Edler and Hertz and the Beginning of Clinical Echocardiography (Ultrasound Cardiography) and M-Mode Imaging

It was the collaboration between cardiologist Inge Edler (1911–2001) and physicist Carl Hellmuth Hertz (1920–1990) (Figure 11-2) at Lund University in Sweden in the early 1950s that is commonly accepted as the beginning of clinical echocardiography.¹³*

Figure 11-2 I. Edler (right) and C. H. Hertz (left) in 1979.

They recorded the first M-mode echocardiogram of the heart in 1953 and are considered the “fathers of clinical echocardiography.”

(From Roelandt JR: Seeing the invisible: A short history of cardiac ultrasound. Eur J Echocardiogr 1:8–11, 2000.)

In a search of a better way to evaluate the mitral valve function before closed mitral commissurotomy, Edler considered the use of something like RADAR (see Table 11-2). He was referred to a young physicist at Lund University, C. Hellmuth Hertz. Hertz was the son of a Nobel Prize winner, and, remarkably, his uncle, Heinrich Hertz, had lent his name to the unit of frequency. Hertz thought that ultrasound (frequencies higher than that audible to the human ear) might be the solution to Edler’s needs.¹⁴ Hertz borrowed an ultrasonic reflectoscope used in nondestructive testing of metals at the ship-building yard in Malmo for a weekend, and Elder tested it on Hertz’s chest and detected movement in the heart. The Siemens Corporation provided them with a reflectoscope in October 1953. They first studied isolated hearts in the laboratory with A-mode (static amplitude), but then devised a recording technique to display motion versus time (M-mode) by photographing the image off the oscilloscope (Figures 11-3 and 11-4). Their machine could produce ultrasound frequencies of 0.5, 1.0, 2.5, and 5.0 MHz, and Hertz opted to use 2.5 in humans as the optimal compromise between penetration and resolution. They first used this machine in a patient on October 29, 1953; they termed it “ultrasound cardiography” (UCG).¹⁹ At first they attributed the reflectors to the posterior wall of the left ventricle and the anterior wall of the left atrium, but, subsequently, Edler demonstrated that the latter came from the anterior leaflet of the mitral valve. He determined this by passing an ice pick in the comparable direction through the chest at the time of an autopsy on a patient he had echoed shortly before death.² He subsequently used UCG to evaluate pericardial effusions and, in 1956, detected a left atrial myxoma (but this was not published until 1960).¹² However, his principal use of UCG was to evaluate mitral stenosis, and he described the value of the E-F slope in quantifying the severity of the stenosis. In the mid-1950s, Edler and Hertz actually experimented with introducing a transducer into the esophagus to overcome the attenuation in lung tissue, but they abandoned this effort because of difficulties in obtaining acoustic coupling between the transducer and the esophageal wall.¹² They published the first review article on UCG in Acta Medica Scandinavia in 1961.²⁰ Hertz left the field of cardiac ultrasound fairly early after he developed ink-jet technology, and Edler made few innovations after 1960.² In 1977, Edler and Hertz were awarded the Lasker Clinical Medicine Research Prize (often referred to as the American Nobel prize in Medicine).

Figure 11-3 The ultrasonoscope initially used by Edler and Hertz for recording their early echocardiograms.

(From Feigenbaum H, Armstrong WF, Ryan T: Feigenbaum’s echocardiography, 6th ed. Philadelphia: Lippincott Williams & Wilkins, 2005.)

Figure 11-4 One of the earliest M-mode echocardiograms of the mitral valve recorded by Edler and Hertz in December 1953.

(From Roelandt JR: Seeing the invisible: A short history of cardiac ultrasound. Eur J Echocardiogr 1:8–11, 2000.)

Others Enter the Field

In the late 1950s, Schmidt and Braun, and Sven Effert in Germany separately began duplicating Elder and Hertz’s work, and in 1959, Effert published the first report identifying an intra-atrial tumor by echocardiography. Between 1961 and 1965, workers in Shanghai and Wuhan, China began to report their use of echocardiography including fetal echocardiography.¹³

The American Experience

In 1957, engineers John Wild and John Reid at the University of Minnesota described ultrasonic echocardiographic imaging of excised hearts.²¹ When Reid went to the University of Pennsylvania, he joined forces with cardiologist Claude Joyner, building an ultrasonoscope that they used to study mitral stenosis. Their report, which appeared in the journal Circulation in 1963, was the first American publication on clinical echocardiography.²² Harvey Feigenbaum in Indianapolis, author of one of the first (first published in 1972) and still published textbooks of echocardiography, first became interested in echocardiography in 1963 (Figure 11-5). He published his first article on use of ultrasound to diagnose pericardial effusions in 1965,²³ and as mentioned later, he collaborated with Reggie Eggleton to develop a mechanical two-dimensional (2D) scanner. Dr. Feigenbaum also may have been the first to train nonphysicians to do echocardiograms, leading to the development of cardiac sonographers.¹³ Feigenbaum believes that Bernie Segal of Philadelphia was the first to use the term echocardiography in print in an article published in 1966.^24,25 The senior author (E.A.H.) recalls cardiologists presenting their primitive, indistinct, and difficult to interpret M-mode images of the mitral valve at preoperative case conferences in the late 1960s and thinking that echocardiography did not have much future. How wrong he was!

Figure 11-5 Feigenbaum using his early M-mode echocardiography, which used a Polaroid camera to record the echocardiogram.

(From Feigenbaum H, Armstrong WF, Ryan T: Feigenbaum’s echocardiography, 6th ed. Philadelphia: Lippincott Williams & Wilkins, 2005.)

Between the 1950s and the 1970s, M-mode was the only clinically useful format for echocardiography, and many advances and applications were described as the equipment and the skill of the echocardiographers improved, peaking in the late 1970s. But then its role rapidly declined with the development of 2D instrumentation.⁹

Two-Dimensional Scanners

Work to develop real-time 2D scanners began in the 1960s, made possible by advances in sonar and radar technology and circuitry. Early pioneering work was done by the previously mentioned Americans Wild and Reid,²⁶ and Howry and Bliss²⁷ in the 1950s. In the later 1960s, Ebina²⁸ in Japan and Asberg²⁹ reported producing ultrasound-generated tomographic images of thoracic structures in humans. Their clinical application was limited because of transducer size.

