ISSN 0326-646X





Sumario Vol. 43 - Nº 1 Enero - Marzo 2014

Is it possible to predict high transvalvular
gradients in aortic valve surgery?

Irene J. Del Corro, Alejandro E. Contreras, Eduardo J. Brenna

Departamento de Ecocardiografía. Servicio de Cardiología
Hospital Privado, Centro Médico de Córdoba.
(5016) Córdoba, Argentina.
E mail

Recibido 21-NOV-13 – ACEPTADO después de revisión el 17-ENERO de 2014.

The authors declare not having a conflict of interest.

Rev Fed Arg Cardiol. 2013; 43(1): 14-17

Print version Imprimir sólo la columna central




Introduction: The aim of this study was to determine the prevalence of patient prosthesis mismatch (PPM), projected or potential in our experience and to assess their ability to predict high transprosthetic gradients after cardiac surgery.
Material and method: Retrospective study. We included patients undergoing aortic valve replacement alone or combined with CABG. The effective orifice area (EOA) index projected was calculated for each valve type and number. The potential PPM was defined as EOA less than 0.85 cm2/m2. A mean gradient greater than 24 mmHg was considered PPM after cardiac surgery obtained in echocardiogram performed on follow-up. By bivariate correlation, Pearson correlation coefficient was obtained between the mean gradient and EOA variables. In addition, taking the mean gradient as gold standardfor  PPM, we calculated the sensitivity and specificity of EOA to predict PPM after cardiac surgery. We used the SPSS 12.0 statistical package.
Results: We included 29 patients, 72.4% men, mean age 65.4 ± 10.9 years, 34.5% were surgery combined with CABG. 41.4% received mechanical prosthesis. 20.7% (6/29) had a mean gradient greater than 24 mmHg. We found no correlation between the projected EOA and transvalvular mean gradients (correlation coefficient 0.27, p = 0.14) The EOA had a sensitivity of 18.7% and specificity of 76.9% to predict transvalvular high mean gradients.
Conclusions: PPM was very frequent as projected EOA; however had a bad correlation with the finding of high mean transvalvular gradients after cardiac surgery.

Key words: Prosthesis patient mismatch. Aortic valve replacement. Correlation.


Aortic valve replacement (AVR) is the second most frequent cardiovascular surgery and in a constant increase due to the population aging [1].

Often, the prosthesis used is small for the patient, a situation described by Rahimtoola [2], in 1978 as prosthesis-patient mismatch (PPM), in which the effective orifice area (EOA) of the prosthetic valve after the implant, is smaller to that of the native valve.

The main consequence of PPM is the generation of high transvalvular gradients through a functionally normal prosthesis. Clinically, it may interfere in the long-term prognosis, increasing mortality [3], making ventricular hypertrophy regression difficult [4] and in turn promoting a permanence of symptoms at strain [5,6]. A prevention strategy is estimating the projected EOA (projected PPM in cases of EOA smaller than 0.85 cm2/m2) to choose the best prosthesis for each patient [7].

The goal of this paper was knowing the prevalence of projected or potential prosthesis-patient mismatch in our experience and evaluating its capacity to predict postoperative high transprosthetic gradients.


Retrospective study, in which the registry of patients operated for aortic valve replacement, alone or in combination with myocardial revascularization surgery, was evaluated consecutively between January 2009 and December 2010. In each patient, the EOA index was estimated retrospectively, projected for every type and number of valve, which leads to previously published normal values divided by the body surface of the patient [8]. An EOA smaller than 0.85cm 2/m/ was considered as potential PPM. Moreover, the echocardiograms performed in the far postoperative term were evaluated, taking into account the valve closure and opening by 2-dimensional mode, ejection fraction by the Teichholz method, transvalvular velocity, peak gradient and mean gradient by continuous Doppler and the presence of valvular or paravalvular regurgitation. Since echocardiograms had been made over the first 6 months of the postoperative period, before a normal valve opening, postoperative PPM was considered at abnormally high transvalvular gradients (>24 mmHg of mean gradient) [9]. Some patients with gradients >24 mmHg but with moderate to severe perivalvular leaks, were not considered as postoperative PPM carriers. The echocardiograph devices used were Vivid 7 and Vivid S5, General Electric Medical Systems.

The categorical variables are expressed in percentage and continuous variables in average and standard deviation. The average of mean gradient was compared between the patients with or without potential PPM with Student’s T test. By the bivariate correlation method, the Pearson correlation coefficient was obtained between the mean gradient variables and EOA. Besides, taking the presence of high mean gradient (>24 mmHg) as gold standard of PPM, the sensibility, specificity and positive and negative predictive value of projected EOA were estimated to predict PPM in the postoperative period. The statistical package SPSS 12.0 was used.


