Jernej Vidmar

Microscopic clot fragment evidence of biochemo-mechanical degradation effects in thrombolysis

INTRODUCTION Although fibrinolytic treatment has been used for decades, the interactions between the biochemical mechanisms and the mechanical forces of the streaming blood remain incompletely understood. Analysis of the blood clot surface in vitro was employed to study the concomitant effect of blood plasma flow and recombinant tissue plasminogen activator (rt-PA) on the degradation of retracted, non-occlusive blood clots. Our hypothesis was that a faster tangential plasma flow removed larger fragments and resulted in faster overall thrombolysis. MATERIALS AND METHODS Retracted model blood clots were prepared in an optical microscopy chamber and connected to an artificial perfusion system with either no-flow, or plasma flow with a velocity of 3 cm/s or 30 cm/s with or without added rt-PA at 2 microg/ml. The clot surface was dynamically imaged by an optical microscope for 30 min with 15s intervals. RESULTS The clot fragments removed during rt-PA mediated thrombolysis ranged in size from that of a single red blood cell to large agglomerates composed of more than a thousand red blood cells bound together by partly degraded fibrin. The average and the largest discrete clot area change between images in adjacent time frames were significantly higher with the faster flow than with the slow flow (14,000 microm(2) and 160,000 microm(2) vs. 2200 microm(2) and 10,600 microm(2)). CONCLUSIONS On the micrometer scale, thrombolysis consists of sequential removal of clot fragments from the clot surface. With increasing tangential plasma flow velocity, the size of the clot fragments and the overall rate of thrombolysis increases.

Impact of altered venous hemodynamic conditions on the formation of platelet layers in thromboemboli

Although it is generally believed that the structure of venous thromboemboli is a homogeneous red blood cell-fibrin clot, their structure may be heterogeneous, with non-uniformly distributed platelet layers, known as the lines of Zahn. We tested (a) whether venous thromboemboli ex vivo contained platelet layers, i.e. the lines of Zahn, and (b) whether, according to mathematical modeling, eddies can arise in the venous system, possibly contributing to platelet aggregation. The structure of venous thromboemboli ex vivo was determined by high-resolution magnetic resonance imaging (MRI) and by immunohistochemistry (IHC). High-resolution ultrasound (US) imaging was employed to determine the popliteal vein geometry and hemodynamics in healthy subjects and in subjects with previous venous thrombosis. The US data were then used as input for numerical simulations of venous hemodynamics. MRI and IHC confirmed that 42 of 49 ex vivo venous thromboemboli were structurally heterogeneous with platelet layers. The peak venous flow velocity was higher in patients with partly recanalized deep vein thrombosis than in healthy subjects in the prone position (46±4cm/s vs. 16±3cm/s). Our numerical simulation showed that partial venous obstruction with stenosis or malfunctioning venous valves creates the conditions for eddy blood flow. Our experimental results and computer simulation confirmed that the heterogeneous structure of venous thromboemboli with twisted platelet layers may be associated with eddy flow at the sites of their formation.

Use of multiparametric magnetic resonance microscopy for discrimination among different processing protocols and anatomical positions of Slovenian dry-cured hams.

A novel multiparametric magnetic resonance microscopy (MRM) approach was applied to the Slovenian Kraški pršut dry-cured ham samples in order to evaluate its potential for discrimination among biceps femoris and semimembranosus muscle from two hams, differing in processing (salting duration) and thus in water and salt content. The approach is based on apparent diffusion coefficient (ADC) mapping as well as on longitudinal (T1) and transversal (T2) nuclear magnetic resonance relaxation time mapping. Three-dimensional maps were acquired and analyzed by one dimensional (1D) ADC, T1, and T2 distributions as well as by paired two-dimensional ADC-T1, ADC-T2 and T1-T2 distributions. The discriminating potential of the applied MRM approach was confirmed by differences among both 1D and 2D distributions of different ham samples. In addition, distribution peak positions highly correlated with the conventionally determined moisture content.

