Color Doppler US findings in the diagnosis of arterial occlusive disease of the lower limb

Transplant renal artery stenosis: utilization of machine learning to identify ancillary sonographic and doppler parameters to predict stenosis in patients with graft dysfunction

PURPOSE

To determine if ancillary sonographic and Doppler parameters can be used to predict transplant renal artery stenosis in patients with renal graft dysfunction.

MATERIALS AND METHODS

IRB-approved, HIPAA-compliant retrospective study included 80 renal transplant patients who had renal US followed by renal angiogram between January 2018 and December 2019. A consensus read of two radiologists recorded these parameters: peak systolic velocity, persistence of elevated velocity, grayscale narrowing, parvus tardus, delayed systolic upstroke, angle of the systolic peak (SP angle), and aliasing. Univariate analysis using t-test or chi-square was performed to determine differences between patients with and without stenosis. P values under 0.05 were deemed statistically significant. We used machine learning algorithms to determine parameters that could better predict the presence of stenosis. The algorithms included logistic regression, random forest, imbalanced random forest, boosting, and CART. All 80 cases were split between training and testing using stratified sampling using a 75:25 split.

RESULTS

We found a statistically significant difference in grayscale narrowing (p = 0.0010), delayed systolic upstroke (p = 0.0002), SP angle (p = 0.0005), and aliasing (p = 0.0024) between the two groups. No significant difference was found for an elevated peak systolic velocity (p = 0.1684). The imbalanced random forest (IRF) model was selected for improved accuracy, sensitivity, and specificity. Specificity, sensitivity, AUC, and normalized Brier score for the IRF model using all parameters were 73%, 81%, 0.82, and 69 in the training set, and 78%, 58%, 0.78, and 80 in the testing set. VIMP assessment showed that the combination of variables that resulted in the most significant change of the training set performance was that of grayscale narrowing and SP angle.

CONCLUSION

Elevated peak systolic velocity did not discriminate between patients with and without TRAS. Adding ancillary parameters into the machine learning algorithm improved specificity and sensitivity similarly in the training and testing sets. The algorithm identified the combination of lumen narrowing coupled with the angle of the systolic peak as better predictor of TRAS. This model may improve the accuracy of ultrasound for transplant renal artery stenosis.

A Toe Flexion NIRS assisted Test for Rapid Assessment of Foot Perfusion in Peripheral Arterial Disease: Feasibility, Validity, and Diagnostic Accuracy

OBJECTIVES

Feasibility, validity, and diagnostic accuracy of a non-invasive dynamic ambulatory test were assessed with near infrared spectroscopy (NIRS) evaluating foot perfusion in peripheral arterial disease (PAD).

METHODS

This was a prospective observational study. Eighty PAD patients (63 males, 71 ± 9 years), including 41 patients with coexisting diabetes, participated. Thirteen healthy subjects (8 males, 26 ± 8 years) were also studied by echo colour Doppler providing 160 diseased and 26 non-diseased limbs. Under identical clinostatic conditions, participants performed a 10-repetition toe flexion tests with NIRS probes on the dorsum of each foot; the area under the curve of the oxygenated haemoglobin trace ("toflex area") was calculated and the ankle-brachial index (ABI) was measured. Time of execution, rate of wrong tests, and adverse reactions were recorded. Within session reliability was assessed by administering the test twice, with a 5 minute interval between tests. The validity was assessed determining whether the toflex area was (a) dependent on the oxygen delivery from the lower limb arteries simulating PAD conditions by a progressive blood flow restriction (40-120% of systolic pressure) in healthy subjects; (b) consistent with the degree of PAD ranked by ABI and correlated with ABI and ankle pressure values in PAD patients. The diagnostic accuracy in detecting PAD was compared with examination using echo colour Doppler ultrasound.

RESULTS

All tests were rapidly, satisfactorily (<1% mistakes), and safely performed. Toflex area values, superimposable in the two sessions (intra-class correlation coefficient 0.92), were comparable to PAD values following blood flow restriction, consistent with PAD severity, correlated with dorsal pedis artery pressure (r = .21; p = .007) and ABI (r = .65; p < .001) in PAD, but not in the presence of diabetes. Toflex area was similar to echo colour Doppler for detecting PAD following receiver operating characteristic curve analysis (area = 0.987, p < .001; toflex area values ≤ -28 arbitrary units, sensitivity/specificity 95.6/100).

CONCLUSION

The toe flexion test enables ambulatory assessment of foot perfusion and PAD detection, even in the presence of non-measurable ABI or diseases affecting the microcirculation.

3-D MAGNETIC RESONANCE ANGIOGRAPHY VERSUS CONVENTIONAL ANGIOGRAPHY IN PERIPHERAL ARTERIAL DISEASE: PILOT STUDY

BACKGROUND

Angiography is usually performed as the preoperative road map for those requiring revascularization for lower extremity peripheral arterial disease (PAD). The alternative investigations are ultrasound, 3-D magnetic resonance angiography (3-D MRA) and computed tomography angiography. This pilot study aimed to assess whether 3-D MRA could replace the gold standard angiography in preoperative planning.

METHODS

Eight patients considered for aortoiliac or infrainguinal arterial bypass surgery were recruited. All underwent both imaging modalities within 7 days. A vascular surgeon and a radiologist each reported on the images from both the 3-D MRA and the angiography, with blinding to patient details and each others reports. Comparisons were made between the reports for the angiographic and the 3-D MRA images, and between the reports of the vascular surgeon and the radiologist.

RESULTS

Compared to the gold standard angiogram, 3-D MRA had a sensitivity of 77% and specificity of 94% in detecting occlusion, and a sensitivity of 72% and specificity of 90% in differentiating high grade (>50%) versus low grade (<50%) stenoses. There was an overall concordance of 78% between the two investigations with a range of 62% in the peroneal artery to 94% in the aorta. 3-D MRA showed flow in 23% of cases where conventional angiography showed no flow.

CONCLUSIONS

In the present pilot study, 3-D MRA had reasonable concordance with the gold standard angiography, depending on the level of the lesion. At times it showed vessel flow where occlusion was shown on conventional angiogram. 3-D MRA in peripheral vascular disease is challenging the gold standard, but is inconsistent at present.