INTRODUCTION — Nephrogenic systemic fibrosis (NSF) is a disorder seen only in patients with advanced kidney disease and is characterized by two primary features [1-3]:
●Thickening and hardening of the skin overlying the extremities and trunk
●Marked expansion and fibrosis of the dermis in association with CD34+ fibrocytes
NSF was originally named nephrogenic fibrosing dermopathy (NFD) because of the characteristic skin findings. However, subsequent studies showed that some patients had fibrosis of deeper structures, including muscle, fascia, lungs, and heart [1]. Because of the systemic findings, NSF is preferred to NFD [4]. (See 'Systemic involvement' below.)
While NSF is only seen among patients with kidney disease, retention of gadolinium (Gd) has been demonstrated in other organs (eg, brain) in patients without and with kidney disease [5]. (See "Patient evaluation before gadolinium contrast administration for magnetic resonance imaging", section on 'Concerns about gadolinium retention multiple organs'.)
This topic will discuss the epidemiology, pathogenesis, clinical manifestations, diagnosis, prevention, and treatment of NSF. General principles of magnetic resonance imaging (MRI), indications for Gd use, and patient evaluation prior to Gd administration are discussed at length elsewhere. (See "Patient evaluation before gadolinium contrast administration for magnetic resonance imaging" and "Principles of magnetic resonance imaging".)
EPIDEMIOLOGY — The cases of nephrogenic systemic fibrosis (NSF) reported in the beginning of the millennium were attributed to the use of the unstable linear gadolinium-based contrast agents (GBCAs; eg, gadodiamide, gadopentetate, or gadoversetamide) administered to patients with dialysis-dependent end-stage kidney disease (ESKD) or acute kidney injury (AKI) [3,6,7]. The linear and macrocyclic GBCAs (eg, gadobenate, gadoterate, or gadoteridol) are more stable as a result of higher thermodynamic stability and low kinetic inertness, both of which minimize the release of the pathogenic free gadolinium (Gd3+) in vivo [8]. Clinician awareness of NSF, adoption of restrictive policies regarding GBCAs for patients with advanced kidney disease, and the use of more stable GBCAs have led to a dramatic reduction in NSF cases after 2007 [3,8-11]. (See 'Pathogenesis' below.)
The American College of Radiology (ACR) has classified the GBCAs into three categories according to their likelihood of causing NSF, with group I being the most likely GBCAs to cause NSF, group II being the least likely to cause it, and group III having limited data suggestive of low risk of NSF (table 1) [12-15]. The details regarding various gadolinium (Gd) groups are discussed at length elsewhere. (See "Patient evaluation before gadolinium contrast administration for magnetic resonance imaging", section on 'Gadolinium group and risk for nephrogenic systemic fibrosis'.)
The incidence rates of NSF are difficult to ascertain, as they vary depending upon the frequency and severity of kidney disease among patients studied, the type and dose of GBCA, and the sample size under study [16-18]. The risk of NSF with the use of more stable linear or macrocyclic GBCAs is substantially lower than with older, unstable agents, but is not zero:
●Group I agents – In one representative study with estimates of NSF presented by chronic kidney disease (CKD) stage, NSF incidence after receipt of gadodiamide (a group I agent) was 0.4 percent among patients with ESKD on maintenance hemodialysis, 0.6 percent among patients with CKD and estimated glomerular filtration rate (eGFR) 15 to 30 mL/min/1.73 m2, 10 percent with CKD and eGFR <5 mL/min/1.73 m2 not on dialysis, and 20 percent among patients with AKI and rising serum creatinine at the time of exposure [18]. Other studies also highlight the risk with group I agents [6,7].
