Medical Blog

Three different calculations were carried out to determine the HF spectral amplitude in the invasive blood pressure signal every 90 seconds of a larger time interval. According to these three calculations, signals were formed by the resulting consecutive HF values in time. In many circumstances, these three signals show the same significant fluctuations. If the fluctuations in these signals did not correspond,
in almost all cases an explanation could be found in a change of the respiratory band width. Therefore the hypothesis can be sustained that all three methods used to calculate the HF spectral components can be applied successfully. A choice can be made whether or not to emphasise respiratory fluctuations. Of course, this hypothesis should be tested on more neonatal data sets. A relation between the spectral
amplitudes and gestational age, age after birth, and behavioural state can be investigated in the future.

The assessment of the autonomic nervous system activity by the HF values is hampered by fluctuations of the respiratory frequency and the respiratory amplitude (depth). The future developments, that allow for real-time filtering with adaptive bandwidth filters, will resolve the problem of the fluctuating respiratory frequency: the actual respiratory frequency can be followed by an adaptation of the filter
coefficients. The fluctuations of respiratory amplitude, however, will still influence the HF spectral amplitude, suggesting changes in the autonomic nervous system that do not exist. Part of the fluctuations in the HF signals, could be attributed to variations in the respiratory frequency. The impact of respiratory amplitude variations on the HF values should be investigated to be able to estimate their influence on the spectral amplitudes. The determination of cross spectra between blood pressure and heart interval as presented in section 6.2 looks promising. The quantification of mutual interactions between blood pressure and heart interval will give us more insight in the origin of the interactions and may refine the underlying physiological model. Further
measurements involving large and different patient groups will be required to be able to draw more definite conclusions on the physiological model.

Spectral equidistancy

A signal built of beat-by-beat parameters (a set of consecutive beat-by-beat values) is inevitably non-equidistant in time. In most of the neonatal signals, the consequences will not be dramatic. The 30% distortion of spectral components mentioned by TenVoorde [TenVoorde, 1994] does not apply to neonates because of their small respiratory sinus arrythmia (see section 5.4), and thus small frequency modulation of the sampling intervals of the beat-to-beat values. If the heartrate variability is larger, the error increases. We used neonatal data to evaluate the impact of the heartrate variability on the spectral components, using several different Fourier algorithms. Our conclusion is that the algorithms that specifically take into account the heart interval (EFT, RFT) give (slightly) different results, compared to the standard DFT/FFT algorithm. The standard DFT/FFT was applied such that the non-equidistancy was ignored. Many institutions follow that method; the boxcar integration we apply in the standard procedure introduces an extra filtering and resampling. The extra filtering reduces the aliasing of higher frequencies and enables us to resample equidistantly. The neonatal data we examined show only small differences with respect to the particular Fourier method chosen. Important is that the respiratory sinus arrhythmia, the main cause of the non-equidistancy of the beat-to-beat values, is much smaller in neonates than in adults.

Stationarity

Stationarity is an important item. If one wants to characterise certain spectral components, resulting from physiological processes in the body. If the underlying physiological processes change during the registration, qualitative characterisation will sometimes be impossible, and quantitative characterisation will only be possible in proportion. Given the fact that all physiological processes in the body change in time, it is desired to limit the registration time to a period in which the body processes hardly change. A trade-off has to be made is between the accuracy of the measurement and the stability of the underlying processes. Physiological changes that we want to exclude from the measurements are, for instance,
−changes in the heartrate base level, e.g., caused by labour or mental stress
−changes in behavioural state or sleep state
−changes in the chemical state of the blood (e.g., pCO2 level changes that drive many central processes)
Tools that check the data on stability should be developed. The clinician could then be informed automatically if the observed variations were the result of a changing process during the measurement. We suggest to use correlations between the signals to detect artefacts, but did not work out procedures by now.

Beat-to-beat values versus the full sampled signal

Usually, beat-by-beat values are used in spectral analysis of the blood pressure variations. Originally, the main reason for this approach was that no calculation power was available to handle the frequency analysis of a full sampled signal, consisting of approximately 30 to 100 times as much data as the beat-by-beat values in a certain time interval. We showed that nowadays the limited calculation power no longer impedes the analysis of the full sampled signal. In that case, problems concerning aliasing and non-equidistancy do not exist any more. The graphs in chapter 5 show such results, and a comparison with the results of the analysis of a beat-to-beat signal. Theoretically, a significant improvement of the results should be obtained, if neither aliasing nor non-equidistant sampling would occur. Part of the improvement, however, is spoiled by the modulations expressed in the full sampled signal at sum and difference frequencies of the heartrate and respiratory rate. It will strongly depend on the application and signal characteristics which method will satisfy most.

