Perfusion-weighted imaging and dynamic 4D angiograms for the estimation of collateral blood flow in lacunar infarction

Abstract

Although lacunar infarction accounts for approximately 25% of ischemic strokes, collateral blood flow through anastomoses is not well evaluated in lacunar infarction. In 111 lacunar infarction patients, we analyzed diffusion-weighted images, perfusion-weighted images, and blood flow on dynamic four-dimensional angiograms generated by use of Signal Processing In NMR-Software. Blood flow was classified as absent (type 1), from periphery to center (type 2), from center to periphery (type 3), and combination of type 2 and 3 (type 4). On diffusion-weighted images, lacunar infarction was found in the basal ganglia (11.7%), internal capsule (24.3%), corona radiata (30.6%), thalamus (24.3%), and brainstem (9.0%). In 58 (52.2%) patients, perfusion-weighted image showed a circumscribed hypoperfusion, in one (0.9%) a circumscribed hyperperfusion, whereas the remainder was normal. In 36 (62.1%) patients, a larger perfusion deficit (>7 mm) was observed. In these, blood flow was classified type 1 in four (11.1%), 2 in 17 (47.2%), 3 in 9 (25.0%), and 4 in six (16.7%) patients. Patients with lacunar infarction in the posterior circulation more often demonstrated blood flow type 2 and less often type 3 (p = 0.01). Detailed examination and graduation of blood flow in lacunar infarction by use of dynamic four-dimensional angiograms is feasible and may serve for a better characterization of this stroke subtype.

Keywords

Anastomoses collateral blood flow lacunar stroke

Introduction

Lacunar infarction (LI) accounts for up to 25% of acute ischemic strokes¹ and is defined as a small subcortical ischemic lesion located in the basal ganglia, thalamus, internal capsule, corona radiata, or brainstem caused by occlusion of a single perforating artery usually due to lipohyalinosis or microatheroma.^2,3 The maximal diameter is classically defined as <15 mm. However, there is a large variety of lacunar lesion size definitions⁴ and more recently, a lesion size <20 mm has been introduced in clinical studies.⁵ The clinical presentation usually comprises five typical so-called lacunar syndromes although about 20 different lacunar syndromes have been reported.³

In patients with acute ischemic stroke, slow collateral blood flow through anastomoses is regarded as an essential condition enabling the survival of potentially salvageable brain tissue but little is known about collateral blood flow between the small perforating arteries of the brain. The existence of anastomoses between the perforating arteries has been examined in meticulous pathoanatomical studies.^6–15 Even a higher frequency of anastomoses between the perforating arteries in the posterior circulation has been reported.¹² Comparable conventional angiography,¹⁶ CT,¹⁷ or MRI studies¹⁸ are sparse and limited to the demonstration of larger perforating branches because of the relatively low resolution. While conventional angiography is the gold standard to characterize the extent of collateral blood flow in large vessel occlusion,¹⁹ its ability to demonstrate corresponding changes in acute LI is very limited due to the small diameter of the perforating arteries and their branches. Furthermore, conventional angiography is usually reserved for patients with large vessel occlusion undergoing endovascular treatment. In the last decade, a large variety of CT and MRI-based approaches for the assessment of collateral blood flow has been investigated.²⁰ Among these, dynamic 4D angiograms derived from dynamic susceptibility contrast (DSC) perfusion raw images have been introduced recently.²¹ Meanwhile, they have been used not only in middle cerebral artery occlusion but also in posterior cerebral artery and basilar artery occlusion.²² Another study demonstrated a good correlation between collateralization grades based on dynamic 4D angiograms in comparison to conventional angiography findings in middle cerebral artery occlusion.²³

In this case series of patients with acute LI admitted to our hospital, we aimed to investigate (1) the frequency and further characteristics of a perfusion deficit in LI, and (2) the ability of dynamic 4D angiograms generated from perfusion-weighted raw images to demonstrate collateral blood flow in LI, (3) to classify patterns of collateral blood flow in LI, (4) to correlate the observed patterns of collateral blood flow with established perfusion maps such as cerebral blood flow (CBF), and cerebral blood volume (CBV), and (5) to compare patterns of collateral blood flow in LI in the anterior and posterior circulation.

Material and methods

Patients

From a prospectively maintained MRI report database (January 2005 to February 2015), we identified 140 patients with acute subcortical infarction who underwent a standard stroke MRI protocol including perfusion-weighted image (PWI) within 24 h after onset of symptoms. In 117 (83.6%) ischemic stroke patients, no other potential etiology of ischemic stroke than LI could be detected. The remaining patients were excluded from the analysis because of large artery atherosclerosis or cardioembolism coexisting as other possible cause of ischemic stroke. Further 6/117 (5.1%) patients were excluded from the analysis due to intravenous thrombolysis before MRI. The demographic details and clinical presentation were abstracted from the case records. Standardized clinical workup included assessment of cardiovascular risk factors, extra- and transcranial Doppler-/duplex sonography, 24 h electrocardiogram monitoring, and transthoracic or transesophageal echocardiography as well as laboratory tests according to stroke unit standard requirements in Germany.²⁴ The study was approved by the local institutional review board (Medizinische Ethikkommission II der Medizinischen Fakultät Mannheim). Patient consent was not required due to the retrospective nature of the study and the lack of patient interaction.

