The risk of miscarriage in the first trimester of pregnancy is fairly high anyway, that it would be very easy for a pregnant women who miscarried and had recently had Interferential to suggest a causal relationship between the two and you would leave yourself wide open to a law suit.
I wouldn't use interferential as a treatment modality anyway, but I certainly wouldn't use it in a pregnant women for exactly the reasons said.
I do not believe you could clinically reason that it would be severely detrimental to the patient if IFT was not used for a shoulder pain and this I think is the key to the argument.
As has already been said, surely the risk of miscarriage (evidence or no evidence) cannot be outweighed by a need for pain relief- there are plenty of alternative treatment choices out there, even if it is a case of no intervention during the few months of pregnancy.
This is similar to the arguments for ACP in pregnancy which I researched recently, again a serious lack of research but ethically you can't do this kind of research on pregnant women!
M