Key Points
-
Aortic valve replacement with a mechanical prosthesis had a lower rate of PPM than AVR with a biological prosthesis in the present study's population.
-
Patients who received biological prostheses and had PPM had worse rates of long-term survival.
-
Patients who received biological prostheses and had PPM had worse QOL physical component scores after long-term follow-up.
Abbreviations and Acronyms
- AVR
-
aortic valve replacement
- EOA
-
effective orifice area
- iEOA
-
indexed effective orifice area
- PPM
-
patient-prosthesis mismatch
- QOL
-
quality of life
- SF-12
-
12-Item Short Form Health Survey
Introduction
Patient-prosthesis mismatch (PPM) represents the mismatch between the prosthetic valve's effective orifice area (EOA) and the patient's hemodynamic requirements.1 Given the considerable heterogeneity of the patient population, PPM is most commonly presented as EOA divided by the patient's body surface area, resulting in indexed EOA (iEOA) values. An iEOA between 0.85 cm2/m2 and 0.65 cm2/m2 is considered moderate PPM, while an iEOA of less than 0.65 cm2/m2 is considered severe PPM.2 The PPM can furthermore be calculated as measured or predicted. If it is measured, it is usually done with a transthoracic echocardiography exam before hospital discharge. This echocardiograph-dependent method, however, is subject to diverse findings. It is also flow-dependent and may lead to an overestimation of PPM in hearts with diminished ejection fractions. To avoid these limitations, predicted PPM, which is calculated using EOA values provided by the manufacturer divided by the patient's body surface area, has been introduced.2,3
Patient-prosthesis mismatch leads to higher transprosthetic gradients that may impair mass myocardial reduction as well as functional capacity and survival after aortic valve replacement (AVR).4,5 Several published studies have emphasized the negative impact of PPM on survival.6-9 Moderate PPM is commonly found after AVR, ranging in frequency from 20% to 70%,10 depending on the definition and method of calculation. Severity of PPM is the primary determinant of its influence on clinical outcomes. Although severe PPM is responsible for worse long-term survival, it is found in fewer than 2% of all AVRs.7
Although mechanical and biological prostheses have a similar rate of valve-related complications, PPM is more frequently observed after biological prosthesis implantation. The impact of structural valve deterioration in a biological prosthesis and stable hemodynamic performance of a mechanical prosthesis on PPM and subsequently on survival has yet to be determined.10,11 Besides the impact of PPM on survival, the question of a patient's quality of life (QOL) arises. Some authors have claimed that PPM after AVR diminishes patients’ physical and mental capacities, especially in patients older than 70 years of age.9,12 Overall, a mechanical prosthesis tends to have a lower risk of PPM than a biological prosthesis, but the clinical outcomes remain unclear.
This study aimed to investigate the impact of an implanted mechanical or biological prosthesis with PPM on long-term survival and QOL after an isolated AVR procedure.
Patients and Methods
Ethical Approval
The clinical studies ethical committee of the University Clinical Centre of Serbia approved the study, and all patients signed the informed consent form (KH151/2020).
Study Design
Between January 2015 and December 2020, 652 consecutive patients underwent an isolated AVR procedure at the University Clinical Centre of Serbia in Belgrade. This observational analysis included elective, urgent, and emergent cases performed for any pathology. Combined surgery was the exclusion criterion. Of the 641 patients who survived the index procedure, 35 refused to participate in research, and 11 were lost to follow-up (Fig. 1).
Citation: Texas Heart Institute Journal 50, 5; 10.14503/THIJ-22-8048
The remaining 595 patients were divided into 2 groups according to whether they had received a biological or mechanical prosthesis and were evaluated for the presence or absence of PPM. The groups with and without PPM were compared with regard to baseline characteristics, operative characteristics, survival, complications, freedom from angina, and QOL for up to 6 years of follow-up. The data were obtained from medical records and through telephone surveys during the follow-up period.
Surgical Procedure
The surgical procedures were performed using cardiopulmonary bypass, and cardioplegic arrest was achieved using a cold, crystalloid cardioplegic solution. Standard median sternotomy, upper ministernotomy through the fourth intercostal space, or anterior minithoracotomy was used for the surgical approach. The following mechanical prostheses were used: St Jude Regent, St Jude Master (Abbott Laboratories); ATS Open Pivot (Medtronic); On-X Heart Valve (CryoLife Inc); and Carbomedics (LivaNova). The biological prostheses used in the study were Hancock (Medtronic); Epic, Trifecta (Abbott Laboratories); CROWN PRT, SoloSmart; and the sutureless Perceval S valve (LivaNova). The choice of the implant procedure was made according to the current guidelines and the surgeon's discretion as well as with the input of the fully informed patient.
