Crossing borders: the role of the endothelial glycocalyx and intravascular haemostasis in vascular complications of diabetes mellitus - Chapter 8: The influence of aspirin dose and glycaemic control on platelet inhibition in patients with 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 will be contacted as soon as possible.

(2)

CHAPTER

8

The influence of aspirin dose and glycaemic control on platelet

inhibition in patients with type 2 diabetes

Bregtje A. Lemkes, Lonneke Bähler, Pieter W. Kamphuisen, An K. Stroobants, Erik J. van den Dool, Joost B. Hoekstra, Rienk Nieuwland, Victor E. Gerdes 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 R36 R37 R38 R39 Chapter 8

Abstract

Background

Low-dose aspirin seems to offer no benefit in primary prevention of cardiovascular disease in type 2 diabetes mellitus (DM2). The anti-platelet effect may be diminished by hyperglycaemia or inadequate dosing of aspirin.

Objectives

To study the effects of both glycaemic control and increasing aspirin dose on platelet response to aspirin in DM2 patients and matched controls.

Patients/methods

Platelet effects of increasing doses of aspirin (30-100-300 mg daily) were prospectively assessed in 94 DM2 patients and 25 matched controls by measuring thromboxane levels in urine (11-dhTxB2) and platelet aggregation using VerifyNow and light transmission aggregometry (LTA). DM2 patients were stratified for glycaemic control (HbA1c ≤53 mmol/mol, 53-69 mmol/mol, ≥69 mmol/mol).

Results

At baseline, median 11-dhTxB2 excretion was higher in the poorly controlled patients (77[IQR 61-109] ng/mmol), and the moderately controlled (84[48-102]) compared to the well controlled patients (64[34-93]) and controls (53[37-73]), p=0.007. 30 mg of aspirin reduced 11-dhTxB2 excretion to 31[23-51], 29[18-43] and 24[16-25] ng/mmol in the poorly, moderately and well controlled patients respectively, and to 19[11-25] ng/mmol in controls, p<0.001. VerifyNow and LTA were also incompletely suppressed in DM2 patients at 30 mg of aspirin, but 100 mg resulted in comparable platelet suppression in all groups, with no additional effect of 300 mg.

Conclusions

DM2 patients with inadequate glycaemic control (HbA1c>53mmol/mol) have higher baseline platelet activity and incomplete suppression of platelet activity with 30 mg of aspirin. However, 100 mg of aspirin leads to optimal inhibition irrespective of glycaemic control, and 300 mg does not further improve platelet suppression.

(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

Aspirin in type 2 diabetes

The effect of low-dose aspirin in primary prevention of cardiovascular disease in patients

with type 2 diabetes mellitus is uncertain1, 2_{. Clinical trials performed so far showed a}

non-significant 10% risk reduction3_{. This reduction seems modest considering the benefit}

of low-dose aspirin in secondary prevention, where it is able to prevent almost a quarter of re-events4_.

Although the reason for the lack of clinical benefit in patients with type 2 diabetes remains unknown, several laboratory studies have demonstrated a diminished anti-platelet effect of aspirin in these patients5-7_{. The anti-platelet effect of aspirin is caused}

by the irreversible acetylation of cyclo-oxygenase (COX)-1 in platelets, which inhibits the conversion of arachidonic acid to thromboxane A2, itself a potent platelet activator. In patients with type 2 diabetes, a high residual formation of thromboxane A2, as measured by its more stable metabolite thromboxane B2 in urine, and high residual platelet

aggregation in response to arachidonic acid have been demonstrated5-7_{. This residual}

activity may be caused by a higher intrinsic platelet activity with worsening glycaemic

control8-10_{, but may also result from incomplete acetylation of COX-1 due to competitive}

glycation. The contribution of hyperglycaemia to a diminished platelet response to aspirin is underlined by similar findings of high on-aspirin platelet activity in patients with type 1 diabetes, who lack other possible confounders such as a high body mass

index (BMI) or dyslipidemia11_{. Thus, the inhibition of platelets by aspirin may be more}

impaired in patients with poor glycaemic control compared to well controlled patients with type 2 diabetes. Moreover, when a reduced response to aspirin is induced by the

competitive glycation of COX-1, a higher dose of aspirin might overcome this problem12_.

We therefore investigated both the effects of glycaemic control and increasing doses of aspirin on platelet inhibition by aspirin in patients with type 2 diabetes, all compared to matched controls.