MRI studies

Magnetic resonance imaging was performed on a 1.5-T MR system (Magnetom Sonata or Avanto, Siemens Medical Systems, Erlangen, Germany) or a 3-T MR system (Magnetom Trio, Siemens Medical Systems, Erlangen, Germany). A standardized protocol was used in all patients including (1) transverse, coronal, and sagittal localizing sequences followed by transverse oblique contiguous images aligned with the inferior borders of the corpus callosum (applied on sequences 2 to 5); (2) T1-weighted images; (3) T2-weighted images; (4) diffusion-weighted images (DWI); (5) fluid-attenuated inversion recovery (FLAIR) images; (5) PWI following the first pass of contrast bolus through the brain; and (6) a 3D time-of-flight MR angiography. The contrast agent gadoteric acid (Dotarem, Guerbet, Aulnay-sous-Bois, France) was bolus injected by a power injector (Spectris MR injection system, Medrad, Volkach, Germany) with a dose of 0.1 mmol/kg of body weight at a rate of 4 ml/s. Parameters of DWI and PWI are displayed in Table 1.

Table 1.

Sequence parameters of DWI and PWI at the department’s MRI scanners.

Sequence	Parameters	MRI scanner
Sequence	Parameters	1.5-T Siemens Sonata	1.5-T Siemens Avanto	3-T Siemens Trio
DWI	FOV	230 × 230	230 × 230	230 × 230
	Matrix	128 × 128	192 × 192	192 × 192
	Number of slices	24	24	24
	ST	5	5	5
	TR	4400	4000	4000
	TE	101	96	91
PWI	FOV	230 × 230	230 × 230	230 × 230
	Matrix	128 × 128	128 × 128	128 × 128
	Number of slices	12	19	24
	ST	6	5	5
	TR	1500	1430	1770
	TE	46	30	32

FOV: field of view (mm × mm); ST = slice thickness (mm); TE = echo time (ms); TR = repetition time (ms).

Postprocessing

Perfusion maps

The postprocessing of the perfusion-weighted raw images was performed by a specific software, Signal Processing In NMR (SPIN, The MRI Institute for Biomedical Research, Detroit, USA).²⁵ Deconvolution with singular value decomposition was used to create quantitative maps of mean transit time (MTT), CBF, and CBV. The position of the arterial input function was automatically determined by using the maximum concentration (Cmax), relative time to peak (TTP), and first moment MTT (fMTT). The concentration–time curve for arteries has short fMTT, short TTP, and high Cmax. Twenty voxels, which best fitted these properties, were selected. Then the concentration–time curves of these voxels were averaged, smoothed, and truncated to avoid the second pass of the tracer.

Dynamic 4D angiograms

Furthermore, perfusion-weighted raw images were used to create a dynamic angiographic representation of blood flow as described recently.²¹ For this purpose, the baseline prebolus image was subtracted from each frame of the raw perfusion data as in digital subtraction angiography by use of the subtract series function in SPIN.

MRI analysis

FLAIR images and dynamic 4D angiograms were analyzed by two independent raters (AF and BM, both with eight years of experience in neuroimaging) blinded to the other MRI sequences and clinical information. Cases with discrepancies were re-reviewed by both readers and discussed until a consensus was reached.

Infarction localization and size

Acute ischemic lesions in the basal ganglia, thalamus, internal capsule, corona radiata, or brainstem were classified as LI attributable to single perforating artery occlusion if the maximal diameter was 20 mm on DWI. Maximal ischemic lesion diameter on DWI as well as maximal diameter of the perfusion deficit on PWI was measured by use of OsiriX, a multidimensional image navigation and display software.²⁶ Furthermore, the area of the DWI lesion on the slice corresponding to the PWI slice used for generation of dynamic 4D angiograms was measured.

Perfusion maps

First PWI was analyzed visually in order to obtain information on hemodynamic alterations, calculated TTP images demonstrating the delay of the contrast agent arrival in the brain parenchyma were used for visual analysis. Time-to-peak maps were classified as showing a normal perfusion pattern or areas of hyper- or hypoperfusion. Furthermore, the perfusion maps were also quantitatively assessed by use of SPIN: a region of interest (ROI) covering the hypoperfused area was placed on the generated maps (TTP, MTT, CBF, CBV) and mirrored to the contralateral unaffected hemisphere. Finally, ratios between the physiological estimates (TTP, MTT, CBF, CBV) of the lesion and of the contralateral mirror ROI were determined. Due to the limited spatial resolution, the analysis of perfusion maps was limited to perfusion lesions with a diameter >7 mm. Furthermore, the area of the PWI lesion on the slice used for generation of dynamic 4D angiograms was measured.

