WO2014107656A1 - Device and method for detecting irregular placement of an extracorporeal vascular access needle - Google Patents

Abstract

A device and method for determining intravascular pressure and irregular needle placement in a hemodialysis procedure include analyzing means for automatically deriving venous access pressure (VAP) at a location of venous needle insertion into the patient and determining a ratio of VAP to mean arterial pressure, where the analyzing means determines multiple VAP values over multiple time periods and determines multiple ratios, one for each VAP value. Detecting means compare each ratio to a predetermined standard ratio derived from a moving average of prior VAP ratios, where the detecting means detects irregular needle placement when elevation of the ratio above the standard ratio is indicated.

Description

DEVICE AND METHOD FOR DETECTING IRREGULAR PLACEMENT OF AN

EXTRACORPOREAL VASCULAR ACCESS NEEDLE

TECHNICAL FIELD

[0001] Embodiments relate to devices and methods for detecting failure in dialysis systems based on pressure measurements.

BACKGROUND

[0002] Proper functioning of the vascular system is essential for the health and fitness of living organisms. The vascular system carries essential nutrients and blood gases to all living tissues and removes waste products for excretion. The vasculature is divided into different regions depending on the organ systems served. If vessels feeding a specific organ or group of organs are compromised, the organs and tissues supplied by those vessels are deleteriously affected and can even fail completely.

[0003] Vessels, especially various types of arteries, not only transmit fluid to various locations, but are also active in responding to pressure changes during the cardiac cycle. With each contraction of the left ventricle of the heart during systole, blood is pumped through the aorta and then distributed throughout the body. Many arteries contain elastic membranes in their walls that assist in expansion of the vessel during systole. These elastic membranes also function in smoothing pulsatile blood flow throughout the vascular system. The vessel walls of such arteries often rebound following passage of the systolic pressure waveform.

[0004] In autoregulation, cerebral blood vessels maintain constant cerebral blood flow by either constricting or dilating over a certain mean arterial blood pressure range so that constant oxygen delivery is maintained to the brain. Vascular failure occurs when the pressure drops too low and the oxygen delivery starts to fall. If the blood pressure gets too high and the vessels can no longer constrict to limit flow, then hyperemia breakthrough or loss of autoregulation can occur. Both of these conditions are pathologic states, and have been described in the literature in terms of mean arterial pressure and cerebral blood flow velocity, but there are others that cannot be explained based on that model. The failure of the model is that it relies upon systemic blood pressure. The pressure of blood in the brain itself is not being measured directly. The resultant pressure curve has an S-shaped curve.

[0005] The force applied to the blood from each heartbeat is what drives the blood forward.

In physics, force is equivalent to mass times acceleration. But when blood is examined on a beat-to- beat variation, each heartbeat delivers about the same mass of blood, unless there is severe loss of blood or a very irregular heart rhythm. Therefore, as a first approximation, the force of flow on the blood at that particular moment is directly proportional to its acceleration.

[0006] Diseased blood vessels lose the ability to stretch. The elasticity or stretch of the blood vessel is very critical to maintaining pulsatile flow. When a muscle is stretched, it is not a passive relaxation. There is a chemical reaction that happens within the muscle itself that causes a micro-contracture to increase the constriction, so that when a bolus of blood comes through with each heartbeat, it stretches the blood vessel wall, but the blood vessel then contracts back and gives the kick forward to maintain flow over such a large surface area. This generates a ripple of waves, starting in the large vessel of the aorta and working its way through the rest of the vessels. As vessels become diseased, they lose the ability to maintain this type of pulsatile flow.

[0007] Further, if vessels are compromised due to various factors such as narrowing or stenosis of the vessel lumen, blood flow becomes abnormal. If narrowing of a vessel is extensive, turbulent flow can occur at the stenosis resulting in damage to the vessel. In addition, blood cannot flow adequately past the point of stenosis, thereby injuring tissues distal to the stenosis. While such vascular injuries can occur anywhere throughout the body, the coronary and cerebral vascular beds are of supreme importance for survival and well-being of the organism. For example, narrowing of the coronary vessels supplying the heart can decrease cardiovascular function and decrease blood flow to the myocardium, leading to a heart attack. Such episodes can result in significant reduction in cardiac function and death.

[0008] Abnormalities in the cerebral vessels can prevent adequate blood flow to neural tissue, resulting in transient ischemic attacks (TIAs), migraines, and stroke. The blood vessels that supply the brain are derived from the internal carotid arteries and the vertebral arteries. These vessels and their branches anastomose through the great arterial circle, also known as the Circle of Willis. From this Circle arise the anterior, middle and posterior cerebral arteries. Other arteries such as the anterior communicating artery and the posterior communicating artery provide routes of collateral flow through the great arterial circle. The vertebral arteries join to form the basilar artery, which itself supplies arterial branches to the cerebellum, brain stem and other brain regions. A blockage of blood flow within the anterior cerebral artery, the posterior cerebral artery, the middle cerebral artery, or any of the other arteries distal to the great arterior circle results in compromised blood flow to the neural tissue supplied by that artery. Since neural tissue cannot survive without normal, constant levels of glucose and oxygen within the blood and provided to neurons by glial cells, blockage of blood flow in any of these vessels leads to death of the nervous tissue supplied by that vessel.

[0009] Strokes result from blockage of blood flow in cerebral vessels due to constriction of the vessel resulting from an embolus or stenosis. Strokes can also arise from tearing of the vessel wall due to any number of circumstances. Accordingly, a blockage can result in ischemic stroke depriving neural tissue distal to the blockage of oxygen and glucose. A tearing or rupture of the vessel can result in bleeding into the brain, also known as a hemorrhagic stroke. Intracranial bleeding exerts deleterious effects on surrounding tissue due to increased intracranial pressure and direct exposure of neurons to blood. Regardless of the cause, stroke is a major cause of illness and death. Stroke is the leading cause of death in women and kills more women than breast cancer.

[0010] Currently, more than three-quarters of a million people in the United States experience a stroke each year, and more than twenty-five percent of these individuals die. Approximately one-third of individuals suffering their first stroke die within the following year. Furthermore, about one -third of all survivors of a first stroke experience additional strokes within the next three years.

[0011] In addition to its terminal aspect, stroke is a leading cause of disability in the adult population. Such disability can lead to permanent impairment and decreased function in any part of the body. Paralysis of various muscle groups innervated by neurons affected by the stroke can lead to confinement to a wheelchair, and muscular plasticity and rigidity. Strokes can leave many patients with little or no ability to communicate either orally or by written means. Often, stroke patients are unable to think clearly and have difficulties naming objects, interacting well with other individuals, and generally functioning within society. [0012] Despite the tremendous risk of stroke, there are presently no convenient and accurate methods to access vascular health. Many methods rely on invasive procedures, such as arteriograms, to determine whether vascular stenosis is occurring. These invasive techniques are often not ordered until the patient becomes symptomatic. For example, carotid arteriograms can be ordered following a physical examination pursuant to the appearance of a clinical symptom. Performing an arteriogram is not without risks due to the introduction of dye materials into the vascular system that can cause allergic responses. Arteriograms also use catheters that can damage the vascular wall and dislodge intraluminal plaque, which can cause an embolic stroke at a downstream site. It would therefore be useful to develop a noninvasive or limited invasive procedure for assessing vascular health.

[0013] Further, in the field of hemodialysis and other techniques where blood is removed from a patient for processing and then returned, it is important to periodically assess the blood flow rate through an arteriovenous fistula, graft, or catheter to monitor the onset of stenosis. This is often accomplished by the reading of access pressures through the venous and arterial access needles. Early detection of stenosis associated with the placement of a fistula, graft, implantable port, or a catheter can permit low cost repairs to be made. On the other hand, if these problems are ignored or not detected, the cost of the revision or replacement of the fistula, graft, implantable port, or catheter can be very high and burdensome to the patient.

[0014] There have been several devices that have been developed to determine pressure inside a dialysis machine or during hemodialysis. For example, as disclosed in U.S. Patent No. 5,454,374 to Omachi, access pressures can be determined through volumetric manipulations involving the determination of a pressure head height of blood in a visual manner. The blood line going to the dialysis machine is used to measure pressure and the problem is one of determining the height between the transducer and the patient's access site.

