Effect of ensemble averaging on amplitude and feature variabilities of Doppler spectrograms recorded in the lower limb arteries

The objective of the present study was to analyse the effect of averaging Doppler blood flow signals in lower limb arteries on amplitude and feature variabilities. Doppler signals recorded in 41 iliac and 35 superficial femoral arteries having different categories of stenosis were averaged over 1–15 cardiac cycles. Based on the relative decreasing rate of an index of variability, results indicated that amplitude variability of the spectrograms was exponentially reduced to 30, 6 and 1 per cent when averaging five, ten and 15 cardiac cycles, respectively. Nine diagnostic features were extracted from Doppler spectrograms and their variations from one cardiac cycle to the next quantified. Based on the relative decreasing rate of these variations, results indicated that feature variability was exponentially reduced to 30, 4 and 1 per cent when averaging five, ten and 15 cardiac cycles, respectively. The effect of averaging on the discriminant power of the features to separate the different categories of stenosis was also investigated by performing t-test analyses. Results showed that averaging between five and ten cardiac cycles provided the better discriminant power for most cases, whereas averaging over more than ten cardiac cycles was of little benefit. Based on the spectral analysis technique used in the present study, we conclude that averaging over ten cardiac cycles is sufficient for the analysis of Doppler spectrograms recorded in the lower limbs.     

Myocardial tissue motion influence on laser Doppler perfusion monitoring using tissue Doppler imaging

Tissue motion of the beating heart generates large movement artifacts in the laser Doppler perfusion monitoring (LDPM) signal. The aim of the study was to use tissue Doppler imaging (TDI) to localise intervals during the cardiac cycle where the influence of movement artifacts on the LDPM signal is minimum. TDI velocities and LDPM signals were investigated on three calves, for normal heartbeat and during occlusion of the left anterior descending coronary artery. Intervals of low tissue velocity (TDI_int<1 cm s⁻¹) during the cardiac cycle were identified. During occlusion, these intervals were compared with low LDPM signal intervals (LDPM_int<50% compared with baseline). Low-velocity intervals were found in late systole (normal and occlusion) and late diastole (normal). Systolic intervals were longer and less sensitive to heart rate variation compared with diastolic ones. The overlap between LDPM_int and TDI_int in relation to TDI_int length was 84±27% (n=14). The LDPM signal was significantly (p<0.001, n=14) lower during occlusion if calculated during minimum tissue motion inside TDI_int), compared with averaging over the entire cardiac cycle without taking tissue motion into consideration. In conclusion, movement artifacts are reduced if the LDPM signal is correlated to the ECG and investigated during minimum wall motion. The optimum interval depends on the application; late systole and late diastole can be used.     

In vivo evaluation of signal processors for laser Doppler tissue flowmeters

The performance of different signal processors for laser Doppler tissue flowmeters was evaluated by the use of a well defined flow model comprising a segment of the feline intestinal wall. The processor that, apart from being based on the calculation of the first moment of the power spectral density, also takes into account the effect of multiple scattering in a number of blood cells gave an output signal that was linearly related to the intestinal wall perfusion as recorded independently by a drop-counting technique. At a recording bandwidth of 12 kHz, this linear relationship was valid for the entire flow range 0–300 ml min⁻¹ 100 g⁻¹ (r=0·98). The processor based on the first moment of the power spectral density alone under-estimated the highest flow rates by about 35 per cent, while within the flow range 0–100 ml min⁻¹ 100 g⁻¹ this processor also gave an output signal linearly related to flow at a recording bandwidth of 12 kHz (r=0·96). When the bandwidth was limited to 4 kHz, the output signals from both processors were linearly related to flow only within the range 0–100 ml min⁻¹ 100 g⁻¹ (r=0·90). The output signals recorded with the 4 kHz systems were, however, generally only about 65 per cent of those recorded with the 12 kHz systems.     

