Association between renal function and platelet reactivity during aspirin therapy in elderly patients with atherosclerotic cardiovascular disease

Background Aspirin is the key treatment in the secondary prevention of atherosclerotic cardiovascular disease. High on-treatment platelet reactivity (HTPR) to aspirin has been reported to partially account for the enhanced risk of thrombotic events. In particular, HTPR has been described more frequently among elderly patients. The aim of this study was to identify the clinical and biological factors associated with HTPR in a real-life elderly population. Methods In this retrospective study, elderly patients with atherosclerotic cardiovascular disease on regular aspirin treatment were enrolled. Cardiovascular risk factors, routine biological parameters, comorbidities, and concomitant medications were recorded. The upper quartile of the platelet aggregation rate, determined by light transmission aggregometry with arachidonic acid, was defined as the HTPR group. Results A total of 304 patients were included (mean age 77 ± 8 years, 76% men). Patients in the HTPR group were older than the patients in the non-HTPR group (mean age: 79 ± 7 vs. 76 ± 8 years, p = 0.008). Patients with moderately decreased estimated glomerular filtration rate (eGFR) had a higher frequency of HTPR than patients with slightly decreased eGFR or normal eGFR (35.8, 22.5, 12.2%, respectively, p < 0.05). In multivariate analysis, an independent risk factor for HTPR was the eGFR (OR: 0.984, 95% CI: 0.980–0.988, p < 0.001). Conclusions Advanced age and decreased eGFR are correlated with poor pharmacodynamic response to aspirin.

Recent studies have shown that aspirin fails to prevent a number of serious cardiovascular events among highrisk patients [6,7]. This has led to the introduction of the concept of high on-treatment platelet reactivity (HTPR) [8]. In particular, HTPR has been described more frequently among elderly patients, conditioned by impaired drug absorption and metabolism and by a baseline of more enhanced platelet reactivity. Most previous studies enrolled patients receiving dual antiplatelet therapy [9,10]. However, there is evidence of overlap in the antiplatelet effects of aspirin and P2Y12 antagonists [11]. The main objective of our study was to identify routinely available clinical and biological factors associated with HTPR in a real-life elderly population of ASCVD patients receiving aspirin as monotherapy. We hypothesized that finding factors associated with HTPR can help identify high-risk patients and eventually modify antiplatelet therapy.

Study design and participants
Elderly patients on regular aspirin treatment in the Department of Geriatrics of Peking University First Hospital were enrolled from March 2014 to December 2019.
The inclusion criteria were as follows: 1. Age: ≥ 60 years; 2. Diagnosis of ASCVD, presence of at least one of the following diseases: coronary heart disease, ischemic stroke, transient ischemic attacks, and peripheral arterial disease; 3. Aspirin monotherapy was used for secondary prevention (50-100 mg daily for more than 1 month) without a change in dose within the previous month. Compliance with aspirin was verified separately by the patient's attending physician and study physician.

Medical records collection
For each patient, detailed medical records were obtained from the electronic medical record system, including age, sex, smoking habit, alcohol habit, body mass index, comorbidities, concomitant medications, and usual biological parameters.

Platelet reactivity
For each patient, platelet aggregation was determined by light transmission aggregometry with arachidonic acid (LTA-AA). Platelet aggregation tests have been routinely performed for several years at our clinical laboratory by experienced technicians. Blood samples were collected within 21-22 h after aspirin intake and analyzed within 1 h from collection. Blood samples anticoagulated with 3.2% sodium citrate were centrifuged at 200×g for 10 min to obtain platelet-rich plasma and further centrifuged at 2000×g for 10 min to obtain platelet-poor plasma. Platelet aggregation was performed by the addition of 0.5 mg/mL arachidonic acid to platelet-rich plasma cuvette using platelet-poor plasma as a reference cuvette. The percentage of platelet aggregation was defined as the maximal light transmittance after arachidonic acid addition measured by the LBY-NJ4 platelet aggregometer (PRECIL, Beijing, China) [12].

Patient group
Patients were divided according to LTA-AA. Patients with platelet aggregation rates in quartile IV were defined as the HTPR group, whereas patients with platelet aggregation rates in quartiles I-III were defined as the non-HTPR group. Patients were also stratified by eGFR, which was calculated using the MDRD (Modification of Diet in Renal Disease) formula [13]. The G1 group included patients with normal eGFR (eGFR ≥90 mL/min/1.73 m 2 ). The G2 group included patients with slightly decreased eGFR (eGFR 60-89 mL/min/1.73 m 2 ). The G3 group included patients with moderately decreased eGFR (eGFR 30-59 mL/min/1.73 m 2 ).

