The evaluation of the imaging of peripheral arteries for diagnostic purposeS comparing optical Coherence tomogrAphy and iNtravascular ultrasound imaging (SCAN) study was a two-center, prospective, non-inferiority study comparing the quality of IVUS and OCT imaging in the assessment of vessel characteristics for diagnosis and to support treatment strategy. The protocol was reviewed and approved by an institutional review board. All subjects signed an informed consent document prior to enrollment. The informed consent document provided to subjects followed guidelines outlined by Good Clinical Practices (GCP), the Declaration of Helsinki, and the International Conference on Harmonization (ICH). The study was registered on the National Institutes of Health website ClinicalTrials.gov (NCT03480685).
Subjects included in this study were adults ( >18 years old) with symptomatic peripheral arterial disease (Rutherford Class 2 or greater) that were scheduled for revascularization of a diseased vessel. Prior to any intervention, the subject was included if angiography determined that the reference vessel was of sufficient size to accommodate either imaging device and contained sufficient volume of plaque to treat with either atherectomy or other method of revascularization. Subjects were excluded only if female and pregnant or breast feeding or unwilling to give informed consent in order to provide plaque and other anatomic characteristics that reflect real-world cases. The procedures were performed on patients with common co-morbidities without acute hemodynamic instability.
Based upon the inclusion/exclusion criteria, twelve (12) subjects diagnosed with peripheral arterial disease of the lower extremities were enrolled, with the goal to acquire at least 120 matched images for analysis. With a sample size of 120 images by both the OCT and IVUS catheters, a two-group 0.05 one-sided matched t-test will have 90% power to reject the null hypothesis that the OCT and IVUS imaging modalities are not equivalent in favor of the alternative hypothesis that the two are equivalent.
Imaging Systems Used
All OCT images were captured with the Pantheris catheter (Avinger, Inc.) and all IVUS images were captured with the Visions PV 0.014P catheter (Royal Philips Corp.). Both devices have marketing clearance by the United States’ Food and Drug Administration (FDA) for use in diagnostic imaging within peripheral vessels. Neither device is considered investigational nor experimental for this indication and, in this study, both were used in accordance with their FDA- cleared indications for use.
The Pantheris catheter is a monorail 7 French device with a working length of 110 cm. It contains a 155 µm optical laser fiber on the shaft of the catheter utilizing frequency domain for image capture. The catheter is connected to a console that displays the images. The device was designed for atherectomy in peripheral vessels, but the OCT-imaging capability of the catheter allows the physician to assess the burden and location of plaque within a vessel, as well as vessel morphology, prior to tissue excision so that the physician can diagnose the patient’s condition and plan the treatment strategy. In addition, the imaging allows the physician to monitor tissue excision in order to avoid healthy tissue and then, using measuring tools in the system’s software, measure vessel lumen diameter to determine appropriate sizing for a balloon or stent to be used in adjunctive treatments. Blood management is accomplished by inflation of two balloons attached to the cannula of the catheter, one adjacent to the laser fiber and the other proximal to the laser fiber. The catheter is provided sterile and is for single use only.
The Visions PV 0.014P IVUS catheter (Visions catheter) has an outer diameter of 5 Fr and a working length of 150 cm. It also is advanced through an indwelling vascular sheath following an 0.014” guide wire. The catheter is connected with an imaging system that displays the vascular images. The IVUS catheter permits the clinician to assess disease markers, such as plaque burden and lesion location and morphology, as well as provide measurements of the lumen of the vessel. It is provided sterile and for single use only.
Image Acquisition
After providing informed consent, the subject was prepared for the diagnostic procedure according to the institution’s and investigator’s standard procedures. Demographic information was recorded for each subject and a radiopaque ruler (Glow N’ Tell, LeMaitre Vascular, Inc. Burlington, MA) was applied to the subject’s leg to provide reference marks for the starting and stopping points of the “target segment” to be imaged by the two devices. The ruler also provided reference points for the angiography cine.
Percutaneous access of the contralateral common femoral artery was obtained and a 7-French vascular sheath inserted up-and-over the aortic bifurcation and to the region of interest of the artery under fluoroscopic guidance. Next an 0.014 inch guide wire was advanced via the lumen of the sheath until its tip was distal to the target segment.
All images captured for review were contained within areas of interest in vessel segments containing disease or anatomical abnormalities within the markings of the radiopaque ruler. First an IVUS catheter was loaded onto the wire and advanced until its imaging transducer resided at the distal end of the target region (the starting point). Using the radiopaque ruler as a reference of the starting point (distal end) and stopping point (proximal end) of the target region, the IVUS catheter was energized for imaging and retracted by the physician through the target segment capturing images within that segment; no pull-back device was used. Following this imaging run, the IVUS catheter was again advanced to the distal point of the area of interest and a snapshot image of the vessel taken (Figure 2), not only as a known point for comparison but also for measurement of the luminal diameter and area. Next the IVUS catheter was retracted to the mid-point of the target segment, which was within the region of disease, and another snapshot taken, followed by withdraw to the proximal point of the run for the final snapshot. Using the software of the IVUS console, shortest and longest diameter and area of the lumen at these three points were measured and recorded. After all images were captured on the IVUS system, the IVUS catheter was removed.
