Left Atrial Anatomy

TTE is a reliable and reproducible method to assess the anatomy of the body of the left atrium (LA). 2D directed M-mode echocardiography^1–3 in the parasternal long or short axis allows for the measurement of LA dimension (Fig. 42-1, A). Overall, this unidimensional index correlates well with angiographically derived LA area and volume^2,3 but may be disproportionately erroneous in common disease states associated with asymmetric atrial dilation, such as mitral valve disease⁴ or situations in which the LA is deformed by extrinsic structures.

Figure 42-1 Transthoracic echo (TTE). Parasternal long-axis view. A, TTE image from parasternal long axis demonstrating measurement of the LA dimension from 2D and M-mode data. B, TTE from apical four-chamber view demonstrating measurement of LA length and LA area.

2D TTE provides more accurate assessment of atrial anatomy. The LA is well seen from both the parasternal and the apical windows. In many laboratories, LA anatomy is characterized by the M-mode or 2D dimension from the parasternal long or short axis and the LA length from the apical four-chamber view (see Fig. 42-1, B). However, because the LA is not spherically shaped and asymmetric enlargement may occur, 2D-derived LA volume provides a more accurate measure of LA size⁵ and is advocated by the American Society of Echocardiography.⁶ It more favorably compares with volumetric methods such as cardiovascular magnetic resonance and cardiac computed tomography.^7,8 With the biplane method,⁹ maximum LA area and length are measured in the apical four-chamber (A_4C, L_4C) (see Fig. 42-1, B) and two-chamber (A_2C, L_2C) orientations with volume derived as:

Population-based studies suggest LA volumes of 41.9 ± 11.9 mL or when normalized for body surface area, 22 ± 5 mL/m².¹⁰ Increasing LA volume predicts development of AF in the general population¹¹ as well as after cardiac surgery¹² and in hypertrophic cardiomyopathy.^13,14 Biplane and Simpson’s methods compare closely.⁹ In situations in which there is focal atrial asymmetry, such as distortion of the LA from an extrinsic structure (e.g., mediastinal tumor, hiatal hernia, or descending thoracic aortic aneurysm), accurate calculation of LA volume may be improved by using a Simpson approach.

TEE is a well-tolerated but moderately invasive diagnostic imaging technique that allows for superior visualization of posterior structures such as the LA and LA appendage. Although TEE may be used to assess the body of the LA from orientations analogous to TTE imaging,¹⁵ its greatest advantage over TTE is in evaluation of the anatomy and function of the left and right atrial appendages, structures poorly seen by TTE.

Identification of Left Atrial Thrombi

Anatomic imaging of the body of the LA may be readily obtained from 2D TTE, but identification or exclusion of LA and LA appendage thrombi is best performed by TEE. The sensitivity of TTE for detection of LA thrombi (Fig. 42-2, A) is only 39% to 63%,^16–18 with very limited success for identifying thrombi in the LA appendage. Even with modified views,¹⁹ visualization of the LA appendage may be accomplished in less than 20% of patients.²⁰ Although identification of LA appendage thrombi using state-of-the-art TTE equipment has been described (see Fig. 42-2, A),²¹ and echocardiographic contrast may be useful,²² this has not been confirmed in a large series.

Figure 42-2 Left atrial appendage thrombus. A, TTE image taken from the apical two-chamber view demonstrating a thrombus (arrow) in the LA appendage. B, TEE in the same patient confirming the LA appendage thrombus. C, TEE in another patient demonstrating multiple thrombi in the LA appendage.

Imaging of the LA appendage is readily accomplished with multiplane TEE (Fig. 42-2, B). The normal length and neck width of the adult LA appendage in the horizontal and vertical imaging planes are 28 ± 5 mm and 16 ± 5 mm, respectively.²³ LA appendage anatomic indexes are dependent on imaging plane, with greater neck width and cross-sectional area when observed at a 135° imaging plane as compared with a 45° or 90° plane,²⁴ consistent with its shape idealized as a special ungula of a right circular cylinder.²⁵ The LA appendage commonly has multiple lobes and trabeculations (Fig. 42-3).

