Skip to main content
NCBI home page
Radiology: Cardiothoracic Imaging logoLink to Radiology: Cardiothoracic Imaging
. 2020 Nov 19;2(6):e200339. doi: 10.1148/ryct.2020200339

Ferumoxytol-enhanced MR Venography of the Central Veins of the Thorax for the Evaluation of Stenosis and Occlusion in Patients with Renal Impairment

Christopher J R Gallo 1, Joseph G Mammarappallil 1, David Y Johnson 1, Hamid Chalian 1, James Ronald 1, Mustafa R Bashir 1, Charles Y Kim 1,✉
PMCID: PMC7977948  PMID: 33778639

Abstract

Purpose

To assess the diagnostic performance of ferumoxytol-enhanced MR venography for the detection of thoracic central vein stenosis or occlusion with conventional venography as the reference standard.

Materials and Methods

In this retrospective study, consecutive patients from May 2012 to December 2018 underwent dedicated ferumoxytol-enhanced MR venography of the thoracic central veins and conventional venography within 6 months for detecting central venous stenosis. The central veins were divided into seven segments for evaluation. MR venography images were evaluated by three radiologists for presence of stenosis or occlusion. Interobserver agreement was assessed using Fleiss κ.

Results

A total of 35 patients were included (mean age, 49 years; age range, 12–75 years; 18 females). Of the 122 total venous segments with corresponding conventional venography, 73 were stenotic or occluded. The sensitivity and specificity for detection of stenosis or occlusion was 99% and 98%, respectively. The sensitivity and specificity for detecting occlusion alone was 96% and 98%, respectively. MR venography readers demonstrated moderate agreement in their ability to grade stenosis or occlusion (κ = 0.59). There were no adverse events related to contrast agent administration.

Conclusion

Ferumoxytol-enhanced MR venography demonstrated excellent sensitivity and specificity for detection of thoracic central vein stenosis or occlusion.

© RSNA, 2020

See also the commentary by Finn in this issue.

Summary

Ferumoxytol-enhanced MR venography demonstrated a very high sensitivity and specificity for the detection of thoracic central vein stenoses and occlusions.

Key Points

  • ■ Ferumoxytol-enhanced MR venography demonstrated excellent sensitivity and specificity (99% and 98%, respectively) for the detection of thoracic central vein stenosis or occlusion.

  • ■ Interobserver agreement between three MR venography readers in their ability to grade stenosis or occlusion was moderate (Fleiss κ = 0.59).

Introduction

The National Kidney Foundation Kidney Disease Outcomes Quality Initiative recommends central venous imaging prior to creation of permanent surgical access in patients with end-stage renal disease (ESRD) with prior central venous catheter placement or suspected of having stenosis (1). For patients in need of endovascular recanalization, preprocedural imaging can also be critically important for intervention planning. While conventional catheter venography is the historic reference standard to evaluate central veins in the setting of superior vena cava syndrome or vascular access planning for patients with ESRD, gadolinium-enhanced MR venography has shown to be highly sensitive and specific for the detection of stenosis or occlusion (2–6). However, an association between gadolinium-based contrast agents and nephrogenic systemic fibrosis has been reported in patients with renal dysfunction (7–9). Therefore, a clinical need exists for high-quality, noninvasive imaging of the central veins of the thorax in patients with impaired renal function.

Ferumoxytol, an ultrasmall superparamagnetic iron oxide particle formulation, is a Food and Drug Administration (FDA)–approved intravenous therapy for iron deficiency anemia in patients with chronic kidney disease (10). This safe and well-tolerated medication has been shown to be an effective intravascular contrast agent for a variety of vascular distributions (11,12). Notably, ferumoxytol demonstrates blood pool pharmacokinetics resulting in prolonged intravascular retention in the range of 14–15 hours. This is particularly favorable for venous imaging, where peak enhancement and washout times can be unpredictable in the setting of stenosis or occlusion (13). Ferumoxytol use in MR angiography of patients with arteriovenous fistulas and a variety of venous structures has been reported to be feasible, safe, and effective (14–16). The purpose of this study was to retrospectively assess the diagnostic performance of ferumoxytol-enhanced MR venography for detection of thoracic central vein stenosis or occlusion, with conventional venography as the reference standard.

