Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Techniques to Assess Tissue Perfusion in Peripheral Arterial Occlusive Disease
Introduction
Determination of lower limb tissue perfusion in patients with peripheral arterial occlusive disease (PAD) or critical limb-threatening ischemia (CLTI) can be challenging. In addition to the clinical examination, additional diagnostic tools, such as the ankle–brachial index (ABI), toe blood pressure, Doppler ultrasound, computed tomography (CT) angiography, magnetic resonance (MR) angiography, and digital subtraction angiography (DSA), are widely used. , However, these imaging techniques focus on the pre- and post-procedural blood flow in the macrovasculature and are not able to determine tissue perfusion or oxygenation of the microvasculature. However, impaired tissue perfusion is a major cause of rest pain, nonhealing ulcers, and infection, especially in patients with concomitant comorbidities such as diabetes mellitus or renal insufficiency. With local measurement of tissue oxygenation in peri-wound areas, healing potential and cause of the ulcer (arterial, venous, or a combination) may be determined more accurately.
Another role for tissue perfusion assessment is in the guidance of treatment of patients with PAD and CLTI. Currently, a completion angiography is performed at the end of an endovascular revascularization procedure to determine the technical success, which is mainly judged on the patency of the femoro-crural arteries and pedal circulation. A technically successful revascularization procedure of the major arteries, however, does not always lead to a clinically successful outcome, such as relief of rest pain and wound healing. Local tissue perfusion measurements may be a better predictor of successful revascularization and may guide the interventionalist during revascularization procedures, especially in multi-level obstructive disease in the femoro-crural arteries.
Several techniques have been introduced that enable tissue perfusion measurements, but most of these have not been broadly implemented in clinical practice so far. In this chapter we provide an overview of the currently available techniques to date using tissue perfusion techniques in vascular patients. An overview of the characteristics, indications, and pros and cons of each technique is summarized in Table 23.1 .
TABLE 23.1
Overview of Characteristics, Indications, and Pros and Cons of Invasive and Noninvasive Techniques to Assess Lower Limb Tissue Perfusion
| Type of Measurements | Application of Technique | Diagnostic Accuracy | Benefits | Limitations |
|---|---|---|---|---|
| Two-dimensional (2D) perfusion angiography | ||||
| Tissue perfusion measurements of a ROI in the lower extremity with DSA and post-processing software | Post-processing of DSA at the start and end of the revascularization procedure | Diagnostic accuracy and predictive value are unknown | No extra ionizing radiation or contrast agents except DSA runs Per-procedural data relevant to the technical success of the procedure |
Invasive technique Motion artifacts |
| Computed tomography (CT) perfusion imaging | ||||
| Tissue perfusion measurements of a ROI of the lower extremity with CT and post-processing software during revascularization or follow-up | CT perfusion imaging at baseline and/or after revascularization | Diagnostic accuracy and predictive value are unknown | High spatial resolution Easily accessible Short acquisition time Quantitative and qualitative analysis of perfusion in the limbs and feet |
Invasive technique Ionizing radiation Motion artifacts |
| Contrast-enhanced ultrasound (CEUS) | ||||
| Perfusion measurements of skeletal muscles of the lower extremity with microbubble contrast agents | CEUS imaging at baseline and/or after revascularization Used in addition to Doppler or duplex ultrasound |
Diagnostic accuracy and predictive value are unknown | No ionizing radiation Real-time visualization of muscle perfusion Easily accessible |
Invasive technique Operator dependent No stand-alone technique Limited transit time of microbubble agents Motion and bone artifacts |
| Magnetic resonance (MR) perfusion imaging (ASL, BOLD, DCE, IVIM) | ||||
| Perfusion measurements of skeletal muscles of the lower extremity with different MR imaging techniques, such as ALS, BOLD, DCE, and IVIM | Measurements at baseline and/or after revascularization | Diagnostic accuracy and predictive value are unknown | No ionizing radiation Possibility to use without contrast agents |
Invasive technique High costs Time consuming Not suitable in case of claustrophobia |
| Near-infrared fluorescence (NIRF) imaging with indocyanine green | ||||
| Skin perfusion measurements of the lower extremity with intravenous indocyanine green contrast agents | NIRF imaging at the start and end of the revascularization procedure | Sensitivity of 67%–100% and specificity of 72%–100% Predictive value is unknown |
No ionizing radiation Intraoperative and real-time visualization relevant to the technical success of revascularization |
