You are on page 1of 6

3404jum553-720 copy_Layout 1 3/17/15 10:09 AM Page 649

ORIGINAL RESEARCH

Relationship Between the Estimated


Glomerular Filtration Rate and Kidney
Shear Wave Speed Values Assessed
by Acoustic Radiation Force Impulse
Elastography
A Pilot Study
Flaviu Bob, MD, PhD, Simona Bota, MD, PhD, Ioan Sporea, MD, PhD, Roxana Sirli, MD, PhD,
Alina Popescu, MD, PhD, Adalbert Schiller, MD, PhD

Objectives—The aim of the study was to establish the relationship between the estimated
glomerular filtration rate (GFR) and kidney shear wave speed values assessed by
acoustic radiation force impulse (ARFI) elastography.
Methods—Our study included 104 patients with or without chronic kidney disease in
which the kidney shear wave speed was evaluated by ARFI elastography and correlated
with the estimated GFR. Five ARFI measurements were performed in the parenchyma
of each kidney. A median value expressed as meters per second was calculated.
Results—Five valid ARFI elastographic measurements were obtained in the right
kidney in all patients and in the left kidney in 97.1% of patients. The mean kidney shear
wave speed values ± SD in the right and left kidneys were similar: 2.17 ± 0.81 versus
2.06 ± 0.75 m/s (P = .30). The mean kidney shear wave speed decreased with the
decrease in the estimated GFR. Statistically significant differences were obtained only when
kidney shear wave speed values obtained in patients with an estimated GFR of greater than
90 mL/min/1.73 m2 were compared to values in patients with stage 4 (estimated GFR,
15–29 mL/min/1.73 m2) and stage 5 (estimated GFR, <15 mL/min/1.73 m2) chronic
kidney disease: 2.32 ± 0.83 versus 1.62 ± 0.75 m/s (P = .03) and 2.32 ± 0.83 versus 1.66
Received May 5, 2014, from the Departments of
± 0.72 m/s (P = .04), respectively. For a cutoff value of 2.26 m/s or lower, kidney shear
Nephrology (F.B., A.S.) and Gastroenterology and
Hepatology (S.B., I.S., R.S., A.P.), Victor Babeș wave speed had 86.7% sensitivity, 48.3% specificity, a 22.1% positive predictive value,
University of Medicine and Pharmacy, Timișoara, and a 95.6% negative predictive value (area under the receiver operating characteristic
Romania. Revision requested June 3, 2014. curve, 0.692; P = .008) for predicting the presence of an estimated GFR of less than 30
Revised manuscript accepted for publication July mL/min/1.73 m2 .
10, 2014.
Address correspondence to Simona Bota, Conclusions—Kidney shear wave speed values obtained by ARFI elastography decrease
MD, PhD, Department of Gastroenterology and with the decrease in the estimated GFR.
Hepatology, Victor Babeș University of Medicine
and Pharmacy, 2 Strada Intrarea Martir Angela Key Words—acoustic radiation force impulse elastography; chronic kidney disease;
Sava, 300742 Timișoara, Romania. glomerular filtration rate; genitourinary ultrasound; kidney shear wave speed
E-mail: bota_simona1982@yahoo.com

Abbreviations
ARFI, acoustic radiation force impulse;
CKD, chronic kidney disease; GFR, glomeru-
lar filtration rate
C hronic kidney disease (CKD) has had increasing incidence
and prevalence in recent years and is associated with high
morbidity and mortality and increased health costs. The best
method for assessing the progression of CKD should be renal
doi:10.7863/ultra.34.4.649 biopsy, but this method is invasive and is therefore associated with
some risks.1 Therefore noninvasive methods are preferred not only

©2015 by the American Institute of Ultrasound in Medicine | J Ultrasound Med 2015; 34:649–654 | 0278-4297 | www.aium.org
3404jum553-720 copy_Layout 1 3/17/15 10:09 AM Page 650

