Alex Frydrychowicz

• Die Rolle des rechten Ventrikels in der Entwicklung der Herzinsuffizienz ist bis heute umstritten. Vor dem Hintergrund neuer transgener Mausmodelle war es Ziel dieser Studie, die MR-basierte Volumenquantifizierung für den rechten Ventrikel der Maus zu validieren und auf zwei Mausmodell mit Rechtsherzinsuffizienz anzuwenden. Gesunde Mäuse unterschiedlichen Alters wurden mittels MR-Bildgebung bei 7T untersucht. Zur Generierung der Herzinsuffizienz wurde eine Infarzierung durch Banding der LAD, zur Erzeugung einer isolierten RechtsherzinsuffizienzDie Rolle des rechten Ventrikels in der Entwicklung der Herzinsuffizienz ist bis heute umstritten. Vor dem Hintergrund neuer transgener Mausmodelle war es Ziel dieser Studie, die MR-basierte Volumenquantifizierung für den rechten Ventrikel der Maus zu validieren und auf zwei Mausmodell mit Rechtsherzinsuffizienz anzuwenden. Gesunde Mäuse unterschiedlichen Alters wurden mittels MR-Bildgebung bei 7T untersucht. Zur Generierung der Herzinsuffizienz wurde eine Infarzierung durch Banding der LAD, zur Erzeugung einer isolierten Rechtsherzinsuffizienz eine Pulmonalarterienstenose operativ induziert. Die Schnittführung der MR-Aufnahmen lag orthogonal zum Ventrikelseptum, um Partialvolumen-Fehler zu minimieren. Die MR-Daten wurden von 2 unabhängigen Auswertern zur Bestimmung der Reproduzierbarkeit analysiert. Der Vergleich linksventrikulärer (LV) und rechtsventrikulärer (RV) Volumina zeigte eine enge Korrelation für die Schlagvolumina (SV) (r=0.97, p<0.0001). Die Bland-Altman-Analyse bestätigte diese enge Übereinstimmung (mittlere Differenz 0.4µl). Wiederholte Messung der RV Parameter zeigte ein hohes Maß an Reproduzierbarkeit (intraobserver- und interobserver-Variabilität jeweils <7%). Die MR-Messung 4 Wochen nach Pulmonalbanding ergab einen signifikanten Anstieg des RV enddiastolischen Volumens (+44%, p<0.005) und vor allem RV endsystolischen Volumens (+133%, p<0.0001). Die RV Ejektionsfraktion war signifikant reduziert (31.2±6.2 % vs. 57.0±2.3 %, p<0.001). Die LV Volumina sowie EF waren unverändert zum Normalkollektiv und bestätigten den Befund der isolierten Rechtsherzinsuffizienz. Neben LV Volumina erlaubt die 3D MR-Datenakquisition auch eine reproduzierbare Analyse des deutlich komplexeren rechten Ventrikels der Maus. Die Studie belegt eindrucksvoll des Potential von hochauflösender MR-Bildgebung zur Aufklärung von Pathomechanismen rechtsventrikulärer Erkrankungen.…
• The predictive importance of the right ventricle (RV) in the failing heart is constantly discussed and MR-Imaging as a uniquely precise method is used to evaluate the underlying processes. The results from left ventricular (LV) murine cardiac MR-Imaging (MRI) led to the interest in assessing the RV by in vivo 7.05T cine-MRI and to demonstrate its potency on two models of heart failure. To demonstrate the feasibility of RV measurements RV volumes and parameters were compared to LV correlates and validated by inter- and intraobserver comparison.The predictive importance of the right ventricle (RV) in the failing heart is constantly discussed and MR-Imaging as a uniquely precise method is used to evaluate the underlying processes. The results from left ventricular (LV) murine cardiac MR-Imaging (MRI) led to the interest in assessing the RV by in vivo 7.05T cine-MRI and to demonstrate its potency on two models of heart failure. To demonstrate the feasibility of RV measurements RV volumes and parameters were compared to LV correlates and validated by inter- and intraobserver comparison. The method was then transferred to mice infarcted by LAD-banding as a model of heart failure and to a relatively unknown model of right heart failure by banding of the pulmonary artery. LAD-banding was performed as previously described by open-chest surgery. For pulmonary banding the pulmonary artery was dissected from other vascular structures and then banded against a prepared 25G needle for reproducibility of results1. All mice were given 4 weeks to recreate from the operation and to develop heart failure. MR studies were done by the known setup of our workgroup on a 7.05T experimental MR scanner (BRUKER Biospec, Germany) with a horizontal bore and a microscopy gradient insert. Adjacent planes were planned orthogonal to the interventricular septum as to depicture both RV and LV in a single measurement with highest possible precision and low geometric assumptions. Volume calculation was achieved by determining endsystolic and enddiastolic slices in every plane. Endo- and epicardial borders were then manually delineated for both left and right ventricle and added up for total ventricular volume (see figures 1-4). After comparison of LV and RV EDV, ESV, SV, CO and EF for validation of the method by one observer, a different observer repeatedly assessed RV volumes of each mouse for determination of inter- and intraobserver variability. After RV evaluation was thereby validated, the method was transferred onto the above models. Comparison of LV and RV in healthy mice showed close correlations (r=0.97, p<0.0001) between SV and CO as expected from human physiology. LV and RV EDV, ESV and EF differed as expected: RV EDV and ESV were significantly higher, RV EF lower than LV correlates (for all p<0.05). Repeated RV assessment for demonstration of inter- and intraobserver variability was done by a different observer: Calculations revealed low differences as shown in tables 1 and 2. Findings in the LAD-banding model were consistent with expectations from human pathophysiological changes with an obvious decrease of RV and LV SV, EF and CO. The relatively unknown banding of the pulmonary artery revealed unsignficantly altered LV parameters whereas RV EF was significantly lower. However, RV SV and CO were only slightly diminished which could be due to a slower progress of RV alterations than expected. We were able to demonstrate the feasibility of high-resolution MRI of the murine right ventricle with good results in direct RV and LV comparison. Inter- and Intraobserver validation of this method showed reasonably low differences which enabled us to successfully transfer this technique on two models of heart failure. These findings open new perspectives to further investigate RV changes both in healthy mice as well as in genetically or surgically engineered mice. Evaluation of the murine RV by MR-Imaging may therefore lead to a better understanding of the development of RV changes under pathophysiological circumstances.…