English (pdf)
Article in xml format
Article references
Automatic translation
Send this article by e-mail
Cited by SciELO 
Similars in
SciELO 
Permalink
Introduction
Carotid artery endarterectomy (CEA) is the first line treatment for symptomatic carotid artery stenosis >50% and selected patients with an asymptomatic stenosis >60% with acceptable perioperative risk and a 5-year life expectancy.1 The benefit of CEA in preventing stroke is mostly counterbalanced by the high-risk baseline cardiovascular profile the typical patient presents, with an increased mortality in the long-term. The 2-year MACE rates after CEA have been described as 6-9%,2),(3 with an average cardiac-related mortality of 2.9% per year after CEA.4 Therefore, identification of patient-related risk factors for adverse outcomes following CEA is paramount.
Atherosclerosis is often a systemic disease, involving multiple vascular beds, such as the coronary arteries, arteries of the lower extremities and carotid arteries.5 Bilateral carotid occlusive disease is estimated to occur in less than 10% of patients with carotid artery disease6),(7 and is considered a high-risk condition for CEA.8 Patients with bilateral carotid disease may have a more severe burden of systemic atherosclerosis, which result in worse outcomes following CEA. Contralateral carotid disease may also pose a technical issue, as these patients may not tolerate carotid cross-clamping and may be prone to suffer intraoperative hemodynamic strokes.
The aim of this review is to provide an overview of the literature regarding the relevance of contralateral carotid occlusive disease in patients submitted to CEA, in terms of short and long-term outcomes.
A MEDLINE search was performed in order to identify articles focused on contralateral carotid occlusive disease in patients submitted to CEA. The following query was used in order to obtain the references: ("Endarterectomy"[MeSH Terms] OR "Endarterectomy, Carotid"[MeSH Terms] OR “carotid endarterectomy” [All Fields] OR “Carotid Endarterectomy” [MeSH Terms]) AND (“bilateral” [All Fields] OR “contralateral” [All Fields]). Additional articles of scientific interest for the purpose of this non-systematic review were included by cross-referencing. Obtained records were screed by two independent authors, blinded to each other, with any discordances being resolved by a third author.
Eligible studies were required to include patients submitted to carotid endarterectomy and with a documented contralateral internal carotid occlusion (CCO), as well as patients without CCO. Data of interest included the preferred anaesthesia type and the protocol regarding shunt use. Main outcomes of interest included intraoperative changes in neurological function or monitoring tests during carotid cross-clamping, adverse neurological events and mortality in the 30 days period and in the long-term. Both prospective and retrospective studies were included.
Search results
The search yielded 1839 records, with 94 being considered for inclusion after title and abstract screening. Twenty-two references were selected for inclusion after full text appraisal the former. The selection flowchart is displayed in Figure 1.
Synthesis of evidence
Intra-operative outcomes
In eight studies, most patients were intervened on under general anesthesia9)-(16, while 4 studies reported the use of general anesthesia in all patients.17)-(20 Regional anesthesia was used in the majority of patients in 4 studies21)-(24 and used in all patients in 2 studies.25),(26 Four authors did not report the type of anesthesia used (Table 1).
Table 1 Type of anesthesia and shunting protocol used by study.
