DE10046514A1 - Process for imaging and diagnosis of thrombi using magnetic resonance imaging using particulate contrast media - Google Patents

Abstract

The invention relates to a new method for imaging and diagnosis of thrombi and the use of particle suspensions for the production of contrast media for the imaging of thrombi by means of magnetic resonance imaging.

Description

The invention relates to a new method for imaging and diagnosis of thrombi and the use of particle suspensions for the production of Contrast media for imaging thrombi using magnetic resonance imaging.

A thrombus is a circumscribed blood fixation that occurs in arteries or veins arises from intravascular blood clotting. Thrombosis, i.e. H. a partial or complete occlusion of the artery or vein by a thrombus can lead to a Anemia and an organ death (infarction). A typical example of arterial thrombosis is that of the coronary arteries (coronary thrombosis). Dissolves a thrombus from the vessel wall, it is flushed away with the blood stream. This Thrombus can be a subordinate smaller vessel in the sense of thromboembolism close. The brain arteries are a typical example of Thromboembolism in arteries. The very common pelvic and thrombosis Leg veins typically lead to pulmonary artery thromboembolism. The Pulmonary artery embolism is particularly feared because it is difficult to detect and often takes a fatal outcome.

So far, thrombi have been diagnosed using various methods. The most Commonly used procedures are catheter x-ray angiography X-ray phlebography and various ultrasound procedures. The angiography will performed increasingly with the help of magnetic resonance imaging (magnetic resonance angiography), showing the blood flow in the vessels. This process is therefore strong flußabhängig. Not all vessels are reached in X-ray phlebography. Of the ultrasonic method is a frequently used method that the duplex Sonography FKDS, which however can only show superficial veins and failed in the area of the pelvis. None of these diagnostic methods provide any unequivocal results and can lead to misdiagnosis.  

In international patent application WO 98/16256 a method for Representation of thrombi by means of magnetic resonance imaging proposed, in which the Contrast agents are applied to patients, which consist of and on chelators coupled integrin-binding molecules exist. The chelators complex paramagnetic metal ions, such as B. Gadolinium ions.

Similar contrast agents are also found in international application WO 95/20603 proposed. The agents described there are peptides to which each Complexing agent is coupled. The complexing agent can complex a metal ion, that either emit radioactive radiation and detect it with a gamma camera can be, or is a gadolinium or other paramagnetic metal ion, which is suitable for use as a contrast medium in magnetic resonance imaging.

However, these patent applications only describe the synthesis of the contrast agents and first in-vitro experiments. There are no in vivo experiments or Magnetic resonance tomograms shown, with which the effectiveness of the funds beyond doubt could be demonstrated. Such contrast agents are also not yet from Pharmaceutical companies have been developed or are available for sale as usable contrast agents purchase.

There is therefore still a need for a new, unambiguous and safe Diagnostic procedures for and for arterial and venous thrombi Suitable contrast agents.

The object of the present invention is therefore a new method for diagnosis to develop arterial and venous thrombi as well as suitable contrast media for a to find such a procedure.

This task is with the new method according to claim 1 and the Use of particle suspensions for the production of contrast media according to Claim 6 solved.  

Surprisingly, it was found that MR blood pool contrast media, such as superparamagnetic iron oxide (SPIO), especially in a formulation with small particles (ultrasmall superparamagnetic iron oxide, USPIO), enrich in an animal model in experimental thrombi and after a time interval of several blood half-lives in magnetic resonance tomograms are visible and diagnostically usable. It was found that this Contrast agents are preferred in the thrombi, but often also in the adjacent one Vessel wall and environment enriched.

In the method according to the invention, the patient is first treated accordingly particulate MR contrast agent administered, and after the enrichment of the Contrast agent in the thrombus and / or the adjacent vessel wall and surroundings (i.e. after a time interval of several blood half-lives) is a Magnetic resonance imaging recorded. A particularly good representation can be achieved with T1-weighted magnetic resonance tomograms. After intracellular uptake of the Contrast agents in the macrophages (phagocytes) of thrombi can also have an effect generated in T2-weighted images.

