Eu,oo.oNuclear Medicine

Eur Nucl Med (1983) 8:60-64

Journal of

© Springer-Verlag 1983

Abnormal Reticuloendothelial Function in Patients with Active Vasculitis and Idiopathic Membranous Glomerulopathy A Study with 99mTc-Labeled Heat-Damaged Autologous Red Blood Cells Fokko J. van der Woude, D. Albertus Piers, Marijke van der Giessen, Philip J. Hoedemaeker, T. Hauw The, and Gjalt K. van der Hem Departments of Internal Medicine, Nuclear Medicine and Pathology, State University Groningen, NL-9713 EZ Groningen, The Netherlands

Abstract. Reticuloendothelial function was assessed in

11 patients with systemic lupus erythematosus, 8 patients with Wegener's granulomatosus, and 20 patients with idiopathic membranous glomerulopathy by using autologous 99mTc-labeled heat-damaged red blood cells. With this method organ uptake could be measured by quantitative scintigraphy. There was no relation between the T~ of the blood disappearance curve and the T± of the splenic2uptake curve. The 7"1of the blood disappearance curve was normal in all three patient groups. However, there was a significant shift from spleen to liver uptake in patients with active systemic lupus erythematosus, active Wegener's granulomatosus, and membranous glomerulopathy in comparison with a control group. There was no relation with age, level of circulating immune complexes, complement level, kidney function, or immunosuppressive treatment. We conclude that an increase of the liver component of reticulo-endothelial function may compensate abnormalities in splenic function. This stresses the importance of quantitative scanning to detect such abnormalities. The study provides evidence for disease related hyposplenism in patients with active systemic lupus erythematosus, active Wegener's granulomatosus, and membranous glomerulopathy. •

2

2

,

Introduction

Involvement of circulating immune complexes in the pathogenesis of vasculitis and glomerulonephritis was suggested by studies on experimental serum sickness about 30 years ago (Germuth 1953). It was shown that in animal models saturation of the reticuloendothelial system contributes to a prolonged circulation of immune complexes (Haakenstad and Mannik 1974) and their subsequent tissue deposition. Frank et al. (1979) found defective Fc receptor function in the spleen in patients with systemic lupus erythematosus (SLE). Reduced splenic uptake of heat damaged red blood cells was described in five patients with SLE and active nephritis and in three patients with Wegener's granulomatosis (WG) (Lockwood et al. 1979). The aim of our study was to determine if a saturated reticuloendothelial system of the spleen exists in vasculitis Offprint requests to: F.J. van der Woude, M.D., Department of Internal Medicine, State University Hospital, Oostersingel 59, 9713 EZ Groningen, The Netherlands

0340-6997/83/0008/0060/$01.00

and glomerulonephritis and whether such a situation is related to disease activity or levels of circulating immune complexes. We therefore performed quantitative scintigraphy with autologous 99mTc-labeled heat damaged red blood cells (HDRBC) in 12 patients with SLE, 8 patients with WG, and 20 patients with idiopathic membranous glomerulopathy (MGP). The latter group was chosen because the involvement of circulating immune complexes in the pathogenesis of the disease (Couser 1981) is doubted and data about RES function in this group are lacking. Materials and Methods Patients

Twelve patients with SLE (11 female, 1 male) satisfied four or more of the ARA preliminary criteria (Cohen et al. 1971). Antibodies against dsDNA, detected in the Crithidia assay were present in all patients (mean age 37 years). Patients with an active urinary sediment, a proteinuria greater than 1 g/day or a creatinine-clearance less than 50 ml/min were considered to have renal involvement. In four of these patients a kidney biopsy had been performed, which revealed diffuse proliferative glomerulo-nephritis in three patients and membranous glomerulopathy in one. Four patients did not fulfil the criteria for renal involvement. According to the number of organ systems clinically involved and to the severity and progression of renal involvement the disease activity was determined before quantitative organ scanning was done. Four patients were treated with corticosteroid therapy alone, three patients were on a combination of corticosteroid and cyclophosphamide, and one patient on azathioprine alone. Three patients received only non-steroidal antiinflammatory drugs and one patient remained untreated• Eight patients with WG (three female, five male) were selected on the basis of manifestations in the upper airways, the lungs, the kidneys, and other organs or tissues, considered either histologically typical of WG or clinically compatible with this diagnosis (Godman and Churg 1954). Mean age was 49 years. In seven of these patients a kidney biopsy had been performed, which revealed focal and local necrosis of the glomeruli and extracapillary proliferation in all. In another patient a lung biopsy specimen showed granulomata and vasculitis, she had no clinical signs of renal involvement. Again clinical activity was assessed before quantitative scintigraphy, five patients had active

