Urgent Carotid Endarterectomy Is Safe in Patients with Few Comorbid Medical Conditions
Article Outline
Recent reports from single institutions have confirmed the efficacy of carotid endarterectomy (CEA) performed in the urgent or emergent setting, although with higher perioperative mortality and morbidity. We determined the results of urgently performed CEA in academic and community hospitals and whether patient or hospital factors affected outcome. The records of patients undergoing CEA in all nonfederal hospitals in the state of Connecticut between 1992 and 2002 were reviewed, and symptomatic patients who presented in an urgent or emergent fashion were compared to patients treated electively. Multivariable logistic regression was used to determine the effect of patient risk factors on perioperative mortality, stroke, and cardiac complications. Patients undergoing urgent CEA (n = 764, 6.3%) had higher perioperative mortality (2.0% vs. 0.3%, p < 0.0001) and stroke (2.9% vs. 1.1%, p < 0.0001) but not cardiac complications (3.0% vs. 2.2%, p = 0.14) compared to patients undergoing elective CEA (n = 11,312). Patients undergoing urgent CEA and with high rates of associated comorbidity had a higher risk of perioperative mortality (7.8% vs. 0.4, p = 0.001), stroke (10.9% vs. 0.8%, p = 0.0002), and cardiac complications (14.1% vs. 0.8%, p < 0.0001) compared to patients presenting urgently but with little comorbidity. Perioperative mortality was associated with performance of the procedure in hospitals with low bed capacity (odds ratio [OR] = 4.6, p = 0.01). Perioperative stroke was associated with renal insufficiency (OR = 5.3, p = 0.04). Perioperative cardiac complications were associated with diabetes (OR = 2.6, p = 0.03) and performance in hospitals with low bed capacity (OR = 5.0, p < 0.01). Urgent admission was associated with age ≥80 (OR = 1.2, p = 0.04), renal disease (OR = 1.8, p = 0.05), and cardiac disease (OR = 1.3, p < 0.01). Urgently performed CEA has higher perioperative mortality and stroke compared with electively performed cases. However, the subset of patients with low rates of associated comorbid medical conditions but urgently needing CEA is associated with low rates of perioperative complications. Patients with severe associated comorbid medical conditions who present urgently for CEA may form a high-risk group of patients to be considered for referral to large treatment centers or possibly alternative therapy.
Introduction
Elective carotid endarterectomy (CEA) is the established standard of care for symptomatic and asymptomatic patients with high-grade carotid artery stenosis. Multiple randomized controlled trials have proven the efficacy of CEA in long-term secondary stroke prevention with acceptably low perioperative mortality and morbidity.1, 2, 3, 4, 5 However, intervention in symptomatic patients who present urgently or emergently remains controversial. For example, the optimal timing of CEA after an ipsilateral ocular or hemispheric event is still not established. In the 1960s, poor results were obtained after urgent interventions in symptomatic patients, and this increased perioperative mortality and morbidity was attributed to the conversion of an ischemic infarct into a disabling hemorrhagic stroke.6, 7 Additional reports of increased perioperative morbidity after early CEA led to recommendations to delay surgery for 4 weeks if possible.8, 9, 10, 11, 12, 13 However, recent reports suggest the benefit of early CEA in preventing recurrent stroke in patients with good neurological recovery and absence of hemorrhage on computed tomographic scan.14, 15 Pooled analysis of the European and North American Symptomatic Carotid Endarterectomy Trial (NASCET) data has suggested a benefit for CEA in symptomatic patients randomized within 2 weeks of symptoms.16 However, optimal timing for performance of CEA in symptomatic patients is not clearly established.
In addition, there are scant data concerning the impact of patient risk factors and hospital-associated factors on outcomes after CEA performed in an urgent or emergent manner. Since several patient risk factors, such as renal insufficiency and cardiac disease, are linked to worse perioperative and long-term outcome after electively performed CEA,17, 18, 19 it is important to determine whether these factors are also associated with increased adverse outcomes after urgently performed CEA.