The effects of aerobic interval training on the left ventricular morphology and function of VLCAD-deficient mice

This study examined the effect of aerobic interval training on cardiac adaptations in VLCAD-deficient mice and determined the effects of the deficiency on the morphology and function of the left ventricle among 53 knockout homozygous VLCAD−/−, 28 heterozygous VLCAD±, and 39 controls VLCAD+/+ male mice (129 SvJ/C57BL6). Echocardiographic images were used to determine the left ventricular (LV) wall thicknesses, during systole and diastole, acquired at a depth setting of 20 mm. Cardiac hypertrophy (as evidenced by increased wall thickness, and decreased left ventricular dimension in diastole and systole) appeared to be a major finding in the VLCAD−/− mouse with, however, normal %FS. The trained mice from all three genotypes exhibited lower body weight compared with their controls. The echocardiographic data of this study demonstrated structural but not functional differences among the three genotypes. This study demonstrated that VLCAD± deficient mice handled interval training similarly to the non-deficient mice. Four VLCAD−/− deficient mice died unexpectedly on the treadmill during the early stages of training. The VLCAD−/− deficient mice that survived adapted to the aerobic interval training similarly to the non-deficient mice. It is unclear whether aerobic interval training is an appropriate training tool for the VLCAD-deficient humans.     

Effect of flow disturbances remaining at the beginning of diastole on intraventricular diastolic flow and colour M-mode Doppler echocardiograms

A computational model of the fluid dynamics of intraventricular flow was used to investigate the importance of the effects of flow disturbances existing within the left ventricle (LV) at the onset of diastole on a diastolic flow field. The simulation started with a quiescent flow state; it continued for a number of cardiac cycles to obtain a cyclically repeatable flow. After the flow became periodic, the initial diastolic flow was not quiescent: flow disturbances, remnants of a systolic flow, were present within the LV. Nevertheless, they faded away during an acceleration phase of diastole and almost ceased by the end of this phase. Consequently, a flow field during a deceleration phase of diastole, characterised by the formation of a vortex ring, was hardly affected by the initial flow disturbances. The propagation velocity of a colour M-mode Doppler echocardiogram obtained by scanning velocity along the LV long axis was 0.58ms⁻¹ in the case where diastolic flow was initially quiescent and 0.56ms⁻¹ in the case where flow disturbances existed at the beginning of diastole. These results indicated that the colour M-mode Doppler echocardiographic technique captures flow dynamics produced purely by ventricular expansion, with little influence from initial diastolic flow disturbances.     

Identification of canine coronary resistance and intramyocardial compliance on the basis of the waterfall model

This study was performed to elucidate the effects of cardiac contraction on coronary pressure-flow relations. On the basis of the waterfall mechanism, a lumped model of the coronary arterial system is presented consisting of a proximal (epicardial) compliance, a coronary resistance, and an intramyocardial compliance. A “back”-pressure, assumed to be proportional (constant k) to left ventricular pressure, impedes flow. From steady-state measurements of circumflex coronary artery flow and inflow pressure, together with left ventricular pressure, the values of the three model parameters and the constant k have been estimated. In the control condition proximal compliance is found to be 1.7×10⁻¹² m⁴s²kg⁻¹, intramyocardial compliance 110×10⁻¹²m⁴s²kg⁻¹, and resistance 7.5×10⁹kgm⁻⁴s⁻¹. The proportionality constant k is close to unity. Effects of changes in left ventricular pressure and inflow pressure and the effect of vasoactive drugs on the parameters are also investigated. Changes in coronary resistance are always opposite to changes in intramyocardial compliance. Sensitivity analysis showed that epicardial compliance plays its major role during isovolumic contraction and relaxation; resistance plays a role throughout the cardiac cycle but is more important in diastole than in systole, whereas intramyocardial compliance plays a role in systole and in early diastole.     