In 1968, Somer³⁰ constructed the first electronic phased-array scanner based on the wavefront theory formulated in the 17th century by Huygens.¹¹ In 1971, this was followed by the description by Nicholas Bom³¹ of Rotterdam of an electronic linear scanner (Figure 11-6) that generated a rectangular image and, in 1974, of an electronic phased-array scanner by F. L. Thurstone and O.T. von Ramm at Duke University.^11,32 At the same time, J. Griffith and W. Henry, at the National Institutes of Health, introduced a mechanical sector scanner.^11,33 This was said to be cumbersome to manipulate, and hence shortly thereafter Reggie Eggleton, working with Feigenbaum in Indianapolis, developed a handheld real-time mechanical 2D scanner³⁴ that was more “user friendly” and commercially successful¹³ (Figure 11-7). In 1976 and 1977, Kisslo and colleagues^35,36 at Duke University reported on the clinical use of the von Ramm and Thurston phased-array scanner for 2D echocardiography (2DE).

Figure 11-6 Twenty-element electronic linear two-dimensional transducer developed by N. Bom in Rotterdam circa 1972.

(From Feigenbaum H, Armstrong WF, Ryan T: Feigenbaum’s echocardiography, 6th ed. Philadelphia: Lippincott Williams & Wilkins, 2005.)

Figure 11-7 First practical handheld mechanical real-time sector scanner developed by Eggleton and Feigenbaum in the mid-1970s using a modified Sunbeam electric toothbrush.

(From Feigenbaum H: Evolution of echocardiography. Circulation 93:1321–1327, 1996.)

Development of Vascular and Cardiac Doppler

In 1956, Satomura reported the application of the Doppler principle in the use of ultrasound to measure blood flow velocity.^11,37 However, in 1961, Franklin, at the University of Washington, was the first to measure Doppler shift of frequency, and hence velocity of flow of blood in vessels using a continuous-wave ultrasonic device.³⁸ Pulse-wave Doppler (in which flow at a defined depth could be analyzed) was nearly simultaneously introduced in 1969 and 1970 by P. N. T. Wells of England, P. A. Peronneau of France, and D. W. Baker in the United States.¹¹ In 1974, F. E. Barber, D. W. Baker, and colleagues at the University of Washington introduced the combination of pulse-wave Doppler with 2D scanning to produce the subsequently widely used “Duplex Scanner.”^11,39 Subsequently, in Norway, but at different institutions, J. Holen⁴⁰ and L. Hatle⁴¹ applied the Bernoulli principle to estimate the pressure decline across stenotic mitral and aortic valves, respectively.

Color-Flow (Doppler) Echocardiography

In 1978, the Swiss-born M. A. Brandestini, working at the University of Washington, described a method (based on pulse-wave Doppler) of color encoding flow velocity data and superimposing it on M-mode images.¹³ His initial observations were extended by Namekawa et al⁴² in Japan who described, in 1982, color Doppler flow imaging (CFI) utilizing autocorrelation to produce real-time images of flow in 2D sector scans. Bommer and Miller in the United States⁴³ and Omoto working with the Namekawa group⁴⁴ described clinical use of 2D color Doppler flow imaging.^9,11

Prelude to Perioperative Transesophageal Echocardiography

The first reported use of intraoperative echocardiography (IOE) was in 1972 when Johnson and his colleagues⁴⁵ at the University of Colorado reported on the use of intraoperative epicardial M-mode echocardiography to evaluate the results after open mitral commissurotomy. In 1976, Wexler and Pohost, of Massachusetts General Hospital, wrote a review article in the journal Anesthesiology on noninvasive techniques for hemodynamic monitoring, in which they described the basic principles and the potential application to anesthesia of the relatively new technique of “echocardiography.”⁴⁶ At that time, this was limited to M-mode technology, but they anticipated that in the future 2D sector scanning would overcome some of the limitations of M-mode scanning. In that same year, at the annual meeting of the American Society of Anesthesiologists (ASA) in October in San Francisco, Rathod and colleagues,⁴⁷ at Cook County Hospital and Loyola University in Chicago, and Paul Barash and colleagues,⁴⁸ at Yale University, described use of transthoracic M-mode echocardiography to monitor the effects of anesthetics on cardiac function during anesthesia in 20 adults and 13 children, respectively. They concluded that this technique was a useful method to assess cardiac function, and Barash et al⁴⁸ predicted that it “has the potential to be a useful tool for clinical anesthesiology that may supplant the invasive monitors currently available.”

In 1978, Strom and colleagues at Albert Einstein College of Medicine in Bronx, New York, published their experience with use of intraoperative epicardial M-Mode scanning during cardiac surgery,⁴⁹ which led to their investigations of transesophageal echocardiography (TEE; see later).

When superior handheld 2D transducers became available, they began to be used epicardially during cardiac surgery after gas sterilization or wrapping them in sterile sheaths. In the early 1980s, Spotnitz and colleagues,⁵¹ at Columbia University College of Medicine in New York City (NYC), reported use of intraoperative epicardial 2D scanning after CPB to detect intracardiac air⁵⁰ and to assess left ventricular ejection fraction before and after cardiac surgery. In 1984, Goldman and colleagues, at Mount Sinai Medical Center in NYC, reported the use of intraoperative epicardial contrast 2DE to evaluate regurgitation after mitral valve surgery⁵² and to assess myocardial perfusion during cardiac surgery.⁵³

Intravascular Ultrasound

The development of TEE had its roots in the search for alternative ultrasound windows because of difficulties encountered in obtaining good ultrasound signals through the chest using the early insensitive transthoracic transducers.⁷ In the early 1960s, this led to the investigation of intravascular probes. In 1960, Cieszynski⁵⁴ inserted a single-element transducer on a catheter into the jugular vein of dogs. In 1963, Omoto et al⁵⁵ reported obtaining static cross-sectional images in patients by slowly rotating a single-element transducer inserted into the right atrium,⁷ and a year later, Kimoto et al⁵⁶ reported obtaining C-scans of the atrial septum in humans from an intravascular catheter.⁹ In 1968, Carleton and Clark⁵⁷ reported similar studies. 1970, Reggie Eggleton⁵⁸ mounted four elements on a catheter and created the first cross-sectional images of intracardiac structures by computer reconstruction of images obtained by slow rotation and ECG triggering,⁵ and in 1972, Nicholas Bom,⁵⁹ in Rotterdam, described a real-time intra- cardiac scanner using a 32-element circular electronically phased-array transducer placed at the tip of a 9 F-catheter. Thereafter, interest faded as sophisticated TEE probes became available, only to re-emerge in the late 1980s (see later).

