Author + information
- Received April 20, 2005
- Revision received August 15, 2005
- Accepted August 22, 2005
- Published online April 18, 2006.
- Rishi Gupta, MD⁎,‡,
- Alex Abou-Chebl, MD⁎,
- Christopher T. Bajzer, MD†,
- H. Christian Schumacher, MD§ and
- Jay S. Yadav, MD⁎,†,1,⁎ ()
- ↵⁎Reprint requests and correspondence:
Dr. Jay S. Yadav, Cleveland Clinic Foundation, 9500 Euclid Avenue, F25, Cleveland, Ohio 44195.
Objectives We sought to determine the frequency, predictors, and consequences of hemodynamic depression (HD) after carotid artery stenting (CAS).
Background Hemodynamic depression has been reported after carotid artery stenting CAS and carotid endarterectomy (CEA).
Methods We retrospectively analyzed data on 500 consecutive CAS procedures performed over a 5-year period. Hemodynamic depression was defined as periprocedural hypotension (systolic blood pressure <90 mm Hg) or bradycardia (heart rate <60 beats/s). Univariate and multivariate binary logistic regression models were used to determine the predictors and consequences of HD and persistent HD.
Results The mean age of the patients was 70.5 ± 10 years, and 69% were men. Hemodynamic depression occurred during 210 procedures (42%), whereas persistent HD developed in 84 procedures (17%). Features that independently predicted HD included lesions involving the carotid bulb (odds ratio [OR] 2.18 [range 1.46 to 3.26], p < 0.0001) or the presence of a calcified plaque (OR 1.89 [range 1.25 to 2.84], p < 0.002). Prior ipsilateral CEA was associated with reduced risk of HD (OR 0.35 [range 0.20 to 0.60], p < 0.0001). Patients who developed persistent HD were at a significantly increased risk of a periprocedural major adverse clinical event (OR 3.05 [range 1.35 to 5.23], p < 0.02) or stroke (OR 3.34 [range 1.13 to 9.90], p < 0.03).
Conclusions Hemodynamic depression is common after CAS, particularly in patients with a calcified plaque in the carotid bulb, but is easily treated with conventional methods. Patients who develop persistent HD are at an increased risk of periprocedural major adverse clinical events and stroke.
Carotid artery stenting (CAS) is being used more frequently as an alternative to carotid endarterectomy (CEA) in high-surgical-risk patients. Hemodynamic depression (HD) has been described after CEA owing to the manipulation of the carotid sinus (1), and there have been reports regarding the frequency and predictors of HD after CAS (2–4) and its possible association with adverse clinical outcomes (4,5). These studies involved small numbers of patients, but all concluded that HD is common with CAS. Disagreement exists as to the clinical relevance of HD and which clinical and angiographic factors may predict it. We performed this study on a large, consecutive group of patients and report the frequency, predictors, and consequences of HD in our cohort of patients.
Subjects and methods
This retrospective study includes 461 consecutive patients who underwent 500 carotid stent procedures for atherosclerotic stenoses between February 1998 and July 2003 and who were prospectively entered into our interventional database. All patients were considered to be at high surgical risk on the basis of age >80 years, prior injury to the laryngeal nerve, the presence of clinically significant coronary artery disease (abnormal stress test, congestive heart failure, more than two major arteries diseased ≥70% or requiring open-heart surgery), severe pulmonary disease, contralateral carotid occlusion, a history of prior ipsilateral CEA, neck exploration, or radiation therapy. Three interventionalists performed all of the CAS procedures. Approval for this study was obtained from the institutional review board of our hospital.
We defined HD as symptomatic or asymptomatic hypotension (systolic blood pressure [SBP] <90 mm Hg) or bradycardia (heart rate [HR]<60 beats/min), regardless of the need for vasopressor support, atropine, or a pacemaker. Vasopressor agents were given to patients in whom fluid boluses or a dose of 60 mg of oral pseudoephedrine failed. Atropine was given for patients who developed a >20 beats/min drop in their heart rate during predilation or transient asystole that did not respond to the patient coughing. Patients who required continuous vasopressor infusion after the procedure were considered to have persistent HD. This definition was selected arbitrarily to represent the population felt to have a more profound hemodynamic change in the periprocedural period.
