Table 2.

Risk of bias for included trials.

Study	Sequence generation	Allocation concealment	Blinding	Incomplete outcome data	Selective outcome reporting	Other sources of bias
Zhong 2014 [26]	(−)	0	0	(−)	(+)	(−)
Zhang 2008 [25]	(−)	0	0	(−)	(−)	(−)
Zhang 2016 [24]	0	0	0	(−)	(−)	(−)
Ruan 2014 [18]	(−)	0	0	(−)	(−)	(−)
Zhong 2011 [23]	(−)	0	0	(−)	(+)	(−)
Xiang 2011 [21]	0	0	0	(−)	(−)	(−)
Wang 2010 [19]	(−)	0	0	(−)	(−)	(−)
Zhou 2005 [20]	0	0	0	(−)	(−)	0
Xu 2014 [22]	(−)	0	0	(−)	(−)	0
Liu 2010 [17]	0	0	0	(−)	(−)	(−)
Liang 2014 [16]	(−)	0	0	(−)	(−)	(−)
Hao 2014 [15]	(−)	0	0	(−)	(−)	(−)
Gao 2015 [14]	(−)	0	0	(−)	(−)	(−)
Li 2015 [27]	0	0	0	(−)	(−)	0
Ya 2010 [28]	0	0	0	(−)	(−)	(−)
Zhang 2016 [29]	(−)	0	0	(−)	(−)	(−)

Note. Low risk of bias: (−); unclear: 0; high risk of bias: (+).