- Email: [email protected]
Correlation Between Doppler Parameters and Renal Cortical Fibrosis in Lupus Nephritis: A Preliminary Observation
Ultrasound in Med. & Biol., Vol. 39, No. 2, pp. 275–282, 2013 Copyright Ó 2013 World Federation for Ultrasound in Medicine & Biology Printed in the USA. All rights reserved 0301-5629/$ - see front matter
http://dx.doi.org/10.1016/j.ultrasmedbio.2012.10.009
Original Contribution
d
CORRELATION BETWEEN DOPPLER PARAMETERS AND RENAL CORTICAL FIBROSIS IN LUPUS NEPHRITIS: A PRELIMINARY OBSERVATION JING GAO,* JAMES CHEVALIER,y YONG HO AUH,* JONATHAN M. RUBIN,z HUI WANG,x LI-NA SUN,x SURYA SESHAN,{ and ROBERT MIN* * Department of Radiology, New York-Presbyterian Hospital, Weill Cornell Medical College, New York, New York, USA; Department of Internal Medicine, Rogosin Institute, Weill Cornell Medical College, New York, New York, USA; z Department of Radiology, University of Michigan Hospital, Ann Arbor, Michigan, USA; x Department of Ultrasound, China-Japan Union Hospital, Jilin University, Changchun, China; and { Department of Pathology, Weill Cornell Medical College, New York, New York, USA
y
(Received 12 July 2012; revised 11 September 2012; in final form 8 October 2012)
Abstract—To assess the relationship between renal Doppler parameters and renal cortical fibrosis in lupus nephritis (LN), we retrospectively reviewed 24 patients with LN underwent both renal color Doppler sonography and renal biopsy. The angle-corrected Doppler parameters, including peak systolic velocity (PSV), end diastolic velocity (EDV) and resistive index (RI) at the main and interlobar renal arteries were measured. The Doppler parameters and PSV and EDV ratios of the interlobar artery to main renal artery were compared with histopathologic analysis of the kidney biopsy specimen. On the basis of renal cortical fibrosis, the 24 cases of LN were divided into two groups: mild (6%–25%) renal cortex fibrosis (n 5 13) and moderate (26%–50%) renal cortex fibrosis (n 5 11). An independent-samples two tailed t test was used to statistically analyze the differences in PSV, EDV and RI between the two groups. Receiver operating characteristic was analyzed for assessing the accuracy of interlobar artery PSV and EDV in predicting moderate renal cortical fibrosis. In our result, both PSV and EDV in moderate renal cortex fibrosis were lower than that in mild renal cortex fibrosis. There were statistically significant differences in PSV and EDV at the interlobar artery, EDV and RI at the main renal artery, and PSV and EDV ratios of the interlobar artery to main renal artery between the two groups (all p , 0.05). The area under receiver operating characteristic curves of PSV and EDV for predicting .26% renal cortical fibrosis was 0.96 and 0.90, respectively. The optimal cutoff values for differentiating .26% renal cortical fibrosis from those ,25% were PSV 30 cm/s (sensitivity 5 0.92; specificity 5 1) and EDV 13 cm/s (sensitivity 5 0.77; specificity 5 1). Therefore, the values of PSV and EDV at the interlobar artery can potentially be used as hemodynamic indicators of renal cortical fibrosis, which may non-invasively assist in monitoring the progression of renal cortical fibrosis in LN, especially in patients with contraindications to renal biopsy. (E-mail: [email protected]) Ó 2013 World Federation for Ultrasound in Medicine & Biology. Key Words: Color Doppler sonography, Histopathology, Lupus nephropathy, Receiver operating characteristic.
a rational treatment strategy for the patient. Therefore, early renal biopsy has been advocated in patients with SLE (Ortega et al. 2010); however, because of its invasive nature, it may be contraindicated in patients with certain conditions such as coagulopathy. In one report, five of the seven patients (70%) who developed significant bleeding after kidney biopsy were patients with LN (Tao et al. 2008). Moreover, severe biopsy complications, such as intra-renal arteriovenous fistula with ‘‘steal’’ or large perinephric hematomas, may eventually lead to or worsen the renal failure (Gao et al. 2007). Serial biopsies to monitor progression of disease would expose patients to these repeated risks and is not practical. Alternatively, sonography, an established non-invasive imaging technique, has
INTRODUCTION Lupus nephritis (LN) is an important predictor of poor outcome in patients with systemic lupus erythematosus (SLE). Renal injury may often remain undiagnosed until nephritic syndrome or acute kidney injuries appear with increased risk of end-stage renal disease (Ortega et al. 2010). Assessing histopathologic categorization, fibrosis, activity and chronicity is of great importance in planning
Address correspondence to: Jing Gao, Department of Radiology, New York-Presbyterian Hospital, Weill Cornell Medical College, 525 East 68th Street, Star 8A, New York, NY 10065. E-mail: jig2001@ med.cornell.edu 275
276
Ultrasound in Medicine and Biology
been applied to the evaluation of chronic renal diseases, including LN, and may help to determine prognosis (Buturovic-Ponikvar et al. 2003; Krumme 2006; Platt et al. 1997). In renal sonography, morphologic changes including the size, parenchymal echogenicity and corticomedullary differentiation of the kidney on grayscale imaging may be detected in a kidney with or without renal failure. Unfortunately, these sonographic appearances in the evaluation of renal parenchyma disease lack specificity (Buturovic-Ponikvar et al. 2003; Krumme 2006). Therefore, resistive index (RI) measured with spectral Doppler has been added to evaluate the vascular resistance and end-organ vascular compliance in the kidneys (Bude et al. 1999a, 1999b). Some publications present a good correlation between the RI and histologic type of LN (Platt et al. 1997; Wang et al. 2007), whereas another article reported that there was no relationship between RI and histopathology € with disease duration of less than 5 years (Ozbek et al. 1995). To date, the usefulness of color Doppler sonography in monitoring progression of LN remains controversial. Yet, as a non-invasive and semi-quantitative parameter, the RI is still considered a useful tool in longitudinally monitoring the progression of renal parenchymal damage and evaluating therapeutic efficacy in LN (Platt et al. 1997; Wang et al. 2007). Recent studies on ultrasound elasticity of the transplanted (Stock et al. 2010; Syversveen et al. 2011; Weiztel et al. 2005) and native kidneys (Gallotti et al. 2010) have provided a new method to assess the biomechanical property (stiffness) of the renal parenchyma in healthy and diseased kidneys. A firmer (i.e., increased stiffness) renal parenchyma, associated with interstitial fibrosis, may gradually develop secondary to insufficient renal perfusion or ischemic insult. Freehand elastography is useful in measuring the strain and strain rate of the renal cortex in a transplanted kidney (Weiztel et al. 2005); however, it is difficult to perform a freehand compression on most native kidneys because of the deep location. On the other hand, the accuracy of shear wave velocity in assessing renal parenchymal stiffness is still not validated (Syversveen et al. 2011) because, at least in part, it can result from varying shear wave velocities owing to anisotropy of the renal cortex (Rubin et al. 1988). The aim of our study was to assess the correlation of the value of Doppler parameters such as peak systolic velocity (PSV) and end diastolic velocity (EDV) with the severity of renal cortical fibrosis in LN. The ultimate goal would be to demonstrate the potential of PSV, EDV or both as hemodynamic predictors for monitoring the progression of renal cortical fibrosis in LN, which may be useful as a marker for renal prognosis.
Volume 39, Number 2, 2013
MATERIALS AND METHODS Patient information We retrospectively reviewed 147 patients (75 in New York–Presbyterian Hospital, Weill Cornell Medical College, New York, NY, USA, and 72 in China–Japan Union Hospital of Jilin University, Changchun, China) who underwent both native kidney biopsy and renal color Doppler sonography between August 2010 and September 2011. We enrolled 24 patients (13 in New York–Presbyterian Hospital and 11 in China–Japan Union Hospital of Jilin University, 5 men and 19 women, age range 18–51 years and mean age 33.5 6 7.8 years) with LN who had spectral Doppler imaging at the lower pole of the left kidney where biopsy samples were taken. The interval between renal sonography and biopsy was within 15 days and most often done on the day of the biopsy. The diagnosis of SLE in all cases was made by the patient’s rheumatologist, using the American College of Rheumatology criteria. The diagnosis of nephritis was made by the patient’s nephrologist, presenting with hematuria, proteinuria or elevated creatinine level. Color Doppler images were reviewed on web-based picture archiving and communication system (PACS, General Electric, Milwaukee, WI, USA) and clinical data and laboratory reports were collected through reviewing electronic medical records of the patients. The retrospective reviews were approved by the local ethics committee in both institutions. A written informed consent was not obtained from patients because the study was retrospective. The study was compliant with Health Insurance Portability and Accountability Act of 1996. Patients with hypertension, high cholesterol, congestive heart failure, significant arrhythmia, aortic valve stenosis, significant renal artery stenosis, abdominal aorta aneurysm, existing intra-renal arteriovenous fistula and hydronephrosis that may affect the value of Doppler velocity and spectral Doppler waveform pattern were excluded from this study. Color Doppler sonography Renal color Doppler sonography was performed with a C41 curved linear array or 4V1 phased-array sector transducer with multiple frequencies (2–4 MHz for grayscale imaging and 1.75–4 MHz for color Doppler imaging; Sequoia 512, Siemens Medical Solutions, Mountain View, CA, USA). To achieve an ultrasound beam parallel to the blood flow direction at the main and intra-renal artery being imaged, the patient was placed in a decubitus position for renal color Doppler sonography. An ultrasound probe covered with transmitting gel was gently placed on the skin over the left kidney.