| Author, year | N | Anesthesia type | Shunt protocol |
| Baker et al, 2000 (ACAS) | 1662 | NR | NR |
| Karmeli et al, 2001 | 94 | Mostly regional | selective |
| Grego et al, 2005 | 1381 | All general | routine |
| Ferguson et al, 1999 (NASCET) | 1415 | Mostly general | selective |
| Ballota et al, 2002 | 336 | All general | selective |
| Rockman et al, 2002 | 2420 | Mostly general | selective |
| Domenig et al, 2003 | 1950 | Mostly general | routine |
| Cinar et al, 2004 | 429 | All general | routine |
| Fitzpatrick et al, 2005 | 221 | Mostly general | selective |
| Dalainas et al, 2007 | 3332 | Mostly general | selective |
| Maatz et al, 2008 | 1960 | NR | routine |
| Bagaev et al, 2010 | 335 | Mostly general | routine |
| Goodney et al, 2012 | 5632 | Mostly general | Both routine and selective |
| Kretz et al, 2012 | 1212 | All general | selective |
| Faggioli et al, 2013 | 1218 | NR | Routine |
| Capoccia et al, 2014 | 1639 | Mostly general | selective |
| Ricotta et al, 2014 | 11614 | Mostly general | selective |
| Taboada et al, 2016 | 434 | All general | selective |
| Kong et al, 2017 | 301 | All general | routine |
| Pothof et al, 2017 | 15487 | Mostly general | selective |
| Turley et al, 2019 | 11948 | NR | NR |
| Clouse et al, 2019 | 80230 | Mostly general | selective |
NR: not reported
Compared with non-CCO patients, patients with CCO were more likely to be intervened on under general anesthesia in 4 studies11),(13),(15),(22, regional anesthesia one study24, while 3 studies reported no difference in the anesthesiologic strategy between the two groups.
The majority of studies where selective shunting was utilized reported higher rates of shunt insertion in patients with CCO (Table 2).
Table 2 Intraoperative outcomes in patients with CCO
| Author, year | N | Outcomes in patients with CCO |
| Intraoperative outcomes | ||
| Karmeli et al, 2001 | 94 | Higher rates of shunt insertion, based on post-clamping neurological symptoms. No difference in anesthesia type. |
| Grego et al, 2005 | 1445 | Higher rates of post-clamping EEG changes |
| Capoccia et al, 2015 | 1639 | Higher rates of shunt use, based on EEG or TCD changes. More likely to be operated on under regional anesthesia |
| Ballota et al, 2002 | 336 | Higher rates of post-clamping EEG changes and shunt use |
| Rockman et al, 2002 | 2420 | Higher rate of shunt use. More likely to be operated on under general anesthesia |
| Cinar et al, 2004 | 429 | No difference in rates of shunt use |
| Fitzpatrick et al, 2005 | 221 | No difference in rates of shunt use; More likely to be operated on under general anesthesia |
| Dalainas et al, 2007 | 3332 | Higher rate of shunt use, based on EEG changes. No difference in anesthesia type |
| Bagaev et al, 2010 | 335 | No difference in anesthesia type |
| Goodney et al, 2012 | 5632 | Higher rate of shunt use. More likely to be operated on under general anesthesia |
| Kretz et al, 2012 | 1212 | Higher rates of shunt use, based on post-clamping neurological symptoms |
| Ricotta et al, 2014 | 11614 | Higher rates of shunt use |
| Taboada et al, 2016 | 434 | Higher rate of shunt use, based on EEG changes |
| Pothof et al, 2017 | 15487 | Higher rates of shunt use. More likely to be operated on under general anesthesia |
EEG: electroencephalogram; TCD: transcranial doppler; CCO: contralateral carotid occlusion
Grego et al. found in a retrospective study of 1445 CEA cases that patients with CCO had more often electroencephalographic (EEG) changes during carotid cross-clamping compared to patients with a patent contralateral ICA, even though there were no significant differences in terms of post-operative neurological events.17 In a study comprising 3332 CEA patients, Ballota et al. denoted an increased rate of shunt insertion, based on post-clamping EEG changes.18 Three other studies where EEG was used as a neuromonitoring tool reported similar findings.19),(23),(24 These findings suggest a disruption of the Willis polygon compensation mechanism during carotid cross-clamping.
Short-term outcomes
Most of the literature focuses on the impact that CCO has on the postoperative (first 30 days) outcomes. Nineteen studies compared postoperative outcomes in patients with a CCO with patients with a patent contralateral internal carotid artery. Among these, 11 studies reported an increased incidence of postoperative neurological events in the presence of a CCO9),(12)-(16),(18),(24),(27)-(29, while 8 studies reported no differences10),(11),(19),(20),(22),(23),(25),(26 (Table 3).