Dynamic 4D angiograms

The presence of blood flow on dynamic 4D angiograms was assessed using a newly implemented blood flow classification for LI: type 1 (no blood flow visible), type 2 (blood flow from the periphery to the center), type 3 (blood flow from the center to the periphery), and type 4 (combination of type 2 and type 3). For a detailed graphic illustration, see Figure 1. Due to the limited spatial resolution, the analysis of dynamic 4D angiograms was limited to perfusion lesions with a diameter >7 mm.

Figure 1.

Schematic illustration of collateral blood flow pathways and corresponding collateral blood flow types on dynamic 4D angiography in lacunar infarction due to small perforating artery occlusion: (1) Anterograde collateral blood flow via an anastomosis between the main stems of two perforating arteries. (2) Retrograde collateral blood flow via small anastomoses at the precapillary level. Type 1: no blood flow visible. Type 2: blood flow from the periphery to the center. Type 3: blood flow from the center to the periphery. Type 4: combination of type 2 and type 3.

Statistical analysis

All statistical analyses were performed using Statistical Product and Service Solutions (SPSS) statistics for Windows (Release 17.0; SPSS, Chicago, IL, USA). Descriptive data were analyzed by use of Chi-square tests and the Mann–Whitney U Test as appropriate. Comparison between patients with LI in the anterior and posterior circulation was performed using Chi-square tests and the Mann–Whitney U Test as appropriate. Comparison of lesion size on DWI and PWI was performed using the Mann–Whitney U Test or the Wilcoxon Test as appropriate. Interrater reliability for the blood flow pattern in LI prior to consensus was assessed using the Cohen kappa coefficient (κ). Differences in TTP, MTT, CBF, and CBV ratios between the different blood flow types were analyzed by use of the Kruskal–Wallis test. All statistics was performed with a 0.05 level of significance.

Results

Baseline characteristics and clinical presentation

In the final analysis, 111 LI patients were included. The median age was 68 years (57–76); 62 (55.9%) patients were male, 49 (44.1%) patients were female. All patients demonstrated a clinical stroke syndrome. Fluctuating or progressive symptoms were observed in 22 (19.8%) patients. Most common clinical symptoms comprised dysarthria (45.9%) and hemiparesis (76.6%) followed by hemihypesthesia (28.8%) whereas other symptoms like dysphasia/aphasia (6.3%), altered vigilance (4.5%), oculomotor dysfunction (4.5%), ataxia (4.5%), gait disturbances (4.5%), and disorientation (2.7%) were observed only occasionally. Patients with LI in the posterior circulation more frequently demonstrated disorientation, dysphasia/aphasia, oculomotor dysfunction, and gait disturbances. Median time from onset to MRI was 328 (IQR 244–542) min; in a subset of 12 (10.8%) patients the exact onset of symptoms could not be determined. Intravenous thrombolysis was performed in eight (7.2%) patients. For details see Table 2.

Table 2.

Demographics, clinical symptoms, and MRI findings in patients with LI in the anterior (AC) and posterior circulation (PC).

n (%) unless noted	All, n = 111	AC, n = 74	PC, n = 37	p
Age, median (IQR)	68 (57–76)	65 (56–75)	71 (64–78.5)	0.03
Sex (male)	62 (55.9)	43 (58.1)	19 (51.4)	0.50
Clinical presentation
Fluctuations	22 (19.8)	16 (21.6)	6 (16.2)	0.50
Altered vigilance	5 (4.5)	0	5 (13.5)	0.001
Dysphasia/Aphasia	7 (6.3)	3 (4.1)	4 (10.8)	0.18
Dysarthria	51 (45.9)	36 (48.6)	15 (40.5)	0.42
Disorientation	3 (2.7)	0	3 (8.1)	0.01
Oculomotor dysfunction	5 (4.5)	0	5 (13.5)	0.001
Hemiparesis	85 (76.6)	60 (81.1)	25 (67.6)	0.11
Hemihypesthesia	32 (28.8)	20 (27.0)	12 (32.4)	0.55
Ataxia	5 (4.5)	3 (4.1)	2 (5.4)	0.75
Gait disturbances	5 (4.5)	0	5 (13.5)	0.001
MRI findings
DWI lesion size, mm, median (IQR)	10 (7.1–12.9)	11.1 (7.75–14.05)	8.4 (5.95–10.8)	0.002
	n = 58	n = 36	n = 22
PWI lesion size, mm, median (IQR)	9.0 (5.6–12)	8.05 (5.7–12.7)	9.4 (5.3–11.5)	0.92
	n = 36	n = 22	n = 14
rTTP, median (IQR)	1.12 (1.09–1.28)	1.12 (1.08–1.28)	1.17 (1.10–1.28)	0.79
rMTT, median (IQR)	1.19 (1.10–1.34)	1.19 (1.10–1.33)	1.19 (1.05–1.36)	0.69
rCBF, median (IQR)	0.69 (0.56–0.88))	0.73 (0.61–0.88)	0.63 (0.41–0.91)	0.39
rCBV, median (IQR)	0.80 (0.64–1.13)	0.80 (0.67–1.03)	0.82 (0.54–1.35)	0.87
Collateral blood flow				0.004
Type 1	4 (11.1)	4 (18.2)	0
Type 2	17 (47.2)	6 (27.3)	11 (78.6)
Type 3	9 (25.0)	9 (40.9)	0
Type 4	6 (16.7)	3(13.6)	3 (21.4)

