京都大学 / 新制・論文博士 / 博士(医学) / 乙第13420号 / 論医博第2228号 / 新制||医||1052(附属図書館) / (主査)教授 川上 浩司, 教授 湊谷 謙司, 教授 山下 潤 / 学位規則第4条第2項該当 / Doctor of Medical Science / Kyoto University / DGAM

Hintergrund: Die funktionelle Mitralklappeninsuffizienz (fMI) tritt häufig sekundär bei Patienten im fortgeschrittenen Stadium der Linksherzinsuffizienz auf. Zur Quantifizierung der fMI werden von den Leitlinien der European Association of cardiovascular Imaging (EACVI) bisher vorwiegend 2D-echokardiographische Methoden, insbesondere die 2D Proximal Isovelocity Surface Area (PISA) als Teil einer semiquantitativen Evaluation zur Schweregrad-Einteilung empfohlen. Allerdings weisen diese Verfahren aufgrund des asymmetrischen Querschnitts des Regurgitationsjets und der proximalen Konvergenz bei fMI deutliche Limitationen in der Quantifizierung auf. Der Unterscheidung einer moderaten von einer schweren fMI kommt eine entscheidende Rolle zu, weil diese Abgrenzung den Zugang zur interventionellen oder chirurgischen Therapie beschränkt. Für die 3D-echokardiographischen Methoden bestehen bislang weder Leitlinienempfehlungen noch feste Grenzwerte zur Unterscheidung von moderater und schwerer fMI. Fragestellung/Hypothese: Das Ziel der Arbeit liegt in einem umfassenden Vergleich der bestehenden 2D- und 3D-echokardiographischen Methoden zur Quantifizierung der fMI. Mithilfe dieser Daten sollen die Unterschiede der verschiedenen Verfahren herausgearbeitet, die Vorteile der 3D Methoden belegt und an der weiteren Validierung der 3D-Methoden mitgewirkt werden. Darüber hinaus sollen Grenzwerte zur Unterscheidung von moderater und schwerer fMI für die unterschiedlichen Verfahren berechnet werden. Material und Methoden: Bei allen Patienten, die für eine perkutane interventionelle Mitralklappentherapie zwischen 01.01.2015 und 31.12.2018 im Herzzentrum Dresden vorgesehen waren, wurden sukzessive 2D- und 3D-Datensätze mit und ohne Farbdoppler anhand eines standardisierten Untersuchungsablauf mittels transösophagealer Echokardiographie durch erfahrene Fachärzte akquiriert. Aus diesem Kollektiv im „Dresdner Register Perkutane Mitralklappentherapie“ wurden anhand von Ausschlusskriterien 105 Patienten mit symptomatischer fMI in die endgültige Analyse eingeschlossen. Die klinischen Charakteristika und Begleiterkrankungen der Patienten wurden erhoben und mit dem Gesamtkollektiv verglichen. Die echokardiographischen Studienparameter wurden entsprechend der Empfehlungen der EACVI bestimmt und als Mittelung aus dreifacher Messung berechnet. Die Analyse der 3D Datensätze wurde offline mit der Multiplanaren Rekonstruktionsmethode durchgeführt. Anschließend erfolgte ein Vergleich der unterschiedlichen Verfahren nach der Bland-Altman-Methode. Für die untersuchten Verfahren wurden außerdem mittels einer Receiver-Operating-Characteristic-Analyse Grenzwerte zur Unterscheidung von moderater und schwerer fMI generiert, welche auf der Einteilung des Schweregrades nach dem Grenzwert des bisherigen Standardverfahrens, der effektiven Regurgitationsfläche (EROA) nach 2D PISA von ≥ 0,2 cm², basieren. Ergebnisse: In den durchgeführten Untersuchungen konnten die 2D Methoden eine akzeptable Validität und gute Reproduzierbarkeit der Messungen belegen. Allerdings wiesen sowohl die EROA 2D PISA als auch die biplanen Vena Contracta Weiten (VCW) eine größere Messvariabilität als die 3D-basierten Verfahren auf. Die 3D-Methoden, insbesondere die 3D hemielliptische PISA, konnten geringere Messungenauigkeiten aufgrund einer besseren Approximation der asymmetrischen Regurgitationsfläche als die 2D-Verfahren belegen. Die 3D VCA konnte durch die direkte Messung in der planimetrischen Aufsicht bei asymmetrischem Querschnitt der Regurgitationsfläche bei fMI valide und reproduzierbare Werte erreichen, die sich auch beim Vorliegen exzentrischer oder multipler Regurgitationsjets stabil zeigten. Für die 3D VCA konnte ein Grenzwert von 0,42 cm² zur Unterscheidung einer moderaten von einer