Peripheral CE-MRA — three-station bolus-chase runoff

Three overlapping stations from the infrarenal aorta to the pedal arch: abdomen and pelvis, thighs, and calves and feet, with a pre-contrast mask acquired at each station in the same geometry for subtraction.

Typical, not policy

Timings, volumes and delays here are representative values drawn from published guidance. Scanner generation, injector, cardiac output and local preference all move them. Confirm against your department's own protocol before you rely on a number.

When to use it

Runoff mapping in claudication or critical limb ischaemia once duplex has established that revascularisation is being considered — the NICE CG147 sequence, where this is the offered alternative to CT angiography.
Patients in whom iodinated contrast is undesirable but gadolinium is acceptable, which in this population is common.
Heavily calcified tibial vessels, where CT angiography is degraded by blooming and MR is not.
Younger patients and those needing repeated mapping, where cumulative radiation from serial CT runoff is the dominant concern.
Assessment of bypass graft patency and of the distal target vessel before planning.

Technique

The whole examination is a race: the bolus is chased distally as the table moves, so the calf station must be reached while enhancement is still arterial and before venous return begins.
Time-resolved imaging of the calf station first, acquired as a separate small dose before the bolus-chase run, is the standard solution where venous contamination is expected — in critical limb ischaemia and in the inflamed, infected or cellulitic foot it is almost mandatory.
A pre-contrast mask at each station in identical geometry is what allows subtraction; if the patient moves between mask and arterial run, the subtraction fails and the station is lost.
Unsubtracted source images should always be reviewed alongside the subtracted maximum intensity projections, because a stent, a clip or a haematoma is visible on the source data and invisible on the projection.

Where it goes wrong

Venous contamination of the calf station is the classic failure and is worst in exactly the patients who need the study most — inflammation, infection and critical ischaemia all accelerate venous return.
Motion between mask and arterial acquisition destroys the subtraction; rest pain that prevents the patient keeping still is a reason to plan analgesia in advance rather than to discover it mid-study.
Metallic stents produce a susceptibility void that mimics occlusion. Without a stent history, a patent stented segment is reported as occluded.
MR angiography systematically overestimates the degree of stenosis relative to catheter angiography, particularly in small distal vessels, so a borderline lesion should not be treated on the MR appearance alone.
A mistimed bolus that fills the centre of k-space too early gives ringing and edge blurring rather than an obviously non-diagnostic image, which makes the error easy to miss.

Clinical questions that reach this study

Peripheral arterial disease — imaging before revascularisation

Contrast

Gadolinium, intravenous

A gadolinium bolus by power injector, typically a single weight-based dose of about 0.1 mmol/kg for a standard macrocyclic extracellular agent, at around 1-2 mL/s with a 20-30 mL saline chaser, timed by a test bolus or MR fluoroscopic triggering in the distal aorta. Some centres use a biphasic injection, slowing the rate for the distal stations so that arterial enhancement is sustained as the table moves.

The injection rate here is deliberately lower than for a single-station neck or renal MRA. A three-station chase needs enhancement sustained over roughly a minute of table travel rather than a single tight peak, so amplitude is traded for duration — which is why the same 0.1 mmol/kg is delivered at 1-2 mL/s rather than 3 mL/s.
Where the agent differs, the dose differs. Gadobenate dimeglumine has roughly twice the intravascular relaxivity of a standard macrocyclic agent, so units using it hold the dose at about 0.1 mmol/kg and gain signal rather than volume — but note that in the EU and UK the EMA 2017 linear-agent review restricted it to LIVER imaging only, so it is not a licensed option for a runoff study in those jurisdictions; gadofosveset, the blood-pool agent that made steady-state imaging after the first pass possible at around 0.03 mmol/kg, was withdrawn from both the European and US markets in 2017 and is named here only because the technique it enabled still appears in older protocols; ferumoxytol-enhanced MRA is what has taken its place in this niche, which is a different acquisition strategy rather than the same one with a different drug.
A high-relaxivity or blood-pool agent is used in some units precisely because the calf station is the one that fails; local agent choice is what makes a given timing scheme work.
Station timing is the study: the abdominopelvic mask and arterial run go first at the triggered delay (commonly around 15-25 s from injection), then thighs, then calves, with the calf station typically reached at roughly 50-90 s. Some protocols acquire the calf station first from a separate small bolus precisely because it is the station a chase most often loses.

Acquisition

Breathing: Breath-hold for the abdominopelvic station; free breathing for the thigh and calf stations, with the legs immobilised and padded.
Preparation: State whether the question is claudication, critical limb ischaemia or graft surveillance, and which leg is symptomatic — the station timing is set to reach the symptomatic calf while the bolus is still arterial. Supply the duplex result: this study is a runoff map to plan intervention, and it is being requested because duplex has already established that revascularisation is on the table. Declare stents, grafts and previous amputations; metallic stents cause a signal void that is read as occlusion if their presence is unknown. A long acquisition with the legs immobilised — assess tolerance, rest pain and analgesia at vetting, because movement between stations is the commonest cause of a failed study.

Phases

Each phase is authored once and shared across every protocol that uses it, so the physiology below is the same wherever you meet it.

Contrast-enhanced MR angiographyAcquisition is synchronised to bolus arrival in the target vessel, which is short and variable (of the order of 12–25 s from injection in timing-bolus studies), using MR fluoroscopic triggering or a test bolus, with a centric or elliptic-centric k-space order. Confirm locally.
Gadolinium shortens blood T1 so that vessels are bright on a heavily T1-weighted acquisition, independently of flow direction or velocity — which is the whole reason it outperforms non-contrast time-of-flight techniques in slow, turbulent or in-plane flow. The transferable principle is that image contrast is set by the moment the CENTRE of k-space is filled, not by when the acquisition starts or ends: fill the centre while the bolus is arterial and the study is arterial even if sampling continues for another half-minute; fill it a few seconds late and the same raw data yield a venous-contaminated study. Triggering and view ordering are therefore not technical preferences but the phase definition itself.

Safety checks this protocol carries

Derived from the contrast agent and phases above, not authored here — which is why they cannot drift apart from what the protocol actually does.

Prior contrast reaction and elective premedication· nurse pre scan
Kidney function and gadolinium-based contrast· radiographer at scan
Intravenous access adequate for the planned injection· radiographer at scan
MR safety screening for implants and foreign bodies· radiographer at scan