Echocardiographic Imaging of Stentless Aortic Valve Prostheses

Leo H.B. Baur, M.D., PH.D., Kathinka Peels, M.D., Jerry Braun, M.D., Arie-Pieter Kappetein, M.D., Marianne Bootsma, M.D., PH.D., Arianne Van Der Ploeg, M.D., Alard Sieders, M.D., Mark Hazekamp, M.D., PH.D., Ernst E. Van Der Wall, M.D., PH.D., and Hans A. Huysmans M.D., PH.D. Department of Cardiology, Atrium Medical Center; Departments of Cardiology and Thoracic Surgery, Leiden University Medical Center; Department of Cardiology, Catherina Hospital, Eindhoven, The Netherlands Homografts and stentless xenografts are increasingly used in aortic valve surgery. Echocardiography technicians and cardiologists have to know what they will nd when performing an echo-Doppler examination in patients who received a stentless valve. We therefore evaluated echocardiographic images of 74 patients who received a Freestyle stentless bioprosthesis with three techniques and a follow-up of 2 years in two high-volume hospitals. Of the patients studied, 81% were operated using the subcoronary technique, 12% using the root-inclusion technique, and 7% using the full-root technique. Results: Transvalvular gradients across the stentless valves were low: 8.0 mmHg when implanted with the subcoronary technique, 8.2 6 5.1 mmHg using the root-inclusion technique, and 6.5 mmHg using the full-root technique. Trivial aortic insuf ciency (grade 1) was observed in 10.7% of the patients (8.9% for the subcoronary technique, 13% for the root-inclusion technique, and 0% for the full-root technique). When the bioprosthesis was implanted using the subcoronary technique or the root-inclusion technique, the prosthesis was placed inside the recipient aortic root. Using these techniques, a lumen between the double layer of the xenograft and the aortic wall could be observed. With the root-replacement technique, the porcine root became the most proximal part of the ascending aorta. As the native aortic wall was removed, in most cases, no double lumen could be observed with imaging of the ascending aorta. (ECHOCARDIOGRAPHY, Volume 17, October 2000) stentless valves, echocardiography, aortic valve disease, cardiac ultrasound During the last decade, renewed interest has been developed in stentless valves for treatment of aortic valve disease. Initiated by the paper of Ross et al.,1 aortic valve replacement with homografts has become the treatment of choice for young patients and patients with endocarditis. However, the limited availability of donor hearts restricts the use of these valves. Therefore, stentless xenografts have been reintroduced with new preservation2 and antimineralization techniques3 in order to copy the excellent hemodynamics of homografts and add immediate availability of all valve sizes. Stentless valves show extremely good ow characteristics in vitro,4 in animal studies,5 and in clinical evaluation in humans.6 Durability has been reported to be better than in stented xenografts,7 accompanied by low gradients and only trivial aortic valve6 regurgitation. These features, combined with the wide availability, make the homograft the valve of choice in many institutions and the stentless aortic bioprosthesis a fair alternative for aortic valve replacement. Three surgical techniques can be used to implant this type of bioprosthesis: orthotopic, subcoronary valve replacement; implantation with the root-inclusion technique; and full-root replacement. Postoperative evaluation of these valves will always be performed by M-mode, two-dimensional (2-D), and Doppler echocardiography. Early hemodynamics of this valve have been reported.8 Echocardiographers who want to perform follow-up studies have to be aware of the typical M-mode, 2-D, and Doppler echocardiographic characteristics that can be observed with each implanted valve technique. Therefore, the present study was undertaken to register the speci c echocardio625

Address for correspondence and reprint requests: Leo H.B. Baur, M.D., Atrium Medical Center, Department of Cardiology, Henri Dunantstreet 5, PO Box 4446, 6401 CX Heerlen, The Netherlands. Fax: 045-57-66133; E-mail: lhbaur@yahoo.com

Vol. 17, No. 7, 2000

ECHOCARDIOGRAPHY: A Jrnl. of CV Ultrasound & Allied Tech.