From January 2009 to December 2010, 49 patients were operated, from whom 20 were excluded for not having an echocardiographic follow-up in our hospital. Thus, 29 patients were included, 72.4% males, average age 65.4±10.9 years. The general characteristics of the population studied are presented in Table 1. Surgeries combined with myocardial revascularization amounted to 34.5%; 41.4% received mechanical prosthesis and the rest biological prosthesis with stent. Body surface was 1.92%±0.18 m2. Number 21 prosthesis were implanted in 34.5%, and the rest were prosthesis 23 to 27. Average EOA was 0.85±0.13 cm2/m2. Sixteen patients (55.2%) had potential PPM, from whom 6.2% (1/16) had severe PPM (<0.65 cn2/m2).







Chronic renal failure


Smoking/former smoker.

13.8% / 37.9%

Coronary artery disease.


Previous myocardial infarction.


Peripheral vascular disease.


Table 1. General characteristics of the population

In the echocardiographic analysis of the follow-up of the 29 patients, they had 61.8±6.9% ejection fraction, transvalvular peak velocity of 2.94±0.74 m/sec, transvalvular mean gradient 20.1±10.9 mmHg. Mean gradient >24 mmHg was present in 20.7% (6/29). From the 16 patients that had potential PPM (16/29, 55.2%), only 3 of them had high gradients in follow-up (3/16, 18.8%). There were no differences in the average of mean gradients between those patients with or without potential PPM (19±11.5 mmHg vs. 21.4±10.5 mmHg, p=0.56)(Figure 1), and we did not find either a correlation between projected EOA and transvalvular mean gradients in the echocardiograms of follow-up, correlation coefficient 0.27, p=0.14 (Figure 2). The projected EOA to predict high transprosthetic gradients in the postoperative period had a sensibility of 18.7% (95 CI, 4.9%-46.3%), specificity 76.9% (95 CI, 45.9%-93.8%), positive predictive value (95 CI, 13.9%-86%) and negative predictive value 43.4% (95 CI, 23-65.1%).

There were 2 patients with mean gradients >24 mmHg, with moderate or severe peri-prosthetic leaks, not considered as patients with PPM.

Figure 1. Comparison of mean gradients. Prosthesis patient mismatch (PPM) according to the projected effective orifice area (smaller or greater than 0.85cm2/m2)


Figure 2. Correlation in follow-up, between the projected effective orifice area and mean transvalvular gradients. Effective orifice area (EOA). Potential prosthesis patient mismatch (PPM).


Prosthesis-patient mismatch (PPM) is defined as the situation in which the effective orifice area of the prosthetic valve is smaller than the native valve[2]. PPM was identified as a non-structural dysfunction, in fact considered as a “hemodynamic abnormality”, with a significant increase in transvalvular gradient, in spite of the normal operation of the implanted valve, being the main consequence [5-6]. According to the guidelines for valve prosthesis follow-up, PPM should be suspected in the cases of velocities greater than 3 m/sec, a ratio between the left ventricular outflow tract velocity and transvalvular velocity between 0.25 to 0.29, with an early peak of aortic jet (<100 ms) and indexed EOA smaller than 0.85 cm2/m2, measured by continuity equation [8]. A simple way of diagnosing PPM in bivalvular mechanical prosthesis could be using mean gradients, the time elapsed between the onset of ejection to the maximal instantaneous gradient and fluoroscopy [9]. Depending on the method used, the incidence of PPM ranges between 20-70%, and taking into account the indexed EOA, the prevalence is 25-50% [10,11]. In our experience, around half of the patients that were operated had potential PPM (55.2%), similar data to those found in bibliography, but potential PPM had no correlation to transvalvular gradients in the postoperative period, a very low sensibility to predict them, and discrete specificity.

Different methods have been proposed to predict PPM. In clinical practice, it is recommended to choose those prostheses with EOA (previously published) divided by the body surface of the patient, resulting in a figure greater than 0.85 cn2/m2. Several studies have analyzed the ability to predict what will happen in the postoperative period. Bleiziffer et al, reports a 53% sensibility and 80% specificity, and positive predictive value of 67%, taking into account as gold standard the estimation of EOA measured by echocardiography at the 6 months of follow-up [12]. Pibarot et al, with the same method reported 62% sensibility and 87% specificity, and had a significant correlation, although regular between EOA estimated in the pre-operative period with the transvalvular gradients in rest (r=0.67) [13]. As it happens in the estimation of native aortic stenosis, there are sources of error in the estimation of the prosthetic valvular area [8]. A recent paper, comparing methods of postoperative evaluation on biological prosthesis, found significant differences in the prevalence of PPM, being 63.1% when defined only based on the EOA (less than 0.85 cm2/m2) vs. 4.6% when the criteria of the prosthesis assessment guidelines are used (velocity and gradient, peak-velocity time, dimensionless index)[14].

