Journal of Rehabilitation Medicine 51-4inkOmslag | Page 17

251
non-affected hand (Fig. 2). Maximum force values
were lower on the affected side in all positions at W2
and W4 (p ≤ 0.003 for all comparisons). However,
maximum force values on the affected side increased
steadily during the follow-up period and approached
the values of the non-affected side at W28 and W30
and were no longer different at W54 (p = 0.09–0.25).
At W2, the median grip force ratio (maximum force
affected/maximum force non-affected hand) was least
for position 1 (0.37, corresponding to 63% difference,
Fig. 2A) and largest for position 5 (0.43, corresponding
to 57% difference, Fig. 2E). At W54, the ratio between
hands ranged from 0.74 to 0.80 for the various handgrip
positions. There was no difference in grip force ratio
between position 1 (narrow) and position 5 (wide) at
any time-point (p ≥ 0.22 for all comparisons). steadily in the affected hand. At W54, maximum force
was 14% lower in the affected hand compared with the
non-affected hand, but the difference between hands
was no longer significant (p = 0.25).
At W2, the maximum force for the 3-finger grip was
55% lower for the affected hand than the non-affected
hand (p < 0.001, Fig. 3B). This difference decreased to
24% at W54, but the maximum force for the affected
side remained lower than the maximum force for the
non-affected side (p = 0.02). A similar pattern was
observed for the recovery of 2-finger force. At W2,
maximum force was 57% lower for the affected side
compared with the non-affected side (p < 0.001, Fig.
3C). This difference decreased to 24% at W54, but
maximum force remained lower on the affected side
compared with the non-affected side (p = 0.01).
Pinch grip strength Rate of force development
Fig. 3 shows the strength recovery of the key grip,
3-finger grip, and 2-finger grip during the 1-year
follow-up. At W2, key grip maximum force was
45% lower in the affected hand compared with the
non-affected hand (p < 0.001, Fig. 3A). Key grip
maximum force remained essentially unchanged for
the non-affected hand during follow-up, but increased Fig. 4A shows the recovery of rate of force develop-
ment in power grip MVC (position 2) during the 1-year
follow-up period. At W2, rate of force development
during the first 500 ms was 62% lower for the affected
hand than for the non-affected hand (p = 0.001, Fig.
4A). Rate of force development during the first 500 ms
in the affected hand increased during the first 6 months
Grip strength after stroke
Fig 2. Recovery of maximum force of power grip during the first year after stroke. Maximum force was measured with a hand dynamometer in
5 positions from (A) narrow grip (position 1) to (E) wide grip (position 5) at 2, 4, 28, 30 and 54 weeks after stroke. Values are mean and error
bars 95% confidence interval (95% CI). Note that the assessment intervals are upscaled to improve readability. Closed circles represent the non-
affected side and open circles the affected side.
J Rehabil Med 51, 2019