Transesophageal Ultrasound (Doppler and M-Mode)

Side and Gosling⁶⁰ were the first to use transesophageal (continuous-wave) Doppler (transducer mounted on a steerable gastroscope) to assess flow in the heart and aorta in 1971. This was followed by similar reports by Olson and Shelton in 1972. In 1974 and 1975, Duck et al⁶¹ and Daigle et al⁶² used transesophageal pulsed Doppler to measure flow in the thoracic aorta.⁷ However, true TEE is said to have its embryonic beginnings in 1976 when Lee Frazin (a Chicago cardiologist) et al⁶³ reported recordings of M-mode echocardiograms of the aortic root and valve, mitral valve, and left atrium from the esophagus by attaching a nonfocused 3.5-mHz transducer attached to a 3-mm coaxial cable (Figure 11-8). They reported superior recordings compared with the transthoracic approach in 38 patients.⁶³ Following this lead, Matsumoto and Oka,⁶⁴ at Albert Einstein College of Medicine, fabricated a stiff transesophageal probe supporting an M-mode transducer (Figure 11-9). They first used this in 1979 to measure ventricular dimensions and volume during mitral valve surgery in a 65-year-old woman.⁶⁴ They subsequently reported intraoperative transesophageal M-mode monitoring in 21 patients undergoing cardiac surgery.⁶⁵ At that time, M-mode TEE was difficult to interpret except by the extremely sophisticated clinician and hence was not widely adopted. However, the development of a new generation of gastroscopes with steerable tips in the late 1960s, onto which echo transducers could be mounted, had a significant positive impact on the development of TEE by facilitating direct contact with the wall of the esophagus,⁷ overcoming the problem encountered by Edler and Hertz a decade earlier.

Figure 11-8 First clinically used transesophageal ultrasound probe used by Frazin and colleagues in the mid-1970s. Consisted of a nonfocused, 3.5-MHz M-mode transducer mounted on the end of a 3-mm coaxial cable.

(From Frazin L, Talano, Stephanides L, et al: Esophageal echocardiography. Circulation 54:102–108, 1976, Figure 1.)

Figure 11-9 First “homemade” transesophageal M-mode probe used to monitor patients during cardiac surgery by Matsumoto and Oka in the Bronx, NY. The shaft consisted of a No. 4 sternal wire covered with a vinyl esophageal stethoscope catheter (A) KB-aerotech 3.5 MHz transducer; (B) Handmade cable.

(From Oka Y, Goldiner PL: Transesophageal echocardiography. Philadelphia: Lippincott, 1992, Figure 1. Originally from Matsumoto M, Oka Y, Lin YT, et al: Transesophageal echocardiography for assessing ventricular performance. N Y State J Med 79:19–21, 1979, Figure 1.)

Transesophageal Two-Dimensional Imaging

Although some primitive 2D TEE mechanical scanners and linear were described between 1977 and 1980 (e.g., Hisanaga et al⁶⁶ and DiMagno et al⁶⁷), it was not until Jacques Souquet⁶⁸ (Figure 11-10), working in conjunction with the Varian Corporation, developed their phased-array transducer (Figure 11-11) mounted on the end of a gastroscope for their 2D sector scanner system (Figures 11-12 and 11-13) that TEE became a practical reality. These new TEE probes were evaluated clinically by cardiologists Michael Schluter and Peter Hanrath in Hamburg, Germany,⁶⁹ and their preliminary results were highlighted at an international conference held in Hamburg in 1981.⁸ The early subsequent use of TEE took different directions in the United States and Europe. Cardiologists in Germany and the Netherlands rapidly began using TEE in awake patients to aid in the diagnosis of a variety of cardiac pathologies. In the United States, cardiologists seemed reluctant to adopt this new technology.⁶ In the early 1980s, Peter Hanrath sent prototypes (first M-mode transducers and then with the 2D transducers mounted on gastroscopes) to cardiologists James Seward at Mayo Clinic and Nelson Schiller (Figure 11-14) at the University of California in San Francisco (UCSF), who passed them on to anesthesiologists. Schiller⁷⁰ spoke with William Hamilton, Chair of the Department of Anesthesiology at UCSF, who put him in contact with two young faculty members, Michael Cahalan and Michael Roizen. At that time (1981–1983), a cardiology fellow from Hanrath’s group in Germany, Peter Kremer (Figure 11-15), came to UCSF and collaborated with Cahalan (Figure 11-16) and Roizen to investigate these new TEE instruments. At the 1982 annual meeting of the ASA, they started the American TEE “revolution” when they presented their results in monitoring cardiac and vascular surgery patients with this new TEE probe, displaying the high-quality recordings and images obtained (Figures 11-17 and 11-18), and describing its usefulness in assessing filling and function of the left ventricle, and in detecting myocardial ischemia and intracardiac air.^71,72 Subsequently, they reported its superiority to PAC in assessing filling in the operating room⁷³ and usefulness in evaluating the hemodynamic changes during anaphylaxis⁷⁴ and during surgery for pheochromocytoma.⁷⁵ Topol and colleagues at Johns Hopkins University demonstrated its usefulness in determining the cause of hypotension immediately after CPB⁷⁶ and documented improvement in myocardial dysfunction immediately after coronary revascularization.⁷⁷ Cucchiara et al,⁷⁸ at the Mayo Clinic, described its usefulness in detecting air embolism during neurosurgery, whereas the UCSF group described its value in monitoring during vascular surgery^79,80 and its superiority over ECG in detecting myocardial ischemia.⁸¹ In 1984, the Hamburg group and Cahalan published an anatomic analysis of six standard TEE views,⁸² which was expanded on a few years later by the Mayo Clinic group.⁸³ Meanwhile, the group at Mount Sinai Medical Center in NYC (Martin Goldman, Joel Kaplan, Daniel Thys, Zaharia Hillel, and Steven Konstadt) was quick to adopt and evaluate perioperative TEE, following the experience of Kaplan working with the early Diasonic TEE probes in the early 1980s.^84–88 In 1987, Cahalan et al⁸⁹ and Clements and deBruijn⁹⁰ at Duke University wrote review articles on the use of TEE in anesthesiology, whereas the group at Mayo Clinic reported on the reproducibility of measurements obtained during intraoperative TEE.⁹¹ In that same year (1987), the second edition of this text edited by Kaplan on Cardiac Anesthesia contained for the first time a 63-page thoroughly referenced chapter on Intraoperative Echocardiography.⁹²

Figure 11-10 Jacques Souquet, PhD, a contemporary portrait. French ultrasound engineer working in both the United States and Europe, largely responsible for the design of the miniaturized phased-array two-dimensional transducer used in the early transesophageal echocardiography probes.

(From Image: A Conversation with…Jacques Souquet, PhD. Available at: http://www.rt-image.com/A_Conversation_with_Jacques_Souquet_PhD_The_diagnostic_capabilities_of_ShearWave/content=8504J05E48BE588440B6967644A0B0441.)