●Group II agents – Multiple studies have reported that the risk of NSF is substantially lower with group II agents [6-8,11,14,19-31]. Although these studies essentially reported no cases of NSF, they were limited by low numbers of patients with an eGFR <15 mL/min/1.73 m2 or who were on dialysis, and for the most part, patients received only one to two exposures of GBCA at a standard dose [8]. Most cases of NSF after group II agents are confounded (ie, a group I agent was administered prior to a group II agent, although the timing between administrations varied, and administration of the group II agent was more temporally related to the NSF); a minority of cases (single digit) were unconfounded (ie, only a group II agent was administered). Representative studies are as follows:
•In a meta-analysis of 16 prospective and retrospective studies including 4931 patients with an eGFR <30 mL/min/1.73 m2 who were administered group II agents, there were no unconfounded cases of NSF identified [11,31]. The number of confounded cases was not reported, and patients with AKI were not included.
•In a large analysis of NSF cases from 2006 to 2016, 25 confounded cases and five unconfounded cases of NSF (three of which were biopsy confirmed) were identified [6]. Of these, there were three confirmed unconfounded cases. However, the proportion of patients with advanced CKD or AKI who received gadobutrol was not reported, and, therefore, the incidence of NSF among patients with advanced kidney disease after administration of a group II GBCA was unclear.
●Group III agents – Data on group III agents are limited, and therefore their safety cannot be ascertained with as much certainty as for the group II agents. However, no unconfounded cases of NSF have been reported among patients with CKD to date [14,15,32-34].
There appears to be no predilection to NSF by sex, race, age, etiology or duration of kidney disease, or time on dialysis [35-39]. Conversely, NSF may be more common in peritoneal dialysis as compared with hemodialysis [40,41]. Additional discussion regarding patient-level risk factors is presented elsewhere. (See "Patient evaluation before gadolinium contrast administration for magnetic resonance imaging", section on 'Patient risk factors for nephrogenic systemic fibrosis'.)
A number of other possible associations with NSF have been found or suggested in patients with kidney disease, including vascular manipulation, increased vascular permeability, presence of edema, infection, medications (including erythropoietin), hyperphosphatemia, and liver transplantation [3,38,40-48]. However, it is difficult to ascertain these associations.
PATHOGENESIS — Gadolinium-containing contrast agents (GBCAs) have been implicated as a cause for nephrogenic systemic fibrosis (NSF) [41,49-55]. A discussion of the various types/groups of gadolinium (Gd) and their varying risk for NSF are discussed at length elsewhere. (See "Patient evaluation before gadolinium contrast administration for magnetic resonance imaging", section on 'Gadolinium group and risk for nephrogenic systemic fibrosis'.)
Free gadolinium (Gd3+) is a poorly soluble and highly toxic lanthanide heavy metal [50]. In order to enable its use in humans, Gd3+ is bound to a large organic molecule called a chelate. Gd3+ may dissociate from the chelate with continued in vivo exposure, such as in patients with end-stage kidney disease (ESKD) before undergoing high-efficiency dialysis. Such dissociation from the chelate has been shown to be enhanced in the presence of iron [56].
Gd3+ has several deleterious effects in vivo [50]. It is similar in size to calcium, which allows it to bind calcium-binding enzymes, thereby inhibiting them. It precipitates in tissues, disrupts calcium ion passage in nerve and muscle cells, and interferes with intracellular enzymes and cell membranes. Gd3+ can also form precipitates with certain anions (such as phosphate) that are elevated in patients with kidney disease [4,57].
How Gd3+ or its precipitates lead to NSF is poorly understood [4,58]. The tissues affected by NSF resemble tissues that are undergoing an injury reaction. Similar to injured tissue, the tissues affected by NSF contain a higher number of macrophages and fibroblasts. It is postulated that the fibrogenic cytokines secreted by the macrophages and fibroblasts lead to a cascade of events similar to wound healing, but with exaggerated fibrosis.
Three proposed mechanisms of exaggerated fibrogenic response are [50,51,59]:
●Activation of the transforming growth factor (TGF) beta-1 pathway – There is a marked increase in TGF-beta-1 messenger RNA levels along with the presence of many CD68+/factor XIIIa+ dendritic cells in the skin and fascia of affected patients [60,61]. According to one theory, a noxious stimulus activates CD68+/factor XIIIa+ dendritic cells in tissue, which then produce TGF-beta-1 [35]. However, TGF-beta-1 itself regulates dendritic cell maturation and antigen presentation. A vicious cycle ensues as this leads to further recruitment and activation of dendritic cells and more TGF-beta-1 production, ultimately resulting in tissue fibrosis.