The prospectively maintained database of all carotid procedures performed in our laboratory was retrospectively queried for the following information: patient demographics (age, gender, risk factor profile, use of beta-blockers), history of contralateral carotid occlusion, history of irradiation to the neck, history of stroke or transient ischemic attack (TIA), stent diameter, diameter of balloon used before and after dilation, arterial lumen diameter, percent stenosis of lesion, lesion location (common carotid, bulb, internal distal to bulb, and so on), presence of calcification and ulceration within the lesion, blood pressure reading recorded every 5 min via the arterial sheath, heart rate recorded every 5 min on a rhythm strip, hypotension, vasopressor support and atropine use, date and time of the procedure, and date and time of discharge. In addition, patients who developed major adverse clinical events (MACE), defined as myocardial infarction (MI), stroke, or death, were recorded within the periprocedural period. Patients with TIA, defined as resolution of a focal neurologic deficit within 24 h, were also recorded but not included as a MACE. To ensure that all periprocedural events were captured and to corroborate the data from the database, the charts and the angiograms of all the patients were also reviewed and analyzed for the same data.
The length of stay of patients was calculated from the time of the procedure to the time of hospital discharge. Sixteen patients had preplanned medical or surgical interventions to be performed following CAS and were not included in the length-of-stay analysis.
The technique of CAS has been published elsewhere (6). Various embolic protection devices were utilized during CAS procedures, including EpiFilter Wire (Boston-Scientific, Fremont, California), Accunet (Guidant Corp., Indianapolis, Indiana), Angioguard (Cordis Corp., Miami Lakes, Florida), and Percusruge (Medtronic Inc., Minneapolis, Minnesota). In addition, all patients had creatine phosphokinase, creatine phosphokinase-MB fraction, and troponin levels drawn before and after the procedure. All CAS procedures were performed without general anesthesia and with continuous arterial pressure and electrocardiographic monitoring. Intravenous anxiolytics and analgesics were used rarely in very anxious patients; all other patients received local access site anesthesia only. All patients were instructed to take their usual oral antihypertensive agents the morning of the procedure. Blood pressure was monitored continuously through the arterial sheath while heart rate was continuously monitored via a rhythm strip. All patients had a SBP >100 mm Hg at the time of insertion of the arterial sheath. Two large-bore peripheral intravenous catheters were placed in every patient, through which normal saline (NS) was infused from the beginning of the procedure; the rate of infusion was adjusted based on a patient’s known left ventricular ejection fraction and fluid status. Atropine and other anti-cholinergic agents were not given prophylactically to any patient nor did patients receive central venous lines or pacemakers prophylactically. Patients were instructed before the procedure to cough hard when asked to, which they practiced before the start of the procedure. If bradycardia or a >20 beats/min decrease in baseline HR occurred with lesion pre-dilation, patients were treated with 0.5 mg of atropine before stent deployment and post-dilation. If hypotension occurred, patients received a 250 to 500 cc bolus of NS. If the hypotension or bradycardia was persistent, patients were treated with additional fluids, atropine, or 2 to 6 μg/kg boluses of norepinephrine. If asystole occurred, patients were asked to cough several times, and the angioplasty balloon was rapidly deflated.
Continuous NS infusions were continued in all patients and were adjusted to maintain normotension. Patients with asymptomatic, persistent hypotension received oral pseudoephedrine 30 to 60 mg two to three times daily. Those with continued hypotension despite fluids and pseudoephedrine generally received vasopressor support with dopamine. In all cases, oral antihypertensive agents were not resumed except in those patients who had significant hypertension or those who were at risk for hyperperfusion syndrome and intracerebral hemorrhage. Strict bed rest was maintained in all patients until the following morning, when orthostatic vital signs were obtained; only those patients without orthostatic hypotension and who were otherwise doing well were discharged. Patients with orthostatic hypotension were observed for another 24 h or until the orthostasis resolved.
Baseline characteristics were compared among the HD, persistent HD, and non-HD groups using the Fisher exact test for categorical variables and the Student ttest for continuous variables. A p value of <0.05 was considered significant. A multivariate binary logistic regression analysis was performed to determine which clinical and angiographic variables would independently predict the development of HD and persistent HD during CAS. In addition, a binary logistic regression model was constructed to determine the variables that independently predicted patients developing a MACE, stroke, MI, death, or TIA. Variables with a p value <0.10 were entered into the logistic regression model, but only variables with a p value <0.05 were considered statistically significant as independent predictors.
