Author + information
- Published online March 5, 2018.
- Liam Ring, MD∗ (, )
- David P. Dutka, MD,
- James Boyd, BSE,
- Karen Parker, BTECH, BSE, MSCT,
- Olaf Wendler, PhD,
- Mark J. Monaghan, PhD and
- Bushra S. Rana, MD
- ↵∗Department of Medicine, University of Cambridge, Addenbrookes Hospital, Hills Road, Cambridge, CB2 0QQ, United Kingdom
Mitral regurgitation is the most common valvular lesion in Europe, frequently assessed with transesophageal and transthoracic echocardiography (TEE/TTE). The spatial resolution of 3-dimensional (3D) TEE is excellent and facilitates measurements similar to those made under direct visualization of the valve (1). A range of reference values for the normal mitral valve annulus (MVA) are essential for clinical practice; whereas normal reference ranges have been established using 3D TTE (2), there is minimal data in the published reports for 3D TEE.
Consecutive patients undergoing TEE between August 2009 and June 2013 were screened. Criteria for inclusion were the following: sinus rhythm; normal left and right ventricular (LV/RV) size and function; normal LV mass; normal left atrial (LA) volume (all defined according to American guidelines ); and no greater than mild mitral, aortic, or tricuspid regurgitation with no valvular stenosis. We excluded patients with elevated pulmonary pressures, coronary disease, or hypertension. All TTE data was obtained from studies completed immediately prior to TEE. The institutional review board of Papworth Hospital approved the analysis.
Images were obtained using an iE33 ultrasound system (Philips, Andover, Massachusetts) with S5-3 and X7-2t transducers and analyzed using commercially available software. TEE studies were completed under conscious sedation, with care taken to ensure blood pressure was maintained throughout. The 3D datasets included the entire annulus and were optimized for depth, sector width, and gain. Frame rates were ≥10 frames/s. The end-systolic frame was chosen for analysis and data obtained as previously described (1). Annular and leaflet parameters were recorded: MVA area; anteroposterior (AP) and intercommissural (IC) diameters; annulus height (AH), annulus height to intercommissural ratio (AHICR, defined as AH/IC); tenting height (the maximal depth of the leaflets below the annular plane), tenting volume, and leaflet surface area. Values were indexed for body surface area (BSA).
Results are expressed as mean ± SD for normally distributed data or median (interquartile range) for non-normally distributed parameters. To determine reference intervals for mitral valve annular parameters, which will be of use in clinical practice, upper and lower reference limits for each parameter were estimated. For normally distributed parameters, cutoffs were estimated as a function of the SD (i.e., mean + 1.96 × SD for the upper reference limit, and mean –1.96 × SD for the lower reference limit). For non-normally distributed parameters, we used square-root transform to normalize the distribution before applying the same methodology. We then converted the values back to yield appropriate upper and lower reference limits. Independent associations of the mitral annular area were determined by entering: sex, age, BSA, indexed LV systolic and diastolic volumes, indexed LA volume, indexed LV mass, ejection fraction, and RV size into a multivariate model as independent variables.
Indexed volumes and mass were used in the multivariate model to exclude the influence of BSA on these parameters. This allows an understanding of how inherent variation in LA and LV volumes (irrespective of a patient’s BSA) relate to the MVA area.
During the study period, 1,252 patients underwent TEE. Of those, 103 patients fulfilled our inclusion criteria, all of whom underwent examination to assess for a cardiac source of embolus or sepsis and were ultimately found to have no significant cardiac lesions.
For results see Table 1. Mitral annular parameters with the exception of AH, AHICR, tenting height, and tenting volume were larger in men, and all parameters were similar after indexing for BSA. The indexed MVA area for the entire cohort was 5.00 ± 0.95 cm2/m2, with an upper reference limit of 6.86 cm2/m2. Independent associates of MVA area were as follow: BSA (beta = 0.513; p < 0.001); indexed LA volume (beta = 0.287; p = 0.001); and indexed LV systolic volume (beta = 0.204; p = 0.040). Sex, age, indexed LV mass, indexed LV diastolic volume, ejection fraction, and RV size were not independently associated with MVA area.
Intraobserver variability (coefficient of variation and 95% limits of agreement calculated using the Bland-Altman method): area: 3.3% ± 0.60 cm2; AP diameter: 2.3% ± 1.3 mm; IC diameter: 2.1% ± 1.6 mm; AH: 4.4% ± 0.6 mm; AHICR: 5.7% ± 1.9%; tenting height: 4.7% ± 0.4 mm; tenting volume: 6.6% ± 0.2 ml; leaflet area: 3.8% ± 0.8 cm2. Interobserver variability: area: 5.0% ± 0.91 cm2; AP diameter: 4.9% ± 2.9 mm; IC diameter: 4.0% ± 3.0 mm; AH: 9.8% ± 1.2 mm; AHICR: 8.2% ± 2.8%; tenting height: 6.3% ± 0.6 mm; tenting volume: 10.0% ± 0.3 ml; leaflet area: 7.7% ± 1.5 cm2.
The present study has defined reference values for the normal MVA in humans using 3D TEE, which will be of use when assessing the mitral valve. Key associates of the MVA area were BSA and indexed LA and LV systolic volumes, which is consistent with previous studies undertaken with TTE (2).
Please note: Dr. Dutka has received unrestricted research support from Sorin and Merck Sharp & Dohme; and grants from the National Institute for Health Research, Medical Research Council, and British Heart Foundation. Dr. Monaghan is on the Speakers Bureau of and receives research support from Philips. Dr. Rana is a proctor for Boston Scientific Corporation. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- 2018 American College of Cardiology Foundation