An unusual case of maxillary sinus carcinoma presenting with heart metastasis is reported. The epicardium, myocardium and endocardium were infiltrated with the tumor, but no evidence of recurrent maxillary sinus carcinoma and other organ metastasis was found. Surgical excision was not possible due to massive invasion of the heart by the tumor mass. The right ventricular inflow tract obstruction was relieved by surgical dilatation.

Background and aim of the study: In investigating the mechanisms of aortic valve disease processes and to accurately construct tissue-engineered heart valve prostheses, a complete understanding of the native interstitial cell population is required. Previously, autopsy samples have been deemed unsuitable for immunocytochemical studies due to the ischemic time before harvesting. In this study, the viability and phenotypic profile of cells explanted from normal porcine heart valves up to 120 h post mortem was examined. Methods: The aortic valve leaflets of porcine hearts were excised at 0, 6, 24, 48, 72, 96 and 120 h post mortem; one half of the tissue was used to explant cells, and the other half was fixed in 3.7% formalin for sectioning. Samples taken at each time point were cultured and cell viability was determined using a trypan blue exclusion assay. Immunocytochemical and immunohistochemical analyses, using specific markers for fibroblasts, myofibroblasts and smooth muscle cells, were used to compare cell phenotypes both in vitro and in situ. Results: Absolute numbers of cells obtained from each leaflet decreased significantly over time; however, cell viability in culture was unaffected up to 96 h. At each time point, explanted cell populations expressed similar phenotypes when compared with histological samples prepared from the same valves. Conclusion: Porcine aortic valve interstitial cells may be explanted up to and including 96 h post mortem, with no statistically significant change in cell viability in vitro, and with a population that phenotypically resembles aortic valve interstitial cells in situ. These data suggest that human aortic valve interstitial cells may be successfully harvested at autopsy for in vitro studies.

Background and aims of the study: Tissue turnover is one of many factors involved in the operational longevity of heart valves. An understanding of how valves remodel their matrix in response to the hemodynamic environment in health and disease is crucial to the design and biological responsiveness of tissue-engineered valve substitutes. Matrix metalloproteinases (MMPs) are proteolytic enzymes involved in matrix remodeling in several tissues, and include interstitial collagenase (MMP-1, MMP-13), the gelatinases (MMP-2, MMP-9) and stromelysin (MMP-3). Methods: Expression of MMPs and their tissue inhibitors (TIMPs) in human aortic, mitral, tricuspid and pulmonary valves from unused donor or transplant recipient hearts was determined by immunohistochemical staining using antibodies against human MMP-1, MMP-2, MMP-3 and MMP-9 and their inhibitors TIMP-1, TIMP-2, TIMP-3. Cell identification was achieved using antibodies against CD31(endothelial cells), smooth muscle alpha-actin (microfilaments) and CD68 (macrophages). Results: MMP-1 was expressed in all valves, whereas MMP-2 was only expressed in aortic and pulmonary leaflets. MMP-3 and MMP-9 were not expressed. TIMP-1 and TIMP-2 were expressed in all leaflets, whereas TIMP-3 was observed only in tricuspid leaflets. Conclusion: Valves have a specific pattern of expression of MMPs and TIMPs, which appears to vary in different heart valves. The functional implications and central mechanisms responsible require further study. These findings have important implications in understanding the dynamic nature of valve remodeling and in aiding the development of tissue-engineered valves.

Background and aims of the study: Prosthetic heart valve placement in young women does not normally prevent successful pregnancy, though either thrombosis in mechanical prostheses or degeneration in biological prostheses may cause risks for both mother and fetus. Although pericardial prostheses avoid the risks of anticoagulation therapy, it has been suggested that pregnancy accelerates structural degeneration in bioprostheses. The study aim was to assess the influence of pregnancy on five-year outcome after pericardial bioprosthesis replacement. Methods: A total of 85 women was studied prospectively over five years from prosthesis implantation, using annual clinical and echocardiographic investigations. Women were allocated to two groups based on pregnancy occurring during follow up: 48 patients (56%; group A) became pregnant, and 37 (44%; group B) did not. Results: At five-year follow up, the clinical events occurred in 29 patients (34.1%): 11 (22.9%) women in group A (mean (+/-SE) linearized rate 3.6+/-0.24%/patient-year (pt-yr)), and 18 (45.6%) in group B (linearized rate 8.6+/-0.30%/pt-yr). Probability of freedom from clinical events was higher (p = 0.0096) in group A (77.9+/-6.1 versus 51.3+/-8.2%). Structural bioprosthesis degeneration, thrombosis or infective endocarditis occurred in 13 patients (27.1%) in group A (linearized rate 2.75+/-0.20%/pt-yr) and 11 (29.7%) in group B (linearized rate 2.5+/-0.18%/pt-yr). Probability of structural bioprosthesis degeneration (e.g. stenosis, calcification or rupture) did not differ (p = 0.6440) between groups (85.4+/-5.1 versus 89.5+/-5.1%), even after age-adjusted analysis (p = 0.3112). Conclusion: Structural changes found at five years after bioprosthesis placement may be attributed to the natural course of the bioprosthesis, and independent of any occurrence of pregnancy.

