
==== Front
JAMA Cardiol
JAMA Cardiol
JAMA Cardiology
2380-6583
2380-6591
American Medical Association

37952176
10.1001/jamacardio.2023.4664
hoi230066
Research
Research
Original Investigation
Online First
Comments
Decline in Estimated Glomerular Filtration Rate After Dapagliflozin in Heart Failure With Mildly Reduced or Preserved Ejection Fraction
A Prespecified Secondary Analysis of the DELIVER Randomized Clinical Trial
Dapagliflozin and eGFR Decline in HFmrEF/HFpEF
Dapagliflozin and eGFR Decline in HFmrEF/HFpEF
Mc Causland Finnian R. MBBCh MMSc 1 2
Claggett Brian L. PhD 2 3
Vaduganathan Muthiah MD MPH 2 3
Desai Akshay MD MPH 2 3
Jhund Pardeep MBBCh PhD 4
Vardeny Orly PharmD 5
Fang James C. MD 6
de Boer Rudolf A. MD 7
Docherty Kieran F. MBBCh 4
Hernandez Adrian F. MD 8
Inzucchi Silvio E. MD 9
Kosiborod Mikhail N. MD 10
Lam Carolyn S. P. MD 11
Martinez Felipe MD 12
Saraiva Jose F. Kerr MD PhD 13
McGrath Martina M. MBBCh 1 2
Shah Sanjiv J. MD 14
Verma Subodh MD 15
Langkilde Anna Maria MD PhD 16
Petersson Magnus MD PhD 16
McMurray John J. V. MD 4
Solomon Scott D. MD 2 3
1 Renal Division, Department of Medicine, Brigham and Women’s Hospital, Boston, Massachusetts
2 Harvard Medical School, Boston, Massachusetts
3 Cardiovascular Division, Department of Medicine, Brigham and Women’s Hospital, Boston, Massachusetts
4 British Heart Foundation Glasgow Cardiovascular Research Centre, School of Cardiovascular and Metabolic Health, University of Glasgow, Glasgow, Scotland
5 Minneapolis VA Center for Care Delivery and Outcomes Research, University of Minnesota, Minneapolis
6 University of Utah School of Medicine, Salt Lake City
7 Erasmus MC, Cardiovascular Institute, Thorax Center, Department of Cardiology, Rotterdam, the Netherlands
8 Duke University Medical Center, Durham, North Carolina
9 Yale School of Medicine, New Haven, Connecticut
10 Saint Luke’s Mid America Heart Institute, University of Missouri, Kansas City
11 National Heart Center Singapore and Duke–National University of Singapore, Singapore
12 National University of Cordoba, Cordoba, Argentina
13 Cardiovascular Division, Instituto de Pesquisa Clínica de Campinas, Campinas, Brazil
14 Northwestern University Feinberg School of Medicine, Chicago, Illinois
15 University of Toronto, Toronto, Ontario, Canada
16 Late-Stage Development, Cardiovascular, Renal and Metabolism, BioPharmaceuticals Research and Development, AstraZeneca, Gothenburg, Sweden
Article Information

Accepted for Publication: October 20, 2023.

Published Online: November 12, 2023. doi:10.1001/jamacardio.2023.4664

Corresponding Author: Finnian R. Mc Causland, MBBCh, MMSc, Renal Division, Department of Medicine, Brigham and Women’s Hospital, 75 Francis St, MRB-4, Boston, MA 02115 (fmccausland@bwh.harvard.edu).
Author Contributions: Dr Mc Causland had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: McCausland, Vaduganathan, Lam, Shah, Langkilde, Solomon.

Acquisition, analysis, or interpretation of data: McCausland, Claggett, Vaduganathan, Desai, Jhund, Vardeny, Fang, de Boer, Docherty, Hernandez, Inzucchi, Kosiborod, Lam, Martinez, Saraiva, McGrath, Verma, Langkilde, Petersson, McMurray.

Drafting of the manuscript: McCausland, Vaduganathan.

Critical review of the manuscript for important intellectual content: Claggett, Vaduganathan, Desai, Jhund, Vardeny, Fang, de Boer, Docherty, Hernandez, Inzucchi, Kosiborod, Lam, Martinez, Saraiva, McGrath, Shah, Verma, Langkilde, Petersson, McMurray, Solomon.

Statistical analysis: McCausland, Claggett, Jhund, Solomon.

Obtained funding: Petersson, Solomon.

Administrative, technical, or material support: Petersson.

Supervision: Claggett, Fang, Kosiborod, Martinez, Langkilde, Solomon.

