Nutrition and Energy Requirements

Daily Energy Expenditure

The Harris-Benedict equations were published in 1919 and did not take into account changes in body weight caused by obesity or edema, thus ideal body weight must be used. The equation

BEE (kcal/day) = 25 x wt. (kg), or 1 kcal/kg per hour

is an approximation of the H-B equations but has been shown to be equivalent [J Am Coll Nutr 3: 51, 1984]. Adjustments to the BEE include fever (x1.1), mild stress (x1.2), moderate stress (x1.4), and severe stress (x1.6) but in reality these adjustments vary widely for individual patients [Crit Care Med 13: 173, 1985]. Studies have shown that these adjustments overestimate BEE by 20 – 60% in the ICU population, thus measurements of energy expenditure are preferred over these equations [Crit Care Med 13: 173, 1985; Anesthesiology 64: 673, 1986; Crit Care Med 18: 1320, 1990]

The BEE was originally measured in the fasting state. To derive the REE (resting expenditure) one must take into account the thermal effect of food, thus

REE = BEE x 1.2

To more accurately calculate REE, measure VO2 and VCO2 using “metabolic carts” (instruments that measure O2 and CO2 exchange across the lungs over the course of 15 – 30 minutes) and plug it into the following equation [Am J Clin Nutr 50: 227, 2989]:

REE = (3.9 x VO2) + (1.1 x VCO2) – 61 (units = kcal/day)

The REE is a good estimate in most patients, but those who are hypermetabolic (ex. sepsis) can be underestimated by ~ 40% when extrapolated over 24 hours [Surgery 118: 154, 1995], thus in hypermetabolic patients REE measured over limited periods (i.e., 30 min) is not as useful. Still, this is the best method of measuring energy requirements. It should be noted, however, that the equipment is expensive and requires specialized personnel, and the calculations are invalidated when inhaled O2 > 50% [Nutr Clin Pract 7: 207, 1992], which is often the case in respiratory failure. Note that when metabolic rate is excessive, -blockers can lower the metabolic rate by as much as 25% [Andrews]

Oxidative Energy Conversion

Fuel VO2 (L/g) VCO2 (L/g) RQ kcal/g Lipid 2.00 1.40 0.70 9.10 Protein 0.96 0.78 0.80 4.00 Glucose 0.74 0.74 1.00 3.70

Non-Protein Calories

Carbohydrates and lipids should be used to provide the necessary calories, with protein reserved for maintaining enzyme and structural protein stores. The proportion of calories that should be designated to carbohydrates vs. lipids is a matter of some debate, but there is no clear evidence favoring one over the other [Nutr Clin Pract 7: 207, 1992]

The central nervous system relies heavily on glucose for fuel, but the human body has limited carbohydrate reserves, thus daily carbohydrates are essential. Keep in mind, however, that excess carbohydrates can be detrimental – 1) they stimulate insulin release which, in time, can impair the ability of the body to mobilize fat stores during times of inadequate nutrition 2) they produce a relatively high amount of CO2 [Chest 88: 512, 1985], which can be an issue in patients with compromised lung function. Interestingly, any energy source can cause hypercapnia if given in excess [Chest 102: 551, 1992]

Most traditional nutritional regimens use lipids for 30% of their caloric intake, however linoleic acid is the only essential fatty acid. If inadequate, scaly dermopathy, cardiac dysfunction, and increased susceptibility to infection result [Linscheer WG. Lipids: Lea & Febiger 47, 1994]. This can be prevented if 0.5% of fatty acids are linoleic acid. Safflower oil is a common source.

