Papillary muscles contribute significantly more to left ventricular work in dilated hearts

Papillary muscles contribute significantly more to left ventricular work in dilated hearts Abstract Aims Left ventricular (LV) dilatation results in increased sphericity and affects position and orientation of papillary muscles (PMs), which may influence their performed work. The aim of this study was to assess the contribution of PM to LV function and its changes with dilatation. Methods and results Fifteen sheep were investigated. Ten animals were subjected to 8 weeks of rapid (180 bpm) pacing, inducing LV dilatation. Five animals served as controls. High-resolution gated computed tomography was performed to assess LV volumes, left ventricular ejection fraction (LVEF), global longitudinal strain (GLS), sphericity index, and PM angle, width and fractional shortening. 18F-fluorodeoxyglucose positron emission tomography (PET) was used to measure glucose metabolism as surrogate of regional myocardial work. Spatial resolution of PET images was maximized by electrocardiogram- and respiratory-gating. 18F-fluorodeoxyglucose uptake was measured in PM and compared with remaining left ventricular myocardium (MYO) to obtain a PM/MYO ratio. Animals with dilated heart had a more spherical left ventricle, with reduced LVEF (P < 0.0001) and GLS (P < 0.0001). In dilated hearts, PET analysis revealed a higher contribution of both PM to LV myocardial work (P < 0.0001); and PM angle towards LV wall correlated with PM work, together with PM width and the LV sphericity index. Sphericity index and posterior PM angle were strongest determinants of posterior PM/MYO ratio (R2 = 0.754; P < 0.0001), while anterior PM/MYO was mostly determined by sphericity index and the PM width (R2 = 0.805; P < 0.0001). Conclusion In dilated hearts, PM contribute relatively more to LV myocardial work. We hypothesize that this is caused by the more cross-sectional orientation of the subvalvular apparatus, which leads to a higher stress on the PM compared with the spherical LV walls. The reduced cross-sectional area of the PM may further explain their increased stress. papillary muscles , left ventricle , dilatation , work Introduction Papillary muscles (PMs) are a continuity of the longitudinal fibres of the myocardium and directly connect the left ventricular (LV) wall with the mitral valve through chordae tendineae. This so-called subvalvular apparatus acts along the longitudinal axis of the left ventricle, almost parallel with the LV wall and is therefore an integral part of the longitudinal function of the left ventricle.1–4 The importance of the PM in the LV systolic function has been suggested before in animal studies,5 where cutting the chordae after mitral valve replacement resulted in immediate shape change and reduced performance of the left ventricle. Preservation of the subvalvular apparatus during mitral valve replacement is recommended by current surgical guidelines6 and has been shown to be associated with a more favourable long-term outcome,7,8 indicating that the contribution to the LV performance might be relevant. However, a proper understanding of the role of the PM within the LV chamber is still lacking, especially in hearts where their geometry and function is altered. Left ventricular dilatation results in a more spherical shape of the chamber and alters the geometry of the subvalvular apparatus due to displacement of the PM.9–13 The subvalvular apparatus thereby acts less parallel with the LV wall and more cross-sectional through the ventricle. However, little is known how these changes influence the contribution of the PM to LV performance. Regional myocardial uptake of the 18F-fluorodeoxyglucose (FDG) tracer assessed with positron emission tomography (PET) is dependent on tissue metabolism, and may therefore, be used as surrogate marker of regional myocardial work distribution.14,15 One of the limitations of clinically reconstructed PET images is the relatively low spatial resolution. However, reconstructions with resolution modelling together with both electrocardiogram (ECG)-triggering and respiratory gating can enhance spatial resolution considerably, allowing to resolve structures smaller than the thickness of a regular LV wall.16,17 The aim of this study was to assess the contribution of PM to LV work and its changes with LV dilatation. We therefore investigated the relation between the work of the PM and the remaining left ventricluar myocardium (MYO) using FDG-PET, in normal animals and in an animal model of dilated cardiomyopathy. Methods Study plan Fifteen female Swifter x Charolais crossbreed sheep were investigated in this study (average age: 12 months; body weight: 42 ± 3 kg). Five animals served as controls with normal LV geometry and function (NORMAL group), while ten were implanted with a dual-chamber pacemaker (Adapta L DDDR; Medtronic, Heerlen, The Netherlands) and subjected to 8 weeks of rapid (180 bpm) pacing to induce biventricular dilatation (DILATED group).18 A detailed description of the experimental set-up can be found in the Supplementary data online, Pacemaker implantation. The study was approved by the animal ethical committee of our institution (project number P146/2012) and complied with the European Commission Directive 2010/63/EU for the protection of animals used for scientific purposes. Animal preparation for imaging For all imaging procedures, the animals were anaesthetized and mechanically ventilated, using the same protocol as during the pacemaker implantation procedure. The heart rate of the paced animals was set to a fixed and reproducible 110 bpm in AAI pacing mode to ensure normal atrioventricular conduction and function. Cardiac computed tomography All animals underwent a contrast-enhanced cardiac computed tomography (CT) scan (128 slice; Siemens Definition Flash, Forchheim, Germany). A detailed description of the CT parameters can be found in the Supplementary data online, Cardiac computed tomography. Image acquisition was performed during breath hold, by discontinuing mechanical ventilation, followed by a recovery period. An ECG was recorded simultaneously to allow retrospective gating. Computed tomography data analysis The acquired CT data was reconstructed into ten cardiac phases, in steps of 10% during the cardiac cycle. Bins with the largest and smallest LV chamber size were used as end-diastolic and end-systolic phase, respectively. To study the size and geometry of the left ventricle in normal and dilated hearts, 2D image planes where reconstructed in four-, two-, and three-chamber views (Figure 1). For analysing the geometry of the subvalvular apparatus in relation to the LV walls and the mitral valve, dedicated image planes where reconstructed, which showed both PMs in a longitudinal cross-section (Figure 2). Figure 1 View largeDownload slide Assessment of structure and function of the LV. In both the NORMAL (A) and DILATED study group (B), the LV cavity was contoured and the sphericity index was calculated as the left ventricular width (W) divided by the left ventricular length (L) at end-diastole. Further, end-diastolic volume (EDV), end-systolic volume (ESV), global longitudinal strain (GLS), and left ventricular ejection fraction (LVEF) where calculated. Measurements were performed in apical four chamber views (shown here) and apical two-chamber views. Figure 1 View largeDownload slide Assessment of structure and function of the LV. In both the NORMAL (A) and DILATED study group (B), the LV cavity was contoured and the sphericity index was calculated as the left ventricular width (W) divided by the left ventricular length (L) at end-diastole. Further, end-diastolic volume (EDV), end-systolic volume (ESV), global longitudinal strain (GLS), and left ventricular ejection fraction (LVEF) where calculated. Measurements were performed in apical four chamber views (shown here) and apical two-chamber views. Figure 2 View largeDownload slide Assessment of structure and function of the subvalvular apparatus. Dedicated modified two-chamber view slices where reconstructed, which were parallel and in-line with both papillary muscles. In both the NORMAL (A) and DILATED hearts (B) the following parameters were determined at end-diastole: angle of papillary muscles (α: anterior and β: posterior), length and fractional shortening of the papillary muscles (including end-systole), and distance between the tips of the papillary muscles (D). This view was also used for measuring the thickness of the LV wall at the PM insertions (white arrowheads, dα: anterior and dβ: posterior). Figure 2 View largeDownload slide Assessment of structure and function of the subvalvular apparatus. Dedicated modified two-chamber view slices where reconstructed, which were parallel and in-line with both papillary muscles. In both the NORMAL (A) and DILATED hearts (B) the following parameters were determined at end-diastole: angle of papillary muscles (α: anterior and β: posterior), length and fractional shortening of the papillary muscles (including end-systole), and distance between the tips of the papillary muscles (D). This view was also used for measuring the thickness of the LV wall at the PM insertions (white arrowheads, dα: anterior and dβ: posterior). Left ventricular geometry and function All geometrical analysis of the CT images was performed offline using an in-house developed, MATLAB-based (MathWorks, Natick, MA, USA) software (TVA, J.-U. Voigt, Leuven, Belgium). Volumes and left ventricular ejection fraction (LVEF) were calculated from the end-systolic and end-diastolic frame of the cine sequences of the four-chamber and two-chamber views using the biplane method of disks. For this, endocardial borders were manually traced, and the PM were regarded as being part of the LV cavity. The LV sphericity index was calculated in end-diastole by dividing the four-chamber LV width at basal level by the LV length (Figure 1, W and L).19,20 Global longitudinal strain (GLS) was assessed offline on the cine sequences of the four-, three-, and two-chamber views, using the Syngo Velocity Vector Imaging software (VVI version 6, Siemens, Erlangen, Germany). Peak systolic longitudinal strain was determined per apical view, and GLS was calculated as the average of all three apical views. Papillary muscle geometry and function Modified two-chamber views in the end-diastolic and end-systolic phase were reconstructed to analyse the subvalvular apparatus geometry and function. The image planes were chosen to ensure that tip and base of both PM were visible in the same slice. First, reference lines were drawn through the middle of the anterior and inferior wall at the level of the PM. Subsequently, reference lines were drawn through the longitudinal axis of both PM, ranging from the PM tip towards the centre of the PM basis and the angle between both lines was measured in end-diastole (Figure 2, α and β). PM length was assessed by measuring the distance between the PM tip and the centre of the PM basis. The fractional shortening of both PM (expressed in percent) was calculated by dividing the difference between end-diastolic and end-systolic length by the end-diastolic length of the PM.12 Papillary muscles end-diastolic width was measured perpendicular to the PM length line at the PM basis. In end-diastole, the distance between the tips of the PM (Figure 2D), and the thickness of the LV wall at the level of the PM insertions were also measured (Figure 2, dα and dβ). 