Journal of the American Academy of Dermatology
Volume 62, Issue 1 , Pages 38-44, January 2010

Gadolinium deposition in nephrogenic systemic fibrosis: An examination of tissue using synchrotron x-ray fluorescence spectroscopy

  • Whitney A. High, MD, MEng

      Affiliations

    • Dermatology, University of Colorado Health Sciences Center, Denver, Colorado
    • Pathology, University of Colorado Health Sciences Center, Denver, Colorado
    • Corresponding Author InformationCorrespondence to: Whitney A. High, MD, MEng, Department of Dermatology, University of Colorado School of Medicine, PO Box 6510, Mail Stop F703, Aurora, CO 80045-0510.
    • ,
  • James F. Ranville, PhD

      Affiliations

    • Chemistry (Geochemistry), Colorado School of Mines, Golden, Colorado
    • ,
  • Mariah Brown, MD

      Affiliations

    • Dermatology, University of Colorado Health Sciences Center, Denver, Colorado
    • ,
  • Tracy Punshon, PhD

      Affiliations

    • Biology, Dartmouth College, Hanover, New Hampshire
    • ,
  • Antonio Lanzirotti, PhD

      Affiliations

    • Center for Advanced Radiation Sources (CARS), University of Chicago, Chicago, Illinois
    • ,
  • Brian P. Jackson, PhD

      Affiliations

    • Chemistry and Earth Sciences, Dartmouth College, Hanover, New Hampshire

Accepted 23 July 2009. published online 09 November 2009.

Article Outline

Background

Nephrogenic systemic fibrosis is a fibrosing disorder associated with gadolinium (Gd)-based contrast agents dosed during renal insufficiency.

Objective

In two patients, Gd deposition in tissue affected by nephrogenic systemic fibrosis was quantified using inductively coupled plasma mass spectrometry. The presence of Gd was confirmed and mapped using synchrotron x-ray fluorescence spectroscopy.

Results

Affected skin and soft tissue from the lower extremity demonstrated 89 and 209 ppm (μg/g, dry weight, formalin fixed) in cases 1 and 2, respectively. In case 2, the same skin and soft tissue was retested after paraffin embedding, with the fat content removed by xylene washes, and this resulted in a measured value of 189 ppm (μg/g, dry weight, paraffin embedded). Synchrotron x-ray fluorescence spectroscopy confirmed Gd in the affected tissue of both cases, and provided high-sensitivity and high-resolution spatial mapping of Gd deposition. A gradient of Gd deposition in tissue correlated with fibrosis and cellularity. Gd deposited in periadnexal locations within the skin, including hair and eccrine ducts, where it colocalized to areas of high calcium and zinc content.

Limitations

Because of the difficulty in obtaining synchrotron x-ray fluorescence spectroscopy scans, tissue from only two patients were mapped. A single control with kidney disease and gadolinium-based contrast agent exposure did not contain Gd.

Conclusions

Gd content on a gravimetric basis was impacted by processing that removed fat and altered the dry weight of the specimens. Gradients of Gd deposition in tissue corresponded to fibrosis and cellularity. Adnexal deposition of Gd correlated with areas of high calcium and zinc content.

Key words: adnexa, calcium, clincopathologic correlation, gadolinium, hypercellularity, mapping, nephrogenic systemic fibrosis, quantification, synchrotron x-ray fluorescence spectroscopy, zinc

Abbreviations used: GBCA, gadolinium-based contrast agent, Gd, gadolinium, ICP-MS, inductively coupled plasma mass spectrometry, MR, magnetic resonance, MRI, magnetic resonance imaging, NSF, nephrogenic systemic fibrosis, SEM/EDS, electron microscopy and energy dispersive spectroscopy, SXRFS, synchrotron x-ray fluorescence spectroscopy

 

Capsule Summary

 


•Synchrotron x-ray fluorescence spectroscopy confirmed gadolinium (Gd) deposition within affected tissue of patients with nephrogenic systemic fibrosis.

•Relative Gd deposition was highest in areas of hypercellularity and fibrosis and the observed gradient correlated with the degree of fibrosis.

•Gd deposition in adnexal structures may relate to high local concentrations of calcium and zinc that may facilitate transmetallation and displacement of Gd from the chelating agent.

•Quantitative Gd assays may yield different results in formalin-fixed tissue versus processed and paraffin-embedded tissue.

