Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2015 Mar 1.
Published in final edited form as: Stroke. 2014 Feb 12;45(3):707–716. doi: 10.1161/STROKEAHA.113.004562

C-reactive Protein as a Prognostic Marker after Lacunar Stroke: The Levels of Inflammatory Markers in the Treatment of Stroke (LIMITS) Study

Mitchell S V Elkind 1,2,3, Jorge M Luna 1,2, Leslie A McClure 4, Yu Zhang 4, Christopher S Coffey 5, Ana Roldan 6, Oscar H Del Brutto 7, Edwin Javier Pretell 8, L Creed Pettigrew 9, Brett C Meyer 10, Jorge Tapia 11, Carole White 12, Oscar R Benavente 13, on behalf of the LIMITS Investigators
PMCID: PMC4114338  NIHMSID: NIHMS603367  PMID: 24523037

Abstract

Background and Purpose

Inflammatory biomarkers predict incident and recurrent cardiac events, but their relationship to stroke prognosis is uncertain. We hypothesized that high-sensitivity C-reactive protein (hsCRP) predicts recurrent ischemic stroke after recent lacunar stroke.

Methods

Levels of Inflammatory Markers in the Treatment of Stroke (LIMITS) was an international, multicenter, prospective ancillary biomarker study nested within Secondary Prevention of Small Subcortical Strokes (SPS3), a Phase III trial in patients with recent lacunar stroke. Patients were assigned in factorial design to aspirin versus aspirin plus clopidogrel, and higher versus lower blood pressure targets. Patients had blood samples collected at enrollment, and hsCRP measured using nephelometry at a central laboratory. Cox proportional hazards models were used to calculate hazard ratios and 95% confidence intervals (HR, 95%CI) for recurrence risks before and after adjusting for demographics, comorbidities, and statin use.

Results

Among 1244 lacunar stroke patients (mean 63.3 ± 10.8 years), median hsCRP was 2.16 mg/L. There were 83 recurrent ischemic strokes (including 45 lacunes), and 115 major vascular events (stroke, myocardial infarction, vascular death). Compared with the bottom quartile, those in the top quartile (hsCRP >4.86 mg/L) were at increased risk of recurrent ischemic stroke (unadjusted HR 2.54, 95%CI 1.30–4.96), even after adjusting for demographics and risk factors (adjusted HR 2.32, 95%CI 1.15–4.68). HsCRP predicted increased risk of major vascular events (top quartile adjusted HR 2.04, 95%CI 1.14–3.67). There was no interaction with randomized antiplatelet treatment.

Conclusions

Among recent lacunar stroke patients, hsCRP levels predict risk of recurrent strokes and other vascular events. HsCRP did not predict response to dual antiplatelets.

Keywords: stroke, lacune, inflammation, C-reactive protein, prognosis

INTRODUCTION

Basic and clinical studies provide evidence that inflammation plays a major role in atherosclerosis and cardiovascular disease,1 but relatively little information is available on serum markers of inflammation as indicators of prognosis after stroke.2 Clinical studies suggest, moreover, that some antiplatelet agents and statins reduce levels of inflammatory markers, and that efficacy of these treatments in preventing cardiac disease may be predicted by these levels.3,4,5 There have been few major multicenter studies of inflammatory markers in predicting outcomes after stroke, however, and no studies of the role of these markers in choosing anti-platelet therapies.

Lacunes, or small subcortical strokes, comprise about 25% of brain infarcts, are especially frequent in Latinos and other U.S. minorities, and are the most common cause of vascular dementia.6,7,8,9 Though infrequently fatal, lacunes are associated with a high risk of recurrence and cognitive impairment. The rate of recurrence among lacunar stroke patients is about 8% per year, slightly higher than other stroke sub-types.10,11,12,13 About 70% of recurrences in these patients are also lacunes, supporting a distinctive pathomechanism.

Major risk factors for lacunar stroke are age, hypertension, and diabetes. About 70–80% of patients have hypertension, and up to 30% have diabetes mellitus.14,15,16 Other potential risk factors include smoking, silent brain infarcts, and white matter hyperintensities on MRI.17,18,19 Stroke risk factors were absent in 18% of patients in one large autopsy study.20 In the Northern Manhattan Stroke Study, the prevalence of hypertension, diabetes, smoking, and hypercholesterolemia did not differ between those with lacunar and non-lacunar stroke.21

Inflammatory mechanisms have been associated with lacunes and their prognosis. A polymorphism of the interleukin-6 gene associated with increased inflammation was an independent risk factor for lacunar stroke in one study.22 This same polymorphism was found to be associated with carotid artery intima-media thickness, as well, providing evidence that polymorphisms related to inflammation may relate to both large and small vessel disease.23 Others have reported that lacunar stroke patients with elevated tumor necrosis factor (TNF)-α and intercellular adhesion molecule 1 (ICAM-1) levels were more likely to experience early neurological deterioration and poor outcome at 3 months.24 In a case-control study, elevated Chlamydia pneumoniae antibody titers were associated with increased risk of lacunar stroke, as well as large vessel atherosclerotic stroke.25 An association between leukocyte count and outcomes after stroke has also been found in those with lacunar stroke, as well as other stroke subtypes.26 These data support the hypothesis that inflammatory mechanisms contribute to risk of lacunar disease and its prognosis.

In response to calls for data regarding the role of inflammation in stroke prognosis,1,2 we designed a prospective observational study to test the hypothesis that inflammatory biomarkers predict recurrence after lacunar stroke. The study was nested within an ongoing secondary stroke prevention trial, Secondary Prevention of Small Subcortical Strokes (SPS3), the results of which have been published.27,28 The primary aim of the Levels of Inflammatory Markers in the Treatment of Stroke study (LIMITS) was to determine whether serum levels of high-sensitivity C-reactive protein (hsCRP) predict recurrent stroke and other vascular events among patients with a recent history of small artery ischemic stroke. A secondary aim was to determine whether hsCRP predicts which patients respond best to dual antiplatelet therapy.

METHODS

Overall Plan

LIMITS was designed as an ancillary study to the SPS3 trial, and the methods for both LIMITS and SPS3 have been described elsewhere.27,29,30 In brief, SPS3 was a multicenter, investigator-initiated, NIH/NINDS-funded Phase III trial focused on secondary prevention of stroke recurrence in patients with small vessel ischemic stroke, or lacunes. The study had a 2 × 2 factorial design with one arm being a blinded comparison of monotherapy vs. combination therapy of antiplatelet agents and the other an open label comparison of two targets of blood pressure control. Participants had a symptomatic lacune proven on MRI within the 6 months prior to randomization. Patients were assigned to two interventions: (i) antiplatelet therapy: aspirin (325 mg/day) monotherapy versus aspirin (325 mg/day) plus clopidogrel (75 mg/day) combination therapy (double blind, placebo control); and (ii) two levels of systolic blood pressure control, “higher” (130–149 mmHg) versus “lower” (<130 mmHg) targets. LIMITS involved collection of plasma and serum samples at baseline and at one-year (up to 18 months) follow-up during the study, and central storage and analysis of samples for inflammatory marker levels. Blood samples were drawn for LIMITS at least 3 weeks after stroke. Sites were required to have the capability to collect, process, and store blood samples at −80°C. Of 84 sites participating in SPS3, 45 (54%) sites recruited patients into LIMITS (see appendix for investigator list.)

Inclusion and Exclusion Criteria

The population for this ancillary study consisted of patients enrolled into the SPS3 trial at sites participating in LIMITS. Patients with a clinical diagnosis of MRI-proven lacunar stroke, and absence of cortical or large subcortical stroke, carotid stenosis or cardioembolic source, were eligible. All patients eligible for SPS3 were eligible for this ancillary study; there were no additional exclusion criteria.27 LIMITS was approved by the Columbia University Medical Center IRB and by IRBs at all participating sites. Participants provided informed consent.

Blood Collection Kits, Phlebotomy, and Local Processing

Materials for collection and shipping of blood specimens were provided to participating sites. A 10 cc blood sample in ethylenediaminetetraacetic acid (EDTA) and a 9.5 ml gel serum separator tube were drawn on the day of randomization, prior to initiation of therapy. Samples were centrifuged locally and samples were aliquotted. Aliquots were frozen locally until shipping to the central lab at Columbia University.

