Introduction
Among patients diagnosed as heart failure with preserved left ventricular ejection fraction (HFpEF), atrial fibrillation (AF) is a common finding at presentation and AF is closely associated to known etiologies of HFpEF as hypertension, diabetes mellitus and ageing [
1‐
3]. Therefore, AF is an important confounder of the diagnosis of HFpEF as well as to the diagnostic criteria used to define HFpEF [
4]. Recently, it has been demonstrated that HFpEF patients with AF differs hemodynamically from those with sinus rhythm having higher pulmonary capillary wedge pressure (PCWP), mean pulmonary artery (PA) pressure and poorer exercise capacity [
5]. Chronic or long-standing AF might in some patients lead to significant structural and functional changes with right and left atrial dilatation along with functional mitral or tricuspid valve insufficiency [
6,
7]. Isolated functional tricuspid regurgitation (FTR) following chronic AF (AF-FTR) is characterized by tricuspid annular dilation and severe right atrial enlargement and accounts for 9% of patients with FTR and is associated with age, female gender and preserved left ventricular ejection fraction (LVEF) [
7,
8]. From a clinical aspect these patients resemble patients categorized as heart failure with preserved LVEF (HFpEF) as they present with typical signs of clinical heart failure, enlarged atria, elevated natriuretic peptides and normal LVEF [
9]. However, there is limited clinical and hemodynamics data available in patients with AF-FTR. A better understanding of these patients is wanted due to adverse prognosis and the fact that most patients will only receive diuretic medical therapy until intractable severe right heart failure appear [
10]. As off today isolated tricuspid valve surgery remains relatively rare and the reluctance to perform isolated tricuspid valve (TV) surgery is likely to reflect the reported increased in-hospital mortality [
11]. In recent years, interest in TV pathology have rapidly expanded in response to reported poor clinical outcome and as a consequence of emerging percutaneous transcatheter treatment options in untreated high-risk patients with severe FTR.[
12] However, less focus has been on the contribution of severe AF-FTR on clinical performance including exercise capacity and hemodynamics. The importance of recognizing significant AF-FTR as a heart failure associated condition is crucial since the prevalence of this type of patients is likely to increase significantly in the future due to the expected increase of AF and HFpEF in the general aging population.
The present study aims to investigate the clinical heart failure characteristics, functional capacity and hemodynamics at rest and during exercise in patients with severe AF-FTR and compare these patients with patients with another HFpEF like condition as cardiac amyloid cardiomyopathy (CA).
Discussion
This study examined the functional capacity and hemodynamic responses to exercise measured invasively in patients with chronic severe AF-FTR with preserved LVEF. For comparison, a group of patients with CA and HFpEF like condition were selected. The main findings are as follows: firstly, the patients with AF-FTR had advanced heart failure symptoms and demonstrated a low exercise capacity which was comparable to CA. Secondly, a reduced CO at rest was noted in addition to an abnormal cardiac output reserve response to stress in AF-FTR patients that was similar to the response seen in the CA group. Thirdly, in the AF-FTR group, severe RA enlargement was demonstrated with significantly increased filling pressures at rest as well during peak exercise. Of notice, the left-sided filling pressures were elevated at rest and particular during peak exercise in both patients with AF-FTR and CA. The LV preload characteristics evaluated by LVMTP differed significantly between the patient groups as the AF-FTR patients seemed to have preserved the LV preload properties in contrast to the expected and demonstrated restrictive LV preload properties in the CA patients. In contrast, the relation between RA filling pressures and CO during stress seems to be reduced in AF-FTR patients as compared to CA.
The risk of developing AF increases substantially with age and some will develop chronic AF. That is now increasingly accepted as a risk marker of FTR development [
7,
23,
24]. The prevalence of AF in patients with moderate-severe FTR is reported as high as 9.2% and the presence of chronic AF is in particular related to the FTR severity [
8,
24]. Chronic AF promotes in particular RA enlargement and TA dilation which increases the risk of development of FTR [
25,
26]. The development of significant FTR in chronic AF seems not to be a valvular disease but rather an abnormality that is the result of disease processes that alters the TA, RA, RV size and function consequently leading to geometrical disturbance in the anatomical structures supporting the TV. Severe FTR among AF patients occur mostly in patients with chronic AF rather than paroxystic AF and a higher prevalence of right-sided heart failure has previous been reported in chronic AF which is in accordance with our findings.[
8]. The presence of FTR may be considered as a mediator or barometer of heart failure severity, but can also be a contributor to heart failure itself. In the present study, the AF-FTR patients shared clinical characteristics with patients defined as classical HFpEF condition. The AF-FTR patients consisted of elderly patients with typical comorbidities seen in AF populations such as hypertension, diabetes mellitus in accordance with preserved LVEF and no signs of valve disease. The AF-FTR patients were highly symptomatic with significantly increased NT-pro-BNP despite all patients were preload optimized with diuretics and had stable rate-controlled AF. The exercise capacity was severely impaired with an average peak consumption of 15 ml/kg/min which was comparable to the CA group. The low exercise capacity could be due to an inadequate chronotropic response following treatment with betablockers. However, this seems unlikely as the average peak heart rate during exercise reached 131 beats per minute accounting for 80–85% of the estimated maximal heart rate. The exercise test was performed in a semi-supine position which might have influenced the maximal exercise capacity achieved but the serum lactate was above 5 mmol/l at peak indicating that the patients were under substantial strain. Overall, the exercise capacity and response in AF-FTR was comparable with patients with clinical heart failure due to restrictive amyloid cardiomyopathy with preserved LVEF.