•Gradients of Gd deposition within tissue must be considered in future research and by clinicians, particularly if quantitative Gd assays are more widely used.

Nephrogenic systemic fibrosis (NSF) is a fibrotic disorder of the skin and viscera that occurs in a subset of patients with renal insufficiency.1 Cutaneous and soft tissue manifestations include skin discoloration and thickening, joint contracture, muscle fibrosis and weakness, and generalized pain.2

In 2006, two observational studies from Europe first suggested an association between NSF and gadolinium (Gd)-based contrast agents (GBCAs) used in medical imaging.3, 4 Later that same year, using electron microscopy and energy dispersive spectroscopy (SEM/EDS), Gd deposition was identified in the affected skin of patients with NSF.5, 6 Electron microscopy has repeatedly demonstrated agglomerations containing Gd within cells, surrounding adnexal structures, and intercalated among collagen bundles of the dermis, subcutis, or both.5, 6, 7, 8, 9 Often deposition of other metals, including iron and calcium, colocalized to these same areas. In 2007, inductively coupled plasma mass spectrometry (ICP-MS) was first used to quantify the amount of Gd in the skin affected by NSF.10 This technique has also been used to quantify Gd in the serum and hair of patients with NSF.11

Although certainly NSF is a systemic disease, we report two cases of NSF with particular pedagogic value for dermatology and dermatopathology. ICP-MS was used to quantify the amount of Gd within affected soft tissue for both cases, and synchrotron x-ray fluorescence spectroscopy (SXRFS) was used to confirm, via a second technology, the presence of this metal, and to provide highly sensitive spatial mapping of relative concentration gradients within tissue. Two-dimensional maps of Gd distribution within tissue were compared with and overlaid on conventional histologic images.

Back to Article Outline

Case presentations 

Case 1 

A 66-year-old woman developed chronic kidney disease in 2005 from poorly controlled diabetes. She began hemodialysis in late 2005, but had switched to peritoneal dialysis by February 2006. In December 2005, and twice in June 2006, including the time she was on peritoneal dialysis, the patient underwent magnetic resonance (MR) imaging (MRI) using GBCAs to evaluate various neurologic symptoms. It is unknown if additional dialysis of any kind followed these contrast administrations. The latter two MRIs used gadodiamide (Omniscan, GE Healthcare, Buckinghamshire, UK) at an unknown dosage. In November 2006, she first reported skin tightening, joint contractures, and pain of the lower extremities. In December 2006, she received a MR angiogram, again with a GBCA, but the particular brand of contrast and supplied dose could not be ascertained (off-label use of a GBCA in MR angiogram often involves up to 3-fold the dose of a standard MRI with contrast).

At the time of her death, in June 2007, the patient's arms and joints of the hands were also affected. Her pain was substantial. She required high doses of narcotic analgesics, even at rest. A biopsy specimen from the affected skin of the lower aspect of her right leg demonstrated fibrosis and a hypercellular dermis. Only Factor XIIIa immunohistochemical staining was performed at the time, but it was diffusely positive in the areas hypercellularity. Staining with Alcian blue demonstrated mucin accumulation. A final diagnosis of NSF was rendered at the tertiary care facility. Release of the original tissue sample was declined as per the policy of the institution, but an autopsy specimen from a similar area of skin was referred for testing upon demise.

Case 2 

An 83-year-old man developed chronic kidney disease in 1992 from uncontrolled hypertension. In January 2005, he underwent a MR angiogram and, shortly thereafter, a MRI, for “arterial investigation” and “shoulder pain,” respectively. His creatinine level at the time of exposure to the GBCA was unknown, but he was on hemodialysis 3 times per week at the time of imaging. In the weeks following, the patient developed the gradual and insidious onset of skin tightening and pain, chiefly of the lower extremities. He could no longer walk and became wheelchair bound. A biopsy specimen from the right thigh demonstrated fibrosis and hypercellularity of the reticular dermis and subcutis, with widening of the subcuticular septae. Immunohistochemical staining demonstrated that many of the infiltrating cells were CD34+ and procollagen I positive. An increased number of Factor XIIIa-positive cells were also observed. Colloidal iron staining demonstrated mucin accumulation. A final diagnosis of NSF was made at a tertiary care facility in Massachusetts.

