Crossing borders: the role of the endothelial glycocalyx and intravascular haemostasis in vascular complications of diabetes mellitus - Chapter 6: Improved glycaemic control by insulin therapy ameliorates the prothrombotic state in type 2 diabetes

(1)

UvA-DARE (Digital Academic Repository)

Crossing borders: the role of the endothelial glycocalyx and intravascular

haemostasis in vascular complications of diabetes mellitus

Lemkes, B.A.

Publication date

2011

Link to publication

Citation for published version (APA):

Lemkes, B. A. (2011). Crossing borders: the role of the endothelial glycocalyx and

intravascular haemostasis in vascular complications of diabetes mellitus.

General rights

It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).

Disclaimer/Complaints regulations

If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You

(2)

CHAPTER

6

Improved glycaemic control by initiation of insulin therapy

ameliorates the prothrombotic state in type 2 diabetes

Bregtje A. Lemkes, Sanne G. Swinnen, J.Hans DeVries, Max Nieuwdorp, Ricardo de Grooth, Joost B.L. Hoekstra, Joost C. Meijers and Frits Holleman

(3)

R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 Chapter 6

Abstract

Objective

We set out to determine whether improved glycaemic control by initiation of basal insulin therapy would improve the prothrombotic state and the hyperglycaemia-induced damage to the endothelial glycocalyx in type 2 diabetes.

Research design and methods

We studied a random sample of 106 participants from the 24 week treat-to-target L2T3 trial in which inadequately controlled patients with type 2 diabetes were randomized to insulin glargine or detemir. Combined measures of coagulation (antithrombin, factor VIII, von Willebrand factor, prothrombin fragment 1+2) and fibrinolysis (d-dimer, plasmin-antiplasmin complex, plasminogen-activator inhibitor-1, clot lysis time) as well as circulating components of the endothelial glycocalyx (syndecan-1, hyaluronan, hyaluronidase) and inflammation (hsCRP) were determined at baseline and at trial completion.

Results

During the trial, HbA1c levels dropped from 8.8 ± 0.9% to 7.2 ± 0.9% (mean ± SD) in this cohort. Prothrombin fragment 1+2, plasminogen-activator inhibitor-1 and clot lysis time decreased significantly (p<0.05) after treatment and plasmin-antiplasmin complexes increased (p<0.001). In multivariate analysis, endpoint HbA1c was the most important predictor of these improvements. Markers of glycobiology and inflammation did not change.

Conclusions

Improvement in glycemic control by initiation of basal insulin therapy ameliorates the prothrombotic state in type 2 diabetes, without affecting markers of inflammation or glycobiology.

(4)

R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39 The pr othr

ombotic state in diabetes

The majority of patients with type 2 diabetes die of cardiovascular disease. Apart from the contribution of diabetes to the development of atherosclerosis, diabetes also leads to a prothrombotic state which predisposes patients with type 2 diabetes to atherothrombotic events1, 2_{. Haemostasis is governed by the interplay between the vascular endothelium,}

the endothelial glycocalyx and the coagulation system, all of which are affected by hyperglycemia3_{. During hyperglycaemia the balance between coagulation and}

fibrinolysis shifts towards a procoagulant state. Worse glycemic control in type 2 diabetes is associated with both increased thrombin generation and decreased fibrinolysis4-6_.

Hyperglycaemia also directly affects the endothelial glycocalyx, which results in enhanced platelet-endothelial cell adhesion and release of coagulation and fibrinolysis factors harboured within the glycocalyx7-9_{. Inflammation driven shedding of one of its}

main components, hyaluronan, has been shown to correlate with plasminogen-activator inhibitor (PAI-1) levels10_{, and an independent association between PAI-1 levels and two}

hyaluronan metabolism genes was found in a population based study11_{. However, while}

the negative effects of hyperglycaemia on these factors in haemostasis have been well documented, less is known about the effects of improving hyperglycaemia. Small scale studies have shown beneficial effects of oral glucose lowering therapies on markers of coagulation and fibrinolysis12, 13_{, but the data on the effects of insulin therapy are}

conflicting, ranging from no effect to beneficial effects on only markers of fibrinolysis in a hospital setting14-16_.

