Reviews/SGLT2 Inhibitors/SGLT2 inhibitor kidney benefit is greatest in patients with high albuminuria
· SG-CK-03 · v2
Borderline: within 4 points of the Likely/Uncertain boundary, so a re-grade could land either side.
The one study that directly compares SGLT2 inhibitor kidney benefit across different starting levels of albuminuria [6] finds that the largest absolute benefit does fall to patients with the most severe albuminuria, so the claim has real support. But the same study shows the size of the drug's proportional effect is much the same whatever the starting albuminuria level. A large part of why high-albuminuria patients gain more in absolute terms is simply that their kidneys are declining faster to begin with [9], so there is more harm to prevent. Meaningful benefit is also seen in people with low or normal albuminuria [15][17], and one analysis even found a low-albuminuria subgroup with greater absolute benefit than its peers [10]. The panel therefore treats the claim as sound when read on the absolute, clinically relevant scale, but as over-reaching if taken to mean benefit is confined to high albuminuria.
Only [6] directly stratifies renal outcomes by baseline albuminuria, concluding that proportional effects are mostly consistent across strata while absolute benefits are greatest among those with severely increased albuminuria (P heterogeneity=0.004 for the renal composite), with no effect modification when albuminuria is fitted as a continuous variable (P heterogeneity=0.94). The risk gradient that drives the absolute finding is confirmed in [9], where the higher-albuminuria group had an eGFR slope of -2.44 mL/min/1.73 m2 per year against a flat trajectory in the lower group. CREDENCE [3] shows a strong renal-composite effect (hazard ratio 0.66) in a uniformly high-albuminuria population but offers no across-strata comparison. Counter-evidence shows benefit outside high albuminuria: eGFR-slope benefit independent of UACR [15], a clear renoprotective effect at UACR 10-30 [17], and a uACR<3 subgroup with greater absolute risk reduction [10]. Several included papers measure albuminuria or proteinuria reduction rather than hard outcomes stratified by baseline albuminuria [1][18]. The contested point: The division is scale-dependent: on the absolute scale [6] supports 'greatest' benefit in severe albuminuria, but on the relative scale the treatment effect is consistent across strata (P heterogeneity=0.94). The second fault line is whether larger absolute benefit reflects a larger drug effect or merely the faster baseline progression of higher-albuminuria patients [9], and whether benefit is confined to high albuminuria at all given clear effects at low/normal levels [10][15][17].
All 20 papers behind this claim with their DOIs and what each does to it, the panel’s arbitration, the evidence gap map and the open research questions. Free, and sent to your inbox.
We’ll also send you other claims from SGLT2 Inhibitors, one at a time, and tell you when a release or a pivotal trial changes a grade. Unsubscribe by replying.
One reviewer's only job is to attack the finding. The challenges below were weighed before the grade was set.
The only directly stratified analysis [6] states proportional effects are 'mostly consistent across patients with different levels of albuminuria' with no effect modification when albuminuria is fitted as a continuous variable (P heterogeneity=0.94); the claim's unqualified 'greatest' is true only on the absolute scale, not the relative treatment effect.
Benefit is demonstrably not confined to high albuminuria: [15] found eGFR-slope benefit 'independent of UACR', [17] showed a clear renoprotective effect at UACR 10-30 and benefit at UACR<30, and [10] identified a uACR<3 subgroup with greater absolute risk reduction than other low-albuminuria patients.
Where it remains uncertain. Proportional effects are mostly consistent across albuminuria levels with no modification when albuminuria is continuous (P heterogeneity=0.94) [6], and benefit persists at low/normal albuminuria [10][15][17], so the larger absolute benefit tracks higher baseline risk [9] rather than a larger pharmacological effect.
Aleatory
Effects at low/normal albuminuria rest on small user counts ([17]: 97 users across UACR<10 and 10-30 strata; [15]: 354 participants), so subgroup estimates carry substantial sampling variability.