Assays for hsCRP

Plasma samples were analyzed in batches blinded to treatment and outcome. HsCRP assays were performed using the Dade-Behring BN-II nephelometric assay system (Dade-Behring, Deerfield, IL), an FDA-approved, national standard assay for this marker.1

Outcomes

Primary outcomes were recurrent ischemic stroke and the combined outcome of a major cardiovascular event (recurrent ischemic stroke, myocardial infarction, or vascular death). Ischemic strokes were defined as a focal neurological deficit persisting for greater than 24 hours, and were ascertained via clinical evaluation and use of CT or MRI. Other vascular events were also defined as in the SPS3 trial.27,29 Additional analyses were performed for the outcome of ischemic or hemorrhagic stroke. Hemorrhagic strokes were defined as neurologic deficits associated with intraparenchymal or subarachnoid hemorrhagic lesions confirmed by CT, MRI, or autopsy.

Statistical Analysis and Sample Size Calculations

Descriptive statistics were calculated and compared for patients in the ancillary LIMITS study and the parent trial, and levels of hsCRP were compared across different patient characteristics. An F-test was used to determine if there were differences in CRP categories for continuous variables, a Chi-Square test for nominal categorical variables, and Cochran-Mantel-Haenszel (CMH) when categorical variables were ordinal.

For the analyses of primary outcomes, hsCRP values at baseline were considered the independent variable of primary interest. Levels of hsCRP were log-transformed to stabilize the variance. Cox proportional hazard regression was used to estimate the unadjusted hazard ratio for hsCRP with time-to-event for ischemic stroke as the dependent variable. Multivariable Cox proportional hazards regression was then used to estimate the adjusted hazard ratio for hsCRP after adjusting for additional potential risk factors, including age, sex, race-ethnicity, region, traditional stroke risk factors defined either dichotomously (hypertension, cardiac disease, diabetes mellitus, smoking) or continuously (body mass index, HDL, LDL). Because statin therapy may influence both levels of hsCRP and outcomes, analyses were also adjusted for statin use. Observations were censored at time of last follow-up visit. Additional analyses utilized quartiles of hsCRP as independent variables, using those in the lowest quartile for each marker as the reference group. In addition, different ranges of standardized levels of risk have been recommended by consensus for primary prevention of cardiac disease: low, medium, and high-risk (CRP < 1, 1–3, and >3 mg/L, respectively).1 There are no consensus levels to be used post-stroke, but the results of small studies suggest that levels are higher after stroke.31,32 We therefore tested as pre-specified thresholds hsCRP <15 and >15 mg/L.

A sample size of 1440 (57% of the initial planned total of 2500 patients to be enrolled in SPS3) was chosen based on feasibility, assuming that about 40 enrolling centers would enroll at least 12 patients annually for 3 years. We computed the detectable hazard ratios based on a required power of 80% and a significance level of 0.05 for comparing the outcome rates in the first and fourth quartiles. Where possible, sample size calculations were based on the same assumptions used in the SPS3 trial (annual 7% rate of recurrent stroke and 10% rate of recurrent ischemic stroke, MI, or death). In addition, a 10% 3-year loss-to-follow up was assumed. Because event rates in SPS3 were lower than expected, the sample size of the parent SPS3 trial was increased to 3000.33 All hypothesis tests performed during the analysis of the primary and secondary endpoints are two-sided and use an alpha of 0.05.

RESULTS

Characteristics of the LIMITS cohort

A total of 1244 patients in SPS3 were enrolled in LIMITS. This represented 41% of the number of participants in the parent SPS3 trial. Participants enrolled in LIMITS were broadly representative of the population enrolled only in SPS3 (Table 1). The mean age of LIMITS participants was 63.3 ± 10.8 years, and 63.4% were men. The race-ethnic distribution differed in that a larger proportion of patients in LIMITS were Hispanic (40.6% in LIMITS versus 23.1% in SPS3 only patients), reflecting an increased number of participating sites from Spain and South America. Other significant differences between participants in LIMITS and SPS3 alone were, in LIMITS, a slightly lower proportion of patients with diagnosed hyperlipidemia and current smoking, a slightly shorter time from qualifying stroke to randomization, and a slightly lower proportion already using aspirin at baseline (Table 1).

Table 1.

Descriptive characteristics of patients enrolled in LIMITS (ancillary study) and SPS3 (parent study) only

Baseline Characteristics LIMITS (N=1244) SPS3 (N=1776) P
Demographics
 Age 63.3±10.8 63.3±10.7 0.85
 Male 789(63.4) 1113(62.7) 0.67
Race-ethnicity
 White 531(42.7) 1007(56.7) <.01
 Hispanic 505(40.6) 411(23.1)
 Black 191(15.4) 301(16.9)
 Other/multiple 17(1.4) 57(3.2)
Region
 North America 679(54.6) 1281(72.1) <.01
 Latin America 428(34.4) 266(15.0)
 Spain 137(11.0) 229(12.9)
Medical History
 Prior symptomatic lacune 140(11.3) 168(9.5) 0.11
 Subcortical TIA 60(4.8) 105(5.9) 0.20
 Hypertension (SPS3 criteria) 1119(90.0) 1590(89.5) 0.71
 Diabetes mellitus 447(35.9) 659(37.1) 0.51
 Hyperlipidemia 576(46.3) 895(50.4) 0.03
 Ischemic heart disease 86(6.9) 139(7.8) 0.35
 Congestive heart failure 5(0.4) 9(0.5) 0.68
 Intermittent claudication 41(3.3) 54(3.0) 0.69
 Current smoking 223(17.9) 394(22.2) <.01
Qualifying stroke at entry
 Rankin score 1.3±0.8 1.3±0.8 0.92
 Days between qualifying stroke and randomization 73.3±49.0 78.6±45.6 <.01
Type of lacunar syndrome
 Pure motor hemiparesis 421(33.9) 574(32.3) 0.17
 Pure sensory stroke 118(9.5) 185(10.4)
 Sensorimotor 403(32.4) 533(30.0)
 Other 301(24.2) 483(27.2)
Use of Aspirin at time of qualifying event 314(25.5) 524(29.8) <.01
Use of statin at baseline 840(67.5) 1241(69.9) 0.17
Use of statin at any time during follow-up 1033(83.0) 1488(83.8) 0.59
Open in a new tab

Levels of hsCRP in LIMITS participants and association with patient characteristics

Among the 1244 patients in LIMITS, median hsCRP was 2.16 mg/L (interquartile range 0.93–4.86), and levels differed by age, sex, smoking, and LDL levels (Table 2). Median hsCRP among current smokers was 3.1 mg/L, compared to 2.0 mg/L among both former and never smokers. Median time between stroke and LIMITS sample collection was 60 days (interquartile range 34–105 days; mean 75 ± 52 days). HsCRP levels were inversely and weakly correlated with proximity to stroke date (r=−0.06, p=0.039).

Table 2.

Distribution of high-sensitivity C-reactive protein levels by baseline characteristics

