagnosis of acute lung injury (ALI) through hospital discharge versus standard care on clinical outcomes (INTACT). The trial was stopped at 40% of goal recruitment (78/200) due to significantly greater mortality in the intervention compared to control group (40 vs 16%, p=0.02). Post-hoc analysis established a relationship between mortality and calorie delivery; early exposure (post randomization days 1-7) to more than 12.5 calories per kilogram (kcal/kg) predicted greater hazards for future hospital mortality. The objective of the current retrospective study was to determine if the relationship observed between mortality and calorie exposure among INTACT participants remained in a comparable population of intubated ICU patients with ALI. Methods: Patients who met eligibility INTACT criteria during the recruitment period but did not participate in the trial were included in this retrospective cohort study. Baseline demographics included age, gender, race/ethnicity, BMI, subjective global assessment, reason for ICU admission, PaO2:FiO2 ratio, sequential organ failure assessment (SOFA) score and hospital days before enrollment. Calorie delivery (including delivery of enteral nutrition (EN), parenteral nutrition (PN), propofol and dextrose containing intravenous fluids) was collected from 24 hours of ALI diagnosis through diet advancement or ICU discharge. Energy needs were estimated at 30 kcal/kg. Overall average energy intake was dichotomized into


days before enrollment was 3.0 (5.3). Patients received nutrition support for 12.5 (7.9) days and mean intake was 13.7 (6.7) kcal/kg, only 6% received PN. Overall mortality was 33% (98/298). Subjects who received >12.5 kcal/kg had a higher likelihood of death (OR 1.822 95% CI 1.061-3.128, p


Nutrition and Metabolism Research Paper Session: Malnutrition, Obesity, Practice Concepts

Abstract of Distinction 2077480 - Impact of a Novel Strategy for Individualized Diet Intervention in Children on Home Mechanical Ventilation. Enid E. Martinez, MD

1,3; Lori J. Bechard, PhD, RD

2; Craig D. Smallwood, RRT

1,3; Robert J. Graham,

MD1,3

; Nilesh M. Mehta, MD1,3

1Division of Critical Care Medicine, Department of Anaesthesia, Perioperative and Pain Medicine, Boston

Children's Hospital, Boston, MA; 3Harvard Medical School, Boston, MA;

2Center for Nutrition, Boston

Children's Hospital, Boston, MA

Purpose: We have previously reported a prevalence of malnutrition, low protein intake and altered body composition in children on home long-term transtracheal mechanical ventilation. Malnutrition and declines in lean body mass (LBM) are associated with poor outcomes in patients with chronic respiratory insufficiency. Individually modifying the energy and protein intake of children on home long-term mechanical ventilation to optimize nutritional status and body composition has not been previously reported. Methods: We performed a prospective interventional study in children aged 1mo to 17 years on home long-term mechanical ventilation. PRE-INTERVENTION: A baseline home visit was conducted where the following assessments were completed: height and weight measurements, body composition by bioelectrical impedance analysis, actual energy intake (AEI) by diet recall, and measured energy expenditure (MEE) by indirect calorimetry. Nutritional status was determined by weight for age z-scores (WAZ) and body mass index (BMI) z-scores. Fat mass percentage (FM%) was compared to age, gender and race matched population norms in children 3 years or older. Feeding state was defined by the AEI:MEE ratio (AEI:MEE 110%, overfed). DIET INTERVENTION: Diets were individually modified and prescribed to meet an AEI that was 90 - 110% of MEE. A.S.P.E.N. recommendations for daily protein intake were followed. The modified diet was continued for approximately 12 weeks, with monthly telephone follow-up. POST-INTERVENTION: All assessments performed during the baseline home visit were repeated during a post intervention home visit. Diet adherence was determined by comparing the prescribed diet and a repeat diet recall. We examined percent differences between pre and post-intervention values for weight, LBM% and FM%. Results: Sixteen of 20 patients, mean age 9.33y (SD 4.9), 8 male, completed the study including diet intervention and a pre and post-intervention home visit. PRE-INTERVENTION: 9/16 (56%) subjects were classified as at risk for malnutrition. Mean FM% was significantly higher in study patients ≥3 years old compared to population norms, (41.6% vs 22.8%, p< .0001). Seven patients were classified as underfed; 4 as overfed. Mean protein intake was 1.5 g/kg/d; protein intake in 8 of 16 patients was below recommended goals for age. DIET INTERVENTION: Prescribed energy intake was increased in 7 patients and decreased in 6 based on feeding state. In patients classified as underfed, mean daily energy intake was significantly increased from baseline visit, 39.8 vs. 44.4 kcal/kg/d, p= .03. Protein intake was increased in all patients. The mean protein intake was increased to 2 g/kg/d, significantly greater than baseline protein intake, p= .02. POST-INTERVENTION: Mean time from diet intervention to post intervention visit was 14 weeks (SD 3.2). There were no significant differences between prescribed and actual energy intake, suggesting adherence to the diet intervention. Percent difference in weight between the pre and post-intervention visits was independent of concurrent changes in LBM% and FM% in individual patients. (Figure 1) Conclusions: Malnutrition and altered body composition were prevalent in a majority of children on home long-term mechanical ventilation in our study. Furthermore, dietary interventions were needed in 81% of this cohort to rectify underfeeding or overfeeding. Body composition assessments should be incorporated into nutrition assessments in this population. Individualized diet modifications based on comprehensive assessments are feasible, well tolerated, and may result in improved patient outcomes.

