Endogenous energy stores are oxidized continuously for fuel. Triglyceride present in adipose tissue is the body’s major fuel reserve and is critical for survival during periods of starvation (Table 4-1). The high energy density and hydrophobic nature of triglycerides make it a fivefold better fuel per unit mass than glycogen. Triglycerides liberate 9.3 kcal/g when oxidized and are stored compactly as oil inside the fat cell. In comparison, glycogen produces only 4.1 kcal/g on oxidation and is stored intracellularly as a gel, containing approximately 2 g of water for every gram of glycogen. Adipose tissue is unable to provide fuel for certain tissues, such as bone marrow, erythrocytes, leukocytes, renal medulla, eye tissues, and peripheral nerves, which cannot oxidize lipids and require glucose for their energy supply. During endurance exercise, glycogen and triglycerides in muscle tissue provide an important source of fuel for working muscles.

Table 4-1 Endogenous Fuel Stores in a 70-kg Man

ENERGY METABOLISM

Energy is required continuously for normal organ function, maintenance of metabolic homeostasis, heat production, and performance of mechanical work. Total daily energy expenditure (TEE) is composed of three components: resting energy expenditure (~70% of TEE); the energy expenditure of physical activity (~20% of TEE); and the thermic effect of feeding (~10% of TEE), which is the temporary increase in energy expenditure that accompanies enteral ingestion or parenteral administration of nutrients.

Resting Energy Expenditure

Resting energy expenditure (REE) represents energy expenditure while a person lies quietly awake in an interprandial state; under these conditions, approximately 1 kcal/kg body weight is consumed/hour in healthy adults. Energy requirements of specific tissues differ dramatically (Table 4-2). The liver, intestine, brain, kidneys, and heart constitute approximately 10% of total body weight but account for approximately 75% of REE. In contrast, skeletal muscle at rest consumes approximately 20% of REE but represents approximately 40% of body weight, and adipose tissue consumes less than 5% of REE but usually accounts for more than 20% of body weight.

Table 4-2 Resting Energy Requirements of a 70-kg Man

Several empirical equations have been generated to estimate resting energy requirements (Table 4-3).¹^–⁴ These equations are useful in healthy subjects because they generate values that are usually within 10% of measured values. These equations are much less accurate, however, for persons who are at extremes in weight or who are ill, because alterations in body composition and metabolic stress influence energy expenditure. Protein-energy malnutrition and hypocaloric feeding without superimposed illness each decrease REE to values 10% to 15% below those expected for actual body size, whereas acute illness or trauma predictably increases energy expenditure (see later).

Table 4-3 Commonly Used Formulas for Calculating Resting Metabolic Rate

Harris-Benedict Equation
Men	66 + (13.7 × W) + (5 × H) − (6.8 × A)
Women	665 + (9.6 × W) + (1.8 × H) − (4.7 × A)

World Health Organization Formula
AGE (YR)	MALE	FEMALE
0-3	(60.9 × W) − 54	(60.1 × W) − 51
3-10	(22.7 × W) − 495	(22.5 × W) + 499
10-18	(17.5 × W) + 651	(12.2 × W) + 746
18-30	(15.3 × W) + 679	(14.7 × W) + 996
30-60	(11.2 × W) + 879	(8.7 × W) + 829
>60	(13.5 × W) + 987	(10.5 × W) + 596

A, age in years; H, height in centimeters; W, weight in kilograms.

Energy Expenditure of Physical Activity

The effect of physical activity on energy expenditure depends on the intensity and duration of daily activities. Highly trained athletes can increase their TEE 10- to 20-fold during athletic events. The activity factors shown in Table 4-4, each expressed as a multiple of REE, can be used to estimate TEE in active patients. The energy expended during physical activity is equal to REE × activity factor × duration of activity in hours/24 hours. TEE represents the summation of energy expended during all daily activities, including rest periods.

Table 4-4 Relative Thermic Effect of Various Levels of Physical Activity

ACTIVITY LEVEL	EXAMPLES	ACTIVITY FACTOR
Resting		1.0
Very light	Standing, driving, typing	1.1-2.0
Light	Walking 2-3 miles/hr shopping, light housekeeping	2.1-4.0
Moderate	Walking 3-4 miles/hr, biking, gardening, scrubbing floors	4.1-6.0
Heavy	Running, swimming, climbing, basketball	6.1-10.0

Adapted from Alpers DA, Stenson WF, Bier DM. Manual of nutritional therapeutics. Boston: Little, Brown; 1995.

INJURY OR ILLNESS	RELATIVE STRESS FACTOR*
Second- or third-degree burns, >40% BSA	1.6-1.8
Multiple trauma	1.5-1.7
Second- or third-degree burns, 20%-40% BSA	1.4-1.5
Severe infections	1.3-1.4
Acute pancreatitis	1.2-1.4
Second- or third-degree burns, 10%-20% BSA	1.2-1.4
Long bone fracture	1.2
Peritonitis	1.2
Uncomplicated postoperative state	1.1

BSA, body surface area.

* A stress factor of 1.0 is assumed for healthy controls.

An alternative and rather simple formula for adult inpatients that also incorporates a gauge of metabolic stress is as follows:

• 20-25 kcal/kg of actual body weight (ABW)/day for unstressed or mildly stressed patients

• 25-30 kcal/ABW/day for moderately stressed patients

• 30-35 kcal/ABW/day for severely stressed patients

In using this formula, adjustments are necessary when the ABW is a misleading reflection of lean body mass. An adjusted ideal body weight should be substituted for ABW in obese individuals who are more than 30% heavier than their ideal body weight (desirable body weight, more commonly referred to as ideal body weight [IBW], appears in Table 4-6). The use of an adjusted IBW helps prevent an overestimation of energy requirements and is calculated as follows:

Table 4-6 Desirable Weight in Relation to Height for Men and Women 25 Years or Older*

Also, in patients who have large artifactual increases in weight because of extracellular fluid retention, such as the patient with ascites, the IBW should be used to estimate energy requirements, rather than the ABW.

Method Without a Stress Factor

Table 4-7 outlines a simple method for estimating total daily energy requirements in hospitalized patients based on body mass index (BMI).⁶ With this method, energy expressed per kilogram is inversely related to BMI. Common sense needs to be applied when using any means to estimate energy expenditure in hospitalized individuals because illness commonly interjects artifacts into these calculations (e.g., ascites).

Table 4-7 Estimated Energy Requirements for Hospitalized Patients Based on Body Mass Index (BMI)*

BODY MASS INDEX (BMI; KG/M²)	ENERGY REQUIREMENTS (KCAL/KG/DAY)^†
<15	35-40
15-19	30-35
20-29	20-25
≥30	15-20

* The lower range within each BMI category should be considered in calculating energy requirements for insulin-resistant or critically ill patients to decrease the risk of hyperglycemia and infection associated with overfeeding.

† These values are recommended for critically ill patients and all obese patients; add 20% of the total calories when estimating energy requirements in non–critically ill patients.

Over the past two decades, the trend generally has been toward a more conservative approach to caloric delivery in acutely ill patients. One reason for this conservatism is that acute illness and its management often exacerbate preexisting diabetes or produces de novo glucose intolerance. As a result, hyperglycemia is a frequent consequence of enteral, and especially parenteral, nutrition. The issue seems to be particularly germane for intensive care unit (ICU) patients, in whom even modest hyperglycemia results in worse clinical outcomes, usually of an infectious nature. Clinical trials of high quality in surgical ICU (SICU)⁷ and medical ICU (MICU)⁸ patients have found that morbidity is substantially and significantly reduced in those randomized to intensive insulin therapy who maintained serum glucose levels below 111 mg/dL compared with those whose glucose values were maintained below 215 mg/dL. Mortality also was significantly lower in those in the SICU randomized to receive tight glucose control, although in the MICU study such reductions in mortality caused by tight glucose control only were realized in those residing in the MICU longer than three days. These clinical observations substantiate years of animal studies showing that even modest hyperglycemia impairs immune function in a variety of tissues.⁹ The clinical benefits of tight glucose control in the ICU, however, have not always been reproducible,¹⁰ and come at the cost of more frequent hypoglycemic episodes,^7,^8,¹⁰ so the issue of how tight glucose control should be remains controversial. Results of a recent meta-analysis of 29 trials in critically ill patients recapitulate the previously observed discrepancies between SICU and MICU patients.¹¹ Overall, the relative risk of septicemia was reduced approximately 25% in those randomized to tight glucose control, but this salutary effect largely was attributable to the SICU patients, in whom the reduction in septicemia was almost 50%. In contrast, no benefit was observed in MICU patients. Also in this meta-analysis, no demonstrable benefit in overall mortality was evident in any of the categories of critically ill patients.

PROTEIN

Twenty different amino acids are found commonly in human proteins. Some amino acids (histidine, isoleucine, leucine, lysine, methionine, phenylalanine, threonine, tryptophan, valine, and possibly arginine) are considered essential because their carbon skeletons cannot be synthesized by the body. Other amino acids (glycine, alanine, serine, cysteine, cystine, tyrosine, glutamine, glutamic acid, asparagine, and aspartic acid) are nonessential in most circumstances because they can be made from endogenous precursors or essential amino acids. In disease states, nonessential amino acids may become essential—so-called conditionally essential amino acids. Thus, infectious morbidity and wound healing appear to be improved in critically ill patients by the inclusion of supplemental glutamine in total parenteral nutrition (TPN) because of cellular depletion of this amino acid. Studies to date suggest that the clinical benefits of providing supplemental quantities of parenteral glutamine only are realized in patients with particularly high severity of illness scores (e.g., high APACHE II or SOFA scores), although the true nature of its efficacy likely will remain controversial until the results of several ongoing clinical trials become known.¹² Similarly, it has been shown that cysteine and tyrosine are essential in some patients with cirrhosis¹³ because of impaired hepatic synthesis.

The body of an average 75-kg man contains approximately 12 kg of protein. In contrast to fat and carbohydrate, there is no storage depot for protein, so excess intake is catabolized and the nitrogen component is excreted. Inadequate protein intake causes net nitrogen losses and, because no depot form of protein exists, there is an obligatory net loss of functioning protein. The U.S. Recommended Daily Allowance (RDA) of protein has been established at 0.8 g/kg/day, which reflects a mean calculated requirement of 0.6 g/kg/day plus an added factor to take into account the biological variance in requirement observed in a healthy population. Intravenously administered amino acids are as effective in maintaining nitrogen balance as oral protein of the same amino acid composition.¹⁴

Individual protein requirements are affected by several factors, such as the amount of nonprotein calories provided, overall energy requirements, protein quality, and the patient’s nutritional status (Table 4-8). Protein requirements increase when calorie intake does not meet energy needs. The magnitude of this increase is directly proportional to the deficit in energy supply. Therefore, nitrogen balance reflects protein intake and energy balance. Thus, correcting a negative nitrogen balance sometimes may be achieved merely by increasing the caloric delivery if the total amount of calories has been inadequate.

Table 4-8 Recommended Daily Protein Intake *

CLINICAL CONDITION	DAILY PROTEIN REQUIREMENT (G/KG IBW)
Normal	0.75
Metabolic stress	1.0-1.6
Hemodialysis	1.2-1.4
Peritoneal dialysis	1.3-1.5

IBW, ideal body weight.

* Additional protein requirements are needed to compensate for excess protein loss in specific patient populations (e.g., patients with burn injuries, open wounds, protein-losing enteropathy, or nephropathy). Lower protein intake may be necessary for patients with renal insufficiency not treated by dialysis and certain patients with liver disease and hepatic encephalopathy.

As metabolic stress (and with it, metabolic rate) increases, nitrogen excretion increases proportionately; quantitatively, the relationship is approximately 2 mg nitrogen (N)/kcal of REE. In part, this increase is explained by the fact that in metabolic stress, a larger proportion of the total substrate that is oxidized for energy is from protein. This has two important implications for managing the nutritional needs of ill patients. The first is that illness, by increasing catabolism and metabolic rate, increases the absolute requirement for protein (see Table 4-8), and does so in a manner that is roughly proportional to the degree of stress. Second, because a greater proportion of energy substrate in acute illness comes from protein, nitrogen balance is more readily achieved if a larger proportion of the total calories are from protein. In healthy adults, as little as 10% of total calories need to come from protein to maintain health, whereas in the ill patient, nitrogen balance is achieved more easily if 15% to 25% of total calories are delivered as protein.

Protein requirements also are determined by the adequacy of essential amino acids in the protein source. Inadequate amounts of an essential amino acid result in inefficient uptake, and therefore proteins of low biological quality increase the protein requirement. In normal adults, approximately 15% to 20% of total protein requirements should be in the form of essential amino acids.

Additional proteins are needed to compensate for excess loss in specific patient populations (e.g., patients with burn injuries, open wounds, and protein-losing enteropathy or nephropathy). Delivering less protein than is needed often is a necessary compromise in patients with acute renal insufficiency who are not adequately dialyzed, because in this situation the rise in azotemia is directly proportional to protein delivery. Once adequate dialysis is available, protein delivery should be increased to the actual projected need, including additional protein to compensate for losses resulting from dialysis (see Table 4-8). Most patients with hepatic encephalopathy respond to simple pharmacologic measures and therefore do not require a protein restriction; however, those who do not respond may benefit from a modest protein restriction (~0.6 g/kg/day).

