A computed tomography-based planning tool for predicting difficulty of minimally invasive aortic valve replacement

A computed tomography-based planning tool for predicting difficulty of minimally invasive aortic... Abstract OBJECTIVES Minimally invasive aortic valve replacement has proven its value over the last decade by its significant advancement and reduction in mortality, morbidity and admission time. However, minimally invasive aortic valve replacement is associated with some on-site difficulties such as limited aortic annulus exposure. Currently, computed tomography scans are used to evaluate the anatomical relationship among the intercostal spaces, ascending aorta and aortic valve prior to surgery. We hypothesized that quantitative measurements of access distance and access angle are associated with outcome and access difficulty. METHODS We introduce a novel minimally invasive aortic valve replacement planning prototype that allows automatic measurements of access angle, access distance and aortic annulus dimensions. The prototype visualizes these measurements on the chest cage as ISO contours. The association of these measures with outcome parameters such as extracorporeal circulation time, aortic cross-clamping time and access difficulty score was assessed. We included 14 patients who received a new valve by ministernotomy. RESULTS The mean access angle was 40.3 ± 5.1°. It was strongly associated with aortic cross-clamping time (Pearson correlation coefficient = 0.60, P = 0.02) and access difficulty score (Spearman’s rank correlation coefficient = 0.57, P = 0.03). Access angles were significantly different between easy and difficult access groups (P = 0.03). There was no significant association between access distance and outcome parameters. CONCLUSIONS Access angle is strongly associated with procedure complexity. The automated presentation of this measure suggests added value of the prototype in clinical practice. Minimally invasive aortic valve replacement, Aortic stenosis, Planning tool, Access difficulty, Computer tomography INTRODUCTION Valve disease is a serious health problem as it carries a poor prognosis, and its prevalence is strongly associated with the population aging [1]. Aortic stenosis is currently the most frequent native valve disease in Europe, which is most often seen in elderly patients [2]. Valve replacement is the conventional therapy for patients with severe aortic stenosis who have symptoms or observable consequences such as left ventricular dysfunction [3]. Traditional minimally invasive aortic valve replacement (mini-AVR) is performed during an open surgical procedure with full sternotomy and with arterial cannulation in the ascendant aorta and venous cannulation in the right atrium to enable extracorporeal circulation by means of a heart–lung machine [4]. Full sternotomy has the advantage of a direct view and access to all the cardiac structures [5]. Operative mortality is quite low, even in elderly patients when properly selected and long-term results are satisfactory [6]. However, risk of surgery is higher in elderly patients with significant comorbidities [4]. During the past decade, surgical techniques improved creating an area for less and minimally invasive procedures. Moreover, a range of procedures classified as mini-AVR has been introduced [7]. When compared with conventional surgery, mini-AVR has shown to reduce postoperative mortality and morbidity. In addition, patients recover faster requiring less rehabilitations resources and shorter admission time [8]. The most common minimally invasive approach for valve replacement is partial upper ministernotomy (MS), followed by the right anterior minithoracotomy (RT) [9]. MS is performed through a vertical incision through skin and sternum with the shape of the character J starting from the sternal angle (manubriosternal joint) moving caudally for 4 cm in the sternum body and completed by a transverse sternal incision [10]. RT is performed through a 5–6-cm skin incision at the level of the 2nd or 3rd intercostal space, starting from the border of the sternum toward the lateral right side [11]. Few studies compared the outcomes of the approaches in terms of quality of life and surgery duration. Only marginal benefits for MS have been reported when compared with RT [12, 13]. The MS approach is generally the preferred approach by surgeons [14, 15]. However, the right anterior minithoracotomy (RT) has been reported to be associated with improved outcomes [12, 16, 17], despite longer duration of surgery when compared with MS [17–19]. The exposure of the aortic root right cusp annulus side is limited for the MS approach. For both techniques, incision location selection is of utmost importance for successful valve replacement [20, 21]. Preoperative planning of patients eligible for mini-AVR is performed using computed tomography angiography (CTA). A previous study proposed a visualization system for surgical planning in the workup for mini-AVR using CTA images. This tool renders the chest cage and the ascending aorta to visually support procedure planning and approach selection [22]. Current planning tools, however, do not yet provide quantitative measurements. In another study, Glauber et al. [21] stated that patients are suitable for RT only if the following criteria are met: (i) the distance from the ascending aorta to the sternum should not exceed 10 cm; (ii) the angle between the sternum midline and the inclination of the ascending aorta should be larger than 45°. It has also been suggested that it is important to measure aortic valve dimensions such as the diameter of the aortic valve annulus, the length of the ascending aorta and the calcifications of the aortic valve and the ascending aorta in the preprocedural planning process because it is strongly recommended to remove all the eccentric calcifications and create a complete decalcification of the aortic annulus [21, 23]. We developed a mini-AVR planning tool that combines 3-dimensional (3D) visualization of the access site in addition to quantitative assessment of important planning measures. This tool provides the access angle and distance from the incision location on the chest cage quantitatively and visually using 3D rendering methods. We believe that both access visualization and quantification give an impression of the difficulty of access by measuring how far and aligned the access with the aortic root. In this tool, the mini-AVR planning tool is validated by assessment of the association of the provided quantitative measures with clinical outcome parameters: surgery time, clamping time and annulus access difficulty score. METHODS Planning tool Automated measures The mini-AVR planning tool provides both distance and access angle from potential incision sites to the aortic root based on the geometrical location of the aortic root landmarks and incision locations [24, 25]. These parameters are determined as follows: access distance is defined as the distance between incision location in the chest bone and the aortic sinotubular junction for both the MS and RT approaches. Access angle is defined as the angle between the aortic root centreline axis and incision–annulus line. The aortic root centre axis is the vector that connects annulus and sinotubular junction centres. The incision–annulus centre axis is the vector that connects the incision with the annulus centre (Fig. 1). Other annulus dimensions such as minimum, maximum diameters, area, perimeter and calcium volume are calculated based on a previously presented automated method, which automatically detected aortic root landmarks and calculated sizing parameters [24]. Figure 1: View largeDownload slide Schematic showing the methodology of calculating the access distance and angle based on the location of 3 landmarks: aortic annulus centre, sinotubular junction and incision point. Figure 1: View largeDownload slide Schematic showing the methodology of calculating the access distance and angle based on the location of 3 landmarks: aortic annulus centre, sinotubular junction and incision point. The mini-AVR planning tool has a specific workflow to calculate access difficulty. This workflow consists of a set of interactive steps. The tool can detect the aortic annulus, sinotubular junction and segment the ribcage from the full body computed tomography (CT) scan, then produce the final measurements. Access distance and angle are presented using ISO contour maps (Fig. 2). To calculate the distance and angle maps, a mesh that represents the ribs is reconstructed. An ISO contour connects points of equal measurement values. Each 2 ISO contours define a range of values between them. The ISO contours of distance and angle were rendered on the 3D CT volume using a blue palette and green palette, respectively. The difference between each 2 successive ISO contours of the distance and angle measurements scale was 5 mm and 10°, respectively, as shown in Fig. 2. The tool allows different rendering modes combined with the ISO contours. The rendering modes include full body volume rendering, chest cage with segmented aorta, skin with segmented aorta and the segmented aorta with calcification. Figure 2: View largeDownload slide Screenshot of the developed tool prototype showing the resulted ISO contours in 3-dimensional views (bottom). These contours represent distance and access angle measurements using highlights from the colour palette; the zoomed-in-colour map represents different measurement ranges. Figure 2: View largeDownload slide Screenshot of the developed tool prototype showing the resulted ISO contours in 3-dimensional views (bottom). These contours represent distance and access angle measurements using highlights from the colour palette; the zoomed-in-colour map represents different measurement ranges. Usage The developed prototype works as follows: first, the user loads 3D CTA images. After choosing the mini-AVR workflow from the main User Interface, the software automatically segments the aortic root. Next, the annulus and sinotubular junction are detected. After the automatic detection, extra tools are made available for the user to correct the detected landmarks if needed. The planning prototype contains a 3D probe, which helps the user to interactively select a location on the 3D-rendered chest cage for local measurements such as access distance and angle measurements. Distance and angle measurements are evaluated for all reconstructed chest cage mesh points. Validation Study population We collected preprocedural 3D volume image data of 14 patients who underwent mini-AVR between January and December 2015 including 3 CT and 11 CTA acquisitions from our institute. All CT scans were performed on a Philips Brilliance 64 slice CT scanner (Philips, Cleveland, OH, USA); imaging parameters were 120 kV and convolution kernel B. The chest, abdomen and pelvis were scanned using 1 bolus of 120-ml contrast Iomeron 400 (Bracco Imaging SpA, Milan, Italy) intravenously infused at a rate of 5 ml/s. In case of dynamic scans, which included 10 cardiac phases, we selected images acquired at 70% of the cardiac cycle, corresponding to mid-diastole. This specific phase mimics the non-beating heart and extracorporeal circulation settings during surgery [26] Also, during this cardiac phase, the aortic valve is closed [27]. All image volumes contained approximately 500–800 slices. Each slice in a volume contained 512 × 512 isotropic pixels with a 16-bit depth. The in-plane image resolution varied from 0.44 to 0.68 mm. The slice thickness for all data sets was 0.9 mm, and the overlap between each 2 successive slices was 0.45 mm. All patients included in this study were operated using the MS approach by 1 cardiothoracic surgeon (A.K.). Image-based parameters Annulus measurements such as minimum, maximum diameter, area, perimeter and calcium volume were assessed only in CTA images. To determine the angle and distance measurements at the incision location used during the surgery, the 3D probe was placed at the centre of the manubriosternal joint that was used by the surgeon to mark the incision. Outcome parameters Extracorporeal circulation time, aortic cross-clamping time (AoX) and access difficulty score were collected as outcome parameters. Access difficulty score was subjectively assessed by the operator. It reflects the complexity of the annulus access while performing the stitches to implant the new valve. The score has 4 grades (very easy, easy, moderate and difficult), and we dichotomized this score to easy (very easy and easy) and difficult (moderate and difficult). Assuring independent scoring, rater was blinded to additional information. Statistical analysis Normally distributed and continuous values are expressed as mean ± standard deviation and median and interquartile range otherwise. Significance of associations between outcome parameters and image-based parameters were tested using the Spearman’s rank correlation coefficient. Significant associated parameters were also evaluated using Pearson product–moment correlation coefficient. Aortic annulus measurements were compared between groups of easy and difficult access. The distribution of access angles and AoX for dichotomized access difficulty categories was visualized with box-and-whisker plots. Scatter plots were used to illustrate the association between access angle and access difficulty score. Significant difference of access angle between easy and difficult access groups was determined by paired t-tests. To predict procedure difficulty, we selected an optimal access angle threshold, based on the receiver operating characteristic curve. Statistical analyses were performed using SPSS (version 19.0, SPSS Inc., Chicago, IL, USA) and MATLAB (Version R2013b, The MathWorks Inc., Natick, MA, USA). Associations with P-values of 0.05 or smaller were considered statistically significant. RESULTS Access distance and access angle measurements were normal for both CT and CTA scans for all patients. Aortic valve stenosis was present in all patients included in this study, and bicuspid aortic valve stenosis was present in 2 (14%) patients. Two patients were scanned using CT, and the remainder was scanned using CTA. Four scans were excluded from annulus dimension measurements. For 2 patients, only non-contrast CT was available, which makes annulus measurements less accurate to assess because of poor contrast between blood and aortic root wall. The other 2 scans were excluded because they were scanned after surgery, which hinders annulus measurements due to the strong blooming effect of the stent at annulus level. Access angle and distance measurements using the 3D probe tool were successful for all patients. Average access angle was 40.3 ± 5.1°, and the distance from incision to the sinotubular junction was 94.9 ± 11.1 mm. The remaining measurements are reported in Table 1. Table 1: Average, standard deviation, median and IQR of the access angle, distance from incision, annulus minimum diameter, maximum diameter, area and perimeter   Mean  Standard deviation  Access angle (degrees)  40.3  5.1  Distance from incision (mm)  94.9  11.1  Annulus minimum diameter (mm)  26.2  3.1  Annulus maximum diameter (mm)  32.4  2.9  Annulus area (mm2)  687.1  122.5  Annulus perimeter (mm), median (IQR)  93.6 (88.7–102.2)    Mean  Standard deviation  Access angle (degrees)  40.3  5.1  Distance from incision (mm)  94.9  11.1  Annulus minimum diameter (mm)  26.2  3.1  Annulus maximum diameter (mm)  32.4  2.9  Annulus area (mm2)  687.1  122.5  Annulus perimeter (mm), median (IQR)  93.6 (88.7–102.2)  IQR: interquartile range. Table 1: Average, standard deviation, median and IQR of the access angle, distance from incision, annulus minimum diameter, maximum diameter, area and perimeter   Mean  Standard deviation  Access angle (degrees)  40.3  5.1  Distance from incision (mm)  94.9  11.1  Annulus minimum diameter (mm)  26.2  3.1  Annulus maximum diameter (mm)  32.4  2.9  Annulus area (mm2)  687.1  122.5  Annulus perimeter (mm), median (IQR)  93.6 (88.7–102.2)    Mean  Standard deviation  Access angle (degrees)  40.3  5.1  Distance from incision (mm)  94.9  11.1  Annulus minimum diameter (mm)  26.2  3.1  Annulus maximum diameter (mm)  32.4  2.9  Annulus area (mm2)  687.1  122.5  Annulus perimeter (mm), median (IQR)  93.6 (88.7–102.2)  IQR: interquartile range. The access angle was significantly associated with AoX, 4 graded and 2 graded access difficulty scores with Spearman correlation coefficient of 0.55 (P = 0.042), 0.59 (P = 0.028) and 0.60 (P = 0.025), respectively. The other image-based measurements were not significantly associated with the outcome parameters (see Table 2). Linear correlation between access angle and all outcome parameters was found to be significant. The Pearson product–moment correlation coefficients are reported in Table 3. Table 2: Spearman correlation coefficient and its significance between image-based measurements (access angle, distance from incision, annulus minimum diameter, maximum diameter, area, perimeter and age) and the outcome parameters [ECC, AoX, access difficulty score (4 grades) and access difficulty score (2 grades)] Spearman correlation coefficient (rho)/P-value  ECC   AoX   Access difficulty score (4 grades)   Access difficulty score (2 grades)     rho  P-value  rho  P-value  rho  P-value  rho  P-value  Access angle (degrees)  0.42  0.13  0.55  0.04  0.59  0.03  0.59  0.03  Distance from incision (mm)  −0.04  0.9  −0.07  0.81  −0.49  0.07  −0.39  0.17  Annulus minimum diameter (mm)  −0.22  0.44  −0.45  0.11  −0.25  0.39  −0.19  0.53  Annulus maximum diameter (mm)  −0.29  0.31  −0.39  0.17  −0.14  0.64  0.02  0.95  Annulus area (mm2)  −0.21  0.48  −0.36  0.21  −0.17  0.56  −0.06  0.85  Annulus perimeter (mm)  −0.17  0.57  −0.3  0.3  −0.15  0.61  −0.06  0.85  Age  −0.19  0.51  −0.26  0.37  −0.53  0.05  −0.5  0.07  Spearman correlation coefficient (rho)/P-value  ECC   AoX   Access difficulty score (4 grades)   Access difficulty score (2 grades)     rho  P-value  rho  P-value  rho  P-value  rho  P-value  Access angle (degrees)  0.42  0.13  0.55  0.04  0.59  0.03  0.59  0.03  Distance from incision (mm)  −0.04  0.9  −0.07  0.81  −0.49  0.07  −0.39  0.17  Annulus minimum diameter (mm)  −0.22  0.44  −0.45  0.11  −0.25  0.39  −0.19  0.53  Annulus maximum diameter (mm)  −0.29  0.31  −0.39  0.17  −0.14  0.64  0.02  0.95  Annulus area (mm2)  −0.21  0.48  −0.36  0.21  −0.17  0.56  −0.06  0.85  Annulus perimeter (mm)  −0.17  0.57  −0.3  0.3  −0.15  0.61  −0.06  0.85  Age  −0.19  0.51  −0.26  0.37  −0.53  0.05  −0.5  0.07  Bold values indicate significant correlation coefficients. AoX: aortic cross-clamping time; ECC: extracorporeal circulation time. Table 2: Spearman correlation coefficient and its significance between image-based measurements (access angle, distance from incision, annulus minimum diameter, maximum diameter, area, perimeter and age) and the outcome parameters [ECC, AoX, access difficulty score (4 grades) and access difficulty score (2 grades)] Spearman correlation coefficient (rho)/P-value  ECC   AoX   Access difficulty score (4 grades)   Access difficulty score (2 grades)     rho  P-value  rho  P-value  rho  P-value  rho  P-value  Access angle (degrees)  0.42  0.13  0.55  0.04  0.59  0.03  0.59  0.03  Distance from incision (mm)  −0.04  0.9  −0.07  0.81  −0.49  0.07  −0.39  0.17  Annulus minimum diameter (mm)  −0.22  0.44  −0.45  0.11  −0.25  0.39  −0.19  0.53  Annulus maximum diameter (mm)  −0.29  0.31  −0.39  0.17  −0.14  0.64  0.02  0.95  Annulus area (mm2)  −0.21  0.48  −0.36  0.21  −0.17  0.56  −0.06  0.85  Annulus perimeter (mm)  −0.17  0.57  −0.3  0.3  −0.15  0.61  −0.06  0.85  Age  −0.19  0.51  −0.26  0.37  −0.53  0.05  −0.5  0.07  Spearman correlation coefficient (rho)/P-value  ECC   AoX   Access difficulty score (4 grades)   Access difficulty score (2 grades)     rho  P-value  rho  P-value  rho  P-value  rho  P-value  Access angle (degrees)  0.42  0.13  0.55  0.04  0.59  0.03  0.59  0.03  Distance from incision (mm)  −0.04  0.9  −0.07  0.81  −0.49  0.07  −0.39  0.17  Annulus minimum diameter (mm)  −0.22  0.44  −0.45  0.11  −0.25  0.39  −0.19  0.53  Annulus maximum diameter (mm)  −0.29  0.31  −0.39  0.17  −0.14  0.64  0.02  0.95  Annulus area (mm2)  −0.21  0.48  −0.36  0.21  −0.17  0.56  −0.06  0.85  Annulus perimeter (mm)  −0.17  0.57  −0.3  0.3  −0.15  0.61  −0.06  0.85  Age  −0.19  0.51  −0.26  0.37  −0.53  0.05  −0.5  0.07  Bold values indicate significant correlation coefficients. AoX: aortic cross-clamping time; ECC: extracorporeal circulation time. Table 3: Pearson correlation between access angle and the outcome parameters [ECC, AoX, access difficulty score (4 grades) and access difficulty score (2 grades)]     ECC  AoX  Access difficulty score (4 grades)  Access difficulty score (2 grades)  Access angle  Pearson correlation  0.56  0.60  0.55  0.57    P-value (2-tailed)  0.04  0.02  0.04  0.03      ECC  AoX  Access difficulty score (4 grades)  Access difficulty score (2 grades)  Access angle  Pearson correlation  0.56  0.60  0.55  0.57    P-value (2-tailed)  0.04  0.02  0.04  0.03  AoX: aortic cross-clamping time; ECC: extracorporeal circulation time. Table 3: Pearson correlation between access angle and the outcome parameters [ECC, AoX, access difficulty score (4 grades) and access difficulty score (2 grades)]     ECC  AoX  Access difficulty score (4 grades)  Access difficulty score (2 grades)  Access angle  Pearson correlation  0.56  0.60  0.55  0.57    P-value (2-tailed)  0.04  0.02  0.04  0.03      ECC  AoX  Access difficulty score (4 grades)  Access difficulty score (2 grades)  Access angle  Pearson correlation  0.56  0.60  0.55  0.57    P-value (2-tailed)  0.04  0.02  0.04  0.03  AoX: aortic cross-clamping time; ECC: extracorporeal circulation time. Access angle measurements were significantly different (P = 0.03) between the easy and difficult groups based on the 2-point access difficulty score. The strongest association was between access angle and AoX. (See the scatter plot in Fig. 3.) In Fig. 4, the box plot shows a clear separation in the access angle between easy and difficult categories. On the other hand, there was an overlap in access angle between very easy and moderate categories on the 4 grades difficulty score. Figure 3: View largeDownload slide A scatter plot of access angle versus AoX showing the fitting line and the confidence interval. AoX: aortic cross-clamping time. Figure 3: View largeDownload slide A scatter plot of access angle versus AoX showing the fitting line and the confidence interval. AoX: aortic cross-clamping time. Figure 4: View largeDownload slide Box-whisker plots representing access angle values per group for the 4-point access difficulty scale (A), 2-point difficulty scale (B) and patient age values per group for the 2-point access difficulty scale (C). Figure 4: View largeDownload slide Box-whisker plots representing access angle values per group for the 4-point access difficulty scale (A), 2-point difficulty scale (B) and patient age values per group for the 2-point access difficulty scale (C). Age was negatively associated with the access difficulty score where access was judged to be easier for older patients. Also, AoX time was reduced with increasing age (Fig. 4). The receiver operating characteristic curve for access angle predicting access difficulty has an area under the curve of 0.86 (Fig. 5). The optimal access angle to distinguish access difficulty for the annulus stitching was 38° with a sensitivity of 78% and a specificity of 80%. Figure 5: View largeDownload slide The ROC curve for access angle showing different sensitivity and specificity of predicting procedure access difficulty for different access angle threshold values. ROC: receiver operating characteristic. Figure 5: View largeDownload slide The ROC curve for access angle showing different sensitivity and specificity of predicting procedure access difficulty for different access angle threshold values. ROC: receiver operating characteristic. DISCUSSION Herein, we have presented a novel mini-AVR tool that combines 3D imaging with quantitative planning measures. This study shows a significant association between access angle and extracorporeal circulation time, AoX and access difficulty. Against our expectation, access distance was not associated with the complexity of the procedure. In our study, it was also shown that older patients were easier to operate and have shorter AoX. Image-based quantitative preoperative planning of mini-AVR was proposed in multiple studies [20, 28]. To the best of our knowledge, this is the 1st study that presents a tool for automated quantitative measurements to assess the difficulty of mini-AVR procedures. Loor et al. proposed a visualization system using CTA images that utilizes multiplanar reconstruction, maximum intensity projections, volume rendering reconstructions and 3D rendering. In their system, image post-processing was performed on a dedicated stand-alone workstation. Similar to our solution, their tool renders the chest cage with the ascending aorta [22]. However, this tool does not provide quantitative measurements to support the planning, and therefore, they have not evaluated the added value of this system in clinical practice. In another study, Glauber et al. [21] proposed the usage of distance from the ascending aorta to the sternum and the angle between the sternum midline and the inclination of the ascending aorta as parameters for planning the prospective approach. However, in this approach, analysis was proposed to be performed manually on 2-dimensional axial and coronal CT images instead of 3D as in our study. Surprisingly, access distance was not associated with the outcome parameters, which contradicts with previous studies [21, 28]. The reason for this discrepancy with previous literature could be due to the treatment selection performed for the patients in this study. Only patients who were considered suitable for MS were included in this study. This may have resulted in the exclusion of patients with extensive distances. Also, the relatively small number of patients may have caused the absence of a statistically significant association. Because of this small number of patients, the difficulty score was dichotomized during the analysis. This study showed that procedures were considered easier to perform for older patients. The negative association found between age and access difficulty score may be explained by the fact that older patients have a larger aortic annulus [29] and smaller access angle probably due to an increased vessel tortuosity, which is common in the elderly [30]. The reported association of access angle and procedure complexity suggests that the access angle could be used to support the choice of type of mini-AVR approach and estimating the onsite complications. In this study, we showed that an access angle of 38° was a good threshold to distinguish easy from difficult MS procedures. For access angle larger than 38°, right anterior minithoracotomy could be a more suitable alternative. However, the comparison of different mini-AVR approaches was beyond the scope of this study. Our design has a number of limitations. This is a single-centre retrospective patient study including only a relatively small number of patients, which may not be a representative sample of the wider population, but it mainly shows the proof of concept and the potential value of the planning tool. The access distance was calculated from the chest bone instead of the skin to the sinotubular junction. The body mass index effect