DWI: diffusion-weighted imaging; PWI: perfusion-weighted imaging; rCBF: relative cerebral blood flow; rCBV: relative cerebral blood volume; rMTT: relative mean transit time; TTP: relative time to peak.

MRI analysis

Diffusion-weighted imaging

Diffusion-weighted imaging demonstrated an acute ischemic infarction in the basal ganglia in 13 (11.7%), internal capsule in 27 (24.3%), corona radiata in 34 (30.6%), thalamus in 27 (24.3%), and brainstem in 10 (9.0%) patients. Figure 2 shows exemplary cases for each location. Ischemic lesions were attributed to the anterior circulation in 74 (66.7%) patients and to the posterior circulation in 37 (33.3%) patients. The right side was affected in 51 (45.9%) patients, the left side in 57 (51.4%) patients while in three (2.7%) patients bilateral thalamic infarctions in the territory of the paramedian arteries were found. On initial DWI, the ischemic lesions had a median diameter of 10 (IQR 7.1–12.9) mm. In comparison, patients with LI in the anterior circulation demonstrated significantly larger ischemic lesions on DWI (median diameter of 11.1 (IQR 7.75–14.05) mm) than patients with LI in the posterior circulation (median diameter of 8.4 (IQR 5.95–10.8) mm, p = 0.002).

Figure 2.

Examples of acute lacunar infarction (LI) in typical locations on DWI and associated perfusion alterations. (a) Acute LI in the right basal ganglia (putamen) on DWI and corresponding area of hypoperfusion on PWI. (b) Acute LI in the right internal capsule on DWI and corresponding area of hypoperfusion on PWI (arrow). (c) Acute LI in the right corona radiata on DWI and corresponding area of hypoperfusion on PWI (arrow). (d) Acute LI in the right thalamus on DWI. Perfusion-weighted imaging shows a corresponding area of hyperperfusion (arrow). (e) Acute LI in the right pons and corresponding area of hypoperfusion on PWI (arrow).

Perfusion weighted imaging

Perfusion weighted imaging showed a circumscribed hypoperfusion in 58 (52.2 %) patients, a corresponding area of hyperperfusion in one (0.9%) patient whereas the remaining perfusion studies were normal. Figure 2 shows exemplary cases for each perfusion pattern. Median time between onset and MRI did not differ between patients with or without a circumscribed hypoperfusion on PWI (304 (IQR 208.5–488) min versus 337.5 (IQR 263.8–594.25) min, p = 0.44). The areas of circumscribed hypoperfusion had a median diameter of 9.0 (IQR 5.6–12) mm. In 40/58 (69.0%) patients the ischemic lesion on DWI was larger than the corresponding perfusion deficit on PWI, while in the remainder the perfusion deficit on PWI was larger than the ischemic lesion on DWI. Median time between onset and MRI did not differ between both groups (300 min (IQR 207.75–430.25) versus 360.5 (IQR 198.75–553.75) min, p = 0.59).

Overall, the ischemic lesion of DWI was significantly larger than the corresponding perfusion deficit on PWI (p = 0.001). In 22 (37.9%) the area of hypoperfusion was smaller, and in 36 (62.1%) larger than 7 mm in diameter. The frequency of an area of hypoperfusion on PWI did not differ significantly between LI in the anterior (36/74 (48.6%)) and posterior circulation (23/37 (62.2%), p = 0.18). However, frequency of a perfusion deficit differed markedly between infarction localizations: basal ganglia in 10/13 (76.9%), internal capsule in 15/27 (55.6%), thalamus in 20/27 (74.1%), corona radiata in 11/34 (32.4%), and brainstem in 3/10 (30.0%) patients (p = 0.003). With regard to the median diameter of the perfusion deficit on PWI, no significant differences were observed between patients with LI in the anterior (8.05 (IQR 5.7–12.7) mm) and posterior circulation (9.4 (IQR 5.3–11.5) mm, p = 0.92).

