MRI Contrast Agents and Nephrogenic Systemic Fibrosis (NSF)*

*Special thanks to Alberto Spinazzi, M.D., Global Medical and Regulatory Affairs, Bracco Diagnostics, Inc.


Even if the first cases of nephrogenic systemic fibrosis (NSF) were identified in 1997, and the first published report of 14 cases appeared in 2000 (1), NSF has only recently received great attention, especially because of its possible association with exposure to gadolinium-based contrast agents (GBCAs). “Nephrogenic” does not mean that the disease is caused by factors originating in the kidney, but that NSF has been observed only in patients with chronic kidney disease, while “systemic” emphasizes the systemic nature of this fibrosing disorder (2). It was previously known as “Nephrogenic Fibrosing Dermopathy”, since its most prominent and visible effects are observed in the skin, where the histopathologic findings closely parallel those observed in wound healing reactions (1, 3, 4, 5). The nomenclature of the disease has been changed to NSF based on autopsy case reports of individual NSF patients that have reported variable degrees of myocardial, pericardial, and pleural fibrosis, along with the involvement of nerves and skeletal muscles (6, 7, 8). The disease is progressive and can be associated with a fatal outcome. There is still no definitive cure.


NSF cannot be detected using a single diagnostic test. A confident diagnosis can usually be reached through the combination of a good clinical history, a good physical exam and the histopathologic examination of a biopsy specimen of involved skin. The main elements that should guide physicians in the diagnostic process are the clinical presentation in the setting of severe renal insufficiency and, more important, confirmatory cutaneous histopathologic findings (9).

So far, NSF has been observed only in patients with acute or chronic severe renal insufficiency (glomerular filtration rate, GFR, <30 mL/min/1.73m2), or acute renal insufficiency of any severity due to the hepato-renal syndrome or in the perioperative liver transplantation period (9). The majority of patients with NSF have a GFR <15 mL/min and are receiving (or have received) either hemodialysis, peritoneal dialysis, or both (9).

The skin changes can mimic progressive systemic sclerosis with a predilection for the extremity involvement that can extend to the torso. It has not been reported to occur on the face, palms, or soles. Unlike scleroderma, NSF usually spares the face. Skin lesions usually begin with swelling, progressing to erythematous papules and coalescing hyperpigmented, browny plaques with peau d’orange surface. Peripheral irregular fingerlike or ameboid projections may be present along with islands of sparing. Bullae and nodules have also been reported. The skin involvement is often symmetrical and bilateral. Upon eye examination, new onset of white-yellow scleral plaques with dilated capillary loops may be seen in patients less than 45 years of age (1, 4, 9, 10 ).

The involved skin and subcutis usually becomes markedly thickened and hardened, with a wooden consistency to palpation. The induration characteristically involves the distal extremities first, gradually proceeding to involve the proximal extremities to the level of the mid-thigh and mid-upper arms (1, 9, 10 ). The involvement of the skin and subcutis overlying joints often causes a decrease in function of the hands and feet first, and then of more proximal joints in affected extremities, with elbows angled inwards and downwards and the body stooped over. Pedal plantar flexion may be sufficiently severe, making walking difficult or impossible. Joint contractures may develop very rapidly, with patients becoming wheelchair bound days to weeks after NSF starts, so that patients may become wheelchair dependent (9, 10).

If renal function is restored, the skin lesions may stabilize or even regress (9). Occasional patients (estimated at less than 5%) have rapidly progressive, fulminant NSF, associated with an accelerated loss of mobility, and often severe pain (9, 10).

Patients with NSF may complain of itching and sharp pain. Itching and pain may be localized in the affected areas, in the rib cage or hips. Loss of appetite and muscle weakness are commonly described (9, 10).

When the above noted signs and symptoms are observed in patients with severe renal insufficiency, a biopsy should be performed to obtain specimens of involved skin.

Histologically, NSF is characterized by dermal fibrosis (11). There is always an increased number of fibrocytes, that are CD34 and procollagen I positive when stained immunohistochemically (11). This dual positivity is characteristic of so-called “circulating fibrocytes,” mesenchymal stem cells of bone marrow origin that participate in wound repair (11, 12).