Research Design and Methods

Study population

We recruited patients with type 2 diabetes and different prespecified levels of glycaemic control: well controlled patients with HbA1c ≤ 53 mmol/mol (7.0%), moderately controlled patients with HbA1c 53-69 mmol/mol (7.0-8.5%) and poorly controlled patients with HbA1c ≥ 69 mmol/mol (8.5%). We also studied age- and sex-matched healthy controls. Eligible subjects were 18 years or older and not using aspirin or any other medication interfering with platelet function (e.g. non-steroidal anti-inflammatory drugs) in the two weeks prior to or during the study. Other exclusion criteria were abnormal platelet count

(5)

disorders, or severe liver or kidney failure. All subjects gave written informed consent and the study was approved by the ethics committee of the Academic Medical Centre. The trial was registered in the Dutch Trial Register (NTR 1273).

Study design

This was a prospective, single centre, open-label clinical trial investigating the effects of three doses of aspirin on COX-1 activity and platelet function. All subjects underwent three treatment periods, sequentially using 30 mg, 100 mg and 300 mg of aspirin for ten days. The dose of 30 mg was chosen since this is the lowest effective dose in prevention

of cerebrovascular events13_{. This low dose would also increase the sensitivity of detecting}

an effect of glycaemic control on the platelet inhibitory effect of aspirin. 100 mg of aspirin is the standard dose in current practice and 300 mg was chosen as the highest dose to increase the platelet inhibitory effect without increasing the risk of bleeding.

Laboratory measurements were performed at baseline and following each treatment period. After an overnight fast, blood was collected between 08.00 and 10.00 am. All subjects were instructed to take their last aspirin dose on the morning of the blood draw. In order to capture a full profile of the effects of aspirin on platelet activity we assessed COX-1 inhibition by measuring thromboxane levels in urine and measured its functional effects by determining platelet aggregation in whole blood using the Verify Now system (aspirin cartridge) and in platelet rich plasma by light transmittance aggregation (LTA). Since platelet function tests show considerable variation, as demonstrated in healthy

subjects using aspirin14_{, we repeated all measurements at day 9 and 10 during the 100 mg}

treatment phase, to determine the reproducibility of the platelet tests. The outcome of all platelet tests following the use of 100 mg of aspirin was calculated as the mean of these two measurements.

Outcome measures

The primary outcome was the difference of the effects of aspirin on thromboxane formation as measured in urine between the study groups. Secondary outcomes were the differences in the effects on platelet function as measured by VerifyNow and light transmittance aggregometry.

Urinary 11-dehydro thromboxane B2 (11-dhTxB2) is the most abundant, and stable,

urinary metabolite of thromboxane A215_{. In patients with type 2 diabetes it has been}

shown that approximately 80% of urinary 11-dhTxB2 is platelet derived9_{. To determine}

urinary 11-dhTxB2 concentrations, first morning urine samples were centrifuged and the supernatant was stored at -80ºC until batch analysis. After thawing and purification according to the manufacturer’s protocol using mixed bed SPE cartridges, 11-dhTxB2 was measured in duplicate using the 11-dhTxB2 EIA kit (Cayman Chemical, Ann Arbor,

(6)

MI). Multiple dilutions were used depending on the expected results. Results were normalized to urinary creatinin and expressed as ng 11-dhTxB2/mmol creatinin. Platelet aggregation in whole blood was determined by the point-of care VerifyNow system (AccuMetrics Inc.; San Diego CA), which uses arachidonic acid (AA) as an aspirin specific platelet agonist to determine the ability of platelets to aggregate. The VerifyNow test was performed according to the manufacturer’s protocol and results are expressed as aspirin reaction units (ARU).

Light transmission aggregometry (LTA) was performed in platelet rich plasma on a PAP-8E aggregometer (Bio/Data Corporation; Horsham, USA). Platelet rich plasma was prepared by centrifugation of citrate-anticoagulated whole blood (0.105 M citrate) for 10 minutes at 185 x g without brake. Agonists were 1, 2, 5 and 10 µmol/L adenosine diphosphate (ADP, Bio/Data corporation), 1 and 2 mmol/L AA (Bio/Data corporation) and 15 µmol/L thrombin receptor activating peptide 6 (TRAP 6, Bachem; Bubendorf, Switzerland).