Overall, the median TTP, MTT, CBF, and CBV ratios were 1.12 (IQR 1.09–1.28), 1.19 (IQR 1.10–1.34), 0.69 (IQR 0.56–0.88), and 0.80 (IQR 0.64–1.13), respectively. Regarding TTP, MTT, CBF, and CBV ratios, there were no significant differences between patients with LI in the anterior and posterior circulation. For details see Table 2.

Dynamic 4D angiograms

With regard to the predefined types of blood flow in LI, 4/36 (11.1%) patients demonstrated type 1 (no blood flow visible), 17/36 (47.2%) patients type 2 (blood flow from the periphery to the center), 9/36 (25.0%) patients type 3 (blood flow from the center to the periphery), and 6/36 (16.7%) patients type 4 (combination of type 3 and type 4). Figure 3 shows exemplary cases of each pattern. Interrater reliability for blood flow patterns in LI prior to consensus was substantial (κ = 0.69, p < 0.001). No significant differences in TTP (p = 0.90), MTT (0.15), CBF (p = 0.45), and CBV ratios (p = 0.53) in patients with blood flow in LI type 1, type 2, type 3, or type 4 were observed. The same holds true for the median diameter (p = 0.71) and area of the ischemic lesion on DWI (p = 0.98) as well as the median diameter (p = 0.91) and area of the perfusion deficit on PWI (p = 0.08). Compared to patients with LI in the anterior circulation, patients with LI in the posterior circulation significantly more often had collateral blood flow type 2 and less often type 3 (p < 0.001). For details see Table 2.

Figure 3.

Examples of collateral blood flow types in lacunar infarction on dynamic 4D angiograms. (a) Type 1 (no blood flow visible). (b) Type 2 (blood flow from the periphery to the center). (c) Type 3 (blood flow from the center to the periphery). (d) Type 4 (combination of type 2 and type 3).

Discussion

The exact underlying pathomechanism in LI is still unexplained and the relevance of perfusion abnormalities as demonstrated in several case reports and smaller case series remains unclear.^27–29 The present study describes three novel and essential findings in LI: (1) a perfusion deficit on PWI is present in only about half of patients and in about two-thirds it is smaller than the ischemic lesion on DWI, (2) by use of dynamic 4D angiograms different collateral blood flow types can be identified, and (3) hyperperfusion also may be a rare finding.

In case of an occlusion of a small perforating artery of the brain, it is generally assumed that this will almost certainly lead to a LI due to the lack of anastomoses between these perforating vessels. Vessel occlusion may be due to lipohyalinosis,^2,3 microatheroma,^2,3 embolism,³ dissection of a small artery³ or arteriolar wall damage, and secondary thrombosis.³⁰ In the present study, approximately half of LI patients demonstrated a perfusion deficit on PWI which is less frequent compared to the results of earlier smaller studies reporting a perfusion deficit in 61.0,³¹ 62.5,²⁷ 67.4,²⁸ and 68.2%.³² Table 3 gives an overview of CT and MRI studies examining perfusion alterations in LI. In particular, with regard to the results of Fisher’s meticulous pathoanatomical studies from the 1960s reporting a frequency of persistent total vessel occlusion in 80–90% of LI,^2,3 this is an unexpected finding. However, this high frequency has been a matter of debate.³⁰ Furthermore, in about two-thirds of patients the area of hypoperfusion on PWI was smaller than the ischemic lesion on DWI. Several pathophysiological, anatomical, and technical considerations might possibly explain this observation.

Table 3.

Overview of CT and MRI studies examining perfusion alterations in LI.

Article	Year	n	Perfusion abnormalities
CTP
Yamada et al.²⁹	2004	26	Hypoperfusion
DSC PWI
Poppe et al.³²	2009	22	Hypoperfusion (match)
Fiebach et al.³³	2010	19	Hypoperfusion (inverse mismatch)
Förster et al.³⁴	2010	5	Hypoperfusion (match)
Förster et al.²⁷	2013	16	Hypoperfusion (match, mismatch, inverse mismatch)
Huang et al.²⁸	2014	43	Hypoperfusion (mismatch, inverse mismatch)

CTP: computed tomography perfusion; DSC PWI: dynamic susceptibility contrast perfusion-weighted imaging.

A longer time from onset to MRI could be associated with a higher frequency of recanalization and consequently absence of a perfusion deficit. Nevertheless, neither in earlier studies^29,31,32 nor in the present one has a significant difference in time between onset and perfusion imaging in patients with and without a perfusion deficit been observed. While this appears counterintuitive on first glance, this may be at least partly explained by persistent hypoperfusion in a subset of cases with persistent occlusion or stenosis of the perforating artery.^2,35

Also, the varying percentage of anatomical localizations could be another reason for the differing frequencies of a perfusion deficit in LI. In the present study, a perfusion deficit was much more common in LI in the basal ganglia, internal capsule, and thalamus than in the corona radiata and brainstem. Since, the perforating arteries supplying the latter have smaller diameters³⁶ and the site of occlusion is a well-known factor influencing recanalization with higher recanalization rates in vessels with smaller diameter,³⁷ it seems plausible that a perfusion deficit can be observed less frequently in the corona radiata and brainstem.

