Diabetes Research and Clinical Practice, 7 (1989) 121-125 Elsevier 121

DIABET 00301

Increased intraglomerular thrombin formation in diabetic microangiopathy

Hisaya Tada, Koji Kuboki, Yasuo Oshima, Shoichi Takeuchi and Sho Isogai Second Department of Internal Medicine, Toho University School of Medicine, Tokyo 143, Japan

(Received 17 August 1988) (Revision received 5 January 1989)

(Accepted 16 March 1989)

Summary

Estimations of soluble fibrin monomer complexes (SFMC) in plasma are a convenient index of thrombin activation. Renal venous-arterial differences in plasma SFMC concentrations were determined in 16 randomly chosen diabetic patients by sampling directly and simultaneously from the renal artery and vein according to the method of Seldinger. In all subjects, SFMC concentrations were higher in the renal vein than in the renal artery, indicating that the kidney is an important source of SFMC. Venous-arterial differences were markedly elevated in patients with severe renal and retinal microangiopathy coupled with hypertension. The hypothesis is advanced that elevated plasma SFMC levels lead to abnormal fibrin deposits in lesioned glomeruli and retinal vessels. It is postulated that plasma SFMC may be a useful parameter for the assessment of diabetic vascular complications.

Key words: Intraglomerular thrombin formation; Soluble fibrin monomer complexes; Microangiopathy, diabetic; Hypertension

Introduction venous-arterial differences of soluble fibrin monomer complexes (V-A SFMC) were deter-

We have previously shown that measurements of mined to assess IGTG in a randomly selected soluble fibrin monomer complexes (S FMC) are a group of 16 diabetic patients. We have found that convenient and reliable index of thrombin activa- IGTG was markedly increased in patients show- tion [ 1,2]. It appears from a search of the litera- ture that rate of intraglomerular thrombin genera- tion (IGTG) has not hitherto been quantitatively measured in the majority of studies dealing with the development of renal disease. Therefore, renal

ing severe proteinuria, and hypertension,

Subjects and methods

proliferative retinopathy,

Address for correspondence: H. Tada, MD, Second Department of Internal Medicine, Toho University School of Medicine, Tokyo 143, Japan.

Clinical details of the subjects are given in Table 1. Sixteen diabetics (10 males and six females) ranging from 40 to 70 years of age were

0168-8277/89/$03.50 0 1989 Elsevier Science Publishers B.V. (Biomedical Division)

TABLE 1

Clinical characteristics of subjects

Case Age Sex Duration of Therapy Urinary protein Serum Retinopathy Blood pressure

(years) diabetes

(years) (g/24 h)

1 63 M 11 2 65 F 6 3 IO M 12 4 66 F 14 5 65 M 9 6 64 F 10 7 70 M 19 8 40 M 13 9 65 M 6

10 50 F 5 11 64 F 17 12 54 M 33 13 63 M 11 14 44 M 18 15 63 F 20 16 49 M 11

creatinine

(mg/dl)

Sulfonylurea so.5 Insulin 1.4 Sulfonylurea so.5 Insulin 0.8 Insulin 5 0.5 Sulfonylurea 1.0 Insulin 1.8 Insulin 0.8 Sulfonylurea 3.9 Sulfonylurea 0.9 Insulin 1.0 Diet alone 1.5 Insulin 4.4 Insulin 8.2 Insulin 6.3 Insulin 1.9

1.0 Pre-proliferative 1.0 Background 1.4 Background 1.6 Proliferative 1.0 Proliferative 1.6 Pre-proliferative 1.9 Background 0.9 Pre-proliferative 1.2 Background 0.9 Background 1.0 Background 1.7 Proliferative 1.6 Proliferative 1.6 Proliferative 1.5 Proliferative 2.1 Proliferative

140170 162/80 150/100 120/80 128/80 150/70 118/83 166192 168/104 170/l 12 135/80 164190 180/88 162190 174154 180/100

122

involved in this study. The duration of diabetes varied from 5 to 33 years and no preselection was made on the basis of therapy, severity of retinop- athy, proteinuria, serum creatinine, or degree of hypertension. A patient was considered to be hypertensive ifthe blood pressure was persistently higher than 160/95 mm Hg.

Data were evaluated when a representative number of patients was obtained in each of the following three groups: urinary protein excretion rate less than 0.5 g/24 h - three patients; 0.5-3.0 g/24 h - eight patients; more than 3.0 g/24 h - five patients.