18F-fluorodeoxyglucose positron emission tomography Animals were scanned for 30 min on a Biograph 16 HireZ PET/CT (Siemens) scanner, 30 min after intravenous injection of 370 MBq (10 mCi) FDG. To reduce myocardial fatty acid metabolism and to stimulate insulin-dependent glucose uptake, plasma glucose levels were titrated using a hyper-insulinemic euglycaemic clamping method as described previously.21 A low dose CT scan for subsequent attenuation correction was done prior to every PET emission scan. Motion and partial volume correction Classical clinical PET scans have a rather low spatial resolution. In order to improve this, PET images were reconstructed using resolution modelling on both ECG and respiratory gated nuclear raw data.16,17 By means of resolution modelling, ECG-triggering and respiratory gating, they can be motion- and partial volume corrected, drastically improving their spatial resolution. A detailed description of the PET motion and partial volume correction procedure can be found in the Supplementary data online, Motion and partial volume correction and PET image resolution. Data analysis Regional FDG uptake was evaluated offline on static images reconstructed from the 30 min scan time using an in-house developed program.16,17 Data were analysed in end-diastole and end-expiration on two long-axis slices of the left ventricle, each parallel and in-line with one of the PM (Figure 4, left panels). On each slice, two regions of interest (ROI) were manually delineated: (i) a region including the PM and the part of the myocardial wall directly perpendicularly adjacent to the base of it (PM); (ii) the remaining MYO. The mean FDG tracer concentration within each ROI was determined and normalized to the ROI area. Finally for both slices, the ratio of the tracer concentrations in the PM and myocardium was calculated (PM/MYO ratio). Clinical proof-of-concept For initial comparison to the animal findings, two patients were investigated, one with a dilated and one with normally-sized left ventricle. Both patients underwent the same FDG PET/CT imaging protocol and data analysis as outlined above. Myocardial scarring and major coronary artery disease was ruled out by medical history and magnetic resonance imaging including late gadolinium enhancement. The study was approved by the ethical commission of the University Hospitals Leuven and subjects gave written informed consent prior to inclusion and any study procedure. Statistical analysis Analysis was performed using SPSS Statistics 20 (IBM, Chicago, IL, USA). All continuous variables are expressed as mean ± standard deviation if normally distributed, or otherwise by median ± interquartile range. Normality was assessed with the use of the Shapiro–Wilk test. All geometrical measurements were analysed in three-fold by a single experienced observer, and average values were used and reported afterwards. Unpaired samples t-tests were used to compare mean differences between the normal and dilated study group. Differences within the PM were examined by using the paired samples t-test. Univariate linear regression analysis with Pearson’s correlation coefficients was used to correlate variables of left ventricle and subvalvular apparatus function and geometry with the PM/MYO ratio. Subsequently, contributors with a significant correlation in the univariate analysis were entered in a multiple linear regression model, using the backwards method, and after evaluation of possible collinearity. 18F-fluorodeoxyglucose activity, LV sphericity index and PM angles were re-assessed by a second experienced observer in order to evaluate the inter-observer agreement using intra-class correlation (ICC). The same variables were also used for intra-observer agreement. Statistical significance was set at a two-tailed probability level of P < 0.05. Results Left ventricular chamber geometry and function The animals subjected to rapid pacing (DILATED group) developed dilated hearts with reduced function compared with controls (NORMAL group) (Table 1, Figures 1 and 2). The DILATED group had significantly higher LV EDV (P = 0.017) and ESV (P = 0.002) and a significantly wider LV chamber (P < 0.0001) while the LV length did not differ significantly (P = 0.815). Consequently, the sphericity index was significantly higher (P < 0.0001). Left ventricular ejection fraction (P < 0.0001) and GLS (P < 0.0001) were significantly decreased. The wall thickness at the PM insertions remained unaltered in dilated hearts, when compared with normal hearts (P = 0.902 for the anterior wall and P = 0.413 for the posterior wall, respectively) (Table 1 and Figure 2). Table 1 Left ventricular geometry NORMAL DILATED P-value Heart rate (bpm) 110 110 N.S. LV EDV (mL) 87 ± 12 113 ± 20 0.017 LV ESV (mL) 29 ± 4 69 ± 23 0.002 LVEF (%) 66 ± 3 39 ± 4 <0.0001 GLS (%) −14.7 ± 0.6 −8.7 ± 2 <0.0001 Sphericity index 0.48 ± 0.03 0.61 ± 0.04 <0.0001 Width (cm) 3.59 ± 0.14 4.66 ± 0.39 <0.0001 Length (cm) 7.56 ± 0.44 7.62 ± 0.46 0.815 Thickness anterior wall (cm) 0.78 ± 0.07 0.784 ± 0.038 0.902 Thickness posterior wall (cm) 0.783 ± 0.081 0.817 ± 0.058 0.413 NORMAL DILATED P-value Heart rate (bpm) 110 110 N.S. LV EDV (mL) 87 ± 12 113 ± 20 0.017 LV ESV (mL) 29 ± 4 69 ± 23 0.002 LVEF (%) 66 ± 3 39 ± 4 <0.0001 GLS (%) −14.7 ± 0.6 −8.7 ± 2 <0.0001 Sphericity index 0.48 ± 0.03 0.61 ± 0.04 <0.0001 Width (cm) 3.59 ± 0.14 4.66 ± 0.39 <0.0001 Length (cm) 7.56 ± 0.44 7.62 ± 0.46 0.815 Thickness anterior wall (cm) 0.78 ± 0.07 0.784 ± 0.038 0.902 Thickness posterior wall (cm) 0.783 ± 0.081 0.817 ± 0.058 0.413 EDV, end-diastolic volume; ESV, end-systolic volume; GLS, global longitudinal strain; LV, left ventricle; LVEF, left ventricular ejection fraction; N.S., not significant. Table 1 Left ventricular geometry NORMAL DILATED P-value Heart rate (bpm) 110 110 N.S. LV EDV (mL) 87 ± 12 113 ± 20 0.017 LV ESV (mL) 29 ± 4 69 ± 23 0.002 LVEF (%) 66 ± 3 39 ± 4 <0.0001 GLS (%) −14.7 ± 0.6 −8.7 ± 2 <0.0001 Sphericity index 0.48 ± 0.03 0.61 ± 0.04 <0.0001 Width (cm) 3.59 ± 0.14 4.66 ± 0.39 <0.0001 Length (cm) 7.56 ± 0.44 7.62 ± 0.46 0.815 Thickness anterior wall (cm) 0.78 ± 0.07 0.784 ± 0.038 0.902 Thickness posterior wall (cm) 0.783 ± 0.081 0.817 ± 0.058 0.413 NORMAL DILATED P-value Heart rate (bpm) 110 110 N.S. LV EDV (mL) 87 ± 12 113 ± 20 0.017 LV ESV (mL) 29 ± 4 69 ± 23 0.002 LVEF (%) 66 ± 3 39 ± 4 <0.0001 GLS (%) −14.7 ± 0.6 −8.7 ± 2 <0.0001 Sphericity index 0.48 ± 0.03 0.61 ± 0.04 <0.0001 Width (cm) 3.59 ± 0.14 4.66 ± 0.39 <0.0001 Length (cm) 7.56 ± 0.44 7.62 ± 0.46 0.815 Thickness anterior wall (cm) 0.78 ± 0.07 0.784 ± 0.038 0.902 Thickness posterior wall (cm) 0.783 ± 0.081 0.817 ± 0.058 0.413 EDV, end-diastolic volume; ESV, end-systolic volume; GLS, global longitudinal strain; LV, left ventricle; LVEF, left ventricular ejection fraction; N.S., not significant. Mitral valve subvalvular apparatus Details may be found in Table 2 and Figure 2. In both, the NORMAL and DILATED group, the anterior PMs were more obliquely attached to the LV wall than the posterior PMs (both P = 0.001). In dilated hearts, however, the angle between the posterior PM and the LV wall was significantly larger (P = 0.001) compared with normal hearts. The angle of the anterior PM was similar in both groups (P = 0.68). In dilated hearts, the distance between the PM tips was higher compared with the normal hearts (P = 0.021). Both the anterior and posterior PM were significantly longer in the DILATED group, compared with NORMAL group (Table 2). In addition, the PM in the DILATED group showed significantly lower fractional shortening (anterior: P = 0.001; posterior: P < 0.0001). The fractional shortening of the anterior and posterior PM correlated significantly with GLS (r = 0.74, P = 0.001 and r = 0.85, P < 0.0001, respectively) (Figure 3). Thickness measurements revealed that in the dilated LV, both the anterior PM (P = 0.005) and posterior PM (P = 0.317) were thinner than in the normal hearts, albeit only significant for the anterior PM (Table 2). Table 2 Papillary muscle geometry and function NORMAL DILATED P-value Anterior PM  Angle, ED (°) 41 ± 5 42 ± 6 0.68  Length, ED (cm) 1.55 ± 0.18 1.77 ± 0.11 0.031  Length, ES (cm) 1.13 ± 0.13 1.50 ± 0.14 0.001  PM fractional shortening (%) 29.04 ± 9.18 13.73 ± 5.72 0.001  Width, ED (cm) 0.97 ± 0.04 0.79 ± 0.11 0.005 Posterior PM  Angle, ED (°) 21 ± 1 32 ± 6 0.0001  Length, ED (cm) 1.47 ± 0.25 1.82 ± 0.17 0.033  Length, ES (cm) 1.18 ± 0.19 1.66 ± 0.18 0.002  PM fractional shortening (%) 19.99 ± 2.22 6.36 ± 4.93 <0.0001  Width, ED (cm) 0.85 ± 0.03 0.78 ± 0.15 0.317 PM tips distance (cm) 2.92 ± 0.34 3.60 ± 0.43 0.021 NORMAL DILATED P-value Anterior PM  Angle, ED (°) 41 ± 5 42 ± 6 0.68  Length, ED (cm) 1.55 ± 0.18 1.77 ± 0.11 0.031  Length, ES (cm) 1.13 ± 0.13 1.50 ± 0.14 0.001  PM fractional shortening (%) 29.04 ± 9.18 13.73 ± 5.72 0.001  Width, ED (cm) 0.97 ± 0.04 0.79 ± 0.11 0.005 Posterior PM  Angle, ED (°) 21 ± 1 32 ± 6 0.0001  Length, ED (cm) 1.47 ± 0.25 1.82 ± 0.17 0.033  Length, ES (cm) 1.18 ± 0.19 1.66 ± 0.18 0.002  PM fractional shortening (%) 19.99 ± 2.22 6.36 ± 4.93 <0.0001  Width, ED (cm) 0.85 ± 0.03 0.78 ± 0.15 0.317 PM tips distance (cm) 2.92 ± 0.34 3.60 ± 0.43 0.021 ED, end-diastole; ES, end-systole; PM, papillary muscle. Table 2 Papillary muscle geometry and function NORMAL DILATED P-value Anterior PM  Angle, ED (°) 41 ± 5 42 ± 6 0.68  Length, ED (cm) 1.55 ± 0.18 1.77 ± 0.11 0.031  Length, ES (cm) 1.13 ± 0.13 1.50 ± 0.14 0.001  PM fractional shortening (%) 29.04 ± 9.18 13.73 ± 5.72 0.001  Width, ED (cm) 0.97 ± 0.04 0.79 ± 0.11 0.005 Posterior PM  Angle, ED (°) 21 ± 1 32 ± 6 0.0001  Length, ED (cm) 1.47 ± 0.25 1.82 ± 0.17 0.033  Length, ES (cm) 1.18 ± 0.19 1.66 ± 0.18 0.002  PM fractional shortening (%) 19.99 ± 2.22 6.36 ± 4.93 <0.0001  Width, ED (cm) 0.85 ± 0.03 0.78 ± 0.15 0.317 PM tips distance (cm) 2.92 ± 0.34 3.60 ± 0.43 0.021 NORMAL DILATED P-value Anterior PM  Angle, ED (°) 41 ± 5 42 ± 6 0.68  Length, ED (cm) 1.55 ± 0.18 1.77 ± 0.11 0.031  Length, ES (cm) 1.13 ± 0.13 1.50 ± 0.14 0.001  PM fractional shortening (%) 29.04 ± 9.18 13.73 ± 5.72 0.001  Width, ED (cm) 0.97 ± 0.04 0.79 ± 0.11 0.005 Posterior PM  Angle, ED (°) 21 ± 1 32 ± 6 0.0001  Length, ED (cm) 1.47 ± 0.25 1.82 ± 0.17 0.033  Length, ES (cm) 1.18 ± 0.19 1.66 ± 0.18 0.002  PM fractional shortening (%) 19.99 ± 2.22 6.36 ± 4.93 <0.0001  Width, ED (cm) 0.85 ± 0.03 0.78 ± 0.15 0.317 PM tips distance (cm) 2.92 ± 0.34 3.60 ± 0.43 0.021 ED, end-diastole; ES, end-systole; PM, papillary muscle. Figure 3 View largeDownload slide Correlation plots of papillary muscle (PM) fractional shortening and global longitudinal strain (GLS). (A): anterior PM. (B): posterior PM. Blue and red markers represent respectively NORMAL and DILATED study groups. Figure 3 View largeDownload slide Correlation plots of papillary muscle (PM) fractional shortening and global longitudinal strain (GLS). (A): anterior PM. (B): posterior PM. Blue and red markers represent respectively NORMAL and DILATED study groups. Papillary muscle work In normal hearts, the FDG activity in both PM regions was slightly higher compared with the remaining myocardium, without differences between anterior and posterior PM (PM/MYO ratio 1.20 ± 0.05 and 1.16 ± 0.06; P = 0.419) (Figure 4, bar chart). In dilated hearts, a reproducible pattern of increased FDG activity at the position of the PMs was observed (Figure 4B and D) with a significantly higher PM/MYO ratio for both PM compared to normal hearts (anterior: 1.47 ± 0.06 vs. 1.20 ± 0.05; P < 0.0001 and posterior: 1.31 ± 0.07 vs. 1.16 ± 0.06; P = 0.002, respectively). The pattern was significantly more pronounced in the anterior PM (P < 0.0001) (Figure 4, bar chart). Figure 4 View largeDownload slide Assessment of papillary muscles work. In two long-axis slices, through each papillary muscle (PM), the 18F-fluorodeoxyglucose (FDG) tracer activity was calculated. Representative views of the anterior (A + B) and posterior PM (C + D) of the NORMAL (A + C) and DILATED (B + D) study group are shown. Note the ‘hotspots’ of increased FDG activity in the PM on the images of the DILATED study group. Activity in PM [region of interest (ROI) with yellow contour] and the remaining LV myocardium (ROI with white contour) was calculated and normalized to the respective ROI area. Contribution of PM to myocardial work is shown as PM/MYO ratio