I, myself, would not consider trying to publish a systematic review/meta-analysis that was done on only 3 studies. I, instead, might consider broadening the research question so that more studies could be included.
If a student asked me your question, I might tell him/her to consider doing the systematic review/meta-analysis if he/she might consider presenting the results at a small research conference.
My self-imposed "minimum number of studies for a systematic review/meta-analysis" rule is 10.
I would agree with Elfatih Malik that a scoping review might be more appropriate and I would highlight the issue you identified: that there are only three studies in this topic and that is a good idea for further research. I would really emphasise that point.