Cerebrovascular Events Complicating Cardiac Catheterization - A Tertiary Care Cardiac Centre Experience.

Background: Cerebrovascular events (CVEs) are one of the rare complications of cardiac catheterization. This prospective single-center study was conducted to assess the incidence, presentations, and outcomes of CVEs in patients undergoing cardiac catheterization.
Methods: Patients undergoing cardiac catheterization who developed CVEs within 48 h of procedure were analyzed prospectively with clinical assessment and neuroimaging.
Results: Out of 55,664 patients, 35 had periprocedural CVEs (0.063%). The incidence of periprocedural CVEs with balloon mitral valvotomy, percutaneous coronary intervention, and coronary angiography was 0.127%, 0.112%, and 0.043%, respectively. A larger proportion of periprocedural CVEs occurred in patients with acute coronary syndrome (ACS, 77.1%) than in patients with stable coronary artery disease (CAD). The majority of CVEs were ischemic type (33 patients, 94.3%). It was most commonly seen in the left middle cerebral artery (MCA) territory. Hemorrhagic CVEs were very rare (2 patients, 5.7%). The majority of the CVEs manifested during or within 24 h of the procedure (31 patients, 88.6%). Neurodeficits persisted during the hospital stay in 20 patients (57.2%), who had longer duration of procedure compared to those with recovered deficits (P = 0.0125). In-hospital mortality occurred in three patients (8.5%) and post-discharge mortality in another 3 (8.5%). Conclusions: Periprocedural CVEs are rare and have decreased over time. They occur in a greater proportion in patients with ACS than in patients with stable CAD, more with interventional than diagnostic procedures. Ischemic event in the left MCA territory is the most common manifestation, commonly seen within 24 h of the procedure. Longer duration of procedure was a risk factor for larger infarcts and hence persistent neurodeficit at discharge. Although a substantial number of patients recover the neurodeficits, periprocedural CVEs are associated with adverse outcomes. Copyright:

INTRODUCTION

Cerebrovascular events (CVEs) resulting from cardiac catheterization is a relatively uncommon complication, despite a large number of cardiac procedures performed worldwide. Interventional procedures such as percutaneous coronary interventions (PCIs) are associated with more incidence of CVEs (0.2%–0.4%),[12] compared to diagnostic procedures such as coronary angiograms (CAG, 0.1%).[34] We conducted a prospective observational study of CVEs in patients undergoing cardiac catheterization to study the incidence, presentation, and outcome of CVEs.

METHODS

Patients who developed CVEs during or within 48 h of cardiac catheterizations at our institute, between January 1, 2018, and February 29, 2020, were studied. Interventional procedures, which are less likely to cause CVEs, such as pacemaker implantation, venous interventions, or isolated right heart interventions were excluded. The study was approved by the Institutional Ethics Committee. Informed consent was obtained from all patients.

Demographic features, cardiovascular risk factors, and medical history were collected. Interventional details such as type, indication, duration, and adjunctive drugs were noted. Clinical features, imaging findings, and outcomes of periprocedural CVEs were also analyzed. Follow-up with either clinical examination or telephonic consultation was done at 3 months, 6 months, and at the end of the study period.

A descriptive statistical analysis was done. The continuous variables, such as age, were expressed in terms of mean ± standard deviation (SD). The categorical measurements were expressed in number (percentage). Comparison between group with persistent and recovered neurodeficits was done using Student's t-test (two-tailed, independent) for continuous variables and Chi-square/Fisher's exact test for categorical variables.

RESULTS

A total of 55,664 patients underwent cardiac catheterization during the study period. Among them, 35 patients had periprocedural CVEs (0.063%). Baseline characteristics of the patients with CVEs are shown in Table 1. Details of the procedures and incidence of CVEs in them are mentioned in Figure 1. The majority of patients with CVEs were in the age group of 51–70 years. It commonly occurred among patients with acute coronary syndrome (ACS, 27 out of 35 patients, 77.1%). Periprocedural CVEs were more frequent with interventional procedures such as balloon mitral valvotomy (BMV) and PCI than diagnostic procedures such as CAG, as shown in Figure 1. Initial cardiac presentations of these patients and complications are depicted in Table 1.

