Search

Background There is an established relationship between the degree of mitral regurgitation (MR) and prognosis. Quantitation of MR severity guides therapeutic approaches. Inconsistent definitions and categorization of MR severity in clinical studies limit meaningful comparisons between trials and compromise development of an effective evidence base. The purpose of this study was to quantify heterogeneity in grading systems for MR severity in the contemporary literature. Methods We performed a systematic review of randomized (RCT) and propensity score (PS) adjusted clinical studies of MV interventions (surgical or percutaneous). A total of 35 articles from 2015-2020 were included (15 RCT, 20 PS). Results There were 22 studies that reported MR severity in numerical categories, either values from the historical “plus” system or numerical MR grades, while 9 studies reported MR severity using text-only descriptive categories. Among the studies that used numerical categories, 2+ MR was defined as moderate in 64% of studies, mild in 27%, and mild-moderate in 9% and 3+ MR was defined as moderate in 14%, moderate-severe in 52%, and severe in 14%. Conclusions There was substantial variability in MR severity definition and reporting in contemporary clinical studies of MV interventions. We recommend the historical “plus” numerical grading system be abandoned and that inclusion and outcome criteria in MR clinical trials be based on American and European guideline-recommended categories as none/trace, mild, moderate, and severe. Adoption of these simple recommendations will improve the consistency and quality of MR clinical trial design and reporting.

Durability of mitral valve repair for ischemic mitral regurgitation (IMR) remains poor. We established a swine model of chronic IMR, and describe the methods and lessons learned from this model.Thirty-five swine underwent percutaneous myocardial infarction with ethanol ablation of the circumflex or obtuse marginal (OM) arteries. Swine were followed with routine echocardiography for the development of severe IMR. Once severe IMR was established, swine underwent mitral valve operations on cardiopulmonary bypass. After operation, swine were survived up to 7 weeks. Angiographic and echocardiographic features of swine who developed severe IMR (IMR swine) and those who did not (non-IMR swine) were compared.The median number of OM arteries was 3, with 2 OM arteries infarcted. Acute survival after the myocardial infarction was 74% (26 of 35) with 3 (9%) early, postoperative deaths. Among the 23 swine with follow-up to determine IMR status, 14 of 23 (61%) developed significant IMR. Among IMR pigs, left ventricular (LV) ejection fraction decreased from 65% pre-myocardial infarction to 45% pre-mitral valve intervention (P.001). Among non-IMR swine, LV ejection fraction decreased nonsignificantly from baseline (60%) to latest follow-up (55%) (P = .443). LV end-diastolic dimension (P = .039), wall motion score (P = .027), global circumferential strain (P = .014), and global longitudinal strain (P = .023) were significantly worse in IMR compared with non-IMR swine.A reproducible percutaneous model of severe IMR in swine is feasible with a guided anesthetic and perioperative approach. This model can serve as a platform to better understand the mechanism of IMR and subsequently to test novel repair techniques.

Resectional and artificial cordal repair techniques are effective strategies for degenerative mitral valve (MV) repair. However, resectional repair requires a tailored approach using various techniques, whereas cordal repair offers a simpler, easily reproducible repair. The approach described in this study approach has evolved from resectional to cordal over time, and outcomes are compared between the eras.Clinical and echocardiographic outcomes of all patients undergoing MV repair for degenerative mitral regurgitation (MR) from January 2004 to September 2017 were reviewed. Patients were stratified by era: from January 2004 to June 2011 (era 1; n = 405), resectional techniques were used in 62% and artificial cordal techniques were used in 38%. From July 2011 to September 2017 (era 2; n = 438), artificial cordal repair was used in 98% of patients. The primary outcome was repair failure, defined as greater than moderate MR or MV reoperation.Of 847 patients with degenerative MR, successful repair was achieved in 843 patients (99.5% repair rate). Leaflet prolapse was posterior in 66%, anterior in 8%, and bileaflet in 26%. Cardiopulmonary bypass time and cross-clamp times were shorter in era 2 (CPB: 109 [IQR, 92-128] minutes vs 97 [IQR, 76-121] minutes; P.001; cross-clamp: 88 [IQR, 73-106] minutes vs. 79 [IQR, 61-99] minutes; P.001). Predismissal echocardiography demonstrated no MR or trace MR in 95%, mild MR in 4.7%, and moderate MR in 0.3% of patients. Operative mortality was similar in the eras (0.5% vs 0.5%; P.999). The rates of 5-year freedom from repair failure (95.1% vs 95.5%; P = .707), stroke (96.8% vs 95.3%; P = .538), and endocarditis (99.3% vs 99.7%; P = .604) were similar between the eras.Artificial cordal repair for all patients with degenerative MR simplifies MV repair and yields equivalent, excellent outcomes compared with a tailored resectional approach.

