Artificial Neural Network Algorithms to Predict Resting Energy Expenditure in Critically Ill Children.
Ontology highlight
ABSTRACT: Introduction
Accurate assessment of resting energy expenditure (REE) can guide optimal nutritional prescription in critically ill children. Indirect calorimetry (IC) is the gold standard for REE measurement, but its use is limited. Alternatively, REE estimates by predictive equations/formulae are often inaccurate. Recently, predicting REE with artificial neural networks (ANN) was found to be accurate in healthy children. We aimed to investigate the role of ANN in predicting REE in critically ill children and to compare the accuracy with common equations/formulae.Study methods
We enrolled 257 critically ill children. Nutritional status/vital signs/biochemical values were recorded. We used IC to measure REE. Commonly employed equations/formulae and the VCO2-based Mehta equation were estimated. ANN analysis to predict REE was conducted, employing the TWIST system.Results
ANN considered demographic/anthropometric data to model REE. The predictive model was good (accuracy 75.6%; R2 = 0.71) but not better than Talbot tables for weight. After adding vital signs/biochemical values, the model became superior to all equations/formulae (accuracy 82.3%, R2 = 0.80) and comparable to the Mehta equation. Including IC-measured VCO2 increased the accuracy to 89.6%, superior to the Mehta equation.Conclusions
We described the accuracy of REE prediction using models that include demographic/anthropometric/clinical/metabolic variables. ANN may represent a reliable option for REE estimation, overcoming the inaccuracies of traditional predictive equations/formulae.
SUBMITTER: Spolidoro GCI
PROVIDER: S-EPMC8618974 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description: Not available
| S-EPMC7521195 | biostudies-literature
Project description:BackgroundTotal energy expenditure (TEE) is the total energy expended by an individual to sustain life, activities, and growth. TEE is formed by four components: resting energy expenditure (REE), activity energy expenditure (AEE), growth-related energy expenditure (GEE), and the thermic effect of feeding (TEF). Some energy expenditure (EE) components may change throughout childhood and cannot be reliably estimated using prediction formulae.ObjectiveTo summarize measured TEE components as reported in the literature in healthy and critically ill children.MethodsWe searched MEDLINE, EMBASE, and CINAHL for studies published between 1946 and 7 September 2023. The primary outcome was energy expenditure. Included studies were published in English and measured one or more of TEE, AEE, GEE, and TEF with Indirect Calorimetry or Doubly Labeled Water in participants between 1 month and 18 years of age. We excluded studies reporting only REE or using predictive equations. Following abstraction, reported values were converted into kcal/kg/day or kcal/day as possible. Weighted mean values were calculated using median or means of EE measurements.ResultsWe found 138 studies, 8163 patients, and 16,636 eligible measurements. The median (IQR) study sample size was 20 (12, 35) patients. TEE was the most evaluated component. The median (IQR) TEE in infants was 73.1 (67.0, 76.5), in children 78.0 (66.0, 81.3), and in adolescents was 44.2 (41.8, 51.9) kcal/kg/day. Very few studies reported on GEE and TEF.ConclusionsThis is one of the first studies that summarizes components of total energy expenditure in different pediatric age groups in healthy and critically ill children. Growth- and feeding-associated energy expenditure are poorly reported in healthy children, while all components of TEE (except REE) are poorly reported in critically ill children.