At rest, CO was reduced in both patient groups with a significantly lower CO in the AF-FTR patients which is in accordance with previous findings [
27]. The CO reserve during exercise was impaired to similar degree in both groups with only a two-fold raise in CO. The impairment of resting and exercise CO is in accordance with a previous study in severe FTR but with some difference in baseline patient characteristics [
28]. In addition, the ratio of ΔVO2/ΔCO was severely reduced as 1 ml/min increase in VO2 was only followed by an approximately 3 ml/min increase in CO which is opposed by 1 ml/min VO2 to 6 ml/min CO relationship in normal subjects [
29].
Due to the regurgitation volume among the AF-FTR patients, the forward output is reduced. Therefore, CO assessed by Fick´s principle during the RHC demonstrated reduced CO at rest and at peak exercise in AF-FTR even though the RV myocardial performance assessed by 2D- and 3D echocardiography demonstrated normal values of RVEF, TAPSE and S´. However, the echocardiographic RV systolic parameters might be overestimating the systolic myocardial performance due to a reduced RV afterload as a consequence of a significant tricuspid regurgitant volume. Based on the RVSWI (lower normal: 8 gm-m/m2/beat) and the CO data some degree of impaired RV performance seems to be present due to the TR itself as well as the coexisting increased pulmonary venous pressure that might reduce the RV compliance and contribute to an abnormal RV performance in AF-FTR patients. It is also noteworthy that the RVSWI did not differ from the control patients representing an infiltrative restrictive cardiomyopathy entity that usually also involves the RV to some extent.
Severe bi-atrial enlargement (in particular RA enlargement) was noted along with moderated elevated resting atrial filling pressures in AF-FTR patients which is in accordance with the findings reported by Borlaug et al. [
28], and in patients with and without AF [
5]. With exercise, the mean RAP raised significantly doubling the pressure and significantly exceeding the level of RAP found in the CA group. The PCWP at peak exercise increased further exceeding 30 mmHg in AF-FTR resembling the same degree of response seen in HFpEF patients [
9,
30]. AF-FTR patients share some common clinical and hemodynamic features with the HFpEF population but differ from those in some aspects as the right and left filling pressures at rest are elevated and the CO is reduced at rest, findings that is not usually seen in HFpEF patients [
9]. The LV preload was assessed by LVMTP which differed significantly between the AF-FTR and CA control group as the slope of the rest-to exercise relation curve of LVMTP and CO were shallower in CA patients in comparison to AF-FTR. This finding could be expected due to the well-known restrictive LV pathophysiology seen in amyloid cardiomyopathy patients. In contrast to the findings by Borlaug et al., we found a positive and steep significant relation between LVMTP and CO comparable to normal subjects indicating that the LV preload is less affected in AF-FTR patients. This is even though the left atrial contraction component is absent due to AF in the FTR patients. However, the increased PCWP at rest and at peak might to some degree compensate the absent atrial contraction component and thereby maintain the LV filling in AF-FTR. Subjects with AF-FTR displayed a minor increase in CO despite a greater increase in RAP than CA control patients. In contrast, the relation between the increase of CO to PCWP was nearly similar in the two patient groups.
In the past recent years different transcatheter technologies have emerged as alternative to conventional surgery to high-risk populations with symptomatic significant TR. Recommendations regarding interventions indications and choice of imaging pre-and during procedure has been suggested and enables to predict risk of therapeutic failure [
31‐
33]. The growing clinical importance of secondary isolated TR contributing to heart failure symptoms and significant altered hemodynamics is supported by the present study. However, whether transcatheter TV interventions can improve the abnormal hemodynamics in AF-FTR and translate this into an improvement of symptoms, physical capacity and even prognosis need to be clarified by future investigations.
Some limitations have to be taken into consideration beyond the experience of a single center and the small cohort of patients. First, the AF-FTR patients studied were all patients that were able to perform a semi-supine bicycle test excluding patients with physical disabilities or even just the inability to perform a stress test which is often seen in this patient population with advanced age. However, the patients were selected from the out-patient clinic where they were referred for unexplained dyspnea, leg edema or cardiac murmur evaluation and not from a patient population referred for invasive hemodynamic evaluation or referred for surgery or percutaneous TR intervention. Second, we used a semi-supine bicycle test which might have reduced the true maximal objective workload, but heart rate achieved and serum lactate level indicates that the patients were stressed sufficiently. Third, the difference in age, diabetes, and hypertension between the two groups might have influenced our results. However, there was no difference in LVEF, levels of NT-proBNP and hemodynamics including PCWP between the two groups at rest and PCWP increased significantly to the same extent in both groups indicating that the differences did not severely influenced our results. Fourth, we did not measure frailty score or lung function in our populations, but all the patients were without pulmonary disease and able to use a semi-supine bicycle test which indicated a reasonable performance. There was no difference in the VO2 consumption and the mean pulmonary artery pressure at rest and during peak exercise indicating that there was no severe difference in the lung function influencing the hemodynamics between the groups. The CA group was chosen as a recognized heart failure population requiring diuretic treatment despite LVEF is in the normal range in order to enlighten how symptomatic the population of AF-FTR is.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.