In early lesions of NSF, collagen bundles may be quite narrow, with abundant edema fluid and/or mucin separating them. Procollagen I positivity is already present, but noted inconspicuously in the perinuclear cytoplasm of the bland dermal fibrocytes. In more advanced disease, collagen bundles become thicker (still generally maintaining clefts of separation between their neighbors) and the cytoplasm of the fibrocytes becomes plump and intensely procollagen I positive (13). The dermis is always involved by the histopathological pattern noted above, while the epidermis is not typically affected by NSF, although some degree of basilar pigmentation and epidermal acanthosis may be noted in advanced disease (13). The subcutaneous septa are markedly widened. In these deeper NSF foci, the widened septa are collagenized in the same manner as described above (13).

Other occasional findings may be a combination of epithelioid CD68-positive histiocytes in the subcutaneous septa, multinucleated giant cells, osteoclast-like giant cells, foci of osteoid deposition, and/or calcified bone spicules (13). Increased numbers of factor XIIIa positive dendritic cells or a coexpression of factor XIIIa and CD68 in the same cell have been observed as well. Vascularity is not typically prominent, although some cases of NSF show evidence of angiogenesis. Microthrombi and vasculitis have never been observed (13). The fibrotic process may extend through the fascia and into the underlying skeletal muscle that can be swelling and hardened as well (13). Patients suspected of NSF should be fully assessed by an experienced dermatologist.

How to minimize the risk of NSF

Step 1: Identify patients at risk

The first step should be to identify patients at risk for NSF, that is, those patients that suffer from severe (stages 4 and 5) chronic kidney disease, that is, GFR below 30 mL/min/1.73 m2, independently of their age, race or gender or acute kidney injury (14). The risk of NSF development in patients with GFR between 30 and 44 mL/min/1.73 m2 (stage 3b) is much smaller, but the American College of Radiology Subcommittee on MR Safety recommends to manage this group of patients as those with stage 4 or 5 chronic kidney disease, since GFR determinations may fluctuate from one day to the next (14). Therefore, it is important to identify patients with GFR<45 mL/min/1.73 m2.

The level of GFR should be estimated from prediction equations that take into account the serum creatinine concentration and some or all of the following variables: age, gender, race, and body size (14). Many experts (including the American College of Radiology Subcommittee on MR Safety) have recommended that an estimated GFR be obtained within six weeks of anticipated GBCA injection in patients who might have reduced renal function. It has been suggested that this would include any patient with a history of renal disease (including a solitary kidney, renal transplant, or renal neoplasm), anyone over the age of 60 years, and with a history of hypertension or diabetes mellitus (14).

The most widely used equations for adult patients are the Modification of Diet in Renal Disease (MDRD) study equation (15) and the Cockcroft-Gault formula (16). Even if both equations provide a marked improvement over serum creatinine alone (17), the MDRD Study equation may perform better than the Cockroft-Gault formula, but the data are very limited (18-20). Both prediction equations assume that the amount of creatinine produced by the patient is equal to the amount being removed by the kidneys. Therefore, both equations are not suitable if renal function is in an unstable condition—that is, in patients with acute renal failure or on dialysis. Results may also deviate from true values in patients with exceptional dietary intake (e.g., vegetarian diet, high protein diet, creatine supplements), extremes of body composition (e.g., very lean, obese, paraplegia), or severe liver disease. Among children, the Schwartz formula provides a clinically useful estimate of GFR (21).

Step 2: Assess risk-benefit of contrast-enhanced MRI in patients at risk

A patient at risk of NSF should receive a GBCA only when a risk–benefit assessment for that particular patient indicates that the benefit of doing so clearly outweighs the potential risk(s). The risk-benefit evaluation should be made by the radiologist in conjunction with the referring physician(s) and should be properly a prospectively documented. History of previous exposures to GBCAs or other factors that are thought to act as possible co-triggers of the disease, such as metabolic acidosis, vascular surgery, thrombotic events, etc. should be taken into account during the risk–benefit assessment of each individual patient.

Patients, or parents or guardians in case of minors, should be properly informed of the benefits, risks, and diagnostic alternatives, based on all the information available at that time, and provide their consent in writing.