This research was partially supported by the Medical Staff Organization’s Boston Children's Hospital House Officer Development Award. The metabolic measurement device was funded by the Boston Children's Hospital Payer-Provider Quality Initiative Grant for the CAPE Program.


Figure 1. Percent difference in weight (kg), lean body mass percent (LBM%), and fat mass percent (FM%) in 16 children on home long-term mechanical ventilation following a diet intervention based on comprehensive nutrition assessment. LBM% and FM% were not available for subject 1.

International Abstract of Distinction 2078614 - An Examination of Measured vs Predicted Resting Metabolic Rate in Older Adults Living in Long-Term Care Facilities Jessica Hicks, MScFN(c)

1; Stephanie Malo, BSc

2; Sayuri Omori, BSc, RD

2; Norine Foley, MSc, RD

3;

Janet Madill, PhD, RD1; Colleen O'Connor , PhD, RD

1

1Food and Nutritional Sciences, Brescia University College at Western University, London, Canada;

2Nutrition Management Services, London, Canada;

3Physical Medicine and Rehabilitation, Schulich

School of Medicine, Western University, London, Canada

Purpose: Obesity and malnutrition have been cited as prevalent issues among residents living in long term care facilities (LTC), suggesting that maintaining long-term energy balance is difficult to achieve. This may be of particular concern for residents >80 years, as their energy needs have not been well studied. While indirect calorimetry is the gold standard, prediction equations are the most common and feasible approach used to estimate energy needs in clinical settings. Although many equations exist, they are associated with both under and over estimates beyond acceptable limits. Furthermore, no prediction equations have been validated for use in persons >80 years. The objective of this pilot study was to determine the accuracy of six commonly used prediction equations, when compared to measured resting metabolic rate (RMR), in older adults living in LTC. Methods: All residents ≥70 years old, living in two LTC facilities in Ontario, Canada from September 2013 to March 2014, whose weight had been stable for the six months prior to the study were invited to participate. Sex, height, and weight data were obtained through chart review for each resident. Body Mass Index (BMI) was calculated. The RMRs (Kcals/24 hours) of participants were measured on a single occasion using a portable metabolic analyzer (FitMate Pro, Cosmed, Chicago, IL). Participants were asked to rest quietly in a supine or semi-recumbent position while measurements were obtained for 15-20 minutes. Measured RMR (RMRm) was compared to estimated RMR (RMRe) for each resident. RMRe was calculated using six prediction equations: Harris Benedict (HB), Mifflin-St Jeor (MSJ), Food and Agriculture Organization/World Health Organization/United Nations University (FAO/WHO/UNU), Schofield, Fredrix, and Owen. The percentage deviation of the RMRe from RMRm was calculated (i.e. [(RMRe - RMRm) / RMRm] * 100). The accuracy of each prediction equation was defined as the number of participants whose RMRe was within the clinically acceptable limit of ±10% of their RMRm. The mean