Definitions and Study End Points
The data were extracted from the University Clinical Centre of Serbia's Aortic Valve Registry, a prospectively maintained clinical registry of all patients undergoing AVR or repair at the institution, and double-checked for accuracy (M.M. and A.M.). All operative survivors were followed regularly, and follow-up was completed in 641 of 652 patients (98.3%). All clinically gathered data, including adverse events during follow-up and cause of death, were registered and reported according to the standardized institutional protocol.
Patient-Prosthesis Mismatch
Patient-prosthesis mismatch was defined as having an iEOA less than 0.85 cm2/m2. An iEOA between 0.85 cm2/m2 and 0.65 cm2/m2 was considered moderate, and an iEOA less than 0.65 cm2/m2 was considered severe. The PPM calculations were performed using the EOA value provided by the prothesis manufacturer divided by the patient's body surface area. Patients with mechanical and biological prostheses were compared according to the presence (iEOA <0.85 cm2/m2) or absence (iEOA >0.85 cm2/m2) of PPM.
QOL Survey
Quality of life was estimated using the 12-Item Short Form Health Survey (SF-12), which is derived from the 36-Item Short Form Health Survey and scored the mental and physical components of the study. The SF-12's physical component investigates physical function, pain levels, and role physical; the cognitive component surveys mental health and social and emotional functioning. The results of these 2 components were scored from 0 to 100, with higher scores representing better mental and physical health.
Statistical Analyses
Descriptive statistics were calculated for baseline demographic and clinical features as well as for treatment outcomes. Graphical and mathematical methods tested the normality of distribution. As appropriate, continuous variables were presented as mean (SD) or median (IQR). Categorical variables were presented as numbers and percentages. Differences between groups were analyzed using the t test or the Mann-Whitney test for continuous variables and the Pearson χ2 test for categorical variables.
A propensity score was developed based on differences in patient age and sex in the PPM and no-PPM groups. Logistic regression analysis with the propensity score as a covariate was performed. Differences in QOL in the PPM and no-PPM groups were analyzed with analysis of covariance, with adjustments made for age and sex. Survival analysis was performed using the Kaplan-Meier method, and the groups were compared using the log-rank test. In addition, Kaplan-Meier survival curves were truncated at a point during follow-up when at least 10% of patients were still at risk to avoid visual misin-terpretation.13 Cox proportional hazards regression was performed with propensity score adjustment. P < .05 was considered statistically significant, and all testing was 2-sided. Statistical analysis was performed using SPSS Statistics for Windows, version 21.0 (IBM Corp).
Results
Of the 595 patients enrolled in the present study, 159 (26.7%) received a biological prosthesis, and 436 (73.3%) received a mechanical prosthesis. The baseline characteristics of patients with biological and mechanical valves are presented in Table I. Patients with biological valves were statistically significantly older than patients with mechanical valves (mean [SD] age, 69.9 [7.7] years vs 62.6 [12.1] years, P < .001). A mechanical prosthesis was more frequently implanted in men than a biological prosthesis (60.4% vs 49.1%, P = .01). In the mechanical prosthesis group, 5.5% of patients had native valve endocarditis compared with 0.6% of patients in the biological prosthesis group (P = .009). There was no statistically significant difference in the other preoperative characteristics and demographics. After applying logistic regression and controlling for propensity score, the mechanical prosthesis still had a lower risk of PPM (odds ratio, 0.02 [95% CI, 0.01-0.03]; P < .001).
The patients’ operative characteristics are presented in Table II. There was no statistically significant difference between the groups in European System for Cardiac Operative Risk Evaluation II, cardiopulmonary bypass, or aortic cross-clamp times. Ejection fraction and mean intensive care unit stays were similar between the groups, as well. Moderate and severe PPM (iEOA <0.85 cm2/m2) were present in 69.8% of patients in the biological prosthesis group compared with 3.7% in the mechanical prosthesis group (P < .001) (Table III). Severe PPM (iEOA <0.65 cm2/m2) was present in 5.1% of patients in the biological group and 1.3% of patients in the mechanical prosthesis group, with a statistically significant difference (P < .001). There was no difference in the number of redo surgeries or the frequency of postoperative endocarditis in the follow-up period between the mechanical and biological prosthesis groups. Freedom from angina at the time of the latest follow-up was also similar between the groups (85.1% vs 88.8%, P = .28).
Separate analysis of PPM with various prosthesis manufacturers and models was conducted in both the mechanical and biological prosthesis groups. In the mechanical valve group, 16 patients had PPM; there was no difference in the frequency of PPM among the different prosthesis models. In the biological valve group, 111 patients had PPM; the prostheses most common found with PPM were the Trifecta (28 [25.2%], P < .001) and Epic valves (25 [22.5%], P < .001) (Table IV).
