Frontiers in Endocrinology · Oct 30, 2026
The HERMES Study: Prospective Assessment of Hepatic and Renal Stiffness Using Two-Dimensional Shear Wave Elastography Following Six Months of SGLT2 Inhibitor Therapy in Adults with Type 2 Diabetes
By Mehmet Turan Ozer, Ahmet Altintas, Omer Tepe et al.
Background. Type 2 diabetes mellitus (T2DM) affects the liver and kidneys. Although sodium–glucose cotransporter-2 (SGLT2) inhibitors have established cardiovascular and renal benefits, prospective evidence on longitudinal change in hepatic and renal stiffness is limited, and concurrent kidney-damage markers are rarely reported. Methods. In this prospective, single-centre, paired-cohort observational study, adults with T2DM initiating SGLT2 inhibitor therapy in routine care underwent two-dimensional shear wave elastography (2D-SWE) of the liver and both renal cortices at baseline and six months, with identical equipment, settings, operator and protocol at both visits. Serial laboratory measurements and spot urine microalbumin were obtained; kidneys were analysed separately. Paired t-tests or Wilcoxon signed-rank tests with Hodges–Lehmann confidence intervals were used; laboratory analyses were Benjamini–Hochberg corrected. No sample-size calculation was performed. Results. In 54 participants, hepatic stiffness fell from 16.62 ± 5.06 to 12.81 ± 5.97 kPa (median change −3.80 kPa, 95% CI −4.94 to −2.75; −22.9%; p < 0.001), right renal stiffness from 13.47 ± 2.95 to 11.44 ± 3.01 kPa (mean −2.03 kPa, −2.67 to −1.39; −15.1%; p < 0.001) and left renal stiffness from 13.24 ± 3.07 to 10.54 ± 2.42 kPa (median −2.33 kPa, −3.26 to −1.64; −20.5%; p < 0.001; n = 53), with reductions in 92.6%, 87.0% and 98.1% of participants. Eleven laboratory parameters changed after correction, and 77–93% of the six-month glycaemic and insulin-secretory change was already present at three months. Spot urine microalbumin (median 12.5 to 12.0 mg/L; p = 0.81) and eGFR did not change, and change in renal stiffness was unassociated with either (ρ = +0.03, −0.07 and −0.15). Of 45 stiffness-versus-laboratory correlations, four were nominally significant but none survived correction; change in hepatic and in left renal stiffness was unassociated with every metabolic parameter. Conclusions. Stiffness fell substantially in the liver and both kidneys, but albuminuria and eGFR did not change and neither correlated with it; the reduction cannot be attributed to reduced kidney damage. Haemodynamic, volume-related and metabolic mechanisms are at least as plausible as structural change. 2D-SWE should be evaluated within a multiparametric hepatorenal ultrasound phenotype.