Energy Expenditure and Nutrition After Pancreatic Surgery

Background and Rationale Patients undergoing pancreatic resection experience surgical stress, systemic inflammation, and frequent exocrine and endocrine insufficiency that together alter energy metabolism and place them at high risk of nutritional deterioration. Predictive equations and simplified weight-based targets are used routinely to prescribe nutrition when indirect calorimetry is unavailable, but their accuracy in this specific surgical population is poorly characterized, and the degree to which the route and adequacy of postoperative feeding influence recovery has not been examined alongside directly measured energy expenditure in the same patients. This study is designed to address both questions within a single prospective cohort, pairing calorimetry-based measurement of resting energy expenditure with structured documentation of nutritional delivery and clinical outcomes.

Design and Setting The study is a single-center prospective observational cohort study. Consecutive eligible adults undergoing total or partial pancreatectomy for pancreatic tumors are enrolled preoperatively after informed consent. Feeding route is not assigned by the protocol; it is determined case by case by the treating surgical and nutrition support teams according to clinical status and tolerance, and is recorded as an observed exposure. All measurements and data abstraction are performed by trained clinical dietitians and drawn from direct measurement and the medical record.

Indirect Calorimetry Procedure Measured resting energy expenditure (mREE) is determined using a Cosmed Q-NRG+ portable indirect calorimeter. The device is calibrated before each session per the manufacturer's automatic gas-calibration procedure. Measurements are obtained by canopy where feasible, with a face mask used only when a canopy cannot be applied, recording VO₂ and VCO₂. Patients are measured supine, awake, and motionless in the morning after an 8-10 hour fast, at a controlled room temperature of 22-24 °C, following a minimum 20-30 minutes of rest. The first five minutes of each recording are discarded, and steady state is defined as at least five consecutive minutes during which the coefficient of variation in VO₂ and VCO₂ does not exceed 10%. Caffeine and strenuous activity are avoided for at least 12 hours beforehand, and supplemental oxygen is suspended or accounted for where clinically feasible. mREE is derived using the abbreviated Weir equation, mREE = (3.94 × VO₂ + 1.11 × VCO₂) × 1.44, and the respiratory quotient is calculated as VCO₂/VO₂. To reduce the influence of single-timepoint sampling within a dynamic postoperative course, calorimetry is scheduled at more than one time point during the early recovery period where the patient's clinical stability permits, with the exact days recorded.

Estimation of Energy Requirements for Comparison Predicted resting energy expenditure is calculated for each patient using the Harris-Benedict and Schofield equations. Harris-Benedict is applied as: men, 66.47 + 13.75 × weight (kg) + 5.0 × height (cm) - 6.75 × age; women, 665.09 + 9.56 × weight (kg) + 1.84 × height (cm) - 4.67 × age. Schofield is applied by sex and age band. For overweight patients, ideal body weight is used in these equations; for obese patients, corrected weight [ideal body weight + 0.25 × (actual - ideal body weight)] is used. Total energy requirements are additionally estimated from the ESPEN weight-based targets of 25 and 30 kcal/kg of ideal body weight (adjusted using a BMI of 22 kg/m² for overweight and 24.9 kg/m² for obese patients) and compared with mREE adjusted by an activity/stress factor of 1.15. Because the two families of estimates target different quantities - basal/resting expenditure for the equations versus total requirements for the weight-based targets - equation comparisons use raw mREE while weight-based comparisons use mREE × 1.15, and this distinction is retained throughout analysis.

Nutritional Intake and Feeding-Route Documentation For each patient, the route of postoperative nutrition (enteral plus oral, parenteral only, or combined enteral and parenteral) is recorded, together with the formulation used, the daily administered volume, and the delivery rate. Daily energy and protein delivery are quantified and expressed relative to individually calculated requirements to derive an energy deficit and a protein deficit for the early postoperative period. Diet-related symptoms (including diarrhea and abdominal pain) are recorded prospectively. These data support the planned subgroup comparison of outcomes across feeding routes and across strata of energy and protein deficit.

Anthropometry and Body Composition Body weight is measured on a calibrated electronic scale to the nearest 0.1 kg. Height is measured with the patient standing upright, unshod and bare-headed, using the Frankfort horizontal plane for head alignment. BMI is calculated as weight (kg) divided by height squared (m²). Weight and BMI are recorded at the preoperative assessment, at discharge, and at the 3-6 month follow-up, and recent percentage weight change is documented. Because body weight in the early postoperative period is confounded by fluid shifts, third-spacing, and edema, preoperative weight is used as the reference exposure in modeling, and objective body composition assessment (bioelectrical impedance or CT-derived skeletal muscle measurement, where available) is incorporated to distinguish loss of fat mass from loss of lean mass.

Nutritional Risk Screening and Biochemical Markers Preoperative nutritional risk is screened using the Perioperative Nutrition Screen (PONS), which weights recent weight loss, reduced intake or anorexia, low BMI, and the presence of a high-metabolic-risk underlying disease. Serum albumin (as a surrogate marker of visceral protein status) and C-reactive protein (as a marker of systemic inflammation) are recorded as exploratory variables at defined time points around the calorimetry measurement, per the hospital's standard laboratory procedures. Clinical variables including comorbidities, tumor site and surgical procedure, postoperative complications, length of stay, and readmissions are abstracted from the medical record.

Follow-up Schedule Assessments are anchored to the perioperative pathway: a preoperative baseline (anthropometry, screening, biochemistry), early postoperative calorimetry and intake documentation during the index admission, a discharge assessment, and a follow-up contact at 3-6 months after discharge for anthropometry and outcome ascertainment. Complication, readmission, and mortality windows follow the postoperative course as specified in the outcome measures.

Statistical Analysis Plan Analyses will be conducted in Jamovi with two-tailed tests at α = 0.05. Distributional normality is assessed by the Shapiro-Wilk test; normally distributed continuous variables are summarized as mean (SD) and non-normal variables as median (interquartile range), with categorical variables as counts and percentages. Agreement between mREE and each estimation method is evaluated by Bland-Altman analysis, reporting mean bias, 95% limits of agreement, and tests for proportional bias, complemented by accuracy indices (percentage within ±10% and ±20%, mean absolute percentage error, and root mean square error) and by linear regression of estimates against mREE. Determinants of mREE are examined using multiple linear regression with prespecified adjustment sets (unadjusted; age and sex; and additionally CRP and albumin). Changes in weight, BMI, and laboratory markers across time points are analyzed with repeated-measures ANOVA or Friedman's test as appropriate, with pairwise comparisons. Where repeated calorimetry measurements are obtained, longitudinal trajectories are analyzed using linear mixed-effects models to accommodate within-patient correlation and variable measurement timing. Feeding route and deficit strata are compared for complications, readmission, and mortality. The single-center design and anticipated sample size are recognized as constraints on power and generalizability, and subgroup analyses (including by surgical type) are treated as exploratory.