Four RCTs (n=140) and 6 open trials (case series) (n=120) were included in the review.
Bupropion (2 RCTs and 1 case series).
One RCT (n=40) found that more patients treated with bupropion (76%) achieved a greater than 30% reduction in the ADHDRS compared with those given placebo (37%). One RCT (n=37) found that more patients treated with bupropion (64%) achieved a CGI-I score equal to 1 or 2 compared with those given methylphenidate (50%) or placebo (27%). The difference between the treatment groups was not significant (P=0.14). One case series (n=19) found that 74% of patients given bupropion achieved a CGI-I score of to 1 or 2.
Venlafaxine (4 case series).
One case series (n=18) found that 50% of patients given venlafaxine achieved a CGI-I score equal to 1 or 2, while a further case series (n=10) found that 78% of patients achieved a CGI-I score of less than or equal to 3 and a 33% reduction in ARS. One case series (n=16) found that 63% of patients given venlafaxine achieved a greater than 30% reduction in the WURS. One case series (n=20) found that 67% of patients had an 'excellent' response to venlafaxine.
Tricyclic antidepressants (1 RCT and 1 case series).
The RCT (n=41) found that significantly more patients treated with desipramine achieved a CGI-I score equal to 1 or 2 compared with those given placebo (68% versus 0%; P<0.0001). The retrospective case series found that 68% of patients given desipramine and 42% of those given nortiptyline achieved a CGI-I score equal to 1 or 2.
Atomoxetine (1 RCT).
The RCT (n=22) found that significantly more patients treated with atomoxetine achieved a greater than 30% reduction in ADHDRS compared with those given placebo (52% versus 9.5%; P<0.01).