Table 5.

GRADE assessment of the quality of the evidence and the strength of the recommendations.

Certainty assessment							No. of patients		Effect		Certainty	Importance
No. of studies	Study design	Risk of bias	Inconsistency	Indirectness	Imprecision	Other considerations	[Intervention]	[Comparison]	Relative (95% CI)	Absolute (95% CI)
VAS
13	Randomised trials	Not serious	Not serious	Seriousa	Not serious	Publication bias strongly suspected dose response gradientb	905	943	–	SMD 0.47 lower (0.74 lower to 0.19 lower)	⊕⊕⊕◯ moderate	Critical
% No/mild pain
6	Non-randomised studies	Not serious	Seriousc	Seriousa	Not serious	Publication bias strongly suspectedb	144/481 (29.9%)	172/543 (31.7%)	OR 0.71 (0.32 to 1.59)	69 fewer per 1000 (from 188 fewer to 108 more)	⊕◯◯◯ very low	CriticaL
Days with no/mild pain
2	Non-randomised studies	Not serious	Not serious	Seriousa	Not serious	None	229	237	–	MD 9 higher (8.93 higher to 9.07 higher)	⊕◯◯◯ very low	Critical
n opioid conumption
3	Non-randomised studies	Not serious	Not serious	Seriousa	Not serious	Strong association dose response gradient	71/218 (32.6%)	125/225 (55.6%)	OR 0.50 (0.33 to 0.76)	171 fewer per 1000 (from 264 fewer to 68 fewer)	⊕⊕⊕◯ moderate	Critical
Change SF-12/SF-36/EQ-5D
9	Non-randomised studies	Not serious	Not serious	Seriousa	Not serious	Publication bias strongly suspected strong association dose response gradientb	548	471	–	SMD 0.39 higher (0.11 higher to 0.68 higher)	⊕⊕◯◯ low	Critical
Any
7	Randomised trials	Not serious	Seriousc	Seriousa	Not serious	Publication bias strongly suspectedb	265/445 (59.6%)	289/499 (57.9%)	OR 1.15 (0.67 to 1.97)	34 more per 1,000 (from 99 fewer to 151 more)	⊕◯◯◯ very low	Important

CI, confidence interval; MD, mean difference; OR, odds ratio; SMD, standardised mean difference.

^a Different etiologies, doses, study design and follow-up times.

^b Publication bias assessed by visual inspection of funnel plots.

^c The results showed a wide variability.