Genitourinar y Imaging Original Research

Beland et al.
Renal Function in Chronic Kidney Disease

Downloaded from www.ajronline.org by 39.225.134.21 on 05/24/15 from IP address 39.225.134.21. Copyright ARRS. For personal use only; all rights reserved

Genitourinary Imaging
Original Research

Renal Cortical Thickness

Measured at Ultrasound: Is It
Better Than Renal Length as an
Indicator of Renal Function in
Chronic Kidney Disease?
Michael D. Beland1
Nicholas L. Walle1
Jason T. Machan2
John J. Cronan1
Beland MD, Walle NL, Machan JT, Cronan JJ

Keywords: chronic kidney disease, estimated glomerular

filtration rate, renal cortex, renal failure, ultrasound
DOI:10.2214/AJR.09.4104
Received December 10, 2009; accepted after revision
January 21, 2010.
WEB
This is a Web exclusive article.
1
Department of Diagnostic Imaging, Rhode Island
Hospital, Warren Alpert Medical School of Brown
University, 593 Eddy St., Providence, RI 02903. Address
correspondence to M. D. Beland (mbeland@lifespan.org).
2
Departments of Biostatistics and Research, Orthopaedics,
and Surgery, Rhode Island Hospital, Warren Alpert Medical
School of Brown University, Providence, RI.

AJR 2010; 195:W146W149

0361803X/10/1952W146
American Roentgen Ray Society

W146

OBJECTIVE. The purpose of our study was to determine whether there is a relationship
between renal cortical thickness or length measured on ultrasound and the degree of renal
impairment in chronic kidney disease (CKD).
MATERIALS AND METHODS. From October to December 2007, 25 patients (13
men and 12 women, mean age 73 years) were identified who had CKD but were not on dialysis. The patients were from a single institution and had undergone renal ultrasound and at
least three serum creatinines within 90 days. The lowest creatinine was used for estimated
glomerular filtration rate (eGFR) calculation using both the Cockcroft-Gault (CG) and the
Modification of Diet in Renal Disease Study (MDRD) equations. Ultrasounds were consensus reviewed by three radiologists (2 attendings and a resident) blinded to specific renal function. Cortical thickness was measured in the sagittal plane over a medullary pyramid, perpendicular to the capsule. Length was measured pole-to-pole. Linear regression was used for
statistical analysis.
RESULTS. Mean cortical thickness was 5.9 mm (range, 3.211.0 mm). Mean length was
10 cm (7.212.4 cm). Mean minimum serum creatinine was 2.1 mg/dL (1.16.1 mg/dL).
Mean eGFR using CG was 34.8 mL/min (10.699.4 mL/min) and 36 mL/min (866 mL/
min) using MDRD. There was a statistically significant relationship between eGFR and cortical thickness using both CG (p < 0.0001) and MDRD (p = 0.005). There was a statistically significant relationship between CG and length (p = 0.003) but not between MDRD and
length (p = 0.08).
CONCLUSION. Cortical thickness measured on ultrasound appears to be more closely
related to eGFR than renal length. Reporting cortical thickness in patients with CKD who are
not on dialysis should be considered.

raditional teaching is that renal

length correlates with renal function in chronic kidney disease
(CKD), and therefore bipolar renal lengths are almost always reported at renal
ultrasound [1]. Previous studies have shown
that renal volume calculated at ultrasound is a
more exact measurement of a functioning kidney than renal length [2, 3]. A more recent
study showed that kidney length and volume
significantly correlated with estimated glomerular filtration rate (eGFR) in the elderly, but
kidney length has a low specificity in predicting renal impairment [4]. However, measuring
the true kidney volume at ultrasound is difficult. Estimates of volume can be made on the
basis of the ellipsoid formula, but this method
has an inherent defect because the kidney is
not actually ellipsoid [5]. In addition, the ellipsoid volume would include the central sinus fat

that does not contain functioning renal tissue

and does vary from patient to patient. These
factors likely contributed to the findings in a
recent study that showed a positive, but weak
association between sonographically determined kidney volume and various indices of
glomerular filtration rate (GFR) [6].
A recent study showed renal volume, and
specifically cortical volume, measured at CT
had a strong positive relationship with renal
function [7]. Additional studies have shown
total renal volumes obtained at CT relate
to renal function [8, 9]. However, there are
drawbacks to using CT, including increased
cost and radiation exposure. In addition, Widjaja et al. [8] showed a significant correlation between ultrasound-measured renal
length and CT-measured renal volume.
In patients with CKD, the renal cortical
echogenicity increases at ultrasound [10].

