Evaluation of Artificial Intelligence-Powered Identification of Large-Vessel Occlusions in a Comprehensive Stroke Center

BACKGROUND AND PURPOSE

Artificial intelligence algorithms have the potential to become an important diagnostic tool to optimize stroke workflow. Viz LVO is a medical product leveraging a convolutional neural network designed to detect large-vessel occlusions on CTA scans and notify the treatment team within minutes via a dedicated mobile application. We aimed to evaluate the detection accuracy of the Viz LVO in real clinical practice at a comprehensive stroke center.

MATERIALS AND METHODS

Viz LVO was installed for this study in a comprehensive stroke center. All consecutive head and neck CTAs performed from January 2018 to March 2019 were scanned by the algorithm for detection of large-vessel occlusions. The system results were compared with the formal reports of senior neuroradiologists used as ground truth for the presence of a large-vessel occlusion.

RESULTS

A total of 1167 CTAs were included in the study. Of these, 404 were stroke protocols. Seventy-five (6.4%) patients had a large-vessel occlusion as ground truth; 61 were detected by the system. Sensitivity was 0.81, negative predictive value was 0.99, and accuracy was 0.94. In the stroke protocol subgroup, 72 (17.8%) of 404 patients had a large-vessel occlusion, with 59 identified by the system, showing a sensitivity of 0.82, negative predictive value of 0.96, and accuracy of 0.89.

CONCLUSIONS

Our experience evaluating Viz LVO shows that the system has the potential for early identification of patients with stroke with large-vessel occlusions, hopefully improving future management and stroke care.

Evaluation of a new disinfection procedure for ultrasound probes using ultraviolet light

Following 183 ultrasound examinations, a randomized trial was conducted to compare three procedures for disinfection of probes under routine conditions: dry wiping with a soft, dry, non-sterile paper towel, antiseptic wiping with a towel impregnated with disinfectant spray and dry wiping followed by a 10 min ultraviolet C (UVC) cycle in a disinfection chamber. After ultrasonography, swabs were taken from transducer heads before and after cleaning and streaked onto plates that were then cultured. The number of colonies per plate was counted and organisms identified. The median microbial reduction was 100% for UVC, 98.4% for antiseptic wiping and 87.5% for dry wiping (P<0.001). The percentage of negative specimens was 88% for UVC, 16% for antiseptic wiping and 4% for dry wiping (P<0.0001). Microbial flora was isolated from 12 probes (6.6%) before cleaning, whereas specimens obtained after cleaning contained no pathogens except in one case after antiseptic wiping. UVC disinfection of ultrasound probe may provide a useful method for reducing the bacterial load under routine conditions.

[Echocardiography in athletes]

Echocardiography of athletes focuses on the left ventricle: in many cases, the interpretation is difficult, showing features of either dilated of hypertrophic cardiomyopathy. Accurate analysis requires careful recording of the left ventricle in the fundamental mode. Diagnostic errors by excess are common: overestimation of the chamber size (inadequate incidence), overestimation of wall thickness (failure to recognise right ventricular bands or false tendons). The body size of the individual must be considered when interpreting the chamber dimensions. The diastolic left ventricular dimension is often < 60 mm but it may attain 70 mm. In these cases, left ventricular function must be accurately determined and shown to be normal, which may prove difficult as the resting left ventricular ejection fraction is physiologically lower in trained athletes. The use of stress echo (exercise left ventricular ejection fraction) and Doppler tissue imaging (left ventricular filling) are often indispensable. Left ventricular wall thickness is rarely > 13 mm. a situation where the possibility of hypertrophic cardiomyopathy has to be assessed. Left atrial dilatation is inconstant and aortic root dilatation is very unusual. The right heart chambers are usually globally dilated and the inferior vena cava may seem abnormally dilated.

[The best of echocardiography in 2005]

Echocardiography continues to evolve in several directions: on the technical front, various tools are being developed and real time 3D is starting to find a place in routine practice as much as contrast is used unequally depending on the country. On the diagnostic front, certain aspects have been particularly emphasised this year: The key role of quantification of valvulopathies in order to guide management, both at rest and on effort. The place of echocardiography in asynchronism, even if a more consensual strategy is needed to take into account the different approaches. The very good prognostic series on stress echography in the context of myocardial ischaemia. Advances in the evaluation of filling pressures have allowed a better definition of the limits, and have refined the understanding of so-called 'diastolic' cardiac failure. Finally, the role of echography to guide and also evaluate the results of foramen ovale or interatrial communication closure.

[Echocardiographic indices of diastolic function put to the practical test]

Assessment of left ventricular filling pressures is now part of routine echocardiographic examination. It is a semi-quantitative evaluation usually based on analysis of transmitral blood flow. Restrictive type flow associated with poor left ventricular contraction is pathognomonic of increased left ventricular filling pressure: conversely, inversion of the mitral flow velocities is generally reassuring except in the presence of left ventricular hypertrophy. In other cases, it is necessary to complete the analysis with indices of left ventricular relaxation. The authors discuss the advantages of pulsed Doppler tissue imaging at the mitral ring over the propagation velocity in colour M-mode, especially when left ventricular contraction is normal or relatively preserved. In particular, the reproducibility of propagation velocity measurements is poor when the velocities are relatively high (> 50 cm/s) because a small error in measurement at these velocities has large repercussion on the value of the slope. Pulmonary vein flow, reputedly difficult to analyse, may be very valuable. Discordance of the different indices is one of the most difficult daily problems to resolve. However, it is possible to give a reliable "normal filling pressures" or "increased filling pressures" conclusion in over 80% of cases. Echocardiography cannot provide a precise study of diastole and terms such as abnormal relaxation, abnormal compliance or diastolic dysfunction should be avoided.