in the bar chart. Asterisks indicate P < 0.05. Figure 4 View largeDownload slide Assessment of papillary muscles work. In two long-axis slices, through each papillary muscle (PM), the 18F-fluorodeoxyglucose (FDG) tracer activity was calculated. Representative views of the anterior (A + B) and posterior PM (C + D) of the NORMAL (A + C) and DILATED (B + D) study group are shown. Note the ‘hotspots’ of increased FDG activity in the PM on the images of the DILATED study group. Activity in PM [region of interest (ROI) with yellow contour] and the remaining LV myocardium (ROI with white contour) was calculated and normalized to the respective ROI area. Contribution of PM to myocardial work is shown as PM/MYO ratio in the bar chart. Asterisks indicate P < 0.05. Determinants of papillary muscle work In linear regression analyses combining normal and dilated hearts, sphericity index, posterior PM angle, LVEF, LV EDV, GLS, distance between PM tips, and fractional shortening of the posterior PM were significantly correlated with the PM/MYO ratio of the posterior PM, which was not the case for the posterior PM width (Table 3). Multivariable linear regression analysis revealed that the sphericity index and posterior PM angle were the strongest determinants of the PM/MYO ratio of the posterior PM (R2 = 0.754, P < 0.0001) (Figure 5, lower panels for the univariable regression plots). Table 3 Determinants of papillary muscle contribution to left ventricular workload (PM/MYO) Univariate Multivariate r P-value R2 P-value Anterior PM  Sphericity index (*) 0.882 <0.0001 0.805 <0.0001  PM width (*) 0.747 0.001  Anterior PM angle 0.360 0.188  LVEF 0.816 <0.0001  LV EDV (*) 0.575 0.025  GLS 0.848 <0.0001  PM tips width 0.474 0.102  PM fractional shortening (*) 0.679 0.005 Posterior PM  Sphericity index (*) 0.805 <0.0001 0.754 <0.0001  Posterior PM angle (*) 0.856 <0.0001  PM width 0.380 0.163  LVEF 0.687 0.005  LV EDV (*) 0.686 0.005  GLS (*) 0.717 0.003  PM tips width 0.622 0.023  PM fractional shortening (*) 0.746 0.001 Univariate Multivariate r P-value R2 P-value Anterior PM  Sphericity index (*) 0.882 <0.0001 0.805 <0.0001  PM width (*) 0.747 0.001  Anterior PM angle 0.360 0.188  LVEF 0.816 <0.0001  LV EDV (*) 0.575 0.025  GLS 0.848 <0.0001  PM tips width 0.474 0.102  PM fractional shortening (*) 0.679 0.005 Posterior PM  Sphericity index (*) 0.805 <0.0001 0.754 <0.0001  Posterior PM angle (*) 0.856 <0.0001  PM width 0.380 0.163  LVEF 0.687 0.005  LV EDV (*) 0.686 0.005  GLS (*) 0.717 0.003  PM tips width 0.622 0.023  PM fractional shortening (*) 0.746 0.001 Combined analysis of NORMAL and DILATED hearts. Asterisk (*) denotes variables that were entered in the multivariate analysis, after being checked for significance in univariate analysis and rendering no collinearity issues. EDV, end-diastolic volume; GLS, global longitudinal strain; LV, left ventricle; LVEF, left ventricular ejection fraction; PM, papillary muscle. Table 3 Determinants of papillary muscle contribution to left ventricular workload (PM/MYO) Univariate Multivariate r P-value R2 P-value Anterior PM  Sphericity index (*) 0.882 <0.0001 0.805 <0.0001  PM width (*) 0.747 0.001  Anterior PM angle 0.360 0.188  LVEF 0.816 <0.0001  LV EDV (*) 0.575 0.025  GLS 0.848 <0.0001  PM tips width 0.474 0.102  PM fractional shortening (*) 0.679 0.005 Posterior PM  Sphericity index (*) 0.805 <0.0001 0.754 <0.0001  Posterior PM angle (*) 0.856 <0.0001  PM width 0.380 0.163  LVEF 0.687 0.005  LV EDV (*) 0.686 0.005  GLS (*) 0.717 0.003  PM tips width 0.622 0.023  PM fractional shortening (*) 0.746 0.001 Univariate Multivariate r P-value R2 P-value Anterior PM  Sphericity index (*) 0.882 <0.0001 0.805 <0.0001  PM width (*) 0.747 0.001  Anterior PM angle 0.360 0.188  LVEF 0.816 <0.0001  LV EDV (*) 0.575 0.025  GLS 0.848 <0.0001  PM tips width 0.474 0.102  PM fractional shortening (*) 0.679 0.005 Posterior PM  Sphericity index (*) 0.805 <0.0001 0.754 <0.0001  Posterior PM angle (*) 0.856 <0.0001  PM width 0.380 0.163  LVEF 0.687 0.005  LV EDV (*) 0.686 0.005  GLS (*) 0.717 0.003  PM tips width 0.622 0.023  PM fractional shortening (*) 0.746 0.001 Combined analysis of NORMAL and DILATED hearts. Asterisk (*) denotes variables that were entered in the multivariate analysis, after being checked for significance in univariate analysis and rendering no collinearity issues. EDV, end-diastolic volume; GLS, global longitudinal strain; LV, left ventricle; LVEF, left ventricular ejection fraction; PM, papillary muscle. Figure 5 View largeDownload slide Correlation plots of PM/MYO ratio versus its determinants. The black dotted regression lines refer to the combined analysis of NORMAL and DILATED hearts (results in the graph in black). The red dotted regression line and the red numbers refer to a separate analysis of DILATED hearts only. Blue and red markers represent respectively NORMAL and DILATED study groups. Figure 5 View largeDownload slide Correlation plots of PM/MYO ratio versus its determinants. The black dotted regression lines refer to the combined analysis of NORMAL and DILATED hearts (results in the graph in black). The red dotted regression line and the red numbers refer to a separate analysis of DILATED hearts only. Blue and red markers represent respectively NORMAL and DILATED study groups. The PM/MYO ratio of the anterior PM showed significant correlation with the same parameters, including the anterior PM width, except for the anterior PM angle (Table 3). However, a significant correlation was still obtained between the PM/MYO ratio and anterior PM angle, when considering only the dilated hearts (Figure 5B). After multivariable linear regression analysis, the sphericity index and anterior PM width remained as strongest determinant of the PM/MYO of the anterior PM (R2 = 0.805, P < 0.0001) (Figure 5, upper panels for the univariable regression plots). Reproducibility Analysis of the intra-observer variability of the LV sphericity index and PM angles showed good reproducibility: [ICC 0.852, 95% confidence interval (CI) 0.772–0.946] and (ICC 0.966, 95% CI 0.936–0.984), respectively. Inter-observer variability also showed good ICC for FDG activity (ICC 0.985, 95% CI 0.866–0.993), LV sphericity index (ICC 0.86, 95% CI 0.722–0.946), and PM angles (ICC 0.945, 95% CI 0.882–0.975). Discussion Main findings of the study In this study, we evaluated the contribution of PM to global LV function in normal and dilated hearts of animals, using myocardial glucose uptake assessed with FDG PET as surrogate of myocardial work.14,15 We found that in dilated hearts: (i) PM have a significantly increased relative contribution to LV work and (ii) this increased contribution is significantly related to geometric changes of the dilated LV. Role of the papillary muscles in normal and dilated hearts In normal hearts, the PM/MYO ratio was slightly higher than one, which allows the interpretation that more work per unit volume is performed by the PM compared with the rest of the MYO. An alternative explanation of our observation of an increased tracer activity in the PM region could be a partial volume effect, since a larger volume of active myocardium is present in the region where the PM comes close to the ventricular wall. Although our efforts of resolution modelling with ECG-triggering and respiratory gating resulted in a dramatic increase of spatial resolution compared with clinical PET reconstructions, this explanation is not unlikely, especially when considering that in normal hearts the direction of force development in PM is more parallel to the LV wall, so that (similar loading assumed) equal work can be expected. In dilated hearts, our findings indicate that there is a significantly increased contribution of the PM to the LV myocardial work. This increased work cannot be explained by an augmented systolic shortening of the PM compared to the rest of the left ventricle. On the contrary, the length of the PM was increased as was the LV size, while the fractional shortening was proportionally reduced compared with the GLS of the left ventricle (Figure 3). Consequently, the higher work of the PM compared with the rest of the left ventricle myocardium must be due to a higher stress caused by the altered orientation of the PM and subvalvular apparatus in the dilated and more spherical hearts, which becomes more cross-sectional to the left ventricle and less parallel with its walls. Furthermore, we found the anterior PMs in dilated hearts to be thinner than in normal hearts, resulting in smaller cross-sectional area and—with this—higher stress. We therefore hypothesize, that the higher work of the PM compared to the rest of the MYO is truly caused by a higher stress. This hypothesis is supported by earlier experimental studies, which have demonstrated that the PM of failing or dilated hearts can generate up to 30% more force as in the normal heart.22,23 Also in dilated hearts, partial volume effects must be considered as potential explanation of our observations. In this case, however, this explanation seems unlikely, since the PM/MYO ratio was much higher. In addition, PM were further away from the LV wall so that they could be clearly distinguished given the augmented resolution of our dual gated PET reconstruction. Finally, LV wall thickness measurements were not significantly different in normal and dilated hearts, so that the partial volume effect in the wall should be comparable in both groups. Determinants of increased papillary muscle work Left ventricular dilatation leads to a more spherical configuration of the LV chamber. Our measurements indicate that this is accompanied by a substantial change in the configuration of the subvalvular apparatus. The PM were displaced more outward, as indicated by the increased distance between their tips. This observation is concordant with previous experimental12,13 and patient studies.9,10 Our data indicate further, that the PM act also no longer parallel, but more angulated to the LV wall. As the angle between PM and the LV wall increases, PM and chordae run more cross-sectional through the LV chamber, which exposes the PM to a higher stress than the curved wall of the spherical ventricle is exposed to.24,25 This hypothesis is supported by our differential findings in the anterior and posterior PM of the dilated hearts. Both the anterior and posterior PM contributed significantly more to the LV myocardial work in the dilated hearts, with the highest contribution delivered by the anterior PM, which had also the larger angle towards the wall (Figure 5). Although on average, the angle of the anterior PM did not differ significantly between normal and dilated hearts, we also found a significant correlation between angle and work contribution in the dilated hearts (Figure 5B, red dotted line). The particular curved shape of the ovine LV chamber26 may have contributed to the differential observations in both PM. While the posterior wall is quite straight, the anterior wall is already relatively curved in normal hearts and this difference remains with LV dilatation.26 Furthermore, our data indicate that in dilated hearts particularly the anterior PM is thinner than normal. Consequently, its width appeared as one of the strongest determinants of the PM/MYO ratio for the anterior PM, which might in part also explain its increased metabolism. Clinical proof-of-concept The FDG PET measurements in our proof-of-concept-patients led to very comparable results as in the animals (Figure 6). In the dilated heart, the contribution of the anterior PM to the LV work was increased compared to the patient with the non-dilated heart. Similarly, also the angle of this PM was clearly larger in the dilated heart than in the non-dilated heart, together with an apparent spherical shape. We therefore assume, that our findings can be applied to the situation in the human heart. Figure 6 View largeDownload slide Clinical validation: assessment of papillary muscles work. Arrangement of panels and colour scheme are similar as in Figure 4. Interestingly, the findings in this proof-of-concept analysis in two patients are comparable to the results of our animal experiments. Figure 6 View largeDownload slide Clinical validation: assessment of papillary muscles work. Arrangement of panels and colour scheme are similar as in Figure 4. Interestingly, the findings in this proof-of-concept analysis in two patients are comparable to the results of our animal experiments. Clinical relevance Observational studies, investigating the effect of mitral valve replacement with and without preservation of the mitral subvalvular apparatus on LV performance, have shown in both animal experiments5 and patients,8 that preserving the subvalvular apparatus results in better post-operative LV shape, ejection fraction and contractility. While the results of these studies have led to the recommendation in surgical guidelines to preserve the subvalvular apparatus,6 the pathophysiologic understanding of the importance of PM function was lacking. Our study results indicate that the contribution of PM to LV performance is relevant, and suggest that their importance is even higher in dilated (e.g. volume loaded) hearts. Therefore, our findings provide evidence for the—so far—empiric practice of preserving the PM during surgical mitral valve replacement. Study limitations The sample size used in this study was small, which is inherent to fundamental animal research. However, the magnitude of observed changes was comparable to other studies in the field and effects were highly significant. Nevertheless, further validation should be performed in a larger cohort. The 11C-acetate tracer (complete oxidative metabolism) would be a valuable alternative to the FDG tracer (glucose metabolism only) used in this study, providing a more general information on the overall myocardial metabolism. However, 11C-acetate has a very short half-life, necessitating an adjacent cyclotron and requires kinetic modelling. Therefore, we used a standard glucose-insuline-clamping technique with FDG to ensure a predominant glucose metabolism in the myocardium. Although sheep are an established animal model, with translational findings to humans,27 particularly the geometry of the normal and dilated chamber may not completely mimic the situation in human pathology. However, in both sheep and humans, a higher work contribution of PM was found in dilated hearts, particularly high in the PM with the largest angle towards the wall. Nevertheless, our proof-of-concept in a small sample of human subjects need further validation in a larger clinical study. Conclusions Our study demonstrated that in dilated hearts, the PMs contribute relatively more to LV work, compared with normal hearts. As a consequence of the LV dilatation, the geometry of the subvalvular apparatus is altered, making the PMs to be thinner and to work more across the LV and less parallel to the LV wall, which results in a higher stress. Our findings emphasise the important contribution of PMs to LV systolic function, and the understanding that they should be preserved, e.g. during mitral repair/replacement. Supplementary data Supplementary data are available at European Heart Journal—Cardiovascular Imaging online. Acknowledgements The authors are indebted to David Célis for his excellent technical assistance during the experiments and care of the animals, and wish to thank Hermien Coppens and Marta Cvijic for assisting with data analysis and preparation of the study. Funding J.D. received the ‘Young Investigator Award—Basic Science’ from the European Association of Cardiovascular Imaging (EACVI) during EuroEcho-Imaging 2016. This work was supported by the KU Leuven [research grant OT/12/084]. The Research Foundation—Flanders (FWO) supported A.T. as a doctoral fellow, K.V. as a post-doctoral fellow, and O.G. and J.-U.V. as senior clinical investigators. A.S.B. received a research grant from the Egyptian Ministry of Higher Education. E.D.P. and S.U. received a research grant from the European Association of Cardiovascular Imaging (EACVI). Conflict of interest: None declared. References 1 Jones CJ , Raposo L , Gibson DG. Functional importance of the long axis dynamics of the human left ventricle . Br Heart J 1990 ; 63 : 215 – 20 . Google Scholar CrossRef Search ADS PubMed 2 Kono T , Sabbah HN , Rosman H , Alam M , Jafri S , Goldstein S. Left ventricular shape is the primary determinant of functional mitral regurgitation in heart failure . J Am Coll Cardiol 1992 ; 20 : 1594 – 8 . Google Scholar CrossRef Search ADS PubMed 3 Sarris GE , Fann JI , Niczyporuk MA , Derby GC , Handen CE , Miller DC. Global and regional left ventricular systolic performance in the in situ ejecting canine heart. Importance of the mitral apparatus . Circulation 1989 ; 80 : I24 – 42 . Google Scholar CrossRef Search ADS PubMed 4 Hansen DE , Cahill PD , DeCampli WM , Harrison DC , Derby GC , Mitchell RS. Valvular-ventricular interaction: importance of the mitral apparatus in canine left ventricular systolic performance . Circulation 1986 ; 73 : 1310 – 20 . Google Scholar CrossRef Search ADS PubMed 5 Yun KL , Niczyporuk MA , Sarris GE , Fann JI , Miller DC. Importance of mitral subvalvular apparatus in terms of cardiac energetics and systolic mechanics in the ejecting canine heart . J Clin Invest 1991 ; 87 : 247 – 54 . Google Scholar CrossRef Search ADS PubMed 6 Vahanian A , Alfieri O , Andreotti F , Antunes MJ , Barón-Esquivias G , Baumgartner H et al. Guidelines on the management of valvular heart disease (version 2012): the Joint Task Force on the Management of Valvular Heart Disease of the European Society of Cardiology (ESC) and the European Association for Cardio-Thoracic Surgery (EACTS) . Eur Heart J 2012 ; 33 : 2451 – 96 . Google Scholar CrossRef Search ADS PubMed 7 Pitarys CJ , Forman MB , Panayiotou H , Hansen DE. Long-term effects of excision of the mitral apparatus on global and regional ventricular function in humans . J Am Coll Cardiol 1990 ; 15 : 557 – 63 . Google Scholar CrossRef Search ADS PubMed 8 Rozich JD , Carabello BA , Usher BW , Kratz JM , Bell AE , Zile MR. Mitral valve replacement with and without chordal preservation in patients with chronic mitral regurgitation. Mechanisms for differences in postoperative ejection performance . Circulation 1992 ; 86 : 1718 – 26 . Google Scholar CrossRef Search ADS PubMed 9 Delgado V , Tops LF , Schuijf JD , de Roos A , Brugada J , Schalij MJ et al. Assessment of mitral valve anatomy and geometry with multislice computed tomography . JACC Cardiovasc Imaging 2009 ; 2 : 556 – 65 . Google Scholar CrossRef Search ADS PubMed 10 Madu EC , Baugh DS , D’Cruz IA , Johns C. Left ventricular papillary muscle morphology and function in left ventricular hypertrophy and left ventricular dysfunction . Med Sci Monit 2001 ; 7 : 1212 – 8 . Google Scholar PubMed 11 Kwan J , Shiota T , Agler DA , Popović ZB , Qin JX , Gillinov MA et al. Geometric differences of the mitral apparatus between ischemic and dilated cardiomyopathy with significant mitral regurgitation: real-time three-dimensional echocardiography study . Circulation 2003 ; 107 : 1135 – 40 . Google Scholar CrossRef Search ADS PubMed 12 Madu EC , D’Cruz IA. The vital role of papillary muscles in mitral and ventricular function: echocardiographic insights . Clin Cardiol 1997 ; 20 : 93 – 8 . Google Scholar CrossRef Search ADS PubMed 13 Timek TA , Dagum P , Lai DT , Liang D , Daughters GT , Tibayan F et al. Tachycardia-induced cardiomyopathy in the ovine heart: mitral annular dynamic three-dimensional geometry . J Thorac Cardiovasc Surg 2003 ; 125 : 315 – 24 . Google Scholar CrossRef Search ADS PubMed 14 Bergman BC , Tsvetkova T , Lowes B , Wolfel EE. Myocardial glucose and lactate metabolism during rest and atrial pacing in humans . J Physiol 2009 ; 587 : 2087 – 99 . Google Scholar CrossRef Search ADS PubMed 15 Russell K , Eriksen M , Aaberge L , Wilhelmsen N , Skulstad H , Remme EW et al. A novel clinical method for quantification of regional left ventricular pressure-strain loop area: a non-invasive index of myocardial work . Eur Heart J 2012 ; 33 : 724 – 33 . Google Scholar CrossRef Search ADS PubMed 16 Turco A , Gheysens O , Nuyts J , Duchenne J , Voigt JU , Claus P et al. Impact of CT-based attenuation correction on the registration between dual-gated cardiac PET and high-resolution CT . IEEE Trans Nucl Sci 2016 ; 63 : 180 – 92 . Google Scholar CrossRef Search ADS 17 Turco A. Partial volume and motion correction in cardiac PET: first results from and in vs ex vivo comparison using animal datasets. In: Turo A, ed. Improved Quantification of Myocardial FDG Uptake combining PET/CT and 4D Anatomical Imaging. Ph.D. Thesis. Leuven: KU Leuven—University of Leuven; 2016 . p. 131 – 46 . 18 Duchenne J , Claus P , Pagouralias ED , Mada RO , Van Puyvelde J , Vunckx K et al. Sheep can be used as animal model of regional myocardial remodelling and controllable work . Cardiol J 2018 [Epub ahead of print], doi:10.5603/CJ.a2018.0007. 19 Agricola E , Oppizzi M , Maisano F , De Bonis M , Schinkel AFL , Torracca L et al. Echocardiographic classification of chronic ischemic mitral regurgitation caused by restricted motion according to tethering pattern . Eur J Echocardiogr 2004 ; 5 : 326 – 34 . Google Scholar CrossRef Search ADS PubMed 20 Magne J , Pibarot P , Dagenais F , Hachicha Z , Dumesnil JG , Sénéchal M. Preoperative posterior leaflet angle accurately predicts outcome after restrictive mitral valve annuloplasty for ischemic mitral regurgitation . Circulation 2007 ; 115 : 782 – 91 . Google Scholar CrossRef Search ADS PubMed 21 Kim JK. Hyperinsulinemic–euglycemic clamp to assess insulin sensitivity in vivo . Methods Mol Biol 2009 ; 560 : 221 – 38 . Google Scholar CrossRef Search ADS PubMed 22 Burch GE , Giles TD. Angle of traction of the papillary muscle in normal and dilated hearts: a theoretic analysis of its importance in mitral valve dynamics . Am Heart J 1972 ; 84 : 141 – 4 . Google Scholar CrossRef Search ADS PubMed 23 Feldman MD , Copelas L , Gwathmey JK , Phillips P , Warren SE , Schoen FJ et al. Deficient production of cyclic AMP: pharmacologic evidence of an important cause of contractile dysfunction in patients with end-stage heart failure . Circulation 1987 ; 75 : 331 – 9 . Google Scholar CrossRef Search ADS PubMed 24 Perloff JK , Roberts WC , Perloff JK , Roberts WC. The mitral apparatus: functional anatomy of mitral regurgitation . Circulation 1972 ; 46 : 227 – 39 . Google Scholar CrossRef Search ADS PubMed 25 Kono T , Sabbah HN , Stein PD , Brymer JF , Khaja F. Left ventricular shape as a determinant of functional mitral regurgitation in patients with severe heart failure secondary to either coronary artery disease or idiopathic dilated cardiomyopathy . Am J Cardiol 1991 ; 68 : 355 – 9 . Google Scholar CrossRef Search ADS PubMed 26 Hill AJ , Iaizzo PA. Comparative cardiac anatomy. In: Iaizzo PA , ed. Handbook of Cardiac Anatomy, Physiology, and Devices . 2nd ed . New York : Springer ; 2009 . p 102 . 27 Houser SR , Margulies KB , Murphy AM , Spinale FG , Francis GS , Prabhu SD et al. Animal models of heart failure: a scientific statement from the American Heart Association . Circ Res 2012 ; 111 : 131 – 50 . Google Scholar CrossRef Search ADS PubMed Published on behalf of the European Society of Cardiology. All rights reserved. © The Author(s) 2018. For permissions, please email: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png European Heart Journal – Cardiovascular Imaging Oxford University Press

Loading next page...
 
/lp/ou_press/papillary-muscles-contribute-significantly-more-to-left-ventricular-cM46gwzehW
Publisher
Oxford University Press
Copyright
Published on behalf of the European Society of Cardiology. All rights reserved. © The Author(s) 2018. For permissions, please email: journals.permissions@oup.com.
