WO2019175612A2

WO2019175612A2 - Methods and device to generate predictors for prognostic characterisation of coronary artery disease

Info

Publication number: WO2019175612A2
Application number: PCT/HU2019/050008
Authority: WO
Inventors: Zsolt KŐSZEGI
Original assignee: Debreceni Egyetem
Priority date: 2018-03-12
Filing date: 2019-03-12
Publication date: 2019-09-19
Also published as: EP3764884A2; WO2019175612A3; HUP1800091A2

Abstract

The present invention relates to methods and a device to generate one or more predictors for prognostic characterisation of coronary artery disease. The method according to the invention comprises the steps of providing (S200a) geometrical parameters describing spatial geometrical relations of a coronary artery (10) segment with a stenosis (12); providing (S200b) intracoronary pressure data characteristic of the coronary artery (10) segment with the stenosis (12); solving (S210) a flow dynamical problem formulated for the coronary artery (10) segment with the stenosis (12) for hyperaemic state and rest state of the coronary artery (10) by making use of said geometrical parameters and pressure data, thereby determining volumetric blood flow (Q) values; and generating (S220) at least one prognostic predictors characteristic of the coronary artery disease by combining the obtained volumetric blood flow (Q) values. The device according to the invention comprises an input data receiving unit for receiving input data, said data being geometrical parameters describing spatial geometrical relations of a coronary artery (10) segment with a stenosis (12) and intracoronary pressure data characteristic of the coronary artery (10) segment with the stenosis (12); a memory unit for storing the data, in data communication with the input data receiving unit; a processing unit in data communication with the memory unit and being configured to execute a computer code for performing steps to receive a flow dynamical problem related to the coronary artery (10) segment with the stenosis (12) and to solve said flow dynamical problem, said computer code being adapted to generate volumetric blood flow (Q) values by solving said flow dynamical problem for hyperaemic state and rest state of the coronary artery (10) by making use of said geometrical parameters and pressure data, and to generate at least one prognostic predictors characteristic of the coronary artery disease by combining the obtained volumetric blood flow (Q) values; an output unit for outputting the one or more prognostic predictors, in data communication with the processing unit.

Description

METHODS AND DEVICE TO GENERATE PREDICTORS FOR PROGNOSTIC

CHARACTERISATION OF CORONARY ARTERY DISEASE

The present invention relates to methods and a device to generate one or more predictors for prognostic characterisation of coronary artery disease.

In many cases, vascular diseases manifest themselves as decreased val- ues of blood flow within blood vessels as a consequence of obstructions of said vessels, e.g. the coronaries („atheroscleroticus stenosis”). In the clinical practice, to prognosticate the degree (severity) of stenoses of blood vessels, generally, two haemodynamic parameters, the so-called fractional flow reserve (FFR) and the coronary flow reserve (CFR) are used. The FFR value is defined as the ratio of two pressure values measured at maximum vasodilation; the first value is meas- ured in one of the coronaries at a location downstream of the stenosis, i.e. in a dis- tal position, the other value is the pressure prevailing/measured in the aorta (at a proximal position). By measuring said pressure ratio, the FFR value provides the extent of decrease in the blood flow within the vessel concerned due to its steno- sis. The CFR value is defined as the ratio of two flow rates measured in one of the coronaries; the first value is measured in said coronary at hyperaemia, the other is measured in the rest state of the same coronary - the CFR value basically charac- terizes the flow capacity of the system of the coronaries as a whole. As it is known, hyperaemia is achieved by making use of various vasodilators. Said pressure val- ues are measured (in a known manner) with pressure wires inserted into the coro- naries and/or the aorta by invasive techniques (intracoronary direct pressure measurement). Similarly, also invasively, the flow rates needed for the derivation of the CFR value can be directly obtained through e.g. intravascular ultrasonic in- spection (by using Doppler measuring wires) during catheterizing.

In the clinical practice, nowadays, the treatment type applicable to coronary disease due to arteriosclerosis (that is, one of a drug therapy, implantation of a stent or a coronary artery bypass operation) is primarily chosen on the basis of the CFR and FFR values which characterize the physiological state of the coronaries. To make decision, as independent prognostic means and in the form of an inde- pendent analysis, coronary dyeing technique (coronary angiography) revealing the anatomical state of the coronaries, as well as its graphical results are also taken into account.

The FFR value obtained for patients as discussed above describes the de- gree of obstruction of the coronaries and/or the flow restraint within the blood ves- sels based on a pressure measurement and/or pressure ratio. By experience, an FFR value less than 0.8 means a significant restraint of flow, and in harmony with clinical recommendations in force at present, it indicates the implantation of a stent to alleviate/cancel stenosis.

The CFR value obtained for patients as discussed above describes the de- gree of obstruction of the coronaries and/or the flow restraint within the blood ves- sels for the whole coronary system on the basis of a flow rate measurement and/or flow rate ratio. By experience, a CFR value less than 2.0 represents insufficient vascular supply to the myocardium and, thus, coronary stenosis. A drawback be- hind applying the CFR value for prognostic characterization is that the insertion of the Doppler measuring wire requires an invasive operation step, and/or the meas- urement itself is rather complicated technologically (i.e. to ensure a flow rate pro- file of high quality, the Doppler measuring wire has to be held in an appropriate position during the whole course of the measurement) and expensive as well.

To measure CFR, thermodilution measuring provides a possible alternative invasive technique which is based on measuring the temperature of the sensor ar- ranged in the pressure wire. Room temperature physiological saline solution is in- troduced into the coronary in the rest state and then during extensive vasodilata- tion. The CFR value is then deduced from the ratio of average transition periods obtained from the measured thermodilution curves by averaging. In the clinical practice, this technique, however, has not spread, since its application is rather elaborate, and as a consequence of the blood flow which flows away at lateral branches, the technique itself is not accurate enough.