Characteristic Overall hsCRP by Quartile N, % hsCRP (standard clinical thresholds) N, %
Median(IQR) Quartile 1 <0.93 mg/L (n=312) Quartile 2 0.93–2.16 mg/L (n=310) Quartile 3 2.16–4.86 mg/L (n=312) Quartile 4 ≥4.86 mg/L (n=310) P <1 mg/L (n=332) 1–3 mg/L (n=417) >3 mg/L (n=495) P
Age
 <50 2.2 (0.9–5.2) 32(25.4) 31(24.6) 29(23.0) 34(27.0) 0.02 33(26.2) 44(34.9) 49(38.9) 0.05
 50–65 2.4 (1.0–5.9) 146(24.5) 131(22.0) 142(23.8) 177(29.7) 153(25.7) 178(29.9) 265(44.5)
 >65–75 2.0 (0.9–4.0) 77(24.0) 89(27.7) 95(29.6) 60(18.7) 85(26.5) 119(37.1) 117(36.4)
 >75 1.7 (0.8–3.9) 57(28.4) 59(29.4) 46(22.9) 39(19.4) 61(30.3) 76(37.8) 64(31.8)
Sex
 Male 2.0 (0.9–4.6) 216(27.4) 202(25.6) 189(24.0) 182(23.1) <.01 230(29.2) 264(33.5) 295(37.4) <.01
 Female 2.4 (1.1–5.7) 96(21.1) 108(23.7) 123(27.0) 128(28.1) 102(22.4) 153(33.6) 200(44.0)
Race
 Hispanic 2.1 (0.9–4.6) 127(25.1) 128(25.3) 129(25.5) 121(24.0) 0.18 136(26.9) 173(34.3) 196(38.8) 0.08
 Non-Hispanic White 2.1 (0.9–4.8) 145(27.3) 126(23.7) 134(25.2) 126(23.7) 155(29.2) 172(32.4) 204(38.4)
 Black 2.7 (1.1–5.8) 38(19.9) 48(25.1) 46(24.1) 59(30.9) 38(19.9) 64(33.5) 89(46.6)
 Other/Multiple 2.1 (1.3–4.5) 2(11.8) 8(47.1) 3(17.6) 4(23.5) 3(17.6) 8(47.1) 6(35.3)
Region
 United States 2.2 (0.9–5.7) 157(25.0) 148(23.6) 146(23.3) 176(28.1) 0.26 163(26.0) 197(31.4) 267(42.6) 0.26
 Canada 1.6 (0.8–4.0) 16(30.8) 15(28.8) 12(23.1) 9(17.3) 18(34.6) 17(32.7) 17(32.7)
 Latin America 2.1 (0.9–4.6) 107(25.0) 108(25.2) 111(25.9) 102(23.8) 115(26.9) 147(34.3) 166(38.8)
 Spain 2.1 (1.0–3.8) 32(23.4) 39(28.5) 43(31.4) 23(16.8) 36(26.3) 56(40.9) 45(32.8)
Hypertension
 Yes 2.2 (0.9–5.0) 275(24.6) 280(25.0) 279(24.9) 285(25.5) 0.17 295(26.4) 374(33.4) 450(40.2) 0.33
 No 2.1 (0.8–4.1) 37(29.6) 30(24.0) 33(26.4) 25(20.0) 37(29.6) 43(34.4) 45(36.0)
Smoking
 Never 2.0 (0.9–4.3) 133(25.5) 140(26.9) 139(26.7) 109(20.9) <.01 141(27.1) 190(36.5) 190(36.5) <.01
 Former 2.0 (0.9–4.6) 134(26.8) 127(25.4) 118(23.6) 121(24.2) 144(28.8) 164(32.8) 192(38.4)
 Current 3.1 (1.2–7.0) 45(20.2) 43(19.3) 55(24.7) 80(35.9) 47(21.1) 63(28.3) 113(50.7)
Ischemic Disease
 Yes 2.1 (1.1–5.7) 17(19.8) 27(31.4) 20(23.3) 22(25.6) 0.68 19(22.1) 32(37.2) 35(40.7) 0.51
 No 2.2 (0.9–4.9) 295(25.5) 283(24.4) 292(25.2) 288(24.9) 313(27.0) 385(33.2) 460(39.7)
Diabetes Mellitus
 Yes 2.2 (0.9–5.0) 110(24.6) 112(25.1) 109(24.4) 116(26.0) 0.66 114(25.5) 151(33.8) 182(40.7) 0.49
 No 2.1 (0.9–4.8) 202(25.3) 198(24.8) 203(25.5) 194(24.3) 218(27.4) 266(33.4) 313(39.3)
Labs
 LDL -- 108.3±40.3 112.7±39.2 112.0±39.1 116.3±39.4 0.10 109.0±40.0 111.7±40.0 115.1±38.7 0.10
 HDL -- 46.3±19.2 45.1±13.1 45.8±20.9 41.8±14.5 <.01 46.2±18.7 45.2±15.4 43.3±17.7 0.05
BMI
 BMI -- 28.1±11.9 28.2±5.0 29.3±6.1 30.3±6.5 <.01 28.1±11.6 28.3±5.0 30.2±6.6 <.01
Open in a new tab

Association of hsCRP with outcomes

There were 83 recurrent ischemic strokes (including 45 lacunes), 16 hemorrhagic strokes, and 115 major vascular events (stroke, MI, vascular death) among LIMITS participants, over a median follow-up of 3.0 years.

Compared to the bottom quartile of hsCRP (<0.93 mg/L), those in the top quartile (≥4.86 mg/L) were at increased risk of recurrent ischemic stroke (unadjusted HR 2.54, 95% CI 1.30–4.96; Table 3). There was a trend toward an intermediate level of increased risk for those in the second and third quartiles, but these increased risks were not significantly elevated. The increased risk for those in the top quartile persisted after adjusting for age, sex, race, region, hypertension, smoking, prior history of stroke, diabetes, body mass index, and lipid levels (model 3 in Table 3), and remained elevated after further adjusting for statin use (adjusted HR 2.32, 95% CI 1.15–4.68). The risk of ischemic or hemorrhagic stroke was of similar magnitude (adjusted HR for those in the top quartile 2.08, 95% CI 1.11–3.89; table 4). Approximately 70% of recurrent ischemic strokes were lacunes; there was no evidence of an independent statistically significant predictive effect on lacunar stroke as an outcome, but the results for lacunar stroke were consistent with the effect on ischemic stroke more generally (adjusted HR 2.27, 95% CI 0.90–5.75).

Table 3.

Risk of recurrent ischemic stroke by levels of high-sensitivity C-reactive protein before and after adjusting for potential confounders

Model 1 Model 2 Model 3 Model 4
HR (95% CI) p HR (95% CI) p HR (95% CI) p HR (95% CI) p
HsCRP continuous (per SD) 1.09[0.93,1.28] 0.27 1.11[0.94,1.32] 0.21 1.11[0.93,1.32] 0.25 1.12[0.93,1.34] 0.22
HsCRP 0.05 0.04 0.09 0.11
 Quartile 1 (Referent) -- -- -- --
 Quartile 2 1.94[0.96,3.95] 2.03[0.99,4.13] 2.08[1.02,4.26] 2.07[1.01,4.23]
 Quartile 3 1.66[0.81,3.40] 1.85[0.90,3.81] 1.91[0.92,3.96] 1.95[0.94,4.05]
 Quartile 4 2.54[1.30,4.96] 2.68[1.36,5.26] 2.46[1.22,4.95] 2.32[1.15,4.68]
HsCRP (Standard clinical thresholds) 0.04 0.03 0.05 0.06
 <1 mg/L (Referent) -- -- -- --
 1–3 mg/L 1.75[0.90,3.38] 1.83[0.94,3.55] 1.82[0.94,3.56] 1.82[0.94,3.55]
 >3 mg/L 2.20[1.19,4.10] 2.36[1.26,4.42] 2.22[1.17,4.24] 2.16[1.13,4.11]
HsCRP 0.55 0.62 0.61 0.65
 ≤15 mg/L (Referent) -- -- -- --
 >15 mg/L 1.29[0.56,2.95] 1.24[0.54,2.85] 1.24[0.54,2.88] 1.22[0.52,2.83]
Open in a new tab

Abbreviations: hsCRP= high-sensitivity C-reactive protein; HR= hazard ratio; 95% CI= 95% confidence interval; SD=standard deviation

Models:
  1. Unadjusted
  2. Adjusted for Demographics (Age, Sex, Race, Region)
  3. Adjusted for Demographics (Age, Sex, Race, Region) and Co-morbidities (Hypertension, Smoking, History of Ischemic Stroke, Diabetes, Body mass index, Low- density lipoprotein and High-density lipoprotein)
  4. Adjusted for Demographics (Age, Sex, Race, Region), Co-morbidities (Hypertension, Smoking, History of Ischemic Stroke, Diabetes, Body mass index, Low- density lipoprotein and High-density lipoprotein), and Statin use at baseline

Table 4.

