Prosthesis-Patient Mismatch: The Complex Interaction between Cardiac Output and Prosthetic Valve Effective Orifice Area

Prosthesis-patient mismatch (PPM) is classified as a non-structural valve dysfunction according to Society of Thoracic Surgeons or Valve Academic Research Consortium 3 standardized definitions.1Généreux P Piazza N Alu MC et al.Valve Academic Research Consortium 3: updated endpoint definitions for aortic valve clinical research.Eur Heart J. 2021; 42 (doi:10.1093/eurheartj/ehaa799.): 1825-1857Google Scholar PPM consists of a prosthetic valve that has a normal function but for which effective orifice area (EOA) is too small in relation to patient’s cardiac output requirement. Body surface area (BSA) has been demonstrated to be the main determinant of stroke volume and cardiac output in normal healthy and non-obese subjects. Hence, EOA indexed to BSA is the universally accepted parameter to assess the presence and severity of PPM. Several studies and meta-analyses reported that severe PPM, defined with the use of indexed EOA, is associated with increased risk of mortality, re-hospitalization, and re-intervention following aortic valve replacement (AVR).2Head S Mokhles M Osnabrugge R 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 27, 186 patients with 133, 141 patient-years.Eur Heart J. 2012; 33 (doi:10.1093/eurheartj/ehs003.): 1518-1529Google Scholar,3Pompeu SM Cavalcanti LRP Rayol SDC et al.Prosthesis-patient mismatch negatively affects outcomes after mitral valve replacement: meta-analysis of 10, 239 patients.Braz J Cardiovasc Surg. 2019; 34 (doi:10.21470/1678-9741-2019-0069.): 203-212Google Scholar However, other studies did not find such associations or suggested that these associations may be related to other confounding factors. In this issue of Structural Heart: The Journal of the Heart Team, Vriesendorp et al. evaluated the validity of using BSA as a proxy for cardiac output when assessing bioprosthetic valve hemodynamic function and PPM in the patients included in the PERIcardial SurGical AOrtic Valve ReplacemeNt (PERIGON) Pivotal Trial.4Vriesendorp MGR, Herrmann H, Head S. OnlyThe clinical implications of body surface area as a poor proxy for cardiac output. Struct Heart. 2021. doi:10.1080/24748706.2021.1968089.Google Scholar This analysis included the subset of 744 patients with normal LV function who underwent AVR with the Medtronic Model 400 bovine pericardial bioprosthetic valve. The authors observed a weak correlation between BSA and cardiac output (r = 0.29) and the relationship between cardiac output and BSA was not proportional. Furthermore, patients with a large BSA, classified as having severe PPM on the basis of the indexed EOA, did not have evidence of valve hemodynamic obstruction, defined as mean gradient ≥20 mmHg and/or Doppler velocity index <0.35. The authors thus concluded that the current definition of PPM results in a systematic overestimation of hemodynamic obstruction in patients with a larger BSA and they thus recommend cautious use of indexed EOA in this subgroup. The study by Vriesendorp et al.4Vriesendorp MGR, Herrmann H, Head S. OnlyThe clinical implications of body surface area as a poor proxy for cardiac output. Struct Heart. 2021. doi:10.1080/24748706.2021.1968089.Google Scholar is based on the premise that the BSA is, de facto, a proxy for stroke volume and cardiac output in patients with AS or with an aortic prosthetic valve BSA is indeed a good proxy for cardiac output solely in normal healthy, normal-weight, young or middle age, subjects free of any cardiovascular disease. In the older patients with history of severe AS who underwent AVR, several factors may lead to a dissociation between BSA and stroke volume or cardiac output. First, a substantial proportion of this population is obese and BSA may result in over-indexation of EOA and thus overestimation of PPM in obese people. The solution that has been proposed by European Association of Cardiovascular Imaging or Valve Academic Research Consortium 3 (VARC-3)1Généreux P Piazza N Alu MC et al.Valve Academic Research Consortium 3: updated endpoint definitions for aortic valve clinical research.Eur Heart J. 2021; 42 (doi:10.1093/eurheartj/ehaa799.): 1825-1857Google Scholar to