●Increase in circulating fibrocytes – A second theory is that Gd deposition in skin in some manner, possibly as a direct toxin, stimulates the bone marrow to produce CD34+ circulating fibroblasts, which accumulate in the affected tissue and produce collagen, even in the absence of tissue injury [35,59]. This is supported by one study showing that tissue fibrosis in NSF is caused by fibrocytes recruited from the circulation rather than by proliferation of resident dendritic cells [4].
●Secretion of cytokines and growth factors by monocytes – Peripheral blood monocytes are thought to secrete proinflammatory and profibrotic cytokines as well as growth factors in response to Gd chelates [62]. Cytokines secreted include interleukin (IL) 13, IL-4, and IL-6, and growth factors include TGF-beta and vascular endothelial growth factor, which are all specific to a monocyte response.
CLINICAL MANIFESTATIONS — Nephrogenic systemic fibrosis (NSF) is characterized by skin involvement in all patients and systemic involvement in some [1]. The median onset of NSF after gadolinium (Gd) exposure is 42 days, with an interquartile range of 19 to 90 days [3]. Disease has been detected anywhere from 1 day to 10 years after exposure.
Skin involvement — Skin disease in NSF typically presents as fibrotic, indurated papules, plaques, or subcutaneous nodules that may or may not be erythematous [1,61,63].
In the majority of cases, the disease is bilateral and symmetric. The skin lesions first appear on the ankles, lower legs, feet, and hands, and then progress proximally to involve the thighs, forearms, and, infrequently, the buttocks [35,37]. Unusual distribution patterns involving the skin overlying the middle [64] and lower abdomen [35,37] have been reported. The head (other than scleral plaques) is usually spared [35,65].
The lesions are often preceded by edema of the region, often mimicking cellulitis. Upon resolution of edema, the involved skin retains a thickened and indurated texture [1]. The skin may have a "cobblestone" [66], "woody" [1,35,60], or peau d'orange appearance (orange peel-like) (picture 1) [1,65]. The lesions may be pruritic or present with sharp pain or a burning sensation [1,36,60]. Confluent dermal plaques with thickening and hardening may also occur.
There may be sclerodactyly or loss of skin folds over the dorsum of the hands and lower extremities [60]. Unlike autoimmune sclerosing conditions, livedo reticularis is not a feature of NSF. Late in the disease, hyperpigmentation, hairlessness, and epidermal atrophy have been described [67].
Systemic involvement — The prevalence of systemic involvement is unknown, but a number of different organ system manifestations have been described:
●Muscle and joint involvement – Bilateral flexion joint contractures can occur in up to 70 percent of patients with advanced disease [3,35,60,68,69]. The limitations in range of motion and flexibility are commonly due to periarticular skin thickening and fibrosis, since there is no evidence of joint damage in the form of synovitis or arthritis (picture 2A-B). Muscle induration may be seen [37,68,70], but strength is normal or only slightly reduced [68].
●Internal organs – Fibrosis has also been identified in a variety of internal organs, including the lungs (with reduced diffusion capacity) [60] and diaphragm (with respiratory failure) [37,71], myocardium [37,60,72], pericardium, pleura [37], and dura mater [72].
●Scleral plaques in young patients – Yellow asymptomatic scleral plaques are common among patients with NSF (picture 3) [1,38]. However, these can also be seen with aging in the general population. Thus, scleral plaques are considered significant if noted in patients younger than 45 years of age.
DIAGNOSIS — The diagnosis of nephrogenic systemic fibrosis (NSF) is based upon histopathologic examination of biopsy tissue from an involved site in a patient with classic clinical findings.
When to suspect NSF — We suspect NSF in patients with kidney disease and history of gadolinium (Gd) exposure who present with classic skin and joint findings. (See 'Clinical manifestations' above.)
