Neurological Outcomes in Patients With Ischemic Stroke Receiving Enoxaparin or Heparin for Venous Thromboembolism Prophylaxis
Subanalysis of the Prevention of VTE After Acute Ischemic Stroke With LMWH (PREVAIL) Study
Background and Purpose— The Prevention of VTE after Acute Ischemic Stroke with LMWH (PREVAIL) study demonstrated that enoxaparin was superior to unfractionated heparin (UFH) in preventing venous thromboembolism in patients with ischemic stroke and was associated with a small but statistically significant increase in extracranial hemorrhage rates. In this PREVAIL subanalysis, we evaluate the long-term neurological outcomes associated with the use of enoxaparin compared with UFH. We also determine predictors of stroke progression.
Methods— Acute ischemic stroke patients aged ≥18 years, who could not walk unassisted, were randomized to receive enoxaparin (40 mg once daily) or UFH (5000 U every 12 hours) for 10 days. Patients were stratified according to baseline stroke severity using the National Institutes of Health Stroke Scale score. End points for this analysis included stroke progression (≥4-point increase in National Institutes of Health Stroke Scale score), neurological outcomes up to 3 months postrandomization (assessed using National Institutes of Health Stroke Scale score and modified Rankin Scale score), and incidence of intracranial hemorrhage.
Results— Stroke progression occurred in 45 of 877 (5.1%) patients in the enoxaparin group and 42 of 872 (4.8%) of those receiving UFH. Similar improvements in National Institutes of Health Stroke Scale and modified Rankin Scale scores were observed in both groups over the 90-day follow-up period. Incidence of intracranial hemorrhage was comparable between groups (20 of 877 [2.3%] and 22 of 872 [2.5%] in enoxaparin and UFH groups, respectively). Baseline National Institutes of Health Stroke Scale score, hyperlipidemia, and Hispanic ethnicity were independent predictors of stroke progression.
Conclusions— The clinical benefits associated with use of enoxaparin for venous thromboembolism prophylaxis in patients with acute ischemic stroke are not associated with poorer long-term neurological outcomes or increased rates of symptomatic intracranial hemorrhage compared with UFH.
Venous thromboembolism (VTE) is a clinically important complication after a stroke. Without VTE prophylaxis, up to 75% of patients with stroke are likely to develop deep vein thrombosis1,2⇓; pulmonary embolism accounts for up to 25% of early deaths after a stroke.3 Therefore, VTE prophylaxis is recommended in consensus guidelines for the management of patients with acute ischemic stroke.4–6⇓⇓
Anticoagulation with unfractionated heparin (UFH) or a low-molecular-weight heparin (LMWH) is commonly prescribed to stroke patients to reduce the risk of VTE. In the International Stroke Trial (IST), UFH was shown to significantly reduce the incidence of pulmonary embolism and stroke recurrence in patients with acute ischemic stroke.7 Subsequent studies have demonstrated that LMWHs are at least as effective as UFH for VTE prophylaxis in patients with acute ischemic stroke,8–10⇓⇓ and they also have improved bioavailability and a more predictable anticoagulation profile compared with UFH.11 The recent Prevention of VTE after Acute Ischemic Stroke with LMWH (PREVAIL) study demonstrated that the LMWH enoxaparin was superior to UFH for VTE prevention in patients with acute ischemic stroke. In this trial, use of enoxaparin reduced the risk for VTE in these patients by 43% compared with UFH (10% versus 18%; relative risk, 0.57; 95% CI, 0.44 to 0.76; P=0.0001) and was associated with similar incidence of total bleeding and a small but significant increase in extracranial hemorrhage (1% versus 0%; P=0.015).12
Several studies have demonstrated that anticoagulation therapy with LMWHs or UFH does not adversely affect functional or neurological outcomes in patients with acute ischemic stroke compared with placebo or aspirin.13–17⇓⇓⇓⇓ However, data from the IST study showed that the beneficial effect of subcutaneous UFH (5000 U or 12 500 U twice daily) in preventing ischemic stroke recurrence was offset by a proportional increase in the rate of intracranial hemorrhage events, leading to no net benefit from UFH therapy.7
In this subanalysis of the PREVAIL trial, we assess whether the reduced risk of VTE associated with use of LMWH, compared with UFH, in patients with acute ischemic stroke is followed by any negative influences in neurological outcomes.