Background and aim of the study: The ideal prosthesis for aortic valve replacement in the small annulus remains controversial, and has yet to be defined. In previous studies, the Medtronic Hall (MH) tilting disc valve showed superior hemodynamic performance in the hemodynamically optimum orientation compared to the St. Jude Medical (SJM) bileaflet valve, especially in smaller-sized valves. Using an animal model, the hemodynamics of 21 mm MH and 23 mm SJM valves, both of which have shown identical performance in previous clinical studies, were compared. Methods: A rotation device holding either a MH or a SJM aortic valve was implanted into eight pigs. The device allowed rotation of the implanted valve without reopening the aorta. In different orientations (best and worst orientation hemodynamically as defined previously), transvalvular pressure gradients and ventricular dimensions were measured using transesophageal echocardiography at constant hemodynamic conditions. Results: In the optimum hemodynamic orientation, pressure gradients of the MH valve (6.3+/-1.7 mmHg) corresponded to those obtained with the SJM valve (6.3+/-3.7 mmHg), whereas in the worst orientation the MH showed a tendency towards higher gradients (14.0+/-2.9 versus 10.3+/-4.0 mmHg) (p = not significant). A significant increase in left ventricular enddiastolic diameter was observed for both valve designs with rotation from the optimal into the worst orientation. Conclusion: In the optimum hemodynamic orientation, the 21 mm MH valve matched the hemodynamic performance of the 23 mm SJM valve. Thus, implantation of the MH valve might be an alternative to root enlargement and implantation of a larger SJM valve in patients with a small aortic annulus, though optimum orientation is required.

Background and aims of the study: Medtronic, Inc. has designed and developed a new bileaflet, aortic prosthesis (Advantage) and, following an in-vitro analysis, a chronic in-vivo 20-week evaluation was performed in sheep. Methods: Fourteen adult male and female sheep underwent implantation of either a 19 mm Advantage or St. Jude Medical (SJM) aortic prosthetic valve for chronic in-vivo evaluation, using a previously reported aortic model technique. Results: There were two operative deaths, and the remaining sheep (eight with Advantage, four with SJM prostheses) underwent chronic hemodynamic and pathologic evaluations at approximately 140 days. One sheep (SJM prosthesis) died unexpectedly at 122 days; postmortem evaluation revealed valvular thrombosis, though the valve was well seated with no excessive pannus formation or paravalvular leaks. Hemodynamic and hematologic data in the remaining sheep were comparable. Pathologic examination of the remaining implanted valves revealed well-seated prostheses, with no evidence of thrombosis, no excessive pannus formation, and no evidence of paravalvular leaks. Conclusion: This preclinical animal study showed that the Advantage prosthesis has acceptable features, including equivalent hemodynamic and hematologic data, and a lack of pathological abnormalities when compared with a commercially available prosthetic bileaflet mechanical valve.

