€¦  · Web viewWord Count: 3604. Abstract. We aimed to explore the morphological evolution of recent small subcortical infarcts (RSSIs) over 15 months. Moreover, we hypothesized

Longitudinal MRI dynamics of recent small subcortical infarcts and

possible predictorsCOVER TITLE: Evolution of recent small subcortical infarcts

Daniela Pinter (PhD)1*, Thomas Gattringer (MD, PhD)1*, Christian Enzinger (MD)1,2,

Thomas Seifert-Held (MD)1, Markus Kneihsl (MD)1, Simon Fandler (MD)1, Alexander

Pichler (MD, PhD)1, Christian Barro (MD)3, Sebastian Eppinger (MD)1, Lukas Pirpamer

(MSc)1, Gerhard Bachmaier (PhD)4, Stefan Ropele (PhD)1, Joanna M Wardlaw (MD)5,6, Jens

Kuhle (MD)3, Michael Khalil (MD, PhD)1 and Franz Fazekas (MD)1

1 Department of Neurology, Medical University of Graz, Austria

2 Division of Neuroradiology, Vascular and Interventional Radiology, Medical University of

Graz, Austria

3 Neurologic Clinic and Policlinic, Departments of Medicine, Biomedicine and Clinical

Research, University Hospital Basel, University of Basel, Basel, Switzerland

4 Institute for Medical Informatics, Statistics and Documentation, Medical University of

Graz, Austria

5 Brain Research Imaging Centre, Centre for Clinical Brain Sciences, University of

Edinburgh, Edinburgh, EH16 4SB, UK

6 UK Dementia Research Institute at the University of Edinburgh, Edinburgh, EH8 9YL, UK

*Contributed equally

Corresponding author: Thomas Gattringer, MD, PhD, Department of Neurology, Medical

University of Graz, Auenbruggerplatz 22, 8036 Graz, Austria, Phone: 0043 316 385 80231,

Email: [email protected]

Figures: 3

Tables: 2

Supplemental Tables: 2

Word Count: 3604

1

Abstract

We aimed to explore the morphological evolution of recent small subcortical infarcts (RSSIs)

over 15 months. Moreover, we hypothesized that quantitative lesion apparent diffusion

coefficient (ADC) values and serum neurofilament light (NfL) levels predict subsequent

lacunar cavitation. We prospectively studied 78 RSSI patients, who underwent pre-defined

follow-up investigations three and 15 months poststroke using 3T MRI including high-

resolution T1 sequences. To identify potential predictors of cavitation, we determined RSSI

size and quantitative ADC values, and serum NfL using the SIMOA technique. The majority

of RSSIs showed cavitation at three months (N=61, 78%) with only minimal changes

regarding cavitation status thereafter. The maximum axial lacunar diameter decreased from

8mm at three to 7mm at 15 months (p<0.05). RSSIs which cavitated had lower lesional ADC

values and were associated with higher baseline NfL levels compared to those without

cavitation, but did not differ regarding lesion size. In logistic regression analysis, only

baseline NfL levels predicted cavitation (p=0.017). In this prospective study using predefined

high-resolution MRI protocols, the majority of RSSIs evolved into lacunes during the first

three months poststroke with not much change thereafter. Serum NfL seems to be a promising

biomarker for more advanced subsequent tissue destruction in RSSI.

Key Words: cavitation, diffusion-weighted MRI, lacunar infarction, neurofilament, recent

small subcortical infarction

2

Introduction

The presumed evolution of recent small subcortical infarcts (RSSIs) into lacunes, i.e. small

apparent CSF-containing holes, has led to the term of lacunar infarction based on careful

histopathologic examinations.1 However, with the possibility of neuroimaging follow-up, it

became apparent that RSSIs do not always evolve into lacunes but can remain as mainly non-

cavitated white matter hyperintensities (WMH) or even disappear after several weeks or

months. This notion is important for several reasons, including the impact on possible

differences in tissue vulnerability and repair, and a possible underestimation of brain tissue

loss from previous RSSIs on cross-sectional imaging. Studies to date report a wide range of

RSSIs that actually undergo cavitation including lacune formation.2–6 Major factors

contributing to this large variability appear to be the use of different neuroimaging techniques,

differences in follow-up time, and differences in definition of ‘lacune formation’. Thus, the

timeline of RSSI cavitation or other fate, and stability thereafter, remain unclear.