[0015] U.S. Patent No. 4,710,163 to Levin et al. discloses a method and system for continuously monitoring patient heart rate and mean arterial blood pressure during hemodialysis and for automatically controlling fluid extraction rate and/or dialysate sodium concentration in the event that blood pressure and/or heart rate indicate onset or impending onset of a patient hypotensive episode. There are three separate machines for performing these functions: an automated blood pressure monitor, an automated patient heart rate monitor, and the hemodialysis machine. The blood pressure monitor is essentially a device for measuring blood pressure based on the blood in the patient's arm, i.e. a cuff that inflates and deflates automatically to read the diastolic and systolic blood pressure readings. This device merely takes the place of an actual technician to take a blood pressure reading. The blood pressure readings are derived from a standard blood pressure cuff on the patient's arm and not from the intravascular blood near the access site for an extracorporeal circuit.

[0016] U.S. Patent No. 6,623,443 to Polaschegg discloses a device that measures and compares the amplitude of pressure pulses within an extracorporeal circuit to determine whether a stenosis has occurred therein. The peak-to-peak amplitude of the pressure waves created by variations in the patient's blood pressure and variations in pressure created by the extracorporeal blood pump are used to indicate the presence of an obstruction in the circuit. A deviation in the peak-to-peak amplitude of the pressure signal from a predetermined standard value indicates a stenosis or loss of occlusion of the roller pump. No standard is defined to indicate a stenosis that represents a significant risk to the patient. No measurements or calculations of intravascular blood pressure occur.

BRIEF DESCRIPTION OF THE DRAWINGS

[0017] FIGURE 1 shows a dialysis circuit used to determine the relationship between blood flow and hemodialysis machine venous drip chamber pressure with hematocrit varied from 38.4% to 18.2%;

[0018] FIGURE 2 shows the venous drip chamber pressure versus blood flow in a hemodialysis machine blood circuit for a range of hematocrit values, including a single curve showing venous needle pressure at a hematocrit of 29.1%, wherein venous needle pressure is 0 mmHg when Qb=0 because the transducer and the venous needle are at the same height, and venous drip chamber pressure is approximately -17 mmHg when Qb=0 because the venous needle is 17 centimeters below the height of the drip chamber transducer;

[0019] FIGURE 3 shows the receiver-operating characteristic (ROC) curves for the January

1999 VAPRT for grafts (1 17) and fistulas (23) combined and grafts alone, an area of 1 represents an ideal test, an area of 0.5 indicates the test has only a 50% probability determining the correct outcome, and an area from 0.80 to 0.90 implies a good test;

[0020] FIGURE 4 shows the distribution of access pressure ratio values within the four possible test groups: true positive, true negative, false positive, and false negative for patients with grafts;

[0021] FIGURE 5 shows the access pressure ratio test results for three separate months of testing, wherein patients were followed for six months after each test for an access failure event;

[0022] FIGURE 6 is a graph showing the relationship between coefficient B in the equation for venous drip chamber pressure with zero venous access pressure VDPo = 0.00042329*Qb + B*Qb 17.325 and hematocrit (Hct);

[0023] FIGURE 7 is a flow chart depicting the inner workings of a device and algorithm used to determine the venous access pressure ratio (VAPR) and monitor for significant variations in VAPR from treatment to treatment according to an embodiment;

[0024] FIGURE 8 is a photograph of a percutaneous transluminal angioplasty;

[0025] FIGURES 9A and B are photographs depicting dialysis machines for use in conjunction with the device in accordance with an embodiment;

[0026] FIGURE 10 is a flowchart depicting a method of detecting a dislodged needle during hemodialysis according to an embodiment.

[0027] FIGURE 11 is a graph depicting venous access pressure ratio (VAPR) versus monitoring time in months, wherein incidences of poor needle placement and the moving average of VAPR are indicated; and

[0028] FIGURE 12 is a flowchart illustrating an algorithm for determining irregular needle placement, wherein after starting the blood flow in the extracorporeal circuit the current venous access pressure ratio is determined and compared to a predetermined standard.

DETAILED DESCRIPTION [0029] As required, detailed embodiments of the present invention are disclosed herein; however, it is to be understood that the disclosed embodiments are merely exemplary of the invention that may be embodied in various and alternative forms. The figures are not necessarily to scale; some features may be exaggerated or minimized to show details of particular components. Therefore, specific structural and functional details disclosed herein are not to be interpreted as limiting, but merely as a representative basis for teaching one skilled in the art to variously employ the present invention.

[0030] Generally, according to embodiments, a detection device and method are provided for detecting variations in intravascular pressure that indicate irregular blood flow, i.e. a suspected blood flow restriction or other blood flow problem, especially when a needle of a hemodialysis device has become dislodged from a patient. The device includes an analyzer for automatically analyzing intravascular pressure upstream of the suspected location of irregular blood flow and comparing the intravascular pressure to a standard, whereby variations in the intravascular pressure during multiple tests is indicative of a blood flow restriction.

[0031] U.S. Patent No. 7,597,666 to Frinak et al. disclosed for the first time a method of detecting an irregular intravascular pressure by measuring extracorporeal pressure taken from a patient and analyzing the extracorporeal pressure with an algorithm to determine intravascular pressure. The intravascular pressure is compared to a standard in order to determine if the patient is at risk of developing a stenosis. Variation of the calculated intravascular pressure multiple times with the standard indicates irregular blood flow and risk of stenosis.

[0032] Dialysis is a very complicated procedure that must be carried out by a team of trained professionals who are responsible for delivering safe and effective care to the patient. It can also be self- administered by a patient in their home, but only after the patient has undergone extensive training. There are many ways that complications can arise during a dialysis session. Many of these potential issues are constrained by alarm circuits and other safeguards built into the dialysis machine.

[0033] Hemodialysis machines utilize two needles, one to remove blood from the patient

(arterial) and one to put the dialyzed blood back into the patient (venous needle). The venous needle can become dislodged from the patient, such as accidentally pulled out of the access, which then allows the blood being pumped back into the patient to run onto the floor. Because of the relatively high blood flows of the dialysis machines (300 to 500 ml of blood per minute), if this dislodgement goes unnoticed the patient can bleed to death in a short amount of time. For example, an average male patient can lose 40% of their blood supply in 8 minutes. Even in a hospital or clinical setting, dislodgement can sometimes occur without any visual detection by a medical staff because a blanket can cover the bloodlines. This issue is even more of a concern when a patient is dialyzed overnight. This can be more convenient for patients who do not want to spend the day in the hospital, with the hemodialysis procedure performed while they are asleep. However, overnight dialysis poses even more of a risk that the dislodgement of the venous line needle during the procedure will go unnoticed. For example, if the patient rolls over during sleep or otherwise significantly moves in the hospital bed, this can cause needle dislodgement. A large quantity of blood can be lost and death can result in many cases. It has been estimated that between 40 and 136 patients die each year in the US due to losing sufficient blood because of needle displacement.

[0034] The current method of detecting dislodgement of a needle is visual monitoring by staff that must instruct the patient not to cover venous lines with a blanket. While many hemodialysis machines do include some sort of alarm to indicate pressure changes in the venous and arterial bloodlines, dislodgement of needles generally do not trigger an alarm, so the dislodgment is often not detected until too late. The reason for this is that small gauge needles that are used to minimize pain to patients create back-pressures that continue to be detected by the machine when the needle is dislodged. This sufficient back-pressure created in the tubing and needle masks the pressure drop at the tip of the needle if it becomes dislodged, such that the drop in the pressure caused by the removal of the needle from the arm, and hence the loss of the pressure required to push the blood into the patient's arm, is not high enough to show a significant change in the pressure as measured by the venous drip chamber transducer, especially if the range of alarm is not set correctly on the machine. Thus, sufficient pressure remains in the circuit between the tubing and the needle so that the measured venous drip pressure does not drop significantly, and no alarm is set off. There is a need for a more reliable method of detecting dislodgement of venous needles from a patient as well as an alarm system to turn off the blood pump on the dialysis machine and alert medical personnel in time to save a patient's life.

[0035] According to an embodiment, a method is provided for detecting a dislodged needle in a hemodialysis procedure by measuring venous drip pressure in a patient, analyzing the venous drip pressure and deriving intravascular blood pressure at the location of needle placement in the patient. The actual pressure may be calculated as seen at the tip of the venous needle, which when dislodged, dramatically decreases to zero or near zero. Hence, the radical change in this calculated pressure when a needle is dislodged allows for the determination that something is wrong with the venous needle and should be investigated. According to another embodiment, a method is also provided of shutting down the dialysis machine and alerting medical personnel of a dislodged needle in a hemodialysis procedure.

[0036] The "detection device" as disclosed herein is intended to include, but is not limited to, any device that is able to detect variations in intravascular pressure that indicate irregular blood flow. In one embodiment, the intravascular pressure is venous pressure that is upstream of the suspected area or location of a blood flow restriction. An example of such a device is a hemodialysis machine.