Risk of recurrent ischemic stroke or cerebral hemorrhage by levels of high-sensitivity C-reactive protein before and after adjusting for potential confounders

Model 1 Model 2 Model 3 Model 4
HR (95% CI) p HR (95% CI) p HR (95% CI) p HR (95% CI) p
HsCRP continuous (per SD) 1.06[0.90,1.25] 0.51 1.07[0.91,1.27] 0.40 1.07[0.90,1.27] 0.44 1.08[0.90,1.28] 0.41
HsCRP 0.06 0.05 0.07 0.06
 Quartile 1 (Referent) -- -- -- --
 Quartile 2 1.94[1.05,3.59] 1.93[1.04,3.59] 1.99[1.07,3.71] 1.99[1.07,3.71]
 Quartile 3 1.37[0.72,2.60] 1.46[0.76,2.79] 1.52[0.79,2.92] 1.55[0.80,2.98]
 Quartile 4 2.09[1.15,3.80] 2.25[1.23,4.11] 2.17[1.16,4.04] 2.08[1.11,3.89]
HsCRP (Standard clinical thresholds ) 0.13 0.08 0.11 0.12
 <1 mg/L (Referent) -- -- -- --
 1–3 mg/L 1.58[0.89,2.80] 1.60[0.90,2.84] 1.61[0.91,2.87] 1.62[0.91,2.88]
 >3 mg/L 1.74[1.01,2.99] 1.87[1.08,3.23] 1.81[1.03,3.19] 1.78[1.01,3.13]
HsCRP 0.89 0.92 0.86 0.90
 ≤15 mg/L (Referent) -- -- -- --
 >15 mg/L 1.06[0.47,2.43] 1.05[0.46,2.40] 1.08[0.47,2.48] 1.06[0.46,2.44]
Open in a new tab

Abbreviations: hsCRP= high-sensitivity C-reactive protein; HR= hazard ratio; 95% CI= 95% confidence interval; SD=standard deviation

Models:
  1. Unadjusted
  2. Adjusted for Demographics (Age, Sex, Race, Region)
  3. Adjusted for Demographics (Age, Sex, Race, Region) and Co-morbidities (Hypertension, Smoking, History of Ischemic Stroke, Diabetes, Body mass index, Low- density lipoprotein and High-density lipoprotein)
  4. Adjusted for Demographics (Age, Sex, Race, Region), Co-morbidities (Hypertension, Smoking, History of Ischemic Stroke, Diabetes Body mass index, Low- density lipoprotein and High-density lipoprotein), and Statin use at baseline

HsCRP was associated with an increased risk of major vascular events, as well (unadjusted HR for those in the top quartile compared to lowest 2.14, 95% CI 1.22–3.75). The results were similar after adjusting for demographics, risk factors, and baseline statin use (adjusted HR 2.04, 95% CI 1.14–3.67; Table 5).

Table 5.

Risk of major vascular events (stroke, myocardial infarction, vascular death) by levels of high-sensitivity C-reactive protein before and after adjusting for potential confounders

Model 1 Model 2 Model 3 Model 4
HR (95%) p HR (95%) p HR (95%) p HR (95%) p
HsCRP continuous (per SD) 1.07[0.92,1.24] 0.39 1.08[0.93,1.25] 0.34 1.08[0.92,1.26] 0.36 1.08[0.92,1.27] 0.34
HsCRP 0.05 0.04 0.07 0.09
 Quartile 1 (Referent) -- -- -- --
 Quartile 2 1.91[1.07,3.41] 1.92[1.07,3.44] 1.95[1.08,3.50] 1.95[1.08,3.50]
 Quartile 3 1.53[0.85,2.77] 1.63[0.90,2.97] 1.60[0.87,2.94] 1.63[0.89,2.99]
 Quartile 4 2.14[1.22,3.75] 2.24[1.27,3.94] 2.12[1.18,3.81] 2.04[1.14,3.67]
HsCRP (Standard Cut-offs) 0.08 0.05 0.11 0.12
 <1 mg/L (Referent) -- -- -- --
 1–3 mg/L 1.50[0.88,2.56] 1.52[0.89,2.60] 1.52[0.89,2.60] 1.53[0.89,2.61]
 >3 mg/L 1.77[1.07,2.93] 1.87[1.13,3.10] 1.75[1.04,2.94] 1.72[1.02,2.90]
HsCRP 0.34 0.38 0.35 0.37
 ≤15 mg/L (Referent) -- -- -- --
 >15 mg/L 1.40[0.71,2.76] 1.36[0.68,2.69] 1.40[0.69,2.72] 1.37[0.69,2.74]
Open in a new tab

Abbreviations: hsCRP= high-sensitivity C-reactive protein; HR= hazard ratio; 95% CI= 95% confidence interval; SD=standard deviation

Models:
  1. Unadjusted
  2. Adjusted for Demographics (Age, Sex, Race, Region)
  3. Adjusted for Demographics (Age, Sex, Race, Region) and Co-morbidities (Hypertension, Smoking, History of Ischemic Stroke, Diabetes, Body mass index, Low- density lipoprotein and High-density lipoprotein)
  4. Adjusted for Demographics (Age, Sex, Race, Region), Co-morbidities (Hypertension, Smoking, History of Ischemic Stroke, Diabetes Body mass index, Low- density lipoprotein and High-density lipoprotein), and Statin use at baseline

Results were similar using clinically recommended hsCRP thresholds of high risk in primary prevention. Compared to those with hsCRP <1 mg/L, those with hsCRP >3 mg/L had an approximately twofold increase in risk of ischemic stroke (adjusted HR 2.16, 95 CI 1.13–4.11; Table 3) and a 70% increase in risk of major vascular events (adjusted HR 1.72, 95% CI 1.02–2.90; Table 5). Using a pre-specified higher threshold (>15 mg/L),31 there was no evidence of a difference of risk (Tables 3–5). HsCRP assessed as a continuous measure was also not significantly associated with risk.

HsCRP levels as a predictor of the effect of antiplatelet and blood pressure therapy

There was no interaction of randomized antiplatelet treatment or blood pressure target with hsCRP levels for stroke risk, indicating a lack of evidence for a differential effect of dual versus single antiplatelet therapy depending on hsCRP level. Furthermore, there was no interaction of hsCRP with statin use at baseline.

DISCUSSION

Our data provide evidence that hsCRP predicts risk of recurrent ischemic stroke, total stroke, and major vascular events among patients with recent lacunar stroke. Patients with the highest levels of hsCRP measured in the subacute phase after lacunar stroke (median 60 days) had more than twice the risk of recurrent ischemic stroke, and approximately twice the risk of total stroke or a major vascular event. There was some evidence that hsCRP levels above 1 mg/L and in the intermediate two quartiles were also associated with an approximate doubling of risk, though the increased risk at those levels was not statistically significant and there was no definite evidence of a linear trend across quartiles. These findings provide indirect evidence that there may be a threshold effect at about 1 mg/L, above which the risk is increased, rather than a continuous relationship between hsCRP and subsequent risk after stroke. This result is also confirmed by our finding of a significantly increased risk for hsCRP levels above the clinically recommended high risk threshold of 3 mg/L, and the absence of an effect for the continuous measure of hsCRP.

C-reactive protein is an acute phase reactant, a member of the pentraxin family of proteins, and part of the innate, non-specific immune response. It is produced in the liver on stimulation by interleukin 6. The high sensitivity assay for CRP has several features that make it ideal both as an epidemiological tool and for clinical purposes, including stability after freeze-thaw cycles, limited diurnal and seasonal variability, and the ability to measure it in the non-fasting state.1,2 Several studies have demonstrated its utility as a measure of future risk of atherosclerotic coronary artery disease, and these findings have been interpreted as a sign that inflammation may be an important contributor to future risk of cardiovascular disease. The fact that even relatively minimal elevations, within the range of hsCRP traditionally considered to be normal (i.e., < 10 mg/L), has been interpreted to mean that even minor inflammation, perhaps related or due to low-grade inflammation within atherosclerosis throughout the arterial tree, contributes to risk of cardiovascular events.1 Our findings are consistent with this hypothesis, though due to its observational design and absence of systematic measures of atherosclerosis, our study cannot shed light on the source of hsCRP.

Relatively few studies have examined hsCRP as a risk factor for ischemic stroke, and the majority of these have focused on first stroke.2,34,35,36,37,38 In the prospective Northern Manhattan Study (NOMAS), hsCRP predicted MI and death, but not first ischemic stroke, after adjusting for potential confounders.38 In a large individual participant meta-analysis of 54 prospective cohort studies (n=160,309), the risk ratio of ischemic stroke per standard deviation increase in logeCRP was 1.44 (95% CI 1.32–1.57) after adjusting for age and sex, but was attenuated to 1.27 (95% CI 1.15–1.40) when further adjusting for other risk factors.39 Non-vascular mortality, including cancer, was increased significantly, and by a greater magnitude, in these analyses (adjusted RR 1.54, 95% CI 1.40–1.68), providing some evidence that elevated CRP may be a marker of illness more generally, rather than a marker of vascular disease risk alone.

