METHOD FOR PROCESSING INFUSION DATA AND AN INFUSION PUMP SYSTEM

Abstract

A distributed infusion pump system is provided with long-term recording of the type of medication, nutrition or hydration, pump output pressure, corresponding infusion flow, infusion flow resistance, and an automatic or manual alarm in the event that thresholds applied to these parameters or their derivatives are exceeded.

Claims

1-24. (canceled)

25. A system for processing infusion data, the system comprising: an infusion pump; one or more sensors; and a computer system operatively coupled to the infusion pump, the computer system configured to: receive downstream pressure measured by the one or more sensors and determining downstream pressure measurements per infusion and per rate; provide downstream pressure statistics based on the downstream pressure measurements per infusion and per rate, wherein the downstream pressure statistics comprise a resistance of flow A and a time constant T; store, in a database, the downstream pressure statistics and linking the stored downstream pressure statistics with infusion attributes; link the infusion attributes with respective health events, such that the database includes a plurality of patient and infusion data over a time period based on the attributes, the health events, and the statistics; query the database to determine respective infusion pump limits for warning and/or alarm for each of a health event type, wherein the infusion pump limits for each of the health event type are based on the infusion attributes, patient data from a plurality of patients, and infusion data from a plurality of infusions using the infusion pump over a time period, wherein an infusion pump limit is temporally reached prior to an occurrence of its corresponding health event type; and generate the warning and/or alarm based on a first infusion pump limit for a first health event type.

26. The system of claim 25, wherein the resistance of flow A is calculated based on $A=\frac{\left(P2-P1\right)*V_w}{\left(F2-F1\right)*V_s},$ wherein, P1 is a first pressure, P2 is a second pressure, F1 is a first flow, F2 is a second flow, V.sub.w is a water viscosity, and V.sub.s is a drug viscosity.

27. The system of claim 25, wherein the time constant T is determined after the resistance of flow A reaches a predetermined limit at which the infusion stops during the measurement and continues or waits for alarm release.

28. The system of claim 25, wherein the time constant T is determined after the pressure reaches a predetermined downstream occlusion level at which the infusion stops during the measurement and continues or waits for alarm release

29. The system of claim 25, wherein the time constant T is determined after a bolus end, and the infusion is stopped until the time constant T is determined.

30. The system of claim 25, wherein the resistance of flow A is determined from different infusions with different rates and drugs administered to a patient.

31. The system of claim 25, wherein the computer system is configured to measure a zero rate downstream pressure before a start of an infusion.

32. The system of claim 25, wherein the attributes are grouped into subcategories.

33. The system of claim 25, wherein the computer system is configured to distribute the warning and/or alarm via a communication means.

34. The system of claim 33, wherein the communication means comprises at least one of a message on a web page, a mobile phone message, an email, or an alarm at the pump.

35. The system of claim 25, wherein the downstream pressure statistics comprise a maximum pressure, an average pressure, and a maximum pressure difference, each corresponding to measuring intervals that vary per therapy.

36. The system of claim 25, wherein the infusion attributes comprise a date, a patient type, a name, a delivery route, a catheter type, a needle type, a drug, and a therapy.

37. The system of claim 25, wherein the health event is triggered by at least one of a catheter block, an infiltration, or an intraneural placement.

38. The system of claim 25, wherein the computer system is configured to compare the downstream pressure measurements with the infusion pump limit for indicating a risk of blockage of the infusion, the infusion pump limit being lower than a value indicating an actual blockage.

39. The system of claim 38, wherein the infusion pump limit is updated based on statistical data of the downstream pressure measurement and/or feedback information.

40. The system of claim 38, wherein the computer system is configured to trigger a blockage alarm if a pressure value is greater than or equal to the infusion pump limit.

41. A method for processing infusion data the method comprising: receiving downstream pressure measured by one or more sensors and determining downstream pressure measurements per infusion and per rate; providing downstream pressure statistics based on the downstream pressure measurements per infusion and per rate, wherein the downstream pressure statistics comprise a resistance of flow A and a time constant T; storing, in a database, the downstream pressure statistics and linking the stored downstream pressure statistics with infusion attributes; linking the infusion attributes with respective health events such that the database includes a plurality of patient and infusion data over a time period based on the attributes, the health events, and the statistics; querying the database to determine respective infusion pump limits for warning and/or alarm for each of a health event type, wherein the infusion pump limits for each of the health event type are based on the infusion attributes, patient data from a plurality of patients, and infusion data from a plurality of infusions using the infusion pump over a time period, wherein an infusion pump limit is temporally reached prior to an occurrence of its corresponding health event type; and generating the warning and/or alarm based on a first infusion pump limit for a first health event type.