In addition, it would help to understand the causes and mechanisms that determine the

morphologic evolution of an RSSI. Diffusion weighted imaging (DWI) appears promising to

predict lesion evolution, as it allows detection of ischemic tissue changes in a quantitative

manner and more pronounced DWI abnormalities might be indicative of a greater likelihood

of complete tissue destruction.7,8 Furthermore, the neurofilament light chain protein (NfL) - a

marker of neuroaxonal damage - could be an interesting biomarker in this field as serum NfL

levels correlated with the occurrence of new clinically silent cerebral small disease related

MRI changes in a longitudinal study of patients with RSSIs.9

To add information on these points, we have analyzed the data from a prospective study of

patients with RSSI related to cerebral small vessel disease (CSVD). Among other

investigations, these individuals underwent baseline and pre-defined MRI follow-up

3

investigations at three and 15 months after stroke, which allowed assessment of the medium-

and long-term morphologic fate of the RSSIs in a standardized fashion. We also hypothesized

that larger RSSI size, lower quantitative apparent diffusion coefficient (ADC) values in the

RSSI, and higher serum NfL levels at baseline would be associated with a higher likelihood

for subsequent lacunar cavitation.

4

Material and Methods

Patients

Starting in May 2012, we invited all consecutive acute stroke patients who were admitted to

our primary and tertiary care university hospital to participate in a prospective lacunar stroke

study if they were ≤75 years of age and diffusion MRI had shown a single RSSI that was

compatible with the clinical stroke symptoms and located in the supply area of a small

perforating brain artery. While 20mm is generally regarded as the maximal axial RSSI

diameter, we expanded the upper size limit to 25mm for this study as some RSSI have larger

diameters in the very acute phase. Evidence for other acute brain infarcts, preexisting

disability (modified Rankin Scale score >1), and contraindications for repeated MRI were

defined as further exclusion criteria.9

Patients underwent a thorough neurological examination and cerebrovascular workup

including ECG, 24h-ECG, echocardiography and duplex sonography of brain-supplying

vessels, plus blood sampling and brain MRI at baseline and at three and 15 months after

stroke as previously reported.9 All patients received acute stroke treatment and secondary

prevention according to standard clinical practice.

Brain MRI acquisition and analysis

At baseline, all study patients underwent brain MRI at 1.5 Tesla (Siemens MAGNETOM

Espree, Siemens Healthcare, Erlangen; Germany) according to a standard protocol9 for the

clinical workup of patients with suspected cerebrovascular events. This included an axial T2-

weighted fast spin echo sequence (0.5x0.5x5mm), an axial fluid-attenuated inversion recovery

(FLAIR) sequence (0.4x0.4x5mm), a sagittal T1-weighted spin echo sequence

(0.6x0.6x5mm), a gradient echo T2* weighted sequence (0.4x0.4x5mm), an axial diffusion-

weighted (DWI) single-shot echo planar imaging sequence (1.2x1.2x5mm) with apparent

5

diffusion coefficient (ADC) maps and a 3D time of flight (TOF) angiography. All axial scans

had a slice thickness of 5mm with 0.5mm gap.

At both follow-ups, brain MRI was performed on a 3 Tesla TimTrio or Prisma scanner

(Siemens Healthcare, Erlangen, Germany). The protocol included high-resolution structural

3D images by means of a T1-weighted MPRAGE sequence with 1mm isotropic resolution (no

gap), T2-weighted (0.8x0.8x3mm, no gap), and FLAIR sequences (0.8x0.8x3mm, no gap),

gradient echo T2* weighted scans (0.8x0.8x2 mm, no gap), and DWI (2x2x2.5mm, no gap)

and intracranial 3D-TOF images.

All MRI scans were reviewed by a neuroradiological expert (CE) according to the ’STandards

for ReportIng Vascular changes on nEuroimaging’ (STRIVE).10 Besides assessing the

location and size of the RSSI, all MRI scans were rated for WMH severity11, lacunes of

presumably vascular origin, old cortical infarcts, microbleeds, old parenchymal hemorrhages,

large vessels pathologies or other concomitant intracranial lesions12 blinded to the clinical

findings. On follow-up MRI scans (FLAIR, T2- and T1-weighted sequences) the investigator

specifically described the evolution of the baseline RSSI (i.e. cavitation to a lacune, WMH, or

absent / "almost vanished" lesion).10 A lacune was defined as a cavitated lesion with a

consistent CSF signal intensity on T1, T2 and FLAIR weighted MRI scans. Hyperintense

lesions in the subcortical grey matter or brainstem were also categorized as WMH if no

lacunar cavitation was noted. Lesions that showed very subtle tissue changes and would not

have been noted without co-registration were defined as "almost vanished". Maximal axial

lesion diameters were computed on DWI sequences for the RSSI and on T1-weighted images

for the respective lacunar lesions. All ratings and measurements were performed on a

dedicated radiological PACS workstation including co-registration.

