- Email: [email protected]
Cerebral infarcts after coronary angiography and percutaneous coronary intervention. A prospective propensity score adjusted comparison of right radial, left radial and femoral approaches
CARREV-01758; No of Pages 6 Cardiovascular Revascularization Medicine xxx (xxxx) xxx
Contents lists available at ScienceDirect
Cardiovascular Revascularization Medicine
Cerebral infarcts after coronary angiography and percutaneous coronary intervention. A prospective propensity score adjusted comparison of right radial, left radial and femoral approaches☆,☆☆,★,★★ Nicola Marchese a,⁎, Massimiliano Copetti b, Vincenzo Inchingolo c, Teresa Popolizio d, Andrea Fontana b, Annalisa Simeone c, Carlo Vigna a a
Unit of Cardiology, Fondazione IRCCS Casa Sollievo della Sofferenza, San Giovanni Rotondo, Italy Unit of Biostatistics, Fondazione IRCCS Casa Sollievo della Sofferenza, San Giovanni Rotondo, Italy c Unit of Neurology, Fondazione IRCCS Casa Sollievo della Sofferenza, San Giovanni Rotondo, Italy d Unit of Radiology, Fondazione IRCCS Casa Sollievo della Sofferenza, San Giovanni Rotondo, Italy b
a r t i c l e
i n f o
Article history: Received 30 August 2019 Received in revised form 7 November 2019 Accepted 11 November 2019 Available online xxxx Keywords: Radial access Femoral access Cerebral infarct
a b s t r a c t Background: New cerebral infarcts (CIs) detected at magnetic resonance imaging (MRI) are reported after cardiac procedures. Clinical and procedural aspects are implicated as potential causal factors. The aim of this study was to evaluate the incidence of new CIs after coronary angiography and percutaneous coronary intervention according to the arterial access site. Methods: 180 patients undergoing elective coronary angiography were studied with cerebral MRI the day before and the day after the procedure. Unadjusted and propensity score (PS) analyses were performed comparing the occurrence of CIs in right radial (RR), left radial (LR) and transfemoral (TF) access groups. Results: New CIs were observed in 14 patients (7.8% of the total sample, one with neurological sequelae). CIs were detected in 15.5% vs 4.9% vs 3.3% of RR, LR and TF groups, respectively (p = .026). In PS adjusted analyses, the RR approach was associated with more CIs compared with the TF approach (odds ratio [OR] estimate from logistic regression adjusted by PS quartiles: 0.158; 95% confidence interval: 0.031 to 0.814; p = .027) and the LR approach (OR: 0.266; 95% confidence interval: 0.066 to 1.080; p = .064). In a secondary analysis, a comparison of RR vs non-RR approach (TF + LR) was performed, showing that post-procedural CIs were more frequent in the RR group (OR: 0.170; 95% confidence interval: 0.050 to 0.574; p = .004). Conclusions: Our study suggests that the RR approach may be associated with a higher rate of new CIs after coronary angiography compared with LR and TF approaches. © 2019 Elsevier Inc. All rights reserved.
1. Introduction
Abbreviations: CIs, cerebral infarcts; MRI, magnetic resonance imaging; PCI, percutaneous coronary intervention; RR, right radial; LR, left radial; TF, transfemoral; PS, propensity score. ☆ All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation. ☆☆ Declaration of competing interest: None. ★ This research did not receive any specific grant from funding agencies in the public, commercial or not-for-profit sectors. ★★ Summary for the annotated table of contents: Radiologically documented cerebral infarcts are frequently reported after left heart procedures, have important clinical implications and vascular access site may play an important role. The present study suggests that right radial approach may be associated with a higher risk of post-procedural cerebral infarcts compared with the left radial and femoral approaches and this association is only partially explained by the presence of arterial tortuosity. ⁎ Corresponding author at: Unit of Cardiology, Casa Sollievo della Sofferenza Hospital IRCCS, Viale Cappuccini 1, 71013 San Giovanni Rotondo, FG, Italy. E-mail address: [email protected] (N. Marchese).
Ischemic stroke after cardiac catheterization is a rare but potentially severe complication, with rates ranging from 0.1% to 0.6% [1]. Stroke occurrence is associated with high in-hospital mortality and great disability in survivors [1]. Clinical (older age, male gender, known cerebrovascular disease, acute coronary syndrome, major cardiovascular risk factors, and coronary atherosclerotic burden) and procedural factors (systematic aortic valve crossing, use of intra-aortic balloon pump, number of catheters used, large guide caliber, and high contrast volumes) have been identified as potential risk factors for post-procedural stroke [2,3]. Fortunately, stroke represents the smallest, albeit dangerous, part of the more multifaceted and recurrent phenomenon of embolic brain injury after cardiac procedures, which may occur subclinically or even be totally silent. Substantially higher rates of clinically silent acute cerebral infarcts (CIs) have been identified at diffusion-weighted magnetic resonance
https://doi.org/10.1016/j.carrev.2019.11.010 1553-8389/© 2019 Elsevier Inc. All rights reserved.
Please cite this article as: N. Marchese, M. Copetti, V. Inchingolo, et al., Cerebral infarcts after coronary angiography and percutaneous coronary intervention. A prospective p..., Cardiovascular Revascularization Medicine, https://doi.org/10.1016/j.carrev.2019.11.010
2
N. Marchese et al. / Cardiovascular Revascularization Medicine xxx (xxxx) xxx
imaging (MRI) after left heart catheterization, with wide variability across studies (2% to 35%) [4–13]. Such CIs have been associated with cognitive decline, dementia and depression, and are considered precursors of symptomatic stroke [14]. Since one supposed mechanism of post-procedural acute brain injury is mobilization of atherosclerotic plaques from the aorta, arterial access route may significantly impact the risk of new CIs. Our objective was to evaluate the prevalence of new CIs, either clinically manifest or silent, after elective coronary angiography or percutaneous coronary intervention (PCI), and to compare the incidence of post-procedural CIs at diffusion-weighted MRI using the three most common vascular access sites for left heart catheterization (right radial [RR] vs left radial [LR] vs transfemoral [TF]). 2. Methods 2.1. Patient population In this single-center prospective study, 180 patients with suspected coronary artery disease scheduled for elective coronary angiography were enrolled. Patients with recent acute coronary syndrome, need for aortic valve crossing, coronary artery bypass grafts, preexisting neurological deficit, history of ischemic or hemorrhagic stroke, ongoing atrial fibrillation, and any contraindication for MRI were excluded. Owing to the limited availability of diffusion-weighted MRI for research purposes in our institution (3 days/week), all eligible patients who gave their written consent to participate were recruited on selected days in the study. The local ethics committee approved the study. 2.2. Coronary angiography Coronary angiography was performed by cardiologists with good expertise in radial access (N300 radial exams/year) using the Seldinger technique with 6F catheters via the RR, LR and TF approach, in a 1:1:1 ratio for the three arms. For RR and LR approaches, 6F 25 cm radial sheaths were used, whereas 6F 10 cm femoral sheaths were used for the TF approach (Radiofocus Introducer II, Terumo Europe, Leuven, Belgium). Unfractionated heparin (2500 IU) was administered intravenously at the end of the diagnostic procedure only to RR and LR patients. If ad hoc PCI was necessary, a total of 5000 IU were
administered to all patients before the procedure and then according to activated clotting time. Coronary angiography was performed with 6F diagnostic catheters using standard choices: 3.5 mm left Judkins and 5.0 mm right Judkins for the RR group; 4.0 mm left Judkins and 4.0 mm right Judkins for the LR and TF groups; with a standard 0.035″, 260 cm exchange guidewire (catheters and exchange wire by Cordis Corporation, Miami, FL, USA). The use of a long (260 cm) exchange guidewire was adopted to restrict wire passages in the aortic arch in all patients. Iohexol, a non-ionic, low-osmolar contrast agent (Omnipaque 350 mg, GE Healthcare Inc., Princeton, NJ, USA) was used. Contrast amount, number of catheters used, procedure and fluoroscopy time, accidental aortic valve crossing with the catheter or exchange guidewire, need for hydrophilic guidewire, presence of vascular tortuosity, and significant aortic calcification were collected. After coronary angiography or intervention, patients were carefully evaluated for the occurrence of neurological and arrhythmic complications with continuous electrocardiographic monitoring for at least 24 h. 2.3. Magnetic resonance imaging All patients provided written informed consent to MRI, which was performed the same day as coronary angiography (baseline MRI) and within 48 h after the cardiac procedure (post-cath MRI) with a 1.5 T system (Siemens MAGNETOM Aera, Erlangen, Germany). The imaging protocol of baseline MRI included diffusion-weighted single-shot echoplanar imaging (TR/TE: 5300/97 ms; NEX: 1; slice thickness 5 mm; Matrix 172\172; diffusion gradient b values of 0 and 1000 S/mm2) and T2 fluid attenuated inversion recovery (FLAIR; TR/TE: 8500/82 ms; NEX: 2; TI: 2240 ms; slice thickness 4 mm; Matrix 320\320). The imaging protocol of post-cath MRI included only diffusion-weighted single-shot echo-planar imaging (using the same parameters as described before). In case of inability to perform baseline evaluation, a complete MRI study was obtained after the cardiac procedure, which included, besides diffusion-weighted and FLAIR, also T1-weighted spin-echo (SE) images (TR/TE: 500/9.9 ms; NEX: 1; slice thickness 5 mm; Matrix 320\320), T2weighted fast spin-echo (FSE) images (TR/TE: 4700/87 ms; NEX: 1; slice thickness 5 mm; Matrix 320\320), and gradient recalled echo (GRE) T2weighted imaging (T2WI) (TR/TE: 950/25 ms; FA: 20; NEX 1; Matrix 256\256). Two blinded experts in neuroradiological images (a neuroradiologist and a neurologist), unaware of the study protocol, were asked
Table 1 Clinical and procedural characteristics of study patients according to arterial approach.