Figure 42-3 TEE of LA appendage. A, Single-lobed LA appendage (LAA). B, LA appendage with an accessory lobe (arrow).

The accuracy of TEE for the identification of LA thrombi has been reported by several investigators and was initially based on older monoplane technology.^20,25–27 For nearly two decades, multiplane TEE technology has been the standard with superior visualization of the LA appendage. In our experience of nearly 200 consecutive patients undergoing intraoperative TEE immediately before mitral valve surgery, TEE sensitivity and specificity for LA thrombi were 100% and 99%, respectively.²⁸ We have found the 90° imaging plane particularly advantageous for assessing the LA appendage for thrombus. Real-time three-dimensional (3D) TEE has been used to characterize LA appendage anatomy.²⁹ The accuracy of this technology for identification of LA appendage thrombus is currently unknown, though case reports and small series suggest a benefit (versus 2D multiplane TEE) when the appendage is heavily trabeculated.³⁰ For patients who are not candidates for conventional TEE, microprobe TEE or intracardiac echo^31,32 may be options, but these technologies have not been as well studied and few data are available on comparative accuracy.

Right Atrial Anatomy

The right atrium has not been as well studied by either TTE or TEE. As is the case with the LA, 2D TTE provides a reasonably comprehensive assessment of right atrial (RA) volume. The primary transthoracic window is the apical four-chamber view, with RA area estimated using a length-diameter ellipsoid formula. Parasternal short-axis and subxiphoid windows provide complementary data. The RA appendage is rarely appreciated from TTE, but TEE at 90° (see Fig. 42-3) and 135° imaging orientations readily allow for visualization of this structure (Fig. 42-4).³³ RA appendage area and length are independent of transducer orientation with areas of 5.4 ± 2.4 cm² and lengths of 4.0 ± 0.9 cm reported for those in sinus rhythm.³³ Care must be taken to avoid mistaking normal structures (such as the Eustachian valve or Chiari network) for thrombi.³⁴

Figure 42-4 TEE of normal right atrial appendage (RAA). A, 92° imaging plane. B, 137° imaging plane.

SVC, Superior vena cava.

TTE and TEE sensitivity and specificity data for identification of RA thrombi are relatively sparse. Schwartzbard and colleagues³⁵ reported on 20 patients with TEE evidence of RA thrombus, including 7 confirmed at surgery. TTE identified thrombus in only 6 patients (30%). All RA appendage thrombi were missed by TTE.

Atrial Mechanical Function

In addition to atrial anatomy, echocardiography is a powerful tool for evaluation of atrial systolic function. The atria are traditionally believed to have two primary functions: as passive conduits or reservoirs for oxygenated blood passing from the vena cava or pulmonary veins to the ventricles, and as booster pumps to augment passive diastolic ventricular filling.

Before the advent of echocardiography, assessment of atrial mechanical function had been limited to invasive catheter-based techniques,³⁶ an impractical approach for serial assessments. M-mode echocardiographic assessment of mitral valve motion (Fig. 42-5) can be used in patients with cardiac arrhythmias.³⁷ Normal anterior mitral leaflet excursion is 13 mm,^37,38 with depressed excursion seen with mitral stenosis, with low cardiac output, and with atrial mechanical dysfunction.

Figure 42-5 TTE 2D guided M-mode demonstrating the normal motion of the mitral valve and atrial systole.

The height of the M-mode anterior leaflet A wave has been used as a measure of atrial mechanical function.

