Gold Nephropathy in Juvenile Rheumatoid Arthritis Fred E. Husserl, MD, Stanton E. Shuler, MD \s=b\ A 2-year-old girl was treated with gold salts for juvenile rheumatoid arthritis. Treatment had to be discontinued when persistent proteinuria was detected. As this case report indicates, close monitoring of the urine is mandatory during treatment with gold salts to detect early signs of toxicity: hematuria followed by casts and then proteinuria as therapy is continued. Histologic examination with electron microscopy will help to differentiate the different forms of gold toxicity. When the findings are consistent with gold-induced renal involvement, therapy should be discontinued. The gold nephropathy usually resolves in time, with no permanent renal damage. (Am J Dis Child 133:50-52, 1979)
has been implicated in of a variety of the renal disorders of which hematuria, proteinuria, and the nephrotic syn¬ drome appear to be the most fre¬ quent.14 These renal disorders have been extensively studied and many pathophysiological mechanisms have been postulated in adults; however, discussion of such renal disorders occurring in children is relatively sparse. We therefore report the clini¬ cal course and pathologic findings of a renal biopsy specimen from a child who had received gold therapy.
Gold therapy production
REPORT OF A CASE A 16-month-old girl was first seen by her pediatrician in April 1969 with a painful right knee, low-grade fever, and an
infected throat for which "antibiotics" were prescribed. These symptoms disap¬ peared; over the ensuing six months, she remained well. In October, persistent fever (39.4 °C) associated with pain in the left wrist and a recurring rash localized to her thorax, abdomen, and lower extremities developed. She was referred to Ochsner From the Department of Internal Medicine, Section of Nephrology (Dr Husserl), and the Departments of Pediatrics and Nuclear Medicine (Dr Shuler), Ochsner Medical Institutions, New Orleans. Reprint requests to Ochsner Clinic, 1514 Jefferson Hwy, New Orleans, LA 70121 (Dr
Shuler).
Medical Institutions in December 1969. On admission, the patient was normoten¬ sive and afebrile, but appeared to be chron¬ ically ill. A maculopapular, evanescent, erythematous rash was localized to the trunk and lower extremities. The left wrist was swollen and tender, but the other joints were normal. The eyes were normal and no lymphadenopathy or hepatospleno¬ megaly were found. The hemoglobin level was 11 g/dl; hema¬ tocrit value, 34%; mean corpuscular volume, 75 cu µ; and a mean corpuscular hemoglobin concentration, 34.4%. The WBC count was 19,700/cu mm, with 71% polymorphonu¬ clear leukocytes, 26% lymphocytes, 2% eo¬ sinophils, and 1% band forms. Urinalysis was normal. The rheumatoid factor was negative, and the ESR was 40 mm/hr. Roentgenograms of the joints were nor¬ mal, but a sodium pertechnetate Tc 99m joint scan demonstrated increased uptake in the left wrist and both ankles, indicating synovitis in these areas. Aspirin therapy was started in December with relatively good initial response, but by May 1970, fever recurred and several joints became painful. The hemoglobin level was 7.5 g/dl; the WBC count, 9,500/cu mm, with a normal differential count. Urinalysis was normal and the BUN level was 10 mg/dl. The ESR was 83 mm/hr. Lupus erythema¬ tosus preparation and antinuclear factor were negative. Prednisone, 30 mg/day, brought improvement of her symptoms. The dose was slowly reduced to a mainte¬ nance dose of 12.5 mg every other day. She did well on this regimen for the next year. Gradually, however, she again became symptomatic and required increasing doses of prednisone. In October 1974, a posterior capsular cataract was found. Urinalysis again was normal, and 24-hour urinary protein excretion was within normal limits. A month later, gold therapy was started with 2.5 mg of gold sodium thiomalate (Myochrysine) given intramus¬ cularly, the dose being gradually increased to 10 mg per week. By May 1975, she had received 217 mg of gold and was doing very well on 2 mg of prednisone every other day. Four months later, 12.5 mg of gold per week was started and was increased in two weeks to 15 mg twice weekly. After two months, in November 1975, treatment with gold had to be discontinued when persis¬ tent proteinuria was detected on repeated urinalyses. Because renal function was
