Proxima Technical Specification Patient-Dedicated Blood Gas, Electrolyte and Glucose Analyser

Transcription

1 Proxima Technical Specification Patient-Dedicated Blood Gas, Electrolyte and Glucose Analyser

2 Sphere Medical Ltd No part of this document may be reproduced or transmitted in any form or by any means, electronic or mechanical, for any purpose without the express permission of Sphere Medical Ltd. UK & International Patents have been granted and are pending on the Proxima System. Contacts and Support The Sphere Medical team is dedicated to providing excellent support. For all technical and safe use questions relating to this document or the Proxima System, contact us at: Sphere Medical Ltd. Harston Mill Harston, Cambridge CB22 7GG United Kingdom Tel: +44 (0) Proxima Technical Specification 2

3 1.1 Technical Specification Envrionmental conditions Proxima System Operating Temperature Proxima System Operating Relative Humidity Proxima System Operating Pressure Range Proxima Monitor Storage and Transport Temperature Storage and Transport Temperature for singleuse components Operating light limits for Proxima System 15 o C to 30 o C 20-60% non-condensing kpa +5 o C to +25 o C (storage), -10 o C to +35 o C (transport) +5 o C to +25 o C (storage), -10 o C to +35 o C (transport) 20 Lux to 5089 Lux Other specifications Protection against electric shock (The power supply must be earthed when in use) Protection against harmful ingress of water or particulate matter Power Supply Unit Electrical Requirement Means of isolating equipment from the supply mains Monitor dimensions Instrument Weight Proxima Monitor externally powered by a Class I Power Supply Unit or internally powered. Power Supply Unit: Class I Proxima Monitor with integrated Power Supply Unit: IPX1 Proxima Sensor Cable: IP2X (IPX4 when mated) Input Voltage: V Input Frequency: Hz Input Current: 1.5A-0.7A Mains plug or appliance inlet are used to disconnect device and must remain readily operable 226mm H x 304mm W x 174mm D 2.9 kg 3

4 1.2 Reference and Reporting Ranges The following tables list the reference ranges, units, reporting ranges and display resolutions. Each institution should establish its own reference range for diagnostic evaluation of patient results. Table 1: Reference Ranges 1,2,3,4,5 Parameter Reference Range ph po mmhg, 11 13kPa pco mmhg, kpa K mmol/l Hct 35-54% 4 chb mmol/l BE ±2 mmol/l 4 - HCO mmol/l 5 Glucose mmol/l 1 Katz A, Ferraro M, Sluss P M, and Lewandrowski K B. Laboratory Reference Values Case Records of the Massachusetts General Hospital New Engl J Med 2004; 351: Crapo R, Jensen R L, Hegewald aand Tashkin P. Arterial Blood Gas Reference Values for sea Level and an Altitude of 1,400 meters. Am J Respir. Crit. Care Med. Vol. 160 pp , Statland B.E. Clinical Decision Levels for Lab Tests. Oradell, NJ: Medical Economic Books Edwards Sharon L. Edwards; Pathophysiology of acid base balance: The theory practice relationship; Intensive and Critical CareNursing (2008) 24,

5 Table 2: Proxima Reporting Range Parameter Units Reporting Range Resolution ph ph H + nmol/l po 2 kpa mmhg pco 2 kpa mmhg K + mmol/l Glucose mmol/l mg/dl Hct % chb mmol/l g/l g/dl BE mmol/l ± HCO 3 - mmol/l NOTE: Haematocrit results are for indication only. 5

6 1.3 Precision in Blood Method A study was conducted for the analytes ph, pco 2, po 2, K +, glucose and Hct. For each analyte, bovine blood samples were prepared at 3 different levels of analyte. Two analyses at each analyte level were taken on 12 devices. After each replicate of a blood sample was analysed on the Proxima System, the same sample was analysed on the reference device (Siemens 1200 blood gas analyser and Hawksley Hematospin). The following tables summarise the results of the Proxima System recovery and inter-device precision testing. ProximaTechnical Description 6

7 1.3.2 Results NOTE: The data are an example of system performance when routine calibration and QC is carried out as specified in the Proxima instructions for use. * WRSD = within-run standard deviation ^ WRCV = within-run coefficient of variation Summary of example of ph precision and recovery in blood data: ph Level 1 Level 2 Level 3 n WRSD * Observed Expected Recovery % 100.4% 100.0% 99.9% WRCV ^ 0.26% 0.21% 0.16% Summary of example of K + (mm) precision and recovery in blood data: Level 1 Level 2 Level 3 n WRSD * Observed Expected Recovery % 98.5% 100.3% 100.1% WRCV ^ 4.51% 1.78% 3.30% K + 7

