Abstract
Aims
We aim to assess the efficacy of congestion relief and safety associated with adding SGLT2i (viz., dapagliflozin 10 mg) to intravenous loop diuretics within 24 h of hospital presentation in patients with acute heart failure (AHF).
Methods and results
A single‐centre open‐label clinical research study enrolled 98 patients admitted with an episode of AHF who were randomized into two groups: (a) receiving SGLT2i once daily in addition to structured intravenous furosemide therapy; (b) receiving structured intravenous furosemide therapy alone. In‐hospital congestion relief was evaluated by body weight change, EVEREST score, lung ultrasound B‐lines, inferior vena cava ultrasound measurement, NT‐proBNP and CD146. Safety was assessed by changes in renal function and serum electrolyte abnormalities. Secondary endpoints included diuresis and natriuresis, hospital care indices and echocardiographic changes in cardiac function at 1‐month. ANCOVA analysis was performed to adjust for imbalance between the two groups regarding chronic kidney disease status and baseline values. The analysis followed an intention‐to‐treat approach. The mean age ± standard deviation in the SGLT2i and control group was 63.63 ± 10.95 years and 65.31 ± 10.82 years, respectively, with 40/49 and 42/49 males. No death occurred in hospital; 1/49 and 2/49 deaths at 30 days were recorded. The adjusted mean change ± standard error (SE) in body weight was −4.90 ± 0.93 kg versus −4.28 ± 0.81 kg in the SGLT2i and control group, respectively. The adjusted mean change ± SE in B‐lines at discharge and at 1 month was −19.93 ± 0.87 versus −18.64 ± 0.79 (P = 0.227) and −19.65 ± 1.54 versus −14.82 ± 1.43 (P = 0.012), respectively. The proportion of worsening renal function was 15/49 and 6/47 (P = 0.048) in the respective treatment groups (SGLT2i and control). The adjusted mean ± SE of 24‐h urinary Na was 248.03 ± 23.69 mmol/day versus 173.83 ± 20.76 mmol/day (P = 0.009). One‐month changes in ultrasound parameters were significantly improved in the SGLT2i group, with median (inter‐quartile range) values of left ventricular ejection fraction and end‐diastolic volume equal to 5% (0.35% to 11.5%) versus 0 (−1% to +5%) and −6.5 mL (−27.5 to +3) versus 4 mL (−11.5 to +10), respectively.
Conclusions
Early initiation of SGLT2i administration in addition to intravenous loop diuretics in patients with AHF would optimize congestion relief and improve clinical outcomes.
Keywords: Acute heart failure, Congestion management, Left ventricular ejection fraction, Lung B‐lines, SGLT2i, Sodium‐glucose cotransporter 2 inhibitor
Introduction
Congestion is the primary cause of hospitalization and readmission in heart failure patients, strongly correlating with adverse outcomes. 1 , 2 , 3 Thus, achieving effective decongestion and preventing residual congestion through early administration of intravenous loop diuretics is considered first‐line therapy for acute heart failure (AHF). 1 , 2 , 3 The secondary goal during an AHF episode is the early implementation of optimal guideline‐directed medical therapy (GDMT) during hospitalization. 1 , 2 , 3 While loop diuretics substantially improve symptoms, they have failed to enhance short‐ and long‐term outcomes in AHF. 4 Furthermore, the diuretic response may be influenced by several factors such as the extent of volume overload and renal function, thus complicating efforts to achieve euvolemia and implement optimal medical therapy. 4
Sodium‐glucose cotransporter 2 inhibitors (SGLT2i) have recently become a cornerstone in treating chronic heart failure; however, their mechanisms of action are not entirely understood. One proposed mechanism concerns their direct effect on the kidneys, promoting natriuresis and water excretion. 5 This has led to studies hypothesizing the potential use of SGLT2i in AHF patients with hypervolaemia, aiming to enhance diuresis for decongestion. Two prospective studies of SGLT2i in AHF reported increased urinary output upon adding SGLT2i to protocol‐based intravenous diuretic regimens but failed to demonstrate quicker or more efficient decongestion. 6 , 7 On the other hand, the EMPULSE trial showed that initiation of empagliflozin in AHF resulted in an early and efficient improvement in congestion with clinical benefit at 90 days. 8 Despite the clear recommendation of AHF guidelines for early introduction of SGLT2i during hospitalization to optimize GDMT and improve prognosis, concerns remain regarding the risks of acute kidney injury and ketoacidosis during the acute decompensation phase. 1 , 2 , 3 This uncertainty, combined with insufficient data about the efficacy of SGLT2i in alleviating congestion in AHF, requires further investigation.