Dynamic model of the short-term regulation of arterial pressure in the cat

The purpose of this study was to characterise the dynamics of the short-term control of arterial pressure in the cat with the aid of a model consisting of a nonlinear negative-feedback control system. The arterial system was described by a three element windkessel model (peripheral resistance, R, aortic characteristic impedance, R_c, and total arterial compliance, C). The resistance regulation was represented by a second-order system with static gain G_R, a damping factor σ and an undamped natural frequency ω_n. The resistance gain, G_R, and the windkessel parameters were obtained from measurements of aortic and venous pressures and cardiac output in two steady states. The parameters σ and ω_n were estimated from mean pressure and mean flow during the transient from control to the new steady state. Pressure reductions averaged 10 per cent and resistance changes averaged 12 per cent. Average windkessel model parameters in the control condition were: C=(25·9±6·1) 10⁻⁶ g⁻¹ cm⁴ s², R_c=(2·51±0·53) 10³ g cm⁻⁴ s⁻¹, R=(40·9±9·8) 10³ g cm⁻⁴ s⁻¹. Average estimates of parameters of the resistance regulator were: G_R=(4·14±2·38) 10⁻³ min ml⁻¹, ω_n = 1·0 ± 1·0 rad s⁻¹, σ=0·41±0·19. A satisfactory fit was found between model predicted and measured pressure. The results suggest that the dynamic short-term control of pressure is underdamped and oscillatory. The amplitude of these oscillations is affected by arterial compliance, suggesting an interaction between the arterial system and short-term resistance regulation.     

Analysis of cardiac left-ventricular volume based on time warping averaging

The cardiac left-ventricular (LV) volume signal, obtained by acoustic quantification, is affected by noise and respiratory modulation, resulting in a large beat-to-beat variability that affects the computation of LV function indices. A new method is proposed to improve the evaluation of LV indices by applying a signal averaging technique based on dynamic time warping to consecutive LV volume waveforms. Volume signals obtained from ten normal young (NY) subjects (mean age±SD: 25±5 years) were used to evaluate the performance of this algorithm. To evaluate its clinical utility, the effects of ageing and pharmacologically induced changes on LV function were assessed by studying, respectively, ten normal (N) adult subjects (age 64±8 years) and ten patients with dilated cardiomyopathy during a control and low-dose dobutamine (10 μg kg⁻¹ min⁻¹) study. Indices of LV function were highly consistent, with a variability of less than 8%, even when only 16 beats were averaged, independently of their selection inside the whole recording. When compared with beat-to-beat measures, the averaging of 16 beats significantly reduced (by more than 50%) the interbeat variability of all indexes. Expected alterations in both diastolic and systolic function were evidenced both with ageing (peak filling atrial contraction and ejection rates: from 275±77 ml s⁻¹, 76±30 ml s⁻¹, 230±70 ml s⁻¹, respectively, in NY, to 160±33 ml s⁻¹, 125±39 ml s⁻¹, 163±54 ml s⁻¹ in N) and with dobutamine (peak filling and ejection rates from 160±72 ml s⁻¹ and 183±86 ml s⁻¹ respectively, in control, to 253±75 ml s⁻¹ and 251±105 ml s⁻¹ with dobutamine). Signal averaging with time warping allows fast and improved assessment of LV function.     

Computer studies of systemic and regional blood flow mechanisms during cardiopulmonary resuscitation

Based on the new concept that intrathoracic and abdominal pressure variations cause blood flow in most of the cardiopulmonary resuscitation (CPR) techniques, two mathematical models were developed to explore related mechanisms of blood flow. The models were based on a representation of the cardiovascular system by resistive, capacitive and inductive elements, and the existence of venous and cardiac unidirectional valves. Cyclic intrathoracic and abdominal pressure variations were simulated by modulating the pressure within the corresponding vessels. It was found that blood flow during CPR is highly dependent on venous valving and aortic valve competence. The systemic blood flow was calculated to be between 10 and 20 per cent of its normal value. The maximum flow under a cyclic pressure of 50 mmHg was 663 m/min⁻¹, which was achieved with a pulse rate of 115 cycles per min and a duty cycle (ratio of artificial systole to cycle duration) of 58 per cent. The coronary blood flow was found to occur only during artificial diastole and was actually reversed during the compression phase. The systemic blood flow increased when pressure variations were delivered to the chest alone or when some phase lag was introduced between the thoracic and abdominal pressure waves. The mathematical model presented provided a tool to study the effect of thoracic and abdominal pressure waves on the circulation in CPR. The information derived from the model can be used to design better methods for CPR.     

