The progressive course of neurological symptoms in anterior choroidal artery infarcts


  • Conflict of interest: None declared.



Infarctions of the anterior choroidal artery affect multiple anatomical structures, leading to a wide spectrum of neurological deficits with frequent symptom fluctuation or progression.


To assess etiological mechanisms, frequency, and predictors of symptom progression, as well as its impact on prognosis.


Anterior choroidal artery infarct patients were prospectively identified via predefined infarct locations with ischemic lesions ≥1·5 cm vertical diameter in cerebral imaging. Definition of neurological progression was ≥2 National Institutes of Health Stroke Scale points in motor function or ≥4 in total National Institutes of Health Stroke Scale. Stroke etiology was determined according to Trial of Org 10172 in Acute Stroke Treatment (TOAST) classification. We assessed demographical data, risk factors, and acute phase parameters in order to find predictors of neurological progression.


Thirty patients fulfilled the inclusion criteria. Eighteen patients (60%) had neurological progression during days 1–3. Despite similar stroke severity at admission (median National Institutes of Health Stroke Scale in progressive infarcts 4·5 versus 4; P = 0·72), patients with progression had more severe deficits at day 3 (median National Institutes of Health Stroke Scale 9 vs. 3·5; P = 0·04) and worse three-month outcome. Only 31% of patients with progression scored <2 in the modified Rankin Scale compared with 89% without progression (P = 0·01) after three-months. No statistically significant differences regarding possible predictors of progression were found. Magnetic resonance imaging findings and etiological assessment suggest overlapping mechanisms of small and large vessel disease.


Neurological deterioration is frequent in anterior choroidal artery infarcts and is associated with worse outcome. While mechanisms of small and large vessel disease seem to overlap in anterior choroidal artery infarction, we were not able to identify predictors of neurological progression.


The anterior choroidal artery (AChA) is a thin artery arising from the most distal segment of the internal carotid artery. Despite its small diameter with on average 0·94 mm [1], this artery and its branches have a long course supplying multiple anatomical structures [1-3] such as the cerebral peduncle, medial temporal lobe, lateral thalamus, lateral geniculate body, optic radiation, posterior limb of internal capsule, posterior part of putamen, choroidal plexus, and paraventricular tail of the caudate nucleus (sometimes classified as paraventricular part of the corona radiata [4]).

In accordance to these anatomical areas, infarctions of the AChA result in not only a variety of neurological symptoms including lacunar syndromes but also visual field deficits, somnolence, conjugate ocular deviations, and cortical syndromes such as aphasia or neglect [2-4]. A fluctuating or progressing course of neurological signs in infarctions of the AChA territory has been described, sometimes with temporarily complete restitution and other times with severe symptom progression [2, 3, 5].

We aimed to assess the rate of progressive AChA infarctions and its impact on prognosis as well as to investigate etiological mechanisms.


All data were collected in the Munich Harlaching Stroke Center. According to the in-house stroke protocol, acute brain imaging is required for all acute strokes and follow-up imaging for all cases without visible acute brain lesion in the initial examination. Screening of patients was done on a daily basis. Inclusion criteria consisted of hospital admission within 24 h of onset and informed consent. The radiological criterion for study participation was acute cerebral infarction restricted to the AChA territory but involving at least three consecutive slices (five-millimeter thickness) in magnetic resonance imaging (MRI) or computed tomography (CT) scans (Fig. 1). We therefore excluded lacunar infarcts that were described as small deep infarcts measuring less than 1·5 mm in diameter [6].

Figure 1.

Infarcts of the anterior choroidal artery in diffusion weighted (patient 1) and FLAIR (patient 2) magnetic resonance imaging.

Areas that are regularly supplied by the AChA were predefined according to Hupperts et al. [4] and included lateral thalamus, posterior limb of the internal capsule, posterior part of putamen, and paraventricular tail of the caudate nucleus. If the admission or follow-up (within 48 h of admission) brain CT or MRI fulfilled the criteria of AChA infarction, patients were asked to participate in the study and informed consent was collected.

Exclusion criteria were preexisting malignant, infectious, or inflammatory disease and traumatic or vascular events within four-weeks before onset.

Baseline parameters were recorded as shown in Table 1. All patients were neurologically assessed including National Institutes of Health Stroke Scale (NIHSS) examination at admission and on day 3. The neurological examination of the patients included all single items of the NIHSS, as well as extended eye movement disorders, swallowing function, and balance. As part of regular acute stroke management, neurological monitoring was based on the assessment of vigilance and verbal communication (according to Glasgow Coma Scale), as well as motor function (according to motor items in NIHSS), and was performed at least every four-hours during the first three-days on the Stroke Unit.