Conflict of Interest Disclosures: Dr McCausland reported receiving grants from the National Institute of Diabetes and Digestive and Kidney Diseases, Lexicon, Novartis, Fifth Eye, and Satellite Healthcare and personal fees from Rubin Anders Scientific, GSK, and Zydus Therapeutics outside the submitted work. Dr Claggett reported receiving consulting fees from Alnylam, Cardurion, Corvia, Cytokinetics, Intellia, Rocket, and CVRX outside the submitted work. Dr Vaduganathan reported receiving personal fees from American Regent, Amgen, AstraZeneca, Bayer AG, Baxter Healthcare, Boehringer Ingelheim, Chiesi, Cytokinetics, Lexicon Pharmaceuticals, Merck, Novartis, Novo Nordisk, Pharmacosmos, Relypsa, Roche Diagnostics, Sanofi, and Tricog Health and grants from AstraZeneca, Galmed, Novartis, Bayer AG, Occlutech, and Impulse Dynamics outside the submitted work. Dr Desai reported receiving grants and consulting fees from AstraZeneca, Abbott, Alnylam, Bayer, and Novartis; consulting fees from Axon, Avidity BioPharma, GlaxoSmithKline, Merck, Medpace, Parexel, Regeneron, Roche, River2Renal, Veristat, Verily, and Zydus outside the submitted work. Dr Jhund reported being employed by and receiving research fees from AstraZeneca; speaker and advisory board fees from Novartis; grants from Boehringer Ingelheim, Roche Diagnostics, Analog Devices Inc; research fees from Bayer, Novo Nordisk; personal fees from ProAdwise, Sun Pharmaceuticals, and Intas Pharma outside the submitted work; and being director GCTP Ltd. Dr Vardeny reported receiving research support from AstraZeneca, Bayer, and Cardurion outside the submitted work. Dr Fang reported receiving steering committee fees from AstraZeneca, serving as deputy editor for the American Heart Association, grants from the National Institutes of Health, and data safety monitoring committee fees from Windtree and Amgen during the conduct of the study. Dr de Boer reported receiving grants from AstraZeneca, Abbott, Boehringer Ingelheim, Cardior Pharmaceuticals GmbH, Novo Nordisk, and Roche and speaker fees from Abbott, AstraZeneca, Bristol Myers Squibb, Cardior Pharmaceuticals GmbH, Novartis, and Roche outside the submitted work. Dr Docherty reported receiving nonfinancial support, grants, and personal fees from AstraZeneca; grants from Boehringer Ingelheim and Roche; personal fees from Pharmacosmos; and advisory board fees from Us2.ai and Bayer outside the submitted work. Dr Hernandez reported receiving grants from AstraZeneca, Boehringer Ingelheim, Novo Nordisk, and Lilly outside the submitted work. Dr Inzucchi reported receiving personal fees from Astra Zeneca, Boehringer Ingelheim, Merck, Pfizer, and Novo Nordisk outside the submitted work. Dr Kosiborod reported receiving personal fees from 35Pharma, Alnylam, Amgen, Applied Therapeutics, Bayer, Boehringer Ingelheim, Cytokinetics, Dexcom, Eli Lilly, Esperion Therapeutics, Imbria Pharmaceuticals, Janssen, Lexicon Pharmaceuticals, Merck, Novo Nordisk, Pharmacosmos, Pfizer, Sanofi, scPharmaceuticals, Structure Therapeutics, Vifor Pharma, and Youngene Therapeutics; and grants from AstraZeneca, Boehringer Ingelheim, and Pfizer; and stock options from Artera Health and Saghmos Therapeutics outside the submitted work. Dr Lam reported receiving research support from Novo Nordisk and Roche Diagnostics; consultant or committee fees from Alleviant Medical, Allysta Pharma, AnaCardio AB, Applied Therapeutics, AstraZeneca, Bayer, Biopeutics, Boehringer Ingelheim, Boston Scientific, Bristol Myers Squibb, CardioRenal, Cytokinetics, Darma Inc, EchoNous Inc, Eli Lilly, Impulse Dynamics, Intellia Therapeutics, Ionis Pharmaceutical, Janssen, Medscape/WebMD Global LLC, Merck, Novartis, Novo Nordisk, Prosciento Inc, Quidel Corporation, Radcliffe Group Ltd, Recardio Inc, ReCor Medical, Roche Diagnostics, Sanofi, Siemens Healthcare Diagnostics, and Us2.ai; being cofounder/nonexecutive director of Us2.ai; having a patent pending for a method for diagnosis and prognosis of chronic heart failure; and having patents for an automated clinical workflow that recognizes and analyzes 2-dimensional and Doppler echo images for cardiac measurements and the diagnosis, prediction, and prognosis of heart disease. Dr Martinez reported receiving personal fees from AstraZeneca during the conduct of the study. Dr Saraiva reported receiving advisory board fees from AstraZeneca, Novartis, Novo Nordisk, Lilly, Merck Sharp Dhome, and Boehringer Ingelheim; grants from AstraZeneca, Novartis, Novo Nordisk, Lilly, Merck Sharp Dhome, and Boehringer Ingelheim outside the submitted work. Dr Shah reported receiving consulting/committee fees and research grants from AstraZeneca during the conduct of the study. Dr Verma reported receiving personal fees from AstraZeneca, Amarin, Amgen, AstraZeneca, Bayer, Boehringer Ingelheim, Canadian Medical and Surgical Knowledge Translation Research Group, Eli Lilly, HLS Therapeutics, Janssen, Novartis, Novo Nordisk, Pfizer, PhaseBio, S & L Solutions Event Management Inc, and Sanofi and grants from Amgen, AstraZeneca, Boehringer Ingelheim, Eli Lilly, Novartis, Novo Nordisk, and PhaseBio outside the submitted work. Dr Langkilde reported being a full-time employee and shareholder of AstraZeneca during the conduct of the study. Dr Petersson reported being a full-time employee and shareholder of AstraZeneca outside the submitted work. Dr McMurray reported receiving institutional research support from AstraZeneca, Bayer, Cardurion, and Novartis; steering committee and travel fees paid to institution from Amgen, Cytokinetics, GSK, and Novartis; advisory fees paid to institution from KBP Biosciences; data safety monitoring board fees from George Clinical PTY Ltd; personal fees from Abbott, Alkem Metabolics, AstraZeneca, Blue Ocean Scientific Solutions Ltd, Boehringer Ingelheim, Canadian Medical and Surgical Knowledge, Emcure Pharmaceuticals Ltd, Eris Lifesciences, European Academy of CME, Hikma Pharmaceuticals, Imagica Health, Intas Pharmaceuticals, J.B. Chemicals & Pharmaceuticals Ltd, Lupin Pharmaceuticals, Medscape/Heart.org, ProAdWise Communications, Radcliffe Cardiology, Sun Pharmaceuticals, The Corpus, Translation Research Group, and Translational Medicine Academy; lecture fees from Alynylam Pharmaceuticals, Bayer, BMS, Ionis Pharmaceuticals, Novartis, Regeneron Pharmaceuticals, and River 2 Renal Corp; consultancy fees from Global Clinical Trial Partners Ltd; and being a director outside the submitted work. Dr Solomon reported receiving grants from Actelion, Alnylam, Amgen, AstraZeneca, Bellerophon, Bayer, BMS, Celladon, Cytokinetics, Eidos, Gilead, GSK, Ionis, Lilly, Mesoblast, MyoKardia, NIH/NHLBI, Neurotronik, Novartis, Novo Nordisk, Respicardia, Sanofi Pasteur, Theracos, and Us2.ai and consulting fees from Abbott, Action, Akros, Alnylam, Amgen, Arena, AstraZeneca, Bayer, Boeringer Ingelheim, BMS, Cardior, Cardurion, Corvia, Cytokinetics, Daiichi Sankyo, GSK, Lilly, Merck, Myokardia, Novartis, Roche, Theracos, Quantum Genomics, Cardurion, Janssen, Cardiac Dimensions, Tenaya, Sanofi Pasteur, Dinaqor, Tremeau, CellProThera, Moderna, American Regent, Sarepta, Lexicon, Anacardio, Akros, and Valo outside the submitted work. No other disclosures were reported.

Meeting Presentation: This paper was presented at the American Heart Association Scientific Sessions 2023; November 12, 2023; Philadelphia, Pennsylvania.

Data Sharing Statement: See Supplement 2.

12 11 2023
2 2024
12 11 2024
9 2 144152
14 9 2023
20 10 2023
Copyright 2023 American Medical Association. All Rights Reserved.
jamacardiol-e234664.pdf

This prespecified analysis of the Dapagliflozin Evaluation to Improve the Lives of Patients With Preserved Ejection Fraction Heart Failure (DELIVER) randomized clinical trial investigates the association of initial changes in estimated glomerular filtration rate with cardiovascular and kidney outcomes in patients with heart failure and mildly reduced or preserved ejection fraction.

Key Points

Question

What is the association of initial changes in estimated glomerular filtration rate (eGFR) with cardiovascular and kidney outcomes among patients with heart failure and mildly reduced ejection fraction (HFmrEF) or preserved ejection fraction (HFpEF)?

Findings

Among 5788 participants of the Dapagliflozin Evaluation to Improve the Lives of Patients With Preserved Ejection Fraction Heart Failure (DELIVER) trial, an initial eGFR decline of greater than 10% (vs ≤10%) was frequent and associated with a higher risk of the primary cardiovascular outcome among those randomized to placebo but not among those randomized to dapagliflozin. An initial eGFR decline greater than 10% was not associated with adverse kidney outcomes among dapagliflozin-treated patients.

Meaning

These data suggest, among patients with HFmrEF or HFpEF, that sodium-glucose cotransporter-2 inhibitors should not be discontinued in response to an initial eGFR decline.

