Table 5.

Evaluation and results of the articles included in the systematic review.

Studies	Outcome Measures	Results
Akbari et al. (2006) [33]	Two measurement time points (pre and post intervention) -Spasticity: MAS -Strength: Maximum force evaluated with dynamometer.	Spasticity: There was a significant decrease in quadriceps spasticity in EG (p < 0.0001), but no change in CG (p = 0.055). There was a significant decrease in gastrocnemius spasticity in both EG (p < 0.0001) and CG (p = 0.041). A significant decrease in spasticity was found in EG compared to CG in both quadriceps (p = 0.034) and gastrocnemius (p = 0.001).
		Strength: There was an increase in strength in all muscles in EG on both the affected side (p < 0.0001) and the non-affected side (p < 0.0001). No differences in strength were found in CG, except in hip and knee extensors (p < 0.0001) and ankle flexors (p = 0.008) on the non-affected side, and hip extensors (p = 0.003) and knee extensors (p < 0.0001) on the affected side. A significant increase in muscle strength was found in EG compared to CG (p < 0.0001), except in knee extensors (p = 0.184).
Coroian et al. (2018) [29]	Four measurement time points (pre and post intervention, 3 months after and 6 months after) -Spasticity: MAS. -Function: UL-FMA -Strength: Maximum force evaluated with dynamometer	Spasticity: No significant differences in spasticity were found between the different time points in both groups (p = 0.4). No significant differences were found between groups (p = 0.98). Function: No significant differences were found pre and post intervention in the total UL-FMA score between the two groups (p = 2). No differences were found in the proximal UL-FMA score. In subsequent time points, no significant differences were found in UL-FMA scores between the two groups. There was a significant improvement for EG in the total UL-FMA score pre and post intervention (p < 0.01), which was maintained at 3 months (p < 0.01) and 6 months (p < 0.01) Strength: No significant differences were found in changes in dynamometer scores in different time points for elbow flexors (p = 0.2), elbow extensors (p = 0.3), wrist flexors (p = 0.1) and wrist extensors (p = 0.1). No significant differences were found between groups in dynamometer scores for elbow flexors (p = 0.2), elbow extensors (p = 0.8), wrist flexors (p = 0.2) or wrist extensors (p = 0.3).
Dehno et al. (2021) [30]	Two intervention measurement time points (pre and post intervention) -Spasticity: MMAS. -Function: UL-FMA. -Strength: On the less affected side, maximum force evaluated with dynamometer. On the most affected side, the Medical Research Council scales.	Spasticity: A significant improvement was found in the MMAS score in EG (p = 0.002). There were no differences in the MMAS score in CG (p = 0.165). A significant change in spasticity was found in EG compared to CG (p = 0.014). Function: Significant improvements were found for both groups between pre and post intervention in the total UL-FMA score (EG p = 0.001, CG p = 0.001). The improvement in the UL-FMA score in the EG was significantly greater than in the CG (p = 0.04). Strength: There was a significant improvement in the strength of the less affected side in EG (p = 0.001). There were no significant differences in the strength of the less affected side in CG (p = 0.106). There was a significant improvement in the strength of the less affected side between EG and CG (p = 0.001). There was a significant improvement in the strength of the most affected side compared to the start of treatment in EG (p = 0.001) and CG (p = 0.001). There was a significant improvement in EG compared to CG (p = 0.029).
Fernandes et al. (2015) [32]	Two measurement time points (pre and post intervention) -Spasticity: MAS. -Balance: BBS	Spasticity: There were no significant differences between or within the groups (p ≥ 0.05). Balance: The results in the BBS showed significant differences in both groups (EG p = 0.002, CG p = 0.008). The comparison between groups after the intervention showed that there was a significant difference, with the EG achieving a greater improvement (p = 0.008).
Fernandez et al. (2016) [34]	Two measurement time points (pre and post intervention) -Spasticity: MAS -Strength: Maximum isometric and dynamic force evaluated in leg press -Gait: TUG. -Balance: BBS.	Spasticity: There were no differences between or within the groups. Strength: Differences were found in the isometric strength of the affected leg after the intervention in EG (p = 0.02). An improvement in isometric strength was found in EG compared to CG, although it was not significant (p = 0.06). There was an improvement in the dynamic strength of both legs after the intervention in EG (p = 0.03). A significant improvement in dynamic strength was found in EG compared to CG (p = 0.03). Gait: An improvement was found in TUG after the intervention in EG (p = 0.01) but not in CG. Significant improvements were found in TUG in EG compared to CG (p = 0.04). Balance: An improvement in BBS after the intervention was found in both EG (p < 0.001) and CG (p = 0.01). Significant improvements were found in EG compared to CG (p < 0.001).