Other biochemical parameters

HbA1c was measured by ion exchange chromatography on a Tosoh-G8 analyzer (Tosoh Bioscience; Tokyo, Japan). Total cholesterol, HDL-cholesterol, triglycerides and high sensitive C-reactive protein (hsCRP) were measured by standard enzymatic methods (Roche Diagnostics; Rotkreuz, Switzerland). LDL-cholesterol was calculated using the Friedewald formula. Haematology parameters were measured on a Sysmex XE-5000 (Sysmex; Kobe Japan); urinary micro-albumin was analyzed using a Modular P-800 system (Roche Diagnostics). For further analysis, plasma aliquots were snap-frozen and stored at -80°C.

Statistical analysis

The study was powered to detect a difference in urinary 11-dhTxB2 excretion between the patients with the worst glycaemic control and those with the best glycaemic control during the use of the lowest dose of aspirin. The sample size calculation indicated that 35 patients per group are needed in order to detect a 30% difference in the effect of 30 mg of aspirin on urinary 11-dhTxB2 levels between the well controlled and poorly controlled patients with 80% power and an alpha level of 0.05.

All data are presented as mean ± SD or median (interquartile range) depending on the distribution of data. Differences in baseline characteristics and aspirin effects between the study groups were tested by one way ANOVA or the Kruskall Wallis test, where appropriate, and differences in proportions were evaluated by Chi Square test. The difference between two individual groups was then tested by an unpaired student’s t-test or Mann Whitney U test. When a difference in outcome measures between groups

(7)

was detected, multivariate analysis was performed to determine the main predictors of the outcome. All baseline characteristics listed in Table 1 were first tested in univariate analysis for covariance with the outcome. Parameters which showed a borderline significant (p<0.10) effect in univariate analyses were then selected as covariates in a multivariate regression model. When the outcome data were skewed, square root transformation was performed prior to regression analysis. Only the covariates which remained significant in multivariate analysis are reported. Within group effects of different doses of aspirin were tested by Wilcoxon Signed Rank test. All correlations were determined by Spearman rank’s correlation.

To determine the reproducibility of our measurements, the outcomes at day nine and ten during the 100 mg treatment phase were correlated to each other by an intraclass correlation coefficient, using a two way random model to determine absolute agreement. In this model, 1 indicates perfect agreement and 0 no agreement between the repeated measurements. All analyses were performed using PASW statistics software version 18.0, a p value of <0.05 was considered statistically significant.

Results

Study population

In total, 121 subjects were included in the study. We analyzed 94 patients with type 2 diabetes and 25 healthy controls. Two subjects were excluded from the final analysis. One patient with type 2 diabetes was excluded because of increased haemolysis due to a myelodysplastic syndrome which did not allow us to classify the patient according to the HbA1c value. One healthy volunteer was excluded from the analyses because of a mild biochemical platelet dysfunction at baseline. In total, eight subjects did not complete the full trial. In four cases, this was related to the study drug, three subjects were having gastric complaints and one subject experienced a mild allergic reaction to the study drug. One healthy subject withdrew consent due to personal circumstances and three diabetes patients were admitted to the hospital during the study, for reasons unrelated to the study drug. No major bleeding events occurred during the trial.

Baseline characteristics are summarized in Table 1. Overall, age and sex were comparable between the groups. BMI and duration of diabetes differed significantly between the diabetes cohorts, showing increased BMI and diabetes duration with higher HbA1c. The majority of patients with diabetes (83%) used metformin, with no differences between groups. There was, however, a difference in the use of sulphonylurea derivatives and insulin between groups. Sulphonylurea derivatives were regularly used in the lower two diabetes cohorts (HbA1c < 69 mmol/mol), whereas insulin use was more frequent in the

(8)

higher two diabetes cohorts (HbA1c > 53 mmol/mol). Use of statins and antihypertensives was comparable between all diabetes cohorts. Low density lipoprotein (LDL) level was higher and triglyceride levels were on average lower in controls compared to the diabetes patients. HsCRP levels were higher in patients with type 2 diabetes as compared to the controls. Microalbuminuria was more frequently present in patients with poor glycaemic control as compared to other groups. Overall, HbA1c decreased 0.1% during the study.

