- Split View
-
Views
-
Cite
Cite
Solbjørg Sagedal, Anders Hartmann, Kristina Sundstrøm, Stine Bjørnsen, Frank Brosstad, Anticoagulation intensity sufficient for haemodialysis does not prevent activation of coagulation and platelets, Nephrology Dialysis Transplantation, Volume 16, Issue 5, May 2001, Pages 987–993, https://doi.org/10.1093/ndt/16.5.987
- Share Icon Share
Abstract
Background. A single bolus of dalteparin at the start of haemodialysis (HD) may prevent clot formation, but subclinical activation of platelets and coagulation may still occur. Consequently, the relationship between clinical clotting events and activation markers of platelets and coagulation before and during HD is of interest.
Methods. The effect of tapered doses of dalteparin during 84 HD sessions (4–4.5 h) was prospectively examined in 12 patients. Six of the patients were treated with warfarin. The initial dalteparin dose was reduced to 50% if no clotting was observed. Clinical clotting was evaluated by inspection of the air trap every hour and by inspection of the dialyser after each session. Anti‐FXa activity was measured for assessment of dalteparin activity. Markers of activated plasma coagulation, (thrombin‐antithrombin (TAT) and prothrombin fragment 1+2 (PF1+2)) and a marker of platelet activation (β‐thromboglobulin, β‐TG), were measured before the start of and after 3 and 4 h of dialysis. Ten pre‐dialytic patients with chronic renal failure served as a control group. A total sof 230 measurements of each parameter were performed.
Results. An anti‐FXa activity above 0.4 IU/ml at the end of HD inhibits overt clot formation for 4 h. This was obtained by an intravenous dalteparin dose of about 5000 IU. TAT and PF1+2 correlated to clinical clotting episodes (r=0.50 and 0.47, P<0.001). β‐TG was not significantly correlated to clinical clotting. All parameters increased during the sessions (TAT, PF1+2, β‐TG, P<0.001). When measurements during clinical clotting episodes were disregarded, all parameters were still markedly increased. Warfarin‐treated patients had lower TAT and PF1+2. Dialysis patients had higher β‐TG values than pre‐dialytic patients.
Conclusion. Despite clinically effective anticoagulation, obtained by dalteparin administration, platelets and coagulation are activated by HD, resulting in a potentially thrombophilic state. Warfarin treatment reduces clinical clot formation and subclinical activation of coagulation.
Introduction
The relationship between the anticoagulant effect of dalteparin and clinical clotting episodes and the dosage of dalteparin necessary to obtain adequate anticoagulation were assessed in a previously published study [1]. The present protocol was designed to examine the activation of platelets and the coagulation system irrespective of clinical clotting episodes. New data are presented in this study from the same dialysis sessions as previously reported [1]. However, in the present study new data from 10 patients with pre‐dialysis end‐stage renal disease (controls) are also included. The markers evaluated were parameters that directly reflect an activation of the haemostatic system: prothrombin fragment 1+2 (PF1+2) and thrombin‐antithrombin complex (TAT), both markers of intravascular thrombin formation [2,3]. The coagulation inhibitor AT was also evaluated.
Platelet activation is important for the initiation of the thrombogenic process, and β‐thromboglobulin (β‐TG) released from alpha‐granules of activated platelets after contact with biomaterials is a sensitive marker of platelet activation [4]. The fibrinolytic system was evaluated by measuring plasmin‐antiplasmin (PAP), a marker of activated fibrinolysis [5]. Dalteparin anticoagulant effect was measured as anti‐FXa activity.
Traditionally, anticoagulation during haemodialysis (HD) is obtained by giving an intravenous (i.v.) bolus of unfractionated heparin followed by continuous i.v. infusion or another bolus during dialysis. However, low molecular weight heparins are frequently used having several potential advantages over unfractionated heparin. They are dialysable and less frequently associated with bleeding complications and heparin‐induced thrombocytopenia [6–8]. In addition, the administration is practical, and the effect on blood lipids may be favourable [9,10].
The aim of the present study was to examine activation of platelets and the coagulation system during HD in patients with and without warfarin and to see if such activation may occur also in the absence of clinical clotting episodes.