8 Summary of example of po 2 precision and recovery in blood data: po 2 Level 1 Level 2 Level 3 n WRSD * (mmhg) WRSD * (kpa) Observed (mmhg) Observed (kpa) Expected (mmhg) Expected (kpa) Recovery % 98.5% 103.4% 103.5% WRCV ^ 4.24% 5.31% 3.31% Summary of example of pco 2 precision and recovery in blood data: pco 2 Level 1 Level 2 Level 3 n WRSD * (mmhg) WRSD * (kpa) Observed (mmhg) Observed (kpa) Expected (mmhg) Expected (kpa) Recovery % 96.5% 94.7% 95.7% WRCV ^ 3.95% 3.70% 4.39% 8

9 Summary of example of glucose (mm) precision and recovery in blood data: Glucose Level 1 Level 2 Level 3 n WRSD * Observed Expected Recovery % 90.7% 95.5% 98.6% WRCV ^ 6.83% 5.88% 7.16% Summary of example of Hct (%) precision and recovery in blood data: Haematocrit Level 1 Level 2 Level 3 n WRSD * Observed Expected Recovery % 98.5% 98.9% 99.3% WRCV ^ 13.22% 9.94% 6.09% 9

10 1.4 Method Comparison The Method Comparison study compared the Proxima with the Siemens 1200 reference device. The Hawksley Hematospin was used as the reference instrument for haematocrit. The study was conducted in a laboratory setting with bovine samples. Results are summarized in the table below. The objective of the study was to compare the Proxima System and the reference device (Siemens 1200 and Hawksley Hematospin) for measurements on bovine blood of ph, pco 2, po 2, K +, glucose and Hct Summary Results Table NOTE: The data are an example of system performance when routine calibration and QC is carried out as specified in the Proxima instructions for use. Example performance data on method comparison is given in the tables below: N Slope Intercept RMSE * r 2 Min Max ph Potassium (mmol/l) po 2 (mmhg) po 2 (kpa) pco 2 (mmhg) pco 2 (kpa) Glucose (mmol/l) Hct (%) *RMSE = Root mean square error 10

11 1.5 Potential Interfering Substances Bench studies were conducted to look for potential interference by exogenous and endogenous substances and demonstrate specificity of the Proxima Sensor. The design of these studies was based on the CLSI guideline EP7 A2 Interference Testing in Clinical Chemistry; Approved Guideline-Second Edition. Summary of example of Interferent Test Data - ph Tested ph (ph units) Interference Interference level tested Effect of Interference (Test - control) Chloride 110mM Chloride 110mM Chloride 120mM Chloride 120mM Acetaminophen 200mM Acetaminophen 200mM Acetaminophen 1324mM Acetaminophen 1324mM Hct 28% 0.005* Hct 29% 0.031* Hct 53% 0.003* Hct 53% 0.032* Salicylic acid 2170mM Salicylic acid 2170mM Salicylic acid 4340mM Salicylic acid 4340mM Acetyl salicylic acid 2170mM 0.014* Acetyl salicylic acid 2170mM * Acetyl salicylic acid 3620mM * Acetyl salicylic acid 3620mM * * For these datasets, the effects of interference was calculated by normalising to the BGA (mean difference between Proxima and the BGA for the test pool, minus the mean difference between Proxima and the BGA for the control pool). This assumption is valid assuming minimal interference for the BGA. 11

12 Summary of example of Interferent Test Data - Potassium Target K + level (mm) Interference Interference level tested Effect of Interference (Test - control) 4.4 Chloride 110mM Chloride 110mM Chloride 120mM Chloride 120mM Hct 29% * 7.0 Hct 28% * 4.4 Hct 52% * 7.0 Hct 52% * * For these datasets, chloride spiking altered the potassium concentration. Therefore, effects of interference was calculated by normalising to the BGA (mean difference between Proxima and the BGA for the test pool, minus the mean difference between Proxima and the BGA for the control pool). This assumption is valid assuming minimal interference for the BGA. 12

13 Summary of example of Interferent Test Data - Carbon Dioxide Target CO 2 (mmhg) Interference Interference level tested Effect of Interference (Test - control) 42.2 Ibuprofen 340mM Ibuprofen 340mM Ibuprofen 2425mM Ibuprofen 2425mM Hct 29% 0.000* 71.0 Hct 28% * 42.0 Hct 52% 1.200* 71.0 Hct 52% * 42.2 Sodium pentothal 20.6µM Sodium pentothal 20.6µM Sodium pentothal 248µM Sodium pentothal 248µM * For these datasets, effects of interference was calculated by normalising to the BGA (mean difference between Proxima and the BGA for the test pool, minus the mean difference between Proxima and the BGA for the control pool). This assumption is valid assuming minimal interference for the BGA. Summary of example of Interferent Test Data Oxygen Target O 2 (mmhg) Interference Interference level tested Effect of Interference (Test - control) 28.3 Isoflurane 253mM Isoflurane 253mM Isoflurane 759mM Isoflurane 759mM Halothane 253mM Halothane 253mM Halothane 759mM Halothane 759mM Nitrous oxide 605.5mmHg Hct 28% * 84.0 Hct 29% * 28.0 Hct 53% 2.033* 84.0 Hct 53% * 13

