Can you clarify what you’re trying to do? If by intrathecal, you mean into the CSF in a region where there isn’t any spinal cord (L4/5), then yes, people do this all the time. You just use an insulin syringe.
You just take 10 ul of injectate, find the interspace between L4/L5 based on landmarks (illiac crest), insert the needle between the interspace, and then inject. If you’re in the right place, you usually see a tail flick. You can do this under anesthesia, or, if you’re good and fast, even awake. @tonellor from @tberta lab does them awake, as does @zhzhj131421. I do it both ways.
Now lumbar enlargement indicates something different. That sounds to me like you want to inject directly into the substance of the spinal cord itself. That is an entirely different procedure. Is that what you’re trying to do? For that, you do need to expose the cord directly and use totally different tools and approaches.
The least invasive is this approach, which I use:
https://www.nature.com/articles/srep14306
We’ve discussed this here before: