Nutritional assessment in the critically ill.

Although many of the measurements and techniques outlined in this article may be epidemiologically useful and correlate with morbidity and mortality, no single indicator is of consistent value in the nutritional assessment of critically ill patients. Measurements such as anthropometrics, total body fat estimation, or delayed hypersensitivity skin testing either are liable to non-nutritional influences or lack accuracy and precision in individual patients. Plasma concentrations of hepatic proteins are affected significantly by the patient's underlying disease state and therapeutic interventions and therefore lack specificity. Although the measurement of these proteins is of little value in the initial nutritional assessment of the critically ill, serial measurement, particularly of plasma pre-albumin, may be useful in monitoring the response to nutritional support. Nitrogen balance is a widely used and valuable nutritional indicator in the critically ill. Direct measurement of urine nitrogen is the preferred test, although nitrogen excretion often is derived from 24-hour urine urea measurement, an inexpensive and easy procedure, but one that is less accurate. More accurate techniques of assessing change in nutritional status, such as IVNAA of total body nitrogen or isotopic measurement of exchangeable potassium or sodium, are more expensive, less available, unsuitable for repeated analyses, and less feasible in severely ill patients. Total body nitrogen measured using IVNAA and total-body potassium, however, are the most accurate ways of measuring body composition in the presence of large amounts of edema fluid. The application of body composition measurements to patient care remains poorly defined because of the many problems encountered with the various techniques, including cost, availability, and radiation exposure. Improved, more sensitive and, preferably, bedside methods for the measurement of body composition are needed. It is of paramount importance that these methods are validated extensively in the critically ill as well as in more stable patients, not only in terms of analytical accuracy, but also to define the point at which altered body composition influences clinical outcome. The biochemical measurement of levels of vitamins, minerals, and trace elements is invaluable in demonstrating specific deficiencies associated with disease and assessing whether long-term nutritional support is adequate. Such measurements rarely are necessary to make the initial clinical decision to give nutritional support, however. The most widely used measures of nutritional state are nitrogen balance and secretory protein concentrations, and these indices improve when sick patients recover.(ABSTRACT TRUNCATED AT 400 WORDS)