schweren fMI generiert werden. Mittels eines direkten Vergleichs der beiden anatomischen Regurgitationsflächen (AROA) konnte erstmals gezeigt werden, dass beide Approximationen der AROA trotz Bestimmung aus unterschiedlichen Bilddatensätzen analoge Werte messen. Dabei besitzen beide AROA-Verfahren eine sehr gute Validität und eine geringe Messabweichung insbesondere für große Beträge der Regurgitationsfläche. Außerdem konnte ein Grenzwert von 0,21 cm² für beide AROAs zur Unterscheidung von moderater und schwerer fMI vorgeschlagen werden. Schlussfolgerungen: Die 2D-echokardiographischen Verfahren zeichnen sich durch eine unkomplizierte, ubiquitär verfügbare Bestimmbarkeit und eine große Anzahl an beschreibenden Studien aus. Die größere Messvariabilität der 2D Methoden besteht aufgrund der suboptimalen geometrischen Näherung mittels der hemisphärischen (2D PISA) oder elliptischen (VCW) Approximation. Die 3D-Verfahren hingegen weisen eine durchweg höhere Messgenauigkeit als die 2D Methoden auf. Am Beispiel der hemielliptischen 3D PISA-Methode zeigt sich, dass mittels einer verbesserten, aber auch aufwendigeren Approximation der asymmetrischen Form der proximalen Konvergenz eine höhere Messgenauigkeit erreicht werden kann. Im Rahmen von Studien wurde mittels (semi-)automatischer Software eine weniger benutzerabhängige Messung der 3D PISA entwickelt. Momentan stellt die 3D VCA aufgrund der exzellenten Reproduzierbarkeit sowie der validen Messung der Regurgitationsfläche die robusteste Methode zur Quantifizierung der fMI dar. Die 3D VCA weist die geringste Unterschätzung der Regurgitationsfläche im Vergleich zu unabhängigen Vergleichsmethoden (MRT; quantitativer Doppler) auf. Darüber hinaus überzeugt der im Rahmen dieser Studie kalkulierte Grenzwert zur Schweregrad-Einteilung der fMI nach 3D VCA mit einer sehr hohen Reliabilität und kann somit bei vergleichbaren Werten wie in früheren Studien zur Bestätigung eines Grenzwerts für die 3D VCA beitragen. Die AROA ist durch die planimetrische Bestimmung der Regurgitationsfläche unabhängig von geometrischen Annahmen und ist somit weniger anfällig für Fluss-bedingte Phänomene. In dieser Arbeit konnte erstmals die universelle und reproduzierbare Bestimmbarkeit der AROA anhand unterschiedlicher Bilddatensätze belegt werden. Bei bisher geringer Datenlage stellt die AROA eine vielversprechende Methode zur Quantifizierung der schweren fMI dar, die noch in weiteren Studien validiert werden muss. In Zukunft wären eine Integration der im Rahmen dieser Arbeit und diverser weiterer Studien belegten 3D-Verfahren in die Schweregrad-Einteilung bei fMI sinnvoll. Es wird deswegen eine zweistufige Beurteilung als Screening mittels 2D PISA und biplaner VCW im TTE und anschließend eine Bestimmung der 3D VCA zusammen mit bereits etablierten Parametern (retrograder Pulmonalvenenfluss; Regurgitationsfraktion) im TEE zur semiquantitativen Schweregrad-Einschätzung vorgeschlagen. Die 3D-echokardiographischen Methoden stellen eine Weiterentwicklung mit Zugewinn an Validität und Messgenauigkeit im Vergleich zu den bestehenden 2D-echokardiographischen Verfahren dar.:Danksagung 3 Inhaltsverzeichnis 4 Tabellenverzeichnis 6 Abbildungsverzeichnis 7 Abkürzungsverzeichnis 10 1. Einleitung 12 1.1 Epidemiologie der Mitralklappeninsuffizienz 12 1.2 Morphologie und Anatomie der Mitralklappe 12 1.3 Ätiologie und Verlaufsformen der MI 12 1.4 Pathophysiologie der funktionellen Mitralklappeninsuffizienz 13 1.4.1 Ischämische Kardiomyopathie (ICM) 14 1.4.2 Nicht-ischämische Kardiomyopathie (nICM) 15 1.4.3 Isolierte linksatriale Dilatation bei Vorhofflimmern 16 1.5 Hämodynamische Auswirkungen bei ausgeprägter MI 16 1.6 Klinik der symptomatischen MI 17 1.7 Diagnostik der MI 18 1.7.1 Echokardiographische Beurteilung bei fMI 18 1.7.2 Echokardiographische Quantifizierung der fMI 20 1.7.3 Magnet-Resonanz-Tomographie zur Quantifizierung der MI 25 1.8 Prognose bei fMI 26 1.9 Therapie der fMI 27 1.9.1 