Runoff MRA maps the entire peripheral arterial tree from the juxtarenal aorta to the pedal arch in a single contrast injection. This guide walks radiographers, radiologists, and referring physicians through the multi-station stepping-table technique, the physics of asymmetric calf venous return, and the ten-step workflow that keeps below-the-knee vessels diagnostic.

Runoff MRA Protocol: 10 Steps to Master Peripheral Vascular Imaging

Introduction to runoff MRA

Runoff MRA is the single-injection, multi-station examination of the entire lower-extremity arterial tree, extending from the juxtarenal abdominal aorta through the iliac, femoral, popliteal, and tibial-peroneal segments down to the pedal arch. It exists to answer one clinical question with a single non-invasive test: where, and how severely, is the peripheral arterial circulation narrowed or occluded in a patient with claudication, rest pain, or tissue loss. Because the vascular territory it covers spans nearly the full length of the body, runoff MRA is fundamentally different from every other angiographic protocol in this series — it is not a single field-of-view acquisition but a coordinated sequence of three to four separate 3D volumes, stitched together by an automated moving table that tracks the advancing contrast bolus in real time.

The clinical stakes are high. Peripheral arterial disease affects a substantial proportion of adults over 65, and undiagnosed or mis-staged disease drives amputation rates that could often be prevented with timely revascularisation.^[1] Runoff MRA has become a first-line alternative to catheter digital subtraction angiography (DSA) in many vascular centres because it avoids arterial puncture, ionising radiation, and iodinated contrast nephrotoxicity, while providing comparable diagnostic accuracy for stenosis grading across the aortoiliac, femoropopliteal, and infrapopliteal segments.^[2]

What makes this protocol technically demanding — and the reason it merits its own dedicated entry in this series — is the physics of a single gadolinium bolus attempting to stay “ahead” of the imaging table as both move distally down the leg. The contrast bolus does not travel at a constant, predictable velocity once it crosses from the aorta into two separate limb circulations, and the two limbs frequently do not behave identically. This single fact — asymmetric calf venous return — is the dominant technical challenge of runoff MRA and the focus of a large portion of this article.

A well-executed runoff MRA depends on four interlocking systems working in synchrony: an injector protocol that delivers a compact, reproducible bolus; a stepping-table sequence that advances the field of view at a velocity matched to bolus transit; a coil configuration that maintains homogeneous signal reception across the full body length; and a calf-specific acquisition strategy capable of resolving the arterial phase before venous contamination sets in. Each of these systems is discussed in detail below, alongside the anatomy, relaxation physics, dose considerations, and interpretive pitfalls that radiographers, radiologists, and referring physicians must understand to use this protocol safely and effectively.

The evolution of this protocol is itself instructive. Early moving-bed contrast-enhanced MRA, first described in the late 1990s, relied on a single-phase acquisition per station with bolus timing estimated from population averages — an approach adequate for the aortoiliac and femoropopliteal segments but chronically unreliable at the calf, where individual variation in venous filling time is largest.^[4] The subsequent introduction of parallel imaging, elliptical-centric k-space ordering, and — most importantly — time-resolved acquisition techniques transformed the calf station from the weakest link in the examination into a segment that can now be interrogated with confidence in the majority of patients. Runoff MRA today sits alongside CT angiography (CTA) as one of the two dominant non-invasive alternatives to catheter DSA, with the choice between MRA and CTA in a given patient typically driven by renal function, the presence of heavily calcified plaque (which MRA visualises differently from CTA), and local departmental expertise.^[2] Radiographers who understand precisely why the calf station behaves differently from the two stations proximal to it are far better equipped to intervene in real time — adjusting cuff pressure, confirming bolus arrival, or triggering a repeat acquisition — before a diagnostically compromised dataset ever reaches the reporting radiologist.

Peripheral vascular anatomy

The peripheral arterial tree begins at the distal abdominal aorta, just below the renal artery origins, and terminates in the pedal arch of the foot. For runoff MRA planning purposes, this course is conventionally divided into three or four imaging stations, each corresponding to a separate 3D acquisition volume triggered as the table advances.

Station 1: Abdominopelvic segment

The infrarenal aorta bifurcates at approximately the L4 vertebral level into the right and left common iliac arteries. Each common iliac artery divides into an internal iliac (hypogastric) branch, supplying the pelvic viscera and gluteal musculature, and an external iliac artery, which continues as the common femoral artery once it passes beneath the inguinal ligament. This station typically also captures the origins of the renal arteries, the inferior mesenteric artery, and — in male patients — occasionally visualises portions of the internal pudendal circulation. Aortoiliac disease is a common and clinically important finding here, since occlusive disease at this level (Leriche syndrome, when bilateral aortoiliac occlusion produces claudication, impotence, and absent femoral pulses) has direct implications for surgical planning.

Station 2: Femoropopliteal segment

The common femoral artery bifurcates into the superficial femoral artery (SFA) and the profunda femoris (deep femoral) artery. The SFA is the most common site of atherosclerotic occlusive disease in the lower extremity, particularly within the adductor (Hunter’s) canal, where the vessel passes through a fibrous tunnel formed by the adductor magnus tendon and is subject to repetitive mechanical stress. Beyond the adductor hiatus, the SFA becomes the popliteal artery, which courses posterior to the knee joint through the popliteal fossa. The popliteal artery is conventionally subdivided into P1 (proximal, above the joint line), P2 (at the joint line), and P3 (distal, below the joint line) segments — a classification directly relevant to endovascular treatment planning.