overcome, at least in part, this limitation is to use lower cutoff values of indexed EOA in obese patients (<0.70 vs. 0.85 for moderate PPM and <0.55 vs. 0.65 cm2/m2 for severe PPM) (Figure 1), and this solution has indeed been applied, to some extent, by the authors in the present study. Second, a substantial proportion of patients present with residual or new onset LV systolic dysfunction and ensuing low flow state following AVR. In an attempt to address this issue and restrict their analysis to patients with normal LV function and flow status, the authors excluded patients with severe LV systolic dysfunction. The data presented in their study,4Vriesendorp MGR, Herrmann H, Head S. OnlyThe clinical implications of body surface area as a poor proxy for cardiac output. Struct Heart. 2021. doi:10.1080/24748706.2021.1968089.Google Scholar however, suggest that at least one-third of the patients had stroke volume <50 mL, which implies that a large proportion of the patients in this cohort were in low-flow state relative to their BSA (i.e. stroke volume index <35 mL/m2). Third, although the authors excluded patients with severe LV systolic dysfunction, patients with mild or moderate dysfunction were likely included. Moreover, as shown in several previous studies, a normal LV ejection does not necessarily imply a normal flow. Indeed, up to 30% of patients may present with a paradoxical low-flow state despite preserved LVEF following AVR.5Anjan VY Herrmann HC Pibarot P et al.Evaluation of flow after transcatheter aortic valve replacement in patients with low-flow aortic stenosis. A secondary analysis of the PARTNER randomized clinical trial.JAMA Cardiol. 2016; 1 (doi:10.1001/jamacardio.2016.0759.): 584-592Google Scholar This paradoxical low-flow pattern may be related to multiple factors including pronounced LV concentric remodeling, LV diastolic dysfunction, atrial fibrillation, concomitant mitral regurgitation or tricuspid regurgitation, right ventricular dysfunction, and/or cardiac amyloidosis. Hence, it is not surprising, and rather expected, that the correlation between BSA and cardiac output or stroke volume is weak in a post AVR cohort. The next and really important question then is: “Does the fact that BSA is not a good proxy for cardiac output following AVR, de facto, invalidate the use of indexed EOA for the identification and quantitation of PPM.” The answer is probably “No” because the purpose of the BSA, in the context of AVR, is not to predict the stroke volume or cardiac output actually measured in the patient but rather to predict the normal “ideal” values of these flow parameters assuming the patient would not be overweight or obese and would have normal geometry and function of cardiac chambers and valves. Hence, the BSA is not meant to be a proxy for the actual cardiac output at a given time-point of the AS patient’s journey but more a proxy for the ideal objective of cardiac output that we are aiming to restore with AVR and other adjuvant therapies. Consistently, the purpose of the indexed EOA measured early after AVR is to determine whether the hemodynamic performance of the prosthetic valve is adequate for – or matches – the normal cardiac output requirements of the patient. The main and immediate physiologic/hemodynamic consequence of PPM is the persistence of high residual transaortic gradients. Severe PPM is, in fact, a residual AS and therefore indicates that the hemodynamic results and so the efficacy of AVR are suboptimal. If the indexed EOA is valid to measure PPM, we should expect this metric to accurately predict the presence of high residual gradient or residual AS following AVR. In this regard, the authors showed that the indexed EOA has excellent accuracy to identify valve hemodynamic obstruction with an area under the ROC curve of 0.84. Hence, despite the fact that BSA is not a good proxy for cardiac output, the indexed EOA, nonetheless, remains a reasonable parameter to assess PPM following AVR. However, we concur with the authors of the present study4Vriesendorp MGR, Herrmann H, Head S. OnlyThe clinical implications of body surface area as a poor proxy for cardiac output. Struct Heart. 