Patients presenting in an urgent or emergent manner represent a high-risk group of patients, the treatment of whom is currently controversial. Carotid angioplasty and stenting (CAS) is an alternative method to treat carotid stenosis in patients considered to be high-risk for surgical complications; however, recent multicenter reports have demonstrated that CAS is not a perfect treatment for all high-risk patients, with different levels of risk and benefit for subgroups of patients;20, 21, 22 some high-risk groups, such as octogenarians, have been demonstrated to have improved outcomes if treated with CEA.22 Since several select series have reported increased adverse outcome after urgently performed CEA, we determined whether some groups of urgently presenting patients have acceptable risk of perioperative complications following CEA in community practice, clarifying the role of this generally low-risk procedure in higher-risk circumstances.
Methods
A database consisting of patient discharge records from all acute-care, nonfederal Connecticut hospitals is maintained by Chime, Inc. (www.cthosp.org; Wallingford, CT). The Connecticut Hospital Association (CHA) Chime Data Program has established and maintains a proprietary health-care information system that incorporates statewide clinical, financial, and patient demographic data dating back to 1980. Reports containing selected variables are available on a fee-for service basis. A previously published algorithm was used to select all CEA procedures performed between 1992 and 2002 from the Chime database.17, 18, 19
Discharge records were selected if they contained the diagnosis-related group (DRG) 5 (extracranial vascular procedures), the International Classification of Diseases, 9th revision, Clinical Modification (ICD-9-CM), procedure code 38.12 (head/neck endarterectomy), and the ICD-9-CM principal diagnosis code 433.xx (occlusion and stenosis of precerebral arteries). The accuracy of the algorithm in this database has previously been verified.17, 18, 19 The hospital records of each CEA procedure were independently reviewed by the CHA and designated as elective, urgent, or emergent according to the physician notes that explicitly specified the nature of the admission; records without designation were excluded from the analysis. Outcomes that were analyzed included in-hospital death, perioperative stroke (ICD-9 997.0: nervous system complications, iatrogenic cerebrovascular infarction, and hemorrhage), and perioperative cardiac complications (ICD-9 997.1: cardiac arrest, insufficiency, cardiorespiratory failure, heart failure, and arrhythmias, during or resulting from a procedure).
Patient demographics identified were age, gender, race, presence of cardiac disease, chronic obstructive pulmonary disease (COPD), diabetes, hypertension, and renal disease. Patients were considered symptomatic if they had a history of amaurosis fugax, transient ischemic attack (TIA), or stroke; patients could have more than one presenting symptom.
A complexity score was used to classify the aggregate severity of patient-associated comorbid conditions. The complexity score in this database is the All Patient Refined Diagnosis Related Groups (APR-DRG) risk mortality variable (3M Health Information Systems, St. Paul, MN); this variable considers patient-associated factors that predict clinical outcome and is stratified into four classes: 1, minor; 2, moderate; 3, major; or 4, extreme severity of illness. Classes 1 and 2 were considered low scores, whereas classes 3 and 4 were considered high scores. This variable was available in the Chime database starting in 1997.
Hospital bed capacity was stratified according to the lower third, middle third, and upper third of all hospital bed capacities in the state; low is ≤132, medium is 133-279, and high is ≥280 beds. Annual hospital CEA volume was defined as low, medium, or high if the annual volume of all CEAs, including elective, urgent, and emergent cases, was ≤10, 10-49, or ≥50, respectively. Other hospital-based factors included hospital catchment area (less than or greater than 75,000 people), hospital site (urban, suburban, or rural), academic status (academic or community), and presence or absence of surgical residents.
The effects of patient comorbidities and hospital-associated factors on outcome were analyzed by univariable analysis; p values were calculated using chi-squared or Fischer's exact test. Logistic regression was used for multivariable analysis. All tests were two-tailed, and p ≤ 0.05 was considered statistically significant (Statview 5.0; SAS Institute, Cary, NC).
