09/660,195 (now U. S. Patent No.), filed September 12,2000.

These applications are commonly-owned and incorporated by reference in their entirety.

Field of Invention This inventions relates to methods and apparatus for treatment of congestive heart failure (CHF) by removal of excessive fluids. such as water. In particular, the invention relates to an extracorporeal circuit with minimized blood residence time.

Background of the Invention Congestive Heart Failure (CHF) is the only form of heart disease still increasing in frequency. According to the American Heart Association, CHF is the"Disease of the Next NIillennium". CHF is a condition that occurs when the heart becomes damaged and reduces blood flow to the organs of the body. If blood flow decreases sufficiently, kidney function becomes impaired and results in fluid retention, abnormal hormone secretion and increased constriction of blood vessels. The fluid overload and associated clinical symptoms resulting from these physiologic changes are the predominant cause for excessive hospital admissions, terrible quality of life and overwhelming costs to the health care system due to CHF.

One possible method for removal of excessive fluid is mechanical fluid removal employing an extracorporeal circuit with a hemofilter. This method is especially useful if CHF is in its final stage and drug treatment is no longer

efficient. Extracorporeal fluid removal is a common method used to treat acute renal disease. Fluid removal treatments are usually combined with either hemodialysis or hemofiltration to also remove solutes normally excreted by the kidney. Currently the most advanced device for this treatment is the PRISMA system from Gambro, which comprises an air free extracorporeal circuit consisting of a blood tubing system with integrated filter, a plurality of injection or sampling ports and pressure measuring domes. Although attempts have been made to construct a streamlined flow path, the blood flow passage in the PRISMAT>'system still has dead zones where fluid or blood stagnates and resides for a prolonged time while blood is otherwise flowing through the system. Zones of fluid stagnation are especially found in the pressure measuring domes of the PRISMA system.

Extracorporeal blood treatment usually requires anticoagulation of blood.

The reason for this is the activation of blood coagulation by shear and by contact of blood to the surface of the extracorporeal circuit. After activation of the clotting system, it takes several minutes until a clot forms. If a fluid path contains poorly perfused dead zones where blood stagnates for a longer period, then blood clots will form at these sites and the clots eventually will block the entire circuit. Other causes for the enhanced formation of blood clots are blood- air interfaces and obstructions in the fluid path, e. g., the commonly used"clot" filters in drip chambers.

Typically, systemic anticoagulation is used such that anticoagulants are not only in the blood in the extracorporeal circuit, but also in the blood in the patient's circulation. This use of anticoagulants increases the risk of bleeding by the patient during and after treatment. Local anticoagulation or anticoagulation free treatment has been reported, but is possible only with additional equipment and monitoring.

The common blood accesses for acute treatment with extracorporeal circuits are central venous catheters. The insertion and use of central venous

catheters are related to several risks that may result in death or severe impairment. In particular, stenosis of the central vessels after use of catheters, which has been well documented, makes frequent insertion of central venous catheters impossible.

The use of peripheral vein access has not been reported with devices used for extracorporeal blood treatments, such as those described above. Peripheral veins tend to collapse during the withdrawal of blood by an extracorporeal blood circuit. The collapse of a vein would cause the blood circuit to issue frequent alarms that would require continuous observation by trained personnel. Thus, the use of perhipheral vein access to Also, extracorporeal systems for the use with adults are designed for blood flows not achievable with peripheral vein access.

Summary of the Invention An extracorporeal circuit has been designed having an optimized streamlined blood passage that is free of obstructions and dead zones where blood can stagnate. The circuit components, e. g., pressure sensors, in the blood are free of obstructions and do not cause blood to stagnate in the circuit or to undergo substantial flow speed changes. The circuit can be used to treat CHF by continuous filtration of blood using peripheral venous blood access, and with minimal or no anticoagulation.

The extracorporeal circuit may be characterized by the normalized residence volume method, derived from the residence time measurements. The normalized volume parameter is used to characterize the flow characteristics of an extracorporeal circuit independently of the total fluid volume capacity of that circuit. When using the 10% to 90% rise of the output function in the normalized volume diagram, the optimized extracorporeal circuit shows superior performance. The present extracorporeal circuit may be applied to remove

excess fluid from a CUT patient allow treatment for up to eight hours using no or minimal anticoagulation.

A practical means to overcome the barriers for the effective treatment of CHF by mechanical fluid removal is described in the U. S. Patent No.