Patients and methods
Patients
Twelve patients on chronic HD were included in the study, nine men and three women. The primary kidney disease causing chronic renal failure was polycystic kidney disease in four of the patients, secondary amyloidosis in two, rapidly progressive glomerulonephritis in one, chronic glomerulonephritis in one, diabetic nephropathy in one, Wegener's granulomatosis in one, reflux nephropathy in one and nephrosclerosis in one. The mean age was 60 years (range 26–77 years) and mean body weight was 70 kg (51–96 kg) (Table 1). Mean time on HD treatment was 9 months (range 3–17 months). Ten of the patients received erythropoietin in a mean dose 12 400 U/week (range 8000–20 000).
Six of the patients were treated with oral warfarin, two because of implanted heart valves (INR=3.3), one because of atrial fibrillation (INR=2.3), two because of a previously malfunctioning jugular cathether (INR=1.6) and one because of previous pulmonary embolism (more than a year ago) and a positive lupus anticoagulant (INR=2.3). The haemoglobin value was stable and the dose of erythropoietin and iron supplement were kept unchanged for the last 2 weeks prior to the study. Patients who received acetylsalicylic acid treatment were excluded from the study. Other exclusion criteria were clinical signs of infection and disseminated malignant disease. None of the patients had bleeding disorders. As a control group, 10 patients with pre‐dialytic chronic renal failure were examined. Their mean age was 54 years (range 26–74 years), mean weight was 77.1 kg (range 58–101 kg) and mean creatinine clearance as estimated from the Cockroft and Gault's formula was 19.0 ml/min (range 8.7–33.1 ml/min). The difference in age between the pre‐dialysis and the dialysis patients was not statistically different (P=0.28). None of the patients in the control group received acetylsalicylic acid or oral warfarin. The protocol was approved by the Regional Ethics Committee and informed consent was obtained from all the patients according to the Helsinki declaration.
Patients | Age (years) | Dry weight (kg) | Months on HD |
12 HD | 60±15 | 70±14 | 8±4 |
10 Pre‐dialytic | 54±16 | 77±15 | 0 |
Patients | Age (years) | Dry weight (kg) | Months on HD |
12 HD | 60±15 | 70±14 | 8±4 |
10 Pre‐dialytic | 54±16 | 77±15 | 0 |
Patients | Age (years) | Dry weight (kg) | Months on HD |
12 HD | 60±15 | 70±14 | 8±4 |
10 Pre‐dialytic | 54±16 | 77±15 | 0 |
Patients | Age (years) | Dry weight (kg) | Months on HD |
12 HD | 60±15 | 70±14 | 8±4 |
10 Pre‐dialytic | 54±16 | 77±15 | 0 |
Dialysis procedure
Dalteparin was given as a single bolus dose into the arterial line at the start of dialysis. The mean applied dalteparin dose at the start of the study was 60.9±19.3 IU/kg in the patient group receiving warfarin and 69.5±11.5 IU/kg in those without oral anticoagulation. The dialysis sessions were 4 h three times per week in seven patients and 4.5 and 5 h three times per week in five patients. A polysulphone hollow fibre dialyser (F6 HPS, Fresenius, Germany), was used during the study. Bicarbonate dialysate was used in all patients. The dialyser and the bubble trap were checked at the end of each dialysis session to ensure that there was no clot formation and the dalteparin dose was kept unchanged. The blood flow rate varied between 230 and 300 ml/min, but was kept close to constant within each patient. Blood flow was recorded by the dialysis machine flowmeter. The dialysate flow was kept constant at 500 ml/min.
Study design
The dalteparin dose was gradually tapered in subsequent dialysis sessions to examine the relationship between anti‐FXa activity, markers of coagulation and platelet activation versus clinical clotting events during dialysis in each of the 12 patients. Nine HD sessions over 3 weeks were studied in eight of the patients and three sessions over 1 week in four patients. The regular dose of dalteparin was given in the first session and then reduced by 25% for each session down to 50% if no clotting was observed. Clinical clotting was evaluated by visual inspection after blood draining of the air trap every h (1=no clotting in the trap, 2=fibrinous ring, 3=clot formation and 4=coagulated system) and by visual inspection of the dialyser at the end of each session (1=clean filter, 2=a few blood stripes (affecting less than 5% of the fibres seen at the surface of the dialyser), 3=many blood stripes (affecting more than 5% of the fibres) and 4=coagulated filter). If clinical signs of clotting grade 3 or 4 were observed, the dalteparin dose was increased by one step (25% of the initial dose). Clinical clotting was compared with simultaneously measured anti‐FXa activity, TAT, PF1+2, PAP, AT, β‐TG, use of oral warfarin and dialysis time. The study was designed to keep Hb, HCT, platelet count, blood flow and ultrafiltration rate constant from 2 weeks before and during the study.