We conducted randomized controlled clinical research aimed to assess the efficacy of congestion relief and safety associated with adding SGLT2i (viz., dapagliflozin) to intravenous loop diuretics within 24 h of hospital presentation in AHF patients.
Methods
Study design
A single‐centre open‐label clinical research study was conducted in a regional tertiary care hospital in western Romania from 1 October 2021 to 30 September 2024. The study was performed according to the CONSORT guidelines 9 and was retrospectively registered with International Standard Randomized Controlled Trial Number (ISRCTN): ISRCTN17064177. The study also complied with the Declaration of Helsinki and was approved by the Hospital Ethics Commission for Clinical Trials (Arad County Clinical Emergency Hospital, number 41/21.08.2021), with all patients providing written informed consent.
Trial patients
Adult patients admitted for an episode of AHF, regardless of ejection fraction or type II diabetes status, were eligible. Inclusion criteria comprised NT‐proBNP ≥ 400 pg/mL, eGFR > 25 mL/min/1.73 m2 and clinical and paraclinical signs of congestion. Exclusion criteria encompassed: infections, type I diabetes, active cancer, severe aortic stenosis, exacerbated chronic obstructive pulmonary disease (COPD), pulmonary embolism and acute coronary syndrome.
Randomization, treatment and study procedures
Patients were randomly assigned to two groups: (a) to receive dapagliflozin 10 mg once daily in addition to a structured intravenous furosemide therapy within 24 h from admission; (b) to receive structured intravenous furosemide therapy alone. Figure 1 shows the study flow diagram according to the CONSORT guidelines. 9 The dapagliflozin treatment was given throughout the hospitalization and was continued as a long‐term therapy. Two patients discontinued dapagliflozin after 3 days from admission for experiencing hypotension (systolic blood pressure < 90 mmHg), and one patient discontinued after 2 weeks due to a 1 mg/dL increase in creatinine (40% from baseline). The analysis followed an intention‐to‐treat (ITT) approach.
Figure 1.
CONSORT flow diagram.
At baseline, each patient was assessed for the following parameters: natriuretic peptide and CD146 serum concentration, standing weight, EVEREST congestion score, lung ultrasound measurement of B‐lines and inferior vena cava diameter. The EVEREST composite congestion score (with a range from 0 to 18) is based on the assessment of simple clinical parameters including dyspnoea, orthopnoea, jugular vein distention, rales, oedema and fatigue; it was introduced by the investigators in the EVEREST trial. 10 Lung ultrasound was performed on a Siemens Acuson SC2000 ultrasound machine using a phased array transducer and an eight‐zone protocol (four zones on each hemidiaphragm) to assess lung congestion (the sum of B‐lines across all zones). For the inferior vena cava (IVC), the maximum diameter was measured. We did not calculate the IVC collapsibility index, but the minimal diameter during inspiration was subtracted from the maximal IVC diameter during expiration; based on these data, collapsibility was assessed as ‘appropriate’ or ‘inappropriate’. This binomial variable was not included in the final analysis, for it did not add statistical value to the other numerical variables concerning the congestion.
Trans‐thoracic echocardiography was conducted for each patient upon admission and at a 1‐month follow‐up. Three patients died before the 1‐month planned echocardiography. Structural and functional parameters assessed included: left ventricular end‐diastolic volume (EDV), systolic pulmonary artery pressure (SPAP), global longitudinal strain (GLS), right ventricular strain and ejection fraction (by 3D echocardiography). Two independent investigators analysed the echocardiographic data.
Serum cluster of differentiation 146 (CD146) was determined for each patient at admission, discharge and 1‐month post‐discharge. For CD146 analysis, blood samples were collected into serum‐separating tubes, centrifuged and stored at −80°C for subsequent analysis. Serum CD146 levels were measured using an optimized enzyme‐linked immunosorbent assay kit (Human MCAM/CD146 ELISA kit PicoKine ‐ Boster Biological Technology, Pleasanton CA, USA #EK1675) and analysed on a Secan Sunrise microplate reader (Texan Austria GmbH, Untersbergstr. 1A, A‐5082 Grodig, Austria). Calibration and standardization of the assay were conducted following the manufacturer's protocol.
