Afferent Arteriolar C3 Disease—A Distinct Pathological Entity

Afferent Arteriolar C3 Disease-A Distinct Pathological Entity Carol A. Pollock, MB, BS, FRACp, Lloyd S. Ibels, MB, BS, FRACp, Robert P. Eckstein, MB, BS, FRCPA, Mark A.B. Thomas, MB, BS, FRACp, Christopher Lauer, MB, BS, FRCPA, and Dennis Moir, MB, BS, FRCPA • Afferent arteriolar C3 deposition was the sole histological abnormality in 79 and the major histological abnormality in an additional 39 of 959 renal biopsies performed over a 10-year period. Of these 79 patients, hematuria was the presenting symptom in 90%, with coincident loin pain in 49%. Urine microscopy of asymptomatic firstdegree relatives revealed hematuria in 44% of children and siblings and 54% of parents, suggesting autosomal dominant inheritance. Arteriolar C3 deposition was confirmed by biopsy in four asymptomatic relatives with hematuria. Generalized thinning of glomerular basement membrane (< 200 nm) was observed in five patients and focal thinning was observed in six patients with coincident afferent arteriolar C3 deposition. Seven other patients were identified as having generalized thinning of glomerular basement membrane in the absence of afferent arteriolar C3 deposition. Renal function was stable and similar in all groups studied over 37.9 ± 23.7 months. No difference in clinical presentation or urinary abnormalities was evident between the groups. No arteriolar C3 deposition was evident in eight autopsy specimens with no known renal disease. It was concluded that afferent arteriolar C3 deposition is a marker of a distinct hereditary pathological entity, with differentiation from thin basement membrane disease not possible on clinical grounds. The medium- and long-term prognoses with respect to renal function are excellent. © 1989 by the National Kidney Foundation, Inc. INDEX WORDS: Afferent arteriolar C3 disease; loin pain/hematuria syndrome; hematuria; glomerulonephritis; thin basement membrane disease; familial nephritis.

I

SOLATED DEPOSITION of the third component of complement (C3) in arterioles or mesangium of renal tissue is increasingly described in association with isolated hematuria l -4 or with the loin pain hematuria syndrome. 5- 10 Thin basement membrane disease may also be associated with hematuria and complement deposition in arteriolar walls. II However, the incidences and natural histories of both of these conditions are not well defined, nor is the prognosis, particularly with regard to renal function. To clarify these issues, those patients whose predominant abnormality on renal biopsy was afferent arteriolar deposition of C3 or generalized thinning of the glomerular basement membrane were reviewed over a lO-year period. METHODS Nine hundred fifty-nine percutaneous renal biopsies were performed between January 1978 and September 1987. To ensure uniform indications for renal biopsy, all were performed by a single physician (L.S.I.) Biopsy specimens were prepared for light microscopy by standard histological techniques and stained with hematoxylin-eosin, periodic acid-Schiff, Wilders silver stain, and Masson's trichrome. Slides for direct immunofluorescent microscopy were prepared by standard methods,12 stained with fluoresceinisothiocynate-labeled antisera to IgG, IgA, IgM, C3, and fibrinogen; and examined with a Nikon fluoroscence microscope (Nikon, Tokyo). For electron microscopy, I-mm cubes were fixed in cold 3.75% gluteraldehyde solution and buffered with sodium cacodylate and su-

crose. After fixing and embedding, I-Jl.m sections were examined with a Siemans Model !Ol electron microscope (Siemans, Munich). One hundred eighteen patients in whom the predominant abnormality assessed retrospectively from past renal biopsies was afferent arteriolar or extraglomerular vascular C3 deposition, with no other specific histological diagnosis made on light microscopy or immunofluorescent examination, were initially included in the review. In all cases, granular deposition of C3 was present in an intensity of + + to + + + positivity (on a scale of 0 to + + +). In biopsies in which both afferent arteriole and extraglomerular blood vessels were present, C3 deposition was similarly distributed. In biopsies in which the plane of section did not include an afferent arteriole (n = 21), isolated C3 deposition in extraglomerular blood vessels was regarded as indicative of coexistent afferent arteriolar C3 deposition. From one to six glomeruli were available for immunofluorescent examination. Prospective review of electron microscopy in 24 of 40 patients with no demonstrable abnormality on light or immunofluorescent microscopic examination and in 86 of the 118 patients with afferent arteriolar C3 deposition was then undertaken, including multiple estimations of glomerular basement membrane thicknesses made from