Thus, in this study we aimed to obtain an integral view of the effect of improving hyperglycaemia by initiation of basal insulin therapy on the endothelial glycocalyx, coagulation, fibrinolysis and inflammation in type 2 diabetes.

Research design and methods

Study population

The study was performed in 106 participants of the L2T3 trial17_{, during which 964}

insulin-naïve patients with type 2 diabetes inadequately controlled on oral glucose lowering drugs started basal insulin therapy. The 106 subjects included in this pre-specified sub analysis were randomly selected from all randomized patients after trial completion. Included subjects were 40-75 years old and had a diabetes duration of more than 1 year. HbA1c levels at study entry were 7.0% - 10.5% (53 mmol/mol – 91 mmol/mol) and oral glucose lowering therapy consisted of metformin 1000 mg/day as a minimum. All subjects gave written informed consent and the study was approved by ethics committees at all centres. The trial was conducted in accordance with the principles of the Declaration of Helsinki (ClinicalTrials.gov number: NCT00405418).

(5)

R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 Chapter 6 Study design

This study was part of a 24-week, multinational, randomized, parallel-group, treat-to-target trial comparing insulin glargine once-daily to detemir twice-daily. The titration target was a fasting plasma glucose (FPG) level of < 5.6 mmol/l. The primary outcome was the percentage of patients reaching HbA1c < 7.0% without symptomatic hypoglycaemia during the trial. An extensive description of the study as well as the primary results have been published elsewhere17, 18_{. In short, non-inferiority of once-daily insulin glargine was}

demonstrated when compared to twice daily insulin detemir. Endpoint insulin dose was lower in the patients allocated to insulin glargine; weight gain was less in the insulin detemir group while hypoglycaemia risk was similar for both treatments. For the purpose of this analysis, a citrated blood sample was taken at baseline and at end of study. To determine the effect of improved glycemic control following insulin therapy, parameters of coagulation, fibrinolysis, glycobiology and inflammation were determined in the pre- and post-treatment blood samples of the selected patients.

Outcomes

The primary outcome of this sub study was change from baseline to 24 weeks in coagulation, fibrinolysis, glycobiology and inflammation parameters in all subjects. Secondary outcomes were differences in these changes between patients who did reach the treatment goals of an HbA1c level < 7% or FPG < 5.6 mmol/l and those who did not, as well as the correlation of baseline measurements of the glycobiology parameters and high sensitive C-reactive protein (hsCRP) to the coagulation and fibrinolysis parameters to determine the association between the endothelial glycocalyx, inflammation and the thrombotic system.

Markers of coagulation

Coagulation status was determined by measuring von Willebrand factor (vWf), prothrombin fragment 1+2, factor VIII and antithrombin. VWF plays a major role in haemostasis for the arrest of blood platelets at the site of injury and binding of factor VIII, but is also an established marker of endothelial dysfunction. Prothrombin fragment 1+2 are released when thrombin is formed from prothrombin and therefore provide an in vivo measure of thrombin formation. Factor VIII is activated by thrombin and maintains the coagulation cascade by enhancing thrombin formation as cofactor for activated factor IX, thereby enhancing activation of factor X, while antithrombin inactivates thrombin and other coagulation enzymes and thereby prevents the formation of fibrin.

(6)

Markers of fibrinolysis

Fibrinolysis was assessed by measuring d-dimer, plasmin-alpha-antiplasmin (PAP) complexes, plasminogen-activator inhibitor (PAI-1) antigen and clot lysis time. D-dimer is a fibrin degradation product, which is dependent on the amount of fibrin that is generated (coagulation) as well as the ability of the fibrinolytic system to degrade the generated fibrin (fibrinolysis). PAI-1 is an inhibitor of tissue plasminogen activator and regulates the formation of plasmin, the most important fibrinolytic enzyme. PAP complexes serve as an indicator of recent in vivo fibrinolytic activity, as a2 antiplasmin is the most important circulating inhibitor of plasmin, while clot lysis time is an ex vivo measure of fibrinolytic capacity.

Markers of glycobiology

To determine turnover of endothelial glycocalyx components we measured syndecan-1, hyaluronan and hyaluronidase. Syndecan-1 is a transmembrane proteoglycan which serves as an anchor protein for the glycocalyx by binding glycosaminoglycans. The most abundant glycosaminoglycan in the glycocalyx is hyaluronan. Hyaluronidase is its regulatory enzyme and hyaluronidase plasma levels reflect hyaluronan catabolism.