Another explanation for lacunar lesions without or only small perfusion deficit might be perivascular edema-mediated lesions, in which there is no occlusion but the tissue around the vessel is damaged by edema fluid.³⁸ Some authors also suggested a different underlying pathomechanism in small subcortical infarction with ischemic lesion growth beyond the hypoperfused area possibly induced by cytotoxic mechanisms at the border of the tissue damage.³³ This hypothesis would explain the large number of patients with a PWI deficit smaller than the ischemic lesion on DWI. Another explanation might be the partial recanalization of a previously occluded perforating artery or a collateral blood flow from adjacent vascular territories.

Finally, a limitation of perfusion MRI is the restricted spatial resolution and susceptibility to bone artifacts especially in the posterior fossa. Thus, small and very small areas of hypoperfusion could have been missed due to methodological reasons, especially in the brainstem.

The observation of a DWI–PWI mismatch in LI patients associated with neurological deterioration in the clinical course in two recently published clinical studies implies some kind of collateral blood flow in these cases.^27,28 Interestingly, the existence of anastomoses between small perforating arteries at different levels of the vascular tree was demonstrated already in a number of different pathoanatomical studies several years ago. In several comprehensive pathoanatomical studies, the presence of anastomoses between the extracerebral segments of major perforating arteries of the anterior circulation supplying the basal ganglia, internal capsule, and corona radiate,^6,7,10,15 as well as of the posterior circulation supplying the hypothalamus,¹¹ mesencephalon and thalamus,^8,9,13 and brainstem¹⁴ has been assessed meticulously. In the anterior circulation, Feekes and colleagues also evaluated the existence of anastomoses at the precapillary level between different neighboring arteries by injection of two different fluorescent dyes.³⁹

In the present study, we observed four different types of blood flow in LI on dynamic 4D angiograms. While some patients demonstrated no relevant blood flow at all (type 1), in the remainder a filling of the vascular territory from the periphery to the center (type 2), from the center to the periphery (type 3), or a combination of both (type 4) was found. Type 1 most likely represents the lack of anastomoses paralleled by an absence of collateral blood flow at all. Interestingly, this type was observed in only 3.6% of the patients suggesting that the absence of anastomoses and collateral blood flow between the perforating arteries of the brain is rather the exception than the rule. Type 2 in turn most likely reflects collateral blood flow from adjacent vascular territories via small anastomoses at the precapillary level as described in the pathoanatomical studies of Feekes et al.³⁹ In the perfusion CT scan-based study of Yamada et al., a similar pathomechanism has been hypothesized with preservation of CBF in LI via the microcirculation from the surrounding brain tissue.²⁹ Similarly, in type 4 the blood flow component from the periphery to the center of the perfusion lesion might represent collateral blood flow from neighboring vascular territories via these small anastomoses. In the present series, blood flow from the periphery to the center seems to be the most common pattern of collateral blood flow in LI either isolated in type 2 (47.2%) or in combination with type 3 blood flow in type 4 (13.9%). Furthermore, this type of blood flow has been observed more often in LI in the posterior circulation in comparison to LI in the anterior circulation. While the explanation of the observed collateral blood flow types 1 and 2 is rather easy, in type 3 and 4 a more detailed consideration seems to be necessary. In type 3 the observed blood flow pattern with filling of the vascular territory from the center to the periphery may be attributable to collateral blood flow via anastomoses between the main stems of two perforating arteries. On the other hand, the observed blood flow pattern might also represent delayed antegrade blood flow across an incomplete vessel occlusion or a circumscribed narrowing of the perforating artery as described previously.³⁵ Thus, in type 3 collateral blood flow and slow antegrade blood flow seem to be indistinguishable and the same holds true for the central blood flow component in type 4. This might also explain the higher frequency of type 3 in LI in the anterior circulation compared to LI in the posterior circulation which stands in contrast with the observation of Marinkovic and colleagues that anastomoses between the main stems of two perforating arteries are far more common in the posterior circulation.¹² Considering the results of the aforementioned pathoanatomical studies, it seems plausible to reason that blood flow type 3 as well as the central blood flow component in type 4 are probably due to delayed antegrade blood flow in most cases of anterior circulation LI. In contrast to this, in posterior circulation LI both blood flow pathways might still be possible.

Another interesting finding of the present study is the corresponding hyperperfusion in one patient with LI in the right thalamus. While hyperperfusion has been described in patients with large artery occlusion such as anterior or middle cerebral artery occlusion⁴⁰ and even in patients with basilar artery occlusion³⁴ after successful recanalization, to our best knowledge no reports on hyperperfusion in LI have been published until today. Although hyperperfusion is usually observed in regions of completed infarction, it has been assumed that hyperperfusion does not contribute to worse clinical outcome.⁴⁰ However, in LI this phenomenon seems to be a rather rare finding with a frequency of less than 1% in this study.