Importance

An initial decline in estimated glomerular filtration rate (eGFR) is expected after initiating a sodium-glucose cotransporter-2 inhibitor (SGLT2i) and has been observed across patients with diabetes, chronic kidney disease, and heart failure.

Objective

To examine the implications of initial changes in eGFR among patients with heart failure with mildly reduced ejection fraction (HFmrEF) or preserved ejection fraction (HFpEF) enrolled in the Dapagliflozin Evaluation to Improve the Lives of Patients With Preserved Ejection Fraction Heart Failure (DELIVER) trial.

Design, Setting, and Participants

This was a prespecified analysis of the results of the DELIVER randomized clinical trial, which was an international multicenter study of patients with EF greater than 40% and eGFR greater than or equal to 25. The DELIVER trial took place from August 2018 to March 2022. Data for the current prespecified study were analyzed from February to October 2023.

Intervention

Dapagliflozin, 10 mg per day, or placebo.

Main Outcomes and Measures

In this prespecified analysis, the frequency of an initial eGFR decline (baseline to month 1) was compared between dapagliflozin and placebo. Cox models adjusted for baseline eGFR and established prognostic factors were fit to estimate the association of an initial eGFR decline with cardiovascular (cardiovascular death or heart failure event) and kidney (≥50% eGFR decline, eGFR<15 or dialysis, death from kidney causes) outcomes, landmarked at month 1, stratified by diabetes.

Results

Study data from 5788 participants (mean [SD] age, 72 [10] years; 3253 male [56%]) were analyzed. The median (IQR) change in eGFR level from baseline to month 1 was −1 (−6 to 5) with placebo and −4 (−9 to 1) with dapagliflozin (difference, −3; P < .001). A higher proportion of patients assigned to dapagliflozin developed an initial eGFR decline greater than 10% vs placebo (1144 of 2892 [40%] vs 737 of 2896 [25%]; odds ratio, 1.9; 95% CI, 1.7-2.1; P difference <.001). An initial eGFR decline of greater than 10% (vs ≤10%) was associated with a higher risk of the primary cardiovascular outcome among those randomized to placebo (adjusted hazard ratio [aHR], 1.33; 95% CI, 1.10-1.62) but not among those randomized to dapagliflozin (aHR, 0.90; 95% CI, 0.74-1.09; P for interaction = .01). Similar associations were observed when alternative thresholds of initial eGFR decline were considered and when analyzed as a continuous measure. An initial eGFR decline of greater than 10% was not associated with adverse subsequent kidney composite outcomes in dapagliflozin-treated patients (aHR, 0.94; 95% CI, 0.49-1.82).

Conclusions and Relevance

Among patients with HFmrEF or HFpEF treated with dapagliflozin, an initial eGFR decline was frequent but not associated with subsequent risk of cardiovascular or kidney events. These data reinforce clinical guidance that SGLT2is should not be interrupted or discontinued in response to an initial eGFR decline.

Trial Registration

ClinicalTrials.gov Identifier: NCT03619213
==== Body
pmcIntroduction

Over the past several years, the sodium-glucose cotransporter-2 inhibitors (SGLT2is) have significantly altered the clinical treatment of patients with type 2 diabetes (T2D),1,2 chronic kidney disease (CKD),1,3,4,5 and heart failure,6,7,8,9,10 with large outcomes trials demonstrating major risk reductions in cardiovascular and kidney outcomes and, in some circumstances, death.

The major pharmacologic action of SGLT2is involves the inhibition of sodium and glucose reabsorption in proximal tubule cells of the kidney, leading to glycosuria, natriuresis, and reduction of intraglomerular pressure.11,12 As such, and similar to that observed with renin-angiotensin system inhibitors, the initiation of SGLT2is is often accompanied by an acute initial decline in estimated glomerular filtration rate (eGFR) that may be observed clinically. Although this has been shown to be reversible upon SGLT2i withdrawal in some studies,13,14 concerns remain that such acute functional changes in eGFR may lead to inappropriate drug cessation, thus preventing patients from benefiting from long-term risk reduction.15

Among those randomly assigned to SGLT2i treatment, post hoc analyses of patients with CKD have observed similar slowing of long-term kidney function decline in those with or without an initial eGFR decline,16,17,18 and analyses from the Empagliflozin Cardiovascular Outcome Event Trial in Type 2 Diabetes Mellitus Patients (EMPA-REG Outcome) program reported that the risk reduction in cardiovascular outcomes with empagliflozin was not modified by the initial eGFR decline.16 In the Dapagliflozin and Prevention of Adverse Outcomes in Heart Failure (DAPA-HF) trial of patients with heart failure and reduced ejection fraction, the development of an eGFR decline greater than 10% was associated with a lower risk of cardiovascular outcomes among those assigned to dapagliflozin, compared with those with a similar decline assigned to placebo.19

In this prespecified analysis of the Dapagliflozin Evaluation to Improve the Lives of Patients With Preserved Ejection Fraction Heart Failure (DELIVER) trial, we evaluated the magnitude and frequency of an initial decline in eGFR, its predictors, and its association with clinical outcomes.

Methods

Data Sharing, Trial Design, and Oversight

Data underlying the findings described in this article may be obtained following AstraZeneca’s data sharing policy.20 The design, statistical analysis plan, and primary outcomes of the DELIVER trial have been published.21,22 Local ethics committees approved the trial, and all patients provided written informed consent. The trial was reviewed by an independent data and safety monitoring committee. This study followed the Consolidated Standards of Reporting Trials (CONSORT) reporting guidelines.

Study Patients

The DELIVER trial enrolled adults 40 years or older with symptomatic heart failure and left ventricular ejection fraction greater than 40%, elevated N-terminal pro–brain natriuretic peptide (NT-proBNP) concentrations, and structural heart disease (left ventricular hypertrophy or left atrial enlargement within 12 months). Patients were randomly assigned in a 1:1 fashion to dapagliflozin, 10 mg once daily, or a matching placebo. Notable exclusion criteria included an eGFR level less than 25 at visit 1 (enrollment) and systolic BP lower than 95 mm Hg on 2 consecutive readings at visit 1 or 2 (randomization). Race was captured on a dedicated demographics case report form and included the following categories: Asian, Black or African American, White, or other race designation (including American Indian or Alaska Native and Native Hawaiian or Other Pacific Islander).

Definitions of Changes in eGFR and Serum Creatinine Level

For this prespecified secondary analysis of initial changes in biomarkers of kidney function, only patients with baseline and 1-month measurements of serum creatinine were included (eTable 1 in Supplement 1). The changes in eGFR level between baseline and month 1, as calculated by the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation (2009),23 were categorized as follows: less than 0% (no decline or increase), 0% to 10% decline, or greater than 10% decline. In sensitivity analyses, additional changes in eGFR level (>25% decline) and serum creatinine (increase ≥0.3 mg/dL; increase ≥0.5 mg/dL; to convert to micromoles per liter, multiply by 88.4) were considered.

