Because it's hard. I helped out at a company among many that are trying to develop a system that would allow you to do this in ambulance and they started testing prototypes 10 years ago and have yet to progress from that stage.
It's not that hard... you just throw some ice packs in the groin and armpits. You won't get them fully into the TTM range in a 20 minute transport, but you can start the process.
Packing with ice packs is a standard component of TTM, generally in conjunction with other therapies (chilled IV fluids, etc). It's also the easiest step to take in a prehospital setting, which is why we do (this is a thing I've done with maybe a dozen patients over the past few years). I'm not just making this stuff up, it's based on actual treatment protocols and algorithms.
We're talking about a naked, sedated patient (so they're already losing a lot of heat due to simple convection, and there's no heat being generated by skeletal muscle movements). The body's metabolism has slowed significantly, and it's producing anywhere near the normal level of "baseline" heat.
Do you have any links to data showing effects on brain temperature from that treatment?
Afaik interest has been low in carrying around enough ice pack capacity to make this viable, and with lacking data for it even working, whereas head-cooling packs with circulating liquid through a cooler takes little space, cools way more efficiently and required no other maintenance after insertion into the patient in the ambulance (and is moved with the patient while continously working).
Overcooling is more of a risk with I've packs than not cooling enough. It's much less of a concern in a prehospital setting though, since we're not with the patient that long. It's more of an issue in the ICU.
This procedure is initiated after the heart has been restarted[1]. You're absolutely correct that it's pointless to do it earlier, especially since the far more important thing is getting the heart beating again.
[1] This is known as "targeted temperature management" (TTM), and is part of the standard post-ROSC (return of spontaneous circulation) algorithm in ACLS (Advanced Cardiac Life Support), if you're looking for terms to Google.
It will but much more slowly. Think of cooling the blood as using the surface area of the arteries and veins as a heat sink to cool the body. Without that blood moving your next best heat sink is the skin so start piling on the ice.
The body itself is the heat sink. The skin is a radiator. The blood is the liquid in a liquid cooling system (moving heat from the sink to the radiator). To your point, liquid cooling systems become _much_ less effective when the liquid stops pumping.
Those spots are not prone to frostbite, since the "coldness" gets distributed through the rest of the body very quickly (apologies for playing a little fast and loose with thermodynamics there). This is why they're used in the first place to cool the person quickly.