Marker of inflammation

Since type 2 diabetes is also considered an inflammatory condition, we also determined HsCRP to establish whether treatment effects on other parameters might be related to changes in the inflammatory state.

Laboratory analysis

All outcome parameters were determined in the Department of Experimental Vascular Medicine of the Academic Medical Centre in Amsterdam, using a fasting, morning plasma sample from blood anticoagulated by trisodium citrate. After centrifugation at high speed all samples were stored at -80°C until analysis. HsCRP, antithrombin, factor VIII and d-dimer were measured on an automated coagulation analyzer (Behring Coagulation System, BSC) using protocols and reagents from the manufacturer (Siemens Healthcare Diagnostics, Marburg, Germany). Antigen levels of vWF were assayed by ELISA using antibodies from Dako (Glostrup, Denmark). Plasma levels of prothrombin fragment 1+2, PAI-1 and PAP complexes were also measured by ELISA (prothrombin fragment 1+2: Siemens Healthcare Diagnostics, PAI-1: Hyphen BioMed, Andrésy, France and PAP: DRG, Marburg, Germany). Clot lysis time was defined as the time between half-maximal lysis and half-maximal clotting and determined as described previously19_{. Syndecan-1}

was measured by ELISA from Diaclone (Sanquin, Amsterdam, the Netherlands), plasma hyaluronan levels were also assayed by ELISA (Echelon Biosciences, Salt Lake City, UT) and plasma hyaluronidase levels were determined with a previously described assay20_.

(7)

R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 Chapter 6 Statistical analysis

Data are presented as mean ± SD or median (interquartile range, IQR) depending on the distribution of data. Differences in patient characteristics before and after treatment were tested by a paired student’s t-test. The difference between outcome parameters at baseline and at study end was tested by a paired t-test or Wilcoxon signed rank test, depending on the distribution of the data. Determinants of change in outcome parameters were established by multivariate regression analysis, including all parameters which showed a borderline significant (p<0.10) effect in univariate analyses as covariates. All baseline characteristics (table 1) as well as changed clinical and laboratory parameters (FPG, HbA1c, insulin dose, treatment allocation, weight, triglycerides and LDL) were included in these analyses. To determine whether the subjects who reached the HbA1c and/or FPG goal differed from those who did not, we compared the percentage change in outcome parameters between these groups by using an unpaired t-test or Mann Whitney U test. We used the percentage change to adjust for any baseline differences between these groups. Correlations were determined by Spearman rank’s correlation. All analyses were performed using Predictive analytics software (PASW) version 18.0 (SPSS inc., Chicago, Il, USA), a p-value of <0.05 was considered statistically significant.

Results

Baseline Characteristics

Two subjects were excluded from the analysis because they did not receive treatment. Baseline characteristics of the included subjects (n=104) are summarized in table 1. Overall, this was a predominantly white population with a mean diabetes duration of 9.1 ± 5.4 years and a BMI of 30.8 ± 5.0 kg/m2_{. After 24 weeks of insulin treatment there}

was a significant drop in HbA1c levels from 8.8 ± 0.9% to 7.2 ± 0.9% (p<0.001). Fasting plasma glucose was lowered from 10.5 ± 2.6 mmol/l to 6.7 ± 2.0 mmol/l (p<0.001). End point insulin dose was 70 ± 54 units (U) or 0.8 ± 0.5U per kilogram bodyweight. During the trial bodyweight increased from 86.7 ± 19.6 kg to 87.9 ± 20.1 kg (p<0.001). Although HDL and LDL levels were unaffected, triglycerides were lowered from 2.5 ± 1.7 mmol/l to 1.9 ± 1.1 mmol/l (p<0.001). No effect of treatment allocation was found for any of the outcome parameters of this sub study, therefore all subjects were analyzed together, irrespective of treatment allocation.

