Insights From the History and Physical Examination in HFpEF or HFrEF: Similarities and Differences

Mark H. Drazner, MD, MSC


JACC Heart Fail. 2021;9(5):398-400. 

The history and physical examination is a time-honored cornerstone of medical care. In addition to promoting a bond between patients and clinicians, both through the sharing of personal details during the history and via "laying on the hands" during the physical examination, the information gleaned is critical for medical decision-making. However, as new imaging and biomarker technologies have emerged, some have questioned the ongoing relevance of the history and physical examination. Partially in response to these concerns, we believed it important that the usefulness of the history and physical examination be studied critically as would be done with any other diagnostic test.[1] A number of such studies have been performed in patients with chronic heart failure and reduced ejection fraction (HFrEF), often based on databases from randomized clinical trials that have tested pharmacological therapies, and they demonstrated that clinical congestion ascertained by the history and physical examination was an independent risk factor for adverse clinical outcomes. Consequently, the pathways through which clinical congestion exert deleterious consequences in patients with HFrEF should be delineated.[2]

In contrast to studies of the prognostic usefulness of the history and physical examination in patients with HFrEF, there has been less investigation of its usefulness in patients with heart failure with preserved ejection fraction (HFpEF). Recently, the TOPCAT (Treatment of Preserved Cardiac Function Heart Failure with an Aldosterone Antagonist) study investigators demonstrated that clinical congestion, as assessed serially via elevated jugular venous pressure (JVP), rales, or peripheral edema, and using a time-updated statistical approach, was independently associated with all-cause mortality and the composite endpoint of cardiovascular mortality and heart failure hospitalization.[3] Because TOPCAT was the first large study to assess the prognostic usefulness of the history and physical examination in patients with HFpEF, additional investigation of this question was warranted.

It is in this context that Jering et al.[4] now provide an analysis assessing the prognostic usefulness of the history and physical examination from the PARAGON-HF (Prospective Comparison of ARNI with ARB Global Outcomes in HF with Preserved Ejection Fraction) study, a trial that compared sacubitril/valsartan versus valsartan in patients with left ventricular ejection fractions ≥45% and symptomatic (mostly New York Heart Association functional class II to III) heart failure. The variables for this analysis, in their order of prevalence in the study population, were dyspnea on exertion, fatigue, peripheral edema, orthopnea, elevated JVP, rales, paroxysmal nocturnal dyspnea, dyspnea at rest, and S3 (Table I). When comparing the prevalence of these findings reported in the PARADIGM-HF (Prospective Comparison of ARNI with ACEI to Determine Impact on Global Mortality and Morbidity in Heart Failure) study[5] versus the PARAGON-HF study, the prevalence of elevated JVP and rales were comparable, but edema (38% vs. 14%) was more common in patients with HFpEF and a S3 (9.5% vs. 2%) was more common in those with HFrEF.

The key findings of the current study were: 1) patients with ≥3 of the history and physical findings at baseline versus those with ≤2 were at increased risk of the primary composite endpoint (total heart failure hospitalizations and cardiovascular death), as well as the endpoint of heart failure hospitalization, in rigorous multivariable analysis that also adjusted for natriuretic peptide levels; 2) there was no heterogeneity of the treatment effects of sacubitril/valsartan versus valsartan between those with a higher or lower burden of the history and physical findings, although when assessing individual findings, those with orthopnea versus those without orthopnea seemed to derive greater benefit of angiotensin receptor neprilysin inhibitors (ARNIs); 3) the burden of the history and physical findings increased before subsequent adverse clinical outcomes occurred; and 4) sacubitril/valsartan did not reduce the overall number of these symptoms and signs over time.

What are potential limitations of the current study? First, trained study personnel performed the history and physical examination. It is hard to know what level of expertise such personnel had when assessing the JVP, an important consideration because accuracy of JVP estimation improves with training and experience.[6] A second consideration is the robustness of the finding that patients with orthopnea were more likely to benefit from ARNI versus valsartan, a conclusion based on an interaction p value of 0.04. The possibility that orthopnea may be a particularly valuable finding from the history and physical examination has previously been suggested, being 1 of only 2 parameters (the other was elevated JVP) associated with an elevated pulmonary capillary wedge pressure in patients with HFrEF and advanced heart failure.[7] However, because of the number of history and physical findings tested for heterogeneity of treatment effects in the current study, the play of chance needs to be considered, and prospective testing of this hypothesis is needed before orthopnea is used to decide whether to initiate an ARNI in a patient with HFpEF. Such considerations gain greater importance with the recent Food and Drug Administration approval for sacubitril/valsartan in some patients with HFpEF.

What gaps remain regarding the role of the history and physical examination in patients with HFpEF? In contrast to HFrEF, few studies have compared the correlation of history and physical findings versus invasively measured hemodynamics in patients with HFpEF. This gap is important because it remains uncertain whether the clinical findings from the history and physical examination are reliable markers of elevated ventricular filling pressures in patients with HFpEF. In the current study, N-terminal pro–B-type natriuretic peptide levels were not higher in those with a greater number of the history and physical findings. Natriuretic peptide levels can be influenced by many other factors besides ventricular filling pressures, including body mass index, and those with more history and physical findings did have a higher body mass index. Furthermore, the inclusion of fatigue, which would not be expected to reflect congestion, as 1 of the history and physical findings may have attenuated the association. Nevertheless, many of the other history and physical findings in the current analysis are classic markers of congestion in patients with HFrEF; in the PARADIGM-HF study, natriuretic peptide levels were higher in participants with an increasing number of the history and physical findings studied,[5] all 4 of which were also included in the PARAGON-HF analysis. The lack of association of the history and physical findings with natriuretic peptide levels in the current study adds to the uncertainty as to whether the former reliably reflects increased ventricular filling pressures in patients with HFpEF.

In the PARAGON-HF study, sacubitril/valsartan versus valsartan did improve dyspnea on exertion but did not decrease the overall burden of the history and physical findings over time. In contrast, sacubitril/valsartan versus enalapril did have that latter effect in the PARADIGM-HF trial, which suggested a decongestive effect of ARNI therapy in patients with HFrEF.[2] The basis of this discrepancy between the 2 trials is uncertain. A lack of power does not appear to be the sole explanation because in the PARAGON-HF study, sacubitril/valsartan did not reduce the prevalence of edema over time (odds ratio: 0.96; p = 0.56), although the prevalence of edema at baseline was 2- to 3-fold more common than that in the PARADIGM-HF subjects. Whether the greater efficacy of sacubitril/valsartan in patients with HFrEF than in those with HFpEF explains this discrepancy is uncertain but plausible. Thus, although the study by Jering et al.[4] demonstrated that the history and physical examination provides important prognostic information in patients with HFpEF, as has been established previously in those with HFrEF, it also reminds us that conclusions reached from study of subjects with 1 of these conditions may not apply to the other.