Nitrogen Balance

Nitrogen (N) balance commonly is used as a proxy measure of protein balance—that is, whether the quantity of protein (or amino acids) taken in is sufficient to prevent any net loss of protein. N balance is calculated as the difference between N intake and N losses in urine, stool, skin, and body fluids. In the clinical setting, it is calculated as follows for adults:

Every 6.25 g of administered protein (or amino acids) contains approximately 1 g of N. The additional 4 g of N loss incorporated into the equation is intended to account for the insensible losses from the other sources listed, and because urinary urea N only accounts for about 80% of total urinary nitrogen. N balance is a suitable surrogate for protein balance because approximately 98% of total body N is in protein, regardless of the health of the individual.

A positive N balance (i.e., intake greater than loss) represents anabolism and a net increase in total body protein, whereas a negative N balance represents net protein catabolism. For example, a negative N balance of 1 g/day represents a 6.25-g/day loss of body protein, which is equivalent to a 30-g/day loss of hydrated lean tissue. In practice, N balance studies tend to be artificially positive because of overestimation of dietary N intake and underestimation of losses caused by incomplete urine collections and unmeasured outputs. It is best to wait at least four days after a substantial change in protein delivery before N balance is examined because a labile N pool exists; this tends to dampen and retard changes that otherwise would be observed as a result of altered protein intake.

CARBOHYDRATE

Complete digestion of the principal dietary digestible carbohydrates—starch, sucrose, and lactose—generate monosaccharides (glucose, fructose, and galactose). In addition, approximately 5 to 20 g of indigestible carbohydrate (soluble and insoluble fibers) are consumed daily. All cells are able to generate energy (adenosine triphosphate) by metabolizing glucose to three-carbon compounds via glycolysis or to carbon dioxide and water via glycolysis and the tricarboxylic acid (TCA) cycle.

There is no absolute dietary requirement for carbohydrate because glucose can be synthesized from endogenous amino acids as well as glycerol. Nevertheless, carbohydrate is an important fuel because of the interactions between carbohydrate and protein metabolism. Carbohydrate intake stimulates insulin secretion, which inhibits muscle protein breakdown,¹⁵ stimulates muscle protein synthesis,¹⁶ and decreases endogenous glucose production from amino acids.¹⁷ In addition, glucose is the required or preferred fuel for red and white blood cells, the renal medulla, eye tissues, peripheral nerves, and the brain. Once glucose requirements for these tissues are met (∼150 g/day), however, the protein-sparing effects of carbohydrate and fat are similar.¹⁸

LIPIDS

Lipids consist of triglycerides (TGs), sterols, and phospholipids. These compounds serve as sources of energy, precursors for steroid hormone, prostaglandin, thromboxane, and leukotriene synthesis, structural components of cell membranes, and carriers of essential nutrients. Dietary lipids are composed mainly of TGs, which contain saturated and unsaturated long-chain fatty acids (FAs) of 16 to 18 carbons. The use of fat as a fuel requires the hydrolysis of endogenous or exogenous TGs and cellular uptake of released FAs. Long-chain FAs are delivered across the outer and inner mitochondrial membranes by a carnitine-dependent transport system.¹⁹ Once inside the mitochondria, FAs are degraded by beta oxidation to acetyl coenzyme A (CoA), which then enters the TCA cycle. Therefore, the ability to use fat as a fuel depends on normally functioning mitochondria. A decrease in the number of mitochondria or oxidative enzymes associated with aging²⁰ or deconditioning favors the use of carbohydrate as fuel.²¹

Essential Fatty Acids

Humans lack the desaturase enzyme needed to produce the n-3 (double bond between carbons 3 and 4) and n-6 (double bond between carbons 6 and 7) FA series. Linoleic acid (C18 : 2, n-6) and linolenic acid (C18 : 3, n-3), therefore, should constitute at least 2% and 0.5%, respectively, of the daily caloric intake to prevent essential FA deficiency (EFAD). Before the advent of parenteral nutrition, EFAD was recognized only in infants and manifested as a scaly rash with a specific alteration in the plasma FA profile (see later). Adults previously were thought not to be susceptible to EFAD because of sufficient essential FA stores in adipose tissue. However, an abnormal FA profile in conjunction with a clinical syndrome of EFAD is now known to occur sometimes in adults with severe short bowel syndrome who are on long-term total parenteral nutrition (TPN) that lacks parenteral lipids.²² Adults who have moderate to severe fat malabsorption (fractional fat excretion >20%) from other causes and who are not TPN-dependent also frequently display a biochemical profile of EFAD,²³ although whether such a biochemical state carries adverse clinical consequences with it is unclear. Moreover, TPN lacking any source of fat may lead to EFAD in adults if no exogenous source of EFAs is available; the plasma pattern of EFAD may be observed as early as 10 days after glucose-based TPN is started and before the onset of any clinical features.²⁴ In this situation, EFAD is probably attributable to the increase in plasma insulin concentrations caused by TPN, because insulin inhibits lipolysis and therefore the release of endogenous essential FAs. The biochemical diagnosis of EFAD is defined as an absolute and relative deficiency in the two EFAs in the plasma FA profile. The full clinical EFAD syndrome includes alopecia, scaly dermatitis, capillary fragility, poor wound healing, increased susceptibility to infection, fatty liver, and growth retardation in infants and children.

MAJOR MINERALS

Major minerals are inorganic nutrients that are required in large (>100 mg/day) quantities, and are important for ionic equilibrium, water balance, and normal cell function. Malnutrition and nutritional repletion can have dramatic effects on major mineral balance. The evaluation of macromineral deficiency and recommended daily allowance (RDA) of minerals for healthy adults are shown in Table 4-9.

Table 4-9 Major Mineral Requirements and Assessment of Deficiency

MICRONUTRIENTS

Micronutrients (the vitamins and trace minerals) are a diverse array of dietary components that are necessary to sustain health. The physiologic roles of micronutrients are as varied as their composition. Some are used in enzymes as coenzymes or prosthetic groups, others as biochemical substrates or hormones and, in some cases, their functions are not well defined. The average daily dietary intake for each micronutrient required to sustain normal physiologic operations is measured in milligrams or smaller quantities. In this way, micronutrients are distinguished from macronutrients (carbohydrates, fats, and proteins) and macrominerals (calcium, magnesium, and phosphorus).

An individual’s dietary requirement for any given micronutrient is determined by many factors, including its bioavailability, the amount needed to sustain its normal physiologic functions, a person’s gender and age, any diseases or drugs that affect the nutrient’s metabolism, and certain lifestyle habits, such as smoking and alcohol use. The U.S. National Academy of Sciences Food and Nutrition Board regularly updates dietary guidelines that define the quantity of each micronutrient that is “adequate to meet the known nutrient needs of practically all healthy persons.” These RDAs underwent revision between 1998 and 2001, and the values for adults appear in Tables 4-10 and 4-11. The formulation of an RDA takes into account the biological variability in the population, and, therefore, RDAs are set 2 SDs above the mean requirement; this allows the requirements of 97% of the population to be met and ingestion of quantities that are somewhat less than the RDA usually are sufficient to meet the needs of a particular individual. A “tolerable upper limit (TUL),” which is “the maximal daily level of oral intake that is likely to pose no adverse health risks,” also has been established for most of the micronutrients (see Tables 4-10 and 4-11). Present recommendations for how much of each micronutrient is needed in individuals on TPN are based on far less data than what were available for the development of the RDAs. Nevertheless, it is important to have guidelines, and Table 4-12 provides such recommendations.

Table 4-10 Salient Features of Vitamins

Table 4-11 Salient Features of Trace Minerals

Table 4-12 Guidelines for Daily Delivery of Parenteral Micronutrients in Adults and Children

MICRONUTRIENT	ADULTS	CHILDREN
Vitamin
A	1000 µg (= 3300 IU)	700 µg
D	5 µg (= 200 IU)	10 µg
E	10 mg (= 10 IU)	7 mg
K	1 mg	200 µg
C	100 mg	80 mg
Folate	400 µg	140 µg
Niacin	40 mg	17 mg
Riboflavin	3.6 mg	1.4 mg
Thiamine	3 mg	1.2 mg
B₆	4 mg	1.0 mg
B₁₂	5 µg	1.0 µg
Pantothenic acid	15 mg	5 mg
Biotin	60 µg	20 µg
Trace Elements
Copper	0.5-1.5 mg	20 µg/kg/day
Chromium	10-15 µg	0.2 µg/kg/day
Manganese	0.1 mg	1.0 µg/kg/day
Zinc	2.5-4.0 mg	50 µg/kg/day
Molybdenum	15 µg	0.25 µg/kg/day
Iodine^*	—	—
Selenium	100 µg	2.0 µg/kg/day
Iron	1-2 mg	1 mg/day

* Naturally occurring contamination of parenteral nutrition appears to provide sufficient quantities of iodine.

Adult vitamin guidelines adapted from American Society of Parenteral and Enteral Nutrition (ASPEN). Board of Directors and the Clinical Guidelines Task Force. Guidelines for the use of parenteral and enteral nutrition in adult and pediatric patients. JPEN J Parenter Enteral Nutr 2002;26:144. Children’s values adapted from Greene HL, Hambidge KM, Schanler R, Tsang RC. Guidelines for the use of vitamins, trace elements, calcium, magnesium, and phosphorus in infants and children receiving total parenteral nutrition: Report of the Subcommittee on Pediatric Parenteral Nutrient Requirements from the Committee on Clinical Practice Issues of the American Society for Clinical Nutrition. Am J Clin Nutr 1988; 48:1324; Am J Clin Nutr 1989; 49:1332; and Am J Clin Nutr 1989; 50:560.

VITAMINS

Vitamins are categorized as fat-soluble (A, D, E, K) or water-soluble (all others) (see Table 4-10). This categorization remains physiologically meaningful; none of the fat-soluble vitamins appear to serve as coenzymes, whereas almost all of the water-soluble vitamins appear to function in that role. Also, the absorption of fat-soluble vitamins is primarily through a micellar route, whereas the water-soluble vitamins are not absorbed in a lipophilic phase in the intestine (see Chapter 100).

TRACE MINERALS

Compelling evidence exists for the essential nature of 10 trace elements in humans—iron, zinc, copper, chromium, selenium, iodine, fluorine, manganese, molybdenum, and cobalt (see Table 4-11). The biochemical functions of trace elements have not been as well characterized as those of the vitamins, but most of their functions appear to be as components of prosthetic groups or as cofactors for a number of enzymes.

Aside from iron, the trace mineral depletion that clinicians are most likely to encounter is zinc deficiency. Zinc depletion is a particularly germane issue to the gastroenterologist because the gastrointestinal tract is a major site for zinc excretion. Chronically excessive losses of gastrointestinal secretions, such as chronic diarrhea in inflammatory bowel disease, is a known precipitant for zinc deficiency, and in this setting zinc requirements often increase several-fold.²⁵ Nevertheless, a biochemical diagnosis of zinc deficiency is problematic, as is true for many of the other essential trace minerals. Accurate laboratory assessment of zinc status is complicated by the very low concentrations of zinc in bodily fluids and tissues, a lack of correlation between serum and red blood cell levels of zinc with levels in the target tissues, and the reality that functional tests have yet to be devised. Furthermore, it is well recognized that in acute illness a shift in zinc occurs from the serum compartment into the liver, further obscuring the diagnostic value of serum zinc levels.^26,²⁷ Thus, it is often best simply to proceed with empirical zinc supplementation in patients whose clinical scenario puts them at high risk of zinc deficiency.

Some reports have indicated that TPN solutions that deliver several-fold more manganese than what is recommended in Table 4-12 may lead to deposition of the mineral in the basal ganglia, with extrapyramidal symptoms, seizures, or both.²⁸ Because the content of manganese varies widely in the different trace element mixtures available for TPN compounding, the health professional needs to be mindful of this issue as protocols for TPN admixtures are developed.

PHYSIOLOGIC AND PATHOPHYSIOLOGIC FACTORS AFFECTING MICRONUTRIENT REQUIREMENTS

Age

An evolution of physiology continues throughout the life cycle, with an impact on the requirements of certain micronutrients with aging; specific RDAs for older adults now have been developed. The mean vitamin B₁₂ status of most populations, for example, declines significantly with older age, in large part because of the high prevalence of atrophic gastritis and its resultant impairment of protein-bound vitamin B₁₂ absorption.²⁹ Approximately 10% to 15% of the older ambulatory population is thought to have significant vitamin B₁₂ depletion because of this phenomenon, and neuropathic degeneration may occur in older individuals whose plasma vitamin B₁₂ levels are in the low-normal range (150 to 300 pg/mL), even in the absence of hematologic manifestations. For this reason, the use of sensitive indicators of cellular depletion of vitamin B₁₂, such as serum methylmalonic acid levels in conjunction with serum levels of vitamin B₁₂, now are recommended for diagnosis.³⁰ Some experts also suggest that older adults should consume a portion of their vitamin B₁₂ requirement in the crystalline form (i.e., as a supplement) rather than relying only on the naturally occurring protein-bound forms found in food.³¹ Older adults also require greater quantities of vitamins B₆ and D and calcium to maintain health compared with younger adults, and these requirements are reflected in the new RDAs (see Tables 4-10 and 4-11).

Malabsorption and Maldigestion

Both fat- and water-soluble micronutrients are absorbed predominantly in the proximal small intestine, with the only exception being vitamin B₁₂. Diffuse mucosal diseases, which affect the proximal portion of the gastrointestinal tract, are therefore likely to result in multiple deficiencies. Even in the absence of proximal small intestinal disease, however, extensive ileal disease, small bowel bacterial overgrowth, and chronic cholestasis may interfere with the maintenance of adequate intraluminal conjugated bile acid concentrations and thereby may impair absorption of fat-soluble vitamins.