Until the present study, most studies of hsCRP as a risk marker after stroke included small selected samples, often from single centers, and with limited duration of follow-up; most also focused on mortality alone, or on combined vascular events, and they incompletely adjusted for other predictors of outcome. In one study, hsCRP levels ≥10.1 mg/L measured within 72 hours of stroke predicted increased mortality over 4 years.31 In another, patients in the highest quintile of hsCRP levels assessed at least 3 months after first ischemic stroke or TIA due to large vessel stenosis had a significantly increased risk of subsequent stroke or MI.40 In a population-based follow-up study after first ischemic stroke of all etiologic subtypes in which several outcomes were assessed, hsCRP measured predominantly within the first 72 hours after stroke was associated with increased risk of mortality over the following 5 years after adjusting for confounding factors.41 There was no increased risk of recurrent stroke alone, nor of the combined outcome of stroke and other vascular events, however. HsCRP in that study, moreover, was significantly associated with stroke severity, a major predictor of post-stroke mortality.

There have been few multicenter studies addressing hsCRP after stroke. In a secondary nested case-control analysis of stored blood specimens from the PROGRESS trial, a multicenter secondary stroke prevention trial, those in the highest tertile of CRP had a modestly increased risk of recurrent ischemic stroke (odds ratio 1.39, 95% CI 1.05–1.85).42 Results were not, however, fully adjusted for all risk factors, including diabetes mellitus.

The present study differs from these other studies in including only lacunar stroke patients and in collecting hsCRP measurements at a much longer time interval after stroke (median 60 days), at a time when they are less likely to be merely reflective of stroke severity. Studies such as NOMAS with access to samples both before and after stroke have shown that hsCRP, an acute phase protein, is elevated acutely after stroke.43 Taken together, these data may be interpreted to indicate that among patients with smaller infarcts, in whom the levels of hsCRP are not as confounded by stroke severity, and in whom hsCRP is measured after the acute phase reaction to stroke has diminished, hsCRP does have prognostic value for recurrent vascular events, including stroke.

The treatment implications of elevations in hsCRP at the time of stroke remain uncertain. In particular, hsCRP levels did not predict a response to dual antiplatelet therapy, although one prespecified hypothesis of the study was that dual antiplatelet therapy would be of greater benefit in those who had elevations in hsCRP. Recent results from a large primary prevention trial using rosuvastatin provide indirect evidence of the utility of measuring hsCRP as a determinant of whether to initiate statin therapy in otherwise low-risk patients.44 The relevance of this finding to patients with stroke is unclear, however. In the present study, the effect of hsCRP on risk of recurrent stroke and other events persisted after adjusting for statin use, and was not influenced by statin use at baseline. Statin use was not randomized, however, and was left to the discretion of the treating physician, though the majority of patients were treated with a statin. Future studies will be needed to determine whether other specific therapies should be used in stroke patients with elevated hsCRP. The present study does provide some evidence, however, that hsCRP may be useful as a prognostic marker for potential risk stratification in future trials.

Our study has several strengths. One major strength of LIMITS is its international multi-center design, but with a central laboratory, thereby minimizing inter-laboratory error. We also had a race-ethnically diverse population, including a significant proportion of Hispanics, who have been traditionally understudied in stroke trials.45 In addition, we were able to assess the role of hsCRP in relation to a specific stroke subtype, using a well-defined population of MRI-proven lacunar strokes, thereby limiting some confounding due to stroke size, etiology, and severity. Third, by nesting LIMITS within a secondary prevention trial, we were able to address interactions of treatment effect and hsCRP level. Future analyses will allow us to determine the effect of both antihypertensive and antiplatelet therapies on hsCRP levels, as well as the association of hsCRP with other outcomes, including cognition, collected in SPS3.

Our study had some limitations, as well. The nesting of LIMITS within SPS3 may have led to selection bias due to the inclusion and exclusion criteria inherent in a clinical trial. Because our study was nested, we also had limited data about chronic inflammatory diseases or clinical infections. The proportion of patients with clinically apparent conditions is likely to be low, however. Rates of infection at the time of stroke range from 6% to 25%, depending upon the population studied and methods used to detect infection.46,47,48,49 The likelihood of infection at time of stroke is related to stroke severity,50 however. Higher rates of infection were also seen in population-based studies than in secondary analyses of clinical trials probably because patients in clinical trials tend to have fewer comorbidities. Patients with permanently disabling strokes were excluded from SPS3, and most patients were likely to be enrolled at a time after acute infections were likely to have resolved.

In conclusion, LIMITS provides evidence that hsCRP levels measured in the subacute phase after recent lacunar stroke predict the risk of recurrent stroke and other vascular events, though they do not predict response to dual antiplatelet therapy. Future studies of this cohort may elucidate whether hsCRP predicts cognitive change after lacunar stroke, and whether hsCRP may be used to stratify patients in future clinical trials.

Acknowledgments

Source of Funding

The SPS3 trial was funded by a cooperative agreement with the NIH/NINDS, and this ancillary study was funded through an ancillary R01 (NINDS NS050724) from the NINDS. Supplemental funding was provided by the Bristol-Myers Squibb-Sanofi Pharmaceutical Partnership.

Appendix

THE LIMITS INVESTIGATORS

CLINICAL SITES IN ORDER OF PARTICIPANTS ENROLLED (n=number participants)

United States, 31 sites (n = 639)