[The best of echocardiography in 2004]

What major trends can we single out from the large amount of literature that appeared in 2004? The role of echography in the investigation of asynchronism has most certainly been better defined. Real time 3D echography, now routinely available, has started to gain respect in different pathologies. The place of filling pressure evaluation in patients' management, as much diagnostic as therapeutic, has again been refined. The prognostic contribution of stress echography has been emphasized in particular this year. Despite a still limited routine role, publications regarding contrast techniques have been quite numerous: not only diagnostic contrast echography, but also contrast and pleiotropic gene transfer in the myocardium. The study of coronary reserve via the transthoracic route, possibly assisted with contrast, should become established as a routine technique.

[Echographic calculation of left ventricular mass in TM mode]

TM sections of the left ventricle (LV) have only been validated in fundamental (F) imaging. We were interested in evaluating the repercussions of new imaging techniques, harmonic (H) and colour tissue Doppler (CTD), on the TM measurement of left ventricular mass (LVM), the reference imaging being F imaging.

METHOD

We performed a prospective study, including 26 patients with a valid TM section. The LV and LVM parameters in F, H and CTD mode according to the Penn and ASE conventions as well as the inter-observer reproducibility were studied.

RESULTS

The correlations for the LVM measurements between F and H, and between F and CTD were high whichever convention was used (r>0.95, p<0.0001). For each observer, the LVM in H and in CTD was always greater than the LVM in F with both conventions (p<0.02). A false diagnosis of LV hypertrophy was made in 27% of patients in H and in 15% of patients in CTD. The best inter-observer reproducibility was obtained in H: the average inter-observer difference (gr.) was 23+/-15 for H, 32+/-19 for F and 59+/-18 for CTD.

CONCLUSION

H and CTD imaging entail an overestimation of LVM, essentially by overestimation of the parietal thickness of the LV. The inter-observer reproducibility was excellent in H and poor in CTD. The use of the harmonic mode for LVM calculation must be validated using new formulae.

[Hemodynamic assessment and prognostic data obtained by echography in cardiac insufficiency]

The evaluation and follow up of insufficiency are matters of "routine" echography. The relevant parameters allow an evaluation of the contractility of the left ventricle (LV), cardiac output, filling pressures, significance of mitral insufficiency (MI), pulmonary pressures, and volumes, and one of the aims is to be able to offer an exhaustive haemodynamic assessment without unnecessarily prolonging the length of the examination. Calculation of the LV ejection fraction is an unavoidable parameter (it is often the inclusion criterion in large studies), recently "rehabilitated" thanks to harmonic imaging. However, the dependence on this index with respect to load conditions, is in fact a very imperfect reflection of LV contractility. The calculation of systolic ejection volume with Doppler (for example from the outflow chamber diameter and sub aortic flow) is a better reflection of LV performance. In the same manner, analysis of the ascending slope of MI flow during isovolumetric contraction (dp/dt) recorded with continuous Doppler allows a reasonably reliable and simple approach to LV contractility, only slightly dependent on load conditions. Numerous parameters allow a reliable evaluation of LV filling pressures: this always relies on the transmitral flow morphology, which has been better interpreted for several years, possibly coupled with a recording of pulmonary venous flow, or even a colour TM mode recording of LV filling or a pulsed tissular Doppler flow recording at the mitral ring. Analysis of the right side of the heart consists of evaluation of the size of the cavities and quantification of the tricuspid flow. Even if the flow is not laminar, it allows a reliable measurement of the pulmonary pressures (by default, pulmonary flow is determined). TM recording in the inferior vena cava has to be systematic to allow evaluation of the right-sided pressures and volumes. Interpretation of these various parameters allows a subtle haemodynamic evaluation in severe cardiac insufficiency, for which the significance is not only diagnostic, but also prognostic: adverse effect of low LVEF; adverse effect of restrictive mitral flow (E/A ratio > 2 and in particular short E deceleration time less than 150 ms), especially if it is not modified by treatment.

[Quantification of valvular regurgitation by the convergence zone method]

The convergence zone method is currently used routinely to quantify valvular regurgitation, particularly mitral insufficiency, but also aortic and tricuspid regurgitation. It is quick to perform, requires relatively little training and remains valid in the presence of atrial fibrillation and in the association of regurgitation-stenosis. It relies on the measurement of output at the level of an isovelocity hemisphere and allows calculation of the regurgitant orifice surface (SOR) and the regurgitant volume per beat (VR). It is reserved for severe regurgitation corresponding to angiographic grade 4: SOR > 40 mm2 and VR > 60 ml for mitral insufficiency, SOR > 30 mm2 and VR > 60 ml for aortic insufficiency, and SOR > 40 mm2 and VR > 40 ml for tricuspid insufficiency.

