0082

MR microscopy to assess clot composition following mechanical thrombectomy predicts recanalization and clinical outcome

Kianush Karimian-Jazi¹, Dominik Vollherbst¹, Daniel Schwarz¹, Manuel Fischer¹, Katharina Schregel¹, Gregor Bauer², Anna Kocharyan³, Volker Sturm¹, Ulf Neuberger¹, Jessica Jesser¹, Christian Herweh¹, Christian Ulfert¹, Tim Hilgenfeld¹, Fatih Seker¹, Fabian Preisner¹, Niclas Schmitt¹, Tobias Charlet¹, Stefan Hamelmann⁴, Felix Sahm⁴, Sabine Heiland¹, Wolfgang Wick¹, Peter Ringleb², Lucas Schirmer³, Martin Bendszus¹, Markus Möhlenbruch¹, and Michael Breckwoldt¹
¹Neuroradiology, University Hospital Heidelberg, Heidelberg, Germany, ²Neurology, University Hospital Heidelberg, Heidelberg, Germany, ³Neurology, University Hospital Mannheim, Mannheim, Germany, ⁴Neuropathology, University Hospital Heidelberg, Heidelberg, Germany

Synopsis

Keywords: Stroke, Stroke

Motivation: Mechanical thrombectomy (MT) is the standard for ischemic stroke with large vessel occlusion (LVO). Clot composition is underexplored in clinical practice, leading to standardized MT regardless of clot type.

Goal(s): This single-center study examined clot composition in 60 LVO stroke patients using high-field MRI at 9.4T ("MR-microscopy").

Approach: MR microscopy correlated with histopathology, and quantifying the hyperdense artery sign (HAS) on pre-interventional CT further stratified clot composition.

Results: MR microscopy successfully identified clot types—red (23%), white (28%), or mixed (48%)—with 95.4% accuracy. White clots required more passes during MT, had worse clinical outcomes, while red clots showed better first-pass recanalization rates.

Impact: This study suggests clot imaging can personalize MT for improved outcomes in LVO stroke patients.

Background

Mechanical thrombectomy (MT) is the standard of care for ischemic stroke patients with large vessel occlusion (LVO) eligible for endovascular treatment.^1,2Clot composition is not routinely assessed in clinical practice as no specific diagnostic value is attributed to it and currently, MT is performed in a standardized “non-personalized” approach, irrespective of the stroke etiology or clot composition. Whether different clot compositions are associated with intrinsic likelihoods of recanalization success or treatment outcome is unknown.

Methods

We performed a prospective, non-randomized, single center study and systematically analyzed the clot composition in 60 consecutive ischemic stroke patients undergoing MT at our institution. Clot specimens were assessed by ex vivo high field MRI at 9.4T (“MR-microscopy”, MR-M) by anatomical and relaxometric MR sequences and high-resolution DVT. Histopathology was performed to validate MR-M results in a subset of 22 specimens. Clot imaging was correlated with preinterventional CT for hyperdense artery sign (HAS) quantification and clinical data.

Results

Ex vivo high field MR microscopy had high accuracy of 95.4% to differentiate the different clot types and 21/22 clots were categorized correctly using histopathology as ground truth. MR microscopy showed red blood cell-rich (red, 23% of clots), platelet-rich (white, 28%) or mixed clots (48%) as distinct morphological patterns. T2* times differed between the clots types as assessed by MR relaxometry. Clot composition could be further stratified by quantifying the HAS on pre-interventional non-enhanced CT. During MT, white clots required significantly more passes to achieve final recanalization and were not amenable to contact aspiration compared to mixed and RBC-rich clots, whereas red clots showed highest probability of first pass recanalization (76.92%) compared to white clots (17.39 %). Clinically, white clots were associated with poorer clinical outcome and performance status at 90 mRS.

Discussion

Our study characterizes the composition of retrieved clots following MT by multimodality quantitative imaging and introduces MR microscopy to assess clot composition. Our study demonstrates that preinterventional clot imaging by quantification of the HAS or MR relaxometry could guide the interventional strategy and predict clinical outcome. Thus, our study could enable a personalized treatment approach to improve radiological MT results and ultimately clinical outcome of stroke patients with LVO undergoing MT. In subsequent studies, we aim to investigate the composition of clots in patients suffering from stroke secondary to pertinent pre-existing conditions, such as cancer. We hypothesize that these clots exhibit a complex structure that may impede mechanical recanalization efforts. Notably, the application of relaxometry for quantification holds substantial potential for pre-interventional clot detection; currently, relaxometry is being measured in artificially generated thrombi within the context of another study, which may standardize this assessment technique.

Acknowledgements

The authors thank Dr S Bonekamp (Neuroradiology Department, University Hospital Heidelberg) for organizational and Manfred Jugold (DKFZ) for technical support of this study.

References

1. Berkhemer OA, Fransen PSS, Beumer D, et al. A randomized trial of Intraarterial treatment for acute ischemic stroke. N Engl J Med 2015;372:11–20.
2. Goyal M, Menon BK, van Zwam WH, et al. Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet 2016;387:1723–31.

Figures

MR microscopy at 9.4 T ex vivo imaged clots following mechanical thrombectomy. Ex vivo magnified photographs and T2- weighted (T2-w), T2*-w, and T1-w images showing different thrombus morphologies after stroke thrombectomy. Red, erythrocyte-rich clots (A) present as hypointense in all three sequences, while white, platelet-rich clots (C) show a weak hyperintense signal, especially in the T1-w images. Mixed clots (B) have a heterogeneous appearance. Classification of the clot composition based on MR microscopy (D) indicates that mixed clots represent the largest fraction.

Correlative imaging, H&E histological staining, QuPath analysis, and close-up images of the clots demonstrate that red blood cell (RBC)-rich clots are composed of a predominant proportion of erythrocytes, whereas almost no erythrocytes can be visualized in platelet-rich clots (A–C). Distribution of clot composition based on histopathology quantification (D). Thresholds for clot composition on histopathology were defined as< 25% of RBCs for platelet-rich, 25–75% for mixed and >75% for RBC-rich clots. Histological evaluation was performed semi-automatically using QuPath.

Representative comparative imaging of red blood cell (RBC)-rich (A), mixed (B) and white clots (C) shows that the hyperdense artery sign (HAS) can be used to differentiate clots types. Regression analysis of HAS and T2 relaxation time shows that lower relaxation times indicate a RBC-rich clot (E). Quantification of digital volume tomography (DVT) demonstrates a significantly lower DVT ratio in RBC-rich clots than in mixed and white, platelet-rich clots (F). Evaluation of MR relaxometry shows significant differences between RBC-rich, mixed, and white clots.

The hyperdense artery sign (HAS) is associated with the number of retrieval attempts (A) and time to final recanalization (B). There is a weak correlation of HAS and retrieval attempts (C). First pass was mainly achieved in red clot occlusions (D). Contact aspiration was mainly successful in red clots, whereas white clots could not be recanalized by contact aspiration only (E). IV thrombolysis with recombinant tissue plasminogen activator did not change maneuver numbers in red (F) and mixed clots (G), whereas it was associated with fewer maneuvers in white clots (H).

Proc. Intl. Soc. Mag. Reson. Med. 32 (2024)

0082

DOI: https://doi.org/10.58530/2024/0082