(8)

Table 1. Baseline characteristics

Clinical characteristics N 104 M/V (%) 56/44 Age (yrs) 58.0 ± 7.8 Race (white/asian/black/other %) 80/13/2/5 BMI (kg/m2) 30.8 ± 5.0

Duration of diabetes (yrs) 9.1 ± 5.4

Systolic Blood Pressure (mmHg) 134 ± 15

Diastolic Blood Pressure (mmHg) 79 ± 10

Presence of macro vascular disease (%) 11

Medication use at baseline

Metformin (%) 100 Sulphonylurea (%) 86 TZD (%) 22 Antithrombotics (%) 40 RAS inhibition (%) 45 Statins (%) 58 Laboratory characteristics HbA1c (%) 8.8 ± 0.9 FPG (mmol/l) 10.5 ± 2.6 Insulin (pmol/l) 98 ± 84 C-peptide (pmol/l) 1118 ± 595 HDL (mmol/l) 1.2 ± 0.4 LDL (mmol/l) 2.5 ± 0.8 Triglycerides (mmol/l) 2.5 ± 1.7 Microalbuminuria (%) 22

Data expressed as mean ± SD. BMI indicates body mass index; TZD, thiazolidinedione; RAS, renin-angiotensin system; HbA1c, hemoglobin-A1c; FPG, fasting plasma glucose; HDL, high-density lipoprotein cholesterol; LDL, low-density lipoprotein cholesterol.

Coagulation

Median levels of all coagulation parameters at baseline and after 24 weeks are listed in table 2. No significant changes were detected in levels of vWf, factor VIII or antithrombin before and after treatment. Prothrombin fragment 1+2 decreased from a median (interquartile range) of 165 (136 to 218) pmol/l to 163 (135 to 194) pmol/l in the entire cohort (p=0.039). In multivariate analysis, FPG < 5.6 mmol/l was found to be the best predictor of the reduction in prothrombin fragment 1+2 (beta 0.233, p=0.025). When comparing subjects who did reach the HbA1c or FPG target to those who did not, significant differences were found in both prothrombin fragment 1+2 and factor VIII levels. Subjects who did reach the HbA1c goal showed a median reduction of -14 (-24 to 7)% in prothrombin fragment 1+2, compared to -1 (-17 to 21)% in those who did not (p=0.025) and a -6 (-14 to 2)% reduction in factor VIII levels compared to a 1 (-10 to

(9)

13)% increase (p=0.046) in patients who were unable to reach the HbA1c goal (figure 1a). Similar differences were found when comparing subjects based on their endpoint FPG levels. Those who had a FPG < 5.6 mmol/l showed a -14 (-26 to 8)% decrease in prothrombin fragment 1+2 compared to a -1 (-16 to 20)% decrease in those who did not (p=0.008) and a -6 (-21 to 3) % reduction in factor VIII compared to a 1 (-16 to 20)% increase in those who did not reach the FPG goal (p=0.046).

Fibrinolysis

All markers of fibrinolysis, with the exception of d-dimer, showed significant improvements after 24 weeks of insulin therapy (table 2). PAP levels were increased from 369 (306 to 466) µg/l to 416 (343 to 515) µg/l (p<0.001), PAI-1 was reduced from 100 (64 to 159) ng/ml to 87 (56 to 139) ng/ml (p=0.007) and clot lysis time was reduced from 70.6 (60.0 to 85.1) min to 64.6 (57.6 to 76.7) min (p=0.001). In multivariate analysis, FPG change and baseline LDL levels were the most important determinants of PAP increase (beta -0.244, p=0.015 and beta -0.397, p<0.001 respectively), while endpoint HbA1c was the strongest determinant of the reduction in clot lysis time (beta 0.242, p=0.039) and PAI-1 levels (beta 0.342, p=0.001). The latter was also significantly influenced by baseline LDL levels (beta 0.214, p=0.039).

When comparing patients who did reach the treatment target to those who did not, clot lysis time reduction was significantly larger in the patients who achieved an HbA1c < 7%, -11 (-20 to 1) %, compared to those who did not, -1 (-10 to 6)%, p =0.007 (figure 1b). PAI-1 level reduction was significantly greater in subjects with an endpoint FPG < 5.6 mmol, -27 (-55 to 7) % compared to those who did not, -4 (-42 to 51)%, p=0.044.