The present study has several limitations. First, this is a retrospective study of moderate sample size. However, to our knowledge this is the largest study on perfusion alterations in LI until now. The rather small number of acute LI patients undergoing PWI collected over a time period of more than 10 years possibly reflects the low appreciation of PWI in LI in daily clinical routine. Second, because of the limited spatial resolution of DSC PWI only perfusion deficits with a diameter larger than 7 mm were analyzed on dynamic 4D angiograms. Thus, the results cannot be generalized to all cases of LI. Third, the study was performed with different MRI scanners and different imaging sequences. However, the MRI sequences, in particular the PWI sequences, have been customized for optimal comparability in daily clinical routine and consequently are generally comparable. Fourth, we did not perform an intrasubject registration of DWI and PWI to a target common space. However, in none of the larger studies focusing on LI and PWI such a registration of DWI and PWI has been performed. Thus, this should not affect the results in general and in turn might facilitate comparison of the different study results. Fifth, our study lacks a gold standard for the assessment of collateral blood flow through anastomoses between small perforating arteries of the brain. However, given the benign clinical course in most cases and the low mortality of patients with LI, neither conventional selective angiography nor autopsy seems to be suitable as diagnostic method for comparison. Finally, the diagnosis of LI is based on the clinical syndrome, characteristic MRI findings, and the exclusion of competing etiologies like cardioembolism or large vessel disease. Since the underlying vascular lesion could only be determined by neuropathological examination, some uncertainty remains regarding the final diagnosis.

In conclusion, dynamic 4D angiograms are feasible to demonstrate different types of collateral blood flow in acute LI and consequently may serve for a better characterization of this stroke subtype and support treatment decisions in daily clinical practice. However, this should be confirmed in larger data sets and serial MRI examinations including PWI.

Footnotes

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Authors’ contributions

AF, BM: contributed equally to this work. AF: study concept and design, data acquisition, statistical analysis, manuscript preparation, manuscript editing. BM: literature research, data acquisition, data analysis, manuscript preparation, manuscript review. JB: data acquisition, data analysis, manuscript preparation, manuscript review. MAZ: data acquisition, manuscript preparation, manuscript review. HW: data acquisition, data analysis, statistical analysis, manuscript preparation, manuscript review. HUK: data acquisition, manuscript preparation, manuscript review. CG: data analysis, manuscript preparation, manuscript review

References

Bamford

Sandercock

Dennis

. Classification and natural history of clinically identifiable subtypes of cerebral infarction. Lancet 1991; 337: 1521–1526.

Fisher

. The arterial lesions underlying lacunes. Acta Neuropathol 1968; 12: 1–15.

Fisher

. Lacunar strokes and infarcts: a review. Neurology 1982; 32: 871–876.

Potter

Marlborough

Wardlaw

. Wide variation in definition, detection, and description of lacunar lesions on imaging. Stroke 2011; 42: 359–366.

Benavente

White

Pearce

. The secondary prevention of small subcortical strokes (SPS3) study. Int J Stroke 2011; 6: 164–175.

Rosner

Rhoton

Jr . Ono M, et al. Microsurgical anatomy of the anterior perforating arteries. J Neurosurg 1984; 61: 468–485.

Marinkovic

Kovacevic

Marinkovic

. Perforating branches of the middle cerebral artery. Microsurgical anatomy of their extracerebral segments. J Neurosurg 1985; 63: 266–271.

Marinkovic

Milisavljevic

Kovacevic

. Anastomoses among the thalamoperforating branches of the posterior cerebral artery. Arch Neurol 1986; 43: 811–814.

Milisavljevic

Marinkovic

Lolic-Draganic

. Anastomoses in the territory of the posterior cerebral arteries. Acta Anat (Basel) 1986; 127: 221–225.

10.

Gomes

Dujovny

Umansky

. Microanatomy of the anterior cerebral artery. Surg Neurol 1986; 26: 129–141.

11.

Marinkovic

Milisavljevic

Marinkovic

. Microanatomy and possible clinical significance of anastomoses among hypothalamic arteries. Stroke 1989; 20: 1341–1352.

12.

Marinkovic

Milisavljevic

Marinkovic

. Anastomoses among the perforating arteries of the brain. Microanatomy and clinical significance. Neurologija 1990; 39: 107–114.

13.

Milisavljevic

Marinkovic

Gibo

. The thalamogeniculate perforators of the posterior cerebral artery: the microsurgical anatomy. Neurosurgery 1991; 28: 523–529.

14.

Marinkovic

Gibo

. The surgical anatomy of the perforating branches of the basilar artery. Neurosurgery 1993; 33: 80–87.

