Transesophageal echocardiography (TEE) plays an important role for real-time procedural guidance during surgical smyectomy (SM) for hypertrophic obstructive cardiomyopathy (HOCM). We aimed to compare (1) interventricular septum (IVS) thickness using 2- (2D) and 3-dimensional (3D) intraoperative TEE and preoperative cardiac magnetic resonance (CMR) and (2) mitral valve (MV) leaflet length using 2D, 3D TEE, automatic quantification of mitral valve (AMVQ) and preoperative CMR. We prospectively studied 50 patients with HOCM (age 59 ± 12 years, 44% men) who underwent SM during 2018 to 2019. The maximal basal, mid, and distal anteroseptum (AS) and inferoseptum (IS) were measured by multiplanar 3D reconstruction on TEE and by short-axis imaging on preoperative CMR and classified as mild (≤18 mm), moderate (18 to 25 mm), or severe (≥25 mm) groups based on AS and IS thickness on CMR. MV leaflet lengths were evaluated by preoperative CMR and intraprocedural 2D TEE, zoom 3D TEE, and AMVQ (EchoPAC, General Electric, Wisconsin). There was a moderate correlation between AS and IS thickness on 3D TEE and CMR ( R 2 = 0.46, p <0.01 and R 2 = 0.41, p <0.01, respectively), with 3D TEE showing an average overestimation of 3.8 and 4.7 mm versus CMR. The 3D TEE overestimated 14 patients (56%) with mild thickness as moderate and 5 patients (22%) with moderate thickness as severe. Assuming 3D TEE as the gold standard, the closest correlation for anterior mitral leaflet length was with CMR (average overestimation by CMR of 0.5 mm [root mean square deviation (RMSE%) 17]), intermediate correlation with 2D TEE (average deviation of 0.6 mm [RMSE% 21]) and no correlation with AMVQ (average deviation of 0.7 mm [RMSE% 24]). In conclusion, 3D TEE overestimates IVS thickness versus CMR in patients with HOCM who underwent SM, with greater discrepancy in those with thinner IVS. There are significant differences in MV lengths measured using different imaging techniques.
Surgical myectomy (SM) is the recommended treatment for patients with hypertrophic obstructive cardiomyopathy (HOCM) with symptomatic left ventricular (LV) outflow tract obstruction (LVOTO) despite optimal medical therapy. It is increasingly being recognized that LVOTO is multifactorial and not limited to the hypertrophied interventricular septum. Structural abnormalities of the mitral valve (MV), such as elongated mitral leaflets, abnormalities of papillary muscles, and malformations of the chordae, often contribute to LVOTO, often requiring an alteration in the surgical approach to relieve LVOTO. To plan an optimal surgical approach to relieve LVOTO, precise morphologic assessment is crucial, including measurements of the interventricular septum (IVS) and MV lengths. This precise planning enables optimal and long-term reduction in LVOTO in addition to preventing complications. such as ventricular septal defect and heart block. Therefore, SM is typically recommended in experienced centers where the full spectrum of surgical skillset, including management of concomitant mitral and subvalvular apparatus in addition to SM, is available. Indeed, performance of SM at lower-volume centers has been associated with failed procedures requiring repeat interventions, higher pacemaker implantation and mortality, longer length of stay, and higher costs. Although surface echocardiography is the mainstay for decision-making about referring a patient for surgery, advanced multimodality imaging, especially transesophageal echocardiography (TEE) and cardiac magnetic resonance imaging (CMR), are often used to make crucial septal and MV measurements to help decide the amount of septal debulking and MV repair that needs to be performed to optimally relieve LVOTO. However, previous reports have demonstrated a weak association between septal measurements on echocardiography and CMR. Although CMR has better spatial resolution and considered the gold standard for septal measurements, intraoperative TEE is routinely used for intraoperative evaluation and guidance. In addition, various TEE techniques are used to measure mitral leaflet length and morphology. However, the accuracy of various intraoperative TEE for septal and MV measurements and correlation with CMR in this setting remains unknown. In patients with severely symptomatic HOCM who underwent SM, this study aimed to compare the IVS thickness and MV lengths measured by various intraoperative TEE techniques (including 3D multiplanar reconstruction [MPR]) with preoperative CMR.