Doppler parameters in lupus nephritis d J. GAO et al.
We started renal sonography with grayscale imaging to measure the size of the kidney and observed the echogenicity, then evaluated for calculus, mass and hydronephrosis in the kidney. Color-flow imaging was initiated at low pulse repetition frequency (PRF), low wall filter and high total color gain for assessing overall vasculature of the kidney, as well as detecting the low velocity flow at the interlobar artery. We then adjusted color Doppler settings by gradually increasing PRF and wall filter until a main renal artery was clearly optimized without aliasing. We use color Doppler as a qualitative tool to assess blood flow in a vessel and tissue and to guide spectral Doppler analysis. However, color Doppler is based on an autocorrelation technique that represents mean flow velocity. Power Doppler does not show flow direction, which cannot distinguish arterial from venous flow. Conversely, spectral Doppler provides semi-quantitatively measure of flow velocity at systole and diastole. Spectral Doppler is more precise than color Doppler in quantitatively assessing blood flow in renal arteries closed to the glomeruli. A 3-mm spectral Doppler gate was placed at the center of the artery to obtain the spectra that were optimized with lowest PRF and wall filter without aliasing, and high gain without obscuring background noise. To obtain the maximal-flow velocity at each artery imaged, the Doppler angle was corrected as the ultrasound beam was made as parallel to the flow direction as possible. The main renal artery velocities were sampled at the renal hilus. The interlobar artery flow velocities were measured adjacent to the medullary pyramids. In our institutions, measuring Doppler velocity and RI at the main renal artery and interlobar artery was performed on the patients with renal failure, chronic medical renal disease or both. As a standard of care for a patient undergoing kidney biopsy, renal color flow imaging and spectral Doppler were routinely performed at the biopsy site to assess the possibility of existing renal vascular abnormalities before the biopsy and vascular complication after the biopsy. The angle-corrected PSV (an absolute maximal value in systole), EDV (an absolute minimal value in diastole), and RI ([PSV–EDV]/PSV) on the spectra were manually measured with electronic calipers and built-in software in the ultrasound scanner (Fig. 1a). To evaluate the relationship between the blood flow ratios of the intra-renal artery to the main renal artery and the severity of renal cortical fibrosis, we measured the PSV ratio of the interlobar artery to the main renal artery. It was defined as PSVratio 5 PSVinterlobar artery =PSVmain renal artery
(1)
Subsequently, EDV ratio was also calculated with the same equation for each patient.
277
Fig. 1. (a) Renal color Doppler sonography was performed on a 49-y-old woman with clinically suspected lupus nephropathy before kidney biopsy. On this longitudinal color duplex imaging of the left kidney, angle-corrected (47 ) Doppler parameters were measured at the interlobar artery. The peak systolic velocity, end diastolic velocity and resistive index (RI) measured 55 cm/s (V1 5 0.55 m/s), 24 cm/s (V2 5 0.24 m/s) and 0.56, respectively. (b) The kidney biopsy samples were taken at the lower pole of the left kidney. The histopathologic result was mild (,25%) renal cortex fibrosis.
Kidney biopsy and histopathology The nephrologists requested and performed each patient’s kidney biopsy using ultrasound guidance. A kidney biopsy was performed for the clinical indication of diagnosis, management and treatment of lupus nephritis. Patients fasted for 8–10 h before the kidney biopsy. A prone position was standard to perform left kidney biopsy. A 4V1 ultrasound transducer with biopsy guide (Ultra-Pro II Needle Guide; CIVCO Medical Instruments, Kalona, IA, USA) was placed into a sterile transducer cover to guide local anesthesia and kidney biopsy with real-time grayscale imaging. The lower pole of the left kidney is commonly selected for kidney biopsy in our institution. An area lacking prominent vessels and absent evidence of vascular
278
Ultrasound in Medicine and Biology
abnormality on both color and spectral Doppler sonography was chosen as the biopsy site. An 18-gauge biopsy needle (Bard Peripheral technologies, Covington, GA, USA) was inserted and advanced to the kidney cortex under ultrasound guidance after local anesthesia. Static images and cine loops of renal sonography and biopsy specimens were stored in the PACS system in the Department of Radiology and the Department of Ultrasound. Biopsy specimens were reviewed by experienced pathologists who had more than 20 years experience each in the interpretation of kidney biopsy specimens and were blinded regarding the results of renal sonography. Biopsy specimens were assigned a class of LN, based on the International Society of Nephrology–Renal Pathology Society classification (Weening et al. 2004). Furthermore, the glomeruli, tubules, interstitium and vessels were evaluated for activity and chronicity. Severity of interstitial fibrosis or tubular atrophy was estimated to the nearest 5% and then divided into four categories: 0, no fibrosis present; 1, mild (,25%) fibrosis; 2, moderate (26%–50%) fibrosis; and 3, severe (.50%) fibrosis. Statistical analysis SPSS software 16.0 (SPSS Inc., Chicago, IL, USA) was applied for statistical analysis. All the Doppler parameters were expressed as mean and SD. An independent-samples two-tailed t test was used to analyze statistically different values of the PSV, EDVand RI at the main renal artery and interlobar artery, as well as PSVand EDV ratios between groups 1 and 2. An independent t test was used to test the difference of SLE duration between the two groups after logarithmically transforming a non-normal distribution of the SLE durations into normal distribution data. Sensitivity and specificity for various cutoff values of interlobar artery PSV and EDV were calculated and graphically displayed using receiver operating characteristic (ROC) curves; p , 0.05 was defined as a statistically significant difference. RESULTS On the basis of renal cortical fibrosis, 24 patients were divided into two groups. They were group 1 with mild (,25%) renal cortex fibrosis (n 5 13), and group 2 with moderate (26%–50%) renal cortex fibrosis (n 5 11). Geographic characteristics and clinical manifestations of the 24 patients with LN are listed in Table 1. There was no significant difference in the ages of the patients or duration of the SLE between the two groups (p . 0.05; Table 1). There were differences in serum creatinine (p , 0.01) and glomerular filtration rate (GFR;
Volume 39, Number 2, 2013
Table 1. Clinical information in 24 cases with lupus nephritis Information Age (year) Gender M/F Year of SLE Urine P/C Creatinine (mg/dL) GFR % % Cortical fibrosis Race African Asian Caucasian Indian Latino
Group 1 (n 5 13) Group 2 (n 5 11) 32.1 6 8.04 3/10 11.77 6 6.02 2.47 6 2.65 0.86 6 0.19 59.85 6 0.55 ,25%
36.4 6 10.8 2/9 11.09 6 9.88 3.68 6 4.79 1.67 6 0.87 49.73 6 14.06 26%–50%
3 7 1 1 1
2 5 1 0 3
t test p . 0.05 p . 0.05 p . 0.05 p , 0.01 p , 0.05
GFR 5 glomerular filtration rate; F 5 female; M 5 male; SLE 5 systemic lupus erythematosus; Urine P/C 5 urine protein to creatinine ratio.
p , 0.05) between the two groups. The GFR was lower in group 2 with moderate renal cortical fibrosis than that in group 1, and creatinine was higher in group 2 than in group 1 (Table 1). For the main renal artery, the PSV, EDV and RI measured 81.17 6 22.65 cm/s, 32.52 6 8.45 cm/s, and 0.60 6 0.05 in Group 1, and 80.80 6 33.99 cm/s, 22.88 6 8.77 cm/s, and 0.69 6 0.06 in group 2. There was no significant difference in PSV at the main renal artery between the two groups (p . 0.05). There was a statistically significant difference in EDV and RI at the main renal artery between the two groups (all p , 0.05). For the interlobar artery, the PSV, EDV and RI were 44.95 6 15.05 cm/s, 18.12 6 6.22 cm/s and 0.60 6 0.03 in group 1 (Fig. 1a, 1b), and 24.72 6 5.08 cm/s, 9.26 6 2.37 cm/s, and 0.61 6 0.07 in group 2 (Fig. 2a, 2b). There was no significant difference in RI between the two groups at the interlobar artery (p . 0.05); however, there was statistically significant difference in PSV and EDV at the interlobar artery (Fig. 3, Table 2) between the two groups (all p , 0.01). In addition, there was also significant difference in flow velocity ratio of the interlobar artery to the main renal artery between the two groups. The PSV and EDV ratios in group 2 were lower than that in group 1 (Table 2). The ROC curves for the PSV and EDV values at the interlobar artery to predict moderate (.26%) renal cortical fibrosis are displayed in Figures 4 and 5. The area under curve, sensitivity and specificity of best cutoff values of the interlobar artery PSV and EDV are listed in Table 3. DISCUSSION Detection of renal parenchymal damage and fibrosis in the early stages of LN or in the silent period is
Doppler parameters in lupus nephritis d J. GAO et al.