From the Departments of Renal Medicine and Pathology, Royal North Shore Hospital, St Leonard, and Hanly Moir Pathology, St Ives, New South Wales. Dr Pollock is the recipient of the Baxter Research Fellowship for 1988. Address reprint requests to L.S. Ibels, MB, BS, FRACP, Department of Renal Medicine, Royal North Shore Hospital, St Leonards NSW 2065, Australia_ © 1989 by the National Kidney Foundation, Inc. 0272-6386/89/1401-0006$3.00/0

American Journal of Kidney Diseases, Vol XIV, No 1 (July), 1989: pp 31-38

31

32

POLLOCK ET AL

Table 1. Criteria for Exclusion of Patients in Whom Afferent Arteriolar C3 Was the Predominant Finding Exclusion Criteria

Systemic disease known to involve the kidney with or without nonspecific renal involvement Diabetes + renal calculi Medullary sponge kidney Lead intoxication Treated renal tuberculosis Systemic illness with rash, arthralgia, and positive ANA (1/160) Abnormal although nonspecific light microscopic changes of a focal nature Interstitial infiltrate ± tubular atrophy ± ischemic changes Mesangial hypercellularity Abnormal electron microscopic changes Electron-dense deposits with normal light microscopy and no immune complexes on immunofluorescence Generalized thinning glomerular basement membrane Irregular thinning glomerular basement membrane Foot process fusion and wrinkling of glomerular basement membrane

No. of Patients

1 3 2

cal, radiological, or histological abnormality, and were similarly investigated. Patients' first -degree relatives were investigated with urine microscopy on a voluntary basis, and renal biopsy was performed in four patients. Hematuria was defined as > 1,000 RBCs/mL. 13 In order to assess whether afferent arteriolar C3 deposition was present in normal renal tissue, renal biopsies from autopsy specimens were sampled 2 to 26 hours (mean, 10.4 hours) after death in eight patients (three male, five female) aged 22 to 74 (mean, 54) years, with no antemortem evidence of renal disease. Statistical analyses were made using a Student's t test.

RESULTS

14

3 5

6 2

measurements of endothelium to epithelium in representative capillary loops removed from mesangial areas. Obliquely sectioned lengths were not included. Measurements of <200 nm were considered pathologically thin. Investigation of these 118 patients included history and physical examination, qualitative urine microscopy, full blood count, BUN, creatinine, creatinine clearance calculated from a 24-hour urinary creatinine estimation, 24-hour urinary protein excretion, serum immunoglobulins, and serum levels of the C3 and C4, ANA, DNA antibody, and rheumatoid factor. Intravenous pyelogram (lVP) was performed in 115 patients and cystoscopy in 102 patients. Other radiological investigations were performed as indicated on an individual basis. To assess as homogenous a group as possible who had solely afferent arteriolar C3 staining and no other abnormality on clinical, biochemical, serological, radiological, or histological examination, 39 patients were then excluded from analysis (Table 1). Patients with controlled hypertension or a history of lower urinary tract infections were not excluded from analysis, although investigations were performed when no infection was demonstrated. Biopsy changes of age-related glomerular sclerosis, vessel wall thickening consistent with hypertension but without corroborative evidence of ischemia (ie, tubular atrophy/interstitial fibrosis), mild increases in mesangial matrix, and trace capillary wall or mesangial IgM were not excluded from analysis. An additional seven of the 959 renal biopsies had generalized thinning of the glomerular basement membrane in the absence of afferent arteriolar C3 deposition according to the above criteria, with no other historical, clinical, biochemical, serologi-

Of 959 renal biopsies performed over the 10year period, 118 (12.3 %; 86 female, 32 male patients) had afferent arteriolar deposition of complement as the most striking abnormality. Thirty-nine patients were then excluded from analysis according to Table 1. Only the predominant cause for exclusion is listed, although in a number of instances abnormalities may coexist. To assess the prevalence of vascular C3 deposition in association with other pathology, an unselected subset of 546 biopsies were reviewed. C3 deposition (+ to + + +) was observed in 479 (87.7 %) cases; with the exception of the case in the current review, all had demonstrable associated light microscopic or immunofluorescent pathology. Thus, 79 patients (59 female, 20 male) with the only abnormality demonstrated being afferent arteriolar C3 deposition remained (Fig 1). Age at the time of biopsy ranged from 18 to 71 years (40.22 ± 12.55 years, mean ± SD) with no significant age difference between male (39.95 ±

Fig 1. Immunofluorescence demonstrating C3 in afferent arteriolar distribution.