Conditions that produce fat malabsorption frequently are associated with selective deficiencies of the fat-soluble vitamins. The early stages of many vitamin deficiencies are not apparent clinically and therefore may go undetected until progression of the deficiency has resulted in significant morbidity. This can be disastrous in conditions such as spinocerebellar degeneration from vitamin E deficiency, which often is irreversible.³² Fat-soluble vitamin deficiencies are well-recognized complications of cystic fibrosis and congenital biliary atresia in which fat malabsorption often is overt, but monitoring also is necessary in conditions associated with more subtle fat malabsorption, such as the latter stages of chronic cholestatic liver disease.^33,³⁴

Restitution of vitamin deficiencies sometimes can be difficult when severe fat malabsorption is present and initial correction may require parenteral administration. In severe fat malabsorption, chemically modified forms of vitamins D and E that largely bypass the need for the lipophilic phase of intestinal absorption are commercially available for oral use and can be helpful. The polyethylene glycol succinate form of vitamin E (Nutr-E-Sol) is very effective in patients with severe fat malabsorption who cannot absorb conventional alpha-tocopherol.³⁵ Similarly, hydroxylated forms of vitamin D (1-hydroxyvitamin D [Hectorol] and 1,25-dihydroxyvitamin D [Rocaltrol]) can be used in patients resistant to the more conventional forms of vitamin D. Intermittent monitoring of serum calcium levels is indicated in the first few weeks of therapy when hydroxylated forms of vitamin D are administered because they are considerably more potent than vitamin D₂ or D₃ and risk of vitamin D toxicity exists. In contrast, water-miscible preparations of fat-soluble vitamins, in which a conventional form of vitamin A or E is dissolved in polysorbate 80 (e.g., Aquasol-E, Aquasol-A), have not been proven to improve overall absorption. At the time of this writing, Aquasol-A is no longer available as an oral supplement.

Maldigestion usually results from chronic pancreatic insufficiency, which if untreated frequently causes fat malabsorption and deficiencies of fat-soluble vitamins. Vitamin B₁₂ malabsorption also can be demonstrated in this setting, but clinical vitamin B₁₂ deficiency is rare unless other conditions known to diminish its absorption also are present, such as atrophic gastritis ²⁹ or chronic administration of proton pump inhibitors (PPIs).³⁶ Whether the long-term administration of PPIs alone warrants occasional checks of vitamin B₁₂ status is a matter of debate. Regardless, malabsorption of vitamin B₁₂ from atrophic gastritis or with PPIs is confined to dietary sources of vitamin B₁₂. Small supplemental doses of crystalline vitamin B₁₂ are absorbed readily in both cases. Histamine H₂ receptor antagonists also inhibit protein-bound vitamin B₁₂ absorption, although the effect generally is believed to be less potent than with the PPIs.³⁷

Many medications may adversely affect micronutrient status. The manner in which drug-nutrient interaction occurs varies; some of the more common mechanisms are described in Table 4-13. A comprehensive discussion of drug-nutrient interactions is beyond the scope of this chapter and the reader is referred to other references for a detailed discourse on this topic.³⁸

Table 4-13 Interactions of Drugs on Micronutrient Status

DRUG(S)	NUTRIENT	MECHANISM(S)
Dextroamphetamine, fenfluramine, levodopa	Potentially all micronutrients	Induces anorexia
Cholestyramine	Vitamin D, folate	Adsorbs nutrient, decreases absorption
PPIs	Vitamin B₁₂	Modest bacterial overgrowth, decreases gastric acid, impairs absorption
Sulfasalazine	Folate	Impairs absorption and inhibits folate-dependent enzymes
Isoniazid	Pyridoxine	Impairs uptake of vitamin B₆
NSAIDs	Iron	Gastrointestinal blood loss
Penicillamine	Zinc	Increases renal excretion

NSAIDs, nonsteroidal anti-inflammatory drugs; PPIs, proton pump inhibitors.

From Goldman L, Ausiello D, Arend W, et al, editors. Cecil Textbook of Medicine. 22nd ed. Philadelphia: WB Saunders; 2004. With permission.

STARVATION

During periods of energy and/or protein deficit, an array of compensatory mechanisms serves to lessen the pathophysiologic impact of these deficits. These responses decrease the metabolic rate, maintain glucose homeostasis, conserve body nitrogen, and increase the uptake of adipose tissue TGs to meet energy needs. To appreciate how acute illness disrupts this compensatory scheme, it is first necessary to understand how the body adapts to starvation in the absence of underlying disease.

During the first 24 hours of fasting, the most readily available energy substrates (i.e., circulating glucose, FAs, and TGs, and liver and muscle glycogen) are used as fuel sources. The sum of energy provided by these stores in a 70-kg man, however, is only about 5000 kJ (1200 kcal) and therefore is less than a full day’s requirements. Hepatic glucose production and oxidation decrease, whereas whole-body lipolysis increases, and the latter provides additional FAs and ketone bodies.³⁹ Oxidation of the FAs released from adipose tissue TGs accounts for approximately 65% of energy consumed during the first 24 hours of fasting.

During the first several days of starvation, obligate glucose-requiring tissues such as the brain and blood cells, which collectively account for about 20% of total energy consumption, can use only glycolytic pathways to obtain energy. Because FAs cannot be converted to carbohydrate, these glycolytic tissues must use glucose or substrates that can be converted to glucose. Glucogenic amino acids derived from skeletal muscle (chiefly alanine and glutamine) are a major source of substrate for this purpose. Approximately 15% of the REE is provided by oxidation of protein.⁴⁰ The relative contribution of gluconeogenesis to hepatic glucose production increases as the rate of hepatic glycogenolysis declines because the latter process becomes redundant; after 24 hours of fasting, only 15% of liver glycogen stores remain.

During short-term starvation (1 to 14 days), several adaptive responses appear that lessen the loss of lean mass. A decline in levels of plasma insulin, an increase in plasma epinephrine levels, and an increase in lipolytic sensitivity to catecholamines stimulate adipose tissue lipolysis.^41,⁴² The increase in FA delivery to the liver, in conjunction with an increase in the ratio of plasma glucagon-to-insulin concentrations, enhances the production of ketone bodies by the liver. A maximal rate of ketogenesis is reached by three days of starvation, and plasma ketone body concentration is increased 75-fold by seven days. In contrast to FAs, ketone bodies can cross the blood-brain barrier and provide most of the brain’s energy needs by seven days of starvation.⁴³ The use of ketone bodies by the brain greatly diminishes glucose requirements and thus spares the need for muscle protein degradation to provide glucose precursors. If early protein breakdown rates were to continue throughout starvation, a potentially lethal amount of muscle protein would be catabolized in less than three weeks. Similarly, the heart, kidney, and skeletal muscle change their primary fuel substrate to FAs and ketone bodies. Other tissues such as bone marrow, renal medulla, and peripheral nerves switch from full oxidation of glucose to anaerobic glycolysis, resulting in increased production of pyruvate and lactate. The latter two compounds can be converted back to glucose in the liver using energy derived from fat oxidation via the Cori cycle, and the resulting glucose is available for systemic consumption. This enables energy stored as fat to be used for glucose synthesis.

Whole-body glucose production decreases by more than 50% during the first few days of fasting because of a marked reduction in hepatic glucose output. As fasting continues, the conversion of glutamine to glucose in the kidney represents almost 50% of total glucose production. Energy is conserved by a decrease in physical activity because of fatigue and a reduction in REE of approximately 10%, resulting from increased conversion of active thyroid hormone to its inactive form and suppressed sympathetic nervous system activity.

During long-term starvation (14 to 60 days), maximal adaptation is reflected by a plateau in lipid, carbohydrate, and protein metabolism. The body relies almost entirely on adipose tissue for its fuel, providing more than 90% of daily energy requirements.⁴⁴ Muscle protein breakdown decreases to less than 30 g/day, causing a marked decrease in urea nitrogen production and excretion. The decrease in osmotic load diminishes urine volume to 200 mL/day, thereby reducing fluid requirements. Total glucose production decreases to approximately 75 g/day, providing fuel for glycolytic tissues (40 g/day) and the brain (35 g/day) while maintaining a constant plasma glucose concentration. Energy expenditure decreases by 20% to 25% at 30 days of fasting and remains relatively constant thereafter, despite continued starvation.

The metabolic response to short- and long-term starvation differs somewhat between lean and obese persons. Obesity is associated with a blunted increase in lipolysis and decrease in glucose production compared with that in lean persons.^45,⁴⁶ In addition, protein breakdown and nitrogen losses are less in obese persons, thereby helping conserve muscle protein.⁴⁷

The events that mark the terminal phase of starvation have been studied chiefly in laboratory animals. Body fat mass, muscle protein, and the sizes of most organs are markedly decreased. The weight and protein content of the brain, however, remain relatively stable. During this final phase of starvation, body fat stores reach a critical level, energy derived from body fat decreases, and muscle protein catabolism is accelerated. Death commonly occurs when there is a 30% to 50% loss of skeletal muscle protein.⁴⁸ In humans, it has been proposed that there are certain thresholds beyond which lethality is inevitable—depletion of total body protein between 30% and 50% and of fat stores between 70% and 95%, or reduction of BMI below 13 kg/m² for men and 11 kg/m² for women.^49,⁵⁰

MALNUTRITION

In the broadest sense, malnutrition implies a sustained imbalance between nutrient availability and nutrient requirements. This imbalance results in a pathophysiologic state in which intermediary metabolism, organ function, and body composition are variously altered. Sustained is an important element of this definition because homeostatic mechanisms and nutrient reserves usually are adequate to compensate for any short-term imbalance.

Customarily, the term malnutrition is used to describe a state of inadequacy in protein, calories, or both, and is more precisely called protein-energy malnutrition (or protein-calorie malnutrition). Occasionally, it is used to describe a state of excessive availability, such as a sustained excess of calories (e.g., obesity) or a vitamin (e.g., vitamin toxicity).

PROTEIN-ENERGY MALNUTRITION

There are different pathways whereby protein-energy malnutrition (PEM) may evolve. Primary PEM is caused by inadequate intake of protein and/or calories or, less commonly, when the protein ingested is of such poor quality that one or more essential amino acids becomes a limiting factor in the maintenance of normal metabolism. Secondary PEM is caused by illness or injury.

Acute illnesses and injuries increase bodily requirements for protein and energy substrate and they impair the digestion, absorption, and uptake of these nutrients in various ways. Consequently, secondary PEM usually arises from multiple factors. Illness and injury also commonly induce anorexia (see later for mechanisms), and so primary and secondary factors often act in concert to create PEM in the setting of illness.

Illness or injury may directly interfere with nutrient assimilation; for example, extensive ileal disease or resection may directly produce fat malabsorption and a caloric deficit. The most common causes of secondary PEM, however, are the remarkable increases in protein catabolism and energy expenditure that occur as a result of a systemic inflammatory response. REE may increase as much as 80% above basal levels in a manner roughly proportional to the magnitude of the inflammatory response, which in turn is roughly proportional to the severity and acuity of the illness. Thus, for example, REE in patients with extensive second- and third-degree burns (the prototype for maximal physiologic stress) may approach twice normal; with sepsis, REE is about 1.5 times normal; and with a localized infection or fracture of a long bone, REE is 25% above normal.⁵ Such stress factors can be used to construct a formula for predicting the caloric needs of ill individuals (see Table 4-5).

Protein catabolism during illness or injury also increases in proportion to the severity and acuity of the insult, and therefore parallels the increase in energy consumption. The magnitude of increase in protein catabolism, however, is proportionately greater than that observed with energy consumption, such that urinary urea N losses, which reflect the degree of protein catabolism in acute illness, are about 2.5 times the basal level with maximal stress.⁵ This increase in catabolism results in a net loss of protein because the rate of synthesis usually does not rise in concert with the rise in catabolism.⁵¹ No known storage form of protein exists in the body and, therefore, any net loss of protein represents a loss of functionally active tissue. A healthy adult typically loses about 12 g N in the urine/day, and excretion may increase to as much as 30 g/day during critical illness. Because 1 g of urinary N represents the catabolism of approximately 30 g of lean mass, it follows that severe illness may produce a daily loss of up to ~0.5 kg of lean mass as a result of excess protein catabolism. Most of this loss comes from the skeletal muscle, where the efflux of amino acids increases two- to sixfold in critically ill patients.⁵²

Mobilization of amino acids from skeletal muscle appears to be an adaptive response. Once liberated, these amino acids, in part, are deaminated and used for gluconeogenesis; they also are taken up by the liver and other visceral organs. The proteolysis of muscle under stress thus enables the body to shift amino acids from the skeletal muscle (the somatic protein compartment) to the visceral organs (the visceral protein compartment), the functions of which are more critical for immediate survival during illness. Nevertheless, with sustained stress, the limitations of this adaptive response become evident, and even the visceral protein compartment sustains a contraction in mass.⁴⁴

Primary versus Secondary Protein-Energy Malnutrition: A Body Compartment Perspective

The type of tissue lost as malnutrition evolves is critical in determining the pathologic ramifications of weight loss. Over 95% of energy expenditure resides in the lean body mass, which therefore contains the bulk of metabolism that sustains homeostasis. It is the maintenance of this body compartment that is most critical for health. Lean body mass can be subdivided further into somatic and visceral protein compartments, blood and bone cells, and extracellular lean mass, such as plasma and bone matrix (Fig. 4-1). In total or semistarvation in otherwise healthy individuals, adipose tissue predominates as a primary energy source; thus, fat mass contracts to a much greater degree proportional to the loss of lean mass.⁴⁴ Alterations in metabolism from injury or illness, however, produce a proportionately greater loss of muscle mass so that it matches or exceeds the loss in fat mass.^53,⁵⁴ Although the lean mass that is lost in illness preferentially is from the somatic protein compartment, with sustained stress there also will be a significant contraction of the visceral protein compartment (Table 4-14). The metabolic forces associated with acute illness and injury are potent, and restoration of muscle mass is unlikely with nutritional support unless the underlying inflammatory condition is corrected. There is increasing interest in attenuating or reversing net catabolism with the use of exogenous anabolic agents in conjunction with nutrition, although to date it remains unclear whether the clinical benefits of using exogenous growth hormone and other anabolic agents in acute illness outweigh their potential side effects.^55,⁵⁶ Another important ramification of the potency of the catabolic state associated with acute illness is that most of the weight that is gained with the provision of nutritional support is the result of increases in fat mass and body water; only minor increases in lean mass are observed until the inflammatory focus is resolved.⁵⁷

Figure 4-1. Body composition analysis by weight in a healthy adult. The speckled segments and gray segment collectively represent lean body mass. The speckled segments alone represent body cell mass.

