Regional Laboratory
AllianceRegional Laboratory
Alliance|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Abstract: JCAHO standards require timely diagnosis and treatment of malnutrition. Such treatment results not only in better patient outcomes, but also in significant financial benefits to hospitals and chronic care facilities. The incidence of malnutrition in hospitalized and chronic care facility patients ranges from 30-50%, but often goes undiagnosed and untreated due to poor physician training and awareness in this area, resulting in increased patient morbidity and mortality. This paper discusses the risk factors and adverse effects associated with malnutrition, the clinical signs and symptoms of malnutrition, and the laboratory tests that are used for diagnosing malnutrition and for monitoring the response to nutritional therapy.
JCAHO standards expect that all patients at nutritional risk will be identified within 24 hours after presentation, that nutritional intervention will be initiated within an additional 24 hours, and that the intervention will be monitored. A delay in the diagnosis of malnutrition (leading to a delay in providing nutritional intervention) is no different from failure to provide metabolic management for a sick patient. Aside from the benefit of meeting JCAHO requirements and avoiding accreditation problems, aggressive programs for the early detection, treatment and monitoring of all forms of malnutrition result in improved patient care with decreased morbidity and mortality. On the financial side, timely diagnosis of malnutrition with appropriate intervention results in decreased pharmacy costs, decreased length of stay in Intensive Care Units, decreased overall hospital length of stay, decreased post-discharge readmission rate, and increased overall survival at home. One study showed a 2.1 day decrease in length of hospital stay resulting from early and persistent nutritional support, and a $2,100 per patient savings when nutritional intervention was initiated within 3 days of hospital admission. Another study showed a financial penalty per patient when there was a 3-5 day delay in initiating nutritional support. Besides these savings, the diagnosis of malnutrition results in an increase in Medicare DRG reimbursement. Costs associated with specialized nutrition support modalities are more than offset by the decreased overall hospital costs and the increased DRG reimbursements that result from timely diagnosis and treatment of malnutrition.
Malnutrition has been found to occur in 40% of free-living persons >60 years old, and in 30-50% of all hospitalized and chronic-care facility patients. Patients may be malnourished at the time of admission, or malnutrition may be acquired during hospitalization when oral intake is inadequate or physically impossible, and/or nutrient losses are excessive. Protein-energy malnutrition (PEM) is particularly common in hospitalized patients with an acute major illness or a surgical procedure concurrent with a chronic medical condition or an extended period of inadequate nutrient intake. One study found that 60% of COPD patients with acute respiratory failure were malnourished, and that malnutrition was more frequent in those patients requiring mechanical ventilation (74%) than in those who did not (43%). It has been estimated that >50% of cases of PEM may go undetected in hospitalized geriatric patients, due to poor physician training in making the diagnosis of malnutrition and in recognizing those who are at risk, lack of awareness that malnutrition may be the presenting feature of a number of treatable diseases in older persons, and lack of knowledge of how best to manage patients with PEM.
Malnutrition is the result of one or more factors, consisting of any combination of (1) increased fuel needs that are not met by normal oral intake, (2) decreased intake of nutrients, (3) abnormalities of absorption, and (4) abnormal nutrient losses. Malnourished hospitalized patients may be divided roughly into two groups, but pure distinctions are rarely seen. The first group are those patients with a kwashiorkor-like malnutrition and hypermetabolism. These patients have 10-40 gram/day urinary nitrogen excretion, elevated acute phase reactants, and elevated or normal BUN. The second group consists of patients with a marasmus-like malnutrition with hypometabolism. They usually have <10 gram/day of urinary nitrogen excretion, normal a C-reactive protein, and a low or normal BUN. Both groups have decreased serum prealbumin. Patients at risk for the hypermetabolic type of malnutrition are those who have major acute physiological stress (major surgery, major trauma, perforation of a viscus, burns, sepsis, pancreatitis, multi-organ failure, etc.). Patients with chronic diseases such as COPD, cancer, AIDS, chronic hepatitis, etc. , as well as elderly persons in chronic care facilities are more at risk for the hypometabolic type of malnutrition.
