Derivation Web

source_66baec80631f4d82

by researka:v2 · 2026-07-30 10:36:00.500750+04:00

# Research Synthesis: SGLT2 Inhibitors Subgroups
## Abstract

Evidence scope: 9/21 retained sources are indirect, review-level, adjacent, or mechanistic and are used only to bound interpretation. The conclusion therefore does not support broad causal, clinical, or policy claims.

This paper synthesizes evidence on SGLT2 inhibitors subgroups across 21 included source papers and 1084 high-confidence extracted claims.

The evidence profile contains 12 direct clinical sources, 8 adjacent, review, or context sources, and 1 mechanistic or model-system source, with a high-density pairwise disagreement map across the evidence base.

Positive study-level signals are summarized in the cardiometabolic and longevity outcome classes, null signals in the longevity, contextual adjacent evidence and cardiometabolic outcome classes, and negative signals in no dominant outcome class. The paper therefore reports a source-directness and outcome-class map rather than a pooled effect.

The conclusion is that SGLT2 inhibitors subgroups remains a bounded evidence case: the retained clinical and mechanistic evidence profile defines the scope for targeted testing, while mixed and null findings limit any unqualified broad clinical claim.

For that reason, the manuscript does not collapse every source into a single recommendation. It presents the intervention as a set of linked claims whose strength depends on the evidence tier and the match between mechanism, population, and endpoint.

## Introduction

This synthesis evaluates evidence on SGLT2 inhibitors subgroups across 21 included source papers and 1084 high-confidence extracted claims. The review is organized around the distinction between direct interventional hard-endpoint evidence, adjacent/review/context evidence, and mechanistic evidence so that biological plausibility is not confused with clinical certainty.

The corpus contains 12 direct clinical sources, 8 adjacent, review, or context sources, and 1 mechanistic or model-system source. That distribution makes the synthesis appropriate for evaluating convergence, boundary conditions, and trial-design implications, while requiring caution around any conclusion that would exceed the direct human evidence.

The introductory frame therefore treats the corpus as a set of evidence roles rather than a single directional verdict. Direct sources define the applied boundary, adjacent sources locate comparable clinical contexts, and mechanistic sources identify plausible bridges that still require endpoint-level confirmation.

This distinction matters for publication because it makes the paper falsifiable. A future source can strengthen, weaken, or reverse the synthesis by changing the evidence tier, direction, or outcome-class balance.

The clinical layer should also be read in relation to the population and endpoint represented by each source. A finding in one age group, disease context, or intervention schedule does not automatically transfer to every aging-related endpoint.

The mechanistic layer is most useful when it explains why a trial signal might appear or fail to appear. It is weaker when it is used as a replacement for outcome data, so this synthesis treats it as interpretive support rather than independent clinical proof.

Null findings have a specific role in this evidence model. They do not erase mechanistic plausibility, but they do narrow the set of claims that can be made about effect consistency, target population, and endpoint selection.

Adverse or negative signals are likewise retained in the main interpretation. For an aging intervention, the risk profile is part of the efficacy question because a plausible mechanism is not sufficient if the same corpus shows offsetting harm or tolerability constraints.

The evidence base also distinguishes breadth from certainty. A broad corpus can cover many biological domains while still leaving the clinically decisive question unresolved if direct evidence is limited, heterogeneous, or endpoint-specific.

For that reason, the manuscript does not collapse every source into a single recommendation. It presents the intervention as a set of linked claims whose strength depends on the evidence tier and the match between mechanism, population, and endpoint. In introduction, interpretation remains limited to the retained endpoint-specific findings. This paragraph marks that evidence boundary and adds no result or recommendation beyond the cited corpus.

## Background

The background evidence for SGLT2 inhibitors subgroups is heterogeneous rather than uniformly confirmatory.

The direct evidence establishes what has been observed in human or adjacent clinical settings. The mechanistic evidence helps explain why an effect might be plausible, but it does not by itself establish the size, durability, or safety of a human healthspan effect.

Across the retained sources, positive signals cluster around the cardiometabolic and longevity outcome classes; null signals around the longevity, contextual adjacent evidence and cardiometabolic outcome classes; and negative or adverse signals around no dominant outcome class. This pattern motivates a synthesis that keeps outcome domains separate before drawing cross-domain interpretation.

Interpretation is deliberately scoped to the retained corpus. Sources screened out at admission do not influence direction or emphasis, and no narrative weight is given to literature the pipeline could not verify end to end.

Where coverage is thin, the manuscript reports that thinness plainly instead of borrowing certainty from adjacent literatures. Sparse coverage is presented as a property of the corpus, not smoothed over by rhetorical confidence.

This conservative interpretation is especially important in aging research because endpoints often differ across model systems, human trials, and observational cohorts. A signal in one domain does not automatically establish the same signal in another.

The study-level structure also prevents selective emphasis. Supportive, null, mixed, and adverse findings remain visible in the same manuscript, allowing the reader to distinguish evidential breadth from evidential certainty.

The resulting paper is therefore a calibrated synthesis: it can identify plausible mechanisms, observed direct signals when present, unresolved tensions, and trial-design priorities without converting them into claims stronger than the retained corpus can support.

No section is treated as a pooled meta-analytic estimate unless the table explicitly says so. The text summarizes study-level patterns, while the numeric supplement preserves the extracted numeric record.

## Methods

### Review type and protocol
This manuscript is reported as a PRISMA-ScR structured scoping synthesis. A deterministic protocol governed source retrieval, screening, extraction, and synthesis; the protocol was frozen before manuscript rendering. The full audit trail is in the supplementary `methods_pack.json` and the timestamped submission directory `synthesis-sglt2_inhibitors_subgroups-v06-DAILY-2026-07-30T06-17-44Z`.

### Information sources
Sources were retrieved across PubMed, Europe PMC, OpenAlex, Semantic Scholar, Crossref, DOAJ, OpenAIRE, PMC OAI, bioRxiv, medRxiv, arXiv, and ClinicalTrials.gov. Retrieval window: 2026-07-30.

### Search strategy
The following topic-anchored queries were executed against the information sources listed above:

- `SGLT2 inhibitors subgroups aging`
- `SGLT2 inhibitors subgroups older adults`
- `SGLT2 inhibitors subgroups randomized controlled trial`
- `SGLT2 inhibitors aging`
- `SGLT2 inhibitors older adults`
- `SGLT2 inhibitors randomized controlled trial`

### Eligibility criteria
- Sources whose primary content addresses sglt2 inhibitors subgroups.
- Sources with extractable quantitative or qualitative findings.
- Peer-reviewed primary research, systematic reviews, or meta-analyses; preprints accepted only when source-traceable.
- Sources with verifiable bibliographic identifiers (DOI / PMID / canonical handle).

### Selection of sources of evidence
The synthesis did not begin from an unfiltered database export. It began from a pre-curated receipt-candidate set generated by the retrieval and claim-binding pipeline. Of 173 records in the receipt-candidate union, 53 were classified as source candidates and 21 were admitted as traceable synthesis sources. Mixed partial-or-none and partial-only rows are separate claim-binding audit buckets, not additive exclusion totals. No additional records were excluded after final source admission.

### source admission funnel

| Admission bucket | n |
|---|---:|
| source candidate union | 173 |
| Classified source candidates | 53 |
| No extractable claims | 22 |
| None-only claim binding | 11 |
| Mixed partial-or-none claim-binding candidates | 49 |
| Partial-only claim-binding candidates | 24 |
| Strict high-confidence sources | 14 |
| Admitted final sources | 21 |

### Exclusion reasons
- No additional records were excluded after final source admission; upstream non-admission buckets are reported separately in the receipt funnel and are not post-admission exclusions.

### Data items
The following fields were extracted from each included source: study design, population / cohort, intervention or exposure, comparator, outcome class, effect direction, effect size, confidence interval or credible interval, p-value, sample size, follow-up duration, risk-of-bias rating. Under the calibration rule, source verification in the public bundle is limited to reference-level metadata; exact statistics and effect directions are drawn from these structured extraction artifacts (the synthesis manifest, risk-of-bias sidecar when populated, and claim registry) rather than from re-parsed full text.

### Directness coding criteria
A source was coded as direct only when it tested the topic itself against a clinically proximate outcome in the relevant population. Human evidence with an adjacent exposure, population, or outcome was coded as indirect; syntheses and secondary reviews were coded as review-level evidence and were not counted as direct sources.

### Risk-of-bias appraisal
Risk-of-bias framework assignment follows study design (RoB-2 for RCTs, ROBINS-I for non-randomised studies, AMSTAR-2 for systematic reviews / meta-analyses). Public appraisal claims are limited to populated `risk_of_bias.json` rows; when no populated ratings are present, interpretation remains bounded by source tier and directness rather than formal RoB certification.

### Synthesis approach
Evidence-tension synthesis: claims grouped by outcome class (cardiometabolic, contextual adjacent evidence, dosing and pharmacokinetics, immune and inflammation, longevity, muscle function, safety and comorbidity); within-class agreement, disagreement, and directness gaps surfaced explicitly. Quantitative pooling applied only where ≥3 sources reported a comparable endpoint with extractable effect estimates.

