Factors Associated with Aspirin Resistance in Hong Kong Chinese Patients with Stable Coronary Heart Disease Using the Multiplate® Analyzer and Serum Thromboxane B2

Introduction: Aspirin resistance may be associated with various conditions. We measured serum thromboxane B2 (TXB2) and platelet function using the Multiplate® Analyzer with arachidonic acid (ASPI) in patients on long term aspirin therapy to identify aspirin resistance and associated factors. Materials and Methods: Chinese patients with stable coronary heart disease had samples for serum TXB2 and ASPI measurement taken before and 1 h after taking a morning dose of 80 mg aspirin. Results: In 266 patients with mean age 66.6 ± 10.7 years, 17% were female and 55% were current or previous smokers. TXB2 and ASPI measurements were significantly higher before the dose than at 1 h post dose, with 46% of subjects having high ASPI values (AUC > 300 AU*min) pre dose compared with 27% at 1 h post dose. TXB2 and ASPI measures of platelet aggregation showed weak correlations, which were only significant before the dose (r = 0.219, p = 0.001). Increased ASPI measurements were associated with white blood cell (WBC) count, haematocrit, platelet count and heart rate at 24 h post dose but only with WBC count, smoking history and heart rate at 1 h post dose. Diabetes was not associated with reduced platelet response to aspirin. The WBC count associated with aspirin resistance was over 6.55 × 109/L by receiver operating characteristic analysis. Conclusions: The antiplatelet response to aspirin was reduced in a large proportion of patients. Patients with higher WBC count within the normal range appear to be at increased risk of aspirin resistance. Higher or more frequent doses of aspirin may be needed in many patients.


Introduction
Aspirin resistance, also known as high on-aspirin residual platelet reactivity (RPR), can be defined as the inability of aspirin to adequately prevent the production of platelet cyclooxygenase-1 (COX-1)-derived thromboxane A 2 (TXA 2 ) and it can be detected by laboratory tests of platelet TXA 2 production by measurement of metabolites such as serum thromboxane B 2 (TXB 2 ) or platelet function testing that depends on platelet thromboxane production [1][2][3]. Laboratory aspirin resistance has been estimated to occur in 5-60% of patients on aspirin treatment, depending on the type of patients studied, different study designs, different tests of platelet function, and different definitions of aspirin resistance [1,[4][5][6][7][8].
Despite aspirin therapy, a considerable number of patients, particularly those with diabetes, continue to have atherothrombotic events, which is referred to as aspirin treatment failure [9,10]. Although the aetiology of aspirin treatment failure is multifaceted, it may be attributable in part to the high on-aspirin RPR or aspirin resistance [1,11,12]. TXB 2 is the major metabolite of platelet TXA 2 and it is derived only from platelet TXA 2 , and thus serum TXB 2 has been suggested to be the most specific test to measure the pharmacological effect of aspirin on platelets [13]. There was a moderate correlation between the serum TXB 2 assay and other assays measuring platelet function and inadequate aspirin response [14,15]. However, platelets may be activated through multiple pathways other than TXA 2 , and thus using this method to measure platelet response to aspirin may underestimate the prevalence of aspirin resistance or "aspirin non-responders".
Platelet function ex vivo has been measured in patients on aspirin treatment by several functional assays including light transmission aggregometry, whole blood aggregometry, Multiplate ® Analyzer, Platelet Function Analyzer-100 ® , VerifyNow Aspirin ® , urinary 11dehydro-TXB 2 and serum TXB 2 measurement, etc. However, these functional assays have variable specificity for measuring the effects of aspirin on platelet function and there is a poor correlation between results from different tests [16][17][18]. The Multiplate ® device performs multiple electrode impedance aggregometry in whole blood, the Platelet Function Analyzer-100 ® measures the time required for platelet aggregation to occlude an aperture, whereas the VerifyNow ® measures platelet aggregation in whole blood based on change in light transmission [19]. The Multiplate impedance platelet aggregometry induced by arachidonic acid (ASPI) measurement has been used in many studies and in participants taking aspirin the highest platelet response variations with arachidonic acid were found in blood samples anticoagulated with hirudin [20,21]. A randomized clinical trial in healthy subjects treated with aspirin or clopidogrel suggested that any one of these three pointof-care devices can be used for assessing the treatment effects of aspirin or clopidogrel on platelet aggregation at the study population level [19].
This study compared the ASPI measurement for the Multiplate ® device with the measurement of serum TXB 2 as the reference standard and examined the association of phenotypic factors on the antiplatelet effect of aspirin, and the prevalence of aspirin resistance in Hong Kong Chinese patients with stable coronary heart disease (CHD).

