Do Women Have a Higher Risk of Adverse Events after Carotid Revascularization? Do Women Have a Higher Risk of Adverse Events after Carotid Revascularization?

Carotid artery stenosis is thought to cause up to 10% of ischemic strokes. Till now, the optimal treatment between carotid endarterectomy (CEA) and carotid artery stenting (CAS) remains debated, in particular for specific subgroups of patients. Available data suggest that female have higher risk of perioperative adverse events, but conflicting results comparing CEA and CAS regarding the benefit for male or female are present in the literature. A systematic review of recent publications on gender-related differences in oper-ative risks is reported. Moreover, a consecutive cohort of 912 symptomatic and asymptom- atic patients undergoing CEA (407, 44.6%) or CAS (505, 55.4%) in a single institution has been evaluated to determine the influence of gender (59.7% male vs. 40.3% female) on the outcomes after both revascularization procedures at 30 days and during 3 years of follow-up. Our experience seems to confirm literature data as regarding female higher risk of restenosis. Female patients had higher periprocedural (2.7% female vs. 0.9% male; p < 0.05) and long-term (11.4% female vs. 4.6% male; p < 0.05) restenosis rate. In conclusion, female anatomic and pathologic parameters should be taken into account for an accurate diagnosis of carotid stenosis and guidelines should be adjusted consequently. (cid:1) (SD). compli-cations, and CEA by Survival, MI, and restenosis using Kaplan-Meier to for patient dropouts and were reported using (SVS) (SE) are reported in Kaplan-Meier analyses. The log-rank was used to determine differences among patients submitted CEA and


Introduction
Carotid artery stenting (CAS) might be a potentially safe and effective therapeutic alternative to carotid endarterectomy (CEA) for carotid artery disease, especially in high-risk patients, because CAS may avoid anesthetic and surgical risk. In this regard, the merit of using CAS has been questioned in specific subgroups of patients [1,2]. The role of gender in the selection of the most effective carotid intervention remains a matter of debate. Historically, large randomized controlled studies looking at CEA had indicated an increased perioperative risk for women when gender subgroup analysis was performed. Additionally, all of the large CEA trials showed decreased or no benefit in women when compared to men mostly because these trials were underpowered to show any utility in the relatively small female population studied and, secondly, because the long-term benefit was undermined by the high perioperative morbidity seen in women [3,4]. The SAPPHIRE trial was the first stenting trial to show noninferiority of CAS to CEA in high-risk patients, but the study did not compare outcomes for gender subgroups and women were underrepresented within the recruited population [5]. More recent trials reported conflicting results regarding risk of peri-and postprocedural adverse events for women and men [1,6]. Then, reports comparing CEA and CAS failed either to analyze the influence of gender or to show a clear benefit for men or women [7].
Thus, to further inform the debate, we sought to conduct a systematic review of recent publications to assess epidemiologic and diagnostic hypotheses, which could underpin gender-related differences. Moreover, we report a retrospective observational study to determine the impact of gender on the outcomes of both carotid interventions in our institution.

Gender: stroke epidemiology and carotid disease 2.1. Gender differences in stroke epidemiology
An extensive review on gender differences in stroke epidemiology has shown that stroke is more common in men than in women and male patients are on average younger than female when they are affected by their first stroke [8]. Western European studies demonstrated that stroke incidence was about 30% higher in men than in women, but the strokes that did occur in women tended to be more severe [9]. During the last decade, an extensive number of papers have been published on epidemiological differences between genders.
A systematic review around gender differences in stroke epidemiology, presented by Appelros et al., including 59 incidence studies from 19 countries and 5 continents, showed that the mean age at first-ever stroke was 68.6 years among men and 72.9 years among women [10]. Stroke incidence and prevalence rates were 33 and 41% higher in male than in female, respectively, with large variations between age bands and between populations. The incidence rates of brain infarction and intracerebral hemorrhage were higher among men, whereas the rate of subarachnoid hemorrhage was higher among women, although this difference was not statistically significant. Stroke tended to be more severe in women with a 1-month case fatality of 24.7% compared with 19.7% for men.
The lower stroke incidence in women has been analyzed in systematic reviews in the last decades. A plausible reason might be the protective role of ovarian estrogen on the cerebral circulation [11][12][13], even if randomized trials currently recommend that postmenopausal hormone therapy should not be used in the primary prevention of stroke [14].
Other possible factors might be genetic factors, but no evidence for this expectation was found in the literature. Conversely, a recent systematic review showed that women with stroke have a higher familiarity of stroke than men [15].
Numerous studies have shown that blood pressure value was higher in men than women of same age [16,17]. Moreover, peripheral artery disease [18,19], ischemic heart disease [18,[20][21][22], and cigarette consumption [18-20, 23, 24] are more frequent among male stroke patients. Women have been shown to be at higher risk than men for atrial fibrillation-related cardioembolic stroke [22,[25][26][27][28]. The Swedish Stroke Register has shown that women with atrial fibrillation receive oral anticoagulant therapy less often than men [24]. The higher prevalence of embolic strokes among women could justify their higher stroke severity.