Figure 11-11 First 32-element 3.5-MHz phased-array transducer developed by Jacquet Souquet that was fitted on an Olympus gastroscope by Souquet.

(From Stumper O, Sutherland GR: Transesophageal echocardiography in congenital heart disease. London: Edward Arnold, 1994, Figure 1.2.)

Figure 11-12 An early Souquet transesophageal echocardiography probe (Souquet phased-array transducer mounted on gastroscope) as used by Hanrath, Schluter, and Kremer in the early 1980s.

(From Schluter M, Langenstein BA, Polster J, et al: Transesophageal cross-sectional echocardiography with a phased array transducer system. Technique and initial clinical results. Br Heart J 48:67–72, 1982, Figure 1A.)

Figure 11-13 The ability to adjust the orientation of the transducer by the external controls on the gastroscope is a key innovative feature of the gastroscope mounted transesophageal echocardiography probe.

(From Schluter M, Langenstein BA, Polster J, et al: Transesophageal cross-sectional echocardiography with a phased array transducer system. Technique and initial clinical results. Br Heart J 48:67–72, 1982, Figure 1B.)

Figure 11-14 Nelson Schiller, MD, in 1983.

He was the head of echo cardiology at the University of California San Francisco (UCSF) and was instrumental in promoting the work of Cahalan, Roizen, and Kremer and supporting their efforts to win acceptance of their publications.

(Photograph courtesy M. K. Cahalan.)

Figure 11-15 Peter Kremer, the cardiology fellow from Hanrath’s group in Germany, who brought the probes to the University of California San Francisco (UCSF) and worked with them there for 2 years. Shown standing before a poster presentation of their work at a cardiology meeting circa 1982.

(Photograph courtesy M. K. Cahalan)

Figure 11-16 Michael K. Cahalan, Professor Anesthesiology, University of Utah.

Pioneered use of transesophageal echocardiography (TEE) perioperatively (together with P. Kremer and M. Roizen) at the University of California in San Francisco in 1982 and helped introduce this modality to cardiologists, surgeons, and anesthesiologists in the United States. He facilitated the collaboration of anesthesiologists with echocardiologists (American Society of Echocardiography) and assisted with the development of postgraduate courses, guidelines, and certifying examinations in perioperative TEE.

(Photograph courtesy M. K. Cahalan.)

Figure 11-17 High-fidelity transesophageal M-mode echocardiogram from the first Hanrath-Souquet probe used in surgery at the University of California San Francisco (UCSF) circa 1981. AW, anterior wall; DD, diastolic distance; et, ejection time; PW, posterior wall; SD, systolic distance.

(Courtesy M. K. Cahalan.)

Figure 11-18 Two-dimensional (2D) image (transgastric midpapillary short-axis view) recording from the first 2D Hanrath-Souquet transesophageal echocardiographic probe used in surgery at the University of California San Francisco (UCSF) circa 1982. alp, anterior lateral papillary muscle; LV, left ventricle; pmp, posterior medial papillary muscle; RV, right ventricle.

(Courtesy M. K. Cahalan.)

In 1986, Goldman and colleagues,⁹³ at Mount Sinai Medical Center in NYC, reported their experience with a new 2D TEE probe that also provided Doppler color-flow imaging (Aloka and Irex), as did a Japanese group.⁹⁴ That same year, the Hewlett-Packard Corporation also introduced a color-flow Doppler TEE probe; and in 1987, deBruijn and Clements, together with pioneering echocardiographer, Joseph Kisslo at Duke University, reviewed their early experience with this new technology.⁹⁵ In that same year, pulse-wave Doppler was added to 2D TEE probes,⁹⁶ although several years earlier the Hamburg group had described results of use of a TEE probe that combined pulse-wave Doppler with M-mode TEE.⁹⁷ In 1989, biplane probes (Figure 11-19) first became available,⁹⁸ although prototypes had been described earlier,⁹⁹ and soon thereafter, multiplane single-transducer TEE probes¹⁰⁰ and smaller diameter probes suitable for pediatric use¹⁰¹ appeared. In 1992, Pandian and colleagues, at Tufts-New England Medical Center, reviewed their early clinical experience with a 5-MHz phased-array multiplane TEE probe (“OmniPlane”).¹⁰²

Figure 11-19 Photograph of single-plane (horizontal or transverse) and biplane (transverse and vertical) probes. The advent of biplane probes greatly increased the views available but were eventually superseded by multiplane probes.

(From Matsuzaki M, Toma Y, Kusukawa R: Clinical applications of transesophageal echocardiography. Circulation 82:709–722, 1990, Figure 1.)

A survey conducted of Society of Cardiovascular Anesthesiologists (SCA) members in early 1988 provides a “snap-shot” of the state of practice of IOE at that time¹⁰³; 20% of the responders reported that IOE studies were being performed at their institutions. The majority of responders were practicing in large teaching hospitals. Anesthesiologists were involved 41% of the time and cardiologists in 33%, although in one third of the former group, cardiologists always assisted. TEE technique was used in 51%, epicardial in 37%, and transthoracic in 22% of cases. At the time of the survey, six manufactures made 2D TEE transducers—Hewlett-Packard, Aloka Toshiba, Diasonics, General Electric, Acusonics, and Hoffrel, but only the first three provide color-flow TEE. Most responders reported using either Diasonics (46%) or Hewlett-Packard (42%); 90% used the 2D mode, 75% M-mode, 60% pulsed-Doppler, 58% color-flow Doppler, and only 23% continuous-wave Doppler, whereas 97% believed that IOE had been or could have been helpful. The authors of the survey noted that few academic medical centers offered formal instruction and certification in IOE, and that a definite need for training had been identified.

Along this same line, in an editorial published in 1989, Kaplan¹⁰⁴ urged caution and restraint in the adoption of this new and complex technology that had the potential to cause numerous complications, and emphasized the need for anesthesiologists who aspire to be echocardiographers to obtain a high level of training.

