MÉTODOS DE ESTIMATIVA DE PESO E ALTURA NA AVALIAÇÃO DE DESNUTRIÇÃO DE PACIENTES COM CÂNCER

 To calculate the estimated and actual height and weight measures in cancer patients, evaluate the correlation and concordance of these measures and to identify the malnutrition patients. 62 cancer patients, treated at the Amaral Carvalho Hospital (JauSP). For classifi cation of nutritional status, the Patient-Generated Subjective Global Assessment, the Body Mass Index and the relationship with usual weight were used. For the estimated weight was used the algorithm Chumlea et al.7 and estimated height, the equation proposed by Chumlea et al.8 and Kwok & Whitelaw24. The Pearson Correlation Coeffi cient and the Intraclass Correlation Coeffi cient was calculated and simple linear regression was conducted. To compare the malnutrition detected by different methods, the Cochran’s Q test were performed. There was a strong and signifi cant correlation between the estimated and actual weight (r=0,95; CI95%=0,92-0,97; p<0,001) and excellent concordance (ρ=0,95). For height measurements, a signifi cant correlation was observed between the actual and estimated measurement the method Chumlea et al.8 (r=0,78; CI95%=0,66-0,86; p<0,001) and Kwok & Whitelaw24 (r=0,85; CI95%: 0,75-0,90; p<0,001). Concordance was “Good” (ρ=0,72) and “Moderate” (ρ=0,67) between the actual height and estimated by Chumlea et al.8 and Kwok & Whitelaw,24 respectively. There was a strong correlation between the actual and estimated weight and height measures. The reproducibility of the equation proposed by Chumlea et al.8 to calculate height was better. The nutritional status, the patients classifi ed as malnutrition was greater when using the Patient-Generated Subjective Global Assessment and the relationship with usual weight.