### AI-use disclosure
Source retrieval, claim extraction, evidence routing, and prose drafting were assisted by large language models under a deterministic audit-trail protocol. Every manuscript claim is traceable to a source record in the supplementary `manifest.json`. Final eligibility and interpretation decisions are author-verified.

### Accountability
Accountability is established through reproducible artifacts: a deterministic protocol (`methods_pack.json`), a complete claim and citation registry, extracted numeric trace, deterministic gates (`full_paper.journal_surface.json`, `pre_submit_gate.json`, `artifact_consistency.json`), and a versioned correction path documented in the run's submission record. Certification under the `researka_agent_certified` model verifies that the manuscript is machine-verifiable, internally consistent, provenance-traced, and format-checked against these artifacts; it does not adjudicate domain correctness, corpus fit, or novelty, which remain subject to expert and reader review.

## Evidence Landscape

### Findings Map

Findings Map completeness note: all 21 admitted manifest rows are surfaced below; outcome class follows endpoint/source context before topic keywords.

Findings Map accounting note: each outcome-class n, direction count, directness count, and source roster is computed from the same source-level rows listed in the detailed table. source-level direction is not a statement that the source abstracts lack directional statistics; it is the conservative coded polarity used for synthesis accounting.

| Evidence domain | Source | Direction | Directness | Tier | Evidence role | Finding |
| --- | --- | --- | --- | --- | --- | --- |
| Animal/Preclinical Context (Cardiometabolic) | Nobayashi 2026: Comparison of SGLT2 Inhibitors for New‐Onset Proteinuria Risk in Patients With Type 2 Diabetes and Preserved Kidney Function | direction=null | directness=animal/preclinical context | A1 | outcome=Animal/Preclinical Context (Cardiometabolic); direction=null | finding=54 extracted claim(s); source-level direction is the coded finding |
| Cardiometabolic | Ahmed 2025: SGLT2 inhibitors versus GLP-1 receptor agonists for major adverse cardiovascular events in type 2 diabetes: a systematic review and meta-analysis of randomized controlled trials | direction=mixed | directness=review | B2 | outcome=Cardiometabolic; direction=mixed | finding=representative statistic P = 0.0003; source-level statistic reported |
| Cardiometabolic | Huang 2026: A Prospective Cohort Study on the Impact of SGLT2 Inhibitors on the 12‑Month Recurrence Risk of Atrial Fibrillation After Catheter Ablation | direction=positive | directness=direct | A1 | outcome=Cardiometabolic; direction=positive | finding=representative statistic P = 0.019; source-level statistic reported |
| Cardiometabolic | Jansz 2026: Kidney outcomes with GLP-1 receptor agonists in people with type 2 diabetes already receiving SGLT2 inhibitors: a target trial emulation study using UK primary care data | direction=positive | directness=direct | A1 | outcome=Cardiometabolic; direction=positive | finding=representative non-significant statistic P = 0.11; not treated as positive or negative directional support unless source direction is coded |
| Cardiometabolic | Kaze 2022: Association of SGLT2 inhibitors with cardiovascular, kidney, and safety outcomes among patients with diabetic kidney disease: a meta-analysis | direction=unclear | directness=review | B1 | outcome=Cardiometabolic; direction=unclear | finding=76 extracted claim(s); source-level direction is the coded finding |
| Cardiometabolic | Kochanowska 2026: Evidence on SGLT2 Inhibitors’ Efficacy in Older and Frail Patients | direction=unclear | directness=direct | A1 | outcome=Cardiometabolic; direction=unclear | finding=representative non-significant statistic P > 0.05; not treated as positive or negative directional support unless source direction is coded |
| Cardiometabolic | Moreno-Perez 2026: Inpatient safety, effectiveness of SGLT2 inhibitors and GLP-1 RAs in type 2 diabetes: ENDOCARE, a pragmatic prospective cohort study | direction=positive | directness=direct | A1 | outcome=Cardiometabolic; direction=positive | finding=representative statistic P < 0.001; source-level statistic reported |
| Cardiometabolic | Sato 2025: Adding Semaglutide to SGLT2 Inhibitors Reduces Liver Enzymes in Patients with Type 2 Diabetes Complicated by Metabolic Dysfunction Associated Steatotic Liver Disease: A Retrospective Observational Study | direction=positive | directness=direct | A1 | outcome=Cardiometabolic; direction=positive | finding=representative statistic P = 0.004; source-level statistic reported |
| Cardiometabolic | Suciu 2025: Do SGLT2 Inhibitors Improve Cardiovascular Outcomes After Acute Coronary Syndrome Regardless of Diabetes? A Systematic Review and Meta-Analysis | direction=positive | directness=review | B1 | outcome=Cardiometabolic; direction=positive | finding=78 extracted claim(s); source-level direction is the coded finding |
| Contextual Adjacent Evidence | Albulushi 2025: Impact of SGLT2 inhibitors on myocardial fibrosis in diabetic HFpEF: a longitudinal study | direction=unclear | directness=direct | A1 | outcome=Contextual Adjacent Evidence; direction=unclear | finding=representative statistic P < 0.001; source-level statistic reported |
| Contextual Adjacent Evidence | Kawanami 2017: SGLT2 Inhibitors as a Therapeutic Option for Diabetic Nephropathy | direction=null | directness=direct | A1 | outcome=Contextual Adjacent Evidence; direction=null | finding=1 extracted claim(s); source-level direction is the coded finding |
| Contextual Adjacent Evidence | Szklarz 2026: Another pleiotropic effect of SGLT2 inhibitors: Is it a new frontier in thyroid function regulation? | direction=null | directness=indirect | B2 | outcome=Contextual Adjacent Evidence; direction=null | finding=3 extracted claim(s); source-level direction is the coded finding |
| Dosing and Pharmacokinetics | Chen 2026: Agent- and Dose-Specific Intestinal Obstruction Safety of GLP-1 Receptor Agonists and SGLT2 Inhibitors: A Network Meta-Analysis of Randomized Trials | direction=null | directness=review | B2 | outcome=Dosing and Pharmacokinetics; direction=null | finding=10 extracted claim(s); source-level direction is the coded finding |
| Immune and Inflammation | Correale 2021: Switch to SGLT2 Inhibitors and Improved Endothelial Function in Diabetic Patients with Chronic Heart Failure | direction=unclear | directness=direct | A1 | outcome=Immune and Inflammation; direction=unclear | finding=representative statistic P < 0.001; source-level statistic reported |
| Longevity | Elrakaybi 2022: Cardiovascular protection by SGLT2 inhibitors – Do anti-inflammatory mechanisms play a role? | direction=null | directness=direct | A1 | outcome=Longevity; direction=null | finding=6 extracted claim(s); source-level direction is the coded finding |
| Longevity | Gao 2026: Repurposing SGLT2 Inhibitors for Cirrhotic Ascites: From Mechanistic Research to Clinical Exploration | direction=null | directness=direct | A1 | outcome=Longevity; direction=null | finding=1 extracted claim(s); source-level direction is the coded finding |
| Longevity | Jiang 2022: Comparative Cardiovascular Outcomes of SGLT2 Inhibitors in Type 2 Diabetes Mellitus: A Network Meta-Analysis of Randomized Controlled Trials | direction=positive | directness=review | B1 | outcome=Longevity; direction=positive | finding=121 extracted claim(s); source-level direction is the coded finding |
| Longevity | Rawish 2026: SGLT2 inhibitors are associated with improved long-term survival in Takotsubo syndrome: insights from large-scale real-world data | direction=unclear | directness=direct | A1 | outcome=Longevity; direction=unclear | finding=67 extracted claim(s); source-level direction is the coded finding |
| Muscle Function | Cersosimo 2025: Impact of SGLT2 inhibitors on endothelial function and echocardiographic parameters in dilated cardiomyopathy | direction=unclear | directness=indirect | B2 | outcome=Muscle Function; direction=unclear | finding=representative statistic P < 0.0001; source-level statistic reported |
| Safety and Comorbidity | Bailey 2022: Renal Protection with SGLT2 Inhibitors: Effects in Acute and Chronic Kidney Disease | direction=null | directness=direct | A1 | outcome=Safety and Comorbidity; direction=null | finding=2 extracted claim(s); source-level direction is the coded finding |
| Safety and Comorbidity | Chen 2024: Effects of SGLT2 inhibitors on cardiac function and health status in chronic heart failure: a systematic review and meta-analysis | direction=unclear | directness=review | B2 | outcome=Safety and Comorbidity; direction=unclear | finding=representative non-significant statistic P = 0.072; not treated as positive or negative directional support unless source direction is coded |

## Results

**Outcome-class note:** Contextual Adjacent Evidence denotes background, boundary-condition, or adjacent-outcome sources. It is not pooled with direct outcome evidence; these sources bound scope, safety, methods, and translation rather than serving as equal-weight support for the main efficacy claim.