Chirurgische Therapieverfahren 27 1.9.2 Perkutane, interventionelle Therapieverfahren 30 1.10 Ziele der Arbeit 33 2. Material und Methoden 34 2.1 Studiendesign und Datenakquisition 34 2.2 Parameter der Patientencharakteristika und Vergleichsgruppe 36 2.3 Echokardiographische Methodik 37 2.3.1 Technische Voraussetzungen der Echokardiographie 37 2.3.2 Echokardiographische Bildakquisition 37 2.3.3 Analyse 3D-Zoom-Datensatz für fMI-Ätiologie & AROA nach Morphe 37 2.3.3 Analyse 2D-Doppler-Datensätze mit PISA- und VCW-Verfahren 39 2.3.4 Analyse 3D-Doppler-Datensätze mit PISA-Methoden 41 2.3.5 Analyse 3D-Doppler-Datensätze mit VCW, VCA & AROA ohne Farbe 43 2.3.6 Analyse 2D-&3D-Datensätze linksventrikuläre Volumina und Funktion 46 2.4 Statistische Methoden und Analysen 47 3. Ergebnisse 51 3.1 Charakteristika des Patientenkollektivs 51 3.2 Deskriptive Statistik der echokardiographischen Methoden 55 3.2.1 Regurgitationsflächen 57 3.2.2 Regurgitationsvolumina 60 3.3 Vergleich der Zusammenhänge der Methoden anhand von Korrelationen 62 3.3.1 Regurgitationsflächen 62 3.3.2 Regurgitationsvolumina 67 3.4 Vergleich echokardiographischer Methoden mittels Bland Altman-Verfahren 69 3.4.1 Regurgitationsflächen 71 3.4.2 Regurgitationsvolumina 81 3.5 Graduierung fMI mittels ROC-Analyse anhand Schweregrad nach 2D PISA 84 3.5.1 Regurgitationsflächen 85 3.5.2 Regurgitationsvolumina 88 4. Diskussion 90 4.1 Charakteristika des Patientenkollektivs 90 4.2 2D-Methoden 95 4.2.1 2D PISA 95 4.2.2 2D VCW und 3D VCW 97 4.2.3 Biplane 2D VCA und biplane 3D VCA 99 4.3 3D-Methoden 101 4.3.1 3D PISA-Methoden 102 4.3.2 3D VCA-Methode 108 4.3.3 AROA-Methoden 113 4.4 Limitationen 118 4.4.1 Limitationen der Bilderfassung 118 4.4.2 Messfaktoren 120 4.4.3 Doppler-bedingte Faktoren 121 4.4.4 Spezielle Charakteristika der fMI 122 4.4.5 Systematische Einschränkungen 123 4.5 Zusammenfassung 124 4.6 Summary 127 Anhang 130 Literaturverzeichnis 133 / Background: Functional mitral valve regurgitation (fMR) secondarily occurs in patients with advanced stages of ischemic or dilated cardiomyopathy. Quantification of fMR is mainly conducted by echocardiography. The guidelines of the European Association of Cardiovascular Imaging (EACVI) recommend 2D echocardiographic methods, especially the 2D Proximal Isovelocity Surface Area (PISA) for the quantification of fMR. However, due to the asymmetric cross section of both the regurgitation jet and the proximal convergence in fMR, these methods exhibit clear limitations in quantification. The discrimination between moderate and severe fMR poses an important distinction, because severe symptomatic fMR represents an indication for interventional or surgical therapy. For 3D echocardiographic methods, there are neither recommended guidelines nor fixed thresholds for differentiating moderate from high-grade fMR. Aim and objectives: The purpose of this thesis is a comprehensive comparison of the existing 2D and 3D echocardiographic methods for the quantification of fMR. Based on this data, the aim was to describe the differences between several methods, to prove the advantages of the 3D methods and to contribute to the further validation of the 3D methods. In addition, it was intended to calculate threshold values for the different methods to differentiate between moderate and severe fMR. Material and methods: In all patients scheduled for percutaneous interventional mitral valve therapy at the Dresden Heart Center between January 1, 2015 and December 31, 2018, successive 2D and 3D data sets with and without color Doppler were acquired by experienced specialists using a standardized examination procedure based on transthoracic (TTE) and transesophageal echocardiography (TEE). From this collective in the 'Dresden Register Percutaneous Mitral Valve Therapy' 105 patients with symptomatic fMR were included in the final analysis based on exclusion criteria. The clinical characteristics and concomitant diseases of the patients