Blood sampling
Blood specimens were taken from the arterial line after lowering the blood flow to 100 ml/min for 1 min. Fibrinogen, Activated partial thromboplastin time (APTT), D‐dimer, Thrombin clotting time and ethanol gelation test were taken at the start and at the end of the first treatment. Hb, HCT, platelets, white cell count, thrombin time and urea were measured at the start and at the end of each dialysis session.
Samples for TAT, PF1+2, PAP, β‐TG and AT were drawn before start without venous compression and after 3 and 4 h of each dialysis session. Blood for TAT, PF1+2, PAP and AT was collected into tubes containing citrate and centrifuged at 1500 g for 10 min at room temperature. For β‐TG assay blood was collected into Diatube H® tubes and immediately cooled in a crushed ice‐water mix for 15 min. Within 1 h after collection the blood sample was prepared by centrifugation at 2500 g for 30 min at 4°C. Anti‐FXa activity was measured after 1, 3 and 4 h in each of the dialysis sessions. Anti‐FXa activity in the dialysis group was not measured before start of dialysis because the pre‐dialytic activity was expected to be below the detection limit (0.1 IU/ml). This is supported by the fact that five of the patients in this study had no detectable anti‐FXa activity after 4 h of dialysis, being in accord with the findings by other investigators [6,10,11]. Blood for anti‐FXa activity was collected into tubes containing citrate and cooled in an ice‐water mix before centrifugation in the same way as β‐TG. In the control group of pre‐dialysis patients creatinine, PAP, TAT, PF1+2, β‐TG and AT were analysed only once, and the blood specimens were collected without stasis and processed as described above.
Laboratory methods
After centrifugation, one‐third the volume of the plasma supernatant from the middle region of the liquid portion was collected and stored at −70°C for later determination of β‐TG. β‐TG was measured with an ELISA (EIA) procedure (Asserachrom®β‐TG kit). AT was measured with a chromogenic assay (Coamatic®, Chromogenix AB, Mølndal, Sweden). PAP was measured with an ELISA procedure (Enzygnost® PAP micro, Behring). PF1+2 and TAT were also measured with an ELISA, (Enzygnost® F1+2 micro and Enzygnost® TAT micro, Behring).
Anti‐FXa activity was measured with a chromogenic assay (Coatest®, Chromogenix AB, Mölndal, Sweden).
Statistics
For non‐parametric distributions, statistical comparisons between two different groups of patients were done by the Mann–Whitney two‐sample rank test and paired comparisons by the Wilcoxon matched‐pairs signed ranks test. For normal distributions parametric t‐tests were used. Parameters not normally distributed were logarithmically transformed before linear regression analysis.
These statistical analyses were performed by use of SPSS and by a computer based statistical soft‐ware program, Graph Pad prism, CA, US.
Results
Mean dalteparin bolus dose during the study was 2665±880 (range 1250–5000 IU) or 39±14 IU/kg body weight (84 dialysis sessions). As previously published, clinical clotting in the bubble trap was significantly correlated to dalteparin effect measured as anti‐FXa activity in the patients not receiving oral anticoagulants and in those receiving warfarin (r=0.67 and 0.32, respectively, P<0.001 for both) [1]. In the patient group without warfarin the same relationship was also found for clotting in the dialysis filter (r=0.30, P=0.03). There were too few clotting episodes to study dialysis filter clotting in the warfarin‐treated patients. Both time and anti‐FXa activity were independently and highly correlated to clinical clotting (grade 2 or 3) in a logistic regression model of repeated measurements performed 3 and 4 h after start of dialysis in the patients without warfarin therapy, as published elsewhere [1].
In a regression analysis between anti‐FXa activity and each of the parameters TAT, PF1+2 and β‐TG TAT and PF1+2, but not β‐TG, were negatively and significantly correlated to anti‐FXa (P<0.001). This means that tapering of the dalteparin dose results in increased levels of TAT and PF1+2.
Table 2 summarizes the mean values for coagulation markers (TAT, PF1+2, AT), fibrinolytic parameter (PAP) plus platelet activation (β‐TG) in 84 HD sessions in all patients at baseline and at the end of each session. All markers showed a significant increase during 4 h of HD.