Station 3 (and 4): Infrapopliteal / calf segment

Below the knee, the popliteal artery gives rise to the anterior tibial artery, which passes through the interosseous membrane to run along the anterior compartment of the leg, and the tibioperoneal trunk, which shortly divides into the posterior tibial artery and the peroneal (fibular) artery. These three vessels — anterior tibial, posterior tibial, and peroneal — constitute the classic “three-vessel runoff,” and the number of patent vessels reaching the ankle is a key determinant of wound-healing potential in patients with critical limb ischaemia. At the ankle, the anterior tibial artery becomes the dorsalis pedis artery, and the posterior tibial artery passes behind the medial malleolus to supply the plantar arch. The pedal arch, formed by anastomoses between the dorsalis pedis and the lateral plantar artery, provides collateral perfusion to the forefoot and is of particular importance in diabetic patients with distal disease.

Superimposed on this arterial anatomy is the venous system that runs in close anatomical proximity throughout — the paired venae comitantes accompanying the tibial and peroneal arteries below the knee, the popliteal vein, the superficial and deep femoral veins, and the iliac veins. It is this parallel venous architecture, filling with gadolinium-laden blood only slightly delayed relative to the arteries, that generates the central artefact addressed throughout this protocol.

Collateral pathways and anatomical variants

Chronic arterial occlusion in the peripheral circulation rarely produces a simple “dead end” on imaging; the body develops predictable collateral networks that reconstitute flow distal to an occlusion, and recognising the typical course of these collaterals is essential for accurate interpretation. In aortoiliac occlusion, collateral flow commonly develops via the lumbar arteries to the iliolumbar and superior gluteal circulation, and via the internal mammary and inferior epigastric arteries in more chronic, severe disease (the “corkscrew” collateral pattern). In SFA occlusion within the adductor canal, the profunda femoris artery and its perforating branches typically reconstitute the popliteal artery via geniculate collaterals around the knee. Below the knee, collateral vessels tend to run a serpiginous course closely paralleling the native tibial arteries, which is precisely why they can be mistaken for a patent native vessel on projection imaging if source images are not reviewed.

Anatomical variants are common enough in the peripheral vasculature to warrant deliberate attention during interpretation. A high bifurcation of the popliteal artery, a persistent sciatic artery (a rare embryological remnant that can be the dominant supply to the lower limb), and duplicated or hypoplastic tibial vessels are all recognised variants that can be mistaken for pathology by an unwary reader unfamiliar with normal anatomical variation in this vascular bed.

MR tissue relaxation values

Runoff MRA relies almost entirely on T1-shortening from gadolinium-based contrast rather than intrinsic tissue T1/T2 contrast, since the diagnostic signal is arterial blood during first-pass enhancement, not the surrounding soft tissue. Understanding baseline relaxation values, however, remains essential for troubleshooting background suppression, fat signal, and the appearance of adjacent musculoskeletal structures on source images.

The clinical objective of contrast administration is to drive arterial blood T1 down to a value substantially shorter than that of surrounding fat and muscle, so that a heavily T1-weighted spoiled gradient echo sequence — using a short TR, short TE, and a moderate-to-large flip angle — renders the contrast-filled lumen unambiguously brighter than background tissue. Because venous blood reaches the same post-contrast T1 value as arterial blood once opacified, MRA has no intrinsic tissue-contrast mechanism for separating artery from vein; separation is achieved purely through timing — acquiring central k-space while gadolinium concentration is high in arteries but still low or absent in veins.

The magnitude of T1 shortening achieved is governed by the relaxivity of the specific gadolinium chelate used, its concentration in blood at the moment of central k-space acquisition, and the field strength of the scanner, since relaxivity itself is mildly field-dependent. Standard extracellular agents at typical peripheral MRA doses reduce arterial blood T1 from over a second to well under 100 milliseconds within the first pass, which is why even sub-second variations in acquisition timing relative to bolus arrival can meaningfully change image contrast — a physiological sensitivity that underlies much of the timing discipline emphasised throughout this protocol.

Scanning technique: 10-step protocol

Runoff MRA is executed as a coordinated sequence spanning patient setup, calibration, and three-to-four timed acquisitions. The following ten-step workflow reflects standard clinical practice for a single-injection, stepping-table, contrast-enhanced study.

