Effects of acute atrial fibrillation and cardioversion on left and right atrial pressures in a dog

Abstract

The occurrence of right-sided congestive heart failure (CHF) in dogs with left-sided heart disease is well-recognized, but its mechanisms are incompletely understood. A 12-year-old Maltese dog was admitted to the clinic for left atrial decompression to treat recurrent CHF due to severe myxomatous mitral valve disease (MMVD). Left atrial decompression was successful but atrial fibrillation (AF) occurred during the procedure. Electric cardioversion restored normal sinus rhythm (NSR) and the dog’s recovery was uneventful. This sequence of events made it possible to study intracameral pressures individually in each atrium in a dog with naturally occurring MMVD during AF and again during NSR. Although pressures in both atria declined following cardioversion, the right atrial pressure declined to a greater degree. These findings indicated a disproportionate effect of AF on right atrial pressure. This difference was noteworthy given the long-standing clinical observation that dogs with MMVD have a higher prevalence of right-sided CHF when AF is present.

Key clinical message:

A dog with MMVD had a greater reduction in right atrial pressure than in left atrial pressure when its AF was cardioverted as part of a cardiac catheterization procedure. This observation proposed a mechanism for the well-known but unexplained observation that dogs with MMVD manifest right-sided CHF disproportionately more often when they have AF.

A basic principle of cardiovascular pathophysiology is that the clinical manifestations of congestive heart failure (CHF) reflect the side of the heart that is affected by a disease process (1). A frequently observed example in veterinary medicine involves the most common heart disease of dogs, myxomatous mitral valve disease (MMVD) (2). Since the mitral valve is a left-sided cardiac structure, advanced MMVD produces signs of left-sided CHF, such as dyspnea due to cardiogenic pulmonary edema (1,2). However, exceptions exist to this simple principle (3,4). In particular, signs of right-sided CHF can occur more often with MMVD when affected dogs also have atrial fibrillation (AF) (3). This paradox has possible explanations, such as differences in venous capacitance of the systemic and pulmonary vasculatures (5) or the existence of comorbidities; e.g., pulmonary arterial hypertension (4). Investigations of hemodynamic changes caused by AF might offer insights into this paradox, but they are rarely undertaken, perhaps because interventional therapies for treating naturally occurring MMVD are a recent development (6,7). In the case reported here, a disproportionate decrease in right atrial (RA) pressure was documented in association with restoration of normal sinus rhythm (NSR) after AF. This finding offered new information that could be associated with the paradox of the higher prevalence of right-sided CHF associated with AF in dogs with left-sided heart disease compared to dogs with the same diseases who are in NSR.

Case description

A 12-year-old, 4.4-kilogram, neutered male Maltese dog with severe MMVD was admitted to the clinic for treatment including left atrial decompression (LAD), a catheter-based procedure that reduces left atrial (LA) pressure through creation of an atrial septal defect (ASD) (6). The dog had experienced recurrent episodes of left-sided CHF requiring hospitalization, most recently 10 d before the procedure. At a recheck 3 d before the procedure, recurrent pulmonary edema was diagnosed and treated on an outpatient basis with injectable furosemide (3.4 mg/kg, SC). An LAD procedure was recommended at that time, as recurrent CHF despite treatment with medications is an indication for LAD (6). At the time of the procedure, the dog was being managed with furosemide (11.4 mg/kg per day PO, with additional 3.4 mg/kg SC boluses q24h or q12h as needed for dyspnea), pimobendan (1.1 mg/kg per day), and sacubitril/valsartan (11.3 mg/kg, PO, q12h). Echocardiography was completed prior to the procedure, and revealed severe LA enlargement [LA:Ao: 2.61; reference value: 1.00 to 1.68 (8)] and no RA enlargement [RA area index: 8.49 cm²/m²; reference value: 4.2 to 10.2 cm²/m² (9)] or other 2-dimensional changes consistent with pulmonary hypertension (4). There was mild tricuspid regurgitation with a measured velocity of 3.21 m/s (reference value: ≤ 3.0 m/s) (4). Preoperative radiographs revealed mild pulmonary interstitial infiltrates consistent with pulmonary edema despite treatments described above. Dobutamine was initiated (5 μg/kg per minute, IV) to improve stability during anesthesia prior to LAD. Left atrial decompression results in an immediate and sometimes profound drop in LA pressure, which in our experience may be excessive if preload is already compromised by diuresis. The dog was premedicated with methadone (0.3 mg/kg IV) and midazolam (0.25 mg/kg IV) and induced with ketamine (1 mg/kg IV) and etomidate (1 mg/kg IV, to effect). Anesthesia was maintained with continuous rate infusions of ketamine (15 to 20 μg/kg per minute), remifentanil (15 to 20 μg/kg per hour), and midazolam (0.3 mg/kg per hour), titrated to effect.