33

AFFERENT ARTERIOLAR C3 DISEASE

Table 2.

Initial and Follow-up Results in Those With Lone Afferent Arteriolar C3 Deposition (n = 79) Follow·up

Initial

Serum creatinine (JLmoI/L) Creatinine clearance (mUmin) Urinary protein excretion (mg/24 h) Urinary RBC excretion (RBC/mL) Urinary WBC excretion (WBC/mL)

Mean ± SD

Range

Mean ± SD

± ± ± ± ±

60-150 50-220 0·1140 250·1.6 x 106 0·22,250

± ± ± ± ±

86 97 109 48,376 1,434

19 24 210 145,725 3,242

10.74 years) and female (40.31 ± 13.20 years) patients. Mean follow-up was 37.9 ± 23.7 months (range, 2 to 95 months). Nineteen patients were followed for > 5 years and 38 patients for >3 years. Initial presentation was due to hematuria in 71 cases (90 %), with coincident loin pain in 39 (49 %) patients. In six cases, loin pain was severe enough to warrant parenteral narcotics, and four patients have undergone successful autotransplantation of one or both kidneys as previously described. 9 . 10 Controlled hypertension was evident in 11 patients (15 %); recurrent lower urinary tract infection was present in 11 patients (15%); and one case of each of asthma, alcoholic liver disease, and controlled thyrotoxicosis was present. Two patients had decreased serum levels of C3 and C4, seven had isolated decrease in C3, and three had isolated decrease in C4. Increase in serum immunoglobulin was evident in 24 cases, with elevations ofIgG in ten, IgM in five, and IgA in 12 patients. Decreased serum immunoglobulins occurred in nine cases, IgM in eight, and IgA in one. Light microscopic changes observed in this group were age-related glomerular sclerosis (27), mild increase in mesangial matrix (6), and thickened arteriolar walls (8). Immunofluorescent abnormalities were strongly staining C3 in afferent arterioles in all, light-to-moderate mesangial C3 deposition in 34, trace mesangial IgM in six, capillary wall IgM in two, and both mesangial and capillary wall IgM in six patients. Initial and follow-up serum creatinine, creatinine clearance, 24-hour urinary protein excretion, and urine microscopy are shown in Table 2; the only overall change occurring over time is a decrease in urinary RBC excretion. At follow-up, creatinine clearance had decreased by > 10% in 16 patients, with a mean decrease of 18.3% ± 7.7% (range, 10% to 34%).

84 100 118 11,384 1,437

21 19 104 26,842 1,890

Range

40·140 66·142 0·830 (0·200,000) (0-11,500)

Of these, follow-up creatmme clearances were > 100 mLlmin in three patients and >85 mLlmin in 11 patients. The two remaining patients both had creatinine clearances of 77 mLlmin, and both had intermittent proteinuria. Twenty-two patients' renal function had improved by > 10%, with a mean increase of33.6% ± 19.3% (range, 10% to 88 %) and follow-up creatinine clearances of 83 and 84 mLlmin in two patients and >95 mLlmin in all others. Abnormal urinary protein excretion (> 150 mgt d) was evident in 15 patients at presentation, with two patients excreting> 1 g/d (1.08 and 1.14 g/d, respectively) and the remaining 13 having a mean loss of 323 ± 190 mg/d. Of these, eight had normal urinary protein loss at follow-up. Thirteen patients with initially normal urinary protein loss developed abnormal urinary protein excretion, with a mean loss of 259 ± 107 mg/d. Increased cast excretion was demonstrated during follow-up in 20 patients (nine hyaline, four hyalogranular, two granular, and five with hyaline or hyalogranular). Sixty-seven first-degree relatives of 24 patients with afferent arteriolar C3 deposition were screened by urine microscopy. All were asymptomatic except for one offspring with loin pain coincident with microscopic hematuria. Forty-five children or siblings had a total of 126 quantitative urine microscopies performed. Fiftythree urine microscopies in 20 children or siblings (44 %) showed increased RBC excretion. Twentytwo parents of 14 patients had a total of 39 quantitative urine microscopies, and 12 parents (55 %) showed increased RBC excretion. Four asymptomatic first-degree relatives (one parent, two siblings, one child) with persistent hematuria underwent renal biopsy and were shown to have moderate-to-strong deposition of afferent arteriolar C3, with one also showing a mild increase in mesangial matrix but no further abnormality on