(U. S. Application Serial No. 09/618,759, filed July 18, 2000), entitled"Method and Apparatus for Peripheral Vein Fluid Removal in Heart Failure", and U. S. Patent No. (U. S. Application Serial No.

09/660, 195 filed September 12, 2000), entitled"Blood Pump Having Disposable Blood Passage Cartridge With Integrated Pressure Sensors,"both of which patents are incorporated by reference.

Summary of the Drawings A preferred embodiment of the invention, the setup for the characterization of hydraulic components by the normalized residence volume method as well as the results of such characterization for an extracorporeal system as described by the invention and for the extracorporeal circuit of the PRISMA device are illustrated in the attached drawings.

Figure 1 illustrates the extracorporeal circuit of an embodiment of the present invention.

Figure 2 illustrates the flow diagram employed for the measurement of the residence time function.

Figure 3 is a graph showing the result of a residence time measurement with conductivity sensors.

Figure 4 is a graph showing the conversion of the graph 3 data from the conductivity versus time domain into the concentration versus volume domain.

Figure 5 is a graph derived from the graph of Figure 4 bv the normalization of the concentration axis with the maximum concentration measured that is the concentration in carboy 302.

Figure 6 is a graph derived from the graph of Figure 5 by the normalization of the volume axis with the volume of the test object calculated from the graph of Figure 5.

Figure 7 is a graph derived from the graph of Figure 6 bv eliminating the effect of the time course of the input conductivity/concentration function. The input function is a step function at zero normalized volume. The test object characterized by this figure is the extracorporeal circuit described by this invention.

Figure 8 is a graph similar to the graph in Figure 7. The test object characterized by this graph is the extracorporeal circuit of the PRISMA'.

Figure 8 shows the normalized residence volume calculation derived from a residence time measurement for the extracorporeal circuit of the PRISMA device.

Detailed Description of the Invention Figure 1 illustrates the fluid path of an embodiment of the present invention. The embodiment consists of a disposable extracorporeal circuit for a treatment device comprising a peristaltic blood pump, protective systems a display and a microprocessor control unit.

Blood is withdrawn from the patient through the withdrawal needle assembly 201. Blood flow is controlled by a roller pump 204. The withdrawal needle assembly is connected to the blood tubing 220 by a pair of matching connectors 230,232. Connector 230 is part of the withdrawal needle assembly and connector 232 is a part of the blood tubing 220. These connectors can either

be an integral part of the connected blood tubing or separate parts glued, welded or mechanical fixated with the tubing.

Blood tubing'20 that is typically 2m (meters) long is connected to a disposable pressure sensor 203. A suitable pressure sensor is disclosed in U. S.

Patent No. (U. S. Application Serial No. 09/660. 195 filed September 12,2000). The opposite end of the pressure sensor 203 is further connected to blood pump tubing 221 that is connected to the disposable pressure sensor 205.

Pressure sensor 205 is connected to blood tubing 222 leading to and permanently connected to the inlet of the blood side of the hemofilter 207.

The outlet of the blood side of the hemofilter is connected to blood tubing 223 that is connected to one side of a disposable pressure sensor 209. The other side of the disposable pressure sensor 209 is connected to the blood tubing 224 that ends with the connector 233. Connector 231 is part of the blood return needle assembly. A filtrate line 212 is connected to the filtrate outlet 211 of the hemofilter 207 on one side and to the filtrate collection bag ? 15 on the other side.

An ultrasonic air detector 206 with sensors in contact with the outer surface of the blood tubing'06 is included, but does not interrupt the smooth flow path in the circuit.

The blood passage in the extracorporeal circuit is an obstruction free smooth blood flow path throughout the circuit. There are no dead zones and no blood-air interfaces in the blood passage. The catheter needle assemblies 201, 210 are preferably designed such that the interface between the needle with a typical inner diameter of 1 mm is connected to the blood tubing with a typical inner diameter of 3 mm by a cone with a smooth blood passage having a cone angle of 10°. Matching connectors 230, 232 and 231, 233 respectively are preferably designed such that the continuous blood flow path with a typical inner diameter of 3mm is not interrupted. For convenience and for compatibility with -existing systems, Luer-lock connectors could be used at this point although these connectors are not obstruction free.

Disposable pressure transducers are positioned in the flow path to be obstruction free and have essentially the same diameter as the blood tubing. To avoid kinking of the blood tubing leading from and to the patient, blood tubing 220 and 209 is preferably made from a harder tubing material than is the blood pump tubing 221. Blood tubing 222 and 223 can be made from either the harder or the softer variety of medical tubing material that is suitable for use in an extracorporeal blood circuit.