14 Summary of example of Interferent Test Data - Glucose Target Glu (mm) Interference Interference level tested Effect of Interferent (Test-Control) 11.8 acetaminophen 1324µM acetaminophen 1324µM Acetyl Salicylic Acid 3620µM Acetyl Salicylic Acid 3620µM Dobutamine 0.32 µm Dobutamine 0.32 µm Dopamine 5.87µM Dopamine 5.87µM Ethanol 86.8mM Ethanol 86.8mM Heparin 3000 IU/litre Heparin 3000 IU/litre Salicylic acid 4340µM Salicylic acid 4340µM ascorbic acid 170µM ascorbic acid 170µM Bilirubin 342µM Bilirubin 342µM conjugated Bilirubin 342µM conjugated Bilirubin 342µM Haemoglobin 2g/l Haemoglobin 2g/l Urea 42.9mM Urea 42.9mM Uric acid 1.4mM Uric acid 1.4mM Triglyceride 37mM Triglyceride 37mM Oxygen 30mmHg Oxygen 30mmHg Haematocrit 20% Haematocrit 20% -0.1 The levels at which acetaminophen interferes have been calculated from the dose response study. It indicated that interference starts at 14

15 0.99mM acetaminophen for high glucose (12mM) and at 0.80mM for low glucose (5mM); compared to the FDA test guideline concentration of 1.324mM (toxic >1.324mM). However, it should be noted that 0.8mM acetaminophen is x4 the highest therapeutic dose of 0.20mM acetaminophen (EP07A2 therapeutic range mM). The study for oxygen depletion indicates that no depletion should be observed for glucose (<10.4mM) down to an oxygen concentration of 50 mmhg (6.7kPa). Summary of example of Interferent Test Data - Haematocrit Target Hct (%) Interference Interference level tested Effect of Interference (Test - control) 25-30% Intralipid 2.76 mmol/l % Intralipid 2.76 mmol/l % Human serum albumin 120 g/l % Human serum albumin 120 g/l % Potassium 7 mmol/l % Potassium 7 mmol/l Proxima Potential Interfering Substances The following substances have been shown to have a potential interference on Proxima sensors: Analyte Interferent Concentration at which interference is observed pco 2 Sodium pentothal >20.6 μmol/l Hct Sodium * 140 ± 17 mmol/l Glucose Oxygen < 80 mmhg/ 10.7 kpa Glucose Acetaminophen 0.8 mm * Significant interference by sodium occurs at sodium concentrations outside 140 ± 17 mmol/l where 140 mmol/l is the normal physiological concentration of sodium in a healthy person. The rest of the substances analysed have no potential interference on Proxima sensors. NOTE: The data are an example of system performance when routine calibration and QC is carried out as specified in the Proxima instructions for use. 15

16 1.6 Precision on Controls Precision on aqueous quality control materials was estimated using 12 Proxima Systems over their use life. Example performance data is given in the tables below. Analyte Level n Mean WRSD * Total SDt ^ ph ph Analyte Level n Mean WRSD * Total SDt ^ pco 2 (kpa) pco 2 (kpa) pco 2 (mmhg) pco 2 (mmhg) Analyte Level n Mean WRSD * Total SDt ^ po 2 (kpa) po 2 (kpa) po 2 (mmhg) po 2 (mmhg) Analyte Level n Mean WRSD * Total SDt ^ K + (mmol/l) K + (mmol/l) Analyte Level n Mean WRSD * Total SDt ^ Glucose (mmol/l) Glucose (mmol/l) * WRSD= within-run standard deviation ^ Total SDt= total standard deviation NOTE: The data are an example of system performance when routine calibration and QC is carried out as specified in the Proxima instructions for use. 16