Epistemic
No primary trial in the corpus was powered to test whether baseline albuminuria modifies kidney benefit; [6] is a post-hoc subgroup analysis and [2] CONFIDENCE is a design paper reporting no results.
Inferential
The claim is true on the absolute scale [6] but not the relative scale (P heterogeneity=0.94); reading one as the other changes the verdict.
The load-bearing papers from a corpus of 20, each classified by what it does to the claim. Every DOI is taken from the corpus the claim was graded on.
Demonstrates substantial renal benefit in a trial enrolling exclusively high-albuminuria patients (UACR >300 to 5000).
“The relative risk of the renal-specific composite of end-stage kidney disease, a doubling of the creatinine level, or death from renal causes was lower by 34% (hazard ratio, 0.66; 95% CI, 0.53 to 0.81; P<0.001)”
Panel auditInternal validity: high · scope of inference: Kidney and cardiovascular outcomes in T2D with albuminuric CKD (UACR >300-5000) on RAS blockade.Key limitations: no albuminuria-stratum comparison · single high-albuminuria population · stopped early
Direct, quantified support for the claim on the absolute scale, with the explicit qualifier that relative/proportional effects do not differ by albuminuria level.
“Greater absolute risk reductions in the renal composite outcome were observed in participants with severely increased albuminuria ( P heterogeneity=0.004).”
Panel auditInternal validity: high · scope of inference: Renal and cardiovascular effects of canagliflozin by baseline albuminuria in high-CV-risk T2D.Key limitations: post-hoc · proportional effects consistent (benefit is absolute not relative) · few severe-albuminuria events (n=760)
Establishes the higher baseline risk that makes absolute benefit larger in high-albuminuria patients.
“It is well known that renal disease progresses more quickly in patients with higher degree of albuminuria than in those with lower levels.”
Panel auditInternal validity: low · scope of inference: Residual renal risk by albuminuria level in CKD patients on RASi plus dapagliflozin.Key limitations: no untreated comparator · quotes trial results secondhand · residual-risk not benefit-magnitude design
Counter-evidence: eGFR-slope benefit in non-diabetic CKD did not depend on baseline albuminuria, tempering an unqualified reading of the claim.
“This effect was independent of UACR.”
Panel auditInternal validity: low · scope of inference: Real-world eGFR-slope effect of SGLT2i in older non-diabetic CKD patients across UACR levels.Key limitations: no control arm · interaction not formally tested · small non-diabetic cohort
4 of 20 papers shown. The full review lists all of them, with the panel’s arbitration →
Claims are graded at each quarterly release, and re-graded between releases only when a pivotal trial lands. Every change is a new version of the claim.
Directly supported on the absolute scale by the single stratified RCT analysis and the risk gradient, but qualified by consistent relative effects and clear benefit in low albuminuria, so conservatively Likely rather than Established.
Release 2026-q4 re-graded every claim from its corpus under protocol v0.19. An earlier release's grade is shown for the record; it came from an earlier protocol and is not a movement.
Assumptions that carry weight but are not themselves established.
The panel records the gap that would most move the claim, and the study that would close it.
No prospective test of effect modification by baseline albuminuria. The claim's gradient rests on a single post-hoc subgroup analysis [6]; the corpus contains no trial prospectively designed to test whether baseline albuminuria modifies hard kidney-outcome benefit. The panel rates this the highest-importance open gap for the claim, and names the study design that would close it: A randomised trial or prospectively pre-specified pooled analysis stratifying hard renal-composite outcomes by baseline albuminuria, powered for interaction on both absolute and relative scales.
Open question: Is the greater absolute kidney benefit in high-albuminuria patients a larger drug effect or a consequence of their faster baseline progression? Why it is still open: [6] shows consistent proportional effects (P heterogeneity=0.94) while [9] shows high-albuminuria patients decline faster (-2.44 versus a flat trajectory), so the corpus cannot separate the two.
Priors is a sceptical first reader, not peer review, and not medical advice.
Release 2026-q4 · graded 23 Oct 2026 · protocol v0.19