For a portion of patients, the two measures of vessel obstruction suggested independently by the CFR and FFR values determined for respective patients dif- fer from one another, i.e. based on the two values, different treatment types would be recommended. This compromises the reliability of the applicability of CFR and FFR values for prognostic description, and also stresses the importance of the fact that it is indispensable to apply a complete description (in terms of both pressure and flow) of and to take into account the pathophysiology of stenosis when the type of treatment is selected.

As a consequence of the above-referred technical limitations of direct flow measurement, a great many experiments have been performed to determine the CFR derived from pressure (or CFRp) and the FFR simultaneously. Such process is disclosed e.g. by E. Shalman et at. in the paper titled„Pressu re-based simulta neous CFR and FFR measurements: understanding the physiology of a stenosed vessel’ (Comput. Biol. Med., September 2001 , 31 (5), pp. 353-363), or in U.S. Pa- tent No. 6,471 ,656 B1 , U.S. Published Patent Appl. Nos. US2003/0032886 A and US2014/0276137 A. In these documents, the authors assume a square depend- ence between the pressure drop (Dr) and the blood flow (Q) through a stenosis in the form of Dr ~ Q², from which the relation of

CFRp = Pav^~ Pdv

(1 ) Pa-Pd comes, wherein p_av and p_a stand for the values of proximal coronary pressure (measured in the aorta) for the hyperaemic and rest states, respectively, while p_dv and p_d represent the values of distal coronary pressure measured in the hyperae- mic and rest states, respectively.

Nevertheless, further investigations have shown that the model for pressure drop through a stenosis in the above approximation (1 ) systematically underesti- mates the CFR value, that is, CFR cannot be determined if merely the pressure is known, and thus it is important to take the properties/effects of the stenosis also into account to a larger degree. In particular, to obtain a more accurate CFR value, it is essential to make use of such a flow model that also incorporates the viscous losses due to the stenosis and the so-called separation losses with the relation of

Dr = f x Q + s x Q², (2a) wherein stands for the friction or viscosity coefficient, s represents the separation or discharge coefficient, Dr is the pressure drop through said stenosis and Q is the blood flow in the coronary artery. If the entire pressure drop is exclusively the con- sequence of said viscous losses, Dr = f x Q holds; if, however, the entire pressure drop is exclusively due to separation losses, Dr = s x Q² holds. Furthermore, ac- cording to various further studies, the ratio f/s, i.e. the ratio of the viscous to sepa- ration losses at a given stenosis cannot be determined merely by pressure meas- urement. In light of this, the CFR value characteristic of the stenosis (i.e. the ratio of hyperaemic blood flow to blood flow at rest in a stenosed coronary can be ex- pressed by the relation of

I hyperaemic Dr

_>pj^ <; hyperaemic Ap ₍2b₎

Dr at rest — ^“ Dr at rest ' ' for any combination of hyperaemic and rest pressure drops. That is, the CFR can be estimated by a whole interval instead of a single value. In literature, said inter- val is referred to as the CFR interval (pb-CFR); further details can be found in a paper by J.M. Ahn et at. titled“Fractional flow reserve and pressure-bounded cor onary flow reserve to predict outcomes in coronary artery disease " (Eur. Fleart J., 17 April 2017) and a paper by F.M. Zimmermann et al. titled“What can intracoro nary pressure measurements tell us about flow reserve? Pressure-Bounded coro nary flow reserve and example application to the randomized DEFER trial." (Cath- eter Cardiovasc. Interv., 15 March 2017). In summary, at present, the best esti- mate for the CFR value by pressure measurement on a patient is given by an in- terval and not a single value of high accuracy; this is a fact which limits significant- ly the applicability of CFR for prognostic description.

In view of the above, an object of the present invention, in its most general aspect, is to develop a technique suitable for generating one or more clinical pa- rameters, i.e. prognostic predictor(s), which characterize the condition and severity of coronary artery diseases and, in turn, classify said diseases, as well as provide means for the objective selection of the treatment type for coronary artery diseas- es. The present invention also aims at elaborating a complex technique, that is, essentially a method and a device, which allows to determine a CFR value accu- rate enough from the prognostic point of view. A yet further object of the present invention is to work out a method and a device, by means of which one or more further blood flow based volumetric value(s) could be obtained that are also suita- ble for being used as predictor(s). In a yet further aspect, the present invention aims at providing a technique, essentially a method and a device, that could be used in everyday clinical practice

- even in-situ - to alleviate or simplify the course of clinical decision-making as to the required action in the case of patients with stenosed coronary arteries. Such a decision should be made in the matter of e.g. implanting a stent or performing a bypass operation.

During the investigations performed, we have come to the astounding con- clusion that reliable prognostic predictors can be generated on the basis of apply- ing the basic laws of liquid flow without any flow rate calculations if a 3D recon- struction technique based on dyeing the coronary arteries (i.e. coronary angi- ography) that reveals the anatomical state of said arteries, or the data obtained in said 3D reconstruction by imaging are combined with pressure data obtained from an FFR measurement performed simultaneously with coronary angiography. Such a prognostic predictor is, especially, the CFR value characteristic of the physiolog- ical state of the coronary arteries, as well as further accurate volumetric values re- lated to the actual blood flow prevailing in coronary arteries.