absolute deviation (RMRm - RMRe) ± standard deviation (SD) was also calculated. Results: Forty-six residents met the inclusion criteria. Twenty-four residents declined, one was excluded, and one did not complete the study test. RMR was measured for 20 residents (14 women, 6 men), with a mean age of 86.6 ± 6.9 years. Mean BMI was 26.7 ± 3.9 kg/m2. Of the six prediction equations evaluated, three estimated RMR within 10% of RMRm for 5/20 (25%) residents, two correctly estimated RMR for 4/20 (20%) and one equation estimated correctly for 2/20 (10%) residents. On average, all equations overestimated energy requirements more frequently and to a greater degree than they underestimated them. The RMRe was overestimated in 8 to 14 of 20 cases, and underestimated in 4 to 8 cases (range of over-estimation=10.1% to 189% of predicted vs. range of underestimation=10.6% to 29.8% of predicted). The largest deviations from RMRm were associated with the FAO/WHO/UNU and Owen equations. The lowest mean absolute deviation from RMRm was associated with the MSJ equation (37 ± 372 Kcals/day), and the highest was associated with the FAO/WHO/UNU equation (256 ± 352 Kcals/day). Conclusions: The performance of six commonly-used equations to estimate energy needs in a small group of older adults living in LTC, was poor. The use of these equations was most-often associated with a clinically significant overestimation of residents' energy needs. Although the sample size was small, the results suggest that these equations may not be appropriate for use in this population. These results may also have implications for menu planning in LTC facilities to ensure residents are not being over fed.

Financial support received from The Canadian Foundation for Dietetic Research.

Abstract of Distinction 2077100 - Appropriateness of a Very High Protein, Low Carbohydrate Formula in Critically Ill Patients With Obesity: A Pilot Study of Design, Safety, and Tolerance Stephen McClave, MD, Professor of Medicine

1; Andrew Bernard, MD

2; Maureen Huhmann, RD

3; Cynthia

C. Lowen, RD3; Juan Ochoa, MD

3

1Medicine, University of Louisville, Louisville, KY;

2Surgery, University of Kentucky, Lexington, KY;

3Medical Unit, Nestle Health Science, Florham, NJ

Purpose: As the weight demographic of the U.S. population increases, so does that of the intensive care unit. For the critically ill patient with obesity, virtually all commercial formulas contain insufficient protein for the caloric content, creating less than optimal feeding opportunities. We designed this prospective pilot study to evaluate tolerance, safety and design of a new, unique enteral formula in critically ill patients with obesity. Methods: In this prospective observational study, patients with a BMI ≥ 30, requiring EN ≥ 3 days were recruited from both Surgical and Medical ICUs at two university hospitals in Kentucky. Study patients received a new innovative formula [1.0 Kcal/mL, 37% protein (100% whey hydrolysate), 31% CHO, with MCT, fish oil and prebiotic fiber] for up to 10 days. Calorie goals designed to meet 60-65% of maintenance requirements were based on 14 Kcal/kg ABW or 70% of MREE; protein goals were calculated at ≥ 2.0g/kg IBW for Class II obesity and ≥ 2.5g/kg IBW for Class III obesity. A “completer” was defined as receiving five days of EN. Daily intake, tolerance and safety data were collected and analyzed using descriptive statistics. Results: Data from 29 subjects were collected; 3 subjects were ineligible, 2 subjects never began EN and 8 subjects received < 5 days of EN. Therefore, a total of 16 subjects (8M,8F) with mean age of 50 years (range of 18-77) entered the analysis, 15 of whom were ventilator-dependent at initiation of trial. Average APACHE II score was 17.3 and average BMI (kg/m2) was 38.8 (range 30.6-67.7). Calorie provision at day 3 was 80% of calculated or measured needs (average of 1270 Kcal/day) and protein provision was 86% of estimated requirements (average of 118 g pro/day). Mean laboratory values were as follows: BUN 29.07 mg/dL, Creat 0.96 mg/DL, Blood Glucose 142.8 mg/dL, Ketones 0.0mg/dL. Four subjects were noted to have ≥ 3 liquid stools/day for partial duration of the study. No subjects had GRV > 400mL and there were no episodes of vomiting. Positive nitrogen balance was recorded in seven of the 16 subjects (of which none exceeded 100% of caloric goal) and negative N2 balance in three. Conclusions: A unique enteral formula comprised of 37% protein (100% whey), 31% CHO, containing MCT, fish oil and prebiotic fiber is safe and well-tolerated in critically ill patients with obesity. The design of this unique formula may facilitate glucose control and allows for achievement of N2 balance without


exceeding the goals of a hypocaloric feeding strategy. Future trials are needed to explore the potential health benefits associated with this type of enteral support in the critically ill patient with obesity.

