Journal of the Neurological Sciences 376 (2017) 211–215

Contents lists available at ScienceDirect

Journal of the Neurological Sciences

journal homepage: www.elsevier.com/locate/jns

Association between PTGS1 polymorphisms and functional outcomes in

Chinese patients with stroke during aspirin therapy: Interaction
with smoking

Huan Cai a,1, Biyang Cai b,1, Lingli Sun b, Hao Zhang b, Shuyu Zhou b, Liping Cao c, Hongquan Guo b, Wen Sun b,
Bernard Yan d, Stephen M. Davis d, Zhizhong Zhang a,⁎, Xinfeng Liu a,⁎
a
Department of Neurology, Jinling Hospital, Southern Medical University, Nanjing 210002, Jiangsu, China
b
Department of Neurology, Jinling Hospital, Medical School of Nanjing University, Nanjing 210002, Jiangsu, China
c
Department of Neurology, The First People's Hospital of Changzhou, Changzhou 213000, Jiangsu, China
d
Melbourne Brain Centre, Royal Melbourne Hospital, University of Melbourne, Melbourne, Australia

a r t i c l e i n f o a b s t r a c t

Article history: Purpose: Prostaglandin-Endoperoxide Synthase 1 (PTGS1) and smoking may play important roles in aspirin
Received 22 September 2016 nonresponsiveness, but the effect of their interaction on stroke outcomes remains largely unknown. We exam-
Received in revised form 3 March 2017 ined the effects of PTGS1 polymorphisms, smoking status, and their interaction on functional outcomes in a co-
Accepted 10 March 2017 hort of Chinese Han patients with stroke during aspirin therapy.
Available online 14 March 2017
Methods: A total of 617 ischemic stroke patients taking aspirin were enrolled. Three single nucleotide polymor-
phisms (SNPs) rs1330344, rs3842788, and rs5788 in PTGS1 were determined for genotyping. Poor functional out-
Keywords:
Prostaglandin-Endoperoxide Synthase 1
comes were defined as a modified Rankin Scale (mRS) of 3–6 at 90-day follow-up. The influence of PTGS1 gene-
Single nucleotide polymorphism smoking interaction on functional outcomes was examined.
Gene-environment interaction Results: Poor functional outcomes occurred in 145 (23.5%) patients. When adjusting multiple factors by logistic
Aspirin regression, CC genotype of rs1330344 was associated with poor functional outcomes (risk ratio [RR] = 1.73;
95% confidence interval [CI]: 1.17–2.37). A similar connection was found in the CGC haplotype (RR = 1.40;
95% CI: 1.08–1.77). Furthermore, we found a significant interaction between rs1330344 and smoking status
(P interaction = 0.018); the interaction effect between the PTGS1 haplotype and smoking also showed statistical sig-
nificance (P interaction = 0.040).
Conclusions: In Chinese Han stroke patients with aspirin therapy, the adverse effect of PTGS1 polymorphisms on
functional outcomes may be modulated by the smoking status. PTGS1 gene-smoking interaction might in part re-
flect the heterogeneity in the prognosis of patients treated with aspirin.
© 2017 Elsevier B.V. All rights reserved.

1. Introduction aggregation. As a key role in aspirin's therapeutic target, COX-1 activity is

Despite received therapeutic doses of aspirin for secondary preven-
tion of stroke, a part of patients still do not respond appropriately to as-
pirin and experience adverse clinical outcomes. This phenomenon has
been described as aspirin nonresponsiveness or aspirin treatment fail-
ure. A number of potentially genetic and environmental factors may
lead alone or together to aspirin nonresponsiveness [1,2].
Aspirin irreversibly acetylates cyclooxygenase-1 (COX-1), preventing
the production of thromboxane A2 (TXA2), and thereby reduces platelet
⁎ Corresponding authors at: Department of Neurology, Jinling Hospital, Southern
Medical University, 305 East Zhongshan Road, Nanjing 210002, Jiangsu Province, China.
E-mail addresses: zhizhongn@gmail.com (Z. Zhang), xfliu2@vip.163.com (X. Liu).
1
These authors contributed equally to this work.
modulated by both genetic and environmental factors. COX-1 enzyme
action and its interaction with aspirin may be influenced by single nucle-
otide polymorphisms (SNPs) or haplotypes in Prostaglandin-Endoperox-
ide Synthase 1 (PTGS1), gene encoding for COX-1 [3]. Since aspirin
therapy is essential to reduce the risk of recurrent stroke and other ische-
mic events, PTGS1 genetic susceptibility may contribute to aspirin
nonresponsiveness [4–7], and thus differ the prognosis of cardiovascular
diseases [8–11]. However, the results from published studies were not
always consistent [12–16].
The reason for this inconformity may be explained by the diversity of
population characteristics, different SNP allele frequencies, and possible
gene-environment interaction. As an important environmental factor,
smoking increased the risk of aspirin nonresponsiveness [17,18]. The
different smoking prevalence in diverse populations may partly explain