A total of 500 CAS procedures were performed on separate occasions in 461 patients. The baseline characteristics of the cohort are shown in Table 1.The overall technical success rate was 99.6%, with two failed procedures due to inability to cross the lesion. Only one patient (0.2%) required a pacemaker after the procedure for third-degree atrioventricular block. Periprocedural HD was noted in 210 procedures (42%). Bradycardia developed during 136 procedures (27%), with 116 (85% of patients with bradycardia) of these occurring during the procedure and 20 (15% of patients with bradycardia) after the procedure. Isolated bradycardia without hypotension occurred in 67 procedures (13%). Hypotension occurred in 134 procedures (27%); 79 (59% of all hypotension patients) developed during the procedure and 55 (41% of all hypotension patients) after the procedure. Vasopressor treatment (one-time bolus or continuous infusion) was required in 89 procedures (18%), whereas atropine was given in 89 procedures (18%). The combination of atropine and vasopressor support was required in 32 procedures (7%).
Persistent HD developed in 84 cases (17% of all interventions but 40% of all HD patients). Patients who developed persistent HD were more likely to have had intraprocedural HD (odds ratio [OR] 1.9 [range 1.6 to 2.1], p < 0.0001). Bradycardia was evident in 41 patients with persistent HD (49%). Patients received intravenous vasopressors for a mean time of 15.9 ± 16.4 h (median = 12 h, range 1 to 96 h).
Hemodynamic depression occurred significantly less often in smokers and in patients with a previous CEA compared with nonsmokers or patients with de novo lesions, respectively (Table 1). A history of diabetes mellitus (DM) and use of beta-blockers were additional variables found to reduce the risk of developing persistent HD. None of the remaining clinical baseline characteristics was associated with HD or persistent HD. Atherosclerotic plaques located at the carotid bulb and plaques showing calcification or ulceration on angiography were significantly associated with HD and persistent HD, whereas stenosis grade, lesion length, eccentric plaque location, balloon/artery ratio, and stent diameter were not (Table 1). Our results were used to construct a model to predict the occurrence of HD. Multivariate binary logistic regression analysis revealed that de novo lesions, plaque location at the carotid bulb, and plaque calcification were independent variables associated with HD (Table 2).In patients who developed persistent HD, plaque ulceration was additionally associated, whereas DM was found to independently reduce the risk.
In a binary logistic regression model, patients with persistent HD were at a significantly higher risk of developing a MACE (OR 3.05 [range 1.35 to 5.23], p < 0.02) or stroke (OR 3.34 [range 1.13 to 9.90], p < 0.03). No clinical or angiographic variable was predictive of MI or death. In a multivariate analysis, patients with ulcerated lesions were at a significantly higher risk of developing a TIA (OR 8.7 [range 1.67 to 15.5], p < 0.01). Patients who developed HD had significantly longer hospitalizations compared with patients without HD (3.4 ± 6.5 days vs. 1.8 ± 1.9 days, p < 0.0001).
The carotid bulb baroreceptors mediate changes in blood pressure and heart rate via the medullary vasomotor nuclei in order to maintain a hemodynamic steady state and adequate cerebral perfusion. Distension of the carotid bulb leads to stimulation of the baroreceptors, which in turn causes reflex inhibition of adrenergic output to the peripheral vasculature and increased cardiac parasympathetic stimulation; hypotension and bradycardia are the consequences (7). Early experiences with CEA showed that transient baroreceptor dysfunction was common (8), and more recent experience has shown that postoperative hypertension, and not hypotension, may be associated with higher rates of stroke or death (9). With CAS, angioplasty and distension of the carotid bulb causes baroreceptor stimulation and HD. This has been reported to occur in 19% to 68% of patients, with unclear clinical consequences (4,10). Because not all patients have HD with CAS, studies inquiring into possible predisposing factors have suggested that older age (11), history of coronary artery disease (10), larger balloon diameters (4), lesion location (3), contralateral carotid stenosis >60% (3), or elevated SBP >180 mm Hg before the procedure (5) may predispose patients to HD. Each of these studies reported predictors based on a small number of patients relative to the number of variables considered in the analysis. Our analysis shows that patients who have lesions involving the carotid bulb or heavy calcification are at a higher risk of developing HD and persistent HD. Other studies have confirmed that lesions involving the carotid bifurcation would predict the presence of HD after CAS (3,12). This is likely to be due to the higher concentration of baroreceptors at this location (12).
