Nephrogenic Systemic Fibrosis

Updated: Feb 08, 2016
  • Author: Noah S Scheinfeld, JD, MD, FAAD; Chief Editor: Dirk M Elston, MD  more...
  • Print


Nephrogenic systemic fibrosis (NSF), also known as nephrogenic fibrosing dermopathy (NFD), is a disease of fibrosis of the skin and internal organs reminiscent but distinct from scleroderma or scleromyxedema. It is caused by gadolinium exposure used in imaging in patients who have renal insufficiency.

Nephrogenic systemic fibrosis always occurs (with the exception of one report in 2 transplant patients whose organ donors’ histories were not noted [1] ) in patients with renal insufficiency who have had imaging studies (eg, magnetic resonance angiography) with gadolinium, a contrast agent used in imaging studies. Gadolinium can be found in tissue samples of nephrogenic systemic fibrosis. Evidence for a link between nephrogenic systemic fibrosis and gadolinium was first described in a case series of 13 patients, all of whom developed nephrogenic systemic fibrosis after being exposed to gadolinium. [2]  Sometimes articles are published that state a patient with renal impairment developed nephrogenic systemic fibrosis, but these are likely instances when an inadequate history has been taken. [3] One case of nephrogenic systemic fibrosis developed 10 years after gadolinium exposure. [4]

Nephrogenic systemic fibrosis resembles scleroderma and eosinophilic fasciitis clinically and scleromyxedema histopathologically. Patients with nephrogenic systemic fibrosis may develop large areas of indurated skin with fibrotic nodules and plaques. Flexion contractures with an accompanying limitation of range of motion also can occur. Although most patients with nephrogenic systemic fibrosis have undergone hemodialysis for renal failure, some have never undergone dialysis and others have received only peritoneal dialysis.

Histopathologically, nephrogenic systemic fibrosis resembles scleromyxedema in that it manifests with a proliferation of dermal fibroblasts and dendritic cells, thickened collagen bundles, increased elastic fibers, and mucin deposition.



The pathophysiology of nephrogenic systemic fibrosis is related to the exposure of patients with renal insufficiency to gadolinium in association with imaging studies. The chelated forms of the less stable gadolinium chelates might have a significant role, but it appears that dissociated gadolinium's gradual release is pivotal in the development of nephrogenic systemic fibrosis and its sometimes delayed onset. [5]

Evidence for a link between nephrogenic systemic fibrosis and gadolinium was first described in a case series of 13 patients, all of whom developed nephrogenic systemic fibrosis after being exposed to gadolinium. [2] The mechanism by which this occurs in not known, but it seems to involve a cell termed a circulating fibrocyte that is stimulated by gadolinium. [6] Endothelin-1/endothelin receptor signaling plays a role in the calcification and fibrosis of nephrogenic systemic fibrosis. [7]

Toll-like receptors (TLR), in particular TLR4 and TLR7, play a role in the development of nephrogenic systemic fibrosis. [8]

Thomsen et al [9] noted that more than 90% of proven nephrogenic systemic fibrosis cases are related to gadodiamide (Omniscan) and some to gadopentetate (Magnevist). [10] As such, gadoversetamide (OptiMARK) and gadopentetate dimeglumine (Magnevist) should not be used for imaging in patients with renal impairment. MultiHance and ProHance, similar brands, should also likely not be used.

The safety of gadopentetate linear product might be no different from macrocyclic preparations such as gadodiamide (Omniscan), but guidelines should be followed in all gadolinium products. [11]

The macrocyclic contrast agents gadobutrol (Gadovist/Gadavist) and gadobenate dimeglumine (MultiHance) should be used only following guidelines. It is possible that Gadovist, Dotarem, and Prohance are safer, but this does not justify changing guidelines. [12]

While evidence data showing a benefit for prompt hemodialysis after gadolinium imaging are lacking, this is a justified precaution. Similar guidelines from the FDA and the American College of Radiology (ACR) state that gadolinium use with an approximate glomerular filtration rate of 30-44 mL/min per 1.73 m2 should be used with extreme caution.

