What is ultrasonic plethysmography and how can we use it in airway disease? Christian Buess ndd Medizintechnik AG, Zürich, Switzerland ERS 2007-1/19
Background Until approx. the year 2000 mainly the following flow sensor principles were used in lung function testing: Differential pressure (variable or fixed resistance) Turbines (rotation speed) Heated wires (cooling effect) All these principles have potential limitations: Hygiene: Difficult to clean Unable to measure very low flows Need of regular calibration Dependency on gas composition and/or temperature ERS 2007-2/19
Operating Principle Ultrasonic flow measurement is commonly used for liquid flow. It is now also used for air flow measurement in lung function diagnostics. Integrated Mouthpiece Ultrasonic Transducer Sound Transparent Windows Exchangeable Flow Tube ERS 2007-3/19
Operating Principle Ultrasonic transit-time measurement. Please note that the principle differs to the Doppler principle used for blood flow measurement. Sound Receiver Transmission of Ultrasonic Pulse Sound Transmission Path ERS 2007-4/19
Operating Principle Transmission of Ultrasonic Pulse Sound Receiver Sound transmission and reception is repeated every 1 to 5 ms. Sampling rates 200 to 1000 Hz. ERS 2007-5/19
Measurement principle (simple) New York Zürich 9 Hours 7 Hours Time difference is proportional to speed of Jet stream Ultrasonic flow sensor: Time difference up to 1 μs ERS 2007-6/19
Measurement principle: Flow upstream transit-time (Zürich NY) downstream transit-time (NY Zürich) t d t u = = L c u cosϕ L c + u cosϕ u = Flow L 2 cosϕ t t d d t t u u Speed of sound depends on temperature and gas composition Flow measurement depends only on flow and geometry No dependency on temperature, gas composition, humidity or pressure ERS 2007-7/19
Measurement principle: Molar Mass upstream transit-time t d = L c u cosϕ Molar Mass downstream transit-time t u = L c + u cosϕ M 4κR = 2 T L t t u u t + t d d 2 speed of sound c R T = κ M Molar mass measurement depends on transit-times and temperature along the sound transmission path Molar mass measurement for free no additional sensor required Molar mass measurement is not specific ERS 2007-8/19
Strengths and Weaknesses No calibration since flow measurement depends only on time measurement and geometry of sensor Completely open, unobstructed tube Good hygienic solution High resolution and dynamic range Molar mass measurement for free Relatively high cost of sensor ERS 2007-9/19
Spirometry: Devices Examples of devices using ultrasound technology Standalone spirometer for office use Flow sensors for spirometry and exercise testing Flow sensor for Intensive Care application ERS 2007-10/19
Application The technology can be used for a large range Neonatology Resolution: < 1 ml/s Range: 500 ml/s Adult Resolution: < 10 ml/s Range: 18 l/s Horses Resolution: < 80 ml/s Range: 100 l/s ERS 2007-11/19
Spirometry: Studies Performance of 70 devices over 3 to 6 month period R. Pérez-Padilla et al.: Long-term stability of portable spirometers. (Respir Care 2006) Performance of 6 devices over 200 days J. Walters et al.: Stability of the EasyOne ultrasonic spirometer for use in general practice. (Respirology 2006) Ultrasonic spirometer used by world-wide BOLD study (Burden of Obstructive Lung Disease). Ultrasonic spirometer used by Platino study in Latin America. ERS 2007-12/19
FRC Measurement: Background Measurement of FRC using multiple breath wash-out / wash-in of a tracer gas by measuring molar mass of in- and expired gas flow. expiratory sidestream MM trace start of tracer wash-in wash-in of tracer gas Computation of FRC, LCI Main stream molar mass signal is synchronous with flow signal Simple instrument setup compared to mass spectrometry Best results are only achieved with side-stream MM sensor ERS 2007-13/19
FRC Measurement: Setup Valve in open position Valve in closed position Test gas supply Ultrasonic Flow / MM Sensor Side-stream sampling tube ERS 2007-14/19
FRC Measurement: Studies Direct comparison of ultrasonic system with mass spectrometer Fuchs et al., Pediatr Pulmol, 2006, 41. r 2 = 1.00, mean difference -0.2% Setup Problems: Gas leaks etc. Most devices and most software are in prototype stage! The above results were obtained using a side-stream MM sensor If applied carefully results compare well with mass spectrometers Additional studies: J. Pillow et al., Schibler et al. etc. ERS 2007-15/19
Tidal Breathing Analysis: Background Comparable to capnography the molar mass analysis of tidal breathing shows the expiratory molar mass curve vs. expired volume. 4.5 4.5 4.0 4.0 3.5 3.0 2.5 S2 S3 3.5 3.0 2.5 2.0 2.0 1.5 1.0 0.5 0.0 Vd Volumen [l] Trial 1 Trial 2 Trial 3 0.0 0.2 0.4 0.6 0.8 1.0 healthy control 1.5 1.0 0.5 0.0 Volumen [l] Trial 1 Trial 2 Trial 3 0.0 0.2 0.4 0.6 0.8 1.0 patient with COPD Several parameters can be computed from the curve data: Slope of phase 2 S 2 Slope of phase 3 S 3 Dead Space volume V d ERS 2007-16/19
Tidal Breathing Analysis: Method 90 seconds of tidal breathing recording flow and molar mass {S 2,S 3,V d, {S 2,S 3,V d, {S 2,S 3,V d, {S 2,S 3,V d, {S 2,S 3,V d, {S 2,S 3,V d, {S 2,S 3,V d, {S 2,S 3,V d, {S 2,S 3,V d, V i,v e,f, } 1 V i,v e,f, } 2 V i,v e,f, } 3 V i,v e,f, } 4 V i,v e,f, } 5 V i,v e,f, } 6 V i,v e,f, } 7 V i,v e,f, } 8 V i,v e,f, } 9 Determination of mean expiratory MM curve S 2,S 3,V d Statistical Analysis Diagnosis ERS 2007-17/19
Tidal Breathing Analysis: Studies Studies analyzing the Molar Mass Index in side-stream MMI = S 3 / S 2 100 25 MMI 20 15 10 5 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Normal controls Very severe to moderate COPD Restricted Beckert, Buess, Magnussen: Not yet published Correlation of MMI and LCI, both determined by ultrasonic flow sensor Fuchs, Gaultier, Buess, Gappa: ERS 2007. ERS 2007-18/19
Conclusion Ultrasonic flow measurement is a new technology with many advantages: No calibration, good hygienic solution and high accuracy Ultrasonic molar mass measurement is the mass spectrometer for everybody its use may have a large potential for measuring multi-breath FRC. Analysis of expiratory molar mass curve during quiet breathing: Potentially interesting for analysis of gas distribution in the lungs. Further investigations are required. ERS 2007-19/19