The frequency analysis of neonatal blood pressure signals differs in many respects from the analysis of signals from adults. In neonates:

• the frequencies of heartrate and respiration are normally much higher,
• the heartrate variability turns out to be less pronounced,
• the number of artefacts in the signals is higher than in adults.

Many studies have been done on heartrate variability in adults and neonates, some on blood pressure variability, but only a few on blood pressure variability in neonates.
The analysis methods used, mainly Fourier transform or a modification of it, and autoregressive model estimates, lead to a variety of representations of the results. All methods have their own characteristics, constraints, and assumptions. In time domain analysis many different indices have been proposed, often only to be used in specific circumstances. The frequency domain methods seem more general; the autoregressive methods require certain model assumptions, like the order of the system. The main problems of the Fourier methods will be discussed in the next sections.

Aliasing

In our analysis the problem of aliasing occurs if the momentary respiratory frequency is higher than half the heartrate. In neonates this situation will occur regularly. The ratio of the respiratory frequency and heartrate in neonates is often higher than in adults. All Fourier techniques are hindered by a relatively high respiratory frequency.
The Lomb method, which is able to handle non-equidistantly sampled signals, shows better results if the major part of the momentary sampling frequencies is larger than half the heartrate.
We found a more fundamental alternative in using the blood pressure filtering (BPF) technique. Using the full blood pressure signal, aliasing will occur only above 32 Hz (half the sampling frequency), which is far beyond our region of interest. The BPF technique is not a final solution. It introduces a problem that is related to the complexity of a physiological signal like the blood pressure signal. The blood pressure
signal consists of pulses that result from different states of heart contraction under strongly changing internal physiological states. These physiological states vary with the heart frequency, and, as a result, they modulate the other variations in the blood pressure signal. Sum and difference frequencies will be present; the difference frequencies being at the same frequency as the aliased frequencies from above half the heartrate when a Fourier technique is used. Using the BPF method, however, they can be explained by a real physical variation. Using one of the other methods, one should always check separately on the respiratory frequency to be attentive to possible spectral aliasing.

Spectral leakage

Spectral leakage implies the leakage of spectral power to neighbouring frequency bins, due to the (generally) non periodicity of the signal in the time domain. In most cases, the application of one of the standard lag-windows gives acceptable results. The HF frequency components we are interested in contain variations, which lead to abroadening of the concerned frequency components. The effect of leakage to the neighbouring bins in that case only marginally affects the result. The LF and VLF components, however, sometimes extend over only a few bins. In that situation one should be aware that the accuracy of the spectral components may decrease considerably.

The invasive techniques, being more accurate, were not readily available for human patients until about 1900 [Frank 1905, Hansen 1949]. In the 18th century, however, Hales described the first measurement of blood pressure with a glass manometer connected to an artery of a living horse [Hales 1733] аnd in 1876 Mayer already used invasive measurements in rabbits for determination of low-frequency blood pressure fluctuations [Traube 1865, Mayer 1876].
Invasive blood pressure measurements are always performed by indwelling of a measurement device in the artery or vein in which the pressure has to be known. The measurement device may either consist of a pressure transducer on the tip of a catheter or of a cannula through which the pressure waves are conducted to a pressure transducer outside the patient. Measurement by a tip-catheter is preferable when high frequency components are investigated, because of its excellent frequency response. However, for use in neonates the catheter has to be very thin [less than 4 French]. Although such catheters exist they still are very expensive. When the device is used for continuous measurements, problems occur due to blood clotting around the tip [Webster 1988]. Clinically, often the possibility of blood sampling is more important than the measurement of accurate blood pressure values. A catheter manometer system [CMS], applied primarily for blood sampling, contains a cannula that is coupled to a pressure transducer. It can be used for at least a week, preventing the baby from being
cannulated again аnd again, аnd allows a continuous monitoring of the arterial blood pressure. A CMS only is applied at clinical indication, taking into account the possible risks: thrombus forming, infection аnd transient ischaemia. The risks of umbilical artery cannulation are higher than the risks of peripheral artery cannulation [Schober 1990, Aldridge аnd Gupta 1992]. Accurate measurements of blood pressure require a transmission of pressure wave frequencies to at least the 10th harmonic of the base frequency [Weindling 1989, Hack et al., 1990a]. In premature neonates this corresponds to a bandwidth of approximately 30 Hz. The transmission of high frequency components is often corrupted by air bubbles or blood clots in the system.