Direct measurement of mean RA pressure during NSR, and prior to LAD, was slightly elevated [4 mmHg; reference value: 1.8 ± 0.2 mmHg (10)]. Left atrial decompression was carried out as reported (6). The LAD procedure was successful; an iatrogenic ASD was created and left-to-right flow across it was documented. However, AF occurred during catheter manipulation in the RA. Left atrial and RA pressures were measured during AF directly, via intracardiac catheterization (Figure 1). In an attempt to convert the acute AF to NSR, cardioversion was done by delivering 10 J of biphasic direct-current energy transcutaneously. This treatment restored NSR immediately. Left atrial and RA pressures were measured again, this time during NSR (Figure 2). Cardiac catheters were withdrawn, incisions were closed, and anesthetic recovery was uneventful. The dog was discharged the following day; antiarrhythmic medications were not instituted. Medical therapy at discharge included furosemide (8.5 mg/kg per day), pimobendan (1.1 mg/kg per day), sildenafil (1.1 mg/kg, PO, q8h), and clopidogrel (18.75 mg, PO, q24h). Sildenafil was given to unload the right ventricle and reduce the risk of postoperative right CHF, which is common practice postoperatively at our clinic following LAD. The dog remained in NSR without requiring antiarrhythmic medical therapy at the last recheck, 23 mo post-procedure.

Figure 1.

Intracameral pressure tracings of the right (RA) and left (LA) atria in a dog during atrial fibrillation. Mean LA pressure: 22 mmHg; mean RA pressure: 18 mmHg. High-amplitude V waves are attributed to mitral and tricuspid regurgitation; their inconsistent amplitude reflects variable ventricular loading conditions, which is a characteristic of atrial fibrillation. Similarly, no A waves are seen.

Figure 2.

Intracameral pressure tracings of the right (RA) and left (LA) atria in a dog in normal sinus rhythm, following cardioversion from atrial fibrillation. Mean LA pressure: 19 mmHg; mean RA pressure: 12 mmHg. High-amplitude V waves are attributed to mitral and tricuspid regurgitation. A waves are attributed to atrial contraction.

Atrial pressures following LAD, during AF, are shown in Figure 1. Pressures were elevated in both atria [mean LA pressure: 22 mmHg, reference value: 6 mmHg (11); mean RA pressure: 18 mmHg]. The effect of AF on atrial waveforms is apparent, with absent A waves and variable peak V wave pressure, depending on coupling intervals. This finding confirmed that AF existed in both atria, which was expected but would not have been verifiable on surface electrocardiograms alone. The large V waves are common in this type of patient, and could be explained by severe atrioventricular valve regurgitation, low atrial compliance (12), and the presence of a left-to-right shunt (the iatrogenic ASD created during LAD) in the case of the RA waveform. Noteworthy changes following cardioversion (Figure 2) included a reduction in atrial pressures and restoration of A waves. Specifically, mean LA pressure decreased to 19 mmHg and mean RA pressure decreased to 12 mmHg. The disparity in the magnitude of reduction in pressures, with the RA pressure (33% decrease) declining substantially more than the LA pressure (14% decrease) despite the presence of left-to-right interatrial shunting that would be expected to contribute to a higher RA pressure (6), was of particular interest.

