Determining factors of functioning in hemodialysis patients using the international classification of functioning, disability and health

This was an exploratory, cross-sectional study conducted between August and December 2019 in the HD unit of the Hospital Santa Casa de Caridade of Diamantina. The study was approved by the Human Ethics Committee of the Universidade Federal dos Vales do Jequitinhonha e Mucuri (Protocol = 3.612.157), and carried out in accordance with the declaration of Helsinki (2013). All the HD patients provided written informed consent to participate in the study.

The study included patients over 18 years old, with ESRD, on HD treatment three times a week for at least 6 months, and with arteriovenous fistula for HD access. Exclusion criteria were HD patients with any contraindications to performing exercise tests or practicing physical exercise. Patients using corticosteroids or non-steroidal anti-inflammatory drugs were also excluded.

GPower software, version 3.1.9.2, was used for sample size calculation, which was estimated based on a pilot study with 10 hemodialysis patients. Using all dependent variables, inputting an effect size of 0.67 (for 60 s sit-to-stand test, i.e., the lowest obtained), 14 possible predictors, probability of error set at 1%, and a power of 99%, a total of 80 patients was estimated.

Handgrip strength (HS): According to the American Society of Hand Therapists (ASHT), the assessment was performed using the Jamar® dynamometer (Asimow Engineering Co, Los Angeles. CA), adjusted to the second position [17]. Volunteers remained in a seated position, with the shoulder in adduction and 90° at the elbow joint, with the forearm in a neutral position [18‐20]. Three measures of HS of the arm without arteriovenous fistula were performed and the average of these measures was used for analysis. An interval of 1 min was given between each measurement [21].

5-repetition and 60-s sit-to-stand tests (5-STS and 60-STS): The volunteers started in a sitting position, with their arms crossed over their chest and their back against the chair. The seat was at a height of approximately 43 cm. The researcher was positioned next to the volunteer, giving instructions and preventing a fall. For the 5-repetition sit-to-stand test, the time to complete the 5 repetitions was registered using a digital timer [22]. For the 60-s sit-to-stand test, the number of repetitions during 60 s was counted and noted [21].

Short physical performance battery (SPPB): The instrument consists of the following tests: static standing balance, walking speed at a normal pace for 4 m, and muscular strength of the lower limbs estimated by the sit-to-stand test without the aid of the arms five times [21, 23]. For each test, the performance was obtained by a score from 0 (worst performance) to 4 (best performance), and the scores were then added together to calculate the final points, resulting in a maximum score of 12 points. A score ≤ 8 points indicates poor physical functioning [24].

Participation scale: This instrument was translated and adapted for the Brazilian population and is based on an interview covering 18 items to measure problems perceived in the main domains of participation [25]. Subjects with a score greater than 12 were classified as having participation restrictions [25, 26].

Human Activity Profile (HAP): HD patients were asked to complete a validated and cross-cultural adapted questionnaire related to physical activity level [27, 28]. The HAP is a scale designed to survey the execution of common physical activities and consists of 94 items ranked in ascending order of energy requirements [29]. The adjusted activity score was recorded and higher scores indicate higher activity levels [29].

Dual-energy X-ray absorptiometry: A densitometer (Lunar Radiation Corporation, Madison, Wisconsin, USA, model DPX) was used to determine total body mass, body fat, appendicular lean mass (ALM), and bone mineral density (BMD) [30]. ALM was obtained as the sum of the muscle mass of the four members. Moreover, the total and lumbar spine (L2-L4) BMD (g/cm³) were also assessed [28].

Biomarkers: Two 10 ml tubes of blood samples without anticoagulant were collected from each patient for the analysis of 25-hydroxyvitamin D (vitamin D), PTH, and CPR. CPR was measured using the turbidimetric method (Biotéctica Indústria e Comércio, Varginha, MG, Brazil) [31]. PTH was determined through electrochemiluminescence (Modular Analytics E170, Roche, Mannheim, Germany). Vitamin D levels were measured using a direct competitive chemiluminescence immunoassay (Architect i2000, Abbott, IL, USA). The observed reference range was 9.3–47.9 ng/mL [32]. The lowest reported value was 4 ng/mL, and the interassay coefficient of variation (CV) was < 20% [33].

Preliminary session: The anamnesis of the HD patients, including personal and disease-related factors (gender, age, comorbidities, duration of HD, diuresis volume, fractional urea clearance - Kt/V, serum levels of hemoglobin - Hb, alkaline phosphatase, and ferritin) were obtained from regular medical records. Subsequently, HD patients were familiarized with all physical functioning tests and with the questionnaires [HAP and the Participation Scale] to be completed in the following session.

The first experimental session: Immediately before the first hemodialysis session of the following week, the HS of the HD patients was assessed. Then, during the hemodialysis session, the HAP and the participation scale questionnaires were applied to assess physical activity level and social participation, respectively.

The reliability of all clinical-functional tests was greater than 80% for all tests [21, 29, 34, 35]. All dependent outcomes were grouped according to ICF [7] domains as follows: (1) Body structure and function – upper limb strength: HS; lower limb strength: 5-STS; and endurance: 60-STS; (2) Activity – SPPB; and (3) Social participation – participation scale; and their association with independent outcomes (age, duration of HD, diuresis volume, BMI, body fat, ALM, BMD total, BMD spine, BMD hip, D vitamin, PTH, CRP, Kt/V, Hb, alkaline phosphatase, ferritin, HAP) was subsequently investigated.

Data analyses were performed using SPSS version 22.0 (SPSS Inc., Chicago, IL, USA) (Additional file 1). The sample distribution and homoscedasticity were assessed using the Shapiro Wilk test and the Levene test. The continuous data were expressed as mean and 95% confidence interval, and categorical variables as absolute number and percentage. Confirmatory analyses of the association, using simple linear regression followed by stepwise multiple linear regression models were performed with independent variables that presented p < 0.20 in the correlation tests (Pearson or Spearman test) (Additional file 2) and p < 0.10 in simple linear regression (Additional file 3). The significance level was set at 5%. The models were adjusted for age, fractional urea clearance, duration of HD, and diuresis volume. To perform the multiple regression analysis, four assumptions were adopted: linearity, distribution of residuals, homoscedasticity, and the absence of multicollinearity. The linearity of the independent variables and residuals was checked by scatter plots and the distribution of residuals was analyzed using the histogram. The homoscedasticity was verified by the scatter plot and characterized by the residuals equally distributed on the regression line. The absence of multicollinearity was defined as variance inflation factor (VIF) values below 10.0. In addition, the auto-correlation of the variables was verified through the Durbin-Watson test and values between 1.5 and 2.5 showed no autocorrelation in the data.