Results
There were 12,076 CEAs performed between 1992 and 2002 in the state of Connecticut and identified as being performed in an elective, urgent, or emergent fashion; 9,964 (82.5%) were performed in asymptomatic patients and 2,112 (17.5%) were performed in symptomatic patients. There were 764 (6.3%) procedures performed in an urgent or emergent fashion in patients who were symptomatic; these are the subject of this study. The demographic factors of these patients and the characteristics of the hospitals in which CEA was performed are listed in Table I. Most urgent procedures (>95%) were performed in hospitals with medium or high elective procedural volume (Table I). Perioperative mortality and stroke, but not cardiac complications, were significantly higher in those patients undergoing urgently performed CEA compared to elective CEA (Table II).
Table I. Analysis of patient demographics, risk factors, and treating hospital characteristics for 764 patients undergoing urgently performed CEA
| Age (years) | |
 <65 | 185 (24.2%) |
| 65-69 | 131 (17.1%) |
| 70-79 | 295 (38.6%) |
| ≥80 | 153 (20%) |
| Male | 465 (60.9%) |
| Caucasian | 728 (95.3%) |
| Cardiac disease | 162 (21.2%) |
| COPD | 97 (12.7%) |
| Diabetes | 202 (26.4%) |
| Hypertension | 540 (70.7%) |
| Renal disease | 14 (1.8%) |
| Symptom | |
| Amaurosis fugax | 135 (17.7%) |
| TIA | 223 (29.2%) |
| Stroke | 464 (60.7%) |
| Hospital bed capacity | |
| Low (≤132) | 125 (16.4%) |
| Medium (133-279) | 273 (35.7%) |
| High (≥280) | 366 (47.9%) |
| Catchment area | |
| <75,000 | 219 (28.7%) |
| ≥75,000 | 545 (71.3%) |
| Site | |
| Urban | 562 (73.5%) |
| Suburban | 158 (20.7%) |
| Rural | 44 (5.8%) |
| Academic setting | |
| Academic | 259 (33.9%) |
| Community | 505 (66.1%) |
| Residents | |
| Yes | 501 (65.6%) |
| No | 263 (34.4%) |
| CEA volume | |
| Low (≤10) | 35 (4.6%) |
| Medium (10-49) | 391 (51.2%) |
| High (≥50) | 338 (44.2%) |
Table II. Perioperative outcome for patients undergoing urgently performed CEA compared to patients having electively performed CEA
| n | Mortality | Stroke | Cardiac complication | |
|---|---|---|---|---|
| Urgent | 764 | 1.96% | 2.88% | 3.01% |
| Elective | 11,312 | 0.33% | 1.13% | 2.19% |
| pa | <0.0001 | <0.0001 | 0.14 |
aChi-squared analysis. |
The effect of patient comorbid conditions and hospital factors on perioperative outcomes after urgently performed CEA was examined using univariable analysis (Table III). Diabetes was associated with an increased risk of perioperative cardiac complications (p = 0.01), and renal disease was associated with increased risk of perioperative stroke (p = 0.05). Performance of the procedure in a hospital with low bed capacity was associated with increased risk of perioperative mortality (p < 0.01) and cardiac complications (p < 0.01) and a trend toward an increased stroke risk (p = 0.07). Performance of an urgently performed CEA in a hospital in a rural setting was associated with increased risk of perioperative cardiac complications (p < 0.01). There was no effect of annual hospital CEA volume on outcome in patients admitted urgently. In addition, there was no difference in outcome after urgent CEA when performed in hospitals performing more than 100 procedures per year compared to hospitals performing fewer than 100 procedures per year (data not shown), a threshold proposed by the Leapfrog Initiative that may be associated with improved outcomes.23