Optimum control of the Hemopump as a left-ventricular assist device

A general framework for designing an optimum control strategy for the Hemopump is described. An objective function was defined that includes four membership functions, each constructed based on the desired values of one of the four members: stroke volume, mean left atrial pressure, aortic diastolic pressure and mean pump rotation speed. The Hemopump was allowed to operate either at a constant speed or at two different speeds during a cardiac cycle. The goal was to maximise the objective function by varying the magnitude and timing of the pump speed. Using a canine circulatory model, it was demonstrated that, in general, different cardiac conditions or different clinical objectives require different operation parameters. For example, when a left ventricle with minor ischaemia was simulated, and the main objective was to increase stoke volume, the objective function was maximised, from a value of 0.877 when the pump was off, to 0.946 when the pump was operated at speed 2 (18 500-revolutions min⁻¹). On the other hand, for a severely ischaemic heart, the optimum pump speed became speed 3 (20 000 revolutions min⁻¹), which maximized the objective function to 0.943 (from 0.707 when the pump was off). The results also suggest that it is more beneficial to operate the Hemopump at two different speeds during a cardiac cycle (a higher speed during systole and early diastole, and a lower speed during late diastole) than to maintain a constant speed throughout the cardiac cycle.     

Angiotensin type 1 (AT1) and type 2 (AT2) receptors mediate the increase in TGF-β1 in thyroid hormone-induced cardiac hypertrophy

Increased thyroid hormone (TH) levels are known to induce cardiac hypertrophy. Some studies have provided evidence for a functional link between angiotensin II (ANG II) and transforming growth factor β1 (TGF-β1) in the heart, both being able to also induce cardiac hypertrophy. However, the contribution of this growth factor activated directly by TH or indirectly by ANG II in cardiac hypertrophy development remains unknown. To analyze the possible role of TGF-β1 in cardiac hypertrophy induced by TH and also to evaluate if the TGF-β1 effect is mediated by ANG II receptors, we employed Wistar rats separated into control, hypothyroid (hypo) and hyperthyroid (T4 − 10) groups combined or not with ANG II receptor blockers (losartan or PD123319). Serum levels of T3 and T4, systolic pressure and heart rate confirmed the thyroid state of the groups. The T4 − 10 group presented a significant increase in cardiac TGF-β1 levels; however, TGF-β1 levels in the hypo group did not change in relation to the control. Inhibition of the increase in cardiac TGF-β1 levels was observed in the groups treated with T4 in association with losartan or PD123319 when compared to the T4 − 10 group. These results demonstrate for the first time the TH-modulated induction of cardiac TGF-β1 in cardiac hypertrophy, and that this effect is mediated by ANG II receptors.     

Attenuated β-adrenoceptor-mediated cardiac contractile responses following androgenic steroid administration to sedentary rats