Invasive technique Expensive imaging cameras Low penetration depth |
| SPECT/CT imaging | ||||
| Perfusion measurements of skeletal muscles of the lower extremity with radionuclide imaging | Measurements performed pre- and/or post-intervention | A cutoff value for high perfusion is 5.35 muscle-to-background ratio with an AUC of 0.92 based on MAE | High image quality | Invasive technique Ionizing radiation High costs Use of radioactive isotopes |
| Hyperspectral imaging (HSI) | ||||
| Skin perfusion measurements with visible light spectroscopy based on oxy- and deoxyhemoglobin concentration | Measurements pre- and/or post-intervention Monitoring of wound healing |
Sensitivity of 80%, specificity of 74%, and positive predictive value of 90% for diabetic foot ulcer healing | Non-invasive technique Fast Handheld Contact free |
Low penetration depth |
| Laser Doppler and laser speckle contrast perfusion techniques (LDPI and LSCI) | ||||
| Skin perfusion measurements using coherent laser light that interferes with the movement of red blood cells in the tissue | Measurements pre- and/or post-intervention | LDPI: Diagnostic accuracy and predictive value are unknown LSCI: Sensitivity of 92.3%, specificity of 75.0%, and positive predictive value of 80.0% for venous ulcer healing |
Noninvasive technique No ionizing radiation Short acquisition time |
Small measurement area Low penetration depth Measurements are affected by temperature and administration of vasoactive medication |
| Micro-lightguide spectrophotometry (O 2 C) | ||||
| Flow measurements of the skin using a combination of laser Doppler flowmetry and spectroscopy | Application at a fixed time pre- and post-intervention or continuously during the intervention | Diagnostic accuracy and predictive value are unknown | Noninvasive technique Fast Real-time visualization |
Small measurement area Variable penetration depth |
| Near-infrared spectroscopy (NIRS) | ||||
| Muscle oxygen saturation measured with red and near-infrared light using the absorption spectra of oxy- and deoxyhemoglobin in the lower limb | Measurements pre- and/or post-intervention or continuously during intervention | Diagnostic accuracy and predictive value are unknown | Noninvasive technique Easily applicable at different locations of the lower limbs and feet |
Differences in commercially available NIRS systems Artifacts due to adipose tissue |
| Skin perfusion pressure (SPP) | ||||
| Tissue perfusion measurements of the skin with a laser probe incorporated in a pressure cuff | Measurements pre- and/or post-intervention to predict wound healing or amputation | Sensitivity of 79.9% and specificity of 78.2% at a cutoff value of 30 mmHg for wound healing | Noninvasive technique Fast |
Cuff placement sometimes difficult and painful Motion artifacts |
| Transcutaneous partial pressure of oxygen (TcP O 2 ) | ||||
| Measurements of partial pressure of oxygen of the skin with a heated sensor-containing electrode | Measurements pre- and/or post-intervention to predict wound healing or amputation | A cutoff value <40 mmHg is associated with a 24% increased risk of wound complications after amputation | Noninvasive technique Low costs |
Time consuming Small measurement area Measurements affected by tissue edema, hair, increased oxygen consumption |
| Thermal imaging | ||||
| Measurement of skin temperature | Measurement pre and/or post-intervention | A decrease of 2°C in temperature had a sensitivity of 100% and specificity of 89% for failure of graft patency after bypass surgery | Noninvasive technique Contact free Fast |
Influence of external factors (e.g., surrounding temperature) on skin temperature |
ASL , arterial spin labeling; AUC , the area under the receiver-operator-curve; BOLD , blood oxygenation level-dependent; DCE , dynamic contrast enhanced; DSA , digital subtraction angiography; IVIM , intra-voxel incoherent motion; MAE , major adverse event; ROI , region of interest; SPECT , single photon emission computed tomography.
Invasive Techniques to Assess Tissue Perfusion
Two-Dimensional Perfusion Angiography
Two-dimensional (2D) perfusion angiography is a technique to determine perfusion of the lower limb and foot using digital subtraction angiography (DSA). These DSA images are analyzed with post-processing software, which can be performed on different workstations during revascularization procedures or post-intervention. The post-processing software converts the DSA images to color-coded images based on the change in pixel density over time. A time density curve (TDC) of contrast volume flow in the leg can be extracted to determine tissue perfusion. After that, a region of interest (ROI) of the feet can be manually selected in the images. A couple of parameters can be derived from the software representing the in- and outflow of contrast in the microcirculation over time. , An example of the images created with 2D perfusion angiography is shown in Figure 23.1 .