In the late 1980s, the group at Duke University demonstrated the utility of routine epicardial echocardiography during surgery for congenital heart disease.^105–107 As soon as smaller probes that could be used in infants and children became available, these began to be used. In 1989, Kyo et al,¹⁰¹ in Japan, were among the first to describe the use of TEE in pediatric cardiac surgery, a practice strongly supported by the work of the Seattle group about which they started publishing in 1993.^107–110 By the end of that decade, 98% of pediatric cardiac surgery centers used intraoperative TEE.¹¹¹

Into the Mainstream

The 1990s witnessed the expansion of intraoperative TEE into the mainstream of clinical practice of cardiac surgery and anesthesia,¹¹² as well as into critical care and noncardiac surgery. From the beginning of perioperative TEE, its value to detect myocardial ischemia based on the appearance or regional wall motion abnormalities was recognized.⁸¹ When only single-plane TEE transducers were available, the midpapillary transgastric short-axis view often was advocated for this purpose. With the appearance of biplane and multiplane probes in 1996, Cahalan’s group¹¹³ documented the limitations of the transgastric short-axis view and the need to use multiple views. In that same year, the group at UCSF also documented the value of TEE in assessing ischemia during CABG surgery¹¹⁴; whereas in 1997, the group at Cleveland Clinic documented the benefit of intraoperative TEE in high-risk CABG.¹¹⁵ In 2000, Aronson and colleagues¹¹⁶ reported on the use of intraoperative low-dose dobutamine echocardiography. Evidence of the increased importance of echocardiography to the practice of cardiac anesthesia is that with its first issue in 2001 (volume 15, number 1, February), Journal of Cardiothoracic and Vascular Anesthesia changed the logo on its cover from a PAC to echocardiographic images, and three TEE images graced the cover of Daniel Thys’ Textbook of Cardiothoracic Anesthesiology published in 2001.¹¹⁷

Contrast Echocardiography and Myocardial Contrast Echocardiography

Contrast enhancement of blood represents an important adjunct to echocardiography. It has been used to identify cardiac structures, enhance identification of the endocardial boarder, detect shunts and valvular regurgitation, and assess myocardial perfusion. Feinstein,¹¹⁸ of Rush University in Chicago, and his international colleagues have reviewed the development of what they refer to as “contrast enhanced ultrasound” imaging and have identified important contributors over the years. Both the radiologist R. Gramiak, of University of Rochester,¹³* and the cardiologist C. R. Joyner, Jr., of the University of Pennsylvania,⁹ have been credited as being the first to incidentally note dense echoes associated with the injection of normal saline or indocyanine green during M-mode echocardiography. Both of them subsequently described using saline contrast to help identify cardiac structures during M-mode echocardiography.¹¹⁹ In 1975, Seward et al,¹²⁰ at the Mayo Clinic, described their initial experience with use of indocyanine green as an echocardiographic contrast agent in more than 300 cases, emphasizing its role in analyzing shunts. Bommer et al showed that when echocontrast agents were injected into coronary arteries during echocardiography, they produced myocardial opacification,⁹ which led to the development of myocardial contrast echocardiography. The echo contrast of these agents was attributed to microbubbles. In these early media, the microbubbles were too large to pass through the capillaries including the lung, and hence direct injection into the left heart was necessary to image the structures in the left heart including the myocardium. In 1984, Feinstein et al described the use of ultrasonic energy (“sonication”) to produce contrast agents consisting of relatively uniform, stable, and small gaseous microbubbles that could pass through capillary beds and allow imaging of the left heart chambers and myocardium via right-sided injection.¹²¹ Over time, superior commercial contrast agents have been developed (e.g., Levovist, Albunex), and newer, more sophisticated agents are under development.¹¹⁸

In 1984, Goldman and colleagues⁵² described the use of intraoperative contrast echocardiography to evaluate mitral regurgitation after mitral valve surgery by injecting saline into the left ventricle and monitoring for any regurgitation of contrast into the left atrium with epicardial 2D imaging. In that same year, this group also was the first to describe intraoperative myocardial contrast epicardial echocardiography (myocardial contrast produced by administration of cardioplegia into the aortic root) to predict the presence of coronary artery disease (CAD).⁵³ In the late 1980s and early 1990s, Aronson and colleagues at the University of Chicago, Kabas and Kisslo at Duke, and Spotnitz and Kaul at University of Virginia used intraoperative contrast echocardiography to assess coronary artery surgery.^122–125 In 2000, Aronson¹²⁶ summarized the progress in measurement of myocardial perfusion by contrast echocardiography in the operating room, and in 2006, Dijkmans and his international colleagues¹²⁷ summarized the state of the development of myocardial contrast echocardiography. In 2007, the U.S. Food and Drug Administration called attention to concern about the safety of some of these agents, and in 2008, the American Society of Echocardiography (ASE) issued a consensus statement on “the clinical applications of ultrasonic contrast agents in echocardiography.”¹²⁸

Epicardial and Epiaortic Imaging

As noted earlier, Edler and Hertz first made epicardial ultrasound recordings of cadaver hearts in 1953, as did Wild and Reid in 1957.²¹ Also, as mentioned earlier, in 1972, Johnson et al⁴⁵ reported on the use of intraoperative epicardial M-mode imaging to evaluate the results after mitral commissurotomy. In the early 1980s, the groups at Columbia University College of Medicine and the Mount Sinai Medical Center in NYC used intraoperative epicardial 2DE to detect intracardiac air,⁵⁰ assess left ventricular ejection fraction,⁵¹ evaluate regurgitation after mitral valve surgery (after injection of contrast into the left ventricle),⁵² and assess myocardial perfusion during cardiac surgery.⁵³ Because suitable small transesophageal probes were not available in the late 1980s and early 1990s, epicardial 2D scanning became the standard for echocardiographic surveillance during pediatric cardiac surgery in that era.¹⁰⁶

With the introduction of TEE, epicardial imaging largely was discarded, but recently, both epicardial and epiaortic imaging have regained attention and are being utilized more frequently in situations in which TEE is not possible or adequate, and to assess the thoracic aorta, especially in the “blind spots” of the TEE.¹²⁹ The ASE/SCA have developed guidelines for epicardial¹³⁰ and epiaortic¹³¹ applications.

Intracardiac Echocardiography

As reviewed earlier, intracardiac ultrasound was explored as an alternate method for examination of the heart during the era of limited transthoracic transducers. This approach largely was discarded with improved transthoracic instruments and, especially, the arrival of TEE. Recently, with improved miniaturization and technology of intravascular probes, and the increase in percutaneous intracardiac procedures (e.g., closure devices, percutaneous mitral valve surgery, electrophysiologic procedures), use of intracardiac ultrasound to monitor and aid in the safer performance of these procedures has become more popular.¹³²

Intravascular Ultrasound

The circular-array catheter-mounted transducers ultimately were miniaturized (1 to 3 mm in diameter) and provided with sophisticated electronics so they could be placed into coronary arteries over a guidewire. Utilizing ultra-high frequencies (40 MHz), they permit a resolution of 150 μm to a depth of 2 to 3 cm. In the late 1980s, Nissen and colleagues,¹³³ at the University of Kentucky, were among the first to exploit these new intravascular ultrasound catheters to study CAD. Such technology is now used routinely to study CAD, and larger intravascular ultrasound probes are used to assist with placement of intracardiac devices, as well as to assess larger vessels.¹³

Echocardiography in Critical Care Medicine: In the Noncardiac Operating Room and Intensive Care Units

Three- and Four-Dimensional Echocardiography

Just as the limitations of 1DE (i.e., M-mode) led to the development of 2DE, the obvious limitations of the latter have led to the search and development of 3D imaging systems.