Cardiovascular and Kidney Outcomes

The primary outcome was the composite of cardiovascular death or a worsening heart failure event (hospitalization or urgent visit). For clinical kidney outcomes, owing to the infrequent occurrence of sustained declines in eGFR level in this trial,24 we evaluated a post hoc kidney composite outcome consisting of either the first occurrence of (1) 50% or greater decline in eGFR level relative to the month 1 value (as the initial decline in eGFR level was calculated from baseline to month 1); (2) development of end-stage kidney disease (from adverse event [AE] reporting or decline in eGFR level to <15); or (3) death due to kidney causes. Prespecified exploratory outcomes included the effect of dapagliflozin on change in eGFR (calculated by the CKD-EPI equation [2009]23 using serum creatinine measurements from visits at baseline and months 1, 4, 12, 24, and 36).

Safety Outcomes

The DELIVER trial collected data on serious AEs, AEs that led to discontinuation of dapagliflozin or placebo, and select other AEs (including kidney-related AEs).9 The frequency of safety events was compared according to categories of eGFR decline (≤10% vs >10%) among patients who received at least 1 dose of randomized treatment. The occurrence of AEs was examined separately in those assigned to placebo and those assigned to dapagliflozin, and according to timing in relation to the calculation of the eGFR decline at month 1 (≤1 month and >1 month).

Statistical Analysis

Data were reported as mean (SD) when normally distributed, as median (IQR) when nonnormally distributed, and as frequencies and percentages for categorical variables. Differences in baseline characteristics were assessed by tests for trend across categories of eGFR decline (<0%, 0%-10%, >10%) using linear regression, the Cuzick nonparametric trend test, and the χ2 test for trend for continuous normal, continuous nonnormal, and categorical data, respectively.

The odds of developing an initial decline in eGFR (>10% vs ≤10% and 25% vs ≤25%) or an increase in serum creatinine (≥0.3 vs <0.3 mg/dL and ≥0.5 vs <0.5 mg/dL) for dapagliflozin vs placebo were initially assessed using unadjusted logistic regression. Subsequently, to examine for potential effect modification by other baseline variables, logistic regression models that adjusted for baseline factors and their corresponding treatment interaction terms were fit (eGFR level, age, sex, race, body mass index [BMI], history of hypertension, history of diabetes, left ventricular ejection fraction, NT-proBNP level, systolic blood pressure, use of a loop diuretic, mineralocorticoid receptor antagonist, angiotensin-converting enzyme inhibitor [ACEi]/angiotensin receptor blocker [ARB], or angiotensin receptor–neprilysin inhibitor). Those with P value interaction terms of <.10 (eGFR level, use of a mineralocorticoid receptor antagonist, or use of an ACEi/ARB) were selected for inclusion in a multivariable-adjusted model.

For the time-to-event outcomes, incidence rates and 95% CIs were calculated per 100 patient-years of follow-up. Unadjusted and adjusted Cox proportional hazards models, landmarked at month 1, were fit to assess the association of eGFR decline greater than 10% (vs ≤10%) with cardiovascular and kidney outcomes. The multivariable model adjusted for the following baseline variables: age, sex, race, eGFR level, BMI), hypertension, left ventricular ejection fraction, log-transformed NT-proBNP level, systolic blood pressure, mineralocorticoid receptor antagonist, ACEi/ARB, and change in systolic blood pressure from baseline to month 1. All Cox models were stratified by the presence or absence of diabetes at baseline. Interaction terms were included to assess if the association of eGFR decline (>10% vs ≤10% from baseline to month 1) with cardiovascular and kidney composite outcomes differed according to randomized treatment assignment. In subgroup analyses, models were fit separately among those assigned to dapagliflozin and those assigned to placebo. The association of eGFR decline as a continuous variable with incidence rates of clinical outcomes was estimated using Poisson regression, with potential nonlinearity accommodated by using restricted cubic splines with 3 knots.

Multivariable mixed-effects models were used to assess for changes in eGFR over time (chronic slope, from month 4 through month 36) according to the presence or absence of an eGFR decline greater than 10% between baseline and month 1. Month 4 was chosen as the starting point for the examination of the chronic slope due to the observed pattern of initial change from baseline to month 1, followed by a rebound from month 1 to month 4. Models included terms for eGFR decline greater than 10%, trial visit, interaction between eGFR decline and trial visit, treatment assignment, the interaction between treatment assignment and visit, age, sex, race, eGFR level, BMI, diabetes, hypertension, left ventricular ejection fraction, log-transformed NT-proBNP level, systolic blood pressure, mineralocorticoid receptor antagonist use, ACEi/ARB use, and change in systolic blood pressure from baseline to month 1. Intercepts and slopes over time were allowed to vary randomly between patients via the inclusion of patient and time as random effects. In subgroup analyses, models were fit separately among those assigned to dapagliflozin and those assigned to placebo.

All analyses were performed at an α level of .05, without correction for multiple hypothesis testing, using Stata/MP, version 16.0 (StataCorp). All P values were 2-sided, and P < .05 was considered statistically significant. Data for the current study were analyzed from February to October 2023.

Results

Patient Characteristics

Overall, 5788 of 6263 patients (92%; mean [SD] age, 72 [10] years; 3255 male [56%]; 2535 female [44%]) had paired measurements of serum creatinine and eGFR level at both baseline and month 1, allowing the initial change in eGFR level to be calculated. Comparisons of patients where an initial eGFR decline was possible vs not possible to calculate are provided in eTable 1 in Supplement 1. Patients self-identified with the following race categories: 171 American Indian or Alaska Native (3%), 1225 Asian (21%), 142 Black or African American (2%), 4068 White (70%), or 182 other race (3%).

During the first month of treatment, 2408 patients (42%) did not have any decline in eGFR level, of which 977 (41%) were assigned to dapagliflozin, and 1431 (59%) were assigned to placebo. A total of 1499 patients (26%) had 0% to 10% decline, of which 771 (51%) were assigned to dapagliflozin, and 728 (49%) were assigned to placebo. A total of 1881 patients (32%) had a decline of greater than 10%, of which 1144 (61%) were assigned to dapagliflozin, and 737 (39%) were assigned to placebo. At baseline, those experiencing higher initial declines vs those experiencing an increase in eGFR were more likely to be older (mean [SD] age, 72 [9] years vs 71 [10] years), have higher systolic blood pressure (mean [SD], 129 [16] mm Hg vs 128 [15] mm Hg), BMI level (mean [SD], 30.2 [6.3] vs 29.6 [6.0]; calculated as weight in kilograms divided by height in meters squared), ejection fraction (mean [SD], 55% [9%] vs 54% [9%]), and eGFR level (mean [SD], 61 [18] vs 58 [19]). They were more likely to have a history of hypertension and diabetes, to be taking an ACEi or ARB, and to be assigned to dapagliflozin (Table 1). Comparisons of baseline characteristics among those who developed an initial decline in eGFR level greater than 10% vs less than or equal to 10% are presented by randomized treatment arm in eTable 2 in Supplement 1.