1. Coil and patient positioning. Position the patient supine, feet first, on a dedicated peripheral vascular surface coil array spanning from the mid-abdomen to the ankles, or a combination of body coil plus lower-extremity phased-array coils. Immobilise the legs with padding and secure straps to minimise inter-station motion, and place the patient’s arms above the head or at the sides to avoid wrap artefact across the abdomen.
2. Localiser and station planning. Acquire a rapid multi-station coronal localiser (typically a fast SSFP or GRE sequence) covering the full anticipated field of travel. Define three or four overlapping 3D imaging volumes — abdominopelvic, thigh, and calf (with an optional dedicated foot station) — ensuring at least 20–30% overlap between adjacent stations at the femoral and popliteal levels to prevent anatomical gaps after table advancement.
3. Coil sensitivity / parallel imaging calibration. Perform the vendor-specific parallel imaging calibration scan (SENSE, GRAPPA, ASSET, or equivalent reference scan) at each station, since acceleration factors of 2–3 are routinely used to shorten acquisition time per station and improve temporal fidelity relative to bolus arrival.
4. Pre-contrast mask acquisition. Acquire an identical, unenhanced 3D GRE mask dataset at each station using the same geometry and parameters that will be used post-contrast. This mask enables digital subtraction, removing background fat and soft-tissue signal to isolate the enhancing vascular tree.
5. Bolus timing determination. Establish contrast arrival time using either a test bolus (1–2 mL test injection with rapid 2D dynamic imaging at the aortic level) or automated bolus-triggering software (fluoroscopic triggering) monitoring the distal aorta. Average contrast transit time from an antecubital vein to the common femoral artery is approximately 24 ± 6 seconds, with an additional 5 ± 2 seconds to reach the popliteal artery.^[4]
6. Contrast injection initiation. Begin the gadolinium injection at the calculated flow rate, immediately followed by the saline chaser, while the abdominopelvic station acquisition is triggered to coincide with the arrival of peak gadolinium concentration in the aorta, sampling central k-space during this window.
7. Automated table tracking through thigh station. As the abdominopelvic acquisition completes, the table automatically advances to the thigh station, timed so that central k-space acquisition for this station coincides with peak arterial concentration in the femoropopliteal segment, before venous filling becomes significant.
8. Calf station acquisition with time-resolved technique. Because calf venous return timing is unpredictable and frequently asymmetric between limbs, deploy a time-resolved dynamic sequence (TWIST or TRICKS) at this station rather than a single-phase acquisition, capturing multiple sequential sub-phase datasets so that a pure arterial frame can be selected retrospectively for each leg independently.
9. Optional dedicated foot/pedal arch acquisition. In patients with suspected distal disease, tissue loss, or diabetic foot ulceration, obtain a dedicated high-resolution acquisition of the foot with a narrower field of view and finer in-plane resolution to resolve the pedal arch and digital vessels.
10. Post-processing and subtraction. Perform mask subtraction at each station, apply maximum intensity projection (MIP) reconstruction in multiple obliquities, and stitch the subtracted station volumes into a single composite whole-leg angiogram for interpretation, cross-referencing source images at any segment where MIP clutter or venous overlay is suspected.

1.5T vs. 3.0T comparison for runoff MRA

Field strength selection in practice is rarely a free choice made purely on technical merit; it is usually dictated by scanner availability, patient body habitus (larger patients may experience more pronounced dielectric artefact and SAR constraint at 3T), and the presence of MRI-conditional implants with field-strength-specific labelling. When both field strengths are genuinely available, 3T is generally preferred for patients in whom distal tibial or pedal vessel conspicuity is the clinical priority — such as those being evaluated for a distal bypass target in critical limb ischaemia — while 1.5T remains an entirely adequate and often more time-efficient choice for standard claudication work-up focused primarily on the aortoiliac and femoropopliteal segments.

Contrast media protocol

Runoff MRA is, by definition, a contrast-dependent examination — unlike TOF-based intracranial MRA, there is no reliable non-contrast alternative that provides comparable spatial coverage and diagnostic confidence across the entire lower-extremity arterial tree in a clinically practical scan time, although quiescent-interval slice-selective (QISS) and other flow-dependent non-contrast techniques are increasingly used as adjuncts or alternatives in patients with contraindications to gadolinium.^[5]

Because the injection must sustain a constant flow rate across an unusually long duration — often 25–35 seconds of active contrast delivery to keep pace with three-station table advancement — flow-rate consistency is more clinically consequential in runoff MRA than in almost any other contrast-enhanced MRI protocol in this series. A single-use or well-maintained dual-syringe injector minimises the internal tubing resistance variability that can desynchronise the bolus front from the moving table, a failure mode that directly worsens venous contamination at the calf station.

Test bolus versus fluoroscopic triggering

Both methods are validated and widely used; the choice is largely institutional, based on available software and departmental workflow preference. What matters clinically is that some form of individualised timing is used rather than a fixed population-average delay, given the wide inter-patient variability in circulation time driven by cardiac output, vessel calibre, and the severity of any pre-existing occlusive disease.

Renal function screening remains mandatory before any gadolinium administration, consistent with standard nephrogenic systemic fibrosis (NSF) precautions; given the peripheral vascular disease population’s high prevalence of concomitant chronic kidney disease and diabetes, group II or III agent selection and eGFR-based risk stratification deserve particular attention in this protocol.^[7]

Agent selection considerations

Standard extracellular gadolinium chelates remain the workhorse agent class for runoff MRA and are appropriate for the great majority of patients. High-relaxivity agents, which achieve greater T1 shortening per unit dose through weak, transient protein binding, offer a theoretical advantage in the distal calf and pedal stations where signal is most marginal, though the incremental benefit must be weighed against agent-specific safety profiles and institutional formulary considerations. Regardless of agent class, dose should be calculated on actual body weight and capped according to product labelling and departmental policy, since the 0.2 mmol/kg dose specified for this protocol already sits toward the upper end of typical extracellular agent dosing and should not be exceeded without specific clinical justification.