Back to Article Outline

Methods 

Inductively coupled plasma mass spectrometry 

To quantify Gd in the skin specimens, ICP-MS analysis was performed using published methods.10 In brief, multiple 30-μm sections were cut from tissue blocks and deparaffinized using twice-distilled xylene and ethanol. After drying, samples were weighed on a shielded balance and placed into digestion vessels for digestion with trace-metal grade nitric acid to fully oxidize all organic material. The fully digested samples were then analyzed using ICP-MS on a Perkin-Elmer 6100 DRC Plus instrument (Perkin-Elmer Life & Analytical Sciences Inc, Wellesley, MA). Total Gd was monitored at dual masses of 158 and 160; an equivalence in concentration between the two masses confirmed that no polyatomic interferences were present during the analysis.

Synchrotron x-ray fluorescence spectroscopy 

SXRFS was used to map the spatial distribution of Gd within the skin specimens. The X26A beamline of the National Synchrotron Light Source at Brookhaven National Laboratory (Upton, NY) was used. X26A is an x-ray microprobe with a focused beam and approximate beam diameter of 10 μm. High-energy x-rays were directed at 50 μm–thick paraffin-embedded sections cut onto pure quartz wafers. Energy dispersive data were collected using a Canberra 9-element germanium array detector (Canberra Industries, Meriden, CT). For mapping, the specimen was rasterized through the focused beam to measure a field 60-μm wide spanning the entire depth of the section for the large pieces of skin and soft tissue (Fig 1, Fig 2), and to measure an area of 600-μm x 600-μm for the examination of the adnexal structures (Fig 3).

  • View full-size image.
  • Fig 1. 

    Mapping of relative concentrations of gadolinium longitudinally through a biopsy specimen of skin and soft tissue using synchrotron x-ray fluorescence spectroscopy revealed greatest deposition to occur in areas of fibrosis and hypercellularity, including within widened subcuticular septae.

  • View full-size image.
  • Fig 2. 

    Mapping of relative concentrations of gadolinium longitudinally in a biopsy specimen taken from patient with a more deep-seated variant of nephrogenic systemic fibrosis revealed greatest deposition to correspond to areas of hypercellularity and fibrosis.

  • View full-size image.
  • Fig 3. 

    Two-dimensional mapping around an eccrine gland and duct (A) demonstrated gadolinium deposition (B) to colocalize to areas of calcium (C) and zinc (D) deposition.

Data regarding the relative Gd content (measured in counts/s) were tabulated and averaged across cross-sectional incremental areas and were converted into bins of relative intensity that were superimposed on photomicrographs of standard histologic sections stained with hematoxylin and eosin (taken immediately before those sections tested by SXRFS).

Back to Article Outline

Results 

Inductively coupled plasma mass spectrometry 

For Case 1, ICP-MS was performed on both formalin-fixed and paraffin-embedded tissue, and demonstrated 89 and 198 ppm (ug/g) Gd, respectively. For Case 2, ICP-MS performed upon the paraffin-embedded tissue taken from the right thigh demonstrated 209 ppm (ug/g) Gd.

Synchrotron x-ray fluorescence spectroscopy 

Analysis of the relative deposition of Gd within tissue yielded striking visual correlations:

•Case 1. The relative quantity of Gd in tissue correlated with the bands of fibrosis within the dermis and the septae of the subcutis (Fig 1). We also mapped the deposition surrounding adnexal structures, including a sweat gland/duct (Fig 3).

•Case 2. The relative quantity of Gd in the dermis was focused within the lower half of the dermis, the same area where the fibrosis and hypercellularity of the tissue was most prominent (Fig 2).

Control tissue 

Control tissue from a skin cancer extirpation occurring in a 70-year-old man with a history of chronic kidney disease (creatinine = 1.7) and 3 prior exposures to GBCAs, but without evidence of NSF, failed to demonstrate appreciable Gd deposition by ICP-MS or by SXRFS. The pure quartz wafers themselves also contained no Gd as evidenced by SXRFS.

Back to Article Outline

Discussion 

Although the exact physiologic pathways leading from Gd deposition to the fibrosis of NSF have not yet been characterized, the association of GBCAs dosed during periods of renal dysfunction and development of NSF has been convincingly demonstrated by multiple investigators.3, 4, 12, 13, 14, 15, 16, 17 In fact, some researchers have hypothesized an early link between Gd deposition and the activation of certain profibrotic enzymes, such as tissue transglutaminases.18 Our own research group was the first to document Gd deposition within skin affected by NSF using SEM/EDS,5 ICP-MS,10 and now, SXRFS. It would appear the demonstration of Gd via multiple analytical techniques has firmly established, albeit largely circumstantially, the concept of transmetallation (in vivo release of Gd from the chelator) under some physiologic conditions.

