Current Limitations of Invasive Exercise Hemodynamics for the Diagnosis of Heart Failure With Preserved Ejection Fraction

Abstract

Background: Exercise hemodynamics can differentiate heart failure with preserved ejection fraction (HFpEF) from noncardiac dyspnea. However, respiratory pressure swings may impact hemodynamic measurements, potentially leading to misdiagnosis of HFpEF. Moreover, threshold values for abnormal hemodynamic response indicative of HFpEF are not universally accepted. Thus, we sought to evaluate the impact of respiratory pressure swings on hemodynamic data interpretation as well as the concordance among 3 proposed exercise hemodynamic criteria for HFpEF: (1) end-expiratory pulmonary artery wedge pressure (PAWPexp) ≥25 mm Hg; (2) PAWPexp/cardiac output slope >2 mm Hg/L per minute; and (3) respiratory-averaged (avg) mean pulmonary artery pressure >30 mm Hg, total pulmonary resistanceavg>3 WU, PAWPavg≥20 mm Hg. Methods: Fifty-seven patients with unexplained dyspnea (70% women, 70±9 years) underwent exercise cardiac catheterization. The difference between end-expiratory and averaged hemodynamic values, as well as the concordance among the 3 hemodynamic definitions of HFpEF, were assessed. Results: End-expiratory hemodynamics measurements were higher than values averaged across the respiratory cycle. During exercise, a larger proportion of patients exceeded the threshold of 25 mm Hg for PAWPexprather than for PAWPavg(70% versus 53%, P<0.01). The concordance of 3/3 HFpEF exercise hemodynamic criteria was recorded in 70% of patients. PAWPexp/cardiac output slope identified HFpEF more frequently than the other 2 criteria (81% versus 64% to 69%), incorporating over 97% of abnormal responses to the latter. Patients with 3/3 positive criteria had worse clinical, gas-exchange, and hemodynamic profiles. Conclusions: Respiratory pressure swings impact on the exercise hemodynamic definitions of HFpEF that provide discordant results in 30% of patients. Equivocal diagnoses of HFpEF might be limited by adopting the most sensitive and inclusive criterion alone (ie, PAWPexp/cardiac output slope).