Abstract
Background and Purpose
Prior meta-analysis showed that carotid endarterectomy (CEA) benefits decline with increasing surgical delay following symptoms. For symptomatic patients in the Carotid Revascularization Endarterectomy versus Stenting Trial (CREST), we assessed if differences in time between symptoms and CEA or stenting (CAS) are associated with differences in risk for periprocedural stroke or death.
Methods
We analyzed the 1180 symptomatic patients in CREST who received their assigned procedure and had clearly defined timing of symptoms. Patients were classified into three groups based on time from symptoms to procedure: <15 days, 15-60 days, and >60 days.
Results
For CEA, risk of periprocedural stroke or death was not significantly different for the two later time periods relative to the earliest time period [(HR=0.74; 96% CI: 0.22-2.49) for 15-60 days and (HR=0.91; 95% CI: 0.25-3.33) for >60 days (p=0.89)]. For CAS, risk of periprocedural stroke or death was also not significantly different for later time periods relative to the earliest time period [(HR=1.12; 95% CI: 0.53-2.40) for 15-60 days and (HR=1.15; 95% CI:0.48-2.75) for >60 days (p=0.93)].
Conclusions
Time from symptoms to CEA or CAS did not alter periprocedural safety, supporting early revascularization regardless of modality.
Keywords: carotid stenosis, carotid stenting, carotid endarterectomy, stroke
INTRODUCTION
A meta-analysis of two trials of carotid endarterectomy versus medical management for symptomatic carotid stenosis showed that time from the last symptomatic event to randomization significantly modified efficacy of surgery.1 The benefits of endarterectomy were greatest for patients operated within 2 weeks after their last ischemic event. Emergent endarterectomy for stroke-in-evolution carries high operative risk, but for stable patients with minor stroke or TIA, surgery in the first week after symptoms does not carry a substantially increased risk.2 Early endarterectomy is thus often encouraged.3 Some have suggested that stenting is safe even if done emergently in patients with symptomatic occlusion4 however, small series suggest that early stenting carries greater risk than delayed stenting.5
This analysis describes the relationship between time from symptoms to revascularization and periprocedural risk of stroke or death for both revascularization methods.
METHODS
CREST randomized 2502 symptomatic and asymptomatic patients to either CEA or CAS. The primary results have been published.6 All governing review boards approved the protocol and every subject gave informed consent.
This analysis focuses on 1180 symptomatic patients receiving the assigned procedure within thirty days of randomization and in whom the timing of qualifying symptoms was documented. Time from qualifying symptoms to revascularization was treated as a categorical variable, with categories of <15 days, 15 to 60 days, and >60 days. The current analysis used the outcome of periprocedural stroke or death. The periprocdural period was defined as 30 days after procedure for those receiving treatment within 30 days of randomization, or 36 days after randomization for those not receiving treatment within 30 days.
RESULTS
The qualifying ischemic event was a transient ischemic attack for 42% (499/1180) of individuals, an ischemic stroke for 41% (487/1180) of individuals, and amaurosis fugax, for the remaining 16% (194/1180).
Median time from symptoms to randomization was 15 days (IQR (interquartile range), 5 to 49) for CEA and 14 days (IQR, 5 to 44) for CAS. Median time from symptoms to revascularization was 22 days (IQR, 8 to 58) for CEA (n=597) and 18 days (IQR, 8 to 51) for CAS (n=583). 59% of patients had endarterectomy within 30 days; 60%, stenting within 30 days. 39% of patients had endarterectomy within 14 days; 43%, stenting within 14 days. A total of 3.2% of endarterectomies were performed within 2 days, and 5.3% of stenting cases were performed within 2 days. Timing of revascularization did not differ significantly by revascularization method (Table 1).
Table 1.