Protein Requirements

Required intake can be predicted by the following equations: 0.8 – 1.0 g/kg/day for normal metabolism, and 1.2 – 1.6 g/kg/day for hypercatabolism [Crim MC. Protein and Amino Acids. Lea & Febiger 3, 1994]. These equations are just estimates, however. To get the true protein requirements one needs to determine the urinary excretion of nitrogen and look at the nitrogen balance. 67% of the nitrogen from protein breakdown is excreted in urine, and protein is 16% nitrogen, thus each gram of urinary nitrogen represents 6.25g of degraded protein:

N balance (in grams) = Protein intake (in grams) / 6.25 – (UUN + 4) (ideal =  4-6 g. Note that if UUN > 30, divide by 6.0 instead of 6.25)

For most patients, urea (UUN) makes up 85% of total UN, however in some ICU patients this number may be < 50%, and the addition of ammonia UN to UUN can give a more accurate estimation [J Parent Ent Nutr 17: 529, 1993]. The clinical significance of this is not yet known.

When trying to achieve  N balance, make sure that non-protein calories are adequate because attempting to achieve  N balance when protein is being degraded for calories is very difficult.

Vitamin Requirements

Daily requirements of vitamins in seriously ill patients may be much higher than those of standard parenteral regimens [J Parent Ent Nutr 11: 229, 1987; Crit Care Med 8: 500, 1980]

Thiamine may be a commonly underprovided vitamin, as 1) the body only stores 30 g (needs 3 g/day) 2) its use is increased significantly in hypercatabolic patients [Intensive Care Med 14: 628, 1988] 3) its excretion is increased by furosemide [Am J Med 91: 151, 1991] and 4) hypomagnesemia leads to decreased conversion to the essential TPP [Acta Med Scand 218: 129, 1985] causing de facto thiamine deficiency. Thiamine deficiency can lead to cardiac dysfunction, Wernicke’s, lactic acidosis, and/or a peripheral neuropathy. Screen with plasma thiamine levels (normal 3.4 – 4.8 g/dL total, 0.8 – 1.1 free, 2.6 – 3.7 phosphorylated), but the most reliable assay to determine thiamine status is the erythrocyte transketolase assay [J Nutr Sci Vitaminol 26: 507, 1980] (transketolase enzyme activity should increase by at least 25%)..

Marino believes that vitamin C and E may be required in excess of what is generally thought to be needed but this data needs to be looked up.

Trace Elements

Iron in the reduced state (II) promotes OH radical formation, thus sequesteration of iron may be the major antioxidant effect of plasma. For this reason, a reduced serum iron level should not prompt replacement therapy unless there is evidence of reduced total body iron (which is suggested by plasma ferritin < 18 g/dL. Ferritin > 100 g/dL suggests that total body iron is adequate).

Selenium is an antioxidant as it is a cofactor for glutathione peroxidase. Plasma levels may fall below normal within 1 week of a critical illness [Crit Care Med 18: 442, 1990]. Supplementation is not routinely provided, thus prolonged parenteral support is often associated with selenium deficiency [J Parent Ent Nutr 16: 54, 1992]. Minimum requirements are 55 and 70 µg/day for men and women, and maximum safe dose is 200 g/day – ICU patients should probably receive close to the maximum dose.

Nutrient Processing

Keep in mind that in the diseased state, malnutrition is often due to erroneous nutrient processing and not insufficient intake. Normal patients convert only 5% of glucose to lactate, but ICU patients can convert as much as 85% [Arch Surg 122: 765, 1987]. Thus, always be aware of the relevant underlying pathophysiologic processes – a study of 20 patients undergoing intraabdominal aortic aneurysm surgery showed that those given LR had a serum lactate increase of < 1 mM, whereas those given D5 solution intraop had an average increase of almost 3 mM by the end of surgery [Anesthesiology 71: 355, 1989] – the paradigm that nutrients will reduce malnutrition in critically ill patients may be misguided.