For quantification of the DWI changes of the RSSI on the baseline MRI we used a semi-

manual region of interest (ROI) approach to outline the RSSI and calculate its mean ADC.

For comparison we identified a matching ROI located in the contralateral hemisphere and 6

same anatomical structure to calculate a corresponding “control” mean ADC. When

identifying this control ROI, care was taken to avoid areas of coexisting morphologic damage

including WMH.

Serum neurofilament assessment

For assessing neurofilament levels in serum, eight millilitres of peripheral blood was taken by

venipuncture within 11 days after the index RSSI (median time from stroke symptom onset to

blood sampling: 4 days, range: 1-11 days) and at the 3 and 15 months follow-up. Serum was

then immediately stored at −80 °C according to international consensus guidelines.13 A Single

Molecule Array (Simoa) assay served to measure serum NfL as described in more detail

elsewhere.14

Statistical Analysis

Demographic and clinical data, ADC values, and NfL levels were analyzed with the Statistical

Package of Social Science (IBM SPSS Statistics 23). The level of significance was set at 0.05.

The Kolmogorov-Smirnov test assessed normality of data distribution. Groups were compared

by the Chi-Square test (for nominal data), the Mann–Whitney U test (for non-normally

distributed variables) or unpaired t-test (for continuous, normally distributed variables).

Correlation analysis was performed using the Spearman or Pearson correlation. A hierarchical

logistic regression analysis was performed to identify baseline markers associated with

cavitation, controlling for age, sex, the interval from stroke symptom onset to MRI

assessment at baseline and time from stroke symptom onset to baseline blood sampling in the

first step and including baseline markers (RSSI diameter on DWI, baseline ADC values and

NfL levels) in the second step.

7

To control for possible effects of small vessel disease, we performed additional analysis

adding deep and periventricular WMH in the second step of the model.

Research ethics and patient consent

The study was approved by the ethics committee of the Medical University of Graz (ID: 24-

260 ex 11/12), and conducted in accordance with the Declaration of Helsinki. All patients

gave written informed consent. Other results from this study cohort on cross-sectional and

longitudinal NfL levels, and their association with the occurrence of new cerebral small vessel

disease related lesions on follow-up MRI have recently been reported elsewhere.9

8

Results

From May 2012 to August 2016, 95 consecutive stroke patients with an MRI DWI-confirmed

RSSI agreed to participate in our study. Blood serum to assess NfL levels was unavailable in

16 patients and one patient had no follow-up MRI (neither at three nor at 15 months) due to

claustrophobia, leading to a final study cohort of 78 patients. The baseline characteristics of

these patients are provided in Table 1. Four patients had concomitant atrial fibrillation, and

one patient had a proximal high-grade vessel stenosis, but these findings did not appear to be

causally related to the RSSI.

Information on intravenous thrombolysis and prestroke vascular medication is given in table

1. Medical treatment for secondary stroke prevention included antiplatelets (n=69), oral

anticoagulation (n=9), statins (n=57), antihypertensives (n=55) and antidiabetics (n=8), and

did not differ between patients with versus without cavitation.

MRI lesion evolution

Out of 74 patients with available MRI at the three-months follow-up, almost four out of five

RSSIs patients showed cavitation at three months (N=61, 78%) and this was confirmed in all

who had a follow up MRI at 15 months (Figure 1). Eleven of 13 patients with another

evolution of the RSSI also retained their morphologic characteristics at 15 months. In only

one patient the RSSI appeared as a WMH at three months but as a lacune at the 15 months

follow-up MRI scan. In another patient, the RSSI evolved to WMH at three months, which

was invisible at the 15 months follow-up.

In patients with RSSI evolving to a lacune, the median axial RSSI diameter was 11.75mm

(IQR 6.57) at baseline. After three months, the median axial diameter of the cavitated area

was 8mm (IQR 7) and further decreased to 7mm (IQR 6.8) at the 15 months follow-up

9

(p=0.008 for the comparison between three and 15 months follow-up). Figure 2 illustrates

different longitudinal RSSI evolutions in three patients, with lacune shrinkage from the three

to the 15 months follow-up in patient A.

Differences in patients with versus without cavitation

The ADC values of the RSSIs at baseline were lower than the contralateral region ADC

values (523.94 +/- 104.14 x 10-6mm²/s versus 759.57 +/- 90.64; x 10-6mm²/s; p<0.0001).