were assessed and compared to the overall collective. The echocardiographic study parameters were determined according to the recommendations of the EACVI and calculated as an average of three repeated measurements. The analysis of the 3D data sets was performed offline using the Multiplanar Reconstruction Method. Subsequently, we performed the comparison of the different procedures using the Bland-Altman method. Furthermore, threshold values for the differentiation of moderate and severe fMR were generated for the investigated methods according to the preexisting limit value of the previous standard method of effective regurgitation area (EROA) according to 2D PISA from ≥ 0.2 cm² by means of a Receiver Operating Characteristic (ROC). Results: In the presented investigations the 2D PISA method convinced with a good temporal resolution of the data sets and prove stable measured values. However, the EROA 2D PISA showed the greatest measurement variability of all methods, which was attributed to a suboptimal geometric approximation of the asymmetric proximal convergence using a hemispherical model. Despite this limitation, the EROA 2D PISA was utilized as a reference method, since this procedure has been investigated in numerous studies analogous to our measurement methodology and is recommended by the EACVI guidelines for distinguishing moderate and severe fMR. The 2D and 3D Vena Contracta Width (VCW) also demonstrated moderate validity and reproducibility, especially as an average from determination in 2 planes (biplane VCW). Moreover, the vena contracta methods showed stable measurements even with eccentric jets. The 3D methods, on the other hand, substantiated lower measurement inaccuracies due to a better approximation of the asymmetric regurgitation surface than the 2D methods. The 3D PISA methods showed the lowest absolute measurement variability of all methods. Despite good results in previous studies, the 3D hemielliptic PISA exhibited the strongest mean deviation compared to the other methods. The 3D VCA was able to achieve valid and reproducible values in particular by direct measurement in planimetric top view with asymmetrical cross section of the regurgitation surface at fMR. Therefore, the 3D VCA could account for stable results even in the presence of eccentric or multiple regurgitation jets. For the 3D VCA a threshold value of 0.42 cm² was generated to distinguish a moderate from a severe fMR. With this data it was proven for the first time that the anatomical regurgitation area (AROA) produces analogous values by determination from different image data sets. The method shows a very good validity with low measurement error. In addition, a limit value of 0.21 cm² for AROA was proposed to distinguish between moderate and severe fMR. Conclusion: Despite a wide range of echocardiographic methods, the EACVI guidelines mainly recommend 2D echocardiographic procedures as part of a semi-quantitative evaluation for classification of severity for fMR. The 2D methods are characterized as uncomplicated, ubiquitously available technology, and many descriptive studies exist. However, they show an underestimation of the regurgitation area and larger measurement inaccuracy than the 3D methods probably due to suboptimal geometric approximations of the asymmetric shape of regurgitation area and proximal convergence. The 3D PISA with improved, but complex approximation of the surface of proximal convergence nevertheless performs an indirect determination of the regurgitation surface, which is more susceptible to flow-related interference in acquisition and measurement. However, some former studies presented approaches to enable a less user-dependent measurement of the 3D PISA using (semi-)automatic software. In previous studies, the 3D VCA was able to demonstrate the smallest underestimation of the regurgitation area compared to independent comparison methods such as MRI or thermodilution. At