WITHDRAWN — 2026-09-09
This page previously served a pooled estimate for this question. That estimate has been withdrawn. The original page is preserved byte for byte and is linked below; nothing has been deleted.
No pooled estimate stands for this question. This page publishes none: the previously served pool has been withdrawn and nothing replaces it.
The pooled estimate OR 0.9074 (95% CI 0.831 to 0.9908) across two trials,
and the claim of a statistically significant reduction that its interval carried by
excluding no effect.
It does not. This is not a presentational correction, a re-rendering, or a restatement of the same result in different terms. The estimate is withdrawn because the object that produced it records, in the same outcome, that the quantities required to make it meaningful were never established.
The object underlying this page holds, in the same outcome block that carried the pooled value:
pool_uniformity.effect_measure[0] "NOT ESTABLISHED" pool_uniformity.estimand[0] "NOT ESTABLISHED" THE_POOL_IS_REFERRED_2026_08_20.state "REFERRED, NOT WITHDRAWN"
And its stated warrant for pooling the trials at all reads, verbatim:
"The extractor recovered a pooled result from this page, so the page pooled these trials."
The page was its own justification. The reason recorded for combining these trials is that a pooled number had already appeared on this page. No source outside this project's own prior output was consulted for it. A conclusion whose only warrant is that it was previously published cannot support a claim about the world, and the interval above should not have been read as evidence of an effect.
Two further properties of the withdrawn estimate, recorded in the same object:
The withdrawn estimate pooled two trials. A published systematic review already cited by this review — Kluger et al. 2018, Cardiorenal Outcomes in the CANVAS, DECLARE-TIMI 58, and EMPA-REG OUTCOME Trials — names three, and CANVAS was not among the two pooled here. A re-derivation of the question additionally reaches VERTIS-CV.
| trial | in the withdrawn pool |
|---|---|
| EMPA-REG OUTCOME (NCT01131676) | yes |
| DECLARE-TIMI 58 (NCT01730534) | yes |
| CANVAS (NCT01032629) | no |
| VERTIS-CV (NCT01986881) | no |
A protocol-faithful re-derivation of this question, built from a protocol registered before it ran and using only values fetched from trial registrations, reaches four trials and gives, under the pre-declared estimator and interval method:
HR 0.9159 (95% CI 0.8242 to 1.0178) — which does not exclude no effect.
THAT RE-DERIVATION IS NOT PUBLISHED AND IS NOT PROMOTED. It has not cleared review and is stated here only so that this withdrawal is not read as implying the opposite result is established. The claim is withdrawn; no replacement claim is being made in its place. A withdrawal that quietly substitutes an unpublished number would repeat, in the other direction, the error being corrected.
Preserved unmodified at archive/SGLT2_MACE_CVOT_REVIEW_2026-09-09.html.
bytes 1742860 sha256 9eb8a23723dc0830fbf3326107845f84086f2e07d90b55aeeac39d40c4491385 git blob (unfiltered) 7e79f52a67246f50f59f0dc79303187f37c280f8
The archived copy is byte-identical to what this URL served immediately before this notice. It is kept because a withdrawal is itself evidence: a reader who followed a citation to the withdrawn estimate must be able to see exactly what was claimed, and a deletion would erase that record and break every inbound link.
Withdrawn 2026-09-09. This notice replaces the page at its canonical address; the address itself is unchanged.