Overall, the larger the axial DWI diameter of the RSSI was, the lower the quantitative ADC

values of the lesion (r=-0.268, p<0.05) and the higher the NfL levels at baseline (r=0.256,

p<0.05) were.

Patients with cavitation of their RSSI at three months had significantly lower ADC values and

significantly higher NfL levels at baseline than those without cavitation at three months

(Table 2, Figure 3).

Patients with subsequent cavitation also had elevated NfL levels at follow-up. However, this

finding disappeared over time when excluding patients with newly evolved yet clinically

silent ischemic lesions within the follow-up period (Table 2).

Higher NfL levels at baseline correlated with a larger RSSI diameter (rs =0.26, p=0.024),

higher scores of deep WMH (rs =0.35, p=0.002) and periventricular WMH (rs =0.23,

p=0.049). No significant correlations between NfL levels at baseline and NIHSS at baseline

(rs =-0.04, p=0.763) or lesional ADC values at baseline were observed (rs =-0.14, p=0.243).

Further information on the relationship between NfL levels at baseline and the various

characteristics of the study cohort can be found in the online supplemental (Table S1 and

Table S2).

10

Prediction of cavitation

In a model including RSSI diameter on DWI, baseline ADC values and NfL levels at baseline,

and controlling for the interval from stroke symptom onset to MRI assessment at baseline and

time from stroke symptom onset to baseline blood sampling, only NfL levels at baseline

predicted RSSI cavitation at three months (p=0.017; OR 1.03).

Similar results were observed if we additionally controlled for age and sex, where again only

NfL levels at baseline predicted lesion cavitation (p=0.035; OR 1.03).

We further added deep and periventricular WMH scores to the model and found stable that

only NfL levels (pg/ml) at baseline predicted lesion cavitation (p=0.043; OR 1.03). Results

also did not change when excluding three patients who had a maximal RSSI diameter of

>20mm (data not shown).

11

Discussion

This prospective cohort study adds some new aspects regarding the frequency, time course,

evolution and predictors of cavitation of CSVD-related RSSI. First, we found that almost 80%

of RSSIs evolved into cavitated lacunes at a prespecified follow-up MRI scan at three months.

Only minimal subsequent changes regarding cavitation status were noted at a second follow-

up scan one year thereafter but the cavitated areas showed significant shrinkage over time.

Second and most notably, we identified lower quantitative lesional ADC values and higher

serum NfL levels at baseline as two potential predictors of RSSI tissue loss.

Although the clinical consequences have still to be determined (e.g. impact on cognitive

status), more refined knowledge on the lesion evolution of RSSIs and specifically tissue loss

including lacune formation, has important implications on the research of the epidemiology

and pathophysiology of CSVD.

We are aware of five previous neuroimaging studies2–6 that have investigated the lesion

evolution of acute “lacunar” infarcts and reported a large heterogeneity with apparent

cavitation including lacune formation occurring in 48-94% of patients. These substantial

differences can be explained by (1) different definitions of ‘cavitation’ including whether

adhering to a strict definition of true lacune formation or any degree of cavitation, (2)

diverging neuroimaging methods used, regarding modality (CT which is often too unspecific

to detect CSVD-related lesions versus MRI), (3) applied MRI sequences (e.g. incorporation of

DWI in the primary diagnosis or not), or (4) possibly MRI field strengths (1.5 or 3 Tesla).

Another important aspect is the time interval when follow-up neuroimaging is performed after

acute infarction, which was considerably different between and within previous studies.

Generally, cavitation was associated with increasing time to follow-up imaging,2,5

highlighting the importance of a pre-defined fixed follow-up assessment to systematically

explore tissue changes. In some patients cavitation has already begun in the first month after 12

RSSIs.3,5 In a small multicenter study investigating two independent cohorts, cavitation was

more common at 90 days (94%) as compared to 30 days (52%), suggesting that cavity

formation might not be complete within 30 days. It is, however, still unclear whether the

proportion of cavitation increases over longer time periods. Therefore, we designed this

prospective study assessing morphological changes of RSSI at two time points (after three

months and one further year thereafter). Interestingly only minor changes occurred in

cavitation status after three months and only one patient showed lacunar cavitation of a white

matter lesion between follow-up 1 and 2. In line with previous studies, RSSIs showed lesion

shrinkage during cavity formation (median axial diameter decreased from 11.5 mm to 8

mm).4–6 Notably, we observed a further decrease of the cavity diameter to 7 mm at the 15

months follow-up. While we currently cannot provide a clear explanation for this process of

lacune shrinkage a similar observation has been already made by others.5 This finding

suggests some morphologic dynamics even at the chronic lacunar stage, which has to be

considered regarding lesion cut-off definitions and deserves further exploration.

