Study Population
Thirty-eight CT pulmonary angiogram (CTPA) scans obtained at the University of Sheffield database (UK) were analyzed to evaluate the dimensions and metrics of the PA. The eligible patients were men or women over 18 years with a diagnosis of New York Heart Association (NYHA) class I-IV HF with reduced or preserved ejection fraction and had no contraindications to CTPA. From this, 34 CT scans were chosen for PA characterization analysis based on the quality of images and measurability at distal segments. The sample consisted of 53% males and 47% females.
CT Pulmonary Angiogram Methods
CT pulmonary angiogram scans acquired from the University of Sheffield were done on light-speed 64-slice MDCT scanners. The parameters of the imaging included 100 mA with automated dose reduction, 120 kV, pitch 1, with a rotation time of 0.5s and 0.625 collimation. The field of view used was 400 mm x 400 mm with an acquisition matrix of 512 x 512. For the imaging, 100 ml of Ultravist and Bayer IV contrast agents were used and administered at 5mL/s. High-resolution CT scans (HRCTs) were reconstructed using the contrast-enhanced acquisitions with 1.25 mm collimation from the apex of the lung to the diaphragm and the methods were carried out in the accordance with the relevant published guidelines.16
Analysis methods
A three-dimensional (3D) model of the PA was constructed through MIMICS medical imaging software using imported patient DICOM images. A “mask” was created in MIMICS to highlight the area of interest in the images. The “mask” was then split to help differentiate PA from extraneous highlighted segments.
The PA was studied in axial, coronal, and sagittal views to allow for more accurate measurements. In the axial view, axes were aligned such that the coronal axis ran parallel to the RPA wall and the sagittal axis lay perpendicular to the RPA wall. In the coronal view, axes were aligned such that the axial axis ran parallel to the RPA wall and the sagittal axis lay perpendicular to the RPA wall. In the sagittal view, the coronal and axial axes intersect at the centre of the circular RPA outline.
The PA metrics evaluated in this study were vessel diameter for both RPA and left pulmonary artery (LPA), the distance of the segment length from the zone where the implantable sensor would be placed in the vessels to the MPA bifurcation, the distance between the sensor and the sensor reader, or what’s called the link distance, the angle of the RPA downturn and the chest circumference of each patient.
The LPA and RPA were divided into 3 zones and the diameter of each zone was determined for analysis. Zone 1 (Proximal) was defined as the section on the RPA and LPA between the main PA bifurcation and the first branch of the RPA and LPA and proximal to the sensor deployment zone. Zone 2 (Sensor) was defined as the section distal to Zone 1 where the Cordella sensor (RPA) and CardioMEMS (LPA) are deployed. Zone 3 (Distal) was defined to be 2cm distal to Zone 2 (Fig. 2A). The diameter of each zone in the RPA was calculated as the average of the horizontal and vertical diameters of each zone which were measured along the axial axis in sagittal view and coronal axis in sagittal view, respectively. The diameter of each zone in the LPA was measured diagonally in between the sagittal and coronal sections in the axial view. For the LPA, only one diameter measurement was constructed instead of the average of two different diameters from two different planes respectively. The segment length from the MPA bifurcation to Zone 2 of each vessel was measured along the coronal axis in axial view (Fig. 3A). The chest circumference was also measured in axial view at PA level using a spline that went along the outer chest wall. The length of the spline was recorded as chest circumference. The link distance (LD) was recorded as the distance from the implantable sensor in each vessel to the skin surface (where a reader device will be located). The Cordella sensor was placed in the RPA at the downturn and CardioMEMS in the LPA. The link distance for the RPA was recorded as the distance from the sensor to the reader that is on the anterior chest surface. The link distance of the LPA was measured as the distance from the sensor to the closest point on the posterior surface of the back of the patient where they will be placed (Fig. 4A, 4C). The RPA downturn is defined as the location in the RPA downstream of the apical bifurcation where the interlobar artery typically turns downward and posterior before further branching into a series of basal arteries feeding the lower lung lobes. The angle of this downturn was measured for analysis (Fig. 5A).
The study was conducted in accordance with ASPIRE ((Assessing the Severity of Pulmonary Hypertension In a Pulmonary Hypertension REferral Centre)) code of ethics approved by Yorkshire and The Humber – Sheffield Ethics Research committee REC:16/YH/0352 and the processing of data complied with the terms of informed consent from the data subjects. The methods were carried out in accordance with the declaration of Helsinki.
Statistical Analysis
Statistical analysis was performed using Microsoft Excel (Microsoft Inc., Redmond, WA, USA). Through this, the mean, median, standard deviation, minimum and maximum were determined. The coefficient of variation (CV) was also calculated to help compare metrics of different nature with different units as it is a statistical measure of the relative dispersion of data points in a data series around a mean.