17 1.7 Operating Principles of Proxima Overview The chemical measurement technologies used in the Proxima are based on electrochemical principles and involve the measurement of current, voltage and conductivity occurring in an amperometric, potentiometric or conductimetric sensor respectively. Each sensor is designed to selectively measure the activity of a specific analyte. The individual approaches used for each analyte are summarised in Table 4. Table 4: Transduction principles and summary descriptions of the chemical measurement technologies used in Proxima. Analyte Transduction principle Measuring device description ph Potentiometric ISFET pco 2 Potentiometric ISFET-based Severinghaus cell po 2 Amperometric Clark electrode Potassium Potentiometric ISFET Haematocrit Conductimetry Conductivity cell Glucose Amperometric Glucose oxidase biosensor The Proxima measurement system is fabricated as a single multi-analyte chip that carries an array of individual sensors, each of which measures a different analyte. The individual sensors use one of three transducer technologies: potentiometric (to measure the potential generated), amperometric (measuring the current generated) and conductimetry (measuring the conductivity). Potentiometric sensors Potentiometry is based on the principle that when a conductor is placed in an electrolyte solution, it will develop an electrical potential with respect to the solution. This is known as the half-cell potential, and is dependent on the electrode material and the composition of the solution. The potentiometry technique involves measurement of the difference in potential (voltage) between an electrode sensitive to the target analyte (sensing electrode) and one which has a half-cell potential independent of the target analyte (reference electrode). The half-cell potential can be strongly affected by any current passing through the electrode due to a phenomenon known as polarisation. It is therefore very important that negligible current passes through the device during measurement. This requires a high input impedance electrical circuit, which can be susceptible to electronic noise and drift. 17

18 Potentiometry is widely used for the measurement of the concentration or activity of ionic species. The activity of an ion solution, ai, generates a potential, E, at the ion sensitive surface, as described by the Nernst equation: E = E0 -RT/zF. ln(ai). where E0 is the potential under standard conditions, T is the temperature z is the charge on the ion and R and F are physical constants. From this it can be calculated that the potential should have a linear relationship to the log of the activity. For a theoretically ideal membrane and a monovalent ion, the signal response is just under 60 mv for each tenfold increase in activity at room temperature. Potentiometry is a well understood and widely used analytical technique based on the ion selective electrode (ISE). An ISE measures the potential generated across a membrane exposed to the target analyte. The oldest and best known ISE is the ph sensor based on a glass membrane. More recently, polymeric membranes have been developed that incorporate ionophores that render the membrane sensitive and specific to a wide range of target ionic species. The potentiometric sensors used in the Proxima are all based on ion selective field effect transistors (ISFETs), where the sensitivity and specificity of the ISE is refined to include the superior transduction properties of a field effect transistor (FET). The principles of ISFETS were first described in the early 1970 s. At the centre of the device is a semiconductor channel along which a given current, the drain current, is passed by applying a driving voltage between the source and drain regions. As an electric field is applied across this region, the conductivity of the channel changes in proportion to the strength of the field applied. As a consequence, the driving voltage required to pass the drain current changes in proportion to the field strength. In an ISFET, the electric field across the gate region is generated by the potential difference between the sensing electrode, known as the gate, and a reference electrode. The advantage of this approach over the traditional ISE is that the high impedance circuit between the reference and the gate is independent of the low input impedance circuit used for measuring the signal, reducing drift and noise. Furthermore, the device is simple and robust. 18

19 Amperometric sensors The amperometric sensors used in Proxima are based on conventional three electrode amperometric cells. The principle of amperometry is used to determine the concentration of an electroactive species by reacting it at a sensing electrode using an electrical potential applied by an external electrical circuit. This reaction will involve either gain or loss of electrons by the electroactive species, known as reduction or oxidation respectively. The resulting current passes through the working electrode. To complete the electrical circuit, a second electrochemical reaction must occur at a counter electrode and there must be an electrolyte solution path to allow ionic conduction between the electrodes. Under controlled conditions, the rate of the reaction of interest, as measured by the amount of current passed through the sensing electrode, is proportional to the concentration of electroactive species present. A third electrode, known as a reference electrode, is used to carefully control the potential applied to the sensing electrode. The amperometric transducers used in Proxima have platinum sensing and counter electrodes and a silver/silver chloride reference electrode. Conductimetric sensors In a conductivity measurement, an excitation signal is applied between two or more electrodes in contact with the test solution and the response signal is measured. These two signals take the form of sine waves and how well they align is used to determine the haematocrit of a blood sample. The conductivity of a solution depends on both the total ion concentration and ease of movement of the ions in the solution under the influence of the excitation signal. How easily the ions move and the ion concentration will affect the response signal. In a solution such as saline the response signal will align with the excitation. The conductimetric sensor used in Proxima applies an excitation signal to the test solution between three platinum disc electrodes spaced at intervals across the sensor array with one electrode at each end the third in the middle. The response signal is proportional to the conductivity of the solution under analysis. By applying the excitation between the central electrode and each of the other two electrodes in turn the sensor may be used to check for the presence of bubbles within the cell. 19