[0037] The "analyzer device" as used herein is intended to include a device that is capable of automatically analyzing the intravascular pressure. Such an analyzer device can be computer-driven. For example, the analyzer can include a device that is associated with a hemodialysis machine, such that it automatically assesses intravascular pressure during hemodialysis. The analyzer can then equate and compare the intravascular pressure to a standard. An equation is used that estimates pressure inside a blood access site and is then used to detect irregular blood flow. In one embodiment, this equation is an algorithm that calculates the ratio between venous blood pressure and mean arterial pressure.

[0038] The term "variation" is intended to include an increase or decrease in the derived intravascular pressure. Any deviation from the standard can be indicative of a problem. Depending upon whether there is an increase or decrease in intravascular pressure, the detection of the deviation helps determine what the problem is at the access site. For example, if there is an increase in intravascular pressure, the problem potentially is something that blocks normal blood flow downstream of the measurement site. The blockage represents a narrowing of a blood vessel that increases the risk for an access failure, a stroke, or a heart attack. If there is a decrease in intravascular pressure, this is indicative of a blockage of normal blood flow upstream of the measurement site. [0039] The term "communication device" as used herein is intended to include a device operably connected to the detecting device for communicating a warning when the detecting device indicates an irregularity of blood pressure of at least two uses of said device. The communicating device can be selected from, but is not limited to, electronic communications, a facsimile, a telephone, a cable modem, and a Tl connection.

[0040] The term "algorithm" as used herein is intended to encompass any computation that enables an individual to ascertain the information necessary for detecting irregular intravascular pressure. In one embodiment, the algorithm is computer driven and follows the general function shown in FIGS. 7A through D. The algorithm can be used as part of an integrated circuit. This circuit enables the algorithm to be more easily incorporated into a dialysis machine. The circuit can be created using technology known to those with skill in the art.

[0041] The methods described herein may be practiced with the following device. The device includes a detection device for detecting irregular intravascular pressure, the device including an analyzer for automatically monitoring intravascular pressure upstream of the suspected location of irregular blood flow, and a device for comparing intravascular pressure to a standard, whereby variation in the intravascular pressure during multiple tests is indicative of irregular blood flow. As disclosed above, the device may be affixed to a hemodialysis machine; however, the device can be affixed to any other device with blood flow. The analyzer is a computer-driven device and may include an algorithm that analyzes intravascular pressure, hemodialysis venous access pressure, and blood pump flow data to identify patients at-risk for access dysfunction, either for thrombosis requiring percutaneous transluminal angioplasty, or surgery to maintain access patency.

[0042] Alternatively, the device can be included as part of a hand-held device. In this embodiment, the device may replace the pressure gauge with a hand-held microprocessor controlled device that measures and records the pressure measurements. An algorithm in the device calculates the average pressure over a predetermined sampling period. The device may also contain a computer database to recall individual patient information and to record current pressure measurements in the patient's database record. Data from the device can be transferred via a communication port to a larger computer system with a more extensive patient database. [0043] Generally, according to at least one embodiment, a method and device may be provided for monitoring and/or detecting failure in a system based on pressure measurements. The present invention has numerous applications which can include, but is not limited to, mechanical, chemical, and biological arts. For instance, in chemical processes, the present invention is useful where pressure changes are indicative of system failure. Additionally, the method and device of the present invention can be used for detecting any variation in blood pressure and forwarding via the communicating device a warning regarding this variation. The device and method therefore can be used in detecting potential access failure, risk of stroke, risk of heart attack, risk of stenosis, and risk of aneurysm.

[0044] More specifically, a method is provided for detecting a dislodged needle in a hemodialysis procedure by measuring venous drip chamber pressure in a patient, analyzing the venous drip pressure and deriving intravascular blood pressure at a location of the venous needle insertion into the patient, comparing the derived intravascular blood pressure to a standard which may have been developed from prior calculations during that particular session, and repeating the measuring, analyzing and deriving, and comparing steps to determine if the derived intravascular blood pressure is within a specified range of the standard, which may indicated that a needle has been dislodged in the hemodialysis procedure. The steps of this method are generally depicted in FIGURE 10.

[0045] The venous drip chamber pressure (VDP) is the pressure that is actually measured in the extracorporeal circuit (outside the body), and is further described below. The intravascular blood pressure is calculated by analyzing the venous drip pressure and the deriving venous access pressure (VAP) in proximity of a location of venous needle's point of access on the body. These steps are further described below. The derived intravascular blood pressure (VAP) is compared to a standard that can be set for the device or derived from prior measurements of VAP during the session or from prior sessions for the patient as further described below. Each of the measuring, analyzing and deriving, and comparing steps may be repeated multiple times during the session when the medical device is in use. More specifically, multiple VAP values are determined over multiple time periods. It may be advantageous from a safety point of view to make these measurements frequently. [0046] Once the intravascular pressure (VAP) has been determined to be within a specified range of the standard, possibly indicating that the needle has become dislodged, an alarm circuit may be activated that then communicates via a communication device a command to alert the medical staff and/or rums off the blood pump of the hemodialysis machine, so that the patient does not continue to lose blood.

[0047] The device may include an alarm that is activated and alerts medical personnel to a problem with the patient. The alarm may provide a warning if the patient's needle came out of the access, i.e. became dislodged. Thus, the venous drip chamber pressure is equal to or close to venous drip chamber at zero access pressure for an alarm to occur. Currently, dialysis machines cannot detect an opening of the venous return line and incidents of severe bleeding have been reported when the venous needle has come out of the access site during dialysis. By detecting a drop in the intravascular pressure of the patient, an alarm can be activated on the detecting device that alerts medical personnel to the patient's condition so that the needle can be replaced and the patient's life can be saved from unnecessary blood loss. The alarm can also wake up the patient if asleep so that the patient can alert medical personnel, and can include a vibrating portion attached to the patient to assist in waking up or alerting the patient.

[0048] The algorithm according to an embodiment calculates the actual pressure as seen at the tip of the needle by removing the pressure caused by the needle and tubing (VDPo) from the measured VDP, which leaves VAP. By building the algorithm into the dialysis machine so that VAP is calculated often, an alarm can be sounded when VAP drops to zero or near zero, thus indicating that the venous needle probably has dislodged. This alarm determination can then a) rum off the machine so that the patient does not lose more blood, and b) sound an alarm to notify either the medical staff or the home care patient that a problem exists.

[0049] The algorithm can be utilized as an alarm system in any device that transports blood from a patient to an extracorporeal circuit and returns the blood to the patient. The algorithm determines the pressure at the point of insertion of the blood into the body based on a pressure reading in the extracorporeal blood circuit along with the rate of fluid flow through the device, the physical properties of the fluid transported through the device and a determination of the pressure inherent in the external circuit beginning from the pressure measuring device to the end of the needle at the point of insertion into the body. The algorithm allows the alarm level to vary with the rate of fluid flow through the device. The device can be utilized as an alarm in plasmapheresis, heart lung machines and any extracorporeal blood treatment or infusion technology circuits. Alarm systems based on the device are not limited to medical applications but can be developed for any fluid transporting device. Alarm levels can be set at any pressure value that provides safe operation of the device.

[0050] The alarm can be a wireless alarm or a hardwired alarm. More specifically, a wireless alarm can send wireless signals to a handheld monitor/device that is carried by medical personnel or to a central monitoring area, such as by the Internet or through communication mechanisms that include, but are not limited to electronic communications, facsimile, telephone, cable modem, and Tl connection. A hardwired alarm can send signals to any device that is in electrical connection with the detecting device of the present invention, such as a central monitoring area. The alarm can also be an audible warning or other similar signal that sends a command to the medical device (such as turn off) and/or wakes up the patient and alerts medical personnel.

[0051] Thus, by performing the method according to disclosed embodiments, if a needle should become dislodged by the patient's movement during sleep or otherwise, the patient's life can be saved by turning off the machine and alerting medical personnel in time.

[0052] A method is also provided for alerting medical personnel of a dislodged needle in a hemodialysis procedure by detecting a drop in intravascular pressure derived from measured venous drip pressure, detecting a dislodged needle, and alerting medical personnel of the dislodged needle. Each of the steps of this method is described above.