ISSN
2047-2404
D.O.I.
10.1093/ehjci/jey043
Publisher site
See Article on Publisher Site

Abstract

Abstract Aims Left ventricular (LV) dilatation results in increased sphericity and affects position and orientation of papillary muscles (PMs), which may influence their performed work. The aim of this study was to assess the contribution of PM to LV function and its changes with dilatation. Methods and results Fifteen sheep were investigated. Ten animals were subjected to 8 weeks of rapid (180 bpm) pacing, inducing LV dilatation. Five animals served as controls. High-resolution gated computed tomography was performed to assess LV volumes, left ventricular ejection fraction (LVEF), global longitudinal strain (GLS), sphericity index, and PM angle, width and fractional shortening. 18F-fluorodeoxyglucose positron emission tomography (PET) was used to measure glucose metabolism as surrogate of regional myocardial work. Spatial resolution of PET images was maximized by electrocardiogram- and respiratory-gating. 18F-fluorodeoxyglucose uptake was measured in PM and compared with remaining left ventricular myocardium (MYO) to obtain a PM/MYO ratio. Animals with dilated heart had a more spherical left ventricle, with reduced LVEF (P < 0.0001) and GLS (P < 0.0001). In dilated hearts, PET analysis revealed a higher contribution of both PM to LV myocardial work (P < 0.0001); and PM angle towards LV wall correlated with PM work, together with PM width and the LV sphericity index. Sphericity index and posterior PM angle were strongest determinants of posterior PM/MYO ratio (R2 = 0.754; P < 0.0001), while anterior PM/MYO was mostly determined by sphericity index and the PM width (R2 = 0.805; P < 0.0001). Conclusion In dilated hearts, PM contribute relatively more to LV myocardial work. We hypothesize that this is caused by the more cross-sectional orientation of the subvalvular apparatus, which leads to a higher stress on the PM compared with the spherical LV walls. The reduced cross-sectional area of the PM may further explain their increased stress. papillary muscles , left ventricle , dilatation , work Introduction Papillary muscles (PMs) are a continuity of the longitudinal fibres of the myocardium and directly connect the left ventricular (LV) wall with the mitral valve through chordae tendineae. This so-called subvalvular apparatus acts along the longitudinal axis of the left ventricle, almost parallel with the LV wall and is therefore an integral part of the longitudinal function of the left ventricle.1–4 The importance of the PM in the LV systolic function has been suggested before in animal studies,5 where cutting the chordae after mitral valve replacement resulted in immediate shape change and reduced performance of the left ventricle. Preservation of the subvalvular apparatus during mitral valve replacement is recommended by current surgical guidelines6 and has been shown to be associated with a more favourable long-term outcome,7,8 indicating that the contribution to the LV performance might be relevant. However, a proper understanding of the role of the PM within the LV chamber is still lacking, especially in hearts where their geometry and function is altered. Left ventricular dilatation results in a more spherical shape of the chamber and alters the geometry of the subvalvular apparatus due to displacement of the PM.9–13 The subvalvular apparatus thereby acts less parallel with the LV wall and more cross-sectional through the ventricle. However, little is known how these changes influence the contribution of the PM to LV performance. Regional myocardial uptake of the 18F-fluorodeoxyglucose (FDG) tracer assessed with positron emission tomography (PET) is dependent on tissue metabolism, and may therefore, be used as surrogate marker of regional myocardial work distribution.14,15 One of the limitations of clinically reconstructed PET images is the relatively low spatial resolution. However, reconstructions with resolution modelling together with both electrocardiogram (ECG)-triggering and respiratory gating can enhance spatial resolution considerably, allowing to resolve structures smaller than the thickness of a regular LV wall.16,17 The aim of this study was to assess the contribution of PM to LV work and its changes with LV dilatation. We therefore investigated the relation between the work of the PM and the remaining left ventricluar myocardium (MYO) using FDG-PET, in normal animals and in an animal model of dilated cardiomyopathy. Methods Study plan Fifteen female Swifter x Charolais crossbreed sheep were investigated in this study (average age: 12 months; body weight: 42 ± 3 kg). Five animals served as controls with normal LV geometry and function (NORMAL group), while ten were implanted with a dual-chamber pacemaker (Adapta L DDDR; Medtronic, Heerlen, The Netherlands) and subjected to 8 weeks of rapid (180 bpm) pacing to induce biventricular dilatation (DILATED group).18 A detailed description of the experimental set-up can be found in the Supplementary data online, Pacemaker implantation. The study was approved by the animal ethical committee of our institution (project number P146/2012) and complied with the European Commission Directive 2010/63/EU for the protection of animals used for scientific purposes. Animal preparation for imaging For all imaging procedures, the animals were anaesthetized and mechanically ventilated, using the same protocol as during the pacemaker implantation procedure. The heart rate of the paced animals was set to a fixed and reproducible 110 bpm in AAI pacing mode to ensure normal atrioventricular conduction and function. Cardiac computed tomography All animals underwent a contrast-enhanced cardiac computed tomography (CT) scan (128 slice; Siemens Definition Flash, Forchheim, Germany). A detailed description of the CT parameters can be found in the Supplementary data online, Cardiac computed tomography. Image acquisition was performed during breath hold, by discontinuing mechanical ventilation, followed by a recovery period. An ECG was recorded simultaneously to allow retrospective gating. Computed tomography data analysis The acquired CT data was reconstructed into ten cardiac phases, in steps of 10% during the cardiac cycle. Bins with the largest and smallest LV chamber size were used as end-diastolic and end-systolic phase, respectively. To study the size and geometry of the left ventricle in normal and dilated hearts, 2D image planes where reconstructed in four-, two-, and three-chamber views (Figure 1). For analysing the geometry of the subvalvular apparatus in relation to the LV walls and the mitral valve, dedicated image planes where reconstructed, which showed both PMs in a longitudinal cross-section (Figure 2). Figure 1 View largeDownload slide Assessment of structure and function of the LV. In both the NORMAL (A) and DILATED study group (B), the LV cavity was contoured and the sphericity index was calculated as the left ventricular width (W) divided by the left ventricular length (L) at end-diastole. Further, end-diastolic volume (EDV), end-systolic volume (ESV), global longitudinal strain (GLS), and left ventricular ejection fraction (LVEF) where calculated. Measurements were performed in apical four chamber views (shown here) and apical two-chamber views. Figure 1 View largeDownload slide Assessment of structure and function of the LV. In both the NORMAL (A) and DILATED study group (B), the LV cavity was contoured and the sphericity index was calculated as the left ventricular width (W) divided by the left ventricular length (L) at end-diastole. Further, end-diastolic volume (EDV), end-systolic volume (ESV), global longitudinal strain (GLS), and left ventricular ejection fraction (LVEF) where calculated. Measurements were performed in apical four chamber views (shown here) and apical two-chamber views. Figure 2 View largeDownload slide Assessment of structure and function of the subvalvular apparatus. Dedicated modified two-chamber view slices where reconstructed, which were parallel and in-line with both papillary muscles. In both the NORMAL (A) and DILATED hearts (B) the following parameters were determined at end-diastole: angle of papillary muscles (α: anterior and β: posterior), length and fractional shortening of the papillary muscles (including end-systole), and distance between the tips of the papillary muscles (D). This view was also used for measuring the thickness of the LV wall at the PM insertions (white arrowheads, dα: anterior and dβ: posterior). Figure 2 View largeDownload slide Assessment of structure and function of the subvalvular apparatus. Dedicated modified two-chamber view slices where reconstructed, which were parallel and in-line with both papillary muscles. In both the NORMAL (A) and DILATED hearts (B) the following parameters were determined at end-diastole: angle of papillary muscles (α: anterior and β: posterior), length and fractional shortening of the papillary muscles (including end-systole), and distance between the tips of the papillary muscles (D). This view was also used for measuring the thickness of the LV wall at the PM insertions (white arrowheads, dα: anterior and dβ: posterior). Left ventricular geometry and function All geometrical analysis of the CT images was performed offline using an in-house developed, MATLAB-based (MathWorks, Natick, MA, USA) software (TVA, J.-U. Voigt, Leuven, Belgium). Volumes and left ventricular ejection fraction (LVEF) were calculated from the end-systolic and end-diastolic frame of the cine sequences of the four-chamber and two-chamber views using the biplane method of disks. For this, endocardial borders were manually traced, and the PM were regarded as being part of the LV cavity. The LV sphericity index was calculated in end-diastole by dividing the four-chamber LV width at basal level by the LV length (Figure 1, W and L).19,20 Global longitudinal strain (GLS) was assessed offline on the cine sequences of the four-, three-, and two-chamber views, using the Syngo Velocity Vector Imaging software (VVI version 6, Siemens, Erlangen, Germany). Peak systolic longitudinal strain was determined per apical view, and GLS was calculated as the average of all three apical views. Papillary muscle geometry and function Modified two-chamber views in the end-diastolic and end-systolic phase were reconstructed to analyse the subvalvular apparatus geometry and function. The image planes were chosen to ensure that tip and base of both PM were visible in the same slice. First, reference lines were drawn through the middle of the anterior and inferior wall at the level of the PM. Subsequently, reference lines were drawn through the longitudinal axis of both PM, ranging from the PM tip towards the centre of the PM basis and the angle between both lines was measured in end-diastole (Figure 2, α and β). PM length was assessed by measuring the distance between the PM tip and the centre of the PM basis. The fractional shortening of both PM (expressed in percent) was calculated by dividing the difference between end-diastolic and end-systolic length by the end-diastolic length of the PM.12 Papillary muscles end-diastolic width was measured perpendicular to the PM length line at the PM basis. In end-diastole, the distance between the tips of the PM (Figure 2D), and the thickness of the LV wall at the level of the PM insertions were also measured (Figure 2, dα and dβ). 