BAUR, ET AL.

graphic ndings in patients who received Freestyle (Medtronic, Inc., Minneapolis, MN, USA) stentless aortic xenografts. Study Patients Seventy-four consecutive patients who received a Freestyle aortic bioprosthesis at the Leiden University Medical Centre, Leiden, the Netherlands, and the Catharina Hospital, Eindhoven, the Netherlands, were evaluated with M-mode, 2-D, and Doppler echocardiography within 4 weeks after operation, 3 to 6 months after operation, and each year during a follow-up period of 2 years. Implantation of the bioprosthesis was performed from June 1993 until September 1997. The indication for operation was aortic stenosis in 59% of the patients, aortic insuf ciency in 34% of the patients, and combined aortic valve disease in 7% of the patients. Valve Characteristics The Freestyle bioprosthesis is a porcine aortic root preserved with gluteraldehyde using physiologic xation (root-pressure xation with zero pressure differential across the valve lea ets). The bioprosthesis is treated with the AOA process to reduce the potential for lea et calci cation. A thin single layer of Dacron fabric covers the porcine myocardium at the right as well as the noncoronary sinus and the in ow edge of the bioprosthesis. Three surgical techniques can be used to implant the bioprosthesis. For orthotopic, subcoronary valve replacement, the two coronary sinuses of the prosthesis are excised. If the root-inclusion technique is used, buttonholes in the two coronary sinuses of the bioprosthesis are made. The prosthesis is then placed within the patients aorta. If the full-root technique is used, the patients aortic root is excised and replaced by the bioprosthesis. The native two coronary ostia are reimplanted in the bioprosthesis. In the current study group, the subcoronary technique was used in 81% of patients, the root-inclusion technique was used in 12% of patients, and the full-root technique was used in 7% of patients. Echocardiography and Doppler Methodology All echocardiograms were made according to guidelines proposed by the American Society of Echocardiography. Echocardiographic images were stored on videotape and analyzed off line.
626

For optimal image, quality patients were examined in left lateral position and recordings were made at complete expiration. Examination included a M-mode tracing, a 2-D echocardiogram in apical four-chamber view, a pulsedDoppler recording of the left ventricular outow velocities, and a continuous-wave Doppler recording of the aortic valve velocities. In addition, a color- ow Doppler image was performed from the parasternal long-axis view and the apical view. The highest peak aortic ow velocity across the prosthesis was measured with continuous-wave Doppler technique. Cardiac cycles with the highest peak velocities were selected for calculations. Doppler measurements of at least three cardiac cycles were averaged. Assessment of aortic regurgitation was done with color- ow imaging and continuous- ow Doppler imaging. Grading of aortic insuf ciency was done according to the criteria used by Perry et al.9 Grading of False Lumen Two independent observers analyzed both parasternal 2-D images and M-mode tracings of the aortic root for the presence or absence of a lumen between the prosthetic valve and the aortic root. If the lumen was more than 2 mm, it was considered to be present. In case of disagreement between the two observers, the images were reviewed by a third observer, who gave the nal decision. Statistical Analysis Statistical analysis was performed using SPSS statistical software (SPSS Inc., Chicago, IL, USA). Two-way analysis of variance was used to evaluate differences in measurements over time. In all statistical analyses, a P value , 0.05 was considered signi cant. The results are reported as the mean value and standard deviation. Results Doppler Measurements Adequate recordings of the aortic jet velocity through the bioprosthesis were obtained in all patients. The values for the mean gradients early after operation, after 3 to 6 months, after 1 year, and after 2 years are shown in Table I and Figure 1. The mean gradient across the bioprosthesis was 8.0 6 5.5 mmHg early after operation and decreased to 4.6 6 3.0 mmHg after 2 years (P , 0.01).
Vol. 17, No. 7, 2000

ECHOCARDIOGRAPHY: A Jrnl. of CV Ultrasound & Allied Tech.

