INTRODUCTION
The assessment of left ventricular (LV) diastolic function is clinically relevant since abnormalities in LV filling have been associated with individuals at higher risk for hospitalisation and mortality and are the main LV functional abnormality among patients with heart failure with preserved ejection fraction (HFpEF). The assessment of left ventricular (LV) diastolic function is clinically relevant since abnormalities during LV filling phase have been associated with individuals at higher risk for hospitalisation and mortality and are the main LV functional abnormality among patients with heart failure with preserved ejection fraction (HFpEF). Echocardiography plays the central role in proposed diagnostic algorithms; however, there is no single parameter that can distinguish among different stages of diastolic dysfunction in sinus rhythm using conventional echocardiography. Atrial fibrillation (AF) can be associated with diastolic dysfunction. In contrast with patients in sinus rhythm, due the loss of organised atrial activity and the consequent loss of the two-peaked pattern in diastole, most of the parameters used by conventional echocardiography are of little use in AF patients.
Vector flow mapping (VFM) is a novel technology that uses speckle tracking and colour Doppler to visualise intraventricular fluid dynamics. The primary advantage of this technique is the ability to provide additional information regarding velocities perpendicular to the echocardiographic Doppler beam, facilitating the quantification of blood flow, without angle dependency. We believed that VFM technology can be used to differentiate normal from failing hearts with preserved ejection fraction (EF) by evaluating early diastole alone.
We aimed to determine whether VFM might be of clinical use during diastolic functional assessments in situations without late diastole, such as during atrial fibrillation.
AIMS
In our study we hypothesised that the three principle VFM parameters; 1) intraventricular velocity gradient, 2) normalised intraventricular pressure gradient, and 3) vortex parameters (sucking pressure and vorticity):
I) Would be closely associated with conventional echocardiographic parameters used to assess diastolic function;
II) Could discriminate between groups with normal diastolic function and groups with mild and advanced diastolic dysfunction (DD) in the sinus rhythm, through the assessment of early diastole only (E wave), with good intra- and interobserver reliability
III) Could discriminate between HFpEF and non-HFpEF patients in AF and control group in sinus rhythm with normal diastolic function.
METHODS
In the first study we included consecutive patients with normal LV EFs who were referred to an echocardiography laboratory for the evaluation of cardiac function. Patients were divided into three groups, based on diastolic function: i) mild DD (impaired relaxation pattern); ii) advanced DD (pseudo-normal or restrictive pattern); and iii) control group (normal diastolic function). We then performed a VFM analysis, examining previously selected principal parameters (intraventricular velocity gradient, normalised intraventricular pressure gradient, vortex sucking pressure, and vorticity), which we believed could be used to describe diastolic function by assessing early diastole alone.
In the second study we tested the same VFM parameters in patients with AF and normal LV EFs. Patients with AF were later divided into two groups: i) AF no-HFpEF, consisting of patients who presented at an annual cardiologic check-up in an out-patient clinic, with no history of heart failure symptoms within the last six months and were considered stable and symptom-free; and ii) AF HFpEF, consisting of patients admitted due to heart failure with preserved EF. A third group consisted of healthy controls in sinus rhythm.
RESULTS
We included 184 consecutive patients of which 124 were in sinus rhythm and 60 in AF. In the first study we tested VFM parameters in 121 patients in sinus rhythm, including 38 with mild DD, 26 with advanced DD, and 57 controls. The results were following: 1) Intraventricular velocities and the calculated intraventricular gradient (GrIV) could discriminate among the three groups, with the highest GrIV values detected in the advanced DD group (13.6 +/- 5.0 vs. 6.8 +/- 2.5 vs. 5.3 +/- 1.9/s, p < 0.001). GrIV values strongly correlated with conventional indices of elevated LV filling pressure, such as E/e’ (r =0.751, p < 0.001). Moreover, the ratio between GrIV and mitral annulus early diastolic velocity (GrIV/e’) was identified as the strongest single predictor for the grade of DD (pseudo-R2 = 0.704, p < 0.001) and was strongly associated with LV filling pressure, 2) during the acceleration time of early diastole, the normalised pressure gradient (nPG) in the left ventricle could distinguish among all three groups (96.35 +/- 61.78 vs. 38.86 +/- 27.95 vs. 20.82 +/- 12.42 mmHg, p < 0.001). During the deceleration of early diastole, the inversion of the gradients was observed, where advanced DD had significantly more negative #nPGLV values compared with the other observed groups (-119.75 +/- 79.23 vs. -36.11 +/- 25.41 in mild DD vs. -60.34 +/- 39.29 mmHg in control group, p < 0.001) and 3) vorticity and vortex sucking pressure were significantly different among groups, with the mild DD group presenting the lowest values and the advanced DD group presenting the highest values.
In the second study, we had tested the same VFM parameters in 60 patients with AF, including 29 with no symptoms of heart failure (AF No-HFpEF) and 31 with symptoms of heart failure (AF HFpEF) and compared them to 60 controls in sinus rhythm. We found that: 1) GrIV was significantly higher in the AF HFpEF group compared with the other two groups (11.9 +/- 8.2 vs. 7.8 +/- 4.3 and 5.4 +/- 1.9/s, p < 0.001), and the ratio between GrIV and mitral annulus early diastolic velocity (GrIV/e’) had a power to discriminate among groups, with the highest values in the AF HFpEF group, followed by the AF No-HFpEF and Control groups (2.0 +/- 1.2 vs 0.9 +/- 0.4 vs. 0.5 +/- 0.2/s, p < 0.001). 2) During the acceleration time in early diastole, patients with AF displayed significantly higher normalised pressure gradients compared with those in the control group; however, no significant difference was identified between the No-HFpEF and HFpEF groups (50.7 +/- 32.7 vs. 72.0 +/- 67.2, p = 0.164). During the deceleration time, the inversion of gradients was observed; however, differences among the groups were not significant, and 3) parameters reflecting vortex strength suggested AF HFpEF group having the most negative sucking pressure and the highest vorticity, however only statistical trend was observed (p = 0.066).
CONCLUSIONS
In our studies we demonstrated that VFM technology can be used to visualise inflow fluid dynamics and to observe and quantify less well-known fluid phenomena during the cardiac cycle in sinus rhythm and AF. First study showed that GrIV was closely associated with conventional echocardiographic indices of elevated left ventricular filling pressure and were able to discriminate between groups with normal diastolic function and patients with mild and advanced DD in sinus rhythm. Particularly, the GrIV/e’ parameter has the potential to be used as a novel DD marker. In contrast, only the normalised intraventricular pressure gradient during the acceleration time of early diastole was able to discriminate among the groups with different stages of DD, whereas the vortex strength parameters appeared to be unaffected by DD. In the second study when tested in patients with AF with normal EF, only GrIV/e’ parameter demonstrated the potential to discriminate between those with and without clinical evidence of heart failure and control group in sinus rhythm, suggesting a potential association between AF and LV diastolic function.
|