MEASUREMENT OF PHYSIOLOGICAL PARAMETER This -invention relates to a method and apparatus for the determination of arterial blood pressure. We have found that determination of the time interval between a heart beat and the associated arterial pulse is related, inter alia, to arterial blood pressure and hence may be used to provide a continuous, non-invasive method for the determination of arterial blood pressure.
Of the two common methods for determination of arterial blood pressure, one is non-invasive but discontinuous and the other is continuous but invasive. The non-invasive method uses a sphygmomanometer which comprises an inflatable rubber cuff connected to a mercury manometer. In use, the cuff is placed around the upper arm and inflated until blood flow in the brachial artery is occluded. A stethoscope is then used, as the pressure in the cuff is gradually reduced, to determine the systolic and diastolic measuring points. The pressure of each of these points is read from the manometer and quoted as the blood pressure. The technique is simple to operate and firmly established in general and clinical medical practice but is not particularly accurate and, since each measuring operation takes some
20-30 seconds to perform, is discontinuous.
_ The continuous method involves the surgical insertion of a catheter in the brachial artery. The catheter is connected to a suitable pressure transducer and this provides a direct, accurate and continuous determination of blood pressure. However, since the method is invasive, it is suitable for use only during relatively major surgery. According to the present invention, a method for the measurement of arterial blood pressure comprises sensing a heart beat, sensing an arterial pulse at at least one location and ascertaining the time interval between at least one heart beat and at least one associated arterial pulse.
The invention also includes apparatus for the measurement of arterial blood pressure, the apparatus comprising first sensing means for sensing a heart beat, second sensing means for sensing an arterial pulse at at least one location, and means for ascertaining the time interval between at least one heart beat and at least one associated arterial pulse.
The present invention, therefore, provides a continuous, non—invasive technique and the output of the apparatus according to the invention is representative, inter alia, either of dynamic
comparative blood pressure or, with suitable calibration, dynamic absolute blood pressure.
The heart beat sensing means may comprise a simple pressure transducer in contact with the chest but it is preferred for reasons of accuracy to utilise electrocardiography. In particular, we have found that the R wave of an electrocardiogram, which corresponds to the beginning of ventricular systole, is a convenient reference for measurement of cardioarterial dela .
The arterial pulse may be sensed at one or more locations, for example at the carotid or temporal arteries in the head and neck, at the brachial radial or ulnar arteries in the arm and wrist, and/or in the fingertip, groin and leg. Sensors which may be used include temperature sensitive devices, piezo-electric transducers, strain gauges, ultrasonic transducers and plethys ographic transducers. Ultrasonic transducers rely on the Doppler effect to detect movement in the arterial wall and/or the flow of blood corpuscles while plethysmographic transducers measure volume changes which result from arterial pulsatile blood flow. Volume changes may be detected by measurement of impedance changes or temperature changes but we prefer to use
- Λ - photoplethysmography, which is an optical technique. In photoplethysmography, light is transmitted into the tissues and the amount of light reflected is inversely proportional to the volume of blood present in the artery, thereby affording a means of detecting changes in volume of the artery with pulse. Conveniently, light in the near infra-red region (wavelength 700- lOOOnm) is used since such light has a relatively high transmittance through tissue but is scattered by blood. Δ light emitting diode which emits in the infra-red may be used as the source of light and a light-sensitive photo-transistor may be used to measure reflected light. An infra-red light emitting diode is particularly convenient because the degree of scattering of light of this wavelength is substantially Independent of the degree of oxygenation of the blood.
The time interval between heartbeats and associated arterial pulses may be ascertained using either an analogue or a digital approach. Preferably the time intervals are electronically processed to give information such as average and instantaneous values of blood pressure and/or to cause operative reactions in other equipment of a diagnostic or therapeutic nature, such as warning devices and the like.