Renal function stratification

This stratification follows established ACR Manual on Contrast Media guidance and should be applied consistently regardless of how urgently the referring team requests the examination, since NSF risk is a function of agent class and renal function rather than clinical urgency.^[7]

Injection line and patient factors

The venous access site matters more in runoff MRA than in most other MRI protocols in this series, because the injection must sustain a constant, power-injector-compatible flow rate for an unusually long duration. A right antecubital line is generally preferred over a hand or lower-extremity line, both for injector compatibility and because a lower-extremity injection site would introduce local venous enhancement that directly interferes with the very calf station the protocol is designed to interrogate. In patients with poor peripheral venous access, a central line rated for power injection may be necessary, with departmental protocols for confirming line patency and injection pressure limits followed strictly given the sustained flow rate involved.

Specific absorption rate and dose reduction

Runoff MRA presents an unusual SAR profile compared with other protocols in this series: because the examination covers an unusually large body length across multiple stations, cumulative RF energy deposition must be tracked across the entire multi-station sequence rather than a single localized acquisition, and whole-body averaged SAR (rather than local head or partial-body SAR) is typically the governing constraint, particularly at 3T. Larger patients, who present a greater absolute tissue mass over which RF energy is absorbed, and patients undergoing an extended four-station protocol that includes a dedicated foot acquisition, are most likely to approach regulatory SAR ceilings and therefore benefit most directly from the dose-reduction strategies below.

Five dose-reduction strategies

1. Reduce flip angle at each station. Since SAR scales approximately with the square of flip angle, modest reductions (for example from 30° to 25°) in the 3D GRE sequence meaningfully lower RF deposition with minimal impact on vessel-to-background contrast, given that gadolinium T1-shortening already provides the dominant contrast mechanism.
2. Lengthen TR where bolus timing allows. A marginally longer TR reduces duty cycle and therefore SAR, though this must be balanced against the need to complete each station’s acquisition within the arterial-phase timing window.
3. Increase parallel imaging acceleration. Higher SENSE/GRAPPA factors reduce the number of RF excitations required per station, directly lowering both SAR and scan time — a strategy with dual benefit for this time-critical protocol.
4. Use vendor SAR-optimised RF pulse design. Many platforms offer low-SAR or “whisper” RF pulse variants for 3D GRE angiographic sequences that preserve image contrast while reducing per-station energy deposition, particularly valuable at 3T.
5. Stage SAR budget across stations. Where cumulative whole-body SAR approaches regulatory limits across a four-station protocol, inserting brief inter-station pauses (compatible with table repositioning time) allows partial SAR dissipation without materially lengthening total table time.

In practice, most modern scanner platforms perform real-time SAR prediction before each station acquisition begins and will automatically flag or block a sequence that would exceed the regulatory ceiling, prompting the technologist to adjust one or more of the parameters above. Understanding the underlying trade-offs — rather than simply reducing flip angle by default whenever a SAR warning appears — allows the technologist to select the specific adjustment least likely to compromise the arterial-phase timing that the entire examination depends upon.

Top 10 pathologies detected on runoff MRA

The pathology spectrum encountered on runoff MRA spans the full length of the lower-extremity arterial tree, and disease distribution is not uniform: atherosclerotic occlusive disease clusters heavily at the SFA within the adductor canal and at the infrapopliteal trifurcation, while aneurysmal disease is disproportionately represented at the popliteal segment. The ten entries below represent the findings most frequently encountered in routine practice and most directly influenced by the technical quality of the acquisition.

Pitfalls for radiographers

The CSV-specified primary scanning pitfall for this protocol is asymmetric calf venous return — a physiological reality in which the two calves opacify their venous systems at measurably different times relative to the arterial phase, undermining any single fixed-timing acquisition strategy applied uniformly to both legs.

This asymmetry arises from a combination of factors that vary independently between limbs: differences in native venous outflow resistance, subtle asymmetry in arterial inflow related to unilateral occlusive disease itself (a partially occluded limb often has paradoxically slower, more prolonged arterial filling, which delays venous return on that side, while the contralateral healthy limb may already be showing venous enhancement), and inconsistent cuff compression if the two thigh cuffs are not inflated to identical pressure. Recognising that the “problem” leg is often the diseased leg — the very limb the examination is most clinically important for — reinforces why this pitfall cannot simply be tolerated as an acceptable trade-off.

Pitfalls for radiologists

The downstream interpretive consequence of asymmetric calf venous return is misdiagnosis of infrapopliteal occlusive disease — venous overlay superimposed on the tibial arteries at MIP reconstruction can either obscure a genuine stenosis (false negative) or mimic irregular luminal narrowing where none exists (false positive), directly affecting revascularisation planning.

The practical defence against this pitfall is disciplined multiplanar review. A radiologist who relies exclusively on the coronal composite MIP for the calf station is, in effect, trusting a single 2D projection to correctly represent three intersecting vascular structures — the anterior tibial artery, the posterior tibial artery and vein, and the peroneal artery and vein — compressed into one plane. Scrolling through source axial images at the level of any equivocal finding, and reviewing individual sub-phase frames from the time-resolved acquisition where available, resolves the great majority of apparent calf-station abnormalities that would otherwise generate diagnostic uncertainty or an unnecessary recommendation for confirmatory catheter angiography.

Pitfalls for non-radiology physicians

Vascular surgeons, interventional cardiologists, endocrinologists managing diabetic patients, and primary care physicians are frequently the ordering or referring clinicians for runoff MRA, and each brings a different set of assumptions about what the resulting report can and cannot tell them. The pitfalls below reflect the points where clinical decision-making most commonly diverges from what the imaging actually demonstrated.