http://dx.doi.org/10.1016/j.jns.2017.03.014
0022-510X/© 2017 Elsevier B.V. All rights reserved.
212 H. Cai et al. / Journal of the Neurological Sciences 376 (2017) 211–215
these discordant results. Moreover, platelet COX-1 expression was pro-
posed to be higher in smokers by up-regulating of platelet COX-1 path-
way [19]. These evidences indicated that not only smoking status but
also its interaction with PTGS1 might influence aspirin responsiveness.
It seems plausible that the interaction between PTGS1 and smoking sta-
tus may affect the prognosis of ischemic stroke patients. Nevertheless,
few studies focused on it. Thus, we investigated the relationship be-
tween PTGS1 polymorphisms and functional outcomes in Chinese Han
patients with stroke taking aspirin, especially the gene-smoking inter-
action analysis.
2. Methods
2.1. Study subjects
Based on Nanjing Stroke Registry Program (NSRP) [20], we consecu-
tively screened acute ischemic stroke patients treated with aspirin be-
tween December 2009 and October 2012 to recruit a cohort. Patients
who met all the following criteria were included: 1) Chinese Han eth-
nicity; 2) aged 18 years or older; 3) diagnosed with acute ischemic
stroke, which was defined as focal neurological dysfunction persisting
≥24 h and confirmed by imaging evidence; 4) treated with enteric-coat-
ed aspirin (100 mg daily) within 48 h after symptom onset; 5) with a
modified Rankin Scale (mRS) score ≤ 1 before the index stroke; 6)
with available blood samples. We excluded patients with any of the fol-
lowing situation: 1) treated with anticoagulants, non-steroid anti-in-
flammatory drugs or other antiplatelet agents besides aspirin; 2)
stroke subtype was classified as cardio-embolism according to the
Trial of Org 10172 in Acute Stroke Treatment (TOAST) criteria; 3)
discontinued or changed the antiplatelet regime before the 90-day fol-
low-up; 4) had a history of bleeding or hematological disorder; 5) diag-
nosed with malignancies; 6) diagnosed with severe liver or kidney
dysfunction. Additional secondary prevention treatments were in ac-
cordance with the approved guideline. This study was approved by
the Ethical Review Board of Jinling Hospital, Nanjing, China. Informed
consent was obtained from all individual participants included in the
study.
2.2. Data collection
Trained physicians collected baseline information, including demo-
graphic characteristics, medical history, laboratory tests, treatment, pre-
treatment mRS and NIH Stroke Scale (NIHSS) scores. Smokers were
those who had smoked ≥100 cigarettes during their lifetime and cur-
rently smoke every day or some days [21]. Functional outcomes mea-
sured by mRS were evaluated at 90 days after stroke onset. Face-to-
face visit or telephone interview was used for the 90-day follow-up by
certified investigators who were blind to the genotype data. Poor func-
tional outcomes were defined as mRS of 3–6, while good functional out-
comes were mRS of 0–2.
2.3. Target SNPs selection and genotyping
We selected target SNPs in two stages. First, we screened all citations
for potentially relevant studies published before December 2015 in
PubMed, using the term [(“aspirin” or “acetylsalicylic acid”) AND
(“COX-1” or “PTGS1” or “cyclooxygenase-1”) AND (“genetic” or “SNP”
or “polymorphism” or “mutation” or “genotype” or “variant”)]. In the
second stage, we further excluded studies irrelevant to aspirin response
or PTGS1. SNPs with minor allele frequencies (MAF) b 5% according to
1000 Genomes database (http://www.ncbi.nlm.nih.gov/variation/
tools/1000genomes/) in Chinese population were excluded. As a result,
three SNPs (rs1330344, rs3842788, rs5788) were included in this study.
Flow diagram of SNPs selection is shown in Supplementary Fig. S1. The