Table 1. Baseline Characteristics According to Categories of Estimated Glomerular Filtration Rate (eGFR) Decline From Baseline to Month 1

Characteristica	eGFR decline	P value for trend	
<0% (n = 2408)	0%-10% (n = 1499)	>10% (n = 1881)	
Age, mean (SD), y	71 (10)	71 (10)	72 (9)	.04	
Sex, No. (%)					
Female	1072 (45)	603 (40)	860 (46)	.56	
Male	1336 (55)	896 (60)	1021 (54)	
Race, No. (%)					
American Indian or Alaska Native	72 (3)	50 (3)	49 (3)	.26	
Asian	533 (22)	340 (23)	352 (19)	
Black or African American	51 (2)	41 (3)	50 (3)	
White	1685 (70)	1023 (68)	1360 (72)	
Otherb	67 (3)	45 (3)	70 (4)	
Geographic region, No. (%)					
Europe and Saudi Arabia	1174 (49)	693 (46)	907 (48)	.09	
Asia	512 (21)	332 (22)	340 (18)	
Latin America	431 (18)	286 (19)	353 (19)	
North America	291 (12)	188 (13)	281 (15)	
Systolic blood pressure, mean (SD), mm Hg	128 (15)	128 (16)	129 (16)	.02	
Heart rate, mean (SD), beats/min	71 (12)	71 (11)	72 (11)	.53	
BMI, mean (SD)c	29.6 (6.0)	29.7 (6.0)	30.2 (6.3)	.003	
Serum creatinine, mean (SD), mg/dL	1.2 (0.4)	1.1 (0.3)	1.1 (0.3)	<.001	
eGFR, mean (SD)d	58 (19)	66 (20)	61 (18)	<.001	
Coronary artery disease, No. (%)	746 (31)	435 (29)	587 (31)	.95	
Left ventricular ejection fraction, mean (SD), %	54 (9)	54 (9)	55 (9)	.03	
NT-proBNP, median (IQR), pg/mL	1007 (627-1741)	972 (599-1619)	1022 (637-1800)	.54	
NYHA classification, No. (%)					
I	0	0	1 (0)	.47	
II	1822 (76)	1157 (77)	1402 (75)	
III	580 (24)	340 (23)	470 (25)	
IV	6 (0)	2 (0)	8 (0)	
Hypertension, No. (%)	2110 (88)	1316 (88)	1704 (91)	.003	
Diabetes, No. (%)	1028 (43)	637 (42)	933 (50)	<.001	
Atrial fibrillation or flutter, No. (%)	1385 (58)	826 (55)	1076 (57)	.76	
Stroke, No. (%)	233 (10)	130 (9)	185 (10)	.92	
Hospitalization for heart failure, No. (%)	988 (41)	580 (39)	772 (41)	.93	
Myocardial infarction, No. (%)	648 (27)	383 (26)	487 (26)	.43	
Loop diuretic at randomization, No. (%)	1870 (78)	1089 (73)	1478 (79)	.66	
ACE inhibitor or ARB at randomization, No. (%)	1719 (71)	1114 (74)	1393 (74)	.04	
Mineralocorticoid-receptor antagonist at randomization, No. (%)	1034 (43)	631 (42)	825 (44)	.58	
β-Blocker at randomization, No. (%)	1985 (82)	1220 (81)	1578 (84)	.25	
ARNI, No. (%)	113 (5)	81 (5)	83 (4)	.73	
Dapagliflozin, No. (%)	977 (41)	771 (51)	1144 (61)	<.001	
Abbreviations: ACE, angiotensin-converting enzyme; ARB, angiotensin receptor blocker; ARNI, angiotensin-neprilysin inhibitor; BMI, body mass index; NT-proBNP, N-terminal pro–brain natriuretic peptide; NYHA, New York Heart Association.

SI conversion factor: To convert serum creatinine to micromoles per liter, multiply by 88.4.

a Plus-minus values are mean (SD).

b Other race included American or Alaska Native and Native Hawaiian or Other Pacific Islander.

c The BMI is the weight in kilograms divided by the square of the height in meters.

d The eGFR at baseline was estimated according to the Chronic Kidney Disease Epidemiology Collaboration (2009) formula.

Initial eGFR Decline From Baseline to Month 1

The median (IQR) change in eGFR level from baseline to month 1 was −4 (−9 to 1) in those assigned to dapagliflozin and −1 (−6 to 5) in those assigned to placebo, with a placebo-corrected difference of −3 (P < .001). The median (IQR) percentage change in eGFR level from baseline to month 1 was −6% (−16% to 3%) in those assigned to dapagliflozin and −1% (−10% to 8%) in those assigned to placebo, with a placebo-corrected difference of −5% (P < .001).

A higher proportion of patients assigned to dapagliflozin experienced a decline in eGFR level of greater than 10% vs those assigned to placebo (1144 of 2892 [40%] vs 737 of 2896 [25%]; odds ratio, 1.9; 95% CI, 1.7-2.1; P difference <.001). Similar patterns were noted in analyses that examined alternative thresholds of changes in eGFR level or serum creatinine from baseline to month 1 (Table 2).

Table 2. Frequency of Threshold-Based Changes in Estimated Glomerular Filtration Rate (eGFR) and Serum Creatinine From Baseline to Month 1

Definition of dip event	Mean (SD) eGFR change	Mean (SD) serum creatinine change, mg/dL	No. of dip events in placebo arm (%)	No. of dip events in dapagliflozin arm (%)	Dip events in dapagliflozin vs placebo, odds ratio (95% CI)	
eGFR dip >10%	−12 (6)	0.3 (0.2)	737 (25)	1144 (40)	1.9 (1.7-2.1)a
1.9 (1.7-2.1)b	
eGFR dip >25%	−20 (8)	0.5 (0.4)	140 (5)	262 (9)	2.0 (1.6-2.4)a
2.0 (1.6-2.4)b	
Delta serum creatinine ≥0.3 mg/dL	−17 (9)	0.5 (0.4)	167 (6)	293 (10)	1.8 (1.5-2.2)a
1.9 (1.6-2.3)b	
Delta serum creatinine ≥0.5 mg/dL	−20 (10)	0.8 (0.5)	57 (2)	101 (3)	1.9 (1.3-2.6)a
1.9 (1.3-2.6)b	
SI conversion factor: To convert serum creatinine to micromoles per liter, multiply by 88.4.

a Unadjusted.

b Adjusted for baseline eGFR, mineralocorticoid use, angiotensin-converting enzyme inhibitor/angiotensin receptor blocker use.

Dapagliflozin and Initial eGFR Decline Greater Than 10% According to Baseline Characteristics

A series of logistic regression models were fit to examine if the treatment effect on the initial decline in eGFR differed according to baseline characteristics (adjusted for the randomized treatment, the variable of interest, and a corresponding interaction term). Although dapagliflozin was consistently associated with higher risk of an initial eGFR decline across the baseline characteristics considered, the risk of an eGFR decline greater than 10% for dapagliflozin (vs placebo) appeared to be higher for those with baseline eGFR level less than or equal to 60, those not taking an ACEi or ARB, and those not taking a mineralocorticoid receptor antagonist at baseline (eFigure 1 in Supplement 1). The effect estimates for dapagliflozin vs placebo for the development of an eGFR decline greater than 10% from models that adjusted for these variables were similar to those obtained from unadjusted models (Table 2).