Androgenic steroids administered in doses at pharmacological levels to sedentary animals have been shown to result in a reduced β-adrenoceptor-mediated increase in systolic cardiac performance when assessed in vivo. Whether the attenuated adrenergic response occurs as a consequence of alterations in either cardiac loads, heart rate, modifications in left ventricular (LV) geometry, or a decrease in myocardial contractile performance has not been determined. In this study the effect of chronic administration (over 3 months) of an androgenic steroid (nandrolone decanoate, 5 mg · kg⁻¹ biweekly) on the response of load-insensitive indices of myocardial contractile function [the slope of the LV systolic stress-strain relationship (LV-Eⁿ _max, where Eⁿ _max is systolic myocardial elastance)] to an adrenergic-inotropic stimulus was examined ex vivo in paced rat hearts. Systolic cardiac performance was assessed at 300 beats · min⁻¹ in isolated constant flow perfused heart preparations both before and during 10^−8.5 mol · l⁻¹ isoproterenol (ISO) infusion (approximate concentration of ISO eliciting 50% maximal inotropic response to ISO). Steroid administration resulted in left-shifted LV systolic and diastolic pressure-volume (P-V ) relationships. The leftshifted P-V relationships were attributed, in part, to increased slopes of these relationships. However, the steroid-mediated increment in the slope of the systolic P-V relationship (systolic chamber elastance, E_max) was not associated with a similar change in LV Eⁿ _max [control 19.2 (SEM 2.1) g · cm⁻², steroid 18.3 (SEM 2.4) g . cm⁻²] as determined in the absence of ISO. Isoproterenol infusion resulted in an increase in both E_max and Eⁿ _max in the control rats, without altering systolic performance in the steroid treated rats. Consequently, in the presence of ISO, the steroid treated rats exhibited a similar E_max, but a reduction in Eⁿ _max compared to the control rats [control 25.6 (SEM 1.9) g · cm⁻², steroid 18.5 (SEM 1.5) g · cm⁻²; P < 0.05]. In conclusion, these results would suggest that chronic high dose androgenic steroid administration produces a decrease in myocardial contractile reserve to β-adrenoceptor stimulation. Accepted: 3 September 1999     

The oxygen binding capacity of human haemoglobin

Summary The internationally standardized HiCN method for measuring the total haemoglobin concentration of the blood (c _Hb ^* ) is based on the chemical composition of the haemoglobin molecule. If all haemoglobin measured in the blood by this method were active, i.e., capable of reversibly binding O₂, the O₂ binding capacity (β) would be 1.39 ml · g⁻¹ The experimental values of β to be found in the literature vary over a wide range, values as low as 1.24 ml · g⁻¹ being no exception. We determined β for 36 blood samples of healthy men, using for the determination of the O₂ content 3 independent methods, i.e., a manometric, a polarographic and a chemical one. Before the measurements the blood was equilibrated with humidified pure O₂ for 3 h. Appropriate corrections for dissolved O₂ were made. The mean β was 1.368 ml · g⁻¹, with a s.d. of 0.017 ml · g⁻¹ (median 1.372 ml · g⁻¹; range 1.322–1.387 ml · g⁻¹). This mean β corresponds to 98.4% active haemoglobin. The inactive haemoglobin consisted of 0.7% HbCO, Hi and SHb (“dyshaemoglobin”) and 0.9% unidentified inactive haemoglobin. In 4 samples a high fraction of unidentified inactive haemoglobin was found (2.8–4.3%). Taking into account that in addition to this fraction a considerable amount of dyshaemoglobin (especially HbCO) may be present in the blood of normal men, it must be concluded that calculating the O₂ carrying capacity of the blood by multiplyingc _Hb ^* by a constant value of β, may be subject to an appreciable error. If β=1.39 ml · g⁻¹ is used, the O₂ carrying capacity of the blood may easily be found 10–15% too high.     

High-intensity exercise decreases muscle buffer capacity via a decrease in protein buffering in human skeletal muscle

We have previously reported an acute decrease in muscle buffer capacity (βm_{in vitro}) following high-intensity exercise. The aim of this study was to identify which muscle buffers are affected by acute exercise and the effects of exercise type and a training intervention on these changes. Whole muscle and non-protein βm_{in vitro} were measured in male endurance athletes (VO_2max = 59.8 ± 5.8 mL kg⁻¹ min⁻¹), and before and after training in male, team-sport athletes (VO_2max = 55.6 ± 5.5 mL kg⁻¹ min⁻¹). Biopsies were obtained at rest and immediately after either time-to-fatigue at 120% VO_2max (endurance athletes) or repeated sprints (team-sport athletes). High-intensity exercise was associated with a significant decrease in βm_{in vitro} in endurance-trained males (146 ± 9 to 138 ± 7 mmol H⁺·kg d.w.⁻¹·pH⁻¹), and in male team-sport athletes both before (139 ± 9 to 131 ± 7 mmol H⁺·kg d.w.⁻¹·pH⁻¹) and after training (152 ± 11 to 142 ± 9 mmol H⁺·kg d.w.⁻¹·pH⁻¹). There were no acute changes in non-protein buffering capacity. There was a significant increase in βm_{in vitro} following training, but this did not alter the post-exercise decrease in βm_{in vitro}. In conclusion, high-intensity exercise decreased βm_{in vitro} independent of exercise type or an interval-training intervention; this was largely explained by a decrease in protein buffering. These findings have important implications when examining training-induced changes in βm_{in vitro}. Resting and post-exercise muscle samples cannot be used interchangeably to determine βm_{in vitro}, and researchers must ensure that post-training measurements of βm_{in vitro} are not influenced by an acute decrease caused by the final training bout.     