(From Mason JB: Gastrointestinal cancer; nutritional support. In Kelsen D, Daly J, Kern S, et al, editors. Principles and Practice of Gastrointestinal Oncology. Philadelphia, Pa: Lippincott Williams & Wilkins; 2002. p 127. With permission.)

Table 4-14 Body Compartment Losses in Simple Starvation versus Metabolic Stress

Cytokines are the most important mediators of the alterations in energy and protein metabolism that accompany illness and injury. In a wide spectrum of systemic illnesses, increased secretion of interleukin-1 (IL-1), tumor necrosis factor-α (TNF-α), IL-6, and interferon-γ (IFN-γ) has been observed to be associated with increased energy expenditure and protein catabolism, as well as the shift of amino acids into the visceral compartments.⁵⁸^–⁶⁰ Such observations concur with in vitro studies in human cells and animal models that have shown remarkably potent effects of these cytokines in this regard (Table 4-15). In the wasting syndrome associated with cancer, proteolysis-inducing factor and lipid-mobilizing factor are two humoral mediators that appear to be unique to cancer cachexia, contributing to protein catabolism and loss of adipose tissue, respectively.⁶¹

Table 4-15 Major Cytokines That Mediate Hypercatabolism and Hypermetabolism Associated with Metabolic Stress

CYTOKINE	CELL SOURCE	METABOLIC EFFECTS
Tumor necrosis factor-α	Monocytes/macrophages, lymphocytes, Kupffer cells, glial cells, endothelial cells, natural killer cells, mast cells	Decreased FFA synthesis
		Increased lipolysis
		Increased peripheral amino acid loss
		Increased hepatic amino acid uptake
		Fever
IL-1	Monocytes/macrophages, neutrophils, lymphocytes, keratinocytes, Kupffer cells	Increased ACTH levels
		Increased hepatic acute-phase protein synthesis
		Fever
IL-6	Monocytes/macrophages, keratinocytes, endothelial cells, fibroblasts, T cells, epithelial cells	Increased acute-phase protein synthesis
IL-6		Fever
IFN-γ	Lymphocytes, pulmonary macrophages	Increased monocyte respiratory burst

ACTH, adrenocorticotropic hormone; FFA, free fatty acid; IL, interleukin; IFN, interferon.

Adapted from Smith M, Lowry S. The hypercatabolic state. In: Shils M, Olson J, Shike M, Ross AC, editors. Modern Nutrition in Health and Disease. Baltimore: Williams & Wilkins; 1999. p 1555.

Protein-Energy Malnutrition in Children

Undernutrition in children differs from that in adults because it affects growth and development. Much of our understanding of undernutrition in children comes from observations made in underdeveloped nations in which poverty, inadequate food supply, and unsanitary conditions lead to a high prevalence of PEM. The Waterlow classification of malnutrition takes into account a child’s weight for height (wasting) and height for age (stunting) (Table 4-16).⁶² The characteristics of the three major clinical PEM syndromes in children—kwashiorkor, marasmus, and nutritional dwarfism—are outlined in Table 4-17.⁶³ Although these three syndromes are classified separately, overlap syndromes often coexist in the same patient.

Table 4-16 Waterlow Classification of Protein-Energy Malnutrition in Children

Table 4-17 Features of Protein-Energy Malnutrition Syndromes in Children

Marasmus

Weight loss and marked depletion of subcutaneous fat and muscle mass are characteristic features of children with marasmus. Ribs, joints, and facial bones are prominent and the skin is thin, loose, and lies in folds. In contrast, the visceral protein compartment is relatively spared, a fact that often is reflected by a normal serum albumin level, which in turn sustains normal oncotic pressure in the vascular compartment, thus minimizing edema.

Kwashiorkor

The word kwashiorkor, from the Ga language of West Africa, means “disease of the displaced child” because it commonly was seen after weaning. The presence of peripheral edema distinguishes children with kwashiorkor from those with marasmus and nutritional dwarfism. Children with kwashiorkor also have characteristic skin and hair changes (see sections on hair and skin changes, later). The abdomen is protuberant because of weakened abdominal muscles, intestinal distention, and hepatomegaly, but ascites is rare. The presence of ascites, therefore, should prompt the clinician to search for liver disease or peritonitis. Children with kwashiorkor typically are lethargic and apathetic, but become very irritable when held. Kwashiorkor is not caused by a relative deficiency in protein intake; rather, it most often occurs when there is physiologic stress, such as infection, in an already malnourished child. Because infection or other acute stress usually is present in kwashiorkor, the metabolic aberrations associated with secondary PEM are in play and contractions of the visceral protein compartment are evident. A decrease in serum proteins such as albumin is common, distinguishing it from pure marasmus. Kwashiorkor is characterized by leaky cell membranes, which permit the movement of potassium and other intracellular ions into the extracellular space, causing water movement and edema.

Nutritional Dwarfism

The child with failure to thrive may be of normal weight for height but have short stature and delayed sexual development. Providing appropriate feeding can stimulate catch-up growth and sexual maturation.

The diagnosis of PEM is different in adults than in children, because adults do not grow in height. Therefore, undernutrition in adults causes wasting rather than stunting. In addition, although pure forms of kwashiorkor and marasmus can occur in adults, contemporary studies of adult PEM in industrialized societies typically focus on hospitalized patients with secondary PEM, coexisting illness or injury, and overlapping features of kwashiorkor and marasmus.

PHYSIOLOGIC IMPAIRMENTS CAUSED BY PROTEIN-ENERGY MALNUTRITION

PEM adversely affects almost every organ system, although the brain usually is spared. Almost all the adverse effects are reversible with nutritional restitution. The following discussion is not exhaustive; rather, it emphasizes impairments that commonly translate into overt morbidity or that are important in the diagnosis of PEM. The effects described reflect what occurs in primary PEM; the superimposition of acute illness and secondary PEM often imposes more complexity.

System Effects

Gastrointestinal Tract

Although PEM alone produces adverse effects on gastrointestinal structure and function, diminished stimulation of the gastrointestinal tract by a lack of ingested nutrients has an independent effect. Thus, sustained absence of nutrients passing through the intestine of healthy, nutritionally replete, parenterally fed individuals alone results in functional atrophy of the small intestinal mucosa, as evidenced by a loss of brush border enzymes and diminished integrity of the epithelial barrier. Villus atrophy also may be observed with lack of intestinal stimulation but, in the absence of PEM, the degree of structural atrophy is minor.⁶⁴

The structural and functional deterioration of the intestinal tract, pancreas, and liver from PEM is described best in children. Marked blunting of the intestinal villi is seen and is usually associated with loss of some or all of the brush border hydrolases. Gastric and pancreatic secretions are reduced in volume and contain decreased concentrations of acid and digestive enzymes, respectively. The volume of bile and the concentrations of conjugated bile acids in bile are reduced. Increased numbers of facultative and anaerobic bacteria are found in the upper small intestine, probably explaining the increased proportion of free bile acids in the intestinal lumen. Malabsorption of carbohydrates, fats, and fat- and water-soluble vitamins may occur, and the degree of steatorrhea is proportional to the severity of the PEM, creating a vicious cycle of further malnutrition. The protuberance of the abdomen that is sometimes evident in advanced malnutrition is thought to arise in part from intestinal hypomotility and gas distention.

Cardiovascular System

Moderate to severe PEM produces quantitative and qualitative declines in the cardiac muscle. Myocardial mass is diminished, although proportionately less than the loss in body weight. Myofibrillar atrophy, edema and, less commonly, patchy necrosis and infiltration with chronic inflammatory cells are seen in the myocardium. These structural changes are associated with impaired myocardial performance. A decrease in stroke volume, cardiac output, and maximal work capacity may be observed, and are most evident under conditions of increased demand. Such functional impairments are sometimes accompanied by bradycardia and, in conjunction with the factors noted, can lead to low blood pressure.

Immune System

The immune system is the most vulnerable to PEM, which explains why several functions of immunity are used diagnostically as indicators of malnutrition, such as total lymphocyte count and delayed skin hypersensitivity. The functional integrity of T lymphocytes, polymorphonuclear leukocytes, and complement is uniformly blunted, whereas impaired B lymphocyte production of antibodies is variably affected. A moderately to severely malnourished patient is an immunocompromised individual. Malnutrition leads to increased susceptibility to infection, which in turn promotes the development of PEM; thus, a vicious cycle is created.

Respiratory System

The diaphragm and other respiratory muscles undergo structural and functional atrophy, diminishing inspiratory and expiratory pressures and vital capacity. These changes in muscular performance, in conjunction with blunted ventilatory drive, impair the ability to sustain ventilation in the severely malnourished individual. In patients with a tracheostomy, adherence of bacteria to the tracheal epithelium correlates with the severity of PEM, exacerbating other compromises in the immune system described earlier.

Endocrine System

Although alterations in hormones are common in PEM, many of the changes can be perceived as serving adaptive functions. The inadequate intake of food leads to a decrease in the availability of circulating glucose and amino acids, low circulating levels of insulin, and increased levels of growth hormone. These alterations, in conjunction with the decreased levels of somatomedins and increased levels of cortisol in PEM, promote skeletal muscle catabolism and, at the same time, enhance incorporation of the liberated amino acids into visceral organs. Urea synthesis is inhibited, decreasing nitrogen loss and enhancing the reutilization of amino acids. The enhancement of lipolysis and gluconeogenesis provides a substrate for energy needs.

Serum levels of triiodothyronine (T₃) and thyroxine (T₄) commonly are decreased in conjunction with increased concentrations of reverse T₃, resembling the pattern observed in the euthyroid sick syndrome. The decreased concentration of T₃ may play a role in decreasing REE and the protein catabolic rate that is observed in primary PEM.

Primary gonadal dysfunction is common in adults with moderate to severe PEM, and results in impaired reproductive potential. Decreased circulating levels of testosterone in men and estrogen in women is evident, and amenorrhea is common. Delayed puberty, or loss of menstrual periods, most often occurs when lean body mass drops below a critical threshold. These changes also can be considered physiologic adaptations, because ensuring immediate survival is more critical than the need for sexual maturation in the child or reproduction in the adult.

Other Effects

Wound Healing

Well-nourished individuals lay down more collagen at the site of a surgical wound than those with even mild malnutrition. Furthermore, nutritional repletion of the malnourished patient before surgery leads to better wound healing than if nutritional needs are addressed only postoperatively.

Skin

Undernutrition often causes dry, thin, and wrinkled skin, with atrophy of the basal layers of the epidermis and hyperkeratosis. Severe malnutrition may cause considerable depletion of skin protein and collagen. Patients with kwashiorkor experience sequential skin changes in different areas. Hyperpigmentation occurs first, followed by cracking and stripping of superficial layers, leaving behind hypopigmented, thin, and atrophic epidermis that is friable and easily macerated.

Hair

Scalp hair becomes thin, sparse, and is easily pulled out. In contrast, the eyelashes become long and luxuriant and there may be excessive lanugo in children. The hair of children with kwashiorkor develops hypopigmentation, with reddish-brown, gray, or blond discoloration. Adults may lose axillary and pubic hair.

Kidneys

Renal mass and function often are well preserved during undernutrition. When malnutrition is severe, however, there are decreases in kidney weight, glomerular filtration rate, ability to excrete acid and sodium, and ability to concentrate urine. Mild proteinuria also may occur.

Bone Marrow

Severe undernutrition suppresses bone marrow red blood cell and white blood cell production, leading to anemia, leukopenia, and lymphocytopenia.

NUTRITIONAL ASSESSMENT TECHNIQUES

Overview

The purpose of nutritional assessment is to identify PEM and other nutritional deficits, even when they are not readily discernible. PEM can be subtle, but most cases are detected when a systematic nutritional assessment is performed. An example of subtle but clinically significant PEM is found in Child-Turcotte-Pugh class A alcoholic cirrhotics. These individuals usually appear well nourished. Indeed, one criterion to determine class A status is a normal serum albumin level. Studies of whole-body nitrogen by in vivo neutron activation analysis, however, have demonstrated that more than half of these class A individuals have less than 80% of expected total body protein,⁶⁵ the threshold level below which patients have increased morbidity associated with malnutrition.⁵⁴

In otherwise healthy people and in those who are chronically ill, PEM usually is defined by comparing an anthropometric measurement, such as weight for height, to established normative standards (see Table 4-6). In contrast, there is no gold standard to define and measure PEM in the acutely ill patient because most parameters used to assess PEM in otherwise healthy persons are altered by illness; weight and the concentration of serum proteins are prime examples.

