eGFR- and Potassium-Guided Stewardship of SGLT2 Inhibitors and Finerenone in Albuminuric Chronic Kidney Disease Associated with Type 2 Diabetes Authors: Chen Su, Xumin Li, Shuang Wang, Dan Liu Published: 2026 ABSTRACT AbstractAlbuminuric chronic kidney disease (CKD) associated with type 2 diabetes remains a high-risk condition despite long-standing renin–angiotensin system (RAS) inhibition. Sodium–glucose cotransporter 2 (SGLT2) inhibitors and finerenone provide complementary kidney and cardiovascular protection through distinct hemodynamic, tubular, inflammatory, and fibrotic mechanisms. The Phase 2, randomized, double-blind Combination Effect of Finerenone and Empagliflozin in Participants with Chronic Kidney Disease and Type 2 Diabetes Using a Urinary Albumin-to-Creatinine Ratio Endpoint (CONFIDENCE) trial subsequently showed that simultaneous finerenone–empagliflozin initiation produced greater short-term albuminuria reduction than either monotherapy, but this biomarker benefit does not establish hard-outcome superiority or define the optimal sequencing strategy. This narrative review proposes a stewardship framework guided by estimated glomerular filtration rate (eGFR) and potassium for adults with albuminuric type 2 diabetes–associated CKD in whom SGLT2 inhibitors and finerenone are being implemented on a RAS-inhibition background. The review emphasizes that kidney protection depends not only on prescribing effective therapies, but also on maintaining exposure through initiation, monitoring, temporary interruption, de-escalation, and re-initiation. A modest early eGFR decline after SGLT2 inhibitor initiation usually reflects a functional hemodynamic response and should not be treated in isolation as drug-related kidney injury. By contrast, finerenone-associated hyperkalemia requires baseline risk stratification, early potassium surveillance, correction of reversible triggers, and structured re-challenge when feasible. Simultaneous initiation may be appropriate in patients with persistent albuminuria, stable eGFR, normal or low-normal potassium, and reliable monitoring access, whereas sequential initiation is often more suitable when potassium risk, hemodynamic instability, or follow-up uncertainty is present. Future studies should move beyond short-term urine albumin-to-creatinine ratio (UACR) lowering to evaluate clinical kidney and cardiovascular outcomes, treatment persistence, potassium-event trajectories, re-initiation after temporary interruption, and implementation fidelity in routine care. Introduction Albuminuric chronic kidney disease (CKD) associated with type 2 diabetes mellitus is characterized by sustained kidney and cardiovascular risk despite conventional risk factor control. Renin–angiotensin system (RAS) inhibition has long provided the therapeutic foundation for proteinuric diabetic kidney disease. Landmark trials with angiotensin-converting enzyme inhibitors and angiotensin receptor blockers established that interrupting the RAS reduces albuminuria and slows diabetic nephropathy progression in overt nephropathy, with supportive evidence extending across earlier albuminuric or high-risk stages.1–5 Later trials of dual RAS blockade defined the limits of single-pathway intensification, showing increased hyperkalemia and acute kidney injury without sufficient clinical benefit.6,7 This history remains instructive for contemporary therapy: albuminuria lowering is clinically meaningful, but mechanistic plausibility must be tested against tolerability, biochemical surveillance, and long-term treatment persistence. The therapeutic landscape has now moved beyond RAS inhibition alone. Sodium–glucose cotransporter 2 inhibitors (SGLT2i) reduce kidney disease progression in type 2 diabetes and CKD, with several trials also demonstrating benefits in selected cardiovascular outcomes, particularly cardiovascular death or heart failure hospitalization-related endpoints.8–10 Nonsteroidal mineralocorticoid receptor antagonism with finerenone has reduced the risk of kidney and cardiovascular outcomes in patients with type 2 diabetes–associated CKD receiving optimized RAS blockade, as demonstrated in the Phase 3 randomized, double-blind, placebo-controlled Finerenone in Reducing Kidney Failure and Disease Progression in Diabetic Kidney Disease (FIDELIO-DKD) and Finerenone in Reducing Cardiovascular Mortality and Morbidity in Diabetic Kidney Disease (FIGARO-DKD) trials, and in their prespecified individual participant-level pooled analysis, Finerenone in Chronic Kidney Disease and Type 2 Diabetes: Combined FIDELIO-DKD and FIGARO-DKD Trial Programme Analysis (FIDELITY).11–13 The addition of glucagon-like peptide-1 receptor agonists, particularly after FLOW, has further expanded the cardiorenal treatment armamentarium.14 Together, these developments have shifted albuminuric type 2 diabetes–associated CKD from a single-drug model toward coordinated multi-mechanism treatment. As outcome evidence has matured, a major unresolved problem is how reliably these therapies are delivered outside trial settings. Kidney protection in routine care is often limited by delayed initiation, premature interruption, dose limitation, and incomplete re-initiation after temporary safety events. Early estimated glomerular filtration rate (eGFR) decline after SGLT2i initiation may be misinterpreted as drug nephrotoxicity, although modest early dips are usually hemodynamic and should be interpreted in clinical context.15 Hyperkalemia during RAS inhibition or finerenone therapy can prompt treatment discontinuation, although many events warrant structured assessment of reversible triggers, temporary dose adjustment or interruption, and planned re-challenge when clinically appropriate.16 Acute illness, prolonged fasting, or surgery may require temporary withholding of SGLT2i, making explicit re-initiation criteria after clinical recovery an important component of safe implementation.17 In this setting, benefit depends not only on initiation, but also on accurate interpretation of early biochemical signals, temporary rather than permanent interruption when appropriate, and active restart after reversible risk periods. The Phase 2, randomized, double-blind Combination Effect of Finerenone and Empagliflozin in Participants with Chronic Kidney Disease and Type 2 Diabetes Using a Urinary Albumin-to-Creatinine Ratio Endpoint (CONFIDENCE) trial brought this sequencing question into sharper focus. In patients with CKD and type 2 diabetes, simultaneous initiation of finerenone and empagliflozin produced a greater reduction in urine albumin-to-creatinine ratio (UACR) at 180 days than either treatment alone, with a 29% greater reduction compared with finerenone and a 32% greater reduction compared with empagliflozin.18 This trial provides randomized evidence that nonsteroidal mineralocorticoid receptor antagonism and SGLT2 inhibition can produce additive short-term albuminuria lowering. Nevertheless, CONFIDENCE was not a long-term hard-outcome trial. It is best interpreted as proof-of-concept evidence for additive albuminuria lowering under protocolized initiation, not as definitive evidence that simultaneous initiation reduces kidney failure, cardiovascular death, or all-cause mortality. This review focuses on adults with albuminuric type 2 diabetes–associated CKD in whom SGLT2 inhibitors and finerenone are implemented against a background of ongoing or intended RAS inhibition. It examines initiation strategy, early eGFR interpretation, potassium surveillance, temporary interruption, de-escalation after biochemical adverse