Extracellular pH defense against lactic acid in untrained and trained altitude residents

The assumption that buffering at altitude is deteriorated by bicarbonate (bi) reduction was investigated. Extracellular pH defense against lactic acidosis was estimated from changes (Δ) in lactic acid ([La]), [HCO₃ ⁻], pH and PCO₂ in plasma, which equilibrates with interstitial fluid. These quantities were measured in earlobe blood during and after incremental bicycle exercise in 10 untrained (UT) and 11 endurance-trained (TR) highlanders (2,600 m). During exercise the capacity of non-bicarbonate buffers (β _nbi = −Δ[La] · ΔpH⁻¹ − Δ[HCO₃ ⁻] · ΔpH⁻¹) amounted to 40 ± 2 (SEM) and 28 ± 2 mmol l⁻¹ in UT and TR, respectively (P < 0.01). During recovery β _nbi decreased to 20 (UT) and 16 (TR) mmol l⁻¹ (P < 0.001) corresponding to values expected from hemoglobin, dissolved protein and phosphate concentrations related to extracellular fluid (ecf). This was accompanied by a larger decrease of base excess after than during exercise for a given Δ[La]. β _bi amounted to 37–41 mmol l⁻¹ being lower than at sea level. The large exercise β _nbi was mainly caused by increasing concentrations of buffers due to temporary shrinking of ecf. Tr has lower β _nbi in spite of an increased Hb mass mainly because of an expanded ecf compared to UT. In highlanders β _nbi is higher than in lowlanders because of larger Hb mass and reduced ecf and counteracts the decrease in [HCO₃ ⁻]. The amount of bicarbonate is probably reduced by reduction of the ecf at altitude but this is compensated by lower maximal [La] and more effective hyperventilation resulting in attenuated exercise acidosis at exhaustion.     

Effect of pulsatile arterial diameter variations on blood flow estimated by Doppler ultrasound

High-resolution measurements of common carotid and femoral arterial diameters have been performed by ultrasound echo devices. When combined with pulsed Doppler measurements of cross-sectional averaged velocity in the same vessels, exact calculations of flow were made possible. The median peak-to-peak pulsatile diameter variations were 0.19 mm (2.8 per cent) in the femoral artery and 0.49mm (6.7 per cent) in the common carotid artery. Flow values were calculated either by taking the time-averaged diameter as a constant value, or by taking into account the dynamic variations in diameter. In comparing the two values, a quantification of the magnitude of error introduced by the averaging of the diameter was made possible. An error in the range 1.5–3.8 per cent was found for the femoral artery, whereas the error in the common carotid artery was in the range 0.4–3.6 per cent despite the larger amplitude of the pulsations in this vessel.     

An Evaluation of the Ensemble Averaged Impedance Cardiogram

The ensemble averaged impedance cardiogram was evaluated during rest and during vocal mental arithmetic stress in samples of female (n = 17) and male (n = 40) undergraduates. Measures of myocardial performance determined from ensemble averaged signals were compared to those determined by simple beat-to-beat averaging over 60-s and 20-s sampling intervals. In addition, the influence of cardiac interval variability on dZ/dt amplitude measures was assessed. As expected, measures derived by the two averaging techniques corresponded closely, with correlations ranging from r = 0.882 to r = 1.000 for the 60-s sampling intervals, and from r = 0.726 to r = 1.000 for the 20-s sampling intervals. Inverse relationships of comparable magnitude were found between cardiac interval variability and dZ/dt amplitude measures determined by both averaging techniques, suggesting that these relationships did not result simply from artifacts introduced by the ensemble averaging technique. The results support the validity of ensemble averaging as a method for deriving impedance cardiographic measures of myocardial performance.     

