3.5 Gradient Echo Pulse Sequences

3.5 Gradient Echo Pulse Sequences

Gradient echo (GRE) pulse sequences represent one of the foundational families of MRI acquisition. Unlike spin echo sequences, which utilize a 180° radiofrequency (RF) rephasing pulse, gradient echo sequences use a combination of low flip angles and gradient reversal to generate echoes. This makes them highly versatile, exceptionally fast, and sensitive to susceptibility changes, which has critical implications for clinical imaging.

Physics of Gradient Refocusing

The primary defining characteristic of a gradient echo sequence is the absence of a 180° RF refocusing pulse. After the initial excitation RF pulse (which is typically a low flip angle, < 90°), the transverse magnetization begins to dephase due to spin-spin interactions (T2 decay) and fixed magnetic field inhomogeneities (B0 inhomogeneities, susceptibility effects, and chemical shift).

To rephase the spins and produce a measurable echo, a bipolar gradient is applied along the frequency encoding axis:

1. Dephasing lobe: A negative gradient lobe is applied first, which rapidly dephases the spins in a controlled manner.
2. Rephasing lobe: The polarity of the gradient is then reversed to positive. This rephasing lobe has twice the area of the dephasing lobe, which forces the precessing spins to rephase.

Because gradients only reverse the phase changes induced by the gradients themselves—and do not correct for fixed field inhomogeneities, susceptibility differences, or chemical shift—the resulting echo is called a gradient echo. The decay of this echo is governed by T2 (T2-star) relaxation* rather than true T2 relaxation. The relationship between T2 and T2* is defined by the equation:

1 / T2* = 1 / T2 + γ * ΔB0

Where γ is the gyromagnetic ratio and ΔB0 represents the local magnetic field inhomogeneity.

Flip Angle Selection and the Ernst Angle

Gradient echo sequences typically employ flip angles between 5° and 70°, which are significantly lower than the 90° excitation pulses used in spin echo sequences. This has a profound effect on image contrast and scan time:

• Low flip angles: Leave a large portion of the magnetization along the longitudinal axis (+z), allowing for rapid longitudinal recovery. This enables the use of very short repetition times (TR), often under 50 ms, drastically reducing scan times.
• Tissue Contrast: Contrast in GRE sequences is determined by a combination of the flip angle, TR, and echo time (TE).
  • T1 Weighting: Large flip angles (typically 30° to 70°) coupled with short TR and short TE maximize T1 contrast by saturating tissues with long T1 relaxation times.
  • T2 Weighting*: Small flip angles (typically 5° to 20°) coupled with long TE (to allow T2* dephasing) and long TR (to minimize T1 saturation) yield T2* weighting.
  • Proton Density (PD) Weighting: Small flip angles (typically 5° to 20°) coupled with short TR and short TE minimize both T1 and T2* contrast.

The relationship between the flip angle, TR, and a tissue's T1 relaxation time is optimized using the Ernst Angle (θE). The Ernst angle is the flip angle that yields the maximum signal intensity (highest SNR) for a specific tissue at a given TR. It is mathematically calculated as:

cos(θE) = e^(-TR / T1)

The Steady State

When the TR is shorter than the T1 and T2 relaxation times of the tissue (typically when TR is less than 50 ms), the spins do not have enough time to fully recover their longitudinal magnetization or fully dephase their transverse magnetization before the next RF pulse is applied. Under these conditions, the system enters the steady state.

In the steady state, there is a continuous coexistence of both longitudinal magnetization and transverse magnetization. The transverse magnetization does not decay to zero between pulses; instead, it is periodically refocused by subsequent RF pulses, establishing a dynamic equilibrium. Image contrast in the steady state is heavily dependent on the ratio of T1 to T2 (T1/T2), rather than their absolute values. Tissues with a high T2/T1 ratio (such as water, cerebrospinal fluid, and fat) appear bright.

Spoiled (Incoherent) vs. Refocused (Coherent) GRE

Steady-state sequences are broadly categorized based on how they handle the residual transverse magnetization before each new RF excitation pulse.

Spoiled (Incoherent) GRE

In spoiled (incoherent) gradient echo sequences, the residual transverse magnetization is deliberately destroyed (spoiled) before the next RF pulse is applied. This prevents the transverse magnetization from contributing to subsequent echoes, ensuring that image contrast is dominated by T1 and proton density recovery. Spoiling can be achieved in two ways:

1. RF Spoiling: The phase of the RF excitation pulse is digitally varied in a pseudo-random sequence for every TR, preventing the establishment of coherent transverse magnetization.
2. Gradient Spoiling: A strong spoiler gradient is applied along the slice-select or readout axis at the end of each TR cycle to rapidly dephase the remaining transverse spins.

Clinical Use: Spoiled GRE (e.g., SPGR, FLASH) is widely used for high-resolution 3D anatomical imaging, dynamic contrast-enhanced (DCE) abdominal imaging, and MR angiography, where pure T1 contrast is required.

Refocused (Coherent) GRE

In refocused (coherent) gradient echo sequences, the residual transverse magnetization is preserved and actively rephased. This is accomplished by applying a rewinder gradient (a phase encoding gradient of equal magnitude but opposite polarity to the original phase encoding step) at the end of each TR cycle. This rewinder gradient brings the transverse spins back into phase coherence before the next RF excitation pulse.

As a result, the signal is a mixture of a gradient echo and a stimulated spin echo. This sequence produces strong T2* contrast. Clinical Use: Coherent GRE (e.g., GRASS, FISP) is utilized for rapid joint imaging, cine cardiac imaging, and MR myelography, as it produces a bright fluid signal.

Steady-State Free Precession (SSFP)

SSFP sequences are designed to read out only the refocused stimulated echo rather than the initial free induction decay (FID). This produces true T2-like contrast, even with very short TRs. A common variant is Balanced SSFP (bSSFP) (e.g., FIESTA, TrueFISP), which uses balanced gradients on all three axes to ensure a net phase shift of zero over each TR cycle. This yields extremely high SNR and high contrast between blood/fluid (bright) and surrounding tissue (dark).

Clinical Applications and Tradeoffs

Gradient echo sequences offer several major advantages:

• High Scan Speed: Enables breath-hold imaging, minimizing respiratory motion artifacts.
• Sensitivity to Flow: Moving spins entering the slice have not experienced repeated RF pulses, making them "fresh" and hyperintense. This is the basis for inflow MRA.
• Sensitivity to Magnetic Susceptibility: Ideal for detecting hemorrhage (due to paramagnetic hemosiderin), calcification, and cortical bone interfaces.

However, GRE sequences also present distinct disadvantages:

• Magnetic Susceptibility Artifacts: Severe signal loss and geometric distortion can occur near metallic implants, dental work, and air-tissue interfaces (e.g., sinuses).
• No Correction for Inhomogeneities: Because there is no 180° RF pulse, field inhomogeneities degrade the signal, making it less suitable for high-uniformity imaging over large volumes.