Table 1. Baseline characteristics of patients with type 2 diabetes mellitus and controls

Controls DM P-value

HbA1c ≤ 53 mmol/_mol 53-69 mmol/_mol ≥ 69 mmol/_mol

N 25 34 35 25 n.a.

Clinical Characteristics

Age (yr) 53.6 (11.6) 59.2 (10.3) 56.0 (12.5) 54.0 (8.9) n.s.

Female sex (%) 44 44 51 44 n.s.

Body Mass Index (m2_/kg) _{25.4 (5.5)} _{28.7 (4.3)} _{31.8 (6.9)} _{32.0 (7.7)} _p<0.001

Duration of diabetes (yr) n.a. 6.7 (4.8) 8.3 (8.0) 10.9 (5.4) p=0.05 SBP (mmHg) 135 (25) 134 (18) 131 (18) 135 (22) n.s. DPB (mmHg) 83 (12) 84 (11) 82 (9) 85 (11) n.s. Medication use Metformin (%) 0 88 74 88 n.s. Sulphonylurea (%) 0 24 23 4 n.s. Insulin (%) 0 38 74 88 p<0.001 Antihypertensives (%) 8 53 63 60 n.s. Statins (%) 0 71 77 64 n.s. Laboratory Characteristics HbA1c (mmol/mol) 39 (36-41) 48 (45-51) 60 (57-63) 84 (76-101) P<0.001 TC (mmol/l) 5.0 (0.9) 4.3 (1.0) 4.3 (0.9) 4.6 (1.5) P=0.04 LDL (mmol/l) 3.1 (0.8) 2.2 (0.9) 2.4 (0.8) 2.6 (1.4) P=0.007 HDL (mmol/l) 1.5 (0.3) 1.3 (0.4) 1.2 (0.4) 1.1 (0.3) P=0.001 Triglycerides (mmol/l) 0.9 (0.6-1.3) 1.5 (0.8-2.1) 1.5 (0.9-2.5) 1.5 (0.9-3.0) P=0.01 CRP (mg/l) 1.0 (0.6-2.3) 1.2 (1.0-3.7) 3.7 (1.1-6.2) 2.2 (1.0-7.1) P=0.02 Hb (mmol/l) 8.8 (0.8) 8.6 (0.7) 8.3 (0.8) 8.6 (1.0) n.s. Thrombocytes (109_/l) _{242 (59)} _{242 (65)} _{243 (55)} _{256 (67)} _n.s. MPV (fl) 10.1 (0.7) 10.5 (1.0) 10.3 (0.9) 10.5 (0.8) n.s. Leucocytes (109_/l) _{5.4 (1.5)} _{6.3 (2.1)} _{6.3 (1.5)} _{6.7 (2.2)} _n.s. Microalbuminuria (%) 4 18 37 48 P=0.001

Data expressed as mean ± SD or as median (interquartile range). SBP indicates systolic blood pressure; DBP, diastolic blood pressure; HbA1c, hemoglobin-A1c; TC, total cholesterol; LDL, low-density cholesterol; HDL, High-low-density cholesterol; CRP, C-reactive protein; Hb, hemoglobin and MPV mean platelet volume.

(9)

Urinary 11-dehydro thromboxane B2 excretion

At baseline, median urinary 11-dhTxB2 excretion was 77 ng/mmol creatinin (IQR 61-109) in the poorly controlled patients, 84 ng/mmol creatinin (IQR 48-102) in the moderately controlled, 64 ng/mmol creatinin (IQR 34-93) in the well controlled patients with diabetes and 53 ng/mmol creatinin (IQR 37-73) in the controls (figure 1). Urinary excretion was significantly lower in the control group when compared to the moderately or poorly controlled patients with diabetes (p=0.015 and p=0.001, respectively), but also in the well controlled compared to the poorly controlled patients with diabetes (p=0.022). In multivariate analysis HbA1c was the strongest predictor of the difference in urinary 11-dhTxb2 excretion between the groups (beta 2.299, p=0.024), although BMI and hsCRP also influenced this association.