Financial support received from Nestlé Health Care Nutrition.

Abstract of Distinction 2077039 - Use of Various Nutrition Screening Tools to Identify Nutrition Risk in Hospitalized Patients Anne Coltman, MS, RD,CNSC; Molly DePrenger, BS; Kelly A. Roehl, MS, RD, CNSC; Melanie Betz, MS, RD, CSG; Diane Sowa, MBA, RD Rush University Medical Center, New Lenox, IL

Purpose: The Joint Commission requires all individuals admitted to a hospital to receive nutrition screening; however, no recommendations have been made regarding the use of a specific tool. Additionally, guidelines exist for nutrition assessment, but these assessments are designed to be complete by a trained nutrition professional and can be time consuming. Therefore, nutrition screening tools were created to quickly identify patients at malnutrition risk who may require a complete nutrition assessment and further nutrition intervention. Currently, no gold standard has been agreed upon for nutrition screening. The objective of this quality improvement project was to determine the prevalence of malnutrition risk using five nutrition screening tools. Methods: A convenience sample of 1346 patients admitted to a large, urban academic center was used. All components required for each screening tool were collected, including weight loss, change in appetite, change in oral intakes, changes in mobility, disease severity, psychological problems, BMI, age, and use of supplements. Each patient was assessed for all characteristics of each screening tool as outlined in Table 1. Rates of specificity, sensitivity, and positive predictive value were calculated using Subjective Global Assessment as the reference standard. Results: SGA identified 32% as malnourished. The NRS 2002 had the highest rate of nutrition risk (Table 2), but the lowest specificity. Alternatively, the MST had the lowest rate of nutrition risk, but the highest rate of specificity. The MNA-SF and MUST were closest in estimation to SGA and had favorable sensitivies and specificities. Conclusions: Different screening tools resulted in determining a large variety of patients at risk for malnutrition. The MNA-SF and MUST found results similar to SGA; both tools include markers of disease severity along with traditional markers of nutrition risk. Inclusion of both markers may be helpful in identifying patients requiring a full nutrition assessment. While nonspecific markers of nutrition, such as weight loss, increase identification of nutrition risk, false positive identifications often result. Further research is required to create and validate one screening tool that concisely measures pertinent characteristics associated with nutrition risk.


Table 1. Characteristics Used in Nutrition Screening Tools.

Screening tool

Weight loss

Decreased appetite/intake

Mobility Disease severity

Psychological problems

BMI Age Use of

supplements/tube feeding

MST X X

MNA-SF X X X X X X

MUST X X X

NRS 2002 X X X X X

SNAQ X X X

MNA-SF, Mini Nutrition Assessment-Short Form; MST, Malnutrition Screening Tool; MUST, Malnutrition Universal Screening Tool; NRS 2002, Nutritional Risk Screening 2002; SNAQ, Short Nutritional Assessment Questionnaire.


Table 2. Rates of Nutrition Risk, Specificity, Sensitivity, and Positive Predictive Value of Nutrition Screening Tools When Compared to Reference Standard (Subjective Global Assessment).

Screening Tool

Number of Patients Identified at Nutrition Risk

Specificity Sensitivity Positive Predictive

Value

MST 204 (15%) 99% 46% 97%

MNA-SF 448 (33%) 89% 80% 77%

MUST 390 (29%) 90% 70% 77%

NRS 2002 751 (56%) 58% 84% 48%

SNAQ 267 (20%) 59% 99% 96%

MNA-SF, Mini Nutrition Assessment-Short Form (classified at risk if score ≤11; patients received score of 0 for disease severity if APACHE II score ≥20); MST, Malnutrition Screening Tool (classified at risk if score ≥2); MUST, Malnutrition Universal Screening Tool (classified at risk if score ≥1; patients received score of 2 for disease severity if APACHE II score ≥20); NRS 2002, Nutritional Risk Screening 2002 (classified at risk if total score ≥3; patient classified as severely ill if APACHE II score ≥10); SNAQ, Short Nutritional Assessment Questionnaire (classified at risk if score ≥2).