Study Population
Eligible patients aged ≥18 years with CHD receiving long-term mono-antiplatelet therapy for ≥3 months with plain, uncoated aspirin (80 mg once daily) were invited to participate in the study in the outpatient clinic at the Prince of Wales Hospital, Hong Kong. The study was performed in accordance with Good Clinical Practice (GCP) and the Principles of the Declaration of Helsinki. The study was approved by the Joint Chinese University of Hong Kong-New Territories East Cluster Clinical Research Ethics Committee with reference number CRE-2014.516-T. Written informed consent was obtained at the time of enrolment from all subjects. This study was reported with respect to the recommendation for reporting randomized clinical trials as the statement of Consolidated Standards of Reporting Randomized Clinical Trials (CONSORT) in this article.
Patients had previously been diagnosed with CHD based on the following criteria: ischaemic electrocardiogram (ECG), positive exercise test, symptoms of angina pectoris, history of acute coronary syndromes, coronary angioplasty, or coronary artery bypass surgery. Patients who had a myocardial infarction, stroke, coronary artery bypass surgery or other revascularization treatment, unstable angina or angioplasty during the previous 3 months were excluded from the study. Patients on ticlopidine, prasugrel, clopidogrel, dipyridamole or other antiplatelet or anticoagulant medications were not recruited. Patients on regular therapy with anti-inflammatory drugs or other drugs containing aspirin or nonsteroidal anti-inflammatory drugs (NSAIDs), or traditional Chinese medicine that have direct effects on the haemostasis were excluded from the study unless they were willing to discontinue these treatments for at least 2 weeks prior to study participation.
The patients were instructed to take aspirin in the morning for at least 7 days before the research visit, if they were not already doing so. Utilizing standard equipment and methodology, anthropometric measurements of body weight and height were obtained. The percentage of total body fat was measured using an impedance device (TANITA Body Composition Analyzer BF-350, TANITA Corporation, Tokyo, Japan). The sitting blood pressure and heart rate was taken using a semiautomatic sphygmomanometer (Critikon Dinamap 8100; GE Medical Systems Information Technologies, Louisville, KY, USA) after resting seated for 5 min.
During the study visit, blood samples were obtained in the morning before the aspirin dose and one hour after the dose to measure the trough and peak levels of the tests for the anti-platelet effects of aspirin. In the study centre, drug adherence was evaluated face-to-face via a personal interview. Demographic information, medical history and recent laboratory test results were collected.

Sample Collection
Participants arrived at the study centre after fasting for at least 10 h. Patients' fasting blood samples were taken immediately before and one hour after a single 80 mg aspirin dose. In a serum separator tube, a 5 mL blood sample was collected and allowed to clot at room temperature before centrifugation at 4 • C. The separated serum was stored at −80 • C in aliquots until TXB 2 analysis. A 3 mL blood sample was collected in a hirudin blood tube for the platelet aggregation test with the Multiplate ® analyzer, which was performed within 3 h of blood collection. During the study, a total of 16 mL blood was taken from each participant.

Multiplate ® Analyzer
The platelet activity of the samples was measured with the hirudin blood using the Multiplate ® Analyzer (Roche Diagnostics International Ltd., CH-6343 Rotkreuz, Switzerland) according to the manufacturer's instructions for the ASPI test. It was analysed within 0.5-3 h after blood collection. Platelet aggregation is measured continuously using electrical impedance for 5 min after addition of 20 µL of reconstituted ASPI reagent containing 0.5 mM arachidonic acid at final concentration to the test cells in channel one and channel two, which serves as a paired control. The impedance is converted to arbitrary aggregation units (AU) and plotted over time, and the area under the aggregation curve (AUC) was calculated as AU × minutes (AU*min).