Despite the inaccuracies inherent in assessing PEM in acutely ill individuals, however, the usefulness of nutritional assessment in this setting has been demonstrated repeatedly. Acutely ill patients who are malnourished sustain higher rates of malnutrition-related morbidity. The presence of PEM, therefore, has a predictive value. Even more importantly, identification of malnourished patients and appropriate nutritional intervention is likely to improve clinical outcome.⁶⁶^–⁷³ Meta-analyses have underscored the importance of performing objective nutritional assessments to categorize inpatients, because individuals who are well-nourished or mildly malnourished seem to realize little benefit from intensive nutritional support.⁷¹^–⁷³

Specific Tools for Assessing Protein-Energy Malnutrition

A comprehensive nutritional assessment requires a history, physical examination, evaluation of anthropometrics or functional measures of nutritional status, and a panel of laboratory blood tests. Some of the more commonly used assessment tools are weight, height, and other anthropometric measures, including skinfold thickness and midarm measurements, functional measures (e.g., hand grip strength or skin testing for evidence of anergy), serum protein concentrations, (e.g., albumin or prealbumin), complete blood count, including absolute lymphocyte count, and 24-hour urinary creatinine and urea nitrogen levels. Less readily available measures of body composition such as bioelectrical impedance and total body potassium can be helpful in the appropriate setting. Some of these measures lack a high degree of specificity but continue to be useful in clinical care because of their prognostic significance. Because no single parameter is sufficiently sensitive or specific to assess PEM, these tools are most effective when used in combination. If the clinician is undertaking a nutritional assessment solely to determine whether a patient falls into a category of moderate to severe PEM, far less than comprehensive assessment usually will suffice; such simple means of categorization are provided in the following sections.

History

Weight Loss

Unintentional weight loss associated with illness is the single most practical predictor of a clinically significant degree of PEM. It is useful to quantify such loss by determining whether the patient has sustained a mild (<5%), moderate (5% to 10%), or severe (>10%) degree of loss over the preceding six months. Because acute illness incites a disproportionately large loss of lean mass, it is not surprising that a more than 10% unintentional loss in body weight usually translates into a 15% to 20% decrease in total body protein.⁵⁴ This degree of unintentional loss is an important threshold because it is associated with impaired physiology, a poor clinical outcome, and extended hospitalization⁷⁴^–⁷⁶; it also defines those individuals who will likely benefit from intensive nutritional support. The clinician should nevertheless be mindful that determination of the magnitude of weight loss by history has limited accuracy; one study has found that one third of patients with true weight loss go undetected by history and one quarter of those who had been weight-stable are miscategorized as having undergone weight loss.⁷⁷ Furthermore, the nutritional significance of changes in body weight can be confounded by changes in hydration status and extracellular fluid accumulation. Because weight loss is an imperfect indicator of PEM, it is useful to obtain other historical clues that can contribute to the identification of these patients (see later).

Food Intake

Has there been a change in habitual diet pattern (number, size, and contents of meals)? What is the reason for altered food intake (e.g., change in appetite, mental status or mood, ability to prepare meals, ability to chew or swallow, gastrointestinal symptoms)?

Evidence of Malabsorption

Are there symptoms that are consistent with malabsorption (e.g., greasy, fatty stools, osteopenia, easy bruisability)?

Evidence of Specific Nutrient Deficiencies

Are there symptoms of specific nutrient deficiencies, including macrominerals, micronutrients, and water? (See Tables 4-9 to 4-11.)

Influence of Disease on Nutrient Requirements

Is the nature of the patient’s underlying illness one that is likely to increase nutrient needs or nutrient losses?

Functional Status

Has the patient’s ability to perform daily activities that determine consumption of wholesome meals changed? Can the patient still shop and prepare meals? Have finances interfered with the ability of the patient to purchase food?

Physical Examination

Hydration Status

The patient should be evaluated for signs of dehydration (e.g., hypotension, tachycardia, postural changes, mucosal xerosis, decreased axillary sweat, dry skin), and excess body fluid (e.g., edema or ascites).

Tissue Depletion

A general loss of adipose tissue can be assumed if there is well-defined bony, muscular, and venous outlines and loose skin folds. A fold of skin, pinched between the forefinger and thumb, can reveal the adequacy of subcutaneous fat. The presence of hollowness in the cheeks, buttocks, and perianal area suggests body fat loss. An examination of the temporalis, interosseous, and quadriceps muscles should be made to judge muscle wasting.

Muscle Function

Strength testing of individual muscle groups can be performed to determine if there is generalized or localized muscle weakness. In addition, an evaluation of myocardial function and of respiratory muscle function using spirometry can be done.

Specific Nutrient Deficiencies

Rapidly proliferating tissues (e.g., oral mucosa, hair, skin, gastrointestinal epithelium, bone marrow) often are more sensitive to nutrient deficiencies than tissues that turn over more slowly (e.g., heart, skeletal muscle, brain; see Tables 4-9 to 4-11).

Anthropometry

Anthropometric techniques are those in which a quantitative measure of the size, weight, or volume of a body part is used to assess protein and calorie status. Historically, one of the most commonly used anthropometric parameters has been weight for height. This is a useful parameter when neither the patient nor family can provide reliable historical information, but is less desirable than a history of unintentional weight loss because it requires the patient’s weight to be judged against a normative standard that has been established in a large control population, and interindividual variability in the population limits the accuracy with which this method correctly predicts PEM in a single person. Table 4-6 displays the 1959 Metropolitan Health desirable body weights that were established with prospective mortality data. The 1959 tables remain preferable to the 1983 tables because of certain assumptions made in the construction of the latter. In the context of the Metropolitan tables, desirable weight for height is defined as that figure associated with maximal longevity. Generally speaking, individuals whose weight is less than 85% of the standard can be considered to have a clinically significant degree of PEM. Of note is that desirable weights in this table are substantially less than average weights in North America.

Body mass index (Table 4-18), defined as weight (in kilograms) divided by height (in meters squared), has been supplanting the use of weight for height, in part because it precludes the need to use normative data tables. BMIs that are outside the desirable range (18.5 to 24.9 kg/m²) help identify patients at increased risk of adverse clinical outcomes. A BMI modestly above the desirable range has been shown to be predictive of adverse outcomes in the surgical management of many diseases⁷⁸^–⁸⁰ and in the medical management of conditions such as alcoholic liver disease.⁸¹ Similarly, a low BMI has been shown to be a robust independent risk factor in surgical and medical patients.⁸² Adult patients who are extremely underweight (BMI < 14 kg/m²) are at high risk of death and should be strongly considered for admission to the hospital for the initiation of intensive nutritional support.

Table 4-18 Classification of Nutritional Status by Body Mass Index in Adults

BODY MASS INDEX (KG/M²)	NUTRITIONAL STATUS
<16.0	Severely malnourished
16.0-16.9	Moderately malnourished
17.0-18.4	Mildly malnourished
18.5-24.9	Normal
25.0-29.9	Overweight
30.0-34.9	Obese (class I)
35.0-39.9	Obese (class II)
≥40	Obese (class III)

The BMI, however, like weight for height, is a surrogate and imperfect measure of body composition. A low BMI (<18.5 kg/m²) is interpreted as an indication of PEM and a high BMI (>24.9 kg/m²) is interpreted as excessive fat mass (overweight or obesity). Although BMI is accurate in this regard for the vast majority of adults, it can be just as misleading as any other measure that relies on body weight without a direct evaluation of body composition.⁸³ For example, the individual with excessive fluid accumulation, where the actual fat and body cell mass are less than that implied by the BMI, and in the muscle-bound athlete, where a high BMI is indicative of an extraordinarily large lean mass, are two examples in which the underlying assumptions inherent in the BMI are false. Gender and race also are confounding variables, although the differences are clinically irrelevant; more important are the remarkable changes in body composition that accompany development, making the interpretation of BMI in childhood and adolescence very complex.⁸⁴

It should be apparent from this discussion that the measurement of relevant body compartments, such as the fat mass or fat-free mass, can reveal important information about nutritional status that is often obscured by measurement of weight alone. Underwater (hydrostatic) weighing, dual energy x-ray absorptiometry (DEXA), air impedance plethysmography, total body potassium, isotopically labeled water dilution, in vivo neutron activation analysis, computed tomography (CT), and magnetic resonance imaging (MRI) are accurate noninvasive or minimally invasive techniques of measuring body compartments.⁸⁵^–⁹¹ All are highly effective but because of their expense, lack of accessibility, and impracticality, their use is relegated largely to the sphere of clinical research. A detailed understanding of these tools is beyond the scope of this chapter; however, the primary use of each, with reference to detailed reviews, is outlined in Table 4-19.

Table 4-19 Advanced Techniques for Measurement of Body Compartments

TECHNIQUE	PRIMARY USE IN BODY COMPARTMENT ANALYSIS
Underwater (hydrostatic) weighing⁸⁴	Proportion of body composed of FMProportion of body composed of LM
Air displacement plethysmography⁸⁵	Proportion of body composed of FMProportion of body composed of LM
Dual energy x-ray absorptiometry⁸⁶	Absolute FM and LM; bone density
Total body potassium⁸⁷	Body cell mass
Isotopically labeled water and NaBr dilution⁸⁸	TBW, ICW, ECW
In vivo neutron activation analysis⁸⁹	Total body protein, absolute FM, absolute LM
Computed tomography⁸⁴	Regional FM/LM
Magnetic resonance imaging⁹⁰	Regional FM/LM

ECW, extracellular water; FM, fat mass; ICW, intracellular water; LM, lean mass; NaBr, sodium bromide; TBW, total body water.

In the clinical setting, simple but less accurate techniques are used to assess body compartments. An approximate measure of whole-body fat mass can be derived from assessing the thickness of the subcutaneous fat, which, in a normally proportioned adult, contains approximately half of the body’s adipose stores. The triceps and subscapular sites are used most commonly for this purpose, and it is best to use the sum of the triceps and subscapular folds because sizable interindividual differences exist in fat distribution. Furthermore, as total body fat changes, the subcutaneous fat at each site responds in a different manner. Similarly, midarm muscle circumference (MAMC) provides a measure of skeletal muscle mass. Table 4-20 contains guidelines for interpretation of skinfold and midarm muscle area based on data from the first two National Health and Nutrition Examination Surveys (NHANES I and II).⁹²

Table 4-20 Normative Standards for Upper Arm Muscle Area and Sum of Triceps and Subscapular Skinfolds

The clinical use of skinfolds and appendicular muscle area has distinct weaknesses. As was true for the weight for height tables, there is considerable interindividual variation in values, so these measurements are more useful in population studies than in an individual. Moreover, these measures are highly operator-dependent.⁹³ Also, although the updated databases defining normative values no longer contain the biases of race and age of older versions, correction factors for hydration and physical activity still are not available.

In practice, I have found the most useful clinical role for the measurement of skinfolds and muscle area is in tracking patients with serial measurements over time as a means of monitoring their recovery from disease or response to a clinical intervention. In this manner, the patient is being compared with himself or herself rather than with some normative value. In gastroenterology, the use of skinfolds and muscle area has been of particular usefulness in the assessment and management of cirrhotic patients because cirrhosis corrupts almost all the other common measures of nutritional status. Abnormally low values for triceps skinfold (TSF) and MAMC are independent predictors of mortality in cirrhotics and their incorporation into a Cox regression model improves the prognostic value of the Child-Turcotte score.⁹⁴ Also, when patients with severe alcoholic hepatitis are treated with anabolic steroids, improvements in MAMC and other measures of the fat-free mass (FFM) correlate with a positive response to treatment.⁹⁵

Interest continues in bioimpedance analysis (BIA) as an inexpensive, relatively easy, noninvasive, and safe means of assessing FFM, body cell mass (BCM), and total body water (TBW). BCM is sometimes perceived as a more important measure of lean mass than FFM because it does not include nonliving lean mass, such as the blood plasma and bone mineral (see Fig. 4-1). Resistance to electrical flow through the body is measured, which is proportional to fat and bone mineral content because these body components have poor conductivity. Because all other components of the body are suffused with electrolyte-laden water that readily conducts an electrical current, calculations of TBW, FFM, and BCM can be made if one abides by some general assumptions that define the water content of each compartment. Those with expertise in its use have found it useful for monitoring the FFM in outpatient studies of renal dialysis and human immunodeficiency virus (HIV) patients.^96,⁹⁷ Acute illness, however, produces shifts in the total amount of body water and its distribution in the different compartments, rendering the technique largely worthless in the inpatient setting.⁹⁸ Furthermore, the algorithms used to calculate body composition contain assumptions about body water that can change with age, obesity, and disease, so BIA must be revalidated within any population in which it is used.

Functional Measures of Protein-Calorie Status

Three different techniques have been developed that exploit the fact that skeletal muscle function is impaired in PEM: fist-grip dynamometry (FGD), adductor pollicis electrical stimulation, and respiratory muscle strength evaluation. FGD uses a hand-held dynamometer to measure the maximal fist-grip force that can be elicited. When examined as a surrogate measure of total body protein in patients awaiting gastrointestinal surgery, FGD correlated strongly with in vivo neutron activation analysis and with MAMC.⁹⁹ Similarly, FGD is excellent for detecting depleted body cell mass in cirrhotic patients,¹⁰⁰ a group in which it is notoriously difficult to perform nutritional assessment. As noted, valid indicators of moderate to severe PEM are strong predictors of clinical outcome in acutely ill patients and FGD is effective in this regard. Preoperative patients whose fist grip strength is less than 85% of age-and gender-corrected standards have a twofold increased risk of perioperative complications compared with those whose FGD is normal.¹⁰¹ In patients undergoing surgery for gastrointestinal cancers, FGD had superior sensitivity and specificity in predicting perioperative morbidity and mortality than a widely used discriminant analysis called the prognostic nutritional index.¹⁰² FGD holds considerable promise for the rapid and convenient assessment of protein-calorie status in inpatients and outpatients. The technique is nevertheless limited by its need for an alert and cooperative patient.