| S-EPMC11357425 | biostudies-literature
Project description:BackgroundData on the influence of age and body mass index (BMI) on energy metabolism of the critically ill are heterogeneous. Due to the increasingly aging critically ill population, investigation on age- and BMI-specific energy metabolism is relevant.MethodsA total of 394 indirect calorimetry measurements were conducted on 348 critically ill adult medical patients, including 46 repeat measurements after 3.6 ± 4.3 days. Measured resting energy expenditure (MREE) was compared for age groups, BMI, and gender. Predicted energy expenditure (PEE) using the Penn State, Swinamer, and Ireton-Jones equations and the ACCP recommendations was also compared with MREE.ResultsThe patients were 65.6 ± 14.5 years old. Their mean Acute Physiology and Chronic Health Evaluation II score was 27.6 ± 7.8. Mean BMI was 27.8 ± 8.4 kg/m2, and 25.6% were obese. MREE adjusted for ideal body weight decreased with increasing age, while it increased with increasing BMI. Age, BMI, and gender are independent determinants of MREE after adjusting for clinical factors (R2 = 0.34). All four prediction equations showed a proportional bias, with the Penn State equation performing acceptably. In 46 patients with repeat indirect calorimetry, there was no significant difference between the first and second MREE (p = 0.62).ConclusionsAge, BMI, and gender are independent determinants of resting energy expenditure in critically ill adults. Variations between measured and predicted energy expenditure are considerable. Should prediction equations be used, their performance in the specific population should be taken into consideration. Repeat indirect calorimetry may not always be necessary. However, this may depend on the length of stay and the extent of stress.
| S-EPMC7493296 | biostudies-literature
Project description:Background: Asthma physiology affects respiratory function and inflammation, factors that may contribute to elevated resting energy expenditure (REE) and altered body composition. Objective: We hypothesized that asthma would present with elevated REE compared to weight-matched healthy controls. Methods: Adults with asthma (n = 41) and healthy controls (n = 20) underwent indirect calorimetry to measure REE, dual-energy X-ray absorptiometry (DEXA) to measure body composition, and 3-day diet records. Clinical assessments included spirometry, fractional exhaled nitric oxide (FENO), and a complete blood count. Results: Asthmatics had greater REE than controls amounting to an increase of ~100 kcals/day, even though body mass index (BMI) and body composition were similar between groups. Inclusion of asthma status and FENO in validated REE prediction equations led to improved estimates. Further, asthmatics had higher white blood cell (control vs. asthma (mean ± SD): 4.7 ± 1.1 vs. 5.9 ± 1.6, p < 0.01) and neutrophil (2.8 ± 0.9 vs. 3.6 ± 1.4, p = 0.02) counts that correlated with REE (both p < 0.01). Interestingly, despite higher REE, asthmatics reported consuming fewer calories (25.1 ± 7.5 vs. 20.3 ± 6.0 kcals/kg/day, p < 0.01) and carbohydrates than controls. Conclusion: REE is elevated in adults with mild asthma, suggesting there is an association between REE and the pathophysiology of asthma.
| S-EPMC8064324 | biostudies-literature
Project description:There is emerging evidence that circadian misalignment may alter energy expenditure, leading to obesity risk among those with irregular schedules [1-5]. It has been reported that energy expenditure is affected by the timing of sleep, exercise, and meals [6]. However, it is unclear whether the circadian system also modulates energy expenditure, independent of behavioral state and food intake. Here, we used a forced desynchrony protocol to examine whether fasted resting energy expenditure (REE) varies with circadian phase in seven participants. This protocol allowed us to uncouple sleep-wake and activity-related effects from the endogenous circadian rhythm, demonstrating that REE varies by circadian phase. REE is lowest at circadian phase ?0°, corresponding to the endogenous core body temperature (CBT) nadir in the late biological night, and highest at circadian phase ?180° in the biological afternoon and evening. Furthermore, we found that respiratory quotient (RQ), reflecting macronutrient utilization, also varies by circadian phase. RQ is lowest at circadian phase ?240° and highest at circadian phase ?60°, which corresponds to biological morning. This is the first characterization of a circadian profile in fasted resting energy expenditure and fasted respiratory quotient (with rhythmic profiles in both carbohydrate and lipid oxidation), decoupled from effects of activity, sleep-wake cycle, and diet in humans. The rhythm in energy expenditure and macronutrient metabolism may contribute to greater weight gain in shift workers and others with irregular schedules.