Serum Thromboxane B 2 Assay
TXB 2 was detected in the serum of 253 patients using Cayman enzyme immunoassay kits (TXB 2 EIA Kit Item no. 501020, Cayman Chemical Company, Ann Arbor, MI, USA) with a detection range of 1.6-1000 pg/mL. All the calibration standards and buffers/reagents for the test were prepared in accordance with the kit's instructions. All samples were diluted in a 1:1 ratio with EIA buffer. The standards were assayed in duplicate, whereas the samples were assayed singly.

Other Assays
Other biochemical assays for glucose, renal and liver function and lipids were measured on a Roche Modular Analytics system (Roche Diagnostics GmbH, Mannheim, Germany) using standard reagent kits supplied by the manufacturer of the analyser. The analytical performance of these assays was within the manufacturer's specifications. Lowdensity lipoprotein cholesterol level was estimated by using the Friedewald formula.

Haematology Assays
Complete blood profile, including white blood cell (WBC) count, was measured using an automated cell counter (GEN-S; Beckman Coulter, Miami, FL, USA).

Definition of Aspirin Resistance with the ASPI Test
In this study, an ASPI AUC value >300 AU*min was used as the conventional cut-off to identify aspirin resistance in patients taking 80 mg aspirin daily [22].

Statistical Analysis
A sample size calculation based on previous similar studies estimated that obtaining results for both tests of anti-platelet activity in 250 patients would allow the study to have at least 80% power to detect a correlation coefficient of at least 0.2 between the two measurements with level of significance of 0.05 in patients with CHD. The study was not powered for validation of a clinical prediction model [23]. Correlations between the results obtained with the two assays, irrespective of aspirin resistance classification, were assessed by the Pearson's correlation coefficient. The strength of the correlation (r-value) was interpreted according to the following common definitions: 0.00 to 0.19 "very weak", 0.20 to 0.39 "weak", 0.40 to 0.59 "moderate", 0.60 to 0.79 "strong", and 0.80 to 1.0 "very strong" [24]. Receiver operating characteristic (ROC) analysis was performed comparing the two tests and factors which may influence them. The relationships between baseline demographic factors and aggregation on the ASPI AUC (AU*min) measurement before and after the aspirin dose were analysed by multiple stepwise linear regression entering platelet count, haematocrit (Hct), WBC count, fasting blood glucose (FBGL), body fat percentage (B fat%), systolic blood pressure (SBP), diastolic blood pressure, heart rate, waist hip ratio (WHR), body mass index (BMI), having DM and cigarette smoking into the equation. Simultaneously, normal distribution and collinearity of explanation variables were detected. Data were analysed using SPSS version 17.0 (SPSS Inc., Chicago, IL, USA).

Results
A total of 1899 potentially eligible patients' medical records were reviewed, and 471 were contacted, with 267 agreeing to participate in the study. In 255 patients, data for the Multiplate ® impedance platelet aggregometry aspirin (ASPI) assessments before and 1 h after the aspirin dose were available ( Figure 1). Only 251 patients had serum TXB 2 levels measured. Tables 1 and 2 presented clinical, demographic, and laboratory information for all patients as well as subjects with and without diabetes mellitus (DM).  The mean (± SD) age was 66.6 ± 10.7 years, the mean body mass index was 25.1 ± 3.5 kg/m 2 , 17% were female and 55% were current or previous smokers. There were no non-compliant patients based on pill counting and face-to-face interviews. The suppressed levels of TXB 2 (mean ± SD, 91.3 ± 90.7 pg/mL pre dose) further confirmed that all patients were compliant with aspirin in this study.
Compared with subjects without DM, those with DM were more likely to have hypertension, peripheral artery disease (PAD) and previous stroke, and to be treated with angiotensin converting enzyme inhibitors (ACEIs) or angiotensin receptor blockers (ARBs) (Tables 1 and 2). Patients with DM were older and had a greater waist circumference and WHR, greater SBP, fasting glucose, triglycerides and lower levels of high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), red blood cell count (RBC), haemoglobin (Hb) and Hct. Platelet count was not different between those with and without DM. The mean (± SD) age was 66.6 ± 10.7 years, the mean body mass index was 25.1 ± 3.5 kg/m 2 , 17% were female and 55% were current or previous smokers. There were no non-compliant patients based on pill counting and face-to-face interviews. The suppressed levels of TXB2 (mean ± SD, 91.3 ± 90.7 pg/mL pre dose) further confirmed that all patients were compliant with aspirin in this study.
Compared with subjects without DM, those with DM were more likely to have hypertension, peripheral artery disease (PAD) and previous stroke, and to be treated with angiotensin converting enzyme inhibitors (ACEIs) or angiotensin receptor blockers (ARBs) (Tables 1 and 2). Patients with DM were older and had a greater waist circumference and WHR, greater SBP, fasting glucose, triglycerides and lower levels of high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), red blood cell count (RBC), haemoglobin (Hb) and Hct. Platelet count was not different between those with and without DM. There were no significant differences in the serum TXB 2 levels and the ASPI measures between patients with and without DM ( Table 3). The serum TXB 2 levels and the ASPI AUC measures were both significantly higher in the samples pre dose, compared with the samples 1 h after the dose of aspirin ( Table 3). The mean values before and 1 h after aspirin for the ASPI test for all patients were 311.8 ± 167.5 AUC and 260.8 ± 122.1 AUC, respectively (p < 0.001). A similar pattern was observed in the DM and non-DM groups. There was no significant relationship between platelet aggregation and age, body mass index, body fat percentage and fasting glucose levels. There were weak correlations between the serum TXB 2 levels and the ASPI measure of platelet aggregation which were significant (p = 0.001) before but not after the dose (r = 0.219 and 0.118, respectively, Figures 2 and 3).
groups. There was no significant relationship between platelet aggregati mass index, body fat percentage and fasting glucose levels. Table 3. Platelet aggregation parameters of subjects before and one hour after ta of CHD patient with and without diabetes mellitus (DM).