The median (IQR) follow-up was 31.8 (1-74) months. When the mechanical and biological prostheses were analyzed overall, mean survival was statistically significantly shorter in the PPM group (57.0 months [95% CI, 51.9-62.2]) than in the no-PPM group (65.2 months [95% CI, 63.1-67.4]; log-rank test P = .008) (Fig. 2). When data were analyzed separately for the biological prosthesis group, mean survival was statistically significantly shorter in patients with PPM (50.2 months [95% CI, 45.2-55.3]) than in patients without PPM (60.1 months [95% CI, 55.7-64.4]; log-rank test P = .04) (Fig. 3). After applying Cox regression with propensity score adjustments, the difference was still statistically significant (hazard ratio, 3.27 [95% CI, 1.00-10.91]; P = .046). Analysis of survival in patients who had received a mechanical prosthesis did not show a difference in mean survival between the PPM group (66.6 months [95% CI, 58.3-74.9]) and the no-PPM group (64.9 months [95% CI, 62.6-67.2]; log-rank test P = .50) (Fig. 4). After applying Cox regression with propensity score adjustments, the difference was not statistically significant (hazard ratio, 0.53 [95% CI, 0.07-3.83]; P = .53).
Citation: Texas Heart Institute Journal 50, 5; 10.14503/THIJ-22-8048
Citation: Texas Heart Institute Journal 50, 5; 10.14503/THIJ-22-8048
Citation: Texas Heart Institute Journal 50, 5; 10.14503/THIJ-22-8048
In total, 87.6% of patients responded to the QOL survey. The SF-12 questionnaires were analyzed separately for patients with mechanical and biological prostheses (Table V). After adjusting for age and sex, there was no statistically significant difference between the mean [SD] physical and mental scores in patients with and without PPM who had received a mechanical prosthesis (physical: 43.9 [9.4] vs 46.9 [8.3], P = .34; mental: 54.8 [4.4] vs 53.5 [5.9], P = .16). Similarly, there was no statistically significant difference between the mean (SD) physical and mental scores in patients with and without PPM who had received a biological prosthesis (physical: 39.4 [8.4] vs 45.7 [10.1], P = .29; mental: 53.9 [6.1] vs 53.1 [8.1], P = .43). Although there was a difference in physical scores in patients with biological valves with and without PPM, the difference was nonsignificant after adjusting for age and sex. There was no statistically significant difference between the mental scores of patients with and without PPM regardless of the type of prosthesis implanted (Table V).
Discussion
The present study analyzed the impact of PPM with a mechanical or biological prosthesis on survival and QOL after AVR. Substantial findings included the following: (1) AVR with a mechanical prosthesis had a markedly lower risk of PPM than AVR with a biological prosthesis, (2) patients with moderate to severe PPM and an implanted biological prosthesis were at the highest risk of long-term mortality, and (3) there was no difference in QOL for patients with biological or mechanical prostheses for up to 6 years of follow-up.
Patient-prosthesis mismatch is associated with a higher risk of poor outcomes after AVR, and its prevention is of paramount importance when selecting a surgical heart valve for implantation.14 Valve manufacturers provide iEOA values as the most appropriate measurement for predicting PPM after implantation.15 A cutoff level of iEOA less than 0.85 cm2/m2 has been introduced to define moderate to severe PPM. In the present study population, PPM was found in 21.5% of all patients who underwent AVR. When analyzed separately, however, 69.8% of patients with biological valves implanted had PPM, while just 3.7% of patients with mechanical valves implanted had PPM (P < .001).
The Quebec group brought PPM into the spotlight, publishing several studies that showed statistically significantly reduced long-term survival in patients with PPM.15,16 The Toronto group confirmed their findings.17 In a large study that enrolled 1,856 patients with mechanical prostheses and 2,275 patients with biological prostheses implanted after AVR, the presence of PPM statistically significantly reduced both short-term and long-term survival. The present study supports these findings, showing lower rates of survival in patients with PPM than in patients without PPM. Although none of the previous studies distinguished outcomes by the presence of a biological or a mechanical prosthesis, the present study revealed lower survival rates with PPM for patients with biological prostheses, while PPM with mechanical prostheses did not affect survival or the physical component of a patient's QOL.