events, and re-initiation after recovery. GLP-1 receptor agonists are discussed only as adjacent therapy because their implementation is not primarily governed by the eGFR–potassium signals that structure this framework. Literature Search Strategy, Study Selection, and Evidence Approach Relevant literature was identified through searches of PubMed/MEDLINE, Embase, Web of Science Core Collection, Scopus, and the Cochrane Library from database inception to June 2026. Additional trial, regulatory, and guideline-related information was checked through ClinicalTrials.gov, regulatory and prescribing-information sources where relevant, and major kidney and diabetes guideline websites. Search terms combined controlled vocabulary and free-text terms related to type 2 diabetes, diabetic kidney disease, albuminuric chronic kidney disease, UACR, eGFR, SGLT2 inhibitors, canagliflozin, dapagliflozin, empagliflozin, finerenone, mineralocorticoid receptor antagonism, renin–angiotensin system inhibition, early eGFR decline, hyperkalemia, treatment interruption, re-initiation, persistence, and implementation. Reference lists of pivotal trials, pooled analyses, post hoc analyses, guidelines, and relevant reviews were also screened manually. This narrative review addresses the implementation of SGLT2 inhibitors and finerenone in adults with albuminuric CKD associated with type 2 diabetes, usually on a background of ongoing or intended RAS inhibition.8,11,12 The evidence base is anchored in randomized kidney and cardiovascular outcome trials and prespecified pooled analyses,9,10,13 with post hoc safety analyses used when they inform early eGFR change or potassium risk. Observational studies are considered for implementation questions, including discontinuation, re-initiation, and persistence-related therapeutic loss, but not as substitutes for randomized efficacy evidence.15,16,19,20 Kidney transplant recipients, dialysis-dependent patients, nonalbuminuric CKD, isolated heart failure without albuminuric CKD, and primarily non-diabetic glomerular disease fall outside the principal scope. Evidence from broader CKD or heart failure populations is used only when it clarifies mechanisms, early eGFR behavior, potassium risk, or transferable implementation principles. Guidelines, regulatory documents, press releases, and conference abstracts were treated as contextual rather than primary evidentiary sources. Baseline characteristics and implementation relevance of the pivotal clinical studies are summarized in Table 1, and additional implementation and hyperkalemia-management evidence is summarized in Supplementary Table S1.Table 1Baseline Characteristics and Implementation Relevance of Key Clinical Studies Informing the Stewardship FrameworkEvidence SourceDesign; NPopulation and AgeBaseline eGFR, Mean or Median as Reported (mL/min/1.73 m2)Baseline UACR, Median Unless Otherwise Stated (mg/g)Background/Eligibility FeaturesIntervention or ComparisonRelevance to This ReviewCREDENCE8,15,21Randomized, double-blind, placebo-controlled kidney-outcome trial; N=4,401Adults with type 2 diabetes and albuminuric CKD; mean age 63.0 yearsMean 56.2Median 927eGFR 30 to <90; UACR >300 to 5000; stable/maximally tolerated ACEi or ARB background therapyCanagliflozin 100 mg vs placeboPrimary SGLT2 inhibitor hard-outcome anchor; supports interpretation of albuminuria response and early eGFR change in albuminuric diabetic CKD.DAPA-CKD9Randomized, double-blind, placebo-controlled kidney-outcome trial; N=4,304Adults with CKD with or without type 2 diabetes; mean age 61.8 years; 67.5% had type 2 diabetesMean 43.1Median 949eGFR 25 to 75; UACR 200 to 5000; ACEi or ARB unless contraindicated or not toleratedDapagliflozin 10 mg vs placeboExtends SGLT2 inhibitor kidney-outcome evidence across albuminuric CKD, including but not limited to diabetic CKD.EMPA-KIDNEY10Randomized, double-blind, placebo-controlled kidney-outcome trial; N=6,609Broad CKD population with or without diabetes; mean age 63.8 years; approximately 54% had no diabetesMean 37.3Median 329eGFR ≥20 to <45 regardless of albuminuria, or eGFR ≥45 to <90 with UACR ≥200; RAS inhibitor use when clinically appropriateEmpagliflozin 10 mg vs placeboBroadens SGLT2 inhibitor evidence to lower eGFR and lower-albuminuria strata; used cautiously because the review focus is albuminuric type 2 diabetes-associated CKD.FIDELIO-DKD11,16Randomized, double-blind, placebo-controlled Phase 3 trial; N=5,734Adults with type 2 diabetes and CKD enriched for kidney risk; mean age 65.6 yearsMean 44.3Median 851Maximum tolerated ACEi or ARB; serum K+ ≤4.8 mmol/L at screening/run-in; albuminuric CKD eligibility based on UACR/eGFR strataFinerenone vs placeboPrincipal finerenone kidney-outcome and potassium-risk evidence; supports baseline potassium stratification, monitoring, interruption, and re-challenge logic.FIGARO-DKD12Randomized, double-blind, placebo-controlled Phase 3 trial; N=7,352 analyzed / 7437 randomizedAdults with type 2 diabetes and CKD with relatively earlier kidney disease than FIDELIO-DKD; mean age 64.1 yearsMean 67.8Median 308 (IQR 108 to 740)Maximum tolerated RAS blockade; serum K+ ≤4.8 mmol/L; eligibility included UACR 30 to <300 with eGFR 25 to 90, or UACR 300 to 5000 with eGFR ≥60Finerenone vs placeboComplements FIDELIO-DKD by supporting finerenone use across earlier CKD and cardiovascular-risk phenotypes.FIDELITY pooled analysis13Prespecified pooled analysis of FIDELIO-DKD and FIGARO-DKD; N=13,026Adults with type 2 diabetes and CKD across a broad kidney-risk spectrum; mean age 64.8 yearsMean 57.6Median 515RAS inhibitor use in 99.8%; baseline SGLT2 inhibitor use 6.7%; serum K+ eligibility inherited from parent trialsFinerenone vs placeboProvides integrated finerenone efficacy and safety estimates across CKD severity; useful for framing external validity and residual risk.CONFIDENCE baseline/main trial18,22Randomized, double-blind, multicenter Phase 2 UACR-endpoint trial; N=818 randomized; baseline table N=801Adults with type 2 diabetes, CKD, and albuminuria; mean age 66.5 yearsMean 54.2Median 583 (IQR 292 to 1140)eGFR 30 to 90; UACR ≥100 to <5,000; clinically maximum tolerated RAS inhibitor for >1 month before screeningFinerenone + empagliflozin vs finerenone + placebo vs empagliflozin + placeboMost direct evidence for simultaneous initiation vs monotherapy; supports the review’s sequencing discussion, but only for short-term albuminuria response.FLOW14Randomized, double-blind, placebo-controlled kidney-outcome trial; N=3,533Adults with type 2 diabetes and CKD; mean age 66.6 yearsMean 47.0Median 567.6ACEi/ARB use unless not tolerated or contraindicated; baseline SGLT2 inhibitor use 15.6%Once-weekly semaglutide vs placeboPositions GLP-1 receptor agonists as adjacent kidney-protective therapy outside the main eGFR-potassium stewardship axis.Note: eGFR and UACR values are shown using the statistic reported by each source; “by group” indicates treatment-arm-specific values.Abbreviations: CKD, chronic kidney disease; eGFR, estimated glomerular filtration rate; UACR, urine albumin-to-creatinine ratio; RAS, renin-angiotensin system; ACEi, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker; SGLT2, sodium-glucose cotransporter 2; GLP-1, glucagon-like peptide-1. Table 1Baseline Characteristics and Implementation Relevance of Key Clinical Studies Informing the Stewardship FrameworkEvidence SourceDesign; NPopulation and AgeBaseline eGFR, Mean or Median as Reported (mL/min/1.73 m2)Baseline UACR, Median Unless Otherwise Stated (mg/g)Background/Eligibility FeaturesIntervention or ComparisonRelevance