Diagnostic Methods Congenital Heart Disease

Two-dimensional echocardiographic features were correlated with surgical findings in 25 patients with Ebstein's anomaly (age 7 days to 71 years). There was excellent agreement between echocardiographic and surgical findings. Echocardiographic observations of anterior leaflet tethering and restriction of motion, as well as small functional right ventricle, were the strongest noninvasively obtained indicators for valve replacement surgery. Plication/annuloplasty (plastic repair) was perforrned in 17 patients. This group differed from the group needing valve-excision surgery in that the anterior leaflet was elongated, was not tethered, and showed large excursion. These observations support the use of two-dimensional echocardiography as an excellent means for preoperative anatomic assessment of Ebstein's anomaly. This method allowed us to determine which patients were amenable to tricuspid valve excision or plastic repair. A noninvasively derived index of anatomic severity is proposed for easier recognition of patient subgroups. Circulation 68, No. 3, 534-544, 1983. EBSTEIN'S ANOMALY of the tricuspid valve is a rare congenital malformation, which has a wide spectrum of morphologic abnormalities.' Surgery is considered when the patient's clinical status deteriorates. Various surgical techniques, which include excision or plastic repair of the tricuspid valve, have been used to help correct functional and structural abnormalities. ' At our institution, the most common surgical approach is atrialized ventricle plication, right atrial reduction, and tricuspid annuloplasty. '0'"I Valve excision has remained necessary in a small percentage of these patients because of the wide anatomic variability of the tricuspid valve deformity in Ebstein's anomaly. However, we believe that valve replacement in this disease entity is not satisfactory. Heretofore, there were no reliable ways to accurately visualize specific morphologic alterations of the tricuspid valve before surgery and therefore no way to From the Division of Cardiovascular Diseases and Internal Medicine, Division of Pediatric Cardiology, and Section of Thoracic, Cardiovascular, Vascular, and General Surgery, Mayo Clinic and Mayo Foundation, Rochester, MN. Address for correspondence: J. B. Seward, M.D., Mayo Clinic, Rochester, MN 55905. Received Jan. 11, 1983; revision accepted April 28, 1983. 534 determine which patients would need valve excision. With the advent of two-dimensional echocardiography the noninvasive diagnosis of Ebstein's anomaly and complete assessment of its morphologic spectrum became feasible.'2t-8 As part of a larger 41⁄2/2 year study,'8 we used preoperative two-dimensional echocardiography in 25 patients scheduled for surgical repair of Ebstein's anomaly. From these observations and from the experience with patients who were not operated on, we have made a detailed assessment of the echocardiographic anatomy of Ebstein's anomaly, with particular emphasis on the surgical subgroup. We have evaluated the two-dimensional echocardiographic morphologic spectrum of Ebstein's anomaly in the patient needing surgery and have sought two-dimensional echocardiographic criteria that would reliably differentiate patients needing tricuspid valve replacement from those needing plastic repair. Materials and methods Between January 1977 and July 1981 (4'/2 years) there were 41 patients with Ebstein's anomaly who underwent a detailed two-dimensional echocardiographic examination at the Mayo Clinic. Twenty-five of these patients (61%) subsequently underwent surgical exploration of the right ventricle, and sufficient anatomic detail was described to permit comparison with CIRCULATION by gest on A uust 7, 2017 http://ciajournals.org/ D ow nladed from DIAGNOSTIC METHODS-CONGENITAL HEART DISEASE that of the two-dimensional echocardiograms. Three additional patients underwent surgery but are not included in the study because surgical exposure of the tricuspid valve was insufficient to permit comparison with the two-dimensional echocardiograms. The 13 remaining patients who were not operated on are presented as a medical group. Preoperative cardiac catheterization was performed in 15 of 25 (60%) of the surgical group and in two of 13 (15%) of the medical group. The two-dimensional echocardiogram was the primary preoperative diagnostic technique in 11 of 38 (29%) patients. Of the 25 patients in the surgical group (table IA), there were 14 male and 11 female patients; their ages ranged from 7 days to 71 years (mean 19 years). Associated anomalies included atrial septal defect (23 patients), patent foramen ovale (one patient), pulmonary stenosis (three patients), pulmonary atresia (one patient), and ventricular septal defect (one patient). Seven of the 25 patients (28%) underwent tricuspid valve replacement (five patients) or isolated valve excision (two patients). The remaining 17 patients (68%) underwent plastic repair (i.e., plication of the atrialized portion of the right ventricle, reduction in size of the tricuspid anulus by posterior annuloplasty, and excision of a portion of the right atrium). 