University of Kentucky, Lexington, KY: Creed Pettigrew, MD, Anand Vaishnav, MD, Peter Sawaya, MD, Anna Fowler, RN, Nedda Hughes, PA, Johnya Rice, RN, Kathy Vanderpool, RN (51); University of California San Diego, San Diego, CA: Brett Meyer, MD, Christy Jackson, MD, Paul Gamble, MD, Nancy Kelly, RN, Janet Warner, RN, Jo Bell, RN (47); Rochester, Stroke Center, Rochester, MN: Irene Meissner, MD, John Graves, MD, Deb Herzig, RN, Jody Covalt, RN (45); St Louis University, St. Louis, MO: Salvador Cruz-Flores, MD, H Douglas Walden, MD, Eve Holzemer, RN (44); University of Arizona, Tucson, Tucson, AZ: Bruce Coull, MD, Lien Howard, MD, Mina Malekniazi, RN, Melissa VanSkiver, Denise Bruck, Stacey Redman, RN (42); Metrohealth Medical Center, Cleveland, OH: Joseph Hanna, MD, Thomas Zipp, MD, Scott Bailey, RN, Dana Cook, RN, Alice Liskay, RN, Dana Simcox, RN, Joan Kappler, RN (38); Mayo Clinic Scottsdale, Scottsdale, AZ: Bart Demaerschalk, MD, Michael Hogan, MD, Daniel Wochos, MD, Judith Wieser, RN, Barbara Cleary, RN, Lori Wood, RN (37); Henry Ford Hospital, Detroit, MI: Angelo Katramados, MD, Brian Silver, MD, Jerry Yee, MD, Krisy Aiello, RN, Kathleen Wilson, RN, Sharon McCarthy, RN (35); Boston University, Boston, MA: Carlos Kase, MD, Irene Gavras, MD, Helena Lau, RN, Matt Ogrodnik, Nancy Allen, RN (34); Berman Center for Clinical Research, Minneapolis, MN: David Anderson, MD, Richard Grimm, MD, Donna Brauer, RN (27); University of South Alabama, Mobile, AL: Dean Naritoku, MD, Richard Zweifler, MD, Michael Culpepper, MD, Mel Parnell, RN, Robin Yunker, RN, Kelly Boots, RN, Renay Drinkard, RN, Rachel Backlin (25); Columbia University Medical Center, New York, NY: Mitch Elkind, MD, Russell John Crew, MD, Jai Radhakrishan, MD, Tania E. Corporan, MD, Julisa Diaz, MD, Rebeca Aragon, BS (25); University of Texas Health Science Center at San Antonio, San Antonio, TX: Oscar Benavente, MD, Robert Hart, MD, Pablo Pergola, MD, Santiago Palacio, MD, Irma Castro, Arlene Farias, Ana Roldan, MD, MS (21); Cooper Health System Dept of Medicine, Camden, NJ: Tom Mirsen, MD, Susan McAllister, MD, Arnaud Bastien, MD, Patricia Niblack, MLT (17); University Hospitals Case Medical Center, Cleveland, OH: Sophia Sundararajan, MD, Mahboob Rahman, MD, Tom Horvath, David Korosec, RN, Chris Murphy, RN (16); Oregon Health & Science University, Portland, OR: Helmi Lutsep, MD, Don Girard, MD, Kali Seisler, Megan Cingel, Megan Ross, Rachel Stone, Darren Larsen, RN, Ann Doherty, RN (15); Medical College of Wisconsin, Milwaukee, WI: Diane Book, MD, Sunu Eapen, MD, Clarence Grimm, MD, Barbara Blaney, Stephanie Rozman, Linda Gaertner, RN, Erin Bradenburg, Laura Loomis, RN, Jolene Monarch- Cotton, RN, Jean Ravavelli-Meyer, RN, Anna Golembieski, RN (15); University of Miami, Miller School, Miami, FL: José Romano, MD, Gustavo Ortiz, MD, Maria del Carmen Lichtenberger, MALS (15); University of Texas Southwestern, Dallas, TX: Mark Johnson, MD, Yinghui Liu, MD, Robert Goldsteen, MD, April Blair, MSW, Gregg Wright, Naomie Gathua, RN (14); Ruan Neurology Clinic and Research Center, Des Moines, IA: Michael Jacoby, MD, David Jones, MD, Jeffrey DeFrancisco, MD, Theresa Hamm, RN (14); University of Rochester Medical Center, Rochester, NY: Scott W. Burgin, MD, Joshua Hollander, MD, Walter Polashenski, MD, Patricia Wallace, RN, Cheryl Weber, RN (12); Helen Hayes Hospital, West Haverstraw, NY: Jason Greenberg, MD, Laura Lennihan, MD, Marjorie King, MD, Laura Tenteromano, RN (9); Research Foundation of SUNY, Buffalo, NY: Lorainne Pereira, MD, Marilou Ching, MD, Robert Sawyer, MD, Kathy Parkes, RN, Cheryl Conover (8); North General Hospital, New York, NY: Jesse Weinberger, MD, Lewis Wright, MD, Dorothy Burch, RN (8); University of Rochester, Rochester, NY: Curtis Benesch, MD, John Bisognano, MD, Ann Leonhardt, RN, Justine Zentner, RN, Molly Hildreth, LVN (6); Marshfield Clinic Dept. of Neurology, Marshfield, WI: Percy Karanjia, MD, Narayana Murali, MD, Richard Dart, MD, Kathleen Mancl, CCRP (5); Wake Forest University, Winston-Salem, NC: David Lefkowitz, MD, Levy Pavel, MD, Nancy Buchheimer, Sara Vaughn, BA, Emily Smith, Jean Satterfield, RN (4); University of Washington at St. Louis, St. Louis, MO: Renee Van Stavern, MD, Angela Brown, MD, Jannie Serna, RN, Jill Newgent, RN, Julie Naylor, Laura Carpenter (4); University of Alberta Stroke Research NACTRC, Edmonton, AB: Ashfaq Shuab, MD, Khurshid Khan, MD, Naeen Dean, MD, Frederika, Herbert, RN, Karen Kastelic, RN (3); Sutter Neuroscience Institute, Sacramento, CA,: Richard Atkinson, MD, Roger Lieberman, MD, Teresa Carter, RN, Pat Zrelak, RN, Nola Kenney, RN (2); St. John’s Mercy Medical Center, St. Louis, MO: William Logan, MD, David Carpenter, MD, Sally Schroer, RN (1);

Canada, 2 sites (n = 52)

Charles LeMoyne Hospital, Greenfield Park, QC: Leo Berger, MD, Sylvain Brunet, MD, Johanne Pontbriand, RN, Martine Mainville, Denise Racicot, RN (29); McGill University - Montreal General, Montreal, QC: Robert Côté, MD, Laurence Green, MD, Lisa Wadup, LVN, Anne-Marie Fontaine, RN (23);

Chile, 2 site (n = 90)

Hospital Clinico Universidad Católica de Chile, Santiago: Jorge Tapia, MD, Ivan Esteban Godoy, MD, Marcela Valdes, RN (46); Universidad Católica, Santiago: Hospital Naval, Viña del Mar: Gonzalo Matamala, MD, Helmut Goecke, MD, Marcela Parra, RN, Jessica Pozo, RN (44);

Ecuador, 1 site (n = 171)

Hospital-Clínica-Kennedy, Guayaquil: Oscar del Brutto, MD, Rocio Santibáñez, MD, Joffre Lara, MD, Mauricio Zambrano (171)

Mexico, 2 sites (n = 18)

Hospital Civil/México, Guadalajara, JAL: José Luis Ruiz Sandoval, MD, Eduardo Salcido Vásquez, MD, Carmen Ruiz (12); Instituto Nacional de Neurología y Neurocirugía, México City: Antonio Arauz, MD, G Amin Cervantes, MD, Adolfo Leyva, MD, Itzel Camacho, MD (6);

Peru, 1 site (n = 151)

Hospital Sabogal, Lima, Perú: Edwin Javier Pretell, MD, José Valdivia, MD, Marissa Pretell (151)

Spain, 6 sites (n = 137)

Hospital de Girona Dr. Josep Trueta, Girona: Joaquín Serena Leal, MD, Mar Castellanos, MD, Verónica Cruz, Mercè Cepeda, RN (36); Hospital Del Sagrat Cor, Barcelona: Adrià Arboix, MD, Antoni Pelegrí, MD, Lorena Blanco (31); Hospital del Bellvitge, Barcelona: Francisco Rubio Borrego, MD, Francisco Gudiol, RN (28); Hospital Universitario German Trias i Pujol, Barcelona: Meritxell Gomis, MD, Juan Arenillas, MD, Antonio Dávalos, MD, Ana Suñol, RN, Silvia Reverté, RN (17); Hospital del Mar, Barcelona: Jaume Roquer, MD, Ana Oliveras Serrano, MD, Jordi Jiménez Conde, MD, Ana Rodríguez, MD, Gemma Romeral, RN (13); Hospital Clínico Universitario de Santiago de Compostela, Santiago de Compostela: José Castillo Sánchez, MD, Miguel Blanco González, MD, Manuel Rodríguez, MD, Isabel Jiménez, Jaime Rodríguez, RN (12);

STATISTICAL AND DATA MANAGEMENT CENTER: (University of Alabama at Birmingham, AL, USA) Leslie McClure, PhD (PI), Christopher Coffey, PhD (co-PI), Jeff Szychowski, PhD, George Howard, PhD, Charles Katholi, PhD, Yu Zhang, MS, Kalyani Peri, MS, Charles Allcorn, Richard Mailhot, Lisa Irby, Fekisha Guyton, MPA, Mary Jo Sewell

SPS3 COORDINATING CENTER: (University of British Columbia, Vancouver, BC, Canada and University of Texas Health Science Center at San Antonio, TX, USA) Oscar Benavente, MD (PI), Robert Hart, MD (Co-PI), Pablo Pergola, MD (Hypertension Co-PI), Ana Roldan, MD, MS (Study Manager for sites in Latin America and Spain), Marie-France Benavente, RN, BScN (Study Manager for sites in North America), Carole White, RN, PhD (Study Manager for sites in North America), Camilla Robu, MBA (Project Administrator), Che Kelly, RN, M Ed (Project Administrator), Robert Talbert, PharmD (Pharmacology Consultant), Eduardo Martinez (Project Data Manager); Neuroradiology Core: Carlos Bazan, MD, Gabriela Pergola, MD; Neuropsychology Core: Lesly Pearce, MS, Raymond Costello, PhD, Claudia Jacova, PhD, Luisa Camelia, PhD, Crystal Mendoza, MA, Brandy Pratt, B. Kin, Steve Holliday, PhD

LIMITS COORDINATING CENTER: (Columbia University, NY, USA) Mitchell S. V. Elkind, MD (LIMITS PI), Jorge Luna.

Footnotes

Disclosures

Dr. Elkind receives personal compensation from the Bristol-Myers Squibb-Pfizer Partnership for consultation related to apixaban and stroke. Dr. Coffey has a consulting contract with ZZ Biotech, LLC. Brett C. Meyer is on Speaker’s Bureaus for Genentech and Boehringer Ingelheim, Inc.

