Things Are Not Always As They Seem Multimodality Exercise Assessment in the Evaluation of Dyspnea

A 68-year-old man with history of hypertension, coronary artery disease (CAD) with prior percutaneous intervention in the left anterior descending, and heart failure with preserved ejection fraction (HFpEF) presented with progressive exertional dyspnea and chest heaviness. He had recently been diagnosed with HFpEF following hospitalization for congestion requiring parenteral diuretics. He complained of marked dyspnea and dizziness walking half a block. The pertinent cardiac medications included low-dose aspirin, clopidogrel, atorvastatin, furosemide, lisinopril, and metoprolol.

Physical examination showed heart rate (HR) 96bpm, blood pressure (BP) 137/82 mmHg, and body mass index 26.5 kg/m². Cardiac examination revealed normal jugular venous pressure, regular rate and rhythm, no gallop, and no peripheral edema. Electrocardiogram, chest x-ray and labs were unremarkable. The echocardiogram demonstrated normal left ventricular size and ejection fraction (EF) (63%), with no regional wall motion abnormalities. There was a grade 1 diastolic dysfunction (E/A 0.9) with septal e’ 6 cm/s, lateral e’ 11 cm/s, E/e’ ratio of 13, mildly enlarged left atrium (volume index 37 ml/m²) and diastolic predominant pulmonary vein Doppler. His right ventricular size and systolic function were normal and estimated right ventricular systolic pressure was 24 mmHg. Mild mitral regurgitation was present with normal valve morphology and a central jet. NT-proBNP was mildly elevated to 323 pg/ml. The HFA-PEFF score was 6/6.

Dr. Borlaug: The patient has been diagnosed with HFpEF, by an HFA-PEFF score of 6/6 and a prior hospitalization for volume overload. However, it is always important to consider alternative causes or contributors to destabilization. Lung auscultation and chest radiography are clear, arguing against pulmonary disease. There is no suggestion of underlying cardiomyopathy, pericardial disease, or primary valvular heart disease based upon physical examination or echocardiography. The patient has typical comorbidities, echocardiographic suggestion of high filling pressures, and a recent hospitalization for congestion, consistent with HFpEF. However, he also has known CAD, which is observed in two-thirds of patients with HFpEF, associated with worse outcomes, and may improve with revascularization.¹ Given his presentation with progressive chest discomfort and dyspnea in the absence of clear volume overload, an evaluation for ischemia is the most important next step.

Patient presentation (continued): An exercise treadmill cardiopulmonary stress test was performed. The patient exercised to a peak VO2 of 18.4 mL/min/kg (69%predicted). The peak respiratory exchange ratio (RER) was 1.26 (maximal effort), minute ventilation/carbon dioxide production (VE/VCO2) slope was 27 (normal breathing efficiency) and HR response was limited (peak 114 BPM, 75% predicted). There were no ischemic changes in ECG during exercise, but left bundle branch block (LBBB) developed at HR of 85 bpm. BP dropped from 124/62 mmHg at rest to 96/52 mmHg during exercise, and patient reported severe dyspnea and dizziness at peak exercise.

Dr. Borlaug: The stress test shows impaired aerobic capacity, confirming the reported symptoms. The ECG did not show ischemic changes, but development of LBBB may signal ischemia in the left anterior descending (LAD) artery territory, which would also fit with the hypotensive response to exercise. Alternatively the LBBB may simply be related to conduction system disease. Noninvasive evaluation for ischemia in the setting of HFpEF is often inaccurate,¹ and the high risk findings on stress testing require further evaluation with coronary angiography.

Dr. Bradley: Transient exercise-induced LBBB occurs in four out of every thousand patients undergoing exercise stress testing.² Several factors, alone or in combination, may increase the likelihood of exercise-induced LBBB. These factors include left bundle (LB) refractory period prolongation, drug effects, myocardial ischemia, electrolyte abnormalities, and structural abnormalities of the cardiac conduction system resulting from an infiltrative process, calcification, or prior infarction.

In a minority of patients, the refractory period of the LB, as opposed to the right bundle, is prolonged. If exercise then causes the HR to exceed a critical value, an electrical impulse arising in the atria may arrive at the LB before the Purkinje cells in the LB have had time to adequately recover from the previous beat. As a result, the impulse is blocked in the LB, thereby producing an aberrant QRS complex in the form of LBBB. Exercise-induced LBBB is commonly seen in the presence of coronary artery disease,² where impaired blood flow to the LB may cause abnormal conduction.