Age (years) Male gender Hypertension Diabetes mellitus Family history of CAD Smokers Hypercholesterolemia Renal failure Prior PCI Heparin (IU) Aspirin use Statin use P2Y12 receptor inhibitors IIb/IIIa inhibitors Total procedure time (min) Fluoroscopy time (s) Contrast used (ml) Catheters used Accidental aortic valve crossing Ad-hoc PCI Arterial tortuosity Significant aortic calcification
Right radial (n = 58)
Left radial (n = 61)
Femoral (n = 61)
p
67.9 ± 9.5 45 (77.6) 51 (87.9) 19 (32.8) 20 (34.5) 6 (10.3) 50 (86.2) 3 (5.2) 13 (22.4) 2500 (2500–5000) 55 (94.8) 49 (84.5) 15 (25.9) 4 (6.9) 21.5 (15.0–41.0) 215 (100–440) 92 (50–170) 2.6 ± 0.8 13 (22.4) 27 (46.6) 8 (13.8) 4 (6.9)
68.0 ± 8.6 47 (77.0) 52 (85.2) 20 (32.8) 26 (42.6) 7 (11.5) 50 (82.0) 6 (9.8) 16 (26.2) 5000 (2500–5000) 60 (98.4) 52 (85.2) 21 (34.4) 2 (3.3) 25.0 (15.0–47.0) 270 (90–600) 128 (50–220) 2.9 ± 0.9 18 (29.5) 32 (52.5) 6 (9.8) 6 (9.8)
68.6 ± 7.0 47 (77.0) 54 (88.5) 21 (34.4) 27 (44.3) 6 (9.8) 54 (88.5) 7 (11.5) 13 (21.3) 5000 (0–5000) 59 (96.7) 52 (85.2) 19 (31.1) 3 (4.9) 30.0 (20.0–45.0) 240 (120–660) 126 (50–200) 2.9 ± 1.1 12 (19.7) 32 (52.5) 2 (3.3) 2 (3.3)
0.929 0.997 0.847 0.975 0.511 0.956 0.580 0.458 0.797 0.040 0.562 0.991 0.594 0.663 0.215 0.395 0.270 0.185 0.422 0.760 0.125 0.347
Values are mean ± standard deviation, median (interquartile range), or n (%). CAD = coronary artery disease, PCI = percutaneous coronary intervention.
Please cite this article as: N. Marchese, M. Copetti, V. Inchingolo, et al., Cerebral infarcts after coronary angiography and percutaneous coronary intervention. A prospective p..., Cardiovascular Revascularization Medicine, https://doi.org/10.1016/j.carrev.2019.11.010
N. Marchese et al. / Cardiovascular Revascularization Medicine xxx (xxxx) xxx
3
Fig. 1. Post-catheterization cerebral infarcts at diffusion-weighted magnetic resonance imaging. Caption: Pre- (A, C) and post-procedural (B, D) diffusion-weighted magnetic resonance imaging showing a new occipital (B) and parietal (D) silent cerebral infarct in two patients (see arrows).
to indicate the presence, number, size and location of new focal brain lesions consistent with CIs. In case of disagreement, the involvement of a senior neuroradiologist was planned. 2.4. Statistical analysis Data were reported as frequencies (percentages) and means ± standard deviation for categorical and continuous variables, respectively. Comparisons among groups were performed using the Pearson chisquare test and Mann–Whitney U or Kruskal–Wallis test. In order to control for possible confounding effects on the association between the three vascular access sites and the risk for CIs occurrence, the propensity score (PS) method was applied [15]. Pairwise PS logistic regression models were built to predict the probability of being in one of the three interventional arms according to the following possible confounders: age, sex, smoking habit, presence of diabetes, dyslipidemia, hypertension, prior myocardial infarction, prior PCI, renal failure, and family history of coronary artery disease. In order to verify that the data can support a comparison of treatment groups that are balanced on all covariates, the distribution of the estimated PSs should be checked for adequate overlap. This can be accomplished by creating overlapping histograms for the treatment groups.
PS quartiles' adjusted logistic regression models were used to assess the effect of the three vascular access sites on CIs occurrence risk, and results were reported as odds ratios (ORs) and 95% confidence intervals. A causal mediation analysis was performed to assess if the association between the three vascular access sites and CIs occurrence risk was mediated by the presence of procedural factors related to the outcome variable (mediators) [16]. The total effect of vascular access sites on CIs was decomposed into the average direct effect (i.e. the effect of the exposure without the action of confounders, as above, and mediator) and the causal mediation effect (i.e. the proportion of the total effect due to mediation), eventually deriving the proportion of the total mediated effect. Two-sided p-values of b0.05 were considered statistically significant. All analyses were performed using the SAS 9.4 statistical software (SAS Institute, Cary, NC, USA). 3. Results 3.1. Univariate analysis A total of 180 patients were recruited (60 patients for each of the three arms). In two patients, the selected RR approach was switched, before the insertion of the arterial sheath, to LR approach in one case
Please cite this article as: N. Marchese, M. Copetti, V. Inchingolo, et al., Cerebral infarcts after coronary angiography and percutaneous coronary intervention. A prospective p..., Cardiovascular Revascularization Medicine, https://doi.org/10.1016/j.carrev.2019.11.010
4
N. Marchese et al. / Cardiovascular Revascularization Medicine xxx (xxxx) xxx
and to TF approach in the other (based on the operator's preference), so the actual number of patients was 58, 61 and 61 for RR, LR and TF groups, respectively. After arterial sheath placement, no approach switch occurred. Baseline and procedural characteristics of patients, according to arterial access site, are listed in Table 1. No significant differences in age, sex, clinical risk factors, and procedural characteristics were found. The number of catheters used was slightly higher in the LR and TF groups, whereas arterial tortuosity was more frequently observed in the RR group, without reaching statistical significance. As per protocol, the use of heparin was lower in the TF group (p = .040). At 24-h electrocardiographic monitoring after the procedure, no major arrhythmias (atrial fibrillation, ventricular or supraventricular tachycardia) were recorded. At post-catheterization diffusion-weighted MRI, a total of 20 CIs (Fig. 1) were detected in 14 patients (7.8% of the total population; one thalamic stroke with neurological sequelae) with substantial differences among groups (RR: 15.5%, LR: 4.9%; TF: 3.3%; p = .026). Patients with and without CIs did not differ significantly in clinical and procedural parameters (Table 2), except for a higher prevalence of arterial tortuosity and massive aortic calcification in patients with post-procedural CIs. Interestingly, all patients with CIs and arterial tortuosity or calcification were in the TR group (tortuosity involving the subclavian or innominate arteries). The radiological characteristics of CIs according to the approach used are summarized in Table 3. Single CIs occurred in 10 patients and multiple CIs in 4 (29%), with involvement of the posterior circulation in 11 (79%). When the side (left or right) of embolic pattern could be identified, this was coherent with the arterial access route in all but one patient. 3.2. Propensity score analysis The PS method was used to account for potential confounding effects on the association between arterial approach and CIs occurrence risk. PS adjusted analyses confirmed the association of the RR approach with a significant higher rate of new CIs compared with the TF (OR estimate from logistic regression adjusted for PS quartiles: 0.158; 95% confidence interval: 0.031 to 0.814; p = .027) and LR approaches (OR
Table 2 Clinical and procedural characteristics of study patients with and without CIs.
Clinical characteristics Age (years) Male gender Hypertension Diabetes mellitus Family history of CAD Smokers (%) Hypercholesterolemia (%) Heparin (IU) Aspirin use P2Y12 receptor inhibitors IIb/IIIa inhibitors Procedural characteristics Total procedure time (min) Fluoroscopy time (sec) Contrast used (ml) Catheters used (n) Accidental aortic valve crossing Ad-hoc PCI Presence of arterial tortuosity Massive aortic calcification
CIs− (n = 166)
CIs+ (n = 14)
p
67.9 ± 8.4 127 (76.5) 145 (87.3) 54 (32.5) 66 (39.8) 17 (10.2) 142 (85.5) 3439 ± 2051 160 (96.4) 50 (30.1) 8 (4.8)
72.0 ± 8.0 12 (85.7) 12 (85.7) 6 (42.9) 7 (50%) 2 (14.3) 12 (85.7) 4143 ± 1574 14 (100%) 5 (35.7) 1 (7.1)
0.115 0.430 0.860 0.431 0.454 0.636 0.986 0.225 0.469 0.663 0.702
25.0 (15.0–45.5) 247.5 (98.75–600.0) 120 (50.0–200.0) 2.8 ± 0.9 39 (23.5)
24.5 (13.0–50.0) 370 (107.5–675.0) 88 (48.5–207.0) 3.1 ± 1.2 4 (28.6)
0.932 0.535
83 (50) 9 (5.4) 8 (4.8)
7 (50) 7 (50) 4 (28.6)
0.958 b0.001 0.001
0.940 0.134 0.669
Values are mean ± standard deviation, median (interquartile range), or n (%). CAD = coronary artery disease, CIs = cerebral infarcts, PCI = percutaneous coronary intervention.
Table 3 Characteristics and distribution of post-procedural cerebral infarcts according to the access site.
