One-Pager

ICU Nutrition and Refeeding

When to feed, first-week targets, PN timing, tube checks, residuals, diarrhea and refeeding syndrome

Reviewed by Alex G, BSN, RN, CEN, CCRN, CMC, NREMT-P · Checked against current sources October 2026

Download one-page PDFLetter landscape, ready to print or keep with your notes

When to feed and by which route

SituationPlan
Cannot eat, resuscitatedEN within 24–48 h (ASPEN/SCCM 2016, ESPEN 2023)
MAP <50, pressors starting or risingHold EN (ASPEN/SCCM 2016)
Shock controlled (MAP ≥60), pressors stable or fallingLow-dose EN, 10–20 mL/h (ESICM 2017); watch for bowel ischemia
Bowel ischemia, obstruction, ACS, active upper GI bleedDelay EN (ESPEN 2023)
Prone, ECMO, paralytics, absent bowel soundsNot reasons to delay EN (ESICM 2017)
EN impossiblePN days 3–7 (ESPEN), after 7 (ASPEN); sooner if malnourished
EN short of goal in week 1No supplemental PN before day 7 (ASPEN 2022); ESPEN: case by case

How much: week 1 vs later

PhaseEnergyProtein
First 7–10 days (ASPEN 2022)12–25 kcal/kg/day1.2–2.0 g/kg/day
Week 1 (ESPEN 2023)≤70% of expenditure early; up to 80–100% of measured after day 3Build to 1.3 g/kg/day
RecoveryMeasured expenditure, or 25–30 kcal/kg/day (ASPEN/SCCM 2016)1.2–2.0 g/kg; more in burns, trauma, CRRT

Trials behind the targets

TrialComparedResult
EPaNIC 2011, n 4,640Supplemental PN from day 8 vs day 1–2Late PN: fewer infections (22.8% vs 26.2%)
CALORIES 2014, n 2,388Early PN vs early EN30-day deaths 33.1% vs 34.2%
NUTRIREA-2 2018, n 2,410Full EN vs PN in shockSame deaths; EN bowel ischemia 2% vs <1%
NUTRIREA-3 2023, n 3,0446 vs 25 kcal/kg/day, less protein, week 1, shockSame deaths; fewer GI complications
EFFORT Protein 2023, n 1,329Protein ≥2.2 vs ≤1.2 g/kg/dayNo benefit; possible harm in AKI

Tubes and aspiration

Residuals, intolerance and diarrhea

Refeeding syndrome (ASPEN 2020)

Phos, K or Mg falls 10–20% (mild), 20–30% (moderate), or >30% or with organ dysfunction (severe) within 5 days of starting calories.

Risk factorModerate: 2 neededSignificant: 1 needed
BMI16–18.5<16
Weight loss5% in 1 month7.5% in 3 months, >10% in 6
Little or no intake5–6 days>7 days
Low K, phos, Mg before feedingMinimally lowModerately or severely low
Fat or muscle lossModerate fat, mild–moderate muscleSevere

Glucose and daily nursing checks

Practice ICU Nutrition and Refeeding questions free in your browser8 exam-style questions with rationales. No account or download needed.

In depth

The one-page summary above is what the PDF prints. These sections go further.

Metabolism of critical illness: why less is more early

Critical illness runs through an early acute phase (roughly days 1–2), a late acute phase (to about day 7) and then recovery or persistent critical illness (ESPEN 2019 and 2023). Cortisol, catecholamines, glucagon and cytokines drive gluconeogenesis and muscle breakdown while insulin resistance blocks glucose uptake. Endogenous energy production supplies 500–1,400 kcal/day that feeding does not switch off (ESPEN 2023), so a full prescription in the first days lands on top of energy the patient is already producing.

Overfeeding shows as hyperglycemia, extra CO₂ to clear, high triglycerides and fatty liver. Early full feeding may also blunt autophagy, the cell's recycling of damaged proteins: in an EPaNIC substudy of 600 patients (Hermans 2013), delaying supplemental PN to day 8 cut ICU-acquired weakness from 43% to 34%.