present, the 3D VCA is the most robust method for quantifying fMR due to its excellent reproducibility and valid measurement of the regurgitation area despite demanding determination. The threshold value for the severity classification of fMR according to 3D VCA, determined by this study, is in a comparable range to values from previous studies and can therefore possibly contribute to confirm a general threshold value for the 3D VCA. Due to the planimetric determination of the regurgitation area, AROA does not require geometric assumptions and is less susceptible to flow-related impact. However, AROA is strongly dependent on the image quality of the data sets and shows a higher measurement inaccuracy for small amounts of the regurgitation area. With little data available so far, AROA represents a promising method for quantifying severe fMR, which still needs to be validated in further studies. In the future, it would be desirable to integrate the 3D methods substantiated in this work and in numerous studies into the severity classification of fMR. There could be a two-tiered assessment, screening by 2D PISA and biplanar VCW in the TTE and then a determination of the 3D VCA together with already established parameters (retrograde pulmonary venous flow; regurgitation fraction) in the TEE for semi-quantitative severity assessment. The 3D echocardiographic methods represent a further development with increased validity and measurement accuracy compared to the existing 2D echocardiographic methods.:Danksagung 3 Inhaltsverzeichnis 4 Tabellenverzeichnis 6 Abbildungsverzeichnis 7 Abkürzungsverzeichnis 10 1. Einleitung 12 1.1 Epidemiologie der Mitralklappeninsuffizienz 12 1.2 Morphologie und Anatomie der Mitralklappe 12 1.3 Ätiologie und Verlaufsformen der MI 12 1.4 Pathophysiologie der funktionellen Mitralklappeninsuffizienz 13 1.4.1 Ischämische Kardiomyopathie (ICM) 14 1.4.2 Nicht-ischämische Kardiomyopathie (nICM) 15 1.4.3 Isolierte linksatriale Dilatation bei Vorhofflimmern 16 1.5 Hämodynamische Auswirkungen bei ausgeprägter MI 16 1.6 Klinik der symptomatischen MI 17 1.7 Diagnostik der MI 18 1.7.1 Echokardiographische Beurteilung bei fMI 18 1.7.2 Echokardiographische Quantifizierung der fMI 20 1.7.3 Magnet-Resonanz-Tomographie zur Quantifizierung der MI 25 1.8 Prognose bei fMI 26 1.9 Therapie der fMI 27 1.9.1 Chirurgische Therapieverfahren 27 1.9.2 Perkutane, interventionelle Therapieverfahren 30 1.10 Ziele der Arbeit 33 2. Material und Methoden 34 2.1 Studiendesign und Datenakquisition 34 2.2 Parameter der Patientencharakteristika und Vergleichsgruppe 36 2.3 Echokardiographische Methodik 37 2.3.1 Technische Voraussetzungen der Echokardiographie 37 2.3.2 Echokardiographische Bildakquisition 37 2.3.3 Analyse 3D-Zoom-Datensatz für fMI-Ätiologie & AROA nach Morphe 37 2.3.3 Analyse 2D-Doppler-Datensätze mit PISA- und VCW-Verfahren 39 2.3.4 Analyse 3D-Doppler-Datensätze mit PISA-Methoden 41 2.3.5 Analyse 3D-Doppler-Datensätze mit VCW, VCA & AROA ohne Farbe 43 2.3.6 Analyse 2D-&3D-Datensätze linksventrikuläre Volumina und Funktion 46 2.4 Statistische Methoden und Analysen 47 3. Ergebnisse 51 3.1 Charakteristika des Patientenkollektivs 51 3.2 Deskriptive Statistik der echokardiographischen Methoden 55 3.2.1 Regurgitationsflächen 57 3.2.2 Regurgitationsvolumina 60 3.3 Vergleich der Zusammenhänge der Methoden anhand von Korrelationen 62 3.3.1 Regurgitationsflächen 62 3.3.2 Regurgitationsvolumina 67 3.4 Vergleich echokardiographischer Methoden mittels Bland Altman-Verfahren 69 3.4.1 Regurgitationsflächen 71 3.4.2 Regurgitationsvolumina 81 3.5 Graduierung fMI mittels ROC-Analyse anhand Schweregrad nach 2D PISA 84 3.5.1 Regurgitationsflächen 85 3.5.2 Regurgitationsvolumina 88 4. Diskussion 90 4.1 Charakteristika des Patientenkollektivs 90 4.2 2D-Methoden 95 4.2.1 2D PISA 95 4.2.2 2D VCW und 3D VCW 97 4.2.3 Biplane 2D VCA und biplane 3D VCA 99 4.3 3D-Methoden 101 4.3.1 3D PISA-Methoden 102 4.3.2 3D VCA-Methode 108 4.3.3 AROA-Methoden 113 4.4 Limitationen 118 4.4.1 Limitationen der Bilderfassung 118 4.4.2 Messfaktoren 120 4.4.3 Doppler-bedingte Faktoren 121 4.4.4 Spezielle Charakteristika der fMI 122 4.4.5 Systematische Einschränkungen 123 4.5 Zusammenfassung 124 4.6 Summary 127 Anhang 130 Literaturverzeichnis 133

