Platelets, Coagulation and Fibrinolysis in Patients with Diabetic Retinopathy

Targeting carbonic anhydrase to treat diabetic retinopathy: emerging evidences and encouraging results

Diabetic retinopathy (DR) is the leading cause of vision loss among working-age populations in developed countries. Current treatment options are limited to tight glycemic, blood pressure control and destructive laser surgery. Carbonic anhydrases (CAs) are a group of enzymes involving in the rapid conversion of carbon dioxide to bicarbonate and protons. Emerging evidences reveal CA inhibitors hold the promise for the treatment of DR. This article summarizes encouraging results from clinical and animal studies, and reviews the possible mechanisms.

Pathobiology and cell interactions of platelets in diabetes

Diabetes is a well-recognised risk factor for atherosclerotic cardiovascular disease and in fact most diabetic patients die from vascular complications. The Diabetes Control and Complications Trial (DCCT) and the U.K. Prospective Diabetes Study (UKPDS) indicate a consistent relationship between hyperglycaemia and the incidence of chronic vascular complications in patients with diabetes. Platelets are essential for haemostasis, and abnormalities of platelet function may cause vascular disease in diabetes. Diabetic patients have hyperreactive platelets with exaggerated adhesion, aggregation and thrombin generation. In summary, the entire coagulation cascade is dysfunctional in diabetes. This review provides a comprehensive overview of the physiological role of platelets in maintaining haemostasis and of the pathophysiological processes that contribute to platelet dysfunction in diabetes and associated cardiovascular diseases, with special emphasis on proteomic approaches and leukocyte-platelet cross-talk.

Platelet prothrombinase activity, a final pathway platelet procoagulant activity, is overexpressed in type 1 diabetes: no relationship with mean platelet volume or background retinopathy

There is overwhelming evidence that platelets from diabetic individuals are hyperreactive, not only when microvascular complications are apparent, but already at an early stage of the disease. There is still controversy about the question of whether primary hyperreactive platelets may contribute to the origin or progression of microangiopathy or whether diabetic platelet hyperfunctionality is just a logical consequence of a continuous low-grade activation of platelets by contact with a diseased microvascular wall. As a consequence of platelet activation, the outer layer of its phospholipid membrane is more procoagulant than in the quiescent state, stimulating thrombin formation in plasma. This platelet function is called platelet procoagulant activity. We studied platelet prothrombinase activity (PPA), a final pathway platelet procoagulant activity of type 1 diabetic platelets, and looked for an eventual correlation with microvascular disease (background retinopathy) and mean platelet volume (MPV). Stypven clotting times (SCTs), reflecting PPA expression, and MPV of citrated platelet-rich plasma (PRP), were measured in 21 patients with type 1 diabetes-10 with and 11 without background retinopathy-under clinically acceptable metabolic control and compared them to 20 disease-free voluntary controls. We also compared PPA expression and MPV in diabetic individuals with and without retinopathy. With the SCT, a selective test adapted for studying PPA in PRP, we found hyperexpression of PPA in all diabetic patients. We found no difference in MPV between diabetic and control PRP. Comparing patients with and without background retinopathy we found no significant difference in PPA expression. From these results, we suggest that the phospholipid surface of diabetic platelets, more than the surface of normal control platelets, stimulate the expression of PPA. This diabetic platelet coagulant anomaly was not related to an increased platelet mass (higher MPV) nor to the presence of microangiopathy. We conclude that PPA hyperexpression is associated with patients with type 1 diabetes, already occurring in an early stage of the disease, and not necessarily a consequence of early-stage microvascular disease, because the anomaly is also demonstrable, in the same degree, in patients with diabetes without microangiopathy.

Association of plasma fibrinogen level and blood pressure with diabetic retinopathy, and renal complications associated with proliferative diabetic retinopathy, in Type 2 diabetes mellitus

AIM

To clarify the association of several clinical parameters, including plasma fibrinogen level, with diabetic retinopathy in patients with Type 2 diabetes mellitus (DM).

METHODS

A total of 294 Japanese patients with Type 2DM were studied; 53 patients with no diabetic retinopathy (NDR), 90 with background diabetic retinopathy (BDR), and 151 with proliferative diabetic retinopathy (PDR). Multiple logistic regression analysis was performed to assess variables independently associated with diabetic retinopathy in two settings: presence of retinopathy of any severity and presence of advanced retinopathy.