| Evidence domain | Corpus slice | Direction profile | Directness | Main limitation |
|---|---|---|---|---|
| SGLT2 Inhibitors Subgroups / Cardiometabolic | n=8; claims=546 | positive=5, negative=0, null=0, mixed=1, unclear=2 (n=8) | 5 direct; 3 review | limited corpus depth in this outcome class |
| SGLT2 Inhibitors Subgroups / Longevity | n=4; claims=195 | positive=1, negative=0, null=2, mixed=0, unclear=1 (n=4) | 3 direct; 1 review | limited corpus depth in this outcome class |
| SGLT2 Inhibitors Subgroups / Contextual Adjacent Evidence | n=3; claims=46 | positive=0, negative=0, null=2, mixed=0, unclear=1 (n=3) | 2 direct; 1 indirect | limited corpus depth in this outcome class |
| SGLT2 Inhibitors Subgroups / Safety and Comorbidity | n=2; claims=66 | positive=0, negative=0, null=1, mixed=0, unclear=1 (n=2) | 1 direct; 1 review | limited corpus depth in this outcome class |
| SGLT2 Inhibitors Subgroups / Animal/Preclinical Context | n=1; claims=54 | positive=0, negative=0, null=1, mixed=0, unclear=0 (n=1) | 1 mechanistic | single-source slice; hypothesis-generating |
| SGLT2 Inhibitors Subgroups / Dosing and Pharmacokinetics | n=1; claims=10 | positive=0, negative=0, null=1, mixed=0, unclear=0 (n=1) | 1 review | single-source slice; hypothesis-generating |
| SGLT2 Inhibitors Subgroups / Immune and Inflammation | n=1; claims=24 | positive=0, negative=0, null=0, mixed=0, unclear=1 (n=1) | 1 direct | single-source slice; hypothesis-generating |
| SGLT2 Inhibitors Subgroups / Muscle Function | n=1; claims=143 | positive=0, negative=0, null=0, mixed=0, unclear=1 (n=1) | 1 indirect | single-source slice; hypothesis-generating |

**Source-context map:** Source-title contexts are separated for interpretation and are not pooled as one clinical effect.
- Dosing and pharmacokinetics context: 1 sources; no extracted directional signal in 1/1 sources.
- Transplant and fibrosis context: 1 sources; significant source statistic in 1/1 sources; receipt-level direction coded unclear.

### Cardiometabolic Outcomes

Across the curated corpus, the cardiometabolic outcome class is the dominant evidence stream and spans direct clinical RCTs, systematic reviews, mechanistic/biomarker studies, and pragmatic cohorts.

The mechanistic substrate underlying the positive functional findings therefore coexists with biomarker-level null signals, suggesting that the clinical RCT-level benefit may operate through pathways that are only partly captured by proteinuria-based biomarkers.

Within-corpus tensions on the cardiometabolic outcome class are evident.

The direction of effect is marked as unclear in the curated record, so the reported significance pattern is presented without an inferred net direction.

Within-corpus tensions surface most clearly as directness gaps. The synthesis adopted a network meta-analytic framework and did not restrict by subgroup beyond the agent and dose strata defined a priori. Directness was characterized as review because the underlying component trials were randomized, while the synthesis itself was observational in structure. No dose-ranging arm, plasma concentration sampling, or formal pharmacokinetic modeling was undertaken; the work instead stratified by reported maintenance dose.

No p-values were extracted for any of the other SGLT2 inhibitor agents in the curated excerpts, and no further effect estimates were captured for dapagliflozin, empagliflozin, or ertugliflozin within the available source. The network meta-analysis did not provide n per arm or follow-up duration in the excerpted material, limiting dose-stratum interpretation across the class. Importantly, the review-level effect direction for the canagliflozin contrast was the only numerically expressed finding in this outcome class, leaving the dose-specific subgroup architecture otherwise qualitative.

Preclinical and pharmacokinetic literature has suggested that SGLT2 inhibitors as a class are minimally absorbed beyond systemic exposure required for renal glucose handling, yet canagliflozin's prolonged intestinal residence time differentiates it from dapagliflozin and empagliflozin.

The source documents therapy with SGLT2i as significantly associated with improved FMD levels even at multivariable stepwise regression analysis (P < 0.001), with additional thresholds recorded at P < 0.01, P < 0.05, and P = 0.0001 across related inflammatory and endothelial analyses.

Exclusion criteria included HbA1c thresholds per the source, indicating that glycemic control was a prespecified covariate in the analytic plan. Effect direction within this source is recorded as unclear, reflecting that downstream inflammatory and endothelial readouts spanned multiple p-value tiers (P < 0.001, P = 0.0001, P < 0.01, P < 0.05) without a single dominant direction label.

Preclinical data from the wider literature would suggest endothelial benefit via reduced oxidative stress and improved nitric oxide signaling, and the present clinical RCT contributes a human-level mechanistic/biomarker anchor to that hypothesis. The source's directness is direct, supporting its use as a primary mechanistic evidence source rather than a secondary epidemiological signal.

Within the immune inflammation outcome class, the corpus contains only one curated source, so within-corpus tensions are necessarily limited. Together these sources provide heterogeneous longevity evidence spanning cardiology, cardio-oncology-adjacent stress cardiomyopathy, and hepatology populations.

Quantitative findings diverge across the four sources in both magnitude and direction. The full per-study endpoint table is summarized in the evidence synthesis.

Within-corpus tensions emerge along the directness dimension rather than as outright contradiction. The integrating sentence from the brief characterizes the longevity signal as context-dependent, and these four sources operationalize that context-dependence through cardiology, Takotsubo cardiomyopathy, and cirrhosis subpopulations. Across the reported timepoints the study delivered 13 named p-values, several of which fell well below conventional significance thresholds and several of which did not, producing a profile that requires careful numeric parsing rather than a uniform direction claim. Additional within-cohort comparisons carry P < 0.001, P = 0.001, P = 0.023, P = 0.027, P = 0.041, and P = 0.044 for various echocardiographic and endothelial contrasts reported in the source. A second cluster of comparisons did not cross conventional significance: P = 0.060, P = 0.066, P = 0.095, P = 0.463, P = 0.589, and P = 0.903, indicating that several secondary endpoints within the same cohort remained statistically inconclusive. The source therefore resolves into a mixed distribution rather than a single direction, and the synthesis reflects this by tagging the effect direction as unclear.

Mechanistically, the within-cohort signal at P < 0.0001 for ΔRHI is consistent with improved microvascular reactivity, which is a pathway adjacent to skeletal-muscle perfusion but not a direct measurement of strength, mass, or gait speed. Preclinical data on SGLT2 inhibition in heart failure models have suggested downstream effects on ketone-body utilization and on endothelial nitric oxide signaling, which can plausibly translate into vascular surrogates such as RHI; however, the curated corpus does not include a dedicated mechanistic human study pairing such pathway readouts with a hard muscle endpoint such as handgrip strength or short physical performance battery score.

Within-corpus tensions for the muscle function class cannot be drawn from paired same-outcome sources because the cross-study disagreement map records no non-orthogonal pairs in this domain. This internal disagreement — between a robust primary endothelial readout and a set of null secondary comparisons — is the central caveat that any downstream claim about SGLT2 inhibitor effects on muscle function in dilated cardiomyopathy must carry. The synthesis therefore presents the muscle function outcome class as hypothesis-generating rather than confirmatory.

Another tension sits inside the cardiometabolic class itself, where the literature on all-cause mortality is overtly split. Kaze 2022 [bundle:6] (cardiometabolic, review), in contrast, returns an unclear/null direction for the same family of cardiovascular and kidney outcomes among patients with diabetic kidney disease [exact source: https://doi.org/10.1186/s12933-022-01476-x]. Resolution would require subgroup-stratified individual-patient-data meta-analysis across the three reviews, with explicit pre-specification of which populations (post-ACS, stable CKD, broad T2D) SGLT2 inhibition actually moves the needle on mortality. Until then, the policy-grade claim "SGLT2 inhibitors reduce all-cause mortality" should be population-qualified rather than universal.

Kochanowska 2026 [bundle:14] reports: Shah et al. reported no significant difference between trials with a mean BMI ≥ 30 kg/m 2 and < 30 kg/m 2 ( p > 0.05) [ 5 ] [exact source: https://doi.org/10.3390/jcm15062219].

Ahmed 2025 [bundle:8] reports: Compared to placebo, SGLT2i significantly reduced MACE HR 0.89, 95% CI 0.84-0.95 [exact source: https://doi.org/10.1186/s12872-025-05455-4].

### Safety and Comorbidity Outcomes

The safety comorbidity outcome class is populated by two registry sources that sit at different points on the directness spectrum. Together these two sources anchor the safety comorbidity subsection, with one direct mechanistic/biomarker RCT family and one aggregated review of indirect clinical endpoints.

Two borderline results, P = 0.072 and P = 0.072, appear in the same source and are flagged in the evidence synthesis as the only non-significant entries in the cardiac meta-analysis.