IMAGING OF STENTLESS AORTIC VALVE PROSTHESES

TABLE I Echocardiographic Imaging During Follow-Up Mean Gradient (mmHg) Subcoronary Technique Discharge 3-6 months postoperative 1 year postoperative 2 years postoperative Root-Inclusion Technique Discharge 3-6 months postoperative 1 year postoperative 2 years postoperative Full-Root Technique Discharge 3-6 months postoperative 1 year postoperative 2 years postoperative 8.0 6 7.7 6 5.4 6 5.4 6 8.2 6 4.7 6 5.1 6 4.0 6 6.5 6 2.4 6 3.5 6 4.1 6 5.0 3.8 3.8 3.7 5.2 3.0 3.6 3.1 5.8 1.2 2.1 4.2 Lumen Visible with M-Mode (% of Pts) 53% 29% 27% 12.5% 44% 25% 0 0 0 0 0 0 Lumen Visible with 2D Echocardiography (% of Pts) 63% 51% 42% 56% 33% 25% 17% 25% 0 0 0 0

Early after operation, an aortic insuf ciency grade I was present in 21 (12%) of the patients. In two patients (1.1%) it was grade 2. These patients had a paravalvular leakage. In one patient, a dehiscence of a suture was present due to endocarditis. After 1 year, the percentage of patients with aortic insuf ciency grade 1 was 14% (n 5 16), while grade 2 was present in two patients (1.7%, paravalvular). In one patient, suture dehiscence was present without insuf ciency. The patient was reoperated. After 2 years, four patients (12%) had aortic incompetence grade 1 and none had grade 2. Presence of a Lumen in the Aortic Root If the subcoronary technique was used, an early paraprosthetic lumen (that is within 4

weeks after surgery) could be observed in 53% of the patients with M-mode and 63% of the patients with 2-D echocardiography. This lumen was usually present at the noncoronary cusp. A typical 2-D image shortly after operation is shown in Figure 2. A small lumen can be ob-

Figure 1. The echocardiographically measured mean gradients of the Freestyle stentless bioprosthesis during a follow-up period of 2 years. One can see an initially somewhat higher gradient, which decreases during follow-up due to the disappearance of the paravalvular lumen.

Figure 2. Typical 2-D echocardiographic image (transesophageal scan) of a stentless valve implanted with the subcoronary technique immediately after operation. Note the paraprosthetic lumen (arrow) at the site of the noncoronary cusp.

Vol. 17, No. 7, 2000

ECHOCARDIOGRAPHY: A Jrnl. of CV Ultrasound & Allied Tech.

627

BAUR, ET AL.

served next to the aortic sinuses. During follow-up, this lumen could still be observed with 2-D echocardiography in 51% of the patients after 3 to 6 months, in 41% of the patients after 1 year, and in 44% of the patients after 2 years. Using M-mode, the lumen could be detected in 28% of the patients after 3 to 6 months, in 27% of the patients after 1 year, and in 12% of the patients after 2 years. The disappearance of the lumen was accompanied by a small decrease of transvalvular gradients from 8.0 6 5.0 mmHg to 5.2 6 3.4 mmHg. Shortly after operation with the root-inclusion technique, a lumen between the aortic root and the bioprosthesis could be observed in 44% of the patients with M-mode and 33% of the patients with 2-D echocardiography. For the most part, the lumen was more extensive than with the subcoronary technique and extended to the right and left coronary cusp. With 2-D echocardiography, the lumen could still be observed in 25% of the patients within 3 to 6 months, in 17% of the patients after 1 year, and in 33% of the patients after 2 years. In addition, the transvalvular gradients decreased with the disappearance of the lumen when using this technique. If the full-root technique was used, none of the patients showed a lumen between the aortic root and the bioprosthesis. At times, a thickened aortic wall could be observed either on M-mode or on 2-D echocardiography. A typical image is shown in Figure 3. Discussion Stentless bioprostheses and homografts are increasingly used in patients with aortic valve disease. Transvalvular gradients have shown

Figure 4. M-mode image of a patient with a paravalvular leakage of a stentless valve implanted with the subcoronary technique. Note the paraprosthetic lumen with diastolic ow indicating paravalvular leakage.