RESULTS

The following parameters were identified as independent factors associated with the presence of diabetic retinopathy (NDR vs. BDR + PDR): type of therapy (P<0.0005), log-transformed plasma fibrinogen level (P < 0.05), mean blood pressure (P < 0.05), and duration of diabetes (P < 0.05). The independent variables associated with advanced retinopathy were type of therapy (P<0.00005), age (P<0.0005) and nephropathy (P<0.05). Body mass index, smoking and hypertensive status, HbA1c and total cholesterol levels were not independently associated.

CONCLUSIONS

These data suggest that in patients with Type 2 DM, an increased blood viscosity due to high fibrinogen level as well as an elevated intravessel pressure play a role in the development of diabetic retinopathy, and that the progression to PDR is influenced or accompanied by the deterioration of renal status.

Coagulation activation in type 2 diabetes mellitus: the higher coronary risk of female diabetic patients

Thrombophilia in diabetic patients is a well-recognized phenomenon which constitutes an additional risk of coronary heart disease. This study included 1980 ethnic Chinese people (835 male, 1145 female); age range: 45 to 69 years, including 280 Type 2 diabetic patients (male 125, female 155). Haemostatic parameters measured were fibrinogen, prothrombin time, activated partial thromboplastin time (APTT), factor VIIc, factor VIIIc, antithrombin III, and plasminogen. Compared with a control group, male diabetic patients showed significantly shorter APTT (25.6 +/- 3.7 vs 27.5 +/- 3.6 s, p < 0.001), and elevated factor VIIIc (171.1 +/- 77.48 vs 131.16 +/- 52.23%, p < 0.0001), whereas female diabetic patients showed significantly shorter APTT (24.9 +/- 4.2 vs 26.5 +/- 3.9 s, p < 0.001) and elevated fibrinogen (10.6 +/- 3.3 vs 9.8 +/- 2.6 mumol 1(-1), p < 0.05), factor VIIc (150.4 +/- 68.7 vs 135.3 +/- 32.3%, p < 0.001), factor VIIIc (190.1 +/- 92.6 vs 141.1 +/- 62.4%, p < 0.0001), and plasminogen (140.3 +/- 41.9 vs 128.4 +/- 38.7%, p < 0.01). This study showed that Chinese diabetic patients had coagulation activation, and that female diabetic patients seemed to constitute a higher risk group for coronary heart disease than males.

Changes in fibrinolytic parameters in male patients with type 2 (non-insulin-dependent) diabetes mellitus

We examined changes in fibrinolytic parameters in male patients with diabetes mellitus (DM) and controls. DM patients were divided into three groups: patients without retinopathy, patients with simple retinopathy, and patients with proliferative retinopathy. Plasma levels of t-PA (tissue plasminogen activator) and t-PA-PAI-1 (plasminogen activator inhibitor-1) complex increased with increase in age, but those of PAI-1 (total and free) did not change in controls. On the other hand plasma levels of PAI-1 decreased with increase in age in DM patients. Plasma levels of t-PA, t-PA-PAI-1 complex, free and total PAI-1 increased with increase in body mass index in controls, but no significant changes were shown in these parameters in DM patients. When compared with controls, plasma levels of t-PA, t-PA-PAI-1 complex and PAI-1 were lower in DM patients. Plasma levels of UK (urokinase) and Lp(a) were higher in DM patients. ELT (euglobulin clot lysis time) was significantly shorter in DM patients than in controls. Patients without retinopathy showed increased fibrinolytic activities compared with those with retinopathy due to the increased levels of t-PA in plasma. These results seem to indicate that blood vessels release larger amounts of t-PA at the early stage of DM, then release being impaired at its advance stage. It is also suggested that the regulatory control mechanisms of fibrinolytic activity associated with mechanisms of fibrinolytic activity associated with change in age and body mass index are different between patients with DM and normal people.

Increased platelet, but unaltered fibrinogen, accumulation in experimental thrombi in alloxan-induced diabetic rabbits

Platelets from diabetic humans and animals have been found previously to be hypersensitive to agonists, including thrombin, in vitro but it is unclear if this hypersensitivity also occurs in vivo and leads to a greater thrombotic tendency. In the present study, the effect of diabetes was examined on thrombus formation and vessel wall responses which result from continuous intimal injury induced by indwelling aortic catheters in rabbits. Platelet and fibrin(ogen) associated with the thrombus and damaged aortae were examined. Control or alloxan-induced diabetic rabbits (9-12 months after initial treatment) were injected with 51Cr-labeled autologous platelets and 125I-labeled fibrinogen (prepared from control rabbits) before insertion of indwelling aortic catheters. The anesthetized rabbits were perfused-fixed after 20 hr or 4 days. The dry weight of thrombus that formed was determined and platelet and fibrin(ogen) accumulation in thrombi and on injured aortae were calculated from the associated 51Cr and 125I, respectively. In diabetic rabbits, more platelets accumulated in the thrombi which formed after either 20 hr or 4 days, although the weight of thrombus and net fibrin(ogen) incorporation into the thrombus were not different from corresponding control rabbits. Net platelet and fibrin(ogen) association with the injured aortae were not different between control and diabetic rabbits. It is likely that the increased platelet accumulation in arterial thrombi in diabetic rabbits which results from continuous injury to aortae is a consequence of hypersensitivity of these platelets to thrombin generated in the thrombus and at the sites of vessel injury.

