Regular Review

Echocardiography BMJ: first published as 10.1136/bmj.297.6656.1071 on 29 October 1988. Downloaded from

Information on morphology andfunction

Ultrasonography has revolutionised diagnostic DOPPLER ULTRASONOGRAPHY and is capable of supplanting cardiac catheterisation for most Information about velocity is derived from the shift in indications apart from coronary . Sector scanning frequency that occurs between transmitted and reflected and M mode produce anatomical images ultrasound. The principle is the same as for the everyday whereas the various Doppler techniques assess abnormalities observation that the sound of a car horn rises in pitch as it of blood flow. approaches and falls as it recedes. Ifred blood cells are moving towards the transducer at the moment when they reflect Technical ultrasound they will shorten its wavelength, and if they are background moving away they will lengthen it (fig 2). The shift in Clinical echocardiography evolved from military origins frequency is in the audible range, and the audiosignal is used about 40 years ago.'-' It was not used routinely until the 1970s, to guide the transducer to obtain the best visual display. but since then there have been important improvements in the quality of the images produced. A rekindling of interest in Doppler ultrasonography because of advances in spectral analysis and clinical interpretation4 led to colour flow mapping being developed in the early 1980s.5 00 00

IMAGING The principle of ultrasonography is similar to that of sonar devices employed by bats and submarines. Ultrasound is high frequency sound above the audible range of 50-20 000 Hz, and Transducer A Transducer B frequencies of 1-9-7-5 MHz are used for examining the . There is a trade offbetween improved penetration at the lower end and resolution at the end of this A FIG 2-The Doppler principle. Ultrasound is transmitted at a constantfrequency. A red higher upper range. cell moving directly towards transducer A reflects a wave of ultrasound and shortens its piezoelectric crystal is made to vibrate by the passage of an wavelength. The red cell is moving directly awayfrom transducerB so the wavelength of electric current, which generates a pulse of ultrasound that the reflected ultrasound is lengthened. The change in wavelength depends on the velocity ofthe red cell and on its direction relative to the ultrasound beam

travels through the body at 1540 m/s. Most ofthe transmitted http://www.bmj.com/ sound energy is scattered or absorbed, but a little is reflected at interfaces between tissues ofdifferent acoustic impedance- Computerised analysis of the returning ultrasound signal for example, between the pericardium and the myocardium. produces a plot ofvelocity against time in which the density of a point on the display is in proportion to the number of red ---.--- cells moving at that velocity...... Figure 3 (top) overleaf shows normal flow across the mitral -., ;------1 valve, which is laminar and therefore has a narrow range of .... blood velocities. As mitral develops peak velocity on 27 September 2021 by guest. Protected copyright. rises, the rate of decay of peak velocity falls, and because flow becomes turbulent a greater range of velocities appear (fig 3 centre and bottom). These changes may be used to measure the degree ofstenosis. The drop in pressure at any time during diastole (and across any other valve or shunt) may be calculated using the modified Bernouilli equation4: pressure drop=4 x (velocity)2. This equation tends to overestimate the pressure drop in aortic FIG 1-Formation ofechocardiographic images. The sector image on the right is built up stenosis at values below about 50 mm Hg and underestimate it from reflected.. ,pulses , ofultrasound transmitted over 120 scan lines in a 900 arc 30 times at values above.6 The rate ofdecay ofpeak transmitral velocity every second. The M mode image on the left is made by looking at one single scan line is expressed as the pressure half time (T½/2). This measure- ment was originally used during cardiac catheterisation, and it Returning reflected pulses are detected by a receiving is the time taken for peak pressure across the mitral valve to crystal, and the magnitude of the received signal controls the halve.' In adapting it for ultrasonography, velocity is related density of a spot placed on the display screen. The position of to the square root of the pressure drop so the pressure half the spot is determined by the time difference between the time becomes the time taken for peak velocity to fall by the transmission and the return of the pulse. To form a typical square root ofhalf(in other words, to 0 7 x the peak velocity). sector (two dimensional) scan pulses are transmitted along Pressure halftime may then be used to estimate the area ofthe about 120 scan lines over a 90° arc 30 times a second. An M orifice in native mitral stenosis (provided that the area is less mode image is constructed by transmitting and recording at than 1-8 cm2) with the empirical formula8: estimated area of only one of these scan lines and displaying the returning the orifice=220/pressure half time. echoes as a graph of depth against time (fig 1). It is important to remember that the pressure gradient