This technique has been explored in several prospective cohort studies in patients with PAD , , , and retrospectively , , , with the purpose to determine its feasibility. Different parameters can be used to determine the hemodynamic changes and improvement in tissue perfusion over time. , , , In three studies, 2D perfusion angiography parameters were significantly correlated with ABI, toe–brachial index, or skin perfusion pressure (SPP). , , The main outcomes of three prospective cohort studies and one retrospective study are summarized in Table 23.2 .
TABLE 23.2
Recent Publications for Tissue Perfusion Imaging Techniques
| Study [year] | Number of subjects | Study design | Measurement protocol | Main outcomes |
|---|---|---|---|---|
| Two-dimensional (2D) perfusion angiography | ||||
| Ikeoka (2020) | 33 patients with CLTI | Prospective cohort study | DSA at the start and end of the intervention with 2D angiography and SPP. The ROI was placed distally to the ankle and included the arterial foot arch. | Only the AT was significantly shortened after EVT in rest and during hyperemia. Only hyperemic AT was significantly correlated with the mean SPP both before and after EVT. |
| Ng (2019) | 47 patients with CLTI | Retrospective study | DSA at the start and end of the intervention with 2D perfusion angiography. Results were compared with ABI and TBI before and after EVT. An ROI was placed on the main run-off the pedal artery of the foot. ABI and TBI measurements were also performed. | Washout phase parameters showed a significant reduction in time required for contrast to decay to a specified percentage after peak. Percentages of contrast decay at specified time intervals after peak were increased significantly. The percentage of contrast decay 4 s after peak demonstrated the highest correlation coefficient with improvement in ABI or TBI. |
| Pärsson (2020) | 37 patients with CLTI | Prospective cohort study | DSA at the start and end of the intervention with 2D angiography. The ROI was placed between the tibio-talar joint and the midtarsal region, including part of the calcaneus. | A significant reduction in contrast AT and TTP was shown. A significant increased WiR was observed. |
| Reekers (2016) | 68 patients with CLTI | Prospective cohort study | DSA at the start and end of the intervention with 2D perfusion angiography. ROI was placed not lower than the middle cuneiform bone. | In most patients, perfusion angiography showed an increase in volume flow, an increase in both AUC and maximal PD, in the foot after successful angioplasty of the below-the-knee arteries. |
| CT perfusion imaging | ||||
| Cindil (2020) | 16 patients with CLTI | Prospective cohort study | CT perfusion imaging 1 to 3 days before and within 1 week after EVT. The TAC was obtained by placing a ROI on the posterior or anterior tibial artery at the ankle level of the untreated limb. The ROIs for analysis were placed on the dermis and muscle area of the sole of the foot and at the abductor halluces muscle of both feet. | The post-treatment BF and BV showed a statistically significant increase in the dermal ROI. The percentage change of BF and BV showed statistical correlations with ABI increase. Intra-observer agreement values showed excellent agreement. |
| Gao (2019) | 19 patients with CLTI | Prospective cohort study | Preoperative and postoperative (within 3 days after EVT) CT perfusion scans in supine position. The TDC was obtained by placing a ROI within the transverse section of the dorsal or plantar artery. The ROIs for tissue perfusion analysis were performed at several locations of the foot. | In the treated limb, the mean value of the BF and the average MSI significantly increased, and the average MTT, TTP, IRFt 0 and T max were significantly shortened. No statistical difference was found for BV and PS. |
| Sah (2019) | 35 patients with PAD | Prospective cohort study | CT perfusion scan immediately before EVT. Hemodynamic assessment was done with different noninvasive techniques such as ABI. The ROI for TAC was drawn onto the popliteal artery. The ROI for tissue perfusion analysis was manually selected around the calf muscles on every slice and adapted to their confines. | BV of the symptomatic limb was significantly different among ABI groups (ABI ≥0.70; 0.60–0.69; ≤0.59), with higher BV in patients with lower ABI. In the asymptomatic limb, there was no significant difference of BV among the ABI groups. |
| Contrast-enhanced ultrasound (CEUS) | ||||