AJR:195, August 2010

Downloaded from www.ajronline.org by 39.225.134.21 on 05/24/15 from IP address 39.225.134.21. Copyright ARRS. For personal use only; all rights reserved

Renal Function in Chronic Kidney Disease

Fig. 175-year-old
woman with chronic
kidney disease.
Longitudinal ultrasound
image of right kidney
shows cortical
thickness measured
perpendicularly from
outer margin of kidney
to corticomedullary
junction (arrow).
Measurement is 0.46 cm.

In addition, the renal cortex often becomes

thinned [11]. Often this finding occurs with a
normal bipolar renal length and an increase
in the relative amount of central sinus fat. To
the best of our knowledge, a relationship between renal function and cortical thickness
has not been well established at ultrasound.
The purpose of our study was to determine
whether there is a relationship between renal
cortical thickness or length measured at ultrasound and the degree of renal impairment
in CKD using two widely accepted computational methods of estimating GFR.
Materials and Methods
Patient Selection
This retrospective study was approved by
our institutional review board and is HIPAA
compliant. A search was performed of our hospital
electronic medical record for patients with the
clinical diagnosis of CKD who were not on dialysis
and also had undergone renal ultrasound over a
3-month period from October 2007 to December
2007. This search was further narrowed to patients
who had at least three serum creatinines and weight
recorded within 90 days of the ultrasound. Patients
with hydronephrosis were excluded. Twenty-five
patients (13 men and 12 women, mean age, 73
years; age range 2690 years) met these criteria
and constitute the study population.

Estimation of Renal Function

The lowest creatinine performed within 90 days
of the ultrasound was used for eGFR calculations.
The lowest creatinine was chosen because it
represents the best recorded renal function during
the study period and helps to minimize the
influence of superimposed acute on chronic renal
insufficiency [10]. The Cockcroft-Gault (CG)
and the Modification of Diet in Renal Disease
Study (MDRD) equations were used for eGFR
calculation, as follows [12]:
The CG equation is eGFR = (140 age)
(Weight in kg) (0.85 if female) / (72 Cr) where
Cr is creatinine.
The equation for Modification of Diet in
Renal Disease (MDRD) for isotope dilution
mass spectrometry (IDMS)-traceable creatinine
measurements is GFR (mL/min/1.73 m2) = 175
(Scr) 1.154 (Age) 0.203 (0.742 if female) (1.212
if African American) (conventional units). Scr is
serum creatinine.

Ultrasound Interpretation
All renal ultrasound studies were performed in
the inpatient setting at our tertiary care hospital.
The examinations were performed using standard
gray-scale B-mode imaging with a 3.5-MHz

AJR:195, August 2010

curvilinear transducer (Logiq 9, GE Healthcare).

The examinations were retrospectively reviewed
at a PACS workstation (Centricity, GE Healthcare)

by three authors (a radiology resident (PGY-4) and

two radiology attending physicians specializing in
ultrasound with 3 and 27 years of experience). All

TABLE 1: Study Population

Patient
No.
Sex

Minimum Maximum
Mean
Age Creatinine Creatinine Creatinine MDRD
Mean Renal Mean Cortical
(y)
(mg/dL)
(mg/dL)
(mg/dL) eGFR CG eGFR Length (cm) Thickness (cm)

