Harry M. Vars Award Candidate Abstracts

Monday, March 24, 2025

Premier Paper Session and Vars Award Competition

Harry M. Vars Award Candidate

Lizl Veldsman, RD, M Nutr, BSc Dietetics¹; Guy Richards, MD, PhD²; Daniel Nel, PhD³; Renée Blaauw, PhD, RD¹

¹Division of Human Nutrition, Department of Global Health, Faculty of Medicine & Health Sciences, Stellenbosch University, Cape Town, Western Cape; ²Department of Surgery, Division of Critical Care, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, Gauteng; ³Centre for Statistical Consultation, Department of Statistics and Actuarial Science, Stellenbosch University, Stellenbosch, Western Cape

Financial Support: This research study was funded by the Fresenius Kabi Jumpstart Research grant.

Background: Critical illness is associated with severe muscle wasting. A combination of protein supplementation and mobilization may attenuate muscle loss. We determined the effect of a combination of intravenous (IV) bolus amino acid (AA) supplementation and in-bed cycling, versus standard care only, on skeletal muscle mass during the first ICU week.

Methods: This randomized controlled trial (ClinicalTrials.gov NCT04099108) was undertaken in a predominantly trauma adult surgical ICU. Participants were randomly assigned into two groups, both of which received standard care nutrition (SCN) and mobilization. The intervention group received an additional IV AA bolus, starting on ICU day 3 on-average, along with a 45-minute in-bed cycling session for a mean of 6 days. Histological change in vastus lateralis (VL) myofiber cross-sectional area (CSA) and ultrasound rectus femoris (RF) muscle CSA was measured from pre-intervention (average ICU day 2) to post-intervention (average ICU day 8) and analyzed using mixed model ANOVA and post hoc least significant difference (LSD) testing. Secondarily, we assessed the change in the protein-to-DNA ratio over time. We further compared responses between participants with and without baseline organ dysfunction (defined as SOFA≥2, due to infection or trauma-related inflammation).

Results: Fifty critically ill patients (90% male, mean age 37±12 years, APACHE II score 13.5±6.6, SOFA score 4.5±3.2, BMI 24.8±4.0 kg/m², modified NUTRIC 2.2±1.8) were included. The control (n = 25) and intervention (n = 25) groups received, on average, 87.62±32.18 and 85.53±29.29 grams of protein per day (1.26±0.41 and 1.29±0.40g/kg/day, respectively) from SCN, and the intervention group an additional 30.43±5.62 grams of AA (0.37±0.06g/kg protein equivalents) from the supplement. There was significant loss of muscle mass over time in all participants (VL myofiber CSA 11.76%, p = 0.011; RF muscle CSA 13.01% p < 0.001). No significant difference was found between groups, but a more compatible trend for greater muscle loss in the control group versus the intervention group (p = 0.051 vs p = 0.067) (Table 1). Protein-to-DNA ratio rose significantly over time (p = 0.007) but was significant only in the intervention group (control 11.33%, p = 0.177; intervention 20.96%, p = 0.007) (Table 1). A mixed model ANOVA adjusting for baseline organ dysfunction revealed a significant study group effect (p = 0.012). Post-hoc LSD comparisons showed that in participants without baseline organ dysfunction, the control group had significantly greater VL myofiber CSA loss (pre: 3803.19 vs. post: 2845.60, p = 0.034), while no significant loss was observed in the intervention group (pre: 4580.19 vs. post: 4393.86, p = 0.787).

Conclusion: Significant muscle mass loss occurs during the first ICU week. While no significant difference was observed between the control and intervention groups, the latter showed a significant increase in the protein-to-DNA ratio, indicating potential benefit on muscle protein content. Additionally, our findings suggest that the intervention may be more effective in patients without baseline organ dysfunction. Future studies should extend the intervention beyond the first ICU week to evaluate whether a longer duration improves muscle CSA and enhances muscle preservation.

Table 1. Change in Vastus Lateralis Myofiber Cross-sectional Area (CSA), Rectus Femoris Muscle CSA and Protein-to-DNA ratio from Pre- to Post-Intervention.

Harry M. Vars Award Candidate

Gabriela de Oliveira Lemos, MD¹; Natasha Mendonça Machado, PhD²; Raquel Torrinhas, PhD³; Dan Linetzky Waitzberg, PhD³

¹University of Sao Paulo School of Medicine, Brasília, Distrito Federal; ²University of Sao Paulo School of Medicine, São Paulo; ³Faculty of Medicine of the University of São Paulo, São Paulo

Financial Support: This study is linked to project no. 2011/09612-3 and was funded by the Fundação de Amparo à Pesquisa do Estado de São Paulo (São Paulo State Research Support Foundation) (FAPESP).