Table 1

Baseline characteristics of the study population

Characteristics	Number of patients (n=35), n (%)
Age (years), mean±SD	56.86±13.08
Sex ratio (male: female)	25:10
Body mass index (kg/m2)	24.50±2.95
HTN	18 (51.4)
DM	18 (51.4)
Dyslipidemia	9 (25.7)
Prior MI	7 (20.0)
Prior CABG	2 (5.7)
Valvular heart disease	3 (8.6)
PAD	1 (2.9)
CKD	3 (8.6)
Smoking	11 (31.4)
LVEF (%)	45.11±9.92
Arrhythmias
Complete heart block	3 (8.6)
Atrial fibrillation	4 (11.5)
Ventricular tachycardia	2 (5.7)
Other complications
Pulmonary edema	5 (14.3)
Cardiogenic shock	3 (8.6)
Left ventricular thrombus	3 (8.6)
Indications for cardiac catheterization
IHD	30 (85.7)
ACS	27 (77.1)
Chronic stable angina	1 (2.9)
Ischemic cardiomyopathy	2 (5.7)
RHD, severe mitral stenosis	3 (8.5)
HF with AF for RFA	1 (2.9)
Coarctation of aorta	1 (2.9)

ACS: Acute coronary syndrome, AF: Atrial fibrillation, CABG: Coronary artery bypass grafting, CKD: Chronic Kidney disease, DM: Diabetes mellitus, HF: Heart failure, HTN: Hypertension, LVEF: Left ventricular ejection fraction, MI: Myocardial infarction, PAD: Peripheral arterial disease, RFA: Radiofrequency ablation, RHD: Rheumatic heart disease, SD: Standard deviation, IHD=Ischemic heart disease

Figure 1

Central illustration - cerebrovascular events complicating cardiac catheterization

The mean procedural duration of CAG and PCI in the cases with CVEs was 25 ± 12.59 min and 56.8 ± 28.1 min, respectively. These durations were longer compared to those of routine procedures, thereby suggesting increased complexity of these procedures and increased manipulation of hardware, which predisposed to CVEs. Among the patients who underwent coronary procedures (CAG and PCI), femoral access was used in 22 patients and radial access in 8 patients. In our study, 16 cases of CVEs were noted during PCI; 9 of them during right coronary artery (RCA) interventions, 6 of them during left coronary artery (LCA) intervention, and 1 during PCI to anomalous RCA originating from the left coronary sinus. CVE complicated 3 patients (0.15%) of BMV in our study.

The most common neurodeficit noted among the patients with periprocedural CVEs was hemiparesis. Transient ischemic attacks were seen in 8 patients (22.9%). The clinical manifestations of CVEs are depicted in Table 2.

Table 2

Clinical manifestations of patients with cerebrovascular events

Clinical presentation of CVEs	Number of patients (n=35), n (%)
Left hemiparesis	9 (22.8)
Right hemiparesis	9 (22.8)
Right faciobrachial monoparesis	2 (5.7)
Cerebellar ataxia	2 (5.7)
Dysarthria	2 (5.7)
Blurring of vision	2 (5.7)
Headache	2 (5.7)
Altered sensorium	2 (5.7)
Isolated left facial palsy	1 (2.9)
Isolated right facial palsy	1 (2.9)
Right homonymous hemianopia	1 (2.9)
Isolated Broca’s aphasia	1 (8.6)
Diplopia	1 (2.9)
Vocal cord palsy	1 (2.9)
Generalized tonic–clonic seizures	1 (2.9)

CVE: Cerebrovascular events

The cross-sectional brain imaging was performed in 31 of 35 patients. 17 underwent computed tomography and 14 underwent magnetic resonance imaging, according to clinician preference and availability. In 4 patients, imaging was not performed, given transient neurodeficit. The scan was normal in 7 patients. 22 patients had an infarct, of which 13 (59%) had isolated middle cerebral artery (MCA) territory infarcts, 3 (13.6%) had isolated posterior cerebral artery territory infarct, 1 (4.6%) had isolated anterior cerebral artery territory infarct, and 1 (4.6%) had a cerebellar infarct. The remaining 4 (18.2%) had multiterritory infarcts. Among those with MCA territory infarct, left-sided (9 patients) was more common than the right (4 patients). Two patients had intracranial hemorrhage; one had right parietal bleed with subarachnoid extension and the other had large temporoparietal bleed with intraventricular extension and mass effect. Overall, 33 patients had ischemic events and 2 had hemorrhagic events, thereby the incidence of each of them being 0.059% and 0.0035%, respectively.