The stressed, hypermetabolic malnourished patient may have normal or abundant adipose tissue reserves with normal or increased total body weight and appearance. Acute physiologic stress can send a well nourished person into a hypermetabolic state in which metabolic expenditures are increased as much as 2.5 times over normal. Even obese patients may become malnourished under these circumstances since fat stores are not utilized as the primary source of energy. Instead, lean body mass is sacrificed via peripheral muscle proteolysis to produce gluconeogenic fuel precursors. Hypermetabolic patients who are not provided adequate nutrition have increased risk of multi-organ failure, which is associated with a 60-95% mortality rate.
The marasmus-like hypometabolic patient tends to be thin with a wasted appearance and very little adipose tissue reserves, and utilizes ketogenic fatty acids as a primary fuel rather than carbohydrates. Recognition of the primary process occurring in a given patient is necessary to determine the best treatment regimen: the hypermetabolic malnourished patient may require aggressive nutrition support to minimize catabolism and protein loss, while aggressive feeding may be undesirable in the hypometabolic patient.
Other risk factors for malnutrition include fluid and nutrient loss from persistent diarrhea, draining fistulas or wounds; extended periods without oral intake due to surgery or ileus; cancer of the GI tract; malabsorption syndromes; alcoholism; tuberculosis; and long term dialysis. In addition to these risk factors, many patients may have marginal metabolic reserve due to poor food intake prior to admission, and may quickly develop PEM if their condition is not addressed. Elderly patients with mental depression, and those with dental prostheses or with problems biting, chewing, or swallowing are especially at risk to be marginally nourished or malnourished at the time of admission. One study also suggests that postmenopausal osteoporosis may be associated with malnutrition.
Studies have repeatedly found that malnutrition is associated with higher incidence of adverse outcomes, including postoperative pneumonia and sepsis, nosocomial pneumonia, opportunistic infections, poor wound healing, wound infections, pressure sores, inability to be weaned off mechanical ventilation, multi-organ failure, and overall mortality. Malnutrition is a major co-morbid cause of death, and in particular is a major independent predictor of adverse outcome in elderly hospitalized patients.
Malnutrition causes decreased immune function, with reductions in cell-mediated immunity, the bactericidal function of neutrophils, the complement components, the IgA secretory antibody response, and interferon production. The number of T-cells (particularly T-4 helper cells) and B-cells is reduced, while null cells are increased in malnourished patients. Malnutrition-induced impairment of immune function not only causes increased incidence of post-operative, nosocomial, and opportunistic infections, but also adversely affects recovery from several types of primary infections (tuberculosis, bacterial diarrhea, measles, viral respiratory infections, Pneumocystis carinii, Candidiasis, etc.).
Malnutrition in COPD patients is associated with a reduction of inspiratory muscle strength and diaphragmatic mass. When viewed together with the increased work of respiration in these patients, it is easy to see why malnutrition is a major cause of adverse outcomes, including failure to wean from mechanical ventilation and increased mortality.
Malnutrition is a major factor in decreased wound healing in surgical patients, trauma victims (including those with hip fractures), and debilitated patients with ischemic ulcers.
Assessment of malnutrition may range from very simple to fairly complex, and no single measurement can accurately diagnose malnutrition. Assessment tools include the physician's overall impression of the patient (subjective global assessment), anthropomorphic measurements (weight, body-mass index, mid-upper arm circumference, skinfold thickness, etc.), and measurement of various biochemical parameters (serum albumin, pre-albumin, transferrin, cholesterol, vitamin and mineral levels, etc.).
