Early neurological deterioration (END) occurs in up to one-third of patients with acute ischemic stroke after hospital admission and is associated with worse outcomes (1). A variety of pathophysiological mechanisms cause END, including progressive infarction (PI), increased intracranial pressure, recurrent cerebral ischemia, and secondary parenchymal bleeding (2). Among them, progressive infarction is the most frequent cause, accounting for ~30% (2).

Small subcortical infarction (SSI), regarded as the small vessel occlusion subtype in stroke classification, usually has a small infarction volume and leads to relatively limited functional deficits (3, 4). However, between 20 and 30% of SSI patients experience END, which severely affects their functional prognosis (5, 6). According to the pathophysiological mechanism, Fisher and Caplan divided SSI into two main types: lacunar infarction (LI) and branch atheromatous disease (BAD) (7, 8). In recent years, several studies have reported that patients with BAD are more likely to develop worsening neurological deficits than patients with LI (9–11). Distinguishing between these two entities is of clinical importance, but knowledge of this distinction is limited. Branch atheromatous disease is usually defined as a lesion extending three or more consecutive slices or an infarct diameter ≥15 mm in diffusion-weighted imaging (DWI). Lacunar infarction is usually considered as a single lesion of <15 mm in the greatest diameter within the territory of the brainstem or basal penetrating arteries (9, 10). However, with the development of neuroimaging, inconsistencies have been found in the radiological differentiation between BAD and LI by the diameter and slice number of the infarction (12, 13). In addition, from the clinical perspective, brainstem infarction belongs to the small vessel occlusion subtype of stroke that rarely extends three slices.

In the present study, we did not attempt to distinguish between BAD and LI to investigate and purely focus on the relationship between the neuroimaging characteristics of anterior circulation small subcortical infarction and progressive infarction.

A total of 86 patients (60 men; mean age, 62.9 y) were recruited for analysis. A total of 31 (36%) patients and 55 (64%) patients were classified as PI and non-progressive infarction (non-PI) patients, respectively.

The baseline demographic, clinical, and laboratory characteristics of the two groups are listed in Table 1. Compared with patients without PI, patients with PI had higher NIHSS scores at discharge, higher mRS scores at 3 months after onset, higher levels of triglycerides, and lower levels of BUN and PT (all p < 0.05). However, there were no significant differences between these two groups in age, sex, the prevalence of hypertension and diabetes, the neutrophil–lymphocyte ratio, fasting glucose levels, other blood lipid levels, other coagulation function indicators, or urine-specific gravity.

The baseline neuroimaging characteristics, including the infarct diameter, the number of lesion slices, and the leukoaraiosis scores, are also shown in Table 1. There were nearly 1.8 times as many patients with PI with more elongated lesions compared with those without PI (extension ≥ three slices in 80.6 vs. 45.5%, p = 0.002) (Figures 1, 2). However, there was no significant difference between the two groups regardless of the infarct diameters or the proportion of infarcts ≥15mm. In addition, the leukoaraiosis scores showed no significant difference between the PI group and the non-PI group.

A multivariable logistic regression model adjusted for age and sex revealed that an extension ≥ three slices was independently related to PI (Model 1: OR = 4.781, 95% CI 1.677–13.627). Furthermore, an extension ≥ three slices was also independently associated with PI after adjustment for triglyceride levels, BUN levels, and PT (Model 2: OR = 4.867, 95% CI 1.6–14.864). At last, after adjusting for relevant confounders, including fasting glucose levels, total cholesterol levels, triglyceride levels, BUN levels, PT, D-dimer levels, an infarct diameter ≥15 mm, and leukoaraiosis scores, an extension ≥ three slices remained significantly associated with PI (Model 3: OR = 3.584, 95% CI 1.034–12.420) (Table 2).