All
Non-DM D Values are given as mean ± SD, *** p < 0.001 compared between pre dose by 0.0001 compared between pre dose by paired T-test.
There were weak correlations between the serum TXB2 levels and t of platelet aggregation which were significant (p = 0.001) before but not 0.219 and 0.118, respectively, Figures 2 and 3).  In total, there were 120 (46%) subjects with ASPI values >300 before the aspirin dose, including 75 (48%) from the non-DM group and 45 (41%) from the DM group. After the aspirin dose there were a total of 71 (27%) subjects with ASPI values >300, including 45 (29%) from the non-DM group and 26 (25%) from the DM group. There were 60 (22.6%) subjects with ASPI values >300 before the dose that became <300 after the dose of aspirin, including 34 (21.8%) from the non-DM group and 26 (23.9%) from the DM group, without any significant difference between the DM and the non-DM groups. There were 12 (4.5%) subjects with values <300 before the dose that became >300 after the dose, 4 (2.6%) from the non-DM group and 8 (7.3%) from the DM group, without any significant difference between groups. There were 10 subjects with missing values for one of the two ASPI AUC values, either pre dose or post dose.  In total, there were 120 (46%) subjects with ASPI values >300 before including 75 (48%) from the non-DM group and 45 (41%) from the DM aspirin dose there were a total of 71 (27%) subjects with ASPI values > (29%) from the non-DM group and 26 (25%) from the DM group. Ther subjects with ASPI values > 300 before the dose that became < 300 after rin, including 34 (21.8%) from the non-DM group and 26 (23.9%) from without any significant difference between the DM and the non-DM gr 12 (4.5%) subjects with values < 300 before the dose that became > 300 (2.6%) from the non-DM group and 8 (7.3%) from the DM group, witho difference between groups. There were 10 subjects with missing values ASPI AUC values, either pre dose or post dose.
The regression diagnostics shows that the model studentized resid  Table 4).   Table 4). On the ASPI measurement 1 h post dose, only the WBC count (B = 21.73, p < 0.0001), smoking history (B = 43.894, p = 0.002) and heart rate (B = 1.918, p = 0.026) were significantly associated ( Table 5). The regression diagnostics also shows that the model studentized  WBC count, Hct and platelet count were significantly associated with ASPI AUC values >300 prior to aspirin dosing based on receiver operating characteristic (ROC) analysis ASPI AUC values >300 pre dose with WBC count having the most significant relationship (Figure 4).
On the ASPI measurement 1 h post dose, only the WBC count (B = 21.73, p < 0.0001 smoking history (B = 43.894, p = 0.002) and heart rate (B = 1.918, p = 0.026) were signif cantly associated ( Table 5). The regression diagnostics also shows that the model stu dentized residuals of histogram followed a normal distribution (Std.Dev = 0.996) an none of the included variables had collinearity (VIF < 10). WBC count, Hct and platelet count were significantly associated with ASPI AUC values > 300 prior to aspirin dosing based on receiver operating characteristic (ROC analysis ASPI AUC values > 300 pre dose with WBC count having the most significan relationship (Figure 4). A cut-off of 6.55×10 9 /L for WBC count was most associated with aspirin resistanc (area under the ROC curve: 0.679, p < 0.0001, n = 255; sensitivity: 0.625; specificity: 0.669) The TXB2 level associated with aspirin resistance identified by ASPI AUC values 300 on ROC analysis was of borderline significance (p = 0.037) with low values for sens tivity and specificity (sensitivity: 0.456; specificity: 0.451) and a cut-off of 44.6 pg/mL fo serum TXB2 to indicate aspirin resistance ( Figure 5). A cut-off of 6.55 × 10 9 /L for WBC count was most associated with aspirin resistance (area under the ROC curve: 0.679, p < 0.0001, n = 255; sensitivity: 0.625; specificity: 0.669).
The TXB 2 level associated with aspirin resistance identified by ASPI AUC values >300 on ROC analysis was of borderline significance (p = 0.037) with low values for sensitivity and specificity (sensitivity: 0.456; specificity: 0.451) and a cut-off of 44.6 pg/mL for serum TXB 2 to indicate aspirin resistance ( Figure 5). A cut-off of 44.6 pg/mL for serum TXB2 was best associated with aspirin resistance with the area under the ROC curve of 0.488 (area under receiver operating characteristic 0.488, p = 0.037, n = 247; sensitivity: 0.456; specificity: 0.451).