So far, knowledge about baseline clinical or morphological correlates associated with lesion

evolution is scarce. To better understand the causes and mechanisms that determine the

morphologic evolution of an RSSI, we investigated possible differences between patients with

versus without cavitation. In line with most previous studies,3–5 we did not identify any

demographic, clinical or imaging variables to predict cavitation. In contrast to this, a previous

study found that deep brain atrophy was associated with cavitation, while hypertension or

diabetes were associated with not cavitating RSSI development.2

In a rodent stroke model15 longer duration of ischemia was associated with cavity formation

indicating that the degree of ischemia may be explanatory for that. It is therefore interesting

that the NIHSS and mRS at baseline as well as at both follow-ups, and also RSSI size were

not different between patients with or without cavitation. However, patients who developed

13

cavitation had lower quantitative lesional ADC values indicative of more severe diffusion

abnormality and local ischemic tissue destruction.7,8,16–19 In line with this finding, ADC values

have been shown to be better predictors of stroke outcome compared to lesion volume

alone.20,21

Therefore it is intriguing that in our multivariate analysis the ADC value was no longer

significantly associated with cavity formation. This is consistent with another study, which

has reported that the amount of ADC reduction in the acute stroke phase could not predict

subsequent tissue outcome.22 It could also be speculated, however, that in our specific setting

NfL levels are reflective of ADC values and outperform this morphologic information as the

small size of the acute lesions may preclude robust measurements, especially in the presence

of a possibly inhomogeneous ADC pattern defining the subsequent area of complete tissue

necrosis.

The only independent predictor for cavitation was the serum NfL level at baseline. Upon

neuroaxonal damage, NfL is released into the extracellular space and subsequently the

cerebrospinal fluid and blood.23 With the recent advent of robust serum assays, serum NfL

might also act as a promising biomarker for stroke, where lumbar puncture is usually not

performed. We previously observed that serum NfL was elevated in RSSI patients over the

level of healthy age-matched controls.9 Thus, serum NfL seems to represent a sensitive

marker for acute ischemic cerebrovascular injury, even in patients with small subcortical

infarcts. We here observed that baseline NfL was positively associated with subsequent

cavitation. This finding might be explained by more extensive initial tissue damage releasing

higher amounts of NfL to the blood, more active diffuse small vessel disease or by ongoing

inflammatory/immunological processes promoting subsequent tissue rarefaction.24 NfL levels

at 3 and 15 months after stroke were also higher in patients with cavitated RSSIs. However,

14

after excluding patients with new clinically silent ischemic lesions at follow-up this difference

was no longer significant.

The main strength of our study is that it was specially designed to determine longitudinal

morphological changes after a symptomatic RSSI by using two prespecified MRI follow-up

assessments at fixed intervals with a uniform 3 Tesla protocol and the incorporation of high-

resolution T1 sequences. Moreau and colleagues3 have highlighted that a comprehensive MRI

acquisition with the inclusion of thin T1 scans clearly increases the sensitivity of detecting

lacunes (and avoiding misclassification as other lesions) over the single analysis of T2 or

FLAIR scans - an approach we also applied here. Moreover, the analysis of these T1 scans

with regard to maximal axial lacunar diameters should have largely excluded imprecise

measurements as well as lesion overestimation compared to T2 or FLAIR-weighted images.

Besides the strengths of our work, our study has also some important limitations. The sample

size was moderate and consistent follow-up assessments were missing in a few patients.

Nevertheless, this is the largest cohort with uniform longitudinal MRI data at 3 Tesla thus far.

There has also been some intended selection of patients as we only included RSSI patients

<75 years of age without preexisting disability (mRS<1) to increase the chance of regular

follow-up assessments and to reduce the extent of coexisting morphologic damage. Therefore,

our results might not be applicable to the entire cohort of RSSI/lacunar stroke patients,

however the mean age of our patients was comparable with previously published research on

this topic. ADC was calculated from routine cerebrovascular MRI obtained at baseline, which

was performed on a 1.5T instead of a 3 T MRI. Given the large voxel size of our clinical

routine scans, we abstained from lesion volumetrics and rather used the axial diameter of the

RSSI and WMH grades of a well-established score. Reassuringly, both measures have been

shown to strongly correlate with respective lesion volumes.25,26 Furthermore, the interval of

both baseline MRI and blood drawing to determine serum NfL levels from stroke onset varied 15

between patients. It is known that both ADC and NfL levels change with the interval from the

acute stroke and this may have influenced the absolute values observed.9,27 Therefore, we

controlled for both intervals in our multivariate regression analysis. However, neither the

interval from acute stroke to MRI nor to blood sampling were significantly different between

patients with cavitating and non-cavitating RSSI at three months. This is important as we have

observed an increase in NfL levels with increasing distance from the RSSI.9 Therefore, future

studies should aim at the acute MRI to be performed most close to the onset of symptoms and

at multiple blood samplings over the first days after stroke to account for these issues and to

define the time interval when NfL levels might be most predictive for the morphologic fate of

the RSSI. Due to these limitations the ultimate clinical value of NfL still has to be determined.

