The Effect of Medical Treatments on Stroke Risk in Asymptomatic Carotid Stenosis
Background and Purpose—Recent evidence suggests current best medical treatment may be sufficient to prevent stroke in patients with asymptomatic carotid stenosis. If this is the case, then it is important to determine risk reduction provided by treatments. Using Asymptomatic Carotid Emboli Study (ACES) prospective data, the effect of current treatment and risk factors on future stroke and transient ischemic attack risk were determined.
Methods—Four-hundred seventy-seven patients with asymptomatic carotid stenosis were followed-up every 6 months for 2 years. Changes in risk factors and stroke prevention therapies were reviewed at each visit. Using time-dependent Cox regression, the relationship between current treatment over time was determined and presented as hazard ratios and 95% confidence intervals for risk of stroke, transient ischemic attack, and cardiovascular death end points.
Results—On multivariate analysis, antiplatelets (P=0.001) and lower mean blood pressure (P=0.002) were independent predictors of reduced risk of ipsilateral stroke and transient ischemic attack. Antiplatelets (P<0.0001) and antihypertensives (P<0.0001) were independent predictors of a lower risk of any stroke or cardiovascular death.
Conclusions—Antiplatelet therapy and blood pressure control are the most important factors in reducing short-term stroke and cardiovascular risk in patients with asymptomatic carotid stenosis. More prospective data are required for medical treatments in asymptomatic carotid stenosis in particular for current statin usage.
The optimal treatment for primary prevention in patients with asymptomatic carotid stenosis is controversial. Carotid endarterectomy reduced the risk of stroke in 2 large trials. However, the absolute benefit was small, mainly because of the low risk of stroke in the medically treated group ofμ2% per annum. Recent evidence suggests that with newer more effective medical therapy, stroke risk is lower.1 There has been a reduction in stroke risk over the past 2 decades, which may mean there is little benefit from carotid endarterectomy.2,3
If this patient group is treated without surgery, then it is important to establish which medical approaches add to stroke prevention. Prospective studies have related risk factors (RFs) and medication at baseline to future risk,3–5 but there are little data on how taking into account changes in current medication influences risk in asymptomatic carotid stenosis. We used prospective Asymptomatic Carotid Emboli Study (ACES) data to address this question. ACES was an international, multi-center study of patients with asymptomatic carotid stenosis, with clinical follow-up over 2 years.6 Current medications, blood pressure (BP), and recurrent stroke or transient ischemic attack (TIA) were recorded at each 6-month follow-up visit. We evaluated treatment changes every 6 months and compared the effects of baseline treatments with current treatments and RFs over the 6-month intervals on the risk of stroke, TIA, and cardiovascular disease (CVD) death within the 2-year follow-up.
Inclusion and exclusion criteria and RF definitions for ACES have been published previously.6 RF treatment was left to the discretion of local physicians.
The predefined primary end point of ACES was ipsilateral stroke or TIA over 2 years of follow-up.6 A secondary end point was any stroke or CVD death.
Treatments and RFs were compared to examine changes over follow-up and were stratified according to their values at baseline. Current treatment at the start of each 6-month period was used to define risk over the subsequent 6 months until next clinical follow-up using time-dependent Cox regression to calculate hazard ratios and 95% confidence intervals for primary and secondary end points. P<0.05 was considered significant. SPSS statistical software (version 18.0) was used. See online-only Data Supplement for further details of analysis.
During follow-up, there were the following end points: ipsilateral stroke or TIA in 32; ipsilateral stroke alone in 10; any stroke in 18; and stroke death or CVD in 37; 29.1% of patients were using all 3 medical therapies and none of these patients experienced ipsilateral stroke during follow-up.
The proportions of demographics, treatments, and RFs at each visit are shown in the Table and as follows.
Lipid-lowering therapy. The proportion of patients using any lipid-lowering or statin therapy changed significantly (both P<0.0001) between visits (Table). Lipid-lowering therapy increased from baseline (65.8%) to 24 months (72.1%). Because statins were the predominant lipid-lowering therapy, only effects of statins are presented in further analyses.
