Profiles of lacunar and nonlacunar stroke

Abstract

Objective: Determining which small deep infarcts (SDIs) are of lacunar, arterial, or cardioembolic etiology is challenging, but important in delivering optimal stroke prevention therapy. We sought to distinguish lacunar from nonlacunar causes of SDIs using a gene expression profile.

Methods: A total of 184 ischemic strokes were analyzed. Lacunar stroke was defined as a lacunar syndrome with infarction <15mm in a region supplied by penetrating arteries. RNA from blood was processed on whole genome microarrays. Genes differentially expressed between lacunar (n = 30) and nonlacunar strokes (n = 86) were identified (false discovery rate ≤ 0.05, fold change >|1.5|) and used to develop a prediction model. The model was evaluated by cross-validation and in a second test cohort (n = 36). The etiology of SDIs of unclear cause (SDIs ≥ 15mm or SDIs with potential embolic source) (n = 32) was predicted using the derived model.

Results: A 41-gene profile discriminated lacunar from nonlacunar stroke with >90% sensitivity and specificity. Of the 32 SDIs of unclear cause, 15 were predicted to be lacunar, and 17 were predicted to be nonlacunar. The identified profile represents differences in immune response between lacunar and nonlacunar stroke.

Interpretation: Profiles of differentially expressed genes can distinguish lacunar from nonlacunar stroke. SDIs of unclear cause were frequently predicted to be of nonlacunar etiology, suggesting that comprehensive workup of SDIs is important to identify potential cardioembolic and arterial causes. Further study is required to evaluate the gene profile in an independent cohort and determine the clinical and treatment implications of SDIs of predicted nonlacunar etiology.

Figure 1

Cluster plot of the 41 probesets (40 genes) that distinguish lacunar stroke from non-lacunar stroke. Subjects are shown on the x-axis and genes are shown on the y-axis. Each lacunar stroke is shown by an orange bar, and each non-lacunar stroke is shown by a green bar. Up regulated genes are shown in red, and down regulated genes in blue. Though no single probeset is able to completely separate every single patient with lacunar stroke from non-lacunar stroke, the combined information from each gene in the profile can separate lacunar from non-lacunar stroke for nearly all patients studied.

Figure 2

Probability plot of the predicted diagnosis of lacunar and non-lacunar stroke based on 10-fold cross-validation analysis using the linear discriminant analysis model for the 41 probesets (40 genes). 2A. The predicted probability of lacunar and non-lacunar stroke in the 30 patients diagnosed clinically as lacunar stroke. Eight subjects were predicted to have a gene expression profile similar to those of non-lacunar stroke, and 22 were predicted to be lacunar stroke. 2B. The predicted probability of non-lacunar and lacunar stroke in the 86 patients with non-lacunar stroke. Eighty of the 86 were predicted to be non-lacunar stroke.