TAT and PF1+2 were the only factors significantly correlated to the degree of clinical clotting, the correlation factors being 0.50 and 0.47 (P<0.0001 for both), respectively. However, when looking separately at the patients not receiving warfarin, the TAT, PF1+2 and AT values were significantly correlated to clinical clotting, the correlation factors being 0.40 (P<0.0001), 0.27 (P=0.014) and −0.31, P=0.004 for TAT, PF1+2 and AT, respectively. PAP and β‐TG were not statistically significantly correlated to the degree of clinical clotting.
There were striking differences between the patients receiving warfarin and those who did not. Figure 1 shows the values of TAT, PF1+2 and PAP during HD. The values of TAT and PF1+2 rose to levels more than three times as high in the non‐warfarin‐treated patients (P<0.001). The coagulation markers TAT and PF1+2 were low and within normal levels in the warfarin‐treated patients in contrast to PAP which was higher and above upper normal limits with warfarin treatment. The difference in PAP between the two groups at baseline and during dialysis was not statistically significant. Figure 2 shows that all patients had increasing levels of β‐TG throughout the dialysis sessions (P<0.0001), the levels being higher than normal values and significantly higher in the warfarin‐treated group at baseline (P<0.0001) and after 3 h of dialysis, (P=0.02).
To evaluate subclinical clotting and platelet activation all samples were evaluated separately when no clinical clotting was seen in the bubble trap. Figure 3 shows that TAT and PF1+2 are far above the normal range in the patients without warfarin treatment. The PAP values are above normal range in both groups, being highest in the warfarin‐treated patients. Figure 4 shows that both with and without clinical clotting β‐TG were more than five times higher than normal values both in warfarin‐treated patients and those without warfarin, and was highest in the warfarin‐treated group.
Table 3 shows the mean values of Hb, HCT, blood platelets and white cell count in 12 dialysis patients at baseline and at the end of HD. The difference from baseline to the end of HD was statistically significant only for Hb and HCT, increasing during HD. Table 3 also shows the mean values of D‐dimer, fibrinogen, APTT and thrombin clotting time in the 12 dialysis patients.
Parameters of coagulation in pre‐dialytic versus dialysis patients are presented in Table 4. The difference between the parameters in the 10 control patients and the baseline values in the non‐warfarin‐treated dialysis patients was statistically significant only for TAT and β‐TG.
Biological markers | Baseline | 4 h of HD | Normal range | P‐valuea |
Coagulation | ||||
TAT, mg/l | 2.5±2.7 | 7.6±8.6 | 1.0−4.1 | P<0.001 |
PF1+2, nmol/l | 1.2±0.8 | 1.4±1.1 | 0.4−1.1 | P=0.0013 |
AT% | 111.8±21.0 | 117.1±25.7 | >75 | P<0.001 |
Fibrinolysis | ||||
PAP, mg/l | 706.8±400.4 | 767.1±405.0 | 120−700 | P=0.02 |
Platelet activation | ||||
β‐TG, IU/ml | 158.4±94.1 | 223.9±118.2 | 10−40 | P<0.001 |
Biological markers | Baseline | 4 h of HD | Normal range | P‐valuea |
Coagulation | ||||
TAT, mg/l | 2.5±2.7 | 7.6±8.6 | 1.0−4.1 | P<0.001 |
PF1+2, nmol/l | 1.2±0.8 | 1.4±1.1 | 0.4−1.1 | P=0.0013 |
AT% | 111.8±21.0 | 117.1±25.7 | >75 | P<0.001 |
Fibrinolysis | ||||
PAP, mg/l | 706.8±400.4 | 767.1±405.0 | 120−700 | P=0.02 |
Platelet activation | ||||
β‐TG, IU/ml | 158.4±94.1 | 223.9±118.2 | 10−40 | P<0.001 |
Mean±SD in 84 HD sessions. aBaseline vs 4 h of HD. Wilcoxon test is used except for AT%, where paired t‐test is used. TAT=thrombin‐antithrombin and PF1+2=prothrombin fragment 1+2 (coagulation parameters), AT=antithrombin, PAP=plasmin antiplasmin (fibrinolysis parameter), β‐TG=β‐thromboglobulin (parameter of platelet activation).