Materials and Methods

Patient Population

Institutional review board approval and a waiver of informed consent was obtained for this Health Insurance Portability and Accountability Act–compliant study. A retrospective review of our imaging database was performed to identify all patients who underwent MR venography of the central veins of the thorax from May 2012 to December 2018 for the purpose of detecting central venous stenoses and occlusions. Of these, 72 consecutive studies were performed with ferumoxytol as the contrast agent. These patients were reviewed to identify those in whom conventional venography was performed within 6 months without development of new symptoms in the interval. Patients were excluded if conventional venography was performed outside of 6 months of MR venography or if there was interval development of symptoms attributable to central venous stenosis between conventional venography and MR venography. In total, 35 consecutive patients who underwent ferumoxytol-enhanced MR venography of the central veins and conventional venography (Table 1) were included for analysis. On the basis of institutional protocol, ferumoxytol was used as an imaging agent only for patients with impaired renal function to avoid development of nephrogenic systemic fibrosis.

Table 1:

Patient Demographics and Characteristics

graphic file with name ryct.2020200339.tbl1.jpg

Open in a new tab

MR Venography Technique

For patients with an existing central venous catheter and poor peripheral venous access, as determined by a radiology nurse’s assessment, the central venous catheter was used for contrast agent injection (n = 4). Otherwise, a peripheral intravenous catheter was inserted for contrast agent injection. Imaging was performed with a 1.5-T MRI system (n = 22, Magnetom Avanto or Aera; Siemens Healthcare, Erlangen, Germany) or a 3-T MRI system (n = 13, Magnetom Skyra; Siemens Healthcare). Two dedicated six-channel receive-only body matrix coils and six to nine elements of the posterior integrated channel spine array coil (Siemens Healthcare) were used for signal reception. Patients were positioned supine with arms at their sides. Wraparound artifacts were avoided by using oversampling and a large field of view (28–50 × 34–50 cm). Ferumoxytol (Feraheme; AMAG Pharmaceuticals, Cambridge, Mass) was dosed at 3 mg per kilogram of body weight diluted in normal saline to a dilution of 1:5. After a three-plane localizer image was obtained, ferumoxytol was injected at a rate of 2 mL per second followed by a 20-mL saline flush injected at the same rate. During the initial infusion, time-resolved angiographic imaging of the thorax was performed for approximately 3 minutes (17–22) (Table 2). Three-dimensional (3D) data sets were processed into one coronal maximum intensity projection image for each time point and stored as a cine loop with a maximum of 20 coronal images. Subsequently, high-spatial-resolution imaging was performed. Postcontrast 3D T1-weighted coronal gradient-recalled-echo (volume-interpolated breath-hold examination) imaging was performed after the time-resolved sequence to allow for a steady-state vascular distribution of ferumoxytol to be reached (Table 2). Electrocardiogram gating was used in 16 patients. All patients underwent continuous monitoring of heart rate, blood pressure, and pulse oximetry during the examination and for at least 30 minutes after the ferumoxytol injection.

Table 2:

MRI Sequence Parameters for Ferumoxytol-enhanced Images

graphic file with name ryct.2020200339.tbl2.jpg

Open in a new tab

Conventional Venography

Conventional digital subtraction venography was performed for diagnostic purposes or as part of a planned venous intervention, typically for endovascular management of a stenosis or occlusion delineated at recent MR venography for restoration of flow due to extremity swelling or for central venous catheter insertion for hemodialysis or subsequent surgical access creation (23). The choice of vein(s) for initial catheter insertion was at the discretion of the operator on the basis of prior MR venography findings, symptoms, or planned surgical site, but included the internal jugular vein, external jugular vein, common femoral vein, and arteriovenous access. Conventional venography was performed using conventional techniques, including digital subtraction angiography. Iodinated contrast material (iopamidol 300; Bracco Diagnostics, Princeton, NJ) was used as the contrast agent by using hand injection. As a matter of routine, venography was performed only of the venous segments pertinent to the intervention or suspected pathologic condition, and thus only several central venous segments were typically imaged for each conventional venography examination.

Image Analysis

The central veins were divided into seven segments, including the bilateral subclavian veins, internal jugular veins, brachiocephalic veins, and superior vena cava. The peripheral boundary of the subclavian vein was defined as the site of overlap with the first rib. For assessment of stenosis or occlusion, the highest degree of stenosis or occlusion was determined within each segment and graded as normal (< 50% narrowing), stenotic (50%–95% narrowing), or occluded (> 95% narrowing) based on visual assessment of reduction in luminal diameter, aided by electronic caliper measurement when needed.

MR venography studies were retrospectively analyzed by three attending radiologists (D.Y.J., J.G.M., and H.C.) with cardiothoracic imaging training and 2–4 years of experience each with dedicated noninvasive cardiovascular imaging. Each reader independently interpreted all studies using contrast-enhanced coronal 3D T1-weighted, axial 3D T1-weighted, and coronal time-resolved images for the degree of stenosis. Readers were blinded to all clinical and demographic information. A consensus interpretation was generated for each venous segment. Each reader rated their level of confidence in interpretation of each venous segment on a scale of 1–4: 1, not confident; 2, somewhat confident; 3, moderately confident; or 4, very confident. The presence of clinically significant collateral veins was assessed separately for the left and right sides of the chest and neck.