on the surgical results and access distance measurements was not studied in this research. Only a single rater performed the measurements, and a single surgeon assessed the subjective score for access difficulty. The robustness of these measures and its association with the outcome measures need to be validated by more raters. Also, measurements need to be collected for surgeries done by different operators with different expertise. In only 9 patients, calcium at the level of the annulus could be measured. This is insufficient to perform reliable statistical analysis, which is the reason that it was not reported in our study. We have introduced a mini-AVR planning tool, which combines 3D rendering of the patient’s anatomy with quantitative measures. With this tool, the access distance and angle are automatically determined. We have shown that this access angle is strongly associated with procedure complexity. This study also suggests that older patients are easier to operate than younger patients. Funding This work was supported by the Dutch Technology Foundation STW [grant number 11630], which is part of the Netherlands Organization for Scientific Research (NWO) and which is partly funded by the Ministry of Economic Affairs. Conflict of interest: Jan Baan received a research grant from Edwards and is a proctor for Edwards. Henk Marquering is founder and share holder of Nico-Lab. Other authors have no conflicts of interest to disclose. INFORMED CONSENT The institutional review board granted approval of the study design and waived informed consent since solely data obtained in the context of clinical care are utilized. REFERENCES 1 Vahanian A, Alfieri O, Al-Attar N, Antunes M, Bax J, Cormier B et al.   Transcatheter valve implantation for patients with aortic stenosis: a position statement from the European Association of Cardio-Thoracic Surgery (EACTS) and the European Society of Cardiology (ESC), in collaboration with the European Association of Percutaneous Cardiovascular Interventions (EAPCI). Eur J Cardiothorac Surg  2008; 34: 1– 8. Google Scholar CrossRef Search ADS PubMed  2 Iung B. A prospective survey of patients with valvular heart disease in Europe: the Euro Heart Survey on valvular heart disease. Eur Heart J  2003; 24: 1231– 43. Google Scholar CrossRef Search ADS PubMed  3 Foroutan F, Guyatt GH, O’Brien K, Bain E, Stein M, Bhagra S et al.   Prognosis after surgical replacement with a bioprosthetic aortic valve in patients with severe symptomatic aortic stenosis: systematic review of observational studies. BMJ  2016; 354: i5065. Google Scholar CrossRef Search ADS PubMed  4 Bustamante J, Cánovas S, Fernández ÁL. Minimally invasive aortic valve surgery—new solutions to old problems. In Hirota M (ed). Aortic Stenosis—Etiology Pathophysiology and Treatment . InTech, 2011. https://mts.intechopen.com/books/aortic-stenosis-etiology-pathophysiology-and-treatment/minimally-invasive-aortic-valve-surgery-new-solutions-to-old-problems. 5 Lamelas J, Nguyen TC. Minimally invasive valve surgery: when less is more. Semin Thorac Cardiovasc Surg  2015; 27: 49– 56. Google Scholar CrossRef Search ADS PubMed  6 Rosengart TK, Feldman T, Borger MA, Vassiliades TA, Gillinov AM, Hoercher KJ et al.   Percutaneous and minimally invasive valve procedures: a scientific statement from the American Heart Association Council on Cardiovascular Surgery and Anesthesia, Council on Clinical Cardiology, Functional Genomics and Translational Biology Interdisciplinary Working Group, and Quality of Care and Outcomes Research Interdisciplinary Working Group. Circulation  2008; 117: 1750– 67. Google Scholar CrossRef Search ADS PubMed  7 Malaisrie SC, Barnhart GR, Farivar RS, Mehall J, Hummel B, Rodriguez E et al.   Current era minimally invasive aortic valve replacement: techniques and practice. J Thorac Cardiovasc Surg  2014; 147: 6– 14. Google Scholar CrossRef Search ADS PubMed  8 Cohn LH, Adams DH, Couper GS, Bichell DP, Rosborough DM, Sears SP et al.   Minimally invasive cardiac valve surgery improves patient satisfaction while reducing costs of cardiac valve replacement and repair. Ann Surg  1997; 226: 421– 6. Google Scholar CrossRef Search ADS PubMed  9 Gilmanov D, Solinas M, Farneti PA, Cerillo AG, Kallushi E, Santarelli F et al.   Minimally invasive aortic valve replacement: 12-year single center experience. Ann Cardiothorac Surg  2015; 4: 160– 9. Google Scholar PubMed  10 Perrotta S, Lentini S. Ministernotomy approach for surgery of the aortic root and ascending aorta. Interact CardioVasc Thorac Surg  2009; 9: 849– 58. Google Scholar CrossRef Search ADS PubMed  11 Miceli A, Ferrarini M, Glauber M. Right anterior minithoracotomy for aortic valve replacement. Ann Cardiothorac Surg  2015; 4: 91– 3. Google Scholar PubMed  12 Glower DD, Desai BS, Hughes GC, Milano CA, Gaca JG. Aortic valve replacement via right minithoracotomy versus median sternotomy: a propensity score analysis. Innovations (Phila)  2014; 9: 75– 81. Google Scholar CrossRef Search ADS PubMed  13 Ruttmann E, Gilhofer TS, Ulmer H, Chevtchik O, Kocher A, Schistek R et al.   Propensity score-matched analysis of aortic valve replacement by mini-thoracotomy. J Heart Valve Dis  2010; 19: 606– 14. Google Scholar PubMed  14 Woo YJ, Seeburger J, Mohr FW. Minimally invasive valve surgery. Semin Thorac Cardiovasc Surg  2007; 19: 289– 98. Google Scholar CrossRef Search ADS PubMed  15 Raja SG, Benedetto U, Amrani M. Aortic valve replacement through J-shaped partial upper sternotomy. J Thorac Dis  2013; 5: S662– 8. Google Scholar PubMed  16 Hiraoka A, Totsugawa T, Kuinose M, Nakajima K, Chikazawa G, Tamura K et al.   Propensity score-matched analysis of minimally invasive aortic valve replacement. Circ J  2014; 78: 2876– 81. Google Scholar CrossRef Search ADS PubMed  17 Glauber M, Karimov JH, Farneti PA, Cerillo AG, Santarelli F, Ferrarini M et al.   Minimally invasive mitral valve surgery via right minithoracotomy. Multimed Man Cardiothorac Surg  2009; 2009: doi:10.1510/mmcts.2008.003350. 18 Glauber M, Miceli A, Gilmanov D, Ferrarini M, Bevilacqua S, Farneti PA et al.   Right anterior minithoracotomy versus conventional aortic valve replacement: a propensity score matched study. J Thorac Cardiovasc Surg  2013; 145: 1222– 6. Google Scholar CrossRef Search ADS PubMed  19 Miceli A, Santarpino G, Pfeiffer S, Murzi M, Gilmanov D, Concistré G et al.   Minimally invasive aortic valve replacement with Perceval S sutureless valve: early outcomes and one-year survival from two European centers. J Thorac Cardiovasc Surg  2014; 148: 2838– 43. Google Scholar CrossRef Search ADS PubMed  20 Stelzer P. 3D thinking for mini-AVR. JACC Cardiovasc Imaging  2013; 6: 272– 3. Google Scholar CrossRef Search ADS PubMed  21 Glauber M, Ferrarini M, Miceli A. Minimally invasive aortic valve surgery: state of the art and future directions. Ann Cardiothorac Surg  2015; 4: 26– 32. Google Scholar PubMed  22 Loor G, Desai MY, Roselli EE. Pre-operative 3D CT imaging for virtual planning of minimally invasive aortic valve surgery. JACC Cardiovasc Imaging  2013; 6: 269– 71. Google Scholar CrossRef Search ADS PubMed  23 Loor G, Roselli EE. Imaging and minimally invasive aortic valve replacement. Ann Cardiothorac Surg  2015; 4: 62– 6. Google Scholar PubMed  24 Elattar M, Wiegerinck E, van Kesteren F, Dubois L, Planken N, Vanbavel E et al.   Automatic aortic root landmark detection in CTA images for preprocedural planning of transcatheter aortic valve implantation. Int J Cardiovasc Imaging  2016; 32: 501– 11. Google Scholar CrossRef Search ADS PubMed  25 Elattar MA, van Kesteren F, Wiegerinck EM, Vanbavel E, Baan J, Cocchieri R et al.   Automated CTA based measurements for planning support of minimally invasive aortic valve replacement surgery. Med Eng Phys  2017; 39: 123– 8. Google Scholar CrossRef Search ADS PubMed  26 Cosgrove DM, Sabik JF. Minimally invasive approach for aortic valve operations. Ann Thorac Surg  1996; 62: 596– 7. Google Scholar CrossRef Search ADS PubMed  27 Chun EJ, Choi SI, Lim C, Park K-H, Chang H-J, Choi D-J et al.   Aortic stenosis: evaluation with multidetector CT angiography and MR imaging. Korean J Radiol  2008; 9: 439– 48. Google Scholar CrossRef Search ADS PubMed  28 Miceli A, Murzi M, Gilmanov D, Fugà R, Ferrarini M, Solinas M et al.   Minimally invasive aortic valve replacement using right minithoracotomy is associated with better outcomes than ministernotomy. J Thorac Cardiovasc Surg  2014; 148: 133– 7. Google Scholar CrossRef Search ADS PubMed  29 Vriz O, Aboyans V, D’Andrea A, Ferrara F, Acri E, Limongelli G et al.   Normal values of aortic root dimensions in healthy adults. Am J Cardiol  2014; 114: 921– 7. Google Scholar CrossRef Search ADS PubMed  30 Sugawara J, Hayashi K, Yokoi T, Tanaka H. Age-associated elongation of the ascending aorta in adults. JACC Cardiovasc Imaging  2008; 1: 739– 48. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2018. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved. 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 Interactive CardioVascular and Thoracic Surgery Oxford University Press

A computed tomography-based planning tool for predicting difficulty of minimally invasive aortic valve replacement

Loading next page...