Glycobiology

Syndecan-1 levels were lowered by 3 ng/ml from 27 (17-69) ng/ml to 24 (16-60) ng/ml, which was borderline significant (p=0.063). Both hyaluronan and hyaluronidase levels did not alter significantly after 24 weeks (table 2). None of the (changes in) glycobiology parameters differed between the patients who did reach the treatment goal and those who did not.

Inflammation

No significant differences were found in hsCRP levels before and after treatment (table 2), also when comparing patients with and without target HbA1c or FPG levels.

(10)

95                

Table 2. Outcome parameters at baseline and 24-weeks

Baseline 24 Weeks P-value

Coagulation

Antithrombin (%) 106 (99-114) 104 (97-113) 0.059

Factor VIII (%) 71 (55-84) 65 (52-84) 0.095

Von Willebrand factor (%) 110 (84-146) 112 (83-141) 0.583

Prothrombin fragment (1+2 pmol/l) 165 (136-218) 163 (135-194) 0.039

Coagulation/fibrinolysis

D-dimer (mg/l) 0.3 (0.2-0.4) 0.3 (0.2-0.4) 0.713

Fibrinolysis

Plasmin-antiplasmin complex (µg/l) 369 (306-466) 416 (343-515) <0.001 Plasminogen-activator inhibitor-1 (ng/ml) 100 (64-159) 87 (56-139) 0.007 Clot lysis time (min) 70.6 (60.0-85.1) 64.6 (57.6-76.7) 0.001

Glycobiology Syndecan-1 (ng/ml) 27 (17-69) 24 (16-60) 0.063 Hyaluronan (ng/ml) 65 (44-87) 61 (45-82) 0.448 Hyaluronidase (U/ml) 90 (74-103) 88 (74-104) 0.886 Inflammation HsCRP (mg/l) 1.4 (0.6-3.7) 1.4 (0.6-4.3) 0.385

Data expressed as median (interquartile range), difference tested by Wilcoxon Signed Ranks test. HsCRP indicates high sensitive C-reactive protein.

Figure 1. Percentage change of prothrombin fragment 1+2 levels (a.) and clot lysis time (b.) in subjects with

endpoint HbA1c < 7% and >= 7%. Difference tested by Kruskall Wallis test.

(11)

Correlation between outcome parameters

At baseline, there were significant correlations between hyaluronan and antithrombin (rho -0.306, p=0.004), factor VIII (rho 0.238, p=0.027) and vWf (rho 0.326, p=0.002). Of the fibrinolytic parameters hyaluronan correlated only with clot lysis time (rho -0.295, p=0.007).

Hyaluronidase showed no correlation with any of the markers of coagulation and fibrinolysis and syndecan-1 correlated only to prothrombin fragment 1+2 levels (rho -0.279, p=0.006). HsCRP showed significant correlations to factor VIII (rho 0.306, p=0.002), vWf (rho 0.206, p=0.039), PAI-1 (rho 0.199, p=0.047) and clot lysis time (rho 0.224, p=0.029) at baseline. No significant associations between hsCRP levels and any glycobiology parameters were detected.

Conclusions

We set out to study the effects of improved glycemic control following initiation of insulin therapy on the prothrombotic state in type 2 diabetes mellitus. We found a remarkably consistent improvement in fibrinolysis, some improvement in coagulation but no effect on the markers of the endothelial glycocalyx or inflammation.

In the study HbA1c levels improved by 1.6 % and the FPG levels dropped by 3.8 mmol/l. This was accompanied by improvements in PAI-1, PAP, clot lysis time, factor VIII and prothrombin fragment 1+2 levels, denoting a shift towards a more favourable, less thrombotic state. These data are in line with other studies which have shown beneficial effects of glucose lowering therapies on prothrombin fragment 1+2 levels, PAI-1 levels and clot lysis time in type 2 diabetes13, 21, 22_{. However, these studies were performed with}

oral glucose lowering drugs and it has been shown that various diabetes treatments have differential effects on these biomarkers23_{. The data on the effects of glucose lowering by}

insulin therapy have been more conflicting. Reynolds and colleagues found no effect of insulin glargine on PAI-1 levels, but their sample size (n=20) may have been too small to detect such an effect16_{. Another larger study included 61 patients with type 2 diabetes}

and found no effect on several markers of coagulation during insulin therapy, however they did not measure prothrombin fragment 1+2 and factor VIII, the only coagulation parameters that showed improvements in our cohort, nor did they determine any measures of fibrinolysis15_{. Another study investigated the effect of improved glycemic}

control by insulin therapy in hospitalized patients with type 2 diabetes and found improvements in both PAP complexes and PAI-1 levels14_.