BMJ VOLUME 297 29 OCTOBER 1988 1071 ige*seg. 6 .6.. *@S#*.t9 S.##4 .6 acrossvalves is highlydependentonflow, andlargedifferences may occur from day to day, or even from beat to beat. This is I I particularly a problem in aortic stenosis, and one way to correct for flow is to calculate the estimated area ofthe orifice with the continuity equation9: area of the orifice=subaortic area x subaortic velocity/aortic velocity. BMJ: first published as 10.1136/bmj.297.6656.1071 on 29 October 1988. Downloaded from There are two basic Doppler techniques-continuous wave and pulsed Doppler. A continuous wave probe has one crystal transmitting and a second' crystal receiving continuously. This system is extremely sensitive and allows the measure- ment of high maximal velocities. An important'drawback, however, is that all velocities along the length and width ofthe ultrasound beam are recorded so that regions of abnormal flow cannot be localised accurately. This problem is circum- vented by using pulsed Doppler. With pulsed Doppler a single crystal transmits ultrasound in pulses and is then set to receive- echoes after a time 'delay. In this way reflected signals are recorded only from a particular depth, and localisation within an echocardiographic image is then possible. But pulsed Doppler also has drawbacks. If the frequency shift is high the gating principle means that important

L sampling errors will occur,: and therefore above a certain -64 '"i, :a:* maximum blood velocity measurements are ambiguous. Furthermore, regions of abnormal 'flow' may be missed altogether ifthe transducer'is not aimed in the right direction. This problem is greatly reduced by colour flow Doppler, a two dimensional version ofpulsed Doppler.

COLOUR FLOW DOPPLER The colour flow system calculates mean blood velocity and direction of flow at multiple points down each scan line of an image sector. The velocity information is then superimposed on to the echocardiographic image with colour coding. By convention velocities in a direction towards;the transducer are displayed in red and those" away in blue. Increasing velocities are shown initially in progressively lighter shades or changes of hue and later as a change in colour from blue to red or vice versa.' This ambiguity in. the velocity display is called http://www.bmj.com/ "aliasing," and it aids the visual detection of abnormal flow. Some systems substitute green in place ofred or blue in areas of high variance (areas containing a wide range of velocities), which highlights turbulent flow. Thus colour flow Doppler gives a dynamic representation ofinformation on the velocity of blood throughout the cardiac cycle. The display resembles

superficially'the image produced by cineangiog'raphy, which, on 27 September 2021 by guest. Protected copyright. however, gives information on the volumetric movement of blood rather than on spatial velocity and is not strictly comparable. Observers unf miliar with Doppler are more likely to understand colour flow Doppler than conventional Doppler displays, which helps reduce the need for cardiac catheter- isation. Care must be taken to avoid overinterpretation, and some lesions detected may be striking visually but of little clinical importance. The sensitivity of colour flow Doppler is comparable with pulsed Doppler, and occasionally low intensity regurgitant jets are shown by continuous wave ultrasonography but not by colour flow Doppler. In practice these are rarely ofhaemodynamic importance. FIG 3-Left ventricular inflow recorded with Doppler ulkrasonography. The top figure shows the normal appearance recorded usingpulsed Doppler with the sample wlume at the tips ofthe mitral leaflets at their maximum diastolic excursion. The patient is in sinus rhythm. The early peak (e) represents passive ventricularfilling and the secondpeak (a) atrial systole. The centre figure shows the appearance in mild mitral stenosis when the maximum inflow velociy is increased. Thepatientisinatrialfibrillation so there is-onlya single peak and the rate ofdecline ofvelocity across the mitral valve is reduced. There is mild turbulence shown byan increase in the width ofthe outline ofthe signal. The bottom figure shows the appearance in severe mitral stenosis recorded by continuous wave Doppler. The pressure halftime .(TI½2) is the time between peak left ventricular inflow velocity and 0 7 x the peak left ventricular inflow velocity. In this case it is 280 msec, giving an estimated area ofthe orifice ofabout 0 75 cm2

1072 BMJ VOLUME 297 29 OCTOBER 1988 The ultrasound exa The ribs ai2d lungs mask the heart from ultrasound except at small "window0s" that may be enlarged by asking the subject to lie s c on the 1eft side. Usually sector

and colour flow scans are performed using parasternal and BMJ: first published as 10.1136/bmj.297.6656.1071 on 29 October 1988. Downloaded from apical views as a rapid screen for morphological and haemo- dynaic abnormalities. -The subcostal viewisu in patients with chronic airways obstruction or for the routine imagigoftheinteratrial septum. Mmoderecordings through the aortic valve, mitral valve, and the left ventricle just belo* tips of the mitral valve are taken at "'this stage before performing an. examiation with continuous wave Doppler. Recordings ofmi3tal and aortic flow ar.e mad.e from the apical approach. The aortic valve may';be viewed from the right intercostal spaces, the pulinonary valve from the left para- -sternal position, and the aorta from the suprasternal notch. An average fufll emination takes about halfan hour. Normal ranges for chamber dimensions. using M mode'0 and two dimensional echocardiography", and nojmal flow veloci,ties using Doppler.ultrasonography'2 ar;e available.

