It just occurred to me why I've had no problem reviewing dozens of texts for my classes but I haven't gotten around to evaluating even a single game engine for classroom use (even though I have about five on my to-do list).
Evaluating a textbook is easy. I'm already familiar with most of the content I'd be looking for, so I can skim chapters. I'll typically read a few key areas closely (those that I plan to emphasize the most in class), which gives me a good idea of the author's general approach. I can usually get through a textbook this way in an afternoon or two. If I'm assigning readings in a course, I can read the passage myself the night before we talk about it in class, so the details can be spaced out.
A game engine, by contrast, demands that I learn something that I'm unfamiliar with (the particular capabilities and syntax for that engine) and then create a couple sample projects with it in order to get a feel for how easy it is to develop with. It takes me an afternoon or two just to read the documentation, and then another couple weeks or so to get used to developing games with this particular tool. And if I'm going to demand (or even suggest) the use of a game engine in a course, I should be proficient from the beginning since students will likely have some advanced questions as soon as they start. So, the time commitment is much larger and also must be given up front -- two things that are the bane of a full-time teacher.
I haven't yet found a way around this. I'm not sure there is a way around it; I (and the other teachers I've talked to) use the tools in class that we're familiar with.