Characterization of the Pro-Thrombotic State in CAPD Patients

Abstract

Purpose. To investigate whether a chronic pro-thrombotic tendency, which may contribute to a high rate of atherothrombotic disease, is present in patients treated for continuous peritoneal dialysis (CAPD), and, if so, what its pattern is. We investigated this issue by jointly exploring all the systems involved, the coagulation and fibrinolytic systems and platelets. Methods. Markers of coagulation activation, markers of fibrinolysis activation, and standard fibrinolytic parameters and platelet aggregation induced by different agents were measured in 15 patients treated by CAPD and in 15 matched, healthy controls. All CAPD patients received erythropoietin, were in the stable condition, and did not have acute disease or malignancy. Results. CAPD patients had substantially (p < 0.001) increased levels of prothrombin fragments F1+2, disclosing a low-grade activation of the coagulation system. D-dimer was also significantly (p < 0.05) increased, whereas the levels of t-PA antigen and activity, PAI antigen and activity, and plasminogen were comparable to controls, suggesting that slight secondary (and not primary) activation of fibrinolysis due to coagulation activation took place. Patients had significantly (p < 0.05) elevated levels of fibrinogen. A study of platelet aggregation (induced by adenosine diphosphate, collagen, and epinephrine) did not show platelet hyperactivity in patients. Conclusions. We found that a pro-thrombotic tendency is present in the plasma of CAPD patients. The main reason for a pro-thrombotic state is chronic low-grade activation of the coagulation system and elevated levels of fibrinogen. The fibrinolytic system and platelets seemingly do not contribute to this pro-thrombotic tendency.

Keywords:

• continuous peritoneal ambulatory dialysis
• coagulation
• fibrinolysis
• platelets
• thrombosis

BACKGROUND

In recent years, it has become evident that CAPD patients have a high incidence of atherothrombotic events,[1],[2] probably related to the high incidence of risk factors present in CAPD patients (e.g., arterial hypertension, hyperlipidemia, diabetes mellitus, hyperparathyroidism, vascular calcification).[1],[2] However, other factors are also likely to be involved in the high prevalence of atherothrombotic events. In this regard, the role of the coagulation system, the fibrinolytic system, and platelet function is less well understood. It is generally believed that enhanced coagulation and/or defective fibrinolysis and increased activity of platelets may play a role in the initiation and progression of atherothrombotic diseases.[3–5] Thus, the fibrinolytic system, coagulation system, and platelets could separately or in cooperation contribute to development of atherothrombotic diseases in CAPD patients. Accordingly, it appears to be clinically relevant to investigate all three systems together in CAPD patients and explore whether all or some individual systems are disturbed and, if so, how they are disturbed (i.e., to identify a pattern of impairment).

In this study we measured markers of coagulation activation (thrombin fragments: F1+2), markers of fibrinolysis activation (D-dimer), standard fibrinolytic parameters (t-PA antigen and activity, PAI-1 antigen and activity, plasminogen, and fibrinogen), and platelet aggregation (induced by adenosine diphosphate, collagen and epinephrine) in 15 CAPD patients and 15 healthy, age-, sex-, and weight-matched controls.

METHODS AND PATIENTS

Patients

Fifteen patients (nine female, six male) were recruited to the study from CAPD outpatients of a university department. They ranged in age from 31 to 65 years, with a mean age of 44 ± 8 years. The duration of CAPD was from 5 to 80 months, with a mean of 23 ± 15 months. CAPD was performed using 2.0 or 1.5 liter packs that were exchanged four times a day. The osmotic pressure of CAPD was adjusted in accordance with the extent of ultrafiltration in each patient. The underlying cause of end stage renal disease (ESRD) included glomerulonephritis in 10 patients and nephrosclerosis in 5 patients. Patients had residual urine output, and none had significant proteinuria. Recombinant erythropoietin was administered to all patients for renal anaemia. Peritonitis had not occurred in any of the patients during the three months before the study. Only patients with levels of C-reactive protein ≤10 mg/L and serum albumin levels ≥ 35 g/L were included in the study. All patients had normal levels of antithrombin III and protein C; none had resistance to activated protein C. Furthermore, the patients recruited did not have manifest atherosclerotic disease (myocardial infarction or angina pectoris, stroke, peripheral arterial disease). During the study, patients did not have acute illness and used their regular therapy, which had not been changed at least one month previously. Neither anticoagulants, aspirin, nor statins were used in any of the patients. Using these strict criteria, we excluded the possible influence of factors that can frequently (at least temporarily) influence the measured parameters. Consequently, this approach allowed us to study more convincingly the characteristics of the principal (i.e., basic, pro-thrombotic) state in CAPD patients. The sex-, age-, and weight-matched 15 healthy volunteers (nine female, six male, age 65 [48–80] years) served as controls. All controls had normal renal function, did not use any drug, did not have acute or chronic disease, and did not have manifest atherosclerotic disease. The study was approved by the local ethical committee, and informed consent was obtained from all participants.