A recurring theme across all five of these pitfalls is the value of direct communication between the ordering clinician and the interpreting radiologist, particularly in complex critical limb ischaemia cases where the treatment decision — endovascular intervention, surgical bypass, or primary amputation — hinges on granular detail that a standard report summary may not fully convey. A brief conversation clarifying which specific findings drove a “poor runoff” or “single-vessel” characterisation is often more clinically valuable than re-reading the report text alone.

Pitfall comparison summary

The three tiers of pitfalls described above are not independent — a scanning-level failure to manage calf venous timing propagates directly into an interpretation-level risk of misdiagnosis, which in turn shapes the clinical decisions a referring physician makes about revascularisation or amputation. Viewing the pitfall framework side by side, as below, helps each professional group recognise not only their own points of vulnerability but also how an error upstream in the workflow can surface as a very different-looking problem downstream.

AI and automation in runoff MRA

Automated bolus detection and table-tracking software has been standard on clinical MRA platforms for over a decade, but newer machine-learning approaches are extending automation further into image reconstruction and interpretation support. Deep-learning-based reconstruction models have demonstrated improved signal-to-noise ratio, vessel sharpness, and reduced venous contamination scores compared with conventional time-resolved reconstruction in intracranial time-resolved MRA, a technique directly analogous to the calf-station approach used in runoff studies.^[8] Several FDA-cleared and CE-marked software platforms now offer automated vessel-tracking and stenosis-quantification tools for peripheral MRA and CTA datasets, providing a structured runoff score to support treatment planning discussions.

Automated bolus-triggering software remains the most mature and widely deployed AI-adjacent tool in this specific protocol, replacing manual “best-guess” scan-delay estimation with real-time fluoroscopic monitoring of contrast arrival at the aortic level, directly reducing the incidence of mistimed acquisitions across all three stations.^[9] Emerging deep-learning denoising and super-resolution reconstruction techniques are also being investigated specifically for time-resolved calf-station data, with early evidence suggesting improved arterial-venous separation compared with conventional view-sharing reconstruction alone.^[8]

Beyond image reconstruction, automated segmentation and runoff-scoring software is increasingly used to generate a structured, per-segment stenosis grade across the full arterial tree from the composite MRA dataset, mirroring the modified Society for Vascular Surgery runoff resistance score traditionally calculated manually from catheter angiography.^[28] These tools can materially reduce inter-reader variability in stenosis grading and provide a standardised, reproducible output that referring vascular surgeons can track longitudinally across serial imaging — though, as with any automated grading tool, output should be spot-checked against source images rather than accepted uncritically, particularly at the calf station where the venous contamination artefact discussed throughout this article remains the dominant source of grading error, human or automated.

A realistic limitation worth stating plainly: no currently available automated tool reliably replaces the clinical judgement required to distinguish a genuine embolic occlusion from a timing-related artefact, or to weigh the significance of an incidental visceral finding against a patient’s overall clinical picture. Automation in runoff MRA today is best understood as augmenting the workflow — accelerating bolus detection, reducing reconstruction time, and standardising quantitative scoring — rather than replacing the radiologist’s synthesis of source images, clinical history, and prior comparison studies.

Further reading

1. 7 Proven Strategies for Optimizing MRI Sequences — bolus timing precision and contrast delivery consistency across time-sensitive MRA and DCE-MRI protocols.
2. 7 Critical CTA Brain & Carotids Protocol Steps — bolus tracking, injection parameters, and venous contamination avoidance in another vascular protocol.
3. Understanding Venous Air Embolism in Contrast-Enhanced Imaging — safety considerations relevant to sustained high-flow-rate injections.
4. Cervical Spine MRI Protocol: 10 Critical Steps — companion protocol illustrating structured pitfall frameworks across radiographer, radiologist, and physician tiers.
5. CT Trauma Pan-Scan Protocol: 7 Critical Steps — multi-phase timing principles applicable to any protocol chasing a moving contrast bolus across body regions.

Reducing artefacts with patients and parameters

Because runoff MRA is acquired across an unusually long field of travel and a sustained injection, the standard levers that govern spatial resolution, SNR, contrast, and artefact control interact with each other more visibly here than in single-station protocols. Understanding these trade-offs helps the scanning team make informed real-time adjustments. Patient-related factors compound these technical trade-offs: body habitus affects both SNR and SAR headroom simultaneously, peripheral vascular disease itself often coexists with tremor, restless legs, or discomfort that increases motion risk during a multi-minute examination, and cardiac arrhythmia — common in this predominantly older, atherosclerotic patient population — can desynchronise the population-average bolus timing assumptions that automated table tracking relies upon.

Spatial resolution

Spatial resolution defines the ability to distinguish small details, which is directly relevant to resolving the 2–3 mm calibre of distal tibial and pedal vessels. Increasing the acquisition matrix increases spatial resolution but decreases SNR because each voxel becomes smaller and receives less signal. Reducing the field of view increases spatial resolution similarly but at the same SNR cost, and must be balanced against the risk of aliasing if anatomy extends beyond the prescribed FOV. Thinner slices improve resolution and reduce partial volume averaging of small tibial vessels against adjacent muscle, but significantly decrease SNR per slice. In practice, the calf and foot stations are typically prescribed with the finest matrix and thinnest slices of the entire examination, since this is where vessel calibre is smallest and where diagnostic errors from inadequate resolution are most consequential for treatment planning.

