- Email: [email protected]
Eosinophiluria in atheroembolic renal disease
Eosinophiluria in Atheroembolic Renal Disease DAVIDM. WILSON, M.D., THOMASL. SALAZER,M.D., MICHAELE. FARKOUH,M.D., Rochester, Minnesota
PURPOSE AND PATIENTS: %sinophikia has been reported in acute interstitial nephritis and other renal diseases, but its presence in atheroembolic renal disease (AJIRD) has not been previously established. ARRD has been identified as a cause of acute and chronic renal failure, particularly in elderly patients with advanced atherosclerosis and in those patients who have undergone manipulation or intervention of the abdominal aorta, renal artery, or coronary artery. The deftitive diagnosis is made by renal biopsy. However, many patients are too acutely ill to tolerate renal biopsy and, in recent years, peripheral eosinophilia, hypocomplementemia, and thrombocytopenia have been recognized in association with ARRD. Previous studies have reported that AJIRD is associated with an inactive renal sediment and an absence of urine eosinophils. We reviewed our experience over a 4year period with 24 patients with renal biopsy-proven ARRD. RESULTS: uriIM3 eosinophils were eVahUtted in nine patients to help determine the cause of their renal deterioration. Seven of these patients presented with evidence of vascular disease. Three patients had procedures involving manipulation of the abdominal aorta Physical examination revealed findings of atheroembolism in three of nine patients. Overall, eight of nine patients had a positive Hansel’s stain for eosinophiluria. Six of eight patients had more than 5% of their urinary white cell count as eosinophils. The reason for failure of previous studies to detect eosinophiluria in ARRD is uuclear but may have been related to the use of Wright’s stain instead of Hansel’s stain. CONCLUSION: h the evaluation of acute red insufficiency, eosinophihuia may indicate ARRD in addition to the other hnown causes for this fmding.
From the Department of Medicine, Mayo Clinic-Mayo Foundation, 200 First Street Southwest, Rochester, Minnesota. Requests for reprints should be addressed to David M. Wilson, M.D.,
Manuscript
186
submitted
August 1991
November
29, 1990, and accepted
in revised
The AmericanJournal of Medicine
A
theroembolic renal disease (AERD) has been identified as a cause of acute and chronic renal failure [l], particularly in elderly patients [2,3] with advanced atherosclerosis and those patients who have undergone manipulation or intervention of the abdominal aorta [4,5], renal artery [6], or coronary artery [7,8]. Renal atheroembolism has been reported in 15% of patients with severe aortic atherosclerosis and 31% of patients with large aortic aneurysms [4]. Pathologically, the essential lesion is an extensive occlusion, most notable in the lumens of vessels 50 to 200 p in diameter, with arcuate, intralobular, and terminal arteries most frequently affected [9,10]. The definitive diagnosis of AERD depends on renal biopsy. Clinically, the diagnosis has been difficult to make. AERD is sometimes associated with emboli to the brain, pancreas, thyroid, gastrointestinal tract, and extremities [ll-151. Often, there are few physical findings. The outcome for patients with AERD is variable, and the condition does not always progress to end-stage renal disease [16]. In recent years, peripheral eosinophilia [17,18], hypocomplementemia, and thrombocytopenia [19] have been reported in association with AERD. We noted eosinophiluria in one of our biopsy-proven cases, which led to a review of our experience. Eosinophiluria secondary to a drug-induced hypersensitivity reaction was first reported by Eisenstaedt [20] in 1951. Helgason and Lundqvist [21] reported on its significance in a broader spectrum of renal disease. In 1978, Galpin et al [22] reported on nine patients with methicillin-induced, acute interstitial nephritis with eosinophiluria ranging from 10% to 60%. Further studies also showed a correlation between urine eosinophils and acute interstitial nephritis [23,24]. In 1986, Nolan et al [25] reported on the marked superiority of Hansel’s stain over Wright’s stain in detecting eosinophiluria. Eosinophiluria has been associated with acute interstitial nephritis, rapidly progressive glomerulonephritis, chronic renal failure, IgA nephropathy, pyelonephritis, urinary tract infection, acute prostatitis, post-infectious glomerulonephritis, obstructive nephropathy, and acute rejection of a renal allograft [24-271; its presence in AERD, however, has not been previously established.