2079487 - Diet Rich in Sugar and Fat Are Able to Reduce Regulatory T Cell in C57BL/6 Mice Tatiani Maioli, PhD

1; Mariana C. Miranda, MS

2; Juliana L. Gonçalves, PhD

2; Thais G. Moreira, PhD

2;

Andrezza F. Santiago, PhD2; Ana Maria C. Faria, PhD

2

1nutrição, Universidade Federal de Minas Gerais, Belo Horizonte, Brazil;

2Bioquímica e Imunologia,

Universidade Federal de Minas Gerais, Belo Horizonte, Brazil

Purpose: Unbalanced eating habits are closely related to the increased percentage of body fat and obesity. It is known that hypercaloric diets lead to generation of inflammatory processes, culminating in metabolic disorders and obesity. Obesity is linked to several diseases and metabolic disorders, such as insulin resistance, dyslipidemia and hepatic steatosis, known as metabolic syndrome. Increased consumption of diet containing fat or refined carbohydrate contributes to obesity and related diseases. Therefore, approaches using different experimental diets should be useful to understand the metabolic and inflammatory impairment induced by increased fat deposition. Although many studies have shown the effect of obesity induced by high fat diet, little is known about the progression of metabolic and inflammatory dysfunction induced by adipose tissue expansion in animals fed high-fat-high-sugar (HFS) diets, like humans. The present work aims to describe a mouse model of fat pad expansion, dyslipidemia, glucose intolerance and immunological alterations induced by a HFS diet and demonstrate that this diet seems to be more effective in triggering metabolic disorders associated with inflammation in adipose tissue than HF diets Methods: Transgenic C57BL/6 Foxp3GFP knock in mice were fed for 11 weeks either with: high-fat diet (HF) containing 53% of fat, 28% of sugar and 19% of protein; high-fat-high-sugar diet (HFS) containing 48% of fat, 36% of sugar and 17% of protein or control diet based on the AIN93G. Than blood samples were collected for cholesterol, triglycerides and glucose. Leukocytes from adipose tissue were isolated with collagenase and cells were stained for CD25 and CD4. The amount of Foxp3+ T cells were calculated among CD4+CD25+ cells. All procedures were approved by ethical committee. Results: HF and HSF diet significantly increase body weight gain after three weeks of feeding (Figure 1A) (p


Financial support received from CNPq, FAPEMIG, CAPES & PRPq.

2077252 - Clinician Attitudes Concerning Ethical Practice in Nutrition Care: A.S.P.E.N. Member Perspective. Nader Armanios, RD

1,2; Denise B. Schwartz, MS, RD, CNSC, FADA, FAND, FASPEN

3; David August,

MD4; Eric H. Frankel, MSE, PharmD, BCNSP

5,6; Elizabeth Sussman, PhD, RD

1; Cheryl Monturo, PhD,

MBE, ACNP-BC7; Alessandro Pontes-Arruda, MD, MSc, PhD, FCCM

8,9

1Family and Consumer Sciences, California State University Northridge, Northridge, CA;

2Food and

Nutrition Services, Olive View-UCLA, Sylmar, CA; 3Food and Nutrition Services, Providence Saint Joseph

Medical Center, Burbank, CA; 4Rutgers Cancer Institute of New Jersey, Rutgers Robert Wood Johnson

Medical School, New Brunswick, NJ; 5Neonatal Clinical Lead & Metabolic Support Service, Truman

Medical Center, Kansas City, MO; 6West Texas Clinical Pharmacy Associates, Inc, Lubbock, TX;

7College

of Health Sciences, West Chester University of Pennsylvania, West Chester, PA; 8Christus

College/School of Medicine , Fortaleza, Brazil; 9Asia-Pacific Baxter Healthcare, Singapore, Singapore

Purpose: As healthcare technologies advance, questions about clinician attitudes concerning ethical practice with nutrition therapies arise. The controversy to initiate or continue nutrition support in some patients remains. Withholding nutrition support is appropriate, if the risks/burdens outweigh the benefits as perceived by the informed patient, family, caregiver, or surrogate decision maker. The International Clinical Ethics Section (I.C. Ethics) of A.S.P.E.N. developed a survey to identify current practice and attitudes of A.S.P.E.N. members and assess education needs. We hypothesize that there may be a difference between disciplines in how to improve clinical ethics application in nutrition care. Methods: A literature review of the past five years identified suggested actions by hospitals and clinicians to optimize clinical ethics in nutrition care. Survey questions were assembled with input from 20 I.C. Ethics members, representing international clinicians from medicine, dietetics, pharmacy, and nursing with different levels of experience. A.S.P.E.N. office entered the survey into SurveyMonkey software. The online survey was announced through a weekly newsletter with a 3-week return period. SPSS software was used to run statistical analysis. Chi-square test and Fischer’s exact test were applied to analyze differences among disciplines and additional collected data on ranking of clinician and healthcare institution action statements; p