Preclinical and human biomarker data therefore align on the direction of cardiac decongestion, while the magnitude of clinical event reduction remains aggregated rather than primary in this outcome class.

The mechanism behind the apparent convergence is that natriuretic peptide lowering plausibly tracks reduced wall stress, and lower wall stress plausibly tracks fewer fatal events — but the chain is not guaranteed, and Ioannidis 2005 caution is directly applicable: surrogate associations do not guarantee hard-outcome validity. The boundary condition is one of endpoint distance: when the only evidence is mechanistic biomarker movement, claims of clinical benefit must be hedged; when matched-cohort or randomized mortality data exist alongside, the hard-outcome claim becomes permissible. What would resolve the residual uncertainty is a head-to-head mediation analysis within a single trial demonstrating that the magnitude of NT-proBNP reduction statistically explains the mortality reduction, rather than the two signals merely coexisting.

The implication is that subgroup-level recommendations should be indexed to the specific clinical event being prevented, and that indirect review-level signals should never be fused with direct RCT endpoints into a single magnitude claim. What would resolve this is harmonization of endpoint definitions across the next generation of subgroup trials.

Another tension, almost unique to this corpus, is the co-occurrence of mechanistic/biomarker RCTs reporting substantial biological movement with concurrent clinical RCTs returning null findings on related hard outcomes. The mechanism-level expectation is that fibrosis regression should track kidney-protective downstream effects; the null result on proteinuria onset suggests either that the fibrosis signal does not propagate to the renal outcome in low-risk preserved-function patients, or that the biomarker window is too short. The relevant background citation is ADA 2024, which sets the HbA1c diagnostic floor at 6.5% — a useful boundary because both trials enrolled T2D patients but stratified very differently across baseline renal risk. The boundary condition is risk-stratum dependence: mechanistic biomarker benefit may be detectable across the risk spectrum, but hard-outcome benefit concentrates in the higher-risk tail. Resolution would require trials that pre-specify biomarker mediation in a population where hard outcomes are also accruing at adequate power, rather than the current pattern of biomarker-rich but event-poor mechanistic studies adjacent to event-rich but biomarker-poor outcome trials.

The fifth tension is the indirectness of the muscle-function and contextual other outcome classes against the relatively direct clinical-endpoint evidence base elsewhere in the corpus. The boundary condition is enrollment representativeness; the resolution would be RCTs specifically powered in sarcopenic or frail subgroups using a frailty anchor such as the Studenski 2011 gait-speed cutoff of 0.8 m/s or the Cesari 2009 severe-frailty cutoff of 0.6 m/s. Without such trials, mechanistic endothelial-function gains in dilated cardiomyopathy and the thyroid hypothesis remain suggestive rather than actionable for older adults at the frailty extreme.

### Boundary-condition synthesis

Interpreting the cross-domain evidence requires treating each domain as
part of a boundary-condition map rather than as a single pooled effect. Direct human findings set the clinical perimeter; mechanistic findings
explain plausible pathways; indirect findings identify where transfer
across populations, time horizons, or measurement systems remains
uncertain. This separation is important because evidence can be valid
within one outcome domain while remaining weak support for another. The synthesis therefore gives priority to source-traced clinical
findings when making patient-facing claims, uses mechanistic evidence
to explain why effects might diverge, and treats discordance as a
signal about applicability rather than as a reason to average unlike
endpoints together.

Chen 2024 [bundle:10] reports: The SGLT2 inhibitors group exhibited a significant reduction in pro b-type natriuretic peptide (NT-proBNP) levels by 136.03 pg/ml 95% confidence interval [CI]: -253.36, - 18.70 [exact source: https://doi.org/10.1186/s12933-023-02042-9].

### Longevity Outcomes

Longevity remains a separate Results slice for SGLT2 Inhibitors Subgroups (n=4; claims=195; positive=1, negative=0, null=2, mixed=0, unclear=1 (n=4); 3 direct; 1 review; limited corpus depth in this outcome class) and is not pooled into adjacent endpoint classes. Source-level findings are:
- Rawish 2026 [bundle:9] (SGLT2 inhibitors are associated with improved long-term survival in Takotsubo syndrome: insights from large-scale; 67 extracted claim(s); receipt-level direction is the coded finding; outcome=Longevity; direction=unclear; directness=direct; tier=A1).
- Elrakaybi 2022 [bundle:17] (Cardiovascular protection by SGLT2 inhibitors – Do anti-inflammatory mechanisms play a role?; 6 extracted claim(s); receipt-level direction is the coded finding; outcome=Longevity; direction=null; directness=direct; tier=A1).
- Gao 2026 [bundle:20] (Repurposing SGLT2 Inhibitors for Cirrhotic Ascites: From Mechanistic Research to Clinical Exploration; 1 extracted claim(s); receipt-level direction is the coded finding; outcome=Longevity; direction=null; directness=direct; tier=A1).
- Jiang 2022 [bundle:2] (Comparative Cardiovascular Outcomes of SGLT2 Inhibitors in Type 2 Diabetes Mellitus: A Network Meta-Analysis of; 121 extracted claim(s); receipt-level direction is the coded finding; outcome=Longevity; direction=positive; directness=review; tier=B1).

Rawish 2026 [bundle:9] reports: After matching (yielding well-balanced 524 patients per group), mortality was significantly reduced in the SGLT2i group compared with RAASi + BB alone (HR 0.56, 95% CI 0.36-0.89) [exact source: https://doi.org/10.1093/ehjcvp/pvaf088].

Elrakaybi 2022 [bundle:17] reports: 30% reduction of HF hospitalization compared to placebo which confirm the role of SGLT2 inhibitors in preventing or delaying HF onset [ [15] , [16] , [17] ] [exact source: https://doi.org/10.1016/j.molmet.2022.101549].

Gao 2026 [bundle:20] reports: 3 Once ascites develops, the five-year mortality increases to about 44% [exact source: https://doi.org/10.14218/JCTH.2025.00465].

### Contextual Adjacent Evidence Outcomes

Mechanistically, the three substrates do not converge on a single pathway. The mechanistic substrate underlying the Albulushi 2025 [bundle:12] finding is a clinical RCT interrogating myocardial fibrosis in diabetic HFpEF, with reductions in fibrotic remodeling proposed as the proximal mediator [exact source: https://doi.org/10.1186/s40001-025-02834-7].

Contextual Adjacent Evidence remains a separate Results slice for SGLT2 Inhibitors Subgroups (n=3; claims=46; positive=0, negative=0, null=2, mixed=0, unclear=1 (n=3); 2 direct; 1 indirect; limited corpus depth in this outcome class) and is not pooled into adjacent endpoint classes. Source-level findings are:
- Albulushi 2025 [bundle:12] (Impact of SGLT2 inhibitors on myocardial fibrosis in diabetic HFpEF: a longitudinal study; representative statistic p < 0.001; source-level statistic reported; outcome=Contextual Adjacent Evidence; direction=unclear; directness=direct; tier=A1).
- Kawanami 2017 [bundle:21] (SGLT2 Inhibitors as a Therapeutic Option for Diabetic Nephropathy; 1 extracted claim(s); receipt-level direction is the coded finding; outcome=Contextual Adjacent Evidence; direction=null; directness=direct; tier=A1).
- Szklarz 2026 [bundle:18] (Another pleiotropic effect of SGLT2 inhibitors: Is it a new frontier in thyroid function regulation?; 3 extracted claim(s); receipt-level direction is the coded finding; outcome=Contextual Adjacent Evidence; direction=null; directness=indirect; tier=B2).

Direction reconciliation: receipt-level null or unclear coding is conservative claim-level coding. Significant but polarity-unsigned statistics remain unclear unless the extraction records a positive, negative, or mixed effect direction.

### Dosing and Pharmacokinetics Outcomes

Dosing and Pharmacokinetics remains a separate Results slice for SGLT2 Inhibitors Subgroups (n=1; claims=10; positive=0, negative=0, null=1, mixed=0, unclear=0 (n=1); 1 review; single-source slice; hypothesis-generating) and is not pooled into adjacent endpoint classes. Source-level findings are:
- Chen 2026 [bundle:16] (Agent- and Dose-Specific Intestinal Obstruction Safety of GLP-1 Receptor Agonists and SGLT2 Inhibitors: A Network; 10 extracted claim(s); receipt-level direction is the coded finding; outcome=Dosing and Pharmacokinetics; direction=null; directness=review; tier=B2).

### Immune and Inflammation Outcomes

Immune and Inflammation remains a separate Results slice for SGLT2 Inhibitors Subgroups (n=1; claims=24; positive=0, negative=0, null=0, mixed=0, unclear=1 (n=1); 1 direct; single-source slice; hypothesis-generating) and is not pooled into adjacent endpoint classes. Source-level findings are:
- Correale 2021 [bundle:15] (Switch to SGLT2 Inhibitors and Improved Endothelial Function in Diabetic Patients with Chronic Heart Failure; representative statistic p < 0.001; source-level statistic reported; outcome=Immune and Inflammation; direction=unclear; directness=direct; tier=A1).