The above objects related to the provision of a method to generate one or more prognostic predictors are achieved by methods according to claims 1 and 10. Preferred further variants of the methods according to the invention are set forth in claims 2 to 9 and claims 11 to 16. The objects related to the provision of a device for generating said one or more prognostic predictors are achieved by the device in accordance with claim 17. Preferred further embodiments of said device are de- fined by claims 18 to 21. Furthermore, the above objects are achieved by comput- er program products in accordance with claims 22 and 23.

In what follows, the invention is discussed in more detail with reference to the accompanying drawings, wherein

- Figure 1 illustrates the geometrical model being part of the 3D reconstruction used in the method according to the invention;

- Figure 2 shows a schematic block diagram of the device according to the in- vention, i.e. the flow-pressure-capacity module, suitable for obtaining at least one clinical prognostic predictor; - Figure 3 presents the flow chart of the method according to the present inven- tion to obtain at least one clinical prognostic predictor;

- Figure 4 is a possible 3D graphical representation of the prognostic predictors to be generated by means of a device illustrated in Figure 2 based on real pa- tient related measurement data; and

- Figure 5 is a possible further planar graphical representation of the prognostic predictors to be generated by means of a device illustrated in Figure 2 based on real patient related measurement data, wherein various portions of the dia- gram match to the clinical classification used in literature.

Figure 1 illustrates a geometrical model 20 for a coronary artery 10 with a stenosis 12 used in the spatial (that is, 3D) reconstruction which forms a basis for the method and device to generate one or more clinical prognostic predictors ac- cording to the invention. The data required/used for the 3D reconstruction of the coronary artery 10 are obtained from invasive coronary angiography performed on a patient in accordance with a protocol accepted in the clinical practice. Parame- ters of the model 20 approximating the coronary artery 10 are determined on the basis of a real 3D image of said coronary artery 10. By making use of imaging 3D reconstruction, the actual anatomical state of the coronary artery 10 gets tested and then taken into consideration in the model 20 through the parameters thus de- rived. Said 3D reconstruction can be performed through processing the graphical data of coronary angiography by means of suitable pieces of software commercial- ly available (see e.g. the software QAngio XA Research Edition 1.0 developed by Medis Specials BV). Said graphical data are preferably X-ray images. In particular, said graphical data are provided in the form of two X-ray images which have been taken in directions closing preferably an angle of at least 25° with one another and is evaluable according to the clinical practice. As is apparent to a skilled person in the art, here, any images prepared by (e.g. non-invasive) imaging techniques ac- cepted in the clinical practice for imaging coronary arteries can equally be used; in case of such images, 3D reconstruction can be performed by appropriate pieces of software tailored to the images concerned.

During said 3D reconstruction, an image of the selected coronary artery 10 is constructed from a proximal position 18 corresponding to the coronary aorta to a distal position 16 (not shown in the Figures) corresponding to the location of the pressure wire. In a next step, parameters of the simplified, flow dynamics-based geometrical model 20 corresponding to said segment of the coronary artery 10 are obtained by performing measurements in the previously obtained 3D image. In terms of the pressure drop due to laminar flow, the geometrical model 20 consists of a first straight rigid-wall tube segment with a flow cross-sectional (lumen) area A_P and a length L_P, a second straight rigid-wall tube segment with a flow cross- sectional area As and a length l__s which continuously connects to the first tube segment and represents the coronary artery 10 with the stenosis 12, and a third straight rigid-wall tube segment with a flow cross-sectional area Ad and a length Ld which continuously connects to the second tube segment and represent the distal position 16. In terms of dynamics, i.e. the blood flow becomes turbulent due to the stenosis (flow separation), the model 20 is described by the smallest flow cross- sectional area AMLA of said stenosis 12 measured in the 3D image. Pressure data required to the flow dynamical characterisation of the coronary artery 10, i.e. val- ues of the proximal coronary pressure p_a and the distal coronary pressure pd in the hyperaemic and rest states are provided by the pressure data measured at the proximal position 18 and the distal position 16 according to a known clinical proto- col with the pressure wire introduced into the coronary artery 10 during cardiac catheterizing of the patient.

Then, the thus obtained geometrical and flow dynamical parameters are used to solve the Hagen-Poiseuille and Borda-Carnot equations, i.e. the flow dy- namical problem, that describe the liquid flow within the combined unit of said tube segments according to the model 20 for the volumetric blood flow (Q) in the hy- peraemic and rest states in the form of

Apres = Api(Q) + Ap2(Q) + Ap3(Q) + Ap₄(Q) . (3)

In this latter equation Ap_res = p_a - pd is the pressure difference measured in the coronary artery 10 under study (that is, the intracoronary pressure drop), and

Dri = 8phc _r rC Q is the pressure drop at the proximal position 18 due to internal friction (viscosity) of the blood flow, Dr2 = 8m\xL_s/As X Q is the pressure drop at the location of the stenosis 12 due to viscosity of the blood flow (this, practically, represents the pressure drop component due to the laminar flow of blood),

Dr3 = r^c(1/^_MΐL

'^s the pressure drop at the location of the stenosis 12 (i.e. the discharge loss component) due to delamination of the flowing layers and build-up of a turbulent flow as a consequence of the stenosed coronary artery 10, and

Dr₄ = 8 r\xL_dIA_d xQ is the pressure drop at the distal position 16 due to viscosity of the blood flow; in the above equations h and p stand for the viscosity and the volume density, respectively, of the blood of the patient.

