In general, it is not appropriate to take the mean of the standard deviations (SDs) of pre- and post-intervention measures to use in a meta-analysis. This is because the SDs are dependent on the means and variances of the pre- and post-intervention measures, and combining them in this way would not accurately reflect the true variability of the change scores.


Instead, a more appropriate method for calculating the standard deviation of change scores in a meta-analysis would be to use the standard deviation of the change scores themselves. This can be calculated by subtracting the pre-intervention measure from the post-intervention measure for each individual, and then taking the standard deviation of these change scores.


Alternatively, you can use the Hedges' g as a measure of effect size. Hedges' g is a standardized mean difference, calculated as the difference between the means of the two groups (pre- and post-intervention) divided by a pooled estimate of the standard deviation of the difference scores. This method is robust to small sample size and provide more accurate estimate of mean difference.


It's worth noting that there are other considerations when conducting a meta-analysis, such as selecting appropriate studies, accounting for potential sources of bias, and interpreting the results in the context of the overall literature. It's always recommended to consult with a statistician or an expert in the field before doing meta-analysis.