Table 1. Baseline parameters, clinical course of patients with and without progressiona of neurological symptoms during day 1–3, and outcome after three-months
 Patients with symptom progressionPatients without symptom progressionP-value
n = 18 (60%)n = 12 (40%)
  1. aSymptom progression is defined as worsening in the NIHSS motor score ≥2 or worsening in the NIHSS sum score ≥4 or symptom fluctuation before or after admission fulfilling the same criteria.
  2. IQR, interquartile range; LDL, low-density lipoprotein; mRS, modified Rankin Scale; NIHSS, National Institutes of Health Stroke Scale; SD, standard deviation.
Age in years, mean ± SD72·2 ± 13·865·1 ± 14·80·17
Female gender, n (%)10 (56)5 (42)0·46
Risk factors   
Hypertension, n (%)18 (100)12 (100)
Nicotine abuse, n (%)7 (39)2 (17)0·25
Peripheral arterial disease, n (%)1 (6)2 (17)0·55
Diabetes, n (%)5 (28)00·066
Atrial fibrillation, n (%)1 (6)1 (8)1·00
Blood glucose at admission, mean ± SD147 ± 57117 ± 200·12
Systolic admission blood pressure, mean ± SD177 ± 28169 ± 300·44
Diastolic admission blood pressure, mean ± SD93 ± 2296 ± 230·52
LDL (mg/dl), mean ± SD141 ± 37133 ± 400·41
Lipoprotein (a) (mg/dl), mean ± SD33·0 ± 3520 ± 180·58
Homocysteine (μmol/l), mean ± SD16·4 ± 814·2 ± 50·87
Acute phase parameters on admission, n (%)   
C-reactive protein >1 mg/l15 (83)11 (92)0·51
White blood cell count >10/nl5 (28)6 (50)0·22
Body temperature at admission >36·5°C2 (11)6 (50)0·31
Leukoaraiosis, n (%)17 (94)9 (75)0·28
Internal capsule, n (%)13 (72)8 (67)0·98
Posterolateral division of stiatum, n (%)16 (89)9 (75)0·59
Clinical course   
NIHSS at admission, median (IQR)4·5 (2–7)4·0 (3–6)0·72
NIHSS at day 3, median (IQR)9 (3–10)3·5 (1–4)0·039
mRS at discharge, median (IQR)4 (3·5–5)2 (1–4)0·039
Outcome after three-monthsn = 17n = 11 
mRS at three-months, median (IQR)3 (0·5–4)0 (0–1)0·033
mRS at three-months <2, n (%)5 (29)9 (82)<0·01

Neurological progression was defined as deterioration of ≥2 NIHSS points in motor function or ≥4 points in total NIHSS. Neurological deterioration was also diagnosed if patients reliably reported progressing symptoms before admission consistent with the above-described criteria. Whether symptom progression occurred before or after admission was accounted for in further analyses. Fluctuation of symptoms was used according to the same criteria as described for neurological progression but with improvement of symptoms in the later course.

According to the study protocol, all patients were investigated by electrocardiogram (ECG), duplex sonography of extracranial and intracranial arteries, 24-h ECG monitoring, and transthoracic echocardiography. Transesophageal echocardiography was performed if the standard investigations had not provided sufficient clarity about the etiological mechanism according to Trial of Org 10172 in Acute Stroke Treatment (TOAST) criteria.

Brain MRI was obtained with a PHILIPS (Amsterdam, Netherlands) 1·5-T MRI. The MRI protocol included fluid attenuated inversion recovery (FLAIR), diffusion weighted imaging (DWI), time-of-flight MR angiography, and T2* sequences. A neuroradiologist blinded to clinical data analyzed the MRI scans according to a standardized protocol with regard to the extension of the acute lesion and possible lesions remote from the AChA territory, previous strokes, intracranial arterial stenosis, degree of leukoaraiosis according to the Wahlund score, and number of microbleeds.

Assessment of modified Rankin Scale (mRS) was performed via telephone three-months after stroke onset.

Statistical analysis and ethics

The sample size of 30 confirmed AChA infarcts was planned to ensure valid assumptions about frequency of symptom progression. Statistical analyses were performed using spss (IBM, Armonk, NY, USA), version 14. Correlations between admission parameters and clinical progression were analyzed with the chi-square or Fisher's exact test for differences in proportions and with the Mann–Whitney U-test for continuous variables. An alpha level of 5% was considered to be statistically significant.

The study was approved by the Ethics Committee of the Ludwig-Maximilians-University Munich.


Between May 2006 and February 2008, a total of 1994 ischemic stroke patients were admitted to the hospital. Forty-one (2·1%) of these patients had an ischemic infarction in the AChA territory with a visible continuous lesion in at least three slices (five-millimeter diameter) in MRI or CT. Eight patients could not be included because of delayed recognition of suitability for study participation due to late follow-up MRI scan (n = 5), language barrier (n = 2), and refusal (n = 1).

Thirty-three patients were included in the prospective evaluation. Three initially included patients had to be excluded because radiological assessment revealed infarct extension beyond AChA territory or an additional remote infarct.

The most common neurological deficits were hemiparesis (100%), dysarthria (70%), dysphagia (40%), hemisensory deficits (37%), dysphasic symptoms (20%), and gaze induced nystagmus (7%). We did not detect visual field deficits or ataxia in routine clinical examination.