Transmitral Doppler has become recognized as an accurate technique for quantifying atrial mechanical function. In the absence of aortic regurgitation, left ventricular (LV) filling may be divided into passive and active phases. The initial (early, or E) wave in the transmitral Doppler flow profile represents passive ventricular filling, and the final (atrial, or A) wave represents active filling during atrial systole (Fig. 42-6). This transmitral Doppler flow profile can be influenced by a number of factors, including heart rate, loading conditions, and sample volume position. The most common indexes of interest include peak A-wave velocity, percent A-wave filling, and peak and percent E/A ratios. Peak E- and A-wave velocities vary considerably with sample position. In our laboratory, we assess the transmitral profile at its maximum (between the tips of the mitral leaflets).

Figure 42-6 TTE 2D guided spectral display of transmitral pulsed Doppler echocardiographic flow velocity.

A, Atrial filling wave; E, early filling wave.

In addition to the previously described quantitative measures of transmitral Doppler data, the transmitral Doppler profile may also be used to calculate atrial ejection force, defined as that force which the atrium exerts to propel blood into the left ventricle (LV).³⁹ The atrial ejection force is based on Newtonian mechanics and is proportional to the mitral orifice area and the square of the peak A-wave velocity. As with more traditional E-wave and A-wave indexes, atrial ejection force may be dependent on loading parameters. If calculation of atrial ejection force is desired, a pulsed Doppler sample position at the level of the mitral annulus would be geometrically more appropriate.

Left Atrial Appendage Function

Because the LA appendage is a blind pouch, volumetric ejection must equal inflow during each cardiac cycle. Pollick and Taylor⁴⁰ noted that patients with sinus rhythm who were free of atrial thrombus had a characteristic contractile LA appendage apex and a noncontractile base. 2D TEE-guided pulsed Doppler has allowed for further characterization of LA appendage systolic performance. The pulsed Doppler profile taken at the mouth of the LA appendage displayed characterizing ejection and filling phases (Fig. 42-7, A). In a healthy adult population, peak LA appendage ejection and filling velocities are 46 ± 18 cm/s and 46 ± 17 cm/s, respectively.⁴¹ Investigators have subsequently characterized LA appendage flow patterns as having four distinct morphologic types (see Fig. 42-7): type I, sinus; type II, flutter (regularized sawtooth); type III, fibrillatory (irregular sawtooth) with ejection velocity greater than 0.2 m/s; and type IV, stagnant/absent flow with peak ejection velocity of greater than 0.2 m/s. In rare instances, there may be a discontinuity between the body of the atrium and appendage^42,43 with sinus pattern on transmitral flow but a fibrillatory pattern in the appendage or even a discrepancy among LA appendage lobes.⁴⁴ Although anatomic indexes appear to be dependent on imaging plane, Doppler indexes of LA appendage ejection and inflow are independent of imaging plane.²⁴ Spontaneous echocardiographic contrast, a marker for stasis, is typically seen with type III or IV appendage flow, whereas LA appendage thrombi are most commonly seen with type IV flow.^45,46 As with transmitral Doppler data, LA appendage flow velocities are dependent on Doppler sample position and loading conditions.⁴⁷

Figure 42-7 TEE spectral display of pulsed Doppler echocardiographic flow velocity with the sample volume at the mouth of the LA appendage.

A, Patient in sinus rhythm. B, Patient with atrial fibrillation with minimal atrial appendage mechanical function. C, Patient with atrial flutter. Vertical calibration markings for each represent 0.1-m/s intervals.

RA appendage function has not been as well studied, but can also be readily assessed by TEE using pulsed Doppler in the 90° or 180° orientations. One report indicated normal RA appendage ejection velocity of 40 ± 16 cm/s, significantly lower than LA appendage ejection velocity.³³

Prevalence of Atrial Fibrillation

AF is a common arrhythmia, with an overall prevalence of 0.4% in the United States,⁴⁸ and is responsible for more than 180,000 hospital admissions annually.⁴⁸ Its prevalence increases sharply with advancing age and is found in up to 4% of those over 60 years.⁴⁹ Data from the Framingham Heart Study suggest that an increased transmitral E/A was associated with an increased risk for the subsequent development of AF.⁵⁰ In an earlier Framingham Heart Study, the presence of permanent AF is associated with a near doubling of both overall and cardiovascular mortality.⁵¹ Atrial flutter is a far less common sustained arrhythmia. Analogous detailed epidemiologic studies are not available.

