Nutrition

Requirement for nutritional supplementation in the critically ill depends on where a patient is in their phase of illness:

Key definitions:

  • Hypocaloric nutrition
    Supplemental non-protein nutrition supplementation lower than daily expenditure, resulting in a calorie deficit.
  • Trophic feeding
    Feeding <500kcal/day.
  • Permissive underfeeding
    Feeding 40-60% of daily required calories.

We define these periods by days, but it is likely that time is a proxy for another physiological process. It is reasonable to think that a recovering patient who deteriorates later in their admission (e.g. ICU-acquired sepsis) will return to the acute phase of illness, and that feeding should be adjusted accordingly.

Nutrition Assessment

A subjective global nutrition assessment includes:

  • Premorbid weight and loss
  • Premorbid nutrition
  • Muscle mass
  • Disease affecting GI function
    • Food intolerances
    • GI symptoms
    • Stomas
    • Liver disease
      Ascites.
  • Disease affecting metabolic rate
    • Hypothyroidism
    • Hyperthyroidism
    • Hypoadrenalism
    • Cushing disease

Determining nutrition requirements:

The Reverse Fick Method:

\[VO_2 = (CO \times C_a) - (CO \times C_v)\]

Where:

  • VO2: Oxygen consumption (mL/min)
  • CO: Cardiac Output (L/min)
  • Ca: Arterial Oxygen Content (mL/L)
  • Cv: Venous Oxygen Content (mL/L)
  • Calorie
    Can be provided as either carbohydrate (functionally dextrose) or lipid, ideally in a 70:30 ratio. Approaches to assessing need caloric requirements include:
    • Assume 25-30kCal/kg/day
      • Easy
      • Crude
      • Inaccurate at extremes of weight and metabolic need
    • Reverse Fick equation
      Metabolic rate is proportional to VO2, which can be calculated using data from a PAC.
      • Does not measure pulmonary VO2, and so underestimates total requirements
        Becomes more inaccurate in pulmonary pathology.
    • Indirect calorimetry
      Direct measurement of VO2 and VCO2.
      • Most accurate (practical) technique
        • Assumes all O2 is used for oxidation of substrate
        • Accuracy falls with ↑ PEEP, ↑ FiO2, and with circuit leaks
      • Best in selected use cases
        Extremes of obesity, temperature, estimated metabolic rate.
      • Requires a metabolic cart that is rarely available and requires additional eperteise
    • Estimate expenditure
      Point estimate of VCO2 from ventilator.
      • Affected by changes to respiratory quotient in the critically ill
    • Predictive equations
      Many (i.e., >200) equations exist to evaluate different patient populations, basically all are poorly calibrated.
  • Protein
    • Assume 1-2g/kg/day
      More than this has no benefit and ↑ nitrogen load, leading to encephalopathy via accumulation of:
      • Urea
        With a normally-functioning urea cycle.
      • Ammonia
        Dysfunctional urea cycle.
    • Higher range recommended in protein deficient conditions:
      • Burns
      • Obesity
      • Trauma
      • RRT

Enteral Nutrition

Use of gastric residual volumes (like most of ICU nutrition) to determine EN tolerance is not well supported by evidence. Measured GRVs vary depending on the type of NGT used, and is a poor predictor of aspiration.

Liquid nutrition delivered directly into the digestive tract. Administration of EN is:

  • Associated with no overall change in mortality
  • Benefits:
    • Cheaper
    • Simpler
    • Overall ↓ infection
      Including pneumonia.
    • Possibly ↑ immune function
  • Risks:
    • Sinusitis
      From NGT; not a risk for OGT.
    • Sinus trauma nad epistaxis
    • VAP
      Microaspiration.
    • GI injury
    • Feed intolerance
  • Either continuous or intermittent bolus dosing
    • No difference in mortality outcomes
    • Continuous feeding reduces nursing workload
    • Intermittent feeding may:
      • Be more physiological
      • ↓ Insulin requirement by ↓ insulin resistance
      • ↑ Muscle synthesis

Gastric Residual Volumes

The GRV is the volume aspirated from the stomach during enteral feeding. Using the GRV to guide feeding:

The GRV is only useful with wide-bore NGTs, not fine-bore or jejunal tubes as these cannot be aspirated.