Respiratory muscle strength, typically measured with a bedside spirometer as maximal sustained inspiratory force and/or maximal sustained expiratory force, also has been used as a measure of protein-calorie status,¹⁰³ but is generally considered not reliable because too many non-nutritional factors may alter its measurement.

Electrical stimulation of the ulnar nerve with measurement of various functions of the adductor pollicis muscle also has been explored as an indicator of protein-calorie status. The technique has been shown to distinguish between malnourished subjects and well-nourished controls,¹⁰⁴ and to predict postoperative complications in patients awaiting surgery for gastrointestinal malignancies¹⁰⁵; abnormal muscle function was observed to resolve with nutritional restitution. Use of this method in an ICU setting, however, was found to be very problematic¹⁰⁶ because patients found it uncomfortable and neuromuscular blocking agents interfered with obtaining adequate data. Much more work is needed to establish this method’s validity and appropriate applications.

Although PEM adversely affects the physiology of almost all organ systems, the immune system is in the most sensitive; delayed hypersensitivity skin testing, which assesses the integrity of cell-mediated immunity, has been used most often in this regard. In critically ill patients, skin testing has value in predicting mortality,¹⁰⁷ but its interpretation is fraught with confounding variables, such as older age, systemic infection, and major surgery, each of which will independently depress reactivity. Furthermore, reactivity improves in an unpredictable manner with nutritional restitution, so it is not useful for monitoring the progress of patients.¹⁰⁸ The value of skin testing in assessing nutritional status is used best as part of an array of parameters that collectively assess nutritional status (see later, “Discriminant Analyses of Protein-Calorie Status”).

Biochemical Measures of Protein-Calorie Status

Serum Proteins

The serum concentrations of several proteins that are synthesized in the liver are used as indicators of protein-calorie status—albumin, prealbumin (transthyretin), transferrin, and retinol-binding protein (RBP; Table 4-21). A low concentration of any of these proteins strongly suggests the presence of PEM in an individual who does not have a concurrent illness or injury. Because the half-lives of prealbumin, transferrin, and RBP are considerably shorter than that of albumin, it follows that changes in nutritional status will be reflected more promptly in levels of these three than in albumin.

Table 4-21 Hepatically Synthesized Proteins Used for Assessment of Nutritional Status

A variety of factors alter the serum concentration of each of these proteins. For example, prealbumin levels often are elevated in chronic kidney disease or by glucocorticoid or oral contraceptive administration. The degree to which serum levels of all of these proteins are decreased in cirrhosis increases incrementally with worsening grades of the Child classification, although even patients who are Child class A have a small decrease in albumin compared with healthy individuals.¹⁰⁹

All these proteins behave as negative acute-phase reactants—that is, their serum concentrations drop in response to systemic inflammation, roughly proportional to the magnitude of the inflammatory response. This effect severely curtails their reliability as indicators of PEM in the acutely ill patient.¹¹⁰ Nonetheless, with proper respect for their limited accuracy, they still can be useful; for example, prealbumin has been shown to be an efficient rapid means of screening inpatients for PEM on hospital admission.¹¹¹

Creatinine-Height Index

The amount of creatinine excreted in the urine over a 24-hour period, corrected for the patient’s height, is an excellent means of assessing total skeletal muscle mass. The relationship holds because a relatively constant percentage (~2%) of muscle creatine is converted to creatinine each day. Values that are more than 20% below gender-and-height-adjusted normative values are indicative of moderate to severe PEM (Table 4-22). Updated normative creatine-height index (CHI) values for children aged 3 to 18 years are available.¹¹² In sick persons, the CHI tends to correlate with simple measures such as unintentional weight loss, as well as with highly accurate measures of skeletal muscle, such as DEXA.¹¹³ However, incomplete urine collection, abnormal or unstable renal function, excessive meat or milk ingestion immediately preceding or during the collection, and glucocorticoid administration all can alter creatinine excretion independently of changes in muscle mass.

Table 4-22 Normative Values for Creatinine Excretion Based on Height

Discriminant Analyses of Protein-Calorie Status

As noted, many of the parameters used to measure PEM also can predict important clinical outcomes; each parameter, however, has its own limitations. Multifactorial indices that incorporate various combinations of these parameters have been developed through the use of discriminant analyses. By combining the strengths of several parameters, the goal is to arrive at a prognostic index that is more accurate for determining whether patients have a substantial degree of PEM and for optimizing the ability to predict which patients will have adverse clinical outcomes because of PEM, thereby identifying those who might benefit from intensive attention to their nutritional needs. Table 4-23 lists the prognostic indices that have been most widely studied, along with the outcomes that each has been shown to predict. Because many of the parameters incorporated into these indices are influenced by the severity of disease and by nutritional status, such indices are more properly thought of as assessing an integration of illness severity and the likelihood of malnutrition.

Table 4-23 Prognostic Indices in Hospitalized Patients

INDEX	INCORPORATED PARAMETERS	CORRELATES WITH
Likelihood of malnutrition	Serum folate, serum vitamin C, serum albumin, lymphocyte count, hematocrit, triceps skinfold, arm muscle circumference, weight	Duration of hospitalization
Prognostic nutritional index	Serum albumin, serum transferrin, delayed hypersensitivity, triceps skinfold	Frequency of postoperative complications and mortality
Instant nutritional index	Serum albumin, lymphocyte count	Frequency of postoperative infection
Hospital prognostic index	Serum albumin, delayed hypersensitivity, presence of sepsis or cancer	Hospital mortality

Adapted from Mason J, Rosenberg I. Protein-energy malnutrition. In: Isselbacher K, Braunwald E, Wilson J, et al, editors. Harrison’s principles of internal medicine. 13th ed. New York: McGraw-Hill; 1994. p 440.

Rapid Screening Tools for the Assessment of Targeted Populations

Inexpensive, rapid, and convenient means of assessing nutritional status that are accurate in identifying patients with PEM are of great value, particularly when large numbers of people need to be evaluated. Two such tools have been developed and extensively validated, the subjective global assessment (SGA) and the Mini-Nutritional Assessment (MNA).

Subjective Global Assessment

This initially was intended for use in surgical inpatients as a means of assessing nutritional status and predicting postoperative infections; for the latter, it was found to be a better predictor than serum albumin concentration, delayed skin hypersensitivity, MAMC, CHI, and the prognostic nutritional index.¹¹⁴ A focused history and physical examination are used to categorize patients as well nourished (category A), having mild or moderate malnutrition (category B), or having severe malnutrition (category C; Table 4-24). In spite of the subjective nature of some of its components, there is excellent agreement in independent observers.¹¹⁵ The SGA has been shown to be reliable, even in the hands of first-year medical and surgical residents¹¹⁶ and validated as a predictor of clinical outcomes in chronically institutionalized older adults and in patients with a variety of medical conditions.¹¹⁷^–¹¹⁹

Table 4-24 Subjective Global Assessment (SGA) of Nutritional Status

History

Weight change

Loss in past 6 months: amount = __________ kg; % loss = __________

Change in past 2 weeks: __________ Increase __________ No change __________ Decrease

Dietary intake change:

No change ________ Change ________ Duration = ________ weeks

Dietary status:

__________ Suboptimal solid diet

__________ Hypocaloric liquids

__________ Starvation

Gastrointestinal symptoms (that have persisted for >2 weeks):

__________ None __________ Nausea __________ Vomiting __________ Diarrhea __________ Anorexia

Functional capacity:

__________ No dysfunction __________ Dysfunction Duration = __________ Weeks

Type:

__________ Working suboptimally

__________ Ambulatory but not working

__________ Bedridden

Effect of disease on nutritional requirements:

Primary diagnosis: __________

Metabolic demand: ________ Low stress ________ Moderate stress __________ High stress

Physical Examination (Normal, Moderate, or Severe)

__________ Loss of subcutaneous fat (triceps, chest)

__________ Muscle wasting (quadriceps, deltoids)

__________ Ankle or sacral edema

__________ Ascites

SGA Rating*

A = Well nourished

B = Mild or moderate malnutrition

C = Severe malnutrition

* The ranks of A, B, and C in the SGA are assigned on the basis of subjective weighting. A patient with weight loss and muscle wasting who is currently eating well and gaining weight is classified as well nourished. A patient with moderate weight loss (between 5% and 10%), continued compromise in food intake, continued weight loss, progressive functional impairment, and moderate stress because of illness is classified as moderately malnourished. A patient with severe weight loss (>10%), poor nutrient intake, progressive functional impairment, and muscle wasting usually is classified as having severe malnutrition.

Mini-Nutritional Assessment

This was developed as a rapidly administered screen to detect PEM in geriatric populations. A combination of history, brief physical examination, and simple anthropometrics (BMI, arm and calf circumference) can be obtained in a few minutes. Subjects receive a score that classifies them as being nourished, malnourished, or at risk of malnutrition. The MNA is a valid means of detecting PEM in older adults who are generally healthy and ambulatory as well as those who are frail and institutionalized ^120,¹²¹; in the chronically institutionalized, it possesses considerable predictive value in projecting future morbidity.¹²² One disadvantage of the MNA is that it does not screen for overweight or obesity. Other screening tools designed for geriatric populations, such as the Nutrition Screening Initiative, have the ability to screen for under- and overnutrition but have not been as extensively validated as the MNA.¹²³

AGGRESSIVE NUTRITIONAL SUPPORT IN THE HOSPITALIZED PATIENT

Aggressive nutritional support, here defined as using whatever means is necessary and practical to meet the nutritional needs of the patient, will not benefit every acutely ill patient. In practice, any acutely ill patient who has moderate to severe malnutrition and is unlikely to be able to meet his or her own nutritional needs within 48 hours is a strong candidate for aggressive nutritional support. Another common indication for aggressive nutritional support is when a well-nourished or mildly malnourished inpatient is judged to be unlikely to meet at least 80% of his or her projected calorie or protein goals for the coming 10 days. However, to date, there is insufficient evidence-based science to prove efficacy of this latter indication.

Catabolic forces that accompany acute illness make it difficult to correct nutritional deficits. In those with a high degree of sustained metabolic stress, nutritional support generally will not lead to an increase in the protein compartment of the body. Moreover, a gain in weight may not occur and, when it does, much of the initial gain is from water and an expanded fat mass.¹²⁴ Despite these limitations, however, even in the absence of weight gain or increases in serum protein levels, a course of nutritional support for an appropriate patient can improve physiologic functions and clinical outcome.¹²⁵

The following sections cite some clinical scenarios particularly relevant to gastroenterology for which there exists compelling clinical research that aggressive nutritional support provides benefit to the inpatient.

MALNOURISHED PATIENTS UNDERGOING MAJOR SURGERY

Nutritional support can be beneficial for moderately to severely malnourished patients who are scheduled to undergo major surgery. Aggressive nutritional support for seven or more days before surgery reduces perioperative complications, and sometimes mortality, in malnourished patients.⁶⁶^–⁷³^,¹²⁶^–¹²⁸ In the VA Cooperative Trial,⁶⁶ which encompassed almost 500 subjects about to undergo major abdominal or thoracic surgery, patients who were categorized as severely malnourished and randomized to receive preoperative TPN realized an almost 90% decrease in noninfectious perioperative complications. No benefits were observed in mildly malnourished or well-nourished individuals. In trials of moderately to severely malnourished patients, preoperative nutrition support generally conveys sizeable benefits: one trial that enrolled 90 patients with gastric or colorectal cancers undergoing surgery demonstrated a 35% decline in overall complications and a significant reduction in mortality.¹²⁸ The observation that the benefits of preoperative nutritional support are confined to those with a substantial degree of malnutrition is the same conclusion reached by meta-analyses.^71,⁷² Deferring aggressive nutritional support until after surgery does not appear to have the same ability to diminish perioperative complications. ¹²⁹

Provision of nutrients via an enteral approach also is beneficial. There have been fewer trials done of preoperative enteral support than of preoperative TPN, but it appears that preoperative enteral support conveys the same nutritional ¹³⁰ and clinical¹³¹ benefits as TPN. As with TPN, postoperative enteral nutrition in the absence of aggressive preoperative support is less likely to convey benefit to the patient.¹³²

PATIENTS HOSPITALIZED WITH DECOMPENSATED ALCOHOLIC LIVER DISEASE

The prevalence of moderate to severe PEM is so high in patients admitted for acute alcoholic hepatitis and other forms of decompensated alcoholic liver disease ⁶⁵ that it is best to assume that all such patients are malnourished. Furthermore, patients with acute alcoholic hepatitis usually fall far short of their nutritional needs when allowed to eat freely. Clinical trials have demonstrated that the rates of morbidity, mortality, and the speed of recovery are improved with prompt institution of enteral or parenteral nutrition in these patients.⁶⁷^–⁶⁹ ¹³³

PATIENTS UNDERGOING RADIATION THERAPY

The usefulness of aggressive nutrition support in patients undergoing radiation therapy has been studied most extensively in those who have head and neck and esophageal cancers. There is now reasonable evidence in these patients that placement of a percutaneous endoscopic gastrostomy (PEG) tube and administration of supplemental tube feedings during and after the course of radiation therapy prevents further deterioration of nutritional status.^134,¹³⁵ In patients with head and neck cancers, supplemental PEG feedings also have been shown to improve quality of life (QOL). Although improvements in survival or decreased morbidity have not yet been demonstrated, the improved QOL alone may warrant its use in this setting.