Muscle is the protein store being spent: rectus femoris cross-sectional area fell 17.7% by day 10 in 63 ventilated patients, faster with multiorgan failure, and muscle protein synthesis was depressed on day 1 (Puthucheary, JAMA 2013). Anabolic resistance may explain why extra early protein has not improved outcomes in trials; protein and mobilization likely matter more in recovery.

Enteral nutrients maintain the mucosa, tight junctions and gut immune tissue, the case for using the gut. Digestion also raises splanchnic oxygen demand: in low-flow shock a full feed can outstrip supply and cause nonocclusive bowel ischemia (NUTRIREA-2, below).

Screening, energy and protein targets

ESPEN 2023 counts every patient in the ICU beyond 48 h as at risk of malnutrition. NRS 2002 ≥5 marks high nutrition risk; the modified NUTRIC score (age, APACHE II, SOFA, comorbidities, hospital days before ICU) uses ≥5 of 9. Malnutrition is diagnosed from weight loss, low BMI or low muscle mass plus a cause (GLIM). Resuscitation fluid hides weight loss, so ask the family about intake and weight before admission.

Indirect calorimetry is the reference (ESPEN 2023); equations can miss by up to 60%, and ventilator VCO₂ × 8.19 does better than equations when calorimetry is unavailable. Full weight-based targets are 20–25 kcal/kg/day (ESPEN) or 25–30 (ASPEN/SCCM 2016). In the first 7–10 days ASPEN 2022 suggests 12–25 kcal/kg/day; ESPEN 2023 gives at most 70% of expenditure early, then up to 80–100% of measured expenditure after day 3. Protein: 1.2–2.0 g/kg/day (ASPEN/SCCM 2016, kept in 2022) or 1.3 g/kg/day reached progressively (ESPEN 2023).

Adjusted weight = ideal weight + 20–25% of the excess (ESPEN 2023).
GroupEnergyProtein
Obesity, BMI 30–50 (ASPEN/SCCM 2016)11–14 kcal/kg actual weight2.0 g/kg ideal weight; up to 2.5 if BMI ≥40
Obesity (ESPEN 2023)Calorimetry, or adjusted weight1.3 g/kg adjusted weight
CRRTCount citrate and dextrose1.5–2.0 g/kg; max 2.5 (ASPEN/SCCM 2016) or 1.7 (KDIGO 2012)
BurnsCalorimetry; Curreri overestimates1.5–2 g/kg (Datta 2022)
Alcohol-associated hepatitis35–40 kcal/kg1.2–1.5 g/kg (AASLD–EASL 2014)

Calories from drugs count (ASPEN 2022, ESPEN 2023). Propofol is a 10% lipid emulsion with 1.1 kcal and 0.1 g of fat per mL (label): 30 mcg/kg/min in a 70 kg patient runs 12.6 mL/h, about 330 kcal and 30 g of fat a day. Clevidipine carries 2.0 kcal/mL (label limit 1,000 mL per 24 h), D5W 170 kcal/L and CRRT citrate a few hundred kcal/day. ESPEN caps IV lipid, propofol included, at 1.5 g/kg/day, and the propofol label calls for less lipid in PN.

Enteral access: placement and verification

Large-bore sump tubes (14–18 Fr) decompress the stomach and can feed short term; small-bore 8–12 Fr tubes placed with a stylet suit longer feeding but collapse under suction, so aspirate is often hard to obtain and residual volumes through them are unreliable. Use the oral route with basilar skull or severe midface fractures. Post-pyloric tubes go in blind, with electromagnetic guidance, under fluoroscopy or by endoscopy; a gastrostomy or jejunostomy is considered when tube feeding will last beyond about 4–6 weeks.

A blind small-bore tube can pass into the trachea and lung, often without coughing in a sedated patient and despite a cuffed endotracheal tube. AACN 2016 asks for two or more bedside checks during insertion (signs of respiratory distress, capnography, aspirate pH, aspirate appearance), never the air-bolus or bubbling test, and a radiograph showing the whole course of every blindly placed tube, electromagnetically guided ones included, before first use.