  • Is not well supported by evidence
    • Often delays feeding and total feeding volume
    • Poorly predicts aspiration
    • Prone to inter-observer variance
  • <500mL is a safe volume, supportive of normal GIT function
  • >500mL suggest delayed gastric emptying
  • Originally used to identify aspiration risk
  • Now used to identify feeding intolerance

Equipment

Feeding tube:

  • Wide-bore gastric tube
    • NGT
    • OGT
      Preferential in the intubated patient.
  • Fine-bore NGT
    • Cannot be used for suctioning
    • May be blocked
    • Better tolerated in the awake patient
      Comfort, improved swallow.
    • Aspiration risk
      Multiple reports of inadvertent tracheal placement.
  • Nasojejunal tube
    • Post-pyloric feeding useful with persistent ↓ gastric emptying
    • Difficult to place
      May require endoscopy.
  • PEG/PEJ
    Appropriate if long-term feeding required.

Formulations

Different feed preparations vary the:

Practically, there is little evidence of benefit with disease-specific combinations.

  • Calories
    • Fat
    • Carbohydrate
      ↓ Glucose content may ↓ insulin resistance
  • Protein
  • Osmolality
    ↑ Diarrhoea with ↑ osmolality. Consider switching to a lower concentration feed if the volume state permits.

Parenteral Nutrition

Nutrition given intravenously. PN is used when EN is either contraindicated or has failed, as it is inferior to EN. Key differences include:

Total Parenteral Nutrition refers to the entirety of a patients nutritional needs being met with TPN, as opposed to top-up PN when EN alone is insufficient.

  • Requires central venous access
    • ↑ Risk of CLABSI
      PN as an independent risk factor for infection is probably overstated.
  • ↑ Volume load
    1-2L/day.
  • Electrolyte derangements
    • NAGMA
    • Risk of refeeding syndrome
  • Lipid load
    • Hypertriglyceridaemia
    • Immunosuppression
  • ↑ BSL
  • ↓ Intestinal function
    Monitor LFTs for:
    • Hepatic steatosis
    • Cholestasis
    • Acalculous cholecystitis
  • More expensive

Refeeding syndrome is covered in Refeeding Syndrome.

Formulations

Different feed preparations vary the:

Thiamine, folic acid, and vitamin K are poorly supplemented by standard TPN.

  • Calories
    • Fat
      • Typically soybean oil with glycerol and egg phosphatides
      • Often 30-40% of non-protein energy
      • Some amount required to provide essential fatty acids
      • ↑ Fat ↓ the total volume of TPN required
    • Glucose
      • Preferred carbohydrate
      • >4mg/kg/min may exceed metabolic capacity and lead to ↑ BSL
      • Additional insulin is usually required
    • Nitrogen
      • Preparation-specific
      • Glutamine, tyrosine, and cysteine are often absent due to instability in solution
      • Typically bonded with chlorine, which may lead to NAGMA
  • Micronutrients
  • Electrolytes

An Approach

There is a paucity of strong evidence to guide decision making for nutrition within the ICU. This is one framework.

Principles:

  • Early (<48 hours) enteral nutrition
  • Early dietician involvement

Disease-Specific Considerations

Approach is largely disease-independent, although specific considerations include:

  • Chronic liver disease
    • No change in energy requirements
    • Thiamine deficiency
    • Fat-soluble vitamin deficiency
  • Pancreatitis
    • Favour EN feeding if available
  • Obesity
    • Expert recommendation for:
      • 11-14kcal/kg actual body weight
      • 1.2g/kg/day protein
    • Bariatric surgery patients at risk of micronutrient deficiency

Oxidation of fat produces less CO2 than oxidation of glucose for an equivalent amount of ATP produced. Use of lipid-dominant, low-respiratory quotient feed in the respiratory cripple does appear to ↓ minute ventilation requirements, but despite this does not appear to significantly affect ventilator time.

Assessment

Pathology severe enough to contraindicate EN could include:

  • Obstruction
  • Perforation
  • Mesenteric ischaemia
  • Short gut syndrome
  • Abdominal compartment syndrome
  • Assess nutritional status
  • Assess GIT function
    Contraindications to EN:
    • Severe GIT pathology
    • Active shock state
    • Undergoing resuscitation
  • Determine targets
    • Calorie target
      Assume 25kCal/kg/day.
    • Protein target
      As above.

Albumin is not a reliable marker of nutritional status because it is a negative acute phase reactant, and so production is suppressed in the critically ill.