34

POLLOCK ET AL

Fig 2. Electron micrograph demonstrating generalized thinning of glomerular basement membrane (original magnification x 7,500).

light microscopy, immunofluorescence, or electron microscopy examination. Of the 39 patients excluded from the above analysis, five patients with strongly staining afferent arteriolar C3 deposition had concomitant generalized thinning of the glomerular basement membrane (all measurements <200 nm) (Fig 2). Six additional patients had segmental thinning of glomerular basement membrane alternating with normal or thickened segments (Fig 3) but otherwise

Fig 3. Electron micrograph demonstrating focal thinning of glomerular basement membrane alternating with normal glomerular basement membrane (original magnification x 7,500).

normal clinical, biochemical, radiological, and histological parameters. In particular, no features of Alports syndrome were present. Biochemical characteristics are presented in Table 3. Seven patients in the initial 959 renal biopsies had generalized thinning of the glomerular basement membrane with no significant immunofluorescent or light microscopic changes and are similarly reviewed in Table 3. No significant difference between these three groups was demonstrated in any

35

AFFERENT ARTERIOLAR C3 DISEASE

Table 3.

Clinical and Biochemical Aspects of Patients With Generalized or Focal Thinning of Glomerular Basement Membranes

Age (yr) Follow-up (mo) Hematuria Initial Follow-up Loin Pain Initial Follow-up Serum creatinine (",moI/L) Initial Follow-up Creatinine clearance (mLlmin) Initial Follow-up Urinary protein excretion (mg/d) Initial Follow-up Urinary RBC/mL Initial Follow-up Urinary WBC/mL Initial Follow-up

Generalized Thinning With C3 Deposition (n = 5)

Focal Thinning With C3 Deposition (n = 6)

Generalized Thinning, No C3 Deposition (n = 7)

41.6 ± 5.7 13.8 ± 4.8

46.0 ± 12.1 23.3 ± 21.9

43.3 ± 17.9 13.6 ± 12.9

5 4

6 5

7 7

2 2

2

84 ± 19 78 ± 8

88 ± 19 90 ± 21

77 ± 7 76 ± 16

91 ± 15 100 ± 5

82 ± 5 89 ± 19

93 ± 10 110 ± 33

102 ± 12 74 ± 5

123 ± 64 122 ± 81

143 ± 130 187 ± 155

334,200 ± 707,639 10,550 ± 13,145

19,396 ± 23,365 11,983 ± 17,934

29,250 ± 38,590 24,300 ± 31,340

4,950 ± 4,288 1,450 ± 1,874

941 ± 1,419 1,042 ± 1,111

2,500 ± 2,791 1,035 ± 603

Glomerular basement membrane thickness reviewed prospectively in 86 of 118 patients with vascular C3 deposition and no light microscopic or additional immunofluorescent abnormalities and in 24 of 40 patients with normal light and immunofluorescent microscopic findings. Data presented as mean ± SO.

parameter, either at presentation or at follow-up, and no clinical, biochemical, or urinary parameter distinguished those with lone afferent arteriolar C3 deposition, lone abnormal glomerular basement membranes, or a combination of the two. Patients with generalized thinning of glomerular basement membranes and afferent arteriolar C3 depositions had family screening undertaken, which demonstrated hematuria in two of four parents and three of five children studied. Renal biopsies from the eight autopsy patients demonstrated age-related glomerular sclerosis in three, but no afferent arteriolar or extraglomerular vascular deposition of C3 in any case. DISCUSSION