Structured intravenous furosemide therapy was utilized in both study arms. For loop diuretic‐naive patients, dosages of 20–40 mg furosemide intravenously were administered upon admission, with a similar dose given at 12 h, targeting a 24‐h urine output of 3–4 L (and doubling the loop diuretic dose if congestion persisted). For patients already on oral diuretic therapy, the initial dose doubled their home 24‐h oral dose and was subsequently administered every 12 h (with dose doubling until the maximum loop diuretic dose was reached if congestion persisted and urine output remained <3–4 L after 24 h). The protocol did not include specifications for adjustment of furosemide dosage based on renal function, but such adjustments were considered and recommended by the treating physician based on the clinical judgement and 24‐h diuresis. A 24‐h urine collection was performed 24 h post‐admission to measure natriuresis and urine output induced by intravenous furosemide alone or in combination with SGLT2i.
At discharge, patients underwent evaluations for natriuretic peptide and CD146 serum concentrations, weight change, congestion EVEREST score, and ultrasound assessments of lung B‐lines and inferior vena cava diameter. One‐month post‐discharge, patients were re‐evaluated for any adverse events and underwent echocardiography to measure changes in ejection fraction, EDV and SPAP. Additionally, a blood sample was taken to measure plasma CD146 levels at 1 month.
Primary efficacy and safety endpoints
Congestion relief was evaluated by comparing changes in congestion markers from baseline (at hospital admission) to discharge: weight change, EVEREST score, lung ultrasound B‐lines, inferior vena cava echo measurement, NT‐proBNP and CD146.
Safety endpoints comprised changes in renal function, defined as worsening renal function and serum electrolyte abnormalities. Worsening renal function (WRF) was determined as any increase in creatinine of ≥25% from baseline level, or absolute value ≥0.3 mg/dL. Serum electrolyte abnormalities included hypokalaemia (<3.5 mmol/L), hyperkalaemia (>5.5 mmol/L) and hyponatremia (<130 mmol/L).
Study participants were monitored daily from admission until discharge and reassessed 1 month after discharge for all primary endpoints. Any adverse events and 72‐h values of body weight, B‐lines, creatinine and eGFR were also included in this analysis.
Secondary endpoints
Secondary endpoints encompassed: (1) diuresis and natriuresis (mean diuresis over 3 days, 24‐h urinary sodium output and total furosemide dose per episode of hospitalization); (2) hospitalization duration, 30‐day heart failure readmissions and mortality; (3) 1‐month echocardiographic changes in heart failure function (ejection fraction, EDV and SPAP in both groups; GLS and RV strain in the SGLT2i group).
Sample size and data analysis
A 1:1 ratio was established for the two study groups. The required sample size was determined based on primary endpoints, considering three outcome types: (a) numerical continuous variables (e.g., changes in weight, echo B‐lines, IVC diameter and NT‐proBNP); (b) numerical scores (e.g., EVEREST scores); (c) proportions of safety‐related outcomes (e.g., hyponatremia or hypokalaemia). The R package ‘webPower’ v. 0.9.0, applying J. Cohen's theory, 11 was utilized for the calculations involving proportions and numerical continuous variables. Based on existing literature, 12 , 13 a medium effect size was assumed with d = 0.6 and h = 0.6 for differences in numerical outcomes and observed proportions, respectively; two‐sided tests were employed with alpha = 0.05 and power = 0.8. The resultant sample size indicated a need for 45 subjects in each group. A 10% dropout adjustment was applied, 14 leading to initial randomization of 50 patients in each group. For numerical scores, the R package ‘MKpower’ v. 0.7 (using Monte Carlo simulation for empirical power calculations) confirmed that for a 1‐point difference with a 1.5‐point standard deviation in two samples of 50 subjects each, the resultant power was >0.9. The actual dropout involved one patient from each group (as illustrated in the study flowchart), with no overall decrease in required statistical power.
Descriptive statistics included observed frequency counts with corresponding percentages for categorical variables, and mean ± standard deviation for numerical variables, irrespective of their distribution. Normality was assessed using the Kolmogorov–Smirnov test. When the distribution of values exhibited significant asymmetry, the median with inter‐quartile range (i.e., Q1 and Q3) were additionally provided as descriptive statistics. The chi‐square statistical test (either asymptotic or using Fisher's exact test) was applied to assess statistical significance in the observed proportions' differences among categorical variables. Non‐parametric tests, specifically the Mann–Whitney U test, were employed for comparing numerical value distributions in independent groups. Non‐parametric Wilcoxon signed rank test was applied for paired samples of values.
The baseline assessment revealed significant imbalance between the two study groups in regard with some characteristics, such as the lung B‐lines and CKD proportion. Therefore, analysis of covariance (ANCOVA) and mixed‐effects analysis of variance (ANOVA) were applied to adjust for baseline CKD status and baseline values, with no decrease in the initially planned statistical power. 15 , 16 , 17 For marginal means' estimation, the expected mean squares were based on type III sums of squares. The analysis followed an approach of planned comparisons according to an a priori protocol, so the least significant difference (LSD) technique was used in multiple comparisons, with no additional adjustments. Numerical values had slightly non‐normal distribution; ANOVA and ANCOVA are robust to slight violation of the normality assumptions; Q‐Q plots and residuals' plots were used to confirm the validity of results. Multicollinearity was tested based on the variance inflation factor (VIF).
