Valve size matters: association of prosthesis size with mortality in patients without prosthesis-patient mismatch after surgical aortic valve replacement in a nationwide cohort

Background Surgical aortic valve replacement (SAVR) remains a standard therapy for severe aortic valve disease. Prosthesis–patient mismatch (PPM) occurs when the effective orifice area (EOA) of an implanted aortic valve is too small in relation to the patient’s body surface area. While PPM is known to impair outcomes, the prognostic impact of valve size in patients without PPM remains unclear. Purpose The aim of the current study is to investigate the association between prosthetic valve size and long-term mortality after primary isolated SAVR across ages and valve types in patients without PPM. Methods A nationwide retrospective cohort analysis was performed using the Netherlands Heart Registration aortic valve supplement (2007–2018). Adult patients undergoing primary isolated SAVR were included. Patients with prior surgery, concomitant procedures, endocarditis, or missing prosthesis data were excluded. Analyses were restricted to patients without any degree of PPM, as defined by the VARC-3 criteria. EOA reference values were derived from echocardiographic guidelines (1,2). Cox models were constructed to assess the association of prosthesis size and EOA, indexed EOA (iEOA) and absolute valve size with long-term mortality, adjusted for relevant available baseline characteristics (age, sex, BSA, LVEF, creatine, diabetes, chronic lung disease, neurological dysfunction, recent myocardial infarction, previous PCI, critical preoperative state, non-cardiac vasculopathy, prosthesis type, surgery year). To assess a potential non-linear relationship between prosthesis size and mortality, natural spline functions were used. Results Of 17140 SAVR patients operated in the Netherlands between 2007-2018, 7148 were free of PPM and met inclusion criteria (6084 bioprostheses; 1064 mechanical). Median follow-up was 7.3 years with 98.2% follow-up completeness; 1367 deaths occurred. Smaller prostheses (19–21mm) were independently associated with higher mortality versus larger sizes (Figure 1) in patients under the age of 74 (Figure 2), despite absence of PPM. In figure 2 values below 1 indicate a reduced mortality hazard of a size 27 (median EOA=2.1) valve compared to a 23 valve (median EOA=1.8). Increasing EOA was associated with lower mortality: an EOA of 2.5cm² was associated with a 23% lower adjusted hazard of death versus the cohort median of 1.8 cm². Conclusions In patients without PPM, under the age of 74 at time of primary isolated SAVR, smaller prosthesis size, EOA and iEOA are independently associated with increased mortality rates after SAVR, questioning whether current PPM definitions should be redefined. Surgical planning should prioritize maximizing prosthesis size, including consideration of annular or root enlargement, to optimize long-term outcomes and prospective valve-in-valve options. Future trials comparing SAVR with transcatheter options should be aware of this association in their inclusion process and analyses.