Laboratory findings are generally nonspecific. Concomitant presence of restrictive lung disease, myocardial fibrosis, cardiomyopathy, or fibrosis of the fascia and muscles on various imaging studies in someone with classic skin findings may suggest the diagnosis [35,68].
Confirmation of diagnosis — The diagnosis of NSF can be confirmed by combining clinical findings and histopathology [69]. Thus, in patients whose clinical picture is suspicious for NSF, we obtain a deep incisional or a punch biopsy to capture tissue from subcutaneous fat, fascia, and muscle [1,38,63].
The findings obtained from physical examination and histopathologic evaluation of the skin biopsy are then used to calculate clinical and histologic scores using a system developed from the Yale International NSF Registry (table 2) [69]. Clinical and histologic scores are plotted to establish the diagnosis of NSF (figure 1). Clinical examination and/or biopsy should be repeated in patients with inconsistent or suggestive findings (figure 1). The utility of this scoring system in establishing the diagnosis has not been validated.
Light microscopy findings can vary with disease severity, ranging from a subtle proliferation of spindle-shaped dermal fibrocytes to a thickened fibrosed dermis with long dendritic processes [52]. There may be an abundance of histiocytes and stellate dendritic cells (picture 4A-B) [1,61]. Thick collagen bundles with surrounding clefts are a prominent finding. Inflammation is usually absent, and elastic fibers are preserved. Osseous sclerotic bodies in skin biopsies obtained from hyperpigmented skin lesions years after the development of NSF have been described. These appear as intensely pink or intensely blue ovoid lesions when stained with hematoxylin and eosin or trichrome stain, respectively. They have been shown to contain gadolinium [73,74].
Immunohistochemical staining reveals abundant CD34+ dermal dendritic cells, with the dendritic processes aligning along elastic fibers and around collagen bundles, forming a dense network (picture 5A-B) [61]. Dendritic cells positive for CD68, factor XIII, or both may be present [37,60].
Special testing is available for detecting Gd in tissue specimens, although it is of limited utility [75-77]. An isolated presence of Gd in tissues is not diagnostic of NSF. Conversely, the absence of Gd on histopathology does not exclude a diagnosis of NSF.
Reporting obligations — Health care providers are encouraged to report cases and include information about prior exposures, treatments offered, and outcome of treatment to the International NSF Registry [78].
Similarly, the US Food and Drug Administration (FDA) urges health care providers and patients in the United States to report adverse event information to the FDA via the MedWatch program by phone (1-800-FDA-1088), fax (1-800-FDA-0178), or their website. Affected patients outside the United States should be reported to the appropriate regulatory agency.
DIFFERENTIAL DIAGNOSIS — The thickening and hardening of the skin overlying the extremities and trunk as seen in nephrogenic systemic fibrosis (NSF) can mimic a variety of other disorders, such as systemic sclerosis (SS; scleroderma), scleromyxedema, eosinophilic fasciitis, and calciphylaxis [35,60,69,79].
These conditions can be differentiated from NSF based upon clinical or histologic features:
●Systemic sclerosis (scleroderma) – SS can be distinguished from NSF by the presence of Raynaud phenomenon and serologic tests suggestive of the disease. SS is characterized by positive antinuclear antibodies and either anticentromere or anti-DNA topoisomerase I (Scl-70) antibodies, which are typically negative in NSF. Additional information regarding diagnosis of scleroderma is presented elsewhere. (See "Clinical manifestations and diagnosis of systemic sclerosis (scleroderma) in adults" and "Clinical manifestations and diagnosis of systemic sclerosis (scleroderma) in adults", section on 'Diagnosis'.)
●Scleromyxedema – Scleromyxedema shares some clinical and histopathologic features with NSF. However, it can be distinguished by disease involvement of the head and association with a monoclonal gammopathy, usually of immunoglobulin G lambda type [35,65]. (See "Cutaneous manifestations of internal malignancy", section on 'Scleromyxedema'.)