A catheter manometer system

If a clinical indication for an indwelling catheter exists, routinely blood pressure is measured with a fluid-filled catheter manometer system. The system consists of a cannula, an extension tube, two stopcocks, аnd a pressure transducer chamber, that is connected to an infusion pump for flushing [see figure 4-1]. The cannula is brought into the radial artery of the neonate. In the period following delivery normally a
longer cannula is used, which is brought into the aorta via the umbilical artery. The total length of the system is about 50 cm аnd it is flushed with heparinised 0.65% saline solution at approximately 1 ml/h. This continuous flushing prevents the blood from clotting around the catheter tip. The risks for complications, i.e., partial obstruction of the artery, are acceptable [Hack et al., 1990b]. In the pressure
transducer chamber a strain gauge is used to convert the pressure variations to electric signals.

Although in this thesis we focus on invasive arterial blood pressure measurements, in some situations we may use the non-invasive alternatives. These all have in common that the assessment of the blood pressure is not very accurate, but, on the other hand, an arterial cannulation is not required. The oscillometric method is the oldest one used in man. In 1866 this method was described by Marey [Marey, 1866]. He used a small cuff, snugged around a limb, which was inflated with water above systolic pressure. During deflation he registrated the oscillations of a mercury column on a carbon drum. Pressure was read from a mercury manometer. The point of maximal oscillations was decided to be the mean arterial pressure. In 1896 Riva-Rocci introduced the palpatory method, only suitable to determine the systolic pressure. He used a cuff with air connected to a mercury manometer to measure the pressure applied to the brachial artery. The pulse was determined by palpation of the a. radialis at the wrist аnd the pressure was increased until the pulse disappeared [systolic pressure]. The oscillation of the cuff pressure with the heartrate, when the cuff was smoothly being deflated, was noted. It results from pressure transmission from the artery to the cuff, which is the basis of all oscillometric measurements. The problem where to locate the point of diastolic pressure in the sequence of
oscillations could not be solved. In 1905 a solution was proposed by Korotkoff, who introduced the auscultatory blood pressure measurement [Korotkoff 1905]. He described the characteristic sounds [named after him] heard with a stethoscope in the “elbow” when the pressure in the inflated cuff was lowered through systolic аnd diastolic pressure. He used the same cuff as Riva-Rocci. This method still is the basis of the auscultatory method used today. The method slightly improved after 1940, when intra-arterial measurements had become available that could be used as a “gold standard”. In particular attention was given to the small size of the cuff, which especially in the obese turned out to overestimate the blood pressure level. The American Heart Association recommends for the cuff-bladder dimension a width of 40% of the circumference of the limb, long enough to encircle 80% of the circumference, to be placed such that it is centered over the artery to be compressed. For premature neonates this leads to the use of a smaller cuff.

The automated oscillometric devices currently used still follow the oscillometric principle described by Marey, using a cuff snugged around a limb. A detailed description, characteristic for many of these devices, is given by Ramsey [Ramsey 1991]. The cuff is inflated to approximately 160 mmHg for the first determination, or to 30 mmHg above the previous systolic pressure found. Oscillations in the cuff
pressure are sampled [by a microprocessor] at a constant cuff pressure to improve artefact reduction. If no oscillations occur the pressure is lowered a few mmHg, аnd sufficient sample time is allowed. If oscillations occur the amplitude of two consecutive pulses should only differ a small amount, аnd the time interval between them has to match closely the previous time intervals. When two consecutive pulses meet this criterion their amplitude is averaged аnd stored for later determination of systolic,
diastolic or mean pressure. This criterion avoids many of the artefacts arising from strong respiratory variation, premature ventricular contraction, аnd external motion. If artefacts occur, the measurement time will be increased by the time waiting for two acceptable pulses. If the delay is too long the actual cuff pressure level is skipped. The cuff pressure is lowered again a few mmHg аnd sampling for the oscillations starts again. In general no oscillations are found any more after going several deflation steps through the diastolic pressure. An algorithm is applied to the average values, stored at each deflation step, to determine the systolic аnd diastolic arterial pressures. The mean is also calculated from the averaged values. The systolic [diastolic] pressure normally is associated with the cuff pressure at which the increase [decrease] of the oscillation amplitude has a maximum. The algorithm is an important part of the measuring device, because it has to reduce artefacts from the set oscillation amplitudes of many different patients аnd patient groups.

Problems in blood pressure measurements may occur due to:

• physiologic variation [shock, large blood pressure variation during the measurement, variation in pulse rate [>15%]]
• anatomic variation [conically shaped arm, calcified arteries, subclavian compression];
• cuff compression variation [movement [of cuff or arm], shivering, bumping the cuff, vehicular or helicopter vibration].