The ratio of the blood flows Qactuai and Qrest obtained when equation (3) is solved for the hyperaemic and the rest states, respectively, is a CFR value -from now on, referred to as the CFRp-3D value - calculated by exploiting the intracoro- nary pressure and 3D coronary anatomical data. One can immediately realize that this value is a single value which is characteristic of the condition and severity of the patient’s coronary artery disease and is based on patient pressure values ac- quired by means of the clinically accepted FFR measurement. Thus, compared to the CFR and pb-CFR values used nowadays, said CFRp-3D value is considered to be a much more accurate and reliably applicable prognostic predictor.

Moreover, the calculated CFRp-3D - as the traditional indicator for con- striction of the coronary flow reserve - can be projected onto the pressure-flow re- lation of coronary circulation and, thus, it supplies information on individual patho- physiology of the patient, and can preferentially be used to conclude on an epicar- dial or microvascular excess of the vascular disease, as well as to prognosticate it.

Based on the above relations and using the measured value of the intra- coronary pressure drop in the hyperaemic state and the geometrical data obtained from 3D reconstruction, a further prognostic predictor can be derived as the maxi- mal flow velocity, Velocity_p-3D_vd, calculated for uniform (stationary) flow. With- out going into deeper theoretical explanation, this prognostic predictor can be used to describe the ischaemic behaviour of the stenosed coronary artery under study and corresponds practically to the parameter suggested previously in literature for describing the maximal coronary flow rate (see ..maximal coronary flow capacity”; CFC); for further details, we refer here to e.g. the paper by T. P. van de Hoef et al. titled“Physiological basis and long-term clinical outcome of discordance between fractional flow reserve and coronary flow velocity reserve in coronary stenoses of intermediate severity” (Circ. Cardiovasc. Interv. 2014(7), pp. 301 -311 ).

By applying the inventive technique, pairs of values can be derived which describe the condition of a patient under study. According to extensive investiga- tion performed in the field, the thus obtained patient-specific predictors, i.e. the various pairs of values of CFRp-3D and Velocity_p-3D_vd are suitable for describ- ing, prognosticating and classifying the degree of flow reserve, the ischaemic be- haviour and the susceptibility to cardiac infarct of the patient’s myocardium.

In light of the aforementioned, the inventive technique also allows quantifi- cation, from patient to patient, of the degree of discharge losses Dr3 characterising the flow delamination due to ..turbulent” flow which builds up as a consequence of the stenosis 12. Since, according to the literature, this is closely related to the pro gress of atherosclerotic plaques in the coronary arteries in time, it is preferred to study and describe the benign laminar component of the blood flow separately from the pathologic ..turbulent” flow with flow delamination. The pressure-flow rela- tion obtained from the FFR measurement and the geometrical parameters of the 3D reconstruction can also be used, in combination, as prognostic means charac- teristic of the condition and severity of the patient’s coronary artery disease. Fur- thermore, within flow limitations of a given coronary artery, said data also provide information as to the two kinds of flow resistance, i.e. the pressure losses (or fric- tion losses; Dri, Dr2 and Dr₄) due to laminar flow and the pressure losses (Dr3) due to turbulent flow which builds up as the consequence of flow delamination. Af ter calculating CFRp-3D as discussed above, the pressure-flow relation is pre- sented in a diagram which presents flow ratios (”CFRp-3D”, Qactuai/Qrest) and pres- sure ratios (”FFR”, pd/p_a). In this way, the pressure losses due to flow delamination get represented in an overall manner (see Figure 5). The pressure ratios have to be construed only between the flow value at rest and the CFR value, thus specific flow conditions characteristic of the patient’s coronary artery under study are ob- tained. The pressure decrease due to flow delamination can be characterised in the patient-specific flow range by integrating over the pressure ratios. That is, the

relation allows to define a yet further prognostic predictor, the flow separation re- sistance index (FSi) which is presented graphically in Figure 5 as domain 410 (to be yet discussed later). This latter predictor represents a dimensionless parameter belonging to unique vessel segments, and hence, it is advantageous in the char- acterisation of all stenoses. Therefore, it can also be applied in the therapeutic ap- proach of stenoses (e.g. at decision-making on the type of treatment to be ap- plied/applicable).

In particular, the flow ratios (at rest: Qaactuai/Qrest = 1 , in hyperaemic state: Qactual/Qrest ⁼ CFR) of the coronary arteries obtained by haemodynamic calculation using 3D reconstruction data and intracoronary pressure are presented over the horizontal axis of Figure 5, while the pressure ratios (pd/pa) are presented over the vertical axis of Figure 5 (here, for said pressure ratios, the basal pd/p_a and pd/p_a = FFR holds at rest and in hyperaemic states [vasodilation], respectively). Now, in tegrating pd/p_a - in accordance with relation (4) - over an interval defined by the flow ratios at rest and at CFR, a patient and vessel specific value of the decrease in the pressure ratio is obtained. In Figure 5, the total area of the flow-pressure- capacity diagram is four units, from which those FSi values (see domain 410) can be interpreted in terms of physiology which range from 0 and about 1. According to the extensive investigations performed, these FSi values can be used as a predic- tor characteristic of patient condition to describe (quantitatively) and prognosticate epicardial lesion/diseased condition of the myocardium.

A possible preferred graphical representation of the prognostic predictors generated by the inventive technique is shown in Figure 4. Flere, the parameters CFRp-3D, Velocity_p-3D_vd and FFR form the independent orthogonal axes x, y and z, respectively, of a Cartesian coordinate system. The thus obtained diagram 300 is the already mentioned flow-pressure-capacity (FPC) diagram which is the output of a flow-pressure-capacity module to be discussed below in detail.

