Gradient echo
Drop the 180° refocusing pulse and MRI gets fast enough for breath-holds and dynamic contrast. The cost is T2*: gradient echo sees every field imperfection that spin echo would have rewound, which is sometimes a flaw and sometimes the whole point.
By the ScanFlo Editorial Team · Updated
In this section · MRI fundamentals
What changes when the 180° pulse goes
A spin echo needs time. The 180° pulse has to be played out and the echo has to wait for it. A gradient echo does without the refocusing pulse. After a short excitation, the readout gradient is switched on in one direction to dephase the spins and then reversed to bring them back into phase. The echo arrives early, TE can be a few milliseconds, and TR can be short enough to fit a whole slice into a breath-hold.
The price is paid in what the gradient reversal cannot undo. It rephases only the dephasing that the gradient caused. Dephasing from the field itself, from small inhomogeneities, from air next to tissue, from a metal screw or from iron in old blood, is left in the signal. The contrast is therefore T2*, not T2, and it gets stronger as TE lengthens.
Flip angle takes over from TR
With a TR of tens of milliseconds, a 90° pulse would leave almost no longitudinal magnetisation to excite next time. Gradient echo uses smaller flip angles instead, and the flip angle becomes a contrast control. A large flip angle with a short TR saturates tissue with long T1 and gives T1 weighting. A small flip angle lets everything recover and shifts the weighting towards proton density, or T2* if TE is long.
| Weighting | Flip angle | TE | Where you see it |
|---|---|---|---|
| T1 (spoiled GRE) | Larger, e.g. 60–80° | Short | Dynamic contrast, in/opposed phase |
| T2* | Small, e.g. 10–30° | Longer | Blood products, cartilage |
| PD | Small | Short | Some MSK work |
Those flip angles are orientation values for 2D imaging, not a recipe. 3D T1 sequences such as MPRAGE and VIBE use smaller angles with their own preparation.
When T2* sensitivity is the point
The weakness is sometimes exactly what the referrer wants. Haemosiderin from an old bleed and microbleeds from amyloid or trauma distort the local field and appear as dark blooming on T2*, often larger than the deposit itself. Susceptibility-weighted imaging exaggerates the same effect on purpose. Calcification and gas behave similarly.
The same sensitivity hurts near the skull base, the sinuses, bowel gas, dental work and orthopaedic hardware. If a gradient echo looks wrecked around metal, a spin echo with the same timing will usually rescue it.
Fat and water, in and out of phase
Fat and water protons precess at slightly different frequencies, so after excitation they drift in and out of phase with each other. A gradient echo reads the signal at a moment of your choosing. At 1.5 T, fat and water are opposed at about 2.2 ms and aligned again at about 4.4 ms (at 3 T, roughly half those values). A voxel containing both loses signal on the opposed-phase image. That simple comparison finds fatty infiltration of the liver and characterises adrenal adenomas, and it is the basis of Dixon fat-water separation, covered in fat suppression.
Two families worth recognising
Spoiled gradient echo destroys the leftover transverse signal before each new excitation, which keeps the contrast clean and T1-driven. It is the sequence behind most breath-hold abdominal and dynamic contrast imaging. Balanced steady-state free precession keeps that leftover signal instead, giving bright blood and fluid with very high signal-to-noise ratio in little time. It is the standard for cardiac cine, and it is prone to dark banding artefacts when the field is not uniform.
Try it in ScanFlo: On the brain case, run a T1 spin echo and then a gradient echo with similar timing. Look at the frontal lobes above the sinuses and the temporal lobes near the petrous bone. Then lengthen TE on the gradient echo and watch the signal loss grow. See ScanFlo MRI →