 
/lp/ou_press/a-computed-tomography-based-planning-tool-for-predicting-difficulty-of-dfTT2RdoPW
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved.
ISSN
1569-9293
eISSN
1569-9285
D.O.I.
10.1093/icvts/ivy128
Publisher site
See Article on Publisher Site

Abstract

Abstract OBJECTIVES Minimally invasive aortic valve replacement has proven its value over the last decade by its significant advancement and reduction in mortality, morbidity and admission time. However, minimally invasive aortic valve replacement is associated with some on-site difficulties such as limited aortic annulus exposure. Currently, computed tomography scans are used to evaluate the anatomical relationship among the intercostal spaces, ascending aorta and aortic valve prior to surgery. We hypothesized that quantitative measurements of access distance and access angle are associated with outcome and access difficulty. METHODS We introduce a novel minimally invasive aortic valve replacement planning prototype that allows automatic measurements of access angle, access distance and aortic annulus dimensions. The prototype visualizes these measurements on the chest cage as ISO contours. The association of these measures with outcome parameters such as extracorporeal circulation time, aortic cross-clamping time and access difficulty score was assessed. We included 14 patients who received a new valve by ministernotomy. RESULTS The mean access angle was 40.3 ± 5.1°. It was strongly associated with aortic cross-clamping time (Pearson correlation coefficient = 0.60, P = 0.02) and access difficulty score (Spearman’s rank correlation coefficient = 0.57, P = 0.03). Access angles were significantly different between easy and difficult access groups (P = 0.03). There was no significant association between access distance and outcome parameters. CONCLUSIONS Access angle is strongly associated with procedure complexity. The automated presentation of this measure suggests added value of the prototype in clinical practice. Minimally invasive aortic valve replacement, Aortic stenosis, Planning tool, Access difficulty, Computer tomography INTRODUCTION Valve disease is a serious health problem as it carries a poor prognosis, and its prevalence is strongly associated with the population aging [1]. Aortic stenosis is currently the most frequent native valve disease in Europe, which is most often seen in elderly patients [2]. Valve replacement is the conventional therapy for patients with severe aortic stenosis who have symptoms or observable consequences such as left ventricular dysfunction [3]. Traditional minimally invasive aortic valve replacement (mini-AVR) is performed during an open surgical procedure with full sternotomy and with arterial cannulation in the ascendant aorta and venous cannulation in the right atrium to enable extracorporeal circulation by means of a heart–lung machine [4]. Full sternotomy has the advantage of a direct view and access to all the cardiac structures [5]. Operative mortality is quite low, even in elderly patients when properly selected and long-term results are satisfactory [6]. However, risk of surgery is higher in elderly patients with significant comorbidities [4]. During the past decade, surgical techniques improved creating an area for less and minimally invasive procedures. Moreover, a range of procedures classified as mini-AVR has been introduced [7]. When compared with conventional surgery, mini-AVR has shown to reduce postoperative mortality and morbidity. In addition, patients recover faster requiring less rehabilitations resources and shorter admission time [8]. The most common minimally invasive approach for valve replacement is partial upper ministernotomy (MS), followed by the right anterior minithoracotomy (RT) [9]. MS is performed through a vertical incision through skin and sternum with the shape of the character J starting from the sternal angle (manubriosternal joint) moving caudally for 4 cm in the sternum body and completed by a transverse sternal incision [10]. RT is performed through a 5–6-cm skin incision at the level of the 2nd or 3rd intercostal space, starting from the border of the sternum toward the lateral right side [11]. Few studies compared the outcomes of the approaches in terms of quality of life and surgery duration. Only marginal benefits for MS have been reported when compared with RT [12, 13]. The MS approach is generally the preferred approach by surgeons [14, 15]. However, the right anterior minithoracotomy (RT) has been reported to be associated with improved outcomes [12, 16, 17], despite longer duration of surgery when compared with MS [17–19]. The exposure of the aortic root right cusp annulus side is limited for the MS approach. For both techniques, incision location selection is of utmost importance for successful valve replacement [20, 21]. Preoperative planning of patients eligible for mini-AVR is performed using computed tomography angiography (CTA). A previous study proposed a visualization system for surgical planning in the workup for mini-AVR using CTA images. This tool renders the chest cage and the ascending aorta to visually support procedure planning and approach selection [22]. Current planning tools, however, do not yet provide quantitative measurements. In another study, Glauber et al. [21] stated that patients are suitable for RT only if the following criteria are met: (i) the distance from the ascending aorta to the sternum should not exceed 10 cm; (ii) the angle between the sternum midline and the inclination of the ascending aorta should be larger than 45°. It has also been suggested that it is important to measure aortic valve dimensions such as the diameter of the aortic valve annulus, the length of the ascending aorta and the calcifications of the aortic valve and the ascending aorta in the preprocedural planning process because it is strongly recommended to remove all the eccentric calcifications and create a complete decalcification of the aortic annulus [21, 23]. We developed a mini-AVR planning tool that combines 3-dimensional (3D) visualization of the access site in addition to quantitative assessment of important planning measures. This tool provides the access angle and distance from the incision location on the chest cage quantitatively and visually using 3D rendering methods. We believe that both access visualization and quantification give an impression of the difficulty of access by measuring how far and aligned the access with the aortic root. In this tool, the mini-AVR planning tool is validated by assessment of the association of the provided quantitative measures with clinical outcome parameters: surgery time, clamping time and annulus access difficulty score. METHODS Planning tool Automated measures The mini-AVR planning tool provides both distance and access angle from potential incision sites to the aortic root based on the geometrical location of the aortic root landmarks and incision locations [24, 25]. These parameters are determined as follows: access distance is defined as the distance between incision location in the chest bone and the aortic sinotubular junction for both the MS and RT approaches. Access angle is defined as the angle between the aortic root centreline axis and incision–annulus line. The aortic root centre axis is the vector that connects annulus and sinotubular junction centres. The incision–annulus centre axis is the vector that connects the incision with the annulus centre (Fig. 1). Other annulus dimensions such as minimum, maximum diameters, area, perimeter and calcium volume are calculated based on a previously presented automated method, which automatically detected aortic root landmarks and calculated sizing parameters [24]. Figure 1: View largeDownload slide Schematic showing the methodology of calculating the access distance and angle based on the location of 3 landmarks: aortic annulus centre, sinotubular junction and incision point. Figure 1: View largeDownload slide Schematic showing the methodology of calculating the access distance and angle based on the location of 3 landmarks: aortic annulus centre, sinotubular junction and incision point. The mini-AVR planning tool has a specific workflow to calculate access difficulty. This workflow consists of a set of interactive steps. The tool can detect the aortic annulus, sinotubular junction and segment the ribcage from the full body computed tomography (CT) scan, then produce the final measurements. Access distance and angle are presented using ISO contour maps (Fig. 2). To calculate the distance and angle maps, a mesh that represents the ribs is reconstructed. An ISO contour connects points of equal measurement values. Each 2 ISO contours define a range of values between them. The ISO contours of distance and angle were rendered on the 3D CT volume using a blue palette and green palette, respectively. The difference between each 2 successive ISO contours of the distance and angle measurements scale was 5 mm and 10°, respectively, as shown in Fig. 2. The tool allows different rendering modes combined with the ISO contours. The rendering modes include full body volume rendering, chest cage with segmented aorta, skin with segmented