Bob et al—Glomerular Filtration Rate and ARFI Elastography

to identify patients with a high risk of CKD progression but The patients without kidney disease were selected from
also to examine the disease in longitudinal studies. patients hospitalized in various departments of our hospital
In addition to the widely used proteinuria and glomerular without histories of kidney diseases and with normal serum
filtration rate (GFR), new biomarkers have been studied but biological test results (creatinine and blood urea nitrogen)
are not yet validated for extensive clinical use (eg, neutrophil and also normal urinary test results (defined as absence of
gelatinase–associated lipocalin, kidney injury molecule 1, proteinuria and hematuria). Renal sonographic findings
and cystatin C).2 In addition to these markers, different were normal in these patients, and the difference in length
imaging methods could help in assessing renal disease pro- between the right and left kidneys was less than 15 mm.
gression. Conventional sonography can quantify only the The patients with various CKDs were patients who
size of the kidneys; therefore, an additional quantifiable received diagnoses and treatment in the nephrology
method for assessing the progression of CKD would be department. Patients undergoing hemodialysis or peri-
helpful.1 It is known that today for the staging of liver fibro- toneal dialysis, renal transplant recipients, and patients with
sis, different elastographic techniques are used.3–5 unilateral or bilateral hydronephrosis, kidney stones, or
Acoustic radiation force impulse (ARFI) elastography renal tumors were not invited to participate in the study.
using the Virtual Touch tissue quantification application Because this study was a pilot study we used the con-
(Siemens AG, Erlangen, Germany) is based on the prop- venience sampling technique for patient selection, which
agation of shear waves through soft tissues. The speed of allowed us to obtain basic data and trends regarding the
shear wave propagation is measured and is proportional to use of ARFI elastography in the kidney. All patients included
the stiffness of the tissue.4,6 Despite the extensive use of this in our study signed an informed consent form; the study
method for assessing liver fibrosis, data regarding its use in was approved by the local Ethics Committee and was in
kidney diseases are scarce. Acoustic radiation force impulse accordance with the 1975 Declaration of Helsinki.
elastography has already been used for assessment of the
kidney shear wave speed in renal transplant recipients.7–10 Renal Tests
The reproducibility of the method in nontransplanted kid- In all patients included in the study, serum biological tests
neys has been shown in a previous study.11 (creatinine and blood urea nitrogen) and urinary tests (for
It is known from the study of liver fibrosis that liver testing the presence of proteinuria and hematuria) were per-
stiffness increases with the progression of liver disease. formed on the same day as kidney shear wave speed meas-
Starting from these data, it can be hypothesized that simi- urements. For all patients, serum and urinary tests were
lar changes occur in the kidneys; thus, the progression of available also at the moment of the invitation to participate
kidney disease due to fibrosis as measured by the decrease in the study and repeated on the day of elastographic meas-
in the estimated GFR should lead to an increase in the kid- urements. The results of serum creatinine were used for
ney shear wave speed as measured by ARFI elastography. estimating the GFR by using the 4-variable Modification of
However, some very recently published data12,13 Diet in Renal Disease (MDRD4) formula.14
showed that the kidney shear wave speed decreases with
the decrease in the estimated GFR and is possibly affected Sonographic Examinations
by the renal blood flow.13 Thus, because not many data are For all patients, a sonographic examination was available
available regarding the relationship between the estimated before the invitation to participate in the study, but a new
GFR and kidney shear wave speed values in nontransplanted sonographic examination was performed in each patient
patients, we set out to determine whether kidney shear on the same day as the kidney shear wave speed measure-
wave speed values assessed by ARFI elastography can pre- ment by ARFI elastography, which was performed by one
dict the degree of the estimated GFR decrease and if this of the study team members (F.B. or S.B.). None of the
method could be used as an additional diagnostic test for patients had hydronephrosis, kidney stones, or renal
patients with CKD. tumors. The kidney length and renal parenchymal thick-
ness were recorded. For the patients with CKD, a renal size
Materials and Methods discrepancy of less than 15 mm between the kidneys was
not an exclusion criteria.
Patients
Our pilot study included patients with or without CKD in Acoustic Radiation Force Impulse Elastography
whom kidney shear wave speed was evaluated by ARFI Acoustic radiation force impulse elastography was per-
elastography between November 2012 and June 2013. formed in all patients with an Acuson S2000 ultrasound