REFEEDING SYNDROME

Injudicious and overly aggressive nutritional therapy can have adverse clinical consequences, known as the refeeding syndrome, in those who have suffered sustained inadequacy in food intake.^136,¹³⁷ Early evidence of the refeeding syndrome was reported at the end of World War II, when it was found that refeeding chronically semistarved war victims and research volunteers caused cardiac insufficiency and neurologic complications. More recently, refeeding abnormalities and serious complications have been reported after aggressive refeeding of hospitalized cachectic patients.¹³⁸ The likelihood that a refeeding syndrome will occur depends on the severity of underlying PEM and the rapidity with which nutritional restitution is undertaken. Thus, the syndrome usually can be avoided merely by approaching nutritional repletion in an incremental fashion over the first several days of nutritional therapy and careful monitoring for potential complications, with interventions as indicated.

MINERAL DEPLETION

Of the mineral abnormalities associated with refeeding, phosphate depletion has received the most attention. Refeeding with enteral carbohydrates or glucose-based parenteral formulas stimulates insulin release and intracellular uptake of phosphate.¹³⁹ This phenomenon can cause the extracellular phosphorus concentration to fall below 1 mg/dL within hours of initiating nutritional therapy if adequate phosphate is not given. Severe hypophosphatemia, accompanied by muscle weakness, paresthesias, seizures, coma, cardiopulmonary decompensation, and death, has occurred in patients receiving enteral or parenteral nutritional repletion.^138,¹⁴⁰

Potassium and magnesium are the most abundant intracellular cations. Loss of body cell mass in the malnourished patient causes whole-body potassium and magnesium depletion. However, serum concentrations of these two electrolytes remain normal or near-normal during starvation because of their release from tissue and bone stores. The increases in protein synthesis rates, body cell mass, and glycogen stores during refeeding require increased intracellular potassium and magnesium. In addition, hyperinsulinemia during refeeding increases the cellular uptake of potassium and can cause a rapid decline in its intravascular concentration.¹⁴¹

CARDIOVASCULAR COMPLICATIONS

Chronic undernutrition is associated with decreased cardiac mass, stroke volume, and end-diastolic volume, bradycardia, and fragmentation of cardiac myofibrils.¹⁴²^–¹⁴⁵ In addition, carbohydrate refeeding increases the concentration of circulating insulin, which enhances sodium and water reabsorption by the renal tubule.¹⁴⁶ Moreover, if hypophosphatemia develops, it can impair oxygen unloading from red blood cells and increase susceptibility to ventricular tachyarrhythmias. A serum phosphate concentration less than 2.0 mg/dL further impedes myocardial performance. These factors put the severely malnourished patient at increased risk of fluid retention and congestive heart failure after nutritional therapy, and may precipitate cardiovascular collapse. The cardiovascular complications of thiamine deficiency (wet beriberi) also have been reported to contribute to cardiovascular compromise in the refeeding syndrome.¹⁴⁷ Because thiamine is an essential cofactor for several enzymes in carbohydrate metabolism, it is believed that overzealous administration of carbohydrate to a severely malnourished patient with limited thiamine availability may precipitate an acute thiamine deficiency state. The high concentrations of glucose in TPN seem particularly likely to produce this effect. Ventricular tachyarrhythmias, which can be fatal, occur during the first week of refeeding,¹⁴⁸ and may be preceded by prolongation of the QT interval.

GLUCOSE INTOLERANCE

The adaptive changes during starvation enhance the use of FAs and ketone bodies for fuel while glucose is conserved. In addition, the ability of insulin to stimulate glucose uptake and oxidation by peripheral tissues is impaired.¹⁴¹ Thus, refeeding with high-carbohydrate meals or large amounts of parenteral glucose may not be well tolerated initially and may produce marked elevations in the blood glucose level, glucosuria, dehydration, and hyperosmolar coma.¹⁴⁹

GASTROINTESTINAL DYSFUNCTION

As noted, starvation and malnutrition cause structural and functional deterioration of the gastrointestinal tract, thereby limiting digestion and absorption.^150,¹⁵¹ When malnutrition is severe, oral refeeding has been associated with an increased incidence of diarrhea and death; most of the adverse gastrointestinal consequences of starvation resolve after one to two weeks of refeeding.

CLINICAL RECOMMENDATIONS

The severity of complications during the refeeding of cachectic, chronically semistarved patients dictates caution, particularly during the first week of therapy, when the risk of complications is highest. A careful search for cardiovascular and electrolyte abnormalities should be performed before refeeding.

Judicious resuscitation with fluids and electrolytes may be necessary before beginning feedings to prevent congestive heart failure. Vitamin supplementation should be given routinely. Severely malnourished patients are poikilothermic, so a warm ambient temperature and warming blankets may be necessary to increase the core temperature slowly.

Oral or enteral tube feedings are preferred over parenteral feeding for nutritional restitution because of their fewer serious complications and enhanced gastrointestinal tract recovery. Isotonic feedings should be given in small amounts at frequent intervals to prevent overwhelming the body’s limited capacity for nutrient processing and to prevent hypoglycemia, which can occur during brief nonfeeding intervals. Parenteral supplementation or TPN may be necessary if the intestine cannot tolerate oral or enteral feeding.

Although it is impossible to know the precise nutrient requirements of individual patients, some general guidelines are recommended for the first week of refeeding. Fluid intake for adults initially should be limited to approximately 800 mL/day plus replacement for insensible losses; adjustments are needed in fluid-overloaded or dehydrated patients. Changes in body weight are a useful guide to evaluate the efficacy of fluid administration. Weight gain of more than 0.25 kg/day, or 1.5 kg/week, probably represents fluid accumulation. Daily calorie intake should be approximately 15 to 20 kcal/kg and contain approximately 100 g of carbohydrate and 1.5 g of protein/kg body weight. Sodium should be restricted to approximately 60 mEq or 1.5 g/day, but liberal amounts of phosphorus, potassium, and magnesium should be given to patients who have normal renal function. Daily monitoring of body weight, fluid intake, urine output, and plasma glucose and electrolyte values (including magnesium and phosphate) are critical during early refeeding (first 3 to 5 days) so that appropriate adjustments can be made.

MANAGEMENT OF SEVERE MALABSORPTION: A NUTRITIONAL PERSPECTIVE

The medical management of patients with severe malabsorption, such as that which often occurs in short bowel syndrome or pancreatic insufficiency, can be challenging and frustrating. In many cases, the patient and clinician must be prepared to accept incremental improvements as the management goal rather than complete resolution of symptoms. Great strides and an improved QOL, however, often can be accomplished with only modest improvements in absorptive physiology. Malabsorption syndromes and short bowel syndrome are discussed in greater detail in Chapters 101 and 103, respectively.

CLINICAL CONSIDERATIONS

Initial assessment of the patient with chronic malabsorption is meant to provide a set of rational strategies to improve the patient’s clinical condition and prevent complications. The therapeutic approach depends on the following: the level at which the patient’s intestinal tract functions; the presence of macronutrient, micronutrient, electrolyte, and fluid deficits; the identification of risk factors for future medical complications; the presence of coexisting diseases that hamper the ability to provide nutritional therapy; and an evaluation of factors that affect the patient’s daily activities.

A careful review of medical records, operative reports, and radiologic studies is needed to evaluate the absorptive capacity of the intestine by determining the length and status of the remaining intestine, the site of intestinal disease or resection, and the presence of other conditions that reduce intestinal absorption. An assessment of fluid losses through diarrhea, ostomy output, and fistula volume should be made to help determine fluid requirements. Knowledge of fluid losses also is used to calculate intestinal mineral losses by multiplying fluid loss by the estimated electrolyte concentration in intestinal fluid (Table 4-25). In patients who do not respond to treatment as predicted, dynamic studies of intestinal absorptive function may be helpful for adjusting the treatment program. Such studies may include measuring fat, carbohydrate, or nitrogen balance and evaluating ostomy, fecal, or fistula mineral and fluid losses.

Table 4-25 Electrolyte Concentrations in Gastrointestinal Fluids*

The urgency for medical intervention is determined by the severity of hemodynamic, electrolyte, and nutritional abnormalities. This requires an evaluation for volume depletion, weight loss, and specific nutrient deficiencies. In addition to standard laboratory tests to evaluate whether micronutrient deficiencies are causes of anemia (iron, folate, vitamin B₁₂), prolonged prothrombin time (vitamin K deficiency), and neurologic deficits (vitamin B₁₂, thiamine, vitamin E), more sophisticated measurements to determine vitamin and trace mineral status can be obtained when such deficiencies are suspected clinically. Water-soluble vitamin deficiencies tend to occur as a cluster, as do fat-soluble vitamin deficiencies, so measurements of red blood cell folate and serum 25-hydroxyvitamin D levels are useful as an initial screen because a deficiency of one vitamin raises the suspicion of additional deficiencies.

Bone mineral densitometry may be useful to establish a baseline and to screen for unrecognized bone mineral depletion. Metabolic bone disease is common in chronic malabsorptive disorders, particularly so in individuals on long-term TPN.¹⁵²

An accurate dietary history is important in evaluating nutrient requirements in nutritionally stable patients and for identifying dietary inadequacies in those with nutrient deficiencies. Working in conjunction with a skilled dietitian for the evaluation and treatment phases of management often is invaluable.

TREATMENT

The goals of therapy are to control diarrhea, maintain fluid, electrolyte, and nutritional homeostasis, treat and prevent medical complications, and maximize the QOL. Striving for an optimal QOL rather than insisting on optimal physiology cannot be overemphasized, and often requires tailoring therapy to the specific needs of each patient. For example, maintaining adequate nutritional status with oral feedings at the cost of massive diarrhea and frequent bowel movements may be unacceptable to the patient with an active social or professional life outside the home; in such a patient, parenteral supplementation may be necessary to improve the QOL.

Initial therapy often requires subsequent modification using a trial and error approach because of individual variability in absorptive function, continued intestinal adaptation, and the development of new medical complications or disease progression. Continued clinical monitoring is critical so that medical and nutritional therapy can be adjusted when necessary.

Control of Diarrhea

Diarrhea often is caused by a combination of factors, including increased gastrointestinal secretions, decreased intestinal transit time, and osmotic stimulation of water secretion by unabsorbed contents of the food stream. In addition, unabsorbed bile acids and fatty acids are potent stimuli of fluid and electrolyte secretion by the colonic mucosa. Therefore, therapy for diarrhea involves limiting endogenous secretions, slowing motility, and improving solute absorption.

Normally, the stomach produces approximately 2.5 L of fluid/day, which is absorbed by the small intestine and colon. Gastric secretion therefore may contribute to diarrhea if there is inadequate absorptive capacity or if there is an excess amount of gastric secretion, such as with Zollinger-Ellison syndrome. Immediately following extensive resection of distal bowel, a gastric hypersecretory state occurs, probably from loss of inhibitory peptides such as glucagon-like peptide-2 (GLP-2) and peptide YY ^153,¹⁵⁴; recovery from this hypersecretory state occurs over a period of months. Histamine (H₂) receptor antagonists or proton pump inhibitors are used to reduce gastric secretions, and the presence of acidic jejunostomy or ileostomy contents after meals is a clear indication for acid reduction therapy.¹⁵⁵ Large doses, twice the normal amount used for the treatment of peptic ulcer disease or gastroesophageal reflux, may be required for adequate control in some patients because of reduced drug absorption.

The somatostatin analog, octreotide acetate (Sandostatin), can decrease small intestine secretions. Therapy with octreotide has been shown to decrease ostomy or stool volume, decrease sodium and chloride output, and prolong small intestine transit time in patients with short bowel syndrome.^156,¹⁵⁷ Octreotide therapy, however, usually does not improve absorption of macronutrients and other minerals; in fact, it may exacerbate the degree of fat malabsorption,¹⁵⁸^–¹⁶⁰ presumably by inhibiting pancreatic secretions. In addition, octreotide is expensive, must be given by subcutaneous injection, can decrease appetite and increase the risk of gallstones,¹⁶¹ and diminishes protein synthesis in the intestinal epithelium and exocrine pancreas.¹⁶² Nevertheless, in patients who have persistent large-volume intestinal output despite standard antidiarrheal therapy, a trial of 100 µg of octreotide injected subcutaneously three times daily with meals may be useful. A long-acting form of octreotide, which only requires monthly administration, has not been studied as thoroughly but has been effective in controlling diarrhea in some settings.¹⁶³

Opiates are the most effective means for slowing intestinal motility and act by delaying gastric emptying, decreasing peristalsis of the small intestine and colon, and increasing anal sphincter tone. Loperamide (Imodium) should be tried first because it is metabolized on first pass by the liver and does not easily cross the blood-brain barrier, thereby limiting its side effects and potential for drug dependence. If loperamide is not effective, other opiates such as codeine or deodorized tincture of opium (DTO, 5 to 15 drops every 6 hours) should be considered. In addition, the combination of an anticholinergic drug and an opiate may be beneficial. Capsules containing 25 mg powdered opium and 15 mg of powdered belladonna are a potent combination, although such capsules are not commercially available and require compounding. Diphenoxylate with atropine (Lomotil) is an effective agent, but is rather expensive if large doses are needed. Patients with severe malabsorption often require much larger than conventional doses of antidiarrheal medications. Pushing the dosage to the limits of the therapeutic range, while maintaining diligent attention to the potential for side effects, often is necessary.

Foods and medications that cause diarrhea should be avoided. Traditionally, the recommendation has been to decrease or eliminate lactose-containing foods because of the reduction in intestinal lactase with intestinal resection. Patients with jejunostomies and 15 to 150 cm of jejunum remaining, however, can tolerate 20-g lactose loads as milk or yogurt.¹⁶⁴ Foods that have laxative effects, such as caffeine-containing drinks and diet products containing osmotically active sweeteners (e.g., sorbitol, xylitol, mannitol), should be avoided. Medications that contain magnesium or sorbitol also can contribute to diarrhea.¹⁶⁵ In individuals with short bowel syndrome and uncontrolled diarrhea, soft drinks and fruit juices should be diluted 1 : 1 with water to reduce their high osmotic load.