Table 3 Short and Long-term outcomes in patients with CCO
| Author, year | N | Outcomes in patients with CCO |
| Short-term (30-days) outcomes | ||
| Ferguson et al, 1999 (NASCET) | 1415 | Higher incidence of stroke/death (aRR 2.2 95%CI: 1.1-4.5) |
| Ballota et al, 2002 | 336 | Higher incidence of contralateral TIA (12% vs 1%, p<0.0001) but not stroke nor mortality |
| Rockman et al, 2002 | 2420 | No difference in rates of neurologic events (3.0% vs 2.1%, p=0.34) nor mortality |
| Domenig et al, 2003 | 1950 | No difference in stroke rates (3.6% vs 1.8%, p=NS) nor mortality |
| Cinar et al, 2004 | 429 | Trend towards higher incidence of stroke (3.6%vs 0.5%, p=0.059) nor mortality |
| Fitzpatrick et al, 2005 | 221 | No difference in stroke rates (6.3% vs 2.6%, p=0.39) nor mortality |
| Dalainas et al, 2007 | 3332 | No difference in stroke rates (2% vs 1.8%, p=0.6) nor mortality |
| Maatz et al, 2008 | 1960 | Higher incidence of stroke (5.6% vs 2.1%, p=0.012). Mortality not analyzed |
| Bagaev et al, 2010 | 335 | Higher incidence of stroke in the first 24h (11% vs 3%, p=0.006) but not mortality |
| Goodney et al, 2012 | 5632 | Higher incidence of stroke (4% vs 1.8%, p=0.002) but not mortality |
| Kretz et al, 2012 | 1212 | No difference in neurological event rates (1.2% vs 1.5%; p=NS) nor mortality |
| Faggioli et al, 2013 | 1218 | Higher incidence of neurological events/death (13.5% vs 3%, p=0.001) but not mortality |
| Capoccia et al, 2014 | 1639 | Higher incidence of stroke (4.4% vs 1.2%, p=0.009) but not mortality |
| Ricotta et al, 2014 | 11614 | Higher incidence of stroke (3.1% vs 1.1%, p<0.001) but not mortality |
| Taboada et al, 2016 | 434 | No postoperative strokes in patients with CCO |
| Kong et al, 2017 | 301 | No difference in stroke rates (2.3% vs 2%, p=0.824) nor mortality |
| Pothof et al, 2017 | 15487 | Higher incidence of stroke/death (OR 2.2, 95% CI:1.4-3.6, P = .001), any in-hospital stroke (OR 2.9, 95% CI: 1.7-4.9, p<0.001), in-hospital ipsilateral stroke (OR 2.1, 95% CI: 1.1-4.0, p=0.02), in-hospital contralateral stroke (OR 7.1, 95% CI: 2.8-17.9, p<0.001). |
| Turley et al, 2019 | 11948 | Higher incidence of stroke/death (aOR 1.73 95%CI: 1.08-2.76). |
| Clouse et al, 2019 | 80230 | Higher incidence of non-ipsilateral stroke (aOR 1.9 95%CI: 1.3-2.8). Mortality not analyzed |
| Long-term outcomes | ||
| Gasecki et al, 1999 (NASCET) | 1415 | Higher rates of ipsilateral stroke (HR 2.18, 95% CI: 1.15-4.11), any stroke (HR 1.89, 95% CI: 1.01-3.53) and stroke/death (HR 1.89, 95% CI: 1.06-3.38) |
| Ballota et al, 2002 | 336 | No difference in stroke-free survival rates |
| Grego et al, 2005 | 1381 | No differences in disabling or fatal stroke or mortality |
| Baker et al, 2000 (ACAS trial) | 1662 | No difference in ipsilateral stroke |
| Fitzpatrick et al, 2005 | 221 | No difference in stroke-free survival |
| Taboada et al, 2016 | 434 | No difference in mortality |
| Kong et al, 2017 | 301 | No difference in stroke, myocardial infarction or mortality |
TIA: transient ischemic attack; CCO: contralateral carotid occlusion.