650 J Ultrasound Med 2015; 34:649–654


3404jum553-720 copy_Layout 1 3/17/15 10:09 AM Page 651

Bob et al—Glomerular Filtration Rate and ARFI Elastography

system (Siemens AG) and Virtual Touch tissue quantifi- Five valid ARFI elastographic measurements were
cation version 2.0 software. For assessing kidney shear obtained in the right kidney in all patients and in the left
wave speed, a 4–9-MHz convex array probe was used. kidney in 101 of 104 patients (97.1%). Excessive depth was
The patient was positioned in the left lateral decubitus the reason for ARFI measurement failure in the left kidney
position. The region of interest, a “box” with a predefined in 1 of the 3 patients with valid elastographic measurements.
size (5-mm width × 10-mm length), was chosen in the mid- The mean kidney shear wave speed values obtained
portion of the kidney, in the renal parenchyma (containing in the right and left kidneys were similar in all patients
cortex and medulla). The main sonographic axis was set combined and also in patients without and with CKD as
parallel to the renal pyramid axis (perpendicular to the sur- separate groups (Table 2). Because the kidney shear wave
face of the kidney). speed values were similar in both kidneys in all patient
The kidney shear wave speed measurement was per- groups, further analyses were performed considering
formed with minimal scanning pressure while the patients the kidney shear wave speed values obtained in the right
were asked to stop breathing for a moment to minimize kidney.
breathing motion. In each patient, we aimed for 5 valid The mean kidney shear wave speed decreased with
measurements, and a median value was calculated and the decrease in the estimated GFR (Table 3 and Figure 2).
expressed in meters per second. The kidney shear wave The relationship between the kidney shear wave speed and
speed and depth measurement were displayed on the estimated GFR is graphically represented by a scatterplot
screen (Figure 1). The maximum depth at which ARFI in Figure 2. Statistically significant differences were
elastographic measurements can be performed is 8 cm. obtained only when kidney shear wave speed values obtained
If the measurement was not valid, X.XX was displayed on in patients with an estimated GFR of greater than 90
the screen. If 5 valid ARFI measurements could not be mL/min/1.73m2 were compared to patients with stage 4
obtained after 15 attempts, we considered that measure- (P = .03) and stage 5 (P = .04) CKD. The mean kidney
ment a failure. shear wave speed values in patients with an estimated GFR
of 60 mL/min/1.73m2 or greater were similar in patients
Statistical Analysis without (n = 58) and with (n = 10) kidney disease: 2.26 ±
The statistical analysis was performed with MedCalc ver- 0.81 versus 2.18 ± 0.88 m/s (P = .76).
sion 12.4.0 software (MedCalc, Mariakerke, Belgium). For a cutoff value of 2.26 m/s or less, kidney shear
The distribution of the numeric variables was tested by the wave speed had 86.7% sensitivity, 48.3% specificity, a
Kolmogrov-Smirnov test. Data were presented as mean 22.1% positive predictive value, and a 95.6% negative pre-
± standard deviation or median and range according to dictive value (area under the receiver operating character-
normal or non-normal distributions of the variables. istic curve, 0.692; P = .008) for predicting the presence of
The Student t test was used for group comparisons of con- an estimated GFR of less than 30 mL/min/1.73m2 (stages
tinuous variables with normal distributions; otherwise, the 4 and 5; Figure 3).
Mann-Whitney U test was applied. Qualitative variables
were presented as number and percentage. Areas under the
receiver operating characteristic curves were calculated for Figure 1. Kidney shear wave speed measurement by ARFI elastography.
the kidney shear wave speed to predict the presence of an
estimated GFR of less than 30 mL/min/1.73m2. The opti-
mal cutoff values were obtained by using the Youden index
(sensibility + specificity – 1) from the area under the curve
analysis. The group size and statistical analyses were per-
formed post hoc. P < .05 was considered significant.

Results

We studied 104 patients with or without CKD in which


the kidney shear wave speed was assessed by ARFI elas-
tography. The main characteristics of these patients are
presented in Table 1. The patients without and with CKD
were similar in age, sex, and renal size (Table 1).