to This ReviewCREDENCE8,15,21Randomized, double-blind, placebo-controlled kidney-outcome trial; N=4,401Adults with type 2 diabetes and albuminuric CKD; mean age 63.0 yearsMean 56.2Median 927eGFR 30 to <90; UACR >300 to 5000; stable/maximally tolerated ACEi or ARB background therapyCanagliflozin 100 mg vs placeboPrimary SGLT2 inhibitor hard-outcome anchor; supports interpretation of albuminuria response and early eGFR change in albuminuric diabetic CKD.DAPA-CKD9Randomized, double-blind, placebo-controlled kidney-outcome trial; N=4,304Adults with CKD with or without type 2 diabetes; mean age 61.8 years; 67.5% had type 2 diabetesMean 43.1Median 949eGFR 25 to 75; UACR 200 to 5000; ACEi or ARB unless contraindicated or not toleratedDapagliflozin 10 mg vs placeboExtends SGLT2 inhibitor kidney-outcome evidence across albuminuric CKD, including but not limited to diabetic CKD.EMPA-KIDNEY10Randomized, double-blind, placebo-controlled kidney-outcome trial; N=6,609Broad CKD population with or without diabetes; mean age 63.8 years; approximately 54% had no diabetesMean 37.3Median 329eGFR ≥20 to <45 regardless of albuminuria, or eGFR ≥45 to <90 with UACR ≥200; RAS inhibitor use when clinically appropriateEmpagliflozin 10 mg vs placeboBroadens SGLT2 inhibitor evidence to lower eGFR and lower-albuminuria strata; used cautiously because the review focus is albuminuric type 2 diabetes-associated CKD.FIDELIO-DKD11,16Randomized, double-blind, placebo-controlled Phase 3 trial; N=5,734Adults with type 2 diabetes and CKD enriched for kidney risk; mean age 65.6 yearsMean 44.3Median 851Maximum tolerated ACEi or ARB; serum K+ ≤4.8 mmol/L at screening/run-in; albuminuric CKD eligibility based on UACR/eGFR strataFinerenone vs placeboPrincipal finerenone kidney-outcome and potassium-risk evidence; supports baseline potassium stratification, monitoring, interruption, and re-challenge logic.FIGARO-DKD12Randomized, double-blind, placebo-controlled Phase 3 trial; N=7,352 analyzed / 7437 randomizedAdults with type 2 diabetes and CKD with relatively earlier kidney disease than FIDELIO-DKD; mean age 64.1 yearsMean 67.8Median 308 (IQR 108 to 740)Maximum tolerated RAS blockade; serum K+ ≤4.8 mmol/L; eligibility included UACR 30 to <300 with eGFR 25 to 90, or UACR 300 to 5000 with eGFR ≥60Finerenone vs placeboComplements FIDELIO-DKD by supporting finerenone use across earlier CKD and cardiovascular-risk phenotypes.FIDELITY pooled analysis13Prespecified pooled analysis of FIDELIO-DKD and FIGARO-DKD; N=13,026Adults with type 2 diabetes and CKD across a broad kidney-risk spectrum; mean age 64.8 yearsMean 57.6Median 515RAS inhibitor use in 99.8%; baseline SGLT2 inhibitor use 6.7%; serum K+ eligibility inherited from parent trialsFinerenone vs placeboProvides integrated finerenone efficacy and safety estimates across CKD severity; useful for framing external validity and residual risk.CONFIDENCE baseline/main trial18,22Randomized, double-blind, multicenter Phase 2 UACR-endpoint trial; N=818 randomized; baseline table N=801Adults with type 2 diabetes, CKD, and albuminuria; mean age 66.5 yearsMean 54.2Median 583 (IQR 292 to 1140)eGFR 30 to 90; UACR ≥100 to <5,000; clinically maximum tolerated RAS inhibitor for >1 month before screeningFinerenone + empagliflozin vs finerenone + placebo vs empagliflozin + placeboMost direct evidence for simultaneous initiation vs monotherapy; supports the review’s sequencing discussion, but only for short-term albuminuria response.FLOW14Randomized, double-blind, placebo-controlled kidney-outcome trial; N=3,533Adults with type 2 diabetes and CKD; mean age 66.6 yearsMean 47.0Median 567.6ACEi/ARB use unless not tolerated or contraindicated; baseline SGLT2 inhibitor use 15.6%Once-weekly semaglutide vs placeboPositions GLP-1 receptor agonists as adjacent kidney-protective therapy outside the main eGFR-potassium stewardship axis.Note: eGFR and UACR values are shown using the statistic reported by each source; “by group” indicates treatment-arm-specific values.Abbreviations: CKD, chronic kidney disease; eGFR, estimated glomerular filtration rate; UACR, urine albumin-to-creatinine ratio; RAS, renin-angiotensin system; ACEi, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker; SGLT2, sodium-glucose cotransporter 2; GLP-1, glucagon-like peptide-1. Baseline Characteristics and Implementation Relevance of Key Clinical Studies Informing the Stewardship Framework Note: eGFR and UACR values are shown using the statistic reported by each source; “by group” indicates treatment-arm-specific values. Abbreviations: CKD, chronic kidney disease; eGFR, estimated glomerular filtration rate; UACR, urine albumin-to-creatinine ratio; RAS, renin-angiotensin system; ACEi, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker; SGLT2, sodium-glucose cotransporter 2; GLP-1, glucagon-like peptide-1. Conceptual Framework: From Treatment Addition to Treatment Stewardship The proposed framework is organized around five recurring phases: pre-initiation assessment, initiation strategy, early monitoring, event-driven adjustment, and re-initiation/long-term stewardship. Pre-initiation assessment should confirm persistent albuminuric risk and characterize RAS inhibitor status, recent eGFR trajectory, baseline serum potassium, blood pressure and volume status, intercurrent illness, diuretic exposure, and monitoring capacity.18 Together, these factors help determine whether simultaneous finerenone–empagliflozin initiation is reasonable, whether SGLT2 inhibitor–first sequencing is more appropriate, or whether finerenone should be delayed until potassium and follow-up conditions are suitable. Early monitoring should remain signal-specific. A modest, early, clinically stable eGFR dip after SGLT2 inhibitor initiation usually supports continued therapy with observation, whereas a potassium rise after finerenone requires confirmation, severity classification, correction of reversible triggers, and dose adjustment or temporary interruption when needed.15,16 UACR reduction can serve as an early biologic response marker, but it should be interpreted alongside eGFR trajectory, potassium events, treatment persistence, and longer-term cardiorenal outcomes.21 The framework is intended to function iteratively. Acute illness, surgery, fasting, volume depletion, acute kidney injury, or clinically significant hyperkalemia may justify temporary interruption, but the trigger, restart criteria, and reassessment timing should be explicit. Long-term stewardship therefore evaluates not only whether therapy is prescribed, but whether it is maintained, appropriately interrupted, and reconsidered for restart after reversible risk periods. Evidence Base and Implementation Constraints RAS Inhibition as the Background Foundation The current multi-drug era in diabetic CKD extends, rather than replaces, the RAS-inhibition era. Early randomized trials established RAS blockade as the background standard for proteinuric diabetic kidney disease, with benefits across established nephropathy and earlier albuminuric risk states.1–5 Captopril, losartan, and irbesartan reduced hard or progressive kidney outcomes in diabetic nephropathy, while IRMA-2 and BENEDICT supported benefit in earlier albuminuric or high-risk disease.1–5 The subsequent experience with intensified RAS-pathway blockade is equally instructive. In ONTARGET, combined telmisartan and ramipril reduced proteinuria more than either monotherapy but worsened major renal outcomes overall.6 VA NEPHRON-D, conducted in patients with type 2 diabetes and proteinuric CKD, was stopped early because combined