One attending interventional radiologist (C.Y.K.), with 10 years of experience with conventional venography and MR venography, interpreted the conventional venograms along with the available MR venography images as the reference standard. The reader was blinded to the other readers’ scores. All venous segments imaged by conventional venography were graded for severity of stenosis using the same system as for MR venography examinations. Only venous segments that were imaged at conventional venography were used for diagnostic accuracy assessments. Venous segments containing stents were excluded.

Quantitative Image Analysis

Quantitative image quality analysis consisted of signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), and intraluminal signal heterogeneity calculations for all available venous segments. Using coronal postcontrast 3D T1-weighted images, mean intraluminal signal intensity (SIv) and standard deviation (SDv) were measured by creating three user-defined, nonoverlapping, ovoid regions of interest within each venous segment. Each region of interest encompassed the maximum amount of vessel lumen while excluding vessel wall. Background image noise was determined by measuring the mean background air signal standard deviation (SDbg) of six 500-pixel regions of interest created in artifact-free locations on the same image as ones used for central vein measurements. The SNR was calculated for all nonoccluded segments in each patient as SIv/SDbg. Intraluminal signal heterogeneity was calculated as SDv/SIv (17). Mean signal intensity of muscle (SIm) was measured by creating three user-defined, nonoverlapping, 100-mm2 ovoid regions of interest within the left pectoralis major muscle. The CNR was calculated as (SIv− SIm)/SDbg.

Statistical Analysis

For each reader and for the consensus assessments for each segment, the sensitivity, specificity, positive predictive value, negative predictive value, and diagnostic accuracy were calculated for presence of clinically significant stenosis (< 50% stenosis vs 50%–95% or > 95% stenosis) and for presence of venous occlusion (< 50% stenosis or 50%–95% vs > 95% stenosis). The 95% confidence intervals (CIs) were calculated using generalized estimating equations to account for repeated measures within patients. Interobserver agreement between MR venography readers in their ability to grade thoracic central vein stenosis or occlusion was performed using Fleiss κ, with values classified as follows: less than 0.00, poor agreement; 0.00–0.20, slight agreement; 0.21–0.40, fair agreement; 0.41–0.60, moderate agreement; 0.61–0.80, substantial agreement; and 0.81–1.00, almost perfect agreement (24). Similarly, interobserver agreement between MR venography readers in their ability to grade chest and neck collateral vasculature severity was performed using Fleiss κ. Statistical within-patient differences in SNR, CNR, and intraluminal heterogeneity across venous segments were evaluated using the one-way analysis of variance test with repeated measures with a Greenhouse-Geisser correction, where a P value less than .05 was considered statistically significant. Statistical analysis was performed using SPSS software version 22 (IBM, Chicago, Ill).

Results

In the 35 patients (mean age, 49 years; age range, 12–75 years; 18 female) who underwent both ferumoxytol-enhanced MR venography and conventional venography of the thoracic central veins, 245 venous segments were imaged on MR venography. A total of 126 of these venous segments had corresponding conventional venography as a reference standard, which was performed a mean of 27 days (range, 0–146) apart (Fig 1). Four segments containing stents were excluded for diagnostic accuracy analysis, resulting in diagnostic accuracy assessment of 122 total venous segments. The division of evaluable venous segments was as follows: superior vena cava (n = 29), right brachiocephalic vein (n = 23), right subclavian vein (n = 17), right internal jugular vein (n = 14), left brachiocephalic vein (n = 20), left subclavian vein (n = 11), and left internal jugular vein (n = 8). A total of 73 (60%) of these 122 segments were stenotic or occluded. There were no adverse events related to ferumoxytol contrast agent administration.

Figure 1:

Flow diagram shows patient inclusion and exclusion. MRV = MR venography.

Open in a new tab

Flow diagram shows patient inclusion and exclusion. MRV = MR venography.

Diagnostic Accuracy and Interobserver Agreement

On the basis of consensus interpretations of the three MR venography readers, the sensitivity and specificity for detection of stenosis or occlusion were 99% (95% CI: 91%, 100%) and 98% (95% CI: 87%, 100%), respectively (Figs 2, 3). The separate reader sensitivities ranged from 84% to 99%, and the specificities ranged from 84% to 94% (Table 3). For each reader, between 76% and 90% of incorrectly graded segments were adjacent to a stenotic or occluded segment. The consensus sensitivity and specificity for detecting occlusions were 96% (95% CI: 91%, 99%) and 98% (95% CI: 89%, 100%), respectively. The separate reader sensitivities ranged from 81% to 98% and specificities ranged from 94% to 95%. Interobserver agreement between MR venography readers in their ability to grade stenosis or occlusion was moderate, with a Fleiss κ value of 0.59 (95% CI: 0.51, 0.67). Mean reader confidence (on a scale of 1–4) in the grading of stenosis or occlusion using ferumoxytol-enhanced MR venography was 3.6 for reader 1, 3.5 for reader 2, and 3.7 for reader 3. Interobserver agreement between MR venography readers in the grading of enlarged chest and neck collateral vein severity demonstrated fair agreement with an interclass correlation coefficient value of 0.30.