Downloaded From: http://archpedi.jamanetwork.com/ by a University of Pennsylvania User on 06/21/2015
normal and proteinuria remained less than 300 mg/24 hr, gold therapy was reinsti¬ tuted in January 1976, using a reduced dosage of 5 mg per week. Proteinuria (600 to 1,200 mg/24 hr) recurred in June 1976,
and gold therapy was again stopped after a total dose of 520 mg. In August 1976, therapy with predni¬ sone, 10 mg four times a day, was started and continued for four weeks. The 24-hour urine protein excretion fell to 290 mg on this program. However, when the dose was changed to 40 mg every other day, urine protein excretion rose to 950 to 1,200 mg/24 hr. A renal biopsy was performed in Octo¬ ber 1976. A needle biopsy specimen of the kidney was examined by light and electron micros¬ copy and by immunofluorescence. For light microscopy, the tissue was fixed in a 3% buffered formaldehyde solution. A thick¬ ened basement membrane typical of mem¬ braneous glomerulonephritis was found on hemotoxylin-eosin and PAS stains, without mesangial hypercellularity. On the silvermethenamine preparation, subepithelial spikes were prominent (Fig 1). For electron microscopy, a 50% gluteraldehyde-10% formaldehyde and 40% buffer solution was used. The basement membrane was irregu¬ lar and thickened, containing numerous translucent areas and a few dense deposits, some of which seemed to be incorporated into the basement membrane (Fig 2). A few of the dense deposits were subepithe¬ lial and surrounded by basement mem¬ brane spikes. The foot processes showed partial fusion. No gold inclusions were found in the glomeruli or in the tubules. Direct immunofluorescence revealed a strongly granular pattern with anti-IgG along the glomerular basement membrane and a weaker but positive granular pattern with anti-IgA and complement (C3) in the same location. Following the renal biopsy, prednisone was continued on an alternate-day basis in slowly decreasing dosage, and the 24-hour urine protein excretion gradually dimin¬ ished. By December 1976, while she was taking prednisone, 20 mg every other day, the urine protein excretion had fallen to normal values and has remained so despite further reduction in corticosteroid dosage.
COMMENT The
of proteinuria in a with juvenile rheumatoid ar¬ thritis who has been treated with gold salts may pose a diagnostic dilemma. Clinically, it may be impossible to establish the exact cause of the proteinuria; however, histologie stud¬ ies with electron microscopy are usual¬ ly diagnostic. This point is important in order to establish the method of occurrence
patient
treatment of such
a
times will be helpful in
prognosis.
patient and at estimating the
Anttila and Laaksonen1 in 1969
reported from Finland that 7.4% or 47 of 638 juvenile rheumatoid arthritis patients had a "nephropathy." Of these, the nephropathy was mani¬ fested by proteinuria in 92% (occur¬ ring equally as frequently with gold treatment
as
with other forms of ther¬
apy), hematuria in 38% (slightly but not significantly more frequent in gold-treated patients), abnormal uri¬ nary sediment with leukocytes and casts, and abnormal renal function with a decrease in creatinine clear¬
in 23%. Of the 15 patients they studied by biopsy or autopsy, ten had amyloidosis, an incidence of 1.6% of the total group. Four of the others ance
were
reported as having severe neph-
rosclerosis (one), "glomerulonephritic" changes (one), and chronic pyelo¬
nephritis (two). In a subsequent report in 1972 by Anttila,6 42.5% of 165 patients with juvenile rheumatoid arthritis had pro¬
teinuria, 23% hematuria, and 25.5% leukocyturia; the creatinine clearance
Fig 1.—Presence methenamine,
of spikes in epithelial 1,250).
side of basement membrane
(arrow) (silver-
Fig 2.—Electron micrograph of capillary loop, showing thickened and irregular basement membrane (Bm) containing a few dense deposits (arrow) within translucent areas. There is partial fusion of epithelial foot processes ( 24,300).
