Standard deviation of the mean difference out of Pre/Post Test Descriptive Statistics

[ecommon]

I have several RCT pre/post intervention studies that I am calculating Hedge's g for. All of the studies report the difference of mean and difference of standard deviation between pre/post for each group except one.

I know how to calculate the difference of mean [= Mean(post)-Mean(pre)], but for the life of me cannot figure out how to calculate the difference of SD without having the individual scores.

Any ideas?