Patients who had a restenosis after a prior ipsilateral CEA were less likely to develop HD in our cohort, a finding observed by others (2,13). This is likely from the reduced number of baroreceptors in the carotid bulb after CEA. In addition, scar formation after endarterectomy causes a reduction in arterial compliance and may require more distension for the baroreceptors to be stimulated (14). Patients with a history of smoking were also at a reduced risk from developing HD and persistent HD in our univariate analysis, but this was not found to be an independent risk factor in our logistic regression model. It is plausible for smokers to be protected from HD during CAS, as smoking is known to increase sympathetic tone through the release of epinephrine and norepinephrine, thereby raising blood pressure and heart rate (15). Habitual smoking may impair the baroreceptors through the reduction of arterial compliance of the entire vasculature, which would reduce baroreceptor stimulation (16). Diabetes mellitus was found to be an independent variable that reduced the risk of patients developing persistent HD. It is well known that DM affects cardiovascular autonomic nerves (17). Early observations revealed that patients with DM had an attenuated parasympathetic response with elevations in arterial pressure (18). Since then, studies have shown that the cardiovagal limb of the baroreflex is impaired in patients with DM, whereas sympathetic output may be normal to slightly attenuated (19). This observation is consistent with what we have found, as patients with DM were protected from persistent HD, likely because of an impaired ability to develop reflex bradycardia during balloon inflation and stent deployment.
Another unique finding with our analysis was the observation that patients with a calcified plaque were at a significant risk of developing HD, which predisposed them to adverse cerebrovascular events. The presence of an ulcerated plaque, on the other hand, did not independently predict HD but did predict persistent HD, which in turn was associated with a significant risk of a MACE. This is consistent with the risks associated with plaque excision by endarterectomy. Histopathologic studies after CEA have shown that calcified plaques are more likely to be “stable,” with a low risk of thromboembolism or plaque rupture in the carotid artery (20). In contrast, ulcerated plaques, which may be more friable, have a higher rate of rupture and thus a higher incidence of embolism and strokes in the perioperative period (21). This is the likely explanation for ulcerated lesions being an independent predictor of patients developing TIAs perioperatively. Plaque morphology has not been rigorously studied after CAS because of the lack of after-surgical specimens. The dichotomy between CAS and CEA in MACE risk based on plaque calcification may be explained by the fact that calcified plaques may require higher angioplasty balloon pressures for lesion dilation. In addition, thick calcified plaques may transmit the balloon pressure more efficiently to the baroreceptors, increasing the stretching of the baroreceptors (22). These two factors may result in more profound HD, which we have found to be associated with adverse events.
There has been considerable debate in the literature with regard to HD as a predictor of adverse outcomes periprocedurally. One group reported that patients with large fluctuations (>107 mm Hg change) in SBP after CAS were at a significantly higher risk of stroke, whereas patients with smaller fluctuations (<50 mm Hg) did not experience an adverse neurologic event (5). Another group found that patients with hypotension were at increased risk of in-hospital minor strokes and death at six months when compared with patients who did not have hypotension (4). Our experience with a large cohort reveals that persistent HD, and not HD itself, is a risk factor for the development of adverse cerebrovascular events. Embolization occurs in all carotid stenting procedures, and embolic protection devices markedly reduce the number of particles that reach the brain (23). Not all emboli are eliminated, however, and hypotension may impair the washout of emboli in the distal intracranial circulation and increase the risk of clinically evident cerebral ischemia (24).