20 1.7.2 Hydrogen ion activity (ph) The Proxima System reports the concentration of hydrogen ions on the ph scale, where the ph can be calculated from [H+], the concentration of hydrogen ions in mmol/l. ph = -log[h+] Clinical significance The ph of arterial blood is normally kept within a narrow range around ph 7.4 by the buffer systems in the blood (bicarbonate, phosphate and proteins) closely controlled by respiratory and metabolic mechanisms. Many pathological conditions are accompanied by disturbances of the acid-base balance of the blood and measurement of blood ph is an important component in the determination of these disorders. Arterial blood ph of less than 7.35 is termed acidosis, which may have either respiratory or metabolic primary causes. Respiratory acidosis is caused by conditions that decrease elimination of carbon dioxide through respiration, leading to a decrease in blood ph. These conditions include: 1. Factors that depress the respiratory centre, such as narcotics and barbiturates as well as impairment of the central nervous system by trauma, infection or degenerative disorders. 2. Conditions that affect the respiratory apparatus, such as airway obstruction, chronic obstructive pulmonary disease, severe pulmonary infection or respiratory distress syndrome. Metabolic acidosis may be caused by a range of factors including: 1. Accumulation of organic acids in conditions such as lactic acidosis, diabetic acidosis or toxicity from methanol, or ethylene glycol. 2. Reduction in the excretion of inorganic acids due to renal failure. 3. Excessive loss of bicarbonate due to gastrointestinal fluid loss from diarrhoea or pancreatitis, or due to increased renal excretion (decreased tubular reclamation). 20

21 Arterial blood ph of greater than 7.45 is termed alkalosis, which may have either respiratory or metabolic primary causes. Respiratory alkalosis is caused by conditions that increase the elimination of carbon dioxide through rapid respiration, leading to an increase in blood ph. These conditions include: 1. Non-pulmonary stimulation of the respiratory centre e.g. due to hypoxia, therapeutics such as salycilates and catecholemines, anxiety or gram negative septicemia. 2. Pulmonary disorders including pulmonary embolism, atrial shunt and congestive heart failure (severe stages may cause acidosis). 3. Ventilator induced hyperventilation. The usual primary causes of a metabolic alkalosis are: 1. Excessive vomiting, leading to a significant loss of acids from the system. 2. Iatrogenic factors including antacids, bicarbonate containing intravenous fluids and sodium citrate overload from massive blood transfusion. Factors affecting results Adequate sample should be drawn to ensure a representative arterial blood sample is measured. The ph sensor The Proxima analyser uses an ISFET sensor for the measurement of ph. The sensor array chips are fabricated with an intrinsically ph sensitive silicon nitride gate layer. A silver-silver chloride electrode is used as the reference electrode. All sample measurements for ph are taken and reported at 37ºC. Results reported at patient temperature are calculated using the following equation. Δ ph/ Δ T = (7.4 - ph) T = Temperature in degrees celcius. 21

22 1.7.3 Carbon dioxide tension (pco 2 ) The partial pressure of carbon dioxide (pco 2 ) is a measure of the tension or pressure of carbon dioxide dissolved in the blood. The Proxima analyser reports pco 2 in units of kilo-pascals (kpa) or millimetres of mercury (mmhg). Clinical significance Carbon dioxide is a significant end product of metabolic processes in the body, and is excreted as gaseous carbon dioxide during respiration and as bicarbonate ions by the kidneys. Carbon dioxide reacts with water to form carbonic acid, which in turn dissociates to form bicarbonate and hydrogen ions. CO 2 + H 2 O H 2 CO 3 H + + HCO 3 - The bicarbonate/carbonic acid buffer system is the most important buffer keeping blood plasma in a narrow range around ph 7.4. Measurement of pco 2 together with ph is an important component in the diagnosis of acid-base disorders. Elevated pco 2 readings are a primary indication of respiratory acidosis or a compensatory response to a metabolic alkalosis. Conversely, low pco 2 levels are a primary indication of a respiratory alkalosis or a compensatory response to a metabolic acidosis. The clinical significance of acidosis and alkalosis are discussed in the section on ph above. Factors affecting results Adequate sample should be drawn to ensure a representative arterial blood sample is measured. If the sample is drawn too quickly, this may de-gas the sample leading to a lower concentration being reported. The pco 2 sensor The pco 2 sensor used in Proxima is an ISFET based version of the sensor described by Severinghaus. The gate region of the ph sensitive ISFET is covered by a thin layer of bicarbonate containing electrolyte, which is isolated from the sample by a gas diffusion membrane. As the sensor is brought into contact with the sample, CO 2 readily diffuses across the gas diffusion membrane to equilibrate the concentration in the sensor electrolyte with that in the sample. As the CO 2 in the sensor electrolyte is itself in equilibrium with the bicarbonate concentration, the ph of the sensor electrolyte is proportional to pco 2. The ISFET is used to measure the ph of the sensor electrolyte with a silver-silver chloride electrode used as the reference electrode. Carbon dioxide tension is a temperature-dependent quantity and Proxima measures and reports pco 2 at 37 C. Prioxima Technical Description 22