Clinical Trial Registration-URL: http://www.clinicaltrials.gov. Unique identifier: NCT00059306.

References

  • 1.Pearson TA, Mensah GA, Alexander RW, Anderson JL, Cannon RO, Criqui M, et al. Markers of inflammation and cardiovascular disease: application to clinical and public health practice: a statement for healthcare professionals from the Centers for Disease Control and Prevention and the American Heart Association. Circulation. 2003;107:499–511. doi: 10.1161/01.cir.0000052939.59093.45. [DOI] [PubMed] [Google Scholar]
  • 2.Di Napoli M, Schwaninger M, Cappelli R, Ceccarelli E, Di Gianfilippo G, Donati C, et al. Evaluation of C-reactive protein measurement for assessing risk and prognosis in ischemic stroke: a statement for healthcare professional from the CRP Pooling Project. Stroke. 2005;36:1316–1329. doi: 10.1161/01.STR.0000165929.78756.ed. [DOI] [PubMed] [Google Scholar]
  • 3.Ridker PM, Cushman M, Stampfer MJ, Tracy RP, Hennekens CH. Inflammation, aspirin, and the risk of cardiovascular disease in apparently healthy men. N Engl J Med. 1997;336:973–979. doi: 10.1056/NEJM199704033361401. [DOI] [PubMed] [Google Scholar]
  • 4.Ridker PM, Danielson E, Fonseca FAH, Genest J, Gotto AM, Jr, Kastelein JJP, et al. Rosuvastatin to prevent vascular events in men and women with elevated C-reactive protein. N Engl J Med. 2008;359:2195–2207. doi: 10.1056/NEJMoa0807646. [DOI] [PubMed] [Google Scholar]
  • 5.Heeschen C, Dimmeler S, Hamm CW, van den Brand MJ, Boersma E, Zeiher AM, et al. Soluble CD40 Ligand in acute coronary syndromes. New Engl J Med. 2003;348:1104–1111. doi: 10.1056/NEJMoa022600. [DOI] [PubMed] [Google Scholar]
  • 6.Chamorro A, Sacco R, Mohr JP, Foulkes MA, Kase CS, Tatemichi TK, et al. Clinical-computed tomographic correlations of lacunar infarction in the Stroke Data Bank. Stroke. 1991;22:175–181. doi: 10.1161/01.str.22.2.175. [DOI] [PubMed] [Google Scholar]
  • 7.Wolfe CDA, Rudd AG, Howard R, Coshall C, Stewart J, Lawrence E, et al. Incidence and case fatality rates of stroke subtypes in a multiethnic population: the South London Stroke Register. J Neurol Neurosurg Psych. 2002;72:211–216. doi: 10.1136/jnnp.72.2.211. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Ross G, Petrovitch H, White L, Masaki KH, Li CY, Curb JD, et al. Characterization of risk factors for vascular dementia: The Honolulu-Asia Aging Study. Neurology. 1999;53:337–343. doi: 10.1212/wnl.53.2.337. [DOI] [PubMed] [Google Scholar]
  • 9.Tatemichi TK, Desmond D, Paik M, Figueroa M, Gropen TI, Stern Y, et al. Clinical determinants of dementia related to stroke. Ann Neurol. 1993;33:568–575. doi: 10.1002/ana.410330603. [DOI] [PubMed] [Google Scholar]
  • 10.Bamford JM, Sandercock P, Jones L, Warlow C. The natural history of lacunar infarction: The Oxfordshire Community Stroke Project. Stroke. 1987;18:545–551. doi: 10.1161/01.str.18.3.545. [DOI] [PubMed] [Google Scholar]
  • 11.Mohr JP, Thompson JLP, Lazar RM, Levin B, Sacco RL, Furie KL, et al. A comparison of warfarin and aspirin for the prevention of recurrent ischemic stroke. N Engl J Med. 2001;345:1444–1451. doi: 10.1056/NEJMoa011258. [DOI] [PubMed] [Google Scholar]
  • 12.Petty GW, Brown RDJ, Whisnant JP, Sicks JD, O’Fallon WM, Wiebers DO. Ischemic stroke subtypes: a population-based study of functional outcome, survival, and recurrence. Stroke. 2000;31:1062–1068. doi: 10.1161/01.str.31.5.1062. [DOI] [PubMed] [Google Scholar]
  • 13.Salgado AV, Ferro JM, Gouveia-Oliveira A. Long-term prognosis of first-ever lacunar stroke. Stroke. 1996;27:661–666. doi: 10.1161/01.str.27.4.661. [DOI] [PubMed] [Google Scholar]
  • 14.Arboix A, Marti-Vilalta JL, Garcia JH. Clinical study of 227 patients with lacunar infarcts. Stroke. 1990;21:842–847. doi: 10.1161/01.str.21.6.842. [DOI] [PubMed] [Google Scholar]
  • 15.Lammie AG, Brannan F. Nonhypertensive cerebral small-vessel disease. Stroke. 1997;28:2222–2229. doi: 10.1161/01.str.28.11.2222. [DOI] [PubMed] [Google Scholar]
  • 16.Mast H, Thompson AJ, Lee SH, Mohr JP, Sacco R. Hypertension and diabetes mellitus as determinants of multiple lacunar infarcts. Stroke. 1995;26:30–33. doi: 10.1161/01.str.26.1.30. [DOI] [PubMed] [Google Scholar]
  • 17.Vermeer SE, Prins ND, den Heijer T, Hofman A, Koudstaal PJ, Breteler MM. Silent brain infarcts and the risk of dementia and cognitive decline. New Engl J Med. 2003;348:1215–1222. doi: 10.1056/NEJMoa022066. [DOI] [PubMed] [Google Scholar]
  • 18.Wright CB, Paik MC, Brown TR, Stabler SP, Allen RH, Sacco RL, et al. Total homocysteine is associated with white matter hyperintensity volume: the Northern Manhattan Study. Stroke. 2005;36:1207–11. doi: 10.1161/01.STR.0000165923.02318.22. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Young G, Halliday GM, Kril JJ. Neuropathologic correlates of white matter hyperintensities. Neurology. 2008;71:804–811. doi: 10.1212/01.wnl.0000319691.50117.54. [DOI] [PubMed] [Google Scholar]
  • 20.Lammie GA, Brannan F, Slattery J, Warlow C. Nonhypertensive cerebral small-vessel disease. An autopsy study. Stroke. 1997;11:2222–9. doi: 10.1161/01.str.28.11.2222. [DOI] [PubMed] [Google Scholar]
  • 21.Gan R, Sacco RL, Kargman DE, Roberts JK, Boden-Albala B, Gu Q. Testing the validity of the lacunar hypothesis: The Northern Manhattan Stroke Study experience. Neurology. 1997;48:1204–1211. doi: 10.1212/wnl.48.5.1204. [DOI] [PubMed] [Google Scholar]
  • 22.Revilla M, Obach V, Cervera A, Davalos A, Castillo J, Chamorro A. A -174G/C polymorphism of the interleukin-6 gene in patients with lacunar infarction. Neurosci Lett. 2002;324:29–32. doi: 10.1016/s0304-3940(02)00169-6. [DOI] [PubMed] [Google Scholar]
  • 23.Rundek T, Elkind MS, Pittman JG, Boden-Albala B, Humphries SE, Sacco RL. Increased carotid intima-media thickness is associated with allelic variants of stromelysin-1, interleukin-6 and hepatic lipase genes: The Northern Manhattan Prospective Cohort Study. Stroke. 2002;33:1420–1423. doi: 10.1161/01.STR.0000015558.63492.B6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Castellanos M, Castillo J, Garcia MM, Leira R, Serena J, Chamorro A, Davalos A. Inflammation-mediated damage in progressing lacunar infarctions: a potential therapeutic target. Stroke. 2002;33:982–987. doi: 10.1161/hs0402.105339. [DOI] [PubMed] [Google Scholar]
  • 25.Elkind MSV, Sciacca RR, Tondella MLC, Feikin DR, Fields BS, Homma S, et al. Seropositivity to Chlamydia pneumoniae is associated with risk of first ischemic stroke. Stroke. 2006;37:790–795. doi: 10.1161/01.STR.0000202624.89869.e9. [DOI] [PubMed] [Google Scholar]
  • 26.Elkind MSV, Cheng J, Rundek T, Boden-Albala B, Sacco RL. Leukocyte count predicts outcome after ischemic stroke: The Northern Manhattan Stroke Study. J Stroke Cerebrovasc Dis. 2004;13:220–227. doi: 10.1016/j.jstrokecerebrovasdis.2004.07.004. [DOI] [PubMed] [Google Scholar]