adults compared to pediatrics and neonatal (p=0.002). Figure 2 shows ranked clinician action statements. There was no statistical significance among disciplines and years of experience. Statistical significance was seen between those who were part of an ethics committee and those who are not (p= 0.037). There was a significant difference between clinicians practicing in the USA, and clinicians outside the USA (p= 0.034). The majority of respondents (62%) indicated they will read more on this topic, as a result of completing this survey. Conclusions: There is no statistically significant difference between disciplines in how to improve clinical ethics application in nutrition care. However the survey results suggest the importance of the following action statements: 1) prevent ethical dilemmas with early communication with patient/family and incorporate evidence-based medicine on benefits versus risk/burdens of nutrition support; 2) healthcare institution establishment of a process to obtain advance directives and/or begin early discussion on healthcare wishes and consistent healthcare team approach; 3) engage with our patients/families to understand what matters most to them at the end of life; 4) use of measurable goals of increased advance directives on chart and increase in number of patients with decision makers to indicate improvement; 5) articles in Nutrition in Clinical Practice and Journal of Parenteral and Enteral Nutrition, or other nutrition related journals most helpful to engage and educate nutrition support clinicians on clinical ethics. Further research is needed to strengthen the literature available on this topic.



Nutrition and Metabolism Research Paper Session: Critical Care & Other Critical Health Issues

Abstract of Distinction 2077687 - The Impact of Unintended Macronutrient Intake Inadequacy on Outcomes in Critically Ill Children: Spotlight on Protein Delivery in the Pediatric Intensive Care Unit Nilesh M. Mehta, MD

1,2; Lori J. Bechard, PhD, RD

2; David Zurakowski, PhD

1; Christopher P. Duggan,

MD, MPH2; Daren K. Heyland, MD, MSc

3

1Critical Care Medicine, Department of Anesthesiology, Perioperative and Pain Medicine, Boston

Children's Hospital, Boston, MA; 2Center for Nutrition, Boston Children's Hospital, Boston, MA;

3Clinical

Evaluation and Research Unit, Kingston General Hospital, Kingston, Canada Purpose: Little is known about the impact of adequate nutrition delivery on clinical outcomes in critically ill children. We completed a large international cohort study examining nutrient delivery in pediatric intensive care units, with the aim of describing adequacy of energy and protein delivery and its relationship with 60-day mortality in critically ill children. These findings have implications for bedside practice and potential improvement in clinical outcomes in this vulnerable cohort. Methods: DESIGN: We conducted a prospective multicenter cohort study, including 59 pediatric intensive care units (PICU) from 15 countries. PICUs with ≥ 8 beds were eligible to participate. SUBJECTS AND MEASUREMENTS: Consecutive children (1 month - 18 yrs) mechanically ventilated for ≥ 48 hours were


enrolled. Energy and protein adequacy, timing and route of nutrient delivery, use of enteral nutrition adjuncts, and clinical outcomes were recorded daily by site investigators and transmitted using a remote electronic data capture tool. MAIN OUTCOME: We examined the association between the percent adequacy of energy and protein delivery (defined as nutrient delivered / nutrient prescribed x 100), and 60-day mortality. We used multivariable modeling to examine variables that predicted adequacy of nutrient delivery. Results: We enrolled 1245 subjects (44% female) with median (IQR) age 1.7 (0.4, 7.0) years, median weight for age z-score -0.60, and mortality 6.6%. A majority of the cohort (n=985, 79%) received enteral nutrition (EN), 354 (36%) via the post-pyloric route. Mean (SD) prescribed energy and protein goals were 69 (28) kcals/kg/day and 1.9 (0.7) grams/kg/day, respectively. EN was initiated in 60% of this group by Day 2 and in 80% by Day 3 after admission. However, EN was interrupted in 724 (58%) subjects, on average 8 hours per day. Mean (SD) adequacy (percentage of the prescribed goal that was delivered) for energy and protein delivery in the PICU was 36% (35) and 47% (48) respectively. Figure 1 shows mean and median cumulative protein delivery for the cohort during each day on the PICU. After adjusting for illness severity, 60-day mortality was significantly associated with adequacy of enteral protein intake (P


Figure 1. Daily cumulative protein intake adequacy in mechanically ventilated children (N = 1245).