Biological markers | Baseline | 4 h of HD | Normal range | P‐valuea |
Coagulation | ||||
TAT, mg/l | 2.5±2.7 | 7.6±8.6 | 1.0−4.1 | P<0.001 |
PF1+2, nmol/l | 1.2±0.8 | 1.4±1.1 | 0.4−1.1 | P=0.0013 |
AT% | 111.8±21.0 | 117.1±25.7 | >75 | P<0.001 |
Fibrinolysis | ||||
PAP, mg/l | 706.8±400.4 | 767.1±405.0 | 120−700 | P=0.02 |
Platelet activation | ||||
β‐TG, IU/ml | 158.4±94.1 | 223.9±118.2 | 10−40 | P<0.001 |
Biological markers | Baseline | 4 h of HD | Normal range | P‐valuea |
Coagulation | ||||
TAT, mg/l | 2.5±2.7 | 7.6±8.6 | 1.0−4.1 | P<0.001 |
PF1+2, nmol/l | 1.2±0.8 | 1.4±1.1 | 0.4−1.1 | P=0.0013 |
AT% | 111.8±21.0 | 117.1±25.7 | >75 | P<0.001 |
Fibrinolysis | ||||
PAP, mg/l | 706.8±400.4 | 767.1±405.0 | 120−700 | P=0.02 |
Platelet activation | ||||
β‐TG, IU/ml | 158.4±94.1 | 223.9±118.2 | 10−40 | P<0.001 |
Mean±SD in 84 HD sessions. aBaseline vs 4 h of HD. Wilcoxon test is used except for AT%, where paired t‐test is used. TAT=thrombin‐antithrombin and PF1+2=prothrombin fragment 1+2 (coagulation parameters), AT=antithrombin, PAP=plasmin antiplasmin (fibrinolysis parameter), β‐TG=β‐thromboglobulin (parameter of platelet activation).
Biological markers | Baseline | End of HD | P‐valuea |
Hb, g/dl | 11.5±1.0 | 12.0±1.3 | P<0.001 |
HCT% | 35.8±3.6 | 37.0±4.4 | P<0.001 |
Platelet count, 109/l | 255±89 | 260±79 | NS |
White cell count, 109/l | 8.7±3.4 | 8.4±3.4 | NS |
D‐dimer, mg/l | 0.38±0.72 | 0.54±1.26 | NS |
Fibrinogen, mmol/l | 14.4±3.4 | 15.8±3.8 | NS |
APTT, s | 37±13 | 39±16 | NS |
Thrombin clotting time, s | 21±8 | 26±13 | NS |
Biological markers | Baseline | End of HD | P‐valuea |
Hb, g/dl | 11.5±1.0 | 12.0±1.3 | P<0.001 |
HCT% | 35.8±3.6 | 37.0±4.4 | P<0.001 |
Platelet count, 109/l | 255±89 | 260±79 | NS |
White cell count, 109/l | 8.7±3.4 | 8.4±3.4 | NS |
D‐dimer, mg/l | 0.38±0.72 | 0.54±1.26 | NS |
Fibrinogen, mmol/l | 14.4±3.4 | 15.8±3.8 | NS |
APTT, s | 37±13 | 39±16 | NS |
Thrombin clotting time, s | 21±8 | 26±13 | NS |
Mean±SD in 84 HD sessions. D‐dimer, fibrinogen, APTT and thrombin clotting time is measured in 12 HD sessions. aBaseline vs end of HD. Wilcoxon test is used except for Hb, where paired t‐test is used. APTT is activated partial thromboplastin time.
Biological markers | Baseline | End of HD | P‐valuea |
Hb, g/dl | 11.5±1.0 | 12.0±1.3 | P<0.001 |
HCT% | 35.8±3.6 | 37.0±4.4 | P<0.001 |
Platelet count, 109/l | 255±89 | 260±79 | NS |
White cell count, 109/l | 8.7±3.4 | 8.4±3.4 | NS |
D‐dimer, mg/l | 0.38±0.72 | 0.54±1.26 | NS |
Fibrinogen, mmol/l | 14.4±3.4 | 15.8±3.8 | NS |
APTT, s | 37±13 | 39±16 | NS |
Thrombin clotting time, s | 21±8 | 26±13 | NS |
Biological markers | Baseline | End of HD | P‐valuea |
Hb, g/dl | 11.5±1.0 | 12.0±1.3 | P<0.001 |
HCT% | 35.8±3.6 | 37.0±4.4 | P<0.001 |
Platelet count, 109/l | 255±89 | 260±79 | NS |
White cell count, 109/l | 8.7±3.4 | 8.4±3.4 | NS |
D‐dimer, mg/l | 0.38±0.72 | 0.54±1.26 | NS |
Fibrinogen, mmol/l | 14.4±3.4 | 15.8±3.8 | NS |
APTT, s | 37±13 | 39±16 | NS |
Thrombin clotting time, s | 21±8 | 26±13 | NS |
Mean±SD in 84 HD sessions. D‐dimer, fibrinogen, APTT and thrombin clotting time is measured in 12 HD sessions. aBaseline vs end of HD. Wilcoxon test is used except for Hb, where paired t‐test is used. APTT is activated partial thromboplastin time.
