Aquino B.1, Moreno N.1, Giacomassi I.2, Oliveira T.2, Tavares I.2, Castro A.P.2, Lunardi A.3,4
1Universidade Cidade de Sao Paulo, Master and Doctoral Program in Physical Therapy, Sao Paulo, Brazil, 2Instituto de Assistência Médica ao Servidor Público Estadual, Physical Therapy Service, Sao Paulo, Brazil, 3Universidade da Cidade de Sao Paulo, Master and Doctoral in Physical Therapy, Sao Paulo, Brazil, 4Universidade de São Paulo, Physical Therapy, Sao Paulo, Brazil
Background: Accelerometry is considered the gold standard for evaluation of the level of physical activity in healthy subjects and in patients with chronic or acute diseases. This information has been increasingly used in hospitals in predicting functional outcomes, complications and quality of life in surgical patients, clinical and critically during hospitalization and after discharge. However, the monitors activity as the accelerometer assess level of physical activity based on energy expenditure inferred by movements performed by healthy subjects and patients. The question is that in hospitalized patients who are inactive most of the time, the accelerometer remains a good method of level of physical activity assessment.
Purpose: To test the validity of accelerometer to detect activities of different intensities and respective energy expenditure in hospitalized patients compared to indirect calorimetry.
Methods: We assessed 20 patients not bedridden, admitted consecutively in two wards of a public hospital in São Paulo - Brazil. Patients underwent simulated routine activities (wear clothes, wearing shoes, eat, wash their hair, bathing, walking, climbing stairs, lie down and get up) using a portable calorimeter and two accelerometers (one placed on the wrist and other placed on the waist). The activities were grouped according to the intensity of movement, as very light (eat), light (wash hair) and moderate (up stairs). The kilocalories expended, oxygen consumption and metabolic equivalent during the performance of the simulated activities were analyzed. One-way ANOVA and Spearman correlation were use to compare the assessment methods.
Results: From the 20 patients assessed, 9 (45%) were men, with a median age of 61,5 years (52,5-67,8), weight 71kg (62,8-88,5) and BMI 28,1 Kg/m2 (24,1-32,1). The kilocalories expended, oxygen consumption and metabolic equivalent spent during activities in different intensities (very light, light and moderate) were differently estimated among calorimeter, accelerometry on the wrist and waist (p 0.05). During moderate activities, there was a moderate and positive correlation between oxygen consumption (r=0.64; p 0.05), and kilocalories expended (r=0,63; p 0.05) assessed by calorimetry and accelerometer on the wrist.
Conclusion(s): Our results show that the energy expenditure assessed by accelerometry is different from calorimetry in hospitalized patients. even they have a good correlation.
Implications: Our results show that the accelerometers are not accurate to assess the energy expenditure in hospitalized patients. Probably, the accelerometry is able only to measure number of movements in this population.
Funding acknowledgements: Grant #2015/25763-2, São Paulo Research Foundation (FAPESP)
Topic: Critical care
Ethics approval: Approved by Research Ethics Committee of IAMSPE (Instituto de Assistência Médica ao Servidor Público Estadual), protocol number: 1.668.863.
All authors, affiliations and abstracts have been published as submitted.