The majority of CVEs were observed either during the procedure (14 patients, 40%) or within 24 h (17 patients, 8.6%). They occurred between 24 and 48 h in 4 patients (11.4%).

Neurodeficits were recovered during the hospital stay in 15 patients (42.8%) [Table 3]. On further follow-up, a total of 21 patients had complete improvement of neurodeficits constituting a cumulative recovery rate of 60%. Those with persistent neurodeficits at discharge had longer duration of procedures compared to those with recovered neurodeficit [P = 0.0125, Table 4].

Table 3

Recovery of neurodeficits among patients with cerebrovascular events

Time of recovery of neurodeficits	Number of patients, n (%)
Within 1 h	3 (8.6)
1-24 h	5 (14.3)
After 24 h-before discharge	7 (20)
Persisted at discharge	17 (48.5)
Death	3 (8.6)
Total	35 (100)

Table 4

Comparison between the patients with recovery of neurodeficits at discharge and those with persistent neurodeficit at discharge

Parameters	Neurodeficits recovered (n=15)		Neurodeficits persisted (n=17)		P
	Present	Absent	Present	Absent
HTN	6	9	9	8	0.7
DM	9	6	8	9	0.7
Dyslipidemia	4	11	4	13	0.8
Smoking	4	11	6	11	0.7
Cardiogenic shock	2	13	1	16	0.9
Age	58.47±11.36		58.06±10.14		0.915
Duration of procedure	25.6±24.06		52.35±31.72		0.0125
Type of procedures
CAG	10		5
PCI	4		10
BMV	1		2
Neuroimaging findings
Lacunar Infarcts	2		2
Nonlacunar infarct	6		11
Normal	4		3
Hemorrhage	0		1

DM: Diabetes mellitus, HTN: Hypertension, BMV: Balloon mitral valvotomy, PCI: Percutaneous coronary intervention, CAG: Coronary angiography

In-hospital mortality occurred in 3 patients (8.5%), and another 3 (8.5%) had mortality within 1 week of discharge. Cumulative mortality was in 6 patients (17%), as depicted in the central illustration. Among those with in-hospital mortality, 2 patients had a large hemorrhagic stroke and 1 had a ventricular septal rupture, causing cardiogenic shock. Among the 3 patients with postdischarge mortality, 2 had large infarcts and the other had a large infarct with hemorrhagic transformation.

DISCUSSION

In our study, it was found that interventional procedures are associated with more CVEs, compared to diagnostic angiograms. This is due to prolonged duration, use of multiple catheters, wires, and use of larger catheters.[345] An earlier study by Korn-Lubetzki et al.[2] has reported similar 2.5 times higher incidence of CVEs with interventional procedures compared to diagnostic procedures. We observed that the majority of CVEs occurred among the patients with ACS (77.1%), compared to stable CAD (8.5%). The more risk of CVEs in ACS patients is probably because of thrombotic milieu, unstable hemodynamics, and sometimes, more challenging anatomy. Similarly, higher incidence of CVEs was associated with PCI performed for ACS, compared to those done for stable CAD in Euro Heart Survey PCI Registry.[5]

Table 5 shows the findings noted in other previous studies on CVEs complicating cardiac catheterization. The incidence of CVEs in our study was lesser compared to previous studies,[678910] probably because of a decade gap, improvement in the technology of diagnostic and interventional procedures including catheters, guidewires, stents, and loading with newer antiplatelets such as Ticagrelor and Prasugrel.