Blood Sampling

Whole blood was obtained by venipuncture between 7 a.m. and 8 a.m. Blood taken for platelet aggregation studies was anticoagulated with 3.9% trisodium citrate and immediately prepared for testing. Blood for other analyses was centrifuged for 20 min at 3000 g. Plasma was frozen and stored at −70°C.

Markers of Coagulation and Fibrinolysis Activation and Fibrinolytic Parameters

Prothrombin fragments (F1+2) were measured by ELISA using commercial assays (Behringwerke AG, Marburg, Germany). D-dimer was measured by enzyme-linked immunosorbent assay (Tintelize® t-PA, Biopool, Umeå, Sweden). Fibrinogen was determined by a clotting assay (Multifibren, Behring, Marburg, Germany), plasminogen by kinetic spectrophotometry (Berichrom Plasminogen, Behring, Marburg, Germany), t-PA and PAI-1 antigens by enzyme-linked immunosorbent assay (Imulyse^TM tPA and Imulyse^TM PAI-1, Biopool, Umeå, Sweden), and PAI activity and t-PA activity by amidolytic assay (Spectrolyse®/fibrin, Biopool, Umeå, Sweden).

Antithrombin III and protein C activity were determined by kinetic spectrophotometric assay (Berichrom Antithrombin III and Berichrom-Protein C both Behring, Marburg/Lahn, Germany). Activated protein C resistance was determined according to previously described method.[6]

Platelet Aggregometry

Platelet aggregometry was performed using an APACT 2 aggregometer (Labor, Hamburg, Germany). Platelet-rich plasma (PRP) was prepared by centrifugation of citrated blood at 2000 g for five minutes. The remaining blood was recentrifuged at 2000 g for 10 minutes, and platelet-poor plasma (PPP) was then removed. All samples of PRP were adjusted to obtain a final platelet count of 250,000 ± 50,000 plts/μL. The minimum and maximum transmittance were adjusted by PRP (0% transmission) and PPP (100% transmission), respectively. The light transmittance of PRP increased as platelets aggregated. The ability of PRP to aggregate was tested with adenosine diphosphate (ADP), collagen, and epinephrine. The concentrations of the agonists were as follows: ADP in final concentrations of 4 and 8 μM, collagen 2 μg/mL, and epinephrine 10 μg/mL. Results were expressed as percentages of maximal aggregation obtained after 5 min of stimulation.

Statistical Methods

All variables were expressed as medians with ranges between the first and the third quartiles. Differences between groups were assessed by the Mann Whitney U-test. A p value of <0.05 was taken as statistically significant.

RESULTS

Comparison of Markers of Coagulation and Fibrinolysis Activation between CAPD Patients and Controls

In CAPD patients, significantly increased levels of F1+2 (p < 0.001) and D-dimer (p < 0.05) were found compared to controls. The results are shown in Table 1.

Table 1 Comparison of markers of coagulation and fibrinolysis activation between patients treated by continuous peritoneal dialysis (CAPD) and controls

	Controls (n = 15)	CAPD patients (n = 15)	p
F1+2 (nmol/L)	1.1 (1.0–1.2)	1.75 (1.4–2.2)	<0.0001
D-dimer (ng/mL)	135 (120–150)	240 (120–335)	<0.05

All values are medians with ranges between the first and the third quartiles.