Introdução: A insuficiência cardíaca é uma síndrome clínica grave e freqüente. Nos estágios avançados, pode se apresentar em associação com a insuficiência mitral secundária. O quadro clínico piora e a sobrevida diminui quando a insuficiência mitral está presente. A abordagem cirúrgica da insuficiência mitral secundária tem sido motivo de investigação e a tração quádrupla dos músculos papilares com implante de prótese biológica pode ser uma opção. Objetivo: Avaliar, através da ecocardiografia trans-torácica, a geometria e a função do ventrículo esquerdo após a troca da valva mitral com tração e fixação quádrupla dos músculos papilares, nos pacientes portadores de insuficiência cardíaca terminal com insuficiência mitral secundária. Casuística: Foram operados de forma consecutiva 20 pacientes portadores de insuficiência cardíaca terminal por miocardiopatia dilatada com insuficiência mitral secundária. O sexo masculino predominou: 70%. A idade variou entre 27 e 72 anos, com média de 50,2 +- 9 anos. O número de admissões na enfermaria no ano precedente à operação foi em média 5,4 por paciente e 2,4 na unidade de terapia intensiva. Onze (55%) estavam em uso prolongado de drogas vasoativas. A fração de ejeção do ventrículo esquerdo, determinada pelo método de Teicholz, foi menor ou igual a 30% em todos os pacientes. Dezessete (85%) estavam em classe funcional IV (NYHA). Método: Todos os pacientes foram submetidos à troca da valva mitral com tração e fixação quádrupla dos papilares. Dezoito (90%) receberam biopróteses de pericárdio bovino um tamanho menor que a medida calculada no ato operatório e dois (10%) receberam próteses mecânicas. A plástica tricúspide (DeVega) foi realizada em 12 (60%) pacientes. No seguimento, os exames ecocardiográficos foram agrupados em períodos: três, 6, 12 e 18 meses. As variáveis ecocardiográficas estudadas foram o volume sistólico do ventrículo esquerdo, a fração de ejeção, os diâmetros sistólico e diastólico finais e os volumes sistólico e diastólico finais. O estudo estatístico foi estruturado com a análise de variância para dados repetidos e o teste nãoparamétrico de Friedmann, objetivando a avaliação do comportamento das variáveis ao longo do tempo. A sobrevida foi aferida pelo método de Kaplan-Meyer e a classe funcional avaliada pelo método de McNemar. Resultados: Dois (10%) pacientes faleceram no período imediato: broncopneumonia e falência de múltiplos órgãos. A sobrevida ao final do primeiro ano foi de 85%, do segundo 44%, do terceiro 44%, do quarto 44% e do quinto 44%. Aos 48 meses de seguimento, a classe funcional melhorou (p<0,001), bem como aos 54 meses. A comparação entre os momentos pré e 3 meses, empregando-se a análise de variância para dados repetidos, não revelou alteração significativa para o volume sistólico (p=0,086). Houve acréscimo da fração de ejeção (p=0,008) e decréscimo do diâmetro diastólico final (p=0,038); do diâmetro sistólico final (p=0,008); do volume diastólico final (p=0,029) e do volume sistólico final (p=0,009). Para a avaliação dos momentos pré, 3 e 6 meses, empregou-se o teste não-paramétrico de Friedmann e não houve significância para nenhuma das variáveis ecocardiográficas. Na avaliação dos momentos pré, 3 meses e última avaliação (final), empregando-se a análise de variância para dados repetidos, não houve significância para os dados estudados. Conclusão: Há melhora significativa da fração de ejeção, dos volumes sistólico e diastólico finais e diâmetros sistólico e diastólico finais do ventrículo esquerdo; até o terceiro mês de pós-operatório. A partir de então, as variáveis permanecem estáveis. / Background. We aimed