Next, using the same radiopaque ruler markings as a reference, the OCT-imaging catheter was advanced over the 0.014 guide wire until its imaging window was at the same starting point (distal end) of the target segment that had been used by the IVUS catheter. With the device energized, the OCT catheter was retracted through the exact same target segment of the vessel and the OCT images captured on the Lightbox system on a cine loop. As was performed with the IVUS catheter, the OCT catheter then was re-advanced to the distal point of the lesion or area of interest and a snapshot taken of the vessel, followed by subsequent snapshots at the middle and proximal points of the lesion. Vessel diameter and luminal area were determined from these snapshots using the measuring software of the Pantheris system. When all imaging with the IVUS and OCT catheters was complete, the subject exited the study.
Data procurement addess less than 5 minutes to the expected procedural time and there were no adverse events as a result of capture of these images. The rate of manual pull-back of the OCT-imaging catheter averaged 23 seconds (SD 2.7 seconds) for one investigator and 27 seconds (SD 1.9 seconds) for the other. Over the same distances (proximal to distal capture endpoints), the IVUS catheters captured a mean of 1239 images (SD 298 images).
The IVUS images were downloaded from the system onto a DVD and transferred to a computer hard-drive for processing. The OCT images were captured on the Lightbox and downloaded to a memory stick prior to transfer to a computer hard-drive. The fluoroscopic images taken during each case were downloaded to a DVD by the hospitals’ radiology departments.
Study Endpoints
The primary efficacy of imaging endpoint was met when OCT imaging had the equivalent or higher ranking of IVUS imaging of the visualization of vessel morphology and disease for:
- Structure of the vessel wall – intima, media, external elastic lamina, and adventitia;
- Non-layered structure—presence of disease (plaque) in the vessel structure;
- Abnormal physiology— presence of calcification; and
- Obscuring of the image by artifacts.
The primary safety endpoints were freedom from diagnostic imaging procedure-related and device-associated adverse events at the time of the imaging, as reported by the physician.
Subject participation lasted the duration of the diagnostic imaging procedure. No follow-up images were collected after the procedure.
Image Analysis
The IVUS and OCT images were reviewed for suitability of use in the study by a consultant with over twenty years of experience in the interpretation of angiography and IVUS and OCT images of peripheral arteries. For each run of the target segments, the start, middle, and stop locations of the target segments in both the IVUS and OCT image files were identified from distinct points on the angiographic ruler and from the procedure notes. To gain additional images for review, IVUS and OCT images were matched to locations along the entire target segment to within 1 mm, using the fluoroscopic images as a reference. The imaging series resulted in over 1000 imaging pairs, but for each subject the matched images that best displayed the characteristic to be assessed were chosen for use in the study library.
The images were anonymized and given a unique identification number prior to assessment by the readers. Since the presentation of IVUS and OCT images are visibly different, the readers could not be blinded as to which imaging device was used to generate the image, but it was important that the images not be presented as matched images in sequence. Therefore, the images were provided in a random sequence so that the reader could not match a specific IVUS image to a specific OCT image at the time of the review. For example, Figure 3 shows IVUS and OCT images from the same location in a peripheral artery of one subject in the study. The IVUS image was identified in the series of images as number 228 and the OCT image was identified as image number 14. In this example, after viewing the OCT image, the readers would have viewed 214 other images between seeing the IVUS image from that same subject at that same point in the vessel, which reduced the chance that the reader would recognize the two images as coming from the same vessel location. This resulted in 120 images matched to each imaging catheter for a total of 240 images to be reviewed. Within these 120 images, the characteristic to be ranked appeared either singly on the image or in conjunction with other characteristics, which resulted in 309 scores from each reader and a total of 927 scores (Table 1).
The images were reviewed by three independent readers, who were experienced interventional radiologists. The readers were not present during the cases at which the images were captured nor were they associated, professionally or otherwise, with the hospitals at which the images were collected. All of the readers had experience in the interpretation of IVUS and OCT images. Prior to the image assessment stage, representative images were reviewed with the readers in order to create a consistent nomenclature when identifying different tissue types with both imaging catheters and to standardize the image review process.
The images were provided in eight blocks of 30 images and the readers ranked each for clarity (using a visual analogue scale or VAS) in the diagnosis of vessel condition and presence of plaque pre-procedure. Qualitative interpretation of OCT and IVUS images were ranked using the convention of Eberhardt et al. [11] as follows:
- Layered structure – intima, media, external elastic lamina and adventitia (ranked as 1- clear differentiation of all the vessel wall layers, 2- differentiation of at least three vessel wall layers, 3- differentiation of at least two vessel wall layers, and 4-no differentiation of any vessel wall layer);
- Non-layered structure —presence of plaque in the vessel structure (1- excellent histology-like image quality to 5- unacceptably poor image quality);
- Abnormal physiology—calcification (1- excellent histology-like image quality to 5- unacceptably poor image quality); and
- Effect of artifacts on reading the images (1- no artifacts present, 2- present and tolerable but not limiting diagnostic image quality, 3- intense and does limit diagnostic image quality).
The results of all readings of both imaging modalities were used for intrareader comparison to evaluate the reproducibility and reliability of the respective imaging modality.
Statistical Analyses
Statistical analyses were carried out using StatView software (SAS Institute, Cary, NC). A p value < 0.05 was considered statistically significant. Matched Pairs t-Test were performed to compare means and on paired samples to compare procedural data and complications associated with the imaging techniques. The Mann-Whitney-Wilcoxon test was applied for statistical comparison of the scores of all three readers. A two-way intraclass correlation (ICC) was calculated to assess intra-reader reproducibility.
Safety Reporting
Incidence and severity of procedure-related and device-related adverse events (e.g., vessel spasm, thrombosis, distal embolism, etc.) were evaluated following each scan and documented over the course of the study.