Biological markers | In dialysis, warf‐ | Pre‐dialytic | Normal range | P‐valuea |
Coagulation | ||||
TAT, mg/l | 3.2±2.6 | 11.0±11.0 | 1.0−4.1 | P=0.001 |
PF1+2, nmol/l | 1.7±0.8 | 1.6±0.9 | 0.4−1.1 | NS |
AT% | 117±19 | 111±15 | >75% | NS |
Fibrinolysis | ||||
PAP, mg/l | 596±258 | 738±343 | 120−700 | NS |
Platelet activation | ||||
β‐TG, IU/ml | 116±287 | 86±27 | 10−40 | P=0.002 |
Biological markers | In dialysis, warf‐ | Pre‐dialytic | Normal range | P‐valuea |
Coagulation | ||||
TAT, mg/l | 3.2±2.6 | 11.0±11.0 | 1.0−4.1 | P=0.001 |
PF1+2, nmol/l | 1.7±0.8 | 1.6±0.9 | 0.4−1.1 | NS |
AT% | 117±19 | 111±15 | >75% | NS |
Fibrinolysis | ||||
PAP, mg/l | 596±258 | 738±343 | 120−700 | NS |
Platelet activation | ||||
β‐TG, IU/ml | 116±287 | 86±27 | 10−40 | P=0.002 |
Biological markers | In dialysis, warf‐ | Pre‐dialytic | Normal range | P‐valuea |
Coagulation | ||||
TAT, mg/l | 3.2±2.6 | 11.0±11.0 | 1.0−4.1 | P=0.001 |
PF1+2, nmol/l | 1.7±0.8 | 1.6±0.9 | 0.4−1.1 | NS |
AT% | 117±19 | 111±15 | >75% | NS |
Fibrinolysis | ||||
PAP, mg/l | 596±258 | 738±343 | 120−700 | NS |
Platelet activation | ||||
β‐TG, IU/ml | 116±287 | 86±27 | 10−40 | P=0.002 |
Biological markers | In dialysis, warf‐ | Pre‐dialytic | Normal range | P‐valuea |
Coagulation | ||||
TAT, mg/l | 3.2±2.6 | 11.0±11.0 | 1.0−4.1 | P=0.001 |
PF1+2, nmol/l | 1.7±0.8 | 1.6±0.9 | 0.4−1.1 | NS |
AT% | 117±19 | 111±15 | >75% | NS |
Fibrinolysis | ||||
PAP, mg/l | 596±258 | 738±343 | 120−700 | NS |
Platelet activation | ||||
β‐TG, IU/ml | 116±287 | 86±27 | 10−40 | P=0.002 |
Discussion
We have presented evidence that a single dose of dalteparin effectively inhibits clinical coagulation in the bubble trap or dialyser, and that time of dialysis is an independent variable for clinical clotting [1]. The correlation between the anticoagulant effect measured as anti‐FXa activity and the degree of clinical clotting was statistically significant. An anti‐FXa activity of 0.4 IU/ml or more at the end of dialysis almost abolished the risk of major clotting, which has been considered the primary goal of anticoagulation during HD.
However, clinical clotting is a crude measure of activation of coagulation. Activation of platelets and coagulation may take place without clinical clotting episodes. Subclinical activation of platelet and coagulation was evaluated by excluding measurements performed with simultaneous clinical clotting. We then found that TAT and PF1+2 values were significantly increased in non‐warfarin‐treated patients compared to the warfarin‐treated patients, P<0.00001 (Figure 3). This indicates thrombin generation during HD in spite of a clinically adequate dosage of dalteparin.