Finally, potential effects of certain or intensified medication on cavitation status will need to

be considered in future studies.

16

Acknowledgement and funding sources: Daniela Pinter receives funding from the Austrian

Science Fund: T690-B23.

Author contribution: TG, CE and FF conceived the study and designed the research question.

DP, TG, CE, MKn, SF, AP, SE acquired the data. DP, TG, TSH, LP, CB, GB, JK and MKh

analyzed the data, which was discussed with FF, CE, SR, JMW and MKh. DP and TG wrote

the first draft of the manuscript and CE, JMW and FF co-drafted the final version. DP

performed the statistical analysis. FF supervised the study. All authors critically revised the

manuscript and have read and approved the final manuscript and agreed to be accountable for

all aspects of the work.

Disclosures: None.

17

References

1. Fisher CM. Lacunes: Small, Deep Cerebral Infarcts. Neurology. 1965 Aug;15:774-84.

2. Potter GM, Doubal FN, Jackson CA, et al. Counting cavitating lacunes underestimates

the burden of lacunar infarction. Stroke. 2010;41:267–272.

3. Moreau F, Patel S, Lauzon ML, et al. Cavitation after acute symptomatic lacunar stroke

depends on time, location, and MRI sequence. Stroke. 2012;43:1837–1842.

4. Koch S, McClendon MS, Bhatia R. Imaging evolution of acute lacunar infarction:

Leukoariosis or lacune? Neurology. 2011;77:1091–1095.

5. Loos CMJ, Staals J, Wardlaw JM, et al. Cavitation of deep lacunar infarcts in patients

with first-ever lacunar stroke: A 2-year follow-up study with MR. Stroke.

2012;43:2245–2247.

6. Lee KJ, Jung H, Oh YS, et al. The Fate of Acute Lacunar Lesions in Terms of Shape

and Size. J Stroke Cerebrovasc Dis. 2017 Jun;26(6):1254-1257.

7. Desmond PM, Lovell AC, Rawlinson AA, et al. The value of apparent diffusion

coefficient maps in early cerebral ischemia. AJNR Am J Neuroradiol. 2001

Aug;22(7):1260-7.

8. Eastwood JD, Engelter ST, MacFall JF, et al. Quantitative assessment of the time

course of infarct signal intensity on diffusion-weighted images. AJNR Am J

Neuroradiol. 2003 Apr;24(4):680-7.

9. Gattringer T, Pinter D, Enzinger C, et al. Serum neurofilament light is sensitive to

active cerebral small vessel disease. Neurology. 2017 Nov 14;89(20):2108-2114.

10. Wardlaw JM, Smith EE, Biessels GJ, et al. Neuroimaging standards for research into

18

small vessel disease and its contribution to ageing and neurodegeneration. Lancet

Neurol. 2013 Aug;12(8):822-38.

11. Fazekas F, Chawluk JB, Alavi A. MR signal abnormalities at 1.5 T in Alzheimer’s

dementia and normal aging. AJR Am J Roentgenol. 1987 Aug;149(2):351-6.

12. Fazekas F, Enzinger C, Schmidt R, et al. MRI in acute cerebral ischemia of the young:

the Stroke in Young Fabry Patients (sifap1) Study. Neurology. 2013;81:1914–1921.

13. Teunissen CE, Petzold A, Bennett JL, et al. A consensus protocol for the

standardization of cerebrospinal fluid collection and biobanking. Neurology.

2009;73:1914–1922.

14. Disanto G, Barro C, Benkert P, et al. Serum Neurofilament light: A biomarker of

neuronal damage in multiple sclerosis. Ann. Neurol. 2017;81:857–870.

15. Garcia JH, Liu KF, Ye ZR, et al. Incomplete infarct and delayed neuronal death after

transient middle cerebral artery occlusion. Stroke. 1997;28:2303–2310.

16. Schlaug G, Siewert B, Benfield A, et al. Time course of the apparent diffusion

coefficient (ADC) abnormality in human stroke. Neurology. 1997;49:113–119.