aorta and the segmented aorta with calcification. Figure 2: View largeDownload slide Screenshot of the developed tool prototype showing the resulted ISO contours in 3-dimensional views (bottom). These contours represent distance and access angle measurements using highlights from the colour palette; the zoomed-in-colour map represents different measurement ranges. Figure 2: View largeDownload slide Screenshot of the developed tool prototype showing the resulted ISO contours in 3-dimensional views (bottom). These contours represent distance and access angle measurements using highlights from the colour palette; the zoomed-in-colour map represents different measurement ranges. Usage The developed prototype works as follows: first, the user loads 3D CTA images. After choosing the mini-AVR workflow from the main User Interface, the software automatically segments the aortic root. Next, the annulus and sinotubular junction are detected. After the automatic detection, extra tools are made available for the user to correct the detected landmarks if needed. The planning prototype contains a 3D probe, which helps the user to interactively select a location on the 3D-rendered chest cage for local measurements such as access distance and angle measurements. Distance and angle measurements are evaluated for all reconstructed chest cage mesh points. Validation Study population We collected preprocedural 3D volume image data of 14 patients who underwent mini-AVR between January and December 2015 including 3 CT and 11 CTA acquisitions from our institute. All CT scans were performed on a Philips Brilliance 64 slice CT scanner (Philips, Cleveland, OH, USA); imaging parameters were 120 kV and convolution kernel B. The chest, abdomen and pelvis were scanned using 1 bolus of 120-ml contrast Iomeron 400 (Bracco Imaging SpA, Milan, Italy) intravenously infused at a rate of 5 ml/s. In case of dynamic scans, which included 10 cardiac phases, we selected images acquired at 70% of the cardiac cycle, corresponding to mid-diastole. This specific phase mimics the non-beating heart and extracorporeal circulation settings during surgery [26] Also, during this cardiac phase, the aortic valve is closed [27]. All image volumes contained approximately 500–800 slices. Each slice in a volume contained 512 × 512 isotropic pixels with a 16-bit depth. The in-plane image resolution varied from 0.44 to 0.68 mm. The slice thickness for all data sets was 0.9 mm, and the overlap between each 2 successive slices was 0.45 mm. All patients included in this study were operated using the MS approach by 1 cardiothoracic surgeon (A.K.). Image-based parameters Annulus measurements such as minimum, maximum diameter, area, perimeter and calcium volume were assessed only in CTA images. To determine the angle and distance measurements at the incision location used during the surgery, the 3D probe was placed at the centre of the manubriosternal joint that was used by the surgeon to mark the incision. Outcome parameters Extracorporeal circulation time, aortic cross-clamping time (AoX) and access difficulty score were collected as outcome parameters. Access difficulty score was subjectively assessed by the operator. It reflects the complexity of the annulus access while performing the stitches to implant the new valve. The score has 4 grades (very easy, easy, moderate and difficult), and we dichotomized this score to easy (very easy and easy) and difficult (moderate and difficult). Assuring independent scoring, rater was blinded to additional information. Statistical analysis Normally distributed and continuous values are expressed as mean ± standard deviation and median and interquartile range otherwise. Significance of associations between outcome parameters and image-based parameters were tested using the Spearman’s rank correlation coefficient. Significant associated parameters were also evaluated using Pearson product–moment correlation coefficient. Aortic annulus measurements were compared between groups of easy and difficult access. The distribution of access angles and AoX for dichotomized access difficulty categories was visualized with box-and-whisker plots. Scatter plots were used to illustrate the association between access angle and access difficulty score. Significant difference of access angle between easy and difficult access groups was determined by paired t-tests. To predict procedure difficulty, we selected an optimal access angle threshold, based on the receiver operating characteristic curve. Statistical analyses were performed using SPSS (version 19.0, SPSS Inc., Chicago, IL, USA) and MATLAB (Version R2013b, The MathWorks Inc., Natick, MA, USA). Associations with P-values of 0.05 or smaller were considered statistically significant. RESULTS Access distance and access angle measurements were normal for both CT and CTA scans for all patients. Aortic valve stenosis was present in all patients included in this study, and bicuspid aortic valve stenosis was present in 2 (14%) patients. Two patients were scanned using CT, and the remainder was scanned using CTA. Four scans were excluded from annulus dimension measurements. For 2 patients, only non-contrast CT was available, which makes annulus measurements less accurate to assess because of poor contrast between blood and aortic root wall. The other 2 scans were excluded because they were scanned after surgery, which hinders annulus measurements due to the strong blooming effect of the stent at annulus level. Access angle and distance measurements using the 3D probe tool were successful for all patients. Average access angle was 40.3 ± 5.1°, and the distance from incision to the sinotubular junction was 94.9 ± 11.1 mm. The remaining measurements are reported in Table 1. Table 1: Average, standard deviation, median and IQR of the access angle, distance from incision, annulus minimum diameter, maximum diameter, area and perimeter   Mean  Standard deviation  Access angle (degrees)  40.3  5.1  Distance from incision (mm)  94.9  11.1  Annulus minimum diameter (mm)  26.2  3.1  Annulus maximum diameter (mm)  32.4  2.9  Annulus area (mm2)  687.1  122.5  Annulus perimeter (mm), median (IQR)  93.6 (88.7–102.2)    Mean  Standard deviation  Access angle (degrees)  40.3  5.1  Distance from incision (mm)  94.9  11.1  Annulus minimum diameter (mm)  26.2  3.1  Annulus maximum diameter (mm)  32.4  2.9  Annulus area (mm2)  687.1  122.5  Annulus perimeter (mm), median (IQR)  93.6 (88.7–102.2)  IQR: interquartile range. Table 1: Average, standard deviation, median and IQR of the access angle, distance from incision, annulus minimum diameter, maximum diameter, area and perimeter   Mean  Standard deviation  Access angle (degrees)  40.3  5.1  Distance from incision (mm)  94.9  11.1  Annulus minimum diameter (mm)  26.2  3.1  Annulus maximum diameter (mm)  32.4  2.9  Annulus area (mm2)  687.1  122.5  Annulus perimeter (mm), median (IQR)  93.6 (88.7–102.2)    Mean  Standard deviation  Access angle (degrees)  40.3  5.1  Distance from incision (mm)  94.9  11.1  Annulus minimum diameter (mm)  26.2  3.1  Annulus maximum diameter (mm)  32.4  2.9  Annulus area (mm2)  687.1  122.5  Annulus perimeter (mm), median (IQR)  93.6 (88.7–102.2)  IQR: interquartile range. The access angle was significantly associated with AoX, 4 graded and 2 graded access difficulty scores with Spearman correlation coefficient of 0.55 (P = 0.042), 0.59 (P = 0.028) and 0.60 (P = 0.025), respectively. The other image-based measurements were not significantly associated with the outcome parameters (see Table 2). Linear correlation between access angle and all outcome parameters was found to be significant. The Pearson product–moment correlation coefficients are reported in Table 3. Table 2: Spearman correlation coefficient and its significance between image-based measurements (access angle, distance from incision, annulus minimum diameter, maximum diameter, area, perimeter and age) and the outcome parameters [ECC, AoX, access difficulty score (4 grades) and access difficulty score (2 grades)] Spearman correlation coefficient (rho)/P-value  ECC   AoX   Access difficulty score (4 grades)   Access difficulty score (2 grades)     rho  P-value  rho  P-value  rho  P-value  rho  P-value  Access angle (degrees)  0.42  0.13  0.55  0.04  0.59  0.03  0.59  0.03  Distance from incision (mm)  −0.04  0.9  −0.07  0.81  −0.49  0.07  −0.39  0.17  Annulus minimum diameter (mm)  −0.22  0.44  −0.45  0.11  −0.25  0.39  −0.19  0.53  Annulus maximum diameter (mm)  −0.29  0.31  −0.39  0.17  −0.14  0.64  0.02  0.95  Annulus area (mm2)  −0.21  0.48  −0.36  0.21  −0.17  0.56  −0.06  0.85  Annulus perimeter (mm)  −0.17  0.57  −0.3  0.3  −0.15  0.61  −0.06  0.85  Age  −0.19  0.51  −0.26  0.37  −0.53  0.05  −0.5  0.07  Spearman correlation coefficient (rho)/P-value  ECC   AoX   Access difficulty score (4 grades)   Access difficulty score (2 grades)     rho  P-value  rho  P-value  rho  P-value  rho  P-value  Access angle (degrees)  0.42  0.13  0.55  0.04  0.59  0.03  0.59  0.03  Distance from incision (mm)  −0.04  0.9  −0.07  0.81  −0.49  0.07  −0.39  0.17  Annulus minimum diameter (mm)  −0.22  0.44  −0.45  0.11  −0.25  0.39  −0.19  0.53  Annulus maximum diameter (mm)  −0.29  