279
Fig. 3. Box plot shows that comparison of Doppler parameters including PSV, EDV, and RI at the interlobar artery between the patients with mild renal cortical fibrosis (group 1) and those with moderate renal cortical fibrosis (group 2). There are statistically significant differences in PSV and EDV between the two groups. However, the difference in RI between the two groups is not significant. Vertical scales (cm/s) express Doppler velocities at the interlobar artery.
Fig. 2. (a) Renal color Doppler sonography was performed on a 34-y-old woman with a history of lupus nephritis for 5 years. Before biopsy, longitudinal color duplex imaging of the left kidney, angle-corrected (29 ) peak systolic velocity was 15 cm/s (V1 5 0.150 m/s), end diastolic velocity measured 7.5 cm/s (V2 5 0.075 m/s) and RI was 0.50 at the interlobar artery. (b) The kidney biopsy samples were taken at the lower pole of the left kidney. The histopathologic result was moderate (50%) renal cortex fibrosis. INTERLOB 5 interlobar artery; LOW, lower pole; LT 5 left kidney.
challenging, but finding a way, especially a non-invasive means of diagnosis, would provide tremendous clinical utility. It is reported that renal damage is sometimes irreversible by the time the patient receives medical attention with clinically apparent renal disease and a kidney biopsy is performed (Wen 2011). In our study, 24 patients exhibited renal failure with elevated creatinine and low GFR, and they developed mild to moderate renal cortical fibrosis at the time of the kidney biopsies. The GFR is significantly lower in moderate renal cortical fibrosis than in mild renal cortical fibrosis. Serum creatinine in moderate renal cortical fibrosis is remarkably higher than that in mild renal cortical fibrosis. One can clearly note that a close correlation between renal cortical fibrosis and renal failure is obvious. Limitations of kidney biopsy are a result of its invasive nature, which prevents periodic sampling in the clinical setting of inactive lupus and renal symptoms, and its
small sample size that might not reflect the disease in the entire kidney. As new technologies for the earlier detection of kidney injury in LN such as biomarkers are emerging (Manoharan et al. 2010; Zhang et al. 2008), renal sonography should provide complementary information. The search for a non-invasive technique that can detect and monitor subclinical renal disease continues. Detecting the beginning or early stages of renal hemodynamic disturbances that may lead to end-stage renal disease or determine prognosis of renal cortical fibrosis Table 2. Doppler parameters in 24 cases with lupus nephropathy
Doppler parameter Main renal artery RI PSV EDV Interlobar artery RI PSV EDV PSV ratio, interlobar/ main renal artery EDV ratio, interlobar/ main renal artery
Group 1 (n 5 13) mean 6 SD*
Group 2 (n 5 11) mean 6 SD
p value
0.60 6 0.05 81.17 6 22.65 32.52 6 8.45
0.69 6 0.06 80.80 6 33.99 22.88 6 8.77
p , 0.01 p . 0.05 p , 0.05
0.60 6 0.03 44.95 6 15.05 18.12 6 6.22 0.55 6 0.17
0.61 6 0.07 24.72 6 5.08 9.26 6 2.37 0.31 6 0.12
p . 0.05 p , 0.001 p , 0.001 p , 0.001
0.56 6 0.17
0.38 6 0.16
p , 0.05
EDV 5 end diastolic velocity; PSV 5 peak systolic velocity; RI 5 resistive index. * Group 1 5 membranous lupus nephritis with renal cortical fibrosis ,25%. Group 2 5 Proliferative lupus nephritis with renal cortical fibrosis .26%.
280
Ultrasound in Medicine and Biology
Fig. 4. ROC curve graph of interlobar artery PSV shows the area under curve is 0.96, which indicates a high accuracy of PSV in discriminating greater than 26% renal cortical fibrosis.
may be one candidate. To date, clinically quantifying blood flow velocity near the glomerulus with a noninvasive imaging method still relies on Doppler measurements, including PSV and EDV, at the interlobar artery. Renal hemodynamic conditions, such as renal vascular resistance and compliance, are referenced by RI (Galesic et al. 2004; Langer and Jones 2007; Martinoli et al. 1996; Murphy et al. 2000, Parolini et al. 2009; Tublin et al. 2003). In our study, there was no marked difference of RI at the interlobar artery between renal cortical fibrosis less than 25% and greater than 26% (p . 0.05; Table 2). However, a significant difference in RI was noted at the
Fig. 5. ROC curve graph of interlobar artery EDV for predicting renal cortical fibrosis shows that the area under curve is 0.90. There is also a high accuracy of EDV in distinguishing moderate (.26%) renal cortical fibroses from mild ones (,25%).
Volume 39, Number 2, 2013
main renal artery between the two groups (p , 0.01). To find out how the RI represents the different renal histologic changes in LN, we reviewed the theoretical explanations for understanding the meaning, usefulness and limitations of RI (Bude et al. 1999a, 1999b; Galesic et al. 2004; Murphy et al. 2000; Platt et al. 1997; Tublin et al. 2003). RI is the ratio of the difference between PSV and EDV to PSV. Theoretically, a non-proportional alteration of any or both the PSV and EDV can change the value of RI. These findings were documented in our study. Although there was no significant difference in PSV at the main renal artery between the two groups, RI at the main renal artery was significantly different between the two groups (0.60 vs. 0.69, p , 0.01), which resulted from the value of EDV in group 2 being significantly lower than group 1 (Table 2). Clinically, variant appearances of the Doppler waveform and velocity at the main and interlobar arteries are due to the interactions of many renal hemodynamic factors that are produced, sometimes altered following the changes of an extrarenal (pre- and post-renal) condition, an intra-renal pathology, or both. Extrarenal conditions that can affect renal blood flow and substantially alter Doppler waveform include cardiac output, blood pressure and vascular conditions, such as renal artery stenosis or aortic aneurysms. Those conditions were excluded from our study, and their effects were outside the scope of this study. However, the intra-renal histopathologic parameters are capable of causing intra-renal hemodynamic changes are glomerulopathy, interstitial fibrosis and arteriosclerosis (Bude et al. 1999a), among which renal cortical fibrosis is the main topic of this report. The values of PSV and EDV as semi-quantitative measurements of intra-renal blood flow on spectral Doppler strongly depend on the distension of the small arteries in the kidney, which correlates with renal vascular compliance and vascular resistance. An elevated renal interstitial pressure would counteract the distention of arteries and arterioles, which would have asymmetric effects on systolic and diastolic blood flow. This effect is more dramatic during diastole (Murphy et al. 2000). The same result was demonstrated in our study, in which EDVs at both the main renal and interlobar arteries were lower in group 2 with greater than 26% renal fibrosis than in group 1 with less than 25% renal cortical fibrosis. Our results suggest that decreased renal vascular distension further reduced the effect of cross-section of the renal vessels, mainly on intra-renal blood flow at both systole and diastole (Table 2). The change of PSVand EDV could mainly result from interstitial fibrosis limiting distension of the artery (Bude et al. 1999a, 1999b; Murphy et al. 2000) in group 2. All statistical results in our study indicate that the changes of blood flow in both systole and
Doppler parameters in lupus nephritis d J. GAO et al.