Volume 91
EOSINOPHILURIA
IN ATHEROEMBOLIC
RENAL
DISEASE / WILSON
ET AL
Renal Artery Stenosis
Carotid Endarterectomy
7
a
9
M-64
CAD
Chronic
F-75
Hypertension CVA
cr. - 3.6 Acute
M-77
Polyneuropathy Systemic Vasadiis
Cr. I 2.3 Acute
None
None
None
cm
None
None
None
IgA
Necrotizing GNSegmental Lived0 Reticularis
FSGN CiW Vasculitis
Atheromatous
Emboli
Neuotizing Atheromatous
GN Emboli
FSGN Atheromatous Emboli
Figure 1. Clinical and renal biopsy findings in nine patients with AERD. Acute renal failure is a change in the serum creatinine level greater than 0.5 mg/dL within 2 weeks. AAA = abdominal aortic aneurysm; CAD = coronary artery disease; CVA = cerebrovascular disease; Cr = serum creatinine (mg/dL); Angio = angiogram; FSGN = focal sclerosing glomerulonephritis; C/W = consistent with; GN = glomerulonephritis; spec = specimen.
PATIENTS AND METHODS We introduced Hansel’s stain for evaluation of eosinophiluria in 1985. In order to study eosinophiluria in AERD, we reviewed this finding in all biopsy-proven cases of AERD. There were 24 such cases at the Mayo Clinic from September 1985 to June 1989. All specimens were reviewed and interpreted by the same pathologist, and all patients were evaluated by a staff nephrologist. Patient records were reviewed for age, race, sex, clinical presentation, nature and degree of renal failure, medications at the time of biopsy, peripheral eosinophil count, platelet count, urinalysis with Gram stain, and the degree and presence of eosinophiluria as assessed by Hansel’s stain. Eosinophiluria was defined as 1% or greater of the white blood cells present after centrifuging 15 mL of urine and staining with Hansel’s stain [25]. In nine of 24 patients, urine eosinophils were obtained in the evaluation of these patients’ increasing serum creatinine levels. We evaluated this subset of patients in relation to eosinophiluria, AERD, and their clinical presentation.
RESULTS Clinical data of the nine patients whose urine eosinophils were measured are summarized in Fig-
ures 1 and 2. The mean age of the cohort was 69.2 years. Five patients were men and four were women. All pathologic specimens had evidence of renal atheroembolism. In addition, four of the specimens also had other pathologic diagnoses as listed in Figure 1. Of these nine patients, 89% had eosinophiluria (n = 8). In six of eight patients, more than 5% of the urinary white cell count was eosinophils; in two of eight, eosinophils ranged from 1% to 5%. One of the nine patients had a negative Hansel’s stain. Leukocyturia, proteinuria, and hematuria were variable, but none of the patients had a positive urine Gram stain. Clinically, 33% (n = 3) had procedures involving manipulation of the abdominal aorta, two after abdominal aortic aneurysm repair and one after coronary and carotid arteriography. Thirty-three percent (n = 3) had acute renal failure, 55% (n = 5) had acute on chronic renal failure, and 11% (n = 1) had chronic renal failure. The degree of atherosclerotic vascular disease varied but was present in seven of nine patients. Physical examination revealed atheroembolism in three of nine patients; two had livedo reticularis and two had Hollenhorst plaques on fundoscopic examination. Peripheral eosinophilia (greater than 350/mm3) was noted in 67% of patients (n = 6) at the time of renal biopsy. The abso-
August
1991
The American
Journal
of Medicine
Volume
91
187
EOSINOPHILURIA
IN ATHEROEMBOLIC
RENAL DISEASE
6
5.9
11-20
0
7
5.2
4-10
2+
6
3.6
l-3
2+
9
2.3
0
0
51-100
/ WILSON
ET AL
ND
>5%
263
6%
666
6.6
0
l-5%
224
4%
500
12.5
11-20
0
>5%
599
7%
1253
17.9
4-10
ND
>5%
300
15%
1230
0.2
0
Figure 2. Laboratory findings in nine patients with AERD. Creat = serum creatinine (mg/dL); WBC/HPF = white blood cells per high-power field: Urine RBC = red blood cells per high-power field; platelets X 1,000 = number of platelets/mm3; eosinophils in blood = percent of total white blood cells as eosinophils; absolute eosinophil count = number of eosinophils/mms; absolute WBC count X 1,000 = number of white blood cells/mm3; “occ” = occasional.
lute peripheral eosinophil count ranged from 500 to 1,253/mm3. Only one of the nine patients had marked thrombocytopenia. Patient 6 is the only patient who presented clinically with chronic rather than acute or acute on chronic renal failure. Although he had marked eosinophiluria, his biopsy also showed an IgA nephropathy which has been reported previously in association with eosinophiluria [25]. It is possible that either or both of these lesions were responsible for the eosinophiluria in this case.