Materials and Methods
The PREVAIL study has been published previously.12,18⇓ Briefly, patients aged ≥18 years, with confirmed acute ischemic stroke and who were unable to walk unassisted, were randomized within 48 hours of stroke symptom onset to receive open-label treatment with enoxaparin (40 mg once daily) or UFH (5000 U every 12 hours) for 10 days. The sponsor (sanofi-aventis, Paris, France) generated the randomization schedule, which was implemented centrally by an independent interactive voice-response system. All patients gave written informed consent. The study was conducted according to the Declaration of Helsinki and local regulations. Approval to perform the study was obtained from the Institutional Review Board at each site.
The primary efficacy end point of the PREVAIL study was the occurrence of VTE, defined as the composite of asymptomatic and symptomatic deep vein thrombosis, symptomatic pulmonary embolism, and fatal pulmonary embolism during the study treatment phase. The primary safety end points were symptomatic intracranial hemorrhage, major extracranial hemorrhage, and all-cause mortality up to 48 hours after treatment cessation.
Neurological outcomes were assessed using the National Institutes of Health Stroke Scale (NIHSS) score and the modified Rankin Scale (mRS) score. The following neurological outcomes were monitored: NIHSS scores at baseline, at Days 4, 7, 10, and 14 during hospitalization, at the end of study treatment, and at the 30-day and 90-day follow-up; stroke progression (defined as an increase in NIHSS score of ≥4 points from the lowest score obtained at any point from baseline to treatment period and follow-up into Day 90); neurological worsening attributable to stroke progression (defined as an increased NIHSS score in items reflecting worsening focal neurological deficits with or without concomitant worsening in items reflecting changes in the level of consciousness); stroke recurrence (defined as the development of new neurological deficits due to an event separate from the qualifying stroke and verified by CT scan or brain MRI); and mRS scores at baseline, end of study treatment, and at the 30-day and 90-day follow-up. For the analysis of stroke progression using the NIHSS score, patients were stratified according to the severity of the index stroke: a severe stroke was defined as an NIHSS score of ≥14, whereas a less severe stroke was defined as an NIHSS score of <14.
Bleeding event rates associated with neurological worsening were assessed by recording the incidence of intracranial hemorrhage at the end of treatment plus up to 48 hours afterward, and related mortality. Intracranial hemorrhages were confirmed by head CT scan, brain MRI, or autopsy. Results were classified on the basis of imaging results and clinical presentation as: asymptomatic hemorrhagic transformation, symptomatic hemorrhagic transformation, primary intracranial hemorrhage, subarachnoid hemorrhage, or subdural or epidural hemorrhage.
Various demographic characteristics and risk factors are presented according to treatment group and NIHSS stratification. Mean±SD were calculated for continuous variables; counts and percentages were measured for discrete variables. Multivariate logistic regression analysis was used to assess independent risk factors for stroke progression in patients receiving VTE prophylaxis using stepwise selection. The following variables were entered into this analysis: age (categorized as <65 years, 65 to 75 years, or >75 years); racial origin; sex; country; previous myocardial infarction; obesity; previous stroke occurring >48 hours before inclusion; measured hypertension; hyperlipidemia; previous use of a platelet inhibitor; diabetes; congestive heart failure; baseline NIHSS score (categorized as <7, 7 to 10, 11 to 14, 15 to 18, 19 to 22, >23); previous VTE; and smoking. Treatment interaction with each of these variables was also included in the model. All variables were entered at once in the model, which was constructed using the approach recommended by Hosmer and Lemeshow (including their measures of model fit, diagnostics, etc).19 Because the primary purpose of constructing this model was to assess the sensitivity of the stroke progression effect in an exploratory manner, model variables were retained if P<0.05.