  • 27.Benavente OR, Hart RG, McClure LA, Szychowski JM, Coffey CS, et al. SPS3 Investigators. Effects of clopidogrel added to aspirin in patients with recent lacunar stroke. N Engl J Med. 2012;367:817–825. doi: 10.1056/NEJMoa1204133. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Benavente OR, Coffey CS, Conwit R, Hart RG, McClure LA, et al. SPS3 Study Group. Blood-pressure targets in patients with recent lacunar stroke: the SPS3 randomised trial. Lancet. 2013;382:507–515. doi: 10.1016/S0140-6736(13)60852-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Benavente OR, White CL, Pearce L, Pergola P, Roldan A, Benavente MF, et al. The Secondary Prevention of Small Subcortical Strokes (SPS3) study. Int J Stroke. 2011;6:164–175. doi: 10.1111/j.1747-4949.2010.00573.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Elkind MSV, Luna JM, Coffey CS, McClure LA, Liu KM, Spitalnik S, et al. The Levels of Inflammatory Markers in the Treatment of Stroke (LIMITS) Study: Inflammatory Biomarkers as Risk Predictors after Lacunar Stroke. Int J Stroke. 2010;5:117–25. doi: 10.1111/j.1747-4949.2010.00420.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Muir KW, Weir CJ, Alwan W, Squire IB, Lees KR. C-reactive protein and outcome after ischemic stroke. Stroke. 1999;30:981–985. doi: 10.1161/01.str.30.5.981. [DOI] [PubMed] [Google Scholar]
  • 32.Di Napoli M, Papa F, Bocola V. Prognostic influence of increased C-reactive protein and fibrinogen levels in ischemic stroke. Stroke. 2001;32:133–138. doi: 10.1161/01.str.32.1.133. [DOI] [PubMed] [Google Scholar]
  • 33.McClure LA, Szychowski JM, Benavente O, Coffey CS. Sample size re-estimation in an ongoing NIH-sponsored clinical trial: the secondary prevention of small subcortical strokes experience. Contemp Clin Trials. 2012;33:1088–93. doi: 10.1016/j.cct.2012.06.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Ford ES, Giles WH. Serum C-reactive protein and self-reported stroke: findings from the Third National Health and Nutrition Examination Survey. Thrombosis & Vascular Biology. 2000;20:1052–1056. doi: 10.1161/01.atv.20.4.1052. [DOI] [PubMed] [Google Scholar]
  • 35.Gussekloo J, Schaap MC, Frolich M, Blauw GJ, Westendorp RG. C-reactive protein is a strong but nonspecific risk factor of fatal stroke in elderly persons. Arterioscler Thromb Vasc Biol. 2000;20:1047–1051. doi: 10.1161/01.atv.20.4.1047. [DOI] [PubMed] [Google Scholar]
  • 36.Rost NS, Wolf PA, Kase CS, Kelly-Hayes M, Silbershatz H, Massaro JM, et al. Plasma concentration of C-reactive protein and risk of ischemic stroke and transient ischemic attack. Stroke. 2001;32:2575–2579. doi: 10.1161/hs1101.098151. [DOI] [PubMed] [Google Scholar]
  • 37.Curb JD, Abbott RD, Rodriguez BL, Sakkinen P, Popper JS, Tyano K, et al. C-reactive protein and the future risk of thromboembolic stroke in healthy men. Circulation. 2003;107:2016–2020. doi: 10.1161/01.CIR.0000065228.20100.F7. [DOI] [PubMed] [Google Scholar]
  • 38.Elkind MSV, Luna JM, Moon YP, Liu KM, Spitalnik S, Paik MC, et al. High-sensitivity C-reactive protein predicts mortality but not stroke in a multi-ethnic cohort: The Northern Manhattan Study. Neurology. 2009;73:1300–1307. doi: 10.1212/WNL.0b013e3181bd10bc. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Kaptoge S, Di Angelantonio E, Lowe G, Pepys MB, Thompson SG, Collins R, et al. C-reactive protein concentration and risk of coronary heart disease, stroke, and mortality: an individual participant meta-analysis. Lancet. 2010;375:132–140. doi: 10.1016/S0140-6736(09)61717-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Arenillas JF, Alvarez-Sabin J, Molina CA, Chacón P, Montaner J, Rovira A, et al. C-reactive protein predicts further ischemic events in first-ever transient ischemic attack or stroke patients with intracranial large-artery occlusive disease. Stroke. 2003;3410:2463–2468. doi: 10.1161/01.STR.0000089920.93927.A7. [DOI] [PubMed] [Google Scholar]
  • 41.Elkind MSV, Tai W, Coates K, Paik MC, Sacco RL. Lipoprotein-associated phospholipase A2, C-reactive protein, and outcome after ischemic stroke. Arch Int Med. 2006;166:2073–2080. doi: 10.1001/archinte.166.19.2073. [DOI] [PubMed] [Google Scholar]
  • 42.Woodward M, Lowe GD, Campbell DJ, et al. Associations of inflammatory and hemostatic variables with the risk of recurrent stroke. Stroke. 2005;36:2143–2147. doi: 10.1161/01.STR.0000181754.38408.4c. [DOI] [PubMed] [Google Scholar]
  • 43.Elkind MSV, Leon V, Moon YP, Paik MC, Sacco RL. High-sensitivity C-reactive protein and lipoprotein-associated phospholipase A2 stability before and after stroke and myocardial infarction. Stroke. 2009;40:3233–3237. doi: 10.1161/STROKEAHA.109.552802. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Everett BM, Glynn RJ, MacFadyen JG, Ridker PM. Rosuvastatin in the prevention of stroke among men and women with elevated levels of C-reactive protein: Justification for the Use of Statins in Prevention: An Intervention Trial Evaluating Rosuvastatin (JUPITER) Circulation. 2010;121:143–150. doi: 10.1161/CIRCULATIONAHA.109.874834. [DOI] [PubMed] [Google Scholar]
  • 45.Cruz-Flores S, Rabinstein A, Biller J, Elkind MSV, Griffith P, Gorelick P, et al. Race-Ethnic Disparities in Stroke Care: The American Experience. A Statement for Healthcare Professionals from the American Heart Association/American Stroke Association Council on Stroke. Stroke. 2011;42:2091–2116. doi: 10.1161/STR.0b013e3182213e24. [DOI] [PubMed] [Google Scholar]
  • 46.Johnston KC, Li JY, Lyden PD, Hanson SK, Feasby TE, Adams R, et al. Medical and neurological complications of ischemic stroke: experience from the RANTTAS trial. Stroke. 1999;29:447–453. doi: 10.1161/01.str.29.2.447. [DOI] [PubMed] [Google Scholar]
  • 47.Langhorne P, Stott DJ, Robertson L, MacDonald J, Jones L, McAlpine C. Medical complications after stroke: a multicenter study. Stroke. 2000;31:1223–1229. doi: 10.1161/01.str.31.6.1223. [DOI] [PubMed] [Google Scholar]
  • 48.Sacco RL, DeRosa JT, Haley EC, Levin B, Ordronneau P, Phillips SJ, et al. Glycine antagonist in neuroprotection for patients with acute stroke. GAIN Americas: A randomized controlled trial. JAMA. 2001;285:1719–1728. doi: 10.1001/jama.285.13.1719. [DOI] [PubMed] [Google Scholar]
  • 49.Davenport RJ, Dennis MS, Wellwood I, Warlow C. Complications after acute stroke. Stroke. 1996;27:415–420. doi: 10.1161/01.str.27.3.415. [DOI] [PubMed] [Google Scholar]
  • 50.Grau AJ, Buggle F, Schnitzler P, Spiel M, Lichy C, Hacke W. Fever and infection early after ischemic stroke. J Neurolog Sci. 1999;171:115–20. doi: 10.1016/s0022-510x(99)00261-0. [DOI] [PubMed] [Google Scholar]

RESOURCES