Abstract
Objective
Mechanical prosthetic valves are the main option for aortic valve replacement (AVR) in the young population. Inspiris Resilia tissue valves with a novel tissue treatment technology are slowly taking the place of mechanical valves in many centers across the world. We compared the early outcomes between Inspiris Resilia and mechanical valves in a propensity-matched cohort of young patients undergoing AVR.
Methods
Using retrospective data collection, we compared 95 patients with an Inspiris valve (I) with 69 patients receiving a mechanical bileaflet valve (M). Propensity matching gave rise to 40 matched patients in each group.
Results
Mean age of the patients was 56 years in both matched cohorts. Preoperative and operative characteristics of the matched groups of patients showed no significant difference. The in-hospital mortality was 0% in both groups, but a significantly longer hospital stay was observed in the M-group (p = .038). During the follow-up, a higher incidence of bleeding (12.5% vs 2.5%) and systemic embolization (5% vs 2.5%) was seen in the M-group. Both were not significant due to the low event numbers. Last follow-up echocardiographic data showed slightly higher (but non-significant) peak and mean gradients in the I-group compared to the M-group (25 vs 19 mmHg, p = .09) and (16 vs 10 mmHg, p = .06), respectively.
Conclusion
The Inspiris Resilia could be a good alternative to the mechanical valves in aortic position, with less postoperative hospital stay, less risk of mid-term thromboembolic and bleeding complications. Randomized controlled studies and registries with longer follow-up are needed for more reliable results.
This is a visual representation of the abstract.
Keywords
Background
Surgical aortic valve replacement (SAVR), either isolated or combined, is considered the gold standard management of degenerative aortic valve disease. Prosthetic aortic valves can be either mechanical or biological. Compared to mechanical valves, biological valves are considered a good option for SAVR as it eliminates the need for lifelong therapeutic anticoagulation with its side effects. 1
In young patients, early degeneration of bioprostheses is considered the main drawback of its use in that age group. American College of Cardiology/American Heart Association guidelines have recently lowered the reasonable age for implantation of a bioprosthetic SAVR from 60 to 50 years old. 2
New preservation technologies for xeno-pericardial prostheses have been implemented to improve long-term durability and hence increase the ability of its implantation in lower age group of patients. 1
Inspiris Resilia aortic valve (Edwards Lifesciences LLC, Irvine, USA) is a new bovine pericardial prosthesis for SAVR. It has been available in the European market since 2017. At this moment, over 250,000 Inspiris valves have been implanted worldwide, and it is becoming one of the most frequently used tissue valves. Inspiris is made of glutaraldehyde-fixed tissue (to prevent oxidation and calcification) and also treated with a permanent capping of free residuals (to reduce free aldehyde remnants) and is further treated by a new process of glycerolization that allows dry storage. Moreover, an expandable stent for smaller-sized annular diameters was incorporated in its manufacture, and that allows the future option of an optimized (potentially one-size-upscaling) valve-in-valve procedure should structural valve deterioration occur. 3
In-vivo animal models revealed a significant reduction in calcification, and the initial few clinical studies have shown promising follow-up results.4,5
Many other clinical studies have shown promising short- and midterm results of its use in both young and old age groups.6,7 Other studies have favorable early and midterm hemodynamic outcomes compared to Magna Ease.3,8
Since the initial clinical studies showed promising results, we compared the early outcomes between Inspiris Resilia and mechanical valves in a propensity-matched cohort of young patients undergoing isolated or combined SAVR.
Patients and Methods
This is a single-center experience in which we retrospectively reviewed 95 patients who were operated for SAVR with Inspiris valve in the period from August 2017 to September 2021. Forty patients of them (I-group) were matched with another 40 patients who received mechanical SAVR (M-group). In the mechanical valve group, valves were either St. Jude Medical, On-X, or ATS bileaflet valves.
To limit confounding, patients were matched using a 1:1 nearest neighbor method without replacement, with a caliper of 0.026 of the pooled standard deviation of the logit PS, using the “MatchIt” package in RStudio (R Foundation for Statistical Computing, Vienna, Austria). Covariates used in the model include age, sex, EuroSCORE II, endocarditis, weight, height, New York Heart Association (NYHA) class, Canadian Cardiovascular Society grade 4 (CCS4), previous cardiac surgery, chronic obstructive pulmonary disease (COPD), insulin-dependent diabetes mellitus (IDDM), renal impairment, dialysis, neurologic motor dysfunction, extra-cardiac arteriopathy, recent myocardial infarction (MI), critical state, urgency, and weight of intervention. Matched data were compared using the log-rank (Mantel-Cox) test, paired student's t-test, Wilcoxon signed rank test, or the McNemar-Bowker test.