Role of the Funding Source
The PREVAIL Steering Committee was responsible for the design of the study, modifications to the study protocol, and blinded adjudication of major hemorrhage events. The protocol for the PREVAIL study was written by the Steering Committee and revised on the basis of discussions with the sponsor (sanofi-aventis). The sponsor generated the randomization schedule, which was then implemented centrally by an independent interactive voice-response system. Data were collected by the sponsor, and data entry was performed by a contract research organization (Parexel, Waltham, Mass). The data were maintained by the contract research organization and analyzed by the sponsor according to the statistical analysis plan, which was reviewed by the Steering Committee. The Steering Committee had full access to the data and vouches for its integrity and completeness. To the best of our knowledge, publication of the data was part of the initial agreement between the sponsor and the Steering Committee regardless of the final results of the trial. An independent data safety monitoring board ensured the proper conduct of the study. The statistician (Min Chen, MS) performed all data analyses and vouches for the accuracy of the analyses. The Steering Committee was responsible for interpretation of the data.
Of the 1762 patients with acute ischemic stroke who were enrolled and randomly assigned to treatment in the PREVAIL study, 877 were treated with enoxaparin and 872 were treated with UFH. The mean duration of treatment in both groups was 10.5 days (SD, 3.2).12 The baseline characteristics of the patients who were enrolled into the study were similar in the two treatment groups12 and also between the patients in the two stroke severity strata within each treatment group (Table 1).
Stroke progression occurred in 45 patients (5.1%) receiving enoxaparin and 42 (4.8%) receiving UFH in the 3 months after randomization. The majority of stroke progression occurred in the first 14 days after randomization, with a similar time to first occurrence of stroke progression in the two treatment groups.
The cumulative incidence of stroke progression in the 3 months after randomization was greater in patients with a severe stroke at baseline compared with those with a less severe stroke, occurring in 10.8% versus 2.8%, respectively (Table 2). The mean change in NIHSS score (baseline to Day 90) attributable to stroke progression in those patients enrolled with a severe stroke was 9.0 points (data not shown) compared with 5.2 points in those enrolled with a less severe stroke (data available in Supplemental Table I; available at http://stroke.ahajournals.org).
Patients showed similar declines in NIHSS scores over time, whether receiving enoxaparin or UFH (Figure 1A), and assessment of neurological outcomes using mRS scores indicated similar degrees of improvement in neurological function over the 90 days postrandomization in patients treated with enoxaparin and UFH (Figure 1B; data used for this figure are available in Supplemental Table II). The proportion of patients who returned to independence (a score on the mRS of 0 to 2) was similar in the enoxaparin (42.7%) and UFH (45.0%) groups 90 days after randomization.
The incidence of stroke recurrence in PREVAIL up to Day 90 was 1.5% and 1.7% in patients receiving enoxaparin or UFH, respectively (Table 3).
The incidences of total bleeding and all subtypes of intracranial hemorrhage were low and comparable between the two treatment groups. Mortality attributable to intracranial hemorrhage was also low, and similar across both treatment groups (Table 4).
Independent Risk Factor Analysis for Stroke Progression
Ethnic origin, the presence of hyperlipidemia, and baseline NIHSS score were identified as independent risk factors for stroke progression in patients receiving UFH or enoxaparin (Figure 2). Asian patients were less likely to experience stroke progression and Hispanic patients more likely to experience stroke progression compared with whites. The likelihood of stroke progression increased with worsening severity of the index stroke.
This subanalysis of the PREVAIL study demonstrates that the improved efficacy for VTE prevention observed with enoxaparin, compared with UFH, is not associated with a difference in long-term neurological outcomes. Our data show comparable long-term improvements in neurological deficits in patients who received enoxaparin or UFH after an acute ischemic stroke. Improvements in neurological outcomes over time were consistently observed, irrespective of the severity of the index stroke, and stroke recurrence rates were consistently low across both treatment groups. The rates of all subtypes of intracranial hemorrhage and mortality related to these events were low and comparable in patients who received either enoxaparin or UFH. The results of this study also indicate that baseline stroke severity, ethnicity, and hyperlipidemia were all independent predictors of stroke progression after an acute ischemic stroke. The increased risk for stroke progression associated with hyperlipidemia did not appear to be related to use of statin therapy because no differences in progression rates were observed when patients were stratified according to statin use (data not shown).