Patients with isolated hematuria, either microscopic or macroscopic, have subsequently been found to have a variety of glomerular lesions with diverse pathological entities such as proliferative or focal glomerulonephritis, predominantly IgA disease, and increasingly appreciated arteriolar C3 disease. I In the present study, afferent arteriolar

C3 deposition was the sole finding in > 8 % of all biopsies performed over a lO-year period. The overlap in clinical presentation and biochemical, serological, radiological, and urinary sediment abnormalities is wide, and a definitive diagnosis is important because of prognostic implications for both the patients and their family members. IgA nephropathy may progress to endstage renal failure in 20% of patientsl4; Alports syndrome, having variable penetrance and expression, may progress to renal failure in a significant proportion of patients,15 while the prognosis of thin basement membrane disease varies, possibly because of the inclusion of some patients with hereditary nephritis. 16·21 The incidence of isolated afferent arteriolar C3 deposition accounting for hematuria is difficult to assess. In our series it was the predominant finding in 12.3% of959 unselected renal biopsies and was the sole abnormality demonstrated in 8.2 %. In other series of abnormal biopsies performed for isolated hematuria, the incidence varies from 11 % to 16 %. 1.2 The prevalence of this pathological en-

36

tity may not have been appreciated in prior series of patients with isolated hematuria because it was not common practice to perform renal biopsies if renal function and BP were stable, provided urological and radiological investigations were normal. It is probable that large numbers of patients who have been observed with persistent microscopic hematuria and stable renal function have had arteriolar C3 disease, because we have demonstrated no decline in renal function in up to 7.9 years of follow-up. The association of afferent arteriolar C3 deposition with loin pain was initially proposed by Naish et al in 1975 6 to be a distinct clinical and pathological entity; this proposal is supported by our series, in which loin pain, varying from mild to narcotic-requiring, occurred in approximately 50% of the cases. Few treatments have been found to be effective in symptomatic patients, with autotransplantation reserved for those incapacitated by loin pain. 9.10 Complement deposition in renal tissue has been described in association with a wide variety of systemic and glomerular disease. 6.12.22 Although it is a nonspecific finding in association with many well-defined renal diseases and may be nonspecifically deposited in association with renal parenchymal damage evident by light microscopy and indeed was observed in the vast majority (87.7 %) of un selected renal biopsies in this series, complement deposition has not been observed in the tissue of normal kidneys2; this has been confirmed in our small numbers of autopsy studies. Hyaline deposits characteristic of renal artery sclerosis contain C3 with IgM in 50% of cases,23 and although patients with light microscopic changes of arterial sclerosis were not excluded from analysis, the 15 % incidence of hypertension is not different from that observed in the general population.24 However, patients were excluded if histological evidence of ischemia subsequent to hypertension was observed. Lower urinary tract infections were treated and hematuria persisted, thus making infection an unlikely cause of persistent hematuria. Because of controversy regarding whether arteriolar C3 is pathogenic or is deposited in response to nonspecific renal injury, patients were excluded from analysis if any clinical or histological abnormality that might be a secondary cause of hematuria was found. Initial urinary RBC counts were higher at presentation than at follow-up, reflecting the initial timing of referral during an exacerbation

POLLOCK ET AL

of hematuria. Overall serum creatinine and creatinine clearance levels, however, remained stable over 37.9 ± 23.7 months, although they did fluctuate in a significant number of patients. Similarly, urinary protein loss was variable and did not uniformly progress during follow-up. Serum complement and immunoglobulin levels were abnormal in a minority of cases, and as such were unhelpful as an aid in diagnosis. Our results confirm those of Trachtman et all and Barcelo et aI, 2 who found that vascular C3 was the only abnormality to explain microscopic hematuria and that no patients developed proteinuria, hypertension, or renal impairment although microscopic hematuria persisted in the majority of cases. The familial nature of this disease has been suggested in this study by quantitative urine microscopic examinations showing microscopic hematuria in 44 % of children or siblings and 54 % of parents with arteriolar C3 disease and demonstrated on biopsy in four asymptomatic first-degree relatives. This pattern is consistent with autosomal dominant inheritance and exceeds the approximate 5 % prevalence of hematuria in family members seen in studies before the advent of immunofluorescence. 25 .26 The latter reported low prevalence probably reflects the insensitivity of urine analysis as a screen for the presence of renal disease. Quantitative urine microscopy is a sensitive, simple, and reproducible test for the presence of renal disease 13 and, as shown in this study, will often reveal unsuspected abnormalities. The familial nature of this disease was not observed by Trachtman et al, I although they did not screen relatives and relied on historical confirmation of familial renal disease. The female preponderance of index patients, despite the equal female-to-male ratio in relatives screened, reflects the increased tendency for female patients to present with coexistent loin pain. The pathogenic significance of complement deposition in extraglomerular blood vessels remains to be defined. Glomerular basement membrane has been demonstrated to activate complement via the alternate pathway and to mediate renal injury directly. 2728 C3 and IgM deposition in renal arterioles has also been reported to mediate vascular injury in the hemolytic uremic syndrome, implicating C3 deposition as a primary or secondary marker in renal vascular abnormalities. 29 In patients with the loin pain hematuria syndrome, it