angiotensin-converting enzyme inhibitor and angiotensin receptor blocker therapy increased hyperkalemia and acute kidney injury without sufficient evidence of clinical benefit.7 These trials remain a useful caution for contemporary combination therapy: albuminuria lowering is not, by itself, sufficient to justify amplifying a single pathway when biochemical toxicity and hemodynamic kidney injury increase in parallel.6,7 RAS inhibition therefore remains foundational but incomplete. It reduces glomerular hemodynamic stress and proteinuria, yet substantial residual kidney and cardiovascular risk persist. SGLT2 inhibitors and nonsteroidal mineralocorticoid receptor antagonists add mechanisms that are not reducible to RAS blockade: SGLT2 inhibitors act through proximal tubular sodium–glucose cotransport inhibition, restoration of tubuloglomerular feedback, and reduction of intraglomerular pressure,23 whereas finerenone targets mineralocorticoid receptor–mediated inflammatory and fibrotic signaling.24 The practical question is how these complementary mechanisms can be layered onto RAS inhibition while preserving tolerability, treatment persistence, and the possibility of re-initiation after temporary interruption. This history is also relevant to discontinuation decisions. In advanced and progressive CKD, routine discontinuation of RAS inhibitors solely because kidney function is low has not been shown to improve long-term kidney function. STOP-ACEi found no significant between-group difference in eGFR after discontinuation vs continuation of RAS inhibitors.25 Temporary withdrawal may be appropriate during acute illness, severe hyperkalemia, hypotension, or acute kidney injury, but reflexive permanent discontinuation can remove a therapy with established kidney and cardiovascular value. The same logic applies to newer kidney-protective agents: transient biochemical or hemodynamic instability should prompt reassessment, correction of reversible triggers, and re-initiation planning when feasible, rather than withdrawal without reassessment. SGLT2 Inhibitors and the Early eGFR Dip Outcome Evidence and Clinical Positioning SGLT2 inhibitors have a mature hard-outcome evidence base in CKD and a reproducible early functional eGFR signal. CREDENCE established kidney protection with canagliflozin in type 2 diabetes and albuminuric CKD on background RAS blockade;8 DAPA-CKD extended benefit to albuminuric CKD with and without diabetes;9 and EMPA-KIDNEY broadened the trial-tested CKD population for empagliflozin.10 Earlier cardiovascular outcome trials and meta-analytic evidence provided supportive kidney and heart-failure signals, but dedicated kidney-outcome trials remain the central evidence base for albuminuric CKD.26–31 Mechanism of Early eGFR Decline SGLT2 inhibitor initiation is commonly followed by an early decline in eGFR, a pattern generally interpreted as a functional hemodynamic dip rather than structural kidney injury.32 By inhibiting proximal tubular sodium–glucose reabsorption, SGLT2 inhibitors increase sodium chloride delivery to the macula densa, restore tubuloglomerular feedback, promote afferent arteriolar vasoconstriction, and lower intraglomerular pressure.23 A modest early decline in eGFR is therefore consistent with the same glomerular hemodynamic shift thought to contribute to long-term kidney protection. Post hoc analyses from kidney outcome trials support this interpretation. In CREDENCE, canagliflozin produced early, sustained albuminuria reductions that were associated with subsequent kidney and cardiovascular outcomes.21 A separate CREDENCE analysis showed that canagliflozin was associated with an acute, largely reversible eGFR drop, while long-term eGFR trajectories remained more favorable with active treatment despite the early dip.15 Clinical interpretations of the SGLT2i-associated eGFR dip emphasize that small-to-moderate early declines should not be misclassified as acute kidney injury in isolation. Larger, progressive, or clinically discordant declines require evaluation, but an isolated early dip should not be treated as evidence of drug-related nephrotoxicity.32 In clinical implementation, early eGFR decline should be classified before being attributed to drug nephrotoxicity. Interpretation should consider the magnitude and trajectory of decline, blood pressure, volume status, diuretic exposure, intercurrent illness, nephrotoxins, obstruction risk, and concurrent biochemical abnormalities. The key distinction is between an early, proportionate, hemodynamically plausible dip and a progressive or clinically discordant decline suggesting acute kidney injury. Early eGFR Declines Should Prompt Reassessment A modest, clinically stable SGLT2i-associated eGFR dip is usually functional, but a large, progressive, or clinically discordant decline should trigger reassessment, particularly when the decline approaches or exceeds approximately 30% from baseline, because excessive early eGFR decline after SGLT2i initiation has been associated with higher subsequent cardiorenal risk in observational data.33,34 This threshold should therefore be interpreted as a pragmatic reassessment trigger, analogous to the approach often used after RAS-pathway hemodynamic interventions, rather than as an automatic stopping rule. Clinical context should modify the response: frailty, orthostasis, or volume depletion may warrant action despite a smaller decline, whereas a clinically stable isolated dip can usually be observed. Early eGFR decline is therefore best handled as a signal requiring clinical classification rather than as a uniform safety endpoint. The first category is the expected functional dip: modest, early, clinically stable, and potentially accompanied by stable or improved albuminuria. The second category is compounded hemodynamic stress, in which the expected dip is amplified by diuretic exposure, reduced oral intake, intercurrent illness, or hypotension. The third category is evolving or established acute kidney injury, suggested by systemic illness, inflammatory urine sediment, obstruction, sepsis, severe volume depletion, or persistent deterioration despite correction of hemodynamic stressors. The latter two categories should generally prompt closer laboratory reassessment, correction of reversible triggers, and, when clinically indicated, temporary withholding; an isolated expected functional dip usually supports continued therapy with observation. Finerenone and Potassium-Limited Persistence Outcome Evidence Mineralocorticoid receptor overactivation promotes inflammatory and fibrotic cardiorenal signaling, while steroidal mineralocorticoid receptor antagonists (MRAs) are limited in CKD by hyperkalemia and, with spironolactone, antiandrogen-related adverse effects.24 Finerenone, a selective nonsteroidal MRA, first demonstrated albuminuria lowering in ARTS-DN and subsequently reduced kidney and cardiovascular outcomes in FIDELIO-DKD, FIGARO-DKD, and the prespecified FIDELITY pooled analysis.11–13,35 Its principal implementation constraint is potassium surveillance: unlike SGLT2 inhibitors, which are commonly framed by an early functional eGFR dip, finerenone exposure is most often limited by hyperkalemia risk. Hyperkalemia Risk FIDELIO-DKD showed that finerenone increased hyperkalemia risk compared with placebo, and a dedicated post hoc analysis further characterized this risk profile. Risk was not uniform: higher baseline serum potassium, lower eGFR, higher UACR, β-blocker use, and finerenone assignment independently predicted treatment-emergent ≥mild hyperkalemia, whereas diuretic use