Initial eGFR Decline and Risk of Cardiovascular Outcomes

In landmark analyses (beginning at-risk time at 1 month), among those assigned to placebo, the primary cardiovascular outcome occurred in 156 of 720 patients (22%) who experienced an eGFR decline greater than 10% and in 357 of 2140 patients (17%) who did not (adjusted hazard ratio [aHR], 1.33; 95% CI, 1.10-1.62). Among those assigned to dapagliflozin, the primary cardiovascular outcome occurred in 175 of 1135 patients (15%) who experienced an eGFR decline greater than 10% and in 274 of 1739 patients (16%) who did not (aHR, 0.90; 95% CI, 0.74-1.09; P for interaction = .01). The relevant Kaplan-Meier curves examining the risk of the primary cardiovascular outcome according to the presence or absence of an initial eGFR decline greater than 10% and assignment to dapagliflozin vs placebo are shown in Figure 1. Similar patterns of association were noted using other definitions of eGFR decline or increase in serum creatinine (Table 3). The incidence rate for the primary cardiovascular outcome according to the initial decline in continuous eGFR level from baseline to month 1 (both absolute and percentage) according to treatment arms are displayed graphically in Figure 2.

Figure 1. Risk of Primary Cardiovascular Outcome

Kaplan-Meier curve showing the risk of the primary cardiovascular outcome according to the presence or absence of an initial decline in estimated glomerular filtration rate greater than 10% vs not and assignment to dapagliflozin vs placebo.

Table 3. Event Rates and Association of Thresholds of Change in Estimated Glomerular Filtration Rate (eGFR) or Serum Creatinine With Cardiovascular and Kidney Outcomesa

Definition of dip	Cardiovascular composite outcome	P value for interaction	
Placebo
No. of events/No. of patients (%)
event rate (95% CI) per 100 patient years	Dapagliflozin
No. of events/No. of patients (%)
event rate (95% CI) per 100 patient-years	Adjusted HR (95% CI)	
No dip	Dip	No dip	Dip	Placebo	Dapagliflozin	
eGFR dip >10%	357/2140 (17)
8.3 (7.4-9.2)	156/720 (22)
11.2 (9.5-13.0)	274/1739 (16)
7.8 (6.9-8.8)	175/1135 (15)
7.5 (6.4-8.6)	1.33 (1.10-1.62)	0.90 (0.74-1.09)	.01	
eGFR dip >25%	472/2727 (17)
8.6 (7.9-9.4)	41/133 (31)
17.9 (13.2-24.3)	396/2615 (15)
7.4 (6.7-8.2)	53/259 (20)
10.3 (7.8-13.4)	1.68 (1.21-2.33)	1.22 (0.91-1.64)	.12	
Delta serum creatinine ≥0.3 mg/dL	459/2701 (17)
8.4 (7.7-9.2)	54/159 (34)
19.6 (15.0-25.5)	387/2585 (15)
7.3 (6.6-8.1)	62/289 (21)
11.0 (8.6-14.1)	1.70 (1.27-2.28)	1.19 (0.90-1.58)	.07	
Delta serum creatinine ≥0.5 mg/dL	499/2806 (18)
8.9 (8.1-9.7)	14/54 (26)
14.9 (8.8-25.1)	429/2775 (15)
7.5 (6.9-8.3)	20/99 (20)
11.0 (7.1-17.0)	1.14 (0.67-1.95)	1.16 (0.73-1.83)	.97	
Post hoc kidney composite outcome b	
eGFR dip >10%	38/2150 (1.8)
0.8 (0.6-1.1)	17/734 (2.3)
1.1 (0.7-1.8)	23/1744 (1.3)
0.6 (0.4-0.9)	15/1143 (1.3)
6 (0.4-1.0)	1.62 (0.90-2.89)	0.94 (0.49-1.82)	.35	
eGFR dip >25%	49/2747 (1.8)
0.8 (0.6-1.1)	6/137 (4.4)
2.3 (1.0-5.1)	36/2626 (1.4)
0.6 (0.5-0.9)	2/261 (0.8)
0.4 (0.1-1.5)	3.54 (1.45-8.61)	0.45 (0.11-1.90)	.05	
Delta serum creatinine ≥0.3 mg/dL	44/2720 (1.6)
0.8 (0.6-1.0)	11/164 (6.7)
3.2 (1.9-6.3)	34/2595 (1.3)
0.6 (0.4-0.9)	4/292 (1.4)
6.1 (0.2-1.8)	4.60 (2.24-9.41)	0.67 (0.23-1.94)	.02	
Delta serum creatinine ≥0.5 mg/dL	50/2830 (1.8)
0.8 (0.6-1.1)	5/54 (9.3)
5.0 (2.1-12.0)	36/2787 (1.3)
0.6 (0.4-0.8)	2/100 (2.0)
1.0 (0.3-4.0)	5.93 (2.22-15.83)	0.97 (0.22-4.20)	.11	
Abbreviation: HR, hazard ratio.

SI conversion factor: To convert serum creatinine to micromoles per liter, multiply by 88.4.

a All models were landmarked for at-risk time to start at the month 1 visit.

b For the renal composite, the threshold of 50% decline in eGFR was calculated relative to the month 1 measurement.

Figure 2. Incidence Rates of the Primary Outcome

Adjusted incidence rates of the primary outcome occurring after 1 month (landmark analysis) according to change in estimated glomerular filtration rate (eGFR) from baseline to month 1 within randomized treatment groups. A, Absolute change in eGFR. B, Percentage change in eGFR. The shaded regions represent 95% CIs.

Initial eGFR Decline and Risk of Kidney Outcomes

Beginning at-risk time at 1 month, among those assigned to placebo, the post hoc kidney outcome occurred in 17 of 734 patients (2.3%) who experienced an eGFR decline greater than 10% and in 38 of 2150 patients (1.8%) who did not (aHR, 1.62; 95% CI, 0.90-2.89). Among those assigned to dapagliflozin, the post hoc kidney outcome occurred in 15 of 1143 patients (1.3%) who experienced an eGFR decline greater than 10% and in 23 of 1744 patients (1.3%) who did not (aHR, 0.94; 95% CI, 0.49-1.82; P for interaction = .35). Similar patterns of association were noted with other thresholds of changes in eGFR and serum creatinine level, although these analyses were limited by small numbers of events and wide CIs (Table 3).

Initial eGFR Decline and Longer-Term eGFR Slope

Among patients who developed an initial eGFR decline greater than 10%, the average adjusted change (IQR) in kidney function from month 4 was −1.2 (−1.6 to −0.8) per year; the corresponding value in those who did not develop an initial eGFR decline was −0.9 (−1.4 to −0.4) per year, with a difference of 0.3 (−0.2 to 0.8) per year. Although the overall rate of eGFR decline was faster among those assigned to placebo (compared with those assigned to dapagliflozin), the randomized treatment did not appear to modify the association of an initial eGFR decline vs no decline with longer-term eGFR slope (P for interaction = .68) (eFigure 2 in Supplement 1).

Initial eGFR Decline and AEs

The overall frequency of AEs was higher among those who experienced an initial eGFR decline greater than 10%, compared with those without this decline, in both early (≤1 month) and later (>1 month) periods. These findings were generally similar according to the randomized treatment assignment. The only exception was any serious AE from baseline to month 1, where the frequency of events among those with an eGFR decline greater than 10% (vs ≤10%) was higher in the placebo arm, as opposed to the dapagliflozin arm (eTable 3 in Supplement 1).