Supplementary Table S1 shows details of these SNPs. Linkage disequilib-
rium (LD) plot across the PTGS1 region based on the HapMap CHB (Han
Chinese in Beijing, China) data (http://hapmap.ncbi.nlm.nih.gov/) is
presented in Fig. 1. The three target SNPs are located in a LD block of
12 kb.
Genotyping was accomplished by Improved Multiple Ligase Detec-
tion Reaction (iMLDR) [22], with technical assistance from Center for
Genetic and Genomic Analysis, Genesky Biotechnologies Inc. (Shanghai,
China). For assessing confirmation of the genotype data, 5% of the sam-
ples were randomly selected to repeat and the results were 100%
concordant.
2.4. Statistical analysis
Power of the study was calculated by PS Software (Power and Sam-
ple Size Calculation, version 3.1.2). The following parameters were set:
the relative risk of 1.7, a type I error probability of 0.05, and the proba-
bility of the outcome for a control patient of 0.21. Based on our sample
size, we obtained a power of 87.4%.
For comparison between groups at baseline, Student's t-test or
Mann-Whitney U test was conducted as appropriate for continuous var-
iable, while χ2 tests or Fisher's exact tests for categorical data as appro-
priate. Univariate and multivariate logistic regression were performed
to investigate the association between PTGS1 polymorphisms and the
rate of poor functional outcomes. Akaike information criterion (AIC)
was calculated for selecting the best model for a specific SNP. Because
of the common outcomes, the odds ratio (OR) obtained from logistic re-
gression was converted to a risk ratio (RR) in order to better represent
the true relative risk [23].
LD between pairs of loci and Hardy-Weinberg equilibrium were test-
ed with Haploview Version 4.2 (http://www.broad.mit.edu/mpg/
haploview). Haplotypes reconstruction by the EM algorithm, associa-
tion of haplotypes with functional outcomes, and haplotype-smoking
interaction were calculated by SNPStats web tool (http://bioinfo.
iconcologia.net/SNPstats) [24]. All haplotypes with frequencies b 0.01
were excluded.
Tests for interaction with smoking were only focused on SNPs that
were significant for the overall effect. We included a cross-product
term in the logistic regression model containing the main effect of geno-
type and smoking status, treating the P value for the cross-product term
as P interaction. Multiple testing corrections were carried out with the
Bonferroni method when necessary. Statistical analysis was performed
by SPSS version 22.0 (IBM Corp., Armonk, NY).
3. Results
3.1. Baseline characteristics
From December 2009 to October 2012, 678 patients were included
in the cohort, of which 22 (3.2%) lost to follow-up. Additionally, a total
of 39 patients (5.8%) discontinued or changed the antiplatelet regime
before the 90-day follow-up. Finally, 617 patients were enrolled in the
analysis (Supplementary Fig. S2). The baseline data of the study popula-
tion grouped by outcomes is provided in Table 1. Poor functional out-
comes occurred in 145 patients (23.5%). The poor functional outcome
individuals were older than those with good functional outcomes
(60.54 ± 11.66 vs 58.16 ± 12.06 years, P = 0.037). Additionally, the
poor functional outcome group was more likely to have a higher
NIHSS (9 [7–12] vs 3 [1–5], P b 0.001), higher proportion of smokers
(46.2% vs 37.3%, P = 0.055) and hypertension (80.0% vs 67.2%, P =
0.003). No significant heterogeneity with regards to other factors be-
tween two groups was observed.
3.2. PTGS1 polymorphisms and functional outcomes
Genotyping was successfully performed for all samples. As showed
in Supplementary Table S1, the MAFs of rs1330344, rs3842788 and
rs5788 are 38.6%, 8.9%, 5.8%, respectively. There was no significant
 H. Cai et al. / Journal of the Neurological Sciences 376 (2017) 211–215 213