Clinical applications ofechocardiography VALVULAR LESIONS A doctor assessing a ,valvular lesion must know not only the severity ofthe hiemodynamic derangement but also its effect on the heart. Severe aortic regurgitation, for example, may not require an operation until early decompensation begins and is shown by systolic dilatation ofthe left ventricle. Sector scanning detects morphological abnormalities, often allowing diagnosis ofthe underlying causes. Doppler ultrasonography is then used to assess the haemodynamic derangement, and' a combination ofsector scanning and M mode ultrasonography is used to assess cardiac function in response to that derange- ment (fig 4). Continuous wave recordings ofmitral inflow from-the apex are easy to obtain, and they give estimates of the drop in pressure and the area of the orifice' that-correlate well with those collected at simultaneous catheterisation."3 In patients http://www.bmj.com/ with aortic stenosis it may be easy to obtain a spectral display, but great care is needed to ensure that the highest velocities are not missed; recordings 'must be made using multiple approaches.'4 Even a small underestimate ofvelocity may lead to serious errors in es"mating the drop in pressure because of the square factor in the modified Bernouilli equation.

It is important to remember that Doppler is used to on 27 September 2021 by guest. Protected copyright.

FIG 4-Assessmene ofartic valve dysfunction by ultrasonography. The left handfigure shows the appearance ofthe aortic valve in a sector scan. The aortic cusps are heavily calcified. The top figure shows the appearance by continuous wave Dopplerfrom the apical positon. There is diastolic,flow representing mild aortic regurgitation (directed towards the transducer therefore displayedabove the baseline), but the dominant lesion is aortic stenosis. Peak velocity ofsystolicflov is 4-5 m/s, which gives an estimated peak instantaneous pressure drop ofoverSO mm Hg(4 x 4 52). Colourflow mapping(centre) firom theparasternalpositin shows a regurgitantjet that has a small area comparedwith theleft ventriculardiastolic area, confirming mild aortic regurgitation. The bottomfigure shows the M mode recording through the left ventricle. The left venriular septum and. posterior wall are wideed because of left ventricular hyop hy induced by severe obstruction ofthe left ventricularoutflow. Iv=Leftventricularcavity;pw=posterior left ventricular wall; rv=right ventricular cavity; s=intervenrular septum

BMJ VOLUME 297 29 OCTOBER 1988 1073 BMJ: first published as 10.1136/bmj.297.6656.1071 on 29 October 1988. Downloaded from