AERD, greater degrees of eosinophiluria may have a higher predictive value, but this needs to be further defined in future studies. Renal biopsy remains the definitive way to diagnose AERD, although skin biopsies have been suggested as an alternative when livedo reticularis is present [32]. Many of the patients suspected of having AERD are too acutely ill to tolerate renal biopsy. Thrombocytopenia, hypocomplementemia, and peripheral eosinophilia have been reported as important laboratory markers for AERD. Previous studies have shown that 80% of patients with AERD have an associated peripheral eosinophilia; our findings were similar (67%). However, in contrast with previous studies that demonstrated thrombocytopenia in 85% of patients with AERD, only one of nine patients in our series presented with significant thrombocytopenia. This patient was also the only one who did not have eosinophiluria. It is possible that eosinophiluria develops later when the platelet count has recovered. Another possible explanation for the absence of thrombocytopenia in our series is that six of nine patients had underlying chronic renal failure. Although eight of nine patients had acute deterioration of renal function, the effect of the chronic component to the renal failure may modulate the platelet response. The efficacy of any laboratory marker to suggest a pathologic state is related to the predictive value of that marker for the disease. The predictive value of a positive Hansel’s stain for interstitial nephritis is higher when the clinical setting for interstitial
COMMENTS Previous studies have reported that AERD is associated with an “inactive” renal sediment and urine eosinophils were not described [17,28]. Eosinophiluria has been reported only anecdotally in association with AERD [29,30]. In our series, eight of nine patients with biopsy-proven AERD had eosinophiluria using Hansel’s stain. The reason for the failure of previous studies to detect urine eosinophils in AERD is unclear but may have been related to the use of Wright’s stain instead of Hansel’s stain, which is clearly a more sensitive and predictable method for detecting eosinophiluria. The presence of any degree of eosinophiluria is pathologic [31] and has been noted in a wide variety of renal diseases. The significance of the degree of eosinophiluria in AERD has not been established. Seventy-five percent of the patients in our study had greater than 5% urine eosinophils using the Hansel’s stain. In those patients presumed to have
188
August
1991
The American
Journal
of Medicine
Volume
91
EOSlNOPHlLURlA
nephritis is also high [33]. This is likely to be true also for AERD. A positive Hansel’s stain is more likely to be a helpful diagnostic marker when the clinical suspicion is high. The pretest probability for AERD may be dependent on a number of factors including the presence of vascular disease, a prior history of invasive vascular procedures, or advanced age. Eighty-nine percent of our patients had pre-existing vascular disease, and 33% had a history of vascular procedures; the median age was 69.2 years. In a retrospective analysis, there are many concerns as to the applicability of the results to future clinical decisions. Several problems with our study include the following: (1) Although eight of nine patients from whom biopsies were obtained had a positive Hansel’s stain, the incidence of eosinophiluria in AERD is unclear. The other patients with biopsy-proven AERD had no Hansel’s stain performed, and biopsies were not obtained from other patients with suspected or clinically inapparent AERD. (2) In addition to atheroemboli, four of nine patients had other renal diseases on biopsy and their contribution to the presence of eosinophiluria is not clearly established. Of those reported, however, only vasculitis (Patient 8) has previously been described with eosinophiluria. (3) Although none of our patients had a positive urine Gram stain, two of the eight positive cases had no Gram stain performed. Urinary tract infections are known to cause low-grade eosinophiluria, but both of our cases with no Gram stain had high-grade (greater than 5%) eosinophiluria. In conclusion, eosinophiluria was noted in 89% of patients with biopsy-proven AERD. Since many of these patients are acutely ill, noninvasive diagnostic tests are often needed to help establish the diagnosis. Although a prospective study would be needed to delineate the exact predictability of eosinophiluria for AERD, in the setting of a high clinical suspicion for AERD, eosinophiluria may be an important finding.