Fig. 1. Linkage disequilibrium (LD) plot across the Prostaglandin-Endoperoxide Synthase 1 (PTGS1) gene region based on the HapMap CHB (Han Chinese in Beijing, China) data. The depth of
red color represents the magnitude of D′. Adapted from the Haploview program and licensed under the Creative Commons Attribution-Share Alike 4.0 International license.

deviation from Hardy-Weinberg equilibrium. Estimation of LD between functional outcomes are shown in Table 2. Depending on the AIC values,
pairs of loci based on our data is also available (Supplementary Fig. S3, rs1330344 best fitted the recessive model. Univariate logistic regression
Supplementary Table S2). The genotypes and their association with suggested that patients with CC genotype were more likely to have poor
functional outcomes (RR = 1.57; 95% confidence interval [CI]: 1.14–
2.08; P = 0.010). After adjustment for age, sex, smoking status, stroke
Table 1
severity (NIHSS score) and hypertension, the CC genotype of
Baseline data of the study population.
rs1330344 was associated with poor functional outcomes (adjusted
Characteristic mRS of 0–2 mRS of 3–6 P value RR = 1.73; 95% CI: 1.17–2.37; P = 0.007). The association was still sig-
(n = 472) (n = 145)
nificant after correcting for multiple comparisons (adjusted P = 0.021).
Age, years 58.16 ± 12.06 60.54 ± 11.66 0.037 Clinical features of CC genotype were similar to those of CT/TT geno-
Male, n (%) 357 (75.6) 99 (68.3) 0.078
types (Supplementary Table S3). No significant difference in the preva-
Medical history, n (%)
Hypertension 317 (67.2) 116 (80.0) 0.003
lence of genotypes of rs3842788 (P = 0.170) and rs5788 (P = 0.750)
Diabetes mellitus 125 (26.5) 40 (27.6) 0.793 was observed between the two groups.
Previous stroke 73 (15.5) 26 (17.9) 0.479 Three target SNPs (rs1330344, rs3842788, rs5788) were selected for
Smokers 176 (37.3) 67 (46.2) 0.055 haplotype analysis (Supplementary Table S4). After adjusting for multi-
Drinking history 156 (33.1) 37 (25.5) 0.087
ple confounders, the CGC haplotype significantly increased the risk of
Previous aspirin use 98 (20.8) 28 (19.3) 0.704
NIHSS score at baseline 3 (1–5) 9 (7–12) b0.001 poor functional outcomes compared with the TGC haplotype (adjusted
mRS score before the index events, n 0.438 RR = 1.40; 95% CI: 1.08–1.77; P = 0.011). This association was still sig-
(%) nificant after Bonferroni correction (adjusted P = 0.044).
0 433 (91.7) 130 (89.7)
1 39 (8.3) 15 (10.3)
TOAST classifications 0.339
3.3. Interaction analysis
LAA 233 (49.4) 76 (52.4)
SVO 182 (38.6) 47 (32.4)
UND 57 (12.1) 22 (15.2) Since only rs1330344 was significant for the overall effect, we per-
Triglyceride, mmol/L 1.38 1.39 0.944 formed interaction analysis with smoking about it. Fig. 2 shows the effect
(1.04–1.86) (1.07–1.76) of rs1330344 on the risk of poor functional outcomes stratified by smoking
Cholesterol, mmol/L 4.38 ± 1.00 4.28 ± 1.00 0.293
status. Compared with non-smokers with rs1330344 CT/TT genotype,
HDL, mmol/L 1.09 ± 0.25 1.05 ± 0.26 0.174
LDL, mmol/L 2.66 ± 0.88 2.52 ± 0.82 0.083 non-smokers with CC had a 1.18-fold (RR =1.18; 95% CI: 0.64–1.97) in-
Treatment, n (%) creased risk of poor functional outcomes, while smokers with CT/TT had
Statin 458 (97.0) 139 (95.9) 0.486 the RR of 1.20 (95% CI: 0.81–1.69). The risk of poor functional outcomes
Antihypertensive 233 (49.4) 80 (55.2) 0.221
was highest in smoking persons with CC genotype, exhibiting a 3.05-fold
Thrombolysis 7 (1.5) 5 (3.4) 0.165
Stent 122 (25.8) 34 (23.4) 0.561 (95% CI: 1.94–3.94) increased risk than non-smokers with CT/TT geno-
type. Moreover, interaction analysis revealed that interaction between
mRS: modified Rankin Scale; NIHSS: NIH Stroke Scale; TOAST: Trial of Org 10,172 in Acute
Stroke Treatment; LAA: large-artery atherosclerosis; SVO: small-vessel occlusion; UND:
rs1330344 and smoking status was significant (P interaction = 0.018).
other determined and undetermined causes; HDL: high-density lipoprotein; LDL: low- In haplotype-smoking interaction analysis, we also found the interaction
density lipoprotein. effect between the PTGS1 haplotype inferred from three variants
214 H. Cai et al. / Journal of the Neurological Sciences 376 (2017) 211–215

Table 2
PTGS1 genotypes and their association with functional outcomes.