Background Glutaraldehyde-fixed autologous or bovine pericardial patches used for mitral valve leaflet reconstruction have been associated with late calcification. Fresh autologous pericardium (FAP) may be a durable alternative. Methods Transthoracic echocardiography was used to assess valve function (regurgitation, mean pressure gradient, patch pliability, and calcification) in patients undergoing FAP mitral leaflet repairs. Pliability was scored between 1 (similar to native leaflets) and 4 (rigid). Calcification was scored between 1 (echobrightness similar to native leaflets) and 4 (very bright). Results Between 2002 and 2018, 62 consecutive patients (50% male, 51 ± 2 years, 69% infective endocarditis) underwent mitral valve repair with FAP, and Patch placement was on the anterior (31 of 62), posterior (27 of 62), or both (1 of 62) leaflets. Late echocardiographic follow-up was available for 43 of 62 patients (median follow-up, 3.6 years; range, 0.5-6 years). Average pliability scores were unchanged between discharge (1.2 ± 0.1) and follow-up (1.2 ± 0.2, P = .79). Average brightness scores increased modestly (predischarge, 1.6 ± 0.1; follow-up, 1.8 ± 0.1; P = .01). Three patients had recurrent severe mitral regurgitation, and 2 underwent reoperation, 1 at 1 year postoperatively for recurrent endocarditis and 1 at 6 years postoperatively for degenerative disease progression. At reoperation, patches were pliable, free from calcification, and comparable in thickness to adjacent native leaflet. One patient developed suture line leak, which was repaired. No other evidence of patch dehiscence, retraction, or aneurysm was observed. The 10-year freedom from reoperation of 82% and survival rate of 84% are comparable to repair with glutaraldehyde-fixed or bovine pericardial patches. Conclusions FAP is an excellent substrate for complex mitral valve leaflet patch repairs and can be used with the expectation of durable, long-term valve function, without evidence of late patch calcification, stiffness, or aneurysmal degeneration.

Data from The Society of Thoracic Surgeons Adult Cardiac Surgery Database were analyzed to identify trends in patient characteristics and outcomes of mitral valve operations in North America.All patients with isolated primary mitral valve operations with or without tricuspid valve repair, surgical atrial fibrillation ablation, or atrial septal defect closure performed July 2011 to September 2016 were identified. A subgroup analysis assessed patients with degenerative leaflet prolapse (DLP).Isolated primary mitral valve operations were performed on 87,214 patients at 1,125 centers, increasing by 24% between 2011 (n = 14,442) and 2016 (n = 17,907). The most common etiology was DLP (60.7%); 4.3% had functional mitral regurgitation. Preoperatively, 47.3% of patients had an ejection fraction less than 60% and 34.2% had atrial fibrillation. Overall mitral valve repair rate was 65.6%, declining from 67.1% (2011) to 63.2% (2016; p 0.0001). Repair rates were related to etiology (DLP, 82.5%; rheumatic, 17.5%). Of the 29,970 mitral valve replacements, 16.2% were preceded by an attempted repair. Repair techniques included prosthetic annuloplasty (94.3%), leaflet resection (46.5%), and artificial cord implantation (22.7%). Bioprosthetic valves were implanted with increasing frequency (2011, 65.4%; 2016, 75.8%; p 0.0001). Less-invasive operations were performed in 23.0% and concomitant tricuspid valve repair in 15.7%. Unadjusted operative mortality was 3.7% (replacements) and 1.1% (repairs).Patients undergoing primary isolated mitral valve operations commonly have ventricular dysfunction, atrial fibrillation, and heart failure. Although contemporary outcomes are excellent, earlier guideline-directed referral and increased frequency and quality of repair may further improve results of mitral valve operations.