●Eosinophilic fasciitis – Eosinophilic fasciitis is distinguished from NSF by sparing of the hands and feet and by presence of eosinophilia and eosinophilic tissue infiltration, which are typically absent in patients with NSF.
●Calciphylaxis – Calciphylaxis can occur in patients with chronic kidney disease (CKD) and be mistaken for early NSF. However, a tissue biopsy can usually distinguish between these disorders. (See "Calciphylaxis (calcific uremic arteriolopathy)".)
PREVENTION — The best method to prevent nephrogenic systemic fibrosis (NSF) is the avoidance of gadolinium (Gd) in patients with acute kidney injury (AKI) or chronic kidney disease (CKD). If Gd is necessary, then group I agents should be avoided. Patients with AKI or stage 4 or 5 CKD should be informed of the potential risk of NSF associated with gadolinium-based contrast agent (GBCA) administration, the reason GBCA administration is indicated, and whether there are alternative diagnostic studies that would avoid exposure to GBCA while providing comparable information [15].
Once Gd is administered, hemodialysis may be performed for its removal. However, this practice is controversial and the ability of this post-Gd hemodialysis session to prevent NSF is unproven [8]. (See 'Removal of gadolinium with hemodialysis' below.)
Choice of gadolinium product in kidney function impairment — The Gd groups and risk for NSF are discussed at length elsewhere (algorithm 1). (See "Patient evaluation before gadolinium contrast administration for magnetic resonance imaging", section on 'Gadolinium group and risk for nephrogenic systemic fibrosis'.)
Assessment of kidney function prior to gadolinium exposure — The decision-making for Gd use by various thresholds of estimated glomerular filtration rate (eGFR) is discussed at length elsewhere (algorithm 1). (See "Patient evaluation before gadolinium contrast administration for magnetic resonance imaging", section on 'Patient risk factors for nephrogenic systemic fibrosis'.)
Removal of gadolinium with hemodialysis — No high-quality data demonstrate that hemodialysis may prevent NSF. After inadvertent exposure to Gd or appropriate administration of Gd following a risk-benefit discussion, our approach to hemodialysis is based upon the causative role of group I agents in NSF, lack of absolute certainty over the benign nature of group II and III agents, and the high dialyzability of Gd. However, we acknowledge the uncertainty of clinical benefit with our approach, which is outlined below [8]:
●Patients already on hemodialysis – Among patients with end-stage kidney disease (ESKD) or AKI who are already receiving hemodialysis, we dialyze as soon as possible (within hours, not days) after Gd is administered [8]. We attempt to schedule the MRI with Gd just before the patient's next scheduled dialysis session, when possible. Otherwise, we perform an additional dialysis treatment shortly after the imaging study. In addition, we perform a second dialysis session approximately 24 hours after the first session.
However, other experts take a different approach in this setting. For example, some guidelines do not recommend urgent dialysis in patients whose MRI with a group II or III GBCA cannot be obtained just before a regularly scheduled dialysis session, and do not recommend multiday dialysis solely to remove group II or III agents [15,80].
●Patients not already on hemodialysis – Among patients with CKD, severe AKI, or AKI with rising creatinine who are not already on hemodialysis, and among patients receiving peritoneal dialysis, our practice differs based upon the type of Gd administered (see "Patient evaluation before gadolinium contrast administration for magnetic resonance imaging", section on 'Gadolinium group and risk for nephrogenic systemic fibrosis'):
•Group II and III agents – We do not initiate hemodialysis among such patients solely to prevent NSF, regardless of the presence or absence of a functioning access. The rationale is that the risks of placing central venous access and initiating hemodialysis for the first time outweigh the risk of NSF. Although data for group III agents are limited, available studies have reported no NSF cases among patients with CKD. (See 'Epidemiology' above.)
•Group I agents – If group I agents are administered to patients who have eGFR <15 mL/min/1.73 m2 (whether due to CKD or AKI), patients receiving peritoneal dialysis, or patients with AKI and rising serum creatinine at the time of exposure, we perform two hemodialysis sessions 24 hours apart. For such patients, we would place a temporary dialysis catheter for delivery of hemodialysis if they don't already have a functioning hemodialysis access.