Background and aims of the study: This study details the authors' experience with the CarboMedics bileaflet mechanical prosthesis, and assesses the valve's efficacy in terms of early mortality and long-term morbidity. Methods: Between July 1994 and December 2000, a total of 696 CarboMedics valves was implanted in 616 patients. Aortic valve replacement (AVR) was performed in 350 (57%) patients, mitral valve replacement (MVR) in 161 (26%), double (aortic + mitral) valve replacement (DVR) in 79 (13%), and other procedures in 26 (4%). Results: There were 32 operative deaths. Operative mortality was 4.3% for AVR, 6.8% for MVR, and 7.6% for DVR. Linearized rates for late death were 3.4%/patient-year (pt-yr), 5.5%/pt-yr and 7.7%/pt-yr for AVR, MVR and DVR, respectively. Survival at five years was 80.1% for AVR, 73.0% for MVR, and 64.6% for DVR (p = 0.004). Freedom from reoperation at five years was 97.6% for AVR, 98.1% for MVR, and 94.1% for DVR (p = 0.718); freedom from thromboembolism at five years was 96.3% for AVR, 98.6% for MVR, and 88.3% for DVR (p = 0.04). Linearized rates of late thromboembolism were 0.7%/pt-yr, 0.4%/pt-yr and 2.3%/pt-yr, respectively. Freedom from PVE at five years was 98.4% for AVR, 97.0% for MVR, and 89.7% for the DVR group (p = 0.172). Linearized rates of PVE were 0.5%/pt-yr, 0.9%/pt-yr and 1.8%/pt-yr, respectively. Freedom from anticoagulant-related bleeding at five years was 98.0% in the AVR group, 97.8% in the MVR group, and 88.3% in the DVR group (p <0.001); linearized rates of this event were 0.4%/pt-yr, 0.6%/pt-yr and 3.2%/pt-yr, respectively. One valve thrombosis was found after MVR, and another in the mitral position after DVR. Conclusion: Early and intermediate-term follow up of the CarboMedics prosthesis indicates that the clinical performance of this valve is satisfactory, and results are comparable with those obtained for other mechanical valves.

Although, traditionally, surgery has been the treatment of choice for left-sided prosthetic valve thrombosis (PVT), there are significant risks associated with re-do surgery. Thrombolysis is an alternative to surgery and has a success rate in excess of 80%, but is associated with an -10% risk of systemic embolism and a 7% mortality rate. Guidelines for the use of thrombolytic therapy for PVT were produced using data acquired from older valves, and largely without the use of transesophageal echocardiography (TEE). Data from more recent studies suggest that thrombolysis should be regarded as first-line therapy for PVT in all NYHA classes. The use of TEE is recommended to visualize thrombus in suspected cases.

Mitral regurgitation (MR) following endomyocardial biopsy is a rare and severe complication. A 70-year-old man with severe MR due to chordal injury caused by left ventricular endomyocardial biopsy is described. In this patient, a few chordae tendineae of the posterior-median papillary muscle were injured by the biopsy forceps. Due to the chordal rupture, both anterior and posterior leaflets were prolapsed and severe MR developed. MR was successfully treated by artificial chordal replacement using extended polytetrafluoroethylene sutures and ring annuloplasty. This mitral valve repair with artificial chordal replacement was considered suitable to treat MR resulting from iatrogenic chordal injury as the leaflets were not involved in the degenerative process and papillary muscle function was preserved. To avoid MR, the transvenous approach should be used routinely for endomyocardial biopsies; biopsy from the left ventricle is not justified.

Background and aim of the study: It has been reported that patients with mitral valve prolapse syndrome (MVPS) also have a disorder in autonomic or neuroendocrine function which can cause many related symptoms. Although a potential role of the reninangiotensin system in the pathogenesis of MVPS has been addressed, the role of the angiotensinogen (AGT) genetic variant in MVPS has not been studied. Thus, a case-controlled study was performed to investigate the possible relationship between AGT gene polymorphisms and MVPS. Methods: A total of 100 patients with MVP diagnosed by echocardiography and 100 age- and sex-matched normal control subjects was studied. AGT gene M235T and T174M polymorphisms were identified by polymerase chain reaction-based restriction analysis. Results: There was a significant difference in the distribution of AGT gene M235T genotypes (p <0.001) and allelic frequencies (p <0.001) between MVPS cases and controls. An Odds Ratio (OR) for risk of MVPS associated with M235T TT genotype was 8.55 (95% CI 4.51-16.18). An OR for risk of MVPS associated with the T allele at the M235T locus of the AGT gene was 3.27 (95% CI 2.05-5.22). The T174M polymorphism of AGT gene showed no association with MVPS (p = 0.94). Conclusion: These findings suggest that the M235T polymorphism of the AGT gene is associated with MVPS in the Chinese population of Taiwan. The association of the TT genotype with MVPS is more noteworthy than an overall increase in the frequency of the T allele at the M235T locus.