Discussion
In this study, at 24 h post dose, 46% of patients taking aspirin 80 mg daily were found to have ASPI AUC values > 300 AU*min, which is the conventional cut-off to identify aspirin resistance [22,24]. The existence and significance of aspirin resistance has been debated but aspirin resistance determined in certain ways has been associated with an increased risk of major adverse CHD events in patients with stable CHD in severa studies [25][26][27].
The measurements of serum TXB2 showed a very weak correlation with the ASP AUC values and from the ROC analysis, it appears that serum TXB2 would not be usefu to identify patients with aspirin resistance identified by the ASPI AUC. In some previous studies, the measurements of serum TXB2 have been used to identify compliance with aspirin treatment rather than to identify aspirin resistance [11,28].
Having type 2 DM has been identified as a risk factor for aspirin resistance in severa previous studies [29]. In one large study of 900 patients with stable CHD, type 2 DM and BMI were independent determinants of increased arachidonic acid (AA)-induced platele aggregation with the Multiplate ® Analyzer using citrate and hirudin anticoagulated blood and also with the VerifyNow ® Aspirin Assay [30]. Other factors associated with reduced aspirin effect included platelet count, previous myocardial infarction, previous coronary artery bypass grafting, age, smoking and female gender [30]. We did not find a relationship between having type 2 DM and reduced aspirin response in the presen study. This could be due to a number of factors including the relatively good glycaemic control in the patients with DM, as shown by the fasting plasma glucose levels. Glyco sylated Hb (HbA1c) levels were not measured in this study visit but had been considered satisfactory at previous clinic visits. There was also frequent use of other protective