23 All sample measurements for pco 2 are taken and reported at 37ºC. Results reported at patient temperature are calculated using the following equation. Δ log pco 2 = Δ T T = Temperature in degrees celcius. 23

24 1.7.4 Oxygen tension (po 2 ) The partial pressure of oxygen (po 2 ) is a measure of the tension or pressure of oxygen dissolved in the blood. The Proxima analyser reports po 2 in units of kilo-pascals (kpa) or millimetres of mercury (mmhg). Clinical significance Clinical management of cardiopulmonary disorders frequently depends on measurement of oxygen tension in the blood. A supply of oxygen from the air to the body s cells is essential for metabolism. The transport of oxygen into the blood is through pulmonary ventilation and diffusion of oxygen from the alveolar air into pulmonary capillary blood. The capacity of blood to take up oxygen is almost entirely dependent on the concentration of haemoglobin in red blood cells. The oxygen binds reversibly to the haemoglobin, being taken up in high oxygen concentration environments and released in regions of low oxygen. Haemoglobin bound oxygen is transported by the cardiovascular system to the cells, where the oxygen is released. A decreased value of po 2 in arterial blood is known as hypoxemia, for which there are a number of principle causes: 1. Factors that depress the respiratory centre, such as narcotics and barbiturates as well as impairment of the central nervous system by trauma, infection or degenerative disorders. 2. Conditions that obstruct ventilation, such as airway obstruction or chronic obstructive pulmonary disease. 3. Conditions that impair gas exchange between alveolar air and pulmonary capillary blood such as severe pulmonary infection or respiratory distress syndrome. 4. Alteration in the flow of blood within the heart or lungs, such as shunting of venous blood into the arterial system without oxygenation in the lungs or pulmonary embolism. Factors affecting results Adequate sample should be drawn to ensure a representative arterial blood sample is measured. If the sample is drawn too quickly, this may de-gas the sample leading to a lower concentration being reported. 24

25 Oxygen sensor The oxygen sensor used in Proxima operates on the same amperometric principle as that first described by Leyland Clark. The sensor consists of a three electrode amperometric transducer covered with a thin layer of electrolyte solution, which is separated from the sample by a gas diffusion membrane. As the sensor is in contact with the sample, oxygen diffuses into the sensor through the membrane and is electrochemically reduced at the sensing electrode. The sensing electrode has a catalytic platinum surface on which the following reaction takes place: O 2 + 4e + 2H 2 O 4OH A steady state condition is rapidly reached, where current generated from this reaction is directly proportional to the rate that oxygen diffuses to the sensing electrode, itself related to the concentration of oxygen in the sample. Oxygen tension is a temperature-dependent quantity and Proxima measures and reports po 2 at 37 C. All sample measurements for po 2 are taken and reported at 37ºC. Results reported at patient temperature are calculated using the following equation. Δ log po x x po = Δ T 9.72 x 10-9 x po T = Temperature in degrees celcius. 25

26 1.7.5 Potassium Clinical significance Potassium (K + ) is the major intracellular cation, with a concentration of approximately 150 mmol/l in tissue cells, 105 mmol/l in erythrocytes and 4 mmol/l in extracellular fluid. High intracellular concentrations are maintained by Na + /K + adenosine triphosphatase (ATPase) which transports potassium into the cell faster than it can diffuse out. The hormones aldosterone and insulin play a part in control of this process. The Na + /K + ATPase pump, fuelled by glycolysis, plays an important role in maintaining and adjusting the ion gradients that sustain the mechanisms of muscle contractility and nerve impulse transmission. Potassium is acquired by dietary intake and most loss is via excretion by the kidneys in a process controlled by aldosterone. Disturbances in plasma potassium levels can be caused either by compromised kidney function or by factors affecting the redistribution of potassium from extracellular fluid into the cells. As the plasma contains relatively little potassium, small changes in these physiological processes can have a marked effect on plasma K + concentration. Low plasma potassium, known as hypokalemia, can lead to muscle weakness, tachycardia and in extreme cases paralysis and cardiac arrest. Some potential causes of hypokalemia include renal tubular disease, hyperaldosteronism, treatment of diabetes ketoacidosis, hyperinsulinism, metabolic alkalosis and diuretic therapy. High plasma potassium levels, known as hyperkalemia, can lead to confusion, weakness of the respiratory muscles, bradycardia and other cardiac abnormalities. Prolonged hyperkalemia can lead to peripheral vascular collapse and cardiac arrest. Some potential causes of hyperkalemia include renal glomerular disease, adrenocortical insufficiency, diabetic ketoacidosis, sepsis and infusion of potassium containing solutions. Measurement of potassium in the blood is important in the diagnosis and treatment of patients undergoing surgery and those suffering from hypertension, renal failure or impairment, cardiac distress, disorientation, dehydration, nausea and diarrhoea. 26