REFERENCES 1. Meyrier A, Buchet P, Simon P. Femet M, Rainfray M, Callard P. Atheromatous renal disease. Am J Med 1988: 85: 139-45. 2. Sieniewicz DJ, Moore S, Moir FD, McDade DF. Atheromatous emboli to the kidney. Radiology 1969: 92: 1231-40. 3. Brenner BM, Lazarus MJ. Acute renal failure. 2nd ed. New York: Churchill
IN ATHEROEMBOLIC
RENAL
DISEASE / WILSON
ET AL
Livingstone Inc, 1988: 375-7. 4. Thurlbeck W. Castleman B. Atheromatous emboll to the kidneys after aortic surgery. N Engl J Med 1957: 257: 442-7. 5. Pollitt J, Lee BM. Renal failure from atheroembolism after translumbar aortography. J Am Geriatr Sot 1971; 19: 989-95. 6. Harrington JT, Somers SC, Kassirer JP. Atheromatous emboli with progressive renal failure: renal arteriography as the probable inciting factor. Ann Intern Med 1968; 68: 152-60. 7. Ramirez G, O’Neill W Jr, Lambert R. eta/. Cholesterol embolization: a complication of angiography. Arch Intern Med 1978; 138: 1430-2. 8. Drost H. Buis B, Haan D, Hillers JA. Cholesterol embolism as a complication of left heart catheterization: report of seven cases. Br Heart J 1984; 52: 339-42. 9. Kassirer JP. Atheroembolic renal disease. N Engl J Med 1969; 280: 812-8. 10. Regester R. Renal failure secondary to spontaneous atheromatous microembolism. J Tenn Med Assoc 1980; 73: 328-30. 11. Flory CM. Arterial occlusion produced by emboli from eroded aortic atheromatous plaques. Am J Pathol 1945; 21: 549-58. 12. Gore I, Collins DP. Spontaneous atheromatous embolization: review of the literature and report of 16 additional cases. Am J Clin Pathol 1960; 33: 416-26. 13. Carvalal JA, Anderson WR. Weiss L. et a/. Atheroembolism: an etiologic factor in renal insufficiency, gastrointestinal hemorrhages and peripheral vascular diseases. Arch Intern Med 1967; 119: 593-9. 14. Eliot RS. Kanlvh VI, Edwards JE. Atheromatous embolism. Circulation 1964; 30: 611-S. 15. Moldveen-Geronimus M, Merriam JC. Cholesterol embollzatlon: from pathological curiosity to clinical entity. Circulation 1967; 35: 946-53. 16. Smith MC, Ghose MK, Henry AR. The clinical spectrum of renal cholesterol embolization. Am J Med 1981; 71: 174-80. 17. Kasinath B. Eosinophilia as a clue to the diagnosis of atheroembollc renal disease. Arch Intern Med 1987; 147: 1384-5. 18. Kasinath BS, Corwin HL, Bidani AK, Korbit SM. SchwartzMM. Eosinophilia in the diagnosis of atheroembolic renal disease. Am J Nephrol 1987; 7: 173-7. 19. Cosio FG. Zager RA, Sharma HM. Atheroembolic renal disease causes hypocomplementemia. Lancet 1985; 2: 118-21. 20. Eisenstaedt JS. Allergy and drug hypersensitivity of the urinary tract. J Urol 1951; 65: 154-9. 21. Helgason S, Lindqvist B. Eosinophilurta. Stand J Ural Nephrol 1972; 6: 257-9. 22. Galpin JE. Shinaberger JH. Standley TM, eta/. Acute Interstitial nephritis due to methicillin. Am J Med 1978; 65: 756-65. 23. Linton AL, Clark WF, Driedger AA, Turnbull DI. Lindsay RM. Acute interstitial nephritis due to drugs. Ann Intern Med 1980; 93: 735-41. 24. Corwin HL. Korbet SM. Schwartz MM. Clinical correlates of eosinophiluria. Arch Intern Med 1985; 145: 1097-g. 25. Nolan CR Ill, Anger MS, Kelleher SP. Eosinophiluria-a new method of detection and definition of the clinical spectrum. N Engl J Med 1986; 315: 1516-9. 26. Sutton JM. Urinary eosinophils. Arch Intern Med 1986; 146: 2243-4. 27. Spencer ES, Peterson V. The urinary sediment after transplantation. Acta Med Stand 1967; 182: 73-82. 28. Massry SG. Glassock RJ. Textbook of nephrology. 2nd ed. Baltimore: Williams and Wilkins, 1989; 916-9. 29. Case recordsof the MassachusettsGeneral Hospital (case 30-1986). N Engl J Med 1986; 315: 308-15. 30. Case records of the Massachusetts General Hospital (case 9-1990). N Engl J Med 1990; 322: 612-22. 31. Corwin HL. Haber MH. The clInIcal significance of eosinophiluria. Am J Clin Pathol 1987; 88: 520-2. 32. McGowan JA, Greenberg A. Cholesterol atheroembolic renal disease. Am J Nephrol 1986; 6: 135-9. 33. Nolan CR Ill, Kelleher SP. Eosinophiluria. Clin Lab Med 1988; 8: 555-65.