to evaluate mitral valve replacement results and a new technique for complete chordae tendineae adjustment for left ventricular remodeling. Methods. Twenty end-stage idiopathic dilated cardiomyopathy patients with severe functional mitral valve regurgitation underwent mitral valve replacement from July 2000 to December 2003. Three (15%) were in New York Heart Association functional class (FC) III; 17 (85%) were in FC IV. Hospital admissions for congestive heart failure in the 12 months prior to surgery were 5.4 ± 3.1 and 2.4±1.2 in the intensive care. Both anterior and posterior leaflets of the mitral valve were divided to obtain 4 pillars of chordae tendineae. These were displaced with traction toward the left atrium and anchored between the mitral annulus and a valvular prosthesis. To evaluate the left ventricular remodeling doppler echocardiography were performed. The statistical analysis was structured with variance analysis and Friedman´s test. Results. Two (10%) early deaths occurred from bronchopneumonia and multisystem organ failure. Kaplan-Meyer showed survival at one year post-operative was 85%, 2 years was 44%, 3 years was 44%, 4 years was 44% and 5 years was 44%. At 48 and 54 months of follow-up, McNemar test showed improvement in Functional Class (p<0.001). At third month of follow-up, variance analyses showed improvement in ejection fraction (p=0.008) and decreasing in end-diastolic diameter (p=0.038), end-sistolic diameter (p=0.008), end-sistolic volume (p=0.029) and end-diastolic volume (p=0.009). No statistical difference were noted in systolic volume. Comparing pre-operative, third and six months of follow-up, Friedmann test showed no statistical differences for all variables studied. Variance analyses for pre, third and final evaluation showed samething. Conclusion. This new technique of mitral valve replacement, involving the positioning of the chordae tendineae, should improvement in EF and decreasing in DD, SD,SV and DV till third month of follow-up. The variables sustain this changes during follow-up. An improvement in functional class and survival were assignated in this group.

Introdução: A insuficiência cardíaca é uma síndrome clínica grave e freqüente. Nos estágios avançados, pode se apresentar em associação com a insuficiência mitral secundária. O quadro clínico piora e a sobrevida diminui quando a insuficiência mitral está presente. A abordagem cirúrgica da insuficiência mitral secundária tem sido motivo de investigação e a tração quádrupla dos músculos papilares com implante de prótese biológica pode ser uma opção. Objetivo: Avaliar, através da ecocardiografia trans-torácica, a geometria e a função do ventrículo esquerdo após a troca da valva mitral com tração e fixação quádrupla dos músculos papilares, nos pacientes portadores de insuficiência cardíaca terminal com insuficiência mitral secundária. Casuística: Foram operados de forma consecutiva 20 pacientes portadores de insuficiência cardíaca terminal por miocardiopatia dilatada com insuficiência mitral secundária. O sexo masculino predominou: 70%. A idade variou entre 27 e 72 anos, com média de 50,2 +- 9 anos. O número de admissões na enfermaria no ano precedente à operação foi em média 5,4 por paciente e 2,4 na unidade de terapia intensiva. Onze (55%) estavam em uso prolongado de drogas vasoativas. A fração de ejeção do ventrículo esquerdo, determinada pelo método de Teicholz, foi menor ou igual a 30% em todos os pacientes. Dezessete (85%) estavam em classe funcional IV (NYHA). Método: Todos os pacientes foram submetidos à troca da valva mitral com tração e fixação quádrupla dos papilares. Dezoito (90%) receberam biopróteses de pericárdio bovino um tamanho menor que a medida calculada no ato operatório e dois (10%) receberam próteses mecânicas. A plástica tricúspide (DeVega) foi realizada em 12 (60%) pacientes. No seguimento, os exames ecocardiográficos foram agrupados em períodos: três, 6, 12 e 18 meses. As