Background and aims of the study: The posterior mitral leaflet is divided into a variable number of scallops, and little is known about the role of scallopmalcoaptation in ischemic mitral regurgitation. The study aim was to assess whether acute ischemia in the posterolateral wall of the left ventricle would induce scallop separation that would contribute to mitral regurgitation. Methods: Radio-opaque markers were surgically placed in the left ventricle, around the mitral annulus, and at three sites along the posterior mitral leaflet edge in eight sheep. Three-dimensional marker coordinates were obtained by biplane videofluoroscopy at 60 Hz and 0.1 mm resolution before and during echocardiographically verified acute ischemic mitral regurgitation produced by balloon occlusion of the circumflex coronary artery. Results: During systole, the mean (+/-SD) distance between the central and anterolateral markers, both placed on the central scallop of the posterior mitral leaflet, was unaffected by ischemia (7.4+/-2.4 versus 7.4+/-2.5 mm; n = 8; p = NS). In contrast, the systolic distance between the central scallop marker and the posteromedial marker increased by 2.3+/-0.2 mm (p = 0.008) in three hearts with the posteromedial marker on the posteromedial scallop, compared with no separation (0.2+/-0.5 mm; p = NS) in five hearts with both the central and posteromedial markers on the central scallop itself. This result shows systolic separation of the central and posteromedial scallops during acute ischemic mitral regurgitation. Conclusion: During acute left ventricular ischemia, the central and posteromedial scallops of the posterior mitral leaflet can fail to coapt during systole, potentially contributing to the mitral regurgitation observed.

Background and aim of the study: Edge-to-edge repair is an effective, recently introduced method to correct mitral insufficiency by suturing the leaflets at the site of regurgitation, though durability of the method has not been proven. To overcome the limitations of the clinical approach, simulations may be used to predict clinical outcome. In this study, the mechanical stress acting on leaflets imposed by the edge-to-edge suture was evaluated as a means of assessing the clinical risk of late fibrosis or tissue degeneration. Methods: A 3-D finite element simulated the stress pattern following edge-to-edge repair. Valve behavior was evaluated both in systole and in diastole. Both 4-mm and 8-mm edge-to-edge sutures were simulated, as well as annular dilation. Results: Systolic simulations validated the model by comparison with previous models of the mitral valve. Diastolic stresses were negligible in the native mitral valve; after edge-to-edge repair (8-mm suture), circumferential and longitudinal stress values were 308 kPa and 489 kPa, respectively, and comparable with those observed at systolic peak (449 kPa and 617 kPa, respectively). With a 4-mm suture, longitudinal stresses decreased both close to the suture (-41.5%) and in the annular region (-68%), while circumferential stresses increased (+37%) close to the suture and decreased (-27%) in the annular region. A 20% dilation of the annulus was followed by increased stresses in the annular region and close to the suture. Conclusion: Leaflet distortion and altered stress distribution occur on the leaflets after edge-to-edge repair. Diastolic peak stress values were comparable with those calculated in systole. The clinical implication is a doubled exposure of valve components to systolic stresses, as if the heart rate were doubled. The use of a prosthetic annuloplasty ring is favorable in the presence of annular dilation to reduce stresses acting on the leaflets after edge-to-edge repair.

Background and aim of the study: Atrial fibrillation (AF) is the most common sustained cardiac arrhythmia encountered in clinical practice. In developing countries, rheumatic mitral stenosis (MS) is the most frequent underlying condition in patients with AF. Sinus rhythm (SR) is difficult to achieve and maintain in these patients, but would be more easily achieved with reduction of left atrial pressure after successful balloon mitral valvotomy (BMV). Methods: Eighty-five patients with persistent AF following BMV received amiodarone (600 mg once daily for two weeks, 200 mg daily thereafter). Electrical cardioversion was performed in those with persistent AF (at six and 12 weeks of drug therapy). Results: Among patients, 33 (39%) converted with amiodarone alone. Of 52 patients who underwent cardioversion at six weeks, 41 (79%) converted to SR. Overall, 87% of patients converted to SR. None of the 11 patients with persistent AF could be converted to SR, despite a second attempt with direct current (DC) cardioversion at 12 weeks. Those who converted to SR had significantly shorter AF duration (AFD) (2.7+/-1.1 versus 3.2+/-0.7 years) and smaller left atrial (LA) size (50.0+/-7.7 versus 57.9+/-4.7 mm). Patient age, gender, NYHA class, ejection fraction and post-BMV variables were comparable between the two groups. Successful maintenance of SR was possible in 61/74 (82%) patients at a mean follow up of 30.6+/-7.1 months (range: 16-43 months). Again, mean AFD was shorter (1.8+/-0.6 versus 3.0+/-0.7 years) and LA size smaller (48.9+/-7.5 versus 54.7+/-6.9 mm) among those who maintained SR. However, even in patients with AFD > or =2 years, successful conversion and maintenance of SR was possible in 74% and 62% of patients, respectively. Among patients with LA size > or =60 mm (n = 16), the corresponding value were 84% and 77%, respectively. On multivariate analysis, only AFD was a predictor of acute and long-term success. The probability of SR remaining in those with AFD <2 years at 21, 30 and 43 months was 96%, 95% and 94.6%, respectively, while for those with AFD > or =2 years these values were 62%, 48% and 40%. Conclusion: Low-dose amiodarone was safe and effective in restoring and maintaining SR in patients with AF and rheumatic heart disease.