| S-EPMC6300153 | biostudies-literature
Project description:It is controversial whether weight loss reduces resting energy expenditure (REE) to a different magnitude in black and white women. This aim of this study was to determine whether changes in REE with weight loss were different between black and white postmenopausal women, and whether changes in body composition (including regional lean and fat mass) were associated with REE changes within each race. Black (n = 26) and white (n = 65) women (age = 58.2 +/- 5.4 years, 25 < BMI < 40 kg/m(2)) completed a 20-week weight-loss intervention. Body weight, lean and fat mass (total body, limb, and trunk) via dual-energy X-ray absorptiometry, and REE via indirect calorimetry were measured before and after the intervention. We found that baseline REE positively correlated with body weight, lean and fat mass (total, limb, and trunk) in white women only (P < 0.05 for all). The intervention decreased absolute REE in both races similarly (1,279 +/- 162 to 1,204 +/- 169 kcal/day in blacks; 1,315 +/- 200 to 1,209 +/- 185 kcal/day in whites). REE remained decreased after adjusting for changes in total or limb lean mass in black (1,302-1,182 kcal/day, P = 0.043; 1,298-1,144 kcal/day, P = 0.006, respectively), but not in white, women. Changes in REE correlated with changes in body weight (partial r = 0.277) and fat mass (partial r = 0.295, 0.275, and 0.254 for total, limb, and trunk, respectively; P < 0.05) independent of baseline REE in white women. Therefore, with weight loss, REE decreased in proportion to the amount of fat and lean mass lost in white, but not black, women.
| S-EPMC2798012 | biostudies-literature
Project description:Pseudohypoparathyroidism type 1A (PHP1A) is caused by loss-of-function mutations on the maternally inherited GNAS allele and is associated with early-onset obesity, neurocognitive defects, and resistance to multiple hormones. The role of energy intake vs central regulation of energy expenditure in the pathophysiology of obesity remains unclear.The aim of this study was to evaluate resting energy expenditure (REE) in participants with PHP1A.We assessed REE, biochemical, endocrine, and auxological status of 12 participants with PHP1A who had normal or elevated body mass index; controls were a cohort of 156 obese participants.This study took place at Children's Hospital in Philadelphia and Sick Children's Hospital in Toronto.REE as a percent of predicted REE was the outcome measure.PHP1A participants had normal endocrine status while receiving appropriate hormone replacement therapy, but had significantly decreased REE as a percent of predicted REE (using the modified Schofield equation).Our results are consistent with REE being the principal cause of obesity in PHP1A rather than it being caused by excessive energy intake or endocrine dysfunction.
| S-EPMC4803160 | biostudies-literature
Project description:PurposeThe purpose of this study was: 1) To compare Resting energy expenditure (REE) in adult males with Becker's Muscular Dystrophy (BeMD, n = 21, 39 ±12 years) and healthy controls (CTRL, n = 12, 37 ±12 years) 2) Determine whether other physiological parameters correlate with REE in BeMD, and 3) Compare current prediction methods of REE with measured REE.MethodsREE was calculated via indirect calorimetry using continuous, expired gas analysis following an overnight fast. Fat free mass (FFM) and fat mass were measured by bioelectrical impedance. B-mode ultrasound measured Tibialis Anterior (TA) and Gastrocnemius Medialis (GM) anatomical cross sectional area (ACSA). The Bone Specific Physical Activity Questionnaire measured physical activity.ResultsNo difference in REE was found between CTRL and BeMD groups (1913 ±203 & 1786 ±324 Kcal respectively). Other physiological comparisons showed increased fat mass (+54%), decreased TA ACSA (-42%), increased GM ACSA (+25%) as well as reduced respiratory function (FVC -28%; FEV1-27%) in BeMD adults compared to controls. REE estimated from prediction equations (Schofield's) in Muscular Dystrophy were different from measured REE (P<0.05, bias = -728kcal), while the Mifflin equation was no different from measured REE (r2 = 0.58, Bias = -8kcal). Within the present BeMD, REE predicted from FFM (REE = FFM x 34.57-270; r2 = 0.85) and body mass (REE = BM x 15.65 + 421.5; r2 = 0.66), were not different from measured REE (bias equals 0 and 0.2kcals, respectively).ConclusionsDespite no differences in REE between CTRL and BeMD adults, increased fat masses highlights the requirement for explicit nutritional guidelines, as well as maintenance of physical activity levels, where possible. Prediction equations are frequently used in clinical settings, however these have been shown to be less accurate in BeMD; therefore, the equations proposed here should be used where possible.