Abstract of Distinction 2079254 - Accuracy of Predicted Energy Expenditure in Critically Ill Children Katelyn Ariagno, RD, LDN

1; Craig D. Smallwood, BS, RRT

2; Enid E. Martinez, MD

3; Nilesh M. Mehta,

MD3

1Division of Gastroenterology and Nutrition, Boston Children’s Hospital, Boston, MA;

2Department of

Respiratory Care, Boston Children’s Hospital, Boston, MA; 3Division of Critical Care Medicine, Boston

Children’s Hospital, Boston, MA

Purpose: A variety of equations have been used to predict the energy expenditure associated with the dynamic metabolic response to pediatric critical illness. However, these equations may be inaccurate in this cohort, and may result in unintended underfeeding or overfeeding. A new series of equations (Oxford equations) have been developed using a data set that included 10, 552 basal metabolic rate (BMR) values in attempts to represent a more contemporary population. We aimed to examine the accuracy of 4 methods of predicting resting energy expenditure (REE) in critically ill children. Methods: Measured resting energy expenditure (MREE) was obtained by indirect calorimetry (IC) in a cohort of critically ill children under 18 years of age, admitted to the pediatric intensive care unit (PICU). Subjects were selected for testing based on institutional criteria. We used Bland Altman analyses to examine the agreement between MREE by IC and predicted REE by 4 methods; i) World Health Organization (WHO) equation, ii) Schofield-Height and Weight equation, iii) Recommended Dietary Allowances (RDA 1989), and iv) the Oxford equations. Spearman method was used to compare the predicted values to the MREE. For the purpose of this study, the accuracy of these REE prediction methods was deemed reasonable if the predicted REE value was within ±20% of MREE. Results: 120 tests (92 subjects) were performed in this study. Steady state was achieved in 87 tests in 69 mechanically ventilated children (n=39 male, median age: 1.8 years, and median weight: 10.4kg) selected by institutional criteria for IC, and the metabolic data from these tests were included in the analysis. The median (IQR) for PICU length of stay (LOS) at the time of the IC test was 60 (25-133) days. The 4 predictive methods were well correlated with MREE. (Table 19-1 shows correlation coefficient). The percent mean bias (95% limits) for agreement with the MREE for predicting methods were as follows; Schofield 1.3 (-89 to +91), WHO 11.6 (-66 to +89), RDA 64 (-3.7 to +132), and Oxford 14.4 (-60 to +88.8). Figure 1 shows the Bland Altman plots. The prediction methods in this study were reasonably accurate (predicted REE within ±20% of MREE) in only a minority of the cohort.


Conclusions: Commonly used pediatric predictive energy equations, including the recently developed Oxford equations were well correlated but not in agreement with the MREE by IC in critically ill children with longer ICU stay. Use of these equations would result in unintended energy imbalances in a majority of this cohort. RDA method results in significant overfeeding and should not be used in the PICU. The use of the other 3 equations could result in significant underfeeding and overfeeding in critically ill children. This is the first study evaluating the Oxford equations in critically ill children. IC measurements are desirable for accurate determination of REE, but IC testing is not universally available. An accurate predictive method for REE in this unique patient population remains elusive.


Table 1. Correlation Between Prediction Methods and MREE.

REE Method R* Median (IQR) EEE (kcals/day) Tests within ±20% of MREE

Schofield 0.82 739 (230-969) 34%

WHO 0.86 763 (325-1088) 39%

RDA 0.84 985 (615-1600) 10%

Oxford 0.86 718 (343-1061) 38%

Median (IQR) MREE: 492 (307-807) kcals/d *R = Spearman coefficient for correlation between prediction methods and MREE; P value


Abstract of Distinction 2076762 - The Relationship Between Premorbid Obesity, Weight Gain, and Eucaloric Feeding of Critically Ill Surgical Patients Eric J. Charles, MD; Robin T. Petroze, MD, MPH; Robert G. Sawyer, MD, FACS Department of Surgery, University of Virginia Health System, Charlottesville, VA