43

6.1

11.5

8.63

7.53

10.72

9.8

0.56

90

1.2

1.8

1.42

56.91

43.29

10.7

0.86

83

2.7

7.1

4.67

22.70

16.51

10.15

0.61

69

3.5

4.1

3.89

12.95

21.57

9.7

0.70

84

2.1

3.1

2.59

22.45

19.43

9.7

0.56

68

1.7

1.9

1.82

29.93

25.43

7.15

0.39

59

1.8

3.8

2.76

38.77

55.79

9.55

0.82

65

1.2

1.9

1.60

45.16

51.03

11.85

0.77

9.45

0.64

85

1.1

3.3

2.07

47.24

35.67

10

26

1.7

7.3

3.25

48.96

99.38

12.4

1.10

11

82

1.5

5.6

3.39

44.78

39.07

11.25

0.64

12

90

1.2

2.6

1.79

42.17

21.81

9.5

0.46

13

83

1.5

2.7

2.15

44.67

56.80

11.45

0.75

14

74

1.6

2.6

2.03

42.42

46.01

10.1

0.86

15

75

1.9

2.9

2.42

25.76

25.33

8.5

0.47

16

84

2.8

3.4

3.16

16.10

18.50

7.7

0.28

17

78

1.8

6.4

3.97

27.23

37.29

9.55

0.49

18

83

1.3

2.0

1.53

52.68

40.77

11.9

19

75

1.7

3.1

2.53

39.45

33.78

10.15

0.46
0.54

20

70

5.7

6.6

6.20

7.35

12.33

10.9

0.34

21

76

2.6

5.7

3.92

24.12

19.01

8.45

0.41

22

87

1.2

3.2

1.77

57.29

48.06

8.75

0.77

23

42

1.2

4.6

2.77

66.33

45.73

10.65

0.67

24

85

1.1

2.2

1.68

47.21

34.00

10.0

0.61

25

69

2.9

3.5

3.24

21.68

17.85

10.0

0.38

NoteeGFR = estimated glomerular filtration rate, MDRD eGFR = Modification of Diet in Renal Disease study
for isotope dilutionmass spectometry traceable creatinine measurements, CG eGFR = Cockcroft-Gault
equation. See Materials and Methods section for calculation of MDRD eGFR and CG eGFR.

W147

100

100
y = 46.8 + 8.2x, p = 0.0029
r 2 = 0.3

CG

y = 13.5 + 80.6x, p = 0.0001

r 2 = 0.66

CG

measurements were made by consensus agreement.

Renal lengths were measured as the greatest poleto-pole distance in the sagittal plane. The renal
cortical thickness was measured in the sagittal
plane at the level of the mid kidney as described
by Moghazi et al. [11]. The measurement was taken
over a medullary pyramid, perpendicular to the
capsule as the shortest distance from the base of
the medullary pyramid to the renal capsule (Fig. 1).
The readers were blinded to specific renal function,
additional imaging, or any additional clinical
information at the time of image review. Cortical
thickness and length were measured bilaterally.

50

0
0

0.5

0
6

1.5

Results
Details regarding the study population are
given in Table 1. The mean cortical thickness was 5.9 mm (range, 3.211.0 mm). The
mean length was 10 cm (range, 7.212.4
cm). A statistically significant positive relationship was observed between eGFR and
mean cortical thickness using both the CG
and the MDRD equations (CG, p < 0.0001;
MDRD, p = 0.0050). There also was a statistically significant relationship between CG
eGFR and mean renal length (p = 0.0029)
but not MDRD eGFR (p = 0.0756) (Fig. 2).
The strongest relationship, as evidenced by
the highest r 2 value, was for mean cortical
thickness and CG eGFR (r 2 = 66%).
Discussion
Our series showed a statistically significant
relationship between cortical thickness measured at ultrasound and renal function in patients with CKD. Although there was also a
significant relationship between CG eGFR and
renal length, there was not for MDRD eGFR.
Renal length has traditionally been considered a surrogate marker of renal function because renal length decreases with decreasing
renal function. Renal lengths are universally
reported and are usually the only measurements given at renal ultrasound [1]. However,
on the basis of our study, it appears that cortical thickness measured at ultrasound may be
related more closely to eGFR than renal length
in patients with chronic renal failure.

W148

10

12

14

100
y = 8.7 + 4.5x, p = 0.0756
r 2 = 0.13

MDRD

y = 9.1 + 44.7x, p = 0.0050

r 2 = 0.29

50

0
0

Mean Length (cm)

100

Statistical Analysis
The data were entered and stored on a spreadsheet (Excel, Microsoft). Mean cortical thickness
and length were used in analyses. Statistical
analysis and visualization were performed using
Matlab (MathWorks). The relationship between
ultrasound measurements and renal function was
tested using linear regression. Significance was
considered at a p value < 0.05.

50

Mean Cortical Thickness (cm)

MDRD

Downloaded from www.ajronline.org by 39.225.134.21 on 05/24/15 from IP address 39.225.134.21. Copyright ARRS. For personal use only; all rights reserved

Beland et al.