[Quantification of mitral regurgitation]

The echographic methods of quantification of mitral regurgitation are various. Semiquantitative methods using the color Doppler extension of the regurgitant jet are now replaced by more quantitative methods, including PISA, jet width, and regurgitant fraction. Although sometimes difficult, accurate quantification of mitral regurgitation is now possible in a majority of patients using transthoracic echocardiography.

[Value of factor analysis in a wall motion study: preliminary application in the detection of left ventricular segmental contraction ischemic disorders by echocardiography]

PURPOSE

Factor Analysis of Medical Image Sequences (FAMIS) was tested to study the regional wall motion of the left ventricle at echocardiography.

MATERIALS AND METHODS

FAMIS analyzed the time signal curves of each pixel. One flat curve and one curve describing the contraction-relaxation of the left ventricle were first estimated. The contributions of the curve of each pixel to the two previous curves were computed, yielding two "factor images". The spatial distribution of positive and negative coefficients of the second factor image was analyzed. The evaluation was carried out on 222 segments (20 patients, 18 parasternal short-axis views, 17 apical four-chamber views, and 15 apical two-chamber views). A first echocardiographer reviewed the factor images and the reading was compared to the conventional reading of the cine-loops by two other echocardiographers. Each segment was scored as normal, hypokinetic, akinetic, or dyskinetic.

RESULTS

On normal segments, the positive coefficients of the second factor image were on the inner side, the negative coefficients were on the outer side. Dyskinesis yields the opposite pattern. Hypokinesis and akinesis give intermediate images. An absolute concordance was obtained on 71.2% of all segments between the two types of reading. Larger discrepancies were found for akinetic and hypokinetic segments.

CONCLUSION

FAMIS is a promising tool to study regional wall motion of the left ventricle.

Determinants of right ventricular pressure in mild hypertension

OBJECTIVE

Doppler echocardiography was used to define reference values and determinants of tricuspid regurgitation peak velocity (TRV) in hypertensive patients. A TRV value > 2.5 m/s is the threshold usually defining abnormal right ventricular systolic pressure.

DESIGN AND PATIENTS

Doppler echocardiography was performed in 320 consecutive uncomplicated hypertensive patients, without overt pulmonary or heart disease. Doppler echocardiography included LV mass measurement, LV inflow and pulmonary venous flow analysis, LV systolic function and TRV measurements.

RESULTS

Among 320 patients, 255 had normal TRV < 2.5 m/s and 65 had elevated TRV > or = 2.5 m/s. Compared with the normal TRV group, the elevated TRV group was older (60 versus 50 years, P < 0.0001), systolic blood pressure was higher (156 versus 151 mmHg, P = 0.02) and antihypertensive therapy was more frequent (68 versus 51%, P = 0.02); indexed LV mass was higher (45.4 versus 40.6 g/m2.7, P = 0.001), pulmonary D wave peak velocity was higher (42 versus 38 cm/s, P = 0.03). In univariate analysis, age was the most predictive variable of TRV (r = 0.36). In multivariate analysis, three variables were independently related to TRV: age, LV mass, pulmonary D wave (multiple r = 0.47).

CONCLUSION

In mild hypertension, TRV is independently related to age, and to a lesser extent, to LV morphology and LV filling pressure. In clinical practice, age should be taken into account to interpret TRV.

Genetic and environmental influences on left ventricular mass. A family study

The relations between left ventricular mass (LVM) with age, gender, body size, and blood pressure were investigated in healthy adults and children from 149 nuclear families. LVM was strongly correlated with overall weight, especially in the children. Genetic analysis indicated that the segregation of LVM was compatible with an additive polygenic model, with a heritability estimate of 34% before adjustment for weight and 28% after adjustment for weight. Genetic and/or familial environmental factors played a strong role in the correlation of LVM and weight; they accounted for all of the correlation between the 2 traits in adults and 59% of the correlation in children. Spouses exhibited a strong correlation in their weight, which suggested that common family environment may contribute to the family correlations and to the observed heritability of the trait. LVM was strongly correlated with blood pressure before adjustment for weight, but this correlation could be attributed to nonfamilial environment rather than genetic factors. After adjustment for weight, the intertrait correlations between LVM and blood pressure were nonsignificant. Thus, adjustment for weight accounts for all common determinants of LVM and blood pressure.

Can echocardiography identify mildly hypertensive patients at high risk, left untreated based on current guidelines?

OBJECTIVE

To determine whether the decision to treat uncomplicated mild hypertension with drugs, in accordance with the World Health Organization - International Society of Hypertension (WHO/ISH) guidelines based on a series of blood pressure (BP) measurements over 6 months, resulted in the treatment of patients at high risk on the basis of echocardiography.

METHODS

One hundred and eighteen patients with mild hypertension (diastolic blood pressure 90-105 mm Hg and/or systolic blood pressure 140-180 mm Hg) were examined by echocardiography at inclusion and followed up for 6 months by a single physician unaware of the echographic results.

RESULTS

Drug treatment was given to 48 patients, and 70 remained untreated. Treated patients had higher echographic indices than untreated patients (all P<0.05): left ventricular (LV) mass/body surface area (83.0+/-15.6 versus 75.3+/-14.8 g/m2), inter-ventricular septal thickness (9.7+/-1.7 versus 8.5+/-1.3 mm), LV posterior wall thickness (8.4+/-1.1 versus 7.8+/-1.1 mm), relative wall thickness (0.37+/-0.06 versus 0.34+/-0.06). LV geometry was normal in 98 patients, and 20 had LV concentric remodelling. The 10-year coronary disease risk (Framingham equation) was higher in the 20 patients with concentric remodelling than in those with normal LV geometry (10.4 versus 4.2%; P<0.005). Nine of these 20 patients were still untreated at the end of the 6-month follow-up period.