Methods
This was a prospective study of adult patients (aged >18 years) with HOCM who underwent SM for symptomatic LVOTO who underwent preoperative CMR and intraoperative TEE. The study protocol was approved by the institutional review board of the Cleveland Clinic. All patients provided written informed consent. We excluded patients who underwent mid or apical myectomy, planned MV repair or replacement for concomitant intrinsic MV disease, other concomitant valvular surgery, coronary artery bypass grafting, and contraindications to either CMR and/or TEE. Relevant clinical characteristics were extracted from electronic medical records including demographics, clinical presentation, medications, past medical history, and laboratory tests.
We used the following measurement techniques to assess IVS thickness and MV leaflet length.
Intraoperative 2D TEE and 3D TEE images were obtained, according to guidelines, using the General Electric (General Electric, Milwaukee, Wisconsin) Vivid I95 platform. The 3D TEE full-volume loops of the LV were obtained and analyzed offline using the GE software package EchoPAC. MPR of the LV septal wall was created to estimate the maximal anteroseptal (AS) and inferoseptal (IS) wall thickness in late diastole. Maximal wall thickness (MWT) was defined as the measurement representing the greatest dimension ( Figure 1 ).
MV leaflet lengths were evaluated using 2D TEE, zoom 3D TEE with MPR, and offline automatic MV quantification (AMVQ) ( Figure 2 ), according to guidelines. , Anterior mitral length was measured from the most distal extent to its insertion into the posterior aortic wall; posterior mitral length was measured as the most distal extent into the basal LV posterior free wall. On 2D TEE, midesophageal long-axis view, the entire length of anterior mitral length, and posterior mitral length were measured at A2/P2 in diastole. On 3D TEE, the measurements were made offline using MPR on a 3D zoom image from the midesophageal long-axis view in diastole. Orthogonal planes were used to accurately identify the A2/P2 scallops. On AMVQ using the General Electric software package EchoPAC, a semiautomated analysis package of MV anatomy at end-systole provided assessment of leaflet lengths plus coaptation height point. We assumed that the most accurate modality to measure the MV length was 3D TEE using MPR.
CMR was performed on a 1.5-T and 3-T scanners (Philips Achieva, Best, The Netherlands), as previously described. , For assessment of LV wall thickness, balanced steady-state free precession images were acquired in short-axis and 2-, 3-, and 4-chamber views. The images were analyzed using cvi42 software (Circle cardiovascular imaging, Calgary, Alberta, Canada). Measurements for the MWT in AS and IS were obtained at end-diastole in the short-axis views. All patients were examined with a CMR within 3 months of the date of planned surgery at our institution. Also, MWT, measured on CMR, was used to classify patients into 3 groups: mild (≤18 mm), moderate (18 to 25 mm), and severe (≥25 mm). We evaluated how these subgroups were classified with 2D and 3D TEE. MV length was measured in the 3-chamber view when the leaflet was fully elongated in mid-diastole.
These measurements, in echocardiography and CMR, were made by an independent cardiologist (AF) separately at different times (at least 1 month apart) to prevent observer bias.
Continuous and categorical variables were expressed as mean ± SD and frequency (percentage), respectively. For comparison of continuous parameters, paired Student’s t test: 1-way or 2-way analyses of variance were used, as appropriate. A contingency table was used to compare categorical variables expressed as proportions. All tests are 2-tailed and the p value considered significant when <0.05. When required, we provided p values after Bonferroni correction. Relations between mitral leaflet length and other continuous variables were assessed by linear regression. The agreement between modalities data were evaluated by the calculating root mean square error (RMSE) and absolute mean error (ME) as follows:
RMSE=∑(Mod−Ref)2n
ME=∑(Mod−Ref)n
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