Background: Sphingolipids (SLs) contribute to the structural integrity of eukaryotic cell membranes and serve as mediators in signaling pathways of cellular regulation. Accumulation of lipid molecules, including SLs, can lead to lipotoxicity, metabolic dysregulation, insulin resistance, and metabolic diseases. This study aimed to evaluate the relationship between plasmatic SLs and biochemical markers of glucose and cholesterol metabolism after Roux-en-Y gastric bypass (RYGB), along with body composition in a population of women with obesity and type 2 diabetes mellitus (T2DM). This investigation is part of the SURMetaGIT study, registered at www.clinicalTrials.gov (NCT01251016).

Methods: Thirty women with grade II-III obesity and T2DM who underwent RYGB were included. Evaluations were taken at baseline and 3 months after surgery. Insulin therapy patients were excluded. Biochemical (n=30) and metabolomic (n=28) analyses were performed on plasma samples. Body weight (kg), body composition (air displacement plethysmography), and anthropometric measures were assessed. SLs were identified through untargeted metabolomics analysis by liquid chromatography coupled to mass spectrometry. The AnalysisBaseFileConverter and MS-DIAL were applied for data processing and analysis. Statistics were performed in the Jamovi software (2.2.5) and MetaboAnalyst 5.0. Fold change was calculated by the log2 post-surgery mean/pre-surgery mean. Paired t-test was applied for group comparisons according to the normality test. Spearman test was used for correlation analysis. Results with an α error < 0.05 were considered significant.

Results: After RYGB, body weight, BMI, waist, and hip circumference significantly decreased (Figure1). Patients experienced a metabolic improvement in glucose metabolism: fasting plasma glucose (215.8 ±72.4 vs 104 ± 21.7 mg/dL), insulin (21.6 ± 14.7 vs 11.1±9.2), HbA1c (8.9 ± 1.6 vs 6.04 ± 0.5), and C-peptide (4.02 ± 1.3 vs 2.9 ± 0.8), respectively- p< .001). Total cholesterol reduced (192 ± 30.7 vs 149.5 ± 29.2 mg/dL, p< .001) at expenses of pro-atherogenic fractions: non-HDL-c (143 ± 35.5 vs 107 ± 42 mg/dL, p < .001), LDL-c (114 ± 29.5 vs 88.5 ± 30.2 mg/dL, p =.01), VLDL-c (29.5 ± 15 vs 21 ± 7 mg/dL, p =.002), TG (150 ± 85.2 vs 104.5 ± 36.7 mg/dL, p =.004). Figure 2 shows the heatmap of SLs before and after RYGB. Twenty-one of the 34 identified SLs changed significantly after surgery (Figure 3). Cer (d18:1/23:0) and Cer (d42:1) showed moderate negative correlation with anthropometric measures and body fat, but positive with lean mass. On the other hand, SM (d38:2) presented the opposite effect. A poor correlation was observed between glucose metabolism biochemical markers and SLs, however, these same variables presented a strong and robust correlation with cholesterol and its fractions, particularly after RYGB (Table 1).

Conclusion: Body composition and biochemical changes related to RYGB were associated with the SLs remodeling in plasma, and a strong and positive correlation of these lipids was observed with markers of cholesterol metabolism. Sphingolipids, especially SM, can hold a potential role in metabolic changes induced by RYGB.

Table 1. Correlation Analysis Between Plasmatic Sphingolipids And Biochemical Parameters Before And After Roux-en-Y Gastric Bypass.

Figure 1. Body Composition and Anthropometrics Data Before and After RYGB.

Figure 2. Heatmap of Plasmatic Sphingolipids Before and After RYGB.

Figure 3. Sphingolipids with Significant Remodeling After RYGB.

Harry M. Vars Award Candidate

Kaitlyn Daff, MA, RD, LDN¹; Mariana Lund, PhD¹; Peder Lund, PhD¹; Gail Cresci, PhD, RD, FASPEN²

¹Case Western Reserve University, Cleveland, OH; ²Cleveland Clinic Lerner Research Institute, Cleveland, OH

Financial Support: Grant funding provided by NIAAA/NIH- R01AA028043 to GAC.

Background: Chronic ethanol exposure disrupts intestinal homeostasis by increasing barrier permeability, dysregulating immune response, and inducing oxidative stress. Arginine is physiologically important in maintaining intestinal homeostasis as it can be metabolized into multiple biologically active metabolites, including L-citrulline, nitric oxide, and polyamines. L-citrulline is associated with intestinal inflammation and is used as a biomarker for intestinal function and disease. Here, we hypothesized that ethanol disrupts intestinal epithelial arginine metabolism, which may contribute to its impact on inflammation and oxidative stress.