0.5

1.5

Mean Cortical Thickness (cm)

50

0
6

10

12

14

Mean Length (cm)

Fig. 2Graphs show Cockcroft-Gault (CG) (top) and Modification of Diet in Renal Disease (MDRD) study
(bottom) estimated glomerular filtration rate (eGFR) plotted as function of mean cortical thickness (left) and
mean length (right) with best-fit straight line from regression.

As the burden of CKD continues to increase, efforts to reduce the cost of monitoring and managing this disease are needed. Our
study attempted to evaluate the usefulness of
a generally obtainable measurement at ultrasound in the setting of CKD as a correlate to
kidney function (eGFR). Prior studies also
have evaluated imaging measurements as surrogate markers of renal function. A study evaluating 69 patients with suspected unilateral renal artery stenosis showed renal volume was a
better predictor of single-kidney GFR than renal length. They also showed the addition of renal area and parenchymal thickness measured
at ultrasound to length was a better predictor
of both single-kidney GFR and renal volume
than length measured at ultrasound alone [8].
Another study showed a correlation between
eGFR and renal volumes measured at ultrasound in 116 healthy children [13]. Other authors have described kidney volume as a better predictor of renal function than renal length
[2, 3]. This was further supported by a study in
2009 by Sanusi et al. [6] showing a weak but
positive correlation between kidney volume
and various indices of GFR, best with measured creatinine clearance, in 40 patients with
CKD. Their results also showed a significant
correlation with the measured creatinine clear-

ance and the CG and MDRD equations, further

validating these estimates of GFR in CKD [6].
Our study is limited by the small study
sample. We hope the results presented here
will serve as a pilot study prompting further studies with larger patient samples to
validate the results. Future areas of investigation using larger patient samples may include development of a predictive range of
renal function given a particular cortical
thickness. Alternatively, a determination of
a threshold cortical thickness above which
renal function is preserved may be identified. Because of the retrospective design of
our study, measurements were made on the
images after they were obtained. Renal cortex measurements were taken perpendicular
to the renal capsule from the capsule to the
corticomedullary interface. This interface
can be difficult to identify in some patients
in whom there is poor corticomedullary differentiation. To ensure accuracy, these measurements ideally would be made prospectively at the time of the examination. This
also would allow real-time image optimization to possibly make the corticomedullary
interface more apparent. Additional measurements in the transverse plane or an average cortical thickness including the upper

AJR:195, August 2010

Downloaded from www.ajronline.org by 39.225.134.21 on 05/24/15 from IP address 39.225.134.21. Copyright ARRS. For personal use only; all rights reserved

Renal Function in Chronic Kidney Disease

and lower poles may prove to be a better representation of functioning parenchymal volume in future studies. Despite these limitations, we found the cortical thickness was
usually easy to measure on the PACS workstation. The method used in our study also
likely reflects common practice in which ultrasound images are obtained by a technologist and then interpreted by the radiologist
after the patient has left the department.
Another potential limitation of our study
is the use of computational estimates of renal function, rather than measured GFR. Although both formulas have been previously
validated and are widely used, there is continued debate over which formula is best [14, 15].
A study by Rule et al. [16] in 2004 showed that
the MDRD equation systematically underestimated renal function. The underestimation
was much more pronounced for healthy volunteers (29% underestimation) than in patients
with CKD (6.2% underestimation). However, their study was performed with a version
of the MDRD equation before standardization of creatinine estimates between laboratories. The development of the IDMS-traceable MDRD study equation that was used in
our study has allowed the use of standardized
creatinine measurements and should minimize measurement differences [12]. A slightly more recent study directly comparing CG
and MDRD estimates of GFR in patients with
CKD showed the MDRD equation to be more
accurate than the CG equation in patients with
moderate to advanced kidney disease and diabetic nephropathy [17]. Our study only evaluated patients with CKD, which would minimize any potential underestimation of GFR
using the MDRD equation relative to patients
without known kidney disease and suggests
MDRD may be the more accurate estimate
applied to our study population [16].
Patients on dialysis were necessarily excluded from this study. Examining the relationship
between renal function on the basis of serum
creatinine and cortical thickness would be inherently flawed in this group because the creatinine used for calculation would be a measure of