281
Table 3. Receiver operating characteristic analysis of interlobar artery PSV and EDV for predicting greater than 26% renal cortical fibrosis Measure
AUC
Cutoff value
PSV EDV
0.96 0.90
30 cm/s 13 cm/s
Sensitivity (%)
95% CI
Specificity (%)
95% CI
92.3 76.9
0.64–0.99 0.46–0.95
100 100
0.79–0.99 0.71–0.98
AUC 5 area under curve; CI 5 confidence interval.
diastole are closely associated with a certain form of renal cortical fibrosis. The effect of moderate cortical fibrosis is more profound on decreasing vascular distension than in mild cortical fibrosis. The correlation of the change in PSV, EDV or both with degree of renal fibrosis has been reported in transplanted kidney dysfunction (Gao et al. 2011) and hemorrhagic fever renal syndrome (Dang et al. 2011). When using Doppler sonography to assess progression of LN, the values of PSV and EDV should be considered when interpreting Doppler parameters and analyzing the correlations between these measurements and clinical data or biochemical results. In our study, both PSV and EDV at the interlobar artery had a close correlation with renal cortical fibrosis in LN. The value of PSV in group 2 was significantly lower than in group 1 (p , 0.01) and the value of EDV in group 2 was also markedly lower than in group 1 (p , 0.01). Moreover, using interlobar artery PSV 30 cm/s and EDV 13 cm/s as cutoff values to discriminate moderate (.26%) renal cortical fibrosis from mild (,25 %) fibrosis, the sensitivity and specificity can achieve as high as 0.92 and 1, and 0.77 and 1, respectively. The lack of significant difference in RI between the two groups might be explained by the fact that the decreases of both PSVand EDV in group 2 were relatively proportional to that reduction in group 1 (Table 2). In addition, there was a close correlation between Doppler velocity ratio of the intra-renal artery to the main renal artery and the severity of renal cortical fibrosis. In our study, both PSV and EDV ratios in group 2 with greater than 26% renal cortical fibrosis were lower than group 1 with less than 25% renal cortical fibrosis. Again, a decrease of intra-renal blood flow can result from poor distension of interlobar artery as moderate renal cortical fibrosis developed. This discovery on a correlation between Doppler velocity ratio of interlobar to the main renal artery and renal cortical fibrosis would be, to the best of our knowledge, the first report in the renal color Doppler sonography literature. In regard to Doppler parameters at the intra-renal artery, our study is not suggesting that any Doppler parameter could replace kidney biopsy in managing LN, but that some sonographic parameters may add to other current and future non-invasive techniques to help provide diagnostic and prognostic information in patients with LN. This may be particularly useful as complimen-
tary information because scarring may be patchy or irregular in distribution on a kidney biopsy. This patchy irregularity may be missed on a renal needle biopsy. Generally, most native kidney biopsies can be performed only at the lower pole of the left kidney where no bowel or diaphragm lies along a needle track from the skin of biopsy needle insertion to the site of biopsy sample. To avoid bowel injury, the upper and mid poles of the kidney might not be recommended as biopsy sites. However, Doppler sonogram can non-invasively measure the flow velocity parameters at the upper, mid and lower poles of the kidney, which will assess the renal parenchyma as a whole. Therefore, Doppler velocity can be used to non-invasively assess each portion and whole kidney as clinically indicated. In conclusion, it appears that renal color Doppler sonography is useful in providing semi-quantitative hemodynamic information of the kidney, which strongly relates to renal cortical fibrosis. Therefore, PSV and EDV can potentially be used as indicators in assessing severity of renal cortical fibrosis or as markers in tracking progression of renal cortical fibrosis in LN, which can eventually reduce the frequency of renal biopsy. The limitations of our study are inter-observer variation in ultrasound scanning, its retrospective nature, and lack of study on all classes of LN. The diagnosis was based on a sample from a needle biopsy that may have missed significant fibrosis if the fibrosis was not uniformly distributed throughout both kidneys; however, this is also a limitation of renal biopsies. Furthermore, the combination of more than one pathologic change in the interstitial compartment and glomeruli is common in LN, which can potentially affect the value of Doppler parameters. More research on Doppler parameters that have potential implications for the non-invasive investigation of renal cortical fibrosis in LN and correlation between hemodynamic information and ultrasound elastography in LN should be encouraged. Acknowledgments—The authors thank Edmund Quest for technical support.
REFERENCES Bude RO, Rubin JM. Relationship between the resistive index and vascular compliance and resistance. Radiology 1999a;211:411–417.
282
Ultrasound in Medicine and Biology
Bude RO, Rubin JM. Effect of downstream cross-sectional area of an arterial bed on the resistive index and the early systolic acceleration. Radiology 1999b;212:732–738. Buturovic-Ponikvar J, Visnar-Perovic A. Ultrasonography in chronic renal failure. Europ J Radiology 2003;46:115–122. Dang XK, Wang Z, Zhao Y, Guo M. Comparative study of hemodynamics and hemorheology changes of haemorrhagic fever with renal syndrome by ultrasonography. J Clin Ultrasound Med 2011;13: 459–461. Galesic K, Sabljar-Matovinovic M, Tomic M, Brkljacic B. Renal vascular resistance in glomerular disease-correlation of resistance index with biopsy findings. Coll Antropol 2004;28:667–674. Gallotti A, D’Onofrio M, Pozzi Mucelli R. Acoustic Radiation Force Impulse (ARFI) technique in ultrasound with Virtual Touch tissue quantification of the upper abdomen. Radiol Med 2010;115: 889–892. Gao J, Ng A, Shih G, Goldstein M, Kapur S, Wang J, Min R. Intrarenal color doppler sonography: A window to vascular complications of renal transplant. J Ultrasound Med 2007;26:1403. Gao J, Rubin JM, Xiang DY, He W, Auh YH, Wang J, Ng A, Min R. Doppler parameters in renal transplant dysfunction: Correlations with histopathological changes. J Ultrasound Med 2011;30: 169–175. Krumme B. Renal Doppler sonography: Update in clinical nephrology. Nephro Clin Pract 2006;103:24–28. Langer JE, Jones LP. Sonographic evaluation of the renal transplant. Ultrasound Clin 2007;203:82–86. Manoharan A, Madaio M. Biomarkers in lupus nephritis. Rheum Dis Clin N Am 2010;36:131–143, ix. Martinoli C, Crespi G, Bertolotto M, Rollandi GA, Rosenberg I, Pretolesi F, Derchi LE. Interlobular vasculature in renal transplants: A power Doppler US study with MR correlation. Radiology 1996; 200:111–117. Murphy ME, Tublin ME. Understanding the Doppler RI: Impact of renal arterial distensibility on the RI in a hydronephrotic EX Vivo rabbit kidney model. J Ultrasound Med 2000;19:303–314. Ortega LM, Schultz DR, Lenz O, Pardo V, Contreras GN. Lupus nephritis: Pathologic features, epidemiology, and guide to therapeutic decisions. Lupus 2010;19:557–574. € € Erden MI, Aytac SK, Olmez € € Ozbek SS, B€ uy€ ukberber S, Tolunay O U. Image-directed color Doppler ultrasonography of the kidney in systemic lupus nephritis. J Clin Ultrasound 1995;23:17–20.
Volume 39, Number 2, 2013 Parolini CP, Noce A, Staffolani E, Giarrizzo GF, Costanzi S, Splendiani G. Renal resistive index and long-term outcome in chronic nephropathies. Radiology 2009;252:888–896. Platt J, Rubin J, Ellis J. Lupus nephritis: Predictive value of conventional and Doppler US and comparison with serologic and biopsy parameters. Radiology 1997;203:82–86. Rubin JM, Carson PL, Meyer CR. Anisotropic ultrasonic backscatter from the renal cortex. Ultrasound Med Biol 1988;14:507–511. Stock KF, Klein BS, Vo Cong MT, Sarkar O, R€omisch M, Regenbogen C, B€uttner M, Schuster T, Matevossian E, Amann K, Clevert DA, Heemann U, K€uchle C. ARFI-based tissue elasticity quantification in comparison to histology for the diagnosis of renal transplant fibrosis. Clin Hemorheol Microcirc 2010;46:139–148. Syversveen T, Brabrand K, Midtvedt K, Strøm EH, Hartmann A, Jakobsen JA, Berstad AE. Assessment of renal allograft fibrosis by acoustic radiation force impulse quantification—a pilot study. Transpl Int 2011;24:100–105. Tao JL, Li H, Li C, Xu XW, Li JF, Yi N, Liu DY, Qin Y, Cai JF, Liu BY, Xu H, Gao RT, Ye WL, Ye W, Li XM, Li XW. Risk factors of postrenal biopsy bleeding. Acta Acad Med Sin 2008;30:313–317. Tublin ME, Bude RO, Platt JF. The resistive index in renal Doppler sonography: Where do we stand? AJR 2003;180:885–892. Wang CL, Shu KH, Lan JL, Cheng CH, Wu MJ, Chen CH. Duplex Doppler sonography to predict response to therapy in active lupus nephritis. Kuang Tien Med J 2007;2:21–27. Weening JJ, D’Agati VD, Schwartz MM, Seshan SV, Alpers CE, Appel GB, Balow JE, Bruijn JA, Cook T, Ferrario F, Fogo AB, Ginzler EM, Hebert L, Hill G, Hill P, Jennette JC, Kong NC, Lesavre P, Lockshin M, Looi LM, Makino H, Moura LA, Nagata M. The classification of glomerulonephritis in systemic lupus erythematosus revisited. J Am Soc Nephrol 2004;15:241–250. Wen YK. Renal biopsy findings in new-onset systemic lupus erythematosus with clinical renal disease. Int Urol Nephrol 2011;43: 801–806. Weiztel WF, Kim K, Rubin JM, Xie H, O’Donnell M. Renal advances in ultrasound elasticity imaging: Measuring the compliance of arteries and kidneys in end-stage renal disease. Blood Purif 2005;23:10–17. Zhang XL, Jin M, Wu HF, Nadasdy T, Nadasdy G, Harris N, Green-Church K, Nagaraja H, Birmingham DJ, Yu CY, Hebert LA, Rovin BH. Biomarkers of lupus nephritis determined by serial urine proteomics. Kidney Int 2008;74:799–807.