Correale 2021 [bundle:15] reports: Therapy with SGLT2i was significantly associated to improved FMD levels even at multivariable stepwise regression analysis ( p < 0.001) [exact source: https://doi.org/10.1007/s10557-021-07254-3].

### Muscle Function Outcomes

Muscle Function remains a separate Results slice for SGLT2 Inhibitors Subgroups (n=1; claims=143; positive=0, negative=0, null=0, mixed=0, unclear=1 (n=1); 1 indirect; single-source slice; hypothesis-generating) and is not pooled into adjacent endpoint classes. Source-level findings are:
- Cersosimo 2025 [bundle:1] (Impact of SGLT2 inhibitors on endothelial function and echocardiographic parameters in dilated cardiomyopathy; representative statistic P < 0.0001; source-level statistic reported; outcome=Muscle Function; direction=unclear; directness=indirect; tier=B2).

Cersosimo 2025 [bundle:1] reports: At 6 months, it significantly increased to 1.40 ± 0.34 ( P < 0.0001), reflecting an absolute change of 0.25 ± 0.03 (ΔRHI baseline - 6 months) [exact source: https://doi.org/10.2459/JCM.0000000000001733].

## Metabolic-Functional Tradeoff Framework

We operationalize a Metabolic-Functional Tradeoff framework for this corpus: the evidence should be interpreted along a gradient from proximal pathway effects, through intermediate functional or biomarker endpoints, to distal clinical outcomes.

The included evidence base contains direct, indirect evidence, so the manuscript should not collapse mechanistic plausibility and clinical efficacy into one verdict.

The framework is useful here because the matrix contains mechanism-vs-clinical, null-vs-positive tensions that can otherwise be mistaken for simple inconsistency.

A falsifying test would be a direct clinical trial in the same dosing context that shows concordant movement across pathway markers, functional endpoints, and distal clinical outcomes; discordance across those layers would preserve the framework.

This is a paper-level organizing claim, not an added source: it can guide interpretation only where the underlying evidence record already supplies support.

## Cross-Domain Synthesis

Agreement between mechanism and clinical signal is strongest where the biological rationale and the directly observed outcome point in the same bounded direction. For SGLT2 inhibitors subgroups, direct sources such as Huang 2026 [bundle:3], Jansz 2026 [bundle:4], Sato 2025 [bundle:7] define the human evidence perimeter, while mechanistic sources such as Nobayashi 2026 [bundle:11] explain why an effect could occur [exact source: https://doi.org/10.2147/DDDT.S580640] [exact source: https://doi.org/10.1016/j.lanprc.2026.100139] [exact source: https://doi.org/10.2169/internalmedicine.6239-25] [exact source: https://doi.org/10.1111/dom.70625]. Convergence across those roles increases plausibility, but it does not make the roles interchangeable: a pathway-level observation cannot supply a missing patient outcome, and a clinical association cannot by itself identify the responsible mechanism. Nobayashi 2026 [bundle:11] provides animal/preclinical context only [exact source: https://doi.org/10.1111/dom.70625]. Nobayashi 2026 [bundle:11] provides animal/preclinical context only.

Divergence is equally informative. Positive signals represented by Jiang 2022 [bundle:2], Huang 2026 [bundle:3], Jansz 2026 [bundle:4] occur alongside null signals represented by Nobayashi 2026 [bundle:11], Chen 2026 [bundle:16], Elrakaybi 2022 [bundle:17] and negative or adverse signals represented by the retained evidence base [exact source: https://doi.org/10.3389/fendo.2022.802992] [exact source: https://doi.org/10.2147/DDDT.S580640] [exact source: https://doi.org/10.1016/j.lanprc.2026.100139] [exact source: https://doi.org/10.1111/dom.70625] [exact source: https://doi.org/10.3390/ijms27020608] [exact source: https://doi.org/10.1016/j.molmet.2022.101549]. Their outcome distribution spans the cardiometabolic and longevity outcome classes, the longevity, contextual adjacent evidence and cardiometabolic outcome classes, and no dominant outcome class. This pattern rejects a single global verdict. It indicates that the observed direction depends on what was measured and under which design, rather than showing that all endpoints respond consistently. Nobayashi 2026 [bundle:11] provides animal/preclinical context only.

The outcome-class map makes that heterogeneity auditable: Cardiometabolic (mixed=1, null=1, positive=5, unclear=2; direct=5, mechanistic=1, review=3; sources Huang 2026 [bundle:3], Jansz 2026 [bundle:4], Suciu 2025 [bundle:5]); Longevity (null=2, positive=1, unclear=1; direct=3, review=1; sources Jiang 2022 [bundle:2], Rawish 2026 [bundle:9], Elrakaybi 2022 [bundle:17]); Contextual Adjacent Evidence (null=2, unclear=1; direct=2, indirect=1; sources Albulushi 2025 [bundle:12], Szklarz 2026 [bundle:18], Kawanami 2017 [bundle:21]); Safety and Comorbidity (null=1, unclear=1; direct=1, review=1; sources Chen 2024 [bundle:10], Bailey 2022 [bundle:19]) [exact source: https://doi.org/10.2147/DDDT.S580640] [exact source: https://doi.org/10.1016/j.lanprc.2026.100139] [exact source: https://doi.org/10.3390/medicina61101866] [exact source: https://doi.org/10.3389/fendo.2022.802992] [exact source: https://doi.org/10.1093/ehjcvp/pvaf088] [exact source: https://doi.org/10.1016/j.molmet.2022.101549] [exact source: https://doi.org/10.1186/s40001-025-02834-7] [exact source: https://doi.org/10.1186/s13044-025-00282-3] [exact source: https://doi.org/10.3390/ijms18051083] [exact source: https://doi.org/10.1186/s12933-023-02042-9] [exact source: https://doi.org/10.1007/s11892-021-01442-z]. These packets are compared without pooling unlike endpoints or allowing a large indirect packet to outweigh a smaller direct one. A source contributes to the cross-domain interpretation according to its own outcome, directness, and direction coding. Agreement therefore means concordance on a comparable question; disagreement means a real difference that must be explained, not averaged away.

Population is the first boundary on transfer. Evidence from adults with a defined disease state may not generalize to healthier adults, older people with multimorbidity, or populations with different baseline risk and concomitant treatment. Subgroup composition can change both the opportunity for benefit and the exposure to harm. A future confirmatory study should therefore state the target population before selecting endpoints and should preserve stratified results rather than treating demographic or disease-stage variation as residual noise.

Dose and schedule form a separate boundary. Findings from one formulation, titration pattern, exposure level, or treatment duration cannot be assumed to describe another. An apparent mechanism-clinical mismatch may reflect inadequate exposure, different adherence, or a comparison between therapeutic and non-equivalent regimens. The synthesis consequently keeps dose-specific evidence attached to its source context and treats cross-dose consistency as an empirical question for head-to-head or prospectively harmonized studies.

Endpoint distance is the third boundary. Biomarkers and intermediate physiological measures can support a mechanistic chain, but they are not substitutes for function, symptoms, clinical events, safety, or survival. Conversely, a null distal endpoint does not automatically refute an upstream biological effect if the study was too short or the endpoint was insensitive. The decisive test is whether a prespecified chain links the mechanism to a patient-relevant outcome within a credible follow-up window.

Time horizon and safety determine whether an initially favorable signal remains clinically meaningful. Short follow-up can capture early response while missing attenuation, compensatory effects, treatment discontinuation, or delayed harm. Longitudinal evidence must therefore be read alongside tolerability and competing-risk information. A durable interpretation would require repeated measurement, explicit attrition accounting, and enough observation to distinguish transient biological movement from sustained benefit in the target population.

Comparator choice determines what a directional result can mean. Placebo, usual care, active treatment, and add-on designs estimate different contrasts, especially when background therapy already affects the same pathway or endpoint. Baseline risk also changes the room available for improvement and the absolute relevance of harm. Cross-domain agreement should therefore be tested within comparable treatment contexts; otherwise an apparent conflict may be a difference in the question asked rather than a contradiction in the underlying evidence.

Measurement and analysis complete the boundary map. Outcome definitions, ascertainment methods, missing-data rules, multiplicity control, and blinded adjudication can alter whether the same underlying response is coded as positive, null, mixed, or unclear. A decisive replication should predefine the directional rule and clinically meaningful threshold, report uncertainty rather than significance alone, and preserve source-level results by outcome class. Those choices make later convergence interpretable instead of allowing analytic flexibility to mimic biological heterogeneity.

Causal interpretation requires the full sequence to remain intact. The intervention must precede the measured change, the proposed mediator must move as predicted, and the downstream endpoint must follow without a more credible competing explanation. Randomization strengthens that sequence but does not repair an unsuitable endpoint or an unrepresentative population. Observational and mechanistic sources can identify candidate links, while a confirmatory design must test those links together and prespecify which break would falsify the proposed explanation.