CONCLUSION

Rigorous application of the WHO/ISH clinical guidelines in a group of mild hypertensive patients led to the treatment of patients with slightly higher LV mass and more concentric LV geometry than were found in those not treated. However, a high-risk subgroup, with concentric remodelling, was not identified and left untreated.

Vascular and cardiac remodeling in world class professional cyclists

BACKGROUND

Numerous studies have demonstrated that left ventricular (LV) hypertrophy is often associated with conditioning.

METHODS AND RESULTS

The aim of the study was to evaluate cardiac and carotid artery changes induced by professional cycling. We collected M-mode left ventricle and B-mode right common carotid artery data from 149 male professional cyclists before the 1995 "Tour de France" race and 52 male control subjects. LV mass indexed to body surface area in cyclists was double that in control subjects, with no overlap of 95% confidence intervals (cyclists 100.9 to 187 g/m2 and control subjects 51.8 to 96.3 g/m2). Both mean arterial diameter and mean arterial diastolic intima-media thickness (IMT) were 13% higher in cyclists than in control subjects, with overlap of 95% confidence intervals (for arterial IMT 0.45 to 0.65 mm in cyclists and 0.38 to 0.60 mm in control subjects).

CONCLUSIONS

Our results suggest that intense cycling has an effect on the cardiovascular system, more pronounced on the left ventricle and less pronounced on large arteries. Nevertheless, athletic training should be considered as a potential determinant of carotid modification.

[Do WHO-ISH guidelines identify high risk mild hypertensive patients?]

OBJECTIVE

To assess the ability of 1993 WHO-ISH recommendations to identify patients who need drug treatment.

METHODS

268 hypertensive patients with suspected mild hypertension were preselected for this study at their first visit at the referral center. 123 were included after a short in-hospital work-up when they fulfilled the 1993 WHO-ISH criteria for mild hypertension (90-105 mmHg diastolic BP and/or 140-180 mmHg systolic BP). Echocardiography was performed in all patients by the same investigator according to ASE convention. The combined 10-year risk of stroke and coronary heart disease was calculated with the Framingham equation. Patients were then followed up for six months by the same physician blinded to echographic results and risk calculations and applying the WHO-ISH guidelines (monthly BP measurement and subjective assessment of risk). Five patients were excluded, for reasons unrelated to the protocol.

RESULTS

The decision of drug treatment was taken at the 1st, 2nd, 3rd, 4th, 5th, 6th monthly visit after work-up in 2, 6, 25, 7, 2 and 6 patients, respectively. Among these 118 patients, 48 patients (29 male, 19 female) were eventually treated and 70 (49 male, 30 female) remained untreated. BP s at preselection and on a day of work-up were similar in both groups. Patients in whom drug treatment was prescribed were older and had higher lef ventricular mass (LVM) than untreated patients, but only 2 of them (all in the treated group) had LVM values above usual thresholds (LVM > 125 g/m2, in men and women). Stroke and coronary risks were both higher in treated than in untreated patients (p < 0.05). The physician using the guidelines decided to treat only 19 of the 38 patients with a 10-year risk < 10% (true positive), whereas she decided to treat with drugs 12 patients among the 44 with a 10-year risk < 5%.

CONCLUSION

The difference in LVM between untreated and treated patients support the validity of the WHO-ISH guidelines, but the measurement of LVM did not bring much information for managing the individual patient. Application of these guidelines did not satisfactorily identify high risk patients and could lead to over-treatment of low risk patients.

[Can echocardiography identify mildly hypertensive patients at high risk, left untreated, based on current guidelines?]

OBJECTIVE

To determine if the decision to treat uncomplicated mild hypertension with drugs, in accordance with the WHO/ISH guidelines based on a series of blood pressure (BP) measurements over six months, resulted in the treatment of patients at high risk, on the basis of echocardiography.

BACKGROUND

The value of echocardiography in mild hypertension management remains is unclear.

METHODS

One hundred and eighteen patients with mild hypertension (90 to 105 mmHg diastolic BP and/or 140 to 180 mmHg systolic BP) were examined by echocardiography at inclusion and followed up for 6 months by a single physician unaware of the echographic results.

RESULTS

Drug treatment was given to 48 patients, and 70 remained untreated. Treated patients had higher echographic indices than untreated patients (all p < 0.05): LV mass/body surface area (82.8 +/- 15.9 vs 74.7 +/- 15.0 g/m2), interventricular septal thickness (9.7 +/- 1.7 vs 8.5 +/- 1.3 mm). LV posterior wall thickness (8.4 +/- 1.1 vs 7.8 +/- 1.1 mm), relative wall thickness (0.37 +/- 0.06 vs 0.34 +/- 0.06). Left ventricular (LV) geometry was normal in 98 patients, and 20 had LV concentric remodeling. The 10-year coronary disease risk (Framingham equation) was higher in treated patients than in untreated patients (10.0% vs 6.3%; p < 0.002), and in the 20 patients with concentric remodeling than in those with normal LV geometry (10.4%) vs 4.2%; p < 0.005). Nine of these 20 patients were still untreated at the end of the six-month follow-up period.