Diagnostic criteria for carotid stenosis by gender
The North American Symptomatic Carotid Endarterectomy Trial (NASCET) investigators standardized the method of quantifying the degree of carotid stenosis, and they considered arteriography the most predictable method for evaluating carotid stenosis [29].
Carotid duplex ultrasound scan (DUS) compares favorably with arteriography, and many physicians are using carotid DUS as the definitive diagnostic procedure before CEA even if it is mandatory to submit the patient to computer tomography angiography (CTA) or magnetic resonance angiography (MRA) for CAS to explore arch anatomy [30,31]. Diagnostic criteria for carotid DUS have been accepted as predictable from laboratories accredited by the Intersocietal Commission for the Accreditation of Vascular Laboratories with ongoing quality programs.
Nevertheless, carotid DUS frequently resulted in overestimation of disease severity in women. This underlined the issue that women may have higher velocities in their carotid arteries than men for similar carotid stenosis. The question is: Are we overdiagnosing disease in women according to gender differences in carotid stenosis? The purpose of Comerota's study was to examine whether there were velocity differences based on gender in patients with carotid artery disease and whether different velocity criteria should be used in women, especially at clinically relevant thresholds of disease [32].
Patients who underwent carotid arteriography and carotid DUS were the basis for this study. Data from 1019 carotid bifurcations were available. Comparison was performed on the basis of 938 carotid arteries. Analyses were made in 536 male and 402 female carotid arteries. Arteriography was performed on average 23 days after the ultrasound examination, with 74% of arteriographic examinations performed within 30 days and 95% performed within 82 days of carotid DUS. Additionally, the single most diseased artery per patient was analyzed by gender. Peak systolic velocity (PSV) and end-diastolic velocity (EDV) were averaged for data subsets according to 10% intervals of internal carotid artery (ICA) stenoses. For all intervals, PSV and EDV averaged 9 and 6% higher in women than in men. Significant gender differences existed between PSV and EDV for 60 and 70% stenoses. For 70% stenosis, PSV averaged 285 AE 16 cm/s in women and 236 AE 11 cm/s in men (p = 0.01) and EDV averaged 79 AE 7 cm/s in women and 77 AE 6 cm/s in men (p = 0.03). For 60% stenosis, PSV averaged 228 AE 14 cm/s in women and 189 AE 11 cm/s in men (p = 0.03) and EDV averaged 68 AE 6 cm/s in women and 51 AE 4 cm/s in men (p = 0.01).
Williams et al. demonstrated that the diameter of the common carotid arteries (CCA), ICA, and external carotid arteries (ECA) were considerably smaller in women compared to those in men [33]. Schultz et al. showed a remarkable gender-specific difference in the distribution of atherosclerotic plaque. They reviewed 5395 arteriograms from the European Carotid Surgery Trial and compared diameter ratios of the ICA, CCA, and ECA with minimal disease to obtain a real relationship between vessels [34]. Among the 2930 arteriograms available for review, the mean ICA/CCA, ICA/ECA, and outflow/inflow area ratios were larger in women than in men (p < 0.0001). In addition, there were differences in the distribution of carotid plaque, with men more likely to have maximal stenosis in the ICA and women having a greater degree of plaque within the carotid bulb. Moreover, women appeared to have more severe disease in the ECA, which would also bias the distribution of existing flow velocity through the patent ICA.
Hansen et al. showed changes in arterial wall compliance in women who display higher agerelated stiffness of their arteries and develop a higher degree of pulsatility with a higher velocity for any given blood pressure [35,36].
Moreover, the natural history of carotid atherosclerosis emerges to be divergent in women compared with men. Independently from age, the risk for stroke is higher in men than in women [37]. The risk for stroke is greater in men with similar degrees of carotid stenosis [4,29,38] due to the fact that men have greater prevalence of high-risk carotid plaques. Joakimsen et al. showed that atherosclerotic lesions in men were more instable and ultrasound characteristics showed soft and lipid-rich plaques, with more common intraplaque hemorrhage. These characteristics are associated with an increased risk for ischemic events, including myocardial infarction (MI) and stroke [39][40][41].
According to those anatomic and pathologic gender-related differences, the higher velocity profiles observed in women compared with men could be explained. It needs to be adjusted in guidelines for an accurate and proper diagnosis in carotid stenosis in the near future.