The concept of in vivo transmetallation of GBCAs used in medical imaging was generally disfavored until the association between Gd and NSF was discovered. Indeed, the assumption that transmetallation does not occur to any significant extent under all physiologic conditions must be critically re-examined. Even if one considers the association with NSF dubious, an increasingly difficult proposition, the accumulation of Gd in human tissue, as verified by multiple analytical techniques and independent investigators, is troubling in its own right.

SXRFS allows for extraordinarily sensitive mapping of metal deposition in tissue. To our knowledge, this technique, which uses the energy of a subatomic particle accelerator, has not been yet been used in a clinical investigation within dermatopathology. With widely reported detection thresholds ranging from 0.1 to 5 ppm,19, 20 SXRFS is superior to mapping by most other techniques, such as SEM/EDS.21

Although generalizations arising from just these two cases should not be overstated, it was strikingly apparent that gradients of relative deposition of Gd in tissue correlated with fibrosis and hypercellularity, two important histopathologic criteria relied on in the microscopic diagnosis of NSF. This finding is supportive also of an earlier hypothesis, namely, that the histopathologic changes of NSF may be proportional to the retained dose of Gd.5 It is also in agreement with similar observations conducted using SEM/EDS.8, 9 Assuming further that the retained dose of Gd is reflective, generally, of the total exposed dose, this finding is supported by the work of other researchers using case-control data regarding NSF,22 and from details of an industry-controlled rat model of toxicity.23

Admittedly, tissue density in the septae is greater than that of the subcuticular tissue, and this could possibly explain some of the deposition gradients observed in Fig 1; however, our independent measurement of Gd by mass spectrometry also demonstrated an effective “concentration” of the specimen when the fatty content was removed by tissue processing, thereby providing further corroborating evidence to suggest that Gd deposition in fat is likely minimal compared with that of fibrous tissue.

Adnexal deposition observed during 2-dimensional mapping is intriguing in two regards. First, the deposition of Gd in proximity to the eccrine gland/duct (Fig 3) may support further the concept of transmetallation, as earlier unrelated studies demonstrated increased calcium and phosphorous excretion in the sweat of patients with renal insufficiency.24 These elements are suspected coparticipants in the transmetallation and/or precipitation reactions envisioned in the mechanism of tissue deposition of Gd. Indeed, the colocalization of calcium and phosphorous has been noted in other areas of the skin affected by Gd deposition in NSF.8, 9 Second, deposition of Gd in perifollicular areas may be clinically important to the observations of hair loss in NSF,25 and Gd deposition has been documented in hair itself.11 It is well known that zinc and copper are contained in relatively high concentrations in hair and are likely transferred to the hair shaft from the dermal papillae or follicular epithelium.26 Possessing higher thermodynamic stability constants than calcium for all the chelates used in GBCAs,27 zinc and/or copper might serve as an even stronger participant than calcium in any transmetallation reactions that displace Gd from the chelating agent. In our own review of the literature, while mentioned by others,6 we believe periadnexal deposition has perhaps been underappreciated, and merits further investigation.

Finally, applying the observations of this Gd deposition in tissue to the clinical practice of dermatology yields important lessons. Had the sampling of tissue been more superficial, in either case, but particularly in case 2, not only would the characteristic fibrosis and hypercellularity be missed, but also the Gd content of the skin would be underappreciated. One of us (W. A. H.), a practicing dermatologist and dermatopathologist, has cautioned widely against the use of superficial sampling techniques in confirming or refuting a diagnosis of NSF.28

As additional research is performed, the heterogeneous deposition of Gd within tissues must be considered before making conclusions regarding accumulation or diminution. Furthermore, although the deposition of Gd is heterogeneous it is certainly not random, and there is strong evidence that perivascular and periadnexal deposition is observed. When studying trends, or comparing Gd deposition in tissue, considerations of anatomic location, biopsy depth, tissue vascularity, tissue density, and adnexal content of the tissue must be considered to arrive at reasonable and reliable conclusions. Ever cognizant of the constraints of retrospective investigation, assessments that do not standardize for anatomic location and biopsy technique, or analytical techniques that do not generously sample the tissue, may lead to erroneous or biased conclusions. Histologic confirmation of a diagnosis of NSF will, of course, depend on procurement of a representative sample; but any conclusions regarding Gd content of the tissue will be impacted not only by this consideration, but also by the reliability, reproducibility, and detection thresholds of the analytical technique used.