| Duerschmied (2010) | 34 patients with PAD | Prospective cohort study | CEUS before and directly after revascularization and at 3 and 5 months of follow-up. The ROI was the area between the proximal and medial third of gastrocnemius and soleus muscle. ABI, PVR, and improvement in clinical stage (Fontaine) were also recorded. Patients were divided in an EVT and a bypass group. | In the EVT group, median TTP was significantly shortened after the intervention. At follow-up, median TTP decreased slightly in this group. In the bypass group, median TTP decreased significantly directly after and decreased slightly up to 3 to 5 months after surgery. No significant correlation between ABI and TTP was found. |
| Kundi (2017) | 13 patients with PAD and 8 healthy controls | Prospective cohort study | CEUS before and after treadmill exercise. ROI was the greatest circumference of the calf. | PP and TTP were significantly different between PAD patients and healthy controls after exercise but not in rest. Significant change in PP and TTP within controls before and after exercise but not in patients. |
| Meneses (2018) | 12 patients with PAD and 12 healthy controls | Prospective cohort study | CEUS before and after cuff occlusion and plantar flexion exercise. ROI was the medial gastrocnemius muscle of the largest leg circumference. | There were no significant differences between PAD patients and healthy controls at rest. Post-occlusion perfusion was lower in patients than in controls. Post-exercise perfusion was not different, but BV was higher in patients compared with controls. |
| MR perfusion imaging (ASL, BOLD, DCE, IVIM) ∗ | ||||
| Bakermans (2020) | 15 patients with PAD and 18 healthy controls | Prospective cohort study | BOLD MR imaging in supine position. Analysis was performed in the transverse plane outlining the gastrocnemius and soleus muscle. Measurements were done at baseline and after hyperemia, including ABI measurements | The relative amplitude of hyperemic perfusion response was significantly higher in PAD patients compared with healthy controls and showed a strong correlation with the ABI for both groups combined. |
| Galanakis (2020) | 10 patients with CLTI | Prospective cohort study | DCE-MRI perfusion imaging of the affected foot before and within 1 month after EVT. Follow-up included clinical examination, ABI, and DUS at 1, 3, 6, and 12 months. Multiple ROIs were placed in the sagittal plane around the entire foot, on the dermis and muscle tissue. Inter-observer reliability was calculated. | After PTA, perfusion parameters increased significantly. There was no significant correlation between perfusion parameters and ABI. Inter-observer reliability was excellent, with ICCs ranging from 0.91 to 0.95. |
| Grözinger (2014) | 10 patients with PAD | Prospective cohort study | PCASL MRI before and after revascularization. Measurements were done at baseline and after hyperemia (cuff occlusion). ROIs were placed in the transverse plane in the dorsal flexor compartment, the soleus muscle, and in the anterior compartment, the tibialis anterior muscle. ABI and pain-free walking distance were analyzed before and after PTA. | Hyperemic perfusion increased in the soleus muscle and increased significantly in the tibialis anterior muscle. Time to peak perfusion decreased in the soleus muscle and significantly in the tibialis anterior. A similar pattern was detected for the time of hyperemia. The mean ABI significantly increased. Pain-free walking improved in 8 patients. |
| Suo (2018) | 14 patients PAD or CLTI and 10 healthy controls | Prospective cohort study | ASL, BOLD, and IVIM cardiovascular MRI after 15 min of rest. Analyses were performed in the axial plane in the middle of the calf for the anterior and lateral compartment and the soleus and gastrocnemius muscle. | BOLD perfusion T2 ∗ values were significantly lower in patients with PAD compared with age-matched healthy controls. T2 ∗ was significantly lower in patients with CLTI compared with claudicant patients. |
| NIRF imaging with ICG | ||||
| Colvard (2016) | 93 patients with CLTI | Prospective cohort study | NIRF imaging with ICG before and immediately after revascularization. Plantar surface of the foot was measured. Measurements were compared with ABI, improvement of walking distance, and decrease in patient-reported claudication symptoms. | Mean ingress, egress, and peak perfusion of the plantar side of the foot increased significantly. Patients with clinically successful outcomes showed significant improvements in ABI, ingress, egress, and peak perfusion. The overall change in ABI significantly correlated with postoperative changes in ingress and peak perfusion. |
