Standardised mean difference with data transformations in meta analysis?

Is it appropriate to use the SMD when some studies report raw data and some report log transformed data? ie cortisol data 

28 October 2016 2,833 7 View