Although there is no evidence that hemodialysis lowers the risk or severity of NSF, hemodialysis immediately after Gd exposure results in significant removal of the contrast agent.
GBCAs have a molecular weight between 500 and 1000 daltons, and they are neither protein bound nor lipophilic [81]. Thus, their volume of distribution is primarily limited to the extracellular fluid. Collectively, these physical properties make GBCAs amenable to dialysis. This was demonstrated in a study of 70 hemodialysis patients who underwent GBCA-enhanced MRI, with average rates of Gd removal of 78, 96, and 99 percent in the first, second, and third every-other-day dialysis sessions, respectively [82]. Other data also report effective clearance of Gd with hemodialysis [83,84].
There are no studies systematically comparing the incidence of NSF among patients who did and did not receive preventative dialysis. In one study of 12 patients who either had AKI or ESKD, group I GBCA-enhanced MRI was associated with development of NSF despite dialysis performed within two days of exposure among 10 of the 12 patients [85]. Some have questioned the benefit of dialysis for prevention of NSF for reasons including lack of evidence of benefit; low likelihood of NSF with group II agents that are commonly in use since the discovery of NSF; logistical challenges of arranging for dialysis; and concern over rapid distribution of GBCAs within tissues after injection, which may render subsequent dialysis futile [86]. Thus, in light of these collective data, we perform preventative hemodialysis following exposure to GBCAs, as a conservative measure, only among patients who are already on regular hemodialysis either for ESKD, severe AKI, or AKI with a rising creatinine. This is based upon the theoretical benefit of preventing NSF with hemodialysis coupled with negligible risk. However, this risk and benefit analysis does not favor placement of a temporary hemodialysis catheter and performance of dialysis among patients not already on hemodialysis.
TREATMENT — There is no proven effective medical therapy for nephrogenic systemic fibrosis (NSF). However, recovery of kidney function may lead to stabilization, marked improvement, or resolution of the disease [35]. Several treatments individually and in combination have been evaluated, but none have shown consistent benefit. In addition, due to the rarity of this condition, most studies involve a small number of patients and a limited duration of follow-up.
Initial management in all patients — All patients should continue receiving therapies directed toward improving their kidney function, if reversible. We refer patients with NSF who have an irreversible decline in kidney function to be evaluated for a kidney transplant. This practice is based upon data showing stabilization or improvement in NSF among some patients who recover kidney function [35,38,39,58,87]. NSF does not affect the transplant wait time. Details regarding evaluation of a patient for kidney transplant and wait times are discussed elsewhere. (See "Kidney transplantation in adults: Evaluation of the potential kidney transplant recipient" and "Kidney transplantation in adults: The kidney transplant waiting list in the United States".)
We also refer all patients with NSF for intensive physical therapy to prevent or reverse disability related to joint contractures [65,88]. In addition, we ensure that the patients' hands are splinted in a functional position to prevent progressive contractures of their finger joints [88,89].
Additional therapies for skin manifestations — We suggest a trial of extracorporeal photopheresis with ultraviolet A (ECP; also called photochemotherapy), which involves extracorporeal exposure of peripheral blood mononuclear cells to photoactivated 8-methoxypsoralen, followed by reinfusion of the treated cells. ECP induces monocyte-derived tumor necrosis factor (TNF)-alpha, which, in turn, suppresses collagen synthesis and enhances collagenase production. The results with ECP have been inconsistent [66,70,90-92], but we favor a trial based on our limited experience [93].
Alternatively, treatment with imatinib mesylate is also reasonable, though supportive data are limited to case reports. In two patients with NSF, treatment with imatinib appeared to result in improvement in skin findings and a reduction in dermal collagen [94]. Skin findings recurred upon discontinuation of imatinib and subsequently improved after reinitiation of imatinib. In another patient with lung fibrosis attributed to NSF, treatment with imatinib was associated with improvement in pulmonary function [95]. Other studies have similarly demonstrated the antifibrotic effect of imatinib [96,97].