(12)

and FPG change, but not insulin dose or treatment allocation, were predictors of the improvements found. This suggests that hyperglycaemia directly influences factors of coagulation and fibrinolysis. In vitro studies have shown that oxidative stress resulting from hyperglycaemia or hyperglycaemia driven formation of advanced glycation end products influences gene transcription of haemostatic parameters26, 27_{. Also, in patients}

with type 1 diabetes a rise in blood glucose was found to be accompanied by increases in markers of oxidative stress, coagulation activation and inhibition of fibrinolysis28_.

Although we did not determine any parameters of oxidative stress, treatment of hyperglycaemia in type 2 diabetes is known to improve oxidative stress levels and this may have mediated the beneficial effects on coagulation and fibrinolysis we detected. Alternatively, improvements in glycemic control are also likely to result in less dense fibrin clots, which are more readily dissolved, since non-enzymatic glycation of fibrin increases clot density in patients with type 2 diabetes29_{. This could also explain the}

improvements in clot lysis time in this cohort.

Although previous studies have suggested an association between hyperglycaemia and plasma hyaluronan levels in diabetes9, 20, 30_{, we could not detect an effect of improved}

glycemic control on hyaluronan metabolism. This could mean that the alterations in hyaluronan metabolism found in type 2 diabetes are either not reversible or under the influence of other factors than hyperglycaemia. Likewise, no significant change in the plasma syndecan-1 levels was detected. Although these results do not support a beneficial effect of improved glycemic control on the composition of the endothelial glycocalyx in type 2 diabetes, such an effect can not be excluded based on this data. The glycocalyx consists of a wide array of proteoglycans and adhered glycosaminoglycans and, although hyaluronan is its main glycosaminoglycan, other components may be affected. Furthermore, the measurement of circulating components of the glycocalyx may be a good reflection of the effect of acute damage to the layer9, 31_{, but may not be the}

appropriate measure for detecting chronic changes.

Finally, we did not detect any changes in hsCRP levels before and after treatment, although it must be noted that hsCRP levels were generally low. Nonetheless, the lack of change in CRP levels is in line with the results of the recent LANCET trial, in which no effect on CRP levels of insulin glargine over placebo was found in early type 2 diabetes, despite improvements in glycemic control32_.

Several aspects of our study deserve comment. First, this study was embedded in a large clinical trial comparing two basal insulin analogues and therefore its design was dependent on design of the main trial. This design did not allow us to include a healthy comparator- or a placebo-group. Second, the outcome of our trial is based on change in risk markers, but provides no evidence as to the clinical benefit of improved glycemic control by basal insulin therapy. Although PAI-1 levels, PAP levels and clot lysis time are

(13)

independent risk factors for myocardial infarction while factor VIII and clot lysis time are predictive of venous thrombosis33-37_{, thus far only moderate benefit of glucose control in}

prevention of cardiovascular disease has been demonstrated38_.

In summary, improvement in glycemic control by initiation of basal insulin therapy in poorly controlled type 2 diabetes has a beneficial effect on fibrinolysis and some markers of coagulation. This effect was most prominent in patients who reached HbA1c or FPG targets. Since these markers have been found to be predictive of future thrombotic events our data give support to the current treatment goal of an HbA1c level < 7% in type 2 diabetes.

Acknowledgements

This study was supported by sanofi-aventis. M. Nieuwdorp is supported by an Innovational Research Incentive grant from the Netherlands Organization for Health Research and Development 2008 (016.096.044).

(14)

References

1. Carr ME. Diabetes mellitus: a hypercoagulable state. J Diabetes Complications 2001;15(1):44-54. 2. Grant PJ. Diabetes mellitus as a prothrombotic condition. J Intern Med 2007;262(2):157-172. 3. Lemkes BA, Hermanides J, DeVries JH, Holleman F, Meijers JC, Hoekstra JB. Hyperglycemia: a

prothrombotic factor? J Thromb Haemost 2010;8(8):1663-1669.