Symptoms and Clinical Impact

AF is characterized by a lack of organized atrial mechanical and electrical activity. This condition promotes stasis of blood within the atria and increases the risk of atrial thrombus formation. Thromboembolism associated with AF is believed to account for nearly half of all cardiac sources of emboli. In addition to the risk of thrombus formation, loss of organized atrial mechanical activity impairs the atrium’s ability to act as a booster pump and thus impairs ventricular filling (and consequently cardiac output).⁵² The mechanical function of the atria appears to be less severely impaired in atrial flutter, but thrombus formation remains a concern, particularly among patients with alternating AF and flutter or flutter in the setting of mitral stenosis or LV systolic dysfunction.

Important clinical issues related to AF include the following:

Atrial Size and Atrial Fibrillation

LA enlargement is common among patients with AF, particularly those with mitral valve disease and/or systemic hypertension. In addition, LA enlargement decreases the likelihood of long-term maintenance of sinus rhythm.^58,59 Aronow and co-workers⁶⁰ studied 588 patients and found LA enlargement in 57% of patients with permanent AF as compared with only 8% of those with sinus rhythm. Data from the Framingham Heart Study suggest that among those without a history of AF, LA enlargement is among the strongest predictors for the subsequent development of nonvalvular AF.⁵¹ Several groups have reported that LA size increases with sustained AF.^61–63

Although AF promotes further LA enlargement, data suggest that cardioversion of AF and long-term maintenance of sinus rhythm may reverse this process. Using M-mode techniques, DeMaria and associates³⁸ reported a decrease in LA dimension within 1 hour of cardioversion. We studied a group of 21 patients with AF of 5 months’ duration.⁶⁴ Patients with persistent sinus rhythm had significantly smaller left atria 3 months after cardioversion as compared with precardioversion data. In contrast, the group of patients who reverted to AF had no change in LA size. Similarly, Gosselink⁶⁵ and Van Gelder,⁶⁶ with their co-workers, reported on atrial size in 41 and 120 patients, respectively, with permanent AF undergoing precardioversion. At 6 months after cardioversion, both LA and RA volumes decreased among patients with persistent sinus rhythm, with no change in the group who reverted to AF. Similar results have also been seen 6 to 12 months after pulmonary vein isolation using 2D⁶⁷ and real-time 3D TEE LA volume methods.⁶⁸ These data strongly suggest that restoration of sinus rhythm may reverse the process of progressive LA and RA enlargement.

Because atrial enlargement may be deleterious and cardioversion to sinus rhythm may prevent or reverse such dilation, we think that avoidance of cardioversion should not be based solely on an absolute LA dimension or volume. Patients with permanent (more than 1 year) AF, rheumatic mitral valve disease, and prominent LA enlargement (dimension greater than 6.0 cm; volume greater than 44 mL/m²), however, are far less likely to have long-term maintenance of sinus rhythm.⁶⁹

