Extended Application of Percutaneous Pulmonary Valve Implantation

Methods

Between September 2007 and March 2008, a total of 15 patients were referred for pulmonary valve implantation. Two patients were excluded from the study for technical reasons. The demographics of the 13 patients who had the complete procedure are summarized in Table 1.

All procedures were performed under general anesthesia. Cardiac catheterization was performed via percutaneous puncture of the right femoral vein. The femoral artery was also percutaneously accessed for monitoring and simultaneous coronary angiogram to rule out any potential compression on the coronary arteries by the newly implanted valve.

After assessment of the RVOT in multiple projections, a super-stiff exchange guidewire (Back-up Meier, Boston Scientific, Miami, Florida) was positioned in a distal pulmonary artery (PA) branch using a right coronary catheter. This was followed by sizing of the narrowest diameter of the outflow tract by manual inflation of a 24-mm sizing balloon (AGA Medical Corporation, Golden Valley, Minnesota), to determine its suitability for PPVI. In patients with previous conduits, the annulus size was determined by the original size of conduit. In patients with an RVOT patch, the mean size of annulus was 18 mm, and the stretch diameter was 20 mm.

Five of 6 patients with an RVOT patch had the RVOT patch extending into the main PAs (i.e., transannular patch); in these patients, a stent was placed into the RVOT at the level of the narrowest area to anchor the stent to create what we call an artificial conduit.

The stents were selected to cover most of the RVOT. The stents were deployed so that one-third of the stent fell below the RVOT narrowest diameter, with the intent to keep the upper end of the stent 5 to 10 mm below the bifurcation of the main PA (Fig. 1).Three of 4 patients had a 34-mm-long CP stent (NuMED, Inc., Hopkinton, New York). In 1 patient, a 36-mm intrastent LD max (ev3 Inc., Plymouth, Minnesota) was used. All were pre-mounted on a 22 × 40 mm balloon-in-balloon catheter (NuMED, Inc.).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Lateral View of the RV Outflow Patch

(A)Lateral view of a still-frame angiogram of the right ventricular (RV) outflow patch with severe pulmonary regurgitation. (B)As in Aafter pre-stenting. (C)Same patient shown in Aand Bfollowing percutaneous pulmonary valve implantation, showing a competent valve.

The Melody valve (Medtronic Inc., Minneapolis, Minnesota) was prepared using the ensemble delivery system as previously described (2,4). In summary, the Melody valve was withdrawn fully into a protective sheath on the delivery system. The whole assembly was then passed over the guidewire and advanced into the pre-stented RVOT patch or the stenotic conduit. Once in an acceptable position, the covering sheath was withdrawn to expose the Melody valve that was previously crimped on the balloon. The inner and outer balloons were sequentially inflated, thus deploying the Melody valve. Thereafter, both balloons were rapidly deflated and removed, leaving the guidewire in place. Angiographic and hemodynamic evaluations were repeated before the guidewire was finally removed.

The procedure could not be completed in the 2 patients who proved to be unsuitable for the procedure because their RVOT patch stretch diameter was >22 mm, which was larger than the size of the Melody valve available.

Patients were assessed on day 1, as well as 1, 3, and 6 months after the procedure. Assessment included physical examination, 12-lead electrocardiography, chest X-rays (posterior anterior and lateral views), and cross-sectional echocardiography with color Doppler.

Results

Seven females and 6 males, age range 10 to 23 years (mean 14.3 years), weight range 30 to 84 kg (mean 45 kg), had successful PPVI (Table 1). Most of the patients had variant TOF, and the majority of them were treated with RV-to-PA valve conduits. Only 4 patients (numbers 1, 7, 8, and 13 in Table 1) had a previous surgical repair using an RVOT patch. The median procedure time was 102 min from puncture to sheath removal (range 67 to 124 min), and median fluoroscopy time was 21 min (range 11 to 36 min).

After valve implantation, the RV systolic pressure dropped from 61.2 ± 14.9 mm Hg to 37.6 ± 6.7 mm Hg (p < 0.05). The outflow (RV-PA) pressure gradient was also reduced from 39.6 ± 15.1 mm Hg to 12.1 ± 9 mm Hg (p < 0.05). PA diastolic pressure increased from 8.1 ± 2.6 mm Hg to 11.5 ± 2.8 mm Hg (p < 0.05) (Fig. 2).Immediately after implantation, angiography showed no patient with more than mild regurgitation. Furthermore, there was a trend toward reduction in the RV end-diastolic pressure; however, this did not reach statistical significance (data not shown).

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

Changes in Ventricular Systolic Pressure, Pulmonary Valve Gradient, and PA Diastolic Pressure

Changes in (A)right ventricular (RV) systolic pressure, (B)pulmonary valve gradient, and (C)pulmonary artery (PA) diastolic pressure before and after percutaneous pulmonary valve implantation (PPVI).

Echocardiography performed 24 h after the PPVI showed a reduction in RV systolic pressure as estimated from tricuspid valve regurgitation velocity and RVOT gradient. There was also a reduction in the PVR grade, and all patients had no or trace PVR after the procedure (Fig. 3).

• Download figure
• Open in new tab
• Download powerpoint

Figure 3

Continuous-Wave Doppler Tracing of the Pulmonary Valve

(Top)Pulmonary regurgitation before pulmonary valve implantation; (bottom)competent pulmonary valve following percutaneous pulmonary valve implantation.

All patients were discharged within 2 days after the procedure. The mean follow-up was 4 months, and all patients were alive and well. There were no immediate or late complications.

During follow-up, serial echocardiography showed that an early decrease in the RVOT gradient was sustained at the latest follow-up. Chest X-rays, posterior-anterior and lateral views, showed no stent migration or fracture.

Discussion

We have shown that PPVI can be used in patients with chronic PVR in the presence of an RVOT patch. However, this can only be accomplished by prior deployment of an intravascular stent as an artificial conduit that allows subsequent valve implantation.

Gibbs et al. (5) have previously described stenting of the RVOT as palliative for severe RV infundibular stenosis as an alternative to surgical ventricular outflow enlargement. We have modified this technique to allow the use of a stent as an artificial conduit to anchor the Melody valve.

This technique can be applied to patients with an RVOT dimension of <22 mm because the maximum available size of the Melody valve is 22 mm. We expect that the use of a longer stent in the RVOT will not only facilitate implanting the pulmonary valve but may also prevent stent fracture and/or migration of the newly implanted pulmonary valve.

Pre-dilation of the stenotic area is an important step for safety and effectiveness in pre-stenting an RVOT patch. Compliance of the RVOT patch is unpredictable, but balloon sizing allows for determination of the exact location of the waist. The presence of multiple waists may result in stent malposition. Furthermore, balloon sizing may predict potential coronary artery compression if performed with simultaneous coronary angiography. This potential complication has been previously described (6). Because the RVOT patch is relatively more distensible as compared to conduits, we elected to use the more compliant AGA balloon.

These preliminary results suggest that, in selected patients, pre-stenting and pulmonary valve implantation offers a less invasive alternative to re-do surgical operation in patients after TOF repair with an RVOT patch. However, further follow-up is required to study the long-term results of PPVI in this subset of patients.