Table III. Analysis of comorbid conditions and hospital characteristics with perioperative outcomes in patients having urgently performed CEA
| Mortality | p | Stroke | p | Cardiac | p | |
|---|---|---|---|---|---|---|
| Age (years) | ||||||
| <65 | 1.1% | 0.25 | 2.1% | 0.55 | 2.2% | 0.23 |
| 65-69 | 1.5% | 2.3% | 3.8% | |||
| 70-79 | 1.7% | 2.7% | 2% | |||
| >80 | 3.9% | 4.6% | 5.2% | |||
| Gender | ||||||
| Male | 2.1% | 0.64 | 2.4% | 0.29 | 3% | >0.99 |
| Female | 1.7% | 3.7% | 3% | |||
| Race | ||||||
| White | 1.8% | 0.15 | 3% | >0.99 | 3% | >0.99 |
| Other | 5.6% | 0 | 2.8% | |||
| Cardiac disease | ||||||
| Yes | 0.6% | 0.21 | 1.8% | 0.59 | 4.3% | 0.29 |
| No | 2.3% | 3.1% | 2.6% | |||
| COPD | ||||||
| Yes | 4.1% | 0.10 | 4.1% | 0.43 | 4.1% | 0.52 |
| No | 1.6% | 2.7% | 2.8% | |||
| Diabetes | ||||||
| Yes | 2.5% | 0.56 | 3.5% | 0.62 | 5.4% | 0.01∗ |
| No | 1.8% | 2.7% | 2.1% | |||
| Hypertension | ||||||
| Yes | 1.9% | 0.78 | 2.2% | 0.09 | 3.3% | 0.49 |
| No | 2.2% | 4.5% | 2.2% | |||
| Renal disease | ||||||
| Yes | 7.4% | 0.24 | 14.2% | 0.05∗ | 3.1% | >0.99 |
| No | 1.9% | 2.7% | 0 | |||
| Hospital bed capacity | ||||||
| Low | 5.6% | <0.01∗ | 4.8% | 0.07 | 8% | <0.01∗ |
| Medium | 1.1% | 1.1% | 1.5% | |||
| High | 1.4% | 3.5% | 2.5% | |||
| Catchment area | ||||||
| <75,000 | 3.2% | 0.15 | 3.2% | 0.81 | 4.1% | 0.25 |
| ≥75,000 | 1.5% | 2.8% | 2.6% | |||
| Site | ||||||
| Urban | 2.3% | 0.82 | 2.3% | 0.91 | 2.8% | <0.01∗ |
| Suburban | 2.5% | 2.5% | 1.3% | |||
| Rural | 1.8% | 3% | 11.3% | |||
| Academic setting | ||||||
| Academic | 1.5% | 0.78 | 3.1% | 0.82 | 3.1% | 0.99 |
| Community | 2.2% | 2.7% | 3% | |||
| Residents | ||||||
| Yes | 1.4% | 0.17 | 3% | 0.99 | 2.8% | 0.66 |
| No | 3% | 2.7% | 3.4% | |||
| CEA volume | ||||||
| Low | 2.9% | 0.38 | 2.9% | 0.35 | 2.9% | 0.87 |
| Medium | 2.6% | 2.1% | 3.3% | |||
| High | 1.2% | 3.8% | 2.6% | |||
∗Denotes statistical significance, P value ≤ 0.05. |
Since patients undergoing urgent CEA have higher rates of perioperative mortality and stroke compared to electively admitted patients (Table II), we determined whether patients admitted under urgent conditions had higher rates of aggregate comorbidity compared to patients admitted under elective conditions. To determine aggregate patient comorbidity and minimize potential bias due to undercoding individual risk factors, we determined the distribution of complexity scores in these patients. Patients who had CEA performed urgently had a greater percentage of high complexity scores (score 3 or 4) compared to patients having CEA performed electively (19.9% vs. 5.8%, p < 0.0001; Table IVA). These results suggest that patients having CEA performed urgently have higher rates of comorbidity compared with patients having CEA performed electively.