Women gender-specific parameters
• Considerably smaller carotid diameter.
• Higher arterial velocity in carotid artery: PSV and EDV.
• Different distribution of atherosclerotic plaque with greater degree of plaque within the carotid bulb and more severe disease in the ECA.
• Different arterial wall compliance with higher age-related stiffness and higher degree of pulsatility.

Carotid revascularization outcomes in women compared with men
The literature shows that women have a higher risk of perioperative adverse events during carotid revascularization. In the Asymptomatic Carotid Atherosclerosis Study (ACAS), women had a higher rate of perioperative events ( [45]. There were 9865 patients (40.6% female) who underwent CEA (n = 6492) and CAS (n = 3373). The primary end point was a composite of death, stroke, and MI at 30 days. There were no differences in age and ethnicity between genders, but males were more likely to be symptomatic (41.6% M vs. 38.6% F; p < 0.003). There was a higher prevalence of hypertension and chronic obstructive pulmonary disease in female, whereas male had a higher prevalence of coronary artery disease, history of MI, and smoking history. For disease etiology in CAS, restenosis was more common in female (28.7% F vs. 19.7% M; p < 0.0001) and radiation was higher in male (6.2% M vs. 2.6% F; p < 0.0001). Comparing by gender, there were no statistically significant differences in the primary endpoint for CEA (4.07% F vs. 4.06% M) or CAS (6.69% F vs. 6.80% M). There were no differences after stratification by symptomatology and multivariate risk adjustment. These divergent results seem to be associated with different factors. In the last decades, best medical therapy improved significantly with a wide disposability of statins and antiplatelets with beneficial results on patient outcomes [46]. Another factor seems the fact that females were underrepresented in carotid randomized controlled trials [6].
4. Do women have a higher risk of adverse events after carotid revascularization in our experience?

Study design
A database of 912 consecutive patients with symptomatic or asymptomatic carotid artery stenosis undergoing CEA or CAS for carotid revascularization in the Department of Surgery of a single Institution from 2010 to 2017 was analyzed. Carotid stenosis was ≥80% for asymptomatic or ≥50% for symptomatic patients, as detected by DUS and confirmed by CTA or MRA using NASCET criteria [29].
The choice of revascularization technique (CAS/CEA) was based on general guideline recommendations, for example, European Society for Vascular Surgery (ESVS), American College of Cardiology and American Heart Association (ACC/AHA), and the team center experience according to morphologic and clinical data indicating best suitability with the aim of performing CAS and CEA with low procedural risks. Usually, patients with unfavorable aortic arch anatomy, severe peripheral vascular disease precluding femoral access, or extremely tortuous carotid anatomy were excluded from CAS. Similarly, unstable plaque, known allergies to aspirin, clopidogrel, or contrast media, and renal insufficiency (creatinine ≥1.5 mg/dL) were considered exclusion criteria for CAS.