Genetic models Genotype mRS of 0–2 (n = 472) mRS of 3–6 (n = 145) Crude RR (95% CI) Crude P value Adjusted RR (95% CI)a P valuea AIC

rs1330344
Codomindant model TT 175 (37.1%) 49 (33.8%) 1.00 0.034 1.00 0.026 512.7
CT 228 (48.3%) 61 (42.1%) 0.97 (0.69–1.33) 1.05 (0.71–1.50)
CC 69 (14.6%) 35 (24.1%) 1.54 (1.06–2.10) 1.77 (1.15–2.47)
Dominant model TT 175 (37.1%) 49 (33.8%) 1.00 0.470 1.00 0.230 516.6
CT/CC 297 (62.9%) 96 (66.2%) 1.11 (0.82–1.48) 0.010 1.23 (0.87–1.68)
Recessive model CT/TT 403 (85.4%) 110 (75.9%) 1.00 0.010 1.00 0.007 510.7
CC 69 (14.6%) 35 (24.1%) 1.57 (1.14–2.08) 1.73 (1.17–2.37)
rs3842788
Codominant model GG 393 (83.3%) 128 (88.3%) 1.00 0.180 1.00 0.170 516.5
GA 76 (16.1%) 17 (11.7%) 0.75 (0.46–1.15) 0.87 (0.50-1.40)
AA 3 (0.6%) 0 (0.0%) NA NA
Dominant model GG 393 (83.3%) 128 (88.3%) 1.00 0.130 1.00 0.430 517.3
GA/AA 79 (16.7%) 17 (11.7%) 0.72 (0.45–1.12) 0.81 (0.46–1.31
rs5788
Codominant model CC 422 (89.4%) 128 (88.3%) 1.00 0.282 1.00 0.472 518.5
CA 46 (9.8%) 17 (11.7%) 1.16 (0.74–1.72) 1.14 (0.66–1.79)
AA 4 (0.8%) 0 (0.0%) NA NA
Dominant model CC 422 (89.4%) 128 (88.3%) 1.00 0.702 1.00 0.750 517.9
CA/AA 50 (10.6%) 17 (11.7%) 1.09 (0.68–1.63) 1.09 (0.63–1.72)

PTGS1: Prostaglandin-Endoperoxide Synthase 1; NA: not applicable; CI: confidence interval; RR: risk ratio; AIC: akaike information criterion.
a
Adjusted for age, sex, smoking status, stroke severity and hypertension.