Delivering feeds and daily nursing care

Standard polymeric formulas carry about 1.0–1.2 kcal/mL; energy-dense 1.5–2.0 kcal/mL formulas suit fluid restriction, though delivering more calories that way did not lower deaths (TARGET 2018). Peptide formulas are for malabsorption and mixed-fiber formulas for persistent diarrhea (ASPEN care pathway). Continuous infusion is preferred to bolus, with less diarrhea (RR 0.42, ESPEN 2023). Start at 10–20 mL/h (ESICM 2017) and reach goal progressively, not before 48–72 h (ESPEN 2023).

Procedures, imaging, extubation and tube problems routinely cut delivery below the order. Volume-based protocols (ASPEN care pathway) set a 24 h volume and let the nurse raise the rate after an interruption, up to a protocol maximum. Hang closed systems up to 24–48 h, open-system formula 8 h and reconstituted powder 4 h (ASPEN).

Give drugs one at a time with water flushes before and after, never into the feed bag; do not crush extended-release or enteric-coated tablets, and remember that sorbitol in liquid drugs adds to diarrhea. Phenytoin, levothyroxine, fluoroquinolones and warfarin interact with feeds; follow pharmacy timing.

Shock, prone position, ECMO and other special cases

ASPEN/SCCM 2016 withholds EN while MAP is below 50 mm Hg or catecholamines are being started or escalated, and allows cautious feeding as pressors come down. ASPEN's 2022 consensus calls a vasopressor no contraindication with careful monitoring: gastric EN within 48 h if MAP is 60 or more, held below 50, trophic-only or held at high vasopressor doses, and held for residuals over 300 mL if measured (low-quality evidence). ESICM 2017 and ESPEN 2023 delay EN only while shock is uncontrolled, then start low-dose EN once fluids and vasopressors have controlled it; for septic shock ESPEN suggests 20–50% of full nutrition at first. NUTRIREA-2 shows why full doses wait: isocaloric EN in ventilated shock caused more vomiting (34% vs 20%), bowel ischemia (2% vs <1%) and colonic pseudo-obstruction (1% vs <1%) than PN, with no fewer infections.

Suspect bowel ischemia with distension, pain, rising lactate or pressor need, new GI bleeding, ileus or large gastric output: stop the feed and call the team.

SituationApproach (ESPEN 2023; ESICM 2017)
Uncontrolled shock, hypoxemia, hypercapnia or acidosisDelay EN
Overt bowel ischemia, abdominal compartment syndromeDelay EN
Active upper GI bleedingStart once bleeding has stopped
High-output fistulaFeed distal to it, or PN
Prone, ECMO, paralysis, open abdomenEarly EN
Brain or cord injury, stroke, severe pancreatitis, GI surgeryEarly EN
Therapeutic hypothermiaLow dose; increase after rewarming
Intra-abdominal hypertension without ACSLow dose; cut back if pressure rises

Prone: ESICM 2017 advises against delaying EN for prone positioning. A protocol of 25° head-up tilt, faster advancement and erythromycin raised delivered volume from 774 to 1,170 mL/day without more vomiting or VAP (Reignier 2010). Units commonly pause feeds and vent the stomach before each turn (Prone card), then restart promptly.

Residuals, aspiration, intolerance and diarrhea

A residual mixes formula with saliva and gastric juice; the volume drawn back depends on tube tip, position and technique. REGANE (2010, 329 patients) raised the limit from 200 to 500 mL: more of the prescription was delivered in week 1 (88% vs 84%) with no more pneumonia. NUTRIREA-1 (2013, 452 patients) stopped measuring altogether against a 250 mL limit: VAP 16.7% vs 15.8%, more patients reached their calorie goal, and vomiting was more common.

ASPEN/SCCM 2016 therefore advises against routine residuals and, where they are still used, against holding feeds for less than 500 mL without other signs. ESPEN 2023 finds them useful while feeds start, delays EN above 500 mL/6 h and adds a prokinetic, and moves to post-pyloric feeding if large volumes persist.

Aspiration risk rises with an unprotected airway, ventilation, age over 70, reduced consciousness, neurologic deficits, reflux, supine position, poor oral care, low staffing, transport and bolus feeding (ESPEN 2023). Prokinetics (ESPEN 2023): IV erythromycin 100–250 mg three times daily first, metoclopramide 10 mg two or three times daily as the alternative, or both; benefit falls to a third by 72 h, so stop after 3 days. Erythromycin prolongs QT, interacts through CYP3A4 and is an antibiotic; metoclopramide carries a boxed warning for tardive dyskinesia.

