Acoustic radiation force impulse imaging for noninvasive evaluation of renal parenchyma elasticity: preliminary findings

Abstract

Objective: To evaluate the diagnostic value of acoustic radiation force impulse (ARFI) to test the elasticity of renal parenchyma by measuring the shear wave velocity (SWV) which might be used to detect chronic kidney disease (CKD).

Methods: 327 healthy volunteers and 64 CKD patients were enrolled in the study. The potential influencing factors and measurement reproducibility were evaluated in the healthy volunteers. Correlations between SWV and laboratory tests were analyzed in CKD patients.?Receiver-operating characteristic curve (ROC) analyses were performed to assess the diagnostic performance of ARFI.

Results: The SWV of healthy volunteers correlated significantly to age (r = -0.22, P<0.001, n = 327) and differed significantly between men and women (2.06±0.48 m/s vs. 2.2±0.52 m/s, P = 0.018, n = 327). However, it did not correlate significantly to height, weight, body mass index, waistline, kidney dimension and the depth for SWV measurement (n = 30). Inter- and intraobserver agreement expressed as intraclass coefficient correlation were 0.64 (95% CI: 0.13 to 0.82, P = 0.011) and 0.6 (95% CI: 0.31 to 0.81, P = 0.001) (n = 40). The mean SWV in healthy volunteers was 2.15±0.51 m/s, while was 1.81±0.43 m/s, 1.79±0.29 m/s, 1.81±0.44 m/s, 1.64±0.55 m/s, and 1.36±0.17 m/s for stage 1, 2, 3, 4 and 5 in CKD patients respectively. The SWV was significantly higher for healthy volunteers compared with each stage in CKD patients. ARFI could not predict the different stages of CKD except stage 5. In CKD patients, SWV correlated to e-GFR (r = 0.3, P = 0.018), to urea nitrogen (r = -0.3, P = 0.016), and to creatinine (r = -0.41, P = 0.001). ROC analyses indicated that the area under the ROC curve was 0.752 (95% CI: 0.704 to 0.797) (P<0.001). The cut-off value for predicting CKD was 1.88 m/s (sensitivity 71.87% and specificity 69.69%).

Conclusion: ARFI may be a potentially useful tool in detecting CKD.

Figure 1. Volunteer underwent ARFI in prone position.

To avoid compression, the probe was contacted with body surface without pressure.

Figure 2. Measurement of SWV in the middle third of the renal parenchyma.

Figure 3. Bland-Altman plot for interobserver SWV in 40 healthy volunteers.

The difference of the observers is expressed as percentage deviation from the average of both observers (y axis), and the x axis represents the means of the two observers. The average interobserver difference is represented by the middle solid line, while the limit of agreement is represented by the two outer solid lines. The bias of the two observers is −4.7%, and the limit of agreement is between −48.9% and 39.5%. Two values lie outside the range.

Figure 4. Correlation analysis between SWV and age in 327 healthy volunteers.

Figure 5. Box and whiskers plot of SWV in different groups.

The tops and bottoms of the boxes are the first and third quartiles, respectively. The line through the middle of each box represents the mean. The error bars show the minimum and maximum values (range).

Figure 6. Receiver operating characteristic curves (ROC) estimated the diagnostic performance of ARFI.

For a cut-off value of 1.88 m/s, area under the ROC curve: 0.752 (95% CI: 0.704 to 0.797), Standard Error: 0.033, Significance level P<0.001. For a cut-off value of 1.88 m/s, ARFI had 71.87% Sensitivity (95%CI: 59.2 to 82.4), 69.69% Specificity (95%CI: 64 to 75), +LR: 2.37,?−LR: 0.4, for predicting CKD.