The physician's subjective global assessment of nutritional status is of great diagnostic and prognostic value. The assessment includes evaluation of the patient's overall appearance, medical and dietary history (including recent weight loss), functional ability, and ability to use the GI tract. (See pp. 2-3 of the Regional Laboratory Alliance Strategies for Clinical Laboratory Diagnosis No. 31.) Several studies have shown good correlation between the physician's clinical impression and more objective criteria (anthropomorphic measurements and biochemical markers of nutrition). At the very least, the clinical impression of undernutrition should lead to specific testing for malnutrition. It should be kept in mind that subjective and socio-cultural aspects influence the clinical impression of nutritional status.
Body weight, expressed as a percentage of the ideal body weight (IBW) is a major anthropomorphic parameter of nutritional status. Body weight <90% of the IBW is a strong predictor of malnutrition, but malnutrition may be present even when body weight is normal, especially if the patient has retention of body water. Age-adjusted reference values of normal body weight are available (see pp. 7-8 of the Regional Laboratory Alliance Strategies for Clinical Laboratory Diagnosis No. 31.), and more accurately reflect normal nutritional status in older patients than the usual height/weight charts. Tables of reference ranges for triceps skinfold, mid-arm circumference and other anthropomorphic measurements are also available. Anthropomorphic measures are most sensitive and specific when used in conjunction with biochemical markers of nutritional status. Creatinine-height index (CHI) is one parameter that utilizes both an anthropomorphic measure and a laboratory test. It is an indicator of muscle mass that expresses the 24-hour excreted creatinine as a percent of the predicted value for the patient's height and sex.
Several biochemical markers of nutrition are available. These include serum levels of total protein, albumin, prealbumin, transferrin, retinol binding protein, iron, various vitamins, cholesterol, magnesium, zinc, phosphate, calcium, BUN, creatinine, and urine urea nitrogen. Other laboratory tests that aid in the assessment of nutritional status include the RBC count and indices, and the hemoglobin level. Several of these tests will be discussed individually. Since immune function is dependent on adequate nutrition, delayed hypersensitivity skin tests utilizing common antigens can provide indirect information regarding nutritional status.
Prealbumin: Of all the serum proteins, prealbumin (a.k.a. transthyretin, thyroxine-binding prealbumin) comes the closest to being the ideal nutritional marker. It has a short circulation half-life of about 2 days, responds rapidly to improved nutrient intake, reflects decreased intake quickly, and accurately reflects the degree of nutritional deficiency. Measuring prealbumin in all newly admitted patients is considered to be the most cost effective method of detecting PEM. The serum level of prealbumin is a sensitive and fairly specific indicator of malnutrition since it is affected by only a few non-nutritional factors: it is a negative acute phase reactant that decreases when infection is present, and it is elevated with end-stage renal disease, hepatic failure, and glucocorticoid administration. Normal serum level is 16-35 mg/dL. Decision points for the diagnosis of malnutrition have been established: >16 mg/dL - not malnourished; 11-16 mg/dL - mild malnutrition; <11 mg/dL - moderate malnutrition; <7 mg/dl - severe malnutrition. Prealbumin can also be utilized to assess and monitor the adequacy of nutritional support. A successful nutritional regimen is accompanied by a daily increase of prealbumin of not less than 0.9 mg/dL, or by a doubling of the serum level within one week. Achievement of a level of 16-18 mg/dL is an indicator of a good clinical response to nutritional therapy.
Albumin: Serum albumin has long been used as an indicator of nutritional status. However, it has significant drawbacks as a nutritional marker, and several studies have found the albumin level to be no better than global assessment for detecting malnutrition. A normal serum albumin level (3.5 - 5.0 gm/dL) does not rule out PEM; the level may be falsely elevated due to dehydration. Conversely, a serum albumin level <3.2gm/dL is significant, but may not be indicative of malnutrition. Albumin levels <3.0 gm/dL are often associated with significant malnutrition, but overhydration and dehydration greatly affect the concentration of serum albumin (independent of protein losses) making it an unreliable indicator of nutritional status in many cases. The level is also affected by liver and renal disease, and by therapeutic administration of albumin. It has a serum half-life of 21 days, so it cannot be used effectively for monitoring nutritional therapy.