Platelet function in male diabetics with and without macrovascular complications

Changes of platelet aggregation in relation to macroangiopathy and to some of its risk factors were observed in microangiopathy-free, well-controlled type 1 diabetic males. Platelet aggregate ratio was generally lower in patients (n = 77) than in age-matched healthy subjects (n = 48). In the absence of cigarette smoking, hypertension, obesity and hypercholesterolemia (n = 25) in vitro platelet hyperaggregation was found induced with epinephrine, collagen or arachidonic acid, and to a lesser degree with ADP. There was no change in the presence of at least one risk factor in addition to diabetes (n = 29), but there was a further significant increase in platelet aggregation when overt coronary, cerebral or peripheral artery disease was present (n = 23).

Platelet behaviour and haemostatic variables in type 1 (insulin-dependent) diabetic patients with and without albuminuria

Abnormalities of haemostasis have been implicated in the development of both large and small vessel disease in diabetes. Platelet behaviour and coagulation factors were studied in 28 non-diabetic control subjects and 81 Type 1 diabetic patients with different degrees of albuminuria. Twenty-four (30%) patients had macro- or micro-albuminuria. These patients had elevated levels of beta-thromboglobulin compared with normo-albuminuric patients and control subjects (macro-albuminuric 113 (range 60-314), micro-albuminuric 88 (50-220), normo-albuminuric 55 (13-273), control 52 (18-210) micrograms l-1, p less than 0.001). Similar results were found for platelet factor 4 (macro-albuminuric 57 (9-350), micro-albuminuric 78 (12-205), normo-albuminuric 10 (2-135), control 9 (3-95) micrograms l-1, p less than 0.001). There were decreased beta-thromboglobulin:platelet factor 4 ratios in the albuminuric patients compared with control subjects and normo-albuminuric patients (p less than 0.001). There is abnormal platelet activity in Type 1 diabetic patients with elevated albumin excretion rates.

Effect of treatment with an inhibitor of platelet aggregation on the evolution of background retinopathy: 2 years of follow-up

Ophthalmic evolution was studied for 2 years in 17 patients with insulin-dependent diabetes mellitus and background diabetic retinopathy. Nine patients were treated with triflusal, a new platelet antiaggregant drug, and the eight remaining patients, with similar clinical and biological characteristics, were considered the control group. At the end of the study the ophthalmic evolution was different in the two groups. In the control group the degree of fluorescein leakage and the number of microaneurysms increased, while in the triflusal-treated group both parameters were reduced. There were no differences in visual acuity and computerised perimetry between the groups. Our results suggest that platelet antiaggregant therapy can be useful in the treatment of background diabetic retinopathy.

Effect of diabetes on the healing process of synthetic grafts implanted in dogs: a preliminary study

The deleterious effects of diabetes in peripheral vascular surgery and blood access are generally considered to be associated with the greater failure rate of vascular prostheses. To understand better these phenomena, we investigated the healing characteristics of an ePTFE graft in dogs. The animals were made diabetic after total pancreatectomy and the grafts implanted for scheduled periods of 24 h, 48 h, 1 wk and 1 month, either as thoracic (series I) or abdominal (series II) aortic substitutes. The same implantations were performed in non-diabetic dogs used as controls. The luminal surfaces of grafts implanted in diabetic dogs, either as thoracic or abdominal bypasses, proved to be more invaded by platelets as a result of the blood's increased platelet aggregation properties. The presence of endothelial-like cells spreading over the pannus after 1 month implantation was strikingly different in diabetic versus non-diabetic dogs. In the first group, the cells were fragile and less abundant, whereas in the second they were more adhesive, elongated and orientated in the direction of the blood flow.