Multiparametric MRI in characterizing venous thrombi and pulmonary thromboemboli acquired from patients with pulmonary embolism

The structure of thrombi plays an important role in delaying reperfusion and re‐occlusion after intravenous thrombolysis and could influence the performance of mechanical thrombectomy devices. This study aims to distinguish various thrombi groups based on their T2 and apparent diffusion coefficient (ADC) properties.

A comparison of the ADC and T2 mapping in an assessment of blood-clot lysability.

The structural characteristics of blood clots are associated with their susceptibility to thrombolysis. As their morphology can be characterized by MRI, several attempts have been made to link the lysability of blood clots with their MRI properties; however, so far no study has associated a clots lysability with the diffusion properties of the water in the clot. The apparent diffusion coefficient (ADC) is highly sensitive to changes in serum mobility and may be used to distinguish between the non‐retracted and the fully retracted regions of the blood clot. Therefore, the ADC may be a suitable, or even a better, marker for an assessment of the clots retraction and consequently for its lysability than the relaxation time T2. The purpose of this study was to evaluate whether it is possible to predict the outcome of clot thrombolysis by ADC mapping prior to treatment. After two hours of thrombolysis using a recombinant tissue plasminogen activator in plasma, whole‐blood clots were efficiently dissolved in regions with ADC ≥ 0.8 ×  10−9 m2/s or T2 ≥ 130 ms, whereas dissolution was poor and prolonged in regions with ADC < 0.8 × 10−9 m2/s or T2 < 130 ms. An analysis based on a comparison between the initial and final ADC and T2 maps after two hours of thrombolysis showed that the ADC can more accurately detect the different grades of clot retraction than T2 and predict the regions of a clot that are resistant to thrombolysis. Therefore, the ADC could be used as an efficient prognostic marker for the outcome of thrombolysis. However, in vivo studies are needed to test this idea. Copyright

Assessment of the dentin-pulp complex response to caries by ADC mapping

The prognostic potential of apparent diffusion coefficient (ADC) mapping was studied as complemented by high‐resolution 3D T1‐weighted MRI in the assessment of dentin‐pulp complex response to caries. Twenty‐six extracted human teeth, with or without caries lesions of different grades in accord with the International Caries Detection and Assessment System (ICDAS), were analyzed by high‐resolution MRI at 2.35 T. A signal rise in demineralized hard dental tissues in high‐resolution T1‐weighted MR images enabled assessment of the demineralization depth over the whole range of ICDAS scores. ADC maps of the teeth were calculated from corresponding diffusion‐weighted images of four different b values: 0, 132, 317, 635 s/mm2. These maps enabled reliable differentiation between intact (ADC > 1.0·10‐9 m2/s) and affected (ADC < 1.0·10‐9 m2/s) regions of dental pulp. Linear regression analyses of demineralization depth in relation to ICDAS score and then also to average ADC of dental pulp showed that a demineralization depth increase of one millimeter corresponds to an ICDAS score increase of 1.2 and an average ADC decrease of 0.07·10‐9 m2/s. Results of the study indicate that the average ADC value of dental pulp could be used as a potential marker to assess tissue response to caries comparable to that of ICDAS scoring. Copyright

An MRI study of the differences in the rate of thrombolysis between red blood cell-rich and platelet-rich components of venous thrombi ex vivo.

To test whether T1‐weighted MRI can detect the differences in the rate of thrombolysis induced by recombinant tissue plasminogen activator (rt‐PA) between platelet‐rich regions and red blood cell (RBC)‐rich regions of venous thrombi ex vivo.

Application of quantitative magnetization transfer magnetic resonance imaging for characterization of dry-cured hams.