Enteral Feeding

Patients with nausea, vomiting, abdominal pain, or severe diarrhea may be unable to tolerate enteral feeding, regardless of their intestinal absorptive capacity. Specific foods that cause gastrointestinal complaints should be avoided, but it is important to evaluate the validity of these complaints to prevent the unnecessary withdrawal of nutritious foods.

The goal of feeding is to provide the patient with all recommended nutritional requirements. The amount of ingested nutrients needed to reach this goal depends on the normal RDAs, modified by an estimate of absorptive function and intestinal losses; this usually requires ingestion of large amounts of fluid, calories, protein, vitamins, and minerals. Even in patients with severe short bowel syndrome, TPN may not be needed when vitamin and mineral supplements and large amounts of calories and protein are provided enterally.¹⁶⁶ Increasing the time that food is in contact with the intestine may enhance absorption in patients with limited absorptive function. For this reason, total dietary intake should be divided into at least six small meals daily. If this is unsuccessful, defined liquid formulas ingested between meals or administered by continuous tube feedings at night may preclude the need for TPN. Patients with severe malabsorption often must ingest 40 to 60 kcal/kg/day and 1.2 to 1.5 g of protein/kg/day to avoid malnutrition, but this is highly variable. A more practical goal might be to consume however much is necessary to sustain a stable and reasonable weight while avoiding dehydration, electrolyte disorders, and any micronutrient deficiencies. The macro- and micronutrient needs of each patient often are arrived at only by experimentation with different regimens.

Fat Intake

Fat intake need not be restricted in patients whose colon is not in continuity with the food stream, despite the presence of steatorrhea. Although food preferences and tolerances differ from patient to patient, a high-fat, low-carbohydrate diet is comparable to a low-fat, high-carbohydrate diet with regard to total fluid, energy, nitrogen, sodium, potassium, and divalent ion absorption in patients with short bowel syndrome whose colon is not in continuity.¹⁶⁷^–¹⁶⁹ By retaining fat in the diet, the patient’s ability to achieve caloric goals will be enhanced because of the high caloric density of fat. In contrast, in those who have steatorrhea and a colon in continuity with the food stream, unabsorbed fatty acids are potent stimuli to colonic ion and fluid secretion, and limiting fat intake usually diminishes gastrointestinal symptoms, colonic water secretion, hyperoxaluria, and divalent cation losses.^170,¹⁷¹ Patients who have only a portion of the colon in continuity with the food stream behave in a variable manner in this regard.

Medium-chain triglycerides (MCTs) can be very useful as a feeding supplement in patients who have impaired fat absorption, because MCTs are hydrolyzed rapidly and do not require bile salts and micelle formation for absorption. The efficacy of MCTs, however, usually is confined to patients whose colon is in continuity with the food stream, apparently because the colon is a site for MCT absorption.¹⁷² In contrast, there is usually little improvement in overall energy absorption in patients with jejunostomies or ileostomies. Many patients do not find pure MCT oil to be palatable (used as a salad oil or as a spread on toast), and it also may cause nausea, vomiting, or abdominal discomfort. Unfortunately, liquid nutritional supplements designed for oral intake that contain most of their fat calories as MCT have been discontinued because of low demand. A homemade substitute nevertheless can be readily produced by mixing a package of Carnation Instant Breakfast, 8 ounces of cold skim milk, and 2 teaspoons of MCT in an electric blender for a few seconds.

Predigested Liquid Nutritional Supplements

Predigested formulas—monomeric (also called elemental; e.g., Vivonex T.E.N.) and oligomeric (also called semielemental; e.g., Peptamen, Reabilan)—claim benefit for patients with short bowel syndrome. Theoretically, these formulas, which contain nitrogen in the form of free amino acids or small peptides, are absorbed more efficiently than polymeric formulas or whole food. However, when subjected to clinical trials, they have shown no benefit ¹⁶⁹ or a marginal increase in nitrogen absorption.¹⁷³ At present, there is insufficient clinical evidence to justify the routine use of expensive predigested formulas in patients with short bowel syndrome.

Oral Rehydration Therapy

In many patients with severe malabsorption, maintenance of fluid and electrolyte homeostasis constitutes a major problem. These patients may benefit from oral rehydration therapy (ORT), taking advantage of the sodium-glucose cotransporter in the intestinal brush border.¹⁷⁴ Frequent ingestion of small-volume feedings of an isotonic glucose or starch-based electrolyte solution stimulates active sodium transport across the intestine, and water follows passively by solvent drag. Intestinal perfusion studies and clinical trials have indicated that sodium and water absorption are maximal from solutions containing 60 to 75 mmol/L of sodium with equimolar concentrations of glucose; this concentration is somewhat lower than what had been suggested in the past.^175,¹⁷⁶ Nevertheless, many commercially available oral rehydration formulas and sport drinks contain far lower sodium concentrations than what is necessary to optimize water and electrolyte absorption, and should be avoided by patients with short bowel syndrome. Instead, inexpensive and more effective solutions can be made by patients at home (Table 4-26). Daily oral administration of 1 to 2 L of rehydration solutions has been successful in correcting fluid and electrolyte abnormalities and allows intravenous supplementation to be discontinued in patients who have had extensive intestinal resection.¹⁷⁷

Table 4-26 Characteristics of Selected Oral Rehydration Solutions*

Major Minerals

Calcium and magnesium bind to unabsorbed FAs and may be lost in large quantities in patients with chronic steatorrhea, thus contributing to the frequent occurrence of metabolic bone disease in these individuals. The amount of calcium lost is proportional to the degree of steatorrhea. Magnesium is an integral component of bone mineral and is necessary for the release of parathyroid hormone (PTH) and for its actions on target organs.^178,¹⁷⁹ Hypocalcemia refractory to aggressive calcium replacement can be caused by chronic hypomagnesemia. Like calcium, total body magnesium status is not well reflected in serum levels of the cation. Consequently, the percentage of magnesium excreted in urine after intravenous infusion may be the best index of body magnesium stores in patients with short bowel syndrome; excretion of less than 80% of infused magnesium suggests whole-body magnesium depletion.¹⁸⁰

Maintaining magnesium homeostasis often is difficult because magnesium is poorly absorbed and enteral supplementation with magnesium salts increases diarrhea. Enteric-coated magnesium supplements are not used in the setting of intestinal malabsorption because their delayed release reduces contact with the intestine for absorption.¹⁸¹ Soluble magnesium salts, such as magnesium gluconate, are tolerated and absorbed better than other magnesium complexes. In some patients, magnesium is best given in liquid form as magnesium gluconate (Fleming, St. Louis) and can be added to an oral rehydration solution in doses of 18 to 27 mmol (432 to 648 mg of elemental magnesium) daily. This solution should be sipped, not ingested as a bolus, to maximize absorption and avoid diarrhea.

In patients with steatorrhea and an intact colon, supplemental calcium should be given routinely to improve calcium availability and reduce excessive oxalate absorption. Plasma levels of calcium usually are maintained by mobilizing bone stores unless there is concurrent magnesium or vitamin D deficiency; urinary calcium excretion, which should be greater than 50 mg in 24 hours, is a more reliable index of calcium status. Most patients with fat malabsorption require approximately 1.5 to 2 g of elemental calcium daily.

Trace Minerals

With the exceptions of zinc and iron, absorption of trace minerals from ingested foods or liquid formulas often is adequate to prevent overt deficiency syndromes. In patients with malabsorption, zinc deficiency is common and often subclinical. Large dosages of oral zinc supplements may become necessary, because zinc losses are often high and zinc absorption is low. Zinc gluconate is tolerated well and is less likely than zinc sulfate to cause gastric distress. Zinc should not be given with meals because absorption is reduced by several foods, in particular those high in phytates.¹⁸² Daily zinc supplementation of 25 mg plus an additional 100 mg/L (or 100 mg/kg) of ostomy or diarrheal output is necessary to maintain zinc homeostasis.²⁵ Thus, many patients require approximately 150 mg of elemental zinc daily. Although zinc ingestion reduces copper absorption and can cause clinically significant copper deficiency,¹⁸³ additional copper intake usually is not needed.

Treating iron deficiency with oral preparations can be difficult. A liquid form of ferrous sulfate (300 mg/5 mL containing 60 mg of elemental iron) mixed in orange juice four times daily has been found to be efficacious. Diluting ferrous sulfate liquid prevents staining of teeth, and the ascorbic acid present in orange juice enhances iron absorption. Some patients, however, require intermittent administration of parenteral iron, which should be administered in the form of iron sucrose because it is far less likely to produce an anaphylactoid reaction compared with iron dextran.¹⁸⁴

Vitamins

Patients with fat malabsorption usually can absorb adequate amounts of most water-soluble vitamins from their diet, but have difficulty absorbing fat-soluble vitamins. Vitamin K deficiency rarely is a clinical problem unless patients are receiving antibiotics because colonic bacteria synthesize usable forms of vitamin K (vitamin K₂; multiprenyl menaquinones). Large doses of vitamins A, D, and E may be required, however, to maintain normal concentrations of these vitamins. If provision of more than 5 to 10 times the basal requirement of a fat-soluble vitamin still is not effective, it is worthwhile using the more expensive polar forms of fat-soluble vitamins (see discussion earlier, in “Micronutrients” section). Testing blood levels of these vitamins should be used to determine whether repletion has been achieved.

Parenteral Feeding

Just because a patient requires parenteral support for fluid and electrolytes does not necessarily warrant inclusion of all the other nutrients in the intravenous solution. Some patients may be able to meet their other nutritional needs through an oral or enteral route. Long-term parenteral support with fluids and electrolytes (and sometimes with minerals and vitamins) alone is fraught with fewer infectious and metabolic complications than if macronutrients are included in the parenteral solution.

Some general guidelines are useful in deciding which patients require parenteral therapy. Although a postduodenal intestinal remnant of less than 100 cm, or the lack of an ileocecal valve or a colonic segment in continuity, are each associated with a much greater likelihood of permanent dependence on parenteral support,¹⁸⁵ exceptions abound. Adults in whom urine output is less than 1 L/day are at increased risk for developing renal dysfunction and should receive intravenous fluids. Adequate levels of fat-soluble vitamins and certain minerals (e.g., magnesium, potassium, and zinc) are difficult to maintain with oral feedings in patients with severe steatorrhea or large intestinal fluid output, and they may require parenteral supplementation. Magnesium sulfate can be injected intramuscularly at a dose of 12 mmol (290 mg of elemental magnesium) one to three times/week if attempts at oral therapy are unsuccessful. Intravenous infusion of magnesium is preferred, however, because intramuscular injections are painful and can cause sterile abscesses. In some patients, TPN may be life-saving or may be needed to limit diarrhea and achieve an acceptable QOL.

Nutritional Rehabilitation of the Short Bowel

Whether administration of growth hormone (or, more recently, GLP-2) to patients with short bowel syndrome will enhance the absorption of macronutrients and electrolytes continues to be a matter of controversy,^186,¹⁸⁷ but pharmacologic management is not the focus of this chapter (see Chapter 103). More relevant to this discussion is that some of these studies have included oral glutamine (at a dosage of approximately 30 g/day) as part of the growth hormone regimen, the rationale being that glutamine is a preferred fuel substrate of the enterocyte. There is, however, no controlled evidence that oral glutamine contributes to recovery of absorptive function.¹⁸⁶^–¹⁸⁹

KEY REFERENCES

Duggan C, Fontaine O, Pierce N, et al. Scientific rationale for a change in the composition of oral rehydration solution. JAMA. 2004;291:2628-31. (Ref 176.)

Hamilton C, Seidner DL. Metabolic bone disease and parenteral nutrition. Curr Gastroenterol Rep. 2004;6:335-41. (Ref 152.)

Heyland D, MacDonald S, Keefe L, et al. TPN in the critically ill patient: A meta-analysis. JAMA. 1998;16:2013-19. (Ref 72.)

Hill G. Body composition research: implications for the practice of clinical nutrition. JPEN J Parenter Enteral Nutr. 1992;16:197-218. (Ref 54.)

Hoffer LJ. Metabolic consequences of starvation. In: Shils M, Shike M, Ross AC, et al, editors. Modern nutrition in health and disease. 10th ed. Baltimore: Lippincott Williams & Wilkins; 2006:730. (Ref 44.)

Klein S. The myth of serum albumin as a measure of nutritional status. Gastroenterology. 1990;99:1845-6. (Ref 110.)

Mason JB, Roubenoff R. The unbearable lightness of being … a cirrhotic. Gastroenterology. 1993;105:1911-14. (Ref 133.)

Prijatmoko D, Strauss B, Lambert J, et al. Early detection of protein depletion in alcoholic cirrhosis: role of body composition analysis. Gastroenterology. 1993;105:1839-45. (Ref 65.)

Stanga Z, Brunner A, Leuenberger M, et al. Nutrition in clinical practice—the refeeding syndrome: Illustrative cases and guidelines for prevention and treatment. Eur J Clin Nutr. 2008;62:687-94. (Ref 136.)

van den Berghe G, Wilmer A, Hermans G, et al. Intensive insulin therapy in the medical ICU. N Engl J Med. 2006;354:449-61. (Ref 8.)

van den Berghe G, Wouters P, Weekers F, et al. Intensive insulin therapy in the critically ill patients. N Engl J Med. 2001;345:1359-67. (Ref 7.)