4. Aoki I, Shimoyama K, Aoki N et al. Platelet-dependent thrombin generation in patients with diabetes mellitus: effects of glycemic control on coagulability in diabetes. J Am Coll Cardiol 1996;27(3):560-566.

5. Cefalu WT, Schneider DJ, Carlson HE et al. Effect of combination glipizide GITS/metformin on fibrinolytic and metabolic parameters in poorly controlled type 2 diabetic subjects. Diabetes Care 2002;25(12):2123-2128.

6. Osende JI, Badimon JJ, Fuster V et al. Blood thrombogenicity in type 2 diabetes mellitus patients is associated with glycemic control. J Am Coll Cardiol 2001;38(5):1307-1312.

7. Nieuwdorp M, Meuwese MC, Vink H, Hoekstra JB, Kastelein JJ, Stroes ES. The endothelial glycocalyx: a potential barrier between health and vascular disease. Curr Opin Lipidol 2005;16(5):507-511.

8. Vink H, Constantinescu AA, Spaan JA. Oxidized lipoproteins degrade the endothelial surface layer : implications for platelet-endothelial cell adhesion. Circulation 2000;101(13):1500-1502. 9. Nieuwdorp M, van Haeften TW, Gouverneur MC et al. Loss of endothelial glycocalyx during

acute hyperglycemia coincides with endothelial dysfunction and coagulation activation in vivo. Diabetes 2006;55(2):480-486.

10. Devaraj S, Yun JM, Adamson G, Galvez J, Jialal I. C-reactive protein impairs the endothelial glycocalyx resulting in endothelial dysfunction. Cardiovasc Res 2009;84(3):479-484.

11. Lanktree MB, Johansen CT, Anand SS et al. Genetic variation in hyaluronan metabolism loci is associated with plasma plasminogen activator inhibitor-1 concentration. Blood 2010;116(12):2160-2163.

12. Fonseca VA, Reynolds T, Hemphill D et al. Effect of troglitazone on fibrinolysis and activated coagulation in patients with non-insulin-dependent diabetes mellitus. J Diabetes Complications 1998;12(4):181-186.

13. Gregorio F, Ambrosi F, Manfrini S et al. Poorly controlled elderly Type 2 diabetic patients: the effects of increasing sulphonylurea dosages or adding metformin. Diabet Med 1999;16(12):1016-1024.

14. Aso Y, Okumura KI, Yoshida N, Tayama K, Takemura Y, Inukai T. Enhancement of fibrinolysis in poorly controlled, hospitalized type 2 diabetic patients by short-term metabolic control: association with a decrease in plasminogen activator inhibitor 1. Exp Clin Endocrinol Diabetes 2004;112(4):175-180.

15. Knobl P, Schernthaner G, Schnack C et al. Haemostatic abnormalities persist despite glycaemic improvement by insulin therapy in lean type 2 diabetic patients. Thromb Haemost 1994;71(6):692-697.

16. Reynolds LR, Kingsley FJ, Karounos DG, Tannock LR. Differential effects of rosiglitazone and insulin glargine on inflammatory markers, glycemic control, and lipids in type 2 diabetes. Diabetes Res Clin Pract 2007;77(2):180-187.

(15)

17. Swinnen SG, Dain MP, Aronson R et al. A 24-week, randomized, treat-to-target trial comparing initiation of insulin glargine once-daily with insulin detemir twice-daily in patients with type 2 diabetes inadequately controlled on oral glucose-lowering drugs. Diabetes Care 2010;33(6):1176-1178.

18. Swinnen SG, Snoek FJ, Dain MP, DeVries JH, Hoekstra JB, Holleman F. Rationale, design, and baseline data of the insulin glargine (Lantus) versus insulin detemir (Levemir) Treat-To-Target (L2T3) study: A multinational, randomized noninferiority trial of basal insulin initiation in type 2 diabetes. Diabetes Technol Ther 2009;11(11):739-743.

19. Verkleij CJ, Bruijn RD, Meesters EW, Gerdes VE, Meijers JC, Marx PF. The Hemostatic System in Patients With Type 2 Diabetes With and Without Cardiovascular Disease. Clin Appl Thromb Hemost 2010.