Diarrhea is 3 or more loose or liquid stools a day with a stool weight over 200–250 g (or over 250 mL) a day (ESICM working group 2012). Keep feeding (ESICM 2017) while working through causes: laxatives and stool softeners, sorbitol or magnesium in liquid drugs, antibiotics, C. difficile (test for ≥3 unformed stools in 24 h without laxatives in 48 h; IDSA/SHEA 2017), bolus or contaminated feeds, malabsorption, ischemia and overflow around impaction. If it persists, try a mixed-fiber formula, then a peptide formula (ASPEN). Replace K, Mg and bicarbonate losses and protect the skin.

No stool for 3 or more days without obstruction is lower GI paralysis (ESICM working group 2012); start a bowel regimen early with opioids.

Parenteral nutrition and glucose control

ASPEN/SCCM 2016 withholds exclusive PN for 7 days in low-risk patients, starts it as soon as possible in high-risk or severely malnourished ones, and considers supplemental PN after 7–10 days if EN meets under 60% of goals. ASPEN 2022 recommends against supplemental PN before day 7 (strong; the trials excluded malnourished patients) and accepts PN or EN when calories match. ESPEN 2023 starts PN within 3–7 days when EN is contraindicated, allows early progressive PN in severe malnutrition, and weighs supplemental PN case by case in week 1 after every EN strategy has been tried.

Dextrose (3.4 kcal/g) should not exceed 5 mg/kg/min and lipid 1.5 g/kg/day including propofol (ESPEN 2023); amino acids give 4 kcal/g. ASPEN 2022 accepts soybean or mixed-oil emulsions; ESPEN prefers mixed emulsions and allows fish oil 0.1–0.2 g/kg/day. Vitamins and trace elements go in daily because PN contains none (ESPEN 2023); CRRT for more than 2 weeks can cause copper deficiency.

SCCM 2024 starts insulin for persistent glucose ≥180 mg/dL, advises against titrating to 80–139 rather than 140–200, prefers an IV infusion in the acute phase, and checks at least hourly while glucose is unstable; ESPEN 2023 gives insulin above 180 mg/dL. NICE-SUGAR (6,104 patients) found more 90-day deaths with 81–108 mg/dL (27.5% vs 24.9%). In shock, use arterial or venous samples on a blood gas or lab analyzer. When feeds stop with insulin running, cut or stop the insulin per protocol or give dextrose.

Refeeding syndrome

Starvation empties cells of phosphate, potassium, magnesium and thiamine while serum levels can stay normal. Carbohydrate from any source, IV dextrose included, raises insulin, which pulls phosphate into glycolysis and potassium in through Na/K-ATPase, so levels can fall within hours (ASPEN 2020). Low phosphate means less ATP and 2,3-DPG: diaphragm weakness and failure to wean, poor contractility, arrhythmias, hemolysis and seizures. Thiamine is a cofactor of carbohydrate metabolism; deficiency causes Wernicke encephalopathy, lactic acidosis and wet beriberi. ASPEN 2020 doubts volume overload is part of the syndrome, though edema can follow.

ESPEN 2023 calls phosphate under 0.65 mmol/L (about 2.0 mg/dL), or a fall of more than 0.16 mmol/L (about 0.5 mg/dL), refeeding hypophosphatemia: measure electrolytes 2–3 times daily and restrict energy for 48 h. The rule comes from Doig 2015 (339 patients, 13 ICUs): 20 kcal/h for at least 2 days, then 40 and 60 kcal/h and 80% of goal before full feeding. Days alive after ICU discharge did not differ significantly, but 60-day survival was 91% vs 78%.

ASPEN 2020 also lists higher-risk conditions: alcohol or drug use disorder, eating disorders, cancer, malabsorption, bariatric surgery, dysphagia, protracted vomiting, food insecurity, and prolonged fasting or major surgery without nutrition. NICE is more cautious: at most 10 kcal/kg/day, and 5 kcal/kg/day with BMI under 14 or negligible intake for over 15 days.