http://dx.doi.org/10.1016/j.ultrasmedbio.2012.10.009
Original Contribution
d
CORRELATION BETWEEN DOPPLER PARAMETERS AND RENAL CORTICAL FIBROSIS IN LUPUS NEPHRITIS: A PRELIMINARY OBSERVATION JING GAO,* JAMES CHEVALIER,y YONG HO AUH,* JONATHAN M. RUBIN,z HUI WANG,x LI-NA SUN,x SURYA SESHAN,{ and ROBERT MIN* * Department of Radiology, New York-Presbyterian Hospital, Weill Cornell Medical College, New York, New York, USA; Department of Internal Medicine, Rogosin Institute, Weill Cornell Medical College, New York, New York, USA; z Department of Radiology, University of Michigan Hospital, Ann Arbor, Michigan, USA; x Department of Ultrasound, China-Japan Union Hospital, Jilin University, Changchun, China; and { Department of Pathology, Weill Cornell Medical College, New York, New York, USA
y
(Received 12 July 2012; revised 11 September 2012; in final form 8 October 2012)
Abstract—To assess the relationship between renal Doppler parameters and renal cortical fibrosis in lupus nephritis (LN), we retrospectively reviewed 24 patients with LN underwent both renal color Doppler sonography and renal biopsy. The angle-corrected Doppler parameters, including peak systolic velocity (PSV), end diastolic velocity (EDV) and resistive index (RI) at the main and interlobar renal arteries were measured. The Doppler parameters and PSV and EDV ratios of the interlobar artery to main renal artery were compared with histopathologic analysis of the kidney biopsy specimen. On the basis of renal cortical fibrosis, the 24 cases of LN were divided into two groups: mild (6%–25%) renal cortex fibrosis (n 5 13) and moderate (26%–50%) renal cortex fibrosis (n 5 11). An independent-samples two tailed t test was used to statistically analyze the differences in PSV, EDV and RI between the two groups. Receiver operating characteristic was analyzed for assessing the accuracy of interlobar artery PSV and EDV in predicting moderate renal cortical fibrosis. In our result, both PSV and EDV in moderate renal cortex fibrosis were lower than that in mild renal cortex fibrosis. There were statistically significant differences in PSV and EDV at the interlobar artery, EDV and RI at the main renal artery, and PSV and EDV ratios of the interlobar artery to main renal artery between the two groups (all p , 0.05). The area under receiver operating characteristic curves of PSV and EDV for predicting .26% renal cortical fibrosis was 0.96 and 0.90, respectively. The optimal cutoff values for differentiating .26% renal cortical fibrosis from those ,25% were PSV 30 cm/s (sensitivity 5 0.92; specificity 5 1) and EDV 13 cm/s (sensitivity 5 0.77; specificity 5 1). Therefore, the values of PSV and EDV at the interlobar artery can potentially be used as hemodynamic indicators of renal cortical fibrosis, which may non-invasively assist in monitoring the progression of renal cortical fibrosis in LN, especially in patients with contraindications to renal biopsy. (E-mail: [email protected]) Ó 2013 World Federation for Ultrasound in Medicine & Biology. Key Words: Color Doppler sonography, Histopathology, Lupus nephropathy, Receiver operating characteristic.
a rational treatment strategy for the patient. Therefore, early renal biopsy has been advocated in patients with SLE (Ortega et al. 2010); however, because of its invasive nature, it may be contraindicated in patients with certain conditions such as coagulopathy. In one report, five of the seven patients (70%) who developed significant bleeding after kidney biopsy were patients with LN (Tao et al. 2008). Moreover, severe biopsy complications, such as intra-renal arteriovenous fistula with ‘‘steal’’ or large perinephric hematomas, may eventually lead to or worsen the renal failure (Gao et al. 2007). Serial biopsies to monitor progression of disease would expose patients to these repeated risks and is not practical. Alternatively, sonography, an established non-invasive imaging technique, has
INTRODUCTION Lupus nephritis (LN) is an important predictor of poor outcome in patients with systemic lupus erythematosus (SLE). Renal injury may often remain undiagnosed until nephritic syndrome or acute kidney injuries appear with increased risk of end-stage renal disease (Ortega et al. 2010). Assessing histopathologic categorization, fibrosis, activity and chronicity is of great importance in planning
Address correspondence to: Jing Gao, Department of Radiology, New York-Presbyterian Hospital, Weill Cornell Medical College, 525 East 68th Street, Star 8A, New York, NY 10065. E-mail: jig2001@ med.cornell.edu 275
276
Ultrasound in Medicine and Biology
been applied to the evaluation of chronic renal diseases, including LN, and may help to determine prognosis (Buturovic-Ponikvar et al. 2003; Krumme 2006; Platt et al. 1997). In renal sonography, morphologic changes including the size, parenchymal echogenicity and corticomedullary differentiation of the kidney on grayscale imaging may be detected in a kidney with or without renal failure. Unfortunately, these sonographic appearances in the evaluation of renal parenchyma disease lack specificity (Buturovic-Ponikvar et al. 2003; Krumme 2006). Therefore, resistive index (RI) measured with spectral Doppler has been added to evaluate the vascular resistance and end-organ vascular compliance in the kidneys (Bude et al. 1999a, 1999b). Some publications present a good correlation between the RI and histologic type of LN (Platt et al. 1997; Wang et al. 2007), whereas another article reported that there was no relationship between RI and histopathology € with disease duration of less than 5 years (Ozbek et al. 1995). To date, the usefulness of color Doppler sonography in monitoring progression of LN remains controversial. Yet, as a non-invasive and semi-quantitative parameter, the RI is still considered a useful tool in longitudinally monitoring the progression of renal parenchymal damage and evaluating therapeutic efficacy in LN (Platt et al. 1997; Wang et al. 2007). Recent studies on ultrasound elasticity of the transplanted (Stock et al. 2010; Syversveen et al. 2011; Weiztel et al. 2005) and native kidneys (Gallotti et al. 2010) have provided a new method to assess the biomechanical property (stiffness) of the renal parenchyma in healthy and diseased kidneys. A firmer (i.e., increased stiffness) renal parenchyma, associated with interstitial fibrosis, may gradually develop secondary to insufficient renal perfusion or ischemic insult. Freehand elastography is useful in measuring the strain and strain rate of the renal cortex in a transplanted kidney (Weiztel et al. 2005); however, it is difficult to perform a freehand compression on most native kidneys because of the deep location. On the other hand, the accuracy of shear wave velocity in assessing renal parenchymal stiffness is still not validated (Syversveen et al. 2011) because, at least in part, it can result from varying shear wave velocities owing to anisotropy of the renal cortex (Rubin et al. 1988). The aim of our study was to assess the correlation of the value of Doppler parameters such as peak systolic velocity (PSV) and end diastolic velocity (EDV) with the severity of renal cortical fibrosis in LN. The ultimate goal would be to demonstrate the potential of PSV, EDV or both as hemodynamic predictors for monitoring the progression of renal cortical fibrosis in LN, which may be useful as a marker for renal prognosis.
Volume 39, Number 2, 2013
MATERIALS AND METHODS Patient information We retrospectively reviewed 147 patients (75 in New York–Presbyterian Hospital, Weill Cornell Medical College, New York, NY, USA, and 72 in China–Japan Union Hospital of Jilin University, Changchun, China) who underwent both native kidney biopsy and renal color Doppler sonography between August 2010 and September 2011. We enrolled 24 patients (13 in New York–Presbyterian Hospital and 11 in China–Japan Union Hospital of Jilin University, 5 men and 19 women, age range 18–51 years and mean age 33.5 6 7.8 years) with LN who had spectral Doppler imaging at the lower pole of the left kidney where biopsy samples were taken. The interval between renal sonography and biopsy was within 15 days and most often done on the day of the biopsy. The diagnosis of SLE in all cases was made by the patient’s rheumatologist, using the American College of Rheumatology criteria. The diagnosis of nephritis was made by the patient’s nephrologist, presenting with hematuria, proteinuria or elevated creatinine level. Color Doppler images were reviewed on web-based picture archiving and communication system (PACS, General Electric, Milwaukee, WI, USA) and clinical data and laboratory reports were collected through reviewing electronic medical records of the patients. The retrospective reviews were approved by the local ethics committee in both institutions. A written informed consent was not obtained from patients because the study was retrospective. The study was compliant with Health Insurance Portability and Accountability Act of 1996. Patients with hypertension, high cholesterol, congestive heart failure, significant arrhythmia, aortic valve stenosis, significant renal artery stenosis, abdominal aorta aneurysm, existing intra-renal arteriovenous fistula and hydronephrosis that may affect the value of Doppler velocity and spectral Doppler waveform pattern were excluded from this study. Color Doppler sonography Renal color Doppler sonography was performed with a C41 curved linear array or 4V1 phased-array sector transducer with multiple frequencies (2–4 MHz for grayscale imaging and 1.75–4 MHz for color Doppler imaging; Sequoia 512, Siemens Medical Solutions, Mountain View, CA, USA). To achieve an ultrasound beam parallel to the blood flow direction at the main and intra-renal artery being imaged, the patient was placed in a decubitus position for renal color Doppler sonography. An ultrasound probe covered with transmitting gel was gently placed on the skin over the left kidney.