Transthoracic Three-Dimensional Echocardiography

History of 3D imaging in medicine starts in 1961 when Baum and Greenwood¹⁵⁶ used 3D ultrasonography for localization of orbital lesions. All early work in cardiac 3DE focused on methods of acquiring multiple 2D images by moving a standard 2D transducer in space and reconstructing the multiple 2D images into a 3D image. The first such publication came in 1974 from Stanford scientists Dekker, Piziali, and Dong¹⁵⁷ on a system for ultrasound imaging of the human heart in three dimensions. Dekker’s group accomplished this by mounting the transducer onto a movable mechanical arm, which allowed alignment of multiple 2D images and creation of a 3D image (Figure 11-20).

Figure 11-20 Dekker et al’s mechanical arm described in 1974. The transducer (arrow) is held on the mechanical arm, and hence the location of the transducer is known.

(From Houck RC, Cooke JE, Gill EA: Live 3D echocardiography: A replacement for traditional 2D echocardiography? AJR Am J Roentgenol 187:1092–1106, 2006, Figure 2.)

In 1976, the spark-gap technique of 3D reconstruction of the heart was developed by Moritz and Shreve.¹⁵⁸ This technique involved measuring the transit times of spark-generated shock waves to obtain the spatial coordinates of the echocardiographic planes. The microprocessor controlled the operation of the system and performed all the computations required to determine the location and orientation of the ultrasound beam with respect to a known coordinate system. About this same time, Japanese engineers and clinicians with Matsumoto et al,¹⁵⁹ from Osaka University, described technology they used to process 3D images of left ventricle, atrium, and aorta: “We applied our newly developed computerized image processing system for 3-D echocardiographic display with binocular parallax shift and for constructing 2-D echocardiographic images in desired planes from sequential recordings of 2-D echocardiograms recorded with anteroposterior emission of ultrasound beams. This system mainly consisted of a flying spot scanner, a minicomputer and a display cathode ray tube.…The frontal plane 2-D echocardiograms provided useful data.”¹⁵⁹ In 1979, Raab et al¹⁶⁰ developed a magnetic locator that they attached to the transducer and to a central computer that located the transducer in space and subsequently followed its position.

Improvements on these methods led to the freehand imaging, allowing free movement of the transducer at a single or multiple acoustic windows and resulting in a 3D wire-frame image. Images were acquired over short periods during held end-expiration with combination of angulated or rotated scans from parasternal or apical windows. Image quality depended on the patient’s respiration. A magnetic field system was used to track the ultrasound scanhead. Images were digitized and registered with position data and image depth, as well as coordinated with independently acquired ECG. The borders of the left ventricle and associated anatomic structures were manually traced in the selected images.¹⁶⁰

The 3D reconstruction of real-time transthoracic 2D images of the left ventricle, using an apical rotation method, was first described in 1982 by Ghosh and colleagues.¹⁶¹ In their method, the transducer was placed on the patient’s chest wall in the region of the apex to obtain the standard four-chamber view to record 2D left ventricular end-diastolic and end-systolic images. The transducer was then rotated in 30-degree increments from 0 to 180 degrees to obtain various planes passing through the apex. Using spatial coordinate data, 3D perspective images were plotted by the computer in any desired view. Three-dimensional reconstructed volumes closely correlated with those obtained by left ventricular angiography. In that same year, Geiser et al¹⁶² described their technique of dynamic 3D echocardiographic reconstruction of the intact human left ventricle in patients.

The major limitation of these techniques was the inadequate delineation of the endocardium because of the nonperpendicular relation of the ultrasound beam to the left ventricular walls. Reconstruction from TTE images was restricted by the presence of chest wall artifacts, the limited number of acoustic windows, and inadequate image quality in some patients. Despite growing interest in 3DE, the clinical use of this technique was limited by time-consuming, complicated image acquisition and postprocessing, and hence was not available in real time.

Olaf von Ramm, professor of biomedical engineering at Duke University, with his colleague, Stephen Smith, pioneered the development of clinical 3D ultrasound scanners in the late 1980s. In 1987, they patented the first real-time high-speed 3D ultrasound system, capable of acquiring pyramidal volumetric data at frame rates (about 8 per second) sufficient to depict cardiac movement. Their transthoracic transducer, transmitting at frequency 2.5 to 3.5 MHz, included 256 elements configured in a “sparse array” pattern (not all the transducer head was used and connected to individual elements).¹⁶³ Later, this technique was developed commercially by Volumetrics Medical Imaging (Chapel Hill, NC) and existed for several years; however, it never made an impact on clinical practice because of poor image quality.

With real-time 3DE, an entire volume of the heart was obtained using one cardiac cycle, which was a major step up from the thin-slice sector of the 2D imaging. The limitations of the technique were poor image quality, motion artifact, low spatial resolution, low frame rates, limited dimensions of the 3D data set volume, and requirement of ECG and respiratory gating. Basically, images were computer-generated 2D slices derived from the 3D dataset. An off-line workstation had to be used to produce true 3D images.

Despite limitations, reconstructive techniques were used in numerous studies for determination of left ventricular volume, chamber size, mass, ejection fraction, and systolic function, particularly in patients with distorted ventricular shape or aneurysm. Comparison with angiography, thermodilution cardiac output measurements, radionuclide angiography, magnetic resonance imaging, conventional 2DTTE/TEE, and cine-ventriculography showed 3DE to be an accurate, if not superior, noninvasive technique.

In 1996, Hewlett-Packard Company began development of real-time 3DE. A rotational device was developed to be contained within the transducer. By 1998, Hewlett-Packard had a working prototype for transthoracic real-time 3DE and began showing it to customers by 1999. Moving beyond the “sparse array” transducer of van Ramm, in November 2002, Philips Medical Systems introduced the “Live 3D” imaging system, an advanced form of real-time 3D transthoracic imaging. The major advance of that system was improved image quality because of a fully sampled (matrix) or dense array configuration of the transducer. The face of the transducer was completely sampled. This dense-array transducer, called a matrix array, consists of approximately 3000 elements compared with previous transducers with 256 elements. Increased computer processing power and hard drive capacity permitted the system to process, store, and analyze data. Even with this markedly improved ultrasound system, the size of the image volume is not enough to contain the whole left ventricle. Series of the four gated cycles had to be combined to create full-volume images, resulting in maximum frame rates of 25 Hz. 3D left atrial volume measurement could not be obtained from the patient with atrial fibrillation or those who could not hold their breath long enough to acquire a full-volume real-time 3D image. At the 2008 ASE meeting in Toronto, Siemens presented the Acuson SC2000 Volume Imaging Ultrasound System, which acquires full-volume 3D data pyramids—20 images/sec—in a single heartbeat. Acquiring a full pyramid in one heartbeat instead of four consecutive beats could be advantageous, because data would not require any restitching, reanimation, and regating.