Background To determine if preoperative embolic stroke is associated with an increased risk of postoperative stroke among patients undergoing early operation for mitral valve (MV) infective endocarditis (IE), we compared outcomes among patients presenting with and without acute stroke. Methods From 2003 to 2015, 243 consecutive patients underwent surgery for active MV IE. Patients were categorized into 2 groups: 72% (174 of 243 patients) with no preoperative acute stroke (clinical, radiographic or both) and 28% (69 of 243 patients) with stroke. Both preoperative and postoperative strokes were confirmed in all patients with brain computed tomography or magnetic resonance imaging and comprehensive examination by a neurologist. Results Among patients presenting with stroke, 33% (23 of 69 patients) were asymptomatic and had only positive imaging findings. The median time from admission to operation was 5 days. The overall rate of new postoperative stroke was 4% (10 of 243 patients). The rate of postoperative stroke was not different between the 2 groups: 4% (7 of 174 patients) among patients with no preoperative stroke and 4% (3 of 69 patients) with stroke ( p = 0.9). One patient developed a hemorrhagic conversion of an acute infarct. Operative mortality was 7% (13 of 174 patients) among patients with no preoperative stroke and 7% (5 of 69 patients) among patients with stroke ( p = 0.9). Conclusions MV surgery for IE and acute stroke can be performed early with a low risk of postoperative neurologic complications. When indicated, surgical intervention for MV IE complicated by acute stroke should not be delayed.

Investigators seeking to select an animal model for use in preclinical research studies prior to FDA submission for allograft heart valves have several types to choose from. Dogs, pigs, cattle, primates and sheep have all served as successful medium and large animal models, and their murine counterparts have also been integral in the advancement of replacement heart valve research. While the national and international regulatory bodies have not specified a universal animal model for use in preclinical research studies, the ovine and porcine models have become the frontrunners within the peer-reviewed literature. Sheep are an excellent model of bioprosthetic valve calcification, with a robust mineralization response that mimics observations of human clinical disease progression, while swine are the model of choice for valves which pose a risk of thrombotic events. In the rapidly advancing field of tissue engineered cardiovascular products, dogs, primates, rodents, sheep and pigs are all valuable models for the study of scaffold remodeling and recellularization. Once an animal model has been chosen, investigators are recommended to consult with the FDA via submission of a Request for Designation in order to identify the type of product and the appropriate Center to which the product should be submitted. Additionally, the investigator should outline a risk analysis plan that categorizes the failure modes of the product, as well as design an in vivo pre-clinical safety and performance study that will capture data applicable to the assessment of the failure modes enumerated in the risk analysis.

Background Because of cryopreserved heart valve–mediated immune responses, decellularized allograft valves are an attractive option in children and young adults. The objective of this study was to investigate the performance and morphologic features of decellularized pulmonary valves implanted in the right ventricular outflow tract of juvenile sheep. Methods Right ventricular outflow tract reconstructions in juvenile sheep (160 ± 9 days) using cryopreserved pulmonary allografts (n = 6), porcine aortic root bioprostheses (n = 4), or detergent/enzyme–decellularized pulmonary allografts (n = 8) were performed. Valve performance (echocardiography) and morphologic features (gross, radiographic, and histologic examination) were evaluated 20 weeks after implantation. Results Decellularization reduced DNA in valve cusps by 99.3%. Bioprosthetic valves had the largest peak and mean gradients versus decellularized valves ( p = 0.03; p p = 0.01; p = 0.001), which were similar ( p = 0.45; p = 0.40). Regurgitation was minimal and similar for all groups ( p = 0.16). No cusp calcification was observed in any valve type. Arterial wall calcification was present in cryopreserved and bioprosthetic grafts but not in decellularized valves. No autologous recellularization or inflammation occurred in bioprostheses, whereas cellularity progressively decreased in cryopreserved grafts. Autologous recellularization was present in decellularized arterial walls and variably extending into the cusps. Conclusions Cryopreserved and decellularized graft hemodynamic performance was comparable. Autologous recellularization of the decellularized pulmonary arterial wall was consistently observed, with variable cusp recellularization. As demonstrated in this study, decellularized allograft valves have the potential for autologous recellularization.