FIG 5-Small muscular ventriculat septal defect. The left handfige shows the two dimensional echocardiogram in the parasternal long axis view. No morphological abnormality is visible. The right handfigure shovs the appearance after tning on colourflow mapping. Abnormalflow is clearly seen in the righ ventricle with its origin in the ventricular septum estimate the peak instantaneous gradient, which is higher ventricular aneurysm33 and thrombus.34 The origin of the than the peak to peak gradient measured at catheterisation.'5 coronary may sometimes be visualised, but this rarely Other problems are dissociating mitral regurgitation from gives useful information. Pulsed Doppler recordings of left aortic stenosis, tricuspid regurgitation from mitral regurgita- ventricular inflow may give an indication of left ventricular tion, and aortic regurgitation from mitral stenosis as these compliance and ofraised left atrial pressure.35 flow patterns may be spatially close and have similar (although M mode echocardiography is used in patients with cardio- not identical) timing. The technique thus requires skill and myopathies to measure intracardiac dimensions, and sector may be time consuming. It is simplified by simultaneous scanning ultrasonography is used to detect abnormalities imaging or better still by using colour flow. of morphology, motion, and acoustic density.-363 This is Colour flow Doppler guides the placing of the probe and particularly useful in regions inaccessible to M mode echo- rapidly identifies regurgitant. jets. Quantifying the regurgi- cardiography-for example, the left ventricular apex and tation is less certain. The easiest way is with colour flow right heart.39 The degree of acceleration of flow in hyper- Doppler, in which the dimensions ofthe jet (notably the area trophic obstructive cardiomyopathy may be shown4" and and width) correlate well with estimates from injections of the response to treatment followed" using Doppler ultra- dye.'16-1 Continuous wave Doppler may give an estimate of sonography. The physiological features of restrictive severity from the intensity and shape of the jet"9; pulsed cardiomyopathy nlay be shown by using pulsed Doppler.4' Doppler may give an estimate by assessing how far back in the receiving chamber the jet may be detected20 or the degree of PROSTHETIC VALVES reverse flow in the subclavian or aorta.'9 In practice M mode and sector echocardiography have been used to regurgitation is usually assessed using both colour flow and describe prosthetic function by analysing movement of the http://www.bmj.com/ continuous wave Doppler and by determining the chamber occluder or cusps,4243 by timing events in the cardiac cycle,'" dimensions with imaging echocardiography.2' and by measuring the speed ofleft ventricular filling45 or ofleft atrial emptying."6 All these methods lack sensitivity and CONGENITAL HEART DISEASE specificity,47 cannot reliably localise the site of regurgitation, Many congenital lesions ofthe heart may be diagnosed from and are time consuming. Doppler ultrasonography is better. distinctive morphological appearances on sector scanning,226 Conventional Doppler allows measurement of the pressure but colour flow Doppler may quickly show abnormalities of drop across the valve," but interpreting this information is on 27 September 2021 by guest. Protected copyright. flow even without echocardiographic abnormalities (fig 5).27 difficult because normal ranges have not been established The flow map then guides the positioning of the continuous with certainty, velocities are dependent on flow and the wave Doppler probe for quantitativehaemodynamic measure- diameter of the annulus, and there is a large individual ments. Comparison of volumetric flow on the right and left variationamongprostheses. Continuouswave Dopplerdetects sides of the heart allows estimation of the size of shunts.28 regurgitant jets easily,49 but conventional pulsed Doppler is Atrial and ventricular septal defects are shown by colour flow not always able to localise them.49-50 The main influence of Doppler with high sensitivity relative to catheterisation,19 colour flow Doppler has been to improve localisation and to and the estimation ofthe amount ofblood being shunted from allow rapid detection ofsmall regurgitant jets.5' the area of abnormal flow correlates well with the estimation from catheterisation.20 Complex abnormalities in flow may PERICARDIAL DISEASE be difficult to decipher with colour flow Doppler, and Sector scanning is quicker -and easier than M mode differentiation of a muscular from a membranous ventricular echocardiography for detecting, localising, and measuring septal defect may still be a problem. Colour flow Doppler pericardial effusions (fig 6).52 The differentiation of con- makes contrast echocardiography for detecting shunts strictive pericarditis from restrictive cardiomyopathy may be virtually redundant. made by characteristic features seen on two dimensional echocardiography53 but is aided by the pattern of atrial MYOCARDIAL FUNCTION AND DISEASE filling on pulsed Doppler (L Hatle, seventh symposium Doppler ultrasonography, especially colour flow mapping, on echocardiology, Rotterdam, 1987). may differentiate between mitral regurgitation and a ven- tricular septal defect, but imaging echocardiography is used MISCELLANEOUS in diagnosing and assessing motion abnormalities of the Sector scanning may detect abnormal structures as small as regional wall caused by ischaemic heart disease3 32 and 3 mm in diameter, although its reliability is much higher for other complications of ischaemic heart disease such as left diameters above 5 mm.54 It may therefore image intracardiac