Background and aims of the study: Surgery for ischemic mitral regurgitation (IMR) is required in 4-5% of patients subjected to coronary artery surgery, and may be challenging. The study aim was to determine outcome following mitral valve repair and myocardial revascularization for moderate-to-severe IMR. Methods: A total of 102 patients (mean age 68+/-7 years) underwent mitral valve repair for IMR between 1998 and 2001 at the authors' unit. Among patients, 28 had acute and 74 chronic mitral regurgitation (MR). Valve repair was achieved with an annuloplasty ring in all 102 patients, while 99 underwent concomitant myocardial revascularization. Preoperatively, 69 patients had MR grade III-IV, 62 had CCS angina class III-IV, 59 were in NYHA class II-IV, 81 had impaired left ventricular function, and 10 were in cardiogenic shock. Follow up was 100% complete (mean 14+/-7 months; range: 0-38 months). Results: Overall operative mortality was 8.8% (n = 9) (17.8% for acute IMR, 5.4% for chronic, p = 0.048). On multiple logistic regression analysis, cardiogenic shock (p = 0.028) was the only significant risk factor for operative death. There were 11 late deaths. Kaplan-Meier survival at one and three years was 82+/-4% and 79+/-4%, respectively. On Cox proportional hazards regression model, preoperative left ventricular end-systolic diameter (LVESD) >4.5 cm (p = 0.01) and NYHA class III-IV (p = 0.02) were independent adverse predictors of survival. Three patients required reoperation. Kaplan-Meier three-year freedom from reoperation was 97+/-2%. Conclusion: Surgery for IMR carries a considerable, but acceptable, operative risk and provides satisfactory freedom from reoperation and mid-term survival. Cardiogenic shock before surgery is the major determinant of an unfavorable in-hospital outcome. LVESD >4.5 cm and poor preoperative NYHA status limit the probability of late survival. The study results support early surgical intervention for IMR, before ventricular dilatation occurs.

Background and aims of the study: Left ventricular (LV) contraction is slowed in patients with aortic stenosis (AS). Although the possible role of LV systolic function abnormalities in the assessment of AS severity has been evaluated, current echocardiographic techniques cannot offer precise quantification of LV motion velocity. The study aim was to evaluate an automated segmental motion analysis (ASMA) system to assess AS severity. Methods: Twenty-two patients with AS, sinus rhythm and preserved LV ejection fraction were studied prospectively. Patients underwent both conventional Doppler echocardiography to measure transaortic gradient and aortic valve area by the continuity equation, and ASMA of the interventricular septum. The ASMA line graph mode displays changes in area through the cardiac cycle. The RR interval and time from the R-wave to peak maximum area shortening were measured, and an ASMA index was calculated. Results: A significant and strong inverse correlation was found between aortic valve area and ASMA index (r = -0.78; 95% CI -0.90 to -0.55; p <0.001). The area under the ROC curve in the diagnosis of severe AS (aortic valve area < or =0.8 cm2) was 0.97 (95% CI 0.90-1.0). Sensitivity, specificity, positive and negative predictive values and overall accuracy for an ASMA index >0.40 were 100, 91.7, 92.3, 100 and 95.8%, respectively. Conclusion: The ASMA system may be valuable in evaluating AS, as it offers a strong correlation with aortic valve area calculated by the continuity equation, and very high sensitivity and specificity in the diagnosis of severe AS.

Background and aims of the study: Concern persists regarding the long-term effects of aortic valve replacement (AVR) in the small aortic root. Methods: Twenty-eight patients (21 females, seven males; mean age 54.1+/-11.7 years; mean body surface area (BSA) 1.39+/-0.07 m2) who underwent AVR with a 19 mm mechanical valve more than 10 years previously at the authors' institute, were followed up. Long-term echocardiographic evaluations were performed in 18 of 21 survivors. Results: Follow up in the 28 patients was complete for up to 19.2 years (mean 11.4 years); cumulative follow up was 318.1 patient-years. Actuarial survival rates at 5, 10 and 15 years were 88.5+/-6.3%, 72.0+/-9.0% and 72.0+/-9.0%, respectively. In general, NYHA class showed satisfactory improvement. Long-term echocardiography showed significant reductions in left ventricular (LV) wall thickness and mass index compared with preoperative values, despite the existence of pressure gradients. A negative correlation was found between the rate of reduction of LV wall thickness and BSA. Conclusion: Implantation of the 19 mm aortic mechanical prosthesis was viable in patients with a small aortic annulus, especially if their BSA was <1.45 m2.

