Good points made by your previous correspondants.
Sorry to confuse things even more, but valves can be very difficult to radiograph successfully due to their complex shape. We have had experiences with cast feed control valves on boilers that have contained defects even though they were radiographed following production. Sometimes vlaves are so complex in shape that forging isn't possible.
In our industry, our 'big' steam valves have to be shown to have a very low probability of failure. Cast valves come under intense scrutiny for such features as evidence of repairs but then we have to because of the plant we operate.
Really I think you have to put the choice into context. Provided a manufacturer is making the components fully in accordance with the specification, provides all the required documentation and has demonstrated that the required NDT has been carried out and that there are no specific requirements for YOUR plant, and if there are that you are satisfied that the NDT will detect the required defect types, then either a cast valve or a forged valve may be acceptable (provided its the right thickness and material).
If you are worried about failure then you get into a whole different ball park when you have to be concerned about whether it would leak before breaking or fracture.
I can't recall that many instances when a valve (body) has failed dramatically, but there are several instances where plant conditions (erosion etc) has caused the valve to bypass due to erosion of an existing casting defect.
I'm not familar with ANSI B16.34 so I can't be more specific about your particular application. Generally (I hope someone can back me up here) if the standard doesn't say you can't use a particular manufacturing method then it is OK to go for either and the choice then is down to either economics (along with the points I made earlier) or any specific requirements your plant or regulator might have.
Hope this hasn't made things more difficult.
Cheers
Andy