61 disease at the time of investigation. Three patients were untreated, three patients received prednisolone and cyclophosphamide, and two patients received cyclophosphamide alone. Twenty patients (6 female, 14 male) had typical histological features of M G P in their renal biopsy specimens in the absence of systemic disease or known precipitating factors. Antinuclear antibodies were negative in all patients. The mean age was 46 years (range 19-69 years). The glomerular filtration rate was 84 ml/min (range 2-152 ml/min). Proteinuria (biuret method) was negative in 8 patients at the time of investigation; in 12 patients proteinuria ranged from 1.3 to 10 g/24 h (mean 3.8 g/24 h). Of these 12 patients, 4 were treated with the combination of hydrochlorothiazide and indomethacin to decrease the proteinuria (Arisz et al. 1976). Patients in this group were not treated with corticosterioids or immunosuppressive drugs. Patients with an abnormal spleen or liver size were not included in this study. Two SLE patients and one W G patient underwent quantitative scintigraphy twice. Informed consent was obtained from all patients. Control normal subjects (n = 9) were recruited from laboratory staff (one female, eight males). The mean age was 28 years.

obtained by dextran sedimentation of heparinized blood, were incubated with patient sera for 60 rain at 37 ° C. After washing the cells cytocentrifuge slides were made and fixed in acetone. Intracellular IgG and complement were detected using specific antisera and FITC-conjugated sheep anti-rabbit immunoglobulin serum in a double layer system. Cells containing at least six fluorescent granules were regarded as positive. A total of 1,000 P M N cells in fields chosen at random were observed at 10 x 40 magnification. Two observers without prior knowledge of the diagnosis read and scored the results independently.

HDRBC Test Citrate blood (5 ml) was collected from each patient and the samples were centrifuged for 5 rain at 1,500 g in 20 ml saline• The packed cells were then heated for at least 20 rain at 49.5 ° C. The degree of damage was standardized by using plasticity indices measured with the erythrocyte filtrability test as described by Teitel (Teitel 1965); appropriately damaged cells showed an infinite plasticity index. Thereafter the cells were incubated with about 500 gCi 99mTcfor 5 rain at room temperature. A volume of freshly prepared 1% stannous chloride was added and the mixture allowed to stand for 5 rain. The cells were then washed twice in normal saline, resuspended with an equal volume of saline and reinjected. Labeling efficiency was about 90%. Reinjection was done in 60 s, blood samples were taken at 0, 3, 8, 13, 18, and 23 min and the blood radioactivity was determined. The half life (T1) was calculated from the samples taken at 3-23 man. The rejected dose was determined by measuring the syringe before and after injection in a common dose calibrator. The patients lay in a supine position on top of a large-field-of-view gamma camera, so that the camera viewed the liver and spleen posteriorly• Digital images were recorded in l-rain frames during 60 min after injection. Curves using a region of interest technique over spleen and liver areas (corrected for physical decay of the radionuclide) were produced. The T,_ of the spleen uptake curve and the 1-h liver and spleen ubtake in counts per minute per gCi adminstered were calculated. ~ ( liver uptake cpm/_tCi -1 '] x 100%] The quotient L\@leen u p ~ e ~ q m ; C ~ -i]