The evidence and what is unsettled

Results follow the order of the comparison.
TrialComparisonResult
EPaNIC 2011 (n 4,640)Supplemental PN from day 8 vs within 48 hLate PN: infections 22.8% vs 26.2%; faster discharge
Early PN 2013 (n 1,372)PN vs standard care when EN not possible60-day deaths 21.5% vs 22.8%; less muscle wasting
CALORIES 2014 (n 2,388)Early PN vs early EN for 5 days30-day deaths 33.1% vs 34.2%; less vomiting on PN
NUTRIREA-2 2018 (n 2,410)Full EN vs PN in ventilated shock28-day deaths 37% vs 35%; more GI harm on EN
EDEN 2012 (n 1,000)Trophic vs full EN for 6 days, lung injuryVentilator-free days 14.9 vs 15.0; deaths 23.2% vs 22.2%
PermiT 2015 (n 894)40–60% vs 70–100% of calories, equal protein90-day deaths 27.2% vs 28.9%
NUTRIREA-3 2023 (n 3,044)6 vs 25 kcal/kg/day in week 1, shock90-day deaths 41.3% vs 42.8%; ICU-ready a day sooner
EFFORT Protein 2023 (n 1,329)≥2.2 vs ≤1.2 g/kg/day protein60-day deaths 34.6% vs 32.1%; harm signal in AKI
PRECISe 2024 (n 935)2.0 vs 1.3 g/kg/day protein, blindedWorse quality of life; more GI intolerance
NUTRIREA-1 2013 (n 452)No residual checks vs a 250 mL limitVAP 16.7% vs 15.8%; more fed to goal
Refeeding 2015 (n 339)Calorie restriction vs usual feeding60-day survival 91% vs 78%

What changed practice: less feeding in week 1, no routine residuals, a week's wait for supplemental PN, and dose mattering more than route. Still open: protein dose and timing after week 1, whether calorimetry-guided feeding helps, how to feed the malnourished patients most trials excluded, the pressor dose too high for trophic feeds, and whether nutrition plus exercise restores function after the ICU.

Exam pearls

Practice free in your browser3,800+ exam-style questions with rationales. No account, no ads. Also free on the App Store and Google Play.