(rs1330344, rs3842788, rs5788) and smoking status was statistically
significant (P interaction = 0.040, Supplementary Table S5).
4. Discussion
In this study, we found that rs1330344 CC genotype significantly in-
creased the risk of poor functional outcomes at 90 days in a prospective
Chinese ischemic stroke cohort treated with aspirin. A similar connec-
tion was found in the CGC haplotype. Moreover, we observed significant
interactions between PTGS1 genotype/haplotype and smoking status.
Most of the variants in PTGS1 are present at low frequency, which
makes it difficult to detect the true association. Therefore, we chose
three common polymorphisms (rs1330344, rs3842788 and rs5788) in
Chinese subjects. Our finding showed that CC genotype in rs1330344
was an unfavorable prognostic factor for acute ischemic stroke treated
with aspirin. Concordantly, a recent study in China also identified an as-
sociation of rs1330344 with poor vascular outcomes [9] in extracranial
or intracranial stenting patients. Similar results were reported in other
Chinese populations with different characteristics [10,11]. In addition,
we also found an association between the CGC haplotype and poor func-
tional outcomes.
PTGS1 genetic variation in the sequence upstream of the transcrip-
tion site or signal peptide region was more common and most likely
Fig. 2. Risk ratio (RR) of poor functional outcomes according to rs1330344 and smoking
status, non-smokers with CT/TT genotype as a reference. *Significant interaction
between rs1330344 and smoking (P interaction = 0.018).
results in differences in the function of COX-1 protein. The hypothetical
mechanisms include affecting the amount of COX-1 protein, the location
of COX-1 in the cell, the biological activity of its product and changing
the role of COX-1 in relation to aspirin [5,25]. Rs1330344 is located in
the promoter of PTGS1 and appears to be informative and potentially
functional [26]. Allele C in rs1330344 possibly alters putative transcrip-
tion factor (GATA-1, CdxA) binding sites [26,27] and influence patho-
genesis, development and severity of complex diseases through its
effect on COX-1 expression and prostaglandin biosynthesis [28]. Fur-
thermore, it was shown that rs1330344 contributed to aspirin
nonresponsiveness [7]. Therefore, we speculated that rs1330344
might affect clinical outcomes through attenuating antiplatelet effect
by aspirin.
In addition, the interaction between genetic and environmental fac-
tors plays an important role in aspirin nonresponsiveness [1,2]. Previous
study demonstrated a higher frequency of aspirin nonresponsiveness in
smokers [29]. And this phenomenon might be partly explained by a
higher expression of COX-1 in smokers via up-regulating the COX-1
pathway [19]. A study explored its mechanism and found that the
smoking-derived N-nitrosamine, 4-(methylnitrosamino)-1-(3-pyri-
dyl)-1-butanone (NNK) generated reactive oxygen species (ROS).
Then ROS could activate the nuclear factor-κB (NF-κB), leading to in-
creased COX-1 activity and subsequent platelet aggregation [30]. Addi-
tionally, researchers found that smoking reduced aspirin response by
increasing the atherogenic metabolites [31–33] prostaglandin F2α
(PGF2α) [34–36] that could be altered by genetic polymorphisms in
PTGS1 [37]. To sum up, it is biologically plausible that the interaction be-
tween PTGS1 variants and smoking may reduce aspirin response; there-
fore, aspirin possibly fails to prevent recurrent stroke or neurological
deterioration, leading to poor functional outcomes.
There are several limitations that should be addressed. First, this
study was conducted in Chinese Han patients, which may not generalize
these findings to other ethnic groups. Second, platelet aggregation func-
tion was not evaluated in the study which might provide more evidence
for a functional role of rs1330344 CC genotype. Finally, this is an obser-
vational genetic association study derived from a sample of Chinese
acute ischemic stroke patients. The replication of these findings in a
multi-ethnic longitudinal cohort with larger sample size and functional
analysis are warranted.
In conclusion, our result suggested that in Chinese Han patients with
stroke taking aspirin for secondary prevention, PTGS1 polymorphisms
