Comparison of small volume infarcts of lacunar and non-lacunar etiologies


  • Conflicts of interest: The authors have no conflicts of interest to disclose.
  • Funding: The MRI substudy of the ENOS trial was supported by a grant from the Singapore Biomedical Research Council (BMRC grant number 04/1/33/19/348). The UK Medical Research Council funds the overall ENOS trial (grant G0501797). Prof P. Bath is Stroke Association Professor of Stroke Medicine, and Chief Investigator of ENOS.

Correspondence: Deidre A. De Silva, Department of Neurology, Singapore General Hospital, Outram Road, Singapore 169608, Singapore.


Small volume infarcts can have underlying lacunar or non-lacunar etiology, which we hypothesize influences clinical and imaging baseline characteristics and natural history. We studied acute ischemic stroke patients of Asian ethnicity in the MR substudy of the Efficacy of Nitric Oxide in Stroke trial [1]. MR imaging was performed at baseline (within 24 h of onset) and day 90, and analyzed with established protocols [2]. Small volume infarcts were defined as baseline Diffusion-weight Imaging (DWI) lesions ≤5 ml [3] and lacunar etiology as infarction in a penetrating arterial territory with no cardioembolic source or large artery obstruction [4].

There were 37 patients with small volume infarcts (26 lacunar, 11 non-lacunar), none were treated with thrombolysis. Patients with lacunar etiology were younger (P = 0·037). There were no differences in gender, prevalence of hypertension, diabetes, hyperlipidemia and smoking, baseline Scandinavian Stroke Scale, TOAST classification and pre-morbid modified Rankin score (mRS). Lacunar infarcts had smaller baseline lesions of DWI [median (IQR)] [1·3 ml (0·8 to 2·0) vs 1·8 ml (1·3 to 3·0), P = 0·070], perfusion (Tmax ≥ 6 s) deficit [0 ml (0 to 0) vs 3·8 ml (0·2 to 7·1), P < 0·001] and absolute mismatch [−1·1 ml (−2·0 to 0·5) vs 14·3 ml (−0·3 to 106), P = 0·001]. Day 90 infarct volume (P = 0·034), absolute (P = 0·071) and relative infarct growth (P = 0·070) were less variable for lacunar than non-lacunar infarcts using the log-rank test (Fig. 1). However, there was no difference in volumes of day 90 infarct (P = 0·273), absolute (P = 0·295) and relative (P = 0·311) infarct growth. Patients with lacunar etiology were 11·8 times more likely to have good functional outcome (mRS ≤ 2) (P = 0·002), independent of baseline DWI volume.

Figure 1.

Variability of imaging outcome measures in lacunar and non- lacunar infarcts. P-values were obtained using the log-rank test for differences in data distribution (variability) between lacunar infarcts and non-lacunar infarcts.

We conclude that small volume infarcts of lacunar etiology have a distinct profile versus non-lacunar etiology, with no/limited perfusion deficit, less variable infarct growth and better functional outcome. Advanced MR imaging can assist with early etiological determination for small volume infarcts, which will allow for better prognostication and guide treatment decisions.


Patients in this study were part of the Efficacy of Nitric Oxide in Stroke (ENOS) trial, and we acknowledge the support from the ENOS trial steering committee.