With 30 mg of aspirin, residual 11-dhTxB2 excretion remained higher in the patients with type 2 diabetes, displaying a stepwise increase of the 11-dhTxB2 concentration with increasing HbA1c levels (p for trend <0.001), despite larger decreases in the DM groups. Again, in multivariate analysis HbA1c was the strongest predictor of residual urinary 11-dhTxB2 excretion at 30 mg of aspirin. Treatment with aspirin 100 mg further reduced excretion of 11-dhTxB2 in all groups (figure 1). The difference in excretion between the groups were smaller compared to the 30 mg dose, but persisted (p=0.04), based on a significant difference between the poorly controlled patients with diabetes and both the control group and the well controlled patients (p=0.02 and p=0.028, respectively). Increasing the aspirin dose to 300 mg reduced urinary 11-dhTxB2 levels by 1 ng/mmol creatinin in the control group, whereas this decrease was non-significant in the diabetes cohorts.

Verify Now Assay

After treatment with 30 mg of aspirin, platelet aggregation in whole blood as measured by Verify Now was reduced from a median of 644 ARU (IQR 635-653) to 567 ARU (492-595) in the poorly controlled patients, from 654 ARU (646-659) to 511 ARU (449-582) in the moderately controlled patients, from 651 ARU (642-658) to 547 ARU(483-585) in the well controlled patients and from 648 ARU (643-656) to 460 ARU (426-533) in controls (figure 2). There was a significantly higher residual activity of platelets in the diabetes groups compared to the controls (p=0.003). BMI (beta 0.345, p=0.001) was the strongest predictor of the Verify Now outcome after 30 mg of aspirin in multivariate analysis, together with insulin use (beta 0.244, p=0.044).

Increasing the daily dose of aspirin to 100 mg resulted in a further inhibition of whole blood platelet aggregation in all groups (p<0.001 for all groups), with no significant difference between groups. A further increase to 300 mg of aspirin per day reduced ARU by 18 points in the well controlled patients with diabetes (p=0.022), but did not significantly affect the other groups (figure 2).

(10)

Figure 1. Urinary 11-dehydrothromboxane B2 excretion

Data are expressed as median (interquartile range).

Figure 2. Verify Now

Data are expressed as median (interquartile range).

                                                                     

(11)

Light Transmission Aggregometry (LTA)

Platelet aggregation induced by 1 mmol/L arachidonic acid (AA) was completely suppressed by all doses of aspirin in all groups (figure 3a). However, when a higher concentration of 2 mmol/L AA was used to trigger platelet aggregation, 30 mg of aspirin resulted in an incomplete inhibition of aggregation in the poorly controlled patients with type 2 diabetes, whereas aggregation was adequately suppressed in the other groups (figure 3b). Again, both BMI and insulin use were the main predictors of residual AA aggregation in multivariate analysis. Inhibition of platelet aggregation by 2 mmol/L AA was comparable in all groups when the dose of aspirin was increased to 100 and 300 mg per day (figure 3b).

ADP induced aggregation gave similar results between the groups at all concentrations tested, with no significant differences after aspirin dose increments (data not shown). Aggregation induced by TRAP was lowered to 96% (IQR 89-106) in the poorly controlled patients, 91% (IQR 83-99) in the moderately controlled patients, 88% (IQR 78-94) in the well controlled patients and 86% in the controls after the use of 30 mg of aspirin (p=0.015). Increasing the aspirin dose to 100 mg daily resulted in similar TRAP induced aggregation in all groups, namely 84% (IQR 76-93) in the poorly controlled, 84% (IQR 75-90) in the moderately controlled, 83% (IQR 77-91) in the well controlled patients and 87% (IQR 70-93) in the controls. Increasing the dose to 300 mg did not result in any significant changes compared to the 100 mg dose.

Correlation between glycaemic control and outcome measures

Urinary 11-dhTxB2 excretion was positively correlated with HbA1c at all aspirin doses (p<0.01). At baseline, none of the in vitro platelet aggregation tests showed a correlation with HbA1c. All but one (5 µmol/L ADP) of the platelet tests correlated positively with HbA1c during 30 mg of aspirin. These correlations were no longer present when the daily dose of aspirin was increased to 100 mg or 300 mg.

Correlation between repeated measurements during 100 mg treatment period

The intraclass correlation coefficients were moderate (0.198-0.575) but significantly associated for all outcome measures (Table 2).

(12)

131



















                  

















                  



Figure 3. AA aggregation

Optical aggregation induced by 1 (a.) and 2 (b.) mmol/L arachidonic acid. Data are expressed as median (interquartile range).

a.