and SGLT2 inhibitor use were associated with lower risk. Because participants were treated on a background of optimized RAS inhibition, RAS blockade is better interpreted here as the therapeutic background rather than as a within-trial discriminator of hyperkalemia risk. Importantly, clinically consequential hyperkalemia remained uncommon in the trial setting, where potassium monitoring, dose adjustment, temporary interruption, and restart criteria were protocolized.16 For routine care, hyperkalemia should be anticipated and classified, because a single potassium excursion does not necessarily establish long-term intolerance. Management begins before drug initiation, with assessment of baseline serum potassium, eGFR, albuminuria burden, and modifiable potassium-raising contributors, including relevant comedications, dietary changes, acute illness, and volume depletion. After initiation or dose adjustment, serum potassium should be reassessed early and periodically, with monitoring intensity guided by baseline risk and early biochemical changes. De-Escalation and Re-Initiation Finerenone implementation should therefore include a prespecified response to potassium elevation. Mild hyperkalemia should prompt confirmation, review of reversible dietary, medication-related, and intercurrent-illness contributors, and repeat testing. Clinically significant or persistent elevation may require dose reduction or temporary interruption, with adjunctive potassium-lowering measures reserved for selected patients in whom continued cardiorenal protective therapy remains clinically important. Severe hyperkalemia requires urgent management and interruption of potassium-raising therapy, followed by reassessment once serum potassium has normalized.16,36,37 Re-initiation should be prespecified as part of the treatment pathway, not left to ad hoc follow-up. Patients who develop hyperkalemia frequently overlap with those at high residual kidney and cardiovascular risk, so permanent discontinuation after a manageable potassium event may unnecessarily reduce exposure to an outcome-improving therapy.13 A more durable approach is temporary de-escalation followed by reassessment for lower-dose re-challenge, with potassium-lowering support considered when appropriate after eGFR stability, potassium normalization, and correction of reversible precipitants.16 Combination Therapy After CONFIDENCE: Greater Short-Term Albuminuria Lowering, Unresolved Long-Term Sequencing SGLT2 inhibitors and finerenone address overlapping cardiorenal risk through distinct, nonredundant mechanisms. SGLT2 inhibition primarily modifies tubular sodium–glucose handling and tubuloglomerular feedback, thereby lowering intraglomerular pressure, whereas finerenone targets mineralocorticoid receptor–mediated inflammatory and fibrotic signaling in cardiorenal tissues. This mechanistic separation provides the biological rationale for combined therapy, but it does not by itself determine how the two agents should be sequenced in routine care.18,22–24 Potassium safety requires separate interpretation. At the class level, an individual participant data meta-analysis showed that SGLT2 inhibitors reduced serious hyperkalemia without increasing hypokalemia.38 This signal is compatible with a more favorable potassium-risk context when SGLT2 inhibition is present, but it should not be extrapolated to mean that empagliflozin prevents finerenone-associated hyperkalemia. In the prespecified CONFIDENCE hyperkalemia analysis, finerenone–empagliflozin combination therapy did not significantly reduce mean potassium change or the odds of hyperkalemia compared with finerenone alone over 180 days.39 Thus, finerenone-containing regimens require active potassium surveillance even when an SGLT2 inhibitor is co-initiated. CONFIDENCE provides direct randomized evidence that simultaneous finerenone–empagliflozin initiation produces greater short-term UACR reduction than either monotherapy in albuminuric CKD and type 2 diabetes on background RAS inhibition.18 This is a credible early response signal, supported by broader evidence linking albuminuria change to CKD progression risk.21,40 The respective long-term outcome-trial programs of SGLT2 inhibitors and finerenone have demonstrated slower CKD progression and reductions in clinically important composite kidney outcomes that include kidney-failure endpoints.8–13 However, CONFIDENCE remains a biomarker-endpoint trial and does not establish that simultaneous co-initiation provides greater long-term preservation of eGFR or reduces kidney failure, cardiovascular death, or mortality compared with sequential implementation.18 The implication is not categorical superiority of combination therapy, but a clearer basis for testing sequencing strategies. CONFIDENCE therefore supports more precise study of which patients should receive simultaneous initiation, which should start with SGLT2 inhibition before finerenone, and how dual therapy should be monitored, temporarily interrupted, and reintroduced after reversible risk events. Comparison of Simultaneous SGLT2 Inhibitor–Finerenone Initiation vs Sequential Initiation No validated algorithm or head-to-head outcome trial currently establishes a universally preferred simultaneous or sequential initiation strategy. We therefore propose a pragmatic, risk-based framework that balances residual albuminuric risk against potassium, kidney-function, hemodynamic, and monitoring vulnerability. The resulting choice among simultaneous initiation, SGLT2 inhibitor–first sequencing, and conditional finerenone deferral is detailed below and summarized in Figure 1.Figure 1eGFR- and potassium-guided implementation framework for SGLT2 inhibitor and finerenone therapy on a background of RAS inhibition in albuminuric type 2 diabetes-associated CKD. The closed-loop pathway integrates pre-initiation assessment, individualized simultaneous or sequential initiation, early reassessment of eGFR and potassium signals, event-driven temporary withholding or dose adjustment, and re-initiation after recovery. Treatment selection and monitoring are guided by albuminuric risk, eGFR stability, baseline serum potassium, hemodynamic status, recent acute illness or kidney injury, and monitoring feasibility.A flowchart of an eGFR- and potassium-guided framework for SGLT2 inhibitor and finerenone therapy.It begins with ′1. Pre-initiation assessment′ including factors like persistent albuminuria, stable eGFR, RAS inhibitor background, baseline serum potassium, recent AKI or illness, BP or volume, diuretics and monitoring feasibility. Arrows lead to ′2. Initiation strategy′ with options for simultaneous start, SGLT2 inhibitor first, or finerenone deferred, all on a RAS inhibitor background. Next is ′3. Early reassessment′ showing eGFR and serum potassium signals, with expected functional eGFR dip and AKI signal and mild potassium rise or clinically significant hyperkalemia. This leads to ′4. Temporary withholding or de-escalation′ due to acute illness, volume depletion, severe hyperkalemia, surgery or fasting and AKI. Finally, ′5. Re-initiation and long-term stewardship′ involves recovery, restarting when clinically stable, repeating eGFR or potassium, reassessing UACR and preserving exposure.Abbreviations: CKD, chronic kidney disease; eGFR, estimated glomerular filtration rate; RAS, renin–angiotensin system; SGLT2, sodium–glucose cotransporter 2. eGFR- and potassium-guided implementation framework for SGLT2 inhibitor and finerenone therapy on a background of RAS inhibition in albuminuric type 2 diabetes-associated CKD. The