Figure 2a:

Images in a 32-year-old man with end-stage renal disease and severe left arm swelling in setting of left arm arteriovenous fistula. (a) Suspicion of central vein occlusion prompted acquisition of ferumoxytol-enhanced MR venography image at 3 T that clearly depicts central veins and a complete left brachiocephalic vein occlusion (arrowhead). (b) The occlusion is confirmed on a subsequent conventional venogram obtained just prior to intervention (arrowhead).

Open in a new tab

Images in a 32-year-old man with end-stage renal disease and severe left arm swelling in setting of left arm arteriovenous fistula. (a) Suspicion of central vein occlusion prompted acquisition of ferumoxytol-enhanced MR venography image at 3 T that clearly depicts central veins and a complete left brachiocephalic vein occlusion (arrowhead). (b) The occlusion is confirmed on a subsequent conventional venogram obtained just prior to intervention (arrowhead).

Figure 3a:

Images in a 61-year-old woman with end-stage renal disease and a dysfunctional right arm arteriovenous graft in setting of suspected central occlusion who underwent ferumoxytol-enhanced MR venography for diagnosis and procedural planning. (a) MR venography at 1.5 T demonstrates a chronic occlusion at the origin of the right subclavian vein (arrowhead), as well as chronic occlusion of the right brachiocephalic vein. (b) Venography performed at time of endovascular recanalization confirmed chronic occlusion (arrowhead).

Open in a new tab

Images in a 61-year-old woman with end-stage renal disease and a dysfunctional right arm arteriovenous graft in setting of suspected central occlusion who underwent ferumoxytol-enhanced MR venography for diagnosis and procedural planning. (a) MR venography at 1.5 T demonstrates a chronic occlusion at the origin of the right subclavian vein (arrowhead), as well as chronic occlusion of the right brachiocephalic vein. (b) Venography performed at time of endovascular recanalization confirmed chronic occlusion (arrowhead).

Table 3:

Diagnostic Performance of Ferumoxytol-enhanced MR Venography

graphic file with name ryct.2020200339.tbl3.jpg

Open in a new tab

Figure 2b:

Images in a 32-year-old man with end-stage renal disease and severe left arm swelling in setting of left arm arteriovenous fistula. (a) Suspicion of central vein occlusion prompted acquisition of ferumoxytol-enhanced MR venography image at 3 T that clearly depicts central veins and a complete left brachiocephalic vein occlusion (arrowhead). (b) The occlusion is confirmed on a subsequent conventional venogram obtained just prior to intervention (arrowhead).

Open in a new tab

Images in a 32-year-old man with end-stage renal disease and severe left arm swelling in setting of left arm arteriovenous fistula. (a) Suspicion of central vein occlusion prompted acquisition of ferumoxytol-enhanced MR venography image at 3 T that clearly depicts central veins and a complete left brachiocephalic vein occlusion (arrowhead). (b) The occlusion is confirmed on a subsequent conventional venogram obtained just prior to intervention (arrowhead).

Figure 3b:

Images in a 61-year-old woman with end-stage renal disease and a dysfunctional right arm arteriovenous graft in setting of suspected central occlusion who underwent ferumoxytol-enhanced MR venography for diagnosis and procedural planning. (a) MR venography at 1.5 T demonstrates a chronic occlusion at the origin of the right subclavian vein (arrowhead), as well as chronic occlusion of the right brachiocephalic vein. (b) Venography performed at time of endovascular recanalization confirmed chronic occlusion (arrowhead).

Open in a new tab

Images in a 61-year-old woman with end-stage renal disease and a dysfunctional right arm arteriovenous graft in setting of suspected central occlusion who underwent ferumoxytol-enhanced MR venography for diagnosis and procedural planning. (a) MR venography at 1.5 T demonstrates a chronic occlusion at the origin of the right subclavian vein (arrowhead), as well as chronic occlusion of the right brachiocephalic vein. (b) Venography performed at time of endovascular recanalization confirmed chronic occlusion (arrowhead).