was found to be persistently de¬ creased in 6.2%; the phenolsulfonphthalein excretion was below normal in 19.6%. Hematuria and leukocyturia were associated with gold therapy in 40%; the significant increase of phenolsulfonphthalein retention was also associated with gold therapy. Of the 60 biopsy or autopsy specimens in that series, 38.4% were significantly abnor¬
mal, showing glomerular changes (glomerulitis in 21.7%, thickening of the basement membrane in 6.7%, and capsular adhesions in 10%) and tubu¬ lointerstitial changes (tubular atrophy in 13.3%, round cell infiltration in 5%,
and increase in collagen tissue in 11.6%). These findings were more frequent in gold-treated patients, but the difference was not statistically significant. It remains speculative if electron microscopy would have shown a higher incidence of glomerular base¬ ment membrane disease. In a sepa¬ rate report of 52 patients with juve¬ nile arthritis treated with gold by Levinson et al,7 22% developed pro¬ teinuria and 15% hematuria. In Europe, amyloidosis has been found to be a frequent complication of juvenile rheumatoid arthritis"; for unknown reasons, it is rarely observed
Downloaded From: http://archpedi.jamanetwork.com/ by a University of Pennsylvania User on 06/21/2015
in America
or Australia. The main manifestation of this complication is proteinuria, and it is associated with death in 47% of the European and 13% of the United States cases." Gold has been implicated in the production of acute tubular necrosisand acute nephritis,3 but the most frequent finding is proteinuria fol¬ lowed by the nephrotic syndrome. Vaamonde and Hunt4 found 19 re¬
ported cases up to 1960 (all adults) of gold salt therapy associated with the nephrotic syndrome. Subsequent re¬ cent
reports of this association have
appeared in the pediatrie literature. The degree of proteinuria is not proportional to the gold concentration in blood or urine, the total dose of gold administered, or the duration of treat¬ ment,"1 but in
some cases, it is propor¬ tional to the single weekly dose given.'" The course is relatively benign, with disappearance of the proteinuria when the treatment is
discontinued. The pathologic picture is that of a membranous glomerulonephritis.1113 Because this is not always apparent on routine light microscopy, electron mi¬ croscopy is necessary for accurate diagnosis. Gold-induced membranous glomerulonephritis should be sug¬ gested by the existence of electrondense deposits located on the epithe¬ lial side of the basement membrane. Tornroth and Skrifvars11 established different stages in the pathologic evolution. Our patient's disease would be intermediate between stage III, in which the deposits are at times intramembranous, and stage IV, in which there is evidence of electron lucent
deposits. Perhaps (as seen in Fig 2), slow resolution is taking place.11 At the time the biopsy was performed, our patient had not received gold in
more than three months, renal func¬ tion was normal, and proteinuria had decreased to approximately 1 g/24 hr. Gold particles have not been found in these dense deposits by electron
microscopic or energy-dispersive roentgenographic analysis,1013 but they are commonly found within the proximal tubular epithelial cells. Ex¬ perimentally, in animals,14 gold parti¬ cles have been found to localize proximally first in the apical cytoplasm and brush border of the tubular epithe¬ lium, and subsequently to accumulate
in the cell cytosomes, uniformly in the cytoplasm of the cells. This location is apparently shared by other substances and represents a nonspecific way the cell handles foreign substances. In biopsy material, gold particles are found consistently in the proximal