closed-loop pathway integrates pre-initiation assessment, individualized simultaneous or sequential initiation, early reassessment of eGFR and potassium signals, event-driven temporary withholding or dose adjustment, and re-initiation after recovery. Treatment selection and monitoring are guided by albuminuric risk, eGFR stability, baseline serum potassium, hemodynamic status, recent acute illness or kidney injury, and monitoring feasibility. Simultaneous initiation is best suited to patients in whom substantial residual albuminuric risk coexists with low anticipated early safety risk: persistent albuminuria despite RAS inhibition, stable eGFR, normal or low-normal serum potassium, stable blood pressure and volume status, no recent acute kidney injury or intercurrent illness, and reliable access to early laboratory follow-up. This profile approximates the trial-tested setting in which additive short-term albuminuria lowering was demonstrated, but it does not remove the need for early eGFR and potassium surveillance.18,39 Sequential initiation is preferable when baseline potassium is borderline, eGFR is low or unstable, recent illness or diuretic intensification has occurred, or timely laboratory follow-up cannot be assured. In such patients, SGLT2 inhibitor–first sequencing is reasonable because this class has robust kidney-outcome evidence and is associated, at the population level, with a lower risk of serious hyperkalemia.8–10,38 Finerenone can then be added after eGFR, volume status, and serum potassium have been reassessed and are suitable for monitored treatment.16 SGLT2 inhibitor initiation may be deferred during active ketosis or high ketoacidosis risk, perioperative fasting, severe frailty with poor intake or volume vulnerability, recurrent severe genital mycotic infection, or active complicated urinary infection.17,41 Finerenone should be deferred when baseline hyperkalemia, recent potassium instability, uncorrected potassium-raising comedications, or inability to complete early potassium monitoring makes safe exposure unlikely.16 Reversible barriers should be documented as conditional deferrals with explicit reassessment triggers. eGFR-Guided Stewardship Establishing a Stable Baseline Stewardship begins with baseline interpretation before prescribing. A single creatinine value should be interpreted against the recent clinical context, particularly after hospitalization, contrast exposure, diuretic intensification, infection, or volume depletion. When possible, eGFR used for initiation decisions should reflect a clinically stable outpatient baseline, and discordant recent values should prompt review of the trajectory rather than reliance on the most recent measurement alone.42 Albuminuria should also be interpreted longitudinally, because UACR has substantial within-person variability and may be transiently influenced by pre-analytical conditions, sampling context, and intercurrent clinical instability.43 For the purposes of this review, persistent albuminuria despite RAS inhibition defines the population most directly aligned with an SGLT2 inhibitor–finerenone stewardship framework, whereas a single elevated UACR should be confirmed when the clinical context suggests transient elevation. Interpreting Early eGFR Changes After SGLT2 inhibitor initiation, a modest, early eGFR dip should usually be interpreted as an expected functional response rather than as an isolated reason for discontinuation.15 Kidney function reassessment is most useful when baseline eGFR is low, volume reserve is limited, diuretic exposure is intensive, or intercurrent illness is present; in these settings, early laboratory changes should be interpreted against the expected SGLT2 inhibitor–associated dip rather than reflexively prompting treatment withdrawal.32 Finerenone may also be associated with an early change in eGFR, although potassium remains the more prominent implementation constraint. In the prespecified FIDELITY pooled analysis, the kidney and cardiovascular effects of finerenone were broadly consistent across categories of acute eGFR change, supporting interpretation of early eGFR change in relation to subsequent outcome trajectory rather than as an isolated toxicity signal.44 Across both drug classes, short-term functional changes should therefore be classified clinically and weighed against longer-term kidney-outcome evidence. Avoiding Permanent Therapeutic Loss Permanent discontinuation after an expected early eGFR dip, a manageable potassium elevation, or a reversible acute illness can convert a temporary safety response into long-term loss of kidney-protective treatment exposure. For RAS inhibitors specifically, observational studies after low eGFR or hyperkalemia suggest that discontinuation may reduce recurrent hyperkalemia in some analyses, but is also associated with higher mortality and cardiovascular risk signals compared with continuation.19,20,45 Although vulnerable to confounding by illness severity, these data support a practical principle: discontinuation should not be treated as neutral by default. In high-risk albuminuric CKD, stopping kidney-protective therapy should receive clinical scrutiny comparable to that applied when treatment is initiated. Potassium-Guided Stewardship Risk Stratification Before Finerenone Potassium stewardship should begin before finerenone initiation, with baseline serum potassium and eGFR used to determine eligibility and monitoring intensity. In the pivotal FIDELIO-DKD and FIGARO-DKD trials, serum potassium ≤4.8 mmol/L was required at screening/run-in, and patients had eGFR ≥25 mL/min/1.73 m2 within trial-specific albuminuria strata.11,12,16 In routine practice, finerenone should not be initiated when serum potassium is >5.0 mmol/L; values >4.8 to 5.0 mmol/L warrant clinical judgment and closer early monitoring, and initiation is generally not recommended when eGFR is <25 mL/min/1.73 m2.37 High-normal potassium, lower or unstable eGFR, greater albuminuria, prior hyperkalemia, heart failure, untreated metabolic acidosis, and concomitant potassium-raising therapy should prompt closer monitoring; when these risks are prominent, sequential rather than simultaneous SGLT2 inhibitor–finerenone initiation is usually preferred.16,36,37 Dietary counseling should avoid indiscriminate restriction of potassium-containing whole foods. In many patients with CKD, broad restriction of fruits, vegetables, legumes, and other plant foods may reduce fiber intake and cardiometabolic dietary quality without reliably stabilizing serum potassium. Counseling should instead prioritize recurrent hyperkalemia history, potassium-containing salt substitutes, processed potassium additives, and abrupt dietary changes that may destabilize serum potassium.46 Medication reconciliation should focus on modifiable contributors, including potassium supplements, potassium-containing salt substitutes, trimethoprim, nonsteroidal anti-inflammatory drugs, and other potassium-raising or nephrotoxic exposures. Beta-blockers should be recognized as a potential risk marker rather than routinely withdrawn when cardiovascular indications are strong.16,36 Monitoring After Initiation Finerenone trials used protocolized potassium surveillance, dose adjustment, temporary interruption, and restart rules. In routine care, potassium should be reassessed early after initiation or dose escalation—commonly around 4 weeks—and then monitored periodically, with testing frequency individualized according to baseline potassium, eGFR, prior hyperkalemia, albuminuria burden, and recent changes in RAS inhibitor