The incidence of stroke progression was low on both therapies; only 5% of the patients had a worsening of ≥4 points on the NIHSS attributable to progression of ischemic stroke. This rate is lower than reported in several prior trials, including the Trial of ORG 10172 in Acute Stroke Treatment (TOAST) study, which documented a 10% rate of stroke progression at Day 7 using the same definition of progression.14 Patients with baseline NIHSS scores of <14 were at extremely low risk for progression.
To further investigate whether LMWHs and/or UFH affect neurological outcomes after an acute ischemic stroke, we reviewed published studies of these anticoagulants versus placebo or aspirin in which neurological outcomes were assessed (Table 5⇓).7,12–17⇓⇓⇓⇓⇓⇓ In contrast to PREVAIL, which primarily tested the use of anticoagulants for VTE prophylaxis in patients with acute ischemic stroke, all other studies were done to investigate whether anticoagulants delay or decrease neurological worsening or stroke progression. Although these studies evaluated neurological function using various methods at different follow-up time points and used different dosing regimens of anticoagulants, the results consistently suggest that LMWHs and UFH do not have a detrimental impact on functional or neurological outcomes in patients with ischemic stroke7,13–17⇓⇓⇓⇓⇓ compared with aspirin or placebo. Furthermore, in line with the results of PREVAIL, these studies suggested that anticoagulant therapy does not have a negative effect on mortality rates in patients with acute ischemic stroke.
Our data indicate that enoxaparin and UFH are associated with low and comparable rates of all subtypes of intracranial hemorrhage and related mortality despite the association of enoxaparin with a higher incidence of major extracranial bleeding. These extracranial bleeding events were mainly gastrointestinal and did not lead to increased mortality.12 In contrast, previous comparisons of LMWHs and UFH in patients with acute ischemic stroke reported that UFH was associated with an increased incidence of major bleeding events, including hemorrhagic transformation of stroke,8,9⇓ a feature that was not systematically assessed in PREVAIL. However, these previous studies compared LMWHs with UFH administered 3 times daily (3×5000 U), whereas the current study compared enoxaparin with UFH administered twice daily (2×5000 U).
Data from the IST trial, which enrolled nearly 20 000 patients with suspected acute ischemic stroke, showed that 0.3% of patients treated with UFH and 0.2% of patients who did not receive UFH died due to a hemorrhagic stroke (defined as symptomatic intracranial hemorrhage or symptomatic hemorrhagic transformation of the original infarct) during the 14-day treatment period.7 In PREVAIL, 0.5% of enoxaparin- and 0.7% of UFH-treated patients experienced symptomatic intracranial hemorrhage over a similar time period. Taken together, these results support the safety of anticoagulation for VTE prevention in patients with ischemic stroke.
In this subanalysis of PREVAIL data, we identified ethnicity as an independent risk factor for the progression of stroke. Previous studies have shown that there are ethnic differences in the incidence rates of initial and recurrent strokes20,21⇓ and, more recently, that ethnicity is associated with risk for subclinical brain infarcts.22 Ethnic differences in stroke risk factors may contribute to differences in initial susceptibility to stroke, whereas our data suggest a possible impact in stroke recurrence and progression as well. This observation needs to be further evaluated in larger prospective observational cohorts of patients with acute ischemic stroke.23
The design of PREVAIL has some limitations that could impact the validity of some of its conclusions. First, the open-label design may result in a potential bias in the declaration of end points. However, all study end points, including neurological worsening and intracranial hemorrhages, were assessed by a central, blinded adjudication committee, assuring lack of bias on the analysis of the declared end points. Second, although during the PREVAIL study, cases of symptomatic or fatal intracranial hemorrhage were confirmed by an imaging study or at autopsy, routine brain imaging was not performed as part of the study protocol. Therefore, the asymptomatic hemorrhagic stroke data reported