Quantitative Image Analysis

Intraluminal SNR, CNR, and heterogeneity values were calculated in 197 nonoccluded venous segments imaged with ferumoxytol-enhanced MR venography (Table 4). The mean calculated SNR, CNR, and heterogeneity for all venous segments combined was 217.9 ± 134.1 (standard deviation), 168.6 ± 105.1, and 0.07 ± 0.05, respectively. No differences in SNR, CNR, or heterogeneity across venous segments were identified (P = .08, .08, and .57, respectively).

Table 4:

Image Quality Analysis of Ferumoxytol-enhanced MR Venography for Thoracic Central Vein Evaluation

graphic file with name ryct.2020200339.tbl4.jpg

Open in a new tab

Discussion

With National Kidney Foundation guidelines recommending central venous imaging prior to permanent arteriovenous access creation in patients with renal failure with prior catheters, and because patients with ESRD are at increased risk of central venous pathologic conditions, exploration and validation of noninvasive imaging modalities is of great interest (1). Given the reported association between conventional gadolinium agents and nephrogenic systemic fibrosis in patients with renal failure, alternative contrast agents may be considered for this population (25). Ferumoxytol-enhanced MR venography provided excellent sensitivity and specificity values of 99% and 98%, respectively, for the detection of stenosis or occlusion. Image quality was also excellent, as demonstrated by high reader confidence levels. These data suggest that ferumoxytol may be a particularly well-suited MR contrast agent for the evaluation of central venous pathologic features in patients with ESRD who are at risk for gadolinium-associated nephrogenic systemic fibrosis.

Ferumoxytol has been previously reported to be an effective venous imaging agent. Shahrouki et al reported 100% sensitivity and specificity while using ferumoxytol-enhanced MR venography of the chest, abdomen, and pelvis for detection of stenosis or deep venous thrombosis of the thoracic, abdominal, or pelvic veins in 14 patients, with conventional venography as the reference standard (16). Nine of those patients underwent intervention of the thoracic central veins, and less than half of that population was dialysis dependent. In a separate study, 12 pediatric patients underwent ferumoxytol-enhanced MR venography to evaluate a variety of venous structures throughout the body, including the central veins, hepatic veins, or portal vein. There was 100% concordance between ferumoxytol-enhanced MR venography and follow-up imaging or invasive procedures for the detection of stenosis, hemorrhage, or thrombus (26). The current study similarly shows an excellent sensitivity and specificity for detection of venous pathologic features but in a hemodialysis-specific population with dedicated thoracic central vein imaging and a high prevalence of abnormalities. Contrast-enhanced MR venography for thoracic central venous assessment using gadolinium-based contrast agents has also shown excellent diagnostic accuracy with a sensitivity of 93%–100% and specificity ranging from 87% to 100% (5,6,19,27).

Blood pool agents are characterized by a prolonged plasma half-life that results in a prolonged intravascular retention time. This property allows for a markedly longer interval during which vascular imaging can be performed (13,28). After a period of circulation, the contrast agent concentration equilibrates throughout the vascular system, which reduces mixing artifact. Gadofosveset and ferumoxytol are two such blood pool agents that have shown to be useful in vascular imaging, although gadofosveset is currently no longer commercially available (29). In two separate studies, gadofosveset demonstrated subjectively superior image quality compared with conventional gadolinium-enhanced MR venography for central vein assessment (30,31). For lower extremity arterial imaging, steady-state MR angiography with gadofosveset demonstrated 100% agreement with the reference standard digital subtraction angiography, whereas conventional first-pass MR angiography demonstrated 87% agreement (32). Bashir et al compared imaging quality of ferumoxytol versus gadofosveset for imaging the abdominopelvic veins, finding equivalent image quality (15). While no studies have compared ferumoxytol with conventional gadolinium agents for detection of central venous stenoses and occlusions, the standalone results suggest that imaging quality and diagnostic accuracy are at least equivalent or better with ferumoxytol, while avoiding the risk of nephrogenic systemic fibrosis.

While conventional venography is widely considered the reference standard for central vein evaluation, there are a number of limitations that are overcome with MR venography. First, conventional venography only provides visualization of the venous system, extending from the injection site to the heart. Bilateral central venous imaging necessitates contrast agent injection from both the right and left sides of the body (33). Internal jugular veins are also poorly opacified when upper extremity peripheral intravenous lines are injected, requiring multiple access points for complete central venous evaluation. While the internal jugular veins can be partially visualized with US, the central portion may be difficult to assess. This anatomic information is critical to ascertain for potential catheter access. Second, hemodilution of iodinated contrast agents in veins farthest from the injection site (ie, superior vena cava) can preclude adequate visualization, especially in the presence of pathologic or slow flow (28). While the superior vena cava visualization was rated as very good, the superior vena cava had the lowest mean CNR value, which may be due to poorer signal reception in the central chest being furthest from the receive coils or from cardiac motion.