Blood samples were drawn from the right artery and vein by Seldinger’s method [ 31 for measure- ment of SFMC, after an overnight fast. Simul- taneously, peripheral blood was drawn from each patient for determination of serum creatinine, fasting plasma glucose and fibrinogen. Patients were fully instructed about procedures and informed consent was obtained. Because of the complexity of the procedure, the number of patients by necessity remained small. Glucose concentrations were determined by the glucose

oxidase procedure using a Beckman Auto- analyzer (Beckman Instruments Inc., Fullerton, CA, U.S.A.). SFMC was determined according to the method of Shoda and Masukata [4] with the following modifications. Veronal buffer (10 mM, 5 ml) pH 8.0, containing 0.1% prota- mine sulfate and 0.3 M NaCl was heated to 37°C and added to 0.1 ml titrated plasma. After incuba- tion at 37°C for 30 rnin, the solution was filtered through ‘Separax’ cellulose acetate paper (Jookoo Co., Ltd., Tokyo). The filter paper was sub- sequently washed with the same buffer, then with 0.1 M phosphate buffer, pH 7.2, then stained with 0.8% Ponceau 3R in 10% trichloroacetic acid. After destaining with 2% acetic acid, the protein- bound pigment was quantitated spectropho- tometrically at O.D. 500 nm using a purified fibrin monomer as a standard. Protein in 24-h urine collections was quantitatively determined by the method of Kingsbury et al. [ 51.

Data were expressed as mean _+ SD and/or mean + SE. Statistical analysis was done by Bonferroni’s test (Figs. 1 and 2) and Student’s t-test (Fig. 3).

123

TABLE 2

Concentration of fasting peripheral plasma glucose, fibrinogen and soluble fibrin monomer complexes (SFMC) in the renal vein and artery arranged by value of renal venous-arterial differences of soluble fibrin monomer complexes

Case Fasting peripheral venous blood

Plasma glucose

(mgidl)

Renal artery (A) Renal vein (V) Renal V-A SFMC SFMC SFMC

(mg/dl) (mg/dl) (mg/dl) Fibrinogen

(mg/dl)

1 163 2 152 3 142 4 139 5 84 6 114 I 162 8 124 9 156

10 175 11 102 12 103 13 92 14 170 15 132 16 131

Mean + SE 133.8 + 6.9

251 4.6 4.9 0.3 395 15.7 16.0 0.3 310 9.2 10.0 0.8 364 4.7 5.6 0.9 413 19.0 20.0 1.0 464 4.1 5.8 1.1 376 1.3 8.8 1.5 289 13.8 15.5 1.7 286 1.2 8.9 1.7 280 3.0 5.4 2.4 286 12.3 14.9 2.6 439 5.9 10.0 4.1 368 11.5 16.6 5.1 466 10.5 16.3 5.8 549 14.9 23.8 8.9 530 18.3 34.7 16.4

387.0 f 23.2 10.2 + 1.3 13.6 k 1.9 3.4 + 1.0

Results

Urinary protein excretion rates shown in Table 1 correlated positively (r = 0.61, P < 0.05) with the levels of serum creatinine.

Table 2 gives the levels of fasting plasma glu- cose, fibrinogen and V-A SFMC for the 16 patients studied. All of these were reasonably well controlled at the time of sampling since peripheral fasting plasma glucose concentrations did not exceed 175 mg/dl. In spite of a considerable spread of values in all subjects, SFMC concentra- tions were always higher in the renal vein than in the renal artery. The values of V-A SFMC corre- lated positively (r = 0.59, P -c 0.05) with the con- centrations of fibrinogen.

The relationship between V-A SFMC and the three degrees of proteinuria is given in Fig. 1, and similar plots of V-A SFMC against diabetic

I-

i-

I-

I -I.

I *-A .

.

.

.

!I ii l

80.5 0.5<-s3.0 3.0<

Urinary protein (g/24h)

Fig. 1. Renal venous-arterial differences of soluble fibrin monomer complexes (V-A SFMC) in diabetic patients with

various degrees of proteinuria. * P < 0.05.

124

zo-

z 15- \ M E

g lo-

t:

? >

5-

O-

I -,-

f-----*---4 .

. : t

. . :

+ .

. . 3 : . I

Simple Pre-prollferatlve Proliferative

Diabetic retinopathy

Fig. 2. Renal venous-arterial differences of soluble fibrin monomer complexes (V-A SFMC) in patients with diabetic

retinopathy. * P < 0.05.

.

.

.

.

.

; : , 1

Without With

Hypertension

Fig. 3. Renal venous-arterial differences of soluble fibrin monomer complexes (V-A SFMC) in diabetic subjects

without and with hypertension. *P < 0.05.

retinopathy and hypertension are provided in Figs. 2 and 3.

These data indicate that diabetes with severe renal and retinal angiopathy and hypertension is usually associated with high venous-arterial dif- ferences of SFMC.

Discussion

It has been shown in previous experiments that diabetic nephropathy is associated with an en- hancement of IGTG because the urinary excre- tion of fibrinogen degradation products increased [6]. This observation has neither established the existence of a causal relationship nor provided us with information concerning the source of the enhanced SFMC levels. Thus, direct sampling from the renal vein and artery has been an important feature of the present series of experi- ments in that it has permitted us to establish the kidney as one of the significant sources of SFMC.

