Pediatric MRA Protocol Considerations

There is no single correct methodology for performing MRA in the pediatric patient population. Instead, a menu of different techniques exists for radiologists to optimize angiographic evaluation in order to answer the clinical question. Broadly, these techniques can be broken into 2 basic groups: noncontrast sequences and contrast-enhanced (CE) techniques ( Table 1 ). The advent of CE-MRA techniques over the past 2 decades has largely replaced traditional noncontrast techniques for imaging most vascular territories. The marked increase in signal provided by gadolinium-based contrast agents (GBCA) allows MRA to be performed rapidly, typically covering a large field of view in a single breath hold. In addition, the high signal obtained using CE-MRA allows for high spatial resolution, which often aids in depiction of subtle vascular wall abnormalities or peripheral vascular branches. Because of the high spatial resolution, a single three-dimensional (3D) data set can be obtained with CE-MRA that allows reconstruction of oblique imaging planes, along with other postprocessing techniques. Although the advantages of CE-MRA were recognized early in adult populations, this technique has now also become standard in the pediatric population. Nonetheless, there has also been a rejuvenated interest in modern noncontrast MRA techniques over the past several years.

Central to this renewed interest in noncontrast imaging is the recognition of an association between GBCA and nephrogenic systemic fibrosis (NSF) in patients with renal dysfunction. NSF is a rare condition characterized by progressive fibrosis and swelling in the extremities of affected individuals, primarily occurring in the skin and integument. In some individuals, the fibrotic process can be progressive, affecting visceral organs such as the lungs, esophagus, and heart, resulting in substantial morbidity or even mortality. CE-MRA is avoided in patients who have renal impairment, such as end-stage renal disease requiring dialysis, severe chronic kidney disease (estimated glomerular filtration rate [eGFR] <30 mL/min/1.73 m ² ), or acute kidney injury. More recently, patients who show borderline renal function (eGFR 30–40 mL/min/1.73 m ² ) are increasingly urged to avoid GBCA exposure if possible. Although patients with chronic renal disease are encountered more frequently in the adult population, many pediatric disease processes can present with chronic diminished renal function or more acute kidney dysfunction.

Beyond NSF, there are several other reasons that noncontrast MRA is useful in the pediatric population. No gadolinium contrast agent is approved by the US Food and Drug Administration (FDA) for first-pass MRA in children, so the administration of contrast agents is considered off-label use. In addition, dosing for GBCA is typically achieved using a weight-based dosing scheme, typically 0.1 mmol/kg for standard doses and 0.2 mmol/kg for double doses. Given the physical size of infants and children, this dosing can result in a small amount of injected contrast material, which makes accurate bolus tracking/timing challenging when CE-MRA is used. Reliable venous access in a tiny pediatric patient can be challenging because of the small blood vessel size. The ability to hold the breath for 20 to 30 seconds is commonplace in adults and adolescents, allowing motion-free CE-MRA images to be obtained. However, in children and infants, breath holding is not possible without the use of general anesthesia and mechanical ventilation, thereby adding substantial periprocedural complexity and the risk of sedation/anesthesia. On the other hand, noncontrast MRA techniques can be accomplished with respiratory compensation (ie, navigator pulses or mechanical respiratory tracking with a bellow) and may be more easily tolerated by pediatric patients than full anesthesia.

Noncontrast MRA

CE-MRA

The use of GBCA has substantially expanded the use of MRA in clinical applications. Gadolinium chelates alter the local magnetic microenvironment around them, resulting in shortened T1 relaxation times, which translate into faster longitudinal recovery. By using T1-weighted imaging sequences, high vascular signal intensity is achieved with CE imaging. Modern MR scanners use ultrafast 3D T1-weighted gradient echo sequences, often with fat suppression, to maximize the visual impact of contrast enhancement and permit reformatting into other planes or the use of 3D reconstruction techniques.