15.

Morandi

Brassier

Darnault

. Microsurgical anatomy of the anterior choroidal artery. Surg Radiol Anat 1996; 18: 275–280.

16.

Kang

Han

Kwon

. Evaluation of the lenticulostriate arteries with rotational angiography and 3D reconstruction. Am J Neuroradiol 2005; 26: 306–312.

17.

Gotoh

Okada

Satogami

. Evaluation of CT angiography for visualisation of the lenticulostriate artery: difference between normotensive and hypertensive patients. Br J Radiol 2012; 85: e1004–e1008.

18.

Conijn

Hendrikse

Zwanenburg

. Perforating arteries originating from the posterior communicating artery: a 7.0-Tesla MRI study. Eur Radiol 2009; 19: 2986–2992.

19.

Liebeskind

. Collateral circulation. Stroke 2003; 34: 2279–2284.

20.

McVerry

Liebeskind

Muir

. Systematic review of methods for assessing leptomeningeal collateral flow. Am J Neuroradiol 2012; 33: 576–582.

21.

Campbell

Christensen

Tress

. Failure of collateral blood flow is associated with infarct growth in ischemic stroke. J Cereb Blood Flow Metab 2013; 33: 1168–1172.

22.

Förster

Wenz

Kerl

. FLAIR vascular hyperintensities and dynamic 4D angiograms for the estimation of collateral blood flow in posterior circulation occlusion. Neuroradiology 2014; 56: 697–707.

23.

Kim

Son

Ryoo

. A novel magnetic resonance imaging approach to collateral flow imaging in ischemic stroke. Ann Neurol 2014; 76: 356–369.

24.

Weimar

Ringelstein

Diener

. Monitoring stroke units: management, outcome, efficiency. Nervenarzt 2007; 78: 957–966.

25.

Tarabishy

. Cortical calcification in Sturge-Weber Syndrome on MRI-SWI: relation to brain perfusion status and seizure severity. J Magn Reson Imaging 2011; 34: 791–798.

26.

Rosset

Spadola

Ratib

. OsiriX: an open-source software for navigating in multidimensional DICOM images. J Digit Imaging 2004; 17: 205–216.

27.

Förster

Kerl

Wenz

. Diffusion- and perfusion-weighted imaging in acute lacunar infarction: is there a mismatch? PLoS One 2013; 8: e77428.

28.

Huang

Tsai

Lee

. Hemodynamic factors may play a critical role in neurological deterioration occurring within 72 hrs after lacunar stroke. PLoS One 2014; 9: e108395.

29.

Yamada

Yoshimura

Kaku

. Prediction of neurologic deterioration in patients with lacunar infarction in the territory of the lenticulostriate artery using perfusion CT. Am J Neuroradiol 2004; 25: 402–408.

30.

Wardlaw

Dennis

Warlow

. Imaging appearance of the symptomatic perforating artery in patients with lacunar infarction: occlusion or other vascular pathology? Ann Neurol 2001; 50: 208–215.

31.

Das

Settecase

Boulos

. Multimodal CT provides improved performance for lacunar infarct detection. Am J Neuroradiol 2015; 36: 1069–1075.

32.

Poppe

Coutts

Kosior

. Normal magnetic resonance perfusion-weighted imaging in lacunar infarcts predicts a low risk of early deterioration. Cerebrovasc Dis 2009; 28: 151–156.

33.

Fiebach

Hopt

Vucic

. Inverse mismatch and lesion growth in small subcortical ischaemic stroke. Eur Radiol 2010; 20: 2983–2989.

34.

Förster

Ottomeyer

Wolf

. Dynamic susceptibility contrast perfusion MRI identifies persistent vessel pathology in acute pontine stroke. Cerebrovasc Dis 2010; 29: 389–394.

35.

De Reuck

Vander

. The arterial angioarchitecture in lacunar state. Acta Neurol Belg 1976; 76: 142–149.

36.

Djulejic

Marinkovic

Milic

. Common features of the cerebral perforating arteries and their clinical significance. Acta Neurochir (Wien) 2015; 157: 743–754.

37.

Bhatia

Hill

Shobha

. Low rates of acute recanalization with intravenous recombinant tissue plasminogen activator in ischemic stroke: real-world experience and a call for action. Stroke 2010; 41: 2254–2258.

38.

Lammie

Brannan

Wardlaw

. Incomplete lacunar infarction (Type Ib lacunes). Acta Neuropathol 1998; 96: 163–171.

39.

Feekes

Hsu

Chaloupka

. Tertiary microvascular territories define lacunar infarcts in the basal ganglia. Ann Neurol 2005; 58: 18–30.

40.

Kidwell

Saver

Mattiello

. Diffusion-perfusion MRI characterization of post-recanalization hyperperfusion in humans. Neurology 2001; 57: 2015–2021.