Immune Complex Tests Blood samples were drawn on the day of the H D R B C test. Blood was allowed to clot at room temperature and the serum was stored at - 8 0 ° C. C 3 and C 4 levels were measured by radial immunodiffusion with monospecific antisera. Circulating immune complexes were assayed by three different tests: Polyethylene glycol (PEG) precipitation, Clq-ELISA, and the indirect granulocyte phagocytosis test (IGFT), PEG-precipitation was done as follows: 50 gl serum was mixed with 150 ~tl 0.08 M CaC12 and 1 ml 5% PEG in 0.1 M borate buffer, pH 8.4. The mixture was left at 4 ° C overnight and centrifuged for 3 rain at 3,500 g. The clear supernatant was carefully removed and the pellet dissolved in l ml 0.1 N NaOH. Optical density (OD) was measured in a 0.5-cm cuvette at 280 nm. The test was performed in duplicate. The Clq-ELISA test was recently developed in our laboratory and has been described more extensively elsewhere (Van der Giesse Giessen et al. 1980). HRPOqinked antihuman IgG is used to detect immune complexes bound to Clq-coated microtiter plates. The detection limit of heataggregated human IgG appears to be less than I ~g/ml, the test is particularly sensitive for immune complexes formed in antigen excess• The indirect granulocyte phagocytosis test (Van Wingerden et al. 1978) was done as follows: normal granulocytes,

•

-

2

.

.

was called the LS ratio. For statistical analysis the bilateral Wilcoxon test was used. Results There were no significant differences in the T~ blood disappearance curve and T1 spleen uptake curvd between the 2

Table 1. ~ blood disappearance curve and T~ spleen uptake curve (min) in the controls and the patient groups studied Controls

SLE Active

n T1 blood disappearance curve, mean R~nge T~ spleen uptake, mean Range

9 13.3 9.4-17.5 7.6 6.0-9.5

WG Not active

5 9 13.5 11,5 1 2 . 4 - 1 4 . 6 7,6-15.7 8.9 7.0 7.7-11.3 3.9-11.1

There are no significant differences in these parameters between the groups

Active

MGP Not active

5 4 16.3 10.7 1 0 . 0 - 2 5 . 0 9.3-14.0 6.8 8.7 2.0-10.3 9.5-11.0

20 11.3 7.0-15.4 7.7 3.6--16.5

62 counts

)er

minl~Ci

cpm JCi-I

1000,

1200 s

800

1000

ooo

0

600

2£).. 800"

400,

s L L

200 /

0 0

(-

•

(l.,

0

80

5_ 600"

,x

0

I11

O O o

o ~o O

/

400"

i0 2'0 3'o ~0

go

Fig. l. Liver (L) and splenic (S) uptake curves in SLE patient no. 1. during active ( - - - ) disease and 3 months later ( - - ) when all symptoms were resolved and renal function was improved. The spleen uptake increased from 420 to 820 cpm ~tCi, liver uptake decreased from 290 to 220 cpm gCi. In both situations there is a normal T~ blood disappearance curve (12.4 and 13.5 rain)

L (%) S 500t 200J. 0

1001 0

80. 0

0 O

40. ¥

20.

i controls

SLE

WG

~

o

go

time ( m i n u t e s )

60.

o

MGP

/ liver uptake cpm pCi o Fig. 2. The LS ratio i x 100 Yo/in controls \spleen uptake cpm gCi / and in patients groups. The open circles are values obtained in patients with active disease. The patient (*) in the SLE group has the membranous form of SLE glomerulonephritis patient groups and the control group (Table 1). Neither was there a relation between T_, blood disappearance and T~ spleen uptake curve ( r = 0.481. In all patients the blood, spleen, and liver curves were of the form as previously described (Lockwood etal. 1979; Peters etal. 1982). An example of liver and spleen curves is given in Fig. 1. Liver uptake always took place during the first 15 min after injec-

200"

0

cpm ~JCi-1 o

16o

26o

360

460

s6o

liveruptake Fig. 3. The correlation between spleen uptake and liver uptake after 1 h (cpm/gCi) in the four studied groups (o = controls, A=SLE, zx=WG, o = M G P )

