CT perfusion in peripheral arterial disease—hemodynamic differences before and after revascularisation

Hemodynamic assessment was performed as previously described in all patients prior to and after revascularisation [10]. Evidence of hemodynamically significant obstruction was established by non-invasive vascular testing, pulse volume recordings, and duplex ultrasound imaging studies. Maximum ankle-brachial pressure index (ABI) was recorded in a standardised way as reported previously [10]. Further details are given in S1 File.

All CTP examinations were performed as previously described [10]: A 128-slice CT scanner (Somatom Definition Flash, Siemens Healthineers) was used in shuttle mode, allowing for 28.2 cm of axial coverage and 2-s temporal resolution, which included almost all muscles of both lower legs in every subject. After placement on the gantry table, patients’ feet were fixated using tape to prevent spontaneous movement. The tube current was set to 100 mAs, the tube voltage to 100 kV(p). The duration of the CT perfusion scan was 60 s, with a rotation time of 2 s. CTP was started 15 s after injection of 60 ml of contrast medium (CM) (Ultravist 370, Bayer AG) at 5 ml/s. CM was injected into an antecubital vein by a dual-head pump injection device (Stellant D, Medrad), followed by a flush of 50 ml of NaCl at 5 ml/s. The collimation was 64 × 0.6 mm. The reconstruction increment was 10 mm at 10 mm slice width. Image reconstruction was performed with a 512 × 512 pixel matrix and medium smooth B30f kernel. For image post-processing and analysis, the reconstructed images of both legs were transferred to a commercially available computer workstation (Syngo Multimodality Workplace, Siemens Healthineers).

CTP parameters blood flow (BF; ml/100 ml*min⁻¹), blood volume (BV; ml/100 ml), and mean transit time (MTT; s) were calculated with a perfusion software (Syngo Volume Perfusion CT Body, Siemens Healthineers), based on the Patlak analysis. Since veins did not show sufficient CM enhancement in the late phase of the dynamic series and since venous enhancement was furthermore often degraded by pathologic flow within varicose veins in our patient cohort, no outflow curve was obtained. Thus, as previously described in the literature, we acted on a no outflow assumption [11, 15‐17]. Dataset motion correction and a noise reduction algorithm were applied automatically. Processing thresholds or segmentation tissue limits were − 50 HU and + 150 HU to exclude bone, vessel wall calcification, and other hyperdense material. Window width and centre for reference vessel input was + 300 HU and + 150 HU, respectively. The relative threshold for inside and outside was 50%, an adaptive smoothing filter was used. The vendor’s standard algorithmic parameters were applied. 3D colour-coded maps for BF, BV, and MTT were generated with a sequential two-compartment model. BF is defined as the amount of blood flowing through 100 ml of muscle tissue within 1 min. MTT is defined as the average time of contrast agent residence within the muscle tissue. BV is defined as the amount of blood within 100 ml of muscle tissue. BV can be expressed as proportion of the total volume of a dedicated voxel. For every patient, an individual arterial input function was determined by placing a region of interest (ROI) into the popliteal artery. A dedicated free-hand user-defined ROI was drawn around the lower leg muscles on every slice and adapted to their confines. Adjacent bones, vessels, and other soft tissue structures were excluded. All image evaluations were performed by two experienced radiologists in consensus with 8 and 12 years of experience.

Before the start of revascularisation, the length of the occlusion or stenosis was measured and classified on subtracted angiographic images (Allura Xper FD, Philips), as reported previously [10]. The degree of collateralisation was assessed by visual estimation on angiographic images. It was rated as good (1) if the estimated cross-sectional area of collaterals added up to > 50% of the pre-lesion diameter of the collateralized vessel, medium (2) if < 50%, and poor (3) if only faint or no collaterals were visible. The performing interventional radiologist had 20 years of experience in the field of revascularisation procedures.

Demographic, hemodynamic, angiographic, and CTP parameters in patients with stenosis and occlusion were compared by Mann-Whitney U test. Differences of CTP parameters between side of hemodynamically significant obstruction (treated limb), and the other side (untreated limb), as well between pre- and post-revascularisation were analysed by Wilcoxon signed-rank test. The correlation between CTP, hemodynamic, and angiographic parameters was analysed using Spearman’s correlation coefficient. Results were interpreted as almost perfect correlation between ± 0.81 and ± 1.00, substantial between ± 0.61 and ± 0.80, moderate between ± 0.41 and ± 0.60, fair between ± 0.21 and ± 0.40, thereunder no correlation [18]. A p value of < 0.05 was considered statistically significant. All analyses were performed using IBM SPSS Statistics™ 25.0.0 (IBM).

Twenty-four (83%) of our 29 patients (71 ± 11.37 years old, range 48–87 years) were referred for interventional revascularisation of an occlusion, five patients (17%) for a high-grade stenosis. The mean length of the lesion was 8.31 ± 7.62 cm (range 1–25). All lesions were located unilaterally in the common iliac, external iliac, or superficial femoral artery. ABP values indicative of media sclerosis were found in 6 subjects (5 occlusions, 1 stenosis). ABP and ABI values of these subjects were hence excluded from further analysis. Five patients had poor, 14 medium, and 10 patients a good degree of collateralisation.

The mean blood flow before and after revascularisation (BFpre and BFpost, respectively) was 7.42 ± 2.66 and 10.95 ± 6.64 ml/100 ml*min⁻¹, respectively (Fig. 1a). The mean blood volume before and after revascularisation (BVpre and BVpost, respectively) was 0.71 ± 0.35 and 1.25 ± 1.07 ml/100 ml, respectively (Fig. 1b). Post-intervention values of blood flow, blood volume, ABI, and ABP were significantly higher than respective pre-intervention values in the treated limb (p = 0.003, 0.02, < 0.001, and < 0.001, respectively), but not in the untreated limb (p = 0.64, 0.72, 0.36, and 0.57, respectively) (Table 1). ABI, ABP, and the degree of collateralisation were not correlating to any blood flow or blood volume parameter (p > 0.15 for all correlations).