variáveis ecocardiográficas estudadas foram o volume sistólico do ventrículo esquerdo, a fração de ejeção, os diâmetros sistólico e diastólico finais e os volumes sistólico e diastólico finais. O estudo estatístico foi estruturado com a análise de variância para dados repetidos e o teste nãoparamétrico de Friedmann, objetivando a avaliação do comportamento das variáveis ao longo do tempo. A sobrevida foi aferida pelo método de Kaplan-Meyer e a classe funcional avaliada pelo método de McNemar. Resultados: Dois (10%) pacientes faleceram no período imediato: broncopneumonia e falência de múltiplos órgãos. A sobrevida ao final do primeiro ano foi de 85%, do segundo 44%, do terceiro 44%, do quarto 44% e do quinto 44%. Aos 48 meses de seguimento, a classe funcional melhorou (p<0,001), bem como aos 54 meses. A comparação entre os momentos pré e 3 meses, empregando-se a análise de variância para dados repetidos, não revelou alteração significativa para o volume sistólico (p=0,086). Houve acréscimo da fração de ejeção (p=0,008) e decréscimo do diâmetro diastólico final (p=0,038); do diâmetro sistólico final (p=0,008); do volume diastólico final (p=0,029) e do volume sistólico final (p=0,009). Para a avaliação dos momentos pré, 3 e 6 meses, empregou-se o teste não-paramétrico de Friedmann e não houve significância para nenhuma das variáveis ecocardiográficas. Na avaliação dos momentos pré, 3 meses e última avaliação (final), empregando-se a análise de variância para dados repetidos, não houve significância para os dados estudados. Conclusão: Há melhora significativa da fração de ejeção, dos volumes sistólico e diastólico finais e diâmetros sistólico e diastólico finais do ventrículo esquerdo; até o terceiro mês de pós-operatório. A partir de então, as variáveis permanecem estáveis. / Background. We aimed to evaluate mitral valve replacement results and a new technique for complete chordae tendineae adjustment for left ventricular remodeling. Methods. Twenty end-stage idiopathic dilated cardiomyopathy patients with severe functional mitral valve regurgitation underwent mitral valve replacement from July 2000 to December 2003. Three (15%) were in New York Heart Association functional class (FC) III; 17 (85%) were in FC IV. Hospital admissions for congestive heart failure in the 12 months prior to surgery were 5.4 ± 3.1 and 2.4±1.2 in the intensive care. Both anterior and posterior leaflets of the mitral valve were divided to obtain 4 pillars of chordae tendineae. These were displaced with traction toward the left atrium and anchored between the mitral annulus and a valvular prosthesis. To evaluate the left ventricular remodeling doppler echocardiography were performed. The statistical analysis was structured with variance analysis and Friedman´s test. Results. Two (10%) early deaths occurred from bronchopneumonia and multisystem organ failure. Kaplan-Meyer showed survival at one year post-operative was 85%, 2 years was 44%, 3 years was 44%, 4 years was 44% and 5 years was 44%. At 48 and 54 months of follow-up, McNemar test showed improvement in Functional Class (p<0.001). At third month of follow-up, variance analyses showed improvement in ejection fraction (p=0.008) and decreasing in end-diastolic diameter (p=0.038), end-sistolic diameter (p=0.008), end-sistolic volume (p=0.029) and end-diastolic volume (p=0.009). No statistical difference were noted in systolic volume. Comparing pre-operative, third and six months of follow-up, Friedmann test showed no statistical differences for all variables studied. Variance analyses for pre, third and final evaluation showed samething. Conclusion. This new technique of mitral valve replacement, involving the positioning of the chordae tendineae, should improvement in EF and decreasing in DD, SD,SV and DV till third month of follow-up. The variables sustain this changes during follow-up. An improvement in functional class and survival were assignated in this group.