Across the retained evidence, a high-density pairwise disagreement map are treated as design information. Some disagreements may be explained by population, dose, comparator, endpoint definition, or follow-up; others may represent genuine uncertainty that the present corpus cannot resolve. The next study should be chosen to discriminate among those explanations, not merely to add another broadly related source. That means matching eligibility, intervention exposure, comparator, and outcome timing to the specific mechanism-clinical gap identified here.

The resulting interpretation is conditional rather than indecisive. Across 21 curated reference papers, the evidence base for SGLT2 inhibitors subgroups shows a context-dependent profile. Positive signals appear in: cardiometabolic, longevity. Null findings dominate: longevity, contextual other. The synthesis surfaces 113 cross-study disagreements across outcome classes — see Cross-Domain Synthesis. The SGLT2 inhibitors subgroups broad aging-related case as currently constituted is incomplete: mechanistic plausibility coexists with mixed or sparse human-RCT evidence, and the boundary conditions remain to be established. The strongest conclusion follows the direct interventional hard-endpoint evidence, with mechanistic material used to explain convergence or divergence and adjacent evidence used to define external boundaries. Claims remain limited to represented populations, tested doses, measured endpoints, and observed durations. Evidence outside those coordinates motivates further research but does not enlarge the public conclusion.

## Discussion

**Thesis:** Across 21 curated reference papers, the evidence base for SGLT2 shows a context-dependent profile. Positive signals appear in: cardiometabolic, longevity. Null findings dominate: longevity, contextual other. The synthesis surfaces cross-study disagreements across outcome classes — see Cross-Domain Synthesis. The SGLT2 broad aging-related case as currently constituted is incomplete: mechanistic plausibility coexists with mixed or sparse human-RCT evidence, and the boundary conditions remain to be established. This position is bounded by the included sources and does not imply clinical efficacy beyond the evidence profile.

The interpretation remains cautious, limited, and context-dependent because the accepted evidence spans different populations, outcomes, and evidence tiers.

### Evidence Summary

The evidence base for this synthesis comprises 21 included sources. The evidence-tier distribution is: A1 (n=13), B2 (n=5), B1 (n=3). By directness, the breakdown is: direct (n=12), review (n=6), indirect (n=3). 12 of 21 sources carry at least one p-value in their bound claims, providing the quantitative basis for the effect-direction conclusions argued above. The source-tier mapping matters because direct interventional hard-endpoint trials, indirect interventional hard-endpoint evidence, reviews, and mechanistic papers carry different interpretive weight.

Populations covered span 2 distinct summaries across the source set: type 2 diabetes patients; adults. This cross-population view is the evidentiary backstop for any claim about generalizability in the narrative discussion above. Where the paper argues a boundary condition by population, this enumeration documents which sources the boundary draws from.

### Interpretation constraints

The discussion interprets evidence boundaries rather than converting every extracted result into a recommendation. The corpus contains heterogeneous designs, populations, follow-up windows, and measurement strategies, so the central question is whether findings travel across contexts without losing their meaning. Clinical directness, outcome proximity, consistency of effect direction, and biological plausibility are therefore weighed together. Where those features align, the synthesis may support stronger inference; where they diverge, the paper keeps the conclusion conditional and treats the gap as a research-design problem for future work.

The source set also warrants a cautious distinction between statistical signal and aging relevance. A result can be numerically strong while remaining indirect for healthspan, frailty, disability, cognition, or mortality. Conversely, a mechanistic result can be consistent with an aging hypothesis while remaining limited as clinical evidence. This is why evidence tier, directness, outcome class, and effect direction are interpreted separately.

The most decision-relevant uncertainty is context-dependent. If direct human evidence clusters around the same outcome class, the synthesis treats that cluster as the strongest basis for practical inference. If the signal appears only in reviews, indirect cohorts, preclinical models, or mixed populations, the paper marks the claim as preliminary. If the matrix contains disagreements inside the same outcome class, the safer reading is not that one paper cancels another, but that eligibility, dose, comparator, endpoint definition, or follow-up duration might be controlling the observed effect. Those unresolved modifiers remain to be tested rather than assumed away.

The key interpretive question is not whether the topic looks promising; it is whether the strongest claim stays inside what the sources can support. This anchor therefore avoids adding new empirical claims. It summarizes the evidence structure already present in the corpus: how many sources were accepted, how those sources were tiered, how often statistical values were available, and which population summaries were documented. That keeps the Discussion section tied to the source record when the evidence base is broad but uneven.

The resulting stance is deliberately conservative. Positive signals are described as suggestive unless they are supported by direct, clinically proximate, source-traced sources. Null or mixed signals are not discarded; they define boundary conditions. Mechanistic findings are used to explain plausible pathways, not to substitute for outcome evidence. Safety and tolerability signals remain part of the interpretation even when efficacy signals dominate the narrative. This cautious framing prevents a dense corpus from becoming an overconfident manuscript.

This section also constrains how readers should use the paper. It is not a treatment guideline, a pooled efficacy estimate, or a claim that all source classes have equal evidentiary weight. It is a structured map of what the current corpus can and cannot justify. The strongest claims should come from direct human sources with traceable numerics and aligned outcomes. Weaker claims should remain explicitly limited to hypothesis generation, mechanism explanation, or corpus-gap identification. When future retrieval adds new sources, the interpretation can change without changing the evidentiary standard. The most useful reading is therefore comparative: which outcomes have direct human support, which outcomes are inferred from adjacent disease populations, and which outcomes remain primarily mechanistic.

Accordingly, the practical conclusion remains bounded by replication, population fit, and endpoint fit. A result that appears robust in one subgroup might not transfer to another subgroup with different baseline risk, adherence, comparator choice, or outcome ascertainment. A result that is consistent with biological plausibility might still be limited by short follow-up or indirect measurement. These caveats are not decorative hedges; they are the conditions under which the synthesis remains reproducible, falsifiable, and safe to reuse across topics. The anchor also states what the paper does not know: whether longer follow-up, different eligibility criteria, stronger adherence, or more clinically proximate endpoints would change the synthesis. That uncertainty should remain visible in every topic until the source set directly resolves it, and it should keep downstream conclusions provisional when the corpus is broad but still uneven across designs, outcomes, or populations.

**Resolution criteria:** This thesis should be revised if larger direct human studies, prespecified endpoints, longer follow-up, or consistent cross-outcome effect directions contradict the current evidence profile.

## Limitations

**Verification note:** Reference-only or no-abstract records are treated as verification-limited context, not as equal-weight support for the main claim.

The curated corpus draws 21 records, but several evidence types commonly used to anchor cardiometabolic and longevity claims are absent, and the resulting gaps impose hard ceilings on what the headline conclusions can carry.

Several outcome classes rest on a single source, which means within-corpus replication is impossible and any cross-paper pooling would be an unsupported act of synthesis. Conclusions that depend on any of these single-source outcomes inherit the entire fragility of that one dataset.

Population specificity is tight and tilts the evidence base toward diabetic, middle-aged-to-elderly, high-cardiovascular-risk patients, with multiple downstream external-validity limits. Pediatric, pregnant, and pre-diabetic populations are absent entirely; the HbA1c framing in the corpus sits squarely in the ADA 2024 7% / 6.5% adult-target band (ADA 2024), with no enrollment at glycemic extremes.

Endpoint coverage is narrower than the broader SGLT2i literature would suggest, and several routinely cited outcomes are simply not measured in this corpus. Hard cognitive endpoints (incident dementia, mild cognitive impairment, executive-function decline) are absent across all 21 sources.

A pronounced mechanism-to-clinic gap runs through the cardiometabolic, longevity, contextual, safety, and muscle-function outcome classes, where direct interventional hard-endpoint evidence in this corpus is sparse or absent while mechanistic plausibility is asserted. The surrogate-endpoint caveat applies (Ioannidis 2005): several claimed benefits rest on biomarkers (NT-proBNP, FMD, RHI, liver enzymes, HbA1c) without confirmation on adjudicated clinical endpoints in the same enrolled cohort.

### Residual uncertainty

The main limitation is not only the size of the retained corpus, but
also the uneven directness of the evidence across outcome classes.
Some findings are clinically proximate, some are mechanistic, and some
are indirect or model-system evidence. The paper therefore avoids
treating all sources as equivalent. Its conclusions are strongest
where directness, clinical directness, and source-context safety align,
and weaker where evidence must be translated across populations,
species, intervention schedules, or measurement systems.

## Conclusion

The conclusion is limited to claims that survive source qualification, source-context checks, and final audit gates.

### Bounded conclusion

This synthesis supports a bounded interpretation across 21 included sources. The evidence tiers are A1 (n=13), B2 (n=5), B1 (n=3), and directness is direct (n=12), review (n=6), indirect (n=3). Effect directions are unclear (n=7), null (n=7), positive (n=6), mixed (n=1), with 12 sources carrying source-traced p-values and 113 documented cross-source tensions. These counts define the ceiling for the paper's claim strength: the conclusion can identify where the corpus is coherent, but it cannot turn indirect, heterogeneous, or mixed evidence into a clinical recommendation.

Population boundary: the included sources document 2 distinct population summaries: adults; type 2 diabetes patients. Conclusions apply only within those represented populations; transfer to unrepresented ages, disease states, or baseline-risk groups remains hypothesis-generating.