CAS and CEA protocols
For CAS, the patient was given dual antiplatelet therapy beginning 1 day before the procedure. All patients received a 300 mg loading dose of clopidogrel 1 hour before the procedure.
Clinical investigation included a baseline assessment of a physical examination, carotid DUS of the supra-aortic vessels, procedural angiography, and neurological assessment measured using the National Institutes of Health Stroke Scale (NIHSS) [47]. All aortic arch types were included.
All procedures were performed following a standardized protocol in the operating room, equipped with a portable imaging fluoroscopic C-arm (OEC 9900 elite; GE Medical Siemens, Waukesha, WI, USA), by a single vascular team. Two skilled operators with high volume experience (>50 CEA/CAS procedures per year as first operator) performed all procedures [48][49][50]. Iodinated or gadolinium contrast was used in patients with normal creatinine level or creatinine >1.5 mg/dL (132 mmol/L), respectively. All patients received an intravenous heparin bolus (100 units/kg heparin) to achieve intraoperative anticoagulation (activated clotting time (ACT) ≥ 250 s throughout the procedure). In our center, with increasing experience, the number of CAS increased over time allowing CEA to be used for fewer and more complex cases in recent years. All procedures were carried out via femoral access. For CEA, patients were usually maintained on aspirin therapy. CEA was performed under local anesthesia under electroencephalography (EEG) monitoring or transcutaneous oxygen saturation monitoring. Dacron or bovine pericardium graft angioplasty, or eversion endarterectomy were performed.

Procedural follow-up
Postprocedural patient evaluation was performed at the periprocedural (30 days) and at the postprocedural periods, at 3 months, 6 months, and yearly thereafter within 3 years, by a neurologist and a vascular surgeon.
Patient data were captured using a paper case report form. Symptoms status and exact information about clinical adverse events were obtained. Carotid DUS of the supra-aortic vessels was obtained to determine the degree of carotid stenosis.

Endpoints, definitions, and statistical analysis
Primary endpoints were rates of death; stroke; MI; a composite of the incidence of any stroke, MI, or death; and restenosis within 30 days. Secondary endpoints were rates of death; stroke; MI; a composite of the incidence of any stroke, MI, or death; and restenosis, during the followup period, within 36 months after the procedure.
For the composite of any stroke, MI, or death, patients might have had more than one event. For example, fatal stroke events are included in both death and stroke outcomes, and subjects might have had both an ipsilateral and a subsequent contralateral stroke. The diagnosis and quantification of restenosis was performed using carotid DUS. Carotid restenosis was set at ≥40% [51,52].
Categorical variables were reported as number and percentages. For numeric variables, minimum, maximum, mean, and standard deviation were calculated. Continuous data are presented as percentages or mean AE standard deviation (SD). Rates for comorbidities, complications, and 30-day outcomes were compared between male and female patients and between those undergoing CAS and CEA by χ 2 test. Survival, stroke, MI, and restenosis rates were calculated using Kaplan-Meier analysis to compensate for patient dropouts and were reported using current Society for Vascular Surgery (SVS) criteria [53]. Standard errors (SE) are reported in Kaplan-Meier analyses. The log-rank test was used to determine differences among patients submitted to CEA and CAS.
A value of p < 0.05 was considered statistically significant for all measurements. All analyses were performed using STATA™ (STATA Corp., version 14.0, College Station, TX, USA). Patients were required to undergo DUS and neurological examination at 1 month and subsequently at 3, 6, 12, 24, and 36 months after the procedure.
Periprocedural outcome measures in male patients compared with female patients are reported in Table 2. The death rate in the overall population was 0.3% (3/912), with no significant differences in rates between the two groups (0.4% M vs. 0.3% F) and between the two procedures (0.5% CEA vs. 0.2% CAS). Periprocedural outcomes were similar for male and female patients for 30-day death, stroke, MI, and a composite of any stroke, MI, or death rates. For long-term follow-up, 95, 91, 78, 75, and 62% of patients attended their 3-, 6-, 12-, 24-, and 36month follow-up appointments, respectively. Patients were not able to attend follow-up mainly because they moved to remote locations or because they declined further visits for personal reasons. Most recent patients could not attend their 24-and 36-month appointments yet.
Long-term outcomes were similar for male and female patients for 36-month death and stroke. Male patients experienced a slightly higher rate of MI. Female patients had higher long-term restenosis rate (11.4% F vs. 4.6% M; p < 0.05). Female patients undergoing CEA had a higher 36-month rate of stroke (14.0% F vs. 6.7% M; p < 0.05) and restenosis (11.5% F vs.   p < 0.05). No significant differences in long-term outcomes were observed between male and female patients undergoing CAS ( Table 2).
The 36-month risk of any cause mortality was 5.6 (1.6)% (mean (SD)) for male patients and 15.2 (3.3)% for female patients, with a significant difference in Kaplan-Meier estimates at 36 months between the two groups according to log-rank test. Female patients undergoing CAS experience lower 3-year freedom from death rate, with no significance difference with respect to the other groups (82.5% vs. 94.5, 94.2, and 86.4% for F-CAS, M-CAS, M-CEA, and F-CEA patients, respectively, p > 0.05; Figure 1).