tubules and interstitium.1"-13 Occasion¬ ally they have been found in the podocytes and mesangial cells10 or within the parietal and visceral epithelial cells of a glomerular tuft.12 Immunofluorescent studies gener¬ ally show strongly positive staining with anti-IgG and complement, usual¬ ly following a diffuse granular pat¬ tern, but sometimes there is a segmental distribution along the glo¬ merular tuft.1113 Anti-IgM and, rare¬ ly, IgA stain positively; anti-IgD and rheumatoid factor stain negatively.13 The close temporal relationship be¬ tween the onset of proteinuria and the use of gold, the positive immunofluo¬ rescence with anti-IgG and comple¬ ment, and the presence of subepithelial deposits point toward the deposi¬ tion of antigen-antibody complexes as the pathogenetic mechanism of the glomerular lesion. How gold triggers this mechanism remains obscure. It is doubtful that gold acts as a hapten because of its well-established ab¬ sence from these deposits. If gold were to induce the production of autoantibodies against glomerular basement membrane or against tubu¬ lar cells, the immunofluorescence would show a linear pattern and the deposits would be expected in a subendothelial location.15 A direct toxic effect of gold on the kidney has been postulated.1" In a few patients, pro¬ teinuria has been found to appear as increasing doses of gold are used and to disappear as these doses are lowered10; this did not occur in our patient. In one reported case,1" pro¬ teinuria was not found in a goldtreated patient although a kidney biopsy specimen demonstrated fused foot processes and gold localized in the usual pattern. Experimentally, in rats,10 gold nephropathy with typical glomerular changes has been induced by small doses (0.025 mg) of gold sodium thiomalate at weekly intervals, and a toxic, probably nonspecific, tubular lesion was caused by larger weekly doses (1 g) of gold sodium thiomalate. To some, this suggests the possibility
Downloaded From: http://archpedi.jamanetwork.com/ by a University of Pennsylvania User on 06/21/2015
of autoantibody-mediated glomerular lesion induced by subcellular dam¬
age.10
Gold therapy is a useful form of therapy for juvenile rheumatoid ar¬ thritis.7 Close monitoring of the urine is mandatory to detect early signs of toxicity: hematuria followed by casts and then proteinuria as gold therapy is continued.17 Proteinuria alone can be due to intrinsic renal involvement from juvenile rheumatoid arthritis, to gold toxicity, and, rarely in the United
States,
to
amyloidosis. Histologie
ex¬
amination with electron microscopy will help to differentiate these differ¬ ent forms. When the findings are consistent with gold-induced renal involvement, gold therapy should be discontinued. The gold nephropathy usually resolves in time with no permanent renal damage. References 1. Lee JC, Dushkin M, Eyring EJ, et al: Renal lesions associated with gold therapy; light and electron microscopic studies. Arthritis Rheum 8:1-13, 1965. 2. Derot M, Kahn J, Mazalton A, et al: Nephrite anurique aigue mortell apres traitement aurique, chrysocyanose associee. Bull Soc Med Hop Paris 70:234-239, 1954. 3. Mathers RG: Gold nephritis and dermatitis in pulmonary tuberculosis. Br Med J 1:223, 1945. 4. Vaamonde CA, Hunt FR: The nephrotic syndrome as a complication of gold therapy. Arthritis Rheum 13:826-834, 1970. 5. Anttila R, Laaksonen AL: Renal disease in juvenile rheumatoid arthritis. Acta Rheumatol Scand 15:99-111, 1969. 6. Anttila R: Renal involvement in juvenile rheumatoid arthritis: A clinical and histopathological study. Acta Paediatr Scand, suppl 227, 1972, pp 7-73. 7. Levinson JE, Balz GP, Bondi S: Gold therapy. Arthritis Rheum 20(suppl):531-535, 1977. 8. Schnitzer TJ, Ansell BM: Amyloidosis in juvenile chronic polyarthritis. Arthritis Rheum 20(suppl):245-252, 1977. 9. Baum J, Gutowska G: Death in juvenile Arthritis rheumatoid arthritis. Rheum 20(suppl):253-255, 1977. 10. Silverberg DS, Kidd EG, Shnitka TK, et al: Bold nephropathy: A clinical and pathologic study. Arthritis Rheum 13:812-825, 1970. 11. Tornroth T, Skrifvars B: Gold nephropathy prototype of membranous glomerulonephritis. Am J Pathol 75:573-590, 1974. 12. Watanabe I, Whittier FC, Moore J, et al: Gold nephropathy: Ultrastructural, fluorescence, and microanalytic studies of two patients. Arch Pathol Lab Med 100:632-635, 1976. 13. Katz A, Little AH: Gold nephropathy\p=m-\an immunopathologic study. Arch Pathol 96:133-136, 1973. 14. Ganote CE, Beaver DL, Moses HL: Renal gold inclusions: A light and electron microscopic study. Arch Pathol 81:429-438, 1966. 15. Dixon FJ: The pathogenesis of glomerulonephritis. Am J Med 44:493-498, 1968. 16. Nagi AM, Alexander F, Barabas AZ: Gold nephropathy in rats: Light and electron microscopic studies. Exp Mol Pathol 15:354-362, 1971. 17. Vance PW: Gold therapy and nephritis. Arthritis Rheum 20(suppl):548-549, 1977.