In the clinical practice, diagram 300 can be used to accurately describe and/or classify the severity of a diseased condition of the patient. It should be here noted, that by means of domains 301 , 302, 303, 304 covering the plane (xz-plane) defined by the CFRp-3D and FFR axes of said diagram 300, individual species of a patient population, as well as the patient-specific values can be classified. In the diagram 300, the clinically validated threshold values are represented by a solid line; these are 0.8 and 2.4 for the FFR and CFR, respectively, as is illustrated by lines 322 and 321 in the diagram 300. In the plane (xy-plane) defined by the CFRp-3D and Velocity_p-3D_vd axes of said diagram 300, the coronary flow ca- pacity can be evaluated by combining physiologically the absolute flow and CFR; individual domains 308, 306, 305 forming the basis for classification connect to each other along (dashed) lines 327 and 326 and covers thereby the whole plane defined by the CFRp-3D and Velocity_p-3D_vd axes. Similarly, individual domains 308, 301 , 303, 304 of diagram 300 forming the basis for classification connect con- tinuously to each other along lines 328 and 329 in the plane defined by the Veloci- ty_p-3D_vd and FFR axes in the diagram 300.

Besides diagram 300 with three coordinate axes, the pieces of information provided by its xz- and xy-planes can be summarized into a single planar diagram 400 shown in Figure 5. The above-discussed CFR-FFR relation occupies the top left portion of the combined diagram 400 (see domains 401 , 402, 403, 404 in dia- gram 400), while the calculated maximal coronary flow rate-CFR relation occupies the bottom right portion of said combined diagram 400 (see domains 405, 406, 407, 408 in diagram 400). These portions of diagram 400 are separated by a solid line 409 which represents a possible physiological limit. In this way, characteristics of the coronary artery (in harmony with the mapping used in literature) are pre- sented by two codes: a first code represents the CFR-FFR relation from a point of view whether a focal epicardial disease or a diffuse disease also influencing the microvasculature dominates in case of the patient under study; said second code indicates ischaemia of the investigated coronary artery. By means of the patient- and vessel-specific relation illustrating the relation of flow and pressure ratios (i.e. the yz-plane in diagram 300), diagram 400 also contains the flow separation re- sistance index FSi in the form of domain 410 which predicts a prognosis of epicar- dial stenosis. In light of values, pairs of values of the above-discussed prognostic predictors for a patient under study, based on diagrams 300, 400 conclusions can be drawn simply and rapidly as to the severity of the patient’s diseased condition: pairs of values (CFRp-3D; FFR) in domain 401 refer primarily to microvascular disease, pairs of values (CFRp-3D; FFR) in domain 402 refer univocally to normal condition, pairs of values (CFRp-3D; FFR) in domain 403 refer univocally to dis- eased condition, while pairs of values (CFRp-3D; FFR) in domain 404 refer to fo- cal epicardial (but non-significant) disease. Furthermore, pairs of values (CFRp- 3D; Velocity_p-3D_vd) in domain 405 refer to normal coronary flow, pairs of values (CFRp-3D; Velocity_p-3D_vd) in domain 406 refer to slightly decreased coronary flow, pairs of values (CFRp-3D; Velocity_p-3D_vd) in domain 407 refer to interme- diately decreased coronary flow, and pairs of values (CFRp-3D; Velocity_p-3D_vd) in domain 408 markedly refer to ischaemic disease of the coronary artery.

In what follows, a preferred embodiment of the flow-pressure-capacity module is discussed in detail with reference to Figures 2 and 3; said module is a possible practical realization of the technique, that is, the method and the device according to the present invention

Figure 2 illustrates a block diagram of a preferred embodiment of the device according to the invention, in the form of a flow-pressure-capacity module 100. The module 100 has an input data receiving unit 110 for receiving input data, a memory unit 120 for storing data, a processing unit 130, an output unit 140 generating one or more predictors for the prognostic description of coronary artery disease, and - optionally - a display unit 150 for displaying the one or more predictors in a format selected by a user, or is connected to such a display unit 150. Flere, the input data receiving unit 110 is in data communication with said memory unit 120, the memory unit 120 is in data communication with said processing unit 130, the processing unit 130 is in data communication with said output unit 140, and the output unit 140 is in data communication with said display unit 150. The input data receiving unit 110 is configured to receive data 111 obtained by 3D reconstruction on a patient’s coronary artery stenosis to be characterized on the one hand, and further measurement data 112 acquired during examination of the patient on the other hand. Data 111 are preferably geometric data describing the coronary artery 10 with the stenosis 12 (see Figure 1 ) derived by processing invasive coronary angiographic data, in conformity with the geometric parameters required for a complete specification of the geometric model 20 of said coronary artery 10 segment shown in Figure 1. The invasive coronary angiographic data are preferably X-ray images captured by a suitable X- ray device, or versions of X-ray images that have been previously subjected to software pre-processing. Data 112 are conventional FFR pressure data measured according to valid clinical protocols by pressure wire(s) commonly used in the clinical practice. The coronary angiography data and the FFR pressure data are recorded/measured simultaneously during examination/treatment (cardiac catheterizing) of the patient.

It is obvious to a person skilled in the art, that data 111 may also be formed by pieces of information acquired by using other imaging techniques accepted in the clinical practice which are suitable for producing images of the coronary arteries of the heart (e.g. non-invasive CT-angiography), optionally after the necessary pre-processing.

The memory unit 120 stores the data 111 , 112 input into the module 100 via the input data receiving unit 110 and may be provided in the form of any suitable data storage media, such as e.g. hard disk drives, a flash memories, optical data storage units or any other devices suitable for data storage known to a person skilled in the art.