0.31  −0.39  0.17  −0.14  0.64  0.02  0.95  Annulus area (mm2)  −0.21  0.48  −0.36  0.21  −0.17  0.56  −0.06  0.85  Annulus perimeter (mm)  −0.17  0.57  −0.3  0.3  −0.15  0.61  −0.06  0.85  Age  −0.19  0.51  −0.26  0.37  −0.53  0.05  −0.5  0.07  Bold values indicate significant correlation coefficients. AoX: aortic cross-clamping time; ECC: extracorporeal circulation time. Table 2: Spearman correlation coefficient and its significance between image-based measurements (access angle, distance from incision, annulus minimum diameter, maximum diameter, area, perimeter and age) and the outcome parameters [ECC, AoX, access difficulty score (4 grades) and access difficulty score (2 grades)] Spearman correlation coefficient (rho)/P-value  ECC   AoX   Access difficulty score (4 grades)   Access difficulty score (2 grades)     rho  P-value  rho  P-value  rho  P-value  rho  P-value  Access angle (degrees)  0.42  0.13  0.55  0.04  0.59  0.03  0.59  0.03  Distance from incision (mm)  −0.04  0.9  −0.07  0.81  −0.49  0.07  −0.39  0.17  Annulus minimum diameter (mm)  −0.22  0.44  −0.45  0.11  −0.25  0.39  −0.19  0.53  Annulus maximum diameter (mm)  −0.29  0.31  −0.39  0.17  −0.14  0.64  0.02  0.95  Annulus area (mm2)  −0.21  0.48  −0.36  0.21  −0.17  0.56  −0.06  0.85  Annulus perimeter (mm)  −0.17  0.57  −0.3  0.3  −0.15  0.61  −0.06  0.85  Age  −0.19  0.51  −0.26  0.37  −0.53  0.05  −0.5  0.07  Spearman correlation coefficient (rho)/P-value  ECC   AoX   Access difficulty score (4 grades)   Access difficulty score (2 grades)     rho  P-value  rho  P-value  rho  P-value  rho  P-value  Access angle (degrees)  0.42  0.13  0.55  0.04  0.59  0.03  0.59  0.03  Distance from incision (mm)  −0.04  0.9  −0.07  0.81  −0.49  0.07  −0.39  0.17  Annulus minimum diameter (mm)  −0.22  0.44  −0.45  0.11  −0.25  0.39  −0.19  0.53  Annulus maximum diameter (mm)  −0.29  0.31  −0.39  0.17  −0.14  0.64  0.02  0.95  Annulus area (mm2)  −0.21  0.48  −0.36  0.21  −0.17  0.56  −0.06  0.85  Annulus perimeter (mm)  −0.17  0.57  −0.3  0.3  −0.15  0.61  −0.06  0.85  Age  −0.19  0.51  −0.26  0.37  −0.53  0.05  −0.5  0.07  Bold values indicate significant correlation coefficients. AoX: aortic cross-clamping time; ECC: extracorporeal circulation time. Table 3: Pearson correlation between access angle and the outcome parameters [ECC, AoX, access difficulty score (4 grades) and access difficulty score (2 grades)]     ECC  AoX  Access difficulty score (4 grades)  Access difficulty score (2 grades)  Access angle  Pearson correlation  0.56  0.60  0.55  0.57    P-value (2-tailed)  0.04  0.02  0.04  0.03      ECC  AoX  Access difficulty score (4 grades)  Access difficulty score (2 grades)  Access angle  Pearson correlation  0.56  0.60  0.55  0.57    P-value (2-tailed)  0.04  0.02  0.04  0.03  AoX: aortic cross-clamping time; ECC: extracorporeal circulation time. Table 3: Pearson correlation between access angle and the outcome parameters [ECC, AoX, access difficulty score (4 grades) and access difficulty score (2 grades)]     ECC  AoX  Access difficulty score (4 grades)  Access difficulty score (2 grades)  Access angle  Pearson correlation  0.56  0.60  0.55  0.57    P-value (2-tailed)  0.04  0.02  0.04  0.03      ECC  AoX  Access difficulty score (4 grades)  Access difficulty score (2 grades)  Access angle  Pearson correlation  0.56  0.60  0.55  0.57    P-value (2-tailed)  0.04  0.02  0.04  0.03  AoX: aortic cross-clamping time; ECC: extracorporeal circulation time. Access angle measurements were significantly different (P = 0.03) between the easy and difficult groups based on the 2-point access difficulty score. The strongest association was between access angle and AoX. (See the scatter plot in Fig. 3.) In Fig. 4, the box plot shows a clear separation in the access angle between easy and difficult categories. On the other hand, there was an overlap in access angle between very easy and moderate categories on the 4 grades difficulty score. Figure 3: View largeDownload slide A scatter plot of access angle versus AoX showing the fitting line and the confidence interval. AoX: aortic cross-clamping time. Figure 3: View largeDownload slide A scatter plot of access angle versus AoX showing the fitting line and the confidence interval. AoX: aortic cross-clamping time. Figure 4: View largeDownload slide Box-whisker plots representing access angle values per group for the 4-point access difficulty scale (A), 2-point difficulty scale (B) and patient age values per group for the 2-point access difficulty scale (C). Figure 4: View largeDownload slide Box-whisker plots representing access angle values per group for the 4-point access difficulty scale (A), 2-point difficulty scale (B) and patient age values per group for the 2-point access difficulty scale (C). Age was negatively associated with the access difficulty score where access was judged to be easier for older patients. Also, AoX time was reduced with increasing age (Fig. 4). The receiver operating characteristic curve for access angle predicting access difficulty has an area under the curve of 0.86 (Fig. 5). The optimal access angle to distinguish access difficulty for the annulus stitching was 38° with a sensitivity of 78% and a specificity of 80%. Figure 5: View largeDownload slide The ROC curve for access angle showing different sensitivity and specificity of predicting procedure access difficulty for different access angle threshold values. ROC: receiver operating characteristic. Figure 5: View largeDownload slide The ROC curve for access angle showing different sensitivity and specificity of predicting procedure access difficulty for different access angle threshold values. ROC: receiver operating characteristic. DISCUSSION Herein, we have presented a novel mini-AVR tool that combines 3D imaging with quantitative planning measures. This study shows a significant association between access angle and extracorporeal circulation time, AoX and access difficulty. Against our expectation, access distance was not associated with the complexity of the procedure. In our study, it was also shown that older patients were easier to operate and have shorter AoX. Image-based quantitative preoperative planning of mini-AVR was proposed in multiple studies [20, 28]. To the best of our knowledge, this is the 1st study that presents a tool for automated quantitative measurements to assess the difficulty of mini-AVR procedures. Loor et al. proposed a visualization system using CTA images that utilizes multiplanar reconstruction, maximum intensity projections, volume rendering reconstructions and 3D rendering. In their system, image post-processing was performed on a dedicated stand-alone workstation. Similar to our solution, their tool renders the chest cage with the ascending aorta [22]. However, this tool does not provide quantitative measurements to support the planning, and therefore, they have not evaluated the added value of this system in clinical practice. In another study, Glauber et al. [21] proposed the usage of distance from the ascending aorta to the sternum and the angle between the sternum midline and the inclination of the ascending aorta as parameters for planning the prospective approach. However, in this approach, analysis was proposed to be performed manually on 2-dimensional axial and coronal CT images instead of 3D as in our study. Surprisingly, access distance was not associated with the outcome parameters, which contradicts with previous studies [21, 28]. The reason for this discrepancy with previous literature could be due to the treatment selection performed for the patients in this study. Only patients who were considered suitable for MS were included in this study. This may have resulted in the exclusion of patients with extensive distances. Also, the relatively small number of patients may have caused the absence of a statistically significant association. Because of this small number of patients, the difficulty score was dichotomized during the analysis. This study showed that procedures were considered easier to perform for older patients. The negative association found between age and access difficulty score may be explained by the fact that older patients have a larger aortic annulus [29] and smaller access angle probably due to an increased vessel tortuosity, which is common in the elderly [30]. The reported association of access angle and procedure complexity suggests that the access angle could be used to support the choice of type of mini-AVR approach and estimating the onsite complications. In this study, we showed that an access angle of 38° was a good threshold to distinguish easy from difficult MS procedures. For access angle larger than 38°, right anterior minithoracotomy could be a more suitable alternative. However, the comparison of different mini-AVR approaches was beyond the scope of this study. Our design has a number of limitations. This is a single-centre retrospective patient study including only a relatively small number of patients, which may not be a representative sample of the wider population, but it mainly shows the proof of concept and the potential value of the planning tool. The access distance was calculated from the chest bone instead of the