18F-fluorodeoxyglucose positron emission tomography Animals were scanned for 30 min on a Biograph 16 HireZ PET/CT (Siemens) scanner, 30 min after intravenous injection of 370 MBq (10 mCi) FDG. To reduce myocardial fatty acid metabolism and to stimulate insulin-dependent glucose uptake, plasma glucose levels were titrated using a hyper-insulinemic euglycaemic clamping method as described previously.21 A low dose CT scan for subsequent attenuation correction was done prior to every PET emission scan. Motion and partial volume correction Classical clinical PET scans have a rather low spatial resolution. In order to improve this, PET images were reconstructed using resolution modelling on both ECG and respiratory gated nuclear raw data.16,17 By means of resolution modelling, ECG-triggering and respiratory gating, they can be motion- and partial volume corrected, drastically improving their spatial resolution. A detailed description of the PET motion and partial volume correction procedure can be found in the Supplementary data online, Motion and partial volume correction and PET image resolution. Data analysis Regional FDG uptake was evaluated offline on static images reconstructed from the 30 min scan time using an in-house developed program.16,17 Data were analysed in end-diastole and end-expiration on two long-axis slices of the left ventricle, each parallel and in-line with one of the PM (Figure 4, left panels). On each slice, two regions of interest (ROI) were manually delineated: (i) a region including the PM and the part of the myocardial wall directly perpendicularly adjacent to the base of it (PM); (ii) the remaining MYO. The mean FDG tracer concentration within each ROI was determined and normalized to the ROI area. Finally for both slices, the ratio of the tracer concentrations in the PM and myocardium was calculated (PM/MYO ratio). Clinical proof-of-concept For initial comparison to the animal findings, two patients were investigated, one with a dilated and one with normally-sized left ventricle. Both patients underwent the same FDG PET/CT imaging protocol and data analysis as outlined above. Myocardial scarring and major coronary artery disease was ruled out by medical history and magnetic resonance imaging including late gadolinium enhancement. The study was approved by the ethical commission of the University Hospitals Leuven and subjects gave written informed consent prior to inclusion and any study procedure. Statistical analysis Analysis was performed using SPSS Statistics 20 (IBM, Chicago, IL, USA). All continuous variables are expressed as mean ± standard deviation if normally distributed, or otherwise by median ± interquartile range. Normality was assessed with the use of the Shapiro–Wilk test. All geometrical measurements were analysed in three-fold by a single experienced observer, and average values were used and reported afterwards. Unpaired samples t-tests were used to compare mean differences between the normal and dilated study group. Differences within the PM were examined by using the paired samples t-test. Univariate linear regression analysis with Pearson’s correlation coefficients was used to correlate variables of left ventricle and subvalvular apparatus function and geometry with the PM/MYO ratio. Subsequently, contributors with a significant correlation in the univariate analysis were entered in a multiple linear regression model, using the backwards method, and after evaluation of possible collinearity. 18F-fluorodeoxyglucose activity, LV sphericity index and PM angles were re-assessed by a second experienced observer in order to evaluate the inter-observer agreement using intra-class correlation (ICC). The same variables were also used for intra-observer agreement. Statistical significance was set at a two-tailed probability level of P < 0.05. Results Left ventricular chamber geometry and function The animals subjected to rapid pacing (DILATED group) developed dilated hearts with reduced function compared with controls (NORMAL group) (Table 1, Figures 1 and 2). The DILATED group had significantly higher LV EDV (P = 0.017) and ESV (P = 0.002) and a significantly wider LV chamber (P < 0.0001) while the LV length did not differ significantly (P = 0.815). Consequently, the sphericity index was significantly higher (P < 0.0001). Left ventricular ejection fraction (P < 0.0001) and GLS (P < 0.0001) were significantly decreased. The wall thickness at the PM insertions remained unaltered in dilated hearts, when compared with normal hearts (P = 0.902 for the anterior wall and P = 0.413 for the posterior wall, respectively) (Table 1 and Figure 2). Table 1 Left ventricular geometry NORMAL DILATED P-value Heart rate (bpm) 110 110 N.S. LV EDV (mL) 87 ± 12 113 ± 20 0.017 LV ESV (mL) 29 ± 4 69 ± 23 0.002 LVEF (%) 66 ± 3 39 ± 4 <0.0001 GLS (%) −14.7 ± 0.6 −8.7 ± 2 <0.0001 Sphericity index 0.48 ± 0.03 0.61 ± 0.04 <0.0001 Width (cm) 3.59 ± 0.14 4.66 ± 0.39 <0.0001 Length (cm) 7.56 ± 0.44 7.62 ± 0.46 0.815 Thickness anterior wall (cm) 0.78 ± 0.07 0.784 ± 0.038 0.902 Thickness posterior wall (cm) 0.783 ± 0.081 0.817 ± 0.058 0.413 NORMAL DILATED P-value Heart rate (bpm) 110 110 N.S. LV EDV (mL) 87 ± 12 113 ± 20 0.017 LV ESV (mL) 29 ± 4 69 ± 23 0.002 LVEF (%) 66 ± 3 39 ± 4 <0.0001 GLS (%) −14.7 ± 0.6 −8.7 ± 2 <0.0001 Sphericity index 0.48 ± 0.03 0.61 ± 0.04 <0.0001 Width (cm) 3.59 ± 0.14 4.66 ± 0.39 <0.0001 Length (cm) 7.56 ± 0.44 7.62 ± 0.46 0.815 Thickness anterior wall (cm) 0.78 ± 0.07 0.784 ± 0.038 0.902 Thickness posterior wall (cm) 0.783 ± 0.081 0.817 ± 0.058 0.413 EDV, end-diastolic volume; ESV, end-systolic volume; GLS, global longitudinal strain; LV, left ventricle; LVEF, left ventricular ejection fraction; N.S., not significant. Table 1 Left ventricular geometry NORMAL DILATED P-value Heart rate (bpm) 110 110 N.S. LV EDV (mL) 87 ± 12 113 ± 20 0.017 LV ESV (mL) 29 ± 4 69 ± 23 0.002 LVEF (%) 66 ± 3 39 ± 4 <0.0001 GLS (%) −14.7 ± 0.6 −8.7 ± 2 <0.0001 Sphericity index 0.48 ± 0.03 0.61 ± 0.04 <0.0001 Width (cm) 3.59 ± 0.14 4.66 ± 0.39 <0.0001 Length (cm) 7.56 ± 0.44 7.62 ± 0.46 0.815 Thickness anterior wall (cm) 0.78 ± 0.07 0.784 ± 0.038 0.902 Thickness posterior wall (cm) 0.783 ± 0.081 0.817 ± 0.058 0.413 NORMAL DILATED P-value Heart rate (bpm) 110 110 N.S. LV EDV (mL) 87 ± 12 113 ± 20 0.017 LV ESV (mL) 29 ± 4 69 ± 23 0.002 LVEF (%) 66 ± 3 39 ± 4 <0.0001 GLS (%) −14.7 ± 0.6 −8.7 ± 2 <0.0001 Sphericity index 0.48 ± 0.03 0.61 ± 0.04 <0.0001 Width (cm) 3.59 ± 0.14 4.66 ± 0.39 <0.0001 Length (cm) 7.56 ± 0.44 7.62 ± 0.46 0.815 Thickness anterior wall (cm) 0.78 ± 0.07 0.784 ± 0.038 0.902 Thickness posterior wall (cm) 0.783 ± 0.081 0.817 ± 0.058 0.413 EDV, end-diastolic volume; ESV, end-systolic volume; GLS, global longitudinal strain; LV, left ventricle; LVEF, left ventricular ejection fraction; N.S., not significant. Mitral valve subvalvular apparatus Details may be found in Table 2 and Figure 2. In both, the NORMAL and DILATED group, the anterior PMs were more obliquely attached to the LV wall than the posterior PMs (both P = 0.001). In dilated hearts, however, the angle between the posterior PM and the LV wall was significantly larger (P = 0.001) compared with normal hearts. The angle of the anterior PM was similar in both groups (P = 0.68). In dilated hearts, the distance between the PM tips was higher compared with the normal hearts (P = 0.021). Both the anterior and posterior PM were significantly longer in the DILATED group, compared with NORMAL group (Table 2). In addition, the PM in the DILATED group showed significantly lower fractional shortening (anterior: P = 0.001; posterior: P < 0.0001). The fractional shortening of the anterior and posterior PM correlated significantly with GLS (r = 0.74, P = 0.001 and r = 0.85, P < 0.0001, respectively) (Figure 3). Thickness measurements revealed that in the dilated LV, both the anterior PM (P = 0.005) and posterior PM (P = 0.317) were thinner than in the normal hearts, albeit only significant for the anterior PM (Table 2). Table 2 Papillary muscle geometry and function NORMAL DILATED P-value Anterior PM  Angle, ED (°) 41 ± 5 42 ± 6 0.68  Length, ED (cm) 1.55 ± 0.18 1.77 ± 0.11 0.031  Length, ES (cm) 1.13 ± 0.13 1.50 ± 0.14 0.001  PM fractional shortening (%) 29.04 ± 9.18 13.73 ± 5.72 0.001  Width, ED (cm) 0.97 ± 0.04 0.79 ± 0.11 0.005 Posterior PM  Angle, ED (°) 21 ± 1 32 ± 6 0.0001  Length, ED (cm) 1.47 ± 0.25 1.82 ± 0.17 0.033  Length, ES (cm) 1.18 ± 0.19 1.66 ± 0.18 0.002  PM fractional shortening (%) 19.99 ± 2.22 6.36 ± 4.93 <0.0001  Width, ED (cm) 0.85 ± 0.03 0.78 ± 0.15 0.317 PM tips distance (cm) 2.92 ± 0.34 3.60 ± 0.43 0.021 NORMAL DILATED P-value Anterior PM  Angle, ED (°) 41 ± 5 42 ± 6 0.68  Length, ED (cm) 1.55 ± 0.18 1.77 ± 0.11 0.031  Length, ES (cm) 1.13 ± 0.13 1.50 ± 0.14 0.001  PM fractional shortening (%) 29.04 ± 9.18 13.73 ± 5.72 0.001  Width, ED (cm) 0.97 ± 0.04 0.79 ± 0.11 0.005 Posterior PM  Angle, ED (°) 21 ± 1 32 ± 6 0.0001  Length, ED (cm) 1.47 ± 0.25 1.82 ± 0.17 0.033  Length, ES (cm) 1.18 ± 0.19 1.66 ± 0.18 0.002  PM fractional shortening (%) 19.99 ± 2.22 6.36 ± 4.93 <0.0001  Width, ED (cm) 0.85 ± 0.03 0.78 ± 0.15 0.317 PM tips distance (cm) 2.92 ± 0.34 3.60 ± 0.43 0.021 ED, end-diastole; ES, end-systole; PM, papillary muscle. Table 2 Papillary muscle geometry and function NORMAL DILATED P-value Anterior PM  Angle, ED (°) 41 ± 5 42 ± 6 0.68  Length, ED (cm) 1.55 ± 0.18 1.77 ± 0.11 0.031  Length, ES (cm) 1.13 ± 0.13 1.50 ± 0.14 0.001  PM fractional shortening (%) 29.04 ± 9.18 13.73 ± 5.72 0.001  Width, ED (cm) 0.97 ± 0.04 0.79 ± 0.11 0.005 Posterior PM  Angle, ED (°) 21 ± 1 32 ± 6 0.0001  Length, ED (cm) 1.47 ± 0.25 1.82 ± 0.17 0.033  Length, ES (cm) 1.18 ± 0.19 1.66 ± 0.18 0.002  PM fractional shortening (%) 19.99 ± 2.22 6.36 ± 4.93 <0.0001  Width, ED (cm) 0.85 ± 0.03 0.78 ± 0.15 0.317 PM tips distance (cm) 2.92 ± 0.34 3.60 ± 0.43 0.021 NORMAL DILATED P-value Anterior PM  Angle, ED (°) 41 ± 5 42 ± 6 0.68  Length, ED (cm) 1.55 ± 0.18 1.77 ± 0.11 0.031  Length, ES (cm) 1.13 ± 0.13 1.50 ± 0.14 0.001  PM fractional shortening (%) 29.04 ± 9.18 13.73 ± 5.72 0.001  Width, ED (cm) 0.97 ± 0.04 0.79 ± 0.11 0.005 Posterior PM  Angle, ED (°) 21 ± 1 32 ± 6 0.0001  Length, ED (cm) 1.47 ± 0.25 1.82 ± 0.17 0.033  Length, ES (cm) 1.18 ± 0.19 1.66 ± 0.18 0.002  PM fractional shortening (%) 19.99 ± 2.22 6.36 ± 4.93 <0.0001  Width, ED (cm) 0.85 ± 0.03 0.78 ± 0.15 0.317 PM tips distance (cm) 2.92 ± 0.34 3.60 ± 0.43 0.021 ED, end-diastole; ES, end-systole; PM, papillary muscle. Figure 3 View largeDownload slide Correlation plots of papillary muscle (PM) fractional shortening and global longitudinal strain (GLS). (A): anterior PM. (B): posterior PM. Blue and red markers represent respectively NORMAL and DILATED study groups. Figure 3 View largeDownload slide Correlation plots of papillary muscle (PM) fractional shortening and global longitudinal strain (GLS). (A): anterior PM. (B): posterior PM. Blue and red markers represent respectively NORMAL and DILATED study groups. Papillary muscle work In normal hearts, the FDG activity in both PM regions was slightly higher compared with the remaining myocardium, without differences between anterior and posterior PM (PM/MYO ratio 1.20 ± 0.05 and 1.16 ± 0.06; P = 0.419) (Figure 4, bar chart). In dilated hearts, a reproducible pattern of increased FDG activity at the position of the PMs was observed (Figure 4B and D) with a significantly higher PM/MYO ratio for both PM compared to normal hearts (anterior: 1.47 ± 0.06 vs. 1.20 ± 0.05; P < 0.0001 and posterior: 1.31 ± 0.07 vs. 1.16 ± 0.06; P = 0.002, respectively). The pattern was significantly more pronounced in the anterior PM (P < 0.0001) (Figure 4, bar chart). Figure 4 View largeDownload slide Assessment of papillary muscles work. In two long-axis slices, through each papillary muscle (PM), the 18F-fluorodeoxyglucose (FDG) tracer activity was calculated. Representative views of the anterior (A + B) and posterior PM (C + D) of the NORMAL (A + C) and DILATED (B + D) study group are shown. Note the ‘hotspots’ of increased FDG activity in the PM on the images of the DILATED study group. Activity in PM [region of interest (ROI) with yellow contour] and the remaining LV myocardium (ROI with white contour) was calculated and normalized to the