Statistical analysis
The distribution normality of each variable was tested using the Kolmogorov-Smirnov test. Continuous variables were expressed as the mean ± SD or median (interquartile range). Homogeneity of variance was tested for continuous variables. For normally distributed data, comparisons between two groups were performed by Student's unpaired t test, and comparisons among groups were performed using one-way ANOVA, followed by post-hoc analysis of LSD tests (if variances were equal) or Tamhane's T2 tests (if variances were unequal). The nonparametric Mann-Whitney U test or Kruskal-Wallis H test were performed if the data were nonnormal. Categorical variables were described as counts (percentages) and were tested using the chisquare test. Correction for multiple comparisons was performed using the Bonferroni-Holm correction. Correlations between quantitative variables were assessed using Pearson or Spearman correlation coefficients, as appropriate. Multivariate logistic regression analysis (forward: conditional) was performed to investigate risk factors for HTPR. After univariate analysis, variables that presented significant associations with HTPR (p < 0. 10) were entered in the multivariate model. Differences were considered significant when p < 0.05 (two-tailed). The cutoff value was calculated by analyzing the receiver operating characteristic (ROC) curve. Statistical analysis was carried out using SPSS version 16.0 software (SPSS Inc., Chicago, IL, USA).

Results
Finally, 304 patients with sufficient available clinical and biological information were selected for this study. The average age was 77 ± 8 years, and 233 (76.6%) patients were males. There were 120 (39.5%) patients with age ≥ 80 years and 8 (2.6%) patients with age ≥ 90 years.

Clinical features of elderly patients with ASCVD classified by HTPR
The clinical and biological characteristics of the study participants divided according to HTPR are shown in Table 1 and Table 2. Patients in the HTPR group were older than patients in the non-HTPR group (mean age: 79 ± 7 vs. 76 ± 8 years, p = 0.008). In terms of the ASCV D spectrum, patients in the HTPR group had an increased presence of ischemic stroke or transient ischemic attacks, and fewer patients had a history of coronary heart disease than those in the non-HTPR group (p < 0.01). The ratio of angiotensin converting enzyme inhibitors/angiotensin receptor blockers (ACEIs/ARBs) medications was higher in the HTPR group (p = 0.046). There were no significant differences in daily aspirin Linear correlation analysis showed that LTA-AA was significantly correlated with age (r = 0.128, p = 0.025), hemoglobin (r = − 0.216, p < 0.001) and eGFR (r = − 0.230, p < 0.001) for the entire study population. Multivariate regression analysis of HTPR included the following independent variables in the model: age, systolic blood pressure, eGFR, hemoglobin, nitrate use, proton pump inhibitor (PPI) use and ACEI/ARB use. The results showed that eGFR was an independent factor associated with HTPR (OR: 0.984, 95% CI: 0.980-0.988, p < 0.001).

Clinical features of elderly patients with ASCVD classified by renal function
Study participants were stratified by eGFR. Clinical and biologic characteristics with statistical significance are shown in Table 3. In all enrolled patients, 16.1% had a normal eGFR, 52.6% had a slight decrease in eGFR, and 31.3% had a moderate decrease in eGFR. LTA-AA was higher in patients with decreased eGFR than in those with normal eGFR. Patients with moderately decreased eGFR had a higher frequency of HTPR than patients with slightly decreased eGFR or normal eGFR (35.8, 22.5, 12.2%, respectively, p < 0.05). Renal function deteriorates with age. As eGFR decreased, a descending trend of diastolic blood pressure levels and hemoglobin levels, as well as an ascending trend of serum creatinine, urea and uric acid levels appeared. Patients with moderately decreased eGFR had higher MPV (mean platelet volume)/PLT (platelet count) ratios and neutrophil percentages compared to those with slightly decreased eGFR (p < 0.05). Compared with patients with normal eGFR, patients with moderately decreased eGFR had an increased presence of ischemic stroke/transient ischemic attack, peripheral artery stenosis and hyperlipidemia (p < 0.05), and the ratios of ACEIs/ARBs, statins and PPI medications were higher (p < 0.05). The results of ROC analysis showed that eGFR levels in all patients were calculated with 70.3 mL/min/1.73m 2 as a cutoff value to predict HTPR, with an area under the ROC curve of 0.620 (95% CI 0.551-0.689, p = 0.002).