It is not surprising that the measurement of Gd by ICP-MS ranged from 89 to 198 ppm (μg/g) in formalin-fixed (unprocessed) tissue versus paraffin-embedded tissue, respectively. Standard xylene-based tissue processing removes the fat content of the subcutis. It is reasonable to assume that a reduction in mass that has low Gd content would concentrate the sample: exactly what was observed in the skin samples. Tissues without significant fatty content, altered by xylene-based processing, would likely have a lesser discrepancy between the measurements of Gd content for formalin-fixed versus paraffin-embedded specimens.

In conclusion, deposition of Gd in tissue affected by NSF has been confirmed with a new technology, SXRFS. High sensitivity 2-dimensional mapping with SXRFS documented Gd gradients within tissue. Areas of increased cellularity and dermal fibrosis appear to manifest with higher relative amounts of Gd in tissue. Adnexal deposition of Gd was observed within sweat glands/ducts and hair follicles; this may be telling of the mechanisms of deposition, the participants in any transmetallation reactions (calcium and/or zinc), and perhaps, of some alternative mechanisms of excretion for GBCAs when renal function is marginal and drug half-lives are prolonged. Although tissue from only two cases was studied herein, our report provides a valuable proof of concept for use of this technology not only to study Gd deposition in NSF, but also for studying metal deposition as it pertains to other dermatologic ailments.