Non-invasive assessment of cardiac function

Noninvasive assessment of cardiac function by Doppler echocardiography is reviewed. The heart propels the blood through the repeated sequence of systole and diastole. The systolic function is essential to maintain the biological function of the whole body. However, before the heart ejects the blood during systole, the heart must be filled up with blood during the preceding diastole. Thus, the diastolic function is as important as the systolic function. Although the diastolic function is traditionally assessed by hemodynamic parameters obtained in the cardiac catheterization laboratory, it has routinely been assessed by Doppler echocardiography in the echocardiographic laboratory in recent years. Since the concept of diastolic failure has widely spread, the important role of the transmitral flow in assessing the diastolic function has been well recognized. Besides the transmitral flow, the modalities for clinical assessment of the left ventricular diastolic function have been well developed. For example, the pulmonary venous flow has been easily obtained by a transthoracic approach, and the tissue Doppler technique provides important information about the diastolic function, and furthermore the color M-mode is applied for the flow propagation velocity of the left ventricular inflow. These modalities make it possible to assess the left ventricular diastolic function more precisely by Doppler echocardiography. Also, TEI index, strain rate and strain imaging, and wave intensity are mentioned.     

Usefulness of sodium chloride as a nondiffusible indicator in the measurement of extravascular lung thermal volume in dogs

The authors examined the usefulness of sodium chloride as a nondiffusible indicator during the first passage through dogs' lungs, before and after increased-permeability pulmonary oedema produced by an intravenous injection of alloxan. With an injection of a mixture of ice-cold 3 per cent sodium chloride and indocyanine green dye (a nondiffusible reference indicator), the authors simultaneously recorded three dilution curves from the aortic root: dye dilution, thermal and blood electrical conductivity dilution curves in six dogs. The mean transit time of sodium chloride in the conductivity dilution curve was significantly different from, but fairly equal to, that of indocyanine green dye (6·2±1·4 s (mean±SD) against 6·5±1·4s(p<0.01) in the baseline period, and 7·6±1·9 s against 8·4±2·1 s (p<0·01) in the oedema period, respectively). The calculated extravascular lung thermal volume with the thermal and conductivity dilution method (Y, ml kg⁻¹) correlated well with the gravimetrically determined extravascular lung mass in a total of 12 dogs, including six other dogs without intervention (x, g kg⁻¹) (y=0·72x±3·03, r=0·96). The authors conclude that sodium chloride is useful as a nondiffusible indicator in the first passage through the lungs, and that the thermal and conductivity dilution method is also useful for measuring extravascular lung water mass.     

Bone mineral and fat measurement with a novel dual photon absorptiometer

The authors have produced an original dual-photon absorptiometer, with specific algorithms, permitting real-time automatic bone recognition, to reduce examination time (to 5 min). It permits the simultaneous measurement of the mineral density of an area of bone (BMD), the fat percentage of the soft tissues (PFST) that surround it, and the variation in this percentage. It visually represents, on one and the same picture, freely defined in terms of dimension and definition, bone mineral distribution and fat mass distribution. The performances are analysed, using phantoms and taking ten control subjects, in the region of the femoral neck. The variations in PFST in the control subjects are compared with the measurements carried out in MRI in the same region. In all these control subjects, there is a regular decrease in PFST in this region along an upper external/lower internal axis. This decrease has been likened to a regression line, the slope of which, representing between −0·16 per cent of fat per cm and −1·7 per cent per cm, has a mean value of −0·81 per cent per cm. The BMD is thus reduced by 0–2·5 per cent, depending on the subjects.