Prediction of valve prosthesis-patient mismatch prior to aortic valve replacement: which is the best method?

To predict the occurrence of valve prosthesis-patient mismatch (VP-PM) after aortic valve replacement (AVR), the surgeon needs to estimate the postoperative effective orifice area index (EOAI).

To compare different methods of predicting VP-PM.

The effective orifice area (EOA) of 383 patients who had undergone AVR between July 2000 and January 2005 with various aortic valve prostheses was obtained echocardiographically 6 months after the operation. We tested the efficacy of (1) EOAI calculated from the echo data obtained in our own laboratory, (2) indexed geometric orifice area, (3) EOAI estimated from charts provided by prosthesis manufacturers (which are based either on in vitro or on echo data) and (4) EOAI estimated from reference echo data published in the literature to predict VP-PM.

Sensitivity and specificity to predict VP-PM were 53% and 83% (method 1), 80% and 53% (charts based on echo data, parts of method 3) and 71% and 67% (method 4) using reference data derived from echocardiographic examinations. The sensitivity of method 2 and of charts based on in vitro data (parts of method 3) to predict VP-PM was 0-17%. The incidence of severe VP-PM could be reduced from 8.7% to 0.8% after the introduction of the systematic estimation of the EOAI at the time of operation (p = 0.003, method 1).

The best method of predicting VP-PM is the use of mean (SD) EOAs derived from echocardiographic examinations, whereas the use of in vitro data or the geometric orifice area is unreliable. After the surgeon's anticipation of VP-PM prior to AVR, the incidence of VP-PM could be reduced.