Considering the largest studies, Pothof et al. studied the influence of contralateral carotid stenosis or occlusion on perioperative outcomes in a cohort study of 15487 patients from the Vascular Study Group of New England (VSGNE) registry (USA). Multivariable regression analysis demonstrated an independent association between CCO and 30-day stroke/death (OR 2.2, 95% CI:1.4-3.6, P = .001), any in-hospital stroke (OR 2.9, 95% CI: 1.7-4.9, p<0.001), in-hospital ipsilateral stroke (OR 2.1, 95% CI: 1.1-4.0, p=0.02), in-hospital contralateral stroke (OR 7.1, 95% CI: 2.8-17.9, p<0.001), and prolonged length of stay (OR 1.7, 95% CI: 1.4-2.0, p<0.001), but not 30-day mortality (OR 1.1, 95% CI: 0.5-2.8, p=0.8). Interestingly, neither moderate (50-79%) nor severe (80-99%) contralateral carotid stenosis were associated with 30-day stroke/death or in-hospital stroke.15 These findings were corroborated by another large cohort study of patients from the Society for Vascular Surgery Vascular Registry. In this study, Ricotta et al. demonstrated that patients with CCO submitted to CEA have higher rates of perioperative major adverse cardiovascular events (4.2% vs 1.8%, p<0.001) and stroke (3.1% vs 1.1%, p<0.001).14 Turley et al., have also found in a study of 11948 CEA patients from the American College of Surgeons National Surgical Quality Initiative Project (ACS NSQIP) that CCO was associated with higher rates of postoperative stroke.29 The largest study yet is drawn from the Vascular Quality Initiative (VQI, USA), encompassing 80230 patients, where Clouse et al. found an independent association between CCO and postoperative non-ipsilateral stroke (aOR 1.9 95%CI: 1.3-2.8).16
Interestingly, no study found an association between CCO and increased post-operative mortality, which suggests that increased incidence of postoperative events is mainly due to non-fatal stroke.
Among 11 studies where selective shunting was used 6 studies reported higher rates of postoperative neurological events in patients with CCO9),(14)-(16),(18),(24, while among 7 studies where routine shunting was the standard practice, 3 studies reported a higher incidence of postoperative neurological events.12),(27),(28
General anesthesia was utilized in 11 studies, 7 of which reported increased rates of perioperative neurological events in patients with CCO.9),(12)-(16),(18 Five studies reported using mainly regional anesthesia, with only one study reporting a higher incidence of perioperative neurological events in patients with CCO.24
Meanwhile, other studies have not confirmed the association between CCO and adverse outcomes. Rockman et al. reviewed a prospective database of 2420 CEA patients. There was no significant differences in terms of perioperative neurological events between patients with and without CCO (3.0% vs 2.1%, p=0.34).22 Similarly, in a study involving 3332 patients, Dalainas et al, found no significant differences in terms of postoperative stroke rates.23 In a smaller study by Taboada et al., 434 patients were submitted to CEA, among which 40 had CCO, none of whom suffered any perioperative stroke.19
Long-term outcomes
While most studies have focused on perioperative outcomes, a few studies have presented long-term outcomes in patients with CCO (Table 3). In the study by Grego et al., at 6 years, there were no significant differences between patients with and without CCO in terms of disabling or fatal stroke or mortality. However, CCO conferred an increased risk of neurological events in the hemisphere ipsilateral to the occluded carotid artery.17 In a post-hoc analysis of the ACAS trial, patients with and without CCO had similar rates of ipsilateral stroke at 5 years after CEA (5.5% vs 5.0%, p=0.86).30 Taboada et al. did not find any significant differences in mortality between groups, with an average follow-up of 75.5 months.19 Two other studies did not find a difference between patients with a patent contralateral carotid artery or CCO in stroke-free survival.11),(18
By contrast, in a post-hoc analysis of the NASCET trial, at 2 years of follow-up patients with CCO had higher rates of ipsilateral stroke (HR 2.18, 95% CI: 1.15-4.11), any stroke (HR 1.89, 95% CI: 1.01-3.53) and stroke/death (HR 1.89, 95% CI: 1.06-3.38).31
Discussion
Contralateral carotid occlusion has long been considered a risk factor for adverse outcomes following CEA. Overall, the existing evidence in the literature points to a higher incidence of brain hypoperfusion during carotid cross-clamping in this subgroup of patients. In a study by Montisci et al., 71 patients who underwent CEA were evaluated preoperatively with magnetic resonance angiography of the circle of Willis (CoW). The presence of two or more agenesiae in the CoW was significantly associated with carotid cross-clamping intolerance.32 Another study by Banga et al found an increased incidence of immediate neurological events after CEA in patients with an isolated middle cerebral artery (incomplete anterior and posterior semicircle).33 Contralateral carotid occlusion could compromise the compensatory mechanism of CoW, analogous to agenesiae of communicating arteries. This is further suggested in a study by Pennekamp et al., where CCO was independently associated with selective shunt use, even after accounting for the morphology of the CoW.34 It is possible that preoperative CoW imaging and routine shunt use in patients with unfavorable CoW anatomies could lower the incidence of perioperative neurological events.