[0053] The detection device can be used to monitor any type of patient blood access site for increased blood pressure and subsequently reduced blood flow. The types of blood access sites that can be monitored include, but are not limited to, fistulas, grafts, catheters, or any type of permanent blood access port. In catheters and permanent blood access ports, the plastic materials used to construct the devices become coated with layers of protein and fibrous substances that reduce the internal diameter of the blood pathway or these devices may induce the formation of a vascular stenosis downstream of the implantation site. Any reduction in internal diameter of the blood pathway that results in an increase in pressure upstream of the catheter or permanent blood access port can be detected by the algorithm in the device and a warning can be issued once an appropriate alarm level is exceeded.

[0054] Additionally, the disclosed device and method can be applied to monitor the arterial line supplying the dialysis machine. A significant increase in the negative pressure created by the dialysis machine blood pump removing blood from the patient can be used to indicate the presence of an arterial stenosis or an obstruction of the arterial line. Further, the device and method can be utilized to describe the relationship between blood flow, pressure, and hematocrit in any type of system that removes blood from a patient and returns the same blood to the patient. Thus, it can be used in conjunction with a heart-lung machine to determine alarm parameters for blood withdrawal and reinfusion.

[0055] The detection device can be used with intravenous infusion systems to determine the pressure profile for fluid infusion through a known tubing set and needle. A significant increase in the infusion pressure at the specified fluid viscosity and flow rate can be used to determine alarm conditions and prevent infusion of fluid into the tissue if the needle is not inside the lumen of the vein. Further, any industrial system that requires regulation of infusion pressure can utilize the present invention to develop a monitoring system based on the analysis of infusion pressure.

[0056] Occasionally, when a medical professional cannulates a patient, the needle may not be centered in the blood vessel. Instead, the needle tip may end up against the side of the vessel, which results in one of the openings at the needle tip being up against the vessel wall. This unintended misplacement results in less blood being able either pulled into the machine (arterial side) or less blood able to be re-introduced into the body (venous side). These are termed 'poor needle sticks' or just 'bad sticks'. In order to fix the problem, the medical professional only needs to reposition the needle to get the tip away from the vessel wall. However, first they must be made aware of the problem. At present, there is no easy way for medical staff to know there is an issue and, as a result, the patient undergoes the dialysis session at decreased efficiency.

[0057] Therefore, in addition to the embodiments described above, there is a need for a system that enables health care providers to determine the reliability and safety of a patient's vascular access connection to an extracorporeal circuit that is used for therapeutic intervention, and to prompt the modification of said connection when irregular placement of an extracorporeal vascular access needle is detected. Irregular access pressures can occur at the vascular access site where blood is removed from the access, creating high negative pressures and low blood flow through the extracorporeal circuit which could compromise the patient's therapy. Irregular access pressures that occur, due to irregular placement of a vascular access needle, at the site where blood is returned to the patient result in high venous return pressures in the extracorporeal circuit and can result in mechanical destruction of red bloods cells causing anemia in the patient. Therefore, extracorporeal circuits employ pressure limits that restrict the level of venous return pressure in the circuit. In the case where high venous return pressure is caused by irregular placement of the access needle, the pressure can be reduced to a normal level if the healthcare provider identifies the situation and repositions the venous access return needle. If the irregular placement of the access needle is not detected by the healthcare provider, the blood flow through the extracorporeal circuit will have to be reduced to comply with the limits set for the venous return pressure, which could compromise the patient's therapy if treatment time is not extended.

[0058] Accordingly, a detection device and method are provided for recognizing irregular, abnormal, or outlying intravascular pressures as an indicator of irregular placement of an extracorporeal vascular access needle. The device includes an analyzer for automatically determining intravascular pressure at the location of needle placement, and a comparing device that correlates the recorded intravascular pressure to a standard, whereby variation in the expected intravascular pressure during needle placement is indicative of irregular placement of an extracorporeal vascular access needle. Also disclosed is a system for providing a warning of an irregular placement of an extracorporeal vascular access needle that can cause potential health problems. Embodiments include a detecting device as set forth above and a communicating device operatively connected to the detecting device for communicating a warning when the device indicates an irregularity of intravascular pressure during the placement of the extracorporeal vascular access needle. Other aspects of the device and method described above for detecting venous needle dislodgement may also be applicable to the device and method for detecting irregular intravascular pressures due to needle placement.

[0059] With reference to FIG. 7, for the assessment of irregular needle placement, the determination of pressures above a threshold is not as relevant. Instead, the pressures may be determined over the past X trials and then a metric calculated that averages the pressures and may account for trends up or down over time. In one embodiment, the pressure should be calculated as soon as possible after the start of the session, after the patient is cannulated and the pump started. If that pressure is some factor Y greater than the previous "average", then a medical professional should be alerted to reposition the needle. In one embodiment, the factor Y may be at least two standard deviations above the average, although the factor is not limited to this value. The factor could also be adjustable, for example, the factor may be lowered following testing to provide for increased sensitivity of the irregular needle placement indication.

[0060] Although irregular needle placement will present as an increase in the pressure above the average, a decrease below the average could be due to a change in the access site (i.e., the site was repaired, the patient is now using a new access site, or is using a catheter). Changes in the patient's access site are often made without updating information in a patient database. In one embodiment, the warning for an irregular needle placement could include an override for a change in the patient's access site.

[0061] A graph of venous access pressure ratio (VAPR) versus monitoring time is depicted in FIG. 11, wherein a moving average of VAPR and instances of irregular or outlying VAPR are indicated. The embodiments disclosed herein for detection irregular placement of an extracorporeal vascular access needle use measurements of irregular, abnormal, or outlying intravascular pressures to detect irregular placement of extracorporeal access needles. The data points marked "Poor Needle Placement" on the graph are from individual patient treatments where the VAPR greatly exceeded the moving average of the VAPR. The fact that the VAPR values was very close to the moving average of the VAPR for the treatments before and after the single extremely high value of VARP indicates irregular placement of the extracorporeal vascular access needle.

[0062] FIG. 12 is a flowchart depicting an algorithm for determining irregular needle placement. Vascular access pressure data is input from previous dialysis treatments, and a moving average of venous access pressure ratios (mVAPR) is calculated. In one embodiment, blood flow (Qb) is greater than 200 ml/min, venous drip chamber pressure (VDP) is greater than 110 mmHg, and mean arterial pressure (MAP) is greater than 60 mmHg. The current VDP, hematocrit, and MAP are determined, and the current VAPR is calculated. If the current VAPR is greater than a standard VAPR, then a warning is provided that needle placement may be irregular. If the current VAPR is not greater than a standard VAPR, then extracorporeal treatment can proceed. Of course, the algorithm could also be used on the arterial side by evaluating the arterial access pressure ratio (AAPR). Still further, the algorithm can be used to detect the non -optimal placement of any catheter, and is not limited to indicating irregular placement of an access needle.

[0063] Hemodialysis access monitoring programs that measure access flow or intra-access pressure have been developed for early detection of evolving stenotic lesions (1-8). Studies have shown that early detection of stenotic lesions followed by timely corrective procedures reduces the thrombosis rate and improves hemodialysis access survival (1, 3, 9, 10). Access monitoring programs are costly because they require equipment, personnel, data storage, and analysis. The method according to an embodiment includes an inexpensive technique known as the venous access pressure ratio test (VAPRT), and obviates these encumbrances.

[0064] During hemodialysis, blood is drawn from the vascular access through the arterial needle by the hemodialysis machine blood pump. After passage through the dialyzer, the blood traverses the venous drip chamber and returns to the access through the venous needle. The pressure required to infuse blood back into the access through the venous tubing and access needle and to overcome the pressure within the access is recorded as the venous drip chamber pressure (VDP). One component of VDP is the access pressure at the venous needle site (hereafter, termed "venous access pressure" (VAP)). Another component of VDP is the combined pressure required to overcome the resistance to flow through the tubing distal to the drip chamber (low) and through the venous return needle (high). VDP is also a function of needle size, tubing length and blood viscosity, represented by hematocrit. If the venous pressure within an access at the needle site is 0 mmHg, VDP can be defined as VDPo, i.e., the venous drip chamber pressure when the access pressure is zero. Consequently, VDP₀ can be calculated for a given hemodialysis machine, tubing set, and needle size when the blood flow rate and hematocrit are measured. Once VDPo is determined, VAP can be calculated from the measured VDP.

[0065] VAP = VDP - VDP₀ Equation ( 1 )

[0066] An elevation of VAP indicates stenosis in the venous outflow of the access and is associated with increased access failure probability (6, 8, 11, 14). To normalize variations in VAP attributed to changes in mean arterial pressure (MAP), the venous access pressure ratio (VAPR) is calculated by dividing VAP by MAP.