Background Cardiac allometric organ growth after pediatric valve replacement can lead to patient–prosthesis size mismatch and valve re-replacement, which could be mitigated with allogeneic decellularized pulmonary valves treated with collagen conditioning solutions to enhance biological and mechanical performance, termed "bioengineered valves." In this study, we evaluated functional, dimensional, and biological responses of these bioengineered valves compared with traditional cryopreserved valves implanted in lambs during rapid somatic growth. Methods From a consanguineous flock of 13 lambs, the pulmonary valves of 10 lambs (mean weight, 19.6 ± 1.4 kg) were replaced with 7 bioengineered valves or 3 classically cryopreserved valves. After 6 months, the 10 lambs with implanted valves and 3 untreated flock mates were compared by echocardiography, cardiac catheterization, and explant pathology. Results Increases in body mass, valve geometric dimensions, and effective orifice areas were similar in the 2 groups of lambs. The bioengineered valves had higher median cusp-to-cusp coaptation areas (34.6%; interquartile range, 21.00%-35.13%) and were more similar to native valves (43.4%; interquartile range, 42.59%-44.01%) compared with cryopreserved valves (13.2%; interquartile range, 7.07%-13.91%) ( P = .043). Cryopreserved valves cusps, but not bioengineered valve cusps, were thicker than native valve cusps ( P = .01). Histologically, cryopreserved valves demonstrated less than native cellularity, whereas bioengineered valves that were acellular at the time of surgery gained surface endothelium and subsurface myofibroblast interstitial cells in pulmonary artery, sinus wall, and cusp base regions. Conclusions Biological valve conduits can enlarge via passive dilatation without matrix synthesis, but this would result in decreased cusp coaptational areas. Bioengineered valves demonstrated similar annulus enlargement as cryopreserved valves but usually retained larger areas of cuspal coaptation. Heat-shock protein 47-positive (collagen-synthesizing) cells were present in previously acellular bioengineered sinus walls and cusp bases, but rarely in more distal cusp matrices.

Objective This study assesses in a baboon model the hemodynamics and human leukocyte antigen immunogenicity of chronically implanted bioengineered (decellularized with collagen conditioning treatments) human and baboon heart valve scaffolds. Methods Fourteen baboons underwent pulmonary valve replacement, 8 with decellularized and conditioned (bioengineered) pulmonary valves derived from allogeneic (N = 3) or xenogeneic (human) (N = 5) hearts; for comparison, 6 baboons received clinically relevant reference cryopreserved or porcine valved conduits. Panel-reactive serum antibodies (human leukocyte antigen class I and II), complement fixing antibodies (C1q binding), and C-reactive protein titers were measured serially until elective sacrifice at 10 or 26 weeks. Serial transesophageal echocardiograms measured valve function and geometry. Differences were analyzed with Kruskal–Wallis and Wilcoxon rank-sum tests. Results All animals survived and thrived, exhibiting excellent immediate implanted valve function by transesophageal echocardiograms. Over time, reference valves developed a smaller effective orifice area index (median, 0.84 cm 2 /m 2 ; range, 1.22 cm 2 /m 2 ), whereas all bioengineered valves remained normal (effective orifice area index median, 2.45 cm 2 /m 2 ; range, 1.35 cm 2 /m 2 ; P = .005). None of the bioengineered valves developed elevated peak transvalvular gradients: 5.5 (6.0) mm Hg versus 12.5 (23.0) mm Hg ( P = .003). Cryopreserved valves provoked the most intense antibody responses. Two of 5 human bioengineered and 2 of 3 baboon bioengineered valves did not provoke any class I antibodies. Bioengineered human (but not baboon) scaffolds provoked class II antibodies. C1q + antibodies developed in 4 recipients. Conclusions Valve dysfunction correlated with markers for more intense inflammatory provocation. The tested bioengineering methods reduced antigenicity of both human and baboon valves. Bioengineered replacement valves from both species were hemodynamically equivalent to native valves.