August
1991
The American
Journal
of Medicine
Volume
91
189
PURPOSE AND PATIENTS: %sinophikia has been reported in acute interstitial nephritis and other renal diseases, but its presence in atheroembolic renal disease (AJIRD) has not been previously established. ARRD has been identified as a cause of acute and chronic renal failure, particularly in elderly patients with advanced atherosclerosis and in those patients who have undergone manipulation or intervention of the abdominal aorta, renal artery, or coronary artery. The deftitive diagnosis is made by renal biopsy. However, many patients are too acutely ill to tolerate renal biopsy and, in recent years, peripheral eosinophilia, hypocomplementemia, and thrombocytopenia have been recognized in association with ARRD. Previous studies have reported that AJIRD is associated with an inactive renal sediment and an absence of urine eosinophils. We reviewed our experience over a 4year period with 24 patients with renal biopsy-proven ARRD. RESULTS: uriIM3 eosinophils were eVahUtted in nine patients to help determine the cause of their renal deterioration. Seven of these patients presented with evidence of vascular disease. Three patients had procedures involving manipulation of the abdominal aorta Physical examination revealed findings of atheroembolism in three of nine patients. Overall, eight of nine patients had a positive Hansel’s stain for eosinophiluria. Six of eight patients had more than 5% of their urinary white cell count as eosinophils. The reason for failure of previous studies to detect eosinophiluria in ARRD is uuclear but may have been related to the use of Wright’s stain instead of Hansel’s stain. CONCLUSION: h the evaluation of acute red insufficiency, eosinophihuia may indicate ARRD in addition to the other hnown causes for this fmding.
From the Department of Medicine, Mayo Clinic-Mayo Foundation, 200 First Street Southwest, Rochester, Minnesota. Requests for reprints should be addressed to David M. Wilson, M.D.,
Manuscript
186
submitted
August 1991
November
29, 1990, and accepted
in revised
The AmericanJournal of Medicine
A
theroembolic renal disease (AERD) has been identified as a cause of acute and chronic renal failure [l], particularly in elderly patients [2,3] with advanced atherosclerosis and those patients who have undergone manipulation or intervention of the abdominal aorta [4,5], renal artery [6], or coronary artery [7,8]. Renal atheroembolism has been reported in 15% of patients with severe aortic atherosclerosis and 31% of patients with large aortic aneurysms [4]. Pathologically, the essential lesion is an extensive occlusion, most notable in the lumens of vessels 50 to 200 p in diameter, with arcuate, intralobular, and terminal arteries most frequently affected [9,10]. The definitive diagnosis of AERD depends on renal biopsy. Clinically, the diagnosis has been difficult to make. AERD is sometimes associated with emboli to the brain, pancreas, thyroid, gastrointestinal tract, and extremities [ll-151. Often, there are few physical findings. The outcome for patients with AERD is variable, and the condition does not always progress to end-stage renal disease [16]. In recent years, peripheral eosinophilia [17,18], hypocomplementemia, and thrombocytopenia [19] have been reported in association with AERD. We noted eosinophiluria in one of our biopsy-proven cases, which led to a review of our experience. Eosinophiluria secondary to a drug-induced hypersensitivity reaction was first reported by Eisenstaedt [20] in 1951. Helgason and Lundqvist [21] reported on its significance in a broader spectrum of renal disease. In 1978, Galpin et al [22] reported on nine patients with methicillin-induced, acute interstitial nephritis with eosinophiluria ranging from 10% to 60%. Further studies also showed a correlation between urine eosinophils and acute interstitial nephritis [23,24]. In 1986, Nolan et al [25] reported on the marked superiority of Hansel’s stain over Wright’s stain in detecting eosinophiluria. Eosinophiluria has been associated with acute interstitial nephritis, rapidly progressive glomerulonephritis, chronic renal failure, IgA nephropathy, pyelonephritis, urinary tract infection, acute prostatitis, post-infectious glomerulonephritis, obstructive nephropathy, and acute rejection of a renal allograft [24-271; its presence in AERD, however, has not been previously established.