Timing of revascularization overall and by subgroups of treatment, sex and degree of qualifying stenosis.
| Percentage of Patients undergoing revascularization within specified time periods after index stroke, TIA or amaurosis fugax | ||||
|---|---|---|---|---|
| Median time to revascularization (IQR Q1-Q3), days | <15 days | 15-60 days | >60 days | |
| Overall | 20 (8 - 55) | 484 (41.1%) | 432 (36.5%) | 264 (22.4%) |
| Stent (n=583) | 18 (8 - 51) | 251 (43.1%) | 213 (36.5%) | 119 (20.4%) |
| CEA (n=597) | 22 (8 - 58) | 233 (39.0%) | 219 (36.7%) | 145 (24.3%) |
| P value | 0.21 | |||
| Women (n = 410) | 20 (8 - 53) | 176 (42.9%) | 145 (35.4%) | 89 (21.7%) |
| Men (n = 770) | 21 (8 - 56) | 308 (40.0%) | 287 (37.3%) | 175 (22.7%) |
| P value | 0.62 | |||
| Age ≤70 yrs. (n=667) | 20 (8 - 55) | 268 (40.2%) | 249 (37.3%) | 150 (22.5%) |
| Age >70 yrs. (n=513) | 20 (8 - 55) | 216 (42.1%) | 183 (35.7%) | 114 (22.2%) |
| P value | 0.78 | |||
| Qualifying event | ||||
| Stroke (n=487) | 20 (8 - 59) | 198 (40.7%) | 173 (35.5%) | 116 (23.8%) |
| TIA (n=499) | 19 (7 - 46) | 226 (45.3%) | 181 (36.3%) | 92 (18.4%) |
| Am fugax (n=194) | 27.5 (10 - 65) | 60 (30.9%) | 78 (40.2%) | 56 (28.9%) |
| P value | 0.004 | |||
Overall, the risk of a periprocedural stroke or death endpoint was 4.2% (50/1180). The risk of periprocedural stroke or death was not significantly different for either time period (15–60 days; hazard ratio [HR], 0.98; 95% confidence interval [CI], 0.52–1.87) or >60 days (HR, 1.07; 95% CI, 0.52–2.21) relative to those who got the procedure <15 days of qualifying symptom (p=0.97) (Table 2). Figure 1 shows the risk of periprocedural stroke or death relative to time following revascularization for the three time groups. We also performed the same analysis stratified by treatment type. For CEA, risk of periprocedural stroke or death was not significantly different for the two later time periods relative to the earliest time period (HR, 0.74; 95% CI, 0.22–2.49 for 15 to 60 days and HR, 0.91; 95% CI, 0.25–3.33 for >60 days; P=0.89). For CAS, risk of periprocedural stroke or death was also not significantly different for later time periods for CAS relative to the earliest time period (HR, 1.12; 95% CI, 0.53–2.40 for 15 to 60 days and HR, 1.15; 95% CI, 0.48–2.75 for >60 days; P=0.93).
Table 2.
Risk of periprocedural stroke or death by time from qualifying event to procedure for symptomatic patients.
| Periprocedural Stroke or Death | Time from qualifying event to procedure (days) | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Overall | CAS | CEA | ||||||||||
| <15 days | 15-60 days | >60 days | p-value | <15 days | 15-60 days | >60 days | p-value | <15 days | 15-60 days | >60 days | p-value | |
| Events / # subjects | 20/484 | 18/432 | 12/264 | 14/251 | 13/213 | 8/119 | 6/233 | 5/219 | 4/145 | |||
| HR (95% CI)* | reference | 0.98 (0.52-1.87) | 1.07 (0.52-2.21) | 0.97 | reference | 1.12 (0.53-2.40) | 1.15 (0.48-2.75) | 0.93 | reference | 0.74 (0.22-2.49) | 0.91 (0.25-3.33) | 0.89 |
Adjusted for age, sex and type of revascularization (CAS vs CEA)
Figure 1.
Kaplan-Meier curve (periprocedural stroke or death)
DISCUSSION
Time from symptoms to treatment did not affect periprocedural safety of CEA or CAS or the risk of stroke after the periprocedural period. Early revascularization appears to be safe for both procedures.