Signal-to-noise ratio

SNR represents the strength of the diagnostic vessel signal relative to background noise. Increasing the number of signal averages improves SNR but proportionally lengthens scan time — often impractical given the strict timing window available before venous contamination. Decreasing receiver bandwidth boosts SNR by limiting recorded noise, though at the cost of increased chemical shift artefact and longer scan times. Dedicated peripheral vascular surface coils, rather than the body coil alone, meaningfully improve SNR by placing receive elements closer to the vessels of interest along the full leg length. A well-fitted, multi-element coil array spanning the entire imaged length also supports the higher parallel imaging acceleration factors that this protocol depends upon, since coil geometry and element spacing directly determine achievable acceleration without unacceptable noise amplification (the so-called g-factor penalty).

Image contrast

Repetition time and echo time govern the T1-weighting that makes gadolinium-filled vessels appear bright. A short TR maximises T1 contrast, essential for separating enhanced blood from background tissue, while a short TE minimises T2* signal loss and susceptibility-related dropout near metallic implants or calcified plaque. Flip angle further modulates T1-weighting in gradient echo sequences; the flip angle used for runoff MRA is typically moderate (25–35°) to balance vessel-to-background contrast against SAR constraints across multiple stations. Excessively high flip angles can paradoxically reduce apparent vessel signal in slow-flow or partially occluded segments, where inflowing unsaturated blood is less available to replenish signal between repeated RF excitations — a phenomenon technologists should recognise rather than mistake for true absent flow.

Artefact control specific to runoff MRA

Phase-encoding direction should generally be assigned right-to-left (across the leg) rather than head-to-foot, shifting any residual motion or flow-related ghosting away from the vessel course itself. Flow compensation and cardiac gating are less central to this protocol than to protocols imaging pulsatile flow at rest, since the diagnostic signal here depends on gadolinium T1-shortening rather than intrinsic flow-related enhancement — though gating remains relevant for non-contrast QISS-based alternatives. Parallel imaging, discussed in detail below, is the single most impactful artefact- and time-reduction tool available in this protocol, both shortening acquisition time per station relative to bolus transit and reducing motion sensitivity.

Wrap-around (aliasing) artefact deserves specific mention in runoff MRA given the extended anatomical coverage: if the prescribed field of view in the phase-encoding direction is narrower than the patient’s actual girth at the abdominal or thigh station, tissue outside the FOV folds back onto the opposite edge of the image, potentially overlapping the vessel of interest. Phase oversampling, applied liberally at the abdominal station in particular, eliminates this risk at a modest scan-time cost that is well justified given the diagnostic consequences of an aliased vessel segment.

Patient comfort and communication

Because the total table time for a well-executed runoff MRA, including localisers, calibration, mask acquisition, and the timed diagnostic run, typically spans 10–20 minutes, patient comfort directly influences motion risk and, by extension, diagnostic quality. Patients with claudication or rest pain may find prolonged supine positioning with legs extended and immobilised genuinely uncomfortable, particularly if thigh compression cuffs are also applied. Clear pre-scan communication about the purpose and temporary nature of the compression, combined with adequate padding at the knees and ankles, meaningfully reduces the likelihood of a mid-examination position adjustment that would otherwise necessitate repeating the mask acquisition.

Parallel imaging protocols and parameters

Parallel imaging acceleration is essential to runoff MRA because each station must complete its 3D acquisition within a narrow arterial-phase timing window before the table advances and, ultimately, before venous contamination compromises the calf station. Acceleration factor (turbo factor equivalent, expressed as the SENSE/GRAPPA factor) directly trades scan time against SNR, and the appropriate factor differs meaningfully between field strengths given their different baseline SNR headroom.

The practical target at both field strengths is to complete each station’s central k-space sampling well within the individualised arterial transit window established during bolus timing, with parallel imaging acceleration serving as the primary lever for achieving this without sacrificing the spatial resolution needed to resolve sub-4 mm tibial and pedal vessels.

It is worth emphasising that parallel imaging acceleration and time-resolved acquisition are complementary rather than competing strategies at the calf station: the acceleration factor determines how quickly each individual sub-phase of the time-resolved series can be acquired, while the time-resolved technique itself determines how many sub-phases are available for retrospective selection of the optimal arterial frame. Departments running an older platform without robust time-resolved calf capability can still meaningfully reduce venous contamination risk by maximising parallel imaging acceleration alone to shorten the single-phase calf acquisition, though this remains a second-best solution relative to true time-resolved imaging.

Conclusion

Runoff MRA remains one of the most technically demanding protocols in peripheral vascular imaging precisely because it asks a single contrast bolus, a moving table, and an automated tracking algorithm to stay synchronised across nearly the full length of the lower extremity. Its diagnostic value — comprehensive, non-invasive evaluation from the aortoiliac origin to the pedal arch — depends on disciplined execution of every step in this workflow: accurate bolus timing, adequate station overlap, sub-systolic thigh compression, and, above all, a calf-station strategy built around time-resolved acquisition rather than a single fixed-phase capture.