37

AFFERENT ARTERIOLAR C3 DISEASE

has been postulated that the deposition of C3 in arterioles is a renal manifestation of hyper sensitivity,30 although in our series no patient has developed systemic manifestations of hypersensitivity. Thin basement membrane disease is also increasingly described and, although thinning is evident in a wide variety of renal disease,31.32 may, as we have demonstrated in isolation, be responsible for hematuria.1.16·21 However, the incidence varies, with some criteria dependent on generalized thinning,1.2o.21 some dependent on >50% involvement, IS and others with variable thickness but thinned segments. 16 Thus, we have grouped our patients into those with generalized thinning (< 200 nm, as has been accepted criteria for abnormal thinning of glomerular basement membrane)21 and those with segmental thinning of glomerular basement membrane. Not all patients with normal results of light and immunofluorescent microscopy in the current review had electron microscopy results available for review, and it is possible that the true prevalence of thin basement membrane disease may have been underestimated. Eleven patients with abnormalities ofthe glomerular basement membrane had coincident C3 deposition in afferent arterioles, and again the true incidence of coexistent abnormalities may be underestimated. In this group, we demonstrated no deterioration in renal function over 23 ± 20 months, despite persistence of hematuria in the majority of cases. This has also been observed by others,16.1S.21 although the observations have not

been universal. 19 In the two patients studied who had generalized thinning of glomerular basement membranes in association with afferent arteriolar complement deposition, family members had hematuria consistent with an autosomal dominant pattern of inheritance. Others have confirmed the familial nature of thin basement membrane disease ls ; although in our study no biopsies were performed on the affected family members, and thus hematuria may equally have been caused by complement deposition in afferent arterioles. In summary, afferent arteriolar C3 deposition in association with normal light and electron microscopic examination is a distinct pathological entity, with family studies suggesting an autosomal dominant inheritance. The predominant presenting features are loin pain and hematuria. Differentiation from thin basement membrane disease is not possible based on clinical criteria alone, and although the two diseases may coexist, they represent distinct pathological entities. Both conditions are associated with a good medium- to long-term prognosis with respect to renal function, although hematuria persists. The diagnosis is important because it allows reassurance of a good prognosis, particularly relevant to those who have been refused life insurance or employment on the basis of urinalysis findings. The treatment of coexistent loin pain remains a problem in a small number of cases, with autotransplantation being the only effective treatment.

REFERENCES 1. Trachtman H, Weiss RA, Bennett B, et al: Isolated haematuria in children: Indications for a renal biopsy. Kidney Int 25:94-99, 1984 2. Barcelo P, Covarsi A, Ballarin JA, et al: The isolated C3 in the 'vascular pole' in patients with primary haematuria. Proc Eur Dial Transplant Assoc 19:692-695, 1982 3. Saint-Andre Jp, Touzard D, Houssun A, et al: Les depots mesangiaux isoles de C3 peuvent - lis etre responsables d 'une nephropathie hematurique. Ann Pathol 2: 159-162, 1982 4. Grekas D, Morley AR, Wilkinson R, et al: Isolated C3 deposition in patients without systemic disease. Clin Nephrol 21:270-274, 1984 5. Orfila C: Mesangial isolated C3 deposition in patients with recurrent or persistent haematuria. Lab Invest 43:1-8, 1980 6. Naish PF, Aber GM, Boyd WN: C3 deposition in renal arterioles in the loin pain and haematuria syndrome. Br Med J 3:746, 1975 7. Nicholls AJ, Muirhead N, Edward N, et al: Loin pain and haematuria in young women: Diagnostic pitfalls. Br J Urol 54:209-211, 1982