We have been unable, so far, to perform V-A SFMC measurements on normal healthy adults, but it is noteworthy that all of our 16 patients, including those with negligible vascular compli- cations, showed positive V-A differences. Thus, it is perhaps safe to generalize and assume that even under normal conditions the kidney is a site of active SFMC production. The present study pro- vided evidence that SFMC production in glome- ruli was increased in diabetic nephropathy. This finding strongly suggests that intraglomerular overproduction of SFMC was closely related with diabetic glomerular lesions. Moreover, structural changes in the glomerular membrane lead to deposition of fibrin and histologic studies have shown endothelial and mesa&al enlargement, nodule formation, and irregularities in the struc- ture of the basement membrane [7]. Therefore, our finding that IGTG increases as proteinuria progresses in severity (Fig. 1) is in line with histo- logical evidence.

Although we do not know whether enhanced SFMC production by the kidney is a primary event or a secondary event in response to lesions

of other origin, it stands to reason that high SFMC levels will also lead to abnormal fibrin deposits in lesioned tissues other than the kidney. Soluble FMC are pH-dependent unstable compounds with a short biological half-life [ 8,9]. Therefore, if fibrin is deposited in excess in retinal and vitreous vessel walls, this may contribute to tissue hypoxia by impaired blood flow. In addi- tion, since increased red cell aggregation has been known in cases with supranormal levels of fibrinogen [lo], it was suspected that the hyper- fibrinogenemia observed in the patients with enhanced SFMC production in the kidney also contributes to the progression of diabetic retinop- athy by sludged flow. Thus, the data shown in Fig. 2 suggest the existence of a close relationship between high SFMC levels and proliferative retinopathy.

Our observation that hypertensive diabetic patients had higher SFMC levels than their normotensive counterparts (Fig. 3) may be par- tially explained by the reference to renal hemo- dynamics, in particular to renal plasma flow and transcapillary hydraulic pressure [ 111 as factors determining SFMC production. If the latter is in excess, the potentially deleterious effects have already been mentioned.

We had too few subjects to make cross-corre- lations between nephropathy as defined by pro- teinuria and hypertension on the one hand and the degree of retinopathy on the other hand. It would be interesting to see to what extent such an inter- dependence relates to elevated SFMC levels. Previous studies from our laboratory have shown that rigid control of the metabolic condition is capable of lowering high plasma SFMC concen- trations [ 121. In the present series, only eight out of 16 diabetic patients showed distinctly normal SFMC values, i.e., less than 10 mg/dl in the renal artery, in spite of reasonable plasma glucose levels in the majority of patients. This raises the well- known question of what is really understood by

125

adequate control of the diabetic condition. It may well be that a better understanding of the signifi- cance of plasma SFMC levels may make such measurements a useful adjunct to the accurate assessment of metabolic control, especially in relation to diabetic vascular complications.

References

1

2

3

4

5

6

I

8

9

10

11

Isogai, S., Kameyama, M. and Urayama, T. (1982) Plasma soluble fibrin monomer complexes in diabetic microangiopathy. Jap. J. Med. 21, 96-100. Isogai, S., Kikuchi, K., Kameyama, M., Oshima, Y., Urayama, T. and Komoto, M. (1982) Plasma soluble monomer complexes in the development of diabetic retinopathy. Tohoku J. Exp. Med. 137, 409-413. Seldinger, S.I. (1953) Catheter replacement of the needle in percutaneous arteriography. Acta Radiol. 39,368-376. Shoda, H. and Masukata, E. (1980) Calorimetric deter- mination of soluble fibrin monomer complexes. J. Med. Sot. Toho Univ. 27, 669-673. Kingsbury, F.B., Clark, C.P., Williams, G. and Post, A.L. (1926) The rapid determination of albumin in urine. J. Lab. Clin. Med. 11, 981-989. Chan, V., Yeung, C.K. and Chan, T.K. (1982) Anti- thrombin III and fibrinogen degradation product (frag- ment E) in diabetic nephropathy. J. Clin. Pathol. 35, 661-666. Farquhar, A., Macdonald, M.K. and Ireland, J.T. (1972) The role of fibrin deposition in diabetic glomerulosclero- sis: a light, electron and immunofluorescence microscopy study. J. Clin. Pathol. 25, 657-667. Shainoff, J.R. and Page, I.H. (1964) Fibrin deposition and arteriosclerosis. Nat. Conf. Cardiovasc. Dis. 2, 374-375. Kikuchi, K. (1982) Plasma soluble fibrin monomer com- plexes in the development of diabetic retinopathy. J. Japan Diab. Sot. 25, 909-914. Littl, H. (1976)The role of abnormal hemorrheodynamics in the pathogenesis of diabetic retinopathy. Trans. Am. Ophthalmol. Sot. 44, 573-636. Parving, H.H., Viberti, G.C., Keen, H., Christiansen, J.S. and Lassen, N.A. (1983) Hemodynamic factors in the genesis of diabetic microangiopathy. Metabolism 32, 943-949.

12 Tada, H., Takeuchi, S., Takagi, M. and Isogai, S. (1987) The effect of glycemic control on plasma soluble fibrin monomer complexes and fibronectin in diabetic patients. Tohoku J. Exp. Med. 153, 145-150.