in PREVAIL were derived from CT scans performed for different clinical indications and not as part of a systematic assessment. This is likely to lead to a gross underestimation of the incidence of this type of hemorrhagic event and precludes any comparisons between the two agents with regard to rates of asymptomatic intracranial bleeding events, the impact of which as a factor affecting outcome (after thrombolysis) remains controversial.24,25⇓ Third, we cannot draw any conclusions on neurological outcomes in patients with different clinical stroke syndromes given that data on subtypes of acute ischemic stroke were not systematically collected during the PREVAIL study. A previous study has indicated that the type of stroke influences functional outcomes, with lacunar infarcts being associated with the greatest recovery and most favorable outcomes, and total anterior circulation infarcts being associated with the poorest recovery and worst outcomes.26
In conclusion, this subanalysis of the data from the PREVAIL study demonstrates that the benefits of VTE prophylaxis using enoxaparin compared with UFH in patients with acute ischemic stroke are not offset by poorer long-term neurological outcomes or increased rates of symptomatic or fatal intracranial hemorrhage. The results of the current subanalysis, together with those of several published studies, indicate that LMWHs and UFH are valuable for VTE prophylaxis after acute ischemic stroke and their use is not associated with any detrimental effects on neurological outcomes. These findings are particularly relevant in light of the negative results of the Clots in Legs Or Stockings after Stroke (CLOTS) trial 1 study,27 which tested thigh-length graduated compression stockings against placebo in patients with acute ischemic stroke. The improved efficacy of enoxaparin for VTE prevention in comparison with UFH and its overall lack of negative impact in neurological outcomes should make it an attractive option for the prevention of this complication of acute ischemic stroke.
We thank the people who agreed to participate in this trial and the study contributors.
Sources of Funding
The PREVAIL study was sponsored by sanofi-aventis, Paris, France. Statistical analyses were performed by Min Chen, MS, sanofi-aventis, Bridgewater, NJ. Editorial assistance in the preparation of this manuscript was funded by sanofi-aventis. G.F.P. received a research grant from Bristol-Myers-Squibb.
C.S.K., G.W.A., C.B., C.F., A.A.G., W.O’R., and G.F.P. received honoraria (modest) from sanofi-aventis, as part of membership in the PREVAIL Steering Committee. G.F.P. is a consultant for Bristol-Myers-Squibb, Boehringer-Ingelheim, and Pfizer; and serves on a speakers’ bureau for Boehringer-Ingelheim and Pfizer.
PREVAIL Steering Committee: David G. Sherman (deceased), Gregory W. Albers, Cesare Fieschi, Alberto A. Gabbai, Carlos S. Kase, William O’Riordan, Graham F. Pineo, and Christopher Bladin.
- Received April 6, 2009.
- Revision received July 2, 2009.
- Accepted July 20, 2009.
- ↵McCarthy ST, Turner J. Low-dose subcutaneous heparin in the prevention of deep-vein thrombosis and pulmonary emboli following acute stroke. Age Ageing. 1986; 15: 84–88.
- ↵Kelly J, Rudd A, Lewis R, Hunt BJ. Venous thromboembolism after acute stroke. Stroke. 2001; 32: 262–267.
- ↵Adams HP Jr, Adams RJ, Brott T, del Zoppo GJ, Furlan A, Goldstein LB, Grubb RL, Higashida R, Kidwell C, Kwiatkowski TG, Marler JR, Hademenos GJ. Stroke Council of the American Stroke Association. Guidelines for the early management of patients with ischemic stroke: a scientific statement from the Stroke Council of the American Stroke Association. Stroke. 2003; 34: 1056–1083.
- ↵Cardiovascular Disease Educational and Research Trust; Cyprus Cardiovascular Disease Educational and Research Trust; European Venous Forum; International Surgical Thrombosis Forum; International Union of Angiology; Union Internationale de Phlébologie. Prevention and treatment of venous thromboembolism. International consensus statement (guidelines according to scientific evidence). Int Angiol. 2006; 25: 101–161.
- ↵International Stroke Trial Collaborative Group. The International Stroke Trial (IST): a randomised trial of aspirin, subcutaneous heparin, both, or neither among 19435 patients with acute ischaemic stroke. International Stroke Trial Collaborative Group. Lancet. 1997; 349: 1569–1581.