(13)

Table 2. Correlation between repeated measurements during 100 mg of aspirin

ICC 95% CI Urinary 11-dhTxB2 excretion 0.575 0.430 - 0.691 Verify Now 0.512 0.352 - 0.643 Maximal aggregation ADP 1 µmol/L 0.365 0.175 - 0.528 ADP 2 µmol/L 0.467 0.293 - 0.612 ADP 5 µmol/L 0.198 -0.007 - 0.386 ADP 10 µmol/L 0.251 0.054 - 0.429 AA 1 mmol/L 0.209 0.012 - 0.392 AA 2 mmol/L 0.238 0.044 - 0.416 TRAP 15 µmol/L 0.401 0.212 - 0.561

ICC indicates intraclass correlation coefficient (95% confidence interval). ADP indicates adenosine diphosphate, AA arachidonic acid and TRAP thrombin receptor activating peptide.

Discussion

Our study shows that 30 mg of aspirin daily insufficiently suppresses platelet activation in patients with type 2 diabetes compared to matched controls. This effect was most pronounced in the patients with poor glycaemic control. Also, based on a broad panel of platelet tests, we conclude that 100 mg of aspirin adequately suppresses platelet activation in patients with type 2 diabetes, irrespective of their glycaemic control and comparable to the effect in healthy subjects. Finally, further increasing the aspirin dose to 300 mg offers no additional benefit for platelet suppression, while clinical trials have

shown an increase in bleeding events from 3.7% to 9.8%16_.

Our baseline results are in line with earlier studies, which have shown an association between glycaemic control and urinary 11-dhTxB2 excretion. In addition, improving

glycaemic control has been shown to lead to a decreased 11-dhTxB2 excretion8-10_{. The}

baseline difference may contribute to the difference in residual 11-dhTxB2 excretion we found during the use of 30 mg of aspirin in the patients with diabetes at different levels of glycaemic control. It confirms earlier findings in healthy volunteers, where pre-existent platelet hyper-reactivity was found to correlate with hypo-responsiveness to

aspirin17_{. The functional relevance of this residual thromboxane formation is supported}

by the aggregation assays, which likewise showed incomplete suppression at 30 mg. However, in multivariate analysis the platelet aggregation tests are mainly influenced by BMI and insulin use, instead of glycaemic control. Nonetheless, it is difficult to disentangle these clinical characteristics as they are often closely linked in patients with poor metabolic control. Interestingly though, our data suggest that their interference with aspirin effectiveness can already be overcome by increasing the dose to 100 mg.

(14)

Dichiara et al. demonstrated similar findings in a subset of 30 patients with diabetes from the secondary prevention population of the ASPECT trial. They tested 81, 162 and 325 mg and also found no significant differences in platelet activity between the two highest

doses5_{. A cross-sectional study also found an association between aspirin responsiveness,}

glycaemic control and aspirin dose in a group of patients with type 2 diabetes on chronic

aspirin treatment18_{. The authors suggest that higher doses of aspirin may be beneficial}

in patients with type 2 diabetes, but our prospective data show that increasing the dose from 100 to 300 mg does not improve the platelet response to aspirin in these patients and strongly suggests that there will be no additional cardiovascular protection from 300 mg of aspirin.

Although correlating platelet tests to clinical outcome is difficult, Eikelboom and colleagues showed that when using aspirin, all urinary 11-dhTxB2 levels higher than the upper limit of the lowest quartile (15.1 ng/mmol creatinin) were associated with an

increased risk of recurrent myocardial infarction, stroke or cardiovascular death19_{. Only}

our control group had lower levels while using 100 and 300 mg. Our diabetes groups had a residual urinary excretion of 11-dhTxB2 on 30 mg aspirin (24-31 ng/mmol) that would categorize them in the third quartile, corresponding with a non-significant odds ratio (OR) of 1.4 (95% CI, 0.9-2.2), and in the second quartile while on 100 and 300 mg of aspirin (16-23 ng/mmol), OR 1.3 (95% CI, 0.9-2.0). Notably, there was a correlation between residual 11-dhTxB2 excretion and glycaemic control at all aspirin doses in our study, also during the 100 and 300 mg dose. This residual thromboxane formation may not be platelet derived. Instead, the residual urinary 11-dhTxB2 may originate from the vascular endothelium or activated leucocytes and further suppression by high doses of aspirin is unlikely to result into substantial clinical benefit.