It is notable that the ferumoxytol injection used in this study was relatively rapid when compared with the FDA-recommended infusion duration of 15 minutes, as the agent is indicated for iron supplementation, not imaging. A group has previously reported the excellent safety profile of this injection rate, where no severe reactions were encountered (34). However, it remains widely recommended that infusion rates follow manufacturer guidelines on an iron quantity per unit time. Also notable is a black box warning presented by the FDA in 2015, stating that ferumoxytol should not be given to patients who have had an allergic reaction to any intravenous iron replacement product.

The major limitation of this study was that comprehensive bilateral conventional venograms were not obtained in all patients, as venography was only performed as part of usual care prior to intervention, and thus there were relatively fewer internal jugular vein segments and subclavian vein segments imaged. Furthermore, the studied population was particularly enriched with patients with stenoses and occlusions, as patients with normal central veins rarely warrant conventional venography or intervention–thus the nonrepresentative prevalence of steno-occlusive disease will skew the positive predictive value, negative predictive value, and diagnostic accuracy values. Second, the system used for grading segments can lead to disparate interpretations when there is a stenosis or occlusion involving a vascular confluence. In these instances, it may not be clear if the stenosis is more appropriately assigned to the upstream or downstream vessel, or both, resulting in assessment of anatomy rather than visualization of pathologic features. Because the majority of incorrectly graded segments were adjacent to a stenotic or occluded vessel, it is possible that this may have been a cause of misinterpretation; however, this observation could be by statistical chance as well, given the population enriched with pathologic features. As the MR venography and reference standard studies were not performed same day, it is feasible that stenoses or occlusions developed in the interval; however, charts were reviewed to ensure no new symptoms to suggest this possibility. Last, the background noise in MR venography images varied slightly based on the location of the central veins within the field of view. In general, noise in MRI is difficult or impossible to accurately assess without repeat measurements over time, which were not available in this retrospective study, so we used a commonly employed surrogate for noise obtained from the background. This variability likely impacted image quality calculations (ie, SNR), limiting their use as absolute values; nonetheless, they were useful for internal comparisons between vascular territories on the same MRI examination.

In summary, ferumoxytol-enhanced MR venography demonstrated excellent sensitivity and specificity for the detection of thoracic central vein stenosis or occlusion in this studied population with a high prevalence of central venous pathologic conditions. Given that ferumoxytol is an FDA-approved parenteral iron supplement for patients undergoing hemodialysis and does not carry a risk of nephrogenic systemic fibrosis, this agent is well-suited for noninvasive vascular imaging in this patient population of particular need for central venous evaluation. While there is currently increasing acceptance of the use of group 2 gadolinium-based contrast agents in patients with renal dysfunction, a prominent advantage of ferumoxytol is that it is the only commercially available blood pool agent, which is a class of contrast agents particularly well suited for optimal, predictable, and homogeneous opacification of venous structures. While imaging features are excellent with this technique, the relatively high cost of ferumoxytol may impact its role for imaging in the populations of patients without renal failure. Additional studies may be useful to determine the diagnostic accuracy of ferumoxytol for screening in a population with low prevalence of central venous pathologic conditions, diagnostic yield, and diagnostic accuracy compared with conventional gadolinium agents.

Disclosures of Conflicts of Interest: C.J.R.G. Activities related to the present article: disclosed money paid to author from the Radiological Society of North America 2019 Annual Meeting Trainee Research Award. Activities not related to the present article: disclosed no relevant relationships. Other relationships: disclosed no relevant relationships. J.G.M. disclosed no relevant relationships. D.Y.J. disclosed no relevant relationships. H.C. disclosed no relevant relationships. J.R. disclosed no relevant relationships. M.R.B. Activities related to the present article: disclosed no relevant relationships. Activities not related to the present article: disclosed money paid to author from MedPace, ICON, and Corcept Therapeutics; disclosed grants/grants pending to author’s institution from Siemens Healthcare, GE Healthcare, Carmot Therapeutics, Madrigal Pharmaceuticals, Metacrine, NGM Bio, Pinnacle Clinical Research, and Prosciento. Other relationships: disclosed no relevant relationships. C.Y.K. disclosed no relevant relationships.