Patients (n) Lesions Multiple lesions Vascular territory Right carotid (n) Left carotid (n) Vertebrobasilar (n) Right vertebral Left vertebral
Right radial
Left radial
Femoral
Total
9 13 3
3 5 1
2 2 0
14 20 4
3 1 7 2 –
0 1 3 – –
1
4 2 11
1 – –
estimate from logistic regression adjusted for PS quartiles: 0.266; 95% confidence interval: 0.066 to 1.080; p = .064). In a secondary analysis, PS method was used to compare RR vs non-RR approach (TF + LR). Post-procedural CIs were again more frequent in the RR group (OR estimate from logistic regression adjusted for PS quartiles: 0.170; 95% confidence interval: 0.050 to 0.574; p = .004). 3.3. Causal mediation analysis The association between RR approach and new CIs, compared with non-RR approach, was partially mediated (23%) by the presence of arterial tortuosity, but not by aortic calcification (1%). 4. Discussion Among periprocedural complications of coronary angiography and PCI, the neurological ones are the most dreaded. Symptomatic cerebral ischemia following diagnostic or interventional coronary angiography is a well-known serious—though rare—complication [1]. However, also radiologically documented CIs without neurological symptoms seem to be relatively frequent [4–13]. Post-left heart catheterization CIs may result from gaseous emboli or thrombotic clots on guidewires or catheters, but are mainly due to atherosclerotic fragments as a direct consequence of catheter/wire manipulation in the aortic wall. At least three important considerations can be drawn from our data. First, consistently with previous studies, a non-negligible incidence (7.8%) of CIs was observed after cardiac catheterization procedures, although most patients had no neurological symptoms. The relatively wide difference in silent CIs occurrence among studies is probably related to several factors (study design, catheterization technique, imaging modality and its time interval from the index procedure, clinical setting, and patient characteristics). In this respect, a higher incidence of CIs was reported in patients with acute coronary syndrome (35%) [9], after procedures with routine aortic valve crossing (14–22%) [4,12], or with internal mammary artery angiography (19%) [10]. Since our objective was to compare the incidence of CIs using the three most common vascular accesses, the per-protocol exclusion of patients with the above characteristics, along with the adoption of standardized catheters and guidewires, may have reduced the risk for potential confounding factors resulting in a lower observation rate. Further, the radiological characteristics and vascular distribution of these CIs are highly compatible with embolic damage; in addition, CIs location is in accordance with the arterial site, suggesting that the latter plays a role in CIs occurrence. Second, in contrast to previous studies, new CIs were not associated with risk factors [9,13], ad-hoc PCI [13], or procedure/fluoroscopy time [5,6], but rather with the presence of arterial tortuosity and significant aortic calcification. Severe tortuosity can pose technical challenges leading to failure of the transradial approach, and sometimes operators are compelled to complex catheter manipulations. In the attempts to overcome a difficult arterial anatomy, catheters may scratch the arterial wall causing vessel damage or mobilization of atherosclerotic debris, thereby
Please cite this article as: N. Marchese, M. Copetti, V. Inchingolo, et al., Cerebral infarcts after coronary angiography and percutaneous coronary intervention. A prospective p..., Cardiovascular Revascularization Medicine, https://doi.org/10.1016/j.carrev.2019.11.010
N. Marchese et al. / Cardiovascular Revascularization Medicine xxx (xxxx) xxx
promoting embolic brain damage. Accordingly, the presence of severe aortic calcification is a marker of high atherosclerotic burden in an anatomic territory involved with the passage of wires and catheters. In our study, the presence of subclavian or innominate arterial tortuosity/loop and/or severe aortic calcification was associated with CIs in 7 of 12 patients of the RR and LR groups (58.3%), suggesting a role in the development of embolic brain injuries. This is the first study to investigate the impact of vessel tortuosity/aortic calcification on post-procedural ischemic brain damage and to suppose their contributing role in post-procedural CIs. We can also speculate that the procedural factors found to be associated with CIs (number of catheters, procedure and fluoroscopy times) in other studies may be, at least in part, related to the presence of challenging arterial anatomy. The third and most important consideration is that the RR approach was found to be independently associated with post-procedural CIs compared with the LR (though without reaching statistical significance) and TF approaches. In addition, the only case of clinically manifest brain ischemia (thalamic stroke with neurological sequelae) occurred in the RR group. This finding was confirmed when comparing the RR vs nonRR (LR + TF) group. The rationale for considering the LR and TF approaches as a whole group lies in the similar straight and direct access to the ascending aorta, with quite similar catheter movement to engage the coronary ostia. This association is only partially explained by the presence of arterial tortuosity but not by the presence of aortic calcification, as shown by causal mediation analysis. This suggests that the RR approach is associated per se with a higher risk of cerebral embolic damage. To the best of our knowledge, this is a new concept, since the majority of previous studies did not focus on the comparison of different arterial routes. Studies evaluating access site in relation to the risk of clinically manifest neurological events reported a lack of association between radial access and ischemic stroke [17]. In the large MATRIX trial, involving patients with acute coronary syndrome who underwent invasive management, no differences were found between radial (with no differentiation between right and left side) and femoral approaches in terms of ischemic stroke occurrence [18]. In addition, there are very limited data on the association of the RR vs LR access site with cerebral ischemic complications. A meta-analysis did not find statistically significant differences in clinically reported stroke/TIA between the two approaches [19] but, more recently, in a large British database, the LR access was associated with a significant decrease in PCI-related in-hospital stroke [20] compared with the RR approach. As regards silent and radiologically documented post-procedural CIs, the only per-protocol comparison of radial vs femoral approach is provided by the randomized study of Hamon and colleagues [12], which enrolled patients with severe aortic stenosis referred for cardiac catheterization, also including aortic valve crossing. The risk of brain embolic damage did not diverged significantly between the two approaches. Other observational studies with systematic post-procedural MRI did not show any differences between radial and femoral procedures (5,9–11). In contrast, at transcranial Doppler, RR access was found to generate more particulate emboli than femoral [21] and LR access [22]. Several reasons may account for our results. First, steering catheters and wires can be more difficult in the setting of tortuosity of the subclavian and brachiocephalic trunk, with the latter in direct communication with the right carotid artery. In ours as in other studies [23,24], a higher prevalence of arterial tortuosity was observed in the RR group (13.8%) compared with the LR (9.8%) and TF (3.3%) groups. Regardless of the presence of tortuous anatomy or loops, during RR angiography the catheters are forced to pass in a more acute angle between the brachiocephalic trunk and ascending aorta compared with the LR and TF approaches, thereby increasing the risk of plaque mobilization. Second, the Judkins preshaped catheters, originally designed for a femoral approach, may require sometimes-difficult manipulations to reach the coronary ostia using the RR approach. Conversely, the LR approach shares with the TF approach the same orientation in relation to the coronary anatomic plane. An indirect confirmation of this hypothesis
5
comes from transcranial studies where a higher number of microembolic signals during catheter manipulation was demonstrated using the RR approach [21,22]. In the present study, the location of CIs confirms the concept, also based on experimental data, that posterior circulation may be more prone to embolic occlusion [5,25,26]. In addition, due to the less dense posterior vascularization of the brain, a given embolic load may result more commonly in CIs. Another possible explanation is that radial approach itself may promote posterior CIs, considering that vertebral artery arises from subclavian artery. While the clinical impact of stroke is evident, the clinical implications and long-term relevance of silent CIs remain a matter of debate. Further, the increasing number of neuroimaging exams, with consequent incidental finding of CIs, together with the improvements in neuroimaging techniques, has amplified the dimension of the problem. Observational population studies found a correlation between asymptomatic CIs and the risk of stroke [27,28]. In a meta-analysis including 13 studies, the occurrence of silent brain infarcts was associated with a 2-fold increased risk of future stroke [29]. It is plausible that silent CIs may represent a risk marker of other diseases (i.e. atherosclerosis, cardiac arrhythmias, cardiomyopathies) that, in turn, are causally associated with stroke. On the contrary, the correlation with cognitive impairment, dementia and Alzheimer's disease [28,30] may recognize CIs as direct contributors since, as demonstrated by MRI-based studies, cognitive impairment is also triggered by cumulative cerebral damage caused by ischemic injuries [31]. In patients with major depression, the occurrence of CIs was also positively associated with symptom severity [32], higher hospitalization rates, and a higher risk of developing psychiatric and neurological disorders [33]. The nature of these studies does not allow us to establish whether CIs are ‘innocent’ bystander, or rather are linked in a cause-and-effect relationship to these diseases. Since silent CIs are relatively common after cardiac procedures (e.g. transcatheter aortic valve implantation, coronary artery bypass grafting, electrophysiologic procedures), the debate on their prognostic implications has gained more attention in parallel with the worldwide rapidly increasing number of procedures. Although the acute phenomenon has been extensively studied, further studies focused on late implications of post-procedural CIs are warranted. 4.1. Strengths and limitations This is the first study focusing on the role of the three most commonly used arterial access sites for coronary angiography in post-procedural occurrence of CIs. The systematic use of pre and post-procedural MRI, the relatively restrictive entry criteria, and a standardized procedural protocol helped us to minimize potential confounding factors and to analyze very consistent data. In addition, the collection of different data from similar studies (i.e. the presence of arterial tortuosity and aortic calcifications) gave us the opportunity to observe a correlation between these relatively frequent characteristics and CIs. However, several study limitations should be acknowledged. The relative low frequency of events (7.8%), also compared with similar studies, the nonrandomized nature of data, even in the presence of a PS analysis, and the non-consecutive recruitment of patients, may have limited the reliability of our results. Unfortunately, the small number of patients limited the ability of our study to strongly correlate the presence of tortuosity to CIs occurrence. The choice to administer low-dose heparin, to prevent radial artery occlusion, at the end of the diagnostic procedure, together with the routine use of 300 mm wire, may have limited the generalizability of our results. In the absence of data on heparin dosage able to protect against occurrence of CIs after cardiac procedures, it can be speculated that low-dose may be less effective than high-dose. However, the comparison between RR and LR is not affected by our protocol, whereas the use of more heparin in the RR and LR groups favors radial vs femoral approach. The lack of radiological and clinical followup does not allow for a comprehensive evaluation of long-term
Please cite this article as: N. Marchese, M. Copetti, V. Inchingolo, et al., Cerebral infarcts after coronary angiography and percutaneous coronary intervention. A prospective p..., Cardiovascular Revascularization Medicine, https://doi.org/10.1016/j.carrev.2019.11.010
6
N. Marchese et al. / Cardiovascular Revascularization Medicine xxx (xxxx) xxx
consequences of CIs (i.e. cognitive decline, clinically evident cerebral ischemia). A systematic radiological study of the subclavian, carotid and vertebral arteries could have helped us to identify additional embolic sources from atherosclerotic plaques during the exams.