respective ROI area. Contribution of PM to myocardial work is shown as PM/MYO ratio in the bar chart. Asterisks indicate P < 0.05. Figure 4 View largeDownload slide Assessment of papillary muscles work. In two long-axis slices, through each papillary muscle (PM), the 18F-fluorodeoxyglucose (FDG) tracer activity was calculated. Representative views of the anterior (A + B) and posterior PM (C + D) of the NORMAL (A + C) and DILATED (B + D) study group are shown. Note the ‘hotspots’ of increased FDG activity in the PM on the images of the DILATED study group. Activity in PM [region of interest (ROI) with yellow contour] and the remaining LV myocardium (ROI with white contour) was calculated and normalized to the respective ROI area. Contribution of PM to myocardial work is shown as PM/MYO ratio in the bar chart. Asterisks indicate P < 0.05. Determinants of papillary muscle work In linear regression analyses combining normal and dilated hearts, sphericity index, posterior PM angle, LVEF, LV EDV, GLS, distance between PM tips, and fractional shortening of the posterior PM were significantly correlated with the PM/MYO ratio of the posterior PM, which was not the case for the posterior PM width (Table 3). Multivariable linear regression analysis revealed that the sphericity index and posterior PM angle were the strongest determinants of the PM/MYO ratio of the posterior PM (R2 = 0.754, P < 0.0001) (Figure 5, lower panels for the univariable regression plots). Table 3 Determinants of papillary muscle contribution to left ventricular workload (PM/MYO) Univariate Multivariate r P-value R2 P-value Anterior PM  Sphericity index (*) 0.882 <0.0001 0.805 <0.0001  PM width (*) 0.747 0.001  Anterior PM angle 0.360 0.188  LVEF 0.816 <0.0001  LV EDV (*) 0.575 0.025  GLS 0.848 <0.0001  PM tips width 0.474 0.102  PM fractional shortening (*) 0.679 0.005 Posterior PM  Sphericity index (*) 0.805 <0.0001 0.754 <0.0001  Posterior PM angle (*) 0.856 <0.0001  PM width 0.380 0.163  LVEF 0.687 0.005  LV EDV (*) 0.686 0.005  GLS (*) 0.717 0.003  PM tips width 0.622 0.023  PM fractional shortening (*) 0.746 0.001 Univariate Multivariate r P-value R2 P-value Anterior PM  Sphericity index (*) 0.882 <0.0001 0.805 <0.0001  PM width (*) 0.747 0.001  Anterior PM angle 0.360 0.188  LVEF 0.816 <0.0001  LV EDV (*) 0.575 0.025  GLS 0.848 <0.0001  PM tips width 0.474 0.102  PM fractional shortening (*) 0.679 0.005 Posterior PM  Sphericity index (*) 0.805 <0.0001 0.754 <0.0001  Posterior PM angle (*) 0.856 <0.0001  PM width 0.380 0.163  LVEF 0.687 0.005  LV EDV (*) 0.686 0.005  GLS (*) 0.717 0.003  PM tips width 0.622 0.023  PM fractional shortening (*) 0.746 0.001 Combined analysis of NORMAL and DILATED hearts. Asterisk (*) denotes variables that were entered in the multivariate analysis, after being checked for significance in univariate analysis and rendering no collinearity issues. EDV, end-diastolic volume; GLS, global longitudinal strain; LV, left ventricle; LVEF, left ventricular ejection fraction; PM, papillary muscle. Table 3 Determinants of papillary muscle contribution to left ventricular workload (PM/MYO) Univariate Multivariate r P-value R2 P-value Anterior PM  Sphericity index (*) 0.882 <0.0001 0.805 <0.0001  PM width (*) 0.747 0.001  Anterior PM angle 0.360 0.188  LVEF 0.816 <0.0001  LV EDV (*) 0.575 0.025  GLS 0.848 <0.0001  PM tips width 0.474 0.102  PM fractional shortening (*) 0.679 0.005 Posterior PM  Sphericity index (*) 0.805 <0.0001 0.754 <0.0001  Posterior PM angle (*) 0.856 <0.0001  PM width 0.380 0.163  LVEF 0.687 0.005  LV EDV (*) 0.686 0.005  GLS (*) 0.717 0.003  PM tips width 0.622 0.023  PM fractional shortening (*) 0.746 0.001 Univariate Multivariate r P-value R2 P-value Anterior PM  Sphericity index (*) 0.882 <0.0001 0.805 <0.0001  PM width (*) 0.747 0.001  Anterior PM angle 0.360 0.188  LVEF 0.816 <0.0001  LV EDV (*) 0.575 0.025  GLS 0.848 <0.0001  PM tips width 0.474 0.102  PM fractional shortening (*) 0.679 0.005 Posterior PM  Sphericity index (*) 0.805 <0.0001 0.754 <0.0001  Posterior PM angle (*) 0.856 <0.0001  PM width 0.380 0.163  LVEF 0.687 0.005  LV EDV (*) 0.686 0.005  GLS (*) 0.717 0.003  PM tips width 0.622 0.023  PM fractional shortening (*) 0.746 0.001 Combined analysis of NORMAL and DILATED hearts. Asterisk (*) denotes variables that were entered in the multivariate analysis, after being checked for significance in univariate analysis and rendering no collinearity issues. EDV, end-diastolic volume; GLS, global longitudinal strain; LV, left ventricle; LVEF, left ventricular ejection fraction; PM, papillary muscle. Figure 5 View largeDownload slide Correlation plots of PM/MYO ratio versus its determinants. The black dotted regression lines refer to the combined analysis of NORMAL and DILATED hearts (results in the graph in black). The red dotted regression line and the red numbers refer to a separate analysis of DILATED hearts only. Blue and red markers represent respectively NORMAL and DILATED study groups. Figure 5 View largeDownload slide Correlation plots of PM/MYO ratio versus its determinants. The black dotted regression lines refer to the combined analysis of NORMAL and DILATED hearts (results in the graph in black). The red dotted regression line and the red numbers refer to a separate analysis of DILATED hearts only. Blue and red markers represent respectively NORMAL and DILATED study groups. The PM/MYO ratio of the anterior PM showed significant correlation with the same parameters, including the anterior PM width, except for the anterior PM angle (Table 3). However, a significant correlation was still obtained between the PM/MYO ratio and anterior PM angle, when considering only the dilated hearts (Figure 5B). After multivariable linear regression analysis, the sphericity index and anterior PM width remained as strongest determinant of the PM/MYO of the anterior PM (R2 = 0.805, P < 0.0001) (Figure 5, upper panels for the univariable regression plots). Reproducibility Analysis of the intra-observer variability of the LV sphericity index and PM angles showed good reproducibility: [ICC 0.852, 95% confidence interval (CI) 0.772–0.946] and (ICC 0.966, 95% CI 0.936–0.984), respectively. Inter-observer variability also showed good ICC for FDG activity (ICC 0.985, 95% CI 0.866–0.993), LV sphericity index (ICC 0.86, 95% CI 0.722–0.946), and PM angles (ICC 0.945, 95% CI 0.882–0.975). Discussion Main findings of the study In this study, we evaluated the contribution of PM to global LV function in normal and dilated hearts of animals, using myocardial glucose uptake assessed with FDG PET as surrogate of myocardial work.14,15 We found that in dilated hearts: (i) PM have a significantly increased relative contribution to LV work and (ii) this increased contribution is significantly related to geometric changes of the dilated LV. Role of the papillary muscles in normal and dilated hearts In normal hearts, the PM/MYO ratio was slightly higher than one, which allows the interpretation that more work per unit volume is performed by the PM compared with the rest of the MYO. An alternative explanation of our observation of an increased tracer activity in the PM region could be a partial volume effect, since a larger volume of active myocardium is present in the region where the PM comes close to the ventricular wall. Although our efforts of resolution modelling with ECG-triggering and respiratory gating resulted in a dramatic increase of spatial resolution compared with clinical PET reconstructions, this explanation is not unlikely, especially when considering that in normal hearts the direction of force development in PM is more parallel to the LV wall, so that (similar loading assumed) equal work can be expected. In dilated hearts, our findings indicate that there is a significantly increased contribution of the PM to the LV myocardial work. This increased work cannot be explained by an augmented systolic shortening of the PM compared to the rest of the left ventricle. On the contrary, the length of the PM was increased as was the LV size, while the fractional shortening was proportionally reduced compared with the GLS of the left ventricle (Figure 3). Consequently, the higher work of the PM compared with the rest of the left ventricle myocardium must be due to a higher stress caused by the altered orientation of the PM and subvalvular apparatus in the dilated and more spherical hearts, which becomes more cross-sectional to the left ventricle and less parallel with its walls. Furthermore, we found the anterior PMs in dilated hearts to be thinner than in normal hearts, resulting in smaller cross-sectional area and—with this—higher stress. We therefore hypothesize, that the higher work of the PM compared to the rest of the MYO is truly caused by a higher stress. This hypothesis is supported by earlier experimental studies, which have demonstrated that the PM of failing or dilated hearts can generate up to 30% more force as in the normal heart.22,23 Also in dilated hearts, partial volume effects must be considered as potential explanation of our observations. In this case, however, this explanation seems unlikely, since the PM/MYO ratio was much higher. In addition, PM were further away from the LV wall so that they could be clearly distinguished given the augmented resolution of our dual gated PET reconstruction. Finally, LV wall thickness measurements were not significantly different in normal and dilated hearts, so that the partial volume effect in the wall should be comparable in both groups. Determinants of increased papillary muscle work Left ventricular dilatation leads to a more spherical configuration of the LV chamber. Our measurements indicate that this is accompanied by a substantial change in the configuration of the subvalvular apparatus. The PM were displaced more outward, as indicated by the increased distance between their tips. This observation is concordant with previous experimental12,13 and patient studies.9,10 Our data indicate further, that the PM act also no longer parallel, but more angulated to the LV wall. As the angle between PM and the LV wall increases, PM and chordae run more cross-sectional through the LV chamber, which exposes the PM to a higher stress than the curved wall of the spherical ventricle is exposed to.24,25 This hypothesis is supported by our differential findings in the anterior and posterior PM of the dilated hearts. Both the anterior and posterior PM contributed significantly more to the LV myocardial work in the dilated hearts, with the highest contribution delivered by the anterior PM, which had also the larger angle towards the wall (Figure 5). Although on average, the angle of the anterior PM did not differ significantly between normal and dilated hearts, we also found a significant correlation between angle and work contribution in the dilated hearts (Figure 5B, red dotted line). The particular curved shape of the ovine LV chamber26 may have contributed to the differential observations in both PM. While the posterior wall is quite straight, the anterior wall is already relatively curved in normal hearts and this difference remains with LV dilatation.26 Furthermore, our data indicate that in dilated hearts particularly the anterior PM is thinner than normal. Consequently, its width appeared as one of the strongest determinants of the PM/MYO ratio for the anterior PM, which might in part also explain its increased metabolism. Clinical proof-of-concept The FDG PET measurements in our proof-of-concept-patients led to very comparable results as in the animals (Figure 6). In the dilated heart, the contribution of the anterior PM to the LV work was increased compared to the patient with the non-dilated heart. Similarly, also the angle of this PM was clearly larger in the dilated heart than in the non-dilated heart, together with an