[0067] VAPR = VAP/MAP Equation (2)

[0068] The data that yields the determination of VDPo is contained within a central database repository that holds dialysis laboratory data and parameters acquired from hemodialysis machines that directly communicate with computers in the dialysis units. The VAPRT algorithm utilizes an empirical formula to calculate VAP from a dynamic measurement of VDP obtained at treatment and digitally recorded. The VAPRT algorithm analyzes monthly VAPR values and identifies individuals with consistently elevated intra-access pressures at risk for access failure. To eliminate treatment errors such as needle reversal or suboptimal needle placement that cause elevated VDP, an abnormal VAPRT was operationally defined as VAPR>0.55 at three treatments.

[0069] Analysis of the data for the hemodialysis machine circuit yielded the following second order polynomial equation, henceforth referred to as Equation (3):

[0070] VDP₀=0.00042*Qb²+(0.62116^!i:Hct²+0.01203^!i:Hct+0.12754)Qb-17.32509 (3)

[0071] Equation (3) can be used to calculate VDPo for any Qb at known Hct. For example, at

Qb=500 ml/min and Hct 18.2%, VDP₀ is 163 mmHg and increases to 200 mmHg when Hct=38.4%. VAP can be calculated from VDP recorded at HD by Equation (1) and VAPR is calculated by Equation (2). At Hct 38.4%, Qb 500 ml/min, VDP 265 mmHg, VDP₀ 200 mmHg, and MAP 100 mmHg, VAPR=0.65=(265-200)/100. In the case where blood flow (Qb) is equal to zero in Equation (3), the following occurs:

[0072] VDPo=0+0- 17.32509 = 17.32509

[0073] Venous access pressure (VAP) is then calculated using Equation (1).

[0074] VAP=VDP-VDP₀ VAP=VDP-(- 17.32509) VAP=VDP+17.32509

[0075] The constant (-17.32509) is determined by the dialysis machine type and the level of the patient's access site. Clinical studies have shown that the venous drip chamber pressure recorded by the machine and corrected for the height difference between the drip chamber transducer the patient's access gives an accurate value for venous access pressure (8, 22). The algorithm can therefore be incorporated into the dialysis machine. The dialysis machine therefore automatically records the readings. Additionally, a sensor can be placed on the hemodialysis machine to determine the height difference between the venous drip chamber transducer and the level of the patient's access site.

[0076] The VAPRT relies on a nonlinear regression formula to calculate VDPo for specific hemodialysis blood tubing set and access needle when the patient's hemodialysis blood pump flow (Qb) and hematocrit are known. The formula was developed from data analysis obtained during in vitro sham hemodialysis. FIG. 1 shows a diagram of the experimental hemodialysis system. The dialysis machine (Fresenius 2008H, Lexington, Mass., U.S.A.) blood pump was calibrated prior to experiments using the standard maintenance procedure. The exact flow was not measured during the in vitro experiment as the intention a priori was to design a monitoring system that utilized routine dialysis data obtained from each dialysis treatment. The reservoir is filled with 500 ml of human whole blood obtained from the hospital blood bank. The blood pump transports blood from a reservoir through the dialyzer and the venous drip chamber and then to a 15 gauge, 1 -inch backeye access needle. The venous access needle is inserted into a section of large-bore tubing that is open at both ends. One end of the tubing returns blood to the reservoir and the other end is elevated to prevent blood from escaping. This section of the circuit is not designed to simulate an actual access, but to avoid any resistance to flow at the tip of the venous access needle that can be recorded as an increase in VDP. The access needle is positioned 17 cm below the venous drip chamber transducer to simulate the average location of an angioaccess relative to the transducer during a typical hemodialysis treatment. The drip chamber transducer monitors the pressure created by the blood flowing through the circuit. VDPo readings are obtained directly from the hemodialysis machine. A sample of blood is obtained for hematocrit determination from the reservoir. VDPo is recorded as Qb is increased from 0 to 600 ml/mm in 50 ml/mm increments. A separate transducer, placed directly behind the access needle, measures the pressure created by the access needle's intrinsic resistance. The blood is then diluted with matched human plasma to lower hematocrit by approximately 4%. Blood is permitted to circulate at 500 ml/mm for 5 minutes to ensure uniform mixing with the additional plasma before the next sample is obtained for hematocrit measurement. VDPo measurements are repeated for Qb from 0 to 600 ml/mm. The circulated blood is diluted five times, reducing the original hematocrit by approximately 20 percentage points. VDPo measurements were conducted at each of the five dilutions.

[0077] The test monitors for a persistent elevation of the VAPR to identify an access that requires additional evaluation. The algorithm calculates VAPR from VDP and blood pump flow data that is routinely collected during hemodialysis and stored in a computer database. The algorithm determines whether a persistent increase in VAPR is present during sequential treatments.

[0078] To limit variability intrinsic to differences in needle gauge, patients with less than 48 hemodialysis treatments were eliminated from analysis because a smaller gauge needle is frequently used when initially cannulating a new or poorly developed angioaccess. The program extracts the most recent hematocrit and individual treatment data from the computer database and analyzes data for those patients who receive treatments via a graft. The VAPR is calculated each time the blood pressure is measured during hemodialysis, given the following criteria: Qb>200 ml/mm, VDP>20 mmHg and MAP>75 mmHg. Data from the last hour of hemodialysis is excluded to eliminate the effect of ultrafiltration on hematocrit (elevated blood viscosity), blood pressure, and changes in systemic and vascular access resistances. The algorithm then calculates the mean VAPR for each hemodialysis treatment using all available data. In the majority of cases three or four measurements are available. Patients with <10 hemodialysis treatments during a month were excluded. The VAPRT is considered positive when, starting with the eighth treatment of the month; the program determines that the VAPR exceeds the specified cutoff value during three consecutive treatments.

[0079] The invention is further described in detail by reference to the following experimental examples. These examples are provided for the purpose of illustration only, and are not intended to be limiting unless otherwise specified. Thus, the invention should in no way be construed as being limited to the following examples, but rather, should be construed to encompass any and all variations that become evident as a result of the teaching provided herein.

[0080] EXAMPLES

[0081] Example 1 [0082] Criterion for the Venous Access Pressure Ratio Test

[0083] To determine the VAPR cutoff value most predictive of access failure, test data and follow up data were analyzed from 117 patients with grafts who received hemodialysis treatment at three hemodialysis facilities during January 1999. VAPR in these patients were correlated with the presence or development of access dysfunction, stenosis requiring intervention by angioplasty or surgical revision to maintain access patency, or the occurrence of thrombosis within the six months of follow up observation. A six month observation period was selected because data reported showed that primary unassisted patency for grafts at six months is 64% and secondary assisted patency is 70% at six months, which is in accordance with data from Sparks (15) showing a primary patency for grafts of 64% at a median of seven months. The data from these studies indicates that in any six month period 30 to 36% of all grafts can fail. The VAPRT is being used to try and identify grafts in this group before they fail.

[0084] A receiver operator curve (ROC) for VAPRT was constructed with cutoff ratios of 0.2,

0.3, 0.4, 0.45, 0.5, 0.55, 0.6 and 0.8 while other test parameters were held constant. The respective sensitivities and specificities were calculated at each VAPR cutoff level. Areas under the receiver operator (ROC) curves were calculated using Mathcad Plus 6.0 (MathSoft Inc., Cambridge, Mass., U.S.A.). Clinical results were analyzed with StatView for Windows v. 5.0 (SAS Institute, Inc., Cary, N.C., U.S.A.) and DeltaGraph 4.0 (SPSS, Inc., Chicago, III, U.S.A.). Grouping variables for unpaired t- tests were true positive (TP; test predicted intervention or access clotting), true negative (TN; test correctly predicted the absence of an access event), false positive (FP; test falsely predicted an access event would have occurred) and false negative (FN; test falsely predicted that an access event would not occur). The hypothesized difference between groups for all comparisons was zero.

[0085] Clinical Application of Venous Access Pressure Ratio Test

[0086] A total of 359 VAPRT were acquired from ESRD patients in three Greenfield Health

System hemodialysis units over a three month interval following the determination of the optimal VAPR=0.55. The same population's data was retrospectively analyzed from January (n=112), February (n=113) and March (n=134) of 1999. Medical records were examined to identify those individuals who required intervention for an access event, defined as an obviously low access flow (<250 ml/mm), an inability to provide adequate dialysis within the predetermined treatment time or surgical or angioplasty intervention to maintain access patency, from stenosis or thrombosis.