The filter 207 provides a smooth flow path for the blood through the filter passages. A membrane surface area of 0.1 m2 may be used to provide sufficient fluid removal during operation of the extracorporeal circuit. A smooth flow path is achieved by making this filter long and thin. rather than short and thick as in the PRISMA design. The filter may have an effective length of 22. 5 cm and a fiber bundle diameter of 1.2 cm.

A parameter for the characterization of the flow characteristic of an extracorporeal circuit or of its components is the residence time. This method of characterizing an extracorporeal has been described, e. g., by Cooney DO, Infantolino W, Kane R. in"Comparative Studies of Hemoperfusion Devices".

For a passive extracorporeal device, the total residence time of the blood in the device should be minimized to reduce the potential for clotting.

Pressure drop and flow uniformity tests may also be used to characterize the blood flow through an extracorporeal circuit, especially those that filter fluids and/or solutes from blood. See Biomater Med Devices Artif Organs 1979; 7: 443-54. In this method the device to be investigated is sequentially perfused with two fluids with different properties. A step function is produced at the input when the fluids are switched and the resulting function at the output is recorded as a function of time. For an ideal flow, the output function would be a step function as well. In an ideal filter, all portions of the filter membrane are being perfused uniformly. In an ideal filter, the blood is thickened (due to fluid being removed by the filter) uniformly as it passes through the capillaries of the

filter. In a non-ideal filter, the slope of the step-function deviates from the ideal slope because blood moves faster in some capillaries than in others, and/or the blood is thickened less in some capillaries than others. The blood becomes more concentrated and has a higher viscosity in those filter capillaries with slow perfusion, than in other capillaries that have at least an average flow rate of blood. These non-uniformities in the blood flowing, through a filter can be indicated by the slope of the step-function curve. Deviation from the ideal slope indicates non-uniform thickening of the blood and non-uniform blood viscosity.

The deviation of the output function from a step function can be used as a measure for the quality of the flow design.

Figure 2 shows a flow diagram of a test setup that can be employed to determine the characterization of extracorporeal circuits or its elements. Bag 301 is a carboy containing demineralized water with a conductivity of typically 5yS/cm (microSiemens per cm). Carboy 302 contains a salt solution (sodium chloride solution) with a typical conductivity of 30 mS/cm. The container 301 is connected through a conduit 303 with the valve 305. The container 302 is connected to the valve 306 through conduit 304. Valves 305 and 306 are connected with a T to conduit 308 leading to a gear pump 310.

A gear pump 310 is connected to a first temperature compensated conductivity sensor 314 by conduit 312. A test object 330, e. g., a filter or blood path in an extracorporeal circuit, is connected to the first conductivity sensor 314 on one side and the second temperature compensated conductivity sensor 316 on the other side. The outlet of the second conductivity sensor 316 is connected to a mass flow meter 320 through conduit 318, and from the mass flow meter 320 a conduit 326 leads to drain. The pressure drop of the mass flow meter 320 is measured by pressure sensor 321 connected to inlet and outlet of the mass flow meter 320 by lines 322 and 324 respectively. If the test object is a filter or if a filter is part of the test object as shown in Figure 2, the filtrate side is connected

to an air pump 340 through a conduit 342, from conduit 324 a pressure measuring line 344 leads to a pressure sensor 346.

An air pump 340 allows removal of fluid from the filtrate side of filter 330. Fluid in the blood passage side of the filter is pushed through the membrane if the air pressure is higher than the maximum fluid pressure on the fluid (blood) side of the membrane. Because the membrane is hydrophilic, air cannot pass the membrane as gas. Fluid, however, remains in the pores of the membrane. The purpose of this fluid removal from the filtrate side is to avoid any influence from the filtrate space. During the measurement, sodium and chloride ions can diffuse freely through the membrane. With a fluid filled filtrate space the measurement would include the filtrate side that would lead to wrong conclusions about the fluid distribution on the"blood"side.