CONCLUSION

Rigorous application of the WHO-ISH clinical guidelines in a group of mild hypertensive patients, led to the treatment of patients with slightly higher LV mass and more concentric LV geometry than were found in those not treated. However, a high-risk subgroup, with concentric remodeling, was not identified and left untreated.

The age of the patient should be taken into account when interpreting Doppler assessed pulmonary artery pressures

The impact of aging on the systolic artery pressure (SPAP) value is not well known. We selected 134 echocardiographic Doppler examinations considered as normal (presence of a sinus rhythm, absence of chronic obstructive pulmonary disease or pulmonary embolism, normal global or segmental wall motion, no right or left ventricular hypertrophy or dilatation, no significant valvular disease, no pericarditis), with a clearly measurable tricuspid insufficiency allowing us to measure the SPAP with the simplified Bernoulli equation. There was a highly significant (p = 0.0001) correlation (r = 0.47) between SPAP values and the age of the patient. SPAP increased progressively with age from 23 +/- 5 mm Hg between 20 to 29 years old to 32 +/- 6 mm Hg when 80 years old or more. The interpretation of the SPAP should take into account the age.

Doppler-derived pulmonary arterial systolic pressure in patients with known systemic arterial pressures

In the present study, we evaluated Doppler-derived systolic pulmonary artery pressure in 476 hypertensive patients. Clinical interpretation of systolic pulmonary arterial pressure should take into account age, body size, and systolic blood pressure values.

Influence of obesity on the diagnostic value of electrocardiographic criteria for detecting left ventricular hypertrophy

Easily applicable, clinically relevant electrocardiographic criteria are needed to screen large populations for left ventricular (LV) hypertrophy. The aim of this study was to evaluate, in a population of 380 hypertensive patients of both sexes, whether obesity modified the diagnostic performance of Sokolow-Lyon and Cornell voltage criteria by comparing them with echocardiographic evaluations using different indexation methods for LV mass presentation (body surface area and various powers of the height variable). For the population as a whole, Cornell voltage was better correlated to LV mass than was Sokolow-Lyon voltage (r = 0.48 and 0.36, respectively). The poorest performance of Sokolow-Lyon voltage was observed among obese patients (best r = 0.1 and 0.21 in obese women and men, respectively). Sensitivities were assessed at a 95% specificity level. In nonobese patients, using sex-adjusted voltage values (43 and 36 mm in men and women, respectively, for Sokolow-Lyon voltage, and 28 and 25 mm for Cornell voltage), the sensitivities of Cornell voltage and Sokolow-Lyon voltage were similar in men and women (near 22% and 36%, respectively), whatever the indexation method used for LV mass. In obese patients, Cornell voltage sensitivity was similar to that of nonobese patients, whereas Sokolow-Lyon voltage had a much poorer sensitivity (<10%). For simple LV hypertrophy detection criteria, Sokolow-Lyon voltage should be avoided in obese hypertensive patients and replaced by the Cornell voltage criteria, which are not influenced by the presence of obesity.

Common carotid artery stiffness and patterns of left ventricular hypertrophy in hypertensive patients

The aim of this study was to determine the relationship between the lumen diameter and function of the common carotid artery, a vessel representative of the capacitance portion of the circulation, and the different patterns of left ventricular hypertrophy in uncomplicated essential hypertensive patients. Carotid luminal diastolic cross-sectional area, distensibility, and compliance were derived from measurements by a high-definition echotracking system. Left ventricular dimensions were from echocardiography. The 86 hypertensive patients included 31 who had never been treated (group 1), 31 in whom treatment had been stopped for at least 2 weeks (group 2), and 24 treated patients (group 3). In multivariate analysis of the population as a whole, the following relations were statistically independent of age, blood pressure, gender, and group: Left ventricular end-diastolic volume index was positively correlated to carotid luminal cross-sectional area (r = .46, P < .0001) and compliance (r = .47, P < .0001); left ventricular mean wall thickness and mass-volume ratio were negatively correlated to distensibility (r = -.68, P < .0001; r = -.46, P < .0001, respectively) and compliance (r = -.40, P < .0001; r = -.37; P < .001, respectively); and left ventricular mass index was positively correlated to luminal cross-sectional area (r = .23, P < .02) and negatively to distensibility (r = -.26, P < .01). These results indicate that geometric and functional changes in the common carotid artery accompany geometric changes in the left ventricle. More specifically, they suggest that a reduction in distensibility paralleled cardiac concentric hypertrophy and remodeling, whereas an increase in arterial volume paralleled increased left ventricular cavity size.

Which definition for echocardiographic left ventricular hypertrophy?