| Seinturier (2020) | 29 patients with PAD or suspected CLTI | Prospective cohort study | NIRF imaging with ICG of both feet consecutively, 15 min between two ICG administrations. ROI was determined for the whole foot and sub-areas. TBP measurement was performed as a reference. | No correlation between TBP and NIRF ICG parameters. In CLTI there was a correlation between the amplitude of ICG intensity on the forefoot and TBP. NIRF ICG parameters did not show any prognostic value for amputation, revascularization, or death. |
| Settembre (2017) | 101 patients with CLTI | Prospective cohort study | NIRF imaging with ICG before and immediately after revascularization. The dorsum of the foot was measured. ABI and TBP measurements were also performed. | The mean ingress and mean ingress rate increased significantly. |
| SPECT/CT imaging | ||||
| Alvelo (2018) | 42 patients with CLTI and 9 healthy controls | Prospective cohort study | SPECT/CT imaging was performed after an 8-hr fast and 15 min after injection of 99m Tc-tetrofosmin at the ankle and foot. In 8 healthy subjects and 6 patients, imaging was repeated 45 min after the tracer injection. Angiosomes of the foot were segmented. | Perfusion was significantly lower in patients with CLTI compared with healthy controls. Perfusion correlated with ABI for all participants but not for patients with CLTI alone. ICC showed excellent agreement of repeated measurements in both groups. |
| Hashimoto (2017) | 38 patients with suspected PAD | Retrospective study | SPECT/CT imaging was performed 15 min after injection of 99m Tc-tetrofosmin of the lower legs and feet. 3D VOI were selected from toes to knees. Data analysis was repeated to determine intra- and inter-observer reproducibility. | Patients with low perfusion had significantly more MAEs within 1 year. A multivariate analysis identified low perfusion as an independent prognostic factor for MAE. Linear regression analysis showed excellent reproducibility. |
| Hyperspectral imaging (HSI) | ||||
| Chiang (2017) | 150 patients with PAD and 20 healthy controls | Prospective cohort study | HSI was performed over the head of the first metatarsal on the plantar side in supine position. TcPO 2 measurements were performed afterward. | Oxyhemoglobin and oxygen saturation correlated significantly with the severity of disease. Deoxyhemoglobin and oxygen saturation correlated significantly with TcPCO 2 . Intra- and inter-operator reliability was excellent. |
| Grambow (2019) | 24 PAD, 25 patients without PAD, and 25 healthy controls | Prospective cohort study | HSI was performed at the plantar angiosome of the foot with patients in prone position. | HSI analysis revealed significantly reduced values for tissue oxygenation in patients with PAD. Tissue oxygenation did not correlate strongly or significantly with ABI. |
| Laser Doppler (LDPI) and laser speckle contrast-based perfusion imaging (LSCI) | ||||
| Humeau-Heurtier (2017) | 34 patients with PAD and 14 healthy controls | Prospective cohort study | LSCI was performed at the lower anterior part of the leg. TcPO 2 measurements were also performed. | The LSCI values of patients were significantly different compared with controls. LSCI significantly correlated with TcPO 2 . |
| Kikuchi (2019) | 31 patients with CLTI and 23 non-PAD controls | Prospective cohort study | LSCI before and immediately and 3 and 7 days after revascularization. ROI at the plantar side of the foot. | The LSCI value of the medial and lateral plantar surface was significantly increased immediately after the procedure and reached a maximum on day 7 after revascularization. The LSCI value was significantly lower for PAD limbs compared with non-PAD controls. |
| Pawlaczyk-Gabriel (2014) | 216 patients with PAD and CLTI and 27 healthy controls | Prospective cohort study | LDPI was performed after 15-min rest, during thermal stimulation, and post-ischemic hyperemia, at the dorsum of the foot between second and third metatarsal. TcPO 2 was also performed. | Skin blood flow was significantly lower in patients than in controls during thermal stimulation. Skin blood flow was significantly different between CLTI and controls and CLTI and claudicant patients. No significant differences in skin blood flow between patients and controls in rest. |
| Micro–lightguide spectrophotometry (O 2 C) | ||||
| Gyldenløve (2019) | 28 patients with PAD | Prospective cohort study | O 2 C measurements before and after treadmill test and before and after 12 weeks of exercise therapy. Measurements in supine position after 3 min of rest. ROI was the first toe of the index limb and contralateral limb. | Neither oxygen saturation nor flow was affected after a 12-week exercise program. Significant decrease after treadmill test for sO 2 , at the start and at 12 weeks of follow-up. |