In line, when using the cut-off values for VerifyNow (454 ARU) and AA aggregation

(20%) suggested by Breet et al20_{, the 30 mg dose in our study failed to reach this target,}

while the 100 and 300 mg doses were sufficient. However, the positive predictive values

for all platelet tests in their study were low (ranging from 8.1-13.3%)20_{, which once more}

indicates that using laboratory tests to identify individuals at risk is difficult. Moreover, the studies correlating platelet tests to clinical outcome were all based on secondary prevention populations, making it difficult to translate their findings to our primary

prevention population21, 22_{. However, a recent dose finding clinical trial was in agreement}

with our laboratory findings, showing no additional benefit from 300-325 mg over 75-100

mg of aspirin, albeit in a high risk, secondary prevention population23_.

Several strengths and limitations of our study require further comment. First, the strength of our study is that increasing doses of aspirin were given to the same patient, which allows a direct intra-individual comparison of the effects of aspirin. Second, we tested both in vivo COX-1 inhibition by aspirin and performed extensive platelet function

(15)

testing using the most reliable tests available, including a baseline measurement as well as an additional second assessment during the 100 mg treatment. Third, we specifically included patients with different levels of glycaemic control, which allowed us to evaluate the effects of hyperglycaemia on platelet activity.

There are also some limitations. First, our study could not directly address the effect of aspirin on cardiovascular events. Furthermore, it is well known that measurement of

platelet activation in vitro is prone to variation24_{. For instance, Santilli and colleagues}

observed an extensive variation in the inhibition of several platelet function tests

in healthy subjects, who were treated with 100 mg of aspirin daily for eight weeks14_.

Therefore, we included an internal validation of our laboratory measurement of platelet activity in this study, which revealed far from perfect intraclass correlation coefficients, indicating that these tests may classify the platelet response differently on separate test days. Consequently, to avoid misclassification based on outliers, we did not try to identify “responders” or “non-responders”, but instead only compared the group response to aspirin between DM patients and healthy controls. Finally, due to our study design of giving aspirin in ascending doses, carry-over effects of the different dosing regimens may have occurred. However, since all our tests were platelet dependent and for every dose patients were treated for ten days, reflecting the lifespan of a blood platelet, it can be assumed that these effects should be minimal if present at all. Nonetheless, we cannot exclude carry-over effects of aspirin deposition in the bone marrow.

In conclusion, 30 mg of aspirin results in an inadequate platelet inhibition in patients with type 2 diabetes, especially in those patients who have poor glycaemic control. However, residual platelet activity is suppressed by 100 mg of aspirin in all patients with type 2 diabetes. Increasing the dose to 300 mg seems to offer little additional effect on platelet inhibition.

Acknowledgements

This study was funded by an unrestricted Netherlands Foundation for Cardiovascular Excellence grant (2008) to B.L.

We would like to thank Marianne C. Schaap from the department of Clinical Chemistry in the Academic Medical Centre for her excellent laboratory assistance and Dorien G. Dragt, diabetes nurse in the department of Internal medicine in the Academic Medical Centre, for her clinical assistance.

(16)

References

1. Farkouh ME, Fuster V. Diabetes and aspirin: beware of underpowered negative trials. Nat Clin Pract Cardiovasc Med 2009;6(1):1.

2. Pignone M, Alberts MJ, Colwell JA et al. Aspirin for primary prevention of cardiovascular events in people with diabetes: a position statement of the American Diabetes Association, a scientific statement of the American Heart Association, and an expert consensus document of the American College of Cardiology Foundation. Circulation 2010;121(24):2694-2701.

3. DeBerardis G., Sacco M, Strippoli GF et al. Aspirin for primary prevention of cardiovascular events in people with diabetes: meta-analysis of randomised controlled trials. BMJ 2009;339:b4531. 4. Collaboration AT. Collaborative meta-analysis of randomised trials of antiplatelet therapy for

prevention of death, myocardial infarction, and stroke in high risk patients. BMJ 2002;324(7329):71-86.

5. DiChiara J, Bliden KP, Tantry US et al. The effect of aspirin dosing on platelet function in diabetic and nondiabetic patients: an analysis from the aspirin-induced platelet effect (ASPECT) study. Diabetes 2007;56(12):3014-3019.

6. Fateh-Moghadam S, Plockinger U, Cabeza N et al. Prevalence of aspirin resistance in patients with type 2 diabetes. Acta Diabetol 2005;42(2):99-103.