Although there are some conflicting results in the literature, the evidence provided by the largest studies, many of them from national registries (VSGNE, VQI, ACS NSQIP), points to a higher incidence of postoperative stroke in patients with CCO.14)-(16),(29 One pitfall is the lack of etiological description of these postoperative strokes. Hemodynamic strokes account for only 10% of all postoperative strokes after CEA.35 It is unknown if these are more prevalent in the presence of CCO. The use of shunts could theoretically avoid cerebral hypoperfusion in this group of patients. Although a Cochrane meta-analysis found no differences in terms of stroke between no shunting, routine shunting and selective shunting36, Goodney et al. reported in a cohort study of patients with CCO that the incidence of postoperative stroke was lower when surgeons used routine shunting, compared to selective shunters.13 The potential benefit of shunting based on clinical signs of post-clamping brain ischemia in patients under regional anesthesia was studied in the General Anesthesia versus Local Anesthesia for Carotid Surgery (GALA) trial. However, there was no difference in the primary endpoint, consisting of stroke, death or myocardial infarction between randomization and day 30 after CEA, between general and regional anesthesia.37
The Stenting and Angioplasty with Protection in Patients at High Risk for Endarterectomy (SAPPHIRE) trial compared CEA with carotid artery stenting (CAS) in patients with both symptomatic and asymptomatic carotid stenosis and with at least one condition considered to be high risk for surgery, including CCO. In the postoperative period there was no difference in terms of stroke/death between groups, although the authors provided no subgroup analysis for patients with CCO.38 Others authors have reported a low incidence of in-hospital neurological events after CAS in patients with CCO.39 However, this may not apply to centers with low volume of CAS cases. To date there are no randomized controlled trials comparing medical vs interventional treatment in patients with CCO. Therefore, particularly for patients with CCO, indication for carotid revascularization should be based on a case-by-case analysis.
Controversy remains regarding CCO as a long-term predictor of adverse outcomes. The NASCET trial found an association between CCO and long-term stroke and death (31) while another study found an association between CCO and later stroke in the hemisphere ipsilateral to the occlusion (17). Long-term cardiovascular mortality in patients with carotid artery disease is high, as shown in the ACAS trial, where 50% of deaths during follow-up were due to cardiac causes.40 Therefore, while outcomes in the postoperative period seem to be higher in patients with CCO, there may be a catch-up phenomenon in the long-term in patients with a patent contralateral carotid artery.
This review provides an up-to-date insight to the clinical relevance of CCO in patients submitted to CEA. It gathers a large number of studies, some of them from large databases, with an adequate level of evidence. It is nonetheless a narrative review, providing the evidence in individual studies, without a pooled estimate.
The optimal management of patients with CCO should be tailored to the individual risk of stroke on medical therapy, symptomatic status of the patient, comorbidities and life expectancy and surgeon experience, until further randomized controlled studies for this specific population.