has been implicated in of a variety of the renal disorders of which hematuria, proteinuria, and the nephrotic syn¬ drome appear to be the most fre¬ quent.14 These renal disorders have been extensively studied and many pathophysiological mechanisms have been postulated in adults; however, discussion of such renal disorders occurring in children is relatively sparse. We therefore report the clini¬ cal course and pathologic findings of a renal biopsy specimen from a child who had received gold therapy.
Gold therapy production
REPORT OF A CASE A 16-month-old girl was first seen by her pediatrician in April 1969 with a painful right knee, low-grade fever, and an
infected throat for which "antibiotics" were prescribed. These symptoms disap¬ peared; over the ensuing six months, she remained well. In October, persistent fever (39.4 °C) associated with pain in the left wrist and a recurring rash localized to her thorax, abdomen, and lower extremities developed. She was referred to Ochsner From the Department of Internal Medicine, Section of Nephrology (Dr Husserl), and the Departments of Pediatrics and Nuclear Medicine (Dr Shuler), Ochsner Medical Institutions, New Orleans. Reprint requests to Ochsner Clinic, 1514 Jefferson Hwy, New Orleans, LA 70121 (Dr
Shuler).
Medical Institutions in December 1969. On admission, the patient was normoten¬ sive and afebrile, but appeared to be chron¬ ically ill. A maculopapular, evanescent, erythematous rash was localized to the trunk and lower extremities. The left wrist was swollen and tender, but the other joints were normal. The eyes were normal and no lymphadenopathy or hepatospleno¬ megaly were found. The hemoglobin level was 11 g/dl; hema¬ tocrit value, 34%; mean corpuscular volume, 75 cu µ; and a mean corpuscular hemoglobin concentration, 34.4%. The WBC count was 19,700/cu mm, with 71% polymorphonu¬ clear leukocytes, 26% lymphocytes, 2% eo¬ sinophils, and 1% band forms. Urinalysis was normal. The rheumatoid factor was negative, and the ESR was 40 mm/hr. Roentgenograms of the joints were nor¬ mal, but a sodium pertechnetate Tc 99m joint scan demonstrated increased uptake in the left wrist and both ankles, indicating synovitis in these areas. Aspirin therapy was started in December with relatively good initial response, but by May 1970, fever recurred and several joints became painful. The hemoglobin level was 7.5 g/dl; the WBC count, 9,500/cu mm, with a normal differential count. Urinalysis was normal and the BUN level was 10 mg/dl. The ESR was 83 mm/hr. Lupus erythema¬ tosus preparation and antinuclear factor were negative. Prednisone, 30 mg/day, brought improvement of her symptoms. The dose was slowly reduced to a mainte¬ nance dose of 12.5 mg every other day. She did well on this regimen for the next year. Gradually, however, she again became symptomatic and required increasing doses of prednisone. In October 1974, a posterior capsular cataract was found. Urinalysis again was normal, and 24-hour urinary protein excretion was within normal limits. A month later, gold therapy was started with 2.5 mg of gold sodium thiomalate (Myochrysine) given intramus¬ cularly, the dose being gradually increased to 10 mg per week. By May 1975, she had received 217 mg of gold and was doing very well on 2 mg of prednisone every other day. Four months later, 12.5 mg of gold per week was started and was increased in two weeks to 15 mg twice weekly. After two months, in November 1975, treatment with gold had to be discontinued when persis¬ tent proteinuria was detected on repeated urinalyses. Because renal function was
Downloaded From: http://archpedi.jamanetwork.com/ by a University of Pennsylvania User on 06/21/2015