Discussion

Among patients with heart failure and mildly reduced or preserved ejection fraction enrolled in the DELIVER trial, the development of an initial decline in eGFR greater than 10% was more common among patients randomized to dapagliflozin than placebo. However, among those assigned to dapagliflozin, the presence of an initial eGFR decline was not associated with a higher risk of the primary cardiovascular outcome, nor with a difference in the longer-term rate of eGFR decline, compared with patients who did not develop an initial decline. Although AEs were overall more common among those with vs those without an initial eGFR decline, most were similar between dapagliflozin and placebo.

Initially used as a therapy to lower blood glucose in patients with diabetes, early experience with SGLT2is noted a propensity for these medications to cause modest acute elevations in serum creatinine level. Confirmation of these patterns (and reciprocal changes in eGFR level) were obtained in subsequent large clinical outcomes trials. More importantly, these trials also observed that the eGFR decline was reversible after drug discontinuation,13,14 and despite the initial decline, these medications slowed the longer-term decline in kidney function, compared with placebo.4,25 The acute decline in eGFR is felt to be a result of a reduction in intraglomerular pressure, via mechanism(s) that are still to be fully elucidated.11,12 Notably, some smaller mechanistic reports have shown that the acute decline in eGFR level with dapagliflozin occurs without evidence of kidney damage, as assessed by urine kidney injury biomarkers.26,27

However, despite these findings, real-world practice concerns remain regarding the underutilization of SGLT2i therapy across broad populations of patients where there are proven clinical indications.28 With respect to an initial decline in eGFR level, several studies have now examined the association of this phenomenon with clinical outcomes. For example, post hoc analyses of the Canagliflozin and Renal Events in Diabetes with Established Nephropathy Clinical Evaluation (CREDENCE) study (T2D and CKD)17 and the EMPA-REG Outcome study (T2D and established cardiovascular disease)16 reported similar rates of longer-term eGFR decline in patients assigned to active therapy with vs without an initial decline in eGFR, whereas a post hoc analysis of the Evaluation of Ertugliflozin Efficacy and Safety Cardiovascular Outcomes Trial (VERTIS CV; T2D and atherosclerotic cardiovascular disease) reported that patients in the largest (vs lowest) tertile of initial eGFR decline with ertugliflozin actually experienced the slowest rate of decline in longer-term eGFR slope.18 Real-world observational data from the Veterans Affairs database is consistent with these findings. In adjusted analyses, SGLT2i use was associated with a lower risk of cardiovascular and kidney outcomes in those with a higher-than-average probability of eGFR dip greater than 10% or greater than30%. Using inverse probability-weighted models, continued use of SGLT2is vs discontinuation at 6 months was associated with a lower risk of cardiovascular and kidney outcomes, across the categories of initial eGFR decline.15

In analyses from DAPA-HF, an initial decline in eGFR level of greater than 10% between baseline and 2 weeks occurred in 38% of patients assigned to dapagliflozin and 21% assigned to placebo. The association of a dip of greater than 10% with the primary cardiovascular outcome differed by treatment arms (HR, 1.45; 95% CI, 1.19-1.78 among those assigned to placebo and HR, 0.73; 95% CI, 0.59-0.91 among those assigned to dapagliflozin; P for interaction <.001).19 In the present analyses, we expand the knowledge base to those with heart failure and mildly reduced or preserved ejection fraction. We reported similar evidence for effect modification of the association of an initial eGFR decline with the primary cardiovascular outcome, according to randomized treatment arm, with an initial decline only being associated with a higher cardiovascular risk among those assigned to placebo. With respect to the kidney composite outcome, we observed the absence of a higher risk for those assigned to dapagliflozin who experienced an initial eGFR decline. This was largely in contrast to those assigned to placebo, where an initial decline was generally associated with a higher risk of the kidney outcome. Further, when chronic eGFR slope was examined (after month 4), the rate of eGFR decline appeared similar among those who did and did not develop an initial decline in eGFR in the dapagliflozin arm.

Strengths and Limitations

The strengths of our study include the large number of patients with the availability of repeated creatinine measurements, inclusion of patients to a lower eGFR threshold of 25, and detailed follow-up for adjudicated outcomes. However, limitations include some loss of patients without repeated kidney function measurements, generalizing results beyond a predominantly Asian and White population, potential residual confounding due to the consideration of a postrandomization exposure, and the relatively infrequent occurrence of post hoc defined kidney composite outcomes.

Conclusions

In conclusion, results of this prespecified analysis of the DELIVER trial showed that although an initial decline in eGFR level on initiation of dapagliflozin among participants of DELIVER was frequent, it was not associated with adverse cardiovascular outcomes and did not appear to alter the long-term trajectory of eGFR decline, compared with those without an initial eGFR decline. This is in contrast to the adverse cardiovascular and kidney prognostic implications of an initial eGFR decline with placebo. These data will be informative to clinicians and patients and, with appropriate monitoring, should provide reassurance for continuation of these medications in the majority of patients.

Supplement 1. eFigure 1. Odds Ratio for Decline in eGFR >10% for Dapagliflozin vs Placebo in Subgroups

eFigure 2. Kaplan-Meier Curve Showing the Risk of the Primary Cardiovascular Outcome According to the Presence or Absence of an Initial Decline in eGFR >10% vs Not and Assignment to Dapagliflozin vs Placebo

eTable 1. Baseline Characteristics Among Those With and Without Available Data to Calculate the Difference on eGFR From Baseline to Month 1

eTable 2. Baseline Characteristics of Patients With an Initial eGFR Decline >10% From Baseline to Month 1, According to Randomized Treatment Arms

eTable 3. Adverse Events According to eGFR Decline >10% vs ≤10% and Relative to Month 1 Visit (Safety Population)

Click here for additional data file.

Supplement 2. Data Sharing Statement.