5. Conclusions Our study confirms that CIs are not infrequent after elective coronary angiography and PCI, and suggests that use of the RR access is independently associated with a higher rate of new CIs compared with the LR and TF approaches. Additionally, the association of the RR approach with post-procedural CIs is only partially explained by the presence of arterial tortuosity. Further studies are warranted to determine the late clinical implications of silent CIs.
References [1] Kwok CS, Kontopantelis E, Myint PK, Zaman A, Berry C, Keavney B, et al. Stroke following percutaneous coronary intervention: type-specific incidence, outcomes and determinants seen by the British Cardiovascular Intervention Society 2007–12. Eur Heart J 2015;36:1618–28. [2] Aggarwal A, Dai D, Rumsfeld JS, Klein LW, Roe MT. American College of Cardiology National Cardiovascular Data Registry Incidence and predictors of stroke associated with percutaneous coronary intervention. Am J Cardiol 2009;104:349–53. [3] Hoffman SJ, Routledge HC, Lennon RJ, Mustafa MZ, Rihal CS, Gersh BJ, et al. Procedural factors associated with percutaneous coronary intervention-related ischemic stroke. J Am Coll Cardiol Intv 2012;5:200–6. [4] Omran H, Schmidt H, Hackenbroch M, Illien S, Bernhardt P, von der Recke G, et al. Silent and apparent cerebral embolism after retrograde catheterisation of the aortic valve in valvular stenosis: a prospective. randomised study Lancet 2003;361: 1241–6. [5] Lund C, Nes RB, Ugelstad TP, Due-Tønnessen P, Andersen R, Hol PK, et al. Cerebral emboli during left heart catheterization may cause acute brain injury. Eur Heart J 2005;26:1269–75. [6] Büsing KA, Schulte-Sasse C, Flüchter S, Süselbeck T, Haase KK, Neff W, et al. Cerebral infarction: incidence and risk factors after diagnostic and interventional cardiac catheterization - prospective evaluation at diffusion-weighted MR imaging. Radiology 2005;235:177–83. [7] Hamon M, Gomes S, Oppenheim C, Morello R, Sabatier R, Lognoné T, et al. Cerebral microembolism during cardiac catheterization and risk of acute brain injury: a prospective diffusion-weighted magnetic resonance imaging study. Stroke 2006;37: 2035–8. [8] Hamon M, Gomes S, Clergeau MR, Fradin S, Morello R, Hamon M. Risk of acute brain injury related to cerebral microembolism during cardiac catheterization performed by right upper limb arterial access. Stroke 2007;38:2176–9. [9] Murai M, Hazui H, Sugie A, Hoshiga M, Negoro N, Muraoka H, et al. Asymptomatic acute ischemic stroke after primary percutaneous coronary intervention in patients with acute coronary syndrome might be caused mainly by manipulating catheters or devices in the ascending aorta, regardless of the approach to the coronary artery. Circ J 2008;72:51–5. [10] Kim IC, Hur SH, Park NH, Jun DH, Cho YK, Nam CW, et al. Incidence and predictors of silent embolic cerebral infarction following diagnostic coronary angiography. Int J Cardiol 2011;148:179–82. [11] Kim BJ, Lee SW, Park SW, Kang DW, Kim JS, Kwon SU. Insufficient platelet inhibition is related to silent embolic cerebral infarctions after coronary angiography. Stroke 2012;43:727–32.
[12] Hamon M, Lipiecki J, Carrié D, Burzotta F, Durel N, Coutance G, et al. Silent cerebral infarcts after cardiac catheterization: a randomized comparison of radial and femoral approaches. Am Heart J 2012;164:449–54. [13] Deveci OS, Celik AI, Ikikardes F, Ozmen C, Caglıyan CE, Deniz A, et al. The incidence and the risk factors of silent embolic cerebral infarction after coronary angiography and percutaneous coronary interventions. Angiology 2016;67:433–7. [14] Sacco RL, Kasner SE, Broderick JP, Caplan LR, Connors JJ, Culebras A, et al. An updated definition of stroke for the 21st century: a statement for healthcare professionals from the American Heart Association/American Stroke Association. Stroke 2013; 44:2064–89. [15] Yanovitzky I, Zanutto E, Hornik R. Estimating causal effects of public health education campaigns using propensity score methodology. Eval Progr Plann 2005;28: 209–20. [16] Imai K, Keele L, Tingley D. A general approach to causal mediation analysis. Psychol Methods 2010;15:309–34. [17] Sirker A, Kwok CS, Kotronias R, Bagur R, Bertrand O, Butler R, et al. Influence of access site choice for cardiac catheterization on risk of adverse neurological events: a systematic review and meta-analysis. Am Heart J 2016;181:107–19. [18] Valgimigli M, Gagnor A, Calabró P, Frigoli E, Leonardi S, Zaro T, et al. Radial versus femoral access in patients with acute coronary syndromes undergoing invasive management: a randomised multicentre trial. Lancet 2015;385:2465–76. [19] Shah RM, Patel D, Abbate A, Cowley MJ, Jovin IS. Comparison of transradial coronary procedures via right radial versus left radial artery approach: a meta-analysis. Catheter Cardiovasc Interv 2016;88:1027–33. [20] Rashid M, Lawson C, Potts J, Kontopantelis E, Kwok CS, Bertrand OF, et al. Incidence, determinants, and outcomes of left and right radial access use in patients undergoing percutaneous coronary intervention in the United-Kingdom: a National Perspective using the BCIS dataset. JACC Cardiovasc Interv 2018;11:1021–33. [21] Jurga J, Nyman J, Tornvall P, Mannila MN, Svenarud P, van der Linden J, et al. Cerebral microembolism during coronary angiography: a randomized comparison between femoral and radial arterial access. Stroke 2011;42:1475–7. [22] Pacchioni A, Mugnolo A, Penzo C, Nikas D, Saccà S, Agostoni P, et al. Cerebral microembolism during transradial coronary angiography: comparison between single and double catheter strategy. Int J Cardiol 2014;170:438–9. [23] Sciahbasi A, Romagnoli E, Burzotta F, Trani C, Sarandrea A, Summaria F, et al. Transradial approach (left vs right) and procedural times during percutaneous coronary procedures: TALENT study. Am Heart J 2011;161:172–9. [24] Freixa X, Trilla M, Feldman M, Jiménez M, Betriu A, Masotti M. Right versus left transradial approach for coronary catheterization in octogenarian patients. Catheter Cardiovasc Interv 2012;80:267–72. [25] Barbut D, Grassineau D, Lis E, Heier L, Hartman GS, Isom OW. Posterior distribution of infarcts in strokes related to cardiac operations. Ann Thorac Surg 1998;65:1656–9. [26] Segal AZ, Abernethy WB, Palacios IF, BeLue R, Rordorf G. Stroke as a complication of cardiac catheterization: risk factors and clinical features. Neurology 2001;56:975–7. [27] Vermeer SE, Den Heijer T, Koudstaal PJ, Oudkerk M, Hofman A, Breteler MM. Incidence and risk factors of silent brain infarcts in the population-based Rotterdam scan study. Stroke 2003;34:392–6. [28] Debette S, Beiser A, DeCarli C, Au R, Himali JJ, Kelly-Hayes M, et al. Association of MRI markers of vascular brain injury with incident stroke, mild cognitive impairment, dementia, and mortality: the Framingham offspring study. Stroke 2010;41:600–6. [29] Gupta A, Giambrone AE, Gialdini G, Finn C, Delgado D, Gutierrez J, et al. Silent brain infarction and risk of future stroke: a systematic review and meta-analysis. Stroke 2016;47:719–25. [30] Vermeer SE, Prins ND, den Heijer T, Hofman A, Koudstaal PJ, Breteler MM. Silent brain infarcts and the risk of dementia and cognitive decline. N Engl J Med 2003; 348:1215–22. [31] Choi SH, Na DL, Chung CS, Lee KH, Na DG, Adair JC. Diffusion-weighted MRI in vascular dementia. Neurology 2000;54:83–9. [32] Fujikawa T, Yamawaki S, Touhouda Y. Background factors and clinical symptoms of major depression with silent cerebral infarction. Stroke 1994;25:798–801. [33] Yanai I, Fujikawa T, Horiguchi J, Yamawaki S, Touhouda Y. The 3-year course and outcome of patients with major depression and silent cerebral infarction. J Affect Disord 1998;47:25–30.