Doppler parameters in lupus nephritis d J. GAO et al.
We started renal sonography with grayscale imaging to measure the size of the kidney and observed the echogenicity, then evaluated for calculus, mass and hydronephrosis in the kidney. Color-flow imaging was initiated at low pulse repetition frequency (PRF), low wall filter and high total color gain for assessing overall vasculature of the kidney, as well as detecting the low velocity flow at the interlobar artery. We then adjusted color Doppler settings by gradually increasing PRF and wall filter until a main renal artery was clearly optimized without aliasing. We use color Doppler as a qualitative tool to assess blood flow in a vessel and tissue and to guide spectral Doppler analysis. However, color Doppler is based on an autocorrelation technique that represents mean flow velocity. Power Doppler does not show flow direction, which cannot distinguish arterial from venous flow. Conversely, spectral Doppler provides semi-quantitatively measure of flow velocity at systole and diastole. Spectral Doppler is more precise than color Doppler in quantitatively assessing blood flow in renal arteries closed to the glomeruli. A 3-mm spectral Doppler gate was placed at the center of the artery to obtain the spectra that were optimized with lowest PRF and wall filter without aliasing, and high gain without obscuring background noise. To obtain the maximal-flow velocity at each artery imaged, the Doppler angle was corrected as the ultrasound beam was made as parallel to the flow direction as possible. The main renal artery velocities were sampled at the renal hilus. The interlobar artery flow velocities were measured adjacent to the medullary pyramids. In our institutions, measuring Doppler velocity and RI at the main renal artery and interlobar artery was performed on the patients with renal failure, chronic medical renal disease or both. As a standard of care for a patient undergoing kidney biopsy, renal color flow imaging and spectral Doppler were routinely performed at the biopsy site to assess the possibility of existing renal vascular abnormalities before the biopsy and vascular complication after the biopsy. The angle-corrected PSV (an absolute maximal value in systole), EDV (an absolute minimal value in diastole), and RI ([PSV–EDV]/PSV) on the spectra were manually measured with electronic calipers and built-in software in the ultrasound scanner (Fig. 1a). To evaluate the relationship between the blood flow ratios of the intra-renal artery to the main renal artery and the severity of renal cortical fibrosis, we measured the PSV ratio of the interlobar artery to the main renal artery. It was defined as PSVratio 5 PSVinterlobar artery =PSVmain renal artery
(1)
Subsequently, EDV ratio was also calculated with the same equation for each patient.
277
Fig. 1. (a) Renal color Doppler sonography was performed on a 49-y-old woman with clinically suspected lupus nephropathy before kidney biopsy. On this longitudinal color duplex imaging of the left kidney, angle-corrected (47 ) Doppler parameters were measured at the interlobar artery. The peak systolic velocity, end diastolic velocity and resistive index (RI) measured 55 cm/s (V1 5 0.55 m/s), 24 cm/s (V2 5 0.24 m/s) and 0.56, respectively. (b) The kidney biopsy samples were taken at the lower pole of the left kidney. The histopathologic result was mild (,25%) renal cortex fibrosis.
Kidney biopsy and histopathology The nephrologists requested and performed each patient’s kidney biopsy using ultrasound guidance. A kidney biopsy was performed for the clinical indication of diagnosis, management and treatment of lupus nephritis. Patients fasted for 8–10 h before the kidney biopsy. A prone position was standard to perform left kidney biopsy. A 4V1 ultrasound transducer with biopsy guide (Ultra-Pro II Needle Guide; CIVCO Medical Instruments, Kalona, IA, USA) was placed into a sterile transducer cover to guide local anesthesia and kidney biopsy with real-time grayscale imaging. The lower pole of the left kidney is commonly selected for kidney biopsy in our institution. An area lacking prominent vessels and absent evidence of vascular
278
Ultrasound in Medicine and Biology
abnormality on both color and spectral Doppler sonography was chosen as the biopsy site. An 18-gauge biopsy needle (Bard Peripheral technologies, Covington, GA, USA) was inserted and advanced to the kidney cortex under ultrasound guidance after local anesthesia. Static images and cine loops of renal sonography and biopsy specimens were stored in the PACS system in the Department of Radiology and the Department of Ultrasound. Biopsy specimens were reviewed by experienced pathologists who had more than 20 years experience each in the interpretation of kidney biopsy specimens and were blinded regarding the results of renal sonography. Biopsy specimens were assigned a class of LN, based on the International Society of Nephrology–Renal Pathology Society classification (Weening et al. 2004). Furthermore, the glomeruli, tubules, interstitium and vessels were evaluated for activity and chronicity. Severity of interstitial fibrosis or tubular atrophy was estimated to the nearest 5% and then divided into four categories: 0, no fibrosis present; 1, mild (,25%) fibrosis; 2, moderate (26%–50%) fibrosis; and 3, severe (.50%) fibrosis. Statistical analysis SPSS software 16.0 (SPSS Inc., Chicago, IL, USA) was applied for statistical analysis. All the Doppler parameters were expressed as mean and SD. An independent-samples two-tailed t test was used to analyze statistically different values of the PSV, EDVand RI at the main renal artery and interlobar artery, as well as PSVand EDV ratios between groups 1 and 2. An independent t test was used to test the difference of SLE duration between the two groups after logarithmically transforming a non-normal distribution of the SLE durations into normal distribution data. Sensitivity and specificity for various cutoff values of interlobar artery PSV and EDV were calculated and graphically displayed using receiver operating characteristic (ROC) curves; p , 0.05 was defined as a statistically significant difference. RESULTS On the basis of renal cortical fibrosis, 24 patients were divided into two groups. They were group 1 with mild (,25%) renal cortex fibrosis (n 5 13), and group 2 with moderate (26%–50%) renal cortex fibrosis (n 5 11). Geographic characteristics and clinical manifestations of the 24 patients with LN are listed in Table 1. There was no significant difference in the ages of the patients or duration of the SLE between the two groups (p . 0.05; Table 1). There were differences in serum creatinine (p , 0.01) and glomerular filtration rate (GFR;
Volume 39, Number 2, 2013
Table 1. Clinical information in 24 cases with lupus nephritis Information Age (year) Gender M/F Year of SLE Urine P/C Creatinine (mg/dL) GFR % % Cortical fibrosis Race African Asian Caucasian Indian Latino
Group 1 (n 5 13) Group 2 (n 5 11) 32.1 6 8.04 3/10 11.77 6 6.02 2.47 6 2.65 0.86 6 0.19 59.85 6 0.55 ,25%
36.4 6 10.8 2/9 11.09 6 9.88 3.68 6 4.79 1.67 6 0.87 49.73 6 14.06 26%–50%
3 7 1 1 1
2 5 1 0 3
t test p . 0.05 p . 0.05 p . 0.05 p , 0.01 p , 0.05
GFR 5 glomerular filtration rate; F 5 female; M 5 male; SLE 5 systemic lupus erythematosus; Urine P/C 5 urine protein to creatinine ratio.
p , 0.05) between the two groups. The GFR was lower in group 2 with moderate renal cortical fibrosis than that in group 1, and creatinine was higher in group 2 than in group 1 (Table 1). For the main renal artery, the PSV, EDV and RI measured 81.17 6 22.65 cm/s, 32.52 6 8.45 cm/s, and 0.60 6 0.05 in Group 1, and 80.80 6 33.99 cm/s, 22.88 6 8.77 cm/s, and 0.69 6 0.06 in group 2. There was no significant difference in PSV at the main renal artery between the two groups (p . 0.05). There was a statistically significant difference in EDV and RI at the main renal artery between the two groups (all p , 0.05). For the interlobar artery, the PSV, EDV and RI were 44.95 6 15.05 cm/s, 18.12 6 6.22 cm/s and 0.60 6 0.03 in group 1 (Fig. 1a, 1b), and 24.72 6 5.08 cm/s, 9.26 6 2.37 cm/s, and 0.61 6 0.07 in group 2 (Fig. 2a, 2b). There was no significant difference in RI between the two groups at the interlobar artery (p . 0.05); however, there was statistically significant difference in PSV and EDV at the interlobar artery (Fig. 3, Table 2) between the two groups (all p , 0.01). In addition, there was also significant difference in flow velocity ratio of the interlobar artery to the main renal artery between the two groups. The PSV and EDV ratios in group 2 were lower than that in group 1 (Table 2). The ROC curves for the PSV and EDV values at the interlobar artery to predict moderate (.26%) renal cortical fibrosis are displayed in Figures 4 and 5. The area under curve, sensitivity and specificity of best cutoff values of the interlobar artery PSV and EDV are listed in Table 3. DISCUSSION Detection of renal parenchymal damage and fibrosis in the early stages of LN or in the silent period is
Doppler parameters in lupus nephritis d J. GAO et al.