20. Nieuwdorp M, Mooij HL, Kroon J et al. Endothelial glycocalyx damage coincides with microalbuminuria in type 1 diabetes. Diabetes 2006;55(4):1127-1132.

21. Derosa G, Dangelo A, Ragonesi PD et al. Effects of rosiglitazone and pioglitazone combined with metformin on the prothrombotic state of patients with type 2 diabetes mellitus and metabolic syndrome. J Int Med Res 2006;34(5):545-555.

22. Grant PJ. Beneficial effects of metformin on haemostasis and vascular function in man. Diabetes Metab 2003;29(4 Pt 2):6S44-6S52.

23. Fonseca VA, Theuma P, Mudaliar S, Leissinger CA, Clejan S, Henry RR. Diabetes treatments have differential effects on nontraditional cardiovascular risk factors. J Diabetes Complications 2006;20(1):14-20.

24. Miyagawa R, Asakura T, Nakamura T et al. Increased expression of plasminogen activator inhibitor type-1 (PAI-1) in HEPG2 cells induced by insulin mediated by the 3’-untranslated region of the PAI-1 gene and its pharmacologic implications. Coron Artery Dis 2010;21(3):144-150. 25. Stegenga ME, van der Crabben SN, Levi M et al. Hyperglycemia stimulates coagulation, whereas

hyperinsulinemia impairs fibrinolysis in healthy humans. Diabetes 2006;55(6):1807-1812. 26. Iwasaki Y, Kambayashi M, Asai M, Yoshida M, Nigawara T, Hashimoto K. High glucose alone,

as well as in combination with proinflammatory cytokines, stimulates nuclear factor kappa-B-mediated transcription in hepatocytes in vitro. J Diabetes Complications 2007;21(1):56-62. 27. Min C, Kang E, Yu SH, Shinn SH, Kim YS. Advanced glycation end products induce apoptosis and

procoagulant activity in cultured human umbilical vein endothelial cells. Diabetes Res Clin Pract 1999;46(3):197-202.

28. Iorio A, Federici MO, Mourvaki E et al. Impaired endothelial antithrombotic activity following short-term interruption of continuous subcutaneous insulin infusion in type 1 diabetic patients. Thromb Haemost 2007;98(3):635-641.

29. Brownlee M, Vlassara H, Cerami A. Nonenzymatic glycosylation reduces the susceptibility of fibrin to degradation by plasmin. Diabetes 1983;32(7):680-684.

30. Nieuwdorp M, Holleman F, de GE et al. Perturbation of hyaluronan metabolism predisposes patients with type 1 diabetes mellitus to atherosclerosis. Diabetologia 2007;50(6):1288-1293. 31. Rehm M, Bruegger D, Christ F et al. Shedding of the endothelial glycocalyx in patients undergoing

(16)

33. Kamphuisen PW, Eikenboom JC, Rosendaal FR et al. High factor VIII antigen levels increase the risk of venous thrombosis but are not associated with polymorphisms in the von Willebrand factor and factor VIII gene. Br J Haematol 2001;115(1):156-158.

34. Lisman T, de Groot PG, Meijers JC, Rosendaal FR. Reduced plasma fibrinolytic potential is a risk factor for venous thrombosis. Blood 2005;105(3):1102-1105.

35. Meltzer ME, Doggen CJ, de Groot PG, Rosendaal FR, Lisman T. Reduced plasma fibrinolytic capacity as a potential risk factor for a first myocardial infarction in young men. Br J Haematol 2009;145(1):121-127.

36. Meltzer ME, Doggen CJ, de Groot PG, Rosendaal FR, Lisman T. Plasma levels of fibrinolytic proteins and the risk of myocardial infarction in men. Blood 2010;116(4):529-536.

37. Thogersen AM, Jansson JH, Boman K et al. High plasminogen activator inhibitor and tissue plasminogen activator levels in plasma precede a first acute myocardial infarction in both men and women: evidence for the fibrinolytic system as an independent primary risk factor. Circulation 1998;98(21):2241-2247.

38. Kelly TN, Bazzano LA, Fonseca VA, Thethi TK, Reynolds K, He J. Systematic review: glucose control and cardiovascular disease in type 2 diabetes. Ann Intern Med 2009;151(6):394-403.