27 The potassium sensor The potassium sensor used in Proxima uses an ion selective membrane deposited onto the surface of the ISFET gate region. The ion selective membrane contains immobilized valinomycin, an ionophore that shows a high selectivity to potassium ions in comparison to other clinically encountered species. A silver-silver chloride electrode is used as the reference electrode. Factors affecting results Adequate sample should be drawn to ensure a representative arterial blood sample is measured. If the sample is drawn too fast the cell lysis will cause high K + concentration. 27

28 1.7.6 Glucose Clinical significance Glucose is the main source of cellular energy and glucose homeostasis is essential to maintaining health. Within healthy patients, glucose concentration in blood is regulated within a reasonably tight range by a number of hormones that convert excess glucose to glycogen or adipose tissue and release glucose from these sources at times of demand. Possible causes of low blood sugar (hypoglycaemia) include excessive administration of insulin, Addison s disease, severe illnesses affecting the liver, kidneys or thyroid gland, high alcohol consumption and some medications including salicylates and propranolol. High blood sugar (hyperglycaemia) may be caused by a wide range of factors, including uncontrolled diabetes mellitus, pancreatitis or pancreatic cancer, hyperthyroidism, Cushing s syndrome, severe stresses on the body, such as heart attack, stroke, trauma, or severe illnesses and certain medications, including prednisone, estrogens, beta-blockers, glucagon, oral contraceptives, phenothiazines. Hyperglycaemia is particularly prevalent in critically ill patients. Measurement of blood glucose concentrations is important in the diagnosis and management of patients with abnormal glucose levels. The glucose sensor The glucose sensor used in the Proxima is a glucose oxidase based amperometric biosensor. The glucose oxidase enzyme (biological source: aspergillus niger) catalyses the following oxidation reaction of glucose. Glucose + ½O 2 + H 2 O gluconic acid + H 2 O 2 Hydrogen peroxide generated by the reaction diffuses to a platinum electrode where it is oxidised to generate a current that is directly proportional to the concentration of glucose present in the sample. Factors affecting results Adequate sample should be drawn to ensure a representative arterial blood sample is measured. Acetaminophen (paracetamol) may cause cross interfering effect on the glucose sensor. Please see Section 1.5 for further details. 28

29 1.7.7 Haematocrit and calculated Hb The Haematocrit is a measurement of the fractional volume of red blood cells, the ratio of erythrocyte volume to the whole blood volume. The Proxima System reports the haematocrit value in %. Also a calculated Hb value is reported. Both parameters are related by the following formula:hct=chb x The factor assumes a regular mean corpuscular haemoglobin concentration. Clinical significance The haematocrit measures the volume of red blood cells compared to the total blood volume (red blood cells and plasma). This is a key indicator of the body s state of hydration, anaemia or severe blood loss, as well as the blood s ability to transport oxygen. A decreased haematocrit can be due to either overhydration or fluid overload, which increases the plasma volume, or a decrease in the number of red blood cells caused by anaemias or blood loss or decreased production of haemoglobin (e.g. thalassemia) An increased haematocrit can be due to loss of fluids, such as in dehydration (diarrhoea, vomiting, excessive sweating or inadequate water intake) diuretic therapy, and burns, or an increase in red blood cells, such as in cardiovascular and renal disorders, polycythemia vera, and impaired ventilation. Factors affecting results Adequate sample should be drawn to ensure representative arterial blood sample is measured. Avoid drawing the sample too fast as this could cause cell lysis. The Proxima System uses the sample conductance to determine percent hematocrit. The hematocrit is based on whole blood and is therefore dependent on plasma volume. The relative conductivity of the plasma component will affect the reported % Hct. Also the conductivity method used on Proxima does not distinguish red blood cells from other non-conducting elements such as proteins, lipids or white blood cells (when the leukocyte concentration is outside the regular range), which also occupy volume in the sample. Electrolyte concentration: The conductivity of the whole blood sample is dependent upon the concentration of electrolytes in the plasma portion. 29