Please cite this article as: N. Marchese, M. Copetti, V. Inchingolo, et al., Cerebral infarcts after coronary angiography and percutaneous coronary intervention. A prospective p..., Cardiovascular Revascularization Medicine, https://doi.org/10.1016/j.carrev.2019.11.010
Contents lists available at ScienceDirect
Cardiovascular Revascularization Medicine
Cerebral infarcts after coronary angiography and percutaneous coronary intervention. A prospective propensity score adjusted comparison of right radial, left radial and femoral approaches☆,☆☆,★,★★ Nicola Marchese a,⁎, Massimiliano Copetti b, Vincenzo Inchingolo c, Teresa Popolizio d, Andrea Fontana b, Annalisa Simeone c, Carlo Vigna a a
Unit of Cardiology, Fondazione IRCCS Casa Sollievo della Sofferenza, San Giovanni Rotondo, Italy Unit of Biostatistics, Fondazione IRCCS Casa Sollievo della Sofferenza, San Giovanni Rotondo, Italy c Unit of Neurology, Fondazione IRCCS Casa Sollievo della Sofferenza, San Giovanni Rotondo, Italy d Unit of Radiology, Fondazione IRCCS Casa Sollievo della Sofferenza, San Giovanni Rotondo, Italy b
a r t i c l e
i n f o
Article history: Received 30 August 2019 Received in revised form 7 November 2019 Accepted 11 November 2019 Available online xxxx Keywords: Radial access Femoral access Cerebral infarct
a b s t r a c t Background: New cerebral infarcts (CIs) detected at magnetic resonance imaging (MRI) are reported after cardiac procedures. Clinical and procedural aspects are implicated as potential causal factors. The aim of this study was to evaluate the incidence of new CIs after coronary angiography and percutaneous coronary intervention according to the arterial access site. Methods: 180 patients undergoing elective coronary angiography were studied with cerebral MRI the day before and the day after the procedure. Unadjusted and propensity score (PS) analyses were performed comparing the occurrence of CIs in right radial (RR), left radial (LR) and transfemoral (TF) access groups. Results: New CIs were observed in 14 patients (7.8% of the total sample, one with neurological sequelae). CIs were detected in 15.5% vs 4.9% vs 3.3% of RR, LR and TF groups, respectively (p = .026). In PS adjusted analyses, the RR approach was associated with more CIs compared with the TF approach (odds ratio [OR] estimate from logistic regression adjusted by PS quartiles: 0.158; 95% confidence interval: 0.031 to 0.814; p = .027) and the LR approach (OR: 0.266; 95% confidence interval: 0.066 to 1.080; p = .064). In a secondary analysis, a comparison of RR vs non-RR approach (TF + LR) was performed, showing that post-procedural CIs were more frequent in the RR group (OR: 0.170; 95% confidence interval: 0.050 to 0.574; p = .004). Conclusions: Our study suggests that the RR approach may be associated with a higher rate of new CIs after coronary angiography compared with LR and TF approaches. © 2019 Elsevier Inc. All rights reserved.
1. Introduction
Abbreviations: CIs, cerebral infarcts; MRI, magnetic resonance imaging; PCI, percutaneous coronary intervention; RR, right radial; LR, left radial; TF, transfemoral; PS, propensity score. ☆ All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation. ☆☆ Declaration of competing interest: None. ★ This research did not receive any specific grant from funding agencies in the public, commercial or not-for-profit sectors. ★★ Summary for the annotated table of contents: Radiologically documented cerebral infarcts are frequently reported after left heart procedures, have important clinical implications and vascular access site may play an important role. The present study suggests that right radial approach may be associated with a higher risk of post-procedural cerebral infarcts compared with the left radial and femoral approaches and this association is only partially explained by the presence of arterial tortuosity. ⁎ Corresponding author at: Unit of Cardiology, Casa Sollievo della Sofferenza Hospital IRCCS, Viale Cappuccini 1, 71013 San Giovanni Rotondo, FG, Italy. E-mail address: [email protected] (N. Marchese).
Ischemic stroke after cardiac catheterization is a rare but potentially severe complication, with rates ranging from 0.1% to 0.6% [1]. Stroke occurrence is associated with high in-hospital mortality and great disability in survivors [1]. Clinical (older age, male gender, known cerebrovascular disease, acute coronary syndrome, major cardiovascular risk factors, and coronary atherosclerotic burden) and procedural factors (systematic aortic valve crossing, use of intra-aortic balloon pump, number of catheters used, large guide caliber, and high contrast volumes) have been identified as potential risk factors for post-procedural stroke [2,3]. Fortunately, stroke represents the smallest, albeit dangerous, part of the more multifaceted and recurrent phenomenon of embolic brain injury after cardiac procedures, which may occur subclinically or even be totally silent. Substantially higher rates of clinically silent acute cerebral infarcts (CIs) have been identified at diffusion-weighted magnetic resonance
https://doi.org/10.1016/j.carrev.2019.11.010 1553-8389/© 2019 Elsevier Inc. All rights reserved.
Please cite this article as: N. Marchese, M. Copetti, V. Inchingolo, et al., Cerebral infarcts after coronary angiography and percutaneous coronary intervention. A prospective p..., Cardiovascular Revascularization Medicine, https://doi.org/10.1016/j.carrev.2019.11.010
2
N. Marchese et al. / Cardiovascular Revascularization Medicine xxx (xxxx) xxx
imaging (MRI) after left heart catheterization, with wide variability across studies (2% to 35%) [4–13]. Such CIs have been associated with cognitive decline, dementia and depression, and are considered precursors of symptomatic stroke [14]. Since one supposed mechanism of post-procedural acute brain injury is mobilization of atherosclerotic plaques from the aorta, arterial access route may significantly impact the risk of new CIs. Our objective was to evaluate the prevalence of new CIs, either clinically manifest or silent, after elective coronary angiography or percutaneous coronary intervention (PCI), and to compare the incidence of post-procedural CIs at diffusion-weighted MRI using the three most common vascular access sites for left heart catheterization (right radial [RR] vs left radial [LR] vs transfemoral [TF]). 2. Methods 2.1. Patient population In this single-center prospective study, 180 patients with suspected coronary artery disease scheduled for elective coronary angiography were enrolled. Patients with recent acute coronary syndrome, need for aortic valve crossing, coronary artery bypass grafts, preexisting neurological deficit, history of ischemic or hemorrhagic stroke, ongoing atrial fibrillation, and any contraindication for MRI were excluded. Owing to the limited availability of diffusion-weighted MRI for research purposes in our institution (3 days/week), all eligible patients who gave their written consent to participate were recruited on selected days in the study. The local ethics committee approved the study. 2.2. Coronary angiography Coronary angiography was performed by cardiologists with good expertise in radial access (N300 radial exams/year) using the Seldinger technique with 6F catheters via the RR, LR and TF approach, in a 1:1:1 ratio for the three arms. For RR and LR approaches, 6F 25 cm radial sheaths were used, whereas 6F 10 cm femoral sheaths were used for the TF approach (Radiofocus Introducer II, Terumo Europe, Leuven, Belgium). Unfractionated heparin (2500 IU) was administered intravenously at the end of the diagnostic procedure only to RR and LR patients. If ad hoc PCI was necessary, a total of 5000 IU were
administered to all patients before the procedure and then according to activated clotting time. Coronary angiography was performed with 6F diagnostic catheters using standard choices: 3.5 mm left Judkins and 5.0 mm right Judkins for the RR group; 4.0 mm left Judkins and 4.0 mm right Judkins for the LR and TF groups; with a standard 0.035″, 260 cm exchange guidewire (catheters and exchange wire by Cordis Corporation, Miami, FL, USA). The use of a long (260 cm) exchange guidewire was adopted to restrict wire passages in the aortic arch in all patients. Iohexol, a non-ionic, low-osmolar contrast agent (Omnipaque 350 mg, GE Healthcare Inc., Princeton, NJ, USA) was used. Contrast amount, number of catheters used, procedure and fluoroscopy time, accidental aortic valve crossing with the catheter or exchange guidewire, need for hydrophilic guidewire, presence of vascular tortuosity, and significant aortic calcification were collected. After coronary angiography or intervention, patients were carefully evaluated for the occurrence of neurological and arrhythmic complications with continuous electrocardiographic monitoring for at least 24 h. 2.3. Magnetic resonance imaging All patients provided written informed consent to MRI, which was performed the same day as coronary angiography (baseline MRI) and within 48 h after the cardiac procedure (post-cath MRI) with a 1.5 T system (Siemens MAGNETOM Aera, Erlangen, Germany). The imaging protocol of baseline MRI included diffusion-weighted single-shot echoplanar imaging (TR/TE: 5300/97 ms; NEX: 1; slice thickness 5 mm; Matrix 172\172; diffusion gradient b values of 0 and 1000 S/mm2) and T2 fluid attenuated inversion recovery (FLAIR; TR/TE: 8500/82 ms; NEX: 2; TI: 2240 ms; slice thickness 4 mm; Matrix 320\320). The imaging protocol of post-cath MRI included only diffusion-weighted single-shot echo-planar imaging (using the same parameters as described before). In case of inability to perform baseline evaluation, a complete MRI study was obtained after the cardiac procedure, which included, besides diffusion-weighted and FLAIR, also T1-weighted spin-echo (SE) images (TR/TE: 500/9.9 ms; NEX: 1; slice thickness 5 mm; Matrix 320\320), T2weighted fast spin-echo (FSE) images (TR/TE: 4700/87 ms; NEX: 1; slice thickness 5 mm; Matrix 320\320), and gradient recalled echo (GRE) T2weighted imaging (T2WI) (TR/TE: 950/25 ms; FA: 20; NEX 1; Matrix 256\256). Two blinded experts in neuroradiological images (a neuroradiologist and a neurologist), unaware of the study protocol, were asked
Table 1 Clinical and procedural characteristics of study patients according to arterial approach.