Abbreviation: F1+2 = prothrombin fragments.

Comparison of Fibrinolytic Parameters between CAPD Patients and Controls

CAPD patients had comparable plasminogen levels, t-PA antigen and activity, and PAI-1 antigen and activity, whereas fibrinogen was significantly elevated in comparison to controls. The results are shown in Table 2.

Table 2 Comparison of fibrinolytic parameters between patients treated by continuous peritoneal dialysis (CAPD) and controls

	Controls (n = 15)	CAPD patients (n = 15)	p
Plasminogen activity (rel)	0.98 (0.87–1.08)	0.97 (0.87–1.05)	NS
t-PA antigen (ng/mL)	8.6 (4.0–10.6)	6.2 (4.8–8.1)	NS
t-PA activity (IU/mL)	0.6 (0.4–1.1)	0.6 (0.4–1.0)	NS
PAI-1 antigen (ng/mL)	14.5 (11.1–16.7)	10.8 (5.1–13.6)	NS
PAI activity (IU/mL)	0.6 (0.4–0.9)	0.3 (0.0–1.9)	NS
Fibrinogen (g/L)	4.0 (3.4–4.6)	5.4 (4.3–6.6)	<0.05

All values are medians with ranges between the first and the third quartiles.

Abbreviations: t-PA = tissue type plasminogen activator, PAI = plasminogen activator inhibitor, NS = not significant.

Comparison of Platelet Aggregation between CAPD Patients and Controls

Platelet aggregation induced by ADP in final concentrations of 4 and 8 μM, collagen of 2μg/mL, and epinephrine of 10μg/mL were comparable in CAPD patients and controls. The results are shown in Table 3.

Table 3 Comparison of platelet aggregation by different stimulants between patients treated by continuous peritoneal dialysis (CAPD) and controls

Aggregation (%)	Controls (n = 15)	CAPD patients (n = 15)	p
ADP 4 μM	83 (73–88)	87 (68–94)	NS
ADP 8 μM	82 (73–85)	82 (65–90)	NS
Collagen 2μg/mL	75 (69–78)	80 (66–92)	NS
Epinephrine 10μg/m	75 (63–80)	78 (45–90)	NS

All values are medians with ranges between the first and the third quartiles.

Abbreviations: ADP = adenosine diphosphate, NS = not significant.

DISCUSSION

Continuous peritoneal dialysis is widely used, and the high incidence of atherothrombotic events has been clearly disclosed.[1],[2] Factors contributing to this are not completely and definitely recognized. The coagulation and fibrinolytic systems and platelets could have a particular role in the pathogenesis of atherothrombotic diseases.[3],[4] Knowing whether permanent impairment of any of these three systems exists in patients treated with currently used techniques of CAPD, which also include erythropoietin administration, would have important clinical implications. The aim of our study was to investigate this topic. Although performed on a relatively small group of patients (due to the strict inclusion criteria), our study showed consistent results.

First, we explored the presence of markers of the ongoing coagulation process. F1+2 is a sensitive marker of coagulation activation.[7],[8] We found elevated levels of F1+2 in all CAPD patients. The activation of the coagulation system was also observed by others. Some authors reported elevated markers of coagulation activation, and others also found increased activity of activated coagulation factors such as activated factor VII.[9–13] Elevated levels of F1+2 clearly confirm activation of the coagulation pathway. The mechanism underlying this “low-grade” chronic activation of coagulation is not known. Whether activation of coagulation could be further (temporarily or chronically) increased in certain clinical conditions remains unknown.