Dr. Borlaug: Patients with HFpEF often report angina on activity, even in absence of epicardial coronary disease. Recent data have shown that this may be related to supply-demand mismatch due to coronary microvascular dysfunction causing subendocardial ischemia and injury.³ Myocardial injury as reflected by high sensitivity troponin levels is increased in many patients with HFpEF, particularly during exercise.³ The degree of injury present in HFpEF is associated with the severity of myocardial dysfunction and hemodynamic abnormalities. High cardiac filling pressures may be caused by ischemia-induced diastolic dysfunction, and the high filling pressures may then exacerbate ischemia by decreasing the coronary perfusion pressure, creating a vicious cycle.³ Myocardial ischemia may be detected during exercise in the catheterization laboratory through direct sampling of blood from systemic circulation and coronary sinus. As angiography is indicated based upon the treadmill test, it would also be reasonable to perform invasive exercise testing to clarify the hemodynamic abnormalities and evaluate for exercise-induced ischemia.

Patient presentation (continued): At cardiac catheterization, hemodynamics were normal at rest, with right atrial (RA) pressure 6 mmHg, pulmonary artery wedge pressure (PAWP) 12 mmHg, mean PA pressure 19 mmHg, and cardiac output 4.86 L/min (Figure 1A). With leg elevation, there was an increase in PAWP to 19 mmHg. With 20 Watts exercise, there was a further increase in PAWP to 26 mmHg with v waves to 35 mmHg (Figure 1B). Later during the same 20 Watt workload, there was an abrupt rise in PAWP to 35 mmHg with v waves to 85 mmHg, immediately following the development of new-onset LBBB (Figure 1C). Development of LBBB was associated with reduction in BP (Figure 2A), which improved during exercise coincident with transient normalization of QRS (Figure 2B), followed by recurrence of LBBB, again with reduction in BP and increase in PAWP.

Figure 1: Invasive Hemodynamics.

Pulmonary artery wedge pressure (PAWP, red) at rest (A), during low-level exercise prior to the onset of left bundle branch block (LBBB) (B), and during the transition from normal conduction with narrow QRS (green) to LBBB (C). Note the dramatic increase in the amplitude of PAWP V waves beginning within 3 beats following the onset of LBBB.

Figure 2: Hemodynamic changes with onset and resolution of LBBB.

Pressure tracings at slower sweep speed and expanded ordinate showing arterial blood pressure (BP) (black), pulmonary artery wedge pressure (red), and ECG (green) at the initial onset of left bundle branch block (LBBB) (A) and during a transient resolution of LBBB following an increased cycle length due to premature ventricular contraction (B). Note the dramatic reduction in systemic BP with LBBB that instantly improves with normal conduction.

Coronary sinus sampling was performed at rest and peak exercise using a multipurpose catheter in tandem with arterial blood sampling (Figure 3). Coronary sinus O₂ saturation decreased from 34% at rest to 12% with exercise. Coronary sinus lactate concentration increased to 5.38 mM and simultaneously measured arterial lactate was 3.4 mM. Coronary angiography was then performed, revealing a 90% obstruction in mid-LAD proximal to the previously-deployed stent, with severe stenosis confirmed by intravascular ultrasound (Figure 3). Repeat percutaneous intervention with stent deployment was performed in the LAD.

Figure 3: Assessment of myocardial ischemia and coronary anatomy.

Top panel shows the results of arterial and coronary sinus blood sampling. Bottom includes coronary angiogram revealing significant stenosis in mid left anterior descending artery lesion (left) and with significant luminal obstruction by intravascular ultrasound (right).

Dr. Borlaug: Unlike other organs, the heart consumes lactate. Therefore, under normal conditions, arterial lactate content is greater than that measured in the coronary sinus effluent. In this patient, the lactate gradient reversed during exercise, demonstrating unequivocal metabolic evidence that myocardial ischemia developed during exercise. This does not necessarily mean that the ischemia caused the LBBB, but certainly this increases the priority for revascularization.

Like many patients with HFpEF, resting PAWP was normal in this patient, but during leg elevation and low-level exercise there was a clear evidence of HFpEF, with pathologic elevation in PAWP. Notably, this was observed prior to the onset of LBBB, indicating that LBBB cannot account the entirety of his HF syndrome, though its development clearly explained the acute deterioration and prominent symptoms of dyspnea. The LBBB would also not explain his recent hospitalization for fluid overload.

The height of the V wave in the PAWP tracing varies inversely with operant left atrial (LA) compliance, and may be increased due to a stiff left atrium, or when LA volume increases acutely, shifting the LA to the steep portion of its pressure-volume relationship. This shift may occur because of inadequate LA emptying due to systolic dysfunction, increased anterograde filling through the pulmonary veins (as with exercise or ventricular septal defect), or through retrograde LA filling caused by mitral regurgitation (MR), which was most likely in this case given the known association between LBBB and MR.⁴

Patient presentation (continued): Despite revascularization, symptoms of dyspnea persisted. Cardiac magnetic resonance imaging showed normal LV size, mass, and EF, with no pattern of abnormal enhancement to suggest infiltrative disease or myocarditis.