J Ultrasound Med 2015; 34:649–654 651


3404jum553-720 copy_Layout 1 3/17/15 10:09 AM Page 652

Bob et al—Glomerular Filtration Rate and ARFI Elastography

Discussion In a study performed on children with vesicoureteral


reflux, the authors revealed that the kidney shear wave speed
There are little data regarding the use of ARFI elastography was increased in affected kidneys compared to unaffected kid-
for assessment of renal tissue to date. Most published studies neys.19 As already mentioned in “Materials and Methods,” in
were performed on transplanted kidneys.7–10 In nontrans- our study, patients with hydronephrosis were excluded. How-
planted kidneys, we have shown that the method has strong ever, in a recently published study performed on nontrans-
interobserver agreement.11 To proceed with the extensive planted patients, the results were quite similar to ours. In that
use of this method in patients with CKD, it is important to study, the shear wave velocity detected by ARFI was signifi-
prove whether there is any kind of relationship between cantly higher in healthy volunteers compared to patients with
renal function and the kidney shear wave speed assessed CKD, and it correlated with the estimated GFR.12
by ARFI elastography. These data suggest that also factors other than fibro-
In this study, we found that kidney shear wave speed val- sis influence the degree of tissue stiffness at the level of the
ues decreased with the decrease in the estimated GFR, and kidney. It is possible that kidney shear wave speed changes
there was a statistically significant difference between kidney reflect, in addition to fibrosis, the association of several
shear wave speed values in patients with an estimated GFR of renal microlesions and are probably also influenced by
greater than 90 mL/min/1.73m2 (with or without CKD) the renal blood flow because the kidney is a highly vascu-
compared to patients with stage 4 or 5 CKD (estimated GFR, larized organ.1,13 In an in vivo pig kidney, it was shown that
<30 mL/min/1.73m2). These changes are surprising but in after ligation of the renal artery, a decrease in elasticity was
line with data published by other authors12,13 showing that noted, whereas after ligation of the renal vein, a huge
the original hypothesis stated in the introduction was incor- increase in elasticity was measured.20 These experimental
rect. We presumed that, considering the fact that in other tis- data are in concordance with data presented by our group
sues (liver), fibrosis tends to increase tissue stiffness, the same in a case report of a patient with renal vein thrombosis, which
changes would be present in the renal tissue.15,16 showed an increased kidney shear wave speed (3.84 m/s
Regarding renal tissue, in a rat model of glomeru- for the kidney with renal vein thrombosis compared to
losclerosis, a progressive increase in intrarenal elasticity with 1.73 m/s for the contralateral kidney).21
the development of intrarenal disease was been found.17 Compared with a study by Guo et al,12 we tried to find
When comparing kidney shear wave speed values and the a kidney shear wave speed cutoff value to differentiate
degree of interstitial fibrosis in human renal transplant between patients with and without moderate and severe
recipients, there was either no correlation at all10,18 or a alteration of kidney function. In our study, we found that
moderate positive correlation.8 for a cutoff kidney shear wave speed value of 2.26 m/s or

Table 1. Patient Characteristics

Parameter All Patients Patients Without CKD Patients With CKD P


Age, y 57.3 ± 13.2 56.5 ± 12.8 58.4 ± 13.8 .47
Male, n (%) 61 (58.6) 33 (56.9) 28 (60.8) .84
Female, n (%) 43 (41.4) 25 (43.1) 18 (39.2) .84
Body mass index, kg/m2 27.7 ± 5.1 26.8 ± 4.4 29.9 ± 8.7 .02
Right kidney length, mm 105.2 ± 13.5 104.5 ± 11.8 106 ± 15.5 .58
Left kidney length, mm 106.3 ± 14.3 106.1 ± 11.7 106.5 ± 17.2 .91
Right kidney parenchymal thickness, mm 14 (8–21) 14 (10–21) 13 (8–20) .65
Left kidney parenchymal thickness, mm 14 (8–22) 14 (9–22) 13 (8–21) .22
Diagnosis, n (%)
No kidney disease 58 (55.8)
Diabetic nephropathy 16 (15.4)
Hypertensive nephrosclerosis 12 (11.5)
Chronic glomerulonephritis 9 (8.7)
Chronic pyelonephritis 3 (2.9)
CKD of unknown etiology 6 (5.7)
Serum creatinine, mg/dLa 0.9 (0.4–6.2) 0.8 (0.4–1.2) 1.4 (0.7–6.2) <.0001
Estimated GFR, mL/min/1.73 m2 81.3 (7.2–254.5) 96.9 ± 21.4 50.9 ± 24.4 <.0001
Data are presented as mean ± SD and median (range) where applicable.
aNormal values: 0.7 to 1.3 mg/dL.