Quantitative magnetization transfer magnetic resonance imaging (qMT-MRI) was employed to characterize dry-cured ham tissues differing in anatomical positions and processing protocols. Experimentally obtained MR images of dry-cured ham sections were analyzed by the well-established binary-spin-bath (BSB) model. The model enabled an efficient discrimination between a free-water proton pool and a restricted-macromolecular proton pool. Significant differences in restricted pool sizes were found among different ham sections. Values of the restricted pool size obtained by the model were in a good agreement with chemically determined protein content. The study confirmed the feasibility of the applied qMT-MRI as a nondestructive tool for characterization of dry-cured ham tissues.

Analysis of blood clot degradation fragment sizes in relation to plasma flow velocity.

Thrombolytic therapy aims to dissolve blood clots and restore vessel patency. Our hypothesis is that the therapy depends not only on chemical reactions of the fibrinolytic system, but also on mechanical forces exerted by streaming blood on the clot surface. The aim of the study was to analyze the role of mechanical forces in promoting thrombolysis and their relation to the maximum size of removed clot fragments. Non-occlusive whole-blood clots were exposed to flow of plasma containing the thrombolytic agent rt-PA. Plasma, perfusing through the clot, was collected and the sizes of clot degradation fragments in the plasma were analyzed by optical microscopy. Theoretical models for the maximum clot fragment size as a function of blood flow velocity were developed based on the relation between surface or volume clot binding forces and the opposing forces of the streaming plasma in the laminar or turbulent flow regime. The best agreement between experimental results and models was obtained for the volume binding forces and the laminar flow model, in which the maximum clot fragment size was linearly dependent on the plasma flow velocity. Such result could not be obtained if thrombolysis would be purely a biochemical process. Therefore, the result confirms our hypothesis that thrombolysis is also strongly influenced by the mechanical forces of streaming plasma.

Thrombosis of pancreatic arteriovenous malformation induced by diagnostic angiography: case report.

BackgroundWe report on a case of pancreatic arteriovenous malformation (PAVM) that obliterated shortly after diagnostic angiography (DSA). PAVM is a rare anomaly that presents with upper abdominal pain, signs of acute pancreatitis and massive gastrointestinal bleeding. The management of PAVM is rather complex, with complete treatment usually accomplished only by a total extirpation of the affected organ or at least its involved portion. DSA prior to treatment decisions is helpful for characterizing symptomatic PAVM, since it can clearly depict the related vascular networks. In addition, interventional therapy can be performed immediately after diagnosis.Case presentationA 39-old male was admitted due to recurring upper abdominal pain that lasted several weeks. Initial examination revealed the absence of fever or jaundice, and the laboratory tests, including that for pancreatic enzymes, were unremarkable. An abdominal ultrasound (US) showed morphological and Doppler anomalies in the pancreas that were consistent with a vascular formation. A subsequent DSA depicted a medium-sized nidus, receiving blood supply from multiple origins but with no dominant artery. Coil embolization was not possible due to the small caliber of the feeding vessels. In addition, sclerotherapy was not performed so as to avoid an unnecessary wash out to the non-targeted duodenum. Consequently, the patient received no specific treatment for his symptomatic PAVM. A large increase in pancreatic enzymes was noticed shortly after the DSA procedure. Imaging follow-up by means of CT and MRI showed small amounts of peripancreatic fluid along with a limited area of intra-parenchymal necrosis, indicating necrotizing pancreatitis. In the post-angiography follow-up the patient was hemodynamically stable the entire time and was treated conservatively. The symptoms of pancreatitis improved over a few days, and the laboratory findings returned to normal ranges. Long-term follow-up by way of a contrast-enhanced CT revealed no recanalization of the thrombosed PAVM.ConclusionThe factors associated with the obliteration of PAVM during or after DSA are poorly understood. In our case it may be attributed to the low flow dynamics of PAVM, as well as to the local administration of a contrast agent. Asymptomatic PAVM, as diagnosed with non-invasive imaging techniques, should not be evaluated with DSA due to the potential risk of severe complications, such as acute pancreatitis.