normal and proteinuria remained less than 300 mg/24 hr, gold therapy was reinsti¬ tuted in January 1976, using a reduced dosage of 5 mg per week. Proteinuria (600 to 1,200 mg/24 hr) recurred in June 1976,
and gold therapy was again stopped after a total dose of 520 mg. In August 1976, therapy with predni¬ sone, 10 mg four times a day, was started and continued for four weeks. The 24-hour urine protein excretion fell to 290 mg on this program. However, when the dose was changed to 40 mg every other day, urine protein excretion rose to 950 to 1,200 mg/24 hr. A renal biopsy was performed in Octo¬ ber 1976. A needle biopsy specimen of the kidney was examined by light and electron micros¬ copy and by immunofluorescence. For light microscopy, the tissue was fixed in a 3% buffered formaldehyde solution. A thick¬ ened basement membrane typical of mem¬ braneous glomerulonephritis was found on hemotoxylin-eosin and PAS stains, without mesangial hypercellularity. On the silvermethenamine preparation, subepithelial spikes were prominent (Fig 1). For electron microscopy, a 50% gluteraldehyde-10% formaldehyde and 40% buffer solution was used. The basement membrane was irregu¬ lar and thickened, containing numerous translucent areas and a few dense deposits, some of which seemed to be incorporated into the basement membrane (Fig 2). A few of the dense deposits were subepithe¬ lial and surrounded by basement mem¬ brane spikes. The foot processes showed partial fusion. No gold inclusions were found in the glomeruli or in the tubules. Direct immunofluorescence revealed a strongly granular pattern with anti-IgG along the glomerular basement membrane and a weaker but positive granular pattern with anti-IgA and complement (C3) in the same location. Following the renal biopsy, prednisone was continued on an alternate-day basis in slowly decreasing dosage, and the 24-hour urine protein excretion gradually dimin¬ ished. By December 1976, while she was taking prednisone, 20 mg every other day, the urine protein excretion had fallen to normal values and has remained so despite further reduction in corticosteroid dosage.
COMMENT The
of proteinuria in a with juvenile rheumatoid ar¬ thritis who has been treated with gold salts may pose a diagnostic dilemma. Clinically, it may be impossible to establish the exact cause of the proteinuria; however, histologie stud¬ ies with electron microscopy are usual¬ ly diagnostic. This point is important in order to establish the method of occurrence
patient
treatment of such
a
times will be helpful in
prognosis.
patient and at estimating the
Anttila and Laaksonen1 in 1969
reported from Finland that 7.4% or 47 of 638 juvenile rheumatoid arthritis patients had a "nephropathy." Of these, the nephropathy was mani¬ fested by proteinuria in 92% (occur¬ ring equally as frequently with gold treatment
as
with other forms of ther¬
apy), hematuria in 38% (slightly but not significantly more frequent in gold-treated patients), abnormal uri¬ nary sediment with leukocytes and casts, and abnormal renal function with a decrease in creatinine clear¬
in 23%. Of the 15 patients they studied by biopsy or autopsy, ten had amyloidosis, an incidence of 1.6% of the total group. Four of the others ance
were
reported as having severe neph-
rosclerosis (one), "glomerulonephritic" changes (one), and chronic pyelo¬
nephritis (two). In a subsequent report in 1972 by Anttila,6 42.5% of 165 patients with juvenile rheumatoid arthritis had pro¬
teinuria, 23% hematuria, and 25.5% leukocyturia; the creatinine clearance
Fig 1.—Presence methenamine,
of spikes in epithelial 1,250).
side of basement membrane
(arrow) (silver-
Fig 2.—Electron micrograph of capillary loop, showing thickened and irregular basement membrane (Bm) containing a few dense deposits (arrow) within translucent areas. There is partial fusion of epithelial foot processes ( 24,300).