Back to Article Outline

References 

  1. Cowper SE, Robin HS, Steinberg SM, Su LD, Gupta S, LeBoit PE. Scleromyxedema-like cutaneous diseases in renal-dialysis patients. Lancet. 2000;356:1000–1001
  2. Cowper SE. Nephrogenic fibrosing dermopathy: the first 6 years. Curr Opin Rheumatol. 2003;15:785–790
  3. Grobner T. Gadolinium–a specific trigger for the development of nephrogenic fibrosing dermopathy and nephrogenic systemic fibrosis?. Nephrol Dial Transplant. 2006;21:1104–1108
  4. Marckmann P, Skov L, Rossen K, Dupont A, Damholt MB, Heaf JG, et al. Nephrogenic systemic fibrosis: suspected causative role of gadodiamide used for contrast-enhanced magnetic resonance imaging. J Am Soc Nephrol. 2006;17:2359–2362
  5. 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;56:21–26
  6. Boyd AS, Zic JA, Abraham JL. Gadolinium deposition in nephrogenic fibrosing dermopathy. J Am Acad Dermatol. 2007;56:27–30
  7. Thakral C, Alhariri J, Abraham JL. Long-term retention of gadolinium in tissues from nephrogenic systemic fibrosis patient after multiple gadolinium-enhanced MRI scans: case report and implications. Contrast Media Mol Imaging. 2007;2:199–205
  8. Thakral C, Abraham JL. Automated scanning electron microscopy and x-ray microanalysis for in situ quantification of gadolinium deposits in skin. J Electron Microsc (Tokyo). 2007;56:181–187
  9. Abraham JL, Thakral C, Skov L, Rossen K, Marckmann P. Dermal inorganic gadolinium concentrations: evidence for in vivo transmetallation and long-term persistence in nephrogenic systemic fibrosis. Br J Dermatol. 2008;158:273–280
  10. High WA, Ayers RA, Cowper SE. Gadolinium is quantifiable within the tissue of patients with nephrogenic systemic fibrosis. J Am Acad Dermatol. 2007;56:710–712
  11. Saussereau E, Lacroix C, Cattaneo A, Mahieu L, Goulle JP. Hair and fingernail gadolinium ICP-MS contents in an overdose case associated with nephrogenic systemic fibrosis. Forensic Sci Int. 2008;176:54–57
  12. Broome DR, Girguis MS, Baron PW, Cottrell AC, Kjellin I, Kirk GA. Gadodiamide-associated nephrogenic systemic fibrosis: why radiologists should be concerned. AJR Am J Roentgenol. 2007;188:586–592
  13. Sadowski EA, Bennett LK, Chan MR, Wentland AL, Garrett AL, Garrett RW, et al. Nephrogenic systemic fibrosis: risk factors and incidence estimation. Radiology. 2007;243:148–157
  14. Khurana A, Runge VM, Narayanan M, Greene JF, Nickel AE. Nephrogenic systemic fibrosis: a review of 6 cases temporally related to gadodiamide injection (Omniscan). Invest Radiol. 2007;42:139–145
  15. Centers for Disease Control and Prevention (CDC). Nephrogenic fibrosing dermopathy associated with exposure to gadolinium-containing contrast agents–St Louis, Missouri, 2002-2006. MMWR Morb Mortal Wkly Rep. 2007;56:137–141
  16. Pryor JG, Poggioli G, Galaria N, Gust A, Robison J, Samie F, et al. Nephrogenic systemic fibrosis: a clinicopathologic study of six cases. J Am Acad Dermatol. 2007;57:105–111
  17. Khurana A, Greene JF, High WA. Quantification of gadolinium in nephrogenic systemic fibrosis: re-examination of a reported cohort with analysis of clinical factors. J Am Acad Dermatol. 2008;59:218–224
  18. Parsons AC, Yosipovitch G, Sheehan DJ, Sangüeza OP, Greenberg CS, Sane DC. Transglutaminases: the missing link in nephrogenic systemic fibrosis. Am J Dermatopathol. 2007;29:433–436
  19. Flinn JM, Hunter D, Linkous DH, Lanzirotti A, Smith LN, Brightwell J, et al. Enhanced zinc consumption causes memory deficits and increased brain levels of zinc. Physiol Behav. 2005;83:793–803
  20. Miller LM, Wang Q, Telivala T, Smith RJ, Lanzirotti A, Miklossy J. Synchrotron-based infrared and X-ray imaging shows focalized accumulation of Cu and Zn co-localized with beta-amyloid deposits in Alzheimer's disease. J Struct Biol. 2006;155:30–37
  21. Lobinski R, Moulin C, Ortega R. Imaging and speciation of trace elements in biological environment. Biochimie. 2006;88:1591–1604
  22. Marckmann P, Skov L, Rossen K, Heaf JG, Thomsen HS. Case-control study of gadodiamide-related nephrogenic systemic fibrosis. Nephrol Dial Transplant. 2007;22:3174–3178
  23. Sieber MA, Pietsch H, Walter J, Haider W, Frenzel T, Weinmann HJ. A preclinical study to investigate the development of nephrogenic systemic fibrosis: a possible role for gadolinium-based contrast media. Invest Radiol. 2008;43:65–75
  24. Prompt CA, Quinton PM, Kleeman CR. High concentration of sweat calcium, magnesium and phosphate in chronic renal failure. Nephron. 1978;20:4–9
  25. Marckmann P, Skov L, Rossen K, Thomsen HS. Clinical manifestation of gadodiamide-related nephrogenic systemic fibrosis. Clin Nephrol. 2008;69:161–168
  26. Kempson IM, Skinner WM, Kirkbride KP. The occurrence and incorporation of copper and zinc in hair and their potential role as bioindicators: a review. J Toxicol Environ Health. 2007;10:611–622
  27. Idée JM, Port M, Raynal I, Schaefer M, Le Greneur S, Corot C. Clinical and biological consequences of transmetallation induced by contrast agents for magnetic resonance imaging: a review. Fundam Clin Pharmacol. 2006;20:563–576
  28. High WA. Nephrogenic systemic fibrosis and gadolinium-based contrast agents. Exp Rev Dermatol. 2007;2:593–605

 Funding for Dr Brown's participation and for materials used in the study was provided by a mentoring grant from the Women's Dermatology Society. Funding for Dr High's participation in the study was provided by a Career Development Award in Medical Dermatology from the Dermatology Foundation. The X26A beamline is supported by the United States Department of Energy (Geosciences, Grant DE-FG02-92ER14244). No additional funding was needed or accepted.

 Conflicts of interest: None declared.

 Reprints not available from the authors.

PII: S0190-9622(09)00933-5

doi:10.1016/j.jaad.2009.07.018

Journal of the American Academy of Dermatology
Volume 62, Issue 1 , Pages 38-44, January 2010