Background and aim of the study: Trends in aortic valve operations (AVO) may help to predict the future of aortic valve surgery in the context of changing case mix, population demographics, emerging technology and shifting paradigms. Methods: All adults undergoing single AVO between 1979 and 1999 were reviewed retrospectively according to age, gender and other typical clinical variables, surgical complexity, specific operation and immediate outcome. Results: There were 3,917 single AVO. Although coronary artery bypass (CAB) volumes declined by 15.3% between 1996 and 1999, AVO volumes have continued to increase by 11.7% since 1996. Over the entire period, there was no significant change in mean age or percent female gender, but increases in the prevalence of octogenarians and aortic stenosis were noted. During the 1990s, degenerative valve disease predominated and the prevalence of sicker patients according to heart failure class and surgical priority decreased. Trends in surgical complexity included an increase in AVO combined with CAB, but a stable 9:1 distribution of first operations to reoperations. Technology adoption included a decreased prevalence of mechanical valve use at the expense of increased use of tissue valves, especially stented xenografts and homografts. Transient technology adoption included stentless xenografts. Small numbers of pulmonary autografts, aortic valve repairs and valve-sparing aortic replacements were carried out. Predictors of hospital mortality rates for AVO included age 65 years, reoperation and combined AVO. Hospital mortality rates for AVO decreased for most age groups between the 1980s and 1990s, but not during the 1990s. Conclusion: AVO volumes are steadily increasing, apparently as a result of the increase in octogenarians and the start of the 'baby boom' wave. Hospital mortality risk is related to age and surgical complexity, but is modest and has stabilized during the past decade. The prevalence of mechanical valve implants has decreased in favor of tissue valve replacement categories. The fastest growth rates have been with stented xenografts, and especially homografts. This may represent a paradigm shift away from mechanical solutions in favor of tissue solutions for aortic valve disease.

Background and aim of the study: The risk of thromboembolism (TE) risk in a heart valve patient who has lived perhaps for 15 years since implant without TE or thrombosis is unknown. As patients are heterogeneous with regard to embolic risk, these patients must have a lower than average risk; whether this risk reduction could be quantified was the aim of the present study. Methods: If all patients had the same risk, the TE-free curve would be exponential, with a constant hazard (equal to the 'linearized' rate). With a mixture of risks in a population, the population hazard will be a decreasing function of time. By fitting a certain parametric function to the TE-free curve, the mixing distribution can be estimated. Subsequently, given an observation for a particular patient, e.g. zero emboli in 15 years, Bayes' theorem can be used to update the mixing distribution for these patients. Results: Using observed TE-free curves for Starr-Edwards valves, the mixing distribution for TE-free periods from one to 25 years was estimated. Because the risk distributions were skewed, the mean value was higher than the median. The mean (median) risk (% per year) fell from 4.5 (1.8) at implant to 1.7 (0.7) at 15 years for the aortic position, and from 7.0 (3.4) to 2.4 (1.2) for the mitral position. Thus, the average risk after 15 years was approximately 35-40% of the risk at implant for both the aortic and mitral positions, using either the mean or median risk. Conclusion: The mean risk for patients after 15 TE-free years is approximately one-third of the risk of all patients at implant. However, there is a wide range, with some patients still having higher risks. Bayes' theorem is useful for deriving such answers. Linearized rates have limited value for describing TE risk in a heterogeneous population.

The rates of valve-related complications determine the clinical performance of biological and mechanical heart valve prostheses by valve position. The crude rates of valve-related complications within the first year after implantation have not been previously reported. The Food and Drug Administration (FDA) of the United States guidance document for the performance of new heart valve prostheses uses objective performance criteria (OPC) as target values for valve-related complications. The FDA requirement for each valve-related complication is that the rate cannot be greater than twice the OPC. This report provides the first-year rates of valve-related complications from a university database of over 7,000 implants, followed longitudinally between 1982 and 1999, of currently marketed biological and mechanical heart valve prostheses. The aim of this report was to provide an additional method for consideration by regulatory authorities in determination of the standards of performance for pre-market approval of new prosthetic valves for heart valve replacement surgery.