Inversion Recovery (STIR, FLAIR)
- Inversion recovery flips the magnetization upside down first, then waits for one tissue to pass through zero signal — and you take the picture exactly then, so that tissue goes black.
- The timing of that pause is called the inversion time (TI). Choose the TI to match the tissue you want to silence.
- STIR uses a short TI to null fat, leaving fluid and edema gloriously bright. Great for bone marrow and musculoskeletal swelling.
- FLAIR uses a long TI to null free fluid (CSF), so bright lesions next to the ventricles stop hiding in the bright spinal fluid.
- STIR nulls by T1, not by chemistry, so it suppresses fat broadly and evenly — but it is not selective and can null other things that share fat's T1.
Most MRI sequences are about which tissue you make bright. Inversion recovery is the rebel of the family: it is about which tissue you make disappear. And once you understand the one trick behind it, both STIR and FLAIR stop being acronyms to memorize and become the same idea wearing two different hats.
The one trick: flip first, then wait for zero
Normally a scan starts by tipping the magnetization sideways and reading the signal. Inversion recovery adds a step before that: a 180-degree pulse that flips the magnetization completely upside down. Now every tissue starts its recovery from the bottom of the pool and has to climb back up to the top.
Here is the magic. As each tissue climbs from fully-inverted back to fully-relaxed, it has to pass through zero on the way — that brief instant where it has no longitudinal magnetization to give. If you fire your imaging pulse at the exact moment a particular tissue is sitting at zero, that tissue contributes no signal. It goes black. Everything else, which is at some other point in its climb, still has signal to give.
Think of a row of swimmers all pushed to the bottom of a pool at once. They surface at different speeds depending on their tissue type. If you snap the photo at the precise second one swimmer breaks the surface — the moment they're neither down nor up but passing through the waterline — that swimmer is invisible in your shot. The rest are caught mid-stroke and show up fine.
The "exact moment" is the inversion time (TI) — the delay between the flip and the snapshot. Pick the TI, pick who vanishes.
STIR: nulling fat to let swelling shine
STIR stands for Short Tau Inversion Recovery (tau is just another name for TI). Fat recovers quickly, so it reaches its zero-crossing early — at a short TI. Set the TI there and fat goes dark.
Why bother? Because fat is bright and loud on many sequences, and it loves to drown out the subtle bright signal of edema — fluid where it shouldn't be. Bone marrow is full of fat, so a marrow bruise or an early stress fracture can hide in plain sight. Knock the fat down with STIR and that waterlogged, angry tissue lights up like a road flare against a now-dark background.
STIR suppresses fat by its T1 (its recovery speed), not by chemically targeting fat's specific resonance. That makes it robust and uniform even on lopsided magnetic fields where chemical fat-saturation falls apart — one reason STIR is a workhorse for musculoskeletal and whole-spine imaging.
This is also why STIR pairs so naturally with the broader family of fat suppression techniques — it is the T1-based member of that club.
STIR is not selective — it nulls anything whose T1 happens to match fat's. So if you inject gadolinium contrast, enhancing tissue can shorten its T1 toward fat's range and get partly suppressed too. The classic rule: don't run STIR for fat suppression on a post-contrast scan, or you may erase the very enhancement you were hunting for. Use chemical fat-sat after contrast instead.
FLAIR: nulling CSF to unmask the brain
FLAIR stands for FLuid-Attenuated Inversion Recovery. Free fluid — cerebrospinal fluid (CSF) — recovers slowly, so it crosses zero late, at a long TI. Set the TI there and CSF goes black.
The payoff is enormous in the brain. On a plain T2-weighted image, both CSF and many lesions (plaques, edema, gliosis) are bright. A small bright lesion sitting right next to the bright, sloshing ventricles is camouflage at its finest — bright on bright. FLAIR drains the CSF to black, so that periventricular lesion suddenly pops out against dark fluid.
FLAIR is essentially a heavily T2-weighted image with the CSF turned off. It is the bread-and-butter sequence for hunting multiple sclerosis plaques, small infarcts, and subtle gliosis precisely because it strips away the distracting bright spinal fluid.
Same trick, two timings
So STIR and FLAIR are not two unrelated tools — they are the same maneuver with the dial set differently:
| Sequence | TI length | Tissue nulled | What you're hunting |
|---|---|---|---|
| STIR | Short | Fat | Edema, marrow lesions, MSK swelling |
| FLAIR | Long | Free fluid (CSF) | Brain lesions next to the ventricles |
Inversion recovery doesn't make a tissue bright — it makes one tissue vanish so everything else is easier to see. Short TI silences fat (STIR); long TI silences CSF (FLAIR).
If you remember nothing else: flip the magnetization upside down, wait for your troublemaker tissue to pass through zero, and take the picture right then. Pick the wait time, pick who disappears. Everything else about these sequences is just bookkeeping around that single, elegant moment of silence.