Click here for additional data file.
==== Refs
References

1 Wanner C, Inzucchi SE, Lachin JM, ; EMPA-REG OUTCOME Investigators. Empagliflozin and progression of kidney disease in type 2 diabetes. N Engl J Med. 2016;375 (4 ):323-334. doi:10.1056/NEJMoa1515920 27299675
2 Neal B, Perkovic V, Mahaffey KW, ; CANVAS Program Collaborative Group. Canagliflozin and cardiovascular and renal events in type 2 diabetes. N Engl J Med. 2017;377 (7 ):644-657. doi:10.1056/NEJMoa1611925 28605608
3 Bhatt DL, Szarek M, Pitt B, ; SCORED Investigators. Sotagliflozin in patients with diabetes and chronic kidney disease. N Engl J Med. 2021;384 (2 ):129-139. doi:10.1056/NEJMoa2030186 33200891
4 Heerspink HJL, Stefánsson BV, Correa-Rotter R, ; DAPA-CKD Trial Committees and Investigators. Dapagliflozin in patients with chronic kidney disease. N Engl J Med. 2020;383 (15 ):1436-1446. doi:10.1056/NEJMoa2024816 32970396
5 Perkovic V, Jardine MJ, Neal B, ; CREDENCE Trial Investigators. Canagliflozin and renal outcomes in type 2 diabetes and nephropathy. N Engl J Med. 2019;380 (24 ):2295-2306. doi:10.1056/NEJMoa1811744 30990260
6 Packer M, Anker SD, Butler J, ; EMPEROR-Reduced Trial Investigators. Cardiovascular and renal outcomes with empagliflozin in heart failure. N Engl J Med. 2020;383 (15 ):1413-1424. doi:10.1056/NEJMoa2022190 32865377
7 Anker SD, Butler J, Filippatos G, ; EMPEROR-Preserved Trial Investigators. Empagliflozin in heart failure with a preserved ejection fraction. N Engl J Med. 2021;385 (16 ):1451-1461. doi:10.1056/NEJMoa2107038 34449189
8 McMurray JJV, Solomon SD, Inzucchi SE, ; DAPA-HF Trial Committees and Investigators. Dapagliflozin in patients with heart failure and reduced ejection fraction. N Engl J Med. 2019;381 (21 ):1995-2008. doi:10.1056/NEJMoa1911303 31535829
9 Solomon SD, McMurray JJV, Claggett B, ; DELIVER Trial Committees and Investigators. Dapagliflozin in heart failure with mildly reduced or preserved ejection fraction. N Engl J Med. 2022;387 (12 ):1089-1098. doi:10.1056/NEJMoa2206286 36027570
10 Bhatt DL, Szarek M, Steg PG, ; SOLOIST-WHF Trial Investigators. Sotagliflozin in patients with diabetes and recent worsening heart failure. N Engl J Med. 2021;384 (2 ):117-128. doi:10.1056/NEJMoa2030183 33200892
11 Kidokoro K, Cherney DZI, Bozovic A, . Evaluation of glomerular hemodynamic function by empagliflozin in diabetic mice using in vivo imaging. Circulation. 2019;140 (4 ):303-315. doi:10.1161/CIRCULATIONAHA.118.037418 30773020
12 van Bommel EJM, Muskiet MHA, van Baar MJB, . The renal hemodynamic effects of the SGLT2 inhibitor dapagliflozin are caused by post-glomerular vasodilatation rather than preglomerular vasoconstriction in metformin-treated patients with type 2 diabetes in the randomized, double-blind RED trial. Kidney Int. 2020;97 (1 ):202-212. doi:10.1016/j.kint.2019.09.013 31791665
13 Wanner C, Heerspink HJL, Zinman B, ; EMPA-REG OUTCOME Investigators. Empagliflozin and kidney function decline in patients with type 2 diabetes: a slope analysis from the EMPA-REG OUTCOME trial. J Am Soc Nephrol. 2018;29 (11 ):2755-2769. doi:10.1681/ASN.2018010103 30314978
14 Cherney DZI, Ferrannini E, Umpierrez GE, . Efficacy and safety of sotagliflozin in patients with type 2 diabetes and severe renal impairment. Diabetes Obes Metab. 2021;23 (12 ):2632-2642. doi:10.1111/dom.14513 34338408
15 Xie Y, Bowe B, Gibson AK, McGill JB, Maddukuri G, Al-Aly Z. Clinical implications of estimated glomerular filtration rate dip following sodium-glucose cotransporter-2 inhibitor initiation on cardiovascular and kidney outcomes. J Am Heart Assoc. 2021;10 (11 ):e020237. doi:10.1161/JAHA.120.020237 34013739
16 Kraus BJ, Weir MR, Bakris GL, . Characterization and implications of the initial estimated glomerular filtration rate dip upon sodium-glucose cotransporter-2 inhibition with empagliflozin in the EMPA-REG OUTCOME trial. Kidney Int. 2021;99 (3 ):750-762. doi:10.1016/j.kint.2020.10.031 33181154
17 Oshima M, Jardine MJ, Agarwal R, . Insights from CREDENCE trial indicate an acute drop in estimated glomerular filtration rate during treatment with canagliflozin with implications for clinical practice. Kidney Int. 2021;99 (4 ):999-1009. doi:10.1016/j.kint.2020.10.042 33316282
18 Cherney DZI, Cosentino F, Dagogo-Jack S, ; VERTIS CV Investigators. Initial eGFR changes with ertugliflozin and associations with clinical parameters: analyses from the VERTIS CV Trial. Am J Nephrol. 2022;53 (7 ):516-525. doi:10.1159/000524889 35691283
19 Adamson C, Docherty KF, Heerspink HJL, . Initial decline (dip) in estimated glomerular filtration rate after initiation of dapagliflozin in patients with heart failure and reduced ejection fraction: insights from DAPA-HF. Circulation. 2022;146 (6 ):438-449. doi:10.1161/CIRCULATIONAHA.121.058910 35442064
20 AstraZeneca. Our transparency commitments. Accessed October 1, 2023. https://astrazenecagrouptrials.pharmacm.com/ST/Submission/Disclosure
21 Solomon SD, Vaduganathan M, Claggett BL, . Baseline characteristics of patients with HF with mildly reduced and preserved ejection fraction: DELIVER trial. JACC Heart Fail. 2022;10 (3 ):184-197. doi:10.1016/j.jchf.2021.11.006 35241246
22 Solomon SD, McMurray JJV, Claggett B, . Dapagliflozin in heart failure with mildly reduced or preserved ejection fraction. N Engl J Med. 2022;387 (12 ):1089-1098. doi:10.1056/NEJMoa2206286 36027570
23 Levey AS, Stevens LA, Schmid CH, ; CKD-EPI (Chronic Kidney Disease Epidemiology Collaboration). A new equation to estimate glomerular filtration rate. Ann Intern Med. 2009;150 (9 ):604-612. doi:10.7326/0003-4819-150-9-200905050-00006 19414839
24 Mc Causland FR, Claggett BL, Vaduganathan M, . Dapagliflozin and kidney outcomes in patients with heart failure with mildly reduced or preserved ejection fraction: a prespecified analysis of the DELIVER randomized clinical trial. JAMA Cardiol. 2023;8 (1 ):56-65. doi:10.1001/jamacardio.2022.4210 36326604
25 Herrington WG, Staplin N, Wanner C, ; The EMPA-KIDNEY Collaborative Group. Empagliflozin in patients with chronic kidney disease. N Engl J Med. 2023;388 (2 ):117-127.36331190
26 Dekkers CCJ, Petrykiv S, Laverman GD, Cherney DZ, Gansevoort RT, Heerspink HJL. Effects of the SGLT-2 inhibitor dapagliflozin on glomerular and tubular injury markers. Diabetes Obes Metab. 2018;20 (8 ):1988-1993. doi:10.1111/dom.13301 29573529
27 Sen T, Li J, Neuen BL, . Effects of the SGLT2 inhibitor canagliflozin on plasma biomarkers TNFR-1, TNFR-2 and KIM-1 in the CANVAS trial. Diabetologia. 2021;64 (10 ):2147-2158. doi:10.1007/s00125-021-05512-5 34415356
28 Schernthaner G, Shehadeh N, Ametov AS, . Worldwide inertia to the use of cardiorenal protective glucose-lowering drugs (SGLT2i and GLP-1 RA) in high-risk patients with type 2 diabetes. Cardiovasc Diabetol. 2020;19 (1 ):185. doi:10.1186/s12933-020-01154-w 33097060