Conductivity sensors 314, 316 and optionally pressure sensors 321 and 346 are connected to a computer for continuous recording of the signals as function of time. To record the residence time, the function valve 305 is opened and pump 310 operated at a prescribed speed, e. g., 100 mL/min. A baseline value of-5S/cm is established, based on readings from pressure sensors 314 and 316. Next, a continuous recording of the signals from pressure sensors 314 and 316 is started. Also, valve 305 is closed and valve 306 is opened. Fluid with a conductivity of-30 mS/cm flows through the pump 310 to the conductivity sensor 314 causing an increase of the signal as shown by signal tracing 404 in the graph 401 of Figure 3. The conductivity sensor 316 at the outlet records a similar but less steep signal after some delay in time depending on the filling volume of the test object 330 and the flow speed of the pump 310. The signal tracing of the conductivity sensor 316 is shown as 406 in Figure 3: In case the test object contains a filter air pump 340 is employed to pressurize the filtrate side with air at a pressure exceeding the maximum fluid pressure on the blood side of the filter prior to the start of the measurement. This forces all fluid from

the filtrate side to the blood side limiting the fluid volume to the fluid volume of the blood path and the fluid volume trapped in the membrane pores.

Figure 3 includes a diagram 401 that is mathematically converted into the diagram 411 shown by Figure 4 by the following steps: Conductivity 402 is converted into concentration 412 by employing a five order power function derived from tables published in the CRC handbook of chemistry and physics 65th edition for sodium chloride. The time axis 403 is converted into a volume axis 413 by multiplying the time increments between the discrete points of measurement with the corresponding flow values and summation. The results are the input sensor function 414 and the output sensor function 416 in the concentration versus volume domain.

Figure 5 shows a diagram 4^ 1 mathematically derived from the graph 411 of figure 4 by normalizing the concentration axis 412 with the maximum concentration measured. This concentration is equal to the concentration in carboy 302 within the errors of measurement. The result is a graph in the normalized concentration 422 versus volume 423 domain with 424 being the input function and 426 being the output function.

Figure 6 shows a diagram mathematically derived from the graph 421 of Figure 5 by normalizing the volume scale 423 with the calculated volume of the test object. The volume of the test object is calculated from the graph 421 of Figure 5 by subtracting the integral under tracing 424 from the integral under tracing 426 over the volume interval shown in the graph of Figure 5. The result is the graph 431 in the normalized concentration 432 versus normalized volume 433 domain with 434 being the input function and 436 being the output function.

Figure 7 shows the result of the final step of the derivation in graph 441.

The delay time and non-ideal step function of the input signal 434 is removed by subtracting the normalized volume between start and tracing 434 for discrete values of the normalized concentration from the corresponding normalized

volume of the tracing 436 by employing a lookup table. The lookup table is constructed in 1% intervals of the normalized concentration. This procedure results in the tracing 445 of graph 441 showing the output function for an extracorporeal circuit as described by the invention for an ideal step function at normalized volume 0 at the input.

Figure 8 shows the result of the measurement and derivation equivalent to the steps described for Figures 3 to 7 but for the extracorporeal circuit of the PRISMA. The graph 451 shows the tracing 454 in the normalized volume 453 versus normalized concentration 452 domain. The described data has been recorded employing the program LabView (National Instruments). The recorded data has been imported into SigmaPlot 5. 0 (SPSS, Inc.) and the mathematical conversions described above were done with the transform program written for SigmaPlot shown in the appendix.

Comparing graphs 441 and 451 allow for a comparison of extracorporeal circuits independent of the absolute volume and the blood flow. For the quantitation of the 10% to 90% interval is used as shown in the following table : Relative volume Data from Fig. 7 (Invention) Data from Fig. 8 (PRISMA) 10% 0.87 0.78 50% 1.01 1.00 90% 1.02 1.22 90% - 10% 0.15 0.44 The 10% to 90% rise time is more than twice as large for the PRISMA compared to the optimized extracorporeal circuit according to the invention. The volume calculated for the two systems was 37.09 mL for the system according to the invention and 105. 38 for the PRISMA. As mentioned above, this volume includes the fluid trapped in the porous structure of the microporous fibers of the

direct comparison of the flow properties independent of the absolute size and volume of the system.

From graphs 441 (Figure 7) or 451 (Figure 8) and the volumes the residence time can easily be calculated. For the extracorporeal circuit according to the disclosed embodiment of the invention the volume is-37 mL. r 128 sec a factor of 3.5 large. For a flow of 60 mL/min typical for peripheral flow, the resulting residence time for the 90% concentration point (1.12) is 1.12*37/60 = 0.69 min or-41 sec. For the same flow the 90% residence time for the PRISMA is 1.92* 105/60 = 2. 13 min or 128 sec a factor of 3.5 larger.

While the invention has been described in connection with what is presently considered to be the most practical and preferred embodiment. it is to be understood that the invention is not to be limited to the disclosed embodiment, but on the contrary, is intended to cover various modifications and equivalent arrangements included within the spirit and scope of the appended claims