| S-EPMC5218559 | biostudies-literature
Project description:Body mass in humans and animals is strongly associated with the rate of heat production as defined by resting energy expenditure (REE). Beginning with the ancient Greeks up to the present time, philosophers and scientists have endeavored to understand the nature and sources of bodily heat. Today we recognize that body mass consists of organs and tissues, each of which produces a specified amount of heat at rest. An individual organ's REE can now be estimated in vivo as the product of its assumed mass-specific metabolic rate and its imaging-derived mass; whole-body REE reflects the sum of organ and tissue metabolic rates. The sizes of organs and total body mass in adults are governed by two main factors, a person's stature or height, and their level of adiposity. With greater body size, as represented by adult height independent of adiposity, organs remain stable or increase in mass according to distinct "scaling" patterns. Similarly, with greater relative adiposity organs adaptively accommodate to the increase in imposed mechanical and metabolic loading conditions. Through a detailed analysis of these stature and adiposity effects, we show how classical statistical REE prediction models can be mechanistically understood at the anatomic body composition level.
| S-EPMC6410366 | biostudies-literature
Project description:BACKGROUND:Sentinel lymph node biopsy (SLNB) is standard staging procedure for nodal status in breast cancer, but lacks therapeutic benefit for patients with benign sentinel nodes. For patients with positive sentinel nodes, individualized surgical strategies are applied depending on the extent of nodal involvement. Preoperative prediction of nodal status is thus important for individualizing axillary surgery avoiding unnecessary surgery. We aimed to predict nodal status in clinically node-negative breast cancer and identify candidates for SLNB omission by including patient-related and pathological characteristics into artificial neural network (ANN) models. METHODS:Patients with primary breast cancer were consecutively included between January 1, 2009 and December 31, 2012 in a prospectively maintained pathology database. Clinical- and radiological data were extracted from patient's files and only clinically node-negative patients constituted the final study cohort. ANN-based models for nodal prediction were constructed including 15 risk variables for nodal status. Area under the receiver operating characteristic curve (AUC) and Hosmer-Lemeshow goodness-of-fit test (HL) were used to assess performance and calibration of three predictive ANN-based models for no lymph node metastasis (N0), metastases in 1-3 lymph nodes (N1) and metastases in ≥ 4 lymph nodes (N2). Linear regression models for nodal prediction were calculated for comparison. RESULTS:Eight hundred patients (N0, n = 514; N1, n = 232; N2, n = 54) were included. Internally validated AUCs for N0 versus N+ was 0.740 (95% CI = 0.723-0.758); median HL was 9.869 (P = 0.274), for N1 versus N0, 0.705 (95% CI = 0.686-0.724; median HL: 7.421; P = 0.492) and for N2 versus N0 and N1, 0.747 (95% CI = 0.728-0.765; median HL: 9.220; P = 0.324). Tumor size and vascular invasion were top-ranked predictors of all three end-points, followed by estrogen receptor status and lobular cancer for prediction of N2. For each end-point, ANN models showed better discriminatory performance than multivariable logistic regression models. Accepting a false negative rate (FNR) of 10% for predicting N0 by the ANN model, SLNB could have been abstained in 27.25% of patients with clinically node-negative axilla. CONCLUSIONS:In this retrospective study, ANN showed promising result as decision-supporting tools for estimating nodal disease. If prospectively validated, patients least likely to have nodal metastasis could be spared SLNB using predictive models. TRIAL REGISTRATION:Registered in the ISRCTN registry with study ID ISRCTN14341750 . Date of registration 23/11/2018. Retrospectively registered.
| S-EPMC6588854 | biostudies-other