We recorded the postoperative hospital stay as a primary outcome and thromboembolic complications and hemodynamics over the prosthetic valve as secondary outcomes.
The sample size calculation was done by G*Power 3.1.9.2 (Universitat Kiel, Germany). We performed a pilot study (5 cases in each group), and we found that the mean (±SD) of hospital stay (the primary outcome) was 7.8 ± 1.4 days in the Inspiris Resilia group and 10.4 ± 4.3 days in the mechanical valves group. The sample size was based on the following considerations: 0.813 effect size, 95% confidence level, 90% power of the study, group ratio 1:1, and 7 cases were added to each group to overcome dropout. Therefore, we recruited 40 patients in each group.
Postoperatively, patients in the I-group were managed by platelet anti-aggregation only in the form of low-dose aspirin, without systemic anticoagulation unless indicated for another reason (eg, postoperative atrial fibrillation, where patients received vitamin K antagonists or Non-vitamin K antagonist Oral AntiCoagulants (NOACs)).
On the other hand, all patients in the M-group received systemic anticoagulation in the form of vitamin K antagonists, with targeted international normalized ratio (INR) (2-2.25). Low-dose aspirin was added if indicated for associated procedures (eg, associated coronary artery bypass grafting).
For those with an on-X mechanical valve, the targeted INR can be lowered to 1.5 to 2.0 after 3 months postoperatively but must be associated with low-dose aspirin.
During the hospital stay, data was collected to assess the length of intensive care unit (ICU) and hospital stay, in-hospital mortality, and morbidities, including aortic valve-related re-intervention, thromboembolic complications, and anticoagulation-related bleeding.
Before discharge from the hospital, all patients were assessed by echocardiography to evaluate the hemodynamic measurements across the aortic prosthesis, including the peak and mean gradients, effective orifice area, and degree of aortic incompetence, as well as the ejection fraction.
Postoperative mortality was recorded. Patients were assessed in the last follow-up visit for the incidence of postoperative complications, including endocarditis, thromboembolic events, and anticoagulation-related bleeding, as well as the echocardiographic assessment of the hemodynamics across the aortic prosthesis.
The mean follow-up duration was 1.5 years.
The research was subjected to the Ethical Committee approval in the study center (S61955).
Considering the retrograde nature of the research, we couldn’t get an informed consent, for participation in the research, from the patients.
Results
Mean age of the patients was 56 years in both matched cohorts.
Preoperative characteristics, including age, gender, aortic valve pathology, and logistic EuroSCORE II risk calculations, were comparable (Table 1).
Preoperative Characteristics.
Data presented as mean ± SD, median (IQR), or n (%).
Abbreviations: AV, aortic valve; AS, aortic valve stenosis; AI, aortic valve insufficiency; n, number; m, male; NA, not applicable; IQR, interquartile range.
Similarly, the operative characteristics including the surgical access, associated procedures, implant size, and cardiopulmonary bypass and cross-clamp time of the matched groups of patients, showed no significant difference (Table 2).
Procedural Data.
Data presented as mean ± SD, median (IQR), or n (%).
Abbreviations: AV, aortic valve; RAT, right anterior thoracotomy; n, number; min, minute; IQR, interquartile range.
In-hospital mortality was 0% in both groups. During the follow-up period, there is one mortality in each group (Table 3). We observed a shorter total hospital stay in the I-group (p = .038) (Figure 1).
Hospital stay (days).
Postoperative Data.
Data presented as mean ± SD, median (IQR), or n (%).
Abbreviations: AV, aortic valve; AI, aortic valve insufficiency; ICU, intensive care unit; NA, not applicable; TIA, transient ischemic attack; EOA, effective orifice area; IQR, interquartile range.
Last Follow-up Data.
Data presented as mean ± SD, median (IQR), or n (%).
Abbreviations: AI, aortic valve insufficiency; NA, not applicable; TIA, transient ischemic attack; EOA, effective orifice area; IQR, interquartile range.