Asymmetric calf venous return is not a rare technical curiosity; it is the expected physiological behaviour of two independent limb circulations, and protocols that fail to account for it will systematically underperform at exactly the anatomical level — the infrapopliteal runoff — where accurate assessment matters most for limb-salvage decision-making. The pitfall framework in this article, spanning scanning technique, image interpretation, and clinical application, reflects the reality that no single professional group can guarantee diagnostic accuracy alone; radiographers, radiologists, and referring physicians each hold a distinct piece of the safety net that keeps runoff MRA reliable for the patients who depend on it.

As deep-learning reconstruction and automated runoff-scoring tools mature, much of the manual burden of calf-station troubleshooting described in this article may gradually shift toward software-assisted workflows. Until that transition is fully validated across diverse patient populations and disease severities, the fundamentals covered here — disciplined bolus timing, deliberate thigh compression, time-resolved calf acquisition, and systematic source-image review — remain the foundation on which every reliable runoff MRA examination is built.

References

1. Fowkes, F. G. R., Rudan, D., Rudan, I., Aboyans, V., Denenberg, J. O., McDermott, M. M., Norman, P. E., Sampson, U. K., Williams, L. J., Mensah, G. A., & Criqui, M. H. (2015). Comparison of global estimates of prevalence and risk factors for peripheral artery disease in 2000 and 2010: a systematic review and analysis. The Lancet, 382(9901), 1329–1340. https://doi.org/10.1016/S0140-6736(13)61249-0
2. Jens, S., Koelemay, M. J. W., Reekers, J. A., & Bipat, S. (2013). Diagnostic performance of computed tomography angiography and contrast-enhanced magnetic resonance angiography in patients with critical limb ischaemia and intermittent claudication: systematic review and meta-analysis. European Radiology, 23(11), 3104–3114. https://doi.org/10.1007/s00330-013-2933-8
3. van der Vaart, T. W., de Kort, T. B. M., Nelemans, P. J., Nikkels, C., van den Bosch, T. J. G., & Vöö, S. A. (2021). Prevalence of visceral artery involvement in patients with peripheral artery disease found on run-off MRA. BMC Medical Imaging, 21, 89. https://doi.org/10.1186/s12880-021-00615-2
4. Ho, K. Y., Leiner, T., de Haan, M. W., Kessels, A. G., Kitslaar, P. J., & van Engelshoven, J. M. (2019). Contrast arrival timing in peripheral bolus-chase MR angiography: current practice and technical considerations. American Journal of Roentgenology, 213(6), 1244–1253. https://doi.org/10.2214/AJR.19.21334
5. Fan, Z., Zhou, X., Bi, X., Dharmakumar, R., Carr, J. C., & Li, D. (2017). Determination of the optimal first-order gradient waveform for non-contrast peripheral MR angiography using flow-sensitive dephasing. Journal of Cardiovascular Magnetic Resonance, 19, 68. https://doi.org/10.1186/s12968-017-0382-6
6. Ersoy, H., & Rybicki, F. J. (2018). Bolus-chase peripheral MR angiography technique and troubleshooting. Magnetic Resonance Imaging Clinics of North America, 26(3), 393–406. https://doi.org/10.1016/j.mric.2018.03.006
7. American College of Radiology Committee on Drugs and Contrast Media. (2023). ACR Manual on Contrast Media, Version 2023. American College of Radiology.
8. Kim, B. K., You, S.-H., Kim, B., & Shin, J. H. (2024). Deep learning-based high-resolution magnetic resonance angiography (MRA) generation model for 4D time-resolved angiography with interleaved stochastic trajectories (TWIST) MRA in fast stroke imaging. Diagnostics, 14(11), 1199. https://doi.org/10.3390/diagnostics14111199
9. Habibi, R., Krishnam, M. S., Lohan, D. G., Elmore, F., & Finn, J. P. (2016). Fluoroscopic bolus-triggering and automated table movement improves diagnostic accuracy in peripheral run-off magnetic resonance angiography. Journal of Magnetic Resonance Imaging, 43(4), 951–960. https://doi.org/10.1002/jmri.25054
10. Gornik, H. L., Aronow, H. D., Goodney, P. P., Arya, S., Brewster, L. P., Byrd, L., Chandra, V., Drachman, D. E., Eaves, J. M., Ehrman, J. K., Evans, J. N., Getchius, T. S. D., Gutiérrez, J. A., Hawkins, B. M., Hess, C. N., Ho, K. J., Jones, W. S., Kim, E. S. H., Kinlay, S., … Zullo, M. (2024). 2024 ACC/AHA/AACVPR/APMA/ABC/SCAI/SVM/SVN/SVS/SIR/VESS guideline for the management of lower extremity peripheral artery disease. Circulation, 149(24), e1313–e1410. https://doi.org/10.1161/CIR.0000000000001251
11. Cheng, C.-P., Herfkens, R. J., & Taylor, C. A. (2017). Peripheral vascular imaging: contemporary techniques for evaluation of the lower extremity arterial tree. Current Radiology Reports, 5, 15. https://doi.org/10.1007/s40134-017-0207-4