Table IV. High aggregate comorbidity in patients presenting urgently for CEA and effect on outcome
| A: Increased frequency of high complexity scores in patients undergoing urgent CEA compared to patients undergoing elective CEA | ||||
|---|---|---|---|---|
| Urgent | Elective | |||
| Complexity score | n | % | n | % |
| 1 | 99 | 30.8% | 4,809 | 73.9% |
| 2 | 158 | 49.2% | 1,311 | 20.2% |
| 3 | 52 | 16.2% | 325 | 4.9% |
| 4 | 12 | 3.7% | 57 | 0.9% |
| B: Lower rates of perioperative morbidity and mortality in patients with low (class 1 or 2) compared to higher rates in patients with high (class 3 or 4) complexity scores | |||
|---|---|---|---|
| Complexity score | Mortality | Stroke | Cardiac complication |
| 1 | 0% | 1.0% | 0% |
| 2 | 0.6% | 0.6% | 1.3% |
| 3 | 5.8% | 3.9% | 9.6% |
| 4 | 16.7% | 41.6% | 33.3% |
| pa | <0.0001 | <0.0001 | <0.0001 |
aChi-squared analysis. |
We next determined whether there was a difference in outcome in patients undergoing urgent CEA with either low or high rates of comorbidity. The majority of patients (80%) with urgently performed CEA had low complexity scores (scores 1 or 2, Table IVA) and similar perioperative mortality compared to patients undergoing an electively performed CEA (0.4% vs. 0.3%, p = 0.57; Table IVB); there were also no significant differences in perioperative stroke (0.8% vs. 1.1%, p > 0.99) or cardiac complications (0.8% vs. 2.2%, p = 0.19) compared to elective patients. However, some patients (20%) undergoing urgent CEA had high complexity scores (scores 3 or 4) and were also associated with increased perioperative mortality compared to patients undergoing urgent CEA but with low complexity scores (7.8% vs. 0.4%, p = 0.001). Similarly, perioperative stroke (10.9% vs. 0.8%, p = 0.0002) and cardiac complications (14.1% vs. 0.8%, p < 0.0001) were higher in patients who were treated urgently with high complexity scores. These results suggest that the increased perioperative mortality and morbidity after CEA performed urgently (Table II) reflects the subset of patients admitted urgently and having a large number of medical comorbid conditions.
Multivariable logistic regression was used to determine the effects of patient- and hospital-associated risk factors on outcome after urgently performed CEA (Table V). Low hospital bed capacity was the only factor associated with increased risk of perioperative mortality (odds ratio [OR] = 4.55, p = 0.01). Renal disease was associated with increased risk of perioperative stroke (OR = 5.33, p = 0.04); diabetes (OR = 2.62, p = 0.03) and low hospital bed capacity (OR = 5.03, p < 0.001) were associated with increased risk of perioperative cardiac complications.
Table V. Multivariable analysis of patient and hospital factors associated with perioperative outcome after urgently performed CEA
| p | OR | 95% Confidence interval | |
|---|---|---|---|
| Perioperative mortality | |||
| Age ≥80 | 0.11 | 2.43 | 0.81-7.28 |
| Nonwhite | 0.07 | 4.59 | 0.89-23.8 |
| Cardiac disease | 0.19 | 0.25 | 0.03-1.9 |
| COPD | 0.12 | 2.60 | 0.78-8.7 |
| Diabetes | 0.51 | 1.46 | 0.47-4.5 |
| Hypertension | 0.61 | 0.75 | 0.25-2.3 |
| Renal disease | 0.28 | 3.47 | 0.37-32.7 |
| Community hospital | 0.84 | 0.87 | 0.23-3.3 |
| Low hospital bed capacity | 0.01∗ | 4.55 | 1.3-14.9 |
| Perioperative stroke | |||
| Age ≥80 | 0.23 | 1.79 | 0.69-4.6 |
| Nonwhite | 0.97 | <0.01 | - |
| Cardiac disease | 0.31 | 0.52 | 0.15-1.8 |
| COPD | 0.47 | 1.5 | 0.49-4.7 |
| Diabetes | 0.48 | 1.4 | 0.55-3.6 |
| Hypertension | 0.09 | 0.48 | 0.20-1.1 |
| Renal disease | 0.04∗ | 5.33 | 1.07-26.6 |
| Community hospital | 0.42 | 0.67 | 0.25-1.8 |
| Low hospital bed capacity | 0.21 | 1.99 | 0.67-5.8 |
| Perioperative cardiac complication | |||
| Age ≥80 | 0.16 | 1.92 | 0.78-4.7 |
| Nonwhite | 0.89 | 0.87 | 0.11-7.1 |
| Cardiac disease | 0.45 | 1.44 | 0.57-3.6 |
| COPD | 0.47 | 1.52 | 0.49-4.7 |
| Diabetes | 0.03∗ | 2.62 | 1.12-6.2 |
| Hypertension | 0.48 | 1.44 | 0.52-4 |
| Renal disease | 0.98 | <0.1 | - |
| Community hospital | 0.17 | 0.48 | 0.17-1.4 |
| Low hospital bed capacity | <0.01∗ | 5.03 | 1.83-13.7 |
∗Denotes statistical significance, P value ≤ 0.05. |
Lastly, we used multivariable analysis to determine patient-associated risk factors that predicted urgent or emergent admission among all patients undergoing CEA in community practice. Age ≥80 years (OR = 1.22, p = 0.04), a history of cardiac disease (OR = 1.29, p < 0.01), and renal disease (OR = 1.76, p = 0.05) each independently predicted urgent or emergent admission for CEA (Table VI).