10, 11 One remaining patient underwent pulmonary valve replacement, closure of an atrial septal defect, and plication of the right atrium; no tricuspid valve surgery was done. Thus there were 24 patients who had either excision or plastic repair of the tricuspid valve. Additional surgical procedures included closure of an atrial septal defect (23 patients), patent foramen ovale (one patient), or ventricular septal defect (one patient), and correction of other minor anomalies (table 1A). The indication for surgery was based on functional disability. All but two patients were in class III or IV (New York Heart Association). Exceptions included one patient with severe cyanosis (patient 23) and one with associated pulmonary atresia (patient 25). Three patients requiring surgery (Nos. 9, 17, and 20) died after attempted plastic repair. Patient 9 died 1 day postoperatively as a result of uncorrectable low cardiac output. Patient 17 died at surgery after coronary bypass and tricuspid annuloplasty. Patient 20 died suddenly of presumed dysrhythmia 4 months after surgery. All survivors of surgery showed improvement in functional capacity. The medical group consisted of 13 patients (ages 10 months to 85 years, mean 24 years). All but two were in functional class I or II (table lB). Echocardiographic examination was performed with a wideangle, phased-array sector scanner (Varian 3000, with a 2.25 or 3.5 MHz transducer) or a mechanical sector scanner (ATL Mark V, with a 3 or 5 MHz transducer). Images were recorded on videotape for real-time and slow-motion playback and our photographs were obtained from stop-action images displayed on a video screen. The examination technique and validation of structures have been previously reported.'9 A complete detailed echocardiographic examination was available for each patient. Short-axis scans were used to assess the right ventricular outflow tract and, to some extent, the posterior tricuspid valve leaflet. However, the most useful view for the assessment of Ebstein's anomaly was the apical four-chamber view (figure 1), which allowed measurement of intracardiac dimensions20 and comparative assessment of chamber dimensions (i.e., atrialized vs functional right ventricle) and of morphologic features (support apparatus and motion) of the septal and anterior tricuspid valve leaflets. Figure 2 shows how we measured various intracardiac dimensions. Tricuspid anulus dimension was measured as the maximum diastolic distance across the right ventricular inlet at the level of the tricuspid valve anulus as identified by echocardiography. Long-axis right ventricular dimension was measured as the distance from the level of the tricuspid anulus to the apex of the right ventricle in diastole. Atrialized right ventricle was measured as the maximum systolic distance from the tricuspid anulus to the leading edge of the tricuspid valve leaflets. Functional right ventricle was measured as the difference in dimension between the right ventricle and the atrialized right ventricle. A ratio of functional right ventricle (RV) to total right ventricular dimension (RV), fRV/RV x 100, was used as an index of anatomic severity. Right atrial dimension was estimated as the maximum systolic distance from the tricuspid anulus to the posterosuperior wall of the right atrium. Displacement of septal tricuspid leaflet was measured in systole as the distance from the tricuspid anulus to the nearest point of septal attachment. For the purpose of comparison, in this large group of patients of various ages and body sizes, values for dimension and distance were indexed to body surface area. With the above data we undertook three comparison studies, each designed to determine the value of two-dimensional echocardiographic features in the prospective assessment of Ebstein's anomaly. Surgical/echocardiographic features. We compared the two-dimensional echocardiographic features of Ebstein's anomaly with features described by direct observation at surgery (25 patients). This comparison was designed to test the sensitivity and diagnostic accuracy of the echocardiographic examination. Sensitivity of diagnosis is expressed as the ratio of true-positive to true-positive plus false-negative. Diagnostic accuracy is expressed as the ratio of true-positive to true-positive plus falsepositive.21 Medical/surgical two-dimensional echocardiographic groups. The two-dimensional echocardiographic morphologic features of Ebstein' s anomaly in 13 medical group patients were compared with the two-dimensional echocardiographic features in 24 patients who underwent tricuspid valve excision or plastic repair. The purpose of this comparison was to identify features that might potentially distinguish patients needing surgery from those not needing surgery. Plastic repair/excision groups. The echocardiographic features of the 17 patients who underwent plastic repair of the tricuspid valve were compared with those in seven patients who required valve excision. Particular attention was paid to morphologic abnormalities of the tricuspid leaflets and support apparatus, to valve mobility, to the size of the functional right ventricle, and to the relative size of intracardiac structures. We hypothesized that the recognition of the spectrum of anatomic severity of these abnormalities could be used to identify a group of patients most likely to require valve replacement. In this retrospective study, it should be possible to recognize the relative importance of individual observations and to develop an index of anatomic severity for an individual patient.