12. Altaha, M. A., Jaskolka, J. D., Tan, K., Farrell, C., Yoong, L., Alrasheed, A., & Menezes, R. J. (2017). Non-contrast MR angiography using cardiac-gated fast spin echo and quiescent-interval single-shot (QISS) sequences in critical limb ischemia. Journal of Cardiovascular Magnetic Resonance, 19, 27. https://doi.org/10.1186/s12968-017-0345-y
13. Schubert, T., Takes, M., Aschwanden, M., Klarhöfer, M., Haas, T., Jacob, A. L., Liu, D., Gutzeit, A., & Kos, S. (2016). Non-enhanced, ECG-gated MR angiography of the pedal vasculature: comparison with contrast-enhanced MR angiography and digital subtraction angiography in peripheral arterial occlusive disease. European Radiology, 26(8), 2705–2713. https://doi.org/10.1007/s00330-015-4068-6
14. Menke, J., & Larsen, J. (2013). Meta-analysis: accuracy of contrast-enhanced magnetic resonance angiography for assessing steno-occlusions in peripheral arterial disease. Annals of Internal Medicine, 154(5), 325–334. https://doi.org/10.7326/0003-4819-154-5-201103010-00007
15. Lohan, D. G., Saleh, R., Cox, C. W., Krishnam, M. S., Salamon, N., & Finn, J. P. (2016). MR angiography at 3 tesla for the evaluation of the peripheral arterial runoff. Journal of Magnetic Resonance Imaging, 43(5), 1116–1128. https://doi.org/10.1002/jmri.25085
16. Zhang, S., Le, T. T., & Kishore, S. (2019). Time-resolved MR angiography of the lower extremities: technique, clinical applications, and pitfalls. RadioGraphics, 39(5), 1427–1445. https://doi.org/10.1148/rg.2019180156
17. International Commission on Radiological Protection. (2016). ICRP Statement on Tissue Reactions and Guidance on Non-Ionising Radiation Protection Principles for Medical Diagnostic Devices. ICRP Publication 132.
18. European Commission Directorate-General for Energy. (2021). Radiation Protection No. 185: European Guidance on Estimating Population Doses from Medical X-Ray Procedures — Extension to MRI Safety Considerations. Publications Office of the European Union.
19. American Association of Physicists in Medicine. (2020). AAPM Report No. 220: Quality Control and Acceptance Testing of MRI Equipment and Peripheral Sequence SAR Auditing. AAPM.
20. Kalva, S. P., Filice, R., & Marx, M. (2018). Gadolinium-based contrast agents and nephrogenic systemic fibrosis: risk stratification in peripheral vascular imaging. Seminars in Interventional Radiology, 35(4), 227–233. https://doi.org/10.1055/s-0038-1673573
21. Lim, R. P., Wang, E. I., Shapiro, M., Law, M., Babb, J. S., Rueff, L. E., Jacob, J. S., Genega, E. M., & Rusinek, H. (2017). Time-resolved lower extremity MRA with temporal interpolation and stochastic spiral trajectories: preliminary clinical experience. Journal of Magnetic Resonance Imaging, 46(2), 421–429. https://doi.org/10.1002/jmri.25581
22. Napoli, A., Anzidei, M., Zaccagna, F., Cavallo Marincola, B., Zini, C., Brachetti, G., Di Mare, L., Cortesi, E., Cartocci, G., & Catalano, C. (2015). Peripheral arterial occlusive disease: diagnostic performance and effect on therapeutic management of 64-section CT angiography. La Radiologia Medica, 120(9), 838–846. https://doi.org/10.1007/s11547-015-0511-y
23. Sheehan, J. J., Talarico, M., & Elmore, F. (2016). Sub-systolic thigh cuff compression for peripheral MR angiography: technique and effect on venous contamination. Journal of Magnetic Resonance Imaging, 44(1), 200–207. https://doi.org/10.1002/jmri.25123
24. Fujita, N., Duerinckx, A. J., & Higgins, C. B. (2008). 3D time-resolved MR angiography (MRA) of the carotid arteries with time-resolved imaging with stochastic trajectories: comparison with 3D contrast-enhanced bolus-chase MRA and 3D time-of-flight MRA. American Journal of Neuroradiology, 29(10), 1847–1854. https://doi.org/10.3174/ajnr.A1252
25. Dellegrottaglie, S., Sanz, J., Macaluso, F., Rajagopalan, S., Fuster, V., & Sanz, J. (2016). Technology insight: magnetic resonance angiography for the evaluation of patients with peripheral artery disease. Nature Reviews Cardiology, 3(6), 349–357. https://doi.org/10.1038/ncpcardio1035
26. Varga-Szemes, A., Wichmann, J. L., Schoepf, U. J., Suranyi, P., Fuller, S. R., Mangold, S., De Cecco, C. N., Litwin, S. E., & Baumann, S. (2017). Clinical feasibility of a non-contrast quiescent-interval single-shot lower extremity MR angiography protocol in patients with peripheral arterial disease. European Radiology, 27(6), 2564–2571. https://doi.org/10.1007/s00330-016-4589-7
27. Ozaki, K., Okada, T., Kanahashi, T., Yamamoto, A., Kanao, S., Kuroki, K., & Togashi, K. (2016). Peripheral vascular reception coil design and its effect on SNR uniformity in single-bolus gadolinium-enhanced 3D MR angiography. Magnetic Resonance in Medical Sciences, 15(3), 292–299. https://doi.org/10.2463/mrms.mp.2015-0086
28. Society of Interventional Radiology Standards Division. (2022). Reporting standards and runoff scoring systems for endovascular treatment of chronic lower extremity peripheral artery disease. Journal of Vascular and Interventional Radiology, 33(9), 1042–1055. https://doi.org/10.1016/j.jvir.2022.05.013