Timing the scan acquisition to coincide with the appropriate vascular phase of contrast enhancement is key in CE-MRA. In general, venous phase imaging can be obtained at a fixed delay after contrast injection, because venous opacification is related to the distribution of contrast material throughout the circulation. However, for arterial phase imaging, precise timing is required to synchronize image acquisition with the arrival of contrast material. This timing can be accomplished several ways in adults, with automatic bolus tracking techniques/software generally being preferred. This goal can be accomplished in older children who can suspend respiration and have similar cardiovascular physiology to adults. However, in younger children who are unable to cooperate with breath holding and technologist commands, a dynamic postcontrast technique can be used, in which multiple T1-weighted data sets are obtained in rapid succession. This strategy increases the odds that at least 1 of the data sets has a pure arterial phase. The increased speed of acquisition/temporal resolution may necessitate either using parallel imaging techniques or a decrease in spatial resolution.

The use of multiple dynamic phase imaging sequences in the pediatric population has been termed time-resolved MRA in several studies and offers an excellent solution to arterial timing issues in children. Terminology overlaps with time-resolved MRA or keyhole MRA in adults, in whom the term typically refers to techniques that highly undersample k-space to increase scan temporal resolution, and subsequently reconstructing the full k-space data using full k-space acquisitions at the beginning or end of the sequence. The result of these techniques is the ability to rapidly sample the changing tissue contrast of a slice/volume during the administration of contrast and maintain spatial resolution by back-filling the spatial k-space data from other time points, because the vascular system does not move substantially during scan acquisition. Although this latter adult technique may be of benefit in the pediatric population, it has yet to gain major acceptance for pediatric body MRA applications.

Normal and Variant Anatomy

The abdominal and pelvic arterial vasculature comprises primarily the abdominal aorta and its visceral branches and the left and right common iliac arteries and their branches. The anterior branches of the abdominal aorta include the celiac, superior mesenteric, and inferior mesenteric arteries, which comprise the principal blood supply of the viscera. There is a large amount of anatomic variation amongst these 3 vessels. The gonadal and phrenic arteries also arise anteriorly.

Laterally, the renal arteries and middle adrenal arteries come off the aorta on either side. The abdominal aorta is clinically divided into suprarenal and infrarenal segments. The renal arteries are typically seen coming off the aorta at the approximate level of L2. However, variant anatomy of renal arteries is present in up to 30% of the population, with many having multiple renal artery on one or both sides.

Posteriorly, the lumbar arteries come off on either side of the aorta for each lumbar vertebral level. The middle sacral artery also arises posteriorly at the aortoiliac bifurcation. The aorta typically bifurcates at the L4 or L5 level into the right and left common iliac arteries, which in turn bifurcate into internal and external iliac arteries. The internal iliac artery further subdivides into an anterior and posterior division, whereas the external iliac artery continues on to become the common femoral artery and provides the primary blood supply for the lower extremity. The common femoral artery serves as the most common site of arterial access for invasive procedures, which can lead to arterial injury such as dissection, pseudoaneurysm formation, or occlusion ( Fig. 1 ).

Six-year-old girl with limb length discrepancy and history of multiple catheterizations. Postcontrast 3D fat-saturated T1W spoiled gradient echo maximum intensity projection image shows occlusion of the right common femoral artery ( arrow ), reconstituted distally by collaterals.

A persistent sciatic artery is a congenital variation occurring in a small percentage of the population (0.025%–0.04%), in which an embryonic lower limb axial artery (sciatic artery) fails to regress and persists as an elongation of the internal iliac artery ( Fig. 2 ). The sciatic artery normally provides the major blood flow to the lower extremity during development but regresses around the sixth week of gestation into portions of the gluteal and popliteal arteries as well as the tibioperoneal trunk. At the same time that this regression is occurring, the external iliac artery and superficial femoral artery develop from a branch of the fifth lumbar artery. Persistent sciatic artery is classified as complete or incomplete depending on the level of development of the external iliac and superficial femoral arteries. There is a slight male predominance, and although it can be seen bilaterally, it is most commonly noted on the right. Persistent sciatic arteries are prone to atheromatous degeneration, aneurysm formation with or without thrombosis or distal embolization, and local compression.

Fifteen-year-old girl with left leg enlargement and pain. Postcontrast 3D T1W spoiled gradient echo coronal maximum intensity projection ( A ) and postcontrast 3D fat-saturated T1W spoiled gradient echo axial ( B ) images show a large persistent left sciatic artery ( arrow ) with a comparatively diminutive left superficial femoral artery.