The processing unit 130, preferably a processor, is used for deriving one or more prognostic predictors characterizing a coronary artery stenosis defined by the data 111 , 112. To this end, it comprises - in the form of a corresponding computer code - the flow dynamical relation(s) necessary to determine the one or more predictors. In this case, these are the equation (p_a - p_d)^measured = Dri(O) + Ap2(Q) + Ap3(Q) + Dr₄(0) in harmony with relation (3) setting the flow dynamical problem, the definition according to relation (4), as well as the steps for solving said equations and the order of the steps. Based on the data 111 , 112 and using the flow dynamical relation(s), the processing unit 130 determines, by calculation, the sought prognostic predictors, in particular, the CFR value (CFRp-3D value) calculated on the basis of the 3D coronary anatomy data, the maximal flow velocity (Velocity_p-3D_vd) calculated for uniform flow, and the flow separation (resistance) index FSi. The processing unit 130 may also read, beside data 111 , 112, a computer code comprising the flow dynamical problem, the steps for solving it, as well as the order of the steps from the memory unit 120. Upon determining said predictors, the predictors and the measured FFR value of the patient are transferred by the output unit 140 in the form of output data 113, 114, 116 and 115 for further use, e.g. for further processing, to e.g. a clinical information system . A potential use of the output data 113, 114, 116 and 115 takes preferably place by graphically displaying at least a part of said data, and thus, using them for example as means for assisting in the medical decision-making as to the required intervention in the case of the patient.

According to this latter use, a preferred embodiment of the module 100 also comprises a display unit 150. The display unit 150 is preferably a screen, but it may also be the output of a conventional printing device, i.e. one or more printed pages. The display unit 150 presents, preferably visualizes/displays data 113, 114 and 116 together with data 115 representing the measured FFR value in a predefined format which corresponds to the predictor determined according to the aforementioned; in particular, in the form of diagrams 300, 400 or at least one of them. It is obvious to a person skilled in the art that the predictors may be presented in other forms, too.

It is also obvious to a person skilled in the art, that data 111 may also be formed by raw (i.e. not yet processed) coronary angiographic X-ray images. In such a case, the software, that performs determination/derivation of the geometric parameters necessary to establish the geometric model 20 for the given segment of the coronary artery, is also stored in the module 100, in the memory unit 120 itself, and it is executed by the processing unit 130 itself before generating the one or more predictors.

The flow-pressure-capacity module 100 according to the invention may also be implemented as a separate unit, which can be connected in a suitable manner to a clinical information system storing the necessary patient data. Alternatively, the module 100 can be configured as a module that can be integrated into the im- aging device itself, preferably into e.g. the X-ray device, expanding its functions, e.g. in the form of a circuit board designed and built specifically to this purpose. Moreover, the module 100 can alternatively be implemented as a distributed sys- tem, wherein one or more of the subunits is/are located/formed/disposed at differ- ent physical locations and - as necessary - communicate with each other via suit- able communication channels (LAN, WAN, internet, etc.).

Figure 3 is a flow chart of a preferred variant of the method aspect of the inventive technique. According to this, geometric parameters characteristic of a segment of the coronary artery 10 with a stenosis 12 are provided (step S200a), preferably by means of 3D reconstruction based on coronary angiography per- formed on a patient, or any other suitable imaging process. Simultaneously with coronary angiography, intracoronary pressure data characteristic of said segment of the coronary artery 10 with the stenosis 12 are also provided (step S200b), preferably by means of at least one pressure wire introduced into the coronary ar- tery 10 at cardiac catheterization of the patient, and preferably by means of intra- coronary pressure measurement performed in accordance with a valid protocol for FFR measurements applied in the clinical practice. Then, using the available geo- metrical parameters and intracoronary pressure data, a flow dynamical problem defined for the segment of the coronary artery 10 with the stenosis 12 is solved (step S210), thereby determining flow conditions prevailing within the segment of the coronary artery 10 concerned. Here, the flow dynamical problem is defined by a flow dynamical equation in the form of relation (4) for the patient’s hyperaemic and rest states. Based on the flow data obtained by solving said flow dynamical equation, or rather the flow dynamical problem tailored to the segment of the coro- nary artery 10 with the stenosis 12, in particular, the values of the actual (hyper- aemic) blood flow and the blood flow at rest, one or more prognostic predictors describing coronary artery disease are generated as discussed above in the form of pre-defined combinations (see above) of said flow data (step S220). Then, by presenting, preferably graphically, at least one of the obtained predictors (step S230), e.g. a cardiologist treating the patient is provided with means for assisting him/her in the decision-making regarding the type of treatment of the patient’s cor- onary artery disease.

The method according to the invention also includes the period of time nec- essary to perform the 3D reconstruction after (at least) two coronarographic imag- es providing enough information on the degradation of the coronary artery 10 have been selected; the pressure difference can be determined from the morphological data by means of relation (3) at once, i.e. this operation requires essentially no fur- ther time (in comparison with the time period necessary for the 3D reconstruction). An experienced examining operator completes said 3D reconstruction in about a couple of minutes. The technique proposed here and the one or more prognostic predictors generated by the technique, thus, form an easily available and really rapid tool in the course of decision-making for the cardiologist who performs an in- tervention during e.g. operation.