Those used in the experiment described in detail below ultra-small superparamagnetic iron oxide particles (USPIO) consist of one Iron oxide core and a carboxydextran shell. The mean diameter of the Particle is preferably less than 50 nm, particularly preferably about 25 nm Production of such particles is, for. B. in the patents EP 656 368 and WO 98/05430 described. Such a particle containing contrast medium is currently developed by Schering AG.

The iron-containing contrast media are, for example, in a dosage of 200 µmol Fe / kg body weight used.

After waiting several blood half-lives during which the contrast agent Enriched in the thrombi are images with a magnetic resonance scanner added. The thrombi are clearly visible in the images thus obtained.  

If the contrast medium accumulates extracellularly, it can by its T1 effect Make the thrombus visible on T1-weighted images. Will that Contrast agent from macrophages (reagent cells) that are involved in thrombus breakdown (Thrombus organization) regularly in the vessel wall or in the edge of the thrombus immigrate, recorded, the T2 effect of the contrast medium may predominate, whereby the marked tissue is low-signal in T2- or T2 * -weighted images appears.

An animal experiment is described in detail below, in which 25 rabbits Thrombi due to catheter embolization and thrombin injection in the jugular vein externa were generated. After 1, 3, 5, 7 and 9 days (each n = 5), T1w-MP- RAGE and T2 * w-FLASH magnetic resonance imaging at 1.5 Tesla before and 24 hours after intravenous application of ultra-small superparamagnetic iron oxide (USPIO, Particle size approx. 25 nm) measured in a dose of 200 µmol Fe / kg body weight. The X-ray phlebography and histology served as the gold standard. The length of the in 3D Reconstructions of the T1w-MP-RAGE sequence visible thrombus were performed in Relationship to true thrombus length. The structure, the signal intensity and the contrast range of the thrombus in the T2 * w technique was subjective Subjected to analysis with a defined scaling. 25 rabbits with Age-appropriate thrombi served as controls.

This experiment is intended to illustrate the invention without restricting it to this want.  

Example for the detection of experimental thrombi using magnetic resonance imaging after injection of particulate contrast media Material and method contrast agents

There has been a formulation of ultra small superparamagnetic iron oxide particles (USPIO) used. The particles consist of an iron oxide core and a Carboxydextran shell with a total diameter of 25 nm. In the rat it is effective plasma half-life 56 ± 17 minutes and the LD50 35 mmol Fe / kg body weight. in the Rabbits have an estimated plasma half-life of approximately 6 hours. The contrast medium was in a dosage of 200 µmol Fe / kg body weight in a volume of 2-3 ml Ear vein of the healthy side slowly and followed by 2 ml physiological Saline injected.

animal model

The experiments were in accordance with the German animal protection law carried out. Approval from the responsible state authority was obtained. There were Chinchilla bastard rabbits of both sexes (2.6-3.8 kg) were used. All Investigations were carried out under deep anesthesia by subcutaneous injection of 40 mg / kg Ketamine HCl (Ketanest-50, Parke-Davis GmbH, Berlin) and 17 mg / kg xylazine HCl (Rompun 2%; Bayer, Leverkusen). Before thrombus induction, the Neck veins first using x-ray phlebography with injection of a Standard contrast medium documented in an ear vein on two levels. Based on the method used by Wessler became a stagnation thrombus due to blood stasis and hypercoagulability [Wessler S. thrombosis in the presence of vascular stasis. On J Med 1962; 33: 648-666.]. However, the stasis was not due to surgery Ligature of the vein, but brought about by catheter embolization. From one Ear vein was cut out using a sheath and a guidewire to approximately 20 cm long guide catheter (3 French) advanced into the external jugular vein. On the Guide catheter became a micro catheter with an outer diameter of 0.61  Millimeters with the tip of the catheter positioned at the level of the 4th or 5th cervical vertebra. The Embolisat, a mixture of an oily iodine-containing lymphography X-ray contrast media (Lipiodol, Byk Gulden, Constance) and a butyl cyanoacrylate Tissue adhesive (Histoacryl, B. Braun, Melsungen) in the ratio 1: 1, was so slowly injected that, on the one hand, adhesion of the emboliate to the vessel wall and on the other hand, a mixture of embolisate and blood could occur. After this The vessel was closed using a second microcatheter 100 units of bovines Thrombin (Sigma-Aldrich, Chemie GmbH, Steinheim) applied. The embolisate should be for lead to a temporary closure of the vessel within a few days. The embolisate should because of the iodine content both radiopaque on the X-ray phlebographs as well because of the hyperintense on T1-weighted MRI in the glue-fixed blood breakdown products Be visible to images.