Consideration has been given by some authors to aggressively treating or pre-treating hypotension or bradycardia periprocedurally. Leisch et al. (3) showed that pre-treatment of patients who developed asystole after the first balloon inflation with atropine prevented a second episode of bradycardia or asystole in 37 of 39 patients (95%). No patient in their cohort required transvenous pacing. Another group (25) utilized transvenous pacing prophylactically in 33 patients in order to prevent bradycardia and hypotension. Interestingly, 23 of the patients (62%) had an activation of the transvenous pacemaker, confirming how common HD is with CAS. Of 500 consecutive patients treated at our center, only one patient required a pacemaker; all others were managed medically. Because profound vasodilation is a major component of the carotid bulb reflex, volume expansion may prevent or attenuate the hypotensive response. We pre-treated, with volume expansion, all of our patients except those with known severely impaired left ventricular function. None of our patients failed to respond to this treatment, alone or in combination with vasopressors and atropine. On the basis of our experience, we feel that the use of prophylactic transvenous pacing is not warranted and may increase complications such as ventricular arrhythmias or perforation (26). Pharmacologic pre-treatment is also not benign. The routine use of atropine, especially in an elderly population with coronary artery disease, may lead to potential complications, including confusion, urinary retention (27), and arrhythmias (28). To date, no study has been performed to determine whether aggressively treating patients who develop “mild” HD may prevent persistent HD and thus reduce potential adverse vascular events.
There are some weaknesses in our study. The first is that we are unable to analyze the effect of patients taking their blood pressure medications the day of the procedure. This may have contributed to the presence or duration of HD in some patients, although an analysis of patients taking beta-blockers did not have an effect on the multivariate analysis. Second, there were some patients who had a baseline HR near the threshold set for inclusion as HD in our study. In studying the persistent HD group, we excluded any patient with isolated bradycardia and focused on patients requiring continuous vasopressor therapy. Third, some patients with hematomas required transfusions and may have had drops in their blood pressure as a result of blood loss. The majority of these patients became hypotensive during the procedure, and thus we included all of these patients in the HD and persistent HD analysis. Last, we did not record the balloon inflation pressures in the database. We do present the maximum balloon diameters used before and after dilation, however. It is possible that some lesions were not fully dilated to the maximum balloon diameter. Despite these shortcomings, we feel this analysis can aid clinicians performing these procedures in predicting patients at risk of HD or persistent HD.
The development of HD with CAS is common. These hemodynamic changes are more likely to occur in patients with calcified and ulcerated de novo lesions located at the carotid bulb, whereas patients with a prior ipsilateral CEA are less likely to develop HD. Hemodynamic depression and persistent HD can be easily managed with conventional pharmacologic agents and volume expansion. Persistent HD is associated with adverse clinical events in high-surgical-risk patients. Further studies are required to determine whether pre-treating high-risk patients with fluids, vasopressors, or anti-cholinergics may prevent persistent HD and subsequently reduce the risk of adverse vascular events.
The authors would like to thank Lynn Oster, RN, and Patricia McMahon, RN, for their efforts, without which this study would not have been possible.
↵1 Dr. Yadav is the inventor of the Angioguard embolic protection device that was used in some of the carotid stent procedures.
- Abbreviations and Acronyms
- carotid artery stenting
- carotid endarterectomy
- diabetes mellitus
- hemodynamic depression
- heart rate
- major adverse clinical events
- myocardial infarction
- normal saline
- odds ratio
- systolic blood pressure
- transient ischemic attack
- Received April 20, 2005.
- Revision received August 15, 2005.
- Accepted August 22, 2005.
- American College of Cardiology Foundation
- Qureshi A.I.,
- Luft A.R.,
- Sharma M.,
- et al.
- Howell M.,
- Krajcer Z.,
- Dougherty K.,
- et al.
- Yadav J.S.,
- Roubin G.S.,
- Iyer S.,
- et al.
- Mlekusch W.,
- Schillinger M.,
- Sabeti S.,
- et al.
- Grassi G.,
- Seravalle G.,
- Calhoun D.A.,
- et al.
- Mancia G.,
- Groppelli A.,
- Di Rienzo M.,
- Castiglioni P.,
- Parati G.
- Lloyd-Mostyn R.H.,
- Watkins P.J.
- Bennett T.D.,
- Hosking D.J.,
- Hampton J.R.
- McDowell T.S.,
- Chapleau M.W.,
- Hajduczok G.,
- Abboud F.M.
- Hunt J.L.,
- Fairman R.,
- Mitchell M.E.,
- et al.
- Ferguson G.G.,
- Eliasziw M.,
- Barr H.W.,
- et al.,
- North American Symptomatic Carotid Endarterectomy Trial (NASCET) Collaborators
- Dangas G.
- Kastrup A.,
- Groschel K.,
- Krapf H.,
- Brehm B.R.,
- Dichgans J.,
- Schulz J.B.
- Murphy J.J.