68Ga-MAA labelling and quality control A volume of 7 mL 68GaCl3 eluted with 0.1 N HCl (from ABX D-01454 Radeberg, Germany) from a 68Ge/68Ga generator was passed through a PSH+ cartridge to yield 68Ga of high purity. 68Ga-chloride was added to 1.5 M 4-(2-hydroxyethyl) piperazine-1-82 ethanesulfonic acid (HEPES) buffer solution (from ABX D-01454), bringing the pH to 4-5. Since the prewash step was important before adding 68Ga to MAA particles in commercial kits, we conducted the labelling process by washing with saline the MAA particles in our study. After vigorous mixing with MAA (inside 5 mL sterile saline), the 68Ga-MAA suspension was incubated in a heat block at 90°C (7 min). The pH value of the 68Ga-MAA solution in the product vial was evaluated, and a 5-μL sample was titrated onto Whatman No. 40 paper (8 cm long, 1 cm thick) and passed through a mobile phase to calculate the RF values. Two different solutions were used as the mobile phase: methanol/ammonium acetate 10% (1:1) and acetone/glacial acetic acid (9:1). Radioactivity measurements were performed with a dose calibrator. The structure, morphology and chemical composition of 68Ga-MAA were analyzed by Scanning Electron Microscopy (SEM)/Energy Dispersive X-ray (EDX). The resulting 68Ga-labeled MAA particles were examined for structural degradation.
FIGURE 1: Flow-chart of rabbit experiment
Animal experiments The flow-chart of the rabbit experiment is shown in Fig. 1. This preclinical study was conducted following ethical approval from the Animal Experiments Ethics Committee (Jan 11, 2019; 60758568-020/2716). European Union directives were followed. The gender of the animal was not considered to be a factor in the experimental design. All rabbits were cared for and kept in separate cages with 12-h day and night cycles at 25°C during all procedures and fed ad libitum. Twelve New Zealand rabbits (Oryctolagus cuniculus) that weighed 2350-3000 g were divided into 2 groups: PE (n=6) and sham (n=6). In all rabbits, ketamine 35 mg/kg (Ketasol 10 ml, Richterpharma, Austria) and xylazine 5 mg/kg (Rompun 25 ml, Bayer, USA) were injected intramuscularly to induce general anaesthesia. After shaving the right femoral region of the rabbits, the skin was disinfected. Pulmonary embolism was induced in 6 rabbits via gelatin sponge (Galenaspon, Eucare Pharmaceuticalis Ltd, Chennai, India) plugs through the right femoral vein. For 6 rabbits, 5 mL of saline was injected into the right femoral vein, and these animals were included in the sham group. The right femoral vein was surgically exposed. An 18 gauge (green) Angiocath (Gloflon, Global Medikit Ltd, Delhi, India) was placed. From a 80x50x10-mm gelatin sponge, four 2 × 2 × 10-mm gelatin sponges per rabbit were cut under sterile conditions and softened in 5 cc saline. The first six rabbits (PE group) were injected with four 2 × 2 × 10-mm gelatin sponge plugs as in Yang et al. . The remaining 6 rabbits (sham group) were only injected with 5 mL of saline.
Ventilation SPECT/CT with Technegas Technegas was produced from a Technegas generator according to the manufacturer's instructions. After the rabbits were anesthetized, the Technegas was ventilated with the aid of a mouth mask modified from a baby mask. Technegas was passively ventilated to rabbits with the help of an AMBU on the device. After a ventilation period of approximately one minute, imaging was performed in the prone position.
SPECT/CT imaging was conducted using a SPECT/CT device equipped with a dual-head hybrid gamma camera and combined within the same gantry with a flat panel CT with a low dose X-ray tube (2.5 mA, 120 kVp; Philips Brightview XBT; Cleveland, OH, USA). SPECT/CT images of the thorax were obtained immediately after Technegas ventilation. CT images were generated using a 512×512 matrix with a slice thickness of 1 mm. The SPECT images were captured using a low-energy high-resolution collimator (140 keV photoelectric pic, 20% energy window). They were acquired using a 64×64 matrix (64 frames; 30 s per frame). Images in each plane were captured by rotating 360 degrees around the animals. The images were reconstructed using an Astonish filter and the filtered back projection method (cutoff, 3; order, 8) (AutoSPECTPro; Philips Intellispace Portal). Attenuation correction was performed using transverse, sagittal and coronal CT imaging. The SPECT and CT images were fused to generate SPECT/CT images.
68Ga-MAA PET/CT perfusion imaging Immediately after the ventilation study was completed, 68Ga-MAA PET/CT perfusion imaging was performed. The rabbits were anesthetized. Immediately after injection of 37 MBq 68Ga-MAA into the ear marginal vein, imaging was performed in the prone position. The rabbits were examined using a PET/CT scanner (Gemini TF TOF PET-CT; Philips, Cleveland, OH; 3D mode, slice thickness of 5 mm, 4x4x22 mm, 256x256 matrix, transverse FOV 576 mm and axial FOV 180 mm). Whole body emission scans were acquired for 2 minutes per position without intravenous contrast injection. Transmission images were obtained by low-dose CT (90 mA, 100 kV, 16 CT detectors, slice thickness of 5 mm). Attenuation correction was performed for PET images using CT maps. The ordered subsets-expectation maximization algorithm included 33 subsets and 3 iterations. PET images were reconstructed. Transverse, sagittal, and coronal sections (5-mm thickness) were created from PET/CT fusion images and evaluated using Philips Fusion Viewer software (ver. 2.1; Philips Healthcare, Best, The Netherlands).
V/Q scan evaluation for pulmonary emboli For each rabbit, the pulmonary perfusion PET/CT and ventilation SPECT/CT findings were assessed for each of the five lung lobes (right upper, middle, and lower lobes and left upper and lower lobes). Furthermore, 68Ga-MAA PET/CT images were blindly evaluated by a nuclear medicine physician. The nuclear medicine physician marked the perfusion and ventilation defects in a chart showing lung lobes for each rabbit. The EANM guidelines for ventilation/perfusion SPECT were used to diagnose emboli [12,13]. The presence of PE was defined as one or more wedge-shaped mismatch V/Q defects.
Histopathologic examination Pathological evaluation was accepted as the reference standard for this study. Rabbits were sacrificed with high-dose anaesthesia after lung V/Q imaging. The lungs and heart were removed in blocks. The materials were fixed with formaldehyde. Pathological slices were obtained from each lung lobe from the hilus to the periphery at 1-mm intervals. The sections were stained with hematoxylin and eosin. Pulmonary emboli were diagnosed when gelatin sponges were stained with HE in the pulmonary arteries.
Statistical analysis Data were analysed with SPSS 24.0 package software (IBM, Armonk, NY, USA). Continuous variables are presented as the mean ± standard deviation, whereas categorical variables are provided as frequencies and percentages. Pathological results were accepted as the reference. Sensitivity, specificity and accuracy values of combined 68Ga-MAA PET/CT perfusion and Technegas SPECT/BT ventilation imaging for PE were calculated.