Associated diseases and the underlying mechanism of ECG changes

The heart diseases most frequently associated with right atrial enlargement (RAE) are congenital heart disease, chronic obstructive pulmonary disease (COPD) and valvular heart disease with right ventricular involvement.

It has been observed that when the right atrium enlarges, the P wave increases in voltage and becomes peaked. The duration of the P wave does not increase because although right atrial activation may be prolonged, it never exceeds the duration of left atrial activation (Figure 9.1B).

In COPD, pulmonary hypertension, and pulmonary emphysema, the P loop often points to the right and downward (vertical), although not beyond +90°. This explains why the ÂP is generally deviated to the right (P pulmonale). Thus, the projection of the vertical P loop on the frontal and horizontal plane results in a low‐voltage P wave in lead I and a peaked P wave of high voltage in II, III, and aVF (P ≥ 2.5 mm) (Figures 9.2B and 9.3B).

In some congenital heart diseases, such as Fallot’s tetralogy, the P loop points more to the left and forward. This explains why the ÂP is displaced somewhat to the left, and consequently the P wave voltage in lead III is lower than in leads I and II (P congenitale) (Figures 9.2A and 9.3C). Sometimes the voltage in leads I and II is of high amplitude (> 2 mm and peaked). The projection of this forward‐pointing P loop on the horizontal plane produces a P wave in V1 with a voltage that may be high and positive or present with a ± morphology but with a rapid inscription diphase , the opposite to that observed in left atrial enlargement (LAE) . Exceptionally, a pattern or even may appear in V1, but in V2 a predominantly positive morphology is recorded. This pattern in V1 is probably caused by the very enlarged and dilated RA, which adopts a low anterior position, and the electrode of V1 faces the negative part of the P vector. The congenital heart diseases that frequently present with these P wave characteristics in V1–V2 are Ebstein’s disease, severe pulmonary stenosis, and tricuspid atresia. In the case of Ebstein’s disease, the P wave in V1 may present with a large voltage, at times only positive and peaked, while at other times biphasic with a large negative node (P + − − −). Occasionally, the P wave voltage is similar to or even greater than the QRS voltage, with the QRS usually showing an atypical right bundle branch block (RBBB) pattern (Figure 9.4).

Electrocardiographic diagnostic criteria: imaging correlations (Table 9.1 and 9.2)

Although many diagnostic criteria were described in the past based on anatomic correlations or electrophysiological hypotheses (Sodi‐Pallares 1956), the majority have been studied in the last 40 years based on echocardiography, especially with two‐dimensional measurements (Reeves et al. 1981; Kaplan et al. 1994). Recently, CMR standards have been introduced for this purpose (Tsao et al. 2008). Table 9.1 shows the ECG patterns that demonstrate a very high specificity for RAE based on these studies, although the sensitivity never is ≥ 50%.

Combining QRS and P criteria

The combined criteria PV2 > 1.5 mm + ÂQRS > 90° + R/S in V1 > 1 have a correct sensitivity (≈50%) with specificity of 100% (Kaplan et al. 1994).

False positive and false negative diagnoses of right atrial enlargement

The ECG diagnosis of RAE may be very difficult to reach for the following reasons:

• The voltage of the P wave is strongly influenced by extracardiac factors, which may result in increases (hypoxia, sympathetic overdrive, etc.) (false positive) (Figure 9.8) or decreases in voltage (emphysema, other barrier factors, atrial fibrosis, etc.), (false negative) (Bayés de Luna et al. 1978a) (Figure 9.9).
  
• The presence of associated atrial block may result in the transient or permanent disappearance of the ECG criteria for right atrial enlargement (false negative) (Figure 9.10).
  
• On the other hand, a high‐voltage P wave may be seen in patients with exclusively left heart pathology and possible left atrial enlargement (false positive) (pseudo‐P‐pulmonale) (Chou and Helm 1965).

These are some of the reasons why changes in the atriogram are generally not very sensitive (many false negative) for the diagnosis of RAE. Although there are some factors that increase the incidence of false positives, they are usually fewer and therefore the specificity of ECG criteria for RAE is much higher.

Associated diseases and underlying mechanisms of ECG changes

In the past, the most common associated diseases were rheumatic heart diseases, particularly mitral stenosis.In fact, the characteristic bimodal P wave of left atrial enlargement (LAE), seen especially in lead II, has been called “P mitrale.” However, this type of P wave is currently seen in the presence of other causes of LAE, such as cardiomyopathies, especially dilated cardiomyopathy, arterial hypertension, and ischemic heart disease.

The P wave seen in LAE usually has a longer than normal duration due to long distance that the stimulus has to cover as a consequence of the dilated LA more than to the hypertrophy of the LA mass itself (Josephson et al. 1977; Velury and Spodick 1994). The enlarged left atrium first expands toward the back, so the vector of LAE points backward (Figure 9.11). Moreover, because of all that, the P loop is longer and often acquires a figure‐of‐eight shape in the horizontal plane (HP). All that explains that the negative component of P wave is greater than the positive, and therefore, the P wave in lead V1, in case of LA enlargement when the electrode of V1 is well located, usually has a positive–negative morphology with a prominent negative component that exceeds the positive component (Figure 9.12). This is not usually the case in the presence of partial IAB without LA enlargement. However, in cases of bad location of the electrode in V1, too high, a P wave, is recorded with false final negative morphology in the absence of LA enlargement. Finally, if exist associated advanced interatrial block (see later) the P wave is be positive–negative in leads II, III, and VF, and in these cases, LAE is practically always present.