Total Protein: Total protein has many of the same drawbacks as albumin as a nutritional marker, and in addition is affected by conditions which affect levels of serum globulins. For this reason, it is only useful as a marker of nutritional status when used in conjunction with other markers.
Transferrin: Transferrin can be used for the assessment of hepatic protein synthesis, but as a marker of nutritional status it is no more reliable than serum albumin. Decreased transferrin is seen in malnutrition, but serum levels can also be decreased in patients with chronic inflammatory disease or chronic infection. Serum levels of transferrin, like albumin, do not correlate with serum levels of prealbumin.
Cholesterol: A serum cholesterol level below 120 mg/dL is indicative of PEM, and a level below 90 mg/dL is highly predictive of severe PEM. However, a normal cholesterol level does not rule out malnutrition. A rise in the serum cholesterol level in conjunction with nutritional support is indicative of a good clinical response to such support.
Retinol binding protein: Retinol binding protein has many of the same advantages as prealbumin as a nutritional marker, but testing is currently not as readily available as prealbumin testing. It provides much the same screening and monitoring information as prealbumin.
Urine Urea Nitrogen: Urine urea nitrogen (UUN) accounts for 60-90% of the total urinary nitrogen (TUN). The remaining 10-40% of urinary nitrogen is in the form of creatinine, ammonia, porphyrins, nitrosoamines, and amino acids. UUN and TUN provide a means of determining nitrogen balance, and are primarily useful in assessing the adequacy of nutritional support in hypermetabolic patients. Although TUN is preferable due to more accurate reflection of nitrogen losses (especially in critically ill patients), it is expensive and is not available in all laboratories. UUN is easily measured and is an inexpensive test. Formulas to estimate the TUN from the UUN (usually given as UUN + 4 gm) can seriously underestimate nitrogen loss in a critically ill/hypermetabolic patient.
Vitamins: Other than B12 and folate, single vitamin deficiencies are rare. Most often, vitamin deficiencies are multiple, and occur in the setting of generalized malnutrition; they can be expected to resolve with adequate nutritional support. Routine testing for specific vitamin levels other than B12 and folate are not recommended unless there is clinical suspicion of a specific deficiency (e.g., pellagra, rickets, beriberi, etc.).
Iron: Iron deficiency can occur in an otherwise well-nourished person, or it can be a component of generalized malnutrition. Serum iron level is most useful when used in conjunction with other tests (serum ferritin, TIBC, transferrin, % saturation) in the laboratory work-up of anemia.
Trace Minerals: Deficiencies of single trace minerals (zinc, copper, chromium, manganese, molybdenum, selenium) are rare, and the clinical signs and symptoms of deficiency, except for zinc, are usually non-specific. Like vitamin deficiencies, trace mineral deficiencies most often occur in the setting of generalized malnutrition. Risk factors for trace metal deficiencies include a vegetarian diet (especially one high in fiber and phytates that bind metals), prolonged diet of less than 1,000 calories per day, IV nutrition for greater than two weeks without added trace minerals, alcohol consumption, drug treatment, and increased losses secondary to malabsorption, drainage, diarrhea, etc. The measurement of trace metals in body fluids has not proved to be a reliable predictor of the overall trace metal nutriture. For this reason, routine measurement of trace metals in unselected populations is not recommended. The presence of clinical signs and symptoms associated with a deficiency of a trace metal would justify the measurement of that metal.
Zinc is an intracellular ion, and plasma levels are a poor reflection of body zinc stores. Although no marker has been accepted as accurately reflecting true zinc status, periodic measurements of serum zinc are recommended to monitor the adequacy of zinc replacement therapy in certain conditions (burns, persistent diarrhea, fistulas, persistent wound drainage).