1074 BMJ VOLUME 297 29 OCTOBER 1988 tumours, thrombi, and vegetations. Vegetations may, how- aortic dissection63 as entry and exit sites may be visualised ever, be difficult to differentiate from redundant cusp tissue quickly and easily. or calcification. Sector scanning reliably shows dilatation of Myocardial perfusion imaging necessitates injecting micro- the ascending aorta55 and may show the site of an intimal flap bubbles into the coronary arteries or peripheral ."65 in aortic dissection.56 Transthoracic colour flow Doppler may The distribution, the initial intensity, and the rate ofdecay of detect tears related to a dissection57 but is far less useful than enhancement on serial echocardiographic frames is a guide to BMJ: first published as 10.1136/bmj.297.6656.1071 on 29 October 1988. Downloaded from transoesophageal echocardiography. the presence of and the haemodynamic importance of a An important application of Doppler ultrasonography is in coronary stenosis. The technique may be used to monitor estimating . Correlation coefficients of up to angioplasty but is still experimental. 0 97 with estimates from temperature dilution, may be obtained.5859 This allows non-invasive monitoring of changes that occur on exercise or after therapeutic manoeuvres in Limitations ofechocardiography intensive care units. The power ofultrasound techniques is highly dependent on Systolic pressure in the pulmonary artery may be estimated the operator. In about one in five patients echocardiographic by continuous wave Doppler. From the modified Bernouilli images are of reduced quality and in about one in 10 they are equation the maximum velocity of a jet of tricuspid re- poor. The limitation on the sensitivity of the technique gurgitation gives the pressure drop between ventricle and depends on the question being asked. It is possible to detect atrium. When there is no pulmonary stenosis this must equal an effusion or mitral stenosis in virtually all patients, but the pressure change between atrium and pulmonary artery. It extremely good pictures are needed to diagnose vegetations. is not necessary to add a clinical estimate of right atrial Even in good echocardiographic subjects it may not be pressure to obtain close correlation with results gathered at possible to image the root of the pulmonary artery well. Care catheterisation.60 is particularly needed with Doppler ultrasonography, in which an incorrect estimation of pressure drop is all the more misleading for its spurious precision. Future trends Echocardiography, has general technical problems. Arte- Endoscopic transoesophageal echocardiography is only facts may cause overinterpretation, especially if inappro- now beginning to be introduced into Britain, although it has priately high gain settings are used. Although the spatial been used on the continent since 1979. It has the major resolution of sector imaging is good, it is still possible to advantages of an excellent optical window and views that are missabnormalities lateralto thefield ofview. Tissuecharacter- complementary to those obtained with transthoracic echo- isation by echocardiography is still fairly poor -for example, cardiography. Virtually the whole ofthe thoracic aorta may be detecting invasion by intracardiac tumours may be difficult. imaged, and the atria, the interatrial septum, and the There are numerous sources of variability in an echocardio- proximal left coronary artery are particularly well seen. Image graphic scan, which makes the reliable detection of serial resolution is improved so that minor abnormalities may be changes difficult, particularly when the dimension is small as, detected, and vegetations and small thrombi are found more for example, with septal thickness."67 often than with transthoracic echocardiography.55 61' The technique may be used for monitoring changes in left ventricular function in intensive care units or operating Summary

theatres62 or for assessing the surgical repair of mitral Imaging echocardiography is an important extension ofthe http://www.bmj.com/ regurgitation or intracardiac shunts. Transoesophageal echo- clinical examination and will answer most questions in an cardiography is likely to become the investigation ofchoice in emergency-forexample, whetheranenlargedcardiac shadow on the represents ventricular dilatation or an effusion. Doppler ultrasonography is essential for hospitals with an interest in cardiology because it provides direct haemodynamic data that are complementary to imaging. It requires more skill than imaging and may also be time consuming. Colour flow Doppler mapping is speedy and on 27 September 2021 by guest. Protected copyright. simple to use and aids the interpretation of continuous wave Doppler. It is therefore a natural companion to conventional Doppler, but there would have to be a high clinical load to justify its purchase. J B CHAMBERS Research Fellow M J MONAGHAN Principal Clinical Scientist G JACKSON Consultant Cardiac Unit, King's College Hospital, London SE5 9RS JBC holds a British Heart Foundation fellowship. Mr K Packer donated the colour flow system.

1 Keidel WD. Ober eine neue Methode zur Registrierung der Volumanderungen des Herzens am Menschen. ZeitschriftfiirKreislaufforsckung 1950;39:257-71. 2 Edler 1, Hertz CH. The use ofultrasonic reflectoscope for the continuous recording ofmovements ofheart walls. Kungliga Fysiografiska Sdllskapets iLundForhandlinger 1954;24:40-58. 3 Satomura S, Zeneko Z. A study on examining the heart with ultrasonics..Jpn Circj 1956;20:227. FIG 6-Moderate sized pericardial effusion. This sector scan is recorded from the 4 Holen J, Aaslid R, Landmark K, Simonsen S. Determination ofpressure gradient in mitral stenosis subcostal approach. A=Left atrium; b=left ventricle; c=right atrium; d=right with a non-invasive ultrasound Doppler technique. ActaMedScand 1976;199:455-60. ventricle; e=effusion 5 Omoto R, Yokote Y, Takamoto S, et al. The development of real-time two-dimensional Doppler

BMJ VOLUME 297 29 OCTOBER 1988 1075 echocardiography and its clinical significance in acquired valvular diseases with special reference assessment of left ventricular filling and septal and posterior wall dynamics in idiopathic to the evaluation of valvular regurgitation. J7pn Heartj 1984;25:325-40. hypertrophic subaortic stenosis. Circulation 1978;57:512-20. 6 Rijsterborgh H, Roelendt J. Doppler assessment of aortic stenosis: Bernouilli revisited. Ultrasound 38 Borer JS, Henry WJ, Epstein SE. Echocardiographic observations in patients with systermic MedBiol 1987;13:241-8. infiltrative disease involving the heart. AmJ Cardiwl 1977;39:184-8. 7 Libanoff AJ, Rodbard S. Atrioventricular half time measure of mitral valve orifice area. Circulation 39 Maron BJ, Gottdiener JS, Bonow RO, Epstein SE. Hypertrophic cardiomyopathy with unusual 1%8;38: 144-50. locations of left ventricular hypertrophy undetectable by M-mode echocardiography: Identifica- 8 Hatle L, Angelsen B, Tromsdal A. Noninvasive assessment of atrioventricular half-time by tion by wide-angle two-dimensional echocardiography. Circulation 1981;63:409-18.

