SGLT2i improves kidney outcomes regardless of an eGFR dip

Introduction and methods

News - June 8, 2020

Kidney implications of the initial eGFR response to SGLT2 inhibition with empaglifozin: the 'eGFR dip' in EMPA-REG OUTCOME

Presented at the virtual ERA-EDTA congress by Bettina J Kraus (Würzburg, Germany)

The SGLT2 inhibitor empagliflozin reduces kidney risk in patients with T2DM and established CVD. Treatment with empagliflozin induces an initial dip in eGFR. This dip has been observed across the class of SGTL2 inhibitors. Although largely reversible, this initial dip has caused concerns in some clinical settings.

Aims of this post hoc analysis of EMPA-REG OUTCOME were to characterize the initial eGFR dip with empagliflozin use, determine baseline factors that predict an eGFR dip, describe kidney safety and outcomes across predictive factors, and investigate whether the initial eGFR dip affects reduction on kidney outcomes by empagliflozin treatment.

Patients were categorized by % initial eGFR change from baseline to week 4 as eGFR dipper: >10% decline, eGFR intermediate: >0 to ≤10% decline, eGFR non-dipper: no decline.

Main results

  • Empagliflozin causes a shift towards more eGFR dipper >10% (28% in patients on empagliflozin vs. 13% in patients on placebo). Patients experiencing eGFR decline >30% was 1.4% in patients on empagliflozin and 0.9% patients on placebo.
  • Mean baseline eGFR was higher in the eGFR intermediate group compared to the eGFR non-dipper group and the eGFR dipper group had a lower baseline eGFR than the other two groups.
  • For all 3 groups there was a rebound of eGFR at 12 weeks followed by a uniform stabilization of eGFR for the rest of the treatment period.
  • After discontinuation of treatment (median: 3 years), eGFR increased at follow-up in all three categories, whereas eGFR remained unchanged in those who took placebo.
  • Diuretic therapy use and higher KDIGO risk category were predictive of eGFR dip >10% (OD 4.4, 95%CI: 2.3-8.5 in those on diuretic therapy and very high KDIGO risk).
  • Serious adverse events and kidney adverse events were consistent for those on placebo vs. empagliflozin for across subgroups of diuretic therapy (yes/no) and KDIGO risk (low-very high).
  • The composite outcome of incident or worsening nephropathy and hard kidney outcome were consistently reduced with empagliflozin compared to placebo in those with dipping OR ≤2.7 and dipping OR >2.7.
  • An eGFR dip did not have a major effect on risk reduction for incident or worsening nephropathy following week 4. HR with adjustment for eGFR dip >10% from baseline to week 4 was 0.58 (95%CI: 0.50-0.66) and HR with the primary model was 0.61 (95%CI: 0.53-0.69).


This post-hoc analysis of the EMPA-REG OUTCOME showed that approximately 1 in 4 patients on empagliflozin experienced an eGFR dip >10% . Second, diuretic therapy use and higher KDIGO risk were independent predictors for an initial eGFR dip >10% with empagliflozin. Third, empagliflozin was safe across subgroups with higher predictive OR for an eGFR dip. Finally, empagliflozin improved kidney outcomes regardless of an eGFR dip.


In response to a question about the effect of RAASi on dipping, Kraus answered they studied this. First, she said, it is important to note that ~80% of patients in the EMPA-REG OUTCOME trial were using RAAS inhibitors at baseline. Analysis of predictive value of use of ACEi/ARBs showed there was an increased risk of dipping in those on ACEi/ARBs, but there was no interaction with empagliflozin treatment.

There was a question about when to worry with regard to eGFR dip. Kraus said to always take the clinical setting into account and exclude other causes such as AKI. According to her, <20% change in eGFR can be regarded as within the clinical range. In this subanalysis, there were no safety signals with a cut-off of 10%.

‘Is it necessary to adjust the dose of loop diuretics after initiating SGLT2 inhibitors’, someone asked. Kraus replied no difference in volume depletion has been observed between those on empagliflozin vs. placebo, but volume depletion has to be taken into account when adding SGLT2i in a patients on diuretic therapy.

- Our reporting is based on the information provided at the ERA-EDTA congress -

Facebook Comments


We’re glad to see you’re enjoying PACE-CME…
but how about a more personalized experience?

Register for free