In the present study β‐TG was increased before the start of HD in all the patients, the highest values were seen in the warfarin‐treated patients (Figure 2). Increased levels of β‐TG in non‐dialysed patients with impaired renal function have previously been reported by others, and the renal excretion of β‐TG is reported to explain this [2,11,12]. The present study confirmed that the levels of β‐TG were higher than normal in pre‐dialytic patients with chronic renal failure. Adler et al. found increased levels of β‐TG before dialysis in patients on regular HD, but in non‐dialysed patients with chronic renal failure β‐TG was not statistically significantly different from normal [13]. The molecular weight of β‐TG is 36000 Da and the half life is 100 min, and β‐TG pass the glomerular basement membrane and is reabsorbed by the tubuli [13,14]. It has been suggested by many authors that β‐TG cannot be used as a marker of platelet release reaction in vivo in patients with renal insufficiency since β‐TG is excreted in the urine and its plasma level rises steeply with decreasing GFR [15]. Deficient catabolism of β‐TG in the diseased kidney rather than deficient excretion is also suggested to account for the elevated levels of β‐TG in patients with chronic renal insufficiency [16]. In addition to decreased clearance by the kidneys and deficient catabolism, another possible explanation for the increased baseline levels of β‐TG may be that the platelet activation does not normalize until the next dialysis session. Moreover, the increase in β‐TG during the HD sessions in the present study was statistically significant, and this can only be explained by platelet release with secretion of β‐TG from the β‐granulae. Increased levels of β‐TG during HD have also been found by others [8,17,18].
In conclusion, the levels of β‐TG were reduced before the next dialysis session, but not to normal levels, while TAT was reduced to normal levels before the next dialysis sessions in all the patients. The normalization of TAT before the next HD sessions in our patients is in contrast to the reports of Ambühl et al. who found TAT levels at baseline three times higher than the upper normal limit in 39 HD patients [3]. Ambühl et al. did not study pre‐dialytic, uraemic patients, as was done in the present study. The present study shows more than three times higher levels of TAT in pre‐dialytic uraemic patients compared with the dialysis patients before start of dialysis, P=0.0001. This is in contrast to the findings of Sagripanti et al. who found that baseline levels of TAT in 22 HD patients were higher than in patients on conservative treatment [2]. However, the normal range of TAT was not given in the latter report, and with our upper normal limit of 4.1 mg/l the reported mean level is just slightly increased compared with the normal range. There are few studies reporting the levels of coagulation parameters in pre‐dialytic, uraemic patients compared with dialysis patients. There is no rational explanation why TAT in the present study is three times higher in pre‐dialytic patients compared to dialysis patients. The results indicate that pre‐dialytic patients with chronic renal failure as well as dialysis patients have an activated coagulation system. In conclusion, the previous reports of the baseline levels of TAT in HD patients differ. The normal baseline levels of TAT in all the dialysis patients as well as the increased baseline levels of PF1+2 in the non‐warfarin‐treated dialysis patients in the present study are in accordance with the findings of Cella et al. [19,20]. The different half‐lives of TAT and PF1+2 are suggested to account for this discrepancy [19]. However, activation of coagulation in chronic renal insufficiency also between the dialysis sessions is another possible explanation. The rise in TAT and PF1+2 during HD reflects an activation of the coagulation system, and this increase is in accordance with previous reports [3,19].
Warfarin treatment did not inhibit the increase of β‐TG during HD, as was the case with TAT and PF1+2, which were kept in the normal range almost without increment with warfarin treatment. The fact that β‐TG levels were significantly higher in the patients on warfarin treatment both before and during dialysis is in contrast to previously reported lower levels of β‐TG in patients with normal renal function who received standard warfarin therapy [21]. This difference in observations cannot easily be explained.
It is well known that HD patients have an increased rate of cardiovascular disease, and cardiovascular complications account for more than 50% of all deaths in HD patients [22–24]. Activation of coagulation may be associated with cardiovascular events and platelet activation also with growth factors with potential adverse effects on heart and vessel structure [25]. The long‐term consequences may be accelleration of the atherosclerotic process. Abnormalities in the coagulation system are associated with an increased rate of cardiovascular complications, and PF1+2 is reported to correlate strongly with risk factors for cardiovascular disease [26]. As expected, the level of TAT and PF1+2 were lowered by warfarin in the present study. Thus, warfarin therapy reduces the tendency towards hypercoagulability in dialysis patients.
In conclusion, we previously found that a single dose of dalteparin effectively inhibits significant intradialytic clinical clotting for at least 4 h during HD [1]. However, activation markers of platelets and the coagulation system may be prominent even without clinical clotting episodes, and these markers increased during HD. Regression analysis showed that anti‐Fxa activity was negatively and significantly correlated to TAT and to PF1+2. This suggests that higher doses of dalteparin may reduce the activation of plasma coagulation, but our study was not designed to investigate this issue. However, our data also show that the activation of coagulation is reduced with simultaneous oral warfarin treatment. Activation of coagulation and platelets may be associated with cardiovascular complications. Assessments of this aspect of anticoagulation in HD patients seems warranted.
