Thumb and finger forces produced by motor units in the long flexor of the human thumb

Abstract

The uncommonly good proprioceptive performance of the long flexor of the thumb, flexor pollicis longus (FPL), may add significantly to human manual dexterity. We investigated the forces produced by FPL single motor units during a weak static grip involving all digits by spike-triggered averaging from single motor units, and by averaging from twitches produced by intramuscular stimulation. Nine adult subjects were studied. The forces produced at each digit were used to assess how forces produced in FPL are distributed to the fingers. Most FPL motor units produced very low forces on the thumb and were positively correlated with the muscle force at recruitment. Activity in FPL motor units commonly loaded the index finger (42/55 units), but less commonly the other fingers (P < 0.001). On average, these motor units produced small but significant loading forces on the index finger ( approximately 5.3% of their force on the thumb) with the same time-to-peak force as the thumb ( approximately 50 ms), but had no significant effect on other fingers. However, intramuscular stimulation within FPL did not produce significant forces in any finger. Coherence at 2-10 Hz between the thumb and index finger force was twice that for the other finger forces and the coherence to the non-index fingers was not altered when the index finger did not participate in the grasp. These results indicate that, within the long-term coordinated forces of all digits during grasping, FPL motor units generate forces highly focused on the thumb with minimal peripheral transfer to the fingers and that there is a small but inflexible neural coupling to the flexors of the index finger.

Figure 1

Experimental set-up The distal pad of the thumb was positioned in a ring connected to a load cell which measured vertical force. There was an adjustable pillar just proximal to the distal interphalangeal joint of the thumb. Each finger pad was positioned over a load cell embedded in the cylinder to measure horizontal forces. Fine-wire intramuscular electrodes were inserted into flexor pollicis longus (FPL) to record from the single motor units. Feedback of force produced by each digit was available. T, thumb; I, index; M, middle; R, ring; L, little.

Figure 2

Data obtained during grasping for two single motor units in flexor pollicis longus (FPL) and one site of intramuscular stimulation in FPL Left, the triggered averages of change in force under each digit are displayed. Right, for the single motor units, the interspike interval histograms are shown, and insets show the shape of the motor unit potential. The vertical arrow indicates the time of discharge of the unit or the time of intramuscular stimulation. The units in both A and B produced typical forces under the thumb and also significant loading of the index finger. There was a small unloading in the middle, ring and little fingers in A. There was loading in the first three fingers in B. A also shows an example of the average response to intramuscular stimulation at a site where motor unit activity was recorded. Unlike for the spike-triggered averages, there was unloading of the index finger. In both panels, axes for the inset represent 1 ms and 50 μV.

Figure 3

Flexor pollicis longus (FPL) motor unit recruitment threshold and twitch force distribution A, relationship between the force at which a unit in FPL was recruited and the size of its twitch derived by spike-triggered averaging. Data are shown for 55 single motor units in FPL. The correlation between recruitment force and twitch force was statistically significant (P < 0.001). B, distribution of the twitch forces of FPL single motor units during grasping. Data for 55 single motor units in FPL are shown as shaded bars. The majority of units in FPL produced small forces (< 10 mN). Dashed line shows the distribution of the twitch forces produced by single motor units in flexor digitorum profundus (FDP) at the finger pads obtained in the study by Kilbreath et al. (2002).

Figure 4

Twitch forces produced by flexor pollicis longus (FPL) motor units on each of the digits A and B, twitch force produced by single motor units in FPL during grasping (mean ±s.e.m). Apart from the primary loading on the thumb (P < 0.01), there was significant loading on the index finger (P < 0.01). error bars for the fingers are smaller than the symbol size in A. This loading of the index is more clearly evident in B in which twitch forces are plotted relative to the loading on the thumb. C, percentage of units producing loading (•), unloading (○) and having no significant effect (▴) on the finger pads. Loading of the index finger occurred frequently and was less common for the other fingers (for all fingers, P < 0.001). *Significant changes.

Figure 5

Power spectra of forces and coherences between the thumb and each finger A, power spectra for each of the digits for one typical subject. The frequency 2–10 Hz bandwidth covered the main power in the force signal. B, coherence plots for the same subject between the thumb force and that of each finger show a similar pattern of common force output across all digits with several tuned peaks (○, peak coherence). C, mean peak coherence (n = 8, mean ±s.e.m) between the thumb and index finger is approximately twice the value between the thumb and the other fingers.

Figure 6

Coherence between the force of the thumb and each finger when grasping A, across the 2–10 Hz bandwidth, the thumb–index coherence was significantly greater than between the thumb and the other fingers. B, grasping without the index finger did not produce significantly different coherence results between the thumb and the remaining fingers. Data shown are mean ±s.e.m. (n = 8).