42. The method of claim 41, wherein the resistance of flow A is calculated based on $A=\frac{\left(P2-P1\right)*V_w}{\left(F2-F1\right)*V_s},$ wherein, P1 is a first pressure, P2 is a second pressure, F1 is a first flow, F2 is a second flow, V.sub.w is a water viscosity, and V.sub.s is a drug viscosity.

43. The method of claim 41, wherein the time constant T is determined after the resistance of flow A reaches a predetermined limit at which the infusion stops during the measurement and continues or waits for alarm release.

44. The method of claim 41, further comprising comparing the downstream pressure measurements with the infusion pump limit for indicating a risk of blockage of the infusion, the infusion pump limit being lower than a value indicating an actual blockage.

Description

[0024] In the following, preferred embodiments of the present invention will be described with reference to the accompanying drawings, wherein

[0025] FIG. 1 is a graph showing a typical long term pressure recording in parenteral nutrition for various fluids,

[0026] FIG. 2 is a graph showing a transition from a first flow and pressure to a second flow and pressure,

[0027] FIG. 3 shows a typical infusion drug delivery chain,

[0028] FIG. 4 is a graph showing a loading dose and intraneural pressure long drop, and

[0029] FIG. 5 is a graph showing a standard subcutaneous loading dose pressure transient.

[0030] FIG. 3 schematically shows a typical drug delivery chain comprising an infusion set 1, a connector 2, a catheter 3 and a tissue or vein 4 wherein the connector 2 couples the infusion set 1 with the catheter 3 which is introduced into the tissue or vein 4.

[0031] In parenteral nutrition, there are different types of infusion, i.e. for main nutrition and for hydration or combination of both (dilutions). For nutrition the average flow speed for adults is 120 ml/hr, and for hydration it is the maximum pump flow, e.g. 400 ml/hr. Typical parenteral nutrition pressure graphs are shown in FIG. 1. For a long term recording, the pump initially records statistical information about downstream pressure of the infusion such as average pressure, maximum pressure and maximum pressure difference, together with how many peaks happened, and duration per peak, for the entire infusion. When connected to a so-called distributed system as described in U.S. Pat. No. 8,551,038, pressure interval (one hour), statistics and flow profile data are uploaded to a telemedicine server. Pressure statistics include a calculated flow resistance A and a transient time constant τ. Average can be a real average or the result of digital filtering on the pressure data. Pump system (pump screen or web server) can display graphs of statistical curves, or warn the user if abnormal peaks or trends are found.

[0032] The system makes time-stamped recordings of the pressure statistics and/or rates, so that a physician or an automated decision-making system evaluates the increase of pressure below occlusion and consequently informs the patient, who can also monitor the pressure statistic curves on the pump. Thanks to local and remote long term recording and the display of pressure statistics, surgical intervention can be avoided. A clinical trial in a hospital has shown that the recording of the statistics on output pressure for each infusion rate in each patient demonstrates the catheter's tendency to block (several months recording).

[0033] In intravenous infusions such as a parenteral nutrition, pressure is usually between 0.1 and 0.5 bar. Pressure difference peaks of more than 0.5 bar and, hence, below occlusion level have showed Superior Vena Cava stenosis needing venoplasty.

[0034] Resistance of flow in an open end tube is A=P/F where P is a steady state pressure for a flow (infusion rate) F, wherein a general tendency for block, i.e. an increase of resistance of flow such as an increase of pressure at a constant flow, can be observed. In the parenteral nutrition example above, said pressure statistics comprise such resistance of flow.

[0035] In the prior art, resistance of flow at steady state of intravenous infusions is calculated by the formula

[00001]$A=\frac{P2-P1}{F2-F1}.$

[0036] FIG. 2 shows a transition from flow F1 and pressure P1 to flow F2 and pressure P2.

[0037] In another embodiment of the present invention, pressures and corresponding flows recorded in nutrition and hydration (or under different speeds for the same liquid) are used to calculate the resistance of flow giving indications on the catheter condition. So, the resistance of flow A is calculated from different pressure and flow which occur with the different infusions and/or even on different days, and this is new in the art. With respect thereto, it is referred to FIG. 1 which as an example shows graphs representing three different flows F1, F2 and F3 under different pressure P1, P2 and P3, respectively. Due to the differing viscosity of different fluids, the flow resistance calculated is not the real figure, but its trend over time reveals whether or not there is a verge of blockage and the catheter needs cleaning. To compensate this, normalized formula for resistance of flow is calculated as

[00002]$A=\frac{\left(P2-P1\right)*\mathrm{Vw}}{(\left(F2-F1\right)*\mathrm{Vs}}$

where viscosity of water Vw and viscosity of drug Vs have been added which for drugs with high viscosity as immunoglobulin is important.

