METHOD:
The reliability and reproducibility of two HHD methods for evaluating upper limb muscle strength was assessed and compared using a prospective, cross-sectional study.
Participants:
Healthy individuals were recruited by direct call, text message or social network. Participants (male or female) older than 18 years and who signed the informed consent form were included in this study. Exclusion criteria included acute bone, muscle or joint diseases, reduced functional range of motion (ROM)13, presenting severe heart disease or neuromuscular diseases, or cognitive limitations that reduced the understanding of motor commands during the evaluation.
Randomization:
The order of the examiner and evaluation method was randomized for all patients with the use of a draw, which was carried out using a brown envelope.
Instruments:
Muscle strength was evaluated using a previously calibrated digital isometric dynamometer, (model 01165, Lafayette Instrument Company, Sagamore, USA). A goniometer (ISP, Sao Paulo, BR) was used to properly mark the articular position of the segments for each movement evaluated14. All patients had the proximal segment (torso, arm, or forearm) stabilized with the use of an inelastic belt, in order to nullify the effect of synergistic muscle chains.
Procedures:
The examiners were trained according to the evaluation protocol prior to testing. The protocol tests the isometric contraction for ten dominant upper limb muscle groups (shoulder flexors and extensors, shoulder internal rotators and external rotators, shoulder adductors and abductors, elbow flexors and extensors, wrist flexors and extensors). Prior to the measurement, the volunteers were instructed and trained how to perform each of the movements, and a muscle warm-up was also performed for each movement.
The fixed method used a system of suction cups adhered to rigid surfaces that were connected to the dynamometer through a Mulligan inelastic belt15,16. During the non-fixed method, the examiner supported the device with one hand, in a direction contrary to the movement, stabilizing the segment proximal to the moving joint under assessment14,17-20.
For both evaluation methods, the isometric contraction was sustained for 3 seconds. Each movement was repeated thrice; the largest of the three values was used. The dynamometer was placed in the distal region of the forearm, 5 cm from the radial styloid process21. A muscle recovery time of 90 seconds between tests was ensured for all measurements14. If there was visible compensation of synergistic muscles in any of the movements, the volunteer would be instructed on the correct movement, and the measurement would be repeated. The positions adopted for the measurements were based on previous studies (Table 1)14, 17-21.
A minimum resting time of 30 minutes was given between the evaluation of each examiner22.
Variables of interest:
The concept of reliability is related to the capacity of the instrument and the evaluation method to generate similar results, even when used by different examiners. Thus, reliability was obtained by comparing the highest peak torque of the evaluations for two independent examiners (or inter-examiner reliability). On the other hand, the reproducibility of the evaluation method (or intra-examiner reliability) was obtained by analyzing the similarity between the HHD test and retest by the same examiner.
Statistical method:
The data were tabulated and analyzed using SPSS (version 21.0, Statistical Package for the Social Sciences, Chicago, USA). The qualitative variables were expressed in absolute and relative frequency, while the quantitative variables were expressed as mean, median, standard deviation (SD) and 95% confidence interval (95% CI). Analysis of variance (ANOVA) was used to evaluate the different the peak torque means obtained from the two methods. The level of significance level was set at p<0.05.
Prior to recruitment, a sample size calculation based on the results of a previous study21 determined 25 participants were sufficient to detect a 10% variation between measurements and a 3% SD with 80% power with α at 0.05. As this was a cross-sectional study sample loss was not expected.
Reproducibility and reliability of the tests were calculated using the intraclass correlation coefficient (ICC) and categorized using the classifications proposed by Weir (2005)23: “almost perfect” for values from 1.0 to 0.81, “very good” from 0.80 to 0.61, “good” from 0.60 to 0.41, “fair” from 0.40 to 0.21, and “low” from 0.20 to 0.00. A Bland-Altman plot was used for the visualization of the agreement between the quantitative measurements obtained in the tests.