Prosthetic valve implantation is one of the most common cardiac surgical procedures, with over 700 implantations performed in Croatia (1) and over 90.000 in the US annually (2). The incidence of postoperative paravalvular regurgitation varies between 2% to 10% in the aortic position and 7% up to 17% in the mitral position (3-6). The background for paravalvular leak (PVL) is usually dehiscence of sutures resulting from infection, tissue friability or annular calcification. Most of these leaks are not significant and remain clinically silent, however, 1% to 3% will become symptomatic and require reoperation6. Patients with clinically significant PVL may present with symptoms and signs of congestive heart failure, as well as haemolytic anaemia. Reoperation has been the traditional treatment for these patients, but it is associated with significant risks and therefore it is sometimes not considered (7, 8). Also, re-do surgery may not be successful because the original anatomic problem persists. Therefore, alternative percutaneous transcatheter procedures for PVL closure have been developed and emerged as an important and viable alternative (9, 10) with good clinical follow-up results (which primarily depend on the degree of residual regurgitation) (11).

The key examination in the preprocedural planning is a detailed transesophageal echocardiography (TEE), with 3D TEE being particularly important in assessment of mitral PVL. Patients with ongoing infection and large dehiscence involving more than 1/4 of the valve ring are not suitable for percutaneous closure and, if feasible, should still be surgically treated.

We describe a case of a 43-year old male worker, who was diagnosed with symptomatic severe combined aortic valve disease (dominant stenosis and 3+ regurgitation) and underwent aortic valve replacement in 2010. In the operation protocol, a somewhat problematic excision of the native valve was noted, due to significant calcification in the area of the noncoronary and left coronary cusps. Otherwise, the procedure was uneventful and a 23 mm SJM prosthetic valve was implanted. On discharge, transthoracic and transesophageal echocardiography showed moderate 2+ paravalvular regurgitation, and the LDH value was also moderately elevated (629 IU/l). At this point, the systolic left ventricular function was normal (LVEF 60%). During the later course, there were no symptoms or signs of infection; however, echocardiography showed a deteriorating aortic regurgitation with moderately depressed left ventricular function (LVEF 45%). Two years later a re-do procedure was performed, during which the prosthesis was re-sutured. This operation was complicated by arrhythmic instability, prolonged weaning from extracorporeal circulation and need for inotropic support. Immediately after this procedure, TEE and aortography showed persisting severe aortic regurgitation. The location of the leak was between the former noncoronary and left aortic cups (Figure 1A and 1B). The patient made a complete clinical recovery, however, a 4+ aortic regurgitation with moderate dilation and depressed systolic function of the left ventricle persisted. Laboratory findings showed highly significantly elevated LDH of 1680 IU/l.

In July 2013 we performed percutaneous paravalvular leak closure. The procedure was done in general anaesthesia with TEE guidance. The vascular access was through a 7F sheath in the right femoral artery. The defect was passed with a 5F Amplatz 2 diagnostic catheter and straight-tip Terumo wire. The position of the wire was conformed by TEE (Figure 2A). This was exchanged for a preformed super-stiff wire positioned in the left ventricular cavity. A straight long 7F sheath was introduced to the left ventricle and used for delivery of the Amplatzer Vacular plug III (St Jude Medical, Maple Grove, Minnesota, US). According to the estimated defect size, a 12x5mm occluder was used (Figure 3). The plug was positioned and released in the usual fashion (Figure 2B, 2C). Angiographic control showed good position of the plug (Figure 2D, 2E), which did not impinge upon the prosthetic valve leaflets. Due to its elliptic shape, it nicely adapted to the crescent form of the defect. Significant reduction of the degree of regurgitation was observed, with only mild to moderate residual regurgitant jet (Figure 2E, 2F).

The invasive aortic pressure changed immediately from 140/40 mmHg to 120/70 mmHg. The patient was discharged 2 days after the procedure.

The follow-up echocardiography was performed 4 months after the occlusion and showed further reduction of the regurgitation to 1+ degree, possibly due to ensuing endothelialisation. During the follow-up period, a marked reduction in haemolysis parameters (control LDL of 432 IU/l) and favourable clinical course has been observed.

Clinically significant PVL after prosthetic valve implantation is a rare, but potentially serious complication. Percutaneous PVL closure has been introduced as an important alternative to re-do surgery with favourable immediate and late results. In view of the significant number of prosthetic valve implantations in Croatia, an increase in the need for this type of intervention can be expected each year.