Figure 1 – Follow-up overall mortality (Kaplan–Meier survival curve)

Discussion

The Perceval ^® prosthetic valve (Sorin Group, Saluggia, Italy) has been increasingly used in European cardiac surgery centers for treatment of aortic valve disease since its first reports in 2007. As a truly sutureless bioprosthesis with proven excellent hemodynamic outcomes, safety and marked reduction in aortic cross-clamping and CPB times, it has been shown to reduced postoperative morbidity and mortality as well as cost reduction of up to 25% when compared to conventional biological heart valves (8,20), especially when treating older patients and in those with comorbidities. However, several questions are yet unanswered, especially regarding long-term durability, endocarditis risk, and the need for postoperative PPI.

In our retrospective study, we sought out to assess the clinical and hemodynamic performance, safety, and durability of the Perceval^® valve for isolated AVR and in concomitant procedures.

Our cohort is composed essentially by elderly patients, with an average overall age of over 77 years and a small standard deviation of approximately 5 years. Additionally, most patients had one or more preoperative risk factors and/or comorbidity, with an average in-hospital mortality risk measured by an EuroSCORE II of almost 3%, despite the majority having preserved LV function.

The overall total surgical time, aorta cross-clamping and CPB times took into consideration the surgeries where reimplantation of the valve was necessary. The times obtained represent a significant reduction in comparison to the mean aorta cross-clamping and CPB times of 78 and 106 min reported in conventional AVR, according to the Society of Thoracic Surgeons database. This major reduction of over 50% in both times could translate into improved clinical results, especially in patients with comorbidities such as diabetes mellitus (7) or in a medium-high surgical risk profile, as aortic clamping and CPB times are considered independent predictors of morbidity and mortality in heart surgery (10).

The Perceval ^® valve was successfully implanted in 98.99% as in 2 cases (2/198 = 1.01%) conversion to conventional bioprosthesis was required due to valve migration. One way of possibly reducing the likelihood of these events is by strictly obliging the manufacturer’s preoperative echo-doppler aortic root evaluation recommendations (a patient is suitable if the ratio between the diameter of the STJ and the annular diameter is ≤ 1,3) (21).

In 4 cases (2.04%) reimplantation was necessary to readjust the valve, due to minor misplacement and/or paravalvular leaks found in the intraoperative transesophageal echocardiography, which is paramount to assure correct placement and function. Reimplantation through the “x-movement” technique, allowed an “en bloc” bioprosthesis excision, simple re-folding and easy reimplantation of the valve without damage to the leaflets or the bioprosthesis structure. (22). Reasons for these misplacements are still unclear but in one of the cases, migration followed cardiac manual manipulation due to additional manual de-airing. (23).

In our study, we report no significant paravalvular leaks, a result below the already low 1-2% commonly observed in trials with the Perceval ^® valve (6). In comparison, TAVI causes a greater number of moderate-to-severe paravalvular leaks (7-12%), followed by conventional AVR (1,9%), which at 2 years have been shown to be independent predictors of mortality (10,24). Consequently, correct measurement, placement and final visual confirmation of correct valve placement could help avoid these complications.

The most common immediate postoperative complications (Table 3) were in line with the expected for a major cardiac procedure. The relatively high rate of postoperative aminergic support might be related to the fluid restriction protocol used in our department. New-onset atrial fibrillation is usually multifactorial and is the most common dysrhythmic complication occurring after any cardiac surgery, affecting typically between 30% and 50% of the patients, more than we report (25). Although the incidence of acute kidney injury is relatively high (10,5%), only a reduced number of patients required renal replacement technique (3,1%).

The overall incidence rate for PPI of 12.8% is within the interval incidence described in the literature, of 3.1% (26) to 17% (27) although it is above rates for conventional bioprosthesis reported of 3.0 to 11.8% (28) and comparable to the ones reported for TAVI (29). Current best available evidence suggests baseline conducting system disease is the most powerful independent predictor of PPI requirement following AVR. Other patient-related predictors are advanced age, annular calcification, (27). Operative-related factors such as incomplete decalcification of the aortic ring, valve oversizing, valve and guiding sutures position, reoperations, longer perioperative CPB time, and procedural implanting steps and sizing’s learning curve effect are also important (13). In our center, all patients were operated on or supervised by an experienced surgeon in this procedure. That said, standardizing the implantation technique could offer benefits in reducing the PPI incidence.

We report one case of valvular thrombosis (0.51%), which was successfully treated with oral anticoagulants (Vitamin K antagonist). Following the procedure, all patients received antiplatelet treatment according to the standard protocol in use in our center, consisting of acetylsalicylic acid or clopidogrel for a period of 3 to 6 months when sinus rhythm was present.  Antiplatelet and anticoagulation management after sutureless valve placement is not standardized, as no specific recommendations have been made in recent guidelines (17).

We found no structural valve deterioration, endocarditis cases or strokes up to the 5-year follow-up, presumably because the Perceval^® valve allows less manipulation of the aortic root and annulus, zero permanent contact with foreign material such as sutures (13) and it has been shown to present high resistance against endocarditis in comparison with conventional prostheses. Nevertheless, 5 years of follow-up is still a short time compared to the data available for conventional prostheses of up to 25 years (30).

The mean postoperative transvalvular gradients after surgery confirm the excellent hemodynamic performance of the Perceval ^®valve, coherent with other studies (6) and significantly lower when compared to the ones provided by conventional prostheses (28) and TAVIs (29).

ICU and total in-hospital stay were markedly lower when compared to reports of conventional bioprostheses (31) and similar when compared to TAVIs (29).

Early overall mortality (in-hospital or up to 30 days after surgery) was 2.0%, below the one predicted by the initial EuroSCORE, in line with the 2.8% reported for conventional prostheses and significantly lower than with TAVIs (10,30,32,33). Of the four patients who died, none of their deaths was caused directly from a failure in the prosthetic valve or the procedure themselves (cardiogenic and septic shock, neurological complications). These results show good short-term clinical outcomes despite the risk profile and advanced age of the cohort.

Overall cumulative survival at 1, 3 and 5 years and the correspondent Kaplan-Meier curve are the ones expected taking into consideration our cohort’s age, comorbidities, and type of interventions. It is up to par with the mean corresponding cumulative survival of conventional AVR (34) and better than with TAVIs up to 2 years (29), showing safety of use and good mid-term durability, with no structural valve deterioration.

Additionally, the Perceval ^® valve enables standardization and simplification of MIS approaches, in a way conventional prostheses have not yet made possible. MIS is associated with a significant technical difficulty due to reduced visualization that increases aortic cross-clamping and CPB time, extending the learning curve (10).