Doppler ultrasound. Circulation 1979;60:1096-104. 40 Maron BJ, Gottdiener JS, Arce J, Rosing DR, Wesley YE, Epstein SE. Dynamic subaortic BMJ: first published as 10.1136/bmj.297.6656.1071 on 29 October 1988. Downloaded from 9 Otto CM, Pearlman AS, Comess KA, Reamer RP, Janko CL, Huntsman LL. Determination of the obstruction in hypertrophic cardiomyopathy: analysis by pulsed Doppler echocardiographv. stenotic aortic valve area in adults using . J7 Am Coll Cardiol J Am Coll Cardiol 1985;6:1-15. 1986;7:509-17. 41 Appleton CP, Hatle LK, Popp RL. Demonstration of restrictive ventricular physiology by Doppler 10 Feigenbaum H. Echocardiograpts. Philadelphia: Lea and Febiger, 1986. echocardiography. 7Am Coll Cardiol 1988;11:757-68. 11 Schnittger I, Gordon EP, Fitzgerald PJ, Popp RL. Standardised intracardiac measurements of two- 42 Kotler MN, Mintz GS, Panidis I, Morganroth J, Segal BL, Ross J. Noninvasive evaluation of dimensional echocardiography. J Am Coll Cardiol 1983;2:934-8. normal and abnormal prosthetic valve function. _Am Coll Cardiol 1983;2:151-73. 12 Wilson N, Goldberg SJ, Dickinson DF, Scott 0. Normal mtracardiac and great artery blood 43 Forman MB, Phelan BK, Robertson RM, Virmani R. Correlation of two-dimensional echocardi- velocity measurements by pulsed Doppler echocardiography. Br Heartj 1985;53:451-8. ography and pathologic findings in porcine valve dysfunction.J Am Coll Cardiol 1985 ;5:224-30. 13 Hatle L, Brubakk A, Tromsdal A, Angelsen B. Non-invasive assessment of pressure drop on mitral 44 Willerson JT, Kastor JA, Dinsmore RE, et al. Non-invasive assessment of prosthetic mitral stenosis by Doppler ultrasound. Br Heartj 1978;40:131-40. paravalvular and intravalvular regurgitation. BrHeart3' 1972;34:561-8. 14 Hatle L. Noninvasive assessment and differentiation of left ventricular outflow obstruction by 45 Dawkins KD, Cotter L, Gibson DG. Assessment of mitral Bjork-Shilev prosthetic dysfunction Doppler ultrasound. Circulation 1981;64:381-7. using digitised M mode echocardiography. Br HeartJ 1984;51:168-74. 15 Krafchek J, Robertson JH, Radford M, Adams D, Kisslo J. A reconsideration of Doppler assessed 46 Skorton DJ, Child JS, Eber LM. Noninvasive diagnosis of prosthetic mitral valve malfunction. gradients in suspected aortic stenosis. Am Heartj 1985;110:765-72. 1979;131:185-6. 16 Miyatake T, Izumi S, Okamoto M, et al. Semiquantitative grading of severity of mitral 47 Mintz GS, Carlson EB, Kotler MN. Comparison of noninvasive techniques in evaluation of the regurgitation by real-time two-dimensional Doppler flow imaging technique. J Am Coll Cardiol nontissue cardiac valve prosthesis. AmJ Cordial 1982;49:39-44. 1986;7:82-8. 48 Weinstein IR, Marbarger JP, Perez JE. Ultrasonic assessment of the St Jude prosthetic valve: M 17 Helmcke F, Nanda NC, Hsuing MC, Soto B, Adey CK, Goyal RG, Gatewood RP. Color Doppler mode, two-dimensional, and Doppler echocardiography. Circulation 1983;68:897-905. assessment of mitral regurgitation with orthogonal planes. Circulation 1987;75:175-83. 49 Veyrat C, Witchitz S, Lessana A, Ameur A, Abitbol G, Kalmanson D. Valvar prosthetic 18 Sahn DJ, Chung KJ, Tamura T, et al. Factors affecting 'et visualisation by color flow mapping dysfunction. Localisation and evaluation of the dysfunction using the Doppler technique. Doppler echo: in vitro studies. Circulation 1986;74(suppl II):271. BrHeartj 1985;54:273-84. 19 Hatle L, Angelsen B. Pulsed and continuous wave Doppler in diagnosis and assessment of various 50 Sagar KB, Wann S, Paulsen WHJ, Romhilt DW. Doppler echocardiographic evaluation of heart lesions. In: Doppler ultrasound in cardiology. Philadelphia: Lea and Febiger, 1985:97-292. Hancock and Bjork-Shiley prosthetic valves. J Am Coll Cardiol 1986;7:681-7. 20 Ciobanu M, Abbasi AS, Allen M, Hermer A, Spellberg R. Pulsed Doppler echocardiography in the 51 Chambers JB, Jackson G. Initial experience of colour flow Doppler mapping in the assessment of diagnosis and estimation of severity of aortic insufficiency. Amj Cardiol 1982;49:339-43. prosthetic valves. Br Heartj 1988;59:90. 