Transesophageal Three-Dimensional Echocardiography

In the 1980s, when rapid development of TEE began, first attempts at 3D reconstruction from TEE were begun. The improved image resolution afforded by TEE was important for quality 3D image reconstruction. In 1986, Martin et al,¹⁶⁴ of the Department of Anesthesiology and Center for Bioengineering at the University of Washington, in Seattle, used a precision micromanipulator of a TEE-mounted transversely oriented 32-element 3.5-MHz ultrasonic array transducer to generate a 3D echocardiogram. This allowed them to obtain multiple multiplanar short-axis 2D images of the heart with known angular relations between them over a series of cardiac cycles. An off-line computer analysis of the images was used to form 3D reconstruction of the left ventricular cavity at end-diastole and end-systole from which stroke volume was determined.¹⁶⁴ In a canine study in 1989, Martin and Bashein¹⁶⁵ were able to demonstrate that the stroke volumes generated from their 3D reconstructed volumes were comparable with radionuclide and thermodilution measurements (Figure 11-21).

Figure 11-21 Martin and Bashein’s method of measuring stroke volume in dogs with three-dimensional (3-D) transesophageal echocardiographic scanning. ECG, electrocardiogram; LV, left ventricle.

(From Martin RW, Bashein G: Measurement of stroke volume with three-dimensional transesophageal ultrasonic scanning: Comparison with thermodilution measurement. Anesthesiology 70:470–476, 1989, Figure 1.)

The next step in development of 3D TEE was when Wollschläger¹⁶⁶ et al reported on their system for TEE computer tomography. Their development was commercialized by TomTec Technology Company (Munich, Germany) and has found application in research and clinical practice. This technology was partially based on the rotational device used by Hewlett-Packard Corporation for their multiplane TEE probe that they introduced in 1992.

In 1992, Pandian et al,¹⁶⁷ at Tufts-New England Medical Center, were among the first to report clinical results with 3D TEE. They used a 5-MHz, 64-element, and phased-array TEE probe connected to a computer that directed transducer movement at 1-mm increments. A complete cardiac cycle was recorded at each tomographic level with ECG and respiration gating. These were then processed using dedicated 4D software and displayed as a dynamic 3D tissue image of the heart. In 1997, Chen et al,¹⁶⁸ at the Thorax Center in Rotterdam, Netherlands, published a study on the measurements of mitral valve orifice area with 3DE in patients with mitral stenosis. They had developed their own system for TEE acquisition, using TomTec software for off-line reconstruction. Transesophageal acquisition was performed in six patients with a custom-built transducer assembly; the transducer was rotated by a step motor that was commanded by computer algorithm that controlled the acquisition of cross sections within preset ranges of heart cycle length by ECG gating and respiratory phase.

From the mid-1990s to 2000, for most multiplane TEE probes, external acquisition and computer analysis from TomTec were used, but only Hewlett-Packard Corporation integrated the acquisition software into the ultrasound system itself. Hewlett-Packard had been working on the development of 3D TEE using the 2D reconstruction approach and had released such a system in 1995.

In 2007, Philips introduced its new “Live 3D TEE” probe with a matrix-array transducer (x7-2t, 2500 elements). This probe featured the combination of two new novel technologies: xMATRIX (3D power) technology and “Pure Wave” crystal technology that resulted in improved image clarity. “Live 3DTEE” offers not only conventional imaging modes such as 2D multiplane imaging, M-mode, continuous- and pulse-wave Doppler, and color Doppler, but also real-time 3D imaging: Live 2D, 3D Zoom, Full Volume, and 3D Color Full Volume (Figure 11-22). Live 3D displays a fixed pyramidal data set determined by the depth of the initial 2D images and can be used to visualize any cardiac structures located in the near field. Movement of the probe results in real-time change of the 3D image. 3D Zoom displays a truncated but magnified pyramidal data set of variable size. Placement of the probe over the region of interest with minimized sector width improved temporal resolution and optimized image quality. 3D zoom is used to view the left atrial appendage, interatrial septum, and mitral and tricuspid valves. Full volume provides a pyramidal data set that allows the inclusion of the larger cardiac volume at frame rate greater than 30 Hz. It combines a series of subvolumes acquired with ECG gating to create a final, larger, reconstructed full-volume image. This is the only one of these 3D modes that requires ECG-gated reconstruction. 3D color full volume combines grayscale of full-volume data with color Doppler.

Figure 11-22 Three real-time three-dimensional (3D) imaging modes.

(From Hung J, et al: ASE Position Paper. 3D echocardiography: A review of current status and future directions. J Am Soc Echocardiogr 20:213–233, 2007, Figure 3.)

3DE has found particular application in measuring ventricular volumes (together with stroke volume, cardiac output, and ejection fraction), especially of the right ventricle (which is a particular problem with 2DE because of its complex geometric shape), to evaluate mitral valve, aortic valve, and aortic root pathoanatomy, to evaluate complex congenital heart disease, and as a guide for percutaneous interventional procedures including placement of closure devices, transeptal puncture, ablation procedures, and placement of left atrial closure devices. In 2007, the ASE published a position article on the current status and future directions of 3DE¹⁶⁹ and, in 2008, published a practical guide for its use.¹⁷⁰

Four-Dimensional Echocardiography

4DE or real-time 3DE, that is, 3DE plus time, is a recent addition to the volumetric cardiac imaging field, which is receiving significant clinical interest because it may replace the need for more expensive techniques (such as magnetic resonance imaging) and also suggests new areas of application for volumetric cardiac imaging (e.g., in interventions). The current approaches to 4D medical ultrasonic imaging can be separated into two categories: non–real-time and real-time imaging. Real-time imaging refers to the class of imaging that shows the rendering of the ultrasonic volume in a time frame that can be associated with other clinical monitoring technologies (such as the ECG). This property allows clinicians a more intuitive interface for capturing 4D images because movement of the probe is directly reflected in the imaging volume acquired.

