Nephrogenic systemic fibrosis, kidney disease, and gadolinium: is there a link?

N ephrogenic systemic fibrosis (NSF), formerly known as nephrogenic fibrosing dermopathy (NFD), is now a major concern for nephrologists. This entity was first described in 1997 in renal transplant recipients with poor graft function (1). More than 215 cases of NSF have subsequently been described in the NFD/NSF registry, with increasing numbers of cases being reported (2). NSF is a fibrosing disorder that involves predominantly the skin but also affects systemic organs such as the liver, heart, lungs, diaphragm, and skeletal muscle (3). It is associated with severe physical disability and death when multisystem disease supervenes (3). The cause of NSF is unknown; however, underlying kidney dysfunction is present in all cases. Approximately 90% of the patients described in the registry have ESRD and are on either hemodialysis or peritoneal dialysis (2). The rest have chronic kidney disease (CKD) or developed NSF in the setting of acute kidney injury (AKI). Thus, underlying kidney disease is a requisite for NSF to occur. Because not all patients with kidney disease develop NSF, one must hypothesize that a trigger is required to set the “fibrosing process” into motion. What do we know about the histology of tissue fibrosis in NSF? Dermal spindle cells, the predominant cell type found in NSF biopsies, have an immunologic profile (CD34/procollagen I) that is identical to blood-borne cells, circulating fibrocytes (cF), which participate in normal wound healing (4). In the setting of tissue/endothelial injury, they enter tissues and engage in wound healing and scar formation. In NSF, however, this process differs from normal wound healing in that cF engage in this activity in the absence of a clinically evident wound. The disturbed environment of kidney disease may supply abnormal signals, which result in cF entry into normal tissues and induction of fibrosis (4). A logical first step to determine the cause of NSF (and why cF inappropriately enter normal tissues) is to examine the underlying characteristics of the host. Uniformly, every patient who has developed NSF had abnormal kidney function. Parenthetically, restoration of renal function in renal transplant recipients and recovery from AKI are noted to regress or stabilize the fibrotic process. Why does underlying kidney disease promote or facilitate the development of NSF? The dialysis procedure itself initially was a major suspect but no longer because NSF develops in patients who have never undergone dialysis (10%) and is absent in the majority of patients who are on long-term maintenance dialysis. Endothelial injury, common in patients with ESRD and CKD, may be one of the critical risk factors by permitting platelets to interact and attach to injured/exposed endothelium (as occurs in normal wound healing). Along this line of reasoning, vascular trauma and thrombotic events occur commonly in patients with ESRD/CKD. Vascular surgical procedures, central catheter placement, deep venous thrombosis, right atrial clots from indwelling catheters, and thrombosed vascular accesses are frequently present before the development of NSF (5). Also, a variety of previously unsuspected hypercoaguable states are uncovered after diagnosis of NSF. One may speculate that the state of “vascular/endothelial dysfunction” that is present in patients with kidney disease (6) primes them for a second event, or “trigger,” that sets the fibrosing process into motion. The trigger for NSF is unknown, but the magnetic resonance imaging (MRI) contrast agent gadolinium (Gd ) has become the leading suspect. In this issue of the Clinical Journal of American Society of Nephrology, two articles describe Gd exposure before the development of NSF in patients who had ESRD and were on dialysis (7,8). A small population study of patients with ESRD that was conducted during an 18-mo period by Deo et al. (7) notes an NSF incidence of 4.3 cases per 1000 patientyears and a 2.4% risk for each Gd exposure. Yerram et al. (8) describe NSF in a patient who had ESRD and was exposed to multiple doses of Gd , suggesting dosage-related toxicity or requirement of another co-factor (in addition to Gd ) to trigger NSF. Grobner (9) initially observed NSF in five patients with ESRD after Gd contrast exposure, a finding that subsequently was confirmed in another 13 patients with ESRD (10). The NFD/NSF registry data reveal that all patients with available data were exposed to Gd before the development of NSF (3). In a personal communication, Dr. Henrik Thomsen (Copenhagen University, Copenhagen, Denmark; December 12, 2006) noted that Gd -associated NSF has now been reported in most European countries, including Denmark, United Kingdom, Austria, Belgium, The Netherlands, Norway, Sweden, and Switzerland. Two recently published studies document Gd within tissues of five patients with NSF using scanning electron Published online ahead of print. Publication date available at www.cjasn.org.