Comparison of techniques to determine human skeletal muscle voluntary activation

Abstract

Determining volitional activation (VA) can provide insights on the cause of muscle weakness in orthopedic and neurological populations. Two electrical stimulation techniques are traditionally used to quantify VA: interpolation (IT) and superimposition (CAR). IT allows for a more accurate VA estimation, however it requires individuals to be stimulated twice, compared to once for CAR, and thus increases stimulation associated discomfort. To date, there is no agreement on what is the best practical technique for calculating quadriceps VA. This paper aims to address this problem by determining what reference force (i.e., using either peak force or force at the time of stimulation) and type of stimulation (train of pulses (burst), doublet, and twitch) is the best technique to use. Our findings showed that the IT with the force at the time of stimulation as a reference should be used to determine VA and that when a burst was used, the VA ratio computations were more accurate. Additionally, using a twitch with a 2ms pulse duration produced reliable VA calculations and may be an acceptable alternative for pain-sensitive subjects. Accurate assessment of VA deficits can help clinicians design rehabilitation programs that are based on subject-specific strength impairments and are more effective.

Keywords: Electrical stimulation; Maximum volitional isometric contraction; Quadriceps femoris; Voluntary activation.

Conflict of interest statement

Conflict of Interest

The authors report no conflicts of interest.

Copyright © 2017 Elsevier Ltd. All rights reserved.

Figures

Figure 1

A: Schematic illustration of a typical analysis curve at 75% of MVIC. A burst is delivered when subjects maintain force at 75% of MVIC and when at rest. a: Volitional force+ superimposed force, b: Volitional peak force prior stimulation (Fpeak), c: Volitional force at the instantaneous delivery of stimulation (F@stim), d: Superimposed force increment, and e: Maximal resting stimulated force. B. VA equations for IT and CAR methods with Fpeak and F@stim as reference forces.

Figure 2

Mean residuals for IT and ST techniques with Fpeak and F@stim as reference forces. The error bars represent 95% confidence intervals.* Significant difference from ST with MVIC (p < 0.0005). † Significant difference from IT with F@stim (p < 0.0005).

Figure 3

VAR values for the three different types of stimuli plotted against %MVIC. All investigated stimuli was described by curvilinear VAR-%MVIC relationships (A–C). A linear VAR-%MVIC relationship for burst is presented (D).

Figure 4

Identity Plots of MVIC versus MVICpredicted forces for all three types of stimuli at 25%, 50%, 75% of the MVIC. MVICpredicted force values were derived by using the curvilinear VAR-%MVIC force relationship for burst, doublet, twitch stimulus and the linear VAR-%MVIC force relationship for the burst stimulus. The black line represents equality between MVIC and MVICpredicted values (y=x) and not the regression line. Data points lying near the line indicate the agreement between the two measurements. Data points falling above the identity line overestimated the MVIC and underestimated it when falling below.