Purpose: A recently published randomized controlled trial found no statistically significant difference between critically ill surgical patients with eucaloric nutritional support goals compared with hypocaloric goals, in terms of infectious complications, ICU or hospital lengths of stay, and mortality. Not assessed initially was the impact of varying levels of nutritional support during critical illness on weight gain, or the impact of premorbid obesity on infectious outcomes. We hypothesized that eucaloric feeding of critically ill surgical patients would cause an anabolic response leading to unintentional weight gain, and that premorbid obesity would negatively affect infectious outcomes in patients with eucaloric nutritional support goals. Methods: An adjunctive analysis was performed of a single-institution randomized controlled trial. During the original trial, critically ill surgical and trauma patients were randomized to receive either the standard calculated daily caloric requirement of 25-30 kcal/kg/day (eucaloric) or 50% of that value (hypocaloric). Both arms of the study were allotted the same protein goal of 1.5 g/kg/day. Premorbid weight was compared with end-of-study weight to determine the effect of differing nutritional support goals on weight gain and outcomes. Univariate nonparametric regression analysis was completed to determine the impact of premorbid obesity on infectious outcomes. Results: There were no statistically significant differences between patients randomized to eucaloric nutritional support goals compared with hypocaloric goals, in terms of mean actual body weight (88.0 ± 3.4 kg and 97.9 ± 4.6 kg, respectively, p = 0.10), mean BMI (28.1 ± 0.9 kg/m^2 and 32.9 ± 2.0 kg/m^2, respectively, p = 0.18), and number of patients with BMI > 30 (11 and 19, respectively, p = 0.07). The mean calories received daily between eucaloric and hypocaloric patients was significantly different (1338 ± 92 kcal and 982 ± 61 kcal, respectively, p = 0.02), however the mean weight change over study length was noticeable different but did not reach statistical significance (5.0 ± 1.9 kg and 0.6 ± 2.0 kg, respectively, p = 0.15). The subset of patients with a BMI > 30 were further analyzed, again showing no statistically significant differences between those in either arm of the study in terms of mean actual body weight, mean BMI, percentage of patients at risk for refeeding syndrome at admission, or APACHE II score. When infectious outcomes were compared within the obese subset of patients, there were 17 infections in the eucaloric obese subset and 58 in the hypocaloric obese subset, but the mean number of infections per patient (1.5 ± 0.6 and 3.1 ± 1.2, respectively, p = 0.29) and the percentage of patients who acquired an infection (45.5 ± 5 % and 79.0 ± 15 %, respectively, p = 0.07) did not reach statistical significance. Conclusions: Considering that weight gain from fluid overload is independently associated with worse outcomes and increased organ dysfunction in critically ill patients, delivering eucaloric feeding does not appear to influence weight gain differently than patients with lower caloric intake goals. As well, obese patients appear to tolerate a wide range of caloric allocation without an appreciable effect on infectious outcomes.

Financial support received from the National Institutes of Health (NIH 5-T32-AI-078875-03) and NIH National Heart, Lung and Blood Institute (NIH/NHLBI-T32-HL007849).

2077188 - Indirect Calorimetry Can Prevent Underfeeding and Overfeeding in the Pediatric Intensive Care Unit Amy K. Schwartz, RD, CSP, CNSC

1; Heather E. Skillman, MS, RD, CSP, CNSC

1; Matthew A. Haemer,

MD, MPH1,2

; Peter M. Mourani, MD3

1Department of Clinical Nutrition, Children′s Hospital Colorado, Aurora, CO;

2Department of Pediatrics,

Section of Nutrition, University of Colorado School of Medicine, Aurora, CO; 3Department of Pediatrics,

Section of Critical Care, University of Colorado School of Medicine, Aurora, CO


Purpose: Energy requirements of critically ill pediatric patients are both highly variable and difficult to predict. Estimating energy needs based on standard equations can significantly underestimate or overestimate the resting energy expenditure, resulting in underfeeding or overfeeding. Imprecise feeding can lead to undesirable consequences, including increased length of stay. The Americ

CLINICAL NUTRITION WEEK 2015 ABSTRACTS February 14 17, … · 2016. 8. 24. · Results: PN caused...

Documents

Transcript of CLINICAL NUTRITION WEEK 2015 ABSTRACTS February 14 17, … · 2016. 8. 24. · Results: PN caused...