Methods: HT-29 (ATCC) human colon adenocarcinoma cells have a small intestinal epithelial cell phenotype and were used to test the direct effects of ethanol on arginine metabolism. Cells were cultured using RPMI-1640 medium supplemented with 1% (v/v) penicillin-streptomycin and 10% (v/v) fetal bovine serum and seeded in duplicate at a cell density of 0.5 x 10⁶ cells per 6 well plate. Confluent cell monolayers were treated with a physiologically relevant concentration of ethanol (±40mM) for 24 hours prior to cell collection. To assess arginine metabolism, stable isotope tracing experiments were performed. Cells treated with ±40mM ethanol were co-incubated with 200µM ¹⁵N₄- L-Arginine for 24 hours prior to sample collection. Metabolites were extracted from cell pellets with 80% methanol, dried under nitrogen, and resuspended for LC-MS analysis. Liquid chromatography was performed on an Agilent 1290 Infinity II LC system paired with a TSQ Altis Plus Triple Quadrupole Mass Spectrometer (ThermoFisher). RNA was collected from biological duplicates of HT-29 cells analyzed for stable isotope tracing experiments, and mRNA expression of arginine-metabolizing enzymes was assessed by qRT-PCR. Four replicates of each experiment were performed, and a student t-test with p < 0.05 was calculated to determine statistically significant changes between treatments.

Results: The relative abundance of ¹⁵N₄-Arginine (M + 4) was decreased in ethanol-treated cells, suggesting a decreased overall uptake of exogenous arginine into intestinal cells (Relative area 11.5 ± 0.029). In contrast, the abundance of ¹⁵N₀-Arginine (M + 0) was increased from 72.6% (± 0.04) to 75.4% in ethanol-treated cells. Ethanol exposure increased the relative abundance of ¹⁵N₂-Citrulline (M + 2) and ameliorated the production of ¹⁵N₃-Citrulline (M + 3), suggesting a shift away from the linear pathway of citrulline synthesis by direct arginine metabolism and towards production by an alternate metabolic pathway. mRNA expression of the enzyme inducible nitric oxide synthase (iNOS), which directly metabolizes arginine into citrulline, and nitric oxide were significantly decreased in ethanol-treated intestinal cells (p = 0.013; n = 4).

Conclusion: Taken together, these data suggest that ethanol exposure impairs exogenous arginine uptake into HT-29 cells, shifting towards endogenous arginine synthesis as evidenced by the decreased relative abundance of ¹⁵N₄-Arginine and increased ¹⁵N₀-Arginine in ethanol-treated cells. Citrulline production also changed in the ethanol-treated cells, and a decrease in iNOS mRNA suggests this pathway may be involved. This metabolic pathway is essential for cell-redox homeostasis and suggests ethanol-induced disruptions to arginine metabolism as a potential driver of oxidative stress and inflammation. Future investigations will focus on understanding the metabolic fate of arginine during ethanol exposure and how this influences intestinal inflammation and oxidative stress.

Harry M. Vars Award Candidate

Mirielle Pauline, PhD, BSc¹; Caitlin Huynh²; Rohan Persad²; Pamela Wizzard, BSc, RAHT²; Patrick Nation, DVM²; Catherine Field, PhD²; Paul Wales, MD³; Justine Turner, PhD, MD²

¹University of Alberta, St. Albert, AB; ²University of Alberta, Edmonton, AB; ³Cincinnati Children's Hospital Medical Center, Cincinnati, OH

Financial Support: This work was completed with financial support from a commercial entity for Baxter International Healthcare Company.

Background: Infants with intestinal failure on total parenteral nutrition (TPN) rely on intravenous lipid emulsions (ILEs) to survive and grow. While there have been recent advances in the composition of these ILEs, none have been designed specifically for infants. This is particularly relevant to those born preterm with unique polyunsaturated fatty acid (PUFA) requirements. Currently available ILEs, including soy-based (SO-ILE) or mixed emulsion (SO, MCT, OO, FO-ILE), differ markedly in PUFA content, and may supply suboptimal amounts of arachidonic acid (AA), eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA). AA and DHA are both important for infant growth, cognitive development and immune function. Neither ILE contains choline, the dietary precursor for phosphatidylcholine (PC) and important in lipid transport and metabolism, cell membrane structure and neurotransmission. The role of choline for perinatal brain development is well established and choline deficiency could negatively impact end-organ delivery of AA and DHA. Furthermore, EPA can negatively impact PUFA tissue deposition, most notably for AA. A novel ILE, NOV-C, was designed with the unique needs of developing humans in mind, specifically increasing AA and DHA, avoiding high EPA, and adding choline, so as to optimize key end organ deposition into phospholipid (PL) of AA and DHA. In TPN fed neonatal piglets, we compared tissue fatty acid composition obtained with this new ILE versus the currently used ILEs for infants.