Table 5

Previous studies with periprocedural cerebrovascular events

Studies	Number of patients	Number of patients with CVEs, n (%)	Ischemic stroke, n (%)	Hemorrhagic stroke, n (%)	Uncertain, n (%)	Risk factors found in patients with CVEs
Wong et al.[6]	76,903	140 (0.18)				Age=70±12 years Acute MI=25% Prior MI=15.7% PAD=27.1% Renal failure=7.1% HF=22.1% EF=45.3%±13% IABP=4.3% GPIIb/IIIa inhibitors=65.7%
Dukkipati et al.[7]	20,679	92 (0.44)	43 (0.21)	13 (0.06)	36 (0.17)	Age=70±12 years DM=39 (42.4%) HTN=26 (28.2%) Dyslipidemia=47 (51%) PAD=24 (26.1) Prior stroke=21 (22.8%) HF=29 (31.5%) Prior PCI=26 (28.2%) Prior CABG=25 (27.2) IABP=3 (3.3%)
Fuchs et al.[8]	9662	43 (0.44)	21 (0.22)	20 (0.21)	2 (0.001)	Age=72±11 years Venous graft Intervention=38.8% IABP=23.3%
Akkerhuis et al.[9]	8555	31 (0.37)	19 (0.22)	12 (0.14)	1 (0.01)	Age=67±7 years HTN=7 (22.6%) DM=7 (22.6%)
Lazar et al.[10]	6465	27 (0.42)	23 (0.36)	-	-	Female gender=17 (62.9%) PAD=7 (26%) LVEF=50%±1% Extensive CAD=20 (74%)

CVE: Cerebrovascular events, DM: Diabetes mellitus, HTN: Hypertension, CABG: Coronary artery bypass grafting, HF: Heart failure, LVEF: Left ventricular ejection fraction, MI: Myocardial infarction, PAD: Peripheral arterial disease, PCI: Percutaneous coronary intervention, IABP: Indications for intra-aortic balloon pump, EF: Ejection fraction

Ischemic event is the most common type of CVEs complicating cardiac catheterization.[345] It occurred in 0.059% of procedures. The incidence was lesser, compared to the other studies.[678910] It was seen in 0.1% in the British Cardiovascular Intervention Society Study[1] and 0.09% in the study by Korn-Lubetzki et al.[2] In most of the cases, the mechanism of ischemic stroke is directly related to embolism from cardiac catheterization itself. Embolism can be due to manipulation of catheters and wires, dislodging debris made up of calcific material, cholesterol particles, or thrombus from atherosclerotic plaques within the aortic arch, proximal carotid, and vertebral arteries. Further, fresh thrombi may form on catheters and guidewires due to inadequate anticoagulation and can embolize to the cerebral circulation.[1234] Less common causes of periprocedural ischemic CVEs include embolism of left ventricular thrombus, air embolism, periprocedural hypotension, arterial dissection, and fractured guidewire.[1112] In our patients, left ventricular thrombus and cardiogenic shock were found in 3 patients (8.6%) each, which might have predisposed to periprocedural CVEs.

Embolic stroke has been reported in patients undergoing BMV in 1.1%–5.4% of cases. Risk factors are atrial fibrillation (AF), calcified valves, advanced age, prolonged procedure, and previous history of thromboembolism.[13]

In our study, periprocedural CVEs were seen in 3 patients, undergoing BMV. One patient had AF. None had left atrial thrombus. Specks of calcium were present in 2. One had a prolonged procedure due to difficulty in crossing the mitral valve.

Hemorrhagic CVEs rarely occur in patients undergoing cardiac catheterization, because of hemostatic abnormalities induced by thrombolytics, anticoagulants, and antiplatelets, used in ACS and periprocedural period.[89] In a retrospective analysis of 43 cases of periprocedural stroke by Fuchs et al.,[8] ischemic stroke occurred in 48.8% and hemorrhagic stroke in 46.5%. The high incidence of hemorrhagic stroke in that study was attributed to increased use of intra-aortic balloon pump and more intense and prolonged anticoagulation.[8] However, subsequent studies have reported a very low incidence of hemorrhagic stroke ranging from 0.03% to 0.09%.[34] Our study also showed a very low incidence of hemorrhagic stroke of 0.0035%.

In various studies, multiple independent predictors of periprocedural CVEs were identified.[678910] Clinical risk factors include advanced age, hypertension, diabetes mellitus, history of CVEs, renal failure, heart failure, and severity of CAD, including the presence of triple-vessel disease.[1011121314] Procedural risk factors include emergency catheterization, prolonged procedural time, more contrast use, retrograde catheterization of the left ventricle in patients with aortic stenosis, interventions on bypass grafts, use of an intra-aortic balloon pump, and presence of coronary artery thrombus.[14151617] In our study, hypertension was present in 18 patients (51.4%), diabetes mellitus in 18 (51.4%), prior myocardial infarction in 7 (20%), and renal failure in 3.