Left ventricular (LV) hypertrophy is diagnosed on the basis of LV mass measurement at echocardiography. However, various thresholds for defining LV hypertrophy have been published, ranging from 111 to 134 g/m2 and from 100 to 125 g/m2 in men and women, respectively. The aim of our study was to evaluate variations in the prevalence of LV hypertrophy induced by the application of different threshold values among hypertensive subjects. LV mass was calculated in 349 hypertensive patients from an M-mode LV tracing obtained by left parasternal view in 83% and by subcostal view in 17% of patients. The prevalence of LV hypertrophy ranged from 17% to 39%, according to the threshold value applied (from 10% to 47%, and from 19% to 39% in women and men, respectively). As expected, the prevalence of LV hypertrophy in obese patients of both sexes was higher when applying the usual height-indexed threshold (143 and 102 g/m for men and women, respectively) than when applying the usual body surface area-indexed threshold (134 and 110 g/m2 for men and women, respectively). The use of normalized thresholds when comparing different indexation methods (in this case, 145 g/m for men, 120 g/m for women) will minimize these variations in part due to the threshold choice. Considering the clinical and therapeutic implications associated with the presence of LV hypertrophy, better standardization of definitions is needed; this could be based either on better-designed cooperative normality studies or meta-analysis of risk stratification.

[Morphological study by transesophageal echocardiography and clinical aspects of ruptured chordae tendineae in the elderly]

Ruptured mitral chordae tendinae is a classical complication of myxomatous mitral valves or Barlow's syndrome. This complication is controversial in non-myxomatous mitral valve. Of 91 consecutive patients with mitral valve prolapse examined over an 18 months period by transthoracic and transesophageal echocardiography, 42 (18 women and 24 men) with an average age of 76 +/- 8 years (60-93 years) had ruptured mitral chordae tendinae. The thickness of the anterior mitral leaflet measured at the distal third of the valve by transesophageal echocardiography enabled the identification of two groups of patients; group I: > 3 mm (24 patients), average 4.8 +/- 0.8 mm and group II: < or = 3 mm (18 patients), average 2.6 +/- 0.3 mm. The diameter of the mitral ring and left atrium, the length of the anterior mitral leaflet, the left ventricular end diastolic dimensions and fractional shortening, were measured by transthoracic 2D echocardiography (mitral ring) and M mode (other parameters). Ruptured chordae were detected in only 13 cases (31%) by transthoracic echocardiography; 38% were asymptomatic and a chance finding at transesophageal echocardiography. No significant difference was observed between the two groups with respect to age, gender presence of hypertension, dimensions of the cardiac chambers, fractional shortening or localisation of the prolapse related to the ruptured chordae. Fifty-eight per cent of patients in group I were in NYHA functional classes 3-4 as compared to 16% in group II (p < 0.02). The size of the left atrium was significantly greater in group I, 51 +/- 8 mm vs 38 +/- 7 mm (p < 0.001).(ABSTRACT TRUNCATED AT 250 WORDS)

Relations between Doppler tracings of pulmonary regurgitation and invasive hemodynamics in acute right ventricular infarction complicating inferior wall left ventricular infarction

To test the hypothesis that flow characteristics from pulmonary regurgitation (PR) can predict right ventricular (RV) involvement in patients with inferior wall acute myocardial infarction, we prospectively recorded continuous-wave Doppler tracings and right-sided cardiac hemodynamics in 48 consecutive patients with inferior wall acute myocardial infarction and PR. Right heart hemodynamics enabled the identification of 29 patients with (group 1) and 19 without (group 2) RV involvement. In patients with RV involvement, the pulmonary regurgitant flow pattern was characterized by a rapid rise in flow velocity to a peak level followed by an abrupt deceleration in mid-diastole, whereas in patients without RV involvement, the deceleration in mid-diastole was gradual. The pressure half-time of PR (PHTPR) and the lowest mid-diastolic to peak early diastolic velocity ratio were significantly lower in group 1 than in group 2 (91 +/- 31 vs 214 +/- 57 ms [p < 0.001], 0.35 +/- 0.08 vs 0.59 +/- 0.13 [p < 0.001], respectively). The best diagnostic accuracy (95%) was obtained with cut-off values of PHTPR < or = 150 ms and the lowest mid-diastolic to peak early diastolic velocity ratio < or = 0.5: sensitivity 100%, specificity 89%, positive predictive value 94%, and negative predictive value 100%. Using multiple logistic regression analysis, we found that PHTPR was the strongest predictor of RV involvement. Thus, these parameters, derived from pulmonary regurgitant tracings, are useful in the noninvasive bedside diagnosis of RV infarction.

Accuracy and reproducibility of left ventricular mass measurement by subcostal M-mode echocardiography in hypertensive patients and professional bicyclists

In some patients, left ventricular (LV) mass cannot be evaluated by M-mode echocardiography because the parasternal long-axis view is not available. The aim of this study was to determine whether the subcostal view obtained by M-mode echocardiography under 2-dimensional guidance allows accurate and reproducible LV mass determination. Using the cube formula, LV mass was calculated, from parasternal and subcostal views in 96 subjects: 73 hypertensives and 23 professional bicyclists, covering a wide range of LV dimensions. M-mode tracings were read by 2 experienced echocardiographers and the interobserver variability was evaluated. With use of the subcostal view, the interobserver reproducibility, expressed as observer 1-observer 2, was excellent: -0.3 +/- 1.3 mm for LV diastolic diameter, -0.1 +/- 1.0 mm for ventricular diastolic septal thickness, 0.2 +/- 0.6 mm for diastolic free wall thickness and 0.03 +/- 16.7 g for LV mass. In 96% of cases, the difference in LV mass between the 2 observers did not exceed 30 g. With use of the parasternal and subcostal approaches, LV mass was not statistically different (202.6 +/- 2.2 g and 206.5 +/- 2.0 g, respectively) and the difference was < 52 g (clinically significant threshold) in 91 of 96 cases. It is concluded that, using the cube formula, LV mass calculation based on the subcostal view was accurate and not statistically different from that provided by the usual parasternal approach.