Conflicting data have been reported concerning the activation of the fibrinolytic system in CAPD patients. Some authors found signs of hypofibrinolysis[9],[14]; others observed activation of fibrinolysis (i.e., hyperfibrinolysis[10],[11],[15]); whereas some authors did not detect any changes in fibrinolytic activity.[16] These differences were likely due to the presence of infection, inflammation, transperitoneal protein loss with consequent hypoalbuminemia, or inclusion of patients with diabetes, all of which could independently influence fibrinolysis. By using strict inclusion criteria, we avoided these additional effects on the principal fibrinolytic state in CAPD patients. When interpreting the results, it is also important to consider that the activation of fibrinolysis could be primary (not induced by coagulation activation) or secondary to coagulation activation.[17],[18] Secondary activation is not associated with changes in the activity of fibrinolytic activators and/or inhibitors (t-PA and/or PAI-1), whereas primary activation is induced by a change in the stable balance between activators and inhibitors, resulting in increased t-PA activity and/or decreased PAI activity. Increased, decreased, and normal levels of t-PA antigen and activity and PAI-1 antigen and activity have been reported in CAPD patients.[9–16] These disparate results, which were probably due to the above-mentioned conditions, do not allow for a convincing estimate of the principal fibrinolytic state in CAPD patients. In any case, additional data are needed, using strict inclusion criteria, obtained by consideration of the activity of the coagulation system, and distinguishing between primary and secondary activation of fibrinolysis. As mentioned previously, changes in an individual fibrinolytic parameter should be interpreted together with other parameters (particularly t-PA and PAI activity) and markers of fibrinolysis activation. We found comparable levels of t-PA antigen and activity, PAI-1 antigen and activity, and plasminogen. We also measured the levels of D-dimer, a sensitive marker of fibrin-specific fibrinolysis,[19] and found that CAPD patients had elevated levels of D-dimer. This result confirms that fibrinolysis in CAPD patients is activated. Furthermore, it could be concluded that fibrinolysis in CAPD patients is not primarily but rather secondarily activated, due to the activation of plasminogen by thrombin. These findings and the associated explanation of hypercoagulability as the primary event and hyperfibrinolysis as the secondary event in CAPD patients has not yet been described.

In addition, we measured elevated levels of fibrinogen in CAPD patients, as uniformly reported earlier by others.[9–15] Elevated levels of fibrinogen contribute to hypercoagulability and to the pro-thrombotic properties of blood.[20]

In the third part of the study, we investigated whether increased platelet aggregation is present in CAPD patients. The study of platelet aggregation was performed using different agents to stimulate platelet aggregation (ADP, collagen, and norepinephrine). The uniform results obtained by a wide range of different stimulants strongly confirm that platelet aggregation in CAPD patients is not aggravated. In fact, we found that it is comparable to that in controls. Platelet function in CAPD patients has been investigated in several studies. Some authors found platelet hyperactivity, whereas others did not find impairment of platelet aggregation.[21–23] It appears that in older studies, hyperactivity of platelets has been found, whereas in more recent studies, mainly unaffected platelet function was found. Regarding our study, the approach of using different agents enabled us to conclude that current techniques of CAPD do not influence platelet function significantly. Thus, it can be concluded that platelets very likely do not contribute importantly to the development of atherothrombotic diseases in CAPD patients.

Our study is limited by the fact that we could not simply attribute the results obtained solely to the CAPD procedure. Indeed, the observed findings could be influenced by renal failure itself, the administration of erythropoietin, or the procedure of CAPD. However, from the clinical point of view, this issue is not very relevant, as the majority of CAPD patients receive erythropoietin, and of course all have renal failure. Thus, the obtained results could be expected to be found in an “average” stable CAPD patient. We could not identify the exact reason(s) for low-grade coagulation activation in CAPD patients. The following morbidities could be involved: renal failure itself, the peritoneal dialysis procedure, erythropoietin, endothelial injury, tissue injury, low-grade inflammation, or some other factors. Further studies are needed to explore this issue in detail.

CONCLUSIONS

Understanding and considering the pro-thrombotic state and its pattern are clinically relevant in managing CAPD patients. Our study has shown the presence of a chronic low-grade pro-thrombotic tendency in plasma of “average” stable patients treated with current techniques of CAPD and receiving erythropoietin. Coagulation, fibrinolysis, and platelets could play role in the pro-thrombotic state. Putting all our results together, we found that the principal pro-thrombotic state is characterized by hypercoagulation (activated coagulation and elevated fibrinogen), whereas fibrinolysis and platelets do not contribute to the pro-thrombotic state.

These results could help in better understanding the reported mixed and conflicting results concerning a possible thrombotic tendency and its pattern in CAPD patients, and could have potential clinical implications.