[0087] RESULTS

[0088] In vitro Modeling of VAP₀

[0089] Derivation of the Mathematical Model

[0090] Results of the sham dialysis study are shown in FIG. 2. Mathematical modeling of

VDPo data is shown in FIG. 2. The data in FIG. 2 was analyzed by fitting each individual curve with an equation of the form:

[0091] VDP₀=A*Qb²+B*Qb+C Equation (la)

[0092] The constant C represents the value of VDP when Qb=0 and the average value of -

17.325 mmHg was used during further analysis of the data. Because coefficient A varied minimally from 0.0004232 to 0.0004327, an increase of only 1.5 mmHg in VDPQ at Qb=400, a mean value of 0.00042329 was used. Coefficient B varied the most with hematocrit from 0.145289 to 0.231968. The raw data was then fit with Equation (2a).

[0093] VDP₀=0.00042329*Qb²+B*Qb-17.325 Equation (2a)

[0094] B coefficients were obtained for each hematocrit value. FIG. 6 displays the plot of

Coefficient B versus hematocrit and Equation (3 a) was fit to the data.

[0095] B=0.62116*Hct²+0.01203*Hct+0.12754 Equation (3a)

[0096] Equations (2a) and (3a) were combined to yield Equation (4a) that relates VDPo to Qb and Hct.

[0097] VDP₀=0.00042*Qb²+ (0.62116*Hct²+0.01203*Hct+0.12754)*Qb-17.32509 (4a)

[0098] Equation (4a) was evaluated for accuracy using a nonlinear regression program

(DataFit, Oakdale Engineering, Oakdale, Pa., U.S.A.). The adjusted coefficient of multiple 2

determination r =0.99982 validated that Equation (4a) represents an accurate mathematical model of the pressure data for access monitoring by dynamic VAPRT.

[0099] Application of the Mathematical Model

[0100] Analysis of the experimental data for the hemodialysis machine circuit yielded the following second order polynomial equation, henceforth referred to as Equation (3):

[0101] VDPo=0.00042*Qb²+ (0.62116*Hct²+0.01203*Hct+0.12754)*Qb-17.32509 (3)

[0102] The common average intercept, -17.35, was established empirically and is related to the 17 cm difference in height between the needle and drip chamber transducer at Qb=0. When pressure is measured from the transducer proximal to needle, the offset becomes zero, and the relationship between pressure and flow remains curvilinear (FIG. 2, venous needle pressure at Hct=29.1). Thus, VDPo increases in relationship to increasing Qb and hematocrit.

[0103] Equation (3) can be used to calculate VDPo for any Qb at known Hct. For example, at Qb=500 ml/min and Hct 18.2%, VDPo is 163 mmHg and increases to 200 mmHg when Hct=38.4%. VAP can be calculated from VDP recorded at HD by Equation (1) and VAPR is calculated by Equation (2). At Hct 38.4%, Qb 500 ml/min, VDP 265 mmHg, VDP₀ 200 mmHg, and MAP 100 mmHg, VAPR=0.65=(265-200)/100. In the case where blood flow (Qb) is equal to zero in Equation (3), the following occurs:

[0104] VDP₀=0.00042*Qb²+(0.62116^!i:Hct²+0.01203^!i:Hct+0.12754)^!i:Qb-17.32509

[0105] When Qb=0 venous access pressure (VAP) is then calculated using Equation (1).

[0106] VDP₀=0+0-17.32509=-17.32509

[0107] VAP=VDP-VDP₀ VAP=VDP-(- 17.32509) VAP=VDP+17.32509

[0108] The constant -17.32509 is determined by the dialysis machine type and the height of the patient's access site. Clinical studies have shown that the venous drip chamber pressure recorded by the machine and corrected for the height difference between the drip chamber transducer the patient's access gives an accurate value for venous access pressure. The algorithm can therefore be incorporated into the dialysis machine. The dialysis machine therefore can automatically take the readings. Additionally, a sensor can be placed on the machine to determine the height difference between the venous drip chamber transducer and the level of the patient's access site.

[0109] Receiver Operator Curve (ROC) Evaluation

[0110] Patients with grafts (N=l 17) included during the January 1999 test period and whose data were used for ROC analysis had mean treatment blood flows 438±61 ml/mm, hematocrit 34.0±4.2% MAP 102±14 mmHg, VDP values ranging from 48 to 430 mmHg (mean 214±43 mmHg), and mean VAPR 0.64±0.35.

[0111] The receiver operator curve (ROC) is shown in FIG. 3. The area under the curve corresponds to the probability (0.82) of correctly ranking the two test alternatives, persistence of access patency or occurrence of access failure within six months (16, 17). The VAPR cutoff of 0.55 was selected for further clinical testing as it provided a rational compromise between sensitivity (75%) and specificity (83%).

[0112] FIG. 4 shows the distribution of individual treatment mean VAPR values for all patient observations with grafts in January 1999. The monthly mean VAPR for each patient was calculated from the VAPR values obtained at each treatment. Patients who had a TP test by VAPRT had a median VAPR 0.89 (mean 0.91±0.24). This value was significantly different from the other three possibilities, FP, TN, and FN (Table 1). Patients with TN tests had a median VAPR of 0.48 (mean 0.52±0.15), which differed from FP (median VAPR 0.70, mean 0.70±0.13 P<0.0001) but not from FN (median VAPR 0.57, mean 0.62±0.23). All test groups had VAPR values greater than 1.0, in this case VDP-VDPo exceeds the mean arterial pressure for the data obtained during treatment and can indicate a problem with needle placement or needle reversal.

[0113] Assessment of the VAPRT

[0114] FIG. 5 shows the study results of three months of VAPRT for January, February, and

March of 1999. In January 26 out of 112 patients (23%) had a positive VAPRT. During the next three months, thirteen of these patients (50%) experienced access failure, by month six the number increased to nineteen (73%) in the positive test group. For the January test, eight patients that tested negative went on to experience access failure (FN, 7% of population tested). The statistical analysis of the VAPRT are shown in Table 2 and represent the average at three and six months after each test. For the three month follow-up period, the mean test sensitivity of VAPRT was 70±8% while the specificity was 88±2%. These improved to a mean sensitivity of 74±5% and specificity of 96±3% for the six month follow-up period. The VAPRT positive predictive value was 84±10% and the negative predictive value 92±3% for the six month follow-up period.

[0115] DISCUSSION

[0116] The location of an access stenosis, in part, determines the ability of a monitoring system to detect the lesion. In most grafts, a stenotic lesion develops in the region of the venous anastomosis (10, 11, 12, 13). A stenosis in this region or in the central vein impedes blood flow through the access and increase VAP, which is observed as an increase in VDP. VDP measured during treatment is the sum of three components; the pressure created by blood flowing through the tubing and the needle (VAP₀), the static pressure created by the difference in height between the access site and the venous pressure transducer in the dialysis machine and VAP. VDP varies with treatment Qb, VAP, and hematocrit. The difference in height between the access site and the venous pressure transducer also varies, but, in most cases, does not differ by more than 5 cm from the value of 17 cm used in the model. This results in a ±5.1 mmHg variation in VAP and at MAP=100 mmHg a ±0.05 variation in VAPR. VAP also varies with the MAP and changes in MAP are reflected in VDP. Mapping of the access pressure gradient from the arterial to the venous anastomosis has shown that the slope of the mid graft pressure gradient increases with the development of a stenosis (11). Therefore, VDP increases with increasing distance between the venous needle and venous anastomosis.

[0117] Initially it appears that values of VAPR exceeding 1.0 are biologically impossible; however, all tests groups had some VAPR values>1.0, reflecting that physiologically calculated VAP exceeded MAP. For the VAPR data presented in FIG. 4, 9.8% of all values were >1.0, with 27.9% of these in the TP group. Several conditions lead to higher than expected VAPR values. Reversal of arterial and venous needles is probably the most common and occurs in as many as 25% of treatments (18). If a smaller diameter needle is used, without indicating the change in the patient's treatment data, the VAPR values will be falsely elevated. It can also be noted that the small diameter of the venous needle creates turbulent flow in the access that increases resistance to flow through the access. The degree of turbulent flow increases when access flow is reduced due to a venous stenosis and results in increased flow resistance and increased VAP. Lodgment of the venous needle against or partially in the access wall (reduces the needle orifice) or a venous line obstruction produces an increase in the measured VDP and results in episodic high VAPR values. Finally, a difference in MAP in the access extremity from that of the non-access arm that is typically used to monitor blood pressure during hemodialysis (19), which results in an increase in VAPR.