Overview

The primary controversy here is whether PN should be started early (i.e. at 2-7 days), particularly in the critically ill.

Consider EN insufficient if unable to meet >60% of calorie and protein requirements.

  • Early Acute Phase (<2 days)
    • Energy: Strongly consider commencing EN at 30mL/hr and ↑ until at 70% of target
      • No PN if EN target not met.
      • Monitor for refeeding syndrome in the severely malnourished
    • Protein: No target
  • Late Acute Phase (2-7 days)
    • Energy: EN at up to 100% of target
      Consider PN if EN insufficient, particularly in the severely malnourished.
    • Protein: 1.3g/kg/day
      More if indicated.
  • Late phase (>7 days)
    • Energy: EN at up to 100% of target
      PN if EN insufficient.
    • Protein: >1.3g/kg/day

Enteral Nutrition

Approach:

Causes of feed intolerance:

  • Patient:
    • Hiatus hernia
    • Gastroparesis
      • Opioids
      • Anticholinergics
    • Bowel injury
      • Ileus
      • Obstruction
    • Sepsis
    • Electrolyte abnormalities
      ↓ K+, ↑ Mg2+.
  • Delivery:
    • NGT blocked/kinked
    • NGT malpositioned
  • Commence feeding at desired rate
  • Measure GRV PRN
    Only if vomiting or gastric distension.
    • Return volumes <500mL
    • After 1st GRV >500mL:
      An isolated high GRV is common.
      • Return up to 500mL of aspirate
      • Consider prokinetics
        No good evidence of benefit. If used, consider ceasing after 24-72 hours due to risks of tachyphylaxis.
    • After 2nd GRV >500mL:
      EN intolerance indicated by ⩾2 consecutive GRVs >500mL.
      • Return up to 500mL of aspirate
      • Halve EN delivery rate for 6/24
      • Repeat GRV in 6/24
    • After 3rd GRV >500mL:
      • Cease EN
      • Consider post-pyloric (nasojejunal) feeding
      • Consider PN
  • Avoid interruptions
    Avoid interrupting feeding when possible (e.g., for procedures where the airway is already secured).
Prokinetics
Drug Dose Notes
Metoclopramide 10mg Q6H IV
  • Dysrhythmias
  • Dystonia
Erythromycin 200mg Q12H IV
  • Prolonged QT
  • Hepatic dysfunction
  • C. Difficile
Methylnaltrexone 8-12mg Q24H SC
  • Only useful in opioid-induced paresis
  • ↓ Dose in renal failure
  • Greater effect on lower GIT
Neostigmine 2.5mg Q6H SC
  • Bradycardia (↑ risk if IV)
  • GI perforation (if true obstruction)
  • Greater effect on lower GIT

Parental Nutrition

Approach:

  • Commence PN at 80% of target if EN contraindicated or insufficient after 7 days
    If EN insufficient, make up remainder of requirement with PN.
  • Use a dedicated CVC lumen
  • Reassess enteral feeding Q12H

Micronutrients

Ensure adequate supplementation of:

  • Vitamins:
    • Fat soluble:
      • Vitamin A
      • Vitamin E
      • Vitamin K
    • Water soluble:
      Cleared in RRT.
      • Thiamine (B1)
      • Riboflavin (B2)
      • Niacin (B3)
      • Pyridoxine (B6)
      • Vitamin C
  • Trace elements
    Particularly in burns patients.
    • Selenium
    • Zinc
    • Manganese
    • Copper

Key Studies

Caloric targets:

  • PermiT (2015)
    • 894 adult ICU patients receiving EN within 48 hours, with an expected ICU LoS >72 hours
    • Multi-centre (7), block randomised, unblinded RCT
    • 80% power for 8% absolute reduction (!!) in mortality from baseline of 25%
    • Permissive underfeeding vs. standard feeding
      • Underfeeding
        • Caloric goal 40-60% of requirement
        • Protein and fluid supplementation as required
        • Average intake was 835 ±297 kCal/day (46 ±14% of goal)
      • Standard feeding
        • Goal was 70-100% of requirement
        • Average intake was 1299 ±467 kCal/day (71 ±22% of goal)
      • Caloric goal by equations
      • Protein goal 1.2-1.5g/kg/day
      • Over or underfeeding was corrected by adjusting the next days feed target
    • No difference in 90-day mortality (27.2% vs. 28.9%)
    • No difference in any secondary outcome or electrolytes
    • Higher insulin requirement and BSL in feeding group