652 J Ultrasound Med 2015; 34:649–654


3404jum553-720 copy_Layout 1 3/17/15 10:09 AM Page 653

Bob et al—Glomerular Filtration Rate and ARFI Elastography

less, there was good sensitivity and an excellent negative Another important limitation of the technique, as
predictive value for predicting the presence of stage 4 or 5 mentioned by other authors performing ARFI elastogra-
CKD (estimated GFR, <30 mL/min/1.73m2). However, phy in kidneys, is the fixed dimension of the region of inter-
despite the fact that kidney shear wave speed values est box,12 which makes the method difficult especially in
decreased with the decrease in the estimated GFR, there patients with decreased renal parenchyma. In patients with
were no statistically significant differences between the advanced CKD, because of the reduced cortical size, it is
values in patients with stage 3 CKD and those with an esti- possible that the region of interest set in the parenchyma
mated GFR of greater than 60 mL/min/1.73m2 (with or contains medulla as well as cortex, a fact that could repre-
without CKD). Unfortunately, we could not find any dif- sent a potential weakness of our analysis.
ference between the kidney shear wave speed in patients In conclusion, we found that kidney shear wave speed
with or without CKD when the estimated GFR was greater values estimated by ARFI elastography decreased with the
than 60 mL/min/1.73m2, although that finding would be decrease in the estimated GFR, and there was a statistically
useful in clinical practice. significant difference between patients with stage 4 and 5
It is possible that there are different patterns of kidney CKD compared to patients with other types of CKD and
shear wave speed changes in different types of renal diseases, patients with normal kidneys. We also found that for a cut-
an idea that should be investigated further. We also could off value of 2.26 m/s or less, the kidney shear wave speed
not make a correlation with renal biopsy results because
the number of patients with renal biopsy was low, but in a Figure 2. Relationship between kidney shear wave speed and esti-
recently published study by Wang et al,22 it was shown that mated GFR. The line represents the regression line.
the kidney shear wave speed had no correlation with indi-
cators of fibrosis in 45 patients with renal biopsy.
It should be noted that limitations of our study included
the quite small sample size, the convenience sampling
method used to select patients, the relatively small number
of patients with CKD, who had different underlying renal
diseases, and the lack of analysis of the renal blood flow,
which can influence renal elasticity according to previously
published studies.13,23,24 Another limitation was that the
group size and statistical analyses were performed post hoc.

Table 2. Comparison of Shear Wave Speed Values Obtained in the


Right and Left Kidneys

Shear Wave Speed, m/s


Patients Right Kidney Left Kidney P
Figure 3. Performance of kidney shear wave speed for predicting an
All 2.17 ± 0.81 2.06 ± 0.75 .30 estimated GFR of less than 30 mL/min/1.73 m2.
No CKD 2.26 ± 0.81 2.12 ± 0.70 .21
CKD 2.06 ± 0.79 2.02 ± 0.83 .84
Data are presented as mean ± SD.

Table 3. Mean Kidney Shear Wave Speed Values According to


Estimated GFR

Estimated GFR, Patients, Shear Wave


CKD Stage mL/min/1.73 m2 n Speed, m/s

1 or no CKD >90 40 2.32 ± 0.83


2 or no CKD 60–89 28 2.21 ± 0.85
3 30–59 21 2.18 ± 0.69
4 15–29 7 1.62 ± 0.75
5 <15 8 1.66 ± 0.72
Data are presented as mean ± SD where applicable.