Conclusion
Stroke has been shown to be more common in male than in female. Male patients are affected by first stroke on average in younger age with respect to female [8]. Stroke incidence was demonstrated to be about 30% higher in men than in women in Western European studies, but strokes in women tend to be more severe [9].
During the last decade, an extensive number of papers have been published on epidemiological differences between gender. The lower stroke incidence in female has been analyzed in systematic reviews in the last decades. Plausible reasons might be the protective role of ovarian estrogen on the cerebral circulation [11][12][13], genetic factors [15], sensitivity for antiplatelet therapy [54], lower blood pressure values [16,17], lower frequency of peripheral artery diseases [18,19], ischemic heart diseases [20][21][22], and cigarettes consumption [23,24] in female patients with respect to male patients.
Carotid revascularization outcomes in female compared with male showed that the former have a higher risk of perioperative adverse events [55]. The ACAS reported a higher rate of perioperative events for female with respect to male (3.6% F vs. 1.7% M) during CEA [42] with a lower rate of events for female (8.7% F vs. 12.1% M) treated with best medical therapy. Moreover, this study showed a lower 5-year risk reduction for female (17% F vs. 66% M). In more recent trials, carotid revascularization outcomes were similar between genders, but restenosis was more common in female [6,45].
In the study presented herein, 912 consecutive patients with symptomatic and asymptomatic carotid artery stenosis underwent CEA (44.6%) or CAS (55.4%) in a single institution. The death rate in the overall population was 0.3% (3/912), with no significant differences in rates between male and female patients (0.4% M vs. 0.3% F) and between the two procedures (0.5% CEA vs. 0.2% CAS). patients undergoing CEA had a higher 36-month rate of stroke (14.0% F vs. 6.7% M; p < 0.05) and restenosis (11.5% F vs. 2.5% M, p < 0.05). No significant differences in long-term outcomes were observed between male and female patients undergoing CAS. Our monocentric experience seems to confirm literature data as regarding female higher risk of restenosis during CEA.
These gender-associated differences should be taken into account for the treatment of carotid artery disease.
In light of this, in the near future, female anatomic and pathologic parameters should be used, such as carotid diameter which is considerably smaller in female than in male, thus determining higher arterial velocity in carotid artery. The greater degree of atherosclerotic plaque within the carotid bulb, the higher age-related arterial wall stiffness and degree of pulsatility, and the presence of artifacts increasing arterial velocity should be taken into account for an accurate diagnosis of carotid stenosis and guidelines should be adjusted consequently.
Gender subgroup analyses from large randomized trials, such as ACST-2, an international randomized trial comparing CEA with CAS for long-term stroke prevention, which planned the recruitment of 3600 patients by the end of 2019 [56][57][58], may report interesting and important results on the impact of gender on the peri-and postprocedural outcomes after carotid revascularization procedures.