Neurosurgical Considerations

Failure to meet metabolic needs has been associated with increased morbidity and mortality in head injury patients [J Neurosurg 58: 906, 1983]. Head injury patients have a BME elevated 40% above baseline at 3 weeks [Ott, Young, in Andrews’], whereas SCI patients have a BME that is 10-55% lower than baseline [JPEN 13: 277, 1989; JPEN 15: 319,1991]. TBI patients routinely exhibit negative nitrogen balance, weight loss, immune depression, and depressed plasma protein levels despite adequate caloric and protein administration. SAH patients are less well studied but appear to have increased BME as well [Acta Neurochir (Wien) 106: 13, 1990]

Estimated Caloric Needs of Various Neurosurgical Patients@

Procedure/Condition Needs (kcal/kg/day) S/p craniotomy 26 GCS 4-5 40-50 GBC 6-7 30-40 GCS 8-12 30-35 Paraplegic 27 Quadraplegic 23

@Data taken from J Trauma 25: 419, 1985 and JPEN 13: 277, 1989

Initial protein requirements following head injury have been estimated at 1.5 – 2.5 g/kg IBW/day [New Horiz 2: 122, 1994]. The Traumatic Brain Injury Foundation Guidelines suggest that 15% of calories should be given as protein, and this can be assessed by nitrogen balance (measure weekly starting 5 days after protein support is established) – the goal should be 2 – 4 g/day positive.

Traditionally, it has been difficult to achieve positive protein balance in these patients, however two recent studies of IGF showed that using IGF makes this possible and may improve outcomes – in one, 8 of 11 treated patients with IGF-I keeping serum levels > 350 ng/ml achieved moderate-to-good outcome scores at 6 months, compared to only one of five patients with lower concentrations (p < 0.05) [J Neurosurg 86: 779, 1997]. A recent, prospective, randomized, double-blind of 97 TBI patients within 72 hours randomized to continuous IV IGF-1 (0.01 mg/kg/hr), and SQ GH (0.05 mg/kg/day) vs. control showed that the mean glucose concentration was higher in the treatment group (123 vs. 104, p < 0.03) but that a positive nitrogen balance was achieved within the first 24 hours in the treatment group and not in the control group (p < 0.05) [J Neurosurg 105: 843, 2006]

Consensus from the Traumatic Brain Injury Foundation Guidelines [Journal of Neurotrauma 24: 1, 2007]:

  • No class I recommendations can be made (it is a Level II recommendation that full nutritional replacement be instituted by day 7 post-injury)
  • Data support feeding at least by the end of the first week
  • It has not been established that any method of feeding is better than another
  • It has not been established that early feeding prior to 7 days improves outcome

Nutritional Considerations in Brain Injury Patients

  • Caloric needs are increased after surgery, SAH, or trauma, and decreased after SCI and during total paralysis [J Trauma 25: 419, 1985 and JPEN 13: 277, 1989]
  • Data suggest that IGF-1 and GH significantly improve nitrogen balance [J Neurosurg 105: 843, 2006] and may improve outcomes [J Neurosurg 86: 779, 1997]. Keep in mind that serum glucose was slightly increased by these studies.
  • Traumatic Brain Injury Guidelines recommend nutritional support by 7 days post-injury, state that there is no established superior feeding method, but note that there is no class I evidence on this subject.

Spinal reconstruction procedures are known to induce a post-operative catabolic state – because of the combination of catabolism with decreased nutritional intake, these patients at high risk for post-operative nutritional depletion [Clin Orthop 234: 5, 1988; Spine 17 (Suppl 8): S310, 1992; Spine 20: 1359, 1995]. For these reasons, TPN has been suggested as a means of reducing postoperative nutritional depletion and complications in patients undergoing spinal reconstructive surgery [Spine 21: 2676, 1996; Spine 23: 1401, 1998; Spine 24: 355, 1999]. The majority of studies, however, suggest that TPN does not improve post-operative nutritional status [Spine 26: 809, 2001], thus earlier enteral nutrition may be the only viable method by which adequate nutritional repletion can maintained in spinal reconstruction patients.

Nutritional Considerations in Spinal Reconstruction Patients

  • These patients are at high risk for post-operative nutritional depletion
  • The majority of studies, however, suggest that TPN does not improve post-operative nutritional status. [Spine 26: 809, 2001]