Fibrinolytic capacity following stimulation with desmopressin acetate in patients with diabetes mellitus

Tissue plasminogen activator (t-PA), tissue plasminogen activator inhibitor, (PAI), and von Willebrand factor (vWF) were measured in 30 diabetics and 17 control subjects. These studies were performed to clarify the role of obesity in causing abnormalities of the fibrinolytic system in diabetics. The t-PA antigen response measured after the infusion of desmopressin acetate (DDAVP) was similar in all groups. Peak responses to DDAVP for controls, type I diabetics, and type II diabetics were 21.2 +/- 9.5 ng/mL, 27.5 +/- 9.0 ng/mL, and 28.8 +/- 11.0 ng/mL (NS), respectively. These responses did not correlate with the body mass index (BMI) or any other of the indices examined. A significant decrease of t-PA activity as contrasted with t-PA antigen following DDAVP occurred in type II diabetics only. The decrease of t-PA activity strongly correlated with greater basal levels of plasminogen activator inhibitor in these same subjects. The plasma level of plasminogen activator inhibitor correlated with BMI but with no other index examined. In contrast to t-PA activity and PAI, vWF responses to DDAVP inversely correlated to basal vWF concentration in all groups. Basal concentrations of vWF were increased in both type I and II diabetics and showed no relationship to degree of obesity. In summary, these results suggest that type II diabetic subjects have decreased t-PA activity, which is best explained by increased levels of PAI. The increased PAI appears related to obesity and not diabetes per se.

Changes in the factor VIII complex in diabetic ketoacidosis: evidence of endothelial cell damage?

Factor VIII-related antigen and von Willebrand factor are synthesised by and released from vascular endothelium. Acute increases in the plasma concentration of these proteins may reflect endothelial cell damage. We have thus measured the plasma concentration of factor VIII-related antigen and von Willebrand factor, together with procoagulant factor VIII, during the course of acute diabetic ketoacidosis in seven patients. In addition, evidence for qualitative changes in the factor VIII complex was sought. Plasma factor VIII-related antigen and von Willebrand factor were markedly increased (plasma factor VIII-related antigen at presentation, median 2.75 U/ml; von Willebrand factor 2.95 U/ml) and returned toward normal with clinical and biochemical resolution (plasma factor VIII-related antigen at clinical recovery, median 1.80 U/ml; von Willebrand factor 2.05 U/ml). Plasma procoagulant factor VIII followed a similar pattern, but levels were less elevated (plasma procoagulant factor VIII, at presentation, median 1.6 U/ml; at clinical recovery, 1.2 U/ml). Crossed immunoelectrophoresis and sodium dodecyl sulphate-acrylamide electrophoresis with autoradiographic identification of multimeric structure revealed no evidence of structurally abnormal factor VIII-related antigen in diabetic ketoacidosis. However, an extra peak on crossed immunoelectrophoresis ("pre-peak") was a feature in the acute phase ketoacidotic plasma in six subjects, and may represent aggregated factor VIII. Changes in plasma factor VIII are a feature of diabetic ketoacidosis and, whilst not specific to this condition, may be the result of endothelial cell damage.

Platelet aggregation and coagulation factors in insulin dependent diabetics with and without microangiopathy

Abnormalities of platelet aggregation and coagulation have been reported in insulin dependent diabetes mellitus (IDDM), although there is controversy concerning their relationship to microangiopathy. We have studied platelet function and haemostasis in 55 patients with IDDM, 23 without, 14 with mild (background retinopathy) and 18 with severe (proliferative retinopathy, or background retinopathy plus proteinuria) complications. Studies were done on 2 occasions 8 weeks apart and the results compared with 28 control subjects. There was evidence of increased in vivo platelet aggregation in the diabetic group v controls shown by raised values of beta-thromboglobulin (61 +/- 42, mean +/- SD, v 18 +/- 14 micrograms/ml, p less than 0.001), platelet factor 4 (62 +/- 76 v 14 +/- 11 micrograms/ml, p less than 0.01), and platelet micro-aggregates (20 +/- 16 v 12 +/- 11%, p less than 0.01). There was no significant difference in fibrinogen and fibrinopeptide A levels, nor in 'in vitro' tests of platelet aggregation between the groups. Dilute whole blood clot lysis time was increased in the diabetic group v controls (6.4 +/- 2.6 v 4.8 +/- 0.5 hours, respectively, p less than 0.001).(ABSTRACT TRUNCATED AT 250 WORDS)

Plasma levels of platelet secretory proteins

Platelets contain three types of secretory organelles: the dense granules, the alpha granules, and the lysosomes. Most of the proteins secreted from platelets are stored in the alpha granules, whereas the dense granules contain substances such as adenine nucleotides, serotonin, Ca++, and inorganic pyrophosphate types as well as a heparatinase. Three of the secreted alpha granule proteins have been measured by radioimmunoassay and it has been suggested that levels of these proteins in patient plasmas provide an index of in vivo platelet activation and secretion. These three are beta-thromboglobulin, platelet factor 4, and thrombospondin. In this chapter the chemistry of these proteins will be considered briefly, as will their clearance from the circulation, and then the clinical studies will be reviewed critically. Since radioimmunoassays were developed for these proteins (the first was reported in 1975), there has been a profusion of reports on levels of one or another of these proteins in a wide range of disease states, and these reports have indicated secreted platelet protein levels ranging from normal to grossly elevated in a given disease state. Possible reasons for such variability will be discussed.