Age (years) Male gender Hypertension Diabetes mellitus Family history of CAD Smokers Hypercholesterolemia Renal failure Prior PCI Heparin (IU) Aspirin use Statin use P2Y12 receptor inhibitors IIb/IIIa inhibitors Total procedure time (min) Fluoroscopy time (s) Contrast used (ml) Catheters used Accidental aortic valve crossing Ad-hoc PCI Arterial tortuosity Significant aortic calcification
Right radial (n = 58)
Left radial (n = 61)
Femoral (n = 61)
p
67.9 ± 9.5 45 (77.6) 51 (87.9) 19 (32.8) 20 (34.5) 6 (10.3) 50 (86.2) 3 (5.2) 13 (22.4) 2500 (2500–5000) 55 (94.8) 49 (84.5) 15 (25.9) 4 (6.9) 21.5 (15.0–41.0) 215 (100–440) 92 (50–170) 2.6 ± 0.8 13 (22.4) 27 (46.6) 8 (13.8) 4 (6.9)
68.0 ± 8.6 47 (77.0) 52 (85.2) 20 (32.8) 26 (42.6) 7 (11.5) 50 (82.0) 6 (9.8) 16 (26.2) 5000 (2500–5000) 60 (98.4) 52 (85.2) 21 (34.4) 2 (3.3) 25.0 (15.0–47.0) 270 (90–600) 128 (50–220) 2.9 ± 0.9 18 (29.5) 32 (52.5) 6 (9.8) 6 (9.8)
68.6 ± 7.0 47 (77.0) 54 (88.5) 21 (34.4) 27 (44.3) 6 (9.8) 54 (88.5) 7 (11.5) 13 (21.3) 5000 (0–5000) 59 (96.7) 52 (85.2) 19 (31.1) 3 (4.9) 30.0 (20.0–45.0) 240 (120–660) 126 (50–200) 2.9 ± 1.1 12 (19.7) 32 (52.5) 2 (3.3) 2 (3.3)
0.929 0.997 0.847 0.975 0.511 0.956 0.580 0.458 0.797 0.040 0.562 0.991 0.594 0.663 0.215 0.395 0.270 0.185 0.422 0.760 0.125 0.347
Values are mean ± standard deviation, median (interquartile range), or n (%). CAD = coronary artery disease, PCI = percutaneous coronary intervention.
Please cite this article as: N. Marchese, M. Copetti, V. Inchingolo, et al., Cerebral infarcts after coronary angiography and percutaneous coronary intervention. A prospective p..., Cardiovascular Revascularization Medicine, https://doi.org/10.1016/j.carrev.2019.11.010
N. Marchese et al. / Cardiovascular Revascularization Medicine xxx (xxxx) xxx
3
Fig. 1. Post-catheterization cerebral infarcts at diffusion-weighted magnetic resonance imaging. Caption: Pre- (A, C) and post-procedural (B, D) diffusion-weighted magnetic resonance imaging showing a new occipital (B) and parietal (D) silent cerebral infarct in two patients (see arrows).
to indicate the presence, number, size and location of new focal brain lesions consistent with CIs. In case of disagreement, the involvement of a senior neuroradiologist was planned. 2.4. Statistical analysis Data were reported as frequencies (percentages) and means ± standard deviation for categorical and continuous variables, respectively. Comparisons among groups were performed using the Pearson chisquare test and Mann–Whitney U or Kruskal–Wallis test. In order to control for possible confounding effects on the association between the three vascular access sites and the risk for CIs occurrence, the propensity score (PS) method was applied [15]. Pairwise PS logistic regression models were built to predict the probability of being in one of the three interventional arms according to the following possible confounders: age, sex, smoking habit, presence of diabetes, dyslipidemia, hypertension, prior myocardial infarction, prior PCI, renal failure, and family history of coronary artery disease. In order to verify that the data can support a comparison of treatment groups that are balanced on all covariates, the distribution of the estimated PSs should be checked for adequate overlap. This can be accomplished by creating overlapping histograms for the treatment groups.
PS quartiles' adjusted logistic regression models were used to assess the effect of the three vascular access sites on CIs occurrence risk, and results were reported as odds ratios (ORs) and 95% confidence intervals. A causal mediation analysis was performed to assess if the association between the three vascular access sites and CIs occurrence risk was mediated by the presence of procedural factors related to the outcome variable (mediators) [16]. The total effect of vascular access sites on CIs was decomposed into the average direct effect (i.e. the effect of the exposure without the action of confounders, as above, and mediator) and the causal mediation effect (i.e. the proportion of the total effect due to mediation), eventually deriving the proportion of the total mediated effect. Two-sided p-values of b0.05 were considered statistically significant. All analyses were performed using the SAS 9.4 statistical software (SAS Institute, Cary, NC, USA). 3. Results 3.1. Univariate analysis A total of 180 patients were recruited (60 patients for each of the three arms). In two patients, the selected RR approach was switched, before the insertion of the arterial sheath, to LR approach in one case
Please cite this article as: N. Marchese, M. Copetti, V. Inchingolo, et al., Cerebral infarcts after coronary angiography and percutaneous coronary intervention. A prospective p..., Cardiovascular Revascularization Medicine, https://doi.org/10.1016/j.carrev.2019.11.010
4
N. Marchese et al. / Cardiovascular Revascularization Medicine xxx (xxxx) xxx
and to TF approach in the other (based on the operator's preference), so the actual number of patients was 58, 61 and 61 for RR, LR and TF groups, respectively. After arterial sheath placement, no approach switch occurred. Baseline and procedural characteristics of patients, according to arterial access site, are listed in Table 1. No significant differences in age, sex, clinical risk factors, and procedural characteristics were found. The number of catheters used was slightly higher in the LR and TF groups, whereas arterial tortuosity was more frequently observed in the RR group, without reaching statistical significance. As per protocol, the use of heparin was lower in the TF group (p = .040). At 24-h electrocardiographic monitoring after the procedure, no major arrhythmias (atrial fibrillation, ventricular or supraventricular tachycardia) were recorded. At post-catheterization diffusion-weighted MRI, a total of 20 CIs (Fig. 1) were detected in 14 patients (7.8% of the total population; one thalamic stroke with neurological sequelae) with substantial differences among groups (RR: 15.5%, LR: 4.9%; TF: 3.3%; p = .026). Patients with and without CIs did not differ significantly in clinical and procedural parameters (Table 2), except for a higher prevalence of arterial tortuosity and massive aortic calcification in patients with post-procedural CIs. Interestingly, all patients with CIs and arterial tortuosity or calcification were in the TR group (tortuosity involving the subclavian or innominate arteries). The radiological characteristics of CIs according to the approach used are summarized in Table 3. Single CIs occurred in 10 patients and multiple CIs in 4 (29%), with involvement of the posterior circulation in 11 (79%). When the side (left or right) of embolic pattern could be identified, this was coherent with the arterial access route in all but one patient. 3.2. Propensity score analysis The PS method was used to account for potential confounding effects on the association between arterial approach and CIs occurrence risk. PS adjusted analyses confirmed the association of the RR approach with a significant higher rate of new CIs compared with the TF (OR estimate from logistic regression adjusted for PS quartiles: 0.158; 95% confidence interval: 0.031 to 0.814; p = .027) and LR approaches (OR
Table 2 Clinical and procedural characteristics of study patients with and without CIs.
Clinical characteristics Age (years) Male gender Hypertension Diabetes mellitus Family history of CAD Smokers (%) Hypercholesterolemia (%) Heparin (IU) Aspirin use P2Y12 receptor inhibitors IIb/IIIa inhibitors Procedural characteristics Total procedure time (min) Fluoroscopy time (sec) Contrast used (ml) Catheters used (n) Accidental aortic valve crossing Ad-hoc PCI Presence of arterial tortuosity Massive aortic calcification
CIs− (n = 166)
CIs+ (n = 14)
p
67.9 ± 8.4 127 (76.5) 145 (87.3) 54 (32.5) 66 (39.8) 17 (10.2) 142 (85.5) 3439 ± 2051 160 (96.4) 50 (30.1) 8 (4.8)
72.0 ± 8.0 12 (85.7) 12 (85.7) 6 (42.9) 7 (50%) 2 (14.3) 12 (85.7) 4143 ± 1574 14 (100%) 5 (35.7) 1 (7.1)
0.115 0.430 0.860 0.431 0.454 0.636 0.986 0.225 0.469 0.663 0.702
25.0 (15.0–45.5) 247.5 (98.75–600.0) 120 (50.0–200.0) 2.8 ± 0.9 39 (23.5)
24.5 (13.0–50.0) 370 (107.5–675.0) 88 (48.5–207.0) 3.1 ± 1.2 4 (28.6)
0.932 0.535
83 (50) 9 (5.4) 8 (4.8)
7 (50) 7 (50) 4 (28.6)
0.958 b0.001 0.001
0.940 0.134 0.669
Values are mean ± standard deviation, median (interquartile range), or n (%). CAD = coronary artery disease, CIs = cerebral infarcts, PCI = percutaneous coronary intervention.
Table 3 Characteristics and distribution of post-procedural cerebral infarcts according to the access site.