During the hospital stay, there was no significant difference between both groups regarding length of stay in the ICU and postoperative complications, including re-intervention for bleeding, transient ischemic attack (TIA), stroke, and peripheral embolization, as well as pre-discharge echocardiographic data, including gradient across the valve, degree of prosthetic valve incompetence, and ejection fraction (Table 3).
During the follow-up period, a higher incidence of bleeding was observed in the M-group compared to the I-group (12.5% vs 2.5%, respectively) (Figure 2). Four patients in the M-group had anticoagulation-related problems in the form of recurrent gastro-intestinal tract (GIT) bleeding, transient hematuria, pseudoaneurysm formation in the groin, and recurrent bruising. One patient in the M-group had intracranial hemorrhage after a head trauma (4 years postoperatively). One patient in the I-group developed hemorrhagic pericardial effusion that was developed 2 months after discharge from the hospital. This patient was on NOACs because of postoperative newly developed atrial fibrillation (p = .2) (Table 4).
Last follow-up data.
Similarly, a higher embolic complication was observed in the M-group which included 3 patients (7.5%) with embolic manifestations compared with the I-group where none of the patients had thrombo-embolic incidents (Figure 2).
One patient (2.5%) in the M-group had a cerebral ischemic stroke 2 years postoperatively. Meanwhile, none in the I-group had such complications (p = non significant (NS)).
Moreover, 2 patients (5%) in the M-group had peripheral embolization, one developed renal and splenic infarctions 2.5 years postoperatively and the other one had coronary embolization and acute embolic anterior MI 6 years postoperatively. On the other hand, none of the I-group had peripheral embolization (p = .5).
Three patients in the M-group (7.5%) had TIA in the form of transient hemiparesis, diplopia, and syncopal attack without obvious organic lesions. Whereas 2 patients in the I-group (5%) had TIA in the form of transient left hemihypoesthesia and left hemichorea with no apparent explanatory structural cause on brain CT scan (p = NS).
Valve dysfunction was observed in 1 patient in each group. In the I-group, 1 patient had structural valve dysfunction (SVD) in the form of iatrogenic perforation of the leaflets caused by Cor-Knot clips that was used for valve placement 3 years earlier and was reoperated because of severe aortic valve insufficiency. One patient in the M-group had developed slight restriction of the movement of one of the mechanical leaflets that was diagnosed after embolic acute MI 6 years postoperatively.
Last follow-up echocardiographic data showed slightly higher peak and mean gradients across the prosthesis in the I-group compared to the M-group (25 vs 19 mmHg, p = .09) and (16 vs 10 mmHg, p = .06), respectively, although not statistically significant (Figure 3).
Last follow-up echocardiography.
One patient in the M-group developed prosthetic valve endocarditis 2 months after surgery and was reoperated with re-replacement with a freestyle aortic root. In the I-group, none of the patients developed endocarditis till the closure date of the study (p = .3).
Discussion
The age of the patients requiring aortic valve replacement (AVR) is a main determinant of the prosthetic valve type, being either a mechanical or a biological valve. In young adults, traditionally, a mechanical valve was preferred to biological valves in the aortic position because of the lower incidence of the need for re-replacement (6.9% vs 12.1% after 15 years), with the associated risk of lifelong anticoagulation with vitamin K antagonists that can be life-threatening in some cases. 9 Although also in mechanical valves there is a risk of redo surgery, which can be due to thrombus formation, pannus overgrowth, or vegetation development causing valve leaflet dysfunction.