Experimental therapies not in use — A number of other therapies have been described in case reports or small series. Possible efficacy has been noted with pentoxifylline [49], intravenous sodium thiosulfate [98,99], alefacept [100], and high-dose intravenous immune globulin [101]. Topical, intralesional, or oral glucocorticoid therapy and cyclophosphamide have generally shown no benefit [35,90].
PROGNOSIS — Nephrogenic systemic fibrosis (NSF) has a chronic and unremitting course in most patients [3,35,38,58,61,102-104]. Recovery of kidney function may lead to stabilization, marked improvement, or resolution of the disease.
A 2019 review included studies between the years 2000 and 2019 of patients who had been diagnosed with NSF by clinical evaluation and biopsy confirmation [3]. Of the 639 patients included in the study, only four deaths were attributed to NSF. Follow-up data were available in 341 patients, of whom 12 patients were considered cured and 72 patients had partial improvement. Among these 84 patients with improvement in their symptoms, 34 patients had improvement in their kidney function prior to improvement of NSF.
Studies among patients who undergo kidney transplantation suggest that not all patients who have improvement in kidney function also have improvement in NSF [3,105,106]. In one study, for example, the rate of NSF improvement was the same in patients who received a kidney transplant and in those who remained on dialysis (55 and 50 percent, respectively) [106]. In addition, symptoms of NSF recurred with deterioration of kidney function among patients with a kidney transplant. Thus, a kidney transplant may improve quality of life in some patients with NSF, but does not guarantee complete or sustained improvement [35,106].
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Dialysis".)
SUMMARY AND RECOMMENDATIONS
●Overview – Nephrogenic systemic fibrosis (NSF) is a fibrosing disorder seen only in patients with advanced kidney disease. NSF is characterized by two primary features (see 'Introduction' above):
•Thickening and hardening of the skin overlying the extremities and trunk
•Marked expansion and fibrosis of the dermis in association with CD34+ fibrocytes
●Epidemiology – Clinician awareness of NSF, adoption of restrictive policies regarding gadolinium-based contrast agents (GBCAs) for patients with severe kidney disease, and the use of more stable GBCAs have led to a dramatic reduction in NSF cases after 2007. The incidence of NSF is highest with group I agents, and while the risk with group II agents is very low, it likely is not zero. The risk with group III agents is unknown. There appears to be no predilection to NSF by sex, race, age, etiology or duration of kidney disease, or time on dialysis. (See 'Epidemiology' above.)
●Pathogenesis – GBCAs have been implicated as a cause for NSF. Free gadolinium (Gd3+) is a poorly soluble and highly toxic lanthanide heavy metal that precipitates in tissues and triggers tissue injury with an exaggerated fibrogenic response. (See 'Pathogenesis' above.)
●Clinical manifestations – Skin involvement occurs in all patients, while some also have systemic manifestations. The median onset of NSF after gadolinium (Gd) exposure is 42 days, but NSF can be seen up to 10 years after exposure. Skin disease typically presents as fibrotic, indurated papules, plaques, or subcutaneous nodules. The skin may have a "cobblestone," "woody," or peau d'orange appearance (orange peel-like) (picture 1). The lesions may be pruritic or present with sharp pain or a burning sensation. Manifestations of systemic involvement may include muscle induration, joint contracture, and fibrosis of the lungs, pleura, diaphragm, myocardium, pericardium, and dura mater. (See 'Clinical manifestations' above.)
●Diagnosis – We suspect NSF in patients with kidney disease and history of Gd exposure who present with classic skin and joint findings. The diagnosis can be confirmed by combining the clinical findings and histopathologic findings from deep incisional or punch biopsy of the involved subcutaneous fat, fascia, and muscle (picture 4A-B and picture 5A-B). These findings are then used to calculate clinical and histologic scores using a system developed from the Yale International NSF Registry (table 2). These scores are then plotted in a calculator to establish the diagnosis of NSF (figure 1). Clinical examination and/or biopsy should be repeated in patients with inconsistent or suggestive findings. The utility of this scoring system in establishing the diagnosis has not been validated. (See 'Diagnosis' above.)