[0118] To reduce errors in the VAPRT, patient VAPR values must exceed 0.55 for three consecutive treatments. Initial dynamic access pressure testing developed by Schwab used three consecutive treatments that exceeded predefined limits to indicate a positive test. Dialysis treatments at the end of the month were selected for evaluation because the test results were included in a monthly dialysis patient report and patients may have had an access intervention during the early part of the month. The objective was to maintain a minimal false positive rate to prevent unnecessary further evaluation of the patient's access.

[0119] FIG. 2 illustrates the problems that must be resolved when using dynamic measurements of VDP to monitor access pressure. As blood flow increases VDP increases, primarily attributed to augmented resistance created by the venous needle. Elevation of hematocrit also increases VDP. The variability in VDP values from Qb and hematocrit can be reduced if the measurements are made at a fixed, relatively low, blood flow, as demonstrated by Schwab et al (1).

However, the appropriate warning level for VDP varies among individuals depending on the MAP and hematocrit. For example, with a 15 gauge needle and Qb=200 ml/mm, VDPQ is 33 mmHg at hematocrit 20% and 42 mmHg at hematocrit 36%. Using the criteria that a patient is at risk when the access pressure ratio >0.55, a patient with a MAP of 120 mmHg requires an access pressure >66 mmHg (66/120=0.55) to receive a warning for that treatment. Therefore at Qb=200 ml/mm, the

VDP warning level is between 99 (=33+66) mmHg and 108 (=42+66) mmHg for a patient when hematocrit varies between 20% and 36%. Applying the same criteria, a patient with MAP=75 mmHg needs a VDP warning level between 74 and 83 mmHg. It then becomes difficult to select a single VDP warning value for patients at risk for VDP between 74 and 108 mm Hg. By using

Equation (2) to calculate VAPR, the VAPRT adjusts the VDP warning level for each access pressure measurement in relationship to Qb, hematocrit and MAP. Notably, this absolute pressure range of 74 to 108 mm Hg is significantly lower than that originally reported by Schwab et al (1). The major reason for this difference is needle gauge, 15 gauge for the present invention versus 16 gauge for the Schwab investigation. The component of VDP due to flow through the needle is expected to be significantly greater with a 16 gauge needle (6). Presently, the algorithm is limited to 1 inch 15 gauge needles for cannulation until investigation of other needle gauges has been carried out.

[0120] An alternative method of determining the VAPR is to monitor static venous pressures and calculate a static venous access pressure ratio (SVPR) to test for a functionally significant stenosis (8). SVPR is an accurate method for access monitoring, however this method involves training of hemodialysis staff and ongoing monitoring to ensure the validity of the data. The VAPRT does not require specific training and the algorithm examines data currently entered in the patient database and evaluates the patient's access for each dialysis treatment. Finally another method measures static intra-access pressures directly prior to hemodialysis using a hydrophobic filter (22).

[0121] A stenosis on the arterial input side of the access or within the access itself is not detected by the VAPRT because this type of lesion reduces access flow and venous access pressure simultaneously. It is feasible to detect an arterial stenosis by developing a model that examines pre-pump arterial drip chamber pressure (ADP) for values more negative than usual. It is also possible to determine the existence of intra-access lesions if arterial intra-access pressure and VAP can be determined. In this regard, Polaschegg and colleagues (20) described a method for detecting and locating an access stenosis using dynamic arterial and venous access pressure measurements.

[0122] Access flow measurements performed within the dialysis unit can determine whether there is a clinically significant reduction of access flow, indicating the necessity for intervention. However, the location of the flow obstruction cannot be definitively identified. The disadvantages of flow measurements are that they require costly equipment, trained personnel and dialysis time for setup and measurement. Studies by Paulson et al. (17, 21) indicate that a single access flow measurement is a relatively poor indicator of graft failure. To achieve a sensitivity of 80% for predicting thrombosis requires an unacceptably high FP rate of 58%. The FP rate is so high because the threshold access blood flows that are used to predict graft failure often include many grafts that function at low blood flows, on the other hand, some grafts with good flows inexplicably thrombose without any warning.

[0123] Analysis of the data demonstrated that at a sensitivity of 80% the FP rate was 34% for testing grafts. A low FP rate (20% for grafts) was maintained in order not to produce a large number of evaluations that results in interventions by either vascular surgeons or interventional radiologists. It has been suggested that trend analysis can be a better predictor of access failure when using access flow. Trend analysis requires more frequent flow measurements and greatly increases the cost of access flow measurements. The VAPRT calculates a VAPR for each dialysis treatment, rendering it ideal for trend analysis. The current VAPRT models the VAPR trend after the eighth treatment of a month. To minimize spurious alarms, a triplet rule was imposed whereby three consecutive treatments with VAPR>0.55 were necessary to elicit a warning of impending graft failure, and this rule is currently being applied to generate an end-of-month report to assist clinicians in identifying patients with grafts at risk for dysfunction. It is possible to improve the VAPRT test if trend analysis of the all data is included in the algorithm. Greater emphasis can be placed on temporal trends or data filters imposed to exclude clearly erroneous measurements. In addition, analysis of data from two or more consecutive months can increase the power to detect access dysfunction.

[0124] The results of this study demonstrate that the VAPRT is a useful noninvasive screening test that identifies a population of dialysis patients that is at risk for access failure. The key component in implementing this system is computer access to the required treatment and laboratory data. The software algorithm to analyze hemodialysis data is incorporated as a standard end-of-month report and as an Internet-based accessible vascular access monitoring system. All patients exhibiting a warning status are flagged and a database trigger is available on-demand to create a report for any location or time period. Access intervention can be tracked along with warning status, thus permitting immediate follow-up and timely cost-saving interventions.

[0125] Example 2

[0126] An alternative method is provided for measuring access pressure through an access needle that is flow-connected to the vascular system of a patient. The method comprises the steps of: connecting one end of pressure tubing to the outer end of the access needle tubing, with a membrane blocking the flow of blood while permitting the passage of air through to a pressure gauge. The membrane suppresses or dampens the pressure pulses or oscillations through the tubing. Thus, upon opening the access needle tubing to the vascular system, blood flowing into the tubing compresses the air in the pressure tubing, plus the connected gauge, causing pressure from the vascular system to be readable by the gauge while the pressure pulses are attenuated in a simple, nonelectronic manner.

[0127] The "membrane" mentioned above may be a microporous membrane, typically a microporous block or plug positioned within or adjacent to the pressure tubing and capable of providing the damping or attenuation of the pulsatile nature of the pressure from the patient's cardiovascular system at the gauge.

[0128] According to an aspect of the present invention, the internal volume of the pressure tubing is less than the internal volume of the access needle tubing. As the result of this, pressurized blood entering the empty access needle tubing as the pressure is read does not advance completely to the level of the membrane, but is halted by compression of the initial air in the tubing, as well as the residual volume of air within the pressure gauge. This can be accomplished by providing pressure tubing that has a connector at each end, the tubing having a single lumen of reduced diameter from normal flexible tubing, which lumen diameter is typically no more than about one third of the outer diameter of the tubing. Thus, the internal volume of the pressure tubing can be less than the internal volume of the first tube even if the length of the pressure tubing is greater than the length of the first tube, this situation is preferred so that there is adequate tube length to conveniently hold a pressure gauge and to position it at approximately the level of the patient's heart and to read it with ease, and also to reduce the chance that the access needle connection to the patient's access is disturbed as the pressure gauge is connected and handled.

[0129] The set that defines the pressure tubing may carry a microporous member that is capable of preventing the passage of bacteria therethrough. This can be a second microporous member if desired, above and beyond the microporous plug described above that suppresses pressure oscillations through the pressure tubing, thus attenuating the pressure pulses. A conventional 0.2 micron bacterial filter can be used. This uniquely provides both flow blocking and aseptic conditions with commercially available materials. [0130] Alternatively, the microporous member can be a plug that has a bacteria blocking capability similar to conventional 0.2 micron bacterial filters. Also, a membrane -type bacterial filter can have pores that are small enough to provide the desired attenuation of pressure pulses through the pressure tubing, to facilitate reading of the gauge.

[0131] Also, if desired, the pressure tubing can have a bore that is sufficiently narrow and of a length to provide the desired pressure pulse attenuation through the tubing without the need for a porous plug so that, typically, only a bacteria blocking filter membrane is provided, as needed, to protect the patient from bacterial contamination through connection to a nonsterile pressure gauge.

[0132] Further development of the device includes replacement of the pressure gauge with a handheld microprocessor controlled device that measures and records the pressure measurements. An algorithm in the device calculates the average pressure over a predetermined sampling period. The device also contains a computer database to recall individual patient information and to record current pressure measurements in the patient's database record. Data from the device can be transferred via a communication port to a larger computer system with a more extensive patient database.