Specific non-saturable binding of insulin by human platelets

Studies on the nature and characterization of the specific binding of 125I-insulin to intact human platelets have been undertaken. Although under conditions of physiological buffer osmolality, a binding equilibrium was attained in 4-6 h at 4 degrees C, at higher temperatures (17 degrees C or 24 degrees C) equilibrium was reached only in the presence of very high buffer osmolality or 25 mM NaF. Under conditions of normal osmolality and in the absence of NaF, binding at 17 degrees C was not saturable. This phenomenon was specific for insulin or insulin-like hormones (the insulin-like growth factors) and did not occur with 125I-labelled growth hormone, ACTH or beta-endorphin. The non-saturable uptake of insulin appeared due to an energy-dependent specific sequestration or internalization of insulin by mechanisms probably involving the microtubule system. This study emphasizes the need to restrict this non-saturable uptake of insulin if one wishes to adequately study the platelet plasma membrane receptors for insulin. These data also indicate that there is a major interaction of insulin and insulin-like hormones with normal human platelets and support previous demonstrations of major insulin effects on platelet physiology in both normal and diabetic states.

Ca2+-independent secretion is dependent on cytoplasmic ATP in human platelets

The secretion evoked in human platelets by physiological agonists is an energy-requiring process that depends on the metabolic ATP pool. Diacylglycerol, phorbol ester and collagen evoke a secretory response, which is not associated with an increase of cell Ca2+ levels, and is attributed to activation of protein kinase C [(1983) Nature 305, 317-319]. The secretion evoked by these agonists decreased along with cytoplasmic ATP depletion in the same way that the thrombin-induced secretion did. The secretory response was restored by raising again the cytoplasmic ATP levels. These results support the idea that the secretory response takes place by the physiological ATP-dependent mechanisms rather than by membrane perturbations in these instances.

Characterization of leucocyte phagocytic stimulatory material released by activated human platelets

The action of material released by activated human platelets on the phagocytic activity of human neutrophils and monocytes has been examined under conditions where there was a low background level of exposure of the leucocytes to platelet release products ( PRPr ). Incubation with PRPr produced an approximate three-fold stimulation of particle ingestion by both cell types. Material with stimulatory activity passed through ultrafiltration membranes capable of retaining molecules greater than 500 Daltons. This low molecular weight stimulator was inactivated by ATP diphosphopyrolase to suggest it could be one or more of the nucleotide di- or triphosphates in PRPr . Comparison with the action of specific nucleotides revealed only ADP and ATP exerted equivalent stimulatory effects at concentrations consistent with those of nucleotides normally present in PRPr . This effect was mediated by induction of a change in leucocyte behaviour which was sustained in the absence of continued exposure to the stimulatory agent. Stimulation of phagocytosis was also produced by high molecular weight material in PRPr , but this appeared to be mediated by dissociation of low molecular weight stimulator from a high molecular weight complex. These observations extend the range of biological functions known to respond to nucleotides in PRPr , and have implications for the interpretation of leucocyte phagocytic activity assessed in vitro in view of the potential modification of this process by exposure of leucocytes to PRPr during the isolation of leucocytes from peripheral blood.

Stimulation of leukocyte phagocytic activity by the platelet release reaction

The influence of material released by activated human platelets on the phagocytic activity of human neutrophils and monocytes has been investigated under conditions where there was a low background level of exposure of the leukocytes to platelet release products (PRPr). Incubation with PRPr prepared from 400 X 10(9)/l platelets resulted in an approximate 3-fold increase in the proportion of leukocytes which ingested particles. Stimulation was produced by dilutions of PRPr down to 1:4 to indicate that phagocytic activity was affected by material released by concentrations of platelets within the physiological range. This response appeared to reflect stimulation of the overall phagocytic capability of the leukocytes as equivalent ingestion was observed with particles which possessed or lacked the capacity to bind complement to their surface. Enhancement of phagocytic activity was evidently secondary to induction of a change in leukocyte behaviour which was sustained in the absence of continued exposure to preparations of PRPr. These observations extend the range of human biological processes known to be responsive to PRPr, and have implications for the assessment of leukocyte phagocytic activity in vitro as this particular property could be modified by contact with PRPr during the isolation of leukocytes from peripheral blood.