●Differential diagnosis – Although thickening and hardening of the skin overlying the extremities and trunk can simulate a variety of other disorders, such as scleroderma, scleromyxedema, and eosinophilic fasciitis, a history of a temporally associated Gd exposure in the setting of advanced kidney disease is strongly suggestive of NSF. (See 'Differential Diagnosis' above.)
●Prevention – The best method to prevent NSF is the avoidance of Gd. If Gd is necessary in a patient with severe kidney disease, then group I agents should be avoided. Once Gd is administered, we perform hemodialysis for its removal in select patients, despite lack of data demonstrating clinical benefit. Our approach to prevention is as follows (see 'Prevention' above):
•Gadolinium group – The Gd groups and risk for NSF are discussed at length elsewhere (algorithm 1). (See "Patient evaluation before gadolinium contrast administration for magnetic resonance imaging", section on 'Gadolinium group and risk for nephrogenic systemic fibrosis'.)
•Kidney function assessment – The decision-making for Gd use by various thresholds of estimated glomerular filtration rate (eGFR) is discussed at length elsewhere (algorithm 1). (See "Patient evaluation before gadolinium contrast administration for magnetic resonance imaging", section on 'Patient risk factors for nephrogenic systemic fibrosis'.)
•Hemodialysis to remove gadolinium – In patients already receiving hemodialysis (for either end-stage kidney disease [ESKD] or acute kidney injury [AKI]) who have received a GBCA, we suggest hemodialysis for the removal of Gd (Grade 2C). We dialyze as soon as possible (within hours, not days) after Gd is administered. However, this practice is controversial, and its benefits in prevention of NSF are unproven. (See 'Removal of gadolinium with hemodialysis' above.)
An attempt should be made to schedule the MRI with Gd just before the patient's next scheduled dialysis session. Otherwise, we perform an additional dialysis treatment shortly after the imaging study. In addition, we perform a second dialysis session approximately 24 hours after the first session.
Among patients with chronic kidney disease (CKD) or severe AKI who are not already on hemodialysis, and among patients receiving peritoneal dialysis, our practice differs based upon the type of Gd administered to the patient:
-Group II and III agents – We do not perform hemodialysis among these patients regardless of the presence or absence of a functioning access. The rationale is that the risks of placing central venous access and initiating hemodialysis for the first time outweigh the risk of NSF in such patients.
-Group I agents – If group I agents are administered to patients who have eGFR <15 mL/min/1.73 m2 (whether due to CKD or AKI), patients receiving peritoneal dialysis, or patients with AKI and rising serum creatinine at the time of exposure, we suggest placing hemodialysis access (if not present) and performing hemodialysis to remove Gd (Grade 2C). As above, we perform two hemodialysis sessions 24 hours apart.
Hemodialysis immediately after Gd exposure results in significant removal of the contrast agent. However, there is no evidence that hemodialysis lowers the risk or severity of NSF, despite effective removal of Gd.
●Treatment – There is no proven effective medical therapy for NSF. However, recovery of kidney function may lead to stabilization, marked improvement, or resolution of the disease. All patients should continue receiving therapies directed toward improving their kidney function, if reversible. We refer patients with NSF who have an irreversible decline in kidney function to be evaluated for a kidney transplant. Intensive physical therapy may help prevent or reverse disability related to joint contractures. Other therapies such as extracorporeal photopheresis and imatinib mesylate have a limited role in treatment of NSF. (See 'Treatment' above.)
●Prognosis – NSF has a chronic and unremitting course in most patients. A small proportion of patients experience improvement. Cure of NSF is rare. (See 'Prognosis' above.)
ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Ambreen Gul, MD, and Shawn E Cowper, MD, who contributed to earlier versions of this topic review.
آیا می خواهید مدیلیب را به صفحه اصلی خود اضافه کنید؟