Because the therapeutic strategies for AR and ATN differ, the RI, as a non-invasive ultrasound parameter, might provide little help in clinical decision making. Therefore, the uncertainty and controversy over using RI to diagnose kidney allograft AR prompted us to develop a better index. Acoustic radiation force impulse technology enables non-invasive quantification of tissue elasticity and is currently used in the assessment of liver stiffness in patients with chronic hepatitis and fatty liver; this method is closely dependent on histologic changes (Fierbinteanu-Braticevici et al. 2009; Friedrich-Rust et al. 2009). A clinical report in 2011 by Stock et al. included the ARFI-based tissue elasticity quantification for examination of kidney allograft dysfunction. The mean ARFI values exhibited an average increase of more than 15% in five acutely rejected kidneys, whereas no increase was observed in the other three dysfunctional kidneys, including two ATN cases and one drug-related toxicity case (Stock et al. 2011). In our study, we found a 16% significant increase in SWS in the acutely rejected kidneys compared with the non-AR kidneys, including stable functioning and ATN kidneys. Then, the subgroup analysis revealed that SWS distinguished the acutely rejected kidneys from the stable or ATN kidneys. These results suggest that SWS might be a reliable marker for kidney allograft acute rejection.
In this study, SWE effectively diagnosed AR regardless of when it developed. When AR occurs at a very early stage post-transplantation, edema in the kidney is an important pathologic change. When AR occurs more than 1 mo post-transplantation, interstitial fibrosis and glomerular sclerosis can be observed in the kidney allograft rather than a wide range of interstitial inflammatory cells, and the vascular R406 is milder than that in kidneys rejected early (Vogler et al. 2007). The significantly elevated SWS value in our study indicates that the kidney allograft becomes stiffer when acute rejection occurs. To establish a reference range of AR prediction, we chose two cutoff values to identify the absence (<2.23 m/s) and presence (>2.90 m/s) of renal allograft AR. With these two cutoffs, approximately 36% of patients could be correctly diagnosed, with >81% accuracy.
In our study, kidney volume was significantly increased in the AR group. It is reported that patients with proven AR exhibit swelling of the renal pyramids (Gao et al. 2011). The kidney is surrounded by a tough fibrous capsule; hence, our results imply that an increase in intrarenal pressure during AR might be the cause of increased stiffness. In addition, it has been proposed that acute tubular necrosis is an important cause of delayed graft function. Compared with that in the acute rejected kidney, the kidney volume in the ATN group was significantly decreased. Possibly, the swelling of the pyramids in ATN is temporary and decreases after regeneration of the injured tubular epithelial cells, exerting little influence on the increase in renal cortex stiffness. However, the swelling of the renal pyramids that occurs in AR is caused by the intense reaction initiated by the immune response. Without appropriate and timely anti-rejection therapy, AR may persist or even increase. This might be the reason that SWS was significantly higher in the AR group than in the ATN group.
In addition, compared with other organs such as the liver and kidney in situ, the transplanted kidney is a better target organ for p-SWE assessment. Most transplanted kidneys are located in the iliac fossa. Because of the more superficial and consistent location, high-resolution images of transplanted kidneys can be obtained. In addition, the structure of the renal cortex and medulla is easily recognizable, so selecting a ROI in the renal cortex is easier, while avoiding the renal medulla. A study from Jean-Luc\’s group found that the anisotropic ratio had strong variations depending on the medulla observed, but the anisotropic ratio within the cortex remained constant with the increase of pressure (Gennisson et al. 2012). When renal ultrasound elastography is performed, clear identification of the renal segments and their orientation with respect to the ultrasound beam and the resulting shear wave propagation is mandatory (Gennisson et al. 2012). To increase the reliability of measurement, as well as reduce the effect of tissue anisotropy, we chose the middle part of the transplanted kidney as the representative ROI. This method had good reliability in terms of less intra-observer and inter-observer variability, as described in our previous study (He et al. 2014). In contrast to our study, Syversveen et al. (2011) reported very high intra-observer and inter-observer variability; they measured SWS in the middle and lower poles of the kidney, and not in a consistent location, which might explain the high variability. Consistently using the same location rather than different locations for the assessment of intra-observer and inter-observer variability is more reliable (Krouskop et al. 1998). Furthermore, compared with the upper pole and inferior pole of the kidney, the acoustic beam is usually perpendicular to the renal cortex in the middle part of the kidney with relatively constant measurement depth, making the measurement more reliable.