Figure 3. Typical 2-D echocardiographic image (transthoracic scan) of a stentless valve implanted with the fullroot technique immediately after operation. Note the thickened aortic root.

to be low and to remain low during an intermediate follow-up period.6,8 The increased use of stentless valves means that cardiologists have to perform echocardiographic follow-up in an increasing number of these patients. It is important that every echocardiographer and echocardiographic technician be aware of the typical images that can be observed with these valves. A paravalvular lumen is frequently encountered in patients in whom the subcoronary technique and the full-root technique are used. It can, therefore, be considered as a normal nding if no diastolic ow is present between the bioprosthesis and the aortic root. The lumen disappears in almost all patients within a period of 1 year. The lumen is pathologic if ow is present in the lumen (Fig. 4). Presence of diastolic ow means loosening of proximal and distal sutures as well as paravalvular leakage. The lumen has to be considered pathologic if the proximal suture line ruptures or shows loosening. This will result in dynamic obstruction and both systolic and diastolic ow between the bioprosthesis and the native aortic root.10 A paravalvular lumen is not unique for stentless bioprostheses, but can also be observed in patients with homografts11,12 or after Ross procedures. The incidence varies between 13% 73% depending on implantation technique. Although gradients are already low shortly after operation, they tend to decrease further after disappearance of the lumen. We conclude that xenograft-speci c echocardiographic images such as paravalvular lumen exist and can be considered normal if no diastolic ow is present. Suture dehiscence gives a typical image with dynamic obstruction.
Vol. 17, No. 7, 2000

628

ECHOCARDIOGRAPHY: A Jrnl. of CV Ultrasound & Allied Tech.

IMAGING OF STENTLESS AORTIC VALVE PROSTHESES

References
1. 2. 3. Ross DN: Homograft replacement of the aortic valve. Lancet 1962;2:487. Broom ND: Fatigue-induced damage in gluteraldehyde preserved heart valve tissue. J Thorac Cardiovasc Surg 1978;76:202-211. Chen W, Schoen FJ, Levy RJ: Mechanism of ef cacy of 2-amino oleic acid for inhibition of calci cation of gluteraldehyde-pretreated porcine bioprosthetic heart valves. Circulation 1994;90:323-329 Yogonathan AP, Eberhardt CE, Walker PG: Hydrodynamic performance of the Medtronic Freestyle aortic root prosthesis. J Heart Valve Dis 1994;3: 571-580 David TE, Ropchan GC, Butany JW: Aortic valve replacement with stentless porcine bioprostheses. J Cardiac Surgery 1988;3:5501-5505. Westaby S, Amarasna N, Long V, et al: Time related hemodynamic changes after aortic valve replacement with the Freestyle stentless xenograft. Ann Thorac Surg 1995;60:1633-1639.

4.

5. 6.

Hazekamp MG, Gof n YA, Huysmans HA: The value of the stentless biovalve prosthesis: An experimental study. Eur J Cardiothor Surg 1993;7:514-519. 8. Baur LHB, Jin XY, Y Houdas Y, et al: Echocardiographic parameters of the Freestyle stentless bioprosthesis in aortic position: The European experience. J Am Soc Echocardiography 1999;12:729-735. 9. Perry GJ, Helmcke F, Nanda NC, et al: Evaluation of aortic insuf ciency by Doppler color- ow mapping. J Am Coll Cardiol 1987;9:952-959. 10. Van Roosmalen R, Baur LHB, Braun J, et al: Dynamic obstruction: An unusual complication of stentless bioprostheses. Int J Cardiac Imaging 1999;15: 209-214. 11. Barbetseas J, Crawford S, Sa HJ, et al: Doppler echocardiographic evaluation of pseudoaneurysms complicating composite grafts of the ascending aorta. Circulation 1992;85:212-222. 12. Oechslin E, Carrel T, Ritter M, et al: Pseudoaneurysm following aortic homograft: Clinical implications? Br Heart J 1995;74:645-649.

7.

Vol. 17, No. 7, 2000

ECHOCARDIOGRAPHY: A Jrnl. of CV Ultrasound & Allied Tech.

629