| Rother (2017) | 30 patients with CLTI | Prospective cohort study | O 2 C measurements continuously during EVT. Three measurement points were defined: dorsal and plantar side of the foot, and lateral side of the ankle. A control probe was placed on the contralateral leg. ABI measurements were also performed. | The mean sO 2 showed a significant improvement after EVT. The overall flow parameter increased significantly after EVT. |
| Near-infrared spectroscopy (NIRS) | ||||
| Boezeman (2016) | 14 patients with CLTI | Prospective cohort study | Measurements were performed during EVT and 4 weeks later. In patients with arterial ulcers, the probe was placed 2 cm next to it, with a reference probe on the contralateral side. In patients without ulcers, the probes were placed on the dorsum of the feet. ABI and TBI values were also measured. | The mean StO 2 increased significantly after 4 weeks. The mean StO 2 was not significantly different directly before and after revascularization. ABI and TBI values showed no significant correlation with StO 2 values. |
| Boezeman, Boersma (2016) | 61 patients with CLTI and 30 age-matched control patients without PAD | Prospective cohort study | NIRS measurements before EVT at 4 spots: proximal vastus lateralis, distal vastus lateralis, proximal gastrocnemius lateralis, and distal gastrocnemius lateralis. The biceps were used as the reference spot. | Single rSO 2 was significantly lower for every measurement location between PAD and controls. The rSO 2 limb-to-arm ratios were significantly lower in PAD compared with controls at the proximal and distal gastrocnemius lateralis but not at the distal vastus lateralis. |
| Fuglestad (2020) | 40 patients with PAD and 10 healthy controls | Prospective cohort study | NIRS measurements continuously at baseline, after 1 minute of exercise, and during recovery period with a wireless, continuous-wave near-infrared spectrophotometer. The NIRS device was placed over the bilateral gastrocnemius muscle. | Baseline values were comparable between PAD patients and healthy controls. Patients with PAD reached minimum StO 2 significantly earlier than healthy controls. Change in StO 2 was significantly higher in patients with PAD at 1 min of exercise and at exercise minimum. |
| Khurana (2013) | 84 patients with PAD | Prospective cohort study | NIRS measurements using a continuous-wave spectrometer before, during, and after treadmill-walking test until pain caused patients to stop. The medial gastrocnemius muscle was used as the ROI. ABI and ankle systolic blood pressure (SBP) were measured before and 1 min after the treadmill test. | Patients were divided into two groups: SBP <50 mmHg post-exercise and SBP ≥50 mmHg. There were no differences in the decline in calf muscle StO 2 to a minimum value or in the time to reach minimum StO 2 . No significant correlation was found between StO 2 and ABI. |
| Skin perfusion pressure (SPP) | ||||
| Ikeoka (2020) | 33 patients with CLTI | Prospective cohort study | SPP on the dorsal and plantar side of the foot before and after EVT. | Dorsal and plantar SPP increased significantly after EVT. |
| Kimura (2019) | 76 patients with wounds | Prospective cohort study | Patients were classified according to the Society for Vascular Surgery WIfI classification, and SPP was performed after 15 min of rest. | Sensitivity of WIfI classification increased by using SPP, from 65% to 80%. Cutoff values for ischemia grade would be 0, >45 mmHg; 1, 44–35 mmHg; 2, 25–34 mmHg; 3, <25 mmHg. |
| Okamoto (2015) | 156 patients with CLTI | Prospective multicenter cohort | SPP was performed after EVT. | Mean pre- and post-EVT SPP was 28 and 46 mmHg, respectively, and correlated significantly with the 1-year amputation-free survival rate, MAEs, and wound healing. |
| Suzuki (2017) | 998 patients with wounds | Retrospective study | Patients who presented with a wound received SPP on the initial visit. | Patients with a SPP <30 mmHg had a significantly longer wound closure time, 235 days compared with 52 days in patients with SPP >50 mmHg. |
| Transcutaneous partial pressure of oxygen (TcPO 2 ) | ||||
| Andrews (2013) | 307 patients undergoing partial foot amputation | Retrospective, observational study | TcPO 2 was measured at two locations on the dorsum of the foot. Measurements were performed in rest and with the leg elevated or depended. | Sensitivity and specificity of 71% at a cutoff value of 38 mmHg for wound healing after 3 months. |