21 Gehl LG, Mintz GS, Kotler MN, Segal BL. Left atrial volume overload in niitral regurgitation: a 52 Martin RP, Rakowski H, French J, Popp RL. Localisation of pericardial effusion with wide angle two dimensional echocardiographic study. Amj Cardiol 1982;49:33-8. phased array echocardiography. AmJ Cardial 1978;42:904-12. 22 Henry WL, Maron BJ, Griffith JM. Cross-sectional echocardiography in the diagnosisofcongenital 53 Voelkel AG, Pietro DA, Folland ED, Fisher ML, Parisi AF. Echocardiographic features of heart disease. Circulation 1977;56:267-73. constrictive pericarditis. Circulation 1978;58:871-5. 23 Lieppe W, Scallion R, Behar VS, Kisslo JA. Two-dimensional echocardiographic findings in atrial 54 Erbel R, Rohmann S, Drexler M, et al. Improved diagnostic value of echocardiography in patients septal defect. Circulation 1977;56:447-56. with infective endocarditis by transoesophageal approach. A prospective study. Eur Heart J 24 Silverman NH, Schiller NB. Apex echocardiography. A two-dimensional technique for evaluating 1988;9:43-53. congenital heart disease. Circulation 1978;57:503-11. 55 DeMaria AN, Bommer W, Neumann A, Weinert L, Bogren H, Mason DT. Identification and 25 Sahn DJ, Allen HD, McDonald G, Goldberg SJ. Real-time cross-sectional echocardiographic localisation of of the ascending aorta by cross-sectional echocardiography. Circulation diagnosis of coarctation of the aorta. Circulation 1977;56:762-9. 1979;59:755-61. 26 Weyman AE, Hurwitz RA, Girod DA, Dillon JC, Feigenbaum H, Green D. Cross-sectional 56 Nanda NC, Gramiak R, Shah PM. Diagnosis of aortic root dissection by echocardiography. echocardiographic visualization of the stenotic pulmonary valve. Circulation 1977;56:769-74. Circulation 1975;48:11-7. 27 Omoto R. Color atlas of real-time two-dimensional Doppler echocardiography. Tokyo: Shindan-To- 57 Iliceto S, Nanda NC, Rizzon P, et al. Color Doppler evaluation of aortic dissection. Circulation Chiryo, 1987. 1987;75:748-55. 28 Sanders SP, Yeager S, Williams RG. Measurement of systemic and pulmonary blood flow and QP/ 58 Sahn DJ. Determination of cardiac output by echocardiographic Doppler methods: Relative QS ratio using Doppler and two-dimensional echocardiography. AmJ Cardiol 1983;51:952-6. accuracy of various sites for measurement. J Am Coll Cardial 1985;6:663-4. 29 Kyo S, Omoto R, Takamoto S, Takanawa E. Quantitative estimation of intracardiac shunt flow in 59 Chandaratna PA, Nanna M, McKay C, et al. Determination of cardiac output by transcutaneous atrial septal defect by real-time two-dimensional color flow Doppler. Circulation 1984;70:(suppl continuous-wave ultrasonic Doppler computer. AmJ7 Cardial 1984;53:234-7. II):3. 60 Berger M, Haimowitz, van Tosh A, Berdoff RL, Goldberg E. Quantitative assessment of 30 Serwer GA, Armstrong BE, Anderson PAW. Continuous wave Doppler ultrasonographic pulmonary hypertension in patients with tricuspid regurgitation using continuous wave Doppler quantitation of flow. J Pediatr 1982;100:297-9. ultrasound. JAm Coll Cardial 1985;6:359-65. 31 Kerber RE, Marcus ML, Erhart J, Wilson R, Abboud FM. Correlation between echocardio- 61 Roelendt JRTC, Sutherland GR. Oesophageal echocardiography. Br Heart3 1988;60:1-3. graphically demonstrated segmental dyskinesis and regional myocardial perfusion. Circulation 62 Topol EJ, Weiss JL, Guzman PA, et al. Immediate improvement of dysfunctional myocardial 1975;52: 1097-104. segments after coronary revascularisation: detection by intraoperative transoesophageal echo- 32 Weiss JL, Bulkley BH, Hutchins JM, Mason SJ. Two-dimensional echocardiographic recognition cardiography. J Am Coll Cardial 1984;4:1123-34. of myocardial injury in man: comparison with postmortem studies. Circulation 1981;63:401-8. 