NIHSS scores on admission and day 3 as well as functional outcome are shown in Table 1. We classified our patients in groups without symptom progression (n = 12) and with symptom progression (n = 18). Seven patients (23%) of the 18 patients with symptom progression had an increase of the NIHSS sum score of ≥4 between admission and day 3. Ten patients had fluctuations of symptoms already before admission but five of them had also worsening of motor NIHSS of ≥2 during in-hospital stay.

Although admission stroke severity was similar in patients with and without progression {median NIHSS 4·5 [interquartile range (IQR) 2–7] vs. 4 (IQR 3–6); P = 0·72}, patients with symptom worsening or fluctuation had more severe deficits at day 3 [median NIHSS 9 (3–10) vs. 3·5 (1–4); P = 0·04].

Three-month outcome was available in 28 patients (93%). Patients with clinical progression during the acute phase had a worse three-month outcome with only 29% (n = 5) of patients with mRS <2 compared with 82% (n = 9) in patients without progression (P = 0·007).

The negative impact of progression remained unchanged if diagnosis of neurological progression was restricted to patients with in-hospital deterioration: median admission NIHSS in patients with in-hospital deterioration – 3 (IQR 0–5·6) in progressive infarcts vs. 5 (IQR 3–7·25) in nonprogressive infarcts, P = 0·15; NIHSS on day 3 – 9·5 (IQR 3·75–10) vs. 4 (IQR 1·75–6·25), P = 0·022. After three-months, 18% (n = 2) versus 71% (n = 12) had a good outcome according to mRS <2 (P = 0·007).

Transesophageal echocardiography was performed in 15 of 25 patients with otherwise unclear etiology. Etiology according to TOAST criteria could be determined in five patients [dissection (n = 1), cardioembolic (n = 2), and large vessel disease (n = 2)]. Twenty-five patients had a stroke of undetermined etiology. All except one patient had atherosclerosis in duplex sonography including three patients with stenosis >50% or occlusion of brain supplying arteries (one patient with contralateral stenosis). Leukoaraiosis was present in all except of three patients investigated with MRI.

No significant difference in baseline parameters could be identified between the groups with progressive and nonprogressive deficits.


Neurological deterioration occurred in 60% of patients with AChA infarcts. The frequency of neurological progression seems to be higher compared with other infarct sub-types [7, 8], and this acute deterioration is associated with worse outcome at three-months. While infarct etiology could not be determined in the majority of cases, we could neither find reasonable predictors nor convince pathomechanisms of neurological deterioration.

Neurological deterioration in AChA infarct patients is a well-known phenomenon [2, 3, 5]. A more recent study found a progression rate of 25% in AChA infarcts using the same definition of neurological worsening [9]. In a recent analysis of AChA infarcts with MR tractography, fiber disruption and involvement of the corticospinal tract were associated with motor deficits and unfavorable outcome [5].

While it is plausible that involvement of the corticospinal tract is correlated with motor deficits, the progression and/or fluctuation of these neurological deficits remain unclear. Because of the lack of serial MRI-DWI images at different stages of stroke evolution, it is unclear whether neurological deterioration is a result of infarct growth. Infarctions of the AChA may be caused not only by vessel occlusion but also by a disproportion of arterial lumen and length of the artery. In view of its length and the areas supplied by this artery, the caliber of the AChA is critically small [1]. Infarctions may therefore be caused by hemodynamic insufficiency and this could well explain the fluctuation or progression of symptoms. However, there was no association between symptom progression and blood pressure at admission or hemodynamic instability in our sample.

Looking for inflammatory mechanisms possibly being associated with deterioration, we could not identify meaningful differences of acute phase parameters between progressive and nonprogressive infarctions.

Frequency of neurological symptoms was similar to previous studies [2, 4]. Motor deficits were present in all patients and accounted for the majority of the neurological deteriorations.

Several etiological mechanisms have been discussed with not only small vessel disease as the most commonly named [2, 4] but also large vessel disease [3, 4, 10], cardio-embolism [3, 4, 10], and dissection [10]. However, the diameter of the AChA (average 0·94 mm) [1] exceeds the upper limit of small penetrating small vessels. According to Fisher [6], lacunar infarcts are caused by atheroma in arteries ranging in size from 400 to 900 μm or from lipohyalinosis of arteries with lumen <200 μm. Although the horizontal diameters of AChA infarcts in our cohort were usually smaller than 1·5 cm, the vertical extension was more than 1·5 cm by our definition of infarct visibility in at least three consecutive slices. Hence, regardless of a relevant leukoaraiosis in most of our patients (90%), we could not classify these infarcts according to TOAST criteria. Leukoaraiosis and the presence of atherosclerosis in carotid duplex suggest overlapping mechanisms of small and large vessel disease. The TOAST classification may not be the optimal etiological categorization for this specific stroke sub-type.

As limitations of the study, we have to acknowledge that the small sample size is a major limitation with respect to the search for possible predictors of neurological deterioration and that the analysis of inflammatory parameters was limited to routinely available parameters.

Neurological deterioration is frequent in acute AChA infarcts and is associated with worse outcome. While mechanisms of small and large vessel disease seem to overlap in AChA infarction, we were not able to identify predictors of neurological progression. Factors leading to progression and stroke evolution in sequential MRI imaging will now be investigated in a prospective study.