The relationship between epoetin alfa (Epogen) and nephrogenic systemic fibrosis has engendered controversy. Whether epoetin alfa is related to nephrogenic systemic fibrosis or if severe renal impairment merely sets the stage for nephrogenic systemic fibrosis remains unclear. Goveia et al [13] noted in their case control study that 100% of patients with nephrogenic systemic fibrosis (n = 8) were treated with recombinant epoetin after undergoing renal transplantation versus only 6% of control subjects (n = 24). They theorized that epoetin, through its ability to promote endothelial cell proliferation and augment fibrin-induced wound healing, could play a role in the pathogenesis of nephrogenic systemic fibrosis. Saab [14] challenged this conclusion and noted that 88% of patients with nephrogenic systemic fibrosis had a serum creatinine greater than 5 mg/dL as compared with only 21% of control subjects. The fact that these patients had significantly worse renal function as compared with most of the control subjects puts them at higher risk for requiring recombinant epoetin therapy for management of anemia. Because nephrogenic systemic fibrosis is only seen in patients with severe renal insufficiency, the epoetin requirement in this group may simply be a manifestation of decreased renal function.

Gadolinium can be deposited in almost any tissue in the body after its use for imaging studies. Gibson et al [15] noted 2 reports with apparent multiorgan fibrosis with involvement of skeletal muscle, myocardium, the lungs, the kidneys, and the testes. Of interest, a condition that resembles nephrogenic systemic fibrosis is eosinophilia-myalgia syndrome, which is also caused by an exogenous substance.

The amount of gadolinium needed to induce aberrant production of hyaluronic acid seems to be minimal. According to an abstract presented by Dr. Susie Mukherjee reported at the 2007 annual meeting of the British Association of Dermatologists, only tiny concentrations of gadolinium are needed to stimulate hyaluronan synthesis by fibroblasts. Both 10-mmol/L and 1-mmol/L concentrations of gadolinium caused a 2.3-fold increase in hyaluronan synthesis. [16]

Parsons et al [17] performed immunohistochemical studies using antibodies to transglutaminase-2, factor XIIIa, transglutaminase isopeptide, and the histiocyte marker CD68 on 5 archived skin biopsy specimens of nephrogenic systemic fibrosis. Parsons et al found that dermal fibroblasts and histiocytes of nephrogenic systemic fibrosis expressed transglutaminase-2, CD68, factor XIIIa, and transglutaminase isopeptide. They posited that this represented increased expression, activation, or concomitant activation and expression of transglutaminases in nephrogenic systemic fibrosis.

Edward et al [18] found that fibroblasts derived from skin affected by nephrogenic systemic fibrosis synthesize elevated levels of sulphated glycosaminoglycans, in particular hyaluronan, compared with normal control samples, while serum from the one patient with dermatomyositis and from the 2 patients with nephrogenic systemic fibrosis stimulated sulphated glycosaminoglycans synthesis, including hyaluronan synthesis, by both control and patient fibroblasts. Metformin use might influence nephrogenic systemic fibrosis, but this is not proven. [19]

Iron metabolism shapes the development of nephrogenic systemic fibrosis in a mouse model. [20]

Historical theories on the etiology of nephrogenic systemic fibrosis

The following discussion is interesting more for historical purposes, because, since the identification of the nephrogenic systemic fibrosis–gadolinium link in 2006, the understanding of nephrogenic systemic fibrosis has changed radically.

Mackay-Wiggan et al [21]

In 2003, Mackay-Wiggan et al found that all patients in their series had anticardiolipin or antiphospholipid antibodies detected on testing. This implied a role for these antibodies in the development of nephrogenic systemic fibrosis. Mackay-Wiggan et al suggested that although these antibodies occur in 10-29% of patients with end-stage renal disease, the antibodies are more common in patients with nephrogenic systemic fibrosis.

Mackay-Wiggan et al also suggested that the lipid molecule of the antiphospholipid or anticardiolipin antibody may interact with a lipid substance in the dialysis procedure. How it interacts is uncertain. Possibly, it interacts with the dialysis machine's filter or the tubing to stimulate fibroblast or mucin production. This cause would not explain the occurrence of nephrogenic systemic fibrosis in the small subset of end-stage renal disease patients with an onset of the disorder before beginning hemodialysis.

Another theory of Mackay-Wiggan et al is that an accumulated substance intrinsic to acute or chronic renal failure may interact with the antiphospholipid antibody. Yet another theory of Mackay-Wiggan et al is that sudden, severe edema may trigger a fibrotic and mucinous cutaneous reaction that results in this progressive scleromyxedemalike illness. They speculate that perhaps edema coupled with immunosuppression in patients with antiphospholipid antibodies stimulate a physiologic response, resulting in the proliferation of fibroblastlike cells and mucin deposition in the dermis.