Table VI. Multivariable analysis of patient comorbid conditions associated with urgently performed CEA
| Risk factor | p | OR | 95% CI |
|---|---|---|---|
| Age ≥80 | 0.04∗ | 1.22 | 1.01-1.46 |
| Male sex | 0.13 | 1.12 | 0.97-1.31 |
| Cardiac disease | <0.01∗ | 1.29 | 1.08-1.55 |
| COPD | 0.62 | 1.06 | 0.85-1.32 |
| Diabetes | 0.46 | 1.07 | 0.90-1.26 |
| Hypertension | 0.27 | 1.10 | 0.93-1.29 |
| Renal disease | 0.05∗ | 1.76 | 1.00-3.08 |
∗Denotes statistical significance, P value ≤ 0.05. |
Discussion
Patients presenting urgently with symptomatic high-grade carotid stenosis remain a clinical challenge. To our knowledge, we report the largest analysis of urgently performed CEA to date. We demonstrate that patients in the state of Connecticut who are symptomatic and present urgently for CEA have increased perioperative mortality and stroke, but not cardiac complications, compared to patients who present electively. However, increased perioperative mortality and stroke were confined to urgently presenting patients with high, but not low, complexity scores. Perioperative mortality in symptomatic urgent patients was also associated with treatment in low-bed capacity hospitals. These results suggest that high-risk patients who present urgently for CEA are not referred to large treatment centers despite regionalization of care for elective patients.
We identified several patient risk factors associated with poor outcome after urgently performed CEA. Patient risk factors that have traditionally been associated with increased risk of perioperative cardiac complications and stroke were similarly identified in our analysis. For example, diabetes is associated with cardiac complications and renal disease is associated with an increased risk of stroke (Table V). However, although symptomatic presentation is a risk factor for adverse outcome in many studies,24, 25 all patients in this study were symptomatic.
To minimize potential underreporting bias in this analysis of an administrative database, we used a validated aggregate complexity score that reflects patients’ associated medical comorbid conditions and is associated with outcome. Using this complexity score, we demonstrate that symptomatic urgent patients have greater comorbidity compared to elective patients (Table IVA). However, we believe that our critical observation is that increased perioperative mortality and stroke in urgently performed cases were confined to the minority of patients with high rates of medical comorbidity. Conversely, most patients presenting urgently but with low rates of associated comorbid conditions had perioperative outcomes similar to those who underwent electively performed CEA (Table IVB). These results suggest that patients presenting urgently, with symptomatic carotid stenosis but few medical comorbidities, are “good-risk” patients who can be safely treated with CEA, with low rates of perioperative mortality and morbidity similar to those expected after elective cases. In addition, since only 48% of symptomatic and urgent cases are performed in large centers (Table I) compared to 58% of all patients in the state,17 it is likely that urgent presentation inhibits referral to larger treatment centers. Referral of high-risk patients with asymptomatic carotid stenosis for treatment may prevent development of urgent symptoms that increase the risk of adverse outcome.26