Detection of Atrial Thrombi in Atrial Fibrillation

AF is believed to be responsible for almost half of cardiogenic thromboembolism. Several large, multicenter, prospective randomized studies have confirmed the beneficial effect of chronic warfarin anticoagulation (international normalized ratio [INR] 2.0 to 3.0) or the oral thrombin inhibitor dabigatran (110 or 150 mg twice daily) in patients with nonvalvular AF^70–75 for clinical stroke prevention, with one retrospective study suggesting an INR of greater than 2.5 immediately before cardioversion to be particularly beneficial.⁷⁶ Because TTE is so limited for the assessment of atrial thrombi,^16–18 data on the prevalence of LA thrombi were not available until the introduction of TEE. Among patients presenting with AF of greater than 2 days’ or unknown duration, we found atrial thrombi in 13%,^77,78 of which more than 92% were LA thrombi and nearly all involved the LA appendage. These data are similar to those reported by others,^79–85 but higher than the approximately 6% incidence of clinical thromboembolism following cardioversion without prolonged anticoagulation.^86–89 This apparent discrepancy may be because some thrombi may not migrate and some embolic events may be clinically silent.^90,91 Patients at particularly high risk for atrial thrombi (Table 42-1) include those with rheumatic mitral valve disease, depressed LV systolic function, recent thromboembolism,⁹² and TEE evidence of severe LA spontaneous echocardiographic contrast and complex aortic debris.^55,91 Duration of AF and LA dimension are not predictive of LA thrombi.^71,72 The CHADS2 scoring system has also been associated with increased risk of TEE evidence of LA thrombus^57,93: LA thrombi and spontaneous echocardiographic contrast are very rarely found in patients with a CHADS2 score of 0 with a progressive increase in atrial thrombi with increasing CHADS2 score (Fig. 42-8). Despite these provocative data, expedited cardioversion (without anticoagulation and screening TEE) or without a month of therapeutic warfarin is currently not recommended for the CHADS2 = 0 group. Similarly, preliminary data suggest that TEE evidence of LA appendage thrombi is not seen among those with a totally normal TTE.⁹⁴ In contrast to data suggesting that at least moderate mitral regurgitation is protective against clinical thromboembolism,^95,96 we have found that mitral regurgitation is not protective against LA thrombi (see Table 42-1) among those with new-onset AF.⁷⁷

TABLE 42-1 Characteristics of Patients at Risk for Atrial Thrombi*

Figure 42-8 Progressive relationship between both TEE evidence of LA appendage thrombus and severe spontaneous echocardiographic contrast (SEC) and increasing CHADS2 score.

(Adapted from Puwanant S, Varr BC, Shrestha K, et al: Role of the CHADS2 score in the evaluation of thromboembolic risk in patients with atrial fibrillation undergoing transesophageal echocardiography before pulmonary vein isolation. J Am Coll Cardiol 54:2032, 2009.)

Immediate cardioversion is generally advocated for patients with AF of less than 2 days’ duration,⁹⁷ under the assumption that the prevalence of atrial thrombi in this group was very low. This common teaching was challenged when Stoddard and colleagues⁹⁸ reported a 14% prevalence of atrial thrombi among patients with AF of less than 3 days’ duration and a prevalence of 27% in those with a duration of at least 3 days in a predominantly male population. In contrast, we found clinical thromboembolism following cardioversion (without antecedent TEE or prolonged warfarin anticoagulation) of less than 1% among patients with AF of less than 2 days’ duration.⁹⁹ Thus, prolonged warfarin or screening TEE is likely not needed in this group (unless there is a history of thromboembolism, severe LV systolic dysfunction, or mitral stenosis). Although we perform cardioversion of AF of less than 48 hours’ duration without prolonged warfarin or screening TEE, we do initiate therapeutic anticoagulation at presentation (intravenous unfractionated heparin or low molecular weight heparin) rather than delaying anticoagulation until the patient has been in AF for 48 hours. If TEE is performed in this short-duration AF group, evidence of LA appendage thrombi is found in 1.4% to 4% of the group,¹⁰⁰ with thrombus and spontaneous echocardiographic contrast associated with depressed LV systolic function and enlarged LA.

As might be expected, atrial thrombi are more common among AF patients who present with acute thromboembolism. In our experience, residual LA thrombi are found in nearly 50% of these patients.¹⁰¹ Because this group represents a very high clinical risk for whom chronic warfarin (or dabigatran) is indicated, we do not perform TEE to search for thrombi unless cardioversion is desired.

RA thrombi are far less common among patients with AF and represent less than 5% of all atrial thrombi identified by TEE.^77,78,102 RA spontaneous echocardiographic contrast is also distinctly unusual, seen in only 10% of AF patients,⁷⁷ and is highly predictive for RA thrombi.