17. Ackerman J, Neil J. The use of MR-detectable reporter molecules and ions evaluate

diffusion in normal and ischemic brain. NMR Biomed. 2010;23:725–733.

18. Roberts TPL, Rowley HA. Diffusion weighted magnetic resonance imaging in stroke.

Eur J Radiol. 2003 Mar;45(3):185-94.

19. Qian Q, Huang HT, Xu L, et al. Prediction of Infarct Lesion Volumes by Processing

Magnetic Resonance Apparent Diffusion Coefficient Maps in Patients with Acute

Ischemic Stroke. J Stroke Cerebrovasc Dis. 2016;25:2821–2827.

19

20. Rosso C, Colliot O, Pires C, et al. Early ADC changes in motor structures predict

outcome of acute stroke better than lesion volume. J Neuroradiol. 2011 May;38(2):105-

12.

21. Henriques IL, Gutiérrez-Fernández M, Rodríguez-Frutos B, et al. Intralesional patterns

of MRI ADC maps predict outcome in experimental stroke. Cerebrovasc Dis.

2015;39:293–301.

22. An H, Ford AL, Vo K, et al. Signal evolution and infarction risk for apparent diffusion

coefficient lesions in acute ischemic stroke are both time-and perfusion-dependent.

Stroke. 2011;42:1276–1281.

23. Teunissen CE, Khalil M. Neurofilaments as biomarkers in multiple sclerosis. Mult

Scler. 2012 May;18(5):552-6.

24. Traenka C, Disanto G, Seiffge DJ, et al. Serum Neurofilament Light Chain Levels Are

Associated with Clinical Characteristics and Outcome in Patients with Cervical Artery

Dissection. Cerebrovasc Dis. 2015;40:222–227.

25. Gouw A, Van der Flier WM, van Straaten ECW, et al. Simple versus complex

assessment of white matter hyperintensities in relation to physical performance and

cognition: the LADIS study. J Neurol. 2006 Sep;253(9):1189-96

26. Gattringer T, Eppinger S, Pinter D, et al. Morphological MRI characteristics of recent

small subcortical infarcts. Int J Stroke. 2015 Oct;10(7):1037-43.

27. Fiebach JB, Jansen O, Schellinger PD, et al. Serial analysis of the apparent diffusion

coefficient time course in human stroke. Neuroradiology. 2002;44:294–298.

20

Figure 1. Lesion evolution of the RSSI at the three and 15 months follow up.

WMH= white matter hyperintensity.

Figure 2: Three examples of different longitudinal RSSI evolution (patients A-C).

Patient A. Evolution of the RSSI to a lacune at the three months follow-up. Note the subsequent shrinkage of the

lacunar diameter at the 15 months (12mm to 11mm). Patient B. RSSI development to a WMH at follow-up. C.

Example of an almost vanishing RSSI over time.

Figure 3: Mean ADC values (in x 10-6mm²/s) of the RSSIs (left panel, p=0.027) and NfL

levels (in pg/ml) at baseline (right panel, p=0.004) compared between patients with versus

without cavitation at three months follow-up.

21

Table 1. Demographic, clinical and MRI characteristics of the total cohort of RSSI patients at

baseline (n=78), and comparisons regarding these variables in patients with available follow-

up MRI at three months according to cavitation status (n=74).

Total

cohort

(n=78)

Follow-Up

cohort

(n=74)

Cavitation

(n=61)

No

Cavitation

(n=13)

p-value*

Age at baseline (years, SD) 61.0 (11.0) 60.7 (11.1) 61.5 (10.2) 57.2 (14.8) 0.331

Sex, n (% male) 53 (68%) 51 (69%) 44 (72%) 7 (54%) 0.206

Clinical characteristics, n (%)