2021. doi:10.1080/24748706.2021.1968089.Google Scholar that the indexed EOA measured by Doppler-echocardiography following AVR has inherent limitations and generally overestimates the incidence and severity of PPM. The EOA is, indeed, subject to measurement pitfalls and variability and, more importantly, it is flow-dependent. Hence, in the presence of flow-flow state (which is highly prevalent in the present study), the EOA will decrease below its normal value even if the bioprosthetic valve function is normal. Because of this flow-induced reduction in EOA, the indexed EOA is “pseudo-severe” and thus overestimates the severity of PPM (Figure 1). This pseudo-severe PPM phenomenon is similar to the pseudo-severe stenosis phenomenon observed in patients with low-flow, low-gradient native AS. A recent analysis from the PARTNER 2 trial and registry revealed that pseudo-severe PPM is frequent at 30-days following TAVR, and even more following SAVR because of the higher prevalence of low-flow state in this arm.6Ternacle J Pibarot P Herrmann HC et al.Prosthesis-patient mismatch after aortic valve replacement in the PARTNER 2 trial and registry.JACC Cardiovasc Interv. 2021; 14 (doi:10.1016/j.jcin.2021.03.069.): 1466-1477Google Scholar To overcome this issue of PPM overestimation by the measured indexed EOA, it has been proposed to use the predicted indexed EOA, which is the normal reference value of EOA reported in the literature for each given model and size of prosthetic valve, divided by patient’s BSA (Figure 1). The predicted indexed EOA has the advantage over the measured indexed EOA of not being altered by measurement error/variability and by patient’s hemodynamic status, including flow status. In the PARTNER 2 trial, the incidence of severe PPM (using adjusted cut-points in obese patients) in the PARTNER 2 SAVR arm was much lower (1.2% vs. 23.6%) with the predicted versus the measured indexed EOAs and was more strongly and independently associated with mortality and re-hospitalization.7Ternacle J Abbas AE Pibarot P Prosthesis-patient mismatch after transcatheter aortic valve replacement: has it become obsolete?.JACC Cardiovasc Interv. 2021; 14 (doi:10.1016/j.jcin.2021.03.039.): 977-980Google Scholar As highlighted in the present study,4Vriesendorp MGR, Herrmann H, Head S. OnlyThe clinical implications of body surface area as a poor proxy for cardiac output. Struct Heart. 2021. doi:10.1080/24748706.2021.1968089.Google Scholar the BSA is probably not the optimal anthropometric parameter to index the EOA and also cardiac output, LV volumes, LV mass, etc. We should thus consider other alternatives to improve EOA indexation. The use of height may not be the optimal solution because this parameter does not account for the (non-linear) increase in cardiac output that is observed in overweight or obese patients. One potential interesting option would be to index the EOA for the fat-free mass measured by a bio-impedance scaler. Fat free-mass has been shown to be the main determinant of cardiac output in both normal-weight and overweight healthy individual.8Collis T Devereux RB Roman MJ et al.Relations of stroke volume and cardiac output to body composition. The strong heart study.Circulation. 2001; 103: 820-825Google Scholar Another option to explore in future studies is an indexation with the native aortic annulus area measured by 3D echocardiography or contrast-enhanced CT. The EOA indexed to aortic annulus area would therefore provide an assessment of the “EOAbility” of a given model and size of prosthetic valve for a given aortic annulus size. In conclusion, despite its limitations, the EOA indexed to BSA remains the gold standard to identify and quantify PPM following AVR. However, we should use the predicted EOA rather than the measured EOA to determine the indexed EOA and we should apply lower cutoff values of indexed EOA in obese patients to avoid overestimation of PPM (Figure 1). In addition, future studies are needed to validate other anthropometric parameters, i.e. fat-free mass or aortic annulus area, to improve the indexation of EOA and therefore the assessment of PPM. Dr. Pibarot is supported by a Canada Research Chair and Foundation grant (FDN-143225) from Canadian Institutes of Health Research, Ottawa, Ontario, Canada.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.