Discussion
Cardiovascular disease and thrombosis are very common complications in patients with renal dysfunction, possibly due to increased platelet activity. In the present study, we reported an association between HTPR and impaired renal function (using the surrogate of eGFR). Our study revealed that decreased eGFR was an independent risk factor of HTPR. HTPR was more common in elderly patients with mildly/moderately decreased eGFR than in those with preserved renal function. Therefore, we suggested that in elderly patients with impaired renal function undergoing aspirin therapy, platelet reactivity should be monitored more intensely to identify HTPR. In clinical practice, the personalized antithrombotic regimen, rather than routine use of aspirin, should be recommended in elderly patients with renal dysfunction. To date, few studies have been conducted regarding the role of renal function on platelet reactivity. Conflicting results were reached, and most of them were conducted among patients receiving clopidogrel in association with aspirin [14][15][16][17]. However, there is evidence of overlap in the antiplatelet effects of aspirin and P2Y12 antagonist [11]. Regarding patients receiving aspirin monotherapy, results from previous studies were also inconsistent. Using the same technology (LTA-AA) in 169 patients with coronary artery disease, Blann et al. found that patients with aspirin resistance had lower eGFR than patients who were sensitive to aspirin [18]. Their data pointed to an association between worsening renal function and aspirin resistance, which was in line with our study. However, Würtz et al. reported that renal function did not correlate with platelet aggregation evaluated by multiple electrode aggregometry or the VerifyNow [19]. These conflicting results may be due to differences in the test methods of platelet reactivity or the inclusion criteria. The underlying mechanism of platelet hyperactivity in patients with impaired renal function might be related to chronic low-grade inflammation, vascular injury [20] and the accumulation of uremic toxins, such as indoxyl sulfate [21] and homocysteine [22]. Furthermore, pre-activation of platelets and platelet turnover [23] also play roles in the pathogenesis of insufficient platelet inhibition by aspirin in patients with impaired renal function.
Besides the impact of renal function, the mechanisms of HTPR are probably multifactorial. Previous studies have specifically addressed the relation between age and HTPR [9,[24][25][26]. Bobescu et al. reported that low response to aspirin was significantly correlated with age older than 65 years [27]. In our study, all enrolled patients are over 60 years old. There was a positive correlation between age and platelet aggregation rate. Age was  [28]. At present, antiplatelet strategies in elderly patients with ASCVD should be cautiously driven by an individualized approach, balancing thrombotic and bleeding risk. Drug interactions have also been implicated as a potential mechanism of HTPR. PPIs are usually prescribed to patients receiving antiplatelet therapy to decrease the risk of upper gastrointestinal bleeding. Acid suppression with PPIs can increase the potential for mucosal esterases to hydrolyze aspirin to its inactive form. A reduction in gastric absorption results in an increased drug load within the small intestine, where hydrolysis by esterases prior to absorption may reduce bioavailability [29]. This could explain the finding in our study that PPI medication was more common in the HTPR group. Nevertheless, the current guidelines on cardiovascular disease are in favor of PPI prescription in those who are at risk of gastrointestinal bleeding, with the additional benefit that a reduction in dyspepsia may improve adherence [30,31]. It is worth noting that platelet reactivity should be monitored in these patients.

Limitations
There were several limitations in this study. First, the present data were observational, with all the inherent limitations of a retrospective analysis. We did not have baseline pretreatment data for platelet function as that would not have been feasible with this particular study design. Prospective or interventional investigations should be conducted in the future. Some clinical and biological parameters were intertwined and inevitable. Larger trials with subgroup analysis and matched-pair analysis are needed. Second, we only used LTA-AA tests as a surrogate measure of aspirin response, rather than specific assessment of its effect on the therapeutic target (i.e., thromboxane B2).

Conclusions
Among elderly patients receiving antiplatelet therapy with aspirin for secondary prevention of ASCVD, advanced age was correlated with insufficient antiplatelet effects of aspirin. A significant relationship between HTPR and impaired renal function was also observed. Larger trials are needed to assess the clinical impact of this finding and investigate the optimal antithrombotic regimen in elderly patients.