LC doesn't have scripted multithreading per se, but neither do other languages like Python that support a good number of concurrent users, so by itself that's not necessarily prohibitive.
I haven't read the code for LC's latest HTTPd lib, but socket comms in LC have a callback option that can radically improve concurrent throughput when needed. See the "with <message>" option in the Dictionary entries for relevant commands like "open socket".
By using callbacks, the wait times for network I/O are offloaded to the OS, freeing the LC process to continue to handle more incoming connections even as the OS is still buffering the last one.
Where async behavior is useful and you want to keep the LC process lean, callbacks can deliver more than an order of magnitude more throughput than without them.
But all that said, you probably neither want nor need to use any scripted implementation of HTTPd for a public service, whether written in LC or anything else.
You probably don't want to because, no matter how crafty you are, any scripted implementation will be less efficient than compiled object code, and more so when that code is written by an army of specialists as with Apache, NginX, Lighttpd, etc.
And you don't need to because just about any HTTPd implementation will provide a way to extend request handling using a separate process, such as LC. Apache and Lightttpd support CGI, which uses multiprocessing rather than multithreading for concurrency, and keeps request handling discrete for clear, clean codiing. NginX and others support reverse proxies, where you can set up a pool of workers to handle requests, which can be written in just about any language, including LC.
LC's HTTPd library is an excellent choice for what it was designed for: a lightweight implementation suitable for local development and testing.
But in production, as they say, go big or go home.
We have plenty of options for allowing great use of LC on servers, without needing to replicate the generic request brokering part of it that's so well handled by apps designed specifically for that.