skin to the sinotubular junction. The body mass index effect on the surgical results and access distance measurements was not studied in this research. Only a single rater performed the measurements, and a single surgeon assessed the subjective score for access difficulty. The robustness of these measures and its association with the outcome measures need to be validated by more raters. Also, measurements need to be collected for surgeries done by different operators with different expertise. In only 9 patients, calcium at the level of the annulus could be measured. This is insufficient to perform reliable statistical analysis, which is the reason that it was not reported in our study. We have introduced a mini-AVR planning tool, which combines 3D rendering of the patient’s anatomy with quantitative measures. With this tool, the access distance and angle are automatically determined. We have shown that this access angle is strongly associated with procedure complexity. This study also suggests that older patients are easier to operate than younger patients. Funding This work was supported by the Dutch Technology Foundation STW [grant number 11630], which is part of the Netherlands Organization for Scientific Research (NWO) and which is partly funded by the Ministry of Economic Affairs. Conflict of interest: Jan Baan received a research grant from Edwards and is a proctor for Edwards. Henk Marquering is founder and share holder of Nico-Lab. Other authors have no conflicts of interest to disclose. INFORMED CONSENT The institutional review board granted approval of the study design and waived informed consent since solely data obtained in the context of clinical care are utilized. REFERENCES 1 Vahanian A, Alfieri O, Al-Attar N, Antunes M, Bax J, Cormier B et al.   Transcatheter valve implantation for patients with aortic stenosis: a position statement from the European Association of Cardio-Thoracic Surgery (EACTS) and the European Society of Cardiology (ESC), in collaboration with the European Association of Percutaneous Cardiovascular Interventions (EAPCI). Eur J Cardiothorac Surg  2008; 34: 1– 8. Google Scholar CrossRef Search ADS PubMed  2 Iung B. A prospective survey of patients with valvular heart disease in Europe: the Euro Heart Survey on valvular heart disease. Eur Heart J  2003; 24: 1231– 43. Google Scholar CrossRef Search ADS PubMed  3 Foroutan F, Guyatt GH, O’Brien K, Bain E, Stein M, Bhagra S et al.   Prognosis after surgical replacement with a bioprosthetic aortic valve in patients with severe symptomatic aortic stenosis: systematic review of observational studies. BMJ  2016; 354: i5065. Google Scholar CrossRef Search ADS PubMed  4 Bustamante J, Cánovas S, Fernández ÁL. Minimally invasive aortic valve surgery—new solutions to old problems. In Hirota M (ed). Aortic Stenosis—Etiology Pathophysiology and Treatment . InTech, 2011. https://mts.intechopen.com/books/aortic-stenosis-etiology-pathophysiology-and-treatment/minimally-invasive-aortic-valve-surgery-new-solutions-to-old-problems. 5 Lamelas J, Nguyen TC. Minimally invasive valve surgery: when less is more. Semin Thorac Cardiovasc Surg  2015; 27: 49– 56. Google Scholar CrossRef Search ADS PubMed  6 Rosengart TK, Feldman T, Borger MA, Vassiliades TA, Gillinov AM, Hoercher KJ et al.   Percutaneous and minimally invasive valve procedures: a scientific statement from the American Heart Association Council on Cardiovascular Surgery and Anesthesia, Council on Clinical Cardiology, Functional Genomics and Translational Biology Interdisciplinary Working Group, and Quality of Care and Outcomes Research Interdisciplinary Working Group. Circulation  2008; 117: 1750– 67. Google Scholar CrossRef Search ADS PubMed  7 Malaisrie SC, Barnhart GR, Farivar RS, Mehall J, Hummel B, Rodriguez E et al.   Current era minimally invasive aortic valve replacement: techniques and practice. J Thorac Cardiovasc Surg  2014; 147: 6– 14. Google Scholar CrossRef Search ADS PubMed  8 Cohn LH, Adams DH, Couper GS, Bichell DP, Rosborough DM, Sears SP et al.   Minimally invasive cardiac valve surgery improves patient satisfaction while reducing costs of cardiac valve replacement and repair. Ann Surg  1997; 226: 421– 6. Google Scholar CrossRef Search ADS PubMed  9 Gilmanov D, Solinas M, Farneti PA, Cerillo AG, Kallushi E, Santarelli F et al.   Minimally invasive aortic valve replacement: 12-year single center experience. Ann Cardiothorac Surg  2015; 4: 160– 9. Google Scholar PubMed  10 Perrotta S, Lentini S. Ministernotomy approach for surgery of the aortic root and ascending aorta. Interact CardioVasc Thorac Surg  2009; 9: 849– 58. Google Scholar CrossRef Search ADS PubMed  11 Miceli A, Ferrarini M, Glauber M. Right anterior minithoracotomy for aortic valve replacement. Ann Cardiothorac Surg  2015; 4: 91– 3. Google Scholar PubMed  12 Glower DD, Desai BS, Hughes GC, Milano CA, Gaca JG. Aortic valve replacement via right minithoracotomy versus median sternotomy: a propensity score analysis. Innovations (Phila)  2014; 9: 75– 81. Google Scholar CrossRef Search ADS PubMed  13 Ruttmann E, Gilhofer TS, Ulmer H, Chevtchik O, Kocher A, Schistek R et al.   Propensity score-matched analysis of aortic valve replacement by mini-thoracotomy. J Heart Valve Dis  2010; 19: 606– 14. Google Scholar PubMed  14 Woo YJ, Seeburger J, Mohr FW. Minimally invasive valve surgery. Semin Thorac Cardiovasc Surg  2007; 19: 289– 98. Google Scholar CrossRef Search ADS PubMed  15 Raja SG, Benedetto U, Amrani M. Aortic valve replacement through J-shaped partial upper sternotomy. J Thorac Dis  2013; 5: S662– 8. Google Scholar PubMed  16 Hiraoka A, Totsugawa T, Kuinose M, Nakajima K, Chikazawa G, Tamura K et al.   Propensity score-matched analysis of minimally invasive aortic valve replacement. Circ J  2014; 78: 2876– 81. Google Scholar CrossRef Search ADS PubMed  17 Glauber M, Karimov JH, Farneti PA, Cerillo AG, Santarelli F, Ferrarini M et al.   Minimally invasive mitral valve surgery via right minithoracotomy. Multimed Man Cardiothorac Surg  2009; 2009: doi:10.1510/mmcts.2008.003350. 18 Glauber M, Miceli A, Gilmanov D, Ferrarini M, Bevilacqua S, Farneti PA et al.   Right anterior minithoracotomy versus conventional aortic valve replacement: a propensity score matched study. J Thorac Cardiovasc Surg  2013; 145: 1222– 6. Google Scholar CrossRef Search ADS PubMed  19 Miceli A, Santarpino G, Pfeiffer S, Murzi M, Gilmanov D, Concistré G et al.   Minimally invasive aortic valve replacement with Perceval S sutureless valve: early outcomes and one-year survival from two European centers. J Thorac Cardiovasc Surg  2014; 148: 2838– 43. Google Scholar CrossRef Search ADS PubMed  20 Stelzer P. 3D thinking for mini-AVR. JACC Cardiovasc Imaging  2013; 6: 272– 3. Google Scholar CrossRef Search ADS PubMed  21 Glauber M, Ferrarini M, Miceli A. Minimally invasive aortic valve surgery: state of the art and future directions. Ann Cardiothorac Surg  2015; 4: 26– 32. Google Scholar PubMed  22 Loor G, Desai MY, Roselli EE. Pre-operative 3D CT imaging for virtual planning of minimally invasive aortic valve surgery. JACC Cardiovasc Imaging  2013; 6: 269– 71. Google Scholar CrossRef Search ADS PubMed  23 Loor G, Roselli EE. Imaging and minimally invasive aortic valve replacement. Ann Cardiothorac Surg  2015; 4: 62– 6. Google Scholar PubMed  24 Elattar M, Wiegerinck E, van Kesteren F, Dubois L, Planken N, Vanbavel E et al.   Automatic aortic root landmark detection in CTA images for preprocedural planning of transcatheter aortic valve implantation. Int J Cardiovasc Imaging  2016; 32: 501– 11. Google Scholar CrossRef Search ADS PubMed  25 Elattar MA, van Kesteren F, Wiegerinck EM, Vanbavel E, Baan J, Cocchieri R et al.   Automated CTA based measurements for planning support of minimally invasive aortic valve replacement surgery. Med Eng Phys  2017; 39: 123– 8. Google Scholar CrossRef Search ADS PubMed  26 Cosgrove DM, Sabik JF. Minimally invasive approach for aortic valve operations. Ann Thorac Surg  1996; 62: 596– 7. Google Scholar CrossRef Search ADS PubMed  27 Chun EJ, Choi SI, Lim C, Park K-H, Chang H-J, Choi D-J et al.   Aortic stenosis: evaluation with multidetector CT angiography and MR imaging. Korean J Radiol  2008; 9: 439– 48. Google Scholar CrossRef Search ADS PubMed  28 Miceli A, Murzi M, Gilmanov D, Fugà R, Ferrarini M, Solinas M et al.   Minimally invasive aortic valve replacement using right minithoracotomy is associated with better outcomes than ministernotomy. J Thorac Cardiovasc Surg  2014; 148: 133– 7. Google Scholar CrossRef Search ADS PubMed  29 Vriz O, Aboyans V, D’Andrea A, Ferrara F, Acri E, Limongelli G et al.   Normal values of aortic root dimensions in healthy adults. Am J Cardiol  2014; 114: 921– 7. Google Scholar CrossRef Search ADS PubMed  30 Sugawara J, Hayashi K, Yokoi T, Tanaka H. Age-associated elongation of the ascending aorta in adults. JACC Cardiovasc Imaging  2008; 1: 739– 48. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2018. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved. 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

Interactive CardioVascular and Thoracic SurgeryOxford University Press

Published: Apr 12, 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