[Thrombosis of atrioventricular prosthesis. Contribution of transesophageal echocardiography]

The aim of this study was to assess the diagnostic value of transesophageal echocardiography for the detection of thrombosis of a mechanical mitral or tricuspid valve prosthesis. Twelve patients (mean age 54 +/- 12 years) out of a series of 39 patients operated between April 1988 and June 1989 for prosthetic valve dysfunction had valve thrombosis at operation (11 mitral and 1 tricuspid valve prosthesis). Transesophageal echocardiography was routinely performed preoperatively in addition to transthoracic Doppler echocardiography to search for an abdominal mass on the prosthetic valves. The largest diameter of the diastolic jet at the level of the prosthetic valve annulus was measured using transesophageal color flow Doppler in the 8 Starr-Edwards mitral valve prostheses and compared with 5 control valves. The results of transthoracic Doppler echocardiography and transesophageal echocardiography were compared with the operative findings. The specificity of transthoracic echocardiography for the positive diagnosis of prosthetic valve thrombosis was 18%. A thrombosis could be suspected in 10 of the 12 cases by transthoracic echocardiography giving a sensitivity of 83%. Eleven of the 12 abnormal masses on the prostheses were visualised by transesophageal echocardiography, a sensitivity of 91%. Detection of the masses on the arterial side was possible in all cases (10/10) but 5 of the 6 extensions of the thrombus into the ventricle could not be visualised. The diameter of the transprosthetic jet was less than 12 mm in 7 of the 8 thrombosed valves compared with greater than 15 mm in the 5 normal control prostheses.(ABSTRACT TRUNCATED AT 250 WORDS)

Influence of pulsatility on the development of intracardiac jets: an in vitro laser Doppler study

So far, it has been hypothesized that numerical data obtained in steady flow conditions apply to pulsatile flows. In order to study the modifications of the velocity fields due to pulsatility, jets were produced by 8 orifices (with a diameter "D" of 4.4 to 11.3 mm) included in a chamber of 50 mm. The velocity was measured using laser Doppler anemometry with a pulsatile flow ("pf") and compared to the values obtained in steady ("sf"): at maximum velocity, the longitudinal velocity profile is qualitatively similar to this observed in steady flow: it is made of a plateau followed by an hyperbolic velocity decay in the turbulent area. The length of the core ("Lpf") is strongly related to "D" (Lpf = 3.72 D + 5.49, r = .99) and the velocity decay depends on the ratio between the distance "x" from the orifice and "D" (V/Vo = 2.83D/x + 3.46, r = .85, where V is the velocity at "x" and Vo the initial velocity). During the acceleration and the deceleration, the laminar core is disturbed by turbulences. The comparison of "pf" data with "sf" data demonstrated similar diameters at the origin of the jets (Dpf = 0.96 Dsf + .12, r = .99), but significant (p less than .0001) differences both for "L" and "V/Vo": Lpf = .91Lsf + 6.58, r = .97, V/Vopf = .63 V/Vosf + .34, r = .76. Thus, pulsatility modifies velocity fields and the results obtained in steady flow conditions do not apply to pulsatile jets.

[Transesophageal echography in the investigation of left atrial ball thrombi. Apropos of a case of migration of a left atrial thrombus]

The authors report the case of a pediculated left atrial ball thrombus whose embolisation and ejection through the aortic valve was recorded during transesophageal echocardiography. The value and risks of transesophageal echocardiography for the diagnosis of left atrial masses are reviewed in the light of this case and a review of the literature. The prognosis and therapeutic implications of a left atrial ball thrombus are also described.

[Doppler echocardiography of the effect of atrioventricular delay on transmitral flow in dual chamber pacing. Role of associated cardiopathy]

Doppler echocardiography was used to analyse transmitral blood flow in 23 patients undergoing DDD pacing under basal conditions at a pacing rate of 70/min. Changes in the atrioventricular delay led to changes in Doppler parameters corresponding to the different phases of ventricular filing. When the atrioventricular interval was increased, the maximum velocity, the velocity time integral and the duration of the E wave decreased and the maximum velocity, the velocity time integral and duration of the A wave increased. The atrial contribution to left ventricular filing increased by 15 to 46% (p less than 0.001). The changes of the Doppler parameters with respect to the duration of the atrioventricular interval varied according to the patient group studied. Patients without ventricular dilatation with or without hypertrophy had greater maximum velocities and velocity time integrals of the A wave than patients with left ventricular dilatation. However, for the same changes in atrioventricular delay, the A wave and atrial contribution to left ventricular filing were more variable in patients without left ventricular dilatation than those with left ventricular dilatation confirming the greater sensitivity of patients without left ventricular dilatation to the setting of the atrioventricular interval. These results confirm the great variability of transmitral flow with changes in atrioventricular delay. They illustrate the need for appropriate programming of the atrioventricular delay especially in patients in whom the mitral flow is most sensitive to this adjustment.