| Faglia (2007) | 564 patients with DM and CLTI | Retrospective study | TcPO 2 measurements before and after treatment of the ischemic leg on the dorsum of the foot. TcPO 2 was measured for 30 min. | A TcPO 2 of <34 mmHg indicates a need for revascularization. The risk of amputation is <1% with a TcPO 2 >50 mmHg. |
| De Graaff (2003) | 96 patients with suspected CLTI | Randomized controlled trial | Patients randomized for the new diagnostic strategy received TcPO 2 measurements at the first metatarsal at 44°C. A threshold of <35 mmHg was an indication for intervention. | The use of TcPO 2 did not lead to better patient outcome of reduction of interventions in patients with an uncertain diagnosis of CLTI. |
| Salaun (2019) | 455 patients with CLTI | Prospective cohort study | TcPO 2 was performed on the first inter-metatarsal space on the forefoot. Sensor heated to 44°C, determined after 15 min. | The amputation rate was twice as high for patients with a TcPO 2 of <10 mmHg compared with patients with TcPO 2 <30 mmHg. Sensitivity of 65% and specificity of 58% for TcPO 2 at 11 mmHg regarding amputation risk. TcPO 2 performed before revascularization could not predict the outcome of the procedure. |
| Thermal imaging | ||||
| Al Shakarchi (2019) | 25 patients with PAD | Prospective cohort study | Thermal imaging before and after bypass surgery. Pre- and 2 hours and 21 days postoperatively. Images were taken of the dorsal and plantar side of the feet. | The temperature increased postoperatively and was significantly associated with bypass patency. A decrease of 2°C had a sensitivity of 100% and specificity of 89%, a positive predictive value of 75%, and a negative predictive value of 100% for failure of graft patency. |
| Ilo (2020) | 40 patients with PAD | Prospective cohort study | Thermal imaging before and after revascularization. Pre- and 1 month postoperatively. Images were taken after 15–20 min of rest of the dorsal and plantar side of the feet. | The mean temperature increased significantly on the plantar side and on the dorsal side postoperatively. The difference between the vascularized and contralateral foot decreased significantly after treatment. |
| Wallace (2018) | 23 patients with PAD | Prospective cohort study | Thermal imaging of the hands and feet, and the temperature ratio between foot and hand was determined (tABI). Conventional ABI measurements were performed. | There was a strong and significant correlation between the temperature ratio and ABI. Bland–Altman analysis showed excellent agreement between tABI and ABI. |
∗ Outcome parameters of MRI perfusion imaging differ between BOLD, ASL, DCE and IVIM. ABI , ankle–brachial index; ASL , arterial spin labelling; AT , arrival time; AUC , area under the curve; BF , blood flow; BOLD , blood oxygenation level dependent; BV , blood volume; CLTI, critical limb threatening ischemia; DCE , dynamic contrast enhanced; DSA , digital subtraction angiography; DUS , Doppler ultrasound; EVT , endovascular treatment; ICG , indocyanine green; IRFt0 , impulse response function; IVIM , intra-voxel incoherent motion; MAE , major adverse events; MSI , mean slope of increase; MTT , mean transit time; PAD, peripheral arterial disease; PCASL , pseudo continuous arterial spin labeling; PD , peak density; PP , peak perfusion; PS , permeability surface; PTA , percutaneous angioplasty; PVR , pulse volume recording; ROI , region of interest; rSO2 , hemoglobin oxygen saturation; SBP , systolic blood pressure; sO2 , oxygen saturation; StO2 , oxygen saturation; T2∗ , T2 relaxation time; tABI , temperature ankle–brachial index; TAC , time attenuation curve; TBI , toe–brachial index; TBP , toe blood pressure; TDC , time density curve; Tmax , time to peak; TTP , time to peak; VOI , volume of interest; WIfI classification , wound, ischemia, foot infection; WiR , wash in rate.
This technique has some limitations, however. It relies on DSA and cannot be used for regular follow-up to monitor disease progression or ulcer healing. Besides, motion artifacts occur in up to 10% of patients. Standardization of the DSA protocol and region of interest (ROI) placement before and after endovascular treatment (EVT) is extremely important to capture any change in perfusion parameters. 2D perfusion angiography is an invasive technique, and at the moment, no validity or accuracy is determined. One of the ongoing studies to determine repeatability and inter- and intra-observer reliability of 2D perfusion angiography is the REPEAT study (see Key reference: Ipema et al.).