Moreover, compared to pure FFR measurements, the present invention is capable of presenting much broader and detailed data: the consequence of an ep- icardial stenosis can be determined together with the complex status of the micro- vascular condition, the endothelial function can also be evaluated from the rela- tionship between the CFR and the FFR (the CFR/FFR ratio is the indicator for the microvascular function). Furthermore, the present inventive technique enables to directly measure the resistance of the myocardium from the ratio of the distal in- tracoronary pressure to the absolute volumetric flow measurable in the hyperae- mic state. Based on the complex evaluation applied, the clinical decision-making as to a drug therapy or the percutaneous/surgical revascularization gets more pre- cise. The overall flow-pressure-capacity diagram is also suitable for predicting the FFR and CFR values achievable by percutaneous coronary intervention. Namely, implanting a stent will cancel the vessel’s square resistance factor (i.e. component Dr3), and hence, the pressure and flow achievable in surgical intervention can be demonstrated through a virtual stent implantation procedure performed as part of the inventive technique. Real effect of said intervention will be provided by an in- vasive measurement carried out after implanting said stent; if the result of a control FFR measurement is lower than what was previously predicted, an insufficient de- crease in resistance may indicate the necessity of further examinations: by intra- vascular imaging the expansion of the stent (IVUS or OCT) can be checked, or a decision can be made on the necessity of a further stent implantation or a post di- latation.

Moreover, in the post-operative follow-up of patients, a repeated clinical ex- amination would precisely demonstrate the efficiency of the therapy for both the epicardial vessels and the microvasculature. According to an assumption based on data published in literature, the flow separation resistance index FSi is a prog- nostic predictor that is much more efficient than the FFR in predicting the progres- sion of native coronary diseases, especially in case of diffuse vascular diseases. Hence, this approach may serve a starting point for the experimental examinations and prospective clinical research studying the effects of various therapeutic tech- niques of the coronary disease.

Claims

1. A method to generate one or more predictors for prognostic characterisa- tion of coronary artery disease, comprising the steps of

providing (S200a) geometrical parameters describing spatial geometrical relations of a coronary artery (10) segment with a stenosis (12);

providing (S200b) intracoronary pressure data characteristic of the coronary artery (10) segment with the stenosis (12);

formulating a flow dynamical problem related to the coronary artery (10) segment with the stenosis (12);

solving (S210) said flow dynamical problem for hyperaemic state and rest state of the coronary artery (10) by making use of said geometrical parameters and pressure data, thereby determining volumetric blood flow (Q) values;

generating (S220) at least one prognostic predictors characteristic of the coronary artery disease by combining the obtained volumetric blood flow (Q) val- ues.

2. The method according to claim 1 , further comprising generating the ge- ometrical parameters by 3D reconstruction of data obtained by an imaging process performed on the coronary artery (10) segment with the stenosis (12).

3. The method according to claim 2, further comprising performing coronary angiography as the imaging process.

4. The method according to claim 2 or 3, further comprising capturing at least two images on said coronary (10) segment from various camera positions forming a given angle with one another in the imaging process performed on the coronary artery (10) segment with the stenosis (12), and performing said 3D re- construction on the basis of the images captured from the various directions.

5. The method according to any of claims 1 to 4, further comprising gener- ating the intracoronary pressure data by intracoronary pressure measurement per- formed in said rest state and in said hyperaemic state of the coronary artery (10) segment with the stenosis (12), the intracoronary pressure measurement being performed according to a clinical protocol applied for measuring fractional flow re- serve (FFR) known per se.

6. The method according to any of claims 2 to 5, further comprising per- forming the imaging process and the intracoronary pressure measurement simul- taneously.

7. The method according to any of claims 1 to 6, further comprising drawing one or more conclusions based on the at least one prognostic predictor regarding a condition of the coronary artery disease, and optionally, classifying the coronary artery disease on the basis of the one or more conclusions with regard to severity.

8. The method according to any of claims 1 to 7, further comprising formu- lating the flow dynamical problem as

p_a - pd = Api(Q) + Ap2(Q) + Ap3(Q) + Ap₄(Q) ,

wherein p_a is a pressure value measured in said coronary artery (10) upstream of the stenosis (12) and pd is a pressure value measured in said coronary artery (10) downstream of the stenosis (12), and

Dri = 8phc _r rC Q is a pressure drop due to internal friction of the blood flow (Q) upstream of the stenosis (12),

Dr2 = dphcZ^ ^cO is a pressure drop due to internal friction of the blood flow (Q) at the stenosis (12),

Dr3 '^{s a} pressure drop due to flow separation at

the stenosis (12), and

Dr₄ = 8 r\xL_dIA_d xQ is a pressure drop due to internal friction of the blood flow (Q) downstream of the stenosis (12), and wherein

h and p stand for the viscosity and the volume density, respectively, of blood, and the flow cross-sectional area A_P, the length L_P, the flow cross-sectional area As, the length l__s, the flow cross-sectional area Ad, the length Ld and the flow cross-sectional area AMLA are the geometrical parameters characteristic of the cor- onary artery (10) segment with the stenosis (12) obtained by 3D reconstruction.

9. The method according to any of claims 1 to 8, further comprising select- ing the at least one prognostic predictor from a group consisting of the coronary flow reserve (CFRp-3D) defined as a ratio of a blood flow (Qactuai) in the hyperae- mic state to a blood flow (Qrest) in the rest state obtained by solving the flow dy- namical problem, calculated with coronary 3D anatomical data, the maximal flow velocity calculated for uniform flow (Velocity_p-3D_vd) and the flow separation re- sistance index (FSi).

10. A method to generate one or more predictors for prognostic characteri- sation of coronary artery disease, comprising the steps of

acquiring graphical data from a coronary artery (10) segment with a steno- sis (12) by an imaging process performed on the coronary artery (10) segment with the stenosis (12) and generating (S200a) geometrical parameters describing spatial geometrical relations of the coronary artery (10) segment with the stenosis (12) by 3D reconstruction of the thus obtained graphical data;

acquiring (S200b) intracoronary pressure data characteristic of the coronary artery (10) segment with the stenosis (12) simultaneously with performing the im- aging process;

generating (S220) at least one prognostic predictors characteristic of the coronary artery disease by combining the obtained volumetric blood flow (Q) val- ues;

presenting (S230) the one or more prognostic predictor graphically.