Patients (n) Lesions Multiple lesions Vascular territory Right carotid (n) Left carotid (n) Vertebrobasilar (n) Right vertebral Left vertebral
Right radial
Left radial
Femoral
Total
9 13 3
3 5 1
2 2 0
14 20 4
3 1 7 2 –
0 1 3 – –
1
4 2 11
1 – –
estimate from logistic regression adjusted for PS quartiles: 0.266; 95% confidence interval: 0.066 to 1.080; p = .064). In a secondary analysis, PS method was used to compare RR vs non-RR approach (TF + LR). Post-procedural CIs were again more frequent in the RR group (OR estimate from logistic regression adjusted for PS quartiles: 0.170; 95% confidence interval: 0.050 to 0.574; p = .004). 3.3. Causal mediation analysis The association between RR approach and new CIs, compared with non-RR approach, was partially mediated (23%) by the presence of arterial tortuosity, but not by aortic calcification (1%). 4. Discussion Among periprocedural complications of coronary angiography and PCI, the neurological ones are the most dreaded. Symptomatic cerebral ischemia following diagnostic or interventional coronary angiography is a well-known serious—though rare—complication [1]. However, also radiologically documented CIs without neurological symptoms seem to be relatively frequent [4–13]. Post-left heart catheterization CIs may result from gaseous emboli or thrombotic clots on guidewires or catheters, but are mainly due to atherosclerotic fragments as a direct consequence of catheter/wire manipulation in the aortic wall. At least three important considerations can be drawn from our data. First, consistently with previous studies, a non-negligible incidence (7.8%) of CIs was observed after cardiac catheterization procedures, although most patients had no neurological symptoms. The relatively wide difference in silent CIs occurrence among studies is probably related to several factors (study design, catheterization technique, imaging modality and its time interval from the index procedure, clinical setting, and patient characteristics). In this respect, a higher incidence of CIs was reported in patients with acute coronary syndrome (35%) [9], after procedures with routine aortic valve crossing (14–22%) [4,12], or with internal mammary artery angiography (19%) [10]. Since our objective was to compare the incidence of CIs using the three most common vascular accesses, the per-protocol exclusion of patients with the above characteristics, along with the adoption of standardized catheters and guidewires, may have reduced the risk for potential confounding factors resulting in a lower observation rate. Further, the radiological characteristics and vascular distribution of these CIs are highly compatible with embolic damage; in addition, CIs location is in accordance with the arterial site, suggesting that the latter plays a role in CIs occurrence. Second, in contrast to previous studies, new CIs were not associated with risk factors [9,13], ad-hoc PCI [13], or procedure/fluoroscopy time [5,6], but rather with the presence of arterial tortuosity and significant aortic calcification. Severe tortuosity can pose technical challenges leading to failure of the transradial approach, and sometimes operators are compelled to complex catheter manipulations. In the attempts to overcome a difficult arterial anatomy, catheters may scratch the arterial wall causing vessel damage or mobilization of atherosclerotic debris, thereby
Please cite this article as: N. Marchese, M. Copetti, V. Inchingolo, et al., Cerebral infarcts after coronary angiography and percutaneous coronary intervention. A prospective p..., Cardiovascular Revascularization Medicine, https://doi.org/10.1016/j.carrev.2019.11.010
N. Marchese et al. / Cardiovascular Revascularization Medicine xxx (xxxx) xxx
promoting embolic brain damage. Accordingly, the presence of severe aortic calcification is a marker of high atherosclerotic burden in an anatomic territory involved with the passage of wires and catheters. In our study, the presence of subclavian or innominate arterial tortuosity/loop and/or severe aortic calcification was associated with CIs in 7 of 12 patients of the RR and LR groups (58.3%), suggesting a role in the development of embolic brain injuries. This is the first study to investigate the impact of vessel tortuosity/aortic calcification on post-procedural ischemic brain damage and to suppose their contributing role in post-procedural CIs. We can also speculate that the procedural factors found to be associated with CIs (number of catheters, procedure and fluoroscopy times) in other studies may be, at least in part, related to the presence of challenging arterial anatomy. The third and most important consideration is that the RR approach was found to be independently associated with post-procedural CIs compared with the LR (though without reaching statistical significance) and TF approaches. In addition, the only case of clinically manifest brain ischemia (thalamic stroke with neurological sequelae) occurred in the RR group. This finding was confirmed when comparing the RR vs nonRR (LR + TF) group. The rationale for considering the LR and TF approaches as a whole group lies in the similar straight and direct access to the ascending aorta, with quite similar catheter movement to engage the coronary ostia. This association is only partially explained by the presence of arterial tortuosity but not by the presence of aortic calcification, as shown by causal mediation analysis. This suggests that the RR approach is associated per se with a higher risk of cerebral embolic damage. To the best of our knowledge, this is a new concept, since the majority of previous studies did not focus on the comparison of different arterial routes. Studies evaluating access site in relation to the risk of clinically manifest neurological events reported a lack of association between radial access and ischemic stroke [17]. In the large MATRIX trial, involving patients with acute coronary syndrome who underwent invasive management, no differences were found between radial (with no differentiation between right and left side) and femoral approaches in terms of ischemic stroke occurrence [18]. In addition, there are very limited data on the association of the RR vs LR access site with cerebral ischemic complications. A meta-analysis did not find statistically significant differences in clinically reported stroke/TIA between the two approaches [19] but, more recently, in a large British database, the LR access was associated with a significant decrease in PCI-related in-hospital stroke [20] compared with the RR approach. As regards silent and radiologically documented post-procedural CIs, the only per-protocol comparison of radial vs femoral approach is provided by the randomized study of Hamon and colleagues [12], which enrolled patients with severe aortic stenosis referred for cardiac catheterization, also including aortic valve crossing. The risk of brain embolic damage did not diverged significantly between the two approaches. Other observational studies with systematic post-procedural MRI did not show any differences between radial and femoral procedures (5,9–11). In contrast, at transcranial Doppler, RR access was found to generate more particulate emboli than femoral [21] and LR access [22]. Several reasons may account for our results. First, steering catheters and wires can be more difficult in the setting of tortuosity of the subclavian and brachiocephalic trunk, with the latter in direct communication with the right carotid artery. In ours as in other studies [23,24], a higher prevalence of arterial tortuosity was observed in the RR group (13.8%) compared with the LR (9.8%) and TF (3.3%) groups. Regardless of the presence of tortuous anatomy or loops, during RR angiography the catheters are forced to pass in a more acute angle between the brachiocephalic trunk and ascending aorta compared with the LR and TF approaches, thereby increasing the risk of plaque mobilization. Second, the Judkins preshaped catheters, originally designed for a femoral approach, may require sometimes-difficult manipulations to reach the coronary ostia using the RR approach. Conversely, the LR approach shares with the TF approach the same orientation in relation to the coronary anatomic plane. An indirect confirmation of this hypothesis
5
comes from transcranial studies where a higher number of microembolic signals during catheter manipulation was demonstrated using the RR approach [21,22]. In the present study, the location of CIs confirms the concept, also based on experimental data, that posterior circulation may be more prone to embolic occlusion [5,25,26]. In addition, due to the less dense posterior vascularization of the brain, a given embolic load may result more commonly in CIs. Another possible explanation is that radial approach itself may promote posterior CIs, considering that vertebral artery arises from subclavian artery. While the clinical impact of stroke is evident, the clinical implications and long-term relevance of silent CIs remain a matter of debate. Further, the increasing number of neuroimaging exams, with consequent incidental finding of CIs, together with the improvements in neuroimaging techniques, has amplified the dimension of the problem. Observational population studies found a correlation between asymptomatic CIs and the risk of stroke [27,28]. In a meta-analysis including 13 studies, the occurrence of silent brain infarcts was associated with a 2-fold increased risk of future stroke [29]. It is plausible that silent CIs may represent a risk marker of other diseases (i.e. atherosclerosis, cardiac arrhythmias, cardiomyopathies) that, in turn, are causally associated with stroke. On the contrary, the correlation with cognitive impairment, dementia and Alzheimer's disease [28,30] may recognize CIs as direct contributors since, as demonstrated by MRI-based studies, cognitive impairment is also triggered by cumulative cerebral damage caused by ischemic injuries [31]. In patients with major depression, the occurrence of CIs was also positively associated with symptom severity [32], higher hospitalization rates, and a higher risk of developing psychiatric and neurological disorders [33]. The nature of these studies does not allow us to establish whether CIs are ‘innocent’ bystander, or rather are linked in a cause-and-effect relationship to these diseases. Since silent CIs are relatively common after cardiac procedures (e.g. transcatheter aortic valve implantation, coronary artery bypass grafting, electrophysiologic procedures), the debate on their prognostic implications has gained more attention in parallel with the worldwide rapidly increasing number of procedures. Although the acute phenomenon has been extensively studied, further studies focused on late implications of post-procedural CIs are warranted. 4.1. Strengths and limitations This is the first study focusing on the role of the three most commonly used arterial access sites for coronary angiography in post-procedural occurrence of CIs. The systematic use of pre and post-procedural MRI, the relatively restrictive entry criteria, and a standardized procedural protocol helped us to minimize potential confounding factors and to analyze very consistent data. In addition, the collection of different data from similar studies (i.e. the presence of arterial tortuosity and aortic calcifications) gave us the opportunity to observe a correlation between these relatively frequent characteristics and CIs. However, several study limitations should be acknowledged. The relative low frequency of events (7.8%), also compared with similar studies, the nonrandomized nature of data, even in the presence of a PS analysis, and the non-consecutive recruitment of patients, may have limited the reliability of our results. Unfortunately, the small number of patients limited the ability of our study to strongly correlate the presence of tortuosity to CIs occurrence. The choice to administer low-dose heparin, to prevent radial artery occlusion, at the end of the diagnostic procedure, together with the routine use of 300 mm wire, may have limited the generalizability of our results. In the absence of data on heparin dosage able to protect against occurrence of CIs after cardiac procedures, it can be speculated that low-dose may be less effective than high-dose. However, the comparison between RR and LR is not affected by our protocol, whereas the use of more heparin in the RR and LR groups favors radial vs femoral approach. The lack of radiological and clinical followup does not allow for a comprehensive evaluation of long-term
Please cite this article as: N. Marchese, M. Copetti, V. Inchingolo, et al., Cerebral infarcts after coronary angiography and percutaneous coronary intervention. A prospective p..., Cardiovascular Revascularization Medicine, https://doi.org/10.1016/j.carrev.2019.11.010
6
N. Marchese et al. / Cardiovascular Revascularization Medicine xxx (xxxx) xxx
consequences of CIs (i.e. cognitive decline, clinically evident cerebral ischemia). A systematic radiological study of the subclavian, carotid and vertebral arteries could have helped us to identify additional embolic sources from atherosclerotic plaques during the exams.