Flansbjer et al. (2008) [35]	Three measurement time points (pre and post intervention, and 5 months after the intervention) -Spasticity: MAS. -Strength: Dynamic strength on a knee exercise machine. Maximum isometric force assessed with a dynamometer. -Gait: TUG, Fast Gait Speed, and 6MWT.	Spasticity: A significant improvement in the MAS score was found after the intervention in EG (p < 0.01) and CG (p = 0.02), which did not continue at follow-up. There were no significant differences between EG and CG after the intervention or at follow-up. Strength: There were significant improvements in dynamic strength both after the intervention and at follow-up in EG (p < 0.001), for both paretic and non-paretic limbs. For CG, significant improvements in dynamic strength were found after intervention in the non-paretic limb (p < 0.05), but not in the paretic, and at follow-up only non-paretic flexion was significantly higher (p < 0.05) than at baseline. There were significant differences between EG and CG after the intervention (p < 0.001) and at follow-up (p < 0.001). There were significant improvements in isokinetic strength for both limbs both after the intervention and at follow-up in EG (p < 0.01). No differences were found for CG in isokinetic strength at intervention or follow-up. There was a significant difference between EG and CG (p < 0.05) after the intervention for non-paretic limb extension and flexion, and at follow-up for non-paretic limb extension. Gait: For EG, all gait tests improved significantly (p < 0.05) after the intervention, and that change was maintained at follow-up in TUG and 6MWT scores. For CG, only TUG improved significantly (p < 0.05) after the intervention. There were no significant differences between EG and CG between pre and post intervention, but there were significant differences at follow-up for TUG score (p < 0.05).
Mun et al. (2019) [36]	Two measurement time points (pre and post intervention) -Spasticity: Biodex system assessing the resistance to mobilization within different speed movements. -Gait: TUG -Balance: BBS; Platform for calculating load distribution in standing with eyes open and closed.	Spasticity: There was a decrease in spasticity in EG between pre and post intervention at angular velocities of 60°/sec, 180°/sec, and 240°/sec (p < 0.05). There was a decrease in spasticity in CG between pre and post intervention at angular velocities of 180°/sec and 240°/sec (p < 0.05). EG decreased spasticity statistically significantly compared to CG at angular velocities of 180°/sec (p = 0.02) and 240°/sec (p = 0.04). Gait: There was a significant decrease in the TUG score for both groups after the intervention (p < 0.05). No statistically significant differences were observed between the two groups (p = 0.11). Balance: There was a significant improvement in the BBS score for both groups after the intervention (p < 0.05). The BBS score in EG increased significantly compared to the CG (p < 0.01). An increase in weight distribution to the paretic side was found in both groups, both with eyes open and closed, after the intervention (p < 0.05). EG statistically significantly increased weight distribution with both eyes open (p = 0.04) and closed (p = 0.03) compared to CG.
Lattouf et al. (2021) [37]	Two measurement time points (pre and post intervention) -Spasticity: MAS -Gait: 10-m Walk Test; 6MWT -Strength: Maximum force calculated from Brzycki’s equation	Spasticity: There were no differences between or within the groups. Gait: For both groups, a significant difference was found in the time of the 10-m Walk Test (p ≤ 0.00001), with a higher walking speed after the intervention. No differences were found in the time of the 10-m Walk Test between the two groups after the intervention. A significant effect was observed in 6MWT between pre and post treatment for CG (p ≤ 0.0003) and for EG (p ≤ 0.0001). The results showed a statistically significant difference between the two groups (p ≤ 0.01).
		Strength: There was a significant difference between pre and post treatment for both groups (CG p ≤ 0.0001, EG p ≤ 0.0001). EG showed a significantly greater increase after treatment than CG (p ≤ 0.014).
Patten et al. (2013) [31]	Four measurement time points (pre-evaluation, in the rest period, at the end of the interventions, and at 6 months) -Spasticity: Ashworth Scale -Function: WMFT-FAS, UL-FMA, FIM	Spasticity: No significant changes were found in the Ashworth score at the post-intervention assessment or at 6 months (p > 0.05).
		Function: Significant improvements in WMFT-FAS were found after treatment block 1 in both groups (p < 0.05). These differences were significantly greater after the Hybrid Group compared to the Functional Physical Therapy group (p = 0.03). Tests of a period effect revealed greater improvements in WMFT-FAS after Hybrid versus Functional Physical Therapy (p = 0.02), regardless of where they occurred in the order of treatment. Overall, no differences were revealed because of the order of treatment (p = 0.43). A significant increase in WMFT-FAS was observed during the 6-month follow-up period (p = 0.03). No differences were revealed between Order A and Order B at the 6-month follow-up (p > 0.05). A significantly higher proportion of participants (51% vs. 39%) achieved the minimum significant difference of two points or more in the FIM after the Hybrid (p = 0.05). These positive changes were observed in 69% of participants at 6 months (p = 0.05). Post-intervention improvements were detected in both the total score and the shoulder-elbow portions of the UL-FMA, but these were not statistically significant. Significant differences were found for UL-FMA at 6 months after the intervention, with the minimum significant difference reaching 53% of all participants (p = 0.04)

6MWT, 6-Minute Walk Test; BBS, Berg Balance Scale; CG, Control Group; EG, Experimental Group; FIM, Functional Independence Measure; MAS, Modified Ashworth Scale; MMAS, Modified Modified Ashworth Scale; TUG, Timed Up & Go; UL-FMA, Upper Limb Fugl-Meyer Assessment; WMFT-FAS, Wolf Motor Function Test-Functional Abilities Scale.