8. Fletcher P, AI-Khadir AA, Parsons V, et al: The pathology of intrarenal vascular lesions associated with the loin pain haematuria syndrome. Nephron 24:150-154, 1979 9. Sheil AGR, Thomas MAB, Ibels LS, et al: Renal autotransplantation for severe loin pain/haematuria syndrome. Lancet 2:1216-1217,1985 10. Sheil AGR, Ibels LS, Pollock C, et al: Treatment of the loin pain/haematuria syndrome by renal autotransplant. Lancet 2:907-908, 1987 11. Perry GJ, George CRp, Field MJ, et al: Thin membrane nephropathy in patients with haematuria. Kidney Int 33: 137, 1988 (abstr) 12. Wilson C: Immunohistopathology of the kidney, in Rose N, Friedman W (eds): Clinical Immunology, Washington, DC, American Society for Microbiology, 1976, pp 692-700 13. Gyory AZ, Kesson AM, Talbot JM: Microscopy of urine-Now you see it, now you don't! Am Heart J 99:537538, 1980 14. Wallace AC: IgA nephropathy. Pathology 13:401-403, 1981 (editorial) 15. Grunfeld Jp, Bois EP, Hing1ais N: Progressive and non-

38 progressive hereditary chronic nephritis. Kidney Int 4:216228, 1973 16. Coleman M, Haynes ND, Dimopoulos P: Glomerular basement membrane abnormalities associated with apparently idiopathic haematuria. Hum Pathol 17: 1022-1030, 1987 17. Children with symptomless haematuria. Lancet I: 1450, 1984 (editorial) 18. Yoshikawa N, Hashimoto H, Katayama Y: The thin glomerular basement membrane in children with haematuria. J Pathol 142:253-257, 1984 19. Dische FE, Weston MJ, Parsons V: Abnormally thin glomerular basement membranes associated with haematuria, proteinuria or renal failure in adults. Am J Nephrol 5: 103-109, 1985 20. Yum M, Bergstein JM: Basement membrane nephropathy: A new classification for Alport's syndrome and asymptomatic haematuria based on ultrastructural findings. Hum Pathol 14:996-1003, 1983 21. Abe S, Amagasaki Y, Iyori S, et al: Thin basement membrane syndrome in adults. J Clin Pathol 40:318-322, 1987 22. Makori Y: Deposits of immunoglobulins and C3 in the walls of human renal arteries. Clin Exp Immunol 43:254-259, 1980 23. Berger J, Moel LH, Yanarra H: Complement deposition in the kidney. Adv Nephrol 4:37-48, 1974 24. Weber MA, Stokes GS, Moses M, et ai: Hypertension

POLLOCK ET AL

survey in a central city free chest x-ray clinic. Med J Aust 11 :529-533, 1973 25. Ayoub E: Benign recurrent haematuria. Am J Dis Child 109:217-223, 1965 26. Glasgow E: Symptomless haematuria in childhood. Br Med J 2:687-692, 1970 27. Williams JD, Czop JK, Abrahamson DR, et al: Activation of the alternative complement pathway by human glomerular basement membrane. J Immunol 133:394-399, 1984 28. Boyce NW, Holdsworth SR: Evidence for direct renal injury as a consequence of glomerular complement activation. J Immunol 136:2421-2425, 1986 29. Hammar Sp, Bloomer HA, McCloskey D: Adult hemolytic uremic syndrome with renal arteriolar deposition of IgM or C3. Am J Clin Pathol 70:434-439, 1978 30. Bell GM, Williams P, Thomson D: Is the loin pain and haematuria syndrome a renal manifestation of hypersensitivity? Lancet 1:340, 1984 31. Linn JT, Maeda H, Haitori M: Discontinuation of glomerular basement membrane in renal diseases of children. Correlation with glomerular crescents and urinary findings. J Clin Electron Microsc 18: 179-192, 1985 32. Reznik VM, Griswold WR, Vazquez MD: Glomerulonephritis with absent glomerular basement membrane antigens. Am J Nephrol 5:296-298, 1985