30 Other non-conducting elements, Total Protein: it is important to be aware of the total protein level for example when monitoring a patient on a cardiopulmonary bypass pump. All priming fluids cause hemodilution that affects the concentration of plasma proteins, which leads to a fall in the haematocrit (HCT).Also total protein levels may be low in burn patients and in patients receiving large volumes of saline-based fluids. The Haematocrit sensor The Haematocrit is determined conductometrically by measuring the electrical conductance of a whole blood sample. Plasma conducts electrical current and blood cells act as insulators. A sample with a relatively high haematocrit has a larger proportion of its volume filled by the non-conductive red blood cells. The overall conductance of the sample will thus be relatively low. The conductimetric sensor used in Proxima comprises three platinum disc electrodes spaced at intervals across the sensor array with one electrode at each end the third in the middle. An excitation signal is applied between the central electrode and the two end electrodes and the response signal is measured. These two signals take the form of sine waves and how well they align is used to determine the haematocrit of a blood sample. By applying the excitation between the central electrode and each of the other two electrodes in turn the sensor may be used to check for the presence of bubbles within the cell Calculated Base excess Actual base excess (BE) is the concentration of titratable base when the blood is titrated with a strong base or acid to a plasma ph of 7.40 at a pco 2 of 40 mmhg (5.3 kpa) and 37 C at the actual oxygen saturation. Base excess is the deviation in mmol/l of the buffer base amount from the normal level in blood. Buffer base represents the total buffer capacity in the blood, comprised of bicarbonate, haemoglobin, plasma proteins and phosphate. The normal total buffer base level is 48 +/- 2 mmol/l. A negative base excess indicates the presence of metabolic acidosis and a positive base excess indicates the presence of metabolic alkalosis. B.E. allows the calculation of the buffer quantity that needs to be infused in a patient with impaired acid-base balance. The Proxima System calculates the value of base excess from the current analyte readings. Base Excess of Blood: BE (blood)= ( x thb) x [(HCO 3 - act 24.8) + (( x thb) x (ph (37ºC)-7.40))] where thb is a default of 15 g/dl. 30

31 1.7.9 Calculated Standard Bicarbonate HCO 3 - (bicarbonate), the most abundant buffer in the blood plasma, is an indicator of the buffering capacity of blood Regulated primarily by the kidneys, HCO 3 - is the metabolic component of acid-base balance. Changes of the HCO 3 - concentration in connection with ph values are used for the determination of whether an acidosis or alkalosis of metabolic origin is present. An increased level of HCO 3 - may be due to a metabolic alkalosis or a compensatory response in respiratory acidosis. Decreased levels of HCO 3 - are seen in metabolic acidosis and as a compensatory mechanism in respiratory alkalosis. Causes of primary metabolic acidosis are ketoacidosis, lactate acidosis (hypoxia), and diarrhoea. The Proxima System calculates the value of standard bicarbonate from the current analyte readings. Standard bicarbonate is a determination of the plasma HCO 3 - concentration if the blood is equilibrated to a pco 2 of 40 mmhg and po 2 of 100 mmhg at 37ºC using the equation described by VanSlyke and Cullin: Oxygen saturation is estimated using the relationship described by Kelman and Thomas: Where N is: And BE (B) is calculated assuming 100% oxygen saturation. When thb is used it is a default of 15 g/dl. 31

32 Calculated PaO 2 /FiO 2 ratio The PaO 2 /FiO 2 ratio is the ratio of arterial oxygen partial pressure to fraction of inspired oxygen; so it is a comparison between the oxygen level in the blood and the oxygen concentration that is breathed. In normal healthy lungs, the PaO 2 value should equal 5x the fraction of inspired oxygen (FiO 2 ). For example for a patient breathing room air, the FiO 2 is 21% O 2 ; and therefore the PaO 2 should be (5 x 21 = 105 mmhg). A PaO 2 less than 4 to 5 times the FiO 2 suggest poor lung function. PaO 2 /FiO 2 ratio is a widely used clinical indicator of hypoxaemia, which allows comparison across time of severity of hypoxemia on patients breathing various oxygen concentrations. A normal PaO 2 /FiO 2 ratio ranges from 500 to 300. The lower the PaO 2 / FiO 2 ratio, the worse the disease process is. A PaO 2 /FiO 2 ratio less than 300 is consistent with Acute Lung Injury (ALI). And a PaO 2 /FiO 2 ratio less than 200 is consistent with Acute Respiratory Distress Syndrome (ARDS). The calculation for PaO 2 /FiO 2 ratio is as follows: PaO2 / FiO2 po2 FiO2 A patient with a normal PaO 2 breathing in room air (at sea level) will have a PaO 2 /FiO 2 ratio of 100 mmhg/ The patient s FiO 2 at the time of analysis needs to be known and manually entered into the system before the PaO 2 /FiO 2 ratio can be calculated. 32

33 2.1 Contact the Manufacturer The Sphere Medical team is dedicated to providing excellent support. For all technical and safe use questions relating to this manual or the Proxima System, contact us at: Sphere Medical Ltd. Harston Mill Harston, Cambridge CB22 7GG United Kingdom Tel: +44 (0) Contacts and Support