It is also possible that high-risk patients who present urgently with symptomatic carotid stenosis may benefit from alternative therapies, including CAS or possibly nonoperative medical management. The Stenting and Angioplasty with Protection in Patients at High Risk for Endarterectomy (SAPPHIRE) and Stent-Supported Percutaneous Angioplasty of the Carotid Artery vs. Endarterectomy (SPACE) trials reported equivalency of CEA and CAS in medically “high-risk” patients, but the majority of these patients were treated electively.20, 22 Hobson et al.27 reported an eightfold increased risk of stroke in octogenarians undergoing CAS, suggesting that CEA is preferred in elderly patients needing treatment of symptomatic carotid stenosis; these results suggest that patients considered “high-risk” due to age alone and who present urgently should be referred to large treatment centers if possible. However, there are no definitive studies that suggest improved results with CAS or other alternative strategies in high-risk patients who present urgently. Preliminary data from the Carotid ACCULINK/ACCUNET Postapproval Trial to Uncover Rare Events (CAPTURE) suggest increased risk of adverse outcome in urgently presenting symptomatic patients treated with CAS.28
We also examined the impact of several hospital-associated factors on outcome after urgently performed CEA. Performance of urgent CEA in hospitals with low bed capacity, which has previously been correlated with hospitals that perform few numbers of CEAs in the state of Connecticut,17 was independently associated with increased perioperative mortality and cardiac complications (Table V). It is possible that our finding that hospital volume was not associated with increased risk of adverse outcome (Table III) reflects overall severity of illness of urgently presenting patients; however, we also found no difference in outcome when analysis was performed for hospitals treating greater than 100 cases per year compared to hospitals treating fewer than 100 cases per year.23 Increased rates of adverse outcome in hospitals with fewer staffed beds may reflect increased risk in small hospitals with fewer resources needed for this group of patients; smaller hospitals may have fewer readily available intensive services, such as a cardiac catheterization lab or possibly continuous on-site critical care support. It is also possible that outcomes based on individual surgeon volume or experience may be more relevant than hospital volume in high-risk patients who are symptomatic and present urgently.29
The limitations of this study include the retrospective nature of data review and the well-known potential for undercoding variables in administrative databases. However, the accuracy of this large contemporary database has been verified previously,17, 18, 19 and we used a validated complexity score to assess aggregate patient severity of illness. Another limitation is the lack of temporal information in this database regarding the timing of the intervention after patient presentation of hospital admission. Furthermore, the reason for urgent admission, the particular symptoms, degree of stenosis, and preoperative use of antiplatelet agents are not available. In addition, the contribution of surgeon factors as well as whether different surgeons are present in different-sized hospitals are not available. It is also possible that some patients referred for CAS may not have been included in this study; to eliminate this potential source of referral bias, we only included patients treated before 2003, when increased numbers of CAS were performed in the state.
We conclude that CEA can be performed in patients who have few or well-controlled comorbid medical conditions and who present urgently for treatment of symptomatic carotid stenosis, with low rates of perioperative mortality and morbidity. However, patients with large numbers of comorbid conditions (“high-risk” patients) presenting urgently with symptomatic carotid disease have increased rates of perioperative complications after CEA. High-risk patients needing urgent treatment of symptomatic carotid stenosis should be referred to large treatment centers if possible. Additional studies focused on patients presenting urgently and with moderate or high-grade carotid stenosis are needed before additional alternative treatment strategies are warranted.
