Figure 1
Experimental set-up for tremor and proprioceptive testing.
(A) Experimental set-up to measure postural tremor without (left) and with (right) weight. An accelerometer was fixed to the dorsum of the hand, or on the device used to clamp the 300 g weight to the participant’s hand. Electromyography activity (EMG) was recorded with surface electrodes positioned over the extensor carpi radialis and flexor carpi ulnaris muscles (not shown). (B) Experimental set-up to measure kinetic tremor without (left) and with (right) weight. A magnetic rotary position sensor on the opposite side of the experimental set-up (not shown) captured the rotation of the experimental device. EMG was recorded from the extensor carpi radialis and flexor carpi ulnaris muscles (not shown). (C) Experimental set-up for width discrimination testing. (D) Experimental set-up for weight discrimination testing. The weights could be quickly and quietly slid onto and off from the dorsum of the device; they were held in place by a strong magnet embedded into the device.
Figure 2
Causal graph and highlighted paths of the plausible mechanism of essential tremor on function mediated by proprioception and tremor amplitude.
(A) Full causal graph with (B) paths of total effect highlighted, (C) paths of average causally mediated effect highlighted, and (D) paths of average direct effect highlighted.
Table 1
Proprioception, tremor amplitude and functional outcomes.
| OUTCOME | CONTROL (n = 22) | ESSENTIAL TREMOR (n = 20) | BETWEEN-GROUP MEAN DIFFERENCE* | BETWEEN-GROUP MEAN DIFFERENCE (95% CI)† |
|---|---|---|---|---|
| Grip strength (kg) | 32.4 (9.4) | 34.8 (11.8) | 2.3 (–4.3 to 8.9) | – |
| Tactile acuity (g) | 0.17 (0.25) | 0.32 (0.34) | 0.14 (–0.05 to 0.33) | – |
| Postural tremor, unweighted (ms–2) | 0.04 (0.01) | 0.15 (0.14) | 0.10 | 0.03 (0.01 to 0.04) |
| Postural tremor, weighted (ms–2) | 0.05 (0.01) | 0.20 (0.19) | 0.16 | 0.04 (0.03 to 0.05) |
| Kinetic tremor, unweighted (deg) | 0.36 (0.01) | 0.65 (0.36) | 0.30 | 0.14 (0.06 to 0.23) |
| Kinetic tremor, weighted (deg) | 0.35 (0.10) | 0.53 (0.20) | 0.17 | 0.11 (0.04 to 0.18) |
| FTM-A (a.u.; max score 24) | 1.0 (1.2) | 5.6 (3.4) | 4.2 | 4.0 (2.6 to 5.4) |
| FTM-B (a.u.; max score 36) | 3.1 (2.8) | 10.2 (5.0) | 6.0 | 6.3 (3.8 to 8.7) |
| FTM-AB (a.u.; max score 60) | 4.1 (4.0) | 15.8 (7.8) | 10.2 | 10.5 (6.8 to 14.2) |
| TDQ (%) | – | 0.19 (0.11) | – | – |
| ABILHAND logits (a.u.) | 5.11 (1.13) | 2.90 (1.27) | –2.09 | –2.06 (–2.90 to –1.22) |
| Box and Block Test (n) | 69 (11) | 58 (11) | –10 | –9 (–16 to –1) |
| 9-Hole Peg Test (s) | 18.5 (2.9) | 24.5 (6.5) | 5.0 | 3.9 (1.4 to 6.5) |
[i] Within-group mean (SD) for all outcomes. Between-group mean differences (95% CI) are provided for grip strength and tactile acuity. For all other outcomes, between-group mean differences from linear regression (*) and between group mean differences (95% CI) for robust regression (†) are adjusted for grip strength and tactile acuity. Legend: a.u.: arbitrary units; FTM: Fahn-Tolosa-Marin Clinical Rating Scale; TDQ: Tremor Disability Questionnaire.
Figure 3
Proprioceptive outcomes of (A) width discrimination threshold and (B) weight discrimination threshold.
Panels show individual participant data (gray circles), group mean and SD (black circle and error bar), between-group mean difference and 95% CI (black triangle and error bar), and the line of no between-group difference (dashed gray line).