63 Erbel R, Borner N, Steller D, et al. Detection of aortic dissection by transoesophageal 33 Weyman AE, Peskoe SM, Williams ES, Dillon JC, Feigenbaum H. Detection of left ventricular echocardiography. Br HeartJ 1987;58:45-51. aneurysms by cross-sectional echocardiography. Circulation 1976;54:936-44. 64 Tei C, Sakamaki T, Shah PM, et al. Myocardial contrast echocardiography: a reproducible http://www.bmj.com/ 34 Asinger RW, Mikell FL, Sharma B, Hodges M. Observations on detecting left ventricular technique of myocardial opacification for identifying regional perfusion defecits. Circulation thrombus with two-dimensional echocardiography: emphasis on avoidance of false positive 1983;67:585-93. diagnoses. Amj Cardiol 1981;47:145-56. 65 Monaghan MI, Quigley PJ, Metcalfe JM, Jewitt DE. Digital subtraction contrast echocardi- 35 Labovitz AJ, Pearson AC. Evaluation of left ventricular diastolic function: Clinical relevance and ography: a new method for the evaluation of regional myocardial perfusion. Br Heart J recent Doppler echocardiographic insights. Am HeartJ 1987;114:836-51. 1988;59: 12-9. 36 Silverman KJ, Huthchins GM, Weiss JL, Moore GW. Catenoidal shape of the interventricular 66 Martin MA. Reproducibility of M-mode echocardiography. In: Hunter S, Hall R, eds. Clinical septum in idiopathic hypertrophic subaortic stenosis: two dimensional echocardiographic echocardiwgraphy. Tunbridge Wells: Castle House Publications, 1986:1-7. confirmation. Amj Cardiol 1982;49:27-32. 67 Popp RL, Filly K, Brown OR, Harrison DC. Effect of transducer placement on echocardiographic 37 Sutton MG, StJ, Tajik AJ, Gibson DG, Brown DJ, Seward JB, Giuliani ER. Echocardiographic measurement of left ventricular dimensions. AmJ7 Cardial 1975;35:537-40. on 27 September 2021 by guest. Protected copyright. INSTRUCTIONS TO AUTHORS The BMJ3 has agreed to accept manuscripts prepared in accordance with the Drugs should be given their approved names, not their proprietary Vancouver style (BMJ, 6 February 1988, p 401) and will consider any paper names, and the source of any new or experimental preparations should be that conforms to the style. More detailed and specific instructions are given given. below. SI units are used for scientific measurements, but blood pressure should The following include the minimum requirements for manuscripts continue to be expressed in mm Hg. submitted for publication. should be described in the methods section or All material submitted for publication is assumed to be submitted Statistical procedures exclusively to the BMJ' unless the contrary is stated. supported by references. Manuscripts will be acknowledged; letters will not be unless a stamped Tables and illustrations should be separate from the text of the paper. addressed envelope is enclosed. Tables should be simple and should not duplicate information in the text of Papers will normally be refereed and may be statistically assessed before the article. acceptance. References must be in the Vancouver style and their accuracy checked Authors should give their names and initials, their posts at the time they before submission. They should be numbered in the order in which they did the work, and one degree or diploma. All authors must sign their appear in the text. consent to publication. Letters to the editor submitted for publication must be signed Three copies should be submitted. If the manuscript is rejected these personally by all authors, who should include one degree or diploma. will be shredded. The editor reserves the customary right to style and ifnecessary shorten Typing should be on one side of the paper, with double spacing between material accepted for publication and to determine the priority and time of the lines and 5 cm margins at the top and left hand side of the sheet. publication. Abbreviations should not be used in the text. Detailed instructions are given in the BMJ dated 2 January 1988, p 48.

1076 BMJ VOLUME 297 29 OCTOBER 1988