279
Fig. 3. Box plot shows that comparison of Doppler parameters including PSV, EDV, and RI at the interlobar artery between the patients with mild renal cortical fibrosis (group 1) and those with moderate renal cortical fibrosis (group 2). There are statistically significant differences in PSV and EDV between the two groups. However, the difference in RI between the two groups is not significant. Vertical scales (cm/s) express Doppler velocities at the interlobar artery.
Fig. 2. (a) Renal color Doppler sonography was performed on a 34-y-old woman with a history of lupus nephritis for 5 years. Before biopsy, longitudinal color duplex imaging of the left kidney, angle-corrected (29 ) peak systolic velocity was 15 cm/s (V1 5 0.150 m/s), end diastolic velocity measured 7.5 cm/s (V2 5 0.075 m/s) and RI was 0.50 at the interlobar artery. (b) The kidney biopsy samples were taken at the lower pole of the left kidney. The histopathologic result was moderate (50%) renal cortex fibrosis. INTERLOB 5 interlobar artery; LOW, lower pole; LT 5 left kidney.
challenging, but finding a way, especially a non-invasive means of diagnosis, would provide tremendous clinical utility. It is reported that renal damage is sometimes irreversible by the time the patient receives medical attention with clinically apparent renal disease and a kidney biopsy is performed (Wen 2011). In our study, 24 patients exhibited renal failure with elevated creatinine and low GFR, and they developed mild to moderate renal cortical fibrosis at the time of the kidney biopsies. The GFR is significantly lower in moderate renal cortical fibrosis than in mild renal cortical fibrosis. Serum creatinine in moderate renal cortical fibrosis is remarkably higher than that in mild renal cortical fibrosis. One can clearly note that a close correlation between renal cortical fibrosis and renal failure is obvious. Limitations of kidney biopsy are a result of its invasive nature, which prevents periodic sampling in the clinical setting of inactive lupus and renal symptoms, and its
small sample size that might not reflect the disease in the entire kidney. As new technologies for the earlier detection of kidney injury in LN such as biomarkers are emerging (Manoharan et al. 2010; Zhang et al. 2008), renal sonography should provide complementary information. The search for a non-invasive technique that can detect and monitor subclinical renal disease continues. Detecting the beginning or early stages of renal hemodynamic disturbances that may lead to end-stage renal disease or determine prognosis of renal cortical fibrosis Table 2. Doppler parameters in 24 cases with lupus nephropathy
Doppler parameter Main renal artery RI PSV EDV Interlobar artery RI PSV EDV PSV ratio, interlobar/ main renal artery EDV ratio, interlobar/ main renal artery
Group 1 (n 5 13) mean 6 SD*
Group 2 (n 5 11) mean 6 SD
p value
0.60 6 0.05 81.17 6 22.65 32.52 6 8.45
0.69 6 0.06 80.80 6 33.99 22.88 6 8.77
p , 0.01 p . 0.05 p , 0.05
0.60 6 0.03 44.95 6 15.05 18.12 6 6.22 0.55 6 0.17
0.61 6 0.07 24.72 6 5.08 9.26 6 2.37 0.31 6 0.12
p . 0.05 p , 0.001 p , 0.001 p , 0.001
0.56 6 0.17
0.38 6 0.16
p , 0.05
EDV 5 end diastolic velocity; PSV 5 peak systolic velocity; RI 5 resistive index. * Group 1 5 membranous lupus nephritis with renal cortical fibrosis ,25%. Group 2 5 Proliferative lupus nephritis with renal cortical fibrosis .26%.
280
Ultrasound in Medicine and Biology
Fig. 4. ROC curve graph of interlobar artery PSV shows the area under curve is 0.96, which indicates a high accuracy of PSV in discriminating greater than 26% renal cortical fibrosis.
may be one candidate. To date, clinically quantifying blood flow velocity near the glomerulus with a noninvasive imaging method still relies on Doppler measurements, including PSV and EDV, at the interlobar artery. Renal hemodynamic conditions, such as renal vascular resistance and compliance, are referenced by RI (Galesic et al. 2004; Langer and Jones 2007; Martinoli et al. 1996; Murphy et al. 2000, Parolini et al. 2009; Tublin et al. 2003). In our study, there was no marked difference of RI at the interlobar artery between renal cortical fibrosis less than 25% and greater than 26% (p . 0.05; Table 2). However, a significant difference in RI was noted at the
Fig. 5. ROC curve graph of interlobar artery EDV for predicting renal cortical fibrosis shows that the area under curve is 0.90. There is also a high accuracy of EDV in distinguishing moderate (.26%) renal cortical fibroses from mild ones (,25%).
Volume 39, Number 2, 2013
main renal artery between the two groups (p , 0.01). To find out how the RI represents the different renal histologic changes in LN, we reviewed the theoretical explanations for understanding the meaning, usefulness and limitations of RI (Bude et al. 1999a, 1999b; Galesic et al. 2004; Murphy et al. 2000; Platt et al. 1997; Tublin et al. 2003). RI is the ratio of the difference between PSV and EDV to PSV. Theoretically, a non-proportional alteration of any or both the PSV and EDV can change the value of RI. These findings were documented in our study. Although there was no significant difference in PSV at the main renal artery between the two groups, RI at the main renal artery was significantly different between the two groups (0.60 vs. 0.69, p , 0.01), which resulted from the value of EDV in group 2 being significantly lower than group 1 (Table 2). Clinically, variant appearances of the Doppler waveform and velocity at the main and interlobar arteries are due to the interactions of many renal hemodynamic factors that are produced, sometimes altered following the changes of an extrarenal (pre- and post-renal) condition, an intra-renal pathology, or both. Extrarenal conditions that can affect renal blood flow and substantially alter Doppler waveform include cardiac output, blood pressure and vascular conditions, such as renal artery stenosis or aortic aneurysms. Those conditions were excluded from our study, and their effects were outside the scope of this study. However, the intra-renal histopathologic parameters are capable of causing intra-renal hemodynamic changes are glomerulopathy, interstitial fibrosis and arteriosclerosis (Bude et al. 1999a), among which renal cortical fibrosis is the main topic of this report. The values of PSV and EDV as semi-quantitative measurements of intra-renal blood flow on spectral Doppler strongly depend on the distension of the small arteries in the kidney, which correlates with renal vascular compliance and vascular resistance. An elevated renal interstitial pressure would counteract the distention of arteries and arterioles, which would have asymmetric effects on systolic and diastolic blood flow. This effect is more dramatic during diastole (Murphy et al. 2000). The same result was demonstrated in our study, in which EDVs at both the main renal and interlobar arteries were lower in group 2 with greater than 26% renal fibrosis than in group 1 with less than 25% renal cortical fibrosis. Our results suggest that decreased renal vascular distension further reduced the effect of cross-section of the renal vessels, mainly on intra-renal blood flow at both systole and diastole (Table 2). The change of PSVand EDV could mainly result from interstitial fibrosis limiting distension of the artery (Bude et al. 1999a, 1999b; Murphy et al. 2000) in group 2. All statistical results in our study indicate that the changes of blood flow in both systole and
Doppler parameters in lupus nephritis d J. GAO et al.
281
Table 3. Receiver operating characteristic analysis of interlobar artery PSV and EDV for predicting greater than 26% renal cortical fibrosis Measure
AUC
Cutoff value
PSV EDV
0.96 0.90
30 cm/s 13 cm/s
Sensitivity (%)
95% CI
Specificity (%)
95% CI
92.3 76.9
0.64–0.99 0.46–0.95
100 100
0.79–0.99 0.71–0.98
AUC 5 area under curve; CI 5 confidence interval.