McNeill and Barr [22]

In 2002, McNeill and Barr hypothesized that patients with hepatitis C who were undergoing hemodialysis would be at increased risk for this disorder because of increased levels of basic fibroblast growth factor, transforming growth factor-beta1, or both.

Jiménez et al [23] and Ortonne et al [24]

In 2004, Jiménez et al reported on their histopathologic studies. These studies of patients with nephrogenic systemic fibrosis indicated that the fibrotic process of nephrogenic systemic fibrosis affected the subcutaneous tissue, fascia, striated muscles, lungs, and myocardium, in addition to the dermis. The skin contained large numbers of CD68+/factor XIIIa+ dendritic cells and increased expression of transforming growth factor-beta1.

Also in 2004, Ortonne et al suggested the presence of CD45RO+ CD34+ cells with collagen synthesis activity as part of the etiology of nephrogenic systemic fibrosis.

Kucher et al [25]

Kucher et al [25] reviewed 9 biopsy specimens positive for a nephrogenic systemic fibrosis diagnosis and 7 biopsy specimens positive for a scleromyxedema diagnosis.

Immunohistochemical staining for CD34, factor XIIIa, CD31, smooth muscle actin, CD68, and colloidal iron were similar for both conditions. Procollagen-I showed increased expression in scleromyxedema. The significance of this is unclear.




United States

Nephrogenic systemic fibrosis is an uncommon condition. Since 1997, hundreds of cases have been reported to the NFD Registry. Few cases have occurred since guidelines have been instituted by the ACR, FDA, and EU regulatory authorities; however, this author has heard at meetings that a few sporadic cases have occurred. These cases do not seem to have been reported the NSF Registry at Yale. That is, even with guidelines in place, in 2010, 2011, 2012, 2013, a handful of new cases of were reported but many fewer than before gadolinium was defined as the cause. Some of these cases likely involve exposure to gadolinium used before the guidelines were put in place and involve delayed onset of the disease.

Deo et al [26] studied a population of patients with end-stage renal disease in and around Bridgeport, Connecticut over an 18-month period. The incidence of nephrogenic systemic fibrosis was 4.3 cases per 1000 patient-years. Each radiologic study using gadolinium presented a 2.4% risk for developing nephrogenic systemic fibrosis.

Todd et al [27] found that exposure to gadolinium-containing contrast was associated with an increased risk of developing cutaneous changes of nephrogenic systemic fibrosis (odds ratio, 14.7; 95% confidence interval, 1.9-117.0) compared with patients who were not exposed to gadolinium.


Nephrogenic systemic fibrosis is an uncommon condition. Several case series from Europe have been reported. Four cases were described in a report in the British Journal of Dermatology in March 2003. [28] Additional cases have been reported in Germany and Switzerland.

For example, delayed-onset nephrogenic systemic fibrosis was reported in a Korean person with renal failure who demonstrated a "reverse superscan" on bone scintigraphy at age 19 years who had been exposed to gadolinium at age 14 years. [29]

A study in Denmark found a rate of 12 cases of nephrogenic systemic fibrosis per million, which may be the highest prevalence worldwide. [30]

In France, after new guidelines for use of gadolinium were instituted, a study of patients from 2009-2011 found that no cases of nephrogenic systemic fibrosis occurred. [31]


No racial predisposition is reported. Whites, Hispanics, African Americans, and Asians have all been reported with this condition.


No sexual predilection is recognized. Data from the NFD Registry indicate a male-to-female ratio of 1:1. However, in the 2001 article by Cowper et al, [32] the male-to-female ratio in patients with biopsy-proven disease was 9:5.


Nephrogenic systemic fibrosis has been reported in all age groups; persons of any age exposed to gadolinium may develop nephrogenic systemic fibrosis. The first group of persons reported with nephrogenic systemic fibrosis were adults whose ages ranged from 31-74 years. Data from the nephrogenic systemic fibrosis Registry indicate a range of ages from 8-87 years at the time of disease onset, with a mean age of 46.4 years. In 2003, Jan et al [33] reported on 2 pediatric cases. A series from England noted a patient in her 20s with the disease.

Children are generally not less able to develop nephrogenic systemic fibrosis. However, since new guidelines have been implemented, children who receive gadolinium have not developed nephrogenic systemic fibrosis. [34]

The development of nephrogenic systemic fibrosis is not correlated with the duration of renal failure, and it can occur early as well as later. Hancox et al [35] noted a case of nephrogenic systemic fibrosis after 5 days of hemodialysis.