Antihypertensive therapy and hypertension. The proportion of patients using >3 antihypertensives increased from baseline (18.9%) to 24 months (23.6%) (Table). Corresponding with this, there was a reduced proportion of patients who were hypertensive (BP >140/90 mm Hg) from baseline (14.7%) to 24 months (8.5%), and reduced mean systolic BP from baseline (147.5 mm Hg) to 24 months (141.5 mm Hg) over the same time period (Table).
Antithrombotic therapy. The proportion of patients prescribed antithrombotics over the study duration altered significantly (P=0.003). However, overall, antithrombotics were prescribed atμ95% of follow-ups. Anticoagulants or dual antiplatelet therapy were uncommon; therefore, only any antiplatelet (including dual) is presented in further analysis. Antiplatelets altered over follow-ups (P<0.0001); the major change was an increase at 6 months. There was no positive trend between use and study duration.
Smoking. Smoking status of patients varied between visits (P=0.017; Table).
Relationship Between Treatments, Risk Factors, and Clinical End Points
None of the baseline treatments predicted risk of ipsilateral stroke or TIA (Figure 1). However, baseline mean BP weakly predicted any stroke or CVD death (P=0.03). In comparison to baseline measures, univariate associations between current treatments or RFs and subsequent risk are shown in Figure 2A and as follows.
Statins. Statins were associated with a lower risk of ipsilateral stroke or TIA (P=0.04) and of any stroke or CVD death (P<0.0001).
Antihypertensives and BP. Antihypertensives predicted a lower risk of ipsilateral stroke or TIA (P=0.01) and any stroke or CVD death (P<0.0001). BP was strongly associated with ipsilateral stroke or TIA and any stroke or CVD death (all P<0.0001). The association of antihypertensives with ipsilateral stroke or TIA was largely mediated via BP control (Figure 2B).
Antiplatelets. Antiplatelets predicted a lower risk of ipsilateral stroke or TIA and any stroke or CVD death (both P<0.0001).
Smoking. Smoking was a predictor of risk of ipsilateral stroke or TIA (P=0.01) and any stroke or CVD death (P<0.0001).
Multivariate analyses were performed including antihypertensives, antiplatelets, and statins (Figure 2C). Antiplatelets (P<0.0001) were an independent predictor of a reduced risk of ipsilateral stroke or TIA. Statins, antihypertensives, age, and sex were not predictors of ipsilateral stroke or TIA. Antiplatelets (P<0.0001) and antihypertensives (P<0.0001) were independent predictors of a lower risk of any stroke or CVD death. Male sex (hazard ratios, 4.12; 95% confidence interval, 1.26–13.51; P=0.02) also was an independent predictor of any stroke or CVD death. Statins and age were not significant predictors of any stroke or CVD death.
A multivariate analysis also was performed replacing antihypertensives with mean BP to determine the effectiveness of BP control for future risk (Figure 2D). Antiplatelets (P=0.001) and mean BP (P=0.002) were independent predictors of a reduced risk of ipsilateral stroke or TIA. Statins, age, and sex were not significant predictors of ipsilateral stroke or TIA. Antiplatelets (P<0.0001) were independent predictors of a lower risk of any stroke or CVD death. Mean BP (P=0.006) and male sex (hazard ratios, 4.72; 95% confidence interval, 1.45–15.38; P=0.01) also were independent predictors of any stroke or CVD death. Statins and age were not significant predictors of any stroke or CVD death.
These results emphasize the importance of controlling conventional cardiovascular RFs in patients with asymptomatic carotid stenosis.
On univariate analysis hypertension, BP, smoking, statins, and antiplatelets all were associated with the end points. On multivariate analysis, antiplatelets and mean BP were associated with reduced ipsilateral TIA and stroke, whereas antiplatelets, mean BP, and antihypertensives were associated with any stroke or CVD death.