5. Conclusions Our study confirms that CIs are not infrequent after elective coronary angiography and PCI, and suggests that use of the RR access is independently associated with a higher rate of new CIs compared with the LR and TF approaches. Additionally, the association of the RR approach with post-procedural CIs is only partially explained by the presence of arterial tortuosity. Further studies are warranted to determine the late clinical implications of silent CIs.
References [1] Kwok CS, Kontopantelis E, Myint PK, Zaman A, Berry C, Keavney B, et al. Stroke following percutaneous coronary intervention: type-specific incidence, outcomes and determinants seen by the British Cardiovascular Intervention Society 2007–12. Eur Heart J 2015;36:1618–28. [2] Aggarwal A, Dai D, Rumsfeld JS, Klein LW, Roe MT. American College of Cardiology National Cardiovascular Data Registry Incidence and predictors of stroke associated with percutaneous coronary intervention. Am J Cardiol 2009;104:349–53. [3] Hoffman SJ, Routledge HC, Lennon RJ, Mustafa MZ, Rihal CS, Gersh BJ, et al. Procedural factors associated with percutaneous coronary intervention-related ischemic stroke. J Am Coll Cardiol Intv 2012;5:200–6. [4] Omran H, Schmidt H, Hackenbroch M, Illien S, Bernhardt P, von der Recke G, et al. Silent and apparent cerebral embolism after retrograde catheterisation of the aortic valve in valvular stenosis: a prospective. randomised study Lancet 2003;361: 1241–6. [5] Lund C, Nes RB, Ugelstad TP, Due-Tønnessen P, Andersen R, Hol PK, et al. Cerebral emboli during left heart catheterization may cause acute brain injury. Eur Heart J 2005;26:1269–75. [6] Büsing KA, Schulte-Sasse C, Flüchter S, Süselbeck T, Haase KK, Neff W, et al. Cerebral infarction: incidence and risk factors after diagnostic and interventional cardiac catheterization - prospective evaluation at diffusion-weighted MR imaging. Radiology 2005;235:177–83. [7] Hamon M, Gomes S, Oppenheim C, Morello R, Sabatier R, Lognoné T, et al. Cerebral microembolism during cardiac catheterization and risk of acute brain injury: a prospective diffusion-weighted magnetic resonance imaging study. Stroke 2006;37: 2035–8. [8] Hamon M, Gomes S, Clergeau MR, Fradin S, Morello R, Hamon M. Risk of acute brain injury related to cerebral microembolism during cardiac catheterization performed by right upper limb arterial access. Stroke 2007;38:2176–9. [9] Murai M, Hazui H, Sugie A, Hoshiga M, Negoro N, Muraoka H, et al. Asymptomatic acute ischemic stroke after primary percutaneous coronary intervention in patients with acute coronary syndrome might be caused mainly by manipulating catheters or devices in the ascending aorta, regardless of the approach to the coronary artery. Circ J 2008;72:51–5. [10] Kim IC, Hur SH, Park NH, Jun DH, Cho YK, Nam CW, et al. Incidence and predictors of silent embolic cerebral infarction following diagnostic coronary angiography. Int J Cardiol 2011;148:179–82. [11] Kim BJ, Lee SW, Park SW, Kang DW, Kim JS, Kwon SU. Insufficient platelet inhibition is related to silent embolic cerebral infarctions after coronary angiography. Stroke 2012;43:727–32.
[12] Hamon M, Lipiecki J, Carrié D, Burzotta F, Durel N, Coutance G, et al. Silent cerebral infarcts after cardiac catheterization: a randomized comparison of radial and femoral approaches. Am Heart J 2012;164:449–54. [13] Deveci OS, Celik AI, Ikikardes F, Ozmen C, Caglıyan CE, Deniz A, et al. The incidence and the risk factors of silent embolic cerebral infarction after coronary angiography and percutaneous coronary interventions. Angiology 2016;67:433–7. [14] Sacco RL, Kasner SE, Broderick JP, Caplan LR, Connors JJ, Culebras A, et al. An updated definition of stroke for the 21st century: a statement for healthcare professionals from the American Heart Association/American Stroke Association. Stroke 2013; 44:2064–89. [15] Yanovitzky I, Zanutto E, Hornik R. Estimating causal effects of public health education campaigns using propensity score methodology. Eval Progr Plann 2005;28: 209–20. [16] Imai K, Keele L, Tingley D. A general approach to causal mediation analysis. Psychol Methods 2010;15:309–34. [17] Sirker A, Kwok CS, Kotronias R, Bagur R, Bertrand O, Butler R, et al. Influence of access site choice for cardiac catheterization on risk of adverse neurological events: a systematic review and meta-analysis. Am Heart J 2016;181:107–19. [18] Valgimigli M, Gagnor A, Calabró P, Frigoli E, Leonardi S, Zaro T, et al. Radial versus femoral access in patients with acute coronary syndromes undergoing invasive management: a randomised multicentre trial. Lancet 2015;385:2465–76. [19] Shah RM, Patel D, Abbate A, Cowley MJ, Jovin IS. Comparison of transradial coronary procedures via right radial versus left radial artery approach: a meta-analysis. Catheter Cardiovasc Interv 2016;88:1027–33. [20] Rashid M, Lawson C, Potts J, Kontopantelis E, Kwok CS, Bertrand OF, et al. Incidence, determinants, and outcomes of left and right radial access use in patients undergoing percutaneous coronary intervention in the United-Kingdom: a National Perspective using the BCIS dataset. JACC Cardiovasc Interv 2018;11:1021–33. [21] Jurga J, Nyman J, Tornvall P, Mannila MN, Svenarud P, van der Linden J, et al. Cerebral microembolism during coronary angiography: a randomized comparison between femoral and radial arterial access. Stroke 2011;42:1475–7. [22] Pacchioni A, Mugnolo A, Penzo C, Nikas D, Saccà S, Agostoni P, et al. Cerebral microembolism during transradial coronary angiography: comparison between single and double catheter strategy. Int J Cardiol 2014;170:438–9. [23] Sciahbasi A, Romagnoli E, Burzotta F, Trani C, Sarandrea A, Summaria F, et al. Transradial approach (left vs right) and procedural times during percutaneous coronary procedures: TALENT study. Am Heart J 2011;161:172–9. [24] Freixa X, Trilla M, Feldman M, Jiménez M, Betriu A, Masotti M. Right versus left transradial approach for coronary catheterization in octogenarian patients. Catheter Cardiovasc Interv 2012;80:267–72. [25] Barbut D, Grassineau D, Lis E, Heier L, Hartman GS, Isom OW. Posterior distribution of infarcts in strokes related to cardiac operations. Ann Thorac Surg 1998;65:1656–9. [26] Segal AZ, Abernethy WB, Palacios IF, BeLue R, Rordorf G. Stroke as a complication of cardiac catheterization: risk factors and clinical features. Neurology 2001;56:975–7. [27] Vermeer SE, Den Heijer T, Koudstaal PJ, Oudkerk M, Hofman A, Breteler MM. Incidence and risk factors of silent brain infarcts in the population-based Rotterdam scan study. Stroke 2003;34:392–6. [28] Debette S, Beiser A, DeCarli C, Au R, Himali JJ, Kelly-Hayes M, et al. Association of MRI markers of vascular brain injury with incident stroke, mild cognitive impairment, dementia, and mortality: the Framingham offspring study. Stroke 2010;41:600–6. [29] Gupta A, Giambrone AE, Gialdini G, Finn C, Delgado D, Gutierrez J, et al. Silent brain infarction and risk of future stroke: a systematic review and meta-analysis. Stroke 2016;47:719–25. [30] Vermeer SE, Prins ND, den Heijer T, Hofman A, Koudstaal PJ, Breteler MM. Silent brain infarcts and the risk of dementia and cognitive decline. N Engl J Med 2003; 348:1215–22. [31] Choi SH, Na DL, Chung CS, Lee KH, Na DG, Adair JC. Diffusion-weighted MRI in vascular dementia. Neurology 2000;54:83–9. [32] Fujikawa T, Yamawaki S, Touhouda Y. Background factors and clinical symptoms of major depression with silent cerebral infarction. Stroke 1994;25:798–801. [33] Yanai I, Fujikawa T, Horiguchi J, Yamawaki S, Touhouda Y. The 3-year course and outcome of patients with major depression and silent cerebral infarction. J Affect Disord 1998;47:25–30.
Please cite this article as: N. Marchese, M. Copetti, V. Inchingolo, et al., Cerebral infarcts after coronary angiography and percutaneous coronary intervention. A prospective p..., Cardiovascular Revascularization Medicine, https://doi.org/10.1016/j.carrev.2019.11.010