Abstract
Background and Aims
Current assessment of diabetic kidney disease (DKD) is limited to estimated glomerular filtration rate (eGFR) and albuminuria. These are inadequate as DKD often has heterogenous clinical phenotypes. There is need for a marker of intra-renal fibrosis. Native kidney biopsy remains the only reference method in clinical practice for this purpose, but is invasive and impractical for repeated evaluations. Recently, two-dimensional ultrasound shear wave elastography (SWE) has emerged as a non-invasive technique to assess renal parenchymal stiffness with renal fibrosis. We aim to investigate SWE-derived estimates of tissue stiffness with different DKD stages in an Asian population.
Method
In this cross-sectional pilot study, 58 patients with DKD were recruited from a single centre ambulatory Nephrology clinic. Laboratory values were taken within 1 week of undergoing SWE, with DKD staging by the Kidney Disease Improving Global Outcomes (KDIGO) guidelines and eGFR calculated using the CKD-EPI equation. 13 patients had histological diagnoses of DKD; 2 (15.3%) Stage G1-2; 5 (38.5%) Stage G3; 5 (38.5%) Stage G4 and 1 (7.7%) Stage G5 subjects, with native kidney biopsies performed within 4 months of study recruitment.
2D SWE was performed with a 2-5 MHz transducer on an Axiplorer© ultrasound system (Supersonic Imagine, Paris) by a single Nephrologist blinded to laboratory results. Using a previously described protocol, 6 SWE measurements were taken from the cortical mid-pole of bilateral kidneys, and renal elasticity estimated as Young’s Modulus (YM) in kilopascals (kPa), (Figure 1).
Results
Study population were 62.1% male (36/58) and 62.1% ethnic Chinese (36/58), with diabetes duration of 11.7 ± 9.2 years. Median eGFR was 35.0 (40-101) mL/min per 1.73 m2, with 6 (10.3%) DKD Stage G1-2; 34 (58.6%) Stage G3; 13 (22.4%) Stage G4 and 5 (8.6%) Stage G5 patients.
There were moderate correlations between YM values in bilateral kidneys. Left kidney maximal YM generally increased in accordance with DKD stage (Stage G1-2: 20.6 kPa, Stage G3A: 13.5 kPa, Stage G3B: 22.4 kPa, Stage G4-5: 30.9 kPa, p <0.01). Kidney depth correlated moderately with body mass index (BMI). After controlling kidney depth and BMI, there was a moderately positive correlation between right kidney YM and DKD stage (Maximal YM; r = 0.4, p < 0.01, Mean YM; r = 0.31, p = 0.02). eGFR negatively correlated with bilateral kidney maximal YM (right r = -0.2, p = 0.04, and left r = -0.3, p = 0.03, respectively). Importantly, there was a strong correlation between right kidney mean YM and histological grading of interstitial fibrosis and tubular atrophy (r = 0.9, p = 0.01). There is no correlation between kidney elasticity and percentage of sclerosed glomeruli.
Using a cut-off of 13.5 kPa for mean estimated tissue YM, the area under the receiver operator curve was 0.8 to distinguish DKD Stage G1 and G2 from G3A (sensitivity 83.3%, specificity 80.0%).
Conclusion
SWE-derived estimates of renal stiffness appear to increase with DKD stage. The strong correlation with histological markers of fibrosis indicate that observed differences are due to renal parenchymal stiffness. SWE shows promise as a non-invasive marker of renal fibrosis, although large multi-centre studies are required to validate these findings.
Perspectives on the Role of Magnetic Resonance Imaging (MRI) for Noninvasive Evaluation of Diabetic Kidney Disease