or diuretic therapy.37 SGLT2 inhibitors reduce serious hyperkalemia risk at the class level, but this does not eliminate the need for potassium monitoring when finerenone is added.38 This is a population-level hyperkalemia-risk signal, not direct assurance against finerenone-associated hyperkalemia; it should not be used to justify initiating finerenone without surveillance in high-risk patients.39 Responding to Hyperkalemia Hyperkalemia should be managed as a classified biochemical event rather than as a uniform signal of drug intolerance. Initial assessment should confirm the result, exclude pseudohyperkalemia or hemolysis, and evaluate acute kidney injury, newly introduced potassium-raising medications, recent dietary change, constipation, untreated metabolic acidosis, volume depletion, and other factors that reduce renal potassium excretion. Mild elevations can often be managed with correction of reversible triggers and repeat testing, whereas moderate or persistent elevations may require finerenone dose reduction or temporary withholding. Severe hyperkalemia, particularly when accompanied by symptoms, electrocardiographic changes, rapid potassium rise, or acute kidney injury, requires urgent management and reassessment of potassium-raising therapy.36 Potassium binders may preserve treatment exposure in selected hyperkalemia-prone patients. OPAL-HK showed that patiromer lowered potassium and reduced recurrent hyperkalemia in CKD patients receiving RAAS inhibitors,47 while AMBER showed that patiromer enabled more persistent spironolactone use in resistant hypertension with advanced CKD.48 Sodium zirconium cyclosilicate also lowers potassium in randomized outpatient and hemodialysis studies.49,50 These data support potassium-lowering therapy as an exposure-preserving strategy, although binder-enabled finerenone persistence in albuminuric type 2 diabetes-CKD remains unproven. Re-Challenge After Hyperkalemia Re-challenge after hyperkalemia should be planned as part of the treatment pathway rather than left to informal follow-up. Once serum potassium has returned to an acceptable range and reversible contributors have been addressed, persistent albuminuria and high residual kidney or cardiovascular risk should prompt reassessment for possible finerenone re-initiation. Restart may involve a lower dose or retitration, closer early potassium monitoring, diuretic adjustment when clinically appropriate, correction of metabolic acidosis, or potassium binders in selected hyperkalemia-prone patients.37 This approach reduces the likelihood that a manageable biochemical event becomes unnecessary permanent discontinuation, while preserving individualized safety limits. Temporary Withholding, De-Escalation, and Re-Initiation Temporary withholding is an essential but selective component of safe implementation. For SGLT2 inhibitors, interruption is most appropriate when prolonged fasting, surgery, severe acute illness, reduced oral intake with dehydration, insulin deficiency, or marked catabolic stress materially increases ketogenesis or volume vulnerability.51 Perioperative evidence should be framed as a rare but clinically important safety signal: systematic case-report evidence links SGLT2 inhibitor exposure to perioperative ketoacidosis in the context of surgical stress, fasting, reduced carbohydrate intake, insulin deficiency, or intercurrent illness, whereas prospective perioperative cohort data suggest a low absolute incidence under contemporary management.52,53 Because hyperglycemia may be absent or modest, delayed recognition remains a central practical concern despite the low absolute incidence suggested by prospective perioperative data.53 RAS inhibitors and finerenone may also require review, dose adjustment, or temporary withholding during acute kidney injury, severe hypotension, or clinically significant hyperkalemia; however, low eGFR alone should not be treated as an automatic indication for permanent RAS inhibitor withdrawal.25 The rationale should remain drug-specific: SGLT2 inhibitor interruption is usually driven by ketogenesis or volume risk, finerenone interruption by potassium thresholds, and RAS inhibitor interruption by hemodynamic instability, acute kidney injury, or clinically significant hyperkalemia.37,51,54 The preferred model is event-specific interruption with a predefined route back to treatment. The clinical record should document the reason for interruption, criteria for re-exposure, and timing of laboratory reassessment. For SGLT2 inhibitors, re-initiation is generally considered after oral intake and volume status have recovered and no active ketoacidosis-promoting condition remains. For finerenone, re-initiation should be tied to potassium recovery, kidney-function stability, and the feasibility of early potassium reassessment.37 For RAS inhibitors, restart should be reconsidered once the precipitating hemodynamic, kidney-function, or potassium-related event has stabilized, rather than allowing temporary interruption to become unexamined permanent discontinuation.54 Failure to reassess restart after temporary interruption is an important source of exposure loss. In a nationwide cohort of patients with type 2 diabetes, SGLT2 inhibitor discontinuation was common and re-initiation after discontinuation was incomplete, illustrating that interruption and restart are practical implementation problems.55 Embedding restart prompts into discharge summaries, electronic prescribing systems, nephrology follow-up, and primary care communication is therefore a practical strategy that should be evaluated for reducing avoidable long-term loss of kidney-protective therapy. GLP-1 Receptor Agonists as Adjacent Therapy Outside the eGFR–Potassium Axis After FLOW, GLP-1 receptor agonists have become directly relevant to diabetic CKD, with semaglutide being the first agent in this class tested in a dedicated kidney-outcome trial in this population. In patients with type 2 diabetes and CKD, once-weekly semaglutide reduced the risk of major kidney disease events or death from cardiovascular causes by 24% compared with placebo.14 A prespecified FLOW analysis suggested that semaglutide benefits were broadly consistent irrespective of baseline SGLT2 inhibitor use, although the analysis was underpowered to exclude smaller interaction effects because baseline SGLT2 inhibitor use was limited.56 Earlier GLP-1 receptor agonist cardiovascular outcome trials, including LEADER and SUSTAIN-6, showed reductions in major adverse cardiovascular events;57,58 kidney signals in these earlier programs were mainly derived from secondary or exploratory renal outcomes, with albuminuria-based components contributing substantially to the observed benefit.59 Glucagon-like peptide-1 receptor agonist therapy should therefore be positioned adjacent to the eGFR–potassium stewardship axis developed here. Its practical constraints center on gastrointestinal tolerability, dose escalation, nutritional status, weight loss, gallbladder or biliary disease, injection burden, and adherence.60 It is highly relevant to residual metabolic, adiposity-related, kidney, and cardiovascular risk,14,61 but its sequencing requires a separate framework because implementation is not primarily governed by the early eGFR and potassium signals that structure SGLT2 inhibitor–finerenone therapy. Real-World Implementation Gaps The translation of trial evidence into routine care remains incomplete. Albuminuria testing is still underused in patients at risk for diabetic CKD, resulting in underdetection of albuminuric disease and missed opportunities