[Quantitative analysis of systolic function of the right ventricule by Doppler echocardiography]

The recording of the velocity of tricuspid valve regurgitation by continuous wave Doppler enables the calculation of the instantaneous systolic pressure gradient between the right ventricle and right atrium. As right atrial pressure is relatively constant, the rate of acceleration of the regurgitant jet reflects the quality of the rise in pressure in the right ventricle in early diastole, and therefore right ventricular contractility. The authors studied 3 Doppler parameters of the rate of velocity increase of the tricuspid regurgitation; the maximum rate of acceleration (dV/dt max), the maximum derivative of the pressure (dP/dt max) and the mean rate of increase in pressure (T). The interobserver variability of these indices is low (r greater than 0.96); reproducibility is good in patients with sinus rhythm but mediocre in atrial fibrillation. The comparison of the Doppler indices with the right ventricular isotopic fraction in 26 patients with tricuspid regurgitation showed a good correlation (dV/dt max, r = 0.79, p less than 0.0001; dP/dt max, r = 0.69, p less than 0.0001; T, r = 0.60, p = 0.0012). These results show that right ventricular systolic function can be evaluated by continuous wave cardiac Doppler by recording the spectral envelope of tricuspid regurgitation.

[The value of recording the pulmonary insufficiency flow by continuous Doppler for the evaluation of systolic pulmonary artery pressure]

Systolic, diastolic and mean pulmonary artery pressures can be evaluated by Doppler recordings of the maximal velocity of tricuspid regurgitation and early and late diastolic pulmonary regurgitant flow. The aim of this study was to assess the reliability of the calculation of systolic pulmonary artery pressure from pulmonary regurgitant flow by comparing the values with those obtained from the tricuspid regurgitant flow in the same patient. With this objective in mind, we investigated 70 patients with an average age of 45 +/- 34 years, in sinus rhythm, all of whom had tricuspid and pulmonary regurgitant jets which could be recorded with continuous wave Doppler. Systolic pulmonary artery pressure was calculated as follows: from tricuspid regurgitation: maximum pressure gradient + 10 mmHg; from pulmonary regurgitation: 3 x early diastolic gradient - 2 x late diastolic gradient + 10 mmHg. The systolic pulmonary artery pressures calculated from tricuspid and pulmonary regurgitation were: 42 +/- 16 mmHg and 43 +/- 17 mmHg respectively (r = 0.97) with an estimated standard error of 4.7 mmHg. These results show that the recording of pulmonary regurgitation by continuous wave Doppler allows accurate estimation of pulmonary artery pressures. The calculation by the two methods using tricuspid and pulmonary regurgitant jets increases the reliability of the results and provides a means of internal validation of the Doppler technique.

[Study of valvular prostheses by Doppler echography]

Doppler echocardiography is a reliable non-invasive method of following up patients with prosthetic heart valves. Transthoracic M mode and two-dimensional echo allow evaluation of the movement of the mobile element of the prosthesis and of the size of the cardiac chambers. Doppler studies (transprosthetic pressure gradient, valve surface area, trans- and periprosthetic regurgitation) give an indication of the hemodynamic profile of the prosthesis. Transesophageal echocardiography is especially useful when there are abnormalities suggestive of prosthetic valve dysfunction (fever, systemic embolism, clinical symptoms, haemolysis...).

[Localized dilatation of the bronchi. A misunderstood etiology of coronaro-bronchial fistula]

The authors report the case of a coronary to bronchial artery anastomosis secondary to focal bronchiectasis. The diagnosis was made after finding large retroatrial vessels on coronary arteriography. A pulmonary steal syndrome, frequently reported in this condition in the literature, was not present in that particular case. The diagnosis of a coronary to bronchial artery anastomosis should alert the physician to possible underlying cardiac disease (Tetralogy of Fallot, supravalvular aortic stenosis, severe coronary artery disease). A bronchopulmonary etiology (chronic obstructive airways disease, multiple bullae, bronchiectasis) is less commonly found as the presentation is often atypical.

Non-invasive study of dual chamber pacing by pulsed Doppler. Prediction of the haemodynamic response by echocardiographic measurements

The increase in stroke volume with DDD compared with VVI pacing was measured at rest using pulsed Doppler echocardiography in 23 patients at a pacing rate of 70 beats min-1. Stroke volume was assessed by measuring the velocity integral of the flow at the aortic annulus using the apical five-chamber window. Pulsed Doppler echocardiography allowed determination of the least and most favourable AV delay haemodynamically. TVI was also measured at each nominal value of AV delay. The percentage increase in stroke volume was determined in every patient changing from VVI to optimum DDD pacing and was used as a measurement of the 'sensitivity' to optimum DDD pacing; the mean increase was 27 +/- 19%. The increase in stroke volume accompanying the change from DDD pacing with the least favourable to the optimum AV delay was also measured, and used as a measurement of 'sensitivity' to changes in AV delay; the mean increase was 23.7 +/- 16.3%. Clinical and standard echocardiographic parameters were studied in order to determine which variable might best identify the patients more likely to benefit from DDD pacing, and to identify those more sensitive to the AV delay setting. With respect to sensitivity to DDD pacing, three echocardiographic variables were selected by linear discriminant analysis from 11 clinical and echocardiographic variables. These were, in order of importance, left ventricular systolic diameter (LVSD), left ventricular wall thickness (LVWT) and left atrial size (LAS) which allowed the prediction of a good or a bad response to optimal DDD pacing with an accuracy of 91.3%.(ABSTRACT TRUNCATED AT 250 WORDS)