Methods: We compared SO-LE (n = 7), SO,MCT,OO,FO-ILE (n = 7), and NOV-C (n = 8) at an appropriate dose for TPN fed piglets (10g/kg/d). On day 14, serum and tissues (liver, brain, lung, retina, jejunum) were collected and the percentage fatty acids in total PL was determined using gas liquid chromatography. Comparisons were made to normative data from litter and age matched control piglets maintained with the sow, not receiving TPN, presented as a range (CON) (n = 8). Comparison between treatment groups used Kruskal-Wallis (significance p < 0.05), data is presented as median (interquartile range).

Results: Table 1. summarizes key fatty acids, in total PL. AA was increased for NOV-C compared to SO,MCT,OO,FO-ILE in serum (p = 0.003), liver (p < 0.001), lung (p < 0.001), retina (p = 0.008) and jejunum (p < 0.001). EPA was significantly increased for SO,MCT,OO,FO-ILE compared to SO-ILE and NOV-C in serum (p = 0.002), liver (p < 0.001), lung (p < 0.001), retina (p < 0.001) and jejunum (p < 0.001). DHA was significantly increased for NOV-C and SO,MCT,OO,FO-ILE compared to SO-LE in liver (p = 0.001) and jejunum (p < 0.001), and for NOV-C over the other emulsions in lung (p < 0.001) and retina (p = 0.004).

Conclusion: This study in neonatal piglets compares a novel ILE designed specifically for infants that prioritizes adding AA, DHA and choline, with low EPA. This new approach significantly increased levels of AA in serum along tissue deposition in PL of AA and DHA. This novel ILE avoids the high EPA/low AA deposition induced by currently available fish oil ILEs, which has the potential to negatively impact somatic growth, and organ and immune functions. Further analysis into specific phospholipid subclasses, notably PC (given the addition of choline to NOV-C) is underway and will be informative. Furthermore, investigation of the impact of this new ILE on organ and immune functions will be highly relevant going forward.

Table 1. Fatty Acid Composition of Serum, Liver and Brain Phospholipid.

Harry M. Vars Award Candidate

Daffne Baldwin, MNC¹; Alfredo Gutierrez, DR¹; Isabel Medina, PhD, RD²; Martha Marquez, DR¹

¹Instituto Nacional de Pediatría (National Institute of Pediatrics), Ciudad de México (Mexico City), Distrito Federal (Federal District); ²Metodología de la Investigación (Research Methodology), Instituto Nacional de Pediatría (National Institute of Pediatrics), México, Distrito Federal (Federal District)

Financial Support: None Reported.

Background: Loss of muscle mass, particularly in the quadriceps, has been identified as a potential prognostic indicator in critically ill adults. However, its significance in the pediatric ICU population is not completely understood. Understanding the changes in quadriceps femoris mass thickness and the implications of muscle mass depletion may facilitate early interventions to prevent it and reduce complications Objetives: To investigate whether changes in quadriceps thickness, measured by ultrasound, are predictively associated with energy intake, mortality risk, and length of stay (LOS) in a Pediatric Intensive Care Unit (PICU) in a pediatric tertiary level hospital.

Methods: This was a comparative, longitudinal, observational, prospective study. Ultrasound evaluations of quadriceps femoris thickness were conducted every 48 hours during the first week and weekly thereafter until discharge. Nutritional assessments were performed weekly until discharge, with daily dietary assessments. Upon discharge from the ICU, the depletion of accumulated muscle mass was evaluated, and muscle atrophy was determined if the percentage was greater than 10%. Reliability was evaluated using the intraclass correlation coefficient (ICC) (0.98, 95% CI 0.96-0.99). The project was approved by the ethics and research committees of the National Institute of Pediatrics (approval number 2024/009).

Results: The study included 29 patients (44% women) with an average age of 9 years 8 months. Of these, 68% were admitted for medical illness. Upon admission, 55% were well-nourished, 25% were malnourished, and 20% were overweight. Muscle atrophy was observed in 48% of patients by day 7 and in 51% at discharge. The mean length of stay was 16 days, with a higher average for patients with muscular atrophy compared to those without (14 days vs. 11 days, p = 0.35). A correlation was found between fasting days and muscular atrophy (p = 0.017). Figure 1. Mortality prevalence was 10%, and all deceased patients had developed muscular atrophy. Figure 2.

Conclusion: Children in the PICU developed a 51% decrease in quadriceps femoris thickness, particularly affecting malnourished and overweight children. Our findings suggest a correlation between fasting, mortality, and changes in quadriceps thickness measured by ultrasound during the PICU stay.

Table 1. Comparison Atrophy Vs No Atrophy.

Figure 1. Fasting and Change in Quadriceps Thickness.

Figure 2. Thickness of quadriceps femoris measured by POCUS.