was found to be persistently de¬ creased in 6.2%; the phenolsulfonphthalein excretion was below normal in 19.6%. Hematuria and leukocyturia were associated with gold therapy in 40%; the significant increase of phenolsulfonphthalein retention was also associated with gold therapy. Of the 60 biopsy or autopsy specimens in that series, 38.4% were significantly abnor¬
mal, showing glomerular changes (glomerulitis in 21.7%, thickening of the basement membrane in 6.7%, and capsular adhesions in 10%) and tubu¬ lointerstitial changes (tubular atrophy in 13.3%, round cell infiltration in 5%,
and increase in collagen tissue in 11.6%). These findings were more frequent in gold-treated patients, but the difference was not statistically significant. It remains speculative if electron microscopy would have shown a higher incidence of glomerular base¬ ment membrane disease. In a sepa¬ rate report of 52 patients with juve¬ nile arthritis treated with gold by Levinson et al,7 22% developed pro¬ teinuria and 15% hematuria. In Europe, amyloidosis has been found to be a frequent complication of juvenile rheumatoid arthritis"; for unknown reasons, it is rarely observed
Downloaded From: http://archpedi.jamanetwork.com/ by a University of Pennsylvania User on 06/21/2015
in America
or Australia. The main manifestation of this complication is proteinuria, and it is associated with death in 47% of the European and 13% of the United States cases." Gold has been implicated in the production of acute tubular necrosisand acute nephritis,3 but the most frequent finding is proteinuria fol¬ lowed by the nephrotic syndrome. Vaamonde and Hunt4 found 19 re¬
ported cases up to 1960 (all adults) of gold salt therapy associated with the nephrotic syndrome. Subsequent re¬ cent
reports of this association have
appeared in the pediatrie literature. The degree of proteinuria is not proportional to the gold concentration in blood or urine, the total dose of gold administered, or the duration of treat¬ ment,"1 but in
some cases, it is propor¬ tional to the single weekly dose given.'" The course is relatively benign, with disappearance of the proteinuria when the treatment is
discontinued. The pathologic picture is that of a membranous glomerulonephritis.1113 Because this is not always apparent on routine light microscopy, electron mi¬ croscopy is necessary for accurate diagnosis. Gold-induced membranous glomerulonephritis should be sug¬ gested by the existence of electrondense deposits located on the epithe¬ lial side of the basement membrane. Tornroth and Skrifvars11 established different stages in the pathologic evolution. Our patient's disease would be intermediate between stage III, in which the deposits are at times intramembranous, and stage IV, in which there is evidence of electron lucent
deposits. Perhaps (as seen in Fig 2), slow resolution is taking place.11 At the time the biopsy was performed, our patient had not received gold in
more than three months, renal func¬ tion was normal, and proteinuria had decreased to approximately 1 g/24 hr. Gold particles have not been found in these dense deposits by electron
microscopic or energy-dispersive roentgenographic analysis,1013 but they are commonly found within the proximal tubular epithelial cells. Ex¬ perimentally, in animals,14 gold parti¬ cles have been found to localize proximally first in the apical cytoplasm and brush border of the tubular epithe¬ lium, and subsequently to accumulate
in the cell cytosomes, uniformly in the cytoplasm of the cells. This location is apparently shared by other substances and represents a nonspecific way the cell handles foreign substances. In biopsy material, gold particles are found consistently in the proximal