Three-dimensional datasets often can be quite difficult to interpret and require considerable effort and time orienting the observer in relation to anatomic landmarks. When multiple measurements are necessary, as is the case with time-varying volumes, it often is helpful to have semiautomated or fully automated means of quantification, tools that are offered by 4D systems.

Real-time acquisition via 4D (3D plus time) ultrasound obviates the need for slice registration and reconstruction, leaving segmentation as the only barrier to an automated, rapid, and clinically applicable calculation of accurate left ventricular cavity volumes and ejection fraction. Tracking of mitral valve leaflets can determine whether they close fully or whether there is an opening for mitral regurgitation.

In the early 1990s, several studies investigated use of 4DE to study the dynamics of the heart over time utilizing myocardial motion phantoms that were constructed to simulate the motion of myocardial segments using various programmed waveforms and an ECG simulator that provided the necessary R-wave signal in synchrony with the phantom motion. This work was essential for evaluation of new acquisition techniques to provide an accurate method to quantify heart volumes from this type of data.

The next decade brought improvement in technology and new clinical studies. In 2005, Corsi et al¹⁷¹ developed a volumetric analysis technique for quantification of global and regional LV function and studied patients with CAD, cardiomyopathy, and valvular disease. They concluded that volumetric analysis of real-time 3DE data was clinically feasible and allows fast, semiautomated, dynamic measurement of left ventricle volume and automated precise detection of regional wall motion abnormalities. More recent studies have used speckle tracking combined with 4DE to more accurately quantify regional LV strain for strain analysis.¹⁷²

Because the left ventricle is a 3D structure with a complex pattern of wall motion, 4D imaging offers great potential for identification of dyssynchrony. It also could help to identify the optimum pacing site for resynchronization therapy. A recently published study described a novel approach for morphologic and functional quantification of the mitral valve based on a 4D model estimated from ultrasound data.¹⁷³ The 4D model’s parameters are estimated for each patient using the latest discriminative learning and incremental searching techniques (Figure 11-23).

Figure 11-23 Four-dimensional echo visualization.

Advanced automatic contour finding algorithms for an easy, fast, accurate, and highly reproducible functional analysis of the left ventricle.

(From www.emageon.com/downloads/Cardiology/HeartSuite_4D_Echo.pdf.)

Tissue Doppler and Speckle Tracking

In 1989, Isaaz et al¹⁷⁴ described the application of Doppler echocardiography to evaluate myocardial motion, and tissue Doppler imaging was born.¹⁷⁵ By appropriate filtering, they examined the high-amplitude low-velocity signals reflected back from the myocardium instead of the blood with pulsed-wave Doppler. This information has been used to evaluate ventricular function (rate and pattern of filling, and emptying of the ventricle, strain, and strain rate) and to estimate filling pressures. In 1997, Nagueh et al¹⁷⁶ at Baylor University suggested that the ratio of early transmitral blood flow (E) to early movement of the basal myocardium as determined by tissue Doppler (Ea or E′), E/E′, could give a clue to left atrial filling pressure. Color coding (mapping) of tissue Doppler velocities in various parts of the myocardium was introduced by the group at the Royal Infirmary of Edinburgh in 1994¹⁷⁷ and has been used to detect dyssynchrony.

Tissue Doppler recording has the same requirement of close alignment of the ultrasound beam with the movement being interrogated. This limitation has been overcome by the identification of small bright spots in the myocardium on the grayscale image because of backscattering from small structures within the myocardium (less than one wavelength) termed Speckles. This was first described by the innovative engineering group at Duke University in the late 1980s and early 1990s.¹⁷⁸ These speckles can be identified and tracked (i.e., Speckle Tracking) to generate similar information provided by tissue Doppler but without the alignment constraint. These techniques initially were introduced for TTE and are technically demanding but are being used for investigational purposes with clinical TEE, the clinical importance of which are being investigated.¹⁵⁴

Regional Wall Motion and Endocardial Border Detection

The value of echocardiography in detecting regional wall motion abnormalities and especially for early detection of myocardial ischemia has been recognized for many years. This usually was accomplished by leisurely off-line analysis. With the development of perioperative applications (as well as echocardiographic monitoring of stress tests), more accurate online assessment was needed. One of the keys to accurate assessment of wall motion is accurate delineation of the endocardial border. Various techniques have been introduced to facilitate this. Initially, this included bloodstream and myocardial contrast (see earlier). In the early 1990s, Hewlett-Packard Corporation introduced a method of transesophageal real-time automatic (endocardial) border detection using backscatter imaging with lateral gain compensation, which they termed “acoustic quantification.”¹⁷⁹ This subsequently was improved by adding color coding, termed “Color Kinesis” (CK) by the manufacturer.¹⁸⁰ Its use in perioperative TEE was assessed by the group at the University Hospital in Vienna.¹⁸¹ More recently, tissue Doppler imaging enhanced by color coding,^182,183 and Speckle Tracking has been introduced to facilitate detection of abnormal regional myocardial function. Recently, dos Reis et al,¹⁸⁴ from University of Brasilia, briefly reviewed the history of attempts at semiautomatic border detection and proposed a new algorithm combining classic mathematical morphology for binary images, high-boost filtering, image segmentation, and motion estimation.

Handheld or Hand-Carried Echocardiography: The “Ultrasound-Stethoscope”

The concept and reality of a portable lightweight battery-operated echocardiograph machine was introduced by Roelandt et al in 1978,¹⁸⁵ who referred to this as the “Ultrasound-Stethoscope.” Since then, industry has introduced more miniaturized but also more capable portable high-resolution devices that are now even the subject of television advertising. Their use is advocated to complement and expand the history and physical examination, especially in the age of decreased skills in the latter.¹⁸⁶ Mondillo et al¹⁸⁷ have classified these devices, based on their size, capabilities, and cost, into four categories. The top-level devices are now capable of providing 2D, M-mode, color-flow, pulse-wave, and continuous-wave and tissue Doppler. Mondillo et al¹⁸⁷ and Kobal et al¹⁸⁸ have reviewed the capability and limitations of the use of these devices. A key limiting factor is the capabilities of the person conducting and interpreting the examination,¹⁸⁹ and there are controversies about how much training and whether certification is required. Cost implications (vs. benefits) of acquiring the equipment and providing appropriate education are other issues. In 2002, the ASE, the American College of Cardiology, and the American Heart Association addressed the use of hand- carried echocardiography¹⁹⁰ and recommended that users should have a minimum of Level 1 training (75 supervised studies and 150 supervised interpretations), but strongly recommended that they have Level 2 training (150 supervised studies and 300 supervised interpretations), which are far beyond what other proponents of the use of this technology have suggested is necessary.¹⁹¹