The high risk of the lifelong use of anticoagulants after mechanical aortic valve implantation has directed the American guidelines on valvular heart disease to recommend the use of mechanical valves in patients younger than 50 years (class IIa, level of evidence (LOE) B- non randomized studies (NR)). 2 Whereas the European guidelines recommend the use of mechanical valve below the age of 60 years (class IIa, LOE B). 10
For patients between 50 and 65 years of age, it is reasonable to individualize the choice of either a mechanical or bioprosthetic AVR with consideration of individual patient factors and after informed shared decision-making (class IIa, LOE B-NR). 2
It is well known that tissue valves possess many advantages over a mechanical valve, like freedom to live a more active lifestyle, fewer dietary restrictions, and no need for long-term anticoagulants. The newly developed Inspiris Resilia aortic valve (Edwards Lifesciences LLC, Irvine, USA) is a new bovine pericardial prosthesis for SAVR that proved to have lower calcification in an animal model.4,5 Several clinical trials are ongoing and show promising data.6,7
Some studies reported a favorable and stable hemodynamic profile of the Inspiris Resilia aortic valve during 1 year of follow-up with significant left ventricular reverse remodeling as early as 30 days, after its implantation in both aortic stenosis and regurgitation. 11
Meta-analyses and systematic reviews compared the outcomes of mechanical AVR with that of biological AVR and revealed that the clinical outcomes in long-term survival and comorbidities were in favor of mechanical valves in patients aged 50 to 70 years with no contraindications for long-term anticoagulation use.12–14
Previous studies that compared Inspiris Resilia with mechanical valve in aortic position aimed to detect the cost-effectiveness of Inspiris Resilia implantation in adults undergoing AVR. One study proved that Inspiris Resilia group of patients have lower hospital stay compared to mechanical AVR in adults aged 40 to 65 years old (6.6 days vs 10.3 days) with consequently lower hospital costs. 15
Another exploratory analysis was conducted in the United Kingdom, suggesting that the novel tissue valve Inspiris Resilia could be a cost-effective intervention for people over the age of 55 with AS who are suitable for surgical AVR. 16
Since Inspiris-Resilia is a recently introduced valve for clinical use, so there is no long-term data supporting its long-term durability.
In our study, we compared the in-hospital and midterm results of the Inspiris valve versus mechanical valves in the aortic position in young age group of patients.
Hospital Stay
We didn’t detect a significant difference between both groups regarding the ICU stay and postoperative complications, including reintervention for bleeding, TIA, stroke, and peripheral embolization. Also, the predischarge echocardiographic data, including gradient across the valve, degree of prosthetic valve incompetence, and ejection fraction, did not differ between the tissue and the mechanical valve. Both valves deliver a good performance. But the total hospital stay (in days) was significantly lower in the I-group compared to the M-group (p = .038), which can be explained by longer initiation time needed for adjustment of the dose of vitamin K antagonists required to reach the targeted INR in the M-group.
Anticoagulation-Related Complications
We observed a higher incidence of anticoagulant-related complications in the M-group compared to the I-group (12.5% vs 2.5%). Four patients in the M-group had anticoagulation-related problems. Only one patient in the I-group developed hemorrhagic pericardial effusion that was developed 2 months after discharge from the hospital.
Embolic Complications
Higher embolic complications were observed in the M-group which included 3 patients with embolic manifestations compared with the I-group where none of the patients had thrombo-embolic incidents.
One patient (2.5%) in the M-group had a cerebral ischemic stroke 2 years postoperatively. Meanwhile, none in the I-group had such complications (p = NS).
The higher incidence of embolic manifestations in the M-group is due to the higher thrombogenic nature of the mechanical valve than the biological one, with the higher risk of thrombus formation in those patients because the anticoagulant status of that group of patients depends on different factors other than vitamin K antagonists (eg diet, other medications that counteract the effect of vitamin K antagonists).
Valve Dysfunction
Valve dysfunction was observed in 1 patient in each group. In the I-group, 1 patient had SVD in the form of iatrogenic perforation of the leaflets caused by Cor-Knot clips.
One patient in the M-group had developed slight restriction of the movement of one of the mechanical leaflets that was diagnosed after embolic acute MI that could be explained as detached thrombus that was developed on the mechanical valve.
Last Follow-up Echocardiographic Data
Showed slightly higher peak and mean gradients across the prosthesis in the I-group compared to the M-group (25 vs 19 mmHg, p = .09) and (16 vs 10 mmHg, p = .06), respectively, although not statistically significant.
We have 1 patient in the I-group who developed SVD as described before with the resultant moderate to severe aortic regurgitation that led to an increase in the stroke volume and hence a relatively higher gradient over the prosthetic valve (Pg = 65 mmHg, Mg = 32 mmHg). Because of the relatively small number of the studied group, that outlier could be a significant factor participating in the higher gradient over the valve in the I-group.
Study Limitations
The number of patients enrolled in the study is relatively small; a larger cohort would probably lead to statistically relevant differences in observed outcomes.
Also, the data was collected retrospectively, so that a propensity-matched technique was adopted to reduce the differences between both groups. A prospective assessment of this cohort of patients is the main objective of the ongoing INDURE registry dataset.
Conclusion
The Inspiris Resilia could be a good alternative to the mechanical valves in young patients undergoing AVR, with less postoperative hospital stay, less risk of mid-term thromboembolic and bleeding complications. Randomized controlled studies and registries with longer follow-up are needed for more reliable results.
Footnotes
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