[0038] In another embodiment of the present invention, the pump can run a slow flow plus a fast flow and calculate the flow resistance, at the beginning or the end of the infusion wherein the end seems best because sugar intake has been completed and the adult patient is not at risk, i.e. flow resistance calculated within one single infusion.

[0039] The local display at the pump or the remote, mobile or internet application, displays the pressure statistics and corresponding flow and also potentially the flow resistance in a diagram chronologically where the trend appears. The derivative of the pressure curve over time, i.e. the pressure's tendency to increase over time (days, weeks, months or years), is also displayed, for easy computation by an automated system, the patient or the physician, so as to create a figure or drawing of conclusions regarding the potential blockage of the catheter in the near future. The derivative during and after a bolus (high rate specific volume infusion) gives useful information especially at subcutaneous infusions, where cavities (very slow change) or dead space and/or high change (inside a nerve) can be detected.

[0040] The system according to the present invention can also be used in cases other than central venous catheters, e.g. for Duodopa infusions in patients with Parkinson's disease, to prevent the blockage of duodenal stomach or peritoneal catheters, in infusions for peripheral analgesia where the involuntary relocation of the catheter near the nerve shows a sudden change in the output pressure, translated into alterations in the analgesia, so that changes to the anaesthetic dosage are required.

[0041] The resistance of flow is independent of flow. Bolus time may not be generally enough to reach a pressure steady state for the flow rate, so that a measurement of transient fall times is taken for calculating a time constant τ.

[0042] The flow resistance is not only measured in subcutaneous, epidural and intrathecal infusions, as pressure is built up relatively fast after a flow step, whereas a decay is relatively slow. So, according to the present invention, a transient response to flow and/or pressure only in the pressure decay after a high rate down to zero rate is measured.

[0043] Some measured curves in the data have an almost linear slope and some are exponential as shown as an example in FIGS. 4 and 5. So, the time constant in the decay from step to zero rate is measured wherein the found curve is always of the same type. This eliminates another type of error, i.e. the assumption of an ideal flow source in the prior art model, as the flow rate is not ideal, resulting in the introduction of errors. Zero rate is selected so that no forced flow intervenes in the flow decay, also because the steady state pressure of the asymptote Pz is known. This can be the medium pressure measured when after purge the pump is connected to the patient, just before the infusion starts, or measured close to a low infusion rate before a high flow step (bolus/loading dose). So, the transient time constant and resistance of flow are calculated easily and relatively fast.

[0044] This measurement is also used in intravenous infusions, stopping for some seconds a relatively high rate (above 100 ml/hr) usual as in intravenous infusions.

[0045] In subcutaneous infusions, such as regional analgesia, with a catheter placed in surgery, the recording time interval can be 30 seconds, and the transient pressure after a maximum rate bolus is recorded followed by recording the decay of pressure at a generally low following basal rate, to show if the catheter is placed into a nerve which is potentially harmful to the patient. The pressure rise and fall times are different depending on the placement of a peripheral nerve blocking catheter in surgery. This can give information regarding a reposition of the catheter and the avoidance of complications or a prediction for a based infusion and bolus needed, when the patient goes home. FIG. 5 shows a normal transient response during and after a loading dose and bolus; and FIG. 4 shows an intraneural transient response i.e. with limited space and elasticity so that the pressure rises abruptly and decays slowly.

[0046] The decay initial linear portion of the exponential gives points (P,t) defining through linear regression (least squares) a slope angle α (FIG. 5) which in theory (initial slope method) intersects the steady state pressure Pz at time point τ. From that triangle τ calculated by the formula τ=(P−Pz)*sin(α) where P is the starting pressure before decay, when zero infusion rate starts, and Pz is the zero rate pressure known as above. This measurement can last a few seconds. Alternatively for fast decaying curves, point (τ, 37% (P−Px)) defines τ with the pressure Pt=37% of (P−Pz) when the time point t=τ is constant.

[0047] As described above, according to the present invention, means are provided for recording a downstream pressure statistics of an infusion at specific time intervals, with mean, maximum and minimum (or maximum difference) pressure within the interval, wherein the intervals depend on the therapy type, and a calculation of statistical methods so to extract useful medical information, such as resistance of flow A=P/F at steady state, and formula A when pressure and flow differences and a transient time constant τ have been obtained, and where said recording can be provided for more than one infusions on the same patient and same catheter, means for displaying such raw and calculated statistics (in contrast to also uploaded and recorded feedback from reported problems) so that medically useful information can easily be deducted from an automated system or the doctor wherein this information gives parameters for the user to define limits per therapy, and wherein a knowledge base is structured by user feedbacks on limits used and false or correct alarms given by the system so that continuous improvement of results can be achieved through telemedicine.