AJR:195, August 2010

dialysis efficacy rather than native renal function. Patients without known kidney disease
also were not included in this study. It would
be interesting to see if the correlation between
cortical thickness and eGFR is also identified
in these patients. Although this would be a
more difficult study to perform, evaluating the
applicability of measuring cortical thickness to
healthy kidneys would be valuable.
In summary, we have shown renal cortical
thickness measured at ultrasound appears to
relate to the degree of renal impairment in
patients with CKD, and routine reporting of
cortical thickness should be considered in
such patients who are not on dialysis.
References
1. American College of Radiology Website. ACR
practice guideline for the performance of an ultrasound examination of the abdomen and/or retroperitoneum (in collaboration with the American
Institute of Ultrasound in Medicine AIUM).
www.acr.org/SecondaryMainMenuCategories/
quality_safety/guidelines/us/us_abdomen_retro.
aspx. Revised 2007. Accessed April 14, 2010
2. Emamian SA, Nielson MB, Pedersen JF, Ytte L.
Kidney dimensions at sonography: correlation
with age, sex, and habitus in 665 adult volunteers.
AJR 1993; 160:8386
3. Jones TB, Riddick LR, Harpen MD, Dubuisson
RL, Samuels D. Ultrasonographic determination
of renal mass and renal volume. J Ultrasound Med
1983; 2:151154
4. Van Den Noortgate N, Velghe A, Petrovic M, et
al. The role of ultrasonography in the assessment
of renal function in the elderly. J Nephrol 2003;
16:658662
5. Bakker J, Olree M, Kaatee R, et al. Renal volume
measurements: accuracy and repeatability of US
compared with that of MR imaging. Radiology
1999; 211:623628
6. Sanusi AA, Arogundade FA, Famurewa OC, et al.
Relationship of ultrasonographically determined
kidney volume with measured GFR, calculated
creatinine clearance and other parameters in
chronic kidney disease (CKD). Nephrol Dial
Transplant 2009; 24:16901694
7. Muto NS, Kamishima T, Harris AA, et al. Renal

cortical volume measured using automatic contouring software for computed tomography and its
relationship with BMI, age and renal function.
Eur J Radiol [Epub 2009 Nov 13]
8. Widjaja E, Oxtoby JW, Hale TL, Jones PW, Harden PN, McCall IW. Ultrasound measured renal
length versus low dose CT volume in predicting
single kidney glomerular filtration rate. Br J Radiol 2004; 77:759764
9. Herts BR, Sharma N, Lieber M, Freire M, Goldfarb DA, Poggio ED. Estimating glomerular filtration rate in kidney donors: a model constructed
with renal volume measurements from donor CT
scans. Radiology 2009; 252:109116
10. Khati NJ, Hill MC, Kimmel PL. The role of ultrasound in renal insufficiency: the essentials. Ultrasound Q 2005; 21:227244
11. Moghazi S, Jones E, Schroepple J, et al. Correlation of renal histopathology with sonographic
findings. Kidney Int 2005; 67:15151520
12. Levey AS, Coresh J, Greene T, et al. Chronic kidney disease epidemiology collaboration. using
standardized serum creatinine values in the Modification of Diet in Renal Disease study equation
for estimating glomerular filtration rate. Ann Intern Med 2006; 145:247254
13. Adibi A, Adibi I, Khosravi P. Do kidney sizes in
ultrasonography correlate to glomerular filtration
rate in healthy children? Australas Radiol 2007;
51:555559
14. Levey AS, Coresh J, Balk E, et al. National Kidney Foundation practice guidelines for chronic
kidney disease: evaluation, classification, and
stratification. Ann Intern Med 2003; 139:137147
15. Stevens LA, Coresh J, Greene T, Levey AS. Assessing kidney function: measured and estimated
glomerular filtration rate. N Engl J Med 2006;
354:24732483
16. Rule AD, Larson TS, Bergstralh EJ, Slezak JM,
Jacobsen SJ, Cosio FG. Using serum creatinine to
estimate glomerular filtration rate: accuracy in
good health and in chronic kidney disease. Ann
Intern Med 2004; 141:929937
17. Poggio ED, Wang X, Greene T, Van Lente F, Hall
PM. Performance of the Modification of Diet in Renal Disease and Cockcroft-Gault equations in the
estimation of GFR in health and in chronic kidney
disease. J Am Soc Nephrol 2005; 16:459466

W149