may increase the risk of poor functional outcomes and this effect may
 H. Cai et al. / Journal of the Neurological Sciences 376 (2017) 211–215
be modulated by the smoking status. PTGS1 gene-smoking interaction
might in part reflect the heterogeneity in the prognosis of patients treat-
ed with aspirin.
Conflict of interest
The authors declare that they have no conflict of interest.
Acknowledgments
This study was supported by National Natural Science Foundation of
China (81220108008, 81501193 and 81571148) and China Postdoctoral
Science Foundation (2016M592954). We greatly appreciate Miss Ying
Lin and Miss Zhaojun Wang in Jinling Hospital for their technical
support.
Appendix A. Supplementary data
Supplementary data to this article can be found online at http://dx.
doi.org/10.1016/j.jns.2017.03.014.
References
[1] A. Undas, K.E. Brummel-Ziedins, K.G. Mann, Antithrombotic properties of aspirin
and resistance to aspirin: beyond strictly antiplatelet actions, Blood 109 (6)
(2007) 2285–2292.
[2] G. Feher, A. Feher, G. Pusch, G. Lupkovics, L. Szapary, E. Papp, The genetics of anti-
platelet drug resistance, Clin. Genet. 75 (1) (2009) 1–18.
[3] T. Goodman, P. Sharma, A. Ferro, The genetics of aspirin resistance, Int. J. Clin. Pract.
61 (5) (2007) 826–834.
[4] J. Ulehlova, L. Slavik, J. Kucerova, V. Krcova, J. Vaclavik, K. Indrak, Genetic polymor-
phisms of platelet receptors in patients with acute myocardial infarction and resis-
tance to antiplatelet therapy, Genet. Test. Mol. Biomarkers 18 (9) (2014) 599–604.
[5] M.K. Halushka, L.P. Walker, P.V. Halushka, Genetic variation in cyclooxygenase 1: ef-
fects on response to aspirin, Clin. Pharmacol. Ther. 73 (1) (2003) 122–130.
[6] A.O. Maree, R.J. Curtin, A. Chubb, C. Dolan, D. Cox, J. O'Brien, et al., Cyclooxygenase-1
haplotype modulates platelet response to aspirin, J. Thromb. Haemost. 3 (10)
(2005) 2340–2345.
[7] X.L. Li, J. Cao, L. Fan, Q. Wang, L. Ye, C.P. Cui, et al., Genetic polymorphisms of HO-1
and COX-1 are associated with aspirin resistance defined by light transmittance ag-
gregation in Chinese Han patients, Clin. Appl. Thromb. Hemost. 19 (5) (2013)
513–521.
[8] C.R. Lee, K.E. North, M.S. Bray, D.J. Couper, G. Heiss, D.C. Zeldin, Cyclooxygenase poly-
morphisms and risk of cardiovascular events: the Atherosclerosis Risk in Communi-
ties (ARIC) study, Clin. Pharmacol. Ther. 83 (1) (2008) 52–60.
[9] X.Q. Li, N. Ma, X.G. Li, B. Wang, S.S. Sun, F. Gao, et al., Association of PON1, P2Y12 and
COX1 with recurrent ischemic events in patients with Extracranial or intracranial
stenting, PLoS One 11 (2) (2016), e0148891. .
[10] L. Fan, J. Cao, L. Liu, X. Li, G. Hu, Y. Hu, et al., Frequency, risk factors, prognosis, and
genetic polymorphism of the cyclooxygenase-1 gene for aspirin resistance in elderly
Chinese patients with cardiovascular disease, Gerontology 59 (2) (2013) 122–131.
[11] L. Cao, Z. Zhang, W. Sun, W. Bai, W. Sun, Y. Zhang, et al., Impacts of COX-1 gene poly-
morphisms on vascular outcomes in patients with ischemic stroke and treated with
aspirin, Gene 546 (2) (2014) 172–176.
[12] C. Pettinella, M. Romano, L. Stuppia, F. Santilli, R. Liani, G. Davi, Cyclooxygenase-1
haplotype C50T/A-842G does not affect platelet response to aspirin, Thromb.
Haemost. 101 (4) (2009) 687–690.
[13] M. Postula, A. Kaplon-Cieslicka, M. Rosiak, A. Kondracka, A. Serafin, K.J. Filipiak, et al.,
Genetic determinants of platelet reactivity during acetylsalicylic acid therapy in di-
abetic patients: evaluation of 27 polymorphisms within candidate genes, J. Thromb.
Haemost. 9 (11) (2011) 2291–2301.
[14] Q. Li, B.L. Chen, V. Ozdemir, W. Ji, Y.M. Mao, L.C. Wang, et al., Frequency of genetic
polymorphisms of COX1, GPIIIa and P2Y1 in a Chinese population and association
with attenuated response to aspirin, Pharmacogenomics 8 (6) (2007) 577–586.
215
[15] Z.H. Xu, J.R. Jiao, R. Yang, B.Y. Luo, X.F. Wang, F. Wu, Aspirin resistance: clinical sig-
nificance and genetic polymorphism, J. Int. Med. Res. 40 (1) (2012) 282–292.
[16] X. Yi, Q. Zhou, J. Lin, L. Chi, Z. Han, Platelet response to aspirin in Chinese stroke pa-
tients is independent of genetic polymorphisms of COX-1 C50T and COX-2 G765C, J.
Atheroscler. Thromb. 20 (1) (2013) 65–72.
[17] I. Coma-Canella, A. Velasc, Variability in individual responsiveness to aspirin: clinical