Among the patients with periprocedural CVEs complicating coronary intervention, transfemoral access was more often used than transradial access. This is consistent with the observation made in a Japanese multicenter registry, which showed that periprocedural CVEs were more often associated with transfemoral access than transradial access.[18]

We also observed the increased occurrence of CVEs with PCI to RCA than LCA. This is in concordance with the study by Murai et al.,[19] in which multivariate analysis showed that PCI to RCA was associated with increased risk of CVEs than LCA. It is probably due to erosion of atherosclerotic plaque in the aortic arch and its embolization to the carotid artery, as engaging the ostium of RCA needs more manipulation than LCA.[19] Another important reason is probably that the right coronary sinus is in straight alignment with the origin of the carotid arteries, compared to the left coronary sinus. Hence, atheromatous and thrombotic debris from the right coronary sinus can easily embolize to the carotid arteries. This is illustrated in Figure 2.

Figure 2

Trajectory of emboli during right coronary artery intervention

CVEs occur during or within 24 h of the procedure as seen in our study (88.6%) and study by Dukkipati et al. (62%).[7] Hemiparesis was the most common presentation of periprocedural CVEs (51.4%) in our study. The frequent manifestations of CVEs in previous studies were visual disturbance, aphasia, dysarthria, hemiparesis, and altered mental status,[6710] which were also observed in our study. In our study, 20 patients had an infarct on brain imaging, of which 60% had MCA territory infarcts. Similarly, in the study by Fuchs et al.,[8] ischemic strokes most often involved MCA territory.

Periprocedural infarcts were more often on the left side than on the right, probably due to increased risk of embolism of atheromatous and thrombotic debris to the left common carotid artery than the right. This is a novel finding noted in our study and needs further validation. However, the side predilection of CVEs in hypertensive patients was studied by Rodríguez Hernández et al.[20] They found that both atherosclerotic and cardioembolic infarcts were more common on the left side of the brain and probably due to hemodynamic and anatomical factors. The right common carotid artery arises from the innominate artery. The presence of the innominate artery reduces the flow velocity and protects the right common carotid artery from atheroemboli and thromboemboli. In contrast, the left common carotid artery arises directly from the aortic arch and runs more in alignment with the ascending aorta. As a result, energy transfer from systolic emptying forces, mean flow velocity, and oscillating shear stress are significantly greater in the left common carotid artery than in the right, resulting in greater chances of athrosclerosis and embolism.[20] Another possible explanation for this important observation is that the left-sided strokes are recognized more frequently because they lead to obvious clinical manifestations, such as aphasia and motor neurological deficits, whereas right-sided strokes may lead to less obvious symptoms, such as hemineglect or spatial disorientation.[2122] This was supported by the study by Portegies et al.[23]

Neurodeficits persisted in 57.2% at discharge, and the mean duration of procedure was longer compared to those with recovery of neurodeficit. This is probably because those with persistent deficit had more of interventional procedures, thereby predisposing to larger nonlacunar infarcts. Overall mortality was 17%. In the study by Lazar, et al.,[10] 63% had recovery of neurodeficits, and in the study by Fuchs et al.,[8] overall mortality was 56.1% in their study. These findings along with ours suggest that periprocedural CVEs were associated with increased mortality.

Our study had a few limitations. We could analyze only the cases with periprocedural CVEs and not the control group. Thus, the inferential analysis was not done. Hence, further case–control and cohort studies are needed.

CONCLUSIONS

CVE is an uncommon, but a dreaded complication of cardiac catheterization. Over the decades, there is a decrease in the incidence of periprocedural CVEs. It occurs commonly with interventional procedures than with diagnostic procedures, often in the patients with ACS. Most commonly, they manifest as ischemic events in MCA territory, commonly occurring during or within 24 h of the procedure. Prolonged procedure is a risk factor for larger infarcts and hence persistent neurodeficit. Although a substantial number of patients recover from neurodeficits, periprocedural CVEs are associated with increased mortality and morbidity. Two pertinent observations in our study are that CVEs are noted more often during PCI to RCA than LCA and left-sided periprocedural embolic infarcts are more common than the right.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.