Since patients were not randomized to early versus delayed surgery, it is not possible to conclude definitively that timing of revascularization had no effect on risk. Unknown or unmeasured differences between early and later revascularized patients may have confounded the comparison of risks. However, given the current recommendations to revascularize early, intentionally delaying treatment would not be ethical. Further, by not having a treatment group managed purely medically, we cannot conclude that delayed surgery is appropriate simply because it seems to carry no greater risk of complications. Early treatment should be the goal because it ought to lead to less time at risk of artery-to-artery embolism.
Timing should not be used to influence the decision to perform stenting versus endarterectomy in patients with symptomatic carotid stenosis.
Acknowledgments
Sources of Funding: Research reported in this publication was supported by the National Institute of Neurological Disorders and Stroke of the NIH - R01 NS 038384 and U01NS038384. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH. Supplemental support was received from Abbott Vascular, Inc.
Footnotes
Clinical Trial Registration-URL: http://www.clinicaltrials.gov
Unique Identifier: NCT00004732
Disclosures: Meschia, Altafulla, Stotts, Gonzales, Voeks and Howard: None; Hopkins consults for Boston Scientific, Cordis, Abbott and Covidien, receives research funding from Toshiba, has >$10,000 financial interests in Boston Scientific, Valor Medical, Claret Medical, Inc., Augmenix, Silk Road, Ostial, Apama, StimSox, Photolitec, ValenTx, Ellipse, Axtria, NextPlain, MedinaMed, and Ocular, and <$10,000 from Endomation; receives <$10,000 honoraria from Cordis Memorial Healthcare System, Complete Conf. Managementand Covidien, holds a >$10,000 Board, Trustee, or Officer Position at Claret Medical (>$10,000); a <$10,000 Speakers Bureau position at Abbott Vascular and Toshiba (<$10,000); Wechsler consults for Biogen Idec and San Bio, holds stock in Silk Road Medical and is on the ACT-1 Steering Committee;Brott is a paid consultant with 3D Communications and Edwards Lifesciences.
REFERENCES
- 1.Rothwell PM, Eliasziw M, Gutnikov SA, Warlow CP, Barnett HJ. Endarterectomy for symptomatic carotid stenosis in relation to clinical subgroups and timing of surgery. Lancet. 2004;363:915–924. doi: 10.1016/S0140-6736(04)15785-1. [DOI] [PubMed] [Google Scholar]
- 2.Rerkasem K, Rothwell PM. Systematic review of the operative risks of carotid endarterectomy for recently symptomatic stenosis in relation to the timing of surgery. Stroke. 2009;40:e564–572. doi: 10.1161/STROKEAHA.109.558528. [DOI] [PubMed] [Google Scholar]
- 3.Gladstone DJ, Oh J, Fang J, Lindsay P, Tu JV, Silver FL, et al. Urgency of carotid endarterectomy for secondary stroke prevention: results from the Registry of the Canadian Stroke Network. Stroke. 2009;40:2776–2782. doi: 10.1161/STROKEAHA.109.547497. [DOI] [PubMed] [Google Scholar]
- 4.Jovin TG, Gupta R, Uchino K, Jungreis CA, Wechsler LR, Hammer MD, et al. Emergent stenting of extracranial internal carotid artery occlusion in acute stroke has a high revascularization rate. Stroke. 2005;36:2426–2430. doi: 10.1161/01.STR.0000185924.22918.51. [DOI] [PubMed] [Google Scholar]
- 5.Topakian R, Strasak AM, Sonnberger M, Haring HP, Nussbaumer K, Trenkler J, et al. Timing of stenting of symptomatic carotid stenosis is predictive of 30-day outcome. Eur J Neurol. 2007;14:672–678. doi: 10.1111/j.1468-1331.2007.01815.x. [DOI] [PubMed] [Google Scholar]
- 6.Brott TG, Hobson RW, 2nd, Howard G, Roubin GS, Clark WM, Brooks W, et al. Stenting versus endarterectomy for treatment of carotid-artery stenosis. N Engl J Med. 2010;363:11–23. doi: 10.1056/NEJMoa0912321. [DOI] [PMC free article] [PubMed] [Google Scholar]