This study was performed to determine if muscle strength increases (during rehabilitation) in each of the four extremities of hemiparetic stroke patients and to determine whether percentage increases are related to initial strength or time since onset of hemiparesis. The strength of 14 muscle groups was measured with a hand-held dynamometer during an initial assessment and again prior to discharge. For seven of the muscle groups, strength increases were expressed as daily percentage increases relative to initial strength. The strength increased significantly in all four extremities between initial and final assessment. Percentage strength increases were greater in weaker muscles, albeit significantly in only two of seven muscle groups. Percentage strength increases were not correlated with time since onset of hemiparesis. Given the findings of this and previous studies, the clinician can probably conclude with some confidence that the muscle strength of all four extremities of stroke patients will, as a rule, improve during inpatient rehabilitation. It is doubtful that higher percentage increases in strength will allow patients with weaker muscles to gain strength to the point that their muscles are as strong as those of patients who were stronger initially. Key Words: Hemiparesis—Muscle strength-Rehabilitation.Following cerebrovascular accidents (CVAs), the usual physical impairment is a unilateral weakness. Such weakness, although perhaps differing in character and distribution, can result from lesions in a wide variety of locations, cortical and extracortical (1,2). Over the past 10 years, an increasing number of reports have been published in which the muscle strength of patients with stroke has been measured quantitatively by hand grip (3,4), fixed (5-7), hand-held (8-13), or isokinetic dynamometers (3,(14)(15)(16)(17)(18)(19)(20)(21)(22). What is known from these reports is that: (a) strokes do cause weakness, not only of the side contralateral to the brain lesion but also of the side ipsilateral to the lesion (3,14,15,17,18); (b) all muscle groups are not affected equally, and the relative involvement of specific muscle groups may differ from that traditionally expected (6,13); (c) normal mechanisms that alter muscular strength such as length/tension and force/velocity relationships may explain, in part, the limitations in strength noted under specific conditions (18-20); (d) significant relationships such as normally exist between muscle strength and independent variables, such as age, weight, and gender, are maintained on the nonparetic but not on the paretic side (12); (e) muscle strength is related to function (3,10,21,22); (0 muscle strength tends to increase with time following a CVA (4,11,13,14,23); and (g) strength soon after stroke is a predictor of strength at a later time (4,11, 12), but time since onset is not (11,12).What is not yet known is how strength changes in multiple muscle groups (paretic and nonparetic) over a short course of inpatient rehabilitation. Furthermore, the influence ...