A 55 year-old female with non-ischemic cardiomyopathy who initially presented with multiple syncopal episodes and increasing shortness of breath on exertion for the preceding 2 months. Her left ventricular ejection fraction (LVEF) was 20% at that time, with severe aortic stenosis and mild AI. Her aortic valve area was 1.0 cm², with a peak gradient 16.9 mmHg and mean gradient 9.2 mmHg gradient. She underwent AVR with a 21 mm Mitroflow (Sorin, Milan, Italy) bioprosthetic valve with significant improvement in her symptoms (aortic valve area 1.8cm², peak gradient 18.2 mmHg and mean gradient 9.0 mmHg).

Over the next 2 months, the patient had worsening heart failure with a decline in her LVEF to 15%. Her medical regimen was optimized with some improvement in her symptoms initially, however the improvement was short lived. Subsequent echocardiograms showed 15% LVEF, with new mild to moderate intra-valvular AI and mild perivalvular leak. Seven months after her AVR the patient was severely short of breath with declining ventricular function and mild to moderate AI. Her right ventricular function was severely depressed. The decision was made to insert HeartWare HVAD (Framingham, MA) with prosthetic aortic valve repair, Park’s stitch. Her INTERMACS profile was 3. Her pre operative platelet count, prothrombin time, and INR were 138 (1000/μl), 14.5 s, and 1.1. At the time of her LVAD insertion, the repair stitch was able to minimize her ischemic time to help preserve her RV. A pledgeted 4-0 Prolene (Ethicon, Cincinnati, OH) was placed to center of the each aortic valve leaflet. We could not appreciate perivalvular leak intraoperatively, however pledgeted mattressed 2-0 ethibond sutures were passed through from subvalvular leaflets to aortic wall and tied over pledgets. No obvious leaflet prolapse or perforation was noted (Fig. 1). The cardiopulmonary bypass time was 167 min and myocardial ischemic time was 52 min. Post-operatively, the patient had no AI on transesophageal echocardiogram as well as limited ejection through the aortic valve. The anticoagulation management was aspirin 325 mg/day and continuous heparin infusion based on thromboelastogram until Coumadin gets therapeutic, INR 2-3. She recovered well and was discharged home at postoperative day 27. At 4 months post HVAD insertion, the patient was feeling well with only trace AI on transthoracic echocardiogram. She developed lacunar infarction within the ventromedial nucleus of the right thalamus on 3 months after LVAD insertion. She was heart transplanted at 6 months following her LVAD insertion. Her aortic valve was fused and closed. No thrombus formation underneath the aortic valve was noted (Fig. 2).

Little has been known about management of recurrent AI after bioprosthetic AVR at the time of LVAD placement. A central coaptation stitch to these patients potentially is a good option for a bridge to transplant.

Short-term follow up in this patient shows promising results with only trace AI on echocardiogram. Although this technique has been shown to be a durable solution for patients with significant native aortic valve insufficiency [4], the long-term results of this technique especially in bioprosthetic valves remains to be seen to determine the durability of these results in the face of continued bioprosthetic valve degeneration.

Although there have been numerous studies of AI in LVAD patients, we don't know the long-term prognosis of patients who develops acquired AI. Since AI seemed to have negative impact on survival [1, 6], either bioprosthetic AVR or aortic valve repair is recommended for these patients. Patients with AI with following a previous bioprosthetic AVR, no recommendation is currently available. Performing a re-replacement bioprosthetic AVR may be more durable to prevent leaflet fusion, however explanting a previous valve and re-implant new valve in patients having an LVAD placement with concomitant severe RV dysfunction raises concerns about further impairing postoperative RV function. A Park's stitch to a bioprosthetic AVR is relatively easy, timely, safe and durable in LVAD patients being bridged to transplant.

This procedure was able to be performed in a quick and effective manner with short aortic cross clamp time to minimize the risk of RV dysfunction after LVAD placement.

The limitation of this technique in the setting of previous bioprosthetic AVR is no long-term mortality and morbidity data reported, therefore the application of this technique to destination patients should be cautious because of aortic valve leaflet fusion and potential thrombus formation below the fused aortic valve. If patients with AI from degenerating bioprosthetic AVR is listed as bridge to transplant, a Park's stitch to the bioprosthetic valve leaflets might be a good option to consider.

In conclusion, we reported a first successful case with Park's stitch and LVAD placement to previously placed bioprosthetic AVR. Further studies are mandatory to prove longevity and survival of this procedure in this context.