Discussion
In this study, at 24 h post dose, 46% of patients taking aspirin 80 mg daily were found to have ASPI AUC values > 300 AU*min, which is the conventional cut-off to identify aspirin resistance [22,24]. The existence and significance of aspirin resistance has been debated but aspirin resistance determined in certain ways has been associated with an increased risk of major adverse CHD events in patients with stable CHD in several studies [25][26][27].
The measurements of serum TXB 2 showed a very weak correlation with the ASPI AUC values and from the ROC analysis, it appears that serum TXB 2 would not be useful to identify patients with aspirin resistance identified by the ASPI AUC. In some previous studies, the measurements of serum TXB 2 have been used to identify compliance with aspirin treatment rather than to identify aspirin resistance [11,28].
Having type 2 DM has been identified as a risk factor for aspirin resistance in several previous studies [29]. In one large study of 900 patients with stable CHD, type 2 DM and BMI were independent determinants of increased arachidonic acid (AA)-induced platelet aggregation with the Multiplate ® Analyzer using citrate and hirudin anticoagulated blood and also with the VerifyNow ® Aspirin Assay [30]. Other factors associated with reduced aspirin effect included platelet count, previous myocardial infarction, previous coronary artery bypass grafting, age, smoking and female gender [30]. We did not find a relationship between having type 2 DM and reduced aspirin response in the present study. This could be due to a number of factors including the relatively good glycaemic control in the patients with DM, as shown by the fasting plasma glucose levels. Glycosylated Hb (HbA1c) levels were not measured in this study visit but had been considered satisfactory at previous clinic visits. There was also frequent use of other protective treatments, such as statins and ACEIs or ARBs, in the patients with DM.
The relationships between reduced platelet response to aspirin and WBC count, platelet count, and RBC or Hct have been described in several previous studies [31]. In Chinese patients with acute ischaemic stroke, aspirin resistance evaluated by the PFA-100 test was associated with WBC count (r = −0.125, p = 0.041) and LDL-C and predicted an unfavourable stroke outcome at 90 days [32]. In a study in Turkey, patients with stable angina with aspirin resistance had significantly higher WBC count, neutrophil counts, neutrophil-to-lymphocyte ratios, serum uric acid (SUA) levels, high-sensitivity C-reactive protein levels, and fasting blood glucose levels and high SUA levels were an independent predictor of aspirin resistance after multivariate analysis, with a cut-off value of 6.45 mg/dl for SUA to predict aspirin resistance with 79% sensitivity and 65% specificity on ROC analysis [33]. In another study, in patients with acute coronary syndromes, the anti-aggregant effect of aspirin assessed by PFA-100 closure times with ADP (PFA/ADP) and epinephrine (PFA/EPI) were modulated not only by the direct action on platelets, but also by erythrocyte deformability and WBC count [34].
We have demonstrated that WBC count and Hct level were associated with the response to aspirin treatment, which has been indicated by other studies showing the interaction of aggregation impedance and other cellular components [35,36]. We also found an association between aspirin resistance and heart rate. A previous study found that heart rate variability was decreased, and sympathetic activity increased in patients with aspirin resistance and stable CHD [37]. It is plausible that increased sympathetic activity associated with increased heart rate will result in platelet activation. Aspirin resistance has also been associated with hypertension [38,39] but we did not find any association with systolic and diastolic blood pressures in this study. As with diabetes, the treatments used for hypertension may interact with the effect of aspirin on platelets.
Increased platelet count, as in essential thrombocythaemia, has been associated with aspirin resistance and this may be explained by accelerated platelet turnover [40]. Increase platelet turnover and oxidative stress have been suggested to be mechanisms of aspirin resistance with DM in some studies [41]. The mechanism of the association of smoking with aspirin resistance may also be through oxidative stress. The mechanism of the association of higher WBC count and aspirin resistance may be related to aspirin insensitive thromboxane generation as monocytes may generate TXA 2 from increased COX-2 expression in inflammatory conditions [31,42]. The mechanism of the association of higher Hct and aspirin resistance remains unclear, but erythrocyte deformability may be relevant [34].
It would be interesting to follow-up the patients in this study to determine if any of the measurements made influence the cardiovascular outcome, but this was beyond the scope of the study. Further studies would be useful to determine whether increasing the dose or the dosing frequency of aspirin could improve the measure of platelet aggregation. A large clinical trial would be necessary to prove that such action would be beneficial and that would be very difficult to perform. This study has several limitations. We only used the ASPI test with the Multiplate ® Analyzer and using other agonists such as collage, thrombin or ADP may have given different results. Furthermore, ex vivo platelet function tests may not accurately reflect in vivo conditions. We did not measure the differential WBC count, which might have provided more useful information. The sample size of the study was relatively small and would not be sufficient to provide validation of a clinical prediction model. A recent review of this topic concluded that external validation of a clinical prediction model would require a sample size of several thousand participants [23].

Conclusions
In the Chinese patients with CHD who participated in this study, 46% showed evidence of aspirin resistance by the Multiplate ® Analyzer ASPI test 24 h after the last dose of aspirin and 27% showed aspirin resistance at 1 h post dose. The measurements of serum TXB 2 did