Computed Tomography Perfusion Imaging
Unlike 2D perfusion angiography, volumetric CT perfusion protocols consist of both CT angiography and CT perfusion imaging. With post-processing analysis, the changes in tissue perfusion parameters after intravenous injection of an contrast medium can be calculated. During post-processing, an ROI is selected, and a time-attenuation curve or TDC can be obtained. Perfusion parameters can be extracted based on the contrast density values. In most studies, CT perfusion imaging is performed before and within a pre-defined time interval after revascularization with a maximal follow-up time of 7 days. Moreover, CT perfusion has been shown to be feasible and reproducible in lower limb tissue perfusion detection, and changes in perfusion parameters are determined before and after EVT. , Sah et al . performed the largest study to date and included 35 patients who underwent CT perfusion imaging combined with ABI and duplex ultrasound measurements before EVT. The results of these studies are summarized in Table 23.2 .
The main benefit of CT perfusion imaging is the provision of a quantitative and qualitative analysis of the perfusion in the limbs and feet. Compared with MR imaging (MRI) perfusion, CT perfusion imaging provides a higher spatial resolution, is easily accessible, less expensive, and requires a shorter image acquisition time. However, CT perfusion, like 2D perfusion angiography, is sensitive to motion artifacts, especially in PAD patients with pain. In addition, it is an invasive technique that involves radiation exposure.
Contrast-Enhanced Ultrasound
Standard duplex ultrasound enables detection of stenosis or occlusions in the feeding arteries of the lower extremity. Contrast-Enhanced Ultrasound (CEUS) uses intravenously administered microbubbles to assess the microvascular blood flow. Microbubbles traverse the microcirculation and compress and expand under influence of an ultrasound wave. The ultrasound signal is amplified, and the microbubbles can be distinguished easily from blood. , After intravenous administration, the desired skeletal muscle can be imaged with the ultrasound transducer, and time to peak (TTP) intensity and peak perfusion can be determined. The diagnostic accuracy for CEUS measurements of muscle perfusion has not been determined.
A meta-analysis of patients with PAD compared with healthy individuals showed that CEUS was able to differentiate between these groups based on the TTP intensity. However, no correlation was found between CEUS and ABI. More recent studies using CEUS for muscle perfusion found no differences between PAD patients and healthy controls at rest, but significant differences in perfusion were observed after exercise or induced ischemia between these groups. , Limited data are available regarding determination of the change in tissue perfusion after revascularization. Up to now, Duerschmied et al. have suggested that a change in tissue perfusion can be detected directly after revascularization but not during follow-up. Results of the studies are summarized in Table 23.2 . Limitations of the technique are its invasiveness, heterogeneity in measurement protocols, its operator dependency, and the lack of correlation between CEUS and conventional diagnostics.
Magnetic Resonance Perfusion Imaging (ASL, BOLD, DCE, IVIM)
MRI can be used for detection of stenotic lesions without the image being substantially affected by calcification. For MR perfusion imaging, different techniques are used, such as arterial spin labeling (ASL), blood oxygenation level dependent (BOLD) imaging, intra-voxel incoherent motion (IVIM), and dynamic contrast-enhanced MRI (DCE-MRI). The working mechanism of ASL, BOLD, and IVIM MRI has been described in more detail in several studies. , , The advantage of ASL, BOLD, and IVIM MRI techniques is that they do not require exogenous contrast agents.
Versluis et al. showed a moderate to good reproducibility for DCE-MRI and BOLD MRI in 10 patients and controls, which was confirmed by Jiji et al. for DCE-MRI measurements on different days in 11 patients and 16 controls. Grözinger et al. performed ASL MRI and Galanakis et al. performed DCE-MRI before and after EVT and showed significant increases in perfusion parameters. Suo et al. integrated ASL, BOLD, and IVIM MRI and compared these techniques between patients with PAD and age-matched healthy volunteers. These findings were recently confirmed in a study using BOLD imaging. The results of these small prospective studies are summarized in Table 23.2 .
MR BOLD imaging seems the most reliable method to measure tissue perfusion in rest in PAD patients. , The application of this technique might be suitable as a supplement in the diagnosis of patients with calcified arteries and contraindications for the use of contrast agents. Disadvantages of MRI perfusion imaging are the high costs, risk of claustrophobia, and time-consuming measurements.
You're Reading a Preview
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here