Abbreviations:

CI
confidence interval
CNR
contrast-to-noise ratio
ESRD
end-stage renal disease
FDA
Food and Drug Administration
SNR
signal-to-noise ratio
3D
three-dimensional

References

  • 1.KDOQI ; National Kidney Foundation . KDOQI Clinical Practice Guidelines and Clinical Practice Recommendations for Anemia in Chronic Kidney Disease. Am J Kidney Dis 2006;47(5 Suppl 3):S11–S145. [DOI] [PubMed] [Google Scholar]
  • 2.Nael K, Laub G, Finn JP. Three-dimensional contrast-enhanced MR angiography of the thoraco-abdominal vessels. Magn Reson Imaging Clin N Am 2005;13(2):359–380. [DOI] [PubMed] [Google Scholar]
  • 3.Oxtoby JW, Widjaja E, Gibson KM, Uzoka K. 3D gadolinium-enhanced MRI venography: evaluation of central chest veins and impact on patient management. Clin Radiol 2001;56(11):887–894. [DOI] [PubMed] [Google Scholar]
  • 4.Shinde TS, Lee VS, Rofsky NM, Krinsky GA, Weinreb JC. Three-dimensional gadolinium-enhanced MR venographic evaluation of patency of central veins in the thorax: initial experience. Radiology 1999;213(2):555–560. [DOI] [PubMed] [Google Scholar]
  • 5.Tanju S, Sancak T, Düşünceli E, Yağmurlu B, Erden I, Sanlidilek U. Direct contrast-enhanced 3D MR venography evaluation of upper extremity deep venous system. Diagn Interv Radiol 2006;12(2):74–79. [PubMed] [Google Scholar]
  • 6.Thornton MJ, Ryan R, Varghese JC, Farrell MA, Lucey B, Lee MJ. A three-dimensional gadolinium-enhanced MR venography technique for imaging central veins. AJR Am J Roentgenol 1999;173(4):999–1003. [DOI] [PubMed] [Google Scholar]
  • 7.Grobner T. Gadolinium--a specific trigger for the development of nephrogenic fibrosing dermopathy and nephrogenic systemic fibrosis? Nephrol Dial Transplant 2006;21(4):1104–1108. [DOI] [PubMed] [Google Scholar]
  • 8.High WA, Ayers RA, Chandler J, Zito G, Cowper SE. Gadolinium is detectable within the tissue of patients with nephrogenic systemic fibrosis. J Am Acad Dermatol 2007;56(1):21–26. [DOI] [PubMed] [Google Scholar]
  • 9.Perazella MA. How should nephrologists approach gadolinium-based contrast imaging in patients with kidney disease? Clin J Am Soc Nephrol 2008;3(3):649–651. [DOI] [PubMed] [Google Scholar]
  • 10.Provenzano R, Schiller B, Rao M, Coyne D, Brenner L, Pereira BJ. Ferumoxytol as an intravenous iron replacement therapy in hemodialysis patients. Clin J Am Soc Nephrol 2009;4(2):386–393. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Nguyen KL, Yoshida T, Kathuria-Prakash N, et al. Multicenter Safety and Practice for Off-Label Diagnostic Use of Ferumoxytol in MRI. Radiology 2019;293(3):554–564. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Toth GB, Varallyay CG, Horvath A, et al. Current and potential imaging applications of ferumoxytol for magnetic resonance imaging. Kidney Int 2017;92(1):47–66. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Bremerich J, Bilecen D, Reimer P. MR angiography with blood pool contrast agents. Eur Radiol 2007;17(12):3017–3024. [DOI] [PubMed] [Google Scholar]
  • 14.Sigovan M, Gasper W, Alley HF, Owens CD, Saloner D. USPIO-enhanced MR angiography of arteriovenous fistulas in patients with renal failure. Radiology 2012;265(2):584–590. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Bashir MR, Mody R, Neville A, et al. Retrospective assessment of the utility of an iron-based agent for contrast-enhanced magnetic resonance venography in patients with endstage renal diseases. J Magn Reson Imaging 2014;40(1):113–118. [DOI] [PubMed] [Google Scholar]
  • 16.Shahrouki P, Moriarty JM, Khan SN, et al. High resolution, 3-dimensional Ferumoxytol-enhanced cardiovascular magnetic resonance venography in central venous occlusion. J Cardiovasc Magn Reson 2019;21(1):17. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Brauck K, Maderwald S, Vogt FM, Zenge M, Barkhausen J, Herborn CU. Time-resolved contrast-enhanced magnetic resonance angiography of the hand with parallel imaging and view sharing: initial experience. Eur Radiol 2007;17(1):183–192. [DOI] [PubMed] [Google Scholar]
  • 18.Fink C, Puderbach M, Ley S, Zaporozhan J, Plathow C, Kauczor HU. Time-resolved echo-shared parallel MRA of the lung: observer preference study of image quality in comparison with non-echo-shared sequences. Eur Radiol 2005;15(10):2070–2074. [DOI] [PubMed] [Google Scholar]