implications and treatment, Cardiovasc. Hematol. Disord. Drug Targets 7 (4) (2007)
274–287.
[18] B. Labuz-Roszak, K. Pierzchala, K. Tyrpien, Resistance to acetylsalicylic acid in pa-
tients with type 2 diabetes mellitus is associated with lipid disorders and history
of current smoking, J. Endocrinol. Investig. 37 (4) (2014) 331–338.
[19] W.M. Loke, K.L. Sing, C.Y. Lee, W.L. Chong, S.E. Chew, H. Huang, et al., Cyclooxygen-
ase-1 mediated platelet reactivity in young male smokers, Clin. Appl. Thromb.
Hemost. 20 (4) (2014) 371–377.
[20] X. Liu, G. Xu, W. Wu, R. Zhang, Q. Yin, W. Zhu, Subtypes and one-year survival of
first-ever stroke in Chinese patients: the Nanjing stroke registry, Cerebrovasc. Dis.
22 (2–3) (2006) 130–136.
[21] I.T. Agaku, B.A. King, S.R. Dube, Centers for disease C, prevention. Current cigarette
smoking among adults - United States, 2005-2012, MMWR Morb. Mortal. Wkly
Rep. 63 (2) (2014) 29–34.
[22] G. Thomas, R. Sinville, S. Sutton, H. Farquar, R.P. Hammer, S.A. Soper, et al., Capillary
and microelectrophoretic separations of ligase detection reaction products pro-
duced from low-abundant point mutations in genomic DNA, Electrophoresis 25
(10−11) (2004) 1668–1677.
[23] J. Zhang, K.F. Yu, What's the relative risk? A method of correcting the odds ratio in
cohort studies of common outcomes, JAMA 280 (19) (1998) 1690–1691.
[24] X. Sole, E. Guino, J. Valls, R. Iniesta, V. Moreno, SNPStats: a web tool for the analysis
of association studies, Bioinformatics 22 (15) (2006) 1928–1929.
[25] G. Cai, W. Zhou, Y. Lu, P. Chen, Z. Lu, Y. Fu, Aspirin resistance and other aspirin-relat-
ed concerns, Neurol. Sci. 37 (2) (2016) 181–189.
[26] T. Arisawa, T. Tahara, T. Shibata, M. Nagasaka, M. Nakamura, Y. Kamiya, et al., Asso-
ciation between genetic polymorphisms in the cyclooxygenase-1 gene promoter
and peptic ulcers in Japan, Int. J. Mol. Med. 20 (3) (2007) 373–378.
[27] T. Arisawa, T. Tahara, T. Shibata, M. Nagasaka, M. Nakamura, Y. Kamiya, et al., Genet-
ic polymorphisms of cyclooxygenase-1 (COX-1) are associated with functional dys-
pepsia in Japanese women, J. Women's Health 17 (6) (2008) 1039–1043.
[28] H. Kapitanovic Vidak, T. Catela Ivkovic, Z. Vidak, S. Kapitanovic, COX-1 and COX-2
polymorphisms in susceptibility to cerebral palsy in very preterm infants, Mol.
Neurobiol. (2016).
[29] A. Mirkhel, E. Peyster, J. Sundeen, L. Greene, A.D. Michelson, A. Hasan, et al., Frequen-
cy of aspirin resistance in a community hospital, Am. J. Cardiol. 98 (5) (2006)
577–579.
[30] N. Rioux, A. Castonguay, The induction of cyclooxygenase-1 by a tobacco carcinogen
in U937 human macrophages is correlated to the activation of NF-kappaB, Carcino-
genesis 21 (9) (2000) 1745–1751.
[31] H.S. Sohn, H. Kim, I.S. Song, E. Lim, M. Kwon, J.H. Ha, et al., Evidence supporting the
need for considering the effects of smoking on drug disposition and effectiveness in
medication practices: a systematic narrative review, Int. J. Clin. Pharmacol. Ther. 53
(8) (2015) 621–634.
[32] S. Basu, Bioactive eicosanoids: role of prostaglandin F(2alpha) and F(2)-isoprostanes
in inflammation and oxidative stress related pathology, Mol. Cell 30 (5) (2010)
383–391.
[33] Y. Yu, M.B. Lucitt, J. Stubbe, Y. Cheng, U.G. Friis, P.B. Hansen, et al., Prostaglandin
F2alpha elevates blood pressure and promotes atherosclerosis, Proc. Natl. Acad.
Sci. U. S. A. 106 (19) (2009) 7985–7990.
[34] S. Sanderson, J. Emery, T. Baglin, A.L. Kinmonth, Narrative review: aspirin resistance
and its clinical implications, Ann. Intern. Med. 142 (5) (2005) 370–380.
[35] J.D. Morrow, B. Frei, A.W. Longmire, J.M. Gaziano, S.M. Lynch, Y. Shyr, et al., Increase
in circulating products of lipid peroxidation (F2-isoprostanes) in smokers. Smoking
as a cause of oxidative damage, N. Engl. J. Med. 332 (18) (1995) 1198–1203.
[36] M. Reilly, N. Delanty, J.A. Lawson, G.A. Fitz Gerald, Modulation of oxidant stress in
vivo in chronic cigarette smokers, Circulation 94 (1) (1996) 19–25.
[37] J. Helmersson, J. Arnlov, T. Axelsson, S. Basu, A polymorphism in the cyclooxygenase
1 gene is associated with decreased inflammatory prostaglandin F2alpha formation
and lower risk of cardiovascular disease, Prostaglandins Leukot. Essent. Fat. Acids 80
(1) (2009) 51–56.