Step 3: Perform any unenhanced MR sequence that may be helpful before injecting the contrast agent

In the United States, the Food and Drug Administration has requested the prescribing information of all GBCAs to be revised by adding a black boxed warning, according to which the use of GBCAs in risk patients should be avoided, unless the diagnostic information is essential and not available with non-contrast enhanced MRI. Therefore, the MR exam should be properly monitored. All unenhanced MR sequences that may be helpful to make a diagnosis should be performed, and the images evaluated by an experienced radiologist, in order to ensure that the administration of a GBCA is still deemed necessary.

Step 4: Do not expose risk patients to high doses of GBCAs

If the use of a GBCA is still deemed necessary, use the lowest dose needed to reliably provide the diagnostic information being clinically sought. According to boxed warning required by the FDA, the recommended doses should never be exceeded and dosing should not be repeated.

Which agent should be used?

In the USA, as in Europe and Japan, some GBCAs (gadodiamide, Omniscan; gadopentetate dimeglumine, Magnevist; gadoversetamide, OptiMARK) are specifically contraindicated for use in patients at risk of NSF (22). The American College of Radiology recommends to never use these agents in patients with GFR <45 mL/min and in patients with known or suspected acute kidney injury, regardless of calculated GFR VALUES(14). Features of acute kidney injury consist of rapid (over hours to days) and usually reversible decrease in kidney function, commonly in the setting of surgery, severe infection, injury, or drug-induced kidney toxicity (22).

The other approved agents (Ablavar, gadofosveset trisodium; Eovist, gadoxetate disodium; MultiHance, gadobenate dimeglumine; ProHance, gadoteridol) should be used in patients at risk of NSF only if the diagnostic information is essential and not available with non-contrast enhanced MRI or other imaging modalities.

What to do after the MRI exam?

The usefulness of hemodialysis in the prevention of NSF is unknown. However, to enhance and speed up the GBCA elimination, many experts recommend that consideration be given to the performance of several dialysis sessions following GBCA administration in patients with end-stage renal disease on chronic dialysis, with use of prolonged dialysis times and increased flow rates and volumes to assist in the process of GBCA clearance (14). Peritoneal dialysis provides much less potential NSF risk reduction compared to hemodialysis and should not be considered protective (14).

Patients at risk of NSF should be followed up at least for one year after a contrast-enhanced MR exam, to promptly identify any symptom or sign suggestive of NSF and confirm or rule out a diagnosis of NSF. Should a new diagnosis of NSF be made, it is recommended that all the regulatory authorities in the United States, Canada, Europe, Asia and other countries be properly notified.

Additional Information to Consider

A study conducted by Nandwana, et al. (23) was performed to determine the incidence of NSF in patients at high risk, that is, patients with end-stage renal disease (ESRD) who had received MultiHance as part of a pre-transplant recipient evaluation screening for abnormalities which could compromise renal transplant outcome. The study population included patients not undergoing dialysis as well as those undergoing hemodialysis or peritoneal dialysis. Enrollment included 401 patients of which 303 (75.5%) were currently undergoing dialysis. For the remaining 98 (24.4%) patients not undergoing dialysis, the mean eGFR was 17 mL/min/1.73 m2. Patients underwent a combined MR imaging and MR angiography of the abdomen and pelvis. Following the institution’s standard protocol, each patient received a total of 0.15 mmol/kg of MultiHance, a dose 50% higher than the approved dose of 0.10 mmol/kg. The mean volume administered during the index MR examination was 24 mL. A total of 66 patients underwent additional examinations with MultiHance. For this group the mean total cumulative dose of MultiHance was 47 mL. Each patient’s electronic medical records including pathology, transplant notes, progress notes, consultation notes (with and without skin examinations) and additional patient communication were evaluated for NSF or NSF-like symptoms. The study found no case even suggestive of NSF after a mean follow up period of more than two years.

In another study, Shaffer, et al. (24), assessed renal safety of MRI with the administration of MultiHance in patients with decompensated cirrhosis awaiting liver transplantation. The study population included 352 patients of whom 70 (20%) had renal insufficiency defined as pre-dose serum creatinine values ? 1.5mg/dL, and 64 (18.2%) who had Chronic Kidney Disease defined as serum creatinine >1.5mg/dL for ? 3 months. For the entire population, the pre-dose serum creatinine values ranged from 0.360 to 4.860 mg/dL. Following the institution’s protocol, patients received half the approved dose of MultiHance (0.05 mmol/kg) for their abdominal MRI examination. There were no incidences of NSF reported among these 352 cases over a follow up period of 25.8 months.