Volume 91
EOSINOPHILURIA
IN ATHEROEMBOLIC
RENAL
DISEASE / WILSON
ET AL
Renal Artery Stenosis
Carotid Endarterectomy
7
a
9
M-64
CAD
Chronic
F-75
Hypertension CVA
cr. - 3.6 Acute
M-77
Polyneuropathy Systemic Vasadiis
Cr. I 2.3 Acute
None
None
None
cm
None
None
None
IgA
Necrotizing GNSegmental Lived0 Reticularis
FSGN CiW Vasculitis
Atheromatous
Emboli
Neuotizing Atheromatous
GN Emboli
FSGN Atheromatous Emboli
Figure 1. Clinical and renal biopsy findings in nine patients with AERD. Acute renal failure is a change in the serum creatinine level greater than 0.5 mg/dL within 2 weeks. AAA = abdominal aortic aneurysm; CAD = coronary artery disease; CVA = cerebrovascular disease; Cr = serum creatinine (mg/dL); Angio = angiogram; FSGN = focal sclerosing glomerulonephritis; C/W = consistent with; GN = glomerulonephritis; spec = specimen.
PATIENTS AND METHODS We introduced Hansel’s stain for evaluation of eosinophiluria in 1985. In order to study eosinophiluria in AERD, we reviewed this finding in all biopsy-proven cases of AERD. There were 24 such cases at the Mayo Clinic from September 1985 to June 1989. All specimens were reviewed and interpreted by the same pathologist, and all patients were evaluated by a staff nephrologist. Patient records were reviewed for age, race, sex, clinical presentation, nature and degree of renal failure, medications at the time of biopsy, peripheral eosinophil count, platelet count, urinalysis with Gram stain, and the degree and presence of eosinophiluria as assessed by Hansel’s stain. Eosinophiluria was defined as 1% or greater of the white blood cells present after centrifuging 15 mL of urine and staining with Hansel’s stain [25]. In nine of 24 patients, urine eosinophils were obtained in the evaluation of these patients’ increasing serum creatinine levels. We evaluated this subset of patients in relation to eosinophiluria, AERD, and their clinical presentation.
RESULTS Clinical data of the nine patients whose urine eosinophils were measured are summarized in Fig-
ures 1 and 2. The mean age of the cohort was 69.2 years. Five patients were men and four were women. All pathologic specimens had evidence of renal atheroembolism. In addition, four of the specimens also had other pathologic diagnoses as listed in Figure 1. Of these nine patients, 89% had eosinophiluria (n = 8). In six of eight patients, more than 5% of the urinary white cell count was eosinophils; in two of eight, eosinophils ranged from 1% to 5%. One of the nine patients had a negative Hansel’s stain. Leukocyturia, proteinuria, and hematuria were variable, but none of the patients had a positive urine Gram stain. Clinically, 33% (n = 3) had procedures involving manipulation of the abdominal aorta, two after abdominal aortic aneurysm repair and one after coronary and carotid arteriography. Thirty-three percent (n = 3) had acute renal failure, 55% (n = 5) had acute on chronic renal failure, and 11% (n = 1) had chronic renal failure. The degree of atherosclerotic vascular disease varied but was present in seven of nine patients. Physical examination revealed atheroembolism in three of nine patients; two had livedo reticularis and two had Hollenhorst plaques on fundoscopic examination. Peripheral eosinophilia (greater than 350/mm3) was noted in 67% of patients (n = 6) at the time of renal biopsy. The abso-
August
1991
The American
Journal
of Medicine
Volume
91
187
EOSINOPHILURIA
IN ATHEROEMBOLIC
RENAL DISEASE
6
5.9
11-20
0
7
5.2
4-10
2+
6
3.6
l-3
2+
9
2.3
0
0
51-100
/ WILSON
ET AL
ND
>5%
263
6%
666
6.6
0
l-5%
224
4%
500
12.5
11-20
0
>5%
599
7%
1253
17.9
4-10
ND
>5%
300
15%
1230
0.2
0
Figure 2. Laboratory findings in nine patients with AERD. Creat = serum creatinine (mg/dL); WBC/HPF = white blood cells per high-power field: Urine RBC = red blood cells per high-power field; platelets X 1,000 = number of platelets/mm3; eosinophils in blood = percent of total white blood cells as eosinophils; absolute eosinophil count = number of eosinophils/mms; absolute WBC count X 1,000 = number of white blood cells/mm3; “occ” = occasional.
lute peripheral eosinophil count ranged from 500 to 1,253/mm3. Only one of the nine patients had marked thrombocytopenia. Patient 6 is the only patient who presented clinically with chronic rather than acute or acute on chronic renal failure. Although he had marked eosinophiluria, his biopsy also showed an IgA nephropathy which has been reported previously in association with eosinophiluria [25]. It is possible that either or both of these lesions were responsible for the eosinophiluria in this case.