J Ultrasound Med 2015; 34:649–654 653


3404jum553-720 copy_Layout 1 3/17/15 10:09 AM Page 654

Bob et al—Glomerular Filtration Rate and ARFI Elastography

could predict the presence of stage 4 or 5 CKD. Thus, ARFI 15. Bota S, Herkner H, Sporea I, et al. Meta-analysis: ARFI elastography ver-
elastography will not replace the use of the estimated GFR sus transient elastography for the evaluation of liver fibrosis. Liver Int 2013;
but has the potential to be a useful method for assessing 33:1138–1147.
patients with CKD. Future studies are needed to deter- 16. Nierhoff J, Chávez Ortiz AA, Herrmann E, Zeuzem S, Friedrich-Rust M.
mine a possible relationship between kidney shear wave The efficiency of acoustic radiation force impulse imaging for the staging
speed and renal disease progression. of liver fibrosis: a meta-analysis. Eur Radiol 2013; 23:3040–3053.
17. Derieppe M, Delmas Y, Gennisson JL, et al. Detection of intrarenal
References microstructural changes with supersonic shear wave elastography in rats.
Eur Radiol 2012; 22:243–250.
1. Grenier N, Gennisson JL, Cornelis F, Le Bras Y, Couzi L. Renal ultra- 18. Syversveen T, Brabrand K, Midtvedt K, et al. Assessment of renal allograft
sound elastography. Diagn Interv Imaging 2013; 94:545–550. fibrosis by acoustic radiation force impulse quantification: a pilot study.
2. Fassett RG, Venuthurupalli SK, Gobe GC, Coombes JS, Cooper MA, Transpl Int 2011; 24:100–105.
Hoy WE. Biomarkers in chronic kidney disease: a review. Kidney Int 2011; 19. Bruno C, Caliari G, Zaffanello M, et al. Acoustic radiation force impulse
80:806–821. (ARFI) in the evaluation of the renal parenchymal stiffness in paediatric
3. Bota S, Sporea I, Sirli R, Popescu A, Danila M, Costachescu D. Intra- and patients with vesicoureteral reflux: preliminary results. Eur Radiol 2013;
interoperator reproducibility of acoustic radiation force impulse (ARFI) 23:3477–3484.
elastography: preliminary results. Ultrasound Med Biol2012; 38:1103–1108. 20. Gennisson JL, Grenier N, Combe C, Tanter M. Supersonic shear wave
4. Palmeri ML, Wang MH, Dahl JJ, Frinkley KD, Nightingale KR. elastography of in vivo pig kidney: influence of blood pressure, urinary
Quantifying hepatic shear modulus in vivo using acoustic radiation force. pressure and tissue anisotropy. Ultrasound Med Biol 2012; 38:1559–1567.
Ultrasound Med Biol 2008; 34:546–558. 21. Grosu I, Bob F, Sporea I, et al. Assessment of renal vein thrombosis using
5. Friedrich-Rust M, Nierhoff J, Lupsor M, et al. Performance of acoustic renal acoustic radiation force impulse (ARFI) imaging in a systemic lupus
radiation force impulse imaging for the staging of liver fibrosis: a pooled erythematosus (SLE) patient: a case report. Paper presented at: 17th
meta-analysis. J Viral Hepat 2012; 19:e212–e219. National Conference of the Romanian Society of Ultrasound in Medicine
6. Nightingale K, Soo MS, Nightingale R, Trahey G. Acoustic radiation force and Biology; 2014; Timișoara, Romania.
impulse imaging: in vivo demonstration of clinical feasibility. Ultrasound 22. Wang L, Xia P, Lv K, et al. Assessment of renal tissue elasticity by acoustic
Med Biol 2002; 28:227–235. radiation force impulse quantification with histopathological correlation:
7. Stock KF, Klein BS, Vo Cong MT, et al. ARFI-based tissue elasticity quan- preliminary experience in chronic kidney disease. Eur Radiol 2014;
tification in comparison to histology for the diagnosis of renal transplant 24:1694–1699.
fibrosis. Clin Hemorheol Microcirc 2010; 46:139–148. 23. Warner L, Yin M, Glaser KJ, et al. Noninvasive in vivo assessment of renal
8. Stock KF, Klein BS, Vo Cong MT, et al. ARFI-based tissue elasticity quan- tissue elasticity during graded renal ischemia using MR elastography. Invest
tification and kidney graft dysfunction: first clinical experiences. Clin Radiol 2011; 46:509–514.
Hemorheol Microcirc 2011; 49:527–535. 24. Amador C, Urban, Kinnick R, Chen S, Greenleaf JF. Viscoelastic meas-
9. Ozkan F, Yavuz YC, Inci MF, et al. Interobserver variability of ultrasound urements on perfused and non-perfused swine renal cortex in vivo.
elastography in transplant kidneys: correlations with clinical-Doppler In: Proceedings of the 2010 IEEE Ultrasonics Symposium. Piscataway, NJ:
parameters. Ultrasound Med Biol 2013; 39:4–9. Institute of Electrical and Electronics Engineers; 2010:1530–1533.
10. Grenier N, Poulain S, Lepreux S, et al. Quantitative elastography of renal
transplants using supersonic shear imaging: a pilot study. Eur Radiol 2012;
22:2138–46.
11. Bob F, Bota S, Sporea I, et al. Factors which influence the reproducibility
of kidney stiffness assessed by means of acoustic radiation force impulse
(ARFI) elastography [abstract]. Eur J Ultrasound 2013; 34(suppl):S15.
12. Guo LH, Xu HX, Fu HJ, Peng A, Zhang YF, Liu LN. Acoustic radiation
force impulse imaging for noninvasive evaluation of renal parenchyma
elasticity: preliminary findings. PLoS One 2013; 8:e68925.
13. Asano K, Ogata A, Tanaka K, et al. Acoustic radiation force impulse elas-
tography of the kidneys: is shear wave velocity affected by tissue fibrosis or
renal blood flow? J Ultrasound Med 2014; 33:793–801.
14. Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D. A more accu-
rate method to estimate glomerular filtration rate from serum creatinine:
a new prediction equation. Modification of Diet in Renal Disease Study
Group. Ann Intern Med 1999; 130:461–470.

654 J Ultrasound Med 2015; 34:649–654