to identify candidates for kidney-protective therapy.62 Even after albuminuric CKD is identified, implementation may be slowed by concerns about early eGFR decline, genital infections, ketoacidosis, hyperkalemia, polypharmacy, laboratory monitoring, and cost. These concerns are clinically relevant, but in eligible patients they more often justify structured monitoring, reassessment, and conditional initiation planning than open-ended deferral. Layering SGLT2 inhibitors and finerenone onto existing RAS inhibition can also increase medication burden and regimen complexity, particularly in patients with multimorbidity and polypharmacy. Although direct evidence that SGLT2 inhibitor–finerenone co-initiation itself reduces adherence is limited, treatment burden should be considered when selecting simultaneous vs sequential initiation. Within the proposed stewardship framework, pragmatic approaches to limiting treatment burden include shared decision-making, medication reconciliation, clear monitoring plans, and staged initiation when adherence or regimen complexity is a concern. The implementation gap also extends beyond diagnosis. In patients with type 2 diabetes and albuminuric CKD, SGLT2 inhibitor use remains substantially below the population that would be expected to benefit, and finerenone implementation is even less mature because it requires potassium-based eligibility assessment, early laboratory surveillance, and dose adjustment.63 These barriers identify where implementation pathways require clearer operational detail. Observational studies can clarify prescribing patterns, persistence, monitoring completion, and safety signals, but they should not be treated as evidence that routine-care exposure reproduces the long-term efficacy observed in randomized trials. Maintaining protocol-driven stewardship outside clinical trials requires simplification rather than replication of trial infrastructure. A pragmatic pathway can use a limited number of standardized checkpoints: baseline eligibility assessment, scheduled early laboratory reassessment, predefined thresholds for temporary withholding or dose adjustment, and an explicit restart plan after reversible events. These steps can be supported by standing laboratory orders, electronic reminders, pharmacist- or nurse-led follow-up, and shared documentation between primary care and nephrology. Monitoring intensity can then be scaled according to baseline eGFR, potassium risk, clinical stability, and local resource availability.37,55,63 In settings with limited specialist capacity, routine prescribing and scheduled laboratory surveillance can be coordinated through primary care, with nephrology input prioritized for patients with advanced or unstable CKD, recurrent hyperkalemia, acute kidney injury, or complex decisions regarding treatment interruption and re-initiation. Implementation research should evaluate sustained exposure, not initiation alone. Relevant endpoints include the proportion of eligible patients receiving therapy, time from albuminuria detection to treatment initiation, completion of early laboratory monitoring, discontinuation after an expected eGFR dip, permanent discontinuation after manageable hyperkalemia, re-initiation after temporary withholding, and persistence at 6 and 12 months. Without these exposure-quality metrics, prescription rates alone may overestimate the real-world delivery of sustained kidney-protective therapy. Future Research Priorities Future trials should test whether simultaneous or rapid sequential initiation improves kidney and cardiovascular outcomes beyond short-term albuminuria lowering. CONFIDENCE provides the most direct randomized rationale for this question because simultaneous finerenone–empagliflozin initiation produced greater UACR reduction at 180 days than either monotherapy; however, kidney failure, cardiovascular events, mortality, and long-term safety require trials powered and followed for clinical outcomes.18 Candidate trial designs should compare simultaneous initiation, SGLT2 inhibitor–first sequencing, and usual stepwise care, with prespecified stratification by baseline eGFR, serum potassium, albuminuria, heart failure status, and frailty. Implementation trials should complement efficacy trials, because routine-care delivery depends not only on eligibility assessment and initial prescribing but also on sustained exposure, monitoring completion, and re-initiation after interruption.55 Candidate interventions include electronic health record prompts, pharmacist-led medication review, nurse-led potassium surveillance, automated laboratory reminders, and post-discharge re-initiation protocols. Outcomes should extend beyond UACR and eGFR slope to include drug persistence, completion of early laboratory monitoring, re-initiation after temporary withholding, recurrent potassium events, adjudicated acute kidney injury, and patient-reported treatment burden. A further priority is preserving treatment persistence in hyperkalemia-prone patients. Patiromer has randomized evidence for lowering potassium, reducing recurrent hyperkalemia in CKD patients receiving RAAS inhibitors, and enabling more persistent spironolactone use in resistant hypertension with advanced CKD.47,48 Sodium zirconium cyclosilicate also has randomized potassium-lowering evidence.49,50 Whether binder-supported continuation of RAS inhibition or finerenone improves kidney or cardiovascular outcomes remains unproven, particularly in advanced albuminuric CKD where both potential absolute benefit and hyperkalemia risk are high. Finally, more data are needed in patients often underrepresented in protocolized trials, including those with very low eGFR, recurrent acute kidney injury, frailty, multimorbidity, cognitive impairment, and limited access to laboratory monitoring. These groups will shape the external validity of dynamic implementation under routine-care constraints. Conclusions Albuminuric type 2 diabetes–associated CKD is now treated in a multi-mechanism era in which RAS inhibition remains foundational and SGLT2 inhibitors and finerenone provide complementary kidney and cardiovascular protection. CONFIDENCE strengthens the rationale for combined therapy by showing additive short-term albuminuria lowering with simultaneous finerenone–empagliflozin initiation, but it does not establish hard-outcome superiority over sequential strategies. The implementation challenge is to preserve exposure while respecting drug-specific safety signals. A modest, early, clinically stable eGFR dip after SGLT2 inhibitor initiation usually reflects a functional hemodynamic response, whereas finerenone-associated hyperkalemia requires baseline risk assessment, early surveillance, reversible-trigger correction, temporary de-escalation, and structured re-initiation when appropriate. Acute illness, surgery, fasting, volume depletion, acute kidney injury, and clinically significant hyperkalemia may justify temporary withholding, but interruption should remain conditional. This narrative review is limited by the heterogeneity of the underlying evidence and the scarcity of direct comparative data on simultaneous vs sequential initiation, with much of the combination evidence still based on short-term albuminuria outcomes and indirect implementation data. Future progress will depend on trials that compare simultaneous and sequential initiation against clinical kidney and cardiovascular outcomes, and on implementation pathways that integrate initiation, monitoring, interruption, and re-initiation into one continuous therapeutic process.