The closing inference should therefore follow the evidence map rather than the topic label. Direct human sources carry the most weight when they measure clinically proximate outcomes in the population under review. Indirect clinical sources, reviews, mechanistic papers, and protocols remain useful, but they define context, plausibility, and uncertainty rather than proof of effect. Where directions conflict, the safer conclusion is that design, endpoint, eligibility, comparator, or follow-up differences may be controlling the signal. Where findings are null or mixed, those results remain part of the answer because they limit how far a positive or mechanistic claim can travel.

The practical takeaway is bounded and revisable. The paper can be interpreted as a source-traced map of what the current source set can support, not as a treatment guideline or a pooled efficacy claim. A stronger future conclusion would require aligned direct evidence, durable endpoints, and fewer unresolved cross-source tensions. Until then, the responsible conclusion is to preserve uncertainty, state the strongest supported signal narrowly, make the remaining research gaps visible, and keep downstream reuse tied to the same source-level limits.

## What This Synthesis Adds

This synthesis maps 21 included sources on SGLT2 Inhibitors Subgroups across 7 outcome classes and 113 cross-study disagreements. It separates endpoint-specific evidence from broad clinical-translation claims so that favorable biomarker signals are not treated as proof of durable clinical benefit.

The strongest unresolved contrast is the null vs positive between Suciu 2025 [bundle:5] and Kaze 2022 [bundle:6] on cardiometabolic (severity 4/5), which defines the boundary condition future studies must test rather than smooth over [exact source: https://doi.org/10.3390/medicina61101866] [exact source: https://doi.org/10.1186/s12933-022-01476-x].

Prior reviews in the corpus (Jiang 2022 [bundle:2], Suciu 2025 [bundle:5], Kaze 2022 [bundle:6]) emphasize convergent signals on SGLT2 Inhibitors Subgroups [exact source: https://doi.org/10.3389/fendo.2022.802992] [exact source: https://doi.org/10.3390/medicina61101866] [exact source: https://doi.org/10.1186/s12933-022-01476-x]. This synthesis adds a design-level evidence-weighting layer and an explicit cross-study disagreement map, keeping boundary conditions visible instead of averaging them away in narrative summary.

### Boundary-Condition Matrix

| Evidence domain | Direct sources | Indirect / mechanism sources | Direction profile | Interpretation boundary |
|---|---:|---:|---|---|
| muscle function | 0 | 1 | unclear | direct interventional hard-endpoint gap |
| longevity | 3 | 1 | null, positive, unclear | replication gap |
| cardiometabolic | 5 | 4 | mixed, null, positive, unclear | conflict-resolution gap |
| dosing and pharmacokinetics | 0 | 1 | null | direct interventional hard-endpoint gap |
| immune and inflammation | 1 | 0 | unclear | replication gap |
| contextual adjacent evidence | 2 | 1 | null, unclear | replication gap |
| safety and comorbidity | 1 | 1 | null, unclear | replication gap |

### Evidence-Gap Priority

| Priority | Gap | Rationale |
|---|---|---|
| P1 | muscle function: direct interventional hard-endpoint gap | 0 direct and 1 indirect source; direction profile: unclear |
| P2 | longevity: replication gap | 3 direct and 1 indirect sources; direction profile: null, positive, unclear |
| P3 | cardiometabolic: conflict-resolution gap | 5 direct and 4 indirect sources; direction profile: mixed, null, positive, unclear |
| P4 | dosing and pharmacokinetics: direct interventional hard-endpoint gap | 0 direct and 1 indirect source; direction profile: null |
| P5 | immune and inflammation: replication gap | 1 direct and 0 indirect source; direction profile: unclear |

### Next-Study Design Recommendation

The next high-yield study for SGLT2 Inhibitors Subgroups should target the **muscle function** evidence gap, pre-register the primary endpoint, separate clinical from mechanistic endpoints, preserve safety and adherence capture, and include an analysis plan that can falsify the current boundary-condition claim rather than only confirming a favorable direction. Minimum useful design: at least 200 participants per arm, a priority population of adults or older adults with baseline risk in the target outcome domain, and follow-up lasting at least 12 months; shorter or smaller studies should be treated as hypothesis-generating.

## Evidence Snapshot

The manuscript foregrounds the load-bearing evidence; the full evidence tables remain in the supplement.

### Load-Bearing Included Studies

- Huang 2026 [bundle:3]; tier=A1; directness=direct; endpoint=cardiometabolic; direction=positive; representative statistic=P = 0.016.
- Jansz 2026 [bundle:4]; tier=A1; directness=direct; endpoint=cardiometabolic; direction=positive; representative statistic=P = 0.11.
- Sato 2025 [bundle:7]; tier=A1; directness=direct; endpoint=cardiometabolic; direction=positive; representative statistic=P = 0.004.
- Rawish 2026 [bundle:9]; tier=A1; directness=direct; endpoint=longevity; direction=unclear.
- Albulushi 2025 [bundle:12]; tier=A1; directness=direct; endpoint=contextual adjacent evidence; direction=unclear; representative statistic=P < 0.001.
- Moreno-Perez 2026 [bundle:13]; tier=A1; directness=direct; endpoint=cardiometabolic; direction=positive; representative statistic=P < 0.001.
- Kochanowska 2026 [bundle:14]; tier=A1; directness=direct; endpoint=cardiometabolic; direction=unclear; representative statistic=P > 0.05.
- Correale 2021 [bundle:15]; tier=A1; directness=direct; endpoint=immune inflammation; direction=unclear; representative statistic=P < 0.001.
- Elrakaybi 2022 [bundle:17]; tier=A1; directness=direct; endpoint=longevity; direction=null.
- Bailey 2022 [bundle:19]; tier=A1; directness=direct; endpoint=safety comorbidity; direction=null.

### Source Classification Map

Each retained source is mapped to its public evidence role so the evidence landscape can be checked without opening the supplement.

- Huang 2026 [bundle:3]: outcome=cardiometabolic; directness=direct; tier=A1; direction=positive; claims=97.
- Jansz 2026 [bundle:4]: outcome=cardiometabolic; directness=direct; tier=A1; direction=positive; claims=93.
- Sato 2025 [bundle:7]: outcome=cardiometabolic; directness=direct; tier=A1; direction=positive; claims=72.
- Rawish 2026 [bundle:9]: outcome=longevity; directness=direct; tier=A1; direction=unclear; claims=67.
- Albulushi 2025 [bundle:12]: outcome=contextual adjacent evidence; directness=direct; tier=A1; direction=unclear; claims=42.
- Moreno-Perez 2026 [bundle:13]: outcome=cardiometabolic; directness=direct; tier=A1; direction=positive; claims=36.
- Kochanowska 2026 [bundle:14]: outcome=cardiometabolic; directness=direct; tier=A1; direction=unclear; claims=26.
- Correale 2021 [bundle:15]: outcome=immune inflammation; directness=direct; tier=A1; direction=unclear; claims=24.
- Elrakaybi 2022 [bundle:17]: outcome=longevity; directness=direct; tier=A1; direction=null; claims=6.
- Bailey 2022 [bundle:19]: outcome=safety comorbidity; directness=direct; tier=A1; direction=null; claims=2.
- Gao 2026 [bundle:20]: outcome=longevity; directness=direct; tier=A1; direction=null; claims=1.
- Kawanami 2017 [bundle:21]: outcome=contextual adjacent evidence; directness=direct; tier=A1; direction=null; claims=1.
- Nobayashi 2026 [bundle:11]: outcome=cardiometabolic; directness=indirect; tier=A1; direction=null; claims=54.
- Jiang 2022 [bundle:2]: outcome=longevity; directness=review; tier=B1; direction=positive; claims=121.
- Suciu 2025 [bundle:5]: outcome=cardiometabolic; directness=review; tier=B1; direction=positive; claims=78.
- Kaze 2022 [bundle:6]: outcome=cardiometabolic; directness=review; tier=B1; direction=unclear; claims=76.
- Cersosimo 2025 [bundle:1]: outcome=muscle function; directness=indirect; tier=B2; direction=unclear; claims=143.
- Ahmed 2025 [bundle:8]: outcome=cardiometabolic; directness=review; tier=B2; direction=mixed; claims=68.
- Chen 2024 [bundle:10]: outcome=safety comorbidity; directness=review; tier=B2; direction=unclear; claims=64.
- Chen 2026 [bundle:16]: outcome=dosing pharmacokinetics; directness=review; tier=B2; direction=null; claims=10.
- Szklarz 2026 [bundle:18]: outcome=contextual adjacent evidence; directness=indirect; tier=B2; direction=null; claims=3. Nobayashi 2026 [bundle:11] provides animal/preclinical context only.