Discussion

Right-sided CHF is known to be associated with severe MMVD when dogs also have AF. For example, in a series of 155 dogs with degenerative mitral valve disease, right-sided CHF was noted in 13/17 dogs with AF (77%) compared to 10/138 dogs without AF (7%; P < 0.0001) (3). Since AF involves both atria, it would be expected to increase pressures in both atria. Therefore, the higher prevalence of right-sided CHF in dogs with MMVD and AF historically has been unexplained, particularly because such patients often have predominantly left-sided heart disease.

The present case demonstrated a disproportionate rise in RA pressure compared to LA pressure when AF was present. Right atrial pressure was slightly elevated before the onset of AF in this dog (mean: 4 mmHg). During AF, mean LA and RA pressures were severely elevated, although the iatrogenic ASD would be expected to have contributed to increased RA pressure. With restoration of NSR via cardioversion, a disproportionate effect of AF on pressures in the 2 atria became apparent. In fact, the reduction in RA pressure associated with conversion of AF to NSR likely was undermeasured in this case because left-to-right shunting through the iatrogenic ASD would increase RA pressure. The discordance between RA and LA pressure declines following return to NSR in this dog supported the contention that LA pressure elevation is primarily driven by the presence of severe mitral regurgitation in these patients, whereas pressure elevation in the RA is due to the arrhythmia per se. This discordance, as illustrated in the present case, could explain the preponderance of right CHF in animals with MMVD and AF.

Although decreased following cardioversion, the RA pressure remained above the reference interval. This is an expected outcome of the LAD procedure (6). This dog has not developed abdominal effusion or other manifestations of right-sided CHF and remained in NSR at the last recheck, 23 mo postoperatively.

A second important aspect of this case is the response of peracute-onset AF to low-energy cardioversion. In dogs, AF is seldom treated with electrical cardioversion (13), and when it is, AF has usually been present for weeks (median: 30 d) (14). Thus, pretreatment with antiarrhythmic drugs such as amiodarone is recommended. In the present case, AF occurred peracutely. The short duration of AF made successful cardioversion more likely, whereas the presence of MMVD made it less likely (14). The low dose of energy that was effective (10 J) corresponded to the previously reported successful median dose of 1.8 J/kg (13).

A limitation of this report is that LA pressure was unknown before the onset of AF. It is reasonable to expect that LA pressure was elevated based on the presence of severe recurrent left-sided CHF in the animal. However, the abrupt onset of AF prevented LA pressure measurements from being made in NSR. A second limitation is the effects of general anesthesia, cardiovascular drugs, and the LAD procedure, although this is offset to some extent by the fact that RA and LA pressures were recorded together during the same anesthetic event.

This case was exceptional because it provided information that is rarely accessible, since so few dogs with naturally occurring MMVD currently undergo cardiac catheterization with direct hemodynamic monitoring of both atria during AF and NSR. Two key findings are presented. First, the conversion of AF to NSR was associated with changes in atrial pressures of different magnitudes in the RA and the LA, with a greater effect noted in the RA. This within-day comparison of atrial pressures during NSR and during AF in a dog with naturally occurring MMVD provides some observational data that could be related to a well-recognized but poorly explained phenomenon: the occurrence of right-sided CHF in dogs with AF and heart disease that is not limited to the right side of the heart. Second, this case demonstrated that acute-onset AF can respond to low-energy cardioversion even when severe MMVD is present, that the resultant NSR provided immediate hemodynamic benefits, and that NSR was sustained for at least 23 mo. CVJ

Funding Statement

Generous funding support for this work was provided by the Ptarmigan Foundation.