Patients were followed-up every 6 months and changes in treatment were recorded. Therefore, we associated current treatments and current RF with events in each subsequent 6-month period. This showed current treatments such as antiplatelets reduced the risk of stroke or TIA. The estimates for ipsilateral stroke or TIA favored statin and antihypertensives treatment, and for any stroke or CVD death the estimates favored statin treatment.
Antiplatelets are widely used in the secondary prevention of stroke, but evidence for their efficacy in primary prevention is less clear.7,8 One previous trial of aspirin in asymptomatic carotid stenosis9 found no benefit, although this was performed some time ago (1988–1994) when there was less effective treatment of vascular RF, and the annual rate of ischemic event was high, at ~11%. This study9 included patients with ≥50% stenosis compared with the ≥70% stenosis cut-off used in our study. Our data suggest antiplatelets are important in patients with asymptomatic carotid stenosis. Because the majority of patients were only using aspirin, we did not have the power to examine efficacy between different or dual antiplatelets.
Limitations of this study include a lack of knowledge of the treatment duration before recruitment and lack of blood tests to assess efficacy of statin therapy, for example, achieved low-density lipoprotein cholesterol level; therefore, these could not be used to assess treatment efficacy. There was also an increase in statin and aspirin usage and a decrease in smoking from 1999 to 2009. However, the number of patients using statins in ACES is likely to be less than in current clinical practice, which may account for the nonsignificance of statins in the multivariate model. In addition, the study had a small sample size and follow-up for this study was 2 years; further data are required to show the effect of medical therapies and controlling risk factors over longer follow-up in larger studies.
Our data emphasize the importance of BP control in asymptomatic carotid stenosis. Mean BP was related to risk of stroke and TIA. On univariate analysis, antihypertensive use was related to a reduced ipsilateral stroke or TIA risk, but this association disappeared after controlling for mean BP level.
A considerable body of evidence has shown that statins reduce risk in stroke and atherosclerosis; the lack of effect we found with statin therapy may partly reflect the short time duration of follow-up. In the Stroke Prevention by Aggressive Reduction in Cholesterol Levels (SPARCL) trial,10 a reduction in risk only started appearing after 1 to 2 years of follow-up.
In summary, our data show that medical treatments can reduce risk in asymptomatic stenosis over relatively short time periods. This is consistent with recent clinical trials and epidemiological studies suggesting that intensive RF treatment significantly reduces stroke rate in asymptomatic carotid stenosis. Importantly, there were no ipsilateral strokes in any patient receiving all 3 medical therapies. Further trials are required to determine whether more intensive treatments will reduce the risk of stroke below that at which patients may benefit from interventions, such as carotid endarterectomy. Interestingly, in ACES, the risk of stroke was nearer to 1% (10/477) than 2%, suggesting stroke rates already may have decreased since the trials of endarterectomy in asymptomatic carotid stenosis were performed.
Sources of Funding
British Heart Foundation (BHF) grants (ACES RG99/073; renewal RG04/002). M. Shipley is supported by BHF. The Stroke Research Network supported ACES recruitment as part of the National Institutes of Health Research Clinical Research Network.
The online-only Data Supplement is available with this article at http://stroke.ahajournals.org/lookup/suppl/doi:10.1161/STROKEAHA.112.673608/-/DC1.
- Received August 8, 2012.
- Revision received October 23, 2012.
- Accepted October 24, 2012.
- © 2013 American Heart Association, Inc.
- Abbott AL
- Marquardt L,
- Geraghty OC,
- Mehta Z,
- Rothwell PM
- Abbott AL,
- Chambers BR,
- Stork JL,
- Levi CR,
- Bladin CF,
- Donnan GA
- 7.↵Antithrombotic Trialists’ collaboration. Collaborative meta-analysis of randomised trials of antiplatelet therapy for prevention of death, myocardial infarction, and stroke in high risk patients. BMJ.2002;324:71–86.
- Goldstein LB,
- Amarenco P,
- Zivin J,
- Messig M,
- Altafullah I,
- Callahan A,
- et al