Table 2
The effect of essential tremor on function that was causally mediated by proprioceptive discrimination (primary, 1°), where tremor amplitude was also a mediator (secondary, 2°).
| OUTCOME | 1° MEDIATOR | 2° MEDIATOR | ACME (ET) | ACME (AVERAGE) | ADE (ET) | ADE (AVERAGE) | TOTAL EFFECT |
|---|---|---|---|---|---|---|---|
| FTM (a.u.) | Width | Postural | –0.4 (–3.5 to 2.7) | –0.4 (–3.3 to 2.5) | 10.6 (6.1 to 15.0) | 10.6 (6.3 to 14.9) | 10.2 (6.9 to 13.2) |
| Width | Kinetic | 0.1 (–1.2 to 1.3) | 0 (–1.0 to 1.1) | 10.2 (6.6 to 13.8) | 10.2 (6.8 to 13.6) | 10.2 (6.7 to 13.4) | |
| Weight | Postural | –0.4 (–3.2 to 2.4) | –0.4 (–2.9 to 2.0) | 10.6 (6.5 to 14.8) | 10.6 (6.5 to 14.8) | 10.2 (6.9 to 13.3) | |
| Weight | Kinetic | –0.1 (–1.8 to 1.6) | –0.1 (–1.1 to 1.0) | 10.3 (6.9 to 13.8) | 10.3 (6.9 to 13.7) | 10.2 (7.0 to 13.4) | |
| ABILHAND logits (a.u.) | Width | Postural | 0.03 (–0.71 to 0.76) | 0 (–0.73 to 0.73) | –2.01 (–3.11 to –1.03) | –2.09 (–3.01 to –1.09) | –2.09 (–2.83 to –1.37) |
| Width | Kinetic | 0.05 (–0.23 to 0.32) | 0.01 (–0.25 to 0.27) | –2.07 (–2.91 to –1.23) | –2.10 (–2.88 to –1.32) | –2.09 (–2.81 to –1.38) | |
| Weight | Postural | 0.01 (–0.65 to 0.67) | 0.01 (–0.63 to 0.65) | –2.10 (–3.03 to –1.18) | –2.10 (–3.02 to –1.19) | –2.09 (–2.83 to –1.36) | |
| Weight | Kinetic | 0.02 (–0.32 to 0.36) | 0.02 (–0.25 to 0.29) | –2.11 (–2.86 to –1.36) | –2.11 (–2.85 to –1.37) | –2.09 (–2.79 to –1.37) | |
| Box and Block Test (n) | Width | Postural | –1 (–7 to 5) | –1 (–6 to 4) | –9 (–16 to –1) | –9 (–16 to –1) | –10 (–16 to –4) |
| Width | Kinetic | 0 (–3 to 3) | 0 (–3 to 2) | –9 (–16 to –3) | –9 (–16 to –3) | –10 (–16 to –4) | |
| Weight | Postural | 0 (–6 to 5) | 0 (–6 to 5) | –9 (–17 to –2) | –9 (–17 to –2) | –10 (–16 to –4) | |
| Weight | Kinetic | 0 (–2 to 2) | 0 (–2 to 2) | –10 (–16 to –3) | –10 (–16 to –3) | –10 (–16 to –3) | |
| 9–Hole Peg Test (s) | Width | Postural | –0.1 (–2.4 to 2.1) | 0 (–1.8 to 1.9) | 4.8 (1.2 to 8.5) | 5.0 (1.4 to 8.7) | 5.0 (2.1 to 8.0) |
| Width | Kinetic | –0.3 (–2.0 to 1.5) | 0 (–1.1 to 1.1) | 4.8 (1.4 to 8.2) | 5.1 (1.8 to 8.4) | 5.0 (2.2 to 8.0) | |
| Weight | Postural | 0 (–3.3 to 3.2) | 0 (–2.5 to 2.6) | 5.0 (1.4 to 8.6) | 5.0 (1.5 to 8.6) | 5.0 (2.2 to 8.0) | |
| Weight | Kinetic | –0.1 (–2.6 to 2.3) | –0.1 (–1.4 to 1.2) | 5.1 (2.1 to 8.1) | 5.1 (2.2 to 8.1) | 5.0 (2.2 to 8.1) |
[i] Only tremor amplitude during the weighted condition was analysed [50]. Estimates of the average causally mediated effect (ACME), the average direct effect (ADE) and all 95% CI were determined under the homogenous treatment assumption, adjusted for grip strength and tactile acuity. Mean total effects correspond to linear regression between-group mean differences (Table 1). For brevity, only the ACME and ADE for participants with ET and averaged between groups are shown. Legend: a.u.: arbitrary units; FTM: Fahn-Tolosa-Marin Clinical Rating Scale.