Hemofiltration compared to hemodialysis for acute kidney injury: systematic review and meta-analysis

Table 3 Risk of bias of included trials

Trial	Sequence generation	Concealment of allocation	Trial stopped early for benefit	Intention to treat analysis	Post-randomization withdrawals for mortality analysis (parallel group trials) or not crossed over
*Parallel group*
Davenport 1993 [21]	List of random numbers^a	Yes^a	No	Yes	Yes - 1/12 CVVHD(lost to follow up)
Alamartine 1994 [22]^a	n/r	n/r	n/r	n/r	n/r
Pettila 2001 [23]	Computer generated	Yes^a (closed envelopes)	No	Yes	Yes - 1/18 IHD(consent withdrawn)
Morgera 2004 [24]	Alternating patients^a	No	No	Yes	n/r
Daud 2006 [25]	Alternating patients^a	No	No	Yes	n/r
Saudan 2006 [26]	Computer generated	Yes (sequentially numbered sealed opaque envelopes)	Yes^a	Yes	No
Chang 2009 [27]	Alternating patients^a	No	No^a	Yes	No^a
Ratanarat 2009 [28]	List of random numbers^a	n/r	n/r	n/r	n/r
Ratanarat 2012 [29]	List of random numbers	n/r	n/r	n/r	n/r
OMAKI 2012 [30]	Computer generated	Yes (sequentially numbered sealed opaque envelopes)	No	Yes	Yes - 1/39 CVVHD(inclusion mistake)
*Crossover*
Maher 1988 [31]	n/r	n/r	n/r	n/r	n/r
Alarabi 1992 [32]	n/r	n/r	n/r	n/r	n/r
Jeffery 1994 [33]	n/r	n/r	n/r	n/r	n/r
Kellum 1998 [34]	n/r	n/r	n/r	n/r	Yes - 2 CVVH and 1 CVVHD/13 died prior to crossover
Garcia-Fernandez 2000 [35]	n/r	n/r	No	n/r	n/r
Maxvold 2000 [36]	n/r	n/r	n/r	n/r	n/r
Wynkel 2004 [37]	n/r	n/r	n/r	n/r	n/r
Ricci 2006 [38]	n/r	n/r	n/r	n/r	n/r
Davies 2008 [39]	n/r	n/r	n/r	n/r	Yes - 14/45 not crossed over

^aObtained after author contact. Abbreviations: CVVH, continuous venovenous hemofiltration; CVVHD, continuous venovenous hemodialysis; IHD, intermittent hemodialysis; n/r, not reported.

ISSN: 1364-8535