tubules and interstitium.1"-13 Occasion¬ ally they have been found in the podocytes and mesangial cells10 or within the parietal and visceral epithelial cells of a glomerular tuft.12 Immunofluorescent studies gener¬ ally show strongly positive staining with anti-IgG and complement, usual¬ ly following a diffuse granular pat¬ tern, but sometimes there is a segmental distribution along the glo¬ merular tuft.1113 Anti-IgM and, rare¬ ly, IgA stain positively; anti-IgD and rheumatoid factor stain negatively.13 The close temporal relationship be¬ tween the onset of proteinuria and the use of gold, the positive immunofluo¬ rescence with anti-IgG and comple¬ ment, and the presence of subepithelial deposits point toward the deposi¬ tion of antigen-antibody complexes as the pathogenetic mechanism of the glomerular lesion. How gold triggers this mechanism remains obscure. It is doubtful that gold acts as a hapten because of its well-established ab¬ sence from these deposits. If gold were to induce the production of autoantibodies against glomerular basement membrane or against tubu¬ lar cells, the immunofluorescence would show a linear pattern and the deposits would be expected in a subendothelial location.15 A direct toxic effect of gold on the kidney has been postulated.1" In a few patients, pro¬ teinuria has been found to appear as increasing doses of gold are used and to disappear as these doses are lowered10; this did not occur in our patient. In one reported case,1" pro¬ teinuria was not found in a goldtreated patient although a kidney biopsy specimen demonstrated fused foot processes and gold localized in the usual pattern. Experimentally, in rats,10 gold nephropathy with typical glomerular changes has been induced by small doses (0.025 mg) of gold sodium thiomalate at weekly intervals, and a toxic, probably nonspecific, tubular lesion was caused by larger weekly doses (1 g) of gold sodium thiomalate. To some, this suggests the possibility
Downloaded From: http://archpedi.jamanetwork.com/ by a University of Pennsylvania User on 06/21/2015
of autoantibody-mediated glomerular lesion induced by subcellular dam¬
age.10
Gold therapy is a useful form of therapy for juvenile rheumatoid ar¬ thritis.7 Close monitoring of the urine is mandatory to detect early signs of toxicity: hematuria followed by casts and then proteinuria as gold therapy is continued.17 Proteinuria alone can be due to intrinsic renal involvement from juvenile rheumatoid arthritis, to gold toxicity, and, rarely in the United
States,
to
amyloidosis. Histologie
ex¬
amination with electron microscopy will help to differentiate these differ¬ ent forms. When the findings are consistent with gold-induced renal involvement, gold therapy should be discontinued. The gold nephropathy usually resolves in time with no permanent renal damage. References 1. Lee JC, Dushkin M, Eyring EJ, et al: Renal lesions associated with gold therapy; light and electron microscopic studies. Arthritis Rheum 8:1-13, 1965. 2. Derot M, Kahn J, Mazalton A, et al: Nephrite anurique aigue mortell apres traitement aurique, chrysocyanose associee. Bull Soc Med Hop Paris 70:234-239, 1954. 3. Mathers RG: Gold nephritis and dermatitis in pulmonary tuberculosis. Br Med J 1:223, 1945. 4. Vaamonde CA, Hunt FR: The nephrotic syndrome as a complication of gold therapy. Arthritis Rheum 13:826-834, 1970. 5. Anttila R, Laaksonen AL: Renal disease in juvenile rheumatoid arthritis. Acta Rheumatol Scand 15:99-111, 1969. 6. Anttila R: Renal involvement in juvenile rheumatoid arthritis: A clinical and histopathological study. Acta Paediatr Scand, suppl 227, 1972, pp 7-73. 7. Levinson JE, Balz GP, Bondi S: Gold therapy. Arthritis Rheum 20(suppl):531-535, 1977. 8. Schnitzer TJ, Ansell BM: Amyloidosis in juvenile chronic polyarthritis. Arthritis Rheum 20(suppl):245-252, 1977. 9. Baum J, Gutowska G: Death in juvenile Arthritis rheumatoid arthritis. Rheum 20(suppl):253-255, 1977. 10. Silverberg DS, Kidd EG, Shnitka TK, et al: Bold nephropathy: A clinical and pathologic study. Arthritis Rheum 13:812-825, 1970. 11. Tornroth T, Skrifvars B: Gold nephropathy prototype of membranous glomerulonephritis. Am J Pathol 75:573-590, 1974. 12. Watanabe I, Whittier FC, Moore J, et al: Gold nephropathy: Ultrastructural, fluorescence, and microanalytic studies of two patients. Arch Pathol Lab Med 100:632-635, 1976. 13. Katz A, Little AH: Gold nephropathy\p=m-\an immunopathologic study. Arch Pathol 96:133-136, 1973. 14. Ganote CE, Beaver DL, Moses HL: Renal gold inclusions: A light and electron microscopic study. Arch Pathol 81:429-438, 1966. 15. Dixon FJ: The pathogenesis of glomerulonephritis. Am J Med 44:493-498, 1968. 16. Nagi AM, Alexander F, Barabas AZ: Gold nephropathy in rats: Light and electron microscopic studies. Exp Mol Pathol 15:354-362, 1971. 17. Vance PW: Gold therapy and nephritis. Arthritis Rheum 20(suppl):548-549, 1977.