AERD, greater degrees of eosinophiluria may have a higher predictive value, but this needs to be further defined in future studies. Renal biopsy remains the definitive way to diagnose AERD, although skin biopsies have been suggested as an alternative when livedo reticularis is present [32]. Many of the patients suspected of having AERD are too acutely ill to tolerate renal biopsy. Thrombocytopenia, hypocomplementemia, and peripheral eosinophilia have been reported as important laboratory markers for AERD. Previous studies have shown that 80% of patients with AERD have an associated peripheral eosinophilia; our findings were similar (67%). However, in contrast with previous studies that demonstrated thrombocytopenia in 85% of patients with AERD, only one of nine patients in our series presented with significant thrombocytopenia. This patient was also the only one who did not have eosinophiluria. It is possible that eosinophiluria develops later when the platelet count has recovered. Another possible explanation for the absence of thrombocytopenia in our series is that six of nine patients had underlying chronic renal failure. Although eight of nine patients had acute deterioration of renal function, the effect of the chronic component to the renal failure may modulate the platelet response. The efficacy of any laboratory marker to suggest a pathologic state is related to the predictive value of that marker for the disease. The predictive value of a positive Hansel’s stain for interstitial nephritis is higher when the clinical setting for interstitial
COMMENTS Previous studies have reported that AERD is associated with an “inactive” renal sediment and urine eosinophils were not described [17,28]. Eosinophiluria has been reported only anecdotally in association with AERD [29,30]. In our series, eight of nine patients with biopsy-proven AERD had eosinophiluria using Hansel’s stain. The reason for the failure of previous studies to detect urine eosinophils in AERD is unclear but may have been related to the use of Wright’s stain instead of Hansel’s stain, which is clearly a more sensitive and predictable method for detecting eosinophiluria. The presence of any degree of eosinophiluria is pathologic [31] and has been noted in a wide variety of renal diseases. The significance of the degree of eosinophiluria in AERD has not been established. Seventy-five percent of the patients in our study had greater than 5% urine eosinophils using the Hansel’s stain. In those patients presumed to have
188
August
1991
The American
Journal
of Medicine
Volume
91
EOSlNOPHlLURlA
nephritis is also high [33]. This is likely to be true also for AERD. A positive Hansel’s stain is more likely to be a helpful diagnostic marker when the clinical suspicion is high. The pretest probability for AERD may be dependent on a number of factors including the presence of vascular disease, a prior history of invasive vascular procedures, or advanced age. Eighty-nine percent of our patients had pre-existing vascular disease, and 33% had a history of vascular procedures; the median age was 69.2 years. In a retrospective analysis, there are many concerns as to the applicability of the results to future clinical decisions. Several problems with our study include the following: (1) Although eight of nine patients from whom biopsies were obtained had a positive Hansel’s stain, the incidence of eosinophiluria in AERD is unclear. The other patients with biopsy-proven AERD had no Hansel’s stain performed, and biopsies were not obtained from other patients with suspected or clinically inapparent AERD. (2) In addition to atheroemboli, four of nine patients had other renal diseases on biopsy and their contribution to the presence of eosinophiluria is not clearly established. Of those reported, however, only vasculitis (Patient 8) has previously been described with eosinophiluria. (3) Although none of our patients had a positive urine Gram stain, two of the eight positive cases had no Gram stain performed. Urinary tract infections are known to cause low-grade eosinophiluria, but both of our cases with no Gram stain had high-grade (greater than 5%) eosinophiluria. In conclusion, eosinophiluria was noted in 89% of patients with biopsy-proven AERD. Since many of these patients are acutely ill, noninvasive diagnostic tests are often needed to help establish the diagnosis. Although a prospective study would be needed to delineate the exact predictability of eosinophiluria for AERD, in the setting of a high clinical suspicion for AERD, eosinophiluria may be an important finding.