The main finding of our study was that a lesion extending three or more slices was independently associated with progressive infarction in patients with anterior circulation small subcortical infarction. However, other imaging characteristics, including leukoaraiosis and infarct diameter, were not significantly associated with progressive infarction. In addition, our results demonstrated that patients with progressive infarction had more severe functional disabilities than those without progressive infarction. Our innovative points are as follows: (i) we specifically focused on progressive infarction; not only did the symptoms meet the criteria for early neurological deterioration, but PI was also confirmed by DWI, and (ii) we solely focused on patients with anterior circulation small subcortical infarction.

During the past few years, END has been reported to occur more frequently in BAD-related strokes than in LI. However, the distinction between BAD and LI is mostly based on indirect evidence relying on neuroimaging, although the term BAD was originally created based on pathology (7, 8). Although BAD was considered identifiable by an infarct diameter ≥15 mm and an infarct extension ≥ three slices in most examined work, there is currently no consensus as to the definition of BAD (16). The definitions of BAD used in the three studies focusing on the association between BAD and END were different (9–11). Therefore, we did not distinguish between BAD and LI, but directly investigated the imaging predictors for progressive infarction. At last, we demonstrated that a lesion extending three or more slices was independently associated with PI in anterior circulation small subcortical infarction.

At present, there is controversy about whether an infarction > three slices is a risk factor for END, and several studies have demonstrated different findings. Ohara et al. found that a posterior-type infarct, but not the infarct slice number, was an independent predictor for the progression of motor deficits (17). In a similar manner, Nannoni et al. (18) demonstrated a negative association between a lesion extension ≥ three slices and neurological worsening. Nevertheless, Saji et al. (19) revealed that arterial stiffness, infarct size, and ≥ three infarct slices were all independently associated with progressive neurological deficits. The reasons for this discrepancy may be as follows: (i) the slice thickness on DWI was different; some were 5 mm, and some were 7 mm; (ii) the definitions of END in the related studies were symptomatic definitions, whereas multiple mechanisms are involved in END, including progressive infarction, increased intracranial pressure, recurrent cerebral ischemia and secondary parenchymal bleeding, which may affect the results; and (iii) brainstem infarction rarely exceeds three slices due to vascular anatomy (20), but some studies concurrently recruited anterior and posterior circulation infarction, which may make a difference in conclusions. Our present study defined progressive infarction not only based on symptoms but also, more importantly, on the re-examination of MRI to determine whether END was due to progressive infarction. In addition, we purely focused on anterior circulation and small subcortical infarction and did not include posterior circulation infarction. At last, we found a positive relationship between progressive infarction and ≥ three infarct slices using a slice thickness of 5 mm on DWI.

A limitation of our study was the number of patients in each group. A key point in our enrollment criteria was the definition of progressive infarction, which required not only the aggravation of symptoms but also the confirmation by DWI. Although strict enrollment criteria could accurately determine patients with progressive infarction, they also led to a small sample size.

In conclusion, a lesion extending ≥ three slices was an independent predictor for progressive infarction in patients with anterior circulation small subcortical infarction. In addition, patients with progressive infarction tend to have worse outcomes. Therefore, it is important to identify the predictors for progressive infarction.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by Research Ethics Committee of First Affiliated Hospital of Nanchang University. Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements.

Concept and design: JL, XM, QY, and DH. Acquisition, analysis, or interpretation of data: YL, SLu, QH, ZS, SLi, CL, and XL. Drafting of the manuscript: JL and XM. Critical revision of the manuscript for important intellectual content: QY and DH. Statistical analysis: YQ and YW. Obtained funding: JL and DH. All authors have read and approved the final manuscript.

This study was supported by the National Natural Science Foundation of China for Young Scientists (No. 82101405), the National Natural Science Foundation of China (Nos. 81460199 and 82160252), the Natural Science Foundation of Jiangxi Province for Young Scientists (No. 20212BAB216023), the Science and technology project of Jiangxi Health Commission (No. 202110028), Double thousand talents program of Jiangxi province (No. jxsq2019101021), and Young Talent Research and Cultivation Fund of the First Affiliated Hospital of Nanchang University (No. YFYPY202012).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.