1. Cowper SE, et al. Scleromyxoedema-like cutaneous disease in renal-dialysis patients. Lancet 2000;356:1000–1001.

2. Cowper SE, Boyer PJ. Nephrogenic systemic fibrosis: An update. Curr Rheumatol Rep 2006;8:151–157.

3. DeHoratius D, Cowper SE. Nephrogenic systemic fibrosis: An emerging threat among renal patients. Semin Dial 2006;19:191-194.

4. Introcaso CE, et al. Nephrogenic fibrosing dermopathy/nephrogenic systemic fibrosis: A case series of nine patients and review of the literature. Int J Dermatol 2007;6:447–452.

5. Mendoza FA, et al. Description of 12 cases of nephrogenic fibrosing dermopathy and review of the literature. Semin Arthritis Rheum 2006;35: 238-249.

6. Gibson SE, et al. Multiorgan involvement in nephrogenic fibrosing dermopathy; an autopsy case and review of the literature. Arch Path Lab Med 2006;130:209-212.

7. Keyrouz S, Rudnicki SA. Neuromuscular Involvement in nephrogenic systemic fibrosis. J Clin Neuromusc Dis 2007;9:297–302.

8. Ting WW, et al. Nephrogenic fibrosing dermopathy with systemic involvement. Arch Dermatol 2003;139:903–9.

9. Knopp EA, Cowper SE. Nephrogenic systemic fibrosis: Early recognition and treatment. Semin Dial 2008;21:123-8.

10. Cowper SE: Nephrogenic fibrosing dermopathy: The first 6 years. Curr Opin Rheumatol 2003;15:785–790.

11. Cowper SE, et al. Nephrogenic fibrosing dermopathy. Amer J Dermatopathol 2001;23:383–393.

12. Bucala R. Circulating fibrocytes: Cellular basis for NSF. J Am Coll Radiol 2008;5:36-39.

13. Cowper SE, et al., Clinical and histological findings in nephrogenic systemic fibrosis. Eur J Radiol 2008;66:191-9.

14. ACR Manual on Contrast Media - Version 7, 2010.

15. Levey AS, et al. A more accurate method to estimate glomerular filtration rate from serum creatinine: A new prediction equation. Modification of Diet in Renal Disease Study Group. Ann Intern Med 1999;130:461–470.

16. Cockcroft DW, Gault MH. Prediction of creatinine clearance from serum creatinine. Nephron 1976;16:31–41.

17. Levey AS, et al. National Kidney Foundation. National Kidney Foundation practice guidelines for chronic kidney disease: Evaluation, classification, and stratification. Ann Intern Med 2003;139:137–147. (Erratum in Ann Intern Med 2003; 139:605)

18. Stevens LA, et al. Impact of creatinine calibration on performance of GFR estimating equations in a pooled individual patient database. Am J Kidney Dis 2007;50:21–35.

19. Levey AS, et al. Chronic Kidney Disease Epidemiology Collaboration. Expressing the Modification of Diet in Renal Disease Study equation for estimating glomerular filtration rate with standardized serum creatinine values. Clin Chem 2007;53:766–772.

20. Levey AS, et al. Chronic Kidney Disease Epidemiology Collaboration. Using standardized serum creatinine values in the modification of diet in renal disease study equation for estimating glomerular filtration rate. Ann Intern Med 2006;145:247–254.

21. Schwartz GJ, et al. A simple estimate of glomerular filtration rate in children derived from body length and plasma creatinine. Pediatrics 1976;58:259–263.


23. Nandwana SB, et al. Gadobenate dimeglumine administration and nephrogenic systemic fibrosis: Is there a real risk in patients with impaired renal function. Radiology 2015;15:741-747.

24. Shaffer KM, et al. The renal safety of intravenous gadolinium enhanced magnetic resonance imaging in patients awaiting liver transplantation. Liver Transplantation 2015;21:1340-6.

  Shellock R & D Services, Inc. email:
  © 2004- by Shellock R & D Services, Inc. and Frank G. Shellock, Ph.D. All rights reserved. (v3.1.109)