Age- and gender-specific reference values for cardiac chamber geometry and function using three-dimensional echocardiography

Background. Three-dimensional echocardiography (3DE) enables a comprehensive, accurate and reproducible quantification of cardiac chamber size and function without any geometric assumption about their shape. Superior accuracy and reproducibility of 3DE over stabdard two-dimensional (2DE) approach for cardiac chamber volume measurements in comparison to cardiac magnetic resonance (CMR) has been well documented in a number of studies. Both the European Association of Cardiovascular Imaging and the American Society of Echocardiography recommend 3DE, rather than 2DE, for routine clinical assessment of cardiac chamber volumes. However, both Societes also acknowledge that the application of 3DE into routine clinical practice has been hindered by the limited availability of reference values, and particularly the lack of gender- and anthropometric-based analysis. Therefore, identification of reference values for cardiac chamber size, geometry and function has become a prerequisite for the routine clinical application of quantitative 3DE. Research Project Single-centre, prospective, observational cohort study aimed to: (i). comprehensively analyze the four cardiac chamber geometry and function using state-of-the-art 3DE equipment in a large cohort of healthy volunteers; (ii). assess the effects of age, body size and gender on these parameters; and (iii). compare the values measured using 3DE with those obtained by conventional echocardiography in the same subjects and with other cohorts of healthy subjects from published 3DE studies. Methods. 263 healthy volunteers (43±14 years, range 18-75; 58% women) whose data sets have been acquired from October 2011 to July 2013 using a commercially available 3D echo scanner (Vivid E9, GE Vingmed, Horten, NO) equipped with 4V matrix array probe. Data sets were analyzed with different commercially available (EchoPac BT 12, GEVingmed Horten, NO; 4D RV function, TomTec Imaging system, Unterschleissheim, D ) and prototype (EchoPac BT 13, GEVingmed Horten, NO; 4D LA Tomtec Imaging systems, Unterschleissheim, D) analysis softwares. The study was approved by the University of Padua Ethics Committee (protocol # 2380 P approved on 06/10/2011) and signed informed consent has been obtained in all volunteers before the screening for eligibility in the study. Results Study #1: Analysis of left ventricular (LV) size, geometry and function. In 226 consecutive healthy volunteers (125 women, aged 18-76 years), we performed a comprehensive 3DE analysis of LV parameters and compared them with values obtained by conventional echocardiography. Upper reference values (mean+2 standard deviatons) for 3D LV end-diastolic (EDV) and end-systolic (ESV) volumes were 85 ml/m2 and 34 ml/m2 in men, and 72 ml/m2 and 28 ml/m2 in women, respectively. Indexing LV volumes by body surface area did not eliminate gender differences. Lower reference values (mean-2 standard deviations) for ejection fraction (EF) were 54% in men and 57% in women, while for stroke volume (SV) were 25 ml/m2 and 24 ml/m2, respectively. Upper reference values for LV mass were 97 g/m2 in men and 90 g/m2 in women, while for end-diastolic sphericity index were 0.49 and 0.48, respectively. Significant age-dependency of LV parameters was identified and reported across age groups. 3DE LV volumes were larger, EF was similar, SV and mass were significantly smaller in comparison with the corresponding values obtained by conventional echocardiography. Study #2: Analysis of right ventricular (RV) size and function. RV volumes, SV and EF were measured by 3DE in 540 healthy adult volunteers, prospectively enrolled, evenly distributed across age and gender. The relation of age, gender and body size parameters with RV volumes and EF were investigated using bivariate and multiple linear regressions. Analysis was feasible in 507 (94%) subjects (260 women, age 45±16 years, range 18-90). Age, gender, height and weight significantly influenced RV volumes and EF. Gender effect was significant (p<0.01), with RV volumes larger and EF smaller in men than in women. Older age was associated with smaller volumes (EDV, -5 ml/decade; ESV, -3 ml/decade; SV, -2 ml/decade), and higher EF (+1%/decade). Inclusion of body size parameters in the statistical models resulted in improved overall explained variance for volumes (EDV, R2=0.43; ESV, R2=0.35; SV, R2=0.30), while EF was unaffected. Ratiometric and allometric indexing for age, gender and body size resulted in no significant residual correlation between RV geometry measures and height or weight. Study #3: Analysis left atrial size and function. 244 healthy volunteers (43±14 years, range 18-75; 58% women) underwent 3DE and 2DE to measure maximal (Vmax), minimal (Vmin) and preA (VpreA) LA volumes to calculate total, passive and active LA emptying volumes (TotEV, PassEV, ActEV) and fractions (TotEmptFr, PassEmptFr, ActEemptFr). Feasibility of 3DE and 2DE LA volumes was 91% and 96% (p=0.59 ). 3DE LA volumes were larger than 2DE ones (Vmax: 48±11 ml vs. 43±11 ml; Vmin: 18±5 vs. 14±6, respectively, p<0.001). LA TotEmptFr (61±6% vs. 68±9%) and ActEmptFr (30±7% vs. 47±10%) were lower by 3DE than 2DE (p<0.001), whereas PassEmptFr (44±10% vs. 41±11%) was higher by 3DE than 2DE (p= 0.002). 3DE LA volumes indexed by body surface area were similar in both genders and increased with ageing (p=0.002). Study #4: Analysis of right atrial (RA) size and function. 200 healthy volunteers (43±15 years; 44% men) underwent 2DE and 3DE to measure maximal (Vmax), minimal (Vmin) and preA (VpreA) volumes to derive total (TotEV), passive (PassEV) and true (TrueEV) emptying volumes and emptying fractions (TotEmptFr, PassEmptFr, TrueEmptFr). 3DE volumes (Vmax, 52±15 ml vs 41±14 ml, p<0.0001), EVs (TotSV, 33±10 ml vs. 24±9 ml, p<0.0001) and EmptFrs (TotEmptFr, 63±9% vs. 58±9%, p<0.0001) were larger than 2DE ones. Indexed 3D RA volumes were significantly larger in men than in women. Aging was associated with a significant decrease in passive RA function (PassEV, r= -0.26; PassEmptFr, r= -0.38; all p<0.0001) and an increase in active RA function (TrueEV, r= 0.25; p<0.0001; and TrueEmptFr, r= 0.15; p= 0.035) in order to maintain TotEV (r= -0.14, p= 0.05). Conclusions The present research project provides a comprehensive quantitative analysis of the four cardiac chamber geometry and function using 3DE in a relatively large cohort of Caucasian healthy volunteers with a wide age range. The main results can be summarized as follows: (i). Cardiac chamber quantification with 3DE is feasible and reproducible; (ii) Reference values for cardiac chamber size and function by 3DE were found to be significantly different from those obtained with conventional echocardiography, highlighting the importance of applying method-specific reference values for a reliable identification of remodeling and/or dysfunction of cardiac chambers; (iii). Cardiac chamber parameters measured by 3DE showed excellent reproducibility, and were more robust than 2DE indices at repeated measurements; (iii). Most parameters describing cardiac chamber size should be defined according to age and gender, since indexing them only for BSA does not account for all the physiologic variations in geometry and function. Availability of reference values and age- and gender-specific cut-off values should facilitate the implementation of 3DE to identify cardiac chamber remodelling and dysfunction in both clinical routine and research.