An abrupt dilatation of the left atrium, which occurs in acute pulmonary edema, for example, may cause +/−P wave morphology in V1. This disappears when the clinical situation improves (Heikila and Luomanmaki 1970) (Figure 9.13).

ECG diagnostic criteria: Imaging correlations (Table 9.2)

Historically, the majority of ECG criteria was based on necropsic or radiological correlations. Today, they are based on the standards of echocardiography and, more recently, CMR. (Josephson et al. 1977; Bartell et al. 1978; Bosch et al. 1981; Miller et al. 1983; Munuswamy et al. 1984; Bayés de Luna et al. 1985; Hazen et al. 1991; Tsao et al. 2008; Troung et al. 2011). Recently, it has been demonstrated by echo‐speckle tracking echo, that this technique may be a surrogate of CMR (Montserrat et al. 2015) to demonstrate the abnormal fibrotic ultrastructure of LA wall, and even the ECG pattern of A‐IAB may also be a surrogate of decrease of LA strain, and that means that the diagnosis of A‐IAB by ECG is an equivalent to the presence of fibrosis in the LA wall (Lacalzada‐Almeida et al. 2019) (Ciuffo et al. 2020).

As we have already commented, the P wave criteria for diagnosis of LAE have low sensitivity. Having this in mind, we will now comment on the ECG criteria of LAE.

• Criteria based on changes of P wave The imaging–ECG correlations have demonstrated that the best ECG criteria for LAE are the following (see Table 9.2):
  
• P wave morphology in V1(P wave indices) (see Chapter 7). A negative mode in V1 with a duration ≥ 40 ms by itself is a sign frequently seen in LAE and has a relatively good SE and high SP, according the paper of Morris (Morris et al. 1964). The Morris index has a better predictive value. This index is obtained from the product of the amplitude of the negative P wave component in V1 (in mm) and the duration of this component (in ms). A Morris index ≥ −40 mm × ms 40 ms is very specific for LAE (Figure 9.14). In practice, this means that the terminal portion of the P wave in V1 has a depth of at least one small box in the recording paper (−1.0 mm = 0.1 mV) and a duration of at least 0.04 sec. This index is very rarely seen in normal populations (SP ≈ 90%) but the SE usually is very low. In Morris paper (1964) the SE was relatively high because all were patients with value heart disease and we are not sure about the location of electrode of V1.

In fact, recently it has been demonstrated (see before) that the morphology of Ptf V1 depends very much on the location of the electrodes of V1 (Rasmussen et al. 2019). If is too high (3^rd of 2^nd ICS), the morphology of P ± with great negative component is the rule. Also, the P in V1 may present a ± pattern or even all negative in cases of pectus excavatum, and other thoracic abnormalities. This may explain the discordant results obtained as a precursor of stroke. Due to that, as we are doing an observational trial to clarify this problem (Sajeev et al. 2019).

• The criteria of the NYHAC score (duration of P wave in II ≥ 120 ms + ÂP in FP close to 0° + Morris index ≥ −40 mm × ms) have a high SP (> 95%) but the SE is much lower (< 20%) (Bartell et al. 1978).
  
• The combination of P ≥ 120 ms in lead II + negative P mode duration in V1 ≥ 40 ms has a SE of 50% with a good SP (> 85%) (Bosch et al. 1981).
  
• The combination of P ≥ 120 ms in lead II plus a Morris index > 0.04 mm/sec has a high SP but a low SE (Lee et al. 2007).
  
• The distance between two peaks of P wave > 0.04 sec, with the second mode being taller than the first, is very specific (≈ 100%) but not sensitive (Munuswamy et al. 1984).
  
• A positive–negative morphology of the P wave in leads II, III, and VF (±) with a duration ≥ than 0.12 sec, the key criteria for advanced interatrial block with retrograde activation of the left atrium (see later), is a very specific criterion for LAE > 90% globally and 100% in valvular heart disease and cardiomyopathies, but has a very low sensitivity (< 5%) (Bayés de Luna et al. 1985) (Figures 9.16, 9.18 and 9.19). If present, it is a clear marker of AF in the future (Bayés de Luna et al. 1988) and even of stroke (Escobar‐Robledo et al. 2018). However, as commented, its sensitivity for LAE is low (5%) (Table 9.2).

  

  
  
• The P wave duration (P wave index) in leads I, II, and/or III in LA enlargement is ≥ 0.12 sec, and is generally bimodal. In some echocardiographic correlations, this criterion is very specific (88%) for LA enlargement, but the sensitivity is 33% (Munuswamy et al. 1984). In contrast, Lee et al. (2007) used two‐dimensional echocardiography measurements to show different results (higher SE = 69%, lower SP = 49%) that were similar to those found by Tsao et al. (2008) using the CMR correlation (high SE = 60% and low SP = 35%). This is probably associated with the type of population used in the different studies. In the series studied by Tsao et al. (2008) (CMR), the patients probably present with isolated interatrial blocks more frequently, which decreases the specificity and increases the sensitivity of this criterion. The same is found in the series presented by Lee et al. (2007) (2D echo). The frequent presence of partial interatrial block without LA enlargement may explain this low SP (< 50%). However, since isolated partial interatrial bloc usually does not present with an important negative mode of the P wave in V1 (Figure 9.12), when we consider the association of P duration ≥ 120 ms + Morris index, or the Morris index alone, as criterion for LAE, we obtain much higher specificity (see Table 9.2). However, the value of Morris index is questionable (see other parts of this chapter).