EN Rate:

  • EDEN (2012)
    • 1000 intubated adult ICU patients with acute lung injury, ventilated <72 hours who required EN
    • Multicentre (44) unblinded RCT
    • 91% power to detect 2.25 day difference in ventilator free days
    • Trophic vs. full enteral feeding for the first 6 days of ICU admission
      • Trophic feeding
        • EN at 20kCal/hr
        • GRV checked Q12H
      • Full feeding
        • EN at 25mL/hr and advanced to goal rate by protocol (25mL/hr increments)
      • 400mL GRV threshold
    • No difference in ventilator free days, mortality, or infection
    • ↓ Feed intolerance in trophic feeding group
  • TARGET (2018)
    • 4000 invasively ventilated Australasians requiring EN for >24 hours
    • Multicentre (46), double-blind, block randomised RCT
    • 3774 patients gives 80% power for ~4% ↓ in 90 day mortality, assuming control group mortality as low as 20%
    • Energy-dense (1.5kCal/mL) vs. standard (1kCal/mL) feed
      • Feeds targeted based on IBW
      • Energy dense feed had higher total energy but similar protein to standard feed
    • No difference in 90 day mortality or a variety of secondary outcome
      Suggests pausing or decreasing feeds is safe in patients who are unable to meet requirements.
    • Energy dense feed had ↑:
      • GRV (250mL vs. 180mL)
      • Regurgitation and vomiting (19% vs. 16%)
      • Insulin use (56% vs. 49%)
      • Prokinetic use (47% vs. 40%)

Early vs. Late PN:

  • EPaNIC (2011)
    • 4640 nutritionally at-risk non-pregnant Belgian adults admitted to ICU, without profound malnutrition, ketosis, HHS, or moribund
    • Multicentre (7), unblinded, allocation-concealed, stratified (16 disease categories) trial
    • 80% power for 1 day change in ICU length of stay
    • Late PN vs. early PN
      • Late PN
        • PN at day 8 if EN not sufficient to meet caloric goal
        • Mean length of stay 3 days (IQR 2-7)
      • Early PN
        • PN at day 3 to meet 100% of goal
        • Mean length of stay 4 days (IQR 2-9)
      • Dextrose for caloric targeting
      • Tight glycaemic control (4.4-6.1mmol/L)
      • EN at day 2 if PO diet could not be initiated
      • 25-30kCal/kg/day, based on IBW
    • Significant ↓ ICU length of stay in late group (3 vs. 4, p = 0.02)
    • No change in mortality
    • Secondary outcomes all favoured late group:
      • ↓ New infection rate
      • ↓ Duration of mechanical ventilation and RRT
    • Very short duration of ICU admission in both groups
  • Gao et al (2022)
    • 230 adult Chinese patients following elective abdominal surgery across 11 hospitals who are:
      • At risk of malnutrition
      • At <30% of energy target via EN on POD 2
      • Without shock, AKI, or hepatic injury
    • Randomised to early (day 3) vs late (day 8) supplemental PN
    • Significant (8.7% vs 18.4%) in nosocomial infection prior to hospital discharge with no change in mortality
    • Not critically unwell patients

References

  1. Lambell, Kate J., Oana A. Tatucu-Babet, Lee-anne Chapple, Dashiell Gantner, and Emma J. Ridley. Nutrition Therapy in Critical Illness: A Review of the Literature for Clinicians. Critical Care 24, no. 1 (4 February 2020): 35.
  2. Gao X, Liu Y, Zhang L, et al. Effect of Early vs Late Supplemental Parenteral Nutrition in Patients Undergoing Abdominal Surgery: A Randomized Clinical Trial. JAMA Surgery. 2022;157(5):384-393.
  3. Bersten, A. D., & Handy, J. M. (2018). Oh’s Intensive Care Manual. Elsevier Gezondheidszorg.
  4. Energy-Dense versus Routine Enteral Nutrition in the Critically Ill. New England Journal of Medicine. 2018;379(19):1823-1834. doi:10.1056/NEJMoa1811687
  5. Casaer MP, Mesotten D, Hermans G, et al. Early versus Late Parenteral Nutrition in Critically Ill Adults. New England Journal of Medicine. 2011;365(6):506-517. doi:10.1056/NEJMoa1102662