Correspondence and offprint requests to: Dr Solbjørg Sagedal, Med.avd., Rikshospitalet, N‐0027 Oslo, Norway.
Pharmacia & Upjohn AS is acknowledged for financial support. The data were presented in abstract form at the ‘Nordiske Nyredager’ in Helsinki, May 1999. We acknowledge Kåre Osnes for skilful assistance with the statistic calculations and Malin Trommer and Annika Skovlie for excellent technical assistance. The study was supported by grants from the Norwegian Society of Nephrology and Janssen Cilag, Norway.
References
Sagedal S, Hartmann A, Sundstrøm K, Bjørnsen S, Fauchald P, Brosstad F. A single dose of dalteparin effectively prevents clotting during haemodialysis.
Sagripanti A, Cupisti A, Baicchi U, Ferdeghini M, Morelli E, Barsotti G. Plasma parameters of the prothrombotic state in chronic uremia.
Ambühl PM, Wüthrich RP, Korte W, Schmid L, Krapf R. Plasma hypercoagulability in haemodialysis patients: impact of dialysis and anticoagulation.
Matzdorff AC, Kemkes‐Matthes B, Voss R, Pralle H. Comparison of b‐Thromboglobulin, Flow Cytometry, and Platelet Activation.
Sakata Y, Aoki N. Significance of cross‐linking of α2‐plasmin inhibitor to fibrin in inhibition of fibrinolysis and in hemostasis.
Ljungberg B, Jacobson SH, Lins lE, Pejler G. Effective anticoagulation by a low molecular weight heparin (Fragmin) in hemodialysis with a highly permeable polysulfone membrane.
Suzuki T, Kazuo O, Naganuma S et al. Clinical application of Fragmin (FR‐860) in hemodialysis: Multicenter cooperative study in Japan.
Warkentin T, Levine MN, Hirsh J et al. Heparin‐induced thrombocytopenia in patients treated with low‐molecular‐weight heparin or unfractionated heparin.
Deuber HJ, Schulz W. Reduced lipid concentrations during four years of dialysis with low molecular weight heparin.
Schmitt Y, Schneider H. Low‐molecular‐weight heparin (LMWH): influence on blood lipids in patients on chronic haemodialysis.
Wheeler DC. Cardiovascular disease in patients with chronic renal failure.
Kurihara Y, Nakayama H, Nakagawa S. Plasma β‐thromboglobulin and platelet factor 4 in renal insufficiency.
H van Hulsteijn, Ad van Es, Bertina R, Briet E. Plasma beta‐thromboglobulin and platelet factor 4 in renal failure.
Adler AJ, Lundin AP, Friedman EA, Berlyne GM. Effect of hemodialysis on plasma beta thromboglobulin levels.
Dawes J, Smith RC, Pepper DS. The release, distribution and clearance of human b‐thromboglobulin and platelet factor 4.
Zahavi J, Kakkar VV. β‐Thromboglobulin—a specific marker of in‐vivo platelet release reaction.
Guzzo J, Niewiarowski S, Musial J et al. Secreted platelet proteins with antiheparin and mitogenic activities in chronic renal failure.
Knudsen F, Dyerberg J. Platelets and antithrombin III in uraemia: the acute effect of haemodialysis.
Green D, Santhanam S, Krumlovsky FA, del Greco F. Elevated β‐thromboglobulin in patients with chronic renal failure: effect of hemodialysis.
Cella G, Vertolli U, Naso A et al. Tissue factor pathway inhibitor (TFPI) activity in uremic patients during hemodialysis.
Kario K, Matsou T, Yamada T, Nakao K, Shimano C, Matsuo M. Factor VII hyperactivity in chronc dialysis patients.
Lip GYH, Lip PL, Zarifis J et al. Fibrin D‐Dimer and b‐Thromboglobulin as markers of thrombogenesis and platelet activation in atrial fibrillation. Effects of introducing ultra‐low‐dose warfarin and aspirin.
Wanner C, Zimmermann J, Quaschning T, Galle J. Inflammation, dyslipidemia and vascular risk factors in hemodialysis patients.
De Lima JJG, Sesso R, Abensur H et al. Predictors of mortality in long‐term haemodialysis patients with a low prevalence of comorbid conditions.
Comments