tion whereafter a plateau phase was reached. The LS ratio was increased in all the patient groups and the increase was more pronounced when SLE of W G were active (Fig. 2). This LS ratio was significantly ( P < 0.01) increased in patients with active SLE and active W G in comparison with the controls. The patient group with M G P also had a significantly increased LS ratio (P<0.01). In the four groups studied there was an inverse relation between spleen uptake and liver uptake (Fig. 3); liver uptake increased as spleen uptake decreased. Total liver uptake after 1 h was significantly ( P < 0 . 0 1 ) increased in the patients with SLE, WG, and MGP. Again this increase was more pronounced when SLE and W G were active. The reverse was true for the spleen uptake, as can be seen in Fig. 3. There was no relation between the parameters of the H D R B C test and age, kidney function, proteinuria, C3 and C4 levels, or immunosuppressive treatment. To test the variability of the system the H D R B C test was repeated three times after 3-month periods in one of the controls. The LS ratio remained 0.18. Two patients with SLE were tested twice, the results obtained in patient no. 1 are shown in Fig. 1. Patient no. 2 had a decrease in LS ratio from 1.4 to 0.95. One patient with W G (no. 6) was tested twice, the LS ratio decreased from 3.0 to 0.43 as disease activity disappeared. The results of the immune complex assays are shown in Fig. 4. Although there often was a good correlation between disease activity and levels of circulating immune complexes in individual patients with W G or SLE, the transversal approach showed only increase in immune complexes in the granulocyte phagocytosis test and P E G precipitation in these patient groups as a whole. In M G P only the P E G test showed higher immune complex levels. There was no relation between immune complex levels and T_,blood clearance, T1 spleen uptake, LS ratio, or spleen or~liver uptake after Ifa.

63 GFT(no.pos.cetls) 800t o

400] 0m

T

ClqELISA0Jg/mt)

(200)

oo

+.

0

PEG(OD280) 1200800 el

400

o

t controls

SLE

°

+

WG

v

MGP

Fig. 4. Circulating immune complexes in the studied patient groups. The top panel shows the results of the IGFT, the middle panel of the Clq-ELISA, and the bottom panel of the PEG precipitation test. The open circles are values obtained in patients with active disease Discussion The clearance of H D R B C is considered to be a well-standardized test of splenic function (Petit 1977), and a prolonged T, of the blood disappearance curve has been reported in'splenic atrophy (Marsh et al. 1966), inflammatory bowel disease (Petit 1977), SLE (Lockwood et al. 1979), and active rheumatoid arthritis (Williams et al. 1979). 99mTc labeling has important advantages over 51Cr labeling in studies with these cells. 51Cr and 99myc bind exactly to the same components of the red cell (Rehani and Sharma 1981), but 99myc has the advantage of a shorter half-life (about 6 h) and optimal gamma ray emission. This implies that it would be possible to perform quantitative scintigraphy with an acceptable radiation dose (spleen dose 1 rad, total body dose 0.1 rad). Quantitative seintigraphy of liver and spleen is necessary, since there is no correlation between the Ta of the blood disappearance curve and the T~ of the spleer~ uptake curve and since the liver componen~ of the RES function seems to be more important in the patient groups in comparison with controls. In this study no relation was established between T~ of the blood disappearance curve and disease activity an patients with non-renal and renal SLE. The results confirm and extend the findings of Elkon et al. (1980) who studied a patient group with non-renal SLE. We noticed a statisti•

2

cally significant relation between LS ratio and disease activity in SLE and WG patients. This may give us the possibility of using this rather simple investigation for measuring disease activity in SLE and WG. The test could perhaps be used to monitor therapy or to differentiate between infection and relapse. However, we feel that more data are required to draw final conclusions about the possibility of such a practical application of the technique. In patients with MGP there also was a significant increase in LS ratio. As far as we know this is the first time that in MGP, a disease which is supposed to be localized in the kidney only, abnormalities in other organs have been described that are not directly related to decreased kidney function, proteinuria, or hypertension. Not much is known about the factors which influence organ localization of HDRBC. Because it has been suggested that sequestration in the spleen is related to reduced red cell deformability (Mohandas et al. 1979), we used plasticity indexes (Teitel 1965) to standardize the degree of damage of these cells. A difference in cell rigidity was thus eliminated as a cause for the differences between the studied groups. Theoretically, abnormal spleen histology could be another cause of the observed phenomena. Patients with a gross abnormal spleen or liver size, however, were not included in this study and although it is difficult to exclude vasculitis of the spleen vessels in the SLE and WG patients, this seems to be a rather unlikely cause for the similar increase in LS ratio which was observed in the MGP group. Spleen and liver blood flow are important factors in the organ distribution of the injected cells, and in a recent study Peters et al. (1982) could indeed demonstrate that blood flow is important for spleen uptake of HDRBC. However, irreversible trapping seems to be an important factor as well. If liver uptake is flow dependent, one should expect a mathematical relation between the number of HDRBC in the circulation and the liver uptake. However, liver uptake reaches a plateau phase within 15 rain. Thus flow changes in the liver are not a likely cause for the increased liver uptake of HDRBC observed in the patient groups. No relation was found between serum immune complex levels and the H D R B C test. This does not add arguments pro or contra the hypothesis that saturation of Fc receptors on the surface of the macrophages in the spleen is important in the pathogenesis of immune complex mediated diseases. Perhaps HDRBC studies in animal models of immune complex disease could reveal the pathophysiology of the observed phenomena in this study. Regardless of the mechanisms involved, this study demonstrates disease related abnormalities of RES function in spleen and liver in patients with active SLE, active WG, and MGP. Acknowledgements. The skilful technical assistance of Ms. A. van