In cases of risk of prosthesis-patient mismatch, as in the case of small aortic root, the technique of root enhancement is proposed, with a mild increase in extracorporeal circulation times, with good surgical results, with no differences in operative mortality (0.9%) in regard to patients with aortic valve replacement [15]; however, this strategy has not shown differences in long-term mortality with patients with pure replacement and PPM [16].

PPM could be prevented, using a simple strategy at the time of surgery, estimating the minimal EOA [7,10] to choose the prosthesis with best profile, or aortic root enhancement; however, we found that the EOA values published previously, had a scant specificity to predict high gradients in the postoperative period.


The prosthesis-patient mismatch was very frequent according to the projected effective orifice area; however, it had a scant correlation with the finding of high trans-valvular gradients in the postoperative period.



  1. Chambers J. Aortic stenosis. BMJ 2005; 330: 801-2.
    2- Rahmtoola SH. The problem of valve prosthesis–patient mismatch. Circulation 1978; 58: 20-24.
    3- Head SJ, Mokhles MM, Osnabrugge RLJ, et al. The impact of prosthesis-patient mismatch on long-term survival after aortic valve replacement: a systematic review and meta-analysis of 34 observational studies comprising 27186 patients with 133141 patient-years. Eur Heart J 2012; 33: 1518-29.
    4- Tasca G, Brunelli F, Cirillo M, et al. Impact of valve prosthesis–patient mismatch on left ventricular mass regression following aortic valve replacement. Ann Thorac Surg 2005; 79: 505-10.
    5- Pibarot P, Honos GN, Durand LG, et al. The effect of prosthesis–patient mismatch on aortic bioprosthetic valve hemodynamic performance and patient clinical status. Can J Cardiol 1996; 12: 379-87.
    6- Pibarot P, Dumesnil JG, Lemieux M, et al. Impact of prosthesis-patient mismatch on hemodynamic and symptomatic status, morbidity, and mortality after aortic valve replacement with bioprosthetic heart valve. J Heart Valve Dis 1998; 7: 211-28.
    7- Pibarot P, Dumesnil JG. Hemodynamic and clinical impact of prosthesis-patient mismatch in the aortic valve position and its prevention. J Am Coll Cardiol 2000; 36-4: 1131-41.
    8- Zoghbi WA, Chambers JB, Dumesnil JG, et al. Recommendations for evaluation of prosthetic valves with echocardiography and Doppler ultrasound. J Am Soc Echocardiogr 2009; 22 (9): 975-1014.
    9- Muratori M, Montorsi P, Maffessanti F, et al. Dysfunction of bileaflet aortic prosthesis: accuracy of echocardiography versus fluoroscopy. J Am Coll Cardiol Img 2013; 6: 196-205.
    10- Pibarot P, Dumesnil JG. Prosthesis patient mismatch: definition, clinical impact and prevention. Heart 2006; 92 (8): 1022-9.
    11- Tasca G, Mhagna Z, Perotti S. Impact of prosthesis–patient mismatch on the cardiac events and mild term mortality after aortic valve replacement in patients with pure aortic stenosis. Circulation 2006; 113: 570-6.
    12- Bleiziffer S, Eichnger WB, Hettich I, et al. Prediction of valve prosthesis patient mismatch prior to aortic valve replacement. Wich is the best method?. Heart 2007; 93: 615-20.
    13- Pibarot P, Dumesnil JG, Cartier PC, et al. Patient-prosthesis mismatch can be predicted at the time of operation. Ann Thorac Surg 2001; 71: 265-8.
    14- Chacko SJ, Ansari AH, McCarthy PM, et al. Prosthesis-patient mismatch in bovine pericardial aortic valves. Evaluation using 3 different modalities and associated medium-term outcomes. Circ Cardiovasc Imaging 2013; 6: 776-83.
    15- Castro LJ, Arcidi JM, Fisher AL, et al. Routine enlargement of the small aortic root: a preventive strategy to minimize mismatch. Ann Thorac Surg 2002; 74: 31-6.
    16- Kulik A, Al-saigh M, Chan V, et al. Enlargement of the small aortic root during aortic valve replacement: is there a benefit? Ann Thorac Surg 2008; 85: 94-100.


Publication: March  2014

Editorial Electrónica
de FAC

8vo. Congreso Virtual de Cardiología

1º Setiembre al
30 Noviembre, 2013

XXXI Congreso Nacional de Cardiología

30-31 Mayo,
1º Junio, 2013
Organiza: Región Patagónica

Revista de FAC


Contenidos Científicos
y Académicos



Accesos rapidos