### Classification Criteria

- **Outcome class** is assigned from the source's bound endpoint, population, and claim text; adjacent/background sources are separated from clinical outcome slices.
- **Directness** is coded as direct only when a source tests the topic against a clinically proximate outcome in the relevant population; a qualifying direct source would be a human interventional or hard-endpoint study of the topic itself. Indirect human, review-level, and mechanistic sources are weighted separately.
- **Directional signal** is counted within the assigned outcome class only. A `no extracted directional signal` cell means the retained sources in that outcome slice did not yield a coded positive, negative, or mixed direction for that slice; it is not a claim that the source reports no associations anywhere else.
- **Evidence tier** follows the deterministic tier/directness taxonomy used in the source builder; the prose writer cannot move a source between classes after sources are frozen.

### Load-Bearing Tensions

- Severity 4 null vs positive: Suciu 2025 [bundle:5] vs Kaze 2022 [bundle:6]; Suciu 2025 [bundle:5] (positive on mortality) vs Kaze 2022 [bundle:6] (null on mortality) — partial conflict
- Severity 4 null vs positive: Ahmed 2025 [bundle:8] vs Kaze 2022 [bundle:6]; Ahmed 2025 [bundle:8] (positive on mortality) vs Kaze 2022 [bundle:6] (null on mortality) — partial conflict
- Severity 4 null vs positive: Huang 2026 [bundle:3] vs Moreno-Perez 2026 [bundle:13]; Huang 2026 [bundle:3] (positive on body mass index) vs Moreno-Perez 2026 [bundle:13] (null on body mass index) — partial conflict
- Severity 4 null vs positive: Huang 2026 [bundle:3] vs Jansz 2026 [bundle:4]; Huang 2026 [bundle:3] (positive on body mass index) vs Jansz 2026 [bundle:4] (null on body mass index) — partial conflict
- Severity 3 indirectness gap: Chen 2024 [bundle:10] vs Bailey 2022 [bundle:19]; Bailey 2022 [bundle:19] (direct, A1) vs Chen 2024 [bundle:10] (review) on safety comorbidity — direct vs indirect must be kept separate
- Severity 3 indirectness gap: Albulushi 2025 [bundle:12] vs Szklarz 2026 [bundle:18]; Albulushi 2025 [bundle:12] (direct, A1) vs Szklarz 2026 [bundle:18] (indirect) on contextual other — direct vs indirect must be kept separate
- Severity 3 indirectness gap: Suciu 2025 [bundle:5] vs Kochanowska 2026 [bundle:14]; Kochanowska 2026 [bundle:14] (direct, A1) vs Suciu 2025 [bundle:5] (review) on cardiometabolic — direct vs indirect must be kept separate
- Severity 3 indirectness gap: Suciu 2025 [bundle:5] vs Huang 2026 [bundle:3]; Huang 2026 [bundle:3] (direct, A1) vs Suciu 2025 [bundle:5] (review) on cardiometabolic — direct vs indirect must be kept separate

## References

- **Cersosimo 2025.** _Impact of SGLT2 inhibitors on endothelial function and echocardiographic parameters in dilated cardiomyopathy._ Journal of Cardiovascular Medicine (Hagerstown, Md.), 2025. DOI: 10.2459/JCM.0000000000001733 PMID: 40472172.
- **Jiang 2022.** _Comparative Cardiovascular Outcomes of SGLT2 Inhibitors in Type 2 Diabetes Mellitus: A Network Meta-Analysis of Randomized Controlled Trials._ Frontiers in Endocrinology, 2022. DOI: 10.3389/fendo.2022.802992 PMID: 35370961.
- **Huang 2026.** _A Prospective Cohort Study on the Impact of SGLT2 Inhibitors on the 12‑Month Recurrence Risk of Atrial Fibrillation After Catheter Ablation._ Drug Design, Development and Therapy, 2026. DOI: 10.2147/DDDT.S580640 PMID: 41913736.
- **Jansz 2026.** _Kidney outcomes with GLP-1 receptor agonists in people with type 2 diabetes already receiving SGLT2 inhibitors: a target trial emulation study using UK primary care data._ The Lancet. Primary Care, 2026. DOI: 10.1016/j.lanprc.2026.100139 PMID: 42109572.
- **Suciu 2025.** _Do SGLT2 Inhibitors Improve Cardiovascular Outcomes After Acute Coronary Syndrome Regardless of Diabetes? A Systematic Review and Meta-Analysis._ Medicina, 2025. DOI: 10.3390/medicina61101866 PMID: 41155853.
- **Kaze 2022.** _Association of SGLT2 inhibitors with cardiovascular, kidney, and safety outcomes among patients with diabetic kidney disease: a meta-analysis._ Cardiovascular Diabetology, 2022. DOI: 10.1186/s12933-022-01476-x PMID: 35321742.
- **Sato 2025.** _Adding Semaglutide to SGLT2 Inhibitors Reduces Liver Enzymes in Patients with Type 2 Diabetes Complicated by Metabolic Dysfunction Associated Steatotic Liver Disease: A Retrospective Observational Study._ Internal Medicine, 2025. DOI: 10.2169/internalmedicine.6239-25 PMID: 41062312.
- **Ahmed 2025.** _SGLT2 inhibitors versus GLP-1 receptor agonists for major adverse cardiovascular events in type 2 diabetes: a systematic review and meta-analysis of randomized controlled trials._ BMC Cardiovascular Disorders, 2025. DOI: 10.1186/s12872-025-05455-4 PMID: 41454299.
- **Rawish 2026.** _SGLT2 inhibitors are associated with improved long-term survival in Takotsubo syndrome: insights from large-scale real-world data._ European Heart Journal. Cardiovascular Pharmacotherapy, 2026. DOI: 10.1093/ehjcvp/pvaf088 PMID: 41549637.
- **Chen 2024.** _Effects of SGLT2 inhibitors on cardiac function and health status in chronic heart failure: a systematic review and meta-analysis._ Cardiovascular Diabetology, 2024. DOI: 10.1186/s12933-023-02042-9 PMID: 38172861.
- **Nobayashi 2026.** _Comparison of SGLT2 Inhibitors for New‐Onset Proteinuria Risk in Patients With Type 2 Diabetes and Preserved Kidney Function._ Diabetes, Obesity & Metabolism, 2026. DOI: 10.1111/dom.70625 PMID: 41804189.
- **Albulushi 2025.** _Impact of SGLT2 inhibitors on myocardial fibrosis in diabetic HFpEF: a longitudinal study._ European Journal of Medical Research, 2025. DOI: 10.1186/s40001-025-02834-7 PMID: 40624547.
- **Moreno-Perez 2026.** _Inpatient safety, effectiveness of SGLT2 inhibitors and GLP-1 RAs in type 2 diabetes: ENDOCARE, a pragmatic prospective cohort study._ Cardiovascular Diabetology, 2026. DOI: 10.1186/s12933-026-03156-6 PMID: 41913157.
- **Kochanowska 2026.** _Evidence on SGLT2 Inhibitors’ Efficacy in Older and Frail Patients._ Journal of Clinical Medicine, 2026. DOI: 10.3390/jcm15062219 PMID: 41899142.
- **Correale 2021.** _Switch to SGLT2 Inhibitors and Improved Endothelial Function in Diabetic Patients with Chronic Heart Failure._ Cardiovascular Drugs and Therapy, 2021. DOI: 10.1007/s10557-021-07254-3 PMID: 34519913.
- **Chen 2026.** _Agent-and Dose-Specific Intestinal Obstruction Safety of GLP-1 Receptor Agonists and SGLT2 Inhibitors: A Network Meta-Analysis of Randomized Trials._ International Journal of Molecular Sciences, 2026. DOI: 10.3390/ijms27020608 PMID: 41596262.
- **Elrakaybi 2022.** _Cardiovascular protection by SGLT2 inhibitors – Do anti-inflammatory mechanisms play a role?._ Molecular Metabolism, 2022. DOI: 10.1016/j.molmet.2022.101549 PMID: 35863639.
- **Szklarz 2026.** _Another pleiotropic effect of SGLT2 inhibitors: Is it a new frontier in thyroid function regulation?._ Thyroid Research, 2026. DOI: 10.1186/s13044-025-00282-3 PMID: 41491704.
- **Bailey 2022.** _Renal Protection with SGLT2 Inhibitors: Effects in Acute and Chronic Kidney Disease._ Current Diabetes Reports, 2022. DOI: 10.1007/s11892-021-01442-z PMID: 35113333.
- **Gao 2026.** _Repurposing SGLT2 Inhibitors for Cirrhotic Ascites: From Mechanistic Research to Clinical Exploration._ Journal of Clinical and Translational Hepatology, 2026. DOI: 10.14218/JCTH.2025.00465 PMID: 41810110.
- **Kawanami 2017.** _SGLT2 Inhibitors as a Therapeutic Option for Diabetic Nephropathy._ International Journal of Molecular Sciences, 2017. DOI: 10.3390/ijms18051083 PMID: 28524098.

{
  "article_type": "research_synthesis",
  "domain_slug": "longevity",
  "researka_object_type": "submission",
  "researka_submission_id": "a99d76b1-cb59-47ed-833e-7e42dd0d7183",
  "title": "Research Synthesis: SGLT2 Inhibitors Subgroups"
}