11. The method according to claim 10, further comprising performing coro- nary angiography as the imaging process.

12. The method according to claim 10 or 11 , further comprising capturing at least two images on said coronary (10) segment from various camera positions forming a given angle with one another in the imaging process performed on the coronary artery (10) segment with the stenosis (12), and performing said 3D re- construction on the basis of the images captured from the various directions.

13. The method according to any of claims 1 to 4, further comprising gen- erating the intracoronary pressure data by intracoronary pressure measurement performed in said rest state and in said hyperaemic state of the coronary artery (10) segment with the stenosis (12), the intracoronary pressure measurement be- ing performed according to a clinical protocol applied for measuring fractional flow reserve (FFR) known perse.

14. The method according to any of claims 10 to 13, further comprising formulating the flow dynamical problem as

p_a - pd = Api(Q) + Ap2(Q) + Ap3(Q) + Ap₄(Q) ,

Dri = dphc _r rC Q is a pressure drop due to internal friction of the blood flow (Q) upstream of the stenosis (12),

Dr3 '^{s a} pressure drop due to flow separation at

the stenosis (12), and

Dr₄ = 8 r\xL_dIA_dx Q is a pressure drop due to internal friction of the blood flow (Q) downstream of the stenosis (12), and wherein

15. The method according to any of claims 10 to 14, further comprising drawing one or more conclusions based on the at least one prognostic predictor regarding a condition of the coronary artery disease, and optionally, classifying the coronary artery disease on the basis of the one or more conclusions with regard to severity.

16. The method according to any of claims 10 to 15, further comprising se- lecting the at least one prognostic predictor from a group consisting of the coro- nary flow reserve (CFRp-3D) defined as a ratio of a blood flow (Qactuai) in the hy- peraemic state to a blood flow (Qrest) in the rest state obtained by solving the flow dynamical problem, calculated with coronary 3D anatomical data, the maximal flow velocity calculated for uniform flow (Velocity_p-3D_vd) and the flow separation re- sistance index (FSi).

17. A device to generate one or more predictors for prognostic characteri- sation of coronary artery disease, comprising

- an input data receiving unit (110) for receiving input data (111 , 112), said data (111 , 112) being geometrical parameters describing spatial geometrical relations of a coronary artery (10) segment with a stenosis (12) and intracoronary pressure data characteristic of the coronary artery (10) segment with the stenosis (12);

- a memory unit (120) for storing the data (111 , 112), in data communication with the input data receiving unit (110);

- a processing unit (130) in data communication with the memory unit (120) and being configured to execute a computer code for performing steps to receive a flow dynamical problem related to the coronary artery (10) segment with the ste- nosis (12) and to solve said flow dynamical problem, said computer code being adapted to generate volumetric blood flow (Q) values by solving said flow dynam- ical problem for hyperaemic state and rest state of the coronary artery (10) by making use of said geometrical parameters and pressure data, and to generate at least one prognostic predictors characteristic of the coronary artery disease by combining the obtained volumetric blood flow (Q) values;

- an output unit (140) for outputting the one or more prognostic predictors, in data communication with the processing unit (130).

18. The device according to claim 17, further comprising a display unit (150) connected to the output unit (140) for presenting the one or more prognostic predictors.

19. The device according to claim 17 or 18, wherein the input data receiving unit (110), the memory unit (120), the processing unit (130) and the output unit (140) are arranged to form a single module (100), said module (100) being con- nectable to an imaging apparatus for deriving the geometrical data through imag- ing the coronary artery (10) segment with the stenosis (12).

20. The device according to claim 17 or 18, wherein at least one of the in- put data receiving unit (110), the memory unit (120), the processing unit (130) and the output unit (140) is configured to form a subunit in a distributed system.

21 . The device according to any of claims 17 to 20, wherein the flow dy- namical problem being defined as

p_a - pd = Api(Q) + Ap2(Q) + Ap3(Q) + Ap₄(Q) ,

Dri = dphc r rC Q is a pressure drop due to internal friction of the blood flow (Q) upstream of the stenosis (12),

Dr2 = dphcZ^ ^ cO is a pressure drop due to internal friction of the blood flow (Q) at the stenosis (12),

Dr3 = '^{s a} pressure drop due to flow separation at

the stenosis (12), and

Dr₄ = 8 r\xL_dIA_dx Q is a pressure drop due to internal friction of the blood flow (Q) downstream of the stenosis (12), and wherein h and p stand for the viscosity and the volume density, respectively, of blood, and the flow cross-sectional area A_P, the length L_P, the flow cross-sectional area As, the length l__s, the flow cross-sectional area Ad, the length Ld and the flow cross-sectional area AMLA are the geometrical parameters characteristic of the spa- tial geometrical relations of the coronary artery (10) segment with the stenosis (12) obtained by 3D reconstruction.

22. A computer program product comprising commands stored on a data storage means readable by the device according to any of claims 17 to 21 , said commands, when loaded into a processing unit (130) of said device, causing said device to execute the steps (S200-S220) of the method according to any of claims 1 to 9.

23. A computer program product comprising commands stored on a data storage means readable by the device according to any of claims 17 to 21 , said commands, when loaded into a processing unit (130) of said device, causing said device to execute the steps (S200-S230) of the method according to any of claims

10 to 16.