Dr. Frye: The recent coronary angiography showed progression of disease which was not significant in a prior angiogram performed several months earlier. The challenge for clinicians in the setting of LBBB is to determine if it is a pure electrical phenomenon or associated with ischemia or primary myocardial disease. Severe elevation in V waves in PAWP tracing with exercise suggests MR, which may be ischemic or due to dyssynchrony in setting of LBBB. In this patient multiple mechanisms seem evident, but with the further complexity of dyssynchrony leading to significant MR. This experience emphasizes the importance of seeking evidence of MR in patients with unexplained dyspnea with severe elevations of left sided filling pressures and secondary pulmonary hypertension.

Patient presentation (continued): Exercise testing with echocardiography imaging was performed. LBBB again developed at heart rate of 84bpm, with acute onset severe MR. Incomplete coaptation of mitral valve leaflets developed secondary to abnormal wall motion, with a tethered and restricted posterior mitral leaflet, resulting in torrential MR (Figure 4). Following cessation of exercise as HR reached 54bpm, the QRS width again decreased, MR severity decreased to mild, and symptoms resolved.

Figure 4: Exercise stress echocardiography.

Apical two chamber view with color Doppler showing mild mitral regurgitation at rest (A) increasing to severe mitral regurgitation with 20 Watts exercise (B). Panel C shows zoomed-in view of mitral valve with severe mitral regurgitation at 20 Watts exercise.

Dr. Borlaug: Functional MR as a result of ventricular dyssynchrony improves with cardiac resynchronization therapy in patients with HFrEF,⁴ but less is known about this phenomenon in HFpEF. Since the mitral valve was structurally normal, mitral valve repair or replacement would not be preferred given the risk of cardiac surgery and disruption of the subvalvular apparatus associated with replacement. Implantation of a biventricular pacemaker (CRT-P) would be expected to be most effective.

Dr. Bradley: In this patient with acceleration-dependent LBBB, we reasoned that CRT-P implantation would offer a means to counter the adverse hemodynamic effects of LBBB.

Patient presentation (continued): Following CRT-P, the patient reported markedly improved symptoms, reduction in dyspnea, and no recurrent chest discomfort or dizziness. Repeat stress echocardiogram showed only mild MR at peak exercise, with improvement in exercise capacity from 4.2 METs (79% predicted) prior to CRT-P to 6.3 METs (94% predicted).

DISCUSSION

This case illustrates that even among patients with an established clinical diagnosis of HFpEF, there may be important lesions that cause symptoms that require specialized interventions, outside of the “garden variety” presentation of HFpEF. These pathologies may be of a cardiac nature, as in this patient, or a non-cardiac mechanism that destabilizes an already compromised cardiovascular system, as with anemia, pulmonary infection, or worsening kidney function. The unique aspect of this case was the acute, severe functional MR that developed secondary to rate-related LBBB, which was completely reversed through use of CRT, with substantial improvement in symptoms. There was also anatomic stenosis and metabolic evidence of ischemia that required treatment, but this was ultimately found to not be the cause of LBBB. The correct diagnosis was reached through use of a stepped approach including dynamic exercise testing with both invasive and noninvasive hemodynamic assessments.

The normal closure and competence of the mitral valve requires coordinated motion of mitral leaflets with the annulus, chordae, papillary muscles, posterior left atrial wall and left ventricular free wall.^{4, 5} MR worsens during LBBB due to dyssynchrony-induced delay in contraction of the papillary muscles associated with deformation of mitral valve apparatus, together with prolongation of the isovolumic contraction and relaxation periods. MR was even more poorly tolerated in this case because the LA was stiff owing to chronic HFpEF, and had not dilated as is usually seen with chronic MR. Therefore, each episode of “acute onset” MR led to even greater elevation in V wave amplitude and PAWP than would be expected with chronic MR, where the LA has remodeled to accommodate the regurgitation. CRT has been shown to improve MR in the setting of LBBB in HFrEF by resynchronization of LV dyssynchrony.⁴ The current case shows that it may also be effective in the setting of a patient with HFpEF.

CONCLUSIONS

Most patients presenting with clinical HF and normal EF fall into the category of “garden variety” HFpEF, a syndrome for which there are currently few treatments outside of diuretics. This case illustrates the importance of consideration of additional mechanisms even when the diagnosis of HFpEF is established, including components related to myocardial ischemia, rhythm disturbance, and stress-induced valve insufficiency. Maintaining a high degree of vigilance to search for these additional culprits falling “outside the garden” in HFpEF allows for individualized therapy that can markedly improve clinical status, as evidenced by the present case.