References
- . Endarterectomy for asymptomatic carotid artery stenosis. Executive Committee for the Asymptomatic Carotid Atherosclerosis Study. JAMA. 1995;273:1421–1428
- . Beneficial effect of carotid endarterectomy in symptomatic patients with high-grade carotid stenosis. North American Symptomatic Carotid Endarterectomy Trial Collaborators. N Engl J Med. 1991;325:445–453
- . Prevention of disabling and fatal strokes by successful carotid endarterectomy in patients without recent neurological symptoms: randomised controlled trial. Lancet. 2004;363:1491–1502
- . Role of carotid endarterectomy in asymptomatic carotid stenosis: a Veterans Administration Cooperative Study. Stroke. 1986;17:534–539
- European Carotid Surgery Trialists’ Collaborative Group. Randomised trial of endarterectomy for recently symptomatic carotid stenosis: final results of the MRC European Carotid Surgery Trial (ECST). Lancet. 1998;351:1379–1387
- . Intracranial hemorrhage following surgical revascularization for treatment of acute strokes. J Neurosurg. 1964;21:212–216
- . Joint Study of Extracranial Arterial Occlusion IV: a review of surgical considerations. JAMA. 1969;209:1889–1895
- Early endarterectomy for severe carotid artery stenosis after a non-disabling stroke: results from the North American Symptomatic Carotid Endarterectomy Trial. J Vasc Surg. 1994;20:288–295
- . Early carotid endarterectomy in patients with small, fixed neurologic deficits. J Vasc Surg. 1984;1:795–799
- Timing of carotid endarterectomy after acute stroke. J Vasc Surg. 1990;11:45–52
- . Timing of carotid endarterectomy in patients with recent stroke. Surgery. 1997;122:850–855
- Early carotid endarterectomy in symptomatic patients is associated with poorer perioperative outcomes. J Vasc Surg. 2006;44:480–487
- Causes of increased stroke rate after carotid endarterectomy in patients with previous strokes. Ann Vasc Surg. 1997;11:28–34
- Early carotid endarterectomy after non-disabling ischemic stroke: adequate therapeutic option in selected patients. Eur J Vasc Endovasc Surg. 1998;15:423–428
- Early carotid endarterectomy in selected stroke patients. Ann Vasc Surg. 1999;13:463–467
- . for the CEA Trialist Collaboration. Endarterectomy for symptomatic carotid stenosis in relation to clinical subgroups and timing of surgery. Lancet. 2004;363:915–924
- Hospital based factors predict outcome after carotid endarterectomy. J Surg Res. 2006;134:74–80
- Influence of age and hospital volume on the results of carotid endarterectomy: a statewide analysis of 9918 cases. J Vasc Surg. 1998;27:25–33
- Do vascular surgeons improve outcome of carotid endarterectomy? An analysis of 12,618 elective cases in the state of Connecticut. Vascular. 2004;12:155–165
- Protected carotid-artery stenting versus endarterectomy in high-risk patients. N Engl J Med. 2004;351:1493–1501
- Endarterectomy versus stenting in patients with symptomatic severe carotid stenosis. N Engl J Med. 2006;355:1660–1671
- . 30 day results from the SPACE trial of stent-protected angioplasty versus carotid endarterectomy in symptomatic patients: a randomised non-inferiority trial. Lancet. 2006;368:1239–1247
- . Volume standards for high-risk surgical procedures: potential benefits of the Leapfrog initiative. Surgery. 2001;130:415–422
- . Safety of carotid endarterectomy in a high-risk population: lessons from the VA and Connecticut. J Am Coll Surg. 2006;203:277–282
- . Carotid endarterectomy in octogenarians: does increased age indicate high risk?. J Vasc Surg. 2005;41:231–237
- . Safety of carotid endarterectomy in 2443 elderly patients: lessons from nonagenarians—are we pushing the limit?. J Am Coll Surg. 2000;5:734–741
- . Carotid artery stenting is associated with increased complications in octogenarians: 30-day stroke and death rates in the CREST lead-in phase. J Vasc Surg. 2004;40:1106–1111
- . Capture 3000: report of the largest US multicenter experience in carotid stenting for high surgical risk. American College of Cardiology, 55th Annual Scientific Session, March 11-14, 2006, Atlanta, GA. Annual Scientific Session. 2006;
- . Surgeon volume as an indicator of outcome after carotid endarterectomy: an effect independent of specialty practice and hospital volume. J Am Coll Surg. 2002;195:814–921
H. A. Bazan's current address is: Department of Surgery, Section of Vascular Surgery, LSUHSC, New Orleans, LA 70118, USA.
PII: S0890-5096(08)00084-8
doi:10.1016/j.avsg.2007.12.019
© 2008 Annals of Vascular Surgery Inc. Published by Elsevier Inc All rights reserved.