Zanten and Mrs. R. Schwander-Boer is gratefully acknowledged. We also thank Dr. H. Beekhuis and Dr. A.J.M. Donker for valuable discussions and co-operation. This investigation was supported by grant C-229 from the Dutch Kidney Foundation (Nierstichting Nederland). References Arisz L, Donker AJM, Brenjens JRH, Van der Hem GK (1976) The effect of indomethacin on proteinuria and kidney function in the nephrotic syndrome• Acta Med Scand 199:121-125 Cohen AS, Reynold WE, Franklin EC, Kulka JP, Ropes MW, Shulman LE, Wallace SL (1971) Preliminary criteria for the

64 classification of systemic lupus erythematosus. Bull Rheum Dis 21 : 643-646 Couser WG (1981) What are circulating immune complexes doing in glomerulonephritis? N Engl J Med 304:1230.1231 Elkon KB, Sewell JR, Ryan PFJ, Hughes GRV (1980) Splenic function in non-renal systemic lupus erythematosus. Am J Med 69 : 80-82 Frank MM, Hamburger MI, Lawley TJ (1979) Defective Fc receptor function in lupus erythematosus. N Engl J Med 300:518 523 Germuth FG (1953) A comparative histologic and immunologic study in rabbits of induced hypersensitivity of the serum sickness type. J Exp Med 97:257-282 Giessen M van der, Dokter-Fokkens J, The TH (1980) A solid phase Clq-binding assay for measuring circulating immune complexes (Clq-ELISA). In: Developments in clinical biochemistry vol. 1 : Immunoenzymatic assay techniques. (Ed. Malvano R), Martinus Nijhoff, The Hague, p 223-231 Godman GC, Churg J (1954) Wegener's granulomatosis. Arch Pathol 58:533-535 Haakenstad AO, Mannik M (1974) Saturation of the reticuloendothelial system with soluble immune complexes. J Immunol 112:1939-1948 Lockwood CM, Worlledge S, Nicolas A, Cotton C, Peters D (1979) Reversal of impaired splenic function in patients with nephritis

or vasculitis (or both) by plasma exchange. N Engl J Med 300 : 524-530 Marsh GW, Lewis SM, Szur L (1966) The use of 51Cr labelled heat-damaged red blood cells to study splenic function. Br J Haematol 12:161-166 Mohandas N, Philips WM, Bessis M (1979) Red blood cell deformability and haemolytic anaemias. Semin Hematol 16:95-101 Peters AM, Ryan PJ, Klonizakis I, Elkon KB, Lewis SM, Hughes GRV (1982) Measurement of splenic function in humans using heat-damaged autologous red blood cells. Scand J Haematol, in press Petit JE (1977) Spleen function. Clin Haematol 6:639-645 Rehani MM, Sharma SR (1980) Site of tc-99m binding to the red blood cell: concise communication. JNM 21:676-678 Teitel P (1965) Disk sphere transformation and plasticity alteration of red blood cells. Nature 206:409-410 Williams BD, Lockwood CM, Pussel BA, Cotton C (1979) Defective reticuloendothelial function in rheumatoid arthritis. Lancet I:1311-1316 Wingerden I van, The TH, Giessen M van der, Rfimke Ph (1978) Demonstration of circulating immune complexes in melanoma patients by means of a granulocyte fagocytosis test. Protides of the Biological Fluids, Proceedings 1978, 26, 345-348 Received June 21, 1982 / September 4, 1982