Guidance of Early Cardioversion

Cardioversion of AF is performed in an effort to improve cardiac function, relieve symptoms, and potentially decrease thrombus formation. However, data from several large prospective studies demonstrate no mortality or thromboembolic advantage to aggressive cardioversion and maintenance of sinus rhythm (versus chronic anticoagulation and rate control).^105–107 (For symptomatic patients who are difficult to rate control, however, cardioversion is still advocated.) Unfortunately, successful cardioversion of nonvalvular AF is associated with an approximately 6% incidence of clinical thromboembolism among patients who are not systematically anticoagulated for several weeks before cardioversion.^86–89 Because atrial thrombi are poorly detected by TTE, conventional care of patients with AF of unknown or prolonged (more than 2 days) duration had demanded that these patients receive several weeks of anticoagulation before cardioversion, with one study suggesting that an INR greater than 2.5 immediately before cardioversion further reduces risk.⁷⁶ Cardioversion is then followed by several weeks of anticoagulation while atrial mechanical function recovers¹⁰⁸ and for prophylaxis should the patient revert to AF. Although no randomized and only two prospective studies^79,82,109 have been reported, a month of precardioversion warfarin therapy decreases the risk of a clinical embolic event following cardioversion to less than 1.6%.^{76,79,82,88,89} The majority of these clinical embolic events occur within the first 10 days after cardioversion (Fig. 42-9).^109,110 Use of warfarin, however, carries a risk of major (2%) and minor (10% to 20%) hemorrhagic complications.^79,82,111 In addition, many patients have a subtherapeutic INR during the month before cardioversion. For these individuals, the warfarin dose is increased and the “1-month clock” restarted, an approach supported by TEE studies demonstrating LA appendage thrombi in patients with a transient subtherapeutic INR.^112,113 Finally, conventional therapy leads to a delay in cardioversion for the large majority of patients who do not have an atrial thrombus, and a second hospitalization is needed for cardioversion.

Figure 42-9 Timing of thromboembolic events after cardioversion in patients receiving warfarin, dabigatran 110 mg twice daily (D110), and dabigatran 150 mg twice daily (D150).

Note that the vast majority of events are within the first 10 days after cardioversion.

(Reprinted from Nagarakanti R, Ezekowitz MD, Oldgren J, et al: Dabigatran versus warfarin in patients with atrial fibrillation: an analysis of patients undergoing cardioversion. Circulation 123:131, 2011.)

Rationale and Advantages

A TEE-guided approach to early and safe cardioversion (Fig. 42-10) has several advantages over traditional strategies for hospitalized patients with AF (Box 42-1). Currently, up to 8 weeks of oral anticoagulation is recommended with cardioversion,^97,114,115 including 1 month before and 1 month after cardioversion. This period of anticoagulation exposes patients to a significant risk of a hemorrhagic complication^79,82,111 by doubling the exposure to systemic anticoagulation (versus an expedited TEE approach). For uncertain reasons, the AF population appears to be at increased risk of hemorrhagic complications during the second month of anticoagulation.⁷⁹

Figure 42-10 Schematic of TEE-guided approach to early cardioversion.

Long term, either chronic warfarin or dabigatran can be used as guided by CHADS2 score. AF, Atrial fibrillation; LAA, left atrial appendage; PTT, partial thromboplastin time.

Box 42-1 Benefits and Risks of Tee-Guided Cardioversion

Benefits

Shorter initial duration of atrial fibrillation (AF)

Enhanced recovery of atrial mechanical function

More rapid resolution of symptoms of congestive heart failure

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. Handheld Echocardiography
2. The Role of Echocardiographic Evaluation in Patients Presenting with Acute Chest Pain to the Emergency Department
3. Echocardiographic Evaluation of the Patient with Pulmonary Hypertension
4. Echocardiographic Evaluation of Coronary Blood Flow: Approaches and Clinical Applications

Stay updated, free articles. Join our Telegram channel

Join