The invention will now be described by way of example with reference to the accompanying drawings of which:
Fig. 1 is a block diagram illustrating the equipment and procedures to which subjects were subjected to determine a delay (Td) between heart beats and associated arterial pulses;
Fig. 2 is a block diagram showing an electrocardiograph system; Fig. 3 is a block diagram showning an arterial pulse transducer system;
Figs. 4.1 to 4.7 are plots of Td against average blood pressure for each of seven subjects. Referring first to Fig. 1, a subject under test was connected as shown to an electrocardiograph and to an arterial pulse transducer. In connecting the electrocardiograph, a standard application of bipolar recording was used in which three identical electrodes are attached one to each arm and one to the left leg of the subject. The two arm electrodes were conncted to the inputs of a differential amplifier, the leg one being connected to common.
As shown in Fig. 2, the differential output was fed to a band pass filter with an approximate 3dB bandwidth of 0.08-80 Hz. This filter was used to remove noise and unrelated spurious
signals. In order to reduce 50 Hz mains interference, the output of the band pass filter was fed to a 50 Hz notch filter. This provided some 20dB's of attenuation at 50 Hz. The output of the notch filter was differentiated for convenience to provide a signal based on the rate of change of the ECG. The ZERO crossings of the derived waveform indicate the peaks of the original ECG waveform. The arterial pulse transducer comprised a Texas Instruments' TIL 139 combined gallium arsenide IR-emitting diode/npn silicon phototransistor mounted together in a moulded ABS plastics housing and held against the finger to detect the arterial pulse therein. The emission from this device typically peaks at wavelength of 940nm. At this wavelength, variations in the optical density of the underlying tissue are primarily determined by the pressure pulse. As shown in Fig. 3, the output of the phototransistor was amplified and fed to a band pass filter and a notch filter similar to those described with reference to Fig. 2. The notch filter output was then fed via a variable non-inverting amplifier to a differentiating circuit identical to that used for the electrocardiograph signal.
The electrocardiograph (ECG) and arterial
pulse (AP) transducer differentiator circuits indicated the rate of change of the respective signals and thus of their peaks. The results were recorded on a Medelec UV recorder to allow manual measurement of the values of the time intervals Td.
In use, the subject under test was connected to the equipment as shown in Fig. 1. The arterial pulse signals were obtained by using the AP transducer to detect the subject's finger pulsations. The experiment was divided into three parts: i. Initially measurements were made of the systolic and diastolic arterial blood .pressure with the subject "at rest", using a sphygmomanometer . ii. A period of exercise was undertaken consisting of a short jogging session around a predetermined route, sufficient to raise the subject's blood pressure to a high value. iii. Periodic measurements of systolic and diastolic blood pressure were then made until the subject's blood pressure had returned to the "at rest" value measured in (i). The ECG and AP results were recorded simultaneously with the blood pressure measurements being taken. The Medelec UV recorder used for
the differentiated outputs is an instrument consisting of both an oscilloscope and UV trace recorder. Traces displayed on the oscilloscope screen can also be recorded on UV sensitive paper producing a permanent record of the waveform being displayed. The oscilloscope has four inputs, two of which are connected to ECG and AP differentiator outputs. The other two are calibrated to ground and positioned to be superimposed on the AP and ECG traces to provide a reference. This ground reference is used to determine the zero crossing that defines the peaks of the input waveforms.
A sufficient length of UV trace recording was done to capture six or more cardiac cycles for each sphygmomanometer measurement. On each UV trace, the subject's initials and blood pressure readings were also recorded. For each experiment on a particular subject a set of 7 or 8 UV trace recordings was obtained. The values of Td were measured, using a rule, from the differentiated zero crossing corresponding to the R wave peak of the ECG to the zero crossing corresponding to the first peak of the AP . Together with the blood pressure readings these were entered into a computer program to calculate: i. the individual values of Td in msecs
for each cardiac cycle; i_i . the average value of Td in msecs over each sphygmomanometer measurement; and ill. the average manually measured blood pressure in mm Hg.
Thus, for a given subject, a set of average Td values and a set of average blood pressure values measured manually by sphygmomanometer were derived. Typically, sets of 7 or 8 values were obtained.
These values were then plotted on to graphs of average Td vs average measured blood pressure, and were also used as input data to a linear regression program to test the degree of correlation. The graphs are shown in Figs. 6.1 to 6.7, each graph representing results for one subject. In the graphs, the straight line is derived from linear regression and the deviation of the plotted points for Td vs blood pressure gives an indication of the degree of correlation.