Efficacy and safety of mesenchymal stem cells in the treatment of systemic sclerosis: a systematic review and meta-analysis

Table 3 Summary of findings and certainty of evidence for efficacy

	Summary of findings		Certainty of evidence				Certainty of evidence
	No of participants (No of trials)	Mean difference (95% CI)	Study design	Inconsistency	Imprecision	Small study effects	Certainty of evidence
mRSS
3 m	22 (3)	4.11 (2.19 to 6.02)	Downgraded*	Not downgraded	Not downgraded	Not downgraded	Moderate
6 m	52 (5)	5.09 (3.38 to 6.81)	Downgraded*	Not downgraded	Not downgraded	Not downgraded	Moderate
12 m	19 (2)	6.49 (4.61 to 8.37)	Downgraded*	Not downgraded	Not downgraded	Not downgraded	Moderate
mRSS
SVF	30 (2)	4.65 (0.74 to 8.57)	Downgraded*	Not downgraded	Not downgraded	Not downgraded	Moderate
UC-MSCs	19 (2)	5.08 (3.10 to 7.05)	Downgraded*	Not downgraded	Not downgraded	Not downgraded	Moderate
BMSCs	3 (1)	6.70 (− 0.22 to 13.62)	Downgraded*	Not downgraded	Not downgraded	Downgraded^c	Low
RCS
6 m	30 (2)	1.80(− 3.38 to 6.99)	Downgraded*	Downgraded^a	Not downgraded	Not downgraded	Low
DU
6 m	36 (4)	21.10 (3.63 to 122.56)	Downgraded*	Not downgraded	Downgraded^b	Downgraded^c	Very low
VAS
6 m	92 (3)	7.58 (0.55 to 14.60)	Downgraded*	Downgraded^a	Not downgraded	Downgraded^c	Very low
MHISS
12 m	69 (2)	5.52 (2.41 to 8.62)	Downgraded*	Downgraded^a	Not downgraded	Not downgraded	Low
CHFS
6 m	30 (2)	9.05 (− 27.01 to 45.11)	Downgraded*	Downgraded^a	Downgraded^b	Downgraded^c	Very low

mRSS, modified Rodnan skin score; SVF, stromal vascular fraction; UC-MSCs, umbilical cord mesenchymal stem cells; BMSCs, bone marrow-derived mesenchymal stem cells; RCS, Raynaud’s condition score; DU, digital ulcer; VAS, visual analogue scale; CHFS, cochin hand function scale score; MHISS, mouth handicap in systemic sclerosis scale; m, month
*Downgraded by one level because > 25% of participants in this comparison were from studies at high risk of bias
^aDowngraded by one level because heterogeneity (I²) > 50%
^bDowngraded by one level because the limits of the 95% confidence interval were 20 points different to smallest worthwhile effect
^cDowngraded by one level owing to small study bias

ISSN: 1757-6512