- ↵Diener HC, Ringelstein EB, von Kummer R, Landgraf H, Koppenhagen K, Harenberg J, Rektor I, Csányi A, Schneider D, Klingelhöfer J, Brom J, Weidinger G; PROTECT Trial Group. Prophylaxis of thrombotic and embolic events in acute ischemic stroke with the low-molecular-weight heparin certoparin: results of the PROTECT Trial. Stroke. 2006; 37: 139–144.
- ↵Sherman DG, Albers GW, Bladin C, Fieschi C, Gabbai AA, Kase CS, O'Riordan W, Pineo GF; PREVAIL Investigators. The efficacy and safety of enoxaparin versus unfractionated heparin for the PREvention of Venous thromboembolism after Acute Ischaemic Stroke (PREVAIL study): an open-label randomised comparison. Lancet. 2007; 369: 1347–1355.
- ↵Wong KS, Chen C, Ng PW, Tsoi TH, Li HL, Fong WC, Yeung J, Wong CK, Yip KK, Gao H, Wong HB; FISS-tris Study Investigators. Low-molecular-weight heparin compared with aspirin for the treatment of acute ischaemic stroke in Asian patients with large artery occlusive disease: a randomised study. Lancet Neurol. 2007; 6: 407–413.
- ↵Sherman DG, Soltes S, Samuel R, Chibedi-Deroche D. Enoxaparin versus unfractionated heparin in the prevention of venous thromboembolism after acute ischemic stroke: rationale, design, and methods of an open-label, randomized, parallel-group multicenter trial. J Stroke Cereb Dis. 2005; 14: 95–100.
- ↵Hosmer DW Jr, Lemeshow S. Applied Logistic Regression, 2nd ed. New York: John Wiley & Sons, Inc; 2000.
- ↵Broderick J, Brott T, Kothari R, Miller R, Khoury J, Pancioli A, Gebel J, Mills D, Minneci L, Shukla R. The Greater Cincinnati/Northern Kentucky Stroke Study: preliminary first-ever and total incidence rates of stroke among blacks. Stroke. 1998; 29: 415–421.
- ↵Sacco RL, Boden-Albala B, Gan R, Chen X, Kargman DE, Shea S, Paik MC, Hauser WA. Stroke incidence among white, black, and Hispanic residents of an urban community: the Northern Manhattan Stroke Study. Am J Epidemiol. 1998; 147: 259–268.
- ↵Prabhakaran S, Wright CB, Yoshita M, Delapaz R, Brown T, DeCarli C, Sacco RL. Prevalence and determinants of subclinical brain infarction: the Northern Manhattan Study. Neurology. 2008; 70: 425–430.
- ↵Sacco RL, Boden-Albala B, Abel G, Lin IF, Elkind M, Hauser A, Paik MC, Shea S. Race–ethnic disparities in the impact of stroke risk factors: the Northern Manhattan Stroke Study. Stroke. 2001; 32: 1725–1731.
- ↵Libman R, Kwiakowski T, Lyden P, Grotta JC, Tilley BC, Fagen SC, Levine SR, Broderick JP, Lin Y, Lewandowski C, Frankel MR; NINDS rt-PA Stroke Study Group. Asymptomatic hemorrhagic transformation of cerebral infarction does not worsen long-term outcome. J Stroke Cerebrovasc Dis. 2005; 14: 50–54.
- ↵Dzialowski I, Pexman JH, Barber PA, Demchuk AM, Buchan AM, Hill MD. CASES Investigators. Asymptomatic hemorrhage after thrombolysis may not be benign: prognosis by hemorrhage type in the Canadian alteplase for stroke effectiveness study registry. Stroke. 2007; 38: 75–79.
- ↵Sprigg N, Gray LJ, Bath PM, Lindenstrøm E, Boysen G, De Deyn PP, Friis P, Leys D, Marttila R, Olsson JE, O'Neill D, Ringelstein EB, van der Sande JJ, Turpie AG; TAIST Investigators. Stroke severity, early recovery and outcome are each related with clinical classification of stroke: data from the ‘Tinzaparin in Acute Ischaemic Stroke Trial’ (TAIST). J Neurol Sci. 2007; 254: 54–59.