In neonates additional care has to be taken in the interpretation of blood pressure values. Studies in which oscillometric аnd invasive arterial measurements are compared report a significant disagreement between them. Many report an overestimation of low pressures [Gevers 1994, Diprose 1986] аnd unacceptably large errors for the individual infant [Briassoulis 1986, Wareham 1987]. Furthermore, attention is required to the cuff size used: too small sizes yields too high pressure values, too large sizes only have a small effect. The cuff has to be applied snugly enough аnd all air has to be squeezed out before applying the cuff. The cuff hoses may not be kinked аnd the air system may not leak. The patient must be still аnd quiet, th e cuff should be at heart level [if no hydrostatic compensation is applied] аnd subclavian compression by the arm or retraction of the chest should be avoided.

Some automated devices use microphones or a Doppler signal to determine the systolic, diastolic аnd mean pressure points during deflation of a cuff. The microphone instruments use the abrupt change of frequency content of the Korotkoff sound. A filter circuit is able to detect the muffled end point of the sound at diastolic pressure, which is more reliable than the detection of disappearance of sound [McCutcheon аnd Rushmer, 1967]. Nevertheless, errors may occur due to ambient sounds or sounds generated by patient movement. These errors can be reduced either by ECG triggering or by use of a second microphone that only registrates the ambient sounds. Subtraction of this signal from the original with the Korotkoff sound gives a purer Korotkoff registration. Doppler blood flow detection under the cuff can be used to determine the systolic pressure in specific small arteries. It is useful in the evaluation of peripheral vascular disease [Sumner 1984]. Only recently a continuous non-invasive measurement of finger arterial blood pressure in adults аnd children older than one year has become available [Smith et al., 1985, Peñáz 1973]. Up till now this method is considered to be inappropriate for use with neonates.

Summarizing we can state that automatic non-invasive blood pressure measurements currently are performed by the oscillometric method. In manual blood pressure measurements the auscultatory method is used. In both cases a significant deviation from the intra-arterial pressure cannot be excluded, especially not in case of a patient with several affected organs аnd systems.

Since 1628, when W. Harvey described for the first time the circular motion of blood through the body, the interactions between heartbeat аnd vascular state have been a topic of interest [Harvey, 1628]. Probably the most important signal for these studies is the blood pressure signal. This signal not only contains systolic [maximum] аnd diastolic [minimum] blood pressure values at every heartbeat, but also incorporates information about the volumetric state, peripheral resistance, heart function [contractility], influence of respiration on blood pressure аnd autonomic nervous system influences. The blood pressure level is an important tool in the evaluation of the condition of a critically ill patient. In the critically ill newborn infant not only the early determination of hypo- or hypertension, but also the determination of fluctuations in blood flow аnd blood pressure were found to be important. Among others, Perlman has shown in 1983 that these fluctuations are involved in the pathogenesis of intracranial lesions [Perlman et al., 1983, Miall-Allen 1989, Goldstein аnd Brazy 1990, 1991]. To monitor fluctuations in the blood pressure a continuous registration is needed. Non-invasive methods for continuous blood pressure registration in neonates are not available. The research group therefore remains restricted to the neonates having a clinical indication for an arterial catheter. This is about half of our NICU population.

Physiology

Blood is pumped around in the body by a rhythmic contraction of the heart. In the system circulation it goes from the heart through arterial vessels to the peripheral tissues аnd organs, аnd circulates back to the heart via venous vessels. In the lung blood circulates through the pulmonary artery to small capillary vessels where oxygen/carbon dioxide exchange takes place, аnd back to the heart in the lung veins.
Because of the rhythmic contraction of the heart the pressure in the arteries also fluctuates. The volume аnd the contour of the arterial pulses not only depend on the left ventricular stroke volume аnd the ejection velocity, but also on the relative compliance аnd capacity of the arterial system. Furthermore, they are composed of forward аnd backward waves, reflected from the peripheral system. The waveform thus contains a lot of implicit information about the contractility of the heart [septum] аnd peripheral resistance. Therefore the arterial pressure waveform depends on the measurement position аnd the quantities mentioned above. In infants, the blood pressure values are influenced by the presence of asphyxia, sepsis, infant respiratory distress syndrome [IRDS] аnd patent ductus arteriosus [PDA]. The influence of these pathologies on the contour of the blood pressure signal has not been investigated yet. In adults, altered waveforms have been reported in case of shock, hypo- аnd hypertension, atherosclerosis, vasodilatation, atrial fibrillation, valvular disease, hypertrophic cardiomyopathy, aortic coarctation аnd chronic uremia [O’Rourke, 1982, Gevers, 1994]. Apart from the waveforms, the fluctuations in blood pressure [beat-to-beat аnd slower] are known to contain information about the function of the autonomic nervous system.