apparent spherical shape. We therefore assume, that our findings can be applied to the situation in the human heart. Figure 6 View largeDownload slide Clinical validation: assessment of papillary muscles work. Arrangement of panels and colour scheme are similar as in Figure 4. Interestingly, the findings in this proof-of-concept analysis in two patients are comparable to the results of our animal experiments. Figure 6 View largeDownload slide Clinical validation: assessment of papillary muscles work. Arrangement of panels and colour scheme are similar as in Figure 4. Interestingly, the findings in this proof-of-concept analysis in two patients are comparable to the results of our animal experiments. Clinical relevance Observational studies, investigating the effect of mitral valve replacement with and without preservation of the mitral subvalvular apparatus on LV performance, have shown in both animal experiments5 and patients,8 that preserving the subvalvular apparatus results in better post-operative LV shape, ejection fraction and contractility. While the results of these studies have led to the recommendation in surgical guidelines to preserve the subvalvular apparatus,6 the pathophysiologic understanding of the importance of PM function was lacking. Our study results indicate that the contribution of PM to LV performance is relevant, and suggest that their importance is even higher in dilated (e.g. volume loaded) hearts. Therefore, our findings provide evidence for the—so far—empiric practice of preserving the PM during surgical mitral valve replacement. Study limitations The sample size used in this study was small, which is inherent to fundamental animal research. However, the magnitude of observed changes was comparable to other studies in the field and effects were highly significant. Nevertheless, further validation should be performed in a larger cohort. The 11C-acetate tracer (complete oxidative metabolism) would be a valuable alternative to the FDG tracer (glucose metabolism only) used in this study, providing a more general information on the overall myocardial metabolism. However, 11C-acetate has a very short half-life, necessitating an adjacent cyclotron and requires kinetic modelling. Therefore, we used a standard glucose-insuline-clamping technique with FDG to ensure a predominant glucose metabolism in the myocardium. Although sheep are an established animal model, with translational findings to humans,27 particularly the geometry of the normal and dilated chamber may not completely mimic the situation in human pathology. However, in both sheep and humans, a higher work contribution of PM was found in dilated hearts, particularly high in the PM with the largest angle towards the wall. Nevertheless, our proof-of-concept in a small sample of human subjects need further validation in a larger clinical study. Conclusions Our study demonstrated that in dilated hearts, the PMs contribute relatively more to LV work, compared with normal hearts. As a consequence of the LV dilatation, the geometry of the subvalvular apparatus is altered, making the PMs to be thinner and to work more across the LV and less parallel to the LV wall, which results in a higher stress. Our findings emphasise the important contribution of PMs to LV systolic function, and the understanding that they should be preserved, e.g. during mitral repair/replacement. Supplementary data Supplementary data are available at European Heart Journal—Cardiovascular Imaging online. Acknowledgements The authors are indebted to David Célis for his excellent technical assistance during the experiments and care of the animals, and wish to thank Hermien Coppens and Marta Cvijic for assisting with data analysis and preparation of the study. Funding J.D. received the ‘Young Investigator Award—Basic Science’ from the European Association of Cardiovascular Imaging (EACVI) during EuroEcho-Imaging 2016. This work was supported by the KU Leuven [research grant OT/12/084]. The Research Foundation—Flanders (FWO) supported A.T. as a doctoral fellow, K.V. as a post-doctoral fellow, and O.G. and J.-U.V. as senior clinical investigators. A.S.B. received a research grant from the Egyptian Ministry of Higher Education. E.D.P. and S.U. received a research grant from the European Association of Cardiovascular Imaging (EACVI). Conflict of interest: None declared. References 1 Jones CJ , Raposo L , Gibson DG. Functional importance of the long axis dynamics of the human left ventricle . Br Heart J 1990 ; 63 : 215 – 20 . Google Scholar CrossRef Search ADS PubMed 2 Kono T , Sabbah HN , Rosman H , Alam M , Jafri S , Goldstein S. Left ventricular shape is the primary determinant of functional mitral regurgitation in heart failure . J Am Coll Cardiol 1992 ; 20 : 1594 – 8 . Google Scholar CrossRef Search ADS PubMed 3 Sarris GE , Fann JI , Niczyporuk MA , Derby GC , Handen CE , Miller DC. Global and regional left ventricular systolic performance in the in situ ejecting canine heart. Importance of the mitral apparatus . Circulation 1989 ; 80 : I24 – 42 . Google Scholar CrossRef Search ADS PubMed 4 Hansen DE , Cahill PD , DeCampli WM , Harrison DC , Derby GC , Mitchell RS. Valvular-ventricular interaction: importance of the mitral apparatus in canine left ventricular systolic performance . Circulation 1986 ; 73 : 1310 – 20 . Google Scholar CrossRef Search ADS PubMed 5 Yun KL , Niczyporuk MA , Sarris GE , Fann JI , Miller DC. Importance of mitral subvalvular apparatus in terms of cardiac energetics and systolic mechanics in the ejecting canine heart . J Clin Invest 1991 ; 87 : 247 – 54 . Google Scholar CrossRef Search ADS PubMed 6 Vahanian A , Alfieri O , Andreotti F , Antunes MJ , Barón-Esquivias G , Baumgartner H et al. Guidelines on the management of valvular heart disease (version 2012): the Joint Task Force on the Management of Valvular Heart Disease of the European Society of Cardiology (ESC) and the European Association for Cardio-Thoracic Surgery (EACTS) . Eur Heart J 2012 ; 33 : 2451 – 96 . Google Scholar CrossRef Search ADS PubMed 7 Pitarys CJ , Forman MB , Panayiotou H , Hansen DE. Long-term effects of excision of the mitral apparatus on global and regional ventricular function in humans . J Am Coll Cardiol 1990 ; 15 : 557 – 63 . Google Scholar CrossRef Search ADS PubMed 8 Rozich JD , Carabello BA , Usher BW , Kratz JM , Bell AE , Zile MR. Mitral valve replacement with and without chordal preservation in patients with chronic mitral regurgitation. Mechanisms for differences in postoperative ejection performance . Circulation 1992 ; 86 : 1718 – 26 . Google Scholar CrossRef Search ADS PubMed 9 Delgado V , Tops LF , Schuijf JD , de Roos A , Brugada J , Schalij MJ et al. Assessment of mitral valve anatomy and geometry with multislice computed tomography . JACC Cardiovasc Imaging 2009 ; 2 : 556 – 65 . Google Scholar CrossRef Search ADS PubMed 10 Madu EC , Baugh DS , D’Cruz IA , Johns C. Left ventricular papillary muscle morphology and function in left ventricular hypertrophy and left ventricular dysfunction . Med Sci Monit 2001 ; 7 : 1212 – 8 . Google Scholar PubMed 11 Kwan J , Shiota T , Agler DA , Popović ZB , Qin JX , Gillinov MA et al. Geometric differences of the mitral apparatus between ischemic and dilated cardiomyopathy with significant mitral regurgitation: real-time three-dimensional echocardiography study . Circulation 2003 ; 107 : 1135 – 40 . Google Scholar CrossRef Search ADS PubMed 12 Madu EC , D’Cruz IA. The vital role of papillary muscles in mitral and ventricular function: echocardiographic insights . Clin Cardiol 1997 ; 20 : 93 – 8 . Google Scholar CrossRef Search ADS PubMed 13 Timek TA , Dagum P , Lai DT , Liang D , Daughters GT , Tibayan F et al. Tachycardia-induced cardiomyopathy in the ovine heart: mitral annular dynamic three-dimensional geometry . J Thorac Cardiovasc Surg 2003 ; 125 : 315 – 24 . Google Scholar CrossRef Search ADS PubMed 14 Bergman BC , Tsvetkova T , Lowes B , Wolfel EE. Myocardial glucose and lactate metabolism during rest and atrial pacing in humans . J Physiol 2009 ; 587 : 2087 – 99 . Google Scholar CrossRef Search ADS PubMed 15 Russell K , Eriksen M , Aaberge L , Wilhelmsen N , Skulstad H , Remme EW et al. A novel clinical method for quantification of regional left ventricular pressure-strain loop area: a non-invasive index of myocardial work . Eur Heart J 2012 ; 33 : 724 – 33 . Google Scholar CrossRef Search ADS PubMed 16 Turco A , Gheysens O , Nuyts J , Duchenne J , Voigt JU , Claus P et al. Impact of CT-based attenuation correction on the registration between dual-gated cardiac PET and high-resolution CT . IEEE Trans Nucl Sci 2016 ; 63 : 180 – 92 . Google Scholar CrossRef Search ADS 17 Turco A. Partial volume and motion correction in cardiac PET: first results from and in vs ex vivo comparison using animal datasets. In: Turo A, ed. Improved Quantification of Myocardial FDG Uptake combining PET/CT and 4D Anatomical Imaging. Ph.D. Thesis. Leuven: KU Leuven—University of Leuven; 2016 . p. 131 – 46 . 18 Duchenne J , Claus P , Pagouralias ED , Mada RO , Van Puyvelde J , Vunckx K et al. Sheep can be used as animal model of regional myocardial remodelling and controllable work . Cardiol J 2018 [Epub ahead of print], doi:10.5603/CJ.a2018.0007. 19 Agricola E , Oppizzi M , Maisano F , De Bonis M , Schinkel AFL , Torracca L et al. Echocardiographic classification of chronic ischemic mitral regurgitation caused by restricted motion according to tethering pattern . Eur J Echocardiogr 2004 ; 5 : 326 – 34 . Google Scholar CrossRef Search ADS PubMed 20 Magne J , Pibarot P , Dagenais F , Hachicha Z , Dumesnil JG , Sénéchal M. Preoperative posterior leaflet angle accurately predicts outcome after restrictive mitral valve annuloplasty for ischemic mitral regurgitation . Circulation 2007 ; 115 : 782 – 91 . Google Scholar CrossRef Search ADS PubMed 21 Kim JK. Hyperinsulinemic–euglycemic clamp to assess insulin sensitivity in vivo . Methods Mol Biol 2009 ; 560 : 221 – 38 . Google Scholar CrossRef Search ADS PubMed 22 Burch GE , Giles TD. Angle of traction of the papillary muscle in normal and dilated hearts: a theoretic analysis of its importance in mitral valve dynamics . Am Heart J 1972 ; 84 : 141 – 4 . Google Scholar CrossRef Search ADS PubMed 23 Feldman MD , Copelas L , Gwathmey JK , Phillips P , Warren SE , Schoen FJ et al. Deficient production of cyclic AMP: pharmacologic evidence of an important cause of contractile dysfunction in patients with end-stage heart failure . Circulation 1987 ; 75 : 331 – 9 . Google Scholar CrossRef Search ADS PubMed 24 Perloff JK , Roberts WC , Perloff JK , Roberts WC. The mitral apparatus: functional anatomy of mitral regurgitation . Circulation 1972 ; 46 : 227 – 39 . Google Scholar CrossRef Search ADS PubMed 25 Kono T , Sabbah HN , Stein PD , Brymer JF , Khaja F. Left ventricular shape as a determinant of functional mitral regurgitation in patients with severe heart failure secondary to either coronary artery disease or idiopathic dilated cardiomyopathy . Am J Cardiol 1991 ; 68 : 355 – 9 . Google Scholar CrossRef Search ADS PubMed 26 Hill AJ , Iaizzo PA. Comparative cardiac anatomy. In: Iaizzo PA , ed. Handbook of Cardiac Anatomy, Physiology, and Devices . 2nd ed . New York : Springer ; 2009 . p 102 . 27 Houser SR , Margulies KB , Murphy AM , Spinale FG , Francis GS , Prabhu SD et al. Animal models of heart failure: a scientific statement from the American Heart Association . Circ Res 2012 ; 111 : 131 – 50 . Google Scholar CrossRef Search ADS PubMed Published on behalf of the European Society of Cardiology. All rights reserved. © The Author(s) 2018. For permissions, please email: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

European Heart Journal – Cardiovascular ImagingOxford University Press

Published: Mar 23, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off