Hoffmann et al12 analyzed 632 consecutive patients who underwent AVR procedures with only Hancock II biological prostheses. Patient-prosthesis mismatch was present in 93.8% of patients; 71% of patients had moderate PPM, and 22.8% of patients had severe PPM. The authors found no difference in 5-year survival for the groups with and without PPM. The present study found a similar distribution of moderate PPM in the biological prosthesis group, but severe PPM was found in only 5.1% of patients. Patients with biological prostheses and PPM had statistically significantly lower survival rates after 6 years of follow-up than patients without PPM. One potential explanation for the observed differences is that the present study included several biological prosthesis types and complete follow-up. Sportelli et al4 conducted an observational study that included 152 patients with both mechanical and biological prostheses implanted after AVR. The overall PPM rate was 53%, and 11.7% of patients had severe PPM. The authors reported that PPM had no influence on survival after long-term follow-up, but no separate analyses for biological and mechanical prostheses were reported. In another study, Weber et al11 revealed more frequent PPM in the biological prosthesis group than in the mechanical prosthesis group, but they did not perform a survival analysis for these groups. Severe PPM was rare in the present study, so it was not suitable for the subanalysis of this population. It should be mentioned that none of the listed studies included patients with sutureless biological prostheses, as appeared in the present study.
It should also be mentioned that nearly 37% of the biological valves implanted in the present study's population were Trifecta valves, whose frequent structural valve disease issues have been raised in a few previously published studies. The prosthesis used in these studies, however, was an earlier Trifecta model; the present study used the new Trifecta GT model. The clinical importance of this fact has yet to be determined.18,19 Analysis in the present study revealed a higher frequency of Trifecta and Epic valves in patients with PPM. Larger cohort studies with echocardiograph data as well as EOA and pressure gradients are needed to further facilitate valve choice when PPM is suspected.
Some studies that have enrolled small cohorts have warned about the negative impact of PPM on QOL, especially on the physical component.9,20 Because PPM is associated with higher transprosthetic gradients, with physical exercise, a rise in the gradient may come close to the values shown in mild and moderate native valve aortic stenosis.21 The median values of the QOL measurements in the present study were close to the SF-12's normal values. The values of the QOL mental component did not show a statistically significant difference in either group's patients with or without PPM. The physical component of the questionnaire revealed statistically significantly lower scores in patients with a biological prosthesis and PPM than in patients with a biological prosthesis without PPM. After adjusting for age and sex, however, there was no statistical difference. This difference was not observed in the mechanical prosthesis group, either. The freedom from angina during follow-up also did not differ between the groups. These results are similar to those published by Hoffman et al,12 who found only a difference in the physical component of the QOL survey, and by Urso et al22 (163 patients enrolled), who found lower physical scores in older patients. Sportelli et al4 (152 patients enrolled) and Reskovic Luksic et al23 (46 patients included) failed to demonstrate a difference in QOL for patients who have PPM. Once again, neither of these studies performed subanalyses for biological or mechanical prostheses.
Perioperative results are affected mainly by the type of valve implanted, but the hemodynamic properties of each valve type can also influence the outcome. In every biological prosthesis, structural valve degeneration will happen to some extent over time. Structural valve degeneration in prostheses with PPM and with higher transprosthetic gradients could lead to further augmentation of gradients, especially during physical exercise. The transprosthetic gradients, however, remain the same over time in a mechanical prosthesis.24 Although the surgeon must strive to implant the largest valve possible if PPM is suspected in a biological prosthesis, if this is not possible, then a mechanical valve, a sutureless valve implantation, or a root enlargement procedure should be considered. The root enlargement procedure is a debated contemporary issue. It can be safely performed but requires that both the surgeon and the center where the procedure is performed have advanced experience. According to available publications, however, the volume of procedures does not contribute to operative risk.25 Aortic root enlargement is a valuable tool to help prevent PPM in both mechanical and biological valves. If a risk of PPM is suspected, the surgeon should consider another prosthetic model with a better hemodynamic profile. In the present study, the frequency of PPM with Trifecta and Epic valves raised an additional concern regarding valve choice, but further clarification is needed by way of larger, randomized studies and echocardiographic follow-up. As the clinical indications for transcatheter AVR use expand rapidly, the size of the implanted biological prosthesis should be carefully planned.
Study Limitations
The present study has several substantial limitations. First, the study was observational and lacked randomization. More extensive prospective randomized trials are needed to explore these results and their clinical application. Second, the study was performed in a single center. Third, the response rate for the QOL survey was 87.6%, and data were not available for all patients enrolled in the study. The present study results should only be interpreted as observational and hypothesis-generating because of the nature of their exploratory data analysis.
Conclusions
Patient-prosthesis mismatch is common after biological valve implantation and substantially affects long-term survival and QOL. If the risk of PPM after the implantation of a biological prosthesis is suspected, prospective strategies to avoid PPM at the time of the operation are warranted. Aortic root enlargement or the choice of another prosthetic model with better hemodynamic performance could be considered according to local practice and expertise.