Related one-pagers

Calculators for this topic

Sources

  1. Compher C, Bingham AL, McCall M, et al. Guidelines for the provision of nutrition support therapy in the adult critically ill patient: The American Society for Parenteral and Enteral Nutrition. JPEN J Parenter Enteral Nutr. 2022;46(1):12-41. Link
  2. American Society for Parenteral and Enteral Nutrition. Guidelines for the Provision of Nutrition Support Therapy in the Adult Critically Ill Patient: guideline questions, evidence grades, recommendations and clinical applications (factsheet). 2021. Link
  3. McClave SA, Taylor BE, Martindale RG, et al. Guidelines for the Provision and Assessment of Nutrition Support Therapy in the Adult Critically Ill Patient: Society of Critical Care Medicine (SCCM) and American Society for Parenteral and Enteral Nutrition (A.S.P.E.N.). JPEN J Parenter Enteral Nutr. 2016;40(2):159-211. Link
  4. American Society for Parenteral and Enteral Nutrition. Enteral Nutrition Care Pathway for Critically Ill Adult Patients. 2024. Link
  5. American Society for Parenteral and Enteral Nutrition. Indications for Enteral Nutrition in Patients with Critical Illness: ASPEN Recommendations (practice tool based on Bechtold ML, et al. When is enteral nutrition indicated? JPEN J Parenter Enteral Nutr. 2022;46(7):1470-1496). 2023. Link
  6. Singer P, Reintam Blaser A, Berger MM, et al. ESPEN practical and partially revised guideline: Clinical nutrition in the intensive care unit. Clin Nutr. 2023;42(9):1671-1689. Link
  7. Reintam Blaser A, Starkopf J, Alhazzani W, et al. Early enteral nutrition in critically ill patients: ESICM clinical practice guidelines. Intensive Care Med. 2017;43(3):380-398. Link
  8. Reintam Blaser A, Malbrain MLNG, Starkopf J, et al. Gastrointestinal function in intensive care patients: terminology, definitions and management. Recommendations of the ESICM Working Group on Abdominal Problems. Intensive Care Med. 2012;38(3):384-394. Link
  9. da Silva JSV, Seres DS, Sabino K, et al. ASPEN Consensus Recommendations for Refeeding Syndrome. Nutr Clin Pract. 2020;35(2):178-195. Link
  10. Worthington P, Gura KM, Kraft MD, et al. Update on the use of filters for parenteral nutrition: an ASPEN position paper. Nutr Clin Pract. 2021;36(1):29-39. Link
  11. American Association of Critical-Care Nurses. AACN Practice Alert: Initial and Ongoing Verification of Feeding Tube Placement in Adults (newsroom summary). September 2016. Link
  12. Honarmand K, Sirimaturos M, Hirshberg EL, et al. Society of Critical Care Medicine Guidelines on Glycemic Control for Critically Ill Children and Adults 2024. Crit Care Med. 2024;52(4):e161-e181. Link
  13. McDonald LC, Gerding DN, Johnson S, et al. Clinical Practice Guidelines for Clostridium difficile Infection in Adults and Children: 2017 Update by IDSA and SHEA. Clin Infect Dis. 2018;66(7):e1-e48. Link
  14. Casaer MP, Mesotten D, Hermans G, et al. Early versus late parenteral nutrition in critically ill adults (EPaNIC). N Engl J Med. 2011;365:506-517. Link
  15. Harvey SE, Parrott F, Harrison DA, et al. Trial of the route of early nutritional support in critically ill adults (CALORIES). N Engl J Med. 2014;371:1673-1684. Link
  16. Reignier J, Boisramé-Helms J, Brisard L, et al. Enteral versus parenteral early nutrition in ventilated adults with shock (NUTRIREA-2). Lancet. 2018;391(10116):133-143. Link
  17. Reignier J, Plantefeve G, Mira JP, et al. Low versus standard calorie and protein feeding in ventilated adults with shock (NUTRIREA-3). Lancet Respir Med. 2023;11(7):602-612. Link
  18. Heyland DK, Patel J, Compher C, et al. The effect of higher protein dosing in critically ill patients with high nutritional risk (EFFORT Protein). Lancet. 2023;401(10376):568-576. Link
  19. Bels JLM, Thiessen S, van Gassel RJJ, et al. Effect of high versus standard protein provision on functional recovery in people with critical illness (PRECISe). Lancet. 2024;404(10453):659-669. Link
  20. National Heart, Lung, and Blood Institute ARDS Clinical Trials Network; Rice TW, Wheeler AP, et al. Initial trophic vs full enteral feeding in patients with acute lung injury: the EDEN randomized trial. JAMA. 2012;307(8):795-803. Link
  21. Arabi YM, Aldawood AS, Haddad SH, et al. Permissive underfeeding or standard enteral feeding in critically ill adults (PermiT). N Engl J Med. 2015;372:2398-2408. Link
  22. TARGET Investigators, for the ANZICS Clinical Trials Group. Energy-dense versus routine enteral nutrition in the critically ill. N Engl J Med. 2018;379:1823-1834. Link
  23. Reignier J, Mercier E, Le Gouge A, et al. Effect of not monitoring residual gastric volume on risk of ventilator-associated pneumonia in adults receiving mechanical ventilation and early enteral feeding (NUTRIREA-1). JAMA. 2013;309(3):249-256. Link
  24. Montejo JC, Miñambres E, Bordejé L, et al. Gastric residual volume during enteral nutrition in ICU patients: the REGANE study. Intensive Care Med. 2010;36:1386-1393. Link
  25. Doig GS, Simpson F, Heighes PT, et al. Restricted versus continued standard caloric intake during the management of refeeding syndrome in critically ill adults. Lancet Respir Med. 2015;3(12):943-952. Link
  26. DIPRIVAN (propofol) injectable emulsion prescribing information. Fresenius Kabi; DailyMed, revised January 2025. Link
  27. CLEVIPREX (clevidipine) injectable emulsion prescribing information. Chiesi USA; DailyMed, revised September 2022. Link