The statistical analysis was conducted at a confidence level of 95% and a significance level of 5%. All reported probability values were two‐tailed. Statistical analyses were performed using IBM SPSS v. 20 and R v. 4.4.3 packages.
Results
Patients' characteristics
Between 1 October 2021 and 30 September 2024, 149 patients with AHF were screened, and 100 were randomized into either the dapagliflozin plus intravenous therapy group, or the intravenous therapy alone group. Two patients dropped out (one from each study arm; they could not be reached for re‐evaluation at 1 month), resulting in a total of 98 patients available for analysis. The mean age of the cohort was 64 years, with a mean ejection fraction of 27% and mean NT‐proBNP of 11 573 pg/mL. Overall baseline characteristics were balanced between treatment arms, except for a higher prevalence of chronic kidney disease (CKD) in the control group (26% vs. 10.2%) and greater lung congestion (as assessed by lung B‐lines ultrasound measurement) in the SGLT2i group (35.08 ± 14.34 vs. 25.29 ± 11.23, P < 0.0001) (values and results of statistical tests in Table 1 ).
Table 1.
Patients' characteristics
| Baseline characteristics | SGLT2i group (N = 49) | Control group (N = 49) | P‐value a , b , c |
|---|---|---|---|
| Age (years) a | 63.63 ± 10.95 | 65.31 ± 10.82 | 0.488 |
| Sex, M b | 40 (81.6%) | 42 (85.7%) | 0.585 |
| BMI | |||
| Females a | 27.39 ± 5.08 | 23.73 ± 2.67 | 0.174 |
| Males a | 28.49 ± 5.65 | 29.12 ± 9.03 | 0.864 |
| Medical history | |||
| Atrial fibrillation b | 16 (32.7%) | 20 (40.8%) | 0.402 |
| Type II DM b | 10 (20.4%) | 15 (30.6%) | 0.247 |
| CKD b | 5 (10.2%) | 13 (26.5%) | 0.037* |
| IHD b | 14 (28.6%) | 10 (20.4%) | 0.347 |
| Clinical parameters | |||
| SBP a | 138.67 ± 27.07 | 140.06 ± 22.93 | 0.713 |
| HR a | 100.9 ± 25.37 | 102.41 ± 26.07 | 0.768 |
| Echocardiography markers | |||
| EF (%) a | 26.03 ± 8.81 | 28.13 ± 9.75 | 0.427 |
| EDV a | 209 ± 63.39 | 189.65 ± 56.8 | 0.072 |
| GLS a | 6.22 ± 2.57 | 7.27 ± 3.35 | 0.078 |
| TAPSE a | 16.43 ± 4.24 | 15.24 ± 2.38 | 0.252 |
| SPAP a | 47.8 ± 11.1 | 47.61 ± 12.57 | 0.778 |
| Congestion markers | |||
| EVEREST score a | 10.08 ± 3.23 | 10.9 ± 2.96 | 0.383 |
| B‐lines a | 35.08 ± 14.34 | 25.29 ± 11.23 | <0.001** |
| IVC a | 2.14 ± 0.519 | 2.18 ± 0.37 | 0.525 |
| NT proBNP (pg/mL) c |
10 363.92 ± 7932.22 9316 (4436 to 13 037) |
12 784.55 ± 10 131.97 9851 (3544 to 20 804) |
0.426 |
| CD146 (pg/mL) c |
1335.1 ± 608.1 1287 (782 to 1854) |
1446.73 ± 991.59 1230.5 (749 to 2176) |
0.895 |
| Laboratory investigations | |||
| Creatinine (mg/dL) a | 1.2 ± 0.37 | 1.13 ± 0.29 | 0.539 |
| eGFR (mL/min/1.73 m2) a | 70.35 ± 20.52 | 73.98 ± 20.05 | 0.486 |
| Troponin c |
110.17 ± 297.38 32.7 (21.5 to 52) |
174.9 ± 474.05 43.2 (21 to 74.5) |
0.579 |
| Ferritin (μg/L) a | 145.54 ± 114.28 | 178.6 ± 139.51 | 0.192 |
| HbA1c a | 6.52 ± 1.47 | 6.02 ± 0.8 | 0.138 |
| Glycaemia a | 144.92 ± 69.23 | 119.