It is just as necessary to properly monitor therapy for PEM as it is to provide therapeutic drug monitoring of patients receiving drugs such as digoxin. With monitoring, it is important to look for trends up or down. This will enable the health care team to change the nutrition plan in response to the patient's improved or worsened nutritional state. Avoidance of only one day of unnecessary TPN can save more than $200, making monitoring extremely cost-effective. Prealbumin should be measured daily or every other day during the acute phase of the illness, especially when nutritional therapy is initiated, and during any transition from one nutritional therapeutic mode to another. Once the patient is stable, twice weekly or weekly prealbumin levels are recommended. Electrolytes should be checked daily or every other day until stable, and then testing 1-3 times per week can be utilized. (Stable long-term TPN patients do not require such frequent monitoring.) Patients receiving IV lipids should have a baseline triglyceride level before the initiation of therapy, and then be monitored to maintain the triglyceride level below 400 mg/dL. The patient's tolerance of triglycerides can be assessed by obtaining a 4-hour post-infusion level, which should be <250 mg/dL. Once the IV lipid administration has been adjusted to keep the serum level below 400 mg/dL, weekly monitoring should be sufficient to ensure that hypertriglyceridemia does not occur. Trace minerals and vitamins should be monitored 3-4 times per year in long-term TPN patients. Serum levels of trace minerals and vitamins are not recommended in patients receiving oral nutritional therapy with evidence of good clinical response (i.e., increased prealbumin and other biochemical markers), unless there is clinical evidence of a specific vitamin or mineral deficiency.
Bibliography
Bernstein, L.H., et al. Usefulness of data on albumin and prealbumin concentrations in determining effectiveness of nutritional support. Clin Chem. 1989:35;2. pp. 271-4.
Carlson, T.H. Therapeutic nutrition monitoring. Advance/Laboratory. August, 1996. pp. 70-77.
Chandra, R.K. Nutrition, immunity and infection: present knowledge and future directions. Lancet. 1983:1. 00. 688-91.
“Consensus Guidelines: Laboratory Support in Assessing and Monitoring Nutritional Status.” Standards of Laboratory Practice. The National Academy of Clinical Biochemistry. 1994. pp. 1-28.
Detsky, A.S., et al. What is subjective global assessment of nutritional status? Journal of Parenteral and Enteral Nutrition. 1987:11. pp. 8-13.
Frisancho, A.R. New standards of weight and body composition by frame size and height for assessment of nutritional status of adults and the elderly. Am J Clin Nutr. 1984:40. pp. 808-19.
Kaplan, L.A., & Bernstein, L.H. Medical and economic consequences of protein-energy malnutrition: the role of prealbumin testing. Clin Chem. 1997:37. pp. 47-60.
Laaban, J.P. et al. Nutritional status of patients with chronic obstructive pulmonary disease and acute respiratory failure. Chest. 1993:103(5). pp. 1362-8.
Mattox, T.W. & Teasley-Strausburg, K.M. Overview of biochemical markers used for nutrition support. DICP, The Annals of Pharmacotherapy. 1991:25. pp. 265-70.
Morley, J.E. Why do physicians fail to recognize and treat malnutrition in older persons? J Am Ger Soc. 1991:39. pp. 1139-40.
Muhlethaler, R., et al. The prognostic significance of protein-energy malnutrition in geriatric patients. Age-Ageing. 1995:24(3). pp. 193-7.
Rico, H., et al. Biochemical markers of nutrition in type-I and type-II osteoporosis. J Bone Joint Surg Br. 1995:77(1) . pp. 148-51.
Rollings, R.C. Facts and Formulas. Copyright by Robert C. Rollings. Nashville, TN, 1984. pp. 56-62.
Sahyoun, N.R., et al. Dietary intakes and biochemical indicators of nutritional status in an elderly, institutionalized population. Am J Clin Nutr. 1988:47. pp. 524-533.
Spiekerman, A.M., et al. Determination of malnutrition in hospitalized patients with the use of a group-based reference. Arch Pathol Lab Med. 1993:117(2). pp. 184-6.
Volkert, D., et al. Malnutrition in geriatric patients: diagnostic and prognostic significance of nutritional parameters. Ann Nutr Metab. 1992:36(2). pp. 97-112.