7. Watala C, Golanski J, Pluta J et al. Reduced sensitivity of platelets from type 2 diabetic patients to acetylsalicylic acid (aspirin)-its relation to metabolic control. Thromb Res 2004;113(2):101-113. 8. Davi G, Averna M, Catalano I et al. Platelet function in patients with type 2 diabetes mellitus: the

effect of glycaemic control. Diabetes Res 1989;10(1):7-12.

9. Davi G, Catalano I, Averna M et al. Thromboxane biosynthesis and platelet function in type II diabetes mellitus. N Engl J Med 1990;322(25):1769-1774.

10. Santilli F, Formoso G, Sbraccia P et al. Postprandial hyperglycemia is a determinant of platelet activation in early type 2 diabetes mellitus. J Thromb Haemost 2010;8(4):828-837.

11. Mehta SS, Silver RJ, Aaronson A, Abrahamson M, Goldfine AB. Comparison of aspirin resistance in type 1 versus type 2 diabetes mellitus. Am J Cardiol 2006;97(4):567-570.

12. Gurbel PA, Bliden KP, DiChiara J et al. Evaluation of dose-related effects of aspirin on platelet function: results from the Aspirin-Induced Platelet Effect (ASPECT) study. Circulation 2007;115(25):3156-3164.

13. A comparison of two doses of aspirin (30 mg vs. 283 mg a day) in patients after a transient ischemic attack or minor ischemic stroke. The Dutch TIA Trial Study Group. N Engl J Med 1991;325(18):1261-1266.

14. Santilli F, Rocca B, De CR et al. Platelet cyclooxygenase inhibition by low-dose aspirin is not reflected consistently by platelet function assays: implications for aspirin “resistance”. J Am Coll Cardiol 2009;53(8):667-677.

15. Catella F, Healy D, Lawson JA, FitzGerald GA. 11-Dehydrothromboxane B2: a quantitative index of thromboxane A2 formation in the human circulation. Proc Natl Acad Sci U S A 1986;83(16):5861-5865.

16. Serebruany VL, Steinhubl SR, Berger PB et al. Analysis of risk of bleeding complications after different doses of aspirin in 192,036 patients enrolled in 31 randomized controlled trials. Am J Cardiol 2005;95(10):1218-1222.

(17)

17. Frelinger AL, Li Y, Linden MD et al. Aspirin ‘resistance’: role of pre-existent platelet reactivity and correlation between tests. J Thromb Haemost 2008;6(12):2035-2044.

18. Ertugrul DT, Tutal E, Yildiz M et al. Aspirin Resistance Is Associated with Glycemic Control, the Dose of Aspirin, and Obesity in Type 2 Diabetes Mellitus. J Clin Endocrinol Metab 2010;95(6):2897-2901.

19. Eikelboom JW, Hirsh J, Weitz JI, Johnston M, Yi Q, Yusuf S. Aspirin-resistant thromboxane biosynthesis and the risk of myocardial infarction, stroke, or cardiovascular death in patients at high risk for cardiovascular events. Circulation 2002;105(14):1650-1655.

20. Breet NJ, van Werkum JW, Bouman HJ, Kelder JC, Ten Berg JM, Hackeng CM. High on-aspirin platelet reactivity as measured with aggregation-based, cyclooxygenase-1 inhibition sensitive platelet function tests is associated with the occurrence of atherothrombotic events. J Thromb Haemost 2010;8(10):2140-2148.

21. Krasopoulos G, Brister SJ, Beattie WS, Buchanan MR. Aspirin “resistance” and risk of cardiovascular morbidity: systematic review and meta-analysis. BMJ 2008;336(7637):195-198. 22. Snoep JD, Hovens MM, Eikenboom JC, van der Bom JG, Huisman MV. Association of

laboratory-defined aspirin resistance with a higher risk of recurrent cardiovascular events: a systematic review and meta-analysis. Arch Intern Med 2007;167(15):1593-1599.

23. Mehta SR, Bassand JP, Chrolavicius S et al. Dose comparisons of clopidogrel and aspirin in acute coronary syndromes. N Engl J Med 2010;363(10):930-942.

24. Gorog DA, Sweeny JM, Fuster V. Antiplatelet drug ‘resistance’. Part 2: laboratory resistance to antiplatelet drugs-fact or artifact? Nat Rev Cardiol 2009;6(5):365-373.