NUTRITIONAL CONTROL CONTROL NUTRICIONAL CONUT AUTOMATIC DETECTION OF

NUTRITIONAL CONTROL (CONTROL NUTRICIONAL, CONUT): AUTOMATIC DETECTION OF NUTRITIONAL ALERT Ulibarri JI, Martín G, Mancha A, Fernández G, Díaz A. Clinical Nutrition and Dietetics Unit. “La Princesa” University Hospital. Madrid (Spain) RATIONALE The modernization of diagnostic processes has allowed the implementation of a computerized method for the early detection of undernourishment, which may be automatically applied to any person undergoing a health check through clinical-biochemical analyses. We have again taken up the use of analytical data for the assessment of the nutritional status of patients after showing that they reflect, to an acceptable degree, not only the patient’s health status but also his/her nutritional status. METHODS: Common laboratory databases were used. We have assessed the impact of undernourishment on usual clinical practice, and have selected serum albumin, serum cholesterol and the total lymphocyte count as useful markers of udernourishment. The module generates the NUTRITIONAL ALERT as a further result of the analyses. The method has been validated by comparing it to nutritional status assessment protocols and to other screening systems. Statistical studies of morbidity, mortality, duration of admission and costs have been carried out. Table I. Degree of undernourishment alert as assessed by CONUT Parameter Table II. Operative data values (CONUT/ SGA) CONUT/SGA quotient 95% CI Sensitivity 0, 80 (0, 68 – 0, 91) Specificity 0, 86 (0, 77 – 0, 91) Positive predictive value 0, 81 (0, 70 – 0, 93) Negative predictive value 0, 84 (0, 75 – 0, 94 ) 5, 64 (2, 93 – 10, 96) 0, 24 (0, 13 – 0, 43 ) Values Deficiency Serum albumin * (Score ) Total lymphocytes (Score ) Total cholesterol ‡ (Score ) No deficiency Light Moderate Severe ≥ 3. 50 3. 00 -3. 49 2. 50 -2. 99 <2. 50 (0) (2) (4) (6) ≥ 1600 1200 -1599 800 -1199 <800 (0) (1) (2) (3) ≥ 180 140 -179 100 -139 <100 (0) (1) (2) (3) Interpretation Total score Degree of undernourishment alert 0 -4 5 -8 9 -12 Normal Moderate Severe Positive probability ratio Negative probability ratio RESULTS: The tool automatically alerts of the deterioration caused by disease, therapeutic measures and hospital admission on the patient’s nutritional balance. It allows for individualized follow-up of the nutritional alert and for the epidemiologic control and outcomes in hospitals and health-care centres. Fig. 2 Fig. 3 Fig. 4 Fig. 5 State or European level Ministry level ML CONCLUSIONS This is a useful and valid tool for nutritional screening as shown by a comparison of its results to those of the SGA and other tools, which confirms its usefulness as a simple, accurate, quick and unexpensive screening system that may be applied to hospitals and other health-care centres at a local, national or even European level (Figure 6). Solo Autonomic Level AL consulta NA AL AL AL BL BL Basic level BL Fig. 6 Nutritional Control The three-level data structure environment: Basic level (BL), Autonomic Community level (AL) and Ministry level (ML). REFERENCES • Validation of a new tool for nutritional screening. A. Gonzalez-Madrono, A. Mancha, N. G. P. de Villar, J. I. Ulíbarri. Clinical Nutrition 2005; 24, pg 558 (Abstract P 047). • A tool for Controlling Nutritional Status. First validation in a hospital population Ulibarri JI, Gonzalez-Madroño A, de Villar NGP et al. CONUT: . Nutr. Hosp. (2005) XX (I) 38 -45. • New screening tool for hospital undernutrition: A study of prevalence in 6242 patients. M. Guevara, B. Gonzalez, A. Mancha, G. Fernandez, F. Rodriguez, A. Diaz, I. Mascarell and I. Ulíbarri. ABSTRACT Clinical Nutrition, Volume 22, Supplement 1, August 2003, Page S 9 -Canosa Penaba J. Propuesta de un sistema de información para la detección precoz de la desnutrición hospitalaria. In: Ministerio de Sanidad y Consumo. Libro Blanco de la Desnutrición Clínica en España. Madrid, Acción Médica 2004. ACKNOWLEDGEMENTS –ABBOTT, FRESENIUS-KABI, GRIFOLS, NESTLÉ, NOVARTIS, NUTRICIA and VEGENAT, for their assistance in financing the development and validation of the CONUT project. –The Spanish Society of Parenteral and Enteral Nutrition, SENPE, for supporting the project. –The Hospital Universitario de la Princesa, Madrid, Spain. CORRESPONDENCE G. Martin Peña Unidad de Nutrición Clínica y Dietética, Hospital Universitario La Princesa C/ Diego de León 62, 28006, Madrid, SPAIN, Tel / Fax: 34 915 202 401, E-mail address: gmartinp@telefonica. net