REFERENCES 1. Meyrier A, Buchet P, Simon P. Femet M, Rainfray M, Callard P. Atheromatous renal disease. Am J Med 1988: 85: 139-45. 2. Sieniewicz DJ, Moore S, Moir FD, McDade DF. Atheromatous emboli to the kidney. Radiology 1969: 92: 1231-40. 3. Brenner BM, Lazarus MJ. Acute renal failure. 2nd ed. New York: Churchill
IN ATHEROEMBOLIC
RENAL
DISEASE / WILSON
ET AL
Livingstone Inc, 1988: 375-7. 4. Thurlbeck W. Castleman B. Atheromatous emboll to the kidneys after aortic surgery. N Engl J Med 1957: 257: 442-7. 5. Pollitt J, Lee BM. Renal failure from atheroembolism after translumbar aortography. J Am Geriatr Sot 1971; 19: 989-95. 6. Harrington JT, Somers SC, Kassirer JP. Atheromatous emboli with progressive renal failure: renal arteriography as the probable inciting factor. Ann Intern Med 1968; 68: 152-60. 7. Ramirez G, O’Neill W Jr, Lambert R. eta/. Cholesterol embolization: a complication of angiography. Arch Intern Med 1978; 138: 1430-2. 8. Drost H. Buis B, Haan D, Hillers JA. Cholesterol embolism as a complication of left heart catheterization: report of seven cases. Br Heart J 1984; 52: 339-42. 9. Kassirer JP. Atheroembolic renal disease. N Engl J Med 1969; 280: 812-8. 10. Regester R. Renal failure secondary to spontaneous atheromatous microembolism. J Tenn Med Assoc 1980; 73: 328-30. 11. Flory CM. Arterial occlusion produced by emboli from eroded aortic atheromatous plaques. Am J Pathol 1945; 21: 549-58. 12. Gore I, Collins DP. Spontaneous atheromatous embolization: review of the literature and report of 16 additional cases. Am J Clin Pathol 1960; 33: 416-26. 13. Carvalal JA, Anderson WR. Weiss L. et a/. Atheroembolism: an etiologic factor in renal insufficiency, gastrointestinal hemorrhages and peripheral vascular diseases. Arch Intern Med 1967; 119: 593-9. 14. Eliot RS. Kanlvh VI, Edwards JE. Atheromatous embolism. Circulation 1964; 30: 611-S. 15. Moldveen-Geronimus M, Merriam JC. Cholesterol embollzatlon: from pathological curiosity to clinical entity. Circulation 1967; 35: 946-53. 16. Smith MC, Ghose MK, Henry AR. The clinical spectrum of renal cholesterol embolization. Am J Med 1981; 71: 174-80. 17. Kasinath B. Eosinophilia as a clue to the diagnosis of atheroembollc renal disease. Arch Intern Med 1987; 147: 1384-5. 18. Kasinath BS, Corwin HL, Bidani AK, Korbit SM. SchwartzMM. Eosinophilia in the diagnosis of atheroembolic renal disease. Am J Nephrol 1987; 7: 173-7. 19. Cosio FG. Zager RA, Sharma HM. Atheroembolic renal disease causes hypocomplementemia. Lancet 1985; 2: 118-21. 20. Eisenstaedt JS. Allergy and drug hypersensitivity of the urinary tract. J Urol 1951; 65: 154-9. 21. Helgason S, Lindqvist B. Eosinophilurta. Stand J Ural Nephrol 1972; 6: 257-9. 22. Galpin JE. Shinaberger JH. Standley TM, eta/. Acute Interstitial nephritis due to methicillin. Am J Med 1978; 65: 756-65. 23. Linton AL, Clark WF, Driedger AA, Turnbull DI. Lindsay RM. Acute interstitial nephritis due to drugs. Ann Intern Med 1980; 93: 735-41. 24. Corwin HL. Korbet SM. Schwartz MM. Clinical correlates of eosinophiluria. Arch Intern Med 1985; 145: 1097-g. 25. Nolan CR Ill, Anger MS, Kelleher SP. Eosinophiluria-a new method of detection and definition of the clinical spectrum. N Engl J Med 1986; 315: 1516-9. 26. Sutton JM. Urinary eosinophils. Arch Intern Med 1986; 146: 2243-4. 27. Spencer ES, Peterson V. The urinary sediment after transplantation. Acta Med Stand 1967; 182: 73-82. 28. Massry SG. Glassock RJ. Textbook of nephrology. 2nd ed. Baltimore: Williams and Wilkins, 1989; 916-9. 29. Case recordsof the MassachusettsGeneral Hospital (case 30-1986). N Engl J Med 1986; 315: 308-15. 30. Case records of the Massachusetts General Hospital (case 9-1990). N Engl J Med 1990; 322: 612-22. 31. Corwin HL. Haber MH. The clInIcal significance of eosinophiluria. Am J Clin Pathol 1987; 88: 520-2. 32. McGowan JA, Greenberg A. Cholesterol atheroembolic renal disease. Am J Nephrol 1986; 6: 135-9. 33. Nolan CR Ill, Kelleher SP. Eosinophiluria. Clin Lab Med 1988; 8: 555-65.
August
1991
The American
Journal
of Medicine
Volume
91
189