[0133] Example 3

[0134] This example demonstrates the case where blood flow (Qb) is equal to zero in

Equation (3). The constant term (-17.32509 in Equation (3)) needed to correct for the difference in height between the venous drip chamber and the level of the patient's access site was calculated for three different dialysis machines and clinical data was evaluated to demonstrate the effectiveness of the system.

[0135] The measurement of venous intra-access pressure (VAP) normalized by mean arterial blood pressure (MAP) facilitates detect venous outlet stenosis and correlates with access blood flow.

General use of VAP/MAP is limited by time and special equipment costs. Bernoulli's equation relates differences between VAP (recorded by an external transducer as PT) and the venous drip chamber pressure (VDP) at zero blood pump flow, the difference in height (ΔΗ) between the measuring sites and fluid density determine the pressure due to the difference in height ΔΡΗ-VAP- VDP. They were therefore correlated VDP and PT measurements at six different dialysis units each using one of three different dialysis machines. Both dynamic (i.e. with blood flow) pressures and static pressures were measured. Validation studies showed that changes in mean blood pressure, zero calibration errors, and hydrostatic height between the transducer and drip chamber accounted for 90% of the variance in VDP with ΔΡΗ=-1.6+0.74*ΔΗ (r=0.88, p<0.001). The major determinant of static VAP/MAP was access type and venous outflow problems. In grafts, flow averaged 555±45 mL/min for VAP/MAP>0.5 and 1229±112 mL/min for VAP/MAP<0.5. ΔΡΗ varied from 9.4 to 17.4 mm Hg among the six centers and was related to ΔΗ between the drip chamber and the arm rest of the dialysis chair. Concordance between the values of VAP/MAP calculated from PT and from VDP+PH was excellent. It was concluded that static VDP measurements corrected by an appropriate ΔΡΗ can be used to prospectively monitor prosthetic bridge grafts for stenosis.

[0136] Throughout this application, various publications, including United States patents, are referenced by author and year and patents by number. Full citations for the publications are listed below. The disclosures of these publications and patents in their entireties are hereby incorporated by reference into this application in order to more fully describe the state of the art to which this invention pertains.

[0137] While exemplary embodiments are described above, it is not intended that these embodiments describe all possible forms of the invention. Rather, the words used in the specification are words of description rather than limitation, and it is understood that various changes may be made without departing from the spirit and scope of the invention. Additionally, the features of various implementing embodiments may be combined to form further embodiments of the invention. TABLE 1

Comparison of Monthly Mean Graft VAPR Values for the Different Test Groups

Count Mean Std. Dev. Std. Err

True Positive 27 0.909 0.237 0.046

True Negative 67 0.515 0.149 0.018

False Negative 9 0.616 0.215 0.072

False Positive 14 0.698 0.125 0.033

Mean Difference p-Value

True Positive, True Negative 0.394 O.0001

True Positive, False Negative 0.293 0.0024

True Positive, False Positive 0.211 0.0036

True Negative, False Positive -0.183 <0.0001

True Negative, False Negative -0.102 0.0734

False Positive, False Negative 0.082 0.2595

TABLE 2

Statistical Analysis of Venous Access Pressure Ratio Test for Grafts

Showing Mean Values for Three Months of Testing

Test Period

0-3 mo 0-6 mo

Sensitivity (%) 70 ± 8 74± 5

Specificity (%) 88 ±2 96 ±3

Positive Predictive Value (%) 52 ± 10 84 ± 10

Negative Predictive Value (%) 94 ± 2 92 ± 3

False Positive rate (%) 12±2 4 ±3

References

1. Schwab SJ, Raymond FR, Saeed M, Newman GE, Dennis PA, Bollinger RR: Prevention of hemodialysis fistula thrombosis. Early detection of venous stenosis. Kidney lnt 36:707— 711, 1989.

2. Strauch BS, O'Connell RS, Geoly KL: Forecasting thromboses of vascular access with Doppler color flow imaging. Am J Kidney Dis 19:554-557, 1992.

3. Levy SS, Sherman RA, Nosher JL: Value of clinical screening or detection of asymptomatic hemodialysis vascular access stenoses. Angiology 43:421-424, 1992.

4. Van Stone JC, Jones M, Van Stone J: Detection of hemodialysis access outlet stenosis by measuring outlet resistance. Am J Kidney Dis 23:562-568, 1994.

5. Rehman SU, Pupim LB, Shyr Y, Hakim R, lkizler TA: lntradialytic serial vascular access flow measurements. Am J Kidney Dis 34:471-477, 1999.

6. Besarab A, Sullivan KL, Ross R, Moritz M: The utility of intra-access monitoring in detecting and correcting venous outlet stenoses prior to thrombosis. Kidney Int. 47: 1364-1373, 1995.

7. Koksoy C, Kuzu A, Erden I, Turkcapar AG, Duzgun I, Anadol E: Predictive value of color Doppler ultrasonography in detecting failure of vascular access grafts. Brit J Surg 82:50-55, 1995.

8. Besarab A, Al-Saghir F, Alnabhan N, Lubkowski T, Frinak S: Simplified measurement of intra- access pressure. ASAIO J 42:M682-M687, 1996.

9. Sands JJ, Miranda CL: Prolongation of hemodialysis access survival with elective revision. Clin Nephrol 44:334-337, 1995.

10. Palder SB, Kirkman RL, Whittemore AD, Hakim RM, Lazarus JM, Tilney NL: Vascular access for hemodialysis. Patency rates and results of revision. Ann Surg. 202:235-239, 1985. 11. Sullivan KL, Besarab A, Bonn J, Shapiro MJ, Gardiner GA, Moritz MJ: Hemodynamics of failing dialysis grafts. Radiology 186:867-872, 1993.

12. Beathard GA, Percutaneous transvenous angioplasty in the treatment of vascular access stenosis. Kidney International. 42(6): 1390-7, 1992.

13. Kanterman RY, Vesely TM, Pilgram TK, Guy BW, Windus DW, Picus D: Dialysis access grafts: anatomic location of venous stenosis and results of angioplasty. Radiology. 195(1): 135-9, 1995.

14. Besarab A, Dorrell 5, Moritz M, Michael H, Sullivan K: Determinants of measured dialysis venous pressure and its relationship to true intra-access venous pressure. Trans Am Soc Artif Intern Organs 37:M270-M271, 1991.

15. Sparks SR, VanderLinden JL, Gnanadev DA, Smith JW, Bunt TJ: Superior patency of perforating antecubital vein arteriovenous fistulae for hemodialysis. Annals of Vascular Surgery. 11(2): 165-7, 1997.

16. Metz CE: Basic principles of ROC analysis. Semin Nuclear Med. 8:283-98, 1978.

17. Paulson WD, Ram SJ, Birk CG, Work J: Does blood flow accurately predict thrombosis or failure of hemodialysis synthetic grafts? A meta-analysis. Am J Kidney Dis 34(3):478-85, 1999.

18. Shapiro W, Gurevich L: Inadvertent reversal of hemodialysis lines— A possible cause of decreased hemodialysis efficiency. [Abstract] J Am Soc Nephrol 8: 172 A, 1997.

19. Besarab A, Lubkowski T, Yu A, Frinak S. Determinants of vascular access flow. ASAIO J 47(5):501-506, 2001.

20. Polaschegg HD, Techert F, Wizemann V: Dynamic pressure measurement for detection of blood access stenosis. Edtna-Erca J 24(4):39-44, 1998. 21. Paulson WD, Ram SJ, Birk CG, Zapczynski M, Martin SR, Work J: Accuracy of decrease in blood flow in predicting hemodialysis graft thrombosis. Am J Kidney Dis 35(6): 1089-1095, 2000.

22. Besarab A, Lubkowski T, Frinak 5: A simpler method for measuring intra-access pressure. J Am Soc Nephrol. 11 :202A, 1999.

Claims

WHAT IS CLAIMED IS:

1. A detection device for determining intravascular pressure and irregular needle placement in a hemodialysis procedure, the detection device comprising:

an analyzer for automatically deriving venous access pressure (VAP) at a location of venous needle insertion into the patient and determining a ratio of VAP to mean arterial pressure, where the analyzing means determines multiple VAP values over multiple time periods and determines multiple ratios, one for each VAP value; and

a comparing device for comparing each ratio to a predetermined standard ratio derived from a moving average of prior VAP ratios, where the comparing device indicates irregular needle placement when elevation of the ratio above the standard ratio is detected.