Noninvasive Brain Stimulations for Unilateral Spatial Neglect after Stroke: A Systematic Review and Meta-Analysis of Randomized and Nonrandomized Controlled Trials

Table 1

GRADE evidence profile for RCTs: noninvasive brain stimulations for unilateral spatial neglect after stroke.

Quality assessment						Illustrative comparative risks (95% CI)		Certainty in estimates orquality of evidence
						Assumed risk	Corresponding risk
Number of participants (studies); Range follow-up; Time in weeks	Risk of bias	Inconsistency	Indirectness	Imprecision	Publication bias	Sham	Noninvasive brain stimulations
Overall USN measured by star cancellation test
116 (6); Immediately postintervention; 4 weeks	Serious limitation^1	Serious limitation^2	No serious limitation^3	Serious limitation^4	Undetectable	The mean in change in USN measured by star cancellation test was 45.29 (SD 5.94)	The std. mean in changes in USN measured by star cancellation test in the intervention group was on average 0.51 fewer (1.89 fewer to 0.88 more)	Very low
Overall USN measured by line bisection test
107 (5); Immediately postintervention; 1 month	Serious limitation^1	Serious limitation^2	No serious limitation^3	No serious limitation	Undetectable	The mean in change in USN measured by line bisection test was 35.79 (SD 18.65)	The std. mean in changes in USN measured by line bisection test in the intervention group was on average 2.33 fewer (3.54 fewer to 1.12 fewer)	Low
Overall USN measured by Motor-Free Visual Perception Test
38 (2) 2–4 weeks	Serious limitation^1	No serious limitation	No serious limitation	No serious limitation	Undetectable	The mean in change in USN measured by Motor-Free Visual Perception Test was 16.9 (SD 2.1)	The std. mean in changes in USN measured by Motor-Free Visual Perception Test in the intervention group was on average 1.46 more (0.73 more to 2.20 more)	Moderate
Overall USN measured by Albert test and line crossing test
50 (2) 4 weeks	Serious limitation^1	Serious limitation^2	No serious limitation	No serious limitation	Undetectable	The mean in change in USN measured by Albert test and line crossing test was 27.33 (SD 4.55)	The std. mean in changes in USN measured by Albert test and line crossing test in the intervention group was on average 1.01 more (1 fewer to 3.02 more)	Low

SD = standard error; std. = standardized. Baseline risk estimates for overall USN come from control arm of Fu et al.’s [52] study (lowest risk of bias trial in the meta-analysis). Baseline risk estimates for overall USN come from control arm of Cha et al.’s [51] study (newest trial in the meta-analysis). ¹The majority of the studies were ranked as high risk of bias for both allocation sequence and allocation concealment. ²There was a substantial heterogeneity (²> 70%). ³There was no substantial difference related to the mean age and eligibility criteria throughout the six included studies. ⁴95% CI for absolute effects includes clinically important benefit and no benefit.