diastole are closely associated with a certain form of renal cortical fibrosis. The effect of moderate cortical fibrosis is more profound on decreasing vascular distension than in mild cortical fibrosis. The correlation of the change in PSV, EDV or both with degree of renal fibrosis has been reported in transplanted kidney dysfunction (Gao et al. 2011) and hemorrhagic fever renal syndrome (Dang et al. 2011). When using Doppler sonography to assess progression of LN, the values of PSV and EDV should be considered when interpreting Doppler parameters and analyzing the correlations between these measurements and clinical data or biochemical results. In our study, both PSV and EDV at the interlobar artery had a close correlation with renal cortical fibrosis in LN. The value of PSV in group 2 was significantly lower than in group 1 (p , 0.01) and the value of EDV in group 2 was also markedly lower than in group 1 (p , 0.01). Moreover, using interlobar artery PSV 30 cm/s and EDV 13 cm/s as cutoff values to discriminate moderate (.26%) renal cortical fibrosis from mild (,25 %) fibrosis, the sensitivity and specificity can achieve as high as 0.92 and 1, and 0.77 and 1, respectively. The lack of significant difference in RI between the two groups might be explained by the fact that the decreases of both PSVand EDV in group 2 were relatively proportional to that reduction in group 1 (Table 2). In addition, there was a close correlation between Doppler velocity ratio of the intra-renal artery to the main renal artery and the severity of renal cortical fibrosis. In our study, both PSV and EDV ratios in group 2 with greater than 26% renal cortical fibrosis were lower than group 1 with less than 25% renal cortical fibrosis. Again, a decrease of intra-renal blood flow can result from poor distension of interlobar artery as moderate renal cortical fibrosis developed. This discovery on a correlation between Doppler velocity ratio of interlobar to the main renal artery and renal cortical fibrosis would be, to the best of our knowledge, the first report in the renal color Doppler sonography literature. In regard to Doppler parameters at the intra-renal artery, our study is not suggesting that any Doppler parameter could replace kidney biopsy in managing LN, but that some sonographic parameters may add to other current and future non-invasive techniques to help provide diagnostic and prognostic information in patients with LN. This may be particularly useful as complimen-
tary information because scarring may be patchy or irregular in distribution on a kidney biopsy. This patchy irregularity may be missed on a renal needle biopsy. Generally, most native kidney biopsies can be performed only at the lower pole of the left kidney where no bowel or diaphragm lies along a needle track from the skin of biopsy needle insertion to the site of biopsy sample. To avoid bowel injury, the upper and mid poles of the kidney might not be recommended as biopsy sites. However, Doppler sonogram can non-invasively measure the flow velocity parameters at the upper, mid and lower poles of the kidney, which will assess the renal parenchyma as a whole. Therefore, Doppler velocity can be used to non-invasively assess each portion and whole kidney as clinically indicated. In conclusion, it appears that renal color Doppler sonography is useful in providing semi-quantitative hemodynamic information of the kidney, which strongly relates to renal cortical fibrosis. Therefore, PSV and EDV can potentially be used as indicators in assessing severity of renal cortical fibrosis or as markers in tracking progression of renal cortical fibrosis in LN, which can eventually reduce the frequency of renal biopsy. The limitations of our study are inter-observer variation in ultrasound scanning, its retrospective nature, and lack of study on all classes of LN. The diagnosis was based on a sample from a needle biopsy that may have missed significant fibrosis if the fibrosis was not uniformly distributed throughout both kidneys; however, this is also a limitation of renal biopsies. Furthermore, the combination of more than one pathologic change in the interstitial compartment and glomeruli is common in LN, which can potentially affect the value of Doppler parameters. More research on Doppler parameters that have potential implications for the non-invasive investigation of renal cortical fibrosis in LN and correlation between hemodynamic information and ultrasound elastography in LN should be encouraged. Acknowledgments—The authors thank Edmund Quest for technical support.
REFERENCES Bude RO, Rubin JM. Relationship between the resistive index and vascular compliance and resistance. Radiology 1999a;211:411–417.
282
Ultrasound in Medicine and Biology
Bude RO, Rubin JM. Effect of downstream cross-sectional area of an arterial bed on the resistive index and the early systolic acceleration. Radiology 1999b;212:732–738. Buturovic-Ponikvar J, Visnar-Perovic A. Ultrasonography in chronic renal failure. Europ J Radiology 2003;46:115–122. Dang XK, Wang Z, Zhao Y, Guo M. Comparative study of hemodynamics and hemorheology changes of haemorrhagic fever with renal syndrome by ultrasonography. J Clin Ultrasound Med 2011;13: 459–461. Galesic K, Sabljar-Matovinovic M, Tomic M, Brkljacic B. Renal vascular resistance in glomerular disease-correlation of resistance index with biopsy findings. Coll Antropol 2004;28:667–674. Gallotti A, D’Onofrio M, Pozzi Mucelli R. Acoustic Radiation Force Impulse (ARFI) technique in ultrasound with Virtual Touch tissue quantification of the upper abdomen. Radiol Med 2010;115: 889–892. Gao J, Ng A, Shih G, Goldstein M, Kapur S, Wang J, Min R. Intrarenal color doppler sonography: A window to vascular complications of renal transplant. J Ultrasound Med 2007;26:1403. Gao J, Rubin JM, Xiang DY, He W, Auh YH, Wang J, Ng A, Min R. Doppler parameters in renal transplant dysfunction: Correlations with histopathological changes. J Ultrasound Med 2011;30: 169–175. Krumme B. Renal Doppler sonography: Update in clinical nephrology. Nephro Clin Pract 2006;103:24–28. Langer JE, Jones LP. Sonographic evaluation of the renal transplant. Ultrasound Clin 2007;203:82–86. Manoharan A, Madaio M. Biomarkers in lupus nephritis. Rheum Dis Clin N Am 2010;36:131–143, ix. Martinoli C, Crespi G, Bertolotto M, Rollandi GA, Rosenberg I, Pretolesi F, Derchi LE. Interlobular vasculature in renal transplants: A power Doppler US study with MR correlation. Radiology 1996; 200:111–117. Murphy ME, Tublin ME. Understanding the Doppler RI: Impact of renal arterial distensibility on the RI in a hydronephrotic EX Vivo rabbit kidney model. J Ultrasound Med 2000;19:303–314. Ortega LM, Schultz DR, Lenz O, Pardo V, Contreras GN. Lupus nephritis: Pathologic features, epidemiology, and guide to therapeutic decisions. Lupus 2010;19:557–574. € € Erden MI, Aytac SK, Olmez € € Ozbek SS, B€ uy€ ukberber S, Tolunay O U. Image-directed color Doppler ultrasonography of the kidney in systemic lupus nephritis. J Clin Ultrasound 1995;23:17–20.
Volume 39, Number 2, 2013 Parolini CP, Noce A, Staffolani E, Giarrizzo GF, Costanzi S, Splendiani G. Renal resistive index and long-term outcome in chronic nephropathies. Radiology 2009;252:888–896. Platt J, Rubin J, Ellis J. Lupus nephritis: Predictive value of conventional and Doppler US and comparison with serologic and biopsy parameters. Radiology 1997;203:82–86. Rubin JM, Carson PL, Meyer CR. Anisotropic ultrasonic backscatter from the renal cortex. Ultrasound Med Biol 1988;14:507–511. Stock KF, Klein BS, Vo Cong MT, Sarkar O, R€omisch M, Regenbogen C, B€uttner M, Schuster T, Matevossian E, Amann K, Clevert DA, Heemann U, K€uchle C. ARFI-based tissue elasticity quantification in comparison to histology for the diagnosis of renal transplant fibrosis. Clin Hemorheol Microcirc 2010;46:139–148. Syversveen T, Brabrand K, Midtvedt K, Strøm EH, Hartmann A, Jakobsen JA, Berstad AE. Assessment of renal allograft fibrosis by acoustic radiation force impulse quantification—a pilot study. Transpl Int 2011;24:100–105. Tao JL, Li H, Li C, Xu XW, Li JF, Yi N, Liu DY, Qin Y, Cai JF, Liu BY, Xu H, Gao RT, Ye WL, Ye W, Li XM, Li XW. Risk factors of postrenal biopsy bleeding. Acta Acad Med Sin 2008;30:313–317. Tublin ME, Bude RO, Platt JF. The resistive index in renal Doppler sonography: Where do we stand? AJR 2003;180:885–892. Wang CL, Shu KH, Lan JL, Cheng CH, Wu MJ, Chen CH. Duplex Doppler sonography to predict response to therapy in active lupus nephritis. Kuang Tien Med J 2007;2:21–27. Weening JJ, D’Agati VD, Schwartz MM, Seshan SV, Alpers CE, Appel GB, Balow JE, Bruijn JA, Cook T, Ferrario F, Fogo AB, Ginzler EM, Hebert L, Hill G, Hill P, Jennette JC, Kong NC, Lesavre P, Lockshin M, Looi LM, Makino H, Moura LA, Nagata M. The classification of glomerulonephritis in systemic lupus erythematosus revisited. J Am Soc Nephrol 2004;15:241–250. Wen YK. Renal biopsy findings in new-onset systemic lupus erythematosus with clinical renal disease. Int Urol Nephrol 2011;43: 801–806. Weiztel WF, Kim K, Rubin JM, Xie H, O’Donnell M. Renal advances in ultrasound elasticity imaging: Measuring the compliance of arteries and kidneys in end-stage renal disease. Blood Purif 2005;23:10–17. Zhang XL, Jin M, Wu HF, Nadasdy T, Nadasdy G, Harris N, Green-Church K, Nagaraja H, Birmingham DJ, Yu CY, Hebert LA, Rovin BH. Biomarkers of lupus nephritis determined by serial urine proteomics. Kidney Int 2008;74:799–807.