  • 19.Kim CY, Mirza RA, Bryant JA, et al. Central veins of the chest: evaluation with time-resolved MR angiography. Radiology 2008;247(2):558–566. [DOI] [PubMed] [Google Scholar]
  • 20.Michaely HJ, Herrmann KA, Dietrich O, Reiser MF, Schoenberg SO. Quantitative and qualitative characterization of vascularization and hemodynamics in head and neck tumors with a 3D magnetic resonance time-resolved echo-shared angiographic technique (TREAT)--initial results. Eur Radiol 2007;17(4):1101–1110. [DOI] [PubMed] [Google Scholar]
  • 21.Michaely HJ, Nael K, Schoenberg SO, et al. Renal perfusion: comparison of saturation-recovery TurboFLASH measurements at 1.5T with saturation-recovery TurboFLASH and time-resolved echo-shared angiographic technique (TREAT) at 3.0T. J Magn Reson Imaging 2006;24(6):1413–1419. [DOI] [PubMed] [Google Scholar]
  • 22.Pinto C, Hickey R, Carroll TJ, et al. Time-resolved MR angiography with generalized autocalibrating partially parallel acquisition and time-resolved echo-sharing angiographic technique for hemodialysis arteriovenous fistulas and grafts. J Vasc Interv Radiol 2006;17(6):1003–1009. [DOI] [PubMed] [Google Scholar]
  • 23.Griffin AS, Gage SM, Lawson JH, Kim CY. Early infection risk with primary versus staged Hemodialysis Reliable Outflow (HeRO) graft implantation. J Vasc Surg 2017;65(1):136–141. [DOI] [PubMed] [Google Scholar]
  • 24.Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977;33(1):159–174. [PubMed] [Google Scholar]
  • 25.Information for Healthcare Professionals . Gadolinium-Based Contrast Agents for Magnetic Resonance Imaging Scans (marketed as Omniscan, OptiMARK, Magnevist, ProHance, and MultiHance). Food and Drug Administration Web site. https://web.archive.org/web/20171115034712/https://www.fda.gov/Drugs/DrugSafety/PostmarketDrugSafetyInformationforPatientsandProviders/ucm142905.htm. Updated December 2006. Accessed August 30, 2019.
  • 26.Luhar A, Khan S, Finn JP, et al. Contrast-enhanced magnetic resonance venography in pediatric patients with chronic kidney disease: initial experience with ferumoxytol. Pediatr Radiol 2016;46(9):1332–1340. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Gao K, Jiang H, Zhai RY, Wang JF, Wei BJ, Huang Q. Three-dimensional gadolinium-enhanced MR venography to evaluate central venous steno-occlusive disease in hemodialysis patients. Clin Radiol 2012;67(6):560–563. [DOI] [PubMed] [Google Scholar]
  • 28.Cline B, Hurwitz LM, Kim CY. MR Venography of the Central Veins of the Thorax. Top Magn Reson Imaging 2017;26(4):167–174. [DOI] [PubMed] [Google Scholar]
  • 29.Thakor AS, Chung J, Patel P, et al. Use of blood pool agents with steady-state MRI to assess the vascular system. J Magn Reson Imaging 2017;45(6):1559–1572. [DOI] [PubMed] [Google Scholar]
  • 30.See TC, Patterson AJ, Hilliard NJ, et al. Gadofosveset-enhanced thoracic MR venography: a comparative study evaluating steady state imaging versus conventional first-pass time-resolved dynamic imaging. Acta Radiol 2018;59(4):418–424. [DOI] [PubMed] [Google Scholar]
  • 31.Åkesson M, Lehti L, Höglund P, Åkeson P, Wassélius J. Thoracic magnetic resonance venography using Gadofosveset in patients with venous pathology-A comparative study of image quality and inter-rater variability. Phlebology 2017;32(7):453–458. [DOI] [PubMed] [Google Scholar]
  • 32.Hadizadeh DR, Gieseke J, Lohmaier SH, et al. Peripheral MR angiography with blood pool contrast agent: prospective intraindividual comparative study of high-spatial-resolution steady-state MR angiography versus standard-resolution first-pass MR angiography and DSA. Radiology 2008;249(2):701–711. [DOI] [PubMed] [Google Scholar]
  • 33.Finn JP, Zisk JH, Edelman RR, et al. Central venous occlusion: MR angiography. Radiology 1993;187(1):245–251. [DOI] [PubMed] [Google Scholar]
  • 34.Vasanawala SS, Nguyen KL, Hope MD, et al. Safety and technique of ferumoxytol administration for MRI. Magn Reson Med 2016;75(5):2107–2111. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Radiology: Cardiothoracic Imaging are provided here courtesy of Radiological Society of North America

RESOURCES