Table 7:

GRADE Evidence Profile for Tremor Severity After MRgFUS Neurosurgery

Number of Studies (Design)	Risk of Bias	Inconsistency	Indirectness	Imprecision	Publication Bias	Upgrade Considerations	Quality
Compared With Sham
1 (RCT)38	No serious limitations	No serious limitationsa	No serious limitations	No serious limitationsb	Undetecteda	NA	⊕⊕⊕⊕ High
Compared With Deep Brain Stimulation
2 (Observational)40,46	Serious limitations (−1)c	No serious limitations	No serious limitations	Serious limitations (−1)d	Undetecteda	NA	⊕ Very Low
Compared With Radiofrequency Thalamotomy
1 (Observational)46	Serious limitations (−1)c	No serious limitationsa	No serious limitations	No serious limitations	Undetecteda	NA	⊕ Very Low
MRgFUS Only
6 (Observational)36,37,39,41,42,45	No serious limitationse	No serious limitationsf	No serious limitations	No serious limitationsg	Undetectedh	NA	⊕⊕ Low

Abbreviations: MRgFUS, magnetic resonance-guided focused ultrasound; NA, not applicable; RCT, randomized controlled trial.

^aCannot definitively assess presence or absence because the evidence is derived from a single study.

^bStudy adequately powered, confidence interval for absolute difference in severity scores relatively narrow, mean difference and upper and lower bounds are clinically meaningful, as is relative improvement.

^cRisk of bias was highest owing to retrospective data collection methods. The full risk-of-bias assessment is presented in Appendix 2, Table A2.

^dConsiderable imbalance in group sizes for comparisons, optimal information size criteria not met, and no measures of variance or confidence intervals provided; therefore, uncertainty remains in the precision of estimates.

^eRisk of bias overall judged to be low or unclear for most studies; one study (Zaaroor et al⁴⁵) judged to be at high risk of bias resulting from participant selection. The full risk-of-bias assessment is presented in Appendix 2, Table A2.

^fVariability in exact magnitude of effect, but of questionable importance as all estimates indicate clinically meaningful benefit.

^gSome studies did not meet optimal information size criterion, yet found both statistically significant and clinically meaningful results that are on the same side of the clinical decision threshold.

^hInadequate information reported from studies to formally assess using funnel plot or statistical tests; however, the studies vary in terms of sample size and are generally small.