NIHSS (median, IQR) 2.0 (3.0) 2.0 (3.0) 2.0 (3.0) 2.0 (3.0) 0.885

Clinical symptoms

Hemiparesis 49 (63%) 45 (61%) 37 (61%) 8 (62%) 0.605

Hemisensory symptoms 29 (37%) 28 (38%) 23 (38%) 5 (39%) 0.597

Dysarthria 26 (33%) 25 (34%) 20 (33%) 5 (39%) 0.752

Other symptoms 16 (21%) 16 (22%) 12 (20%) 4 (31%) 0.293

Atrial fibrillation 4 (5%) 3 (4%) 3 (5%) 0 (0%) 0.998

Hypertension 60 (77%) 57 (77%) 48 (79%) 9 (70%) 0.480

Hyperlipidemia 62 (80%) 59 (80%) 49 (80%) 10 (77%) 0.719

Diabetes 13 (17%) 13 (18%) 12 (20%) 1 (8%) 0.442

Smoking 34 (44%) 33 (45%) 27 (44%) 6 (46%) 0.999

History of prior stroke 3 (4%) 3 (4%) 2 (3%) 1 (8%) 0.445

Previous symptomatic RSSI 2 (3%) 2 (3%) 2 (3%) 0

RSSI location, n (%)**

Supratentorial white matter 39 (50%) 35 (47%) 31 (51%) 4 (31%) 0.231

Thalamus 17 (22%) 17 (23%) 13 (22%) 4 (31%) 0.480

Basal ganglia 14 (18%) 13 (18%) 11 (18%) 2 (15%) 0.988

22

Brainstem 11 (14%) 9 (12%) 6 (10%) 3 (23%) 0.189

MRI findings

Deep WMH score, (IQR) 1.0 (2.0) 1.0 (2.0) 1.0 (2.0) 1.0 (1.0) 0.251

Periventricular WMH score,

(IQR)

1.0 (3.0) 1.0 (3.0) 1.0 (3.0) 1.0 (2.0) 0.245

Old lacunar infarcts (≥1) 29 (37%) 22 (30%) 19 (31%) 3 (23%) 0.574

Old cortical infarcts (≥1) 7 (9%) 7 (9%) 5 (8%) 2 (15%) 0.458

Microbleeds (≥1) 13 (17%) 11 (15%) 11 (2%) 0 0.194

NBV (cm³), (IQR) 1456.0

(135.4)

1459.1

(146.7)

1458.2 (152.6) 1460.1

(132.5)

0.722

WMV (cm³), IQR 749.9

(67.1)

752.4 (78.5) 753.0 (80.1) 745.1 (73.7) 0.798

Medication

Previous antithrombotic therapy 18 (23%) 18 (24%) 14 (23%) 4 (31%) 0.722

Previous statins 7 (9%) 7 (9%) 7 (11%) 0 0.341

Intravenous thrombolysis 6 (8%) 5 (7%) 5 (8%) 0 0.579

* p-values refer to the comparison between patients with versus without cavitation at the three months follow-up.

** 3 RSSIs were located in white matter and basal ganglia; NBV = normalized brain volume, WMV=

normalized white matter volume, SD = standard deviation, RSSI = recent small subcortical infarct, NIHSS =

National Institutes of Health Stroke Scale, WMH = white matter hyperintensities, IQR = interquartile range

Percentages are presented column-wise.

23

Table 2. Comparison of MRI findings, NfL levels and clinical data between patients with

versus without cavitation at the three months follow-up.

Cavitation

(n=61)

No Cavitation

(n=13)

p-value

Axial RSSI DWI diameter, mm (IQR)¶ 12.0 (5.5) 10.0 (5.5) 0.530

Days from symptom onset to baseline

MRI, median (IQR)

2.0 (2.0) 2.0 (2.0) 0.651

Mean quantitative lesional ADC* 514.6 (99.8) 586.7 (110.0) 0.027

Mean contralesional ADC* 759.3 (89.6) 730.2 (55.2) 0.283

NfL levels+ all patients

at baseline 85.6 (126.7) 50.1 (51.1) 0.004

at 3 months 97.9 (88.2) (n=56) 47.9 (121.4) (n=11) 0.104

at 15 months 46.0 (31.7) (n=40) 28.5 (40.6) (n=11) 0.042

NfL levels+ in patients without new

ischemic lesions at follow-up

at 3 months 93.1 (78.1) (n=50) 46.5 (117.9) (n=10) 0.057

at 15 months 43.9 (31.6) (n=35) 33.0 (42.5) (n=10) 0.111

Days from symptom onset to baseline

blood sampling, median (IQR)

4.0 (3.0) 5.0 (3.0) 0.434

NIHSS at 3 months, median (IQR) 0.0 (1.0) 0.0 (1.0) 0.328

mRS at 3 months 1.0 (1.0) 1.0 (2.0) 0.920

NIHSS at 15 months 0.0 (1.0) 0.0 (1.0) 0.798

mRS at 15 months 0.0 (1.0) 1.0 (1.0) 0.596

NfL = neurofilament light chain protein, RSSI = recent small subcortical infarct, DWI = diffusion weighted

imaging, ADC = apparent diffusion coefficient, NIHSS = National Institutes of Health Stroke Scale, mRS =

modified Rankin Scale, IQR = interquartile range

24

¶ Three patients had a maximal axial RSSI diameter >20 mm.

*ADC values in x 10-6mm²/s

+NfL levels in pg/ml median (IQR)

25