Experimental design

Fifty thrombus-carrying animals were randomly assigned to the contrast medium group (D) or assigned to the control group (K). A further subdivision was made according to the thrombus age at the time of the first MRI measurement (1, 3, 5, 7 or 9 days). Accordingly, there was an assignment in 5 each Contrast medium groups (D1, D3, D5, D7 and D9) and 5 control groups (K1, K3, K5, K7 and K9). Each subgroup comprised five test animals. In all experimental animals the thrombus was documented by X-ray phlebography prior to the baseline MRI. Animals from the contrast medium group (K) received this after this initial MRI Contrast medium and after a 24-hour interval the progressive MRI. In group D3 z. B. after embolization on day 0, a baseline MRI on day 3 and one Follow-up MRI on day 4. Of the controls, only animals from groups K1, K3 received and K5 a progressive MRI, but not animals of groups K7 and K9, which were initially of the experiments had been measured when the length of the gradient interval was not yet it was clear.

MRI

The investigations were carried out in a 1.5 Tesla magnet (Vision, Siemens, Erlangen) performed with a standard knee coil. The anesthetized animals were in  Supine position positioned in a U-shaped foam pad. A watery one Plastic bags were placed on the ventral side of the neck to improve fat saturation.

T1-weighted images were taken in a coronary slice orientation with a 3D Magnetization-Prepared Rapid Gradient-Echo Imaging Sequence (3D-MP-RAGE) generated [Mugler JPd, Brookeman JR. Three-dimensional magnetization-prepared rapid gradient-echo imaging (3D MP RAGE). Magn Reson Med 1990; 15: 152-7.]. The sequence used suppresses the signal from fat by a sequence shown selective water excitation pulse and that of blood by choosing the Delay time [Moody AR, Pollock JG, O'Connor AR, Bagnall M. Lower-limb deep venous thrombosis: direct MR imaging of the thrombus. Radiology 1998; 209: 349-55.]. The following parameters were used: repetition time 10.3 ms; Echo time 4.0 ms; Flip angle 15 °; Inversion time 20 ms; Delay time 1,000 ms; Number of layers 100; Partitions 80; Acquisitions 1; Layer thickness 0.8 mm; Field of view 100 × 200 mm; Matrix 128 x 256; Acquisition time 5:30 minutes; pixel size 0.78 x 0.78 x 0.8 mm. From the 2D source images, approx. 1.5 cm thick 3D Maximum intensity projection reconstructions (MIP) made. T2 * -weighted axial images were recorded using a 3D gradient echo sequence (Fast Low Angle Shot, FLASH) produces [Frahm J, Haase A, Matthaei D. Rapid three dimensional MR imaging using the FLASH technique. J Comput Assist Tomogr 1986; 10: 363-8.]. The sequence uses the sequence-inherent bright blood effect Venous lumens are hyperintense to fat and muscles. Due to the moderate T2 * - Weighting should reduce intracellularly enriched blood breakdown products or SPIOs be mapped. The following recording parameters were used: repetition time 54 ms; Echo time 18 ms; Flip angle 15 °; Block sealing thickness 80 mm, partitions 40; Layer thickness 2 mm; Image section 80 × 80 mm; Matrix 256 x 256; acquisition time 22:08 minutes; Pixel size 0.31 × 0.31 × 0.2 mm.

pathology

After the progressive MRI, the animals were killed by an overdose of xylazine. The caudal and cranial segments of the external jugular vein and the facial vein were prepared, fixed in 10% formalin for 24 hours and in five vessel cylinders each  of 3 mm in length. After embedding in paraffin, the caudal end became a cut each cylinder to a thickness of 3 µm and colored it with iron blue on Berlin.

image analysis

The histological sections, hard copies of the phlebographies and the MRI were performed evaluated a radiologist. The length of the thrombus in the external jugular vein were measured using phlebography. The histology served as the second Detection method of the thrombus, especially in the case of fresh, occluding thrombi are not washed around by contrast medium. However, it is because of the well known Shrinkage artifacts during histological processing only limited Length determinations suitable.

The length of the thrombus was then measured on the T1-weighted 3D Reconstructions of the MP-RAGE sequence were determined. The thrombus became visible defined if it was hyperintense and clearly distinguishable from the environment. At a incomplete presentation, only the visible thrombus portions were measured. Around an influence of the individual and thrombus-dependent variance of the Eliminating thrombus length was determined by the gold standard Thrombus length normalized to 1.0 and that in the 3D reconstruction of the MP-RAGE Sequence measured thrombus length given in proportions of 1.0, e.g. B. 0.4.

The T2 * weighted gradient echo sequence images were only subjected to a qualitative analysis. A representative layer was selected for the evaluation, which was clearly determined by the reference methods and was characteristic of the individual thrombus. If the thrombus could not be distinguished from the surrounding blood, the reference methods were used to locate the thrombus. The structure of the thrombus was classified as "homogeneous", "heterogeneously disordered" or "heterogeneous-concentric" (central hyperintense, peripheral hypointense). The signal intensity of the "homogeneous" thrombus, the leading signal intensity, ie that of the largest visible thrombus portion of the "heterogeneous-disordered" thorax or the signal intensity of the center in the case of a "heterogeneous-concentric" thrombus was classified on a 5-point scale from signal-free to signal-rich :

• 1. 1 signal-free, signal intensity such as compact bone or background
• 2. 2 low-signal, between 1 and 3  
• 3. 3 medium signal intensity, muscle isointense
• 4. 4 rich in signals, between 3 and 5
• 5. 5 very signal-rich, corresponding to a "light bulb effect"

The contrast range of the thrombus was based on the definitions of the Signal intensity determined as the difference between the minimum and maximum Signal intensity. The result is a 4-point scale.

In order to achieve a contrast enhancement of the wall and To examine the vascular environment, they were placed on a representative layer of the T2 * - weighted sequence classified in muscle isointense (normal) or low-signal.

statistics

The data were separated with the mean and the standard deviation Groups (controls, contrast media) and the thrombus age (days) are summarized or represented graphically (StatView 4.5, SAS Institute Inc., Cary, NC, USA). On Comparison of the first measurement with the course measurement was made with the t-test for Pair comparisons separately for groups D and K. The individual groups, e.g. B. D1 versus K1 were compared with the analysis of variance (ANOVA), for which one Significance check with the Fischer Test (Fischer's Protected Least Significant Difference, PLSD) was carried out. The level of significance was in all cases p <0.05.

Results animal model

Seven animals died during or after catheter embolization due to a shock reaction, presumably due to mechanical irritation of blood pressure regulators, by flushing the embolizate into the lungs or from an unexplained cause. These animals were excluded from the experiment. Thrombi could be produced in the remaining animals in all cases. The embolisate could be recognized in all cases by X-ray phlebography due to the lipiodol content ( Fig. 1 and 2). In 44 of the 50 cases, it was also rich in signals on the 3D MIP reconstruction of the MP-RAGE sequence. In the remaining six cases, it could be seen at least on the 2D source images of the MP-RAGE sequence.

X-ray phlebographically, the length of the thrombus in the external jugular vein of group K1 of the controls was 43 ± 8 mm and in group D1 of the contrast medium animals 36 ± 10 mm. The length decreased significantly in the study period and was 23 ± 10 and 11 ± 8 mm, respectively, in groups K9 and D9 ( Fig. 3). There was never a significantly different length between control and contrast medium groups of the same thrombus age.

Histologically there was a fresh thrombus consisting of groups K1 and D1 tightly packed erythrocytes (hemostasis) interspersed with trabeculae; in the groups K3 and D3 contribute to tightly packed hemostasis with central erythrocyte resolution individual animals; in groups K5 and D5 a resolution of the erythocytes in the Thrombus center in all cases (homogenization) as well as marginal immigration of mononuclear cells; in groups K7 and D7 increasing enforcement of the thrombus by mononuclear cells and the beginning of endothelialization singly sprouting capillaries; and in groups K9 and D9 the almost complete enforcement of the thrombus by mononuclear cells and capillaries the beginning of connective tissue formation.

MRI

Thrombi were visible on the 3D reconstructions of the T1-weighted MP-RAGE sequence as worm-like hyperintense structures ( Figs. 1 and 2). Occasionally, the course measurement of contrast medium animals also showed a slightly increased signal in larger neck vessels in the sense of a residual angiography effect. Regardless of the thrombus age, the relative thrombus length of the 3D MP-RAGE sequence of the initial MRI and phlebography was found to be 0.2 ± 0.3 in group K and 0.1 ± 0.3 in D. In the animals of group D (n = 25), there was a significant increase in the corresponding relative thrombus length from 0.1 ± 0.3 before contrast medium administration to 0.5 ± 0.5 24 hours afterwards over the 24-hour course, p = 0.001 ( Figs. 1, 2 and 4); but not for controls (only groups K1, K3 and K5), p = 0.34. The analysis of the individual day groups shows a dependence of the thrombus portion visible in the MP-RAGE sequence on the thrombus age ( Fig. 4).

At low thrombus age, this was 0.2 ± 0.3 to in groups K1, K3 and K5 0.4 ± 0.5, or older groups in groups K7 and K9 0.1 ± 0.1 and 0. According to Contrast agent administration showed a significant increase in the visible thrombus portion in the following groups: D3, from 0.6 ± 4 to 0.8 ± 0.4, D5, from 0.1 ± 0.1 to 1 ± 0.1, and D7 from 0 to 0.6 ± 0.4. No significant difference was found in contrast medium animals of groups D1 and D9.

The moderate T2 * -weighted FLASh sequence showed a considerable thrombus age-dependent variability in the structure, the signal intensity and the contrast range. A single example can be found in Fig. 5. In controls, the "heterogeneous-disordered" and "heterogeneous-concentric" thrombus structure predominated until day 7 ( Fig. 6). From the 7th day, a "homogeneous" thrombus structure dominates in the controls. The leading signal intensity of the thrombus in the control animals K1 was the "middle", ie isointense to the muscles, but with a clear spread ( Fig. 7). By the third third day, the signal intensity of the thrombus in the control animals tended to increase to "signal-rich" in order to drop to an intensity defined as "low-signal" by the 9th day. The contrast range of fresh thrombi was 1.6 in the K1 group and 0.4 in the K9 group ( Fig. 8). A total of only four animals in group K showed a signal reduction of the thrombus in the moderately T2 * -weighted FLASH sequence after contrast agent administration, which was evaluated as a possible contrast agent effect ( Fig. 5). A comparison of the initial and the course measurement showed no significant change in the thrombus structure ( Fig. 6), the thrombus signal ( Fig. 7) or the contrast range of the thrombus ( Fig. 8) in total or separately according to thrombus age.

Representative layers of the T2 * -weighted sequence were found in 6 out of 50 Animals already have a partial (n = 4) or circular (n = 2) signal deficiency of the wall and the immediate vascular environment. After the administration of contrast medium, all 6 Animals continue to be poor. The wall of 11 other animals was poor in signal and immediate vascular environment only in the follow-up; with 2 animals in the circular and at 9 in the partial expression. Of these 11 animals, one was one Control. The remaining 10 had received contrast media. Of these 10 contrast-enhanced studies were 4 of group D1 and 4 of D3, i.e. H. the  Contrast enhancement of the vein wall and immediate vascular environment was found preferably in young thrombus stages in 4 out of 5 cases.  

Detailed description of the pictures Fig. 1

X-ray phlebography (a) shows incomplete occlusion of the left vein jugularis externa three days after thrombus induction. The external jugular vein is preserved another inflow of medial, the facial vein. The outflow of the contrast medium z. T. over the opposite side. The embolisate (arrow) is radiopaque in phlebography (A). In 3D reconstructions of the MP-RAGE sequence, the position of the embolicate is before (b) and 24 hours after administration of superparamagnetic iron oxide (c) by an arrow marked. The cranial end of the visible part of the thrombus was Arrow head marked (b and c). After administration of contrast medium (c) in the MP-RAGE Sequence of additional cranial thrombus parts compared to the original MRI (b) visible. After the contrast medium had been administered, there was a complete agreement between Thrombus length in the MP-RAGE sequence and the gold standard. It is found also a residual angiography effect of the contrast medium in the other neck vessels.

Fig. 2

In X-ray phlebography (a) five days after embolization there is an embolisate (Arrow) to recognize. Individual parts of the thrombus are located in the cranial jugular vein internally washed around (a, arrowhead). There are ipsilateral collaterals (a, thinner Arrow). The thrombus is in the 3D reconstruction of the native MP-RAGE sequence not and the embolism is weakly recognizable (b, arrow). 24 hours after The administration of contrast medium (c) is the entire thrombus from the embolisate (arrow) to the cranial external jugular vein (arrowhead) and a lateral ear vein visible.

Fig. 3

Length of the thrombus in the external jugular vein. There is a decrease in the Thrombus length from day 1 to 9.  

Fig. 4

Correspondence of the relative thrombus length in 3D reconstruction of the T1 weighted MP-RAGE sequence and X-ray phlebography separately after Contrast medium animals (D) and controls (K) as well as thrombus age (1-9 days). It the outcome of the baseline and 24-hour MRI is shown. missing Agreement is expressed by the value 0, complete by 1. To Administration of DDM34 resulted in a significantly higher agreement in the MPR-RAGE sequence visible thrombus length in groups D3, D5 and D7, but not for fresh (D1) and organized thrombi (D9).

Fig. 5

In the moderately T2 * -weighted FLASH sequence there is before contrast agent administration (a) in the external jugular vein (arrow) a seven day old thrombus with an as "homogeneously" defined thrombus structure and an "average" signal intensity. To The administration of contrast medium is the thrombus "low-signal" (b, arrow). Significant signal loss the spine after SPIO administration (b, arrow head).

Fig. 6

Thrombus structure in the T2 * -weighted FLASH sequence in controls (K) and Animals before contrast agent administration (D) separated by days (1st-9th day). Until the 7th day the heterogeneous thrombus structure predominates heterogeneously disordered and heterogeneous concentric. The homogeneous thrombus structure dominates from the 7th day.

Fig. 7

Thrombus signal intensity in the T2 * -weighted FLASH sequence on a 5- Points scale from signal-free (1), signal-poor (2), muscle-iso-intensive (3), signal-rich (4) and very signal rich (5). There are groups of controls (K) and contrast medium animals shown separately according to the thrombus age. Both the starting and the Progressive MRI are shown - there were no controls on the K7 and K9 Plot-MRT.  

Fig. 8

Contrast range of experimental thrombi in the T2 * -weighted Gradient sequence. It can be seen that the contrast range for controls (K) as well as in contrast medium animals (D) decreases over a period of nine days, d. H. the Thrombus becomes more homogeneous. A significant change in the contrast range no group was found between the initial and follow-up examination.

Claims (8)

1. A method for the visual representation and diagnosis of thrombi, characterized in that the patient is first administered a particulate MR contrast medium, and after the contrast medium has been accumulated in the thrombus and / or the adjacent vessel wall and surroundings, a magnetic resonance tomogram is recorded. 2. The method according to claim 1, characterized in that the particulate Contrast agent contains superparamagnetic iron oxide particles (SPIO). 3. The method according to claim 2, characterized in that the superparamagnetic iron oxide particles ultra small superparamagnetic Iron oxide particles (USPIO) with an average particle diameter of less than 50 nm. 4. The method according to claim 1, characterized in that the MRI is recorded as a T1-weighted image. 5. The method according to claim 1, characterized in that the MRI is recorded as a T2- or T2 * -weighted image. 6. Use of particle suspensions for the production of contrast media for the Representation of thrombi using magnetic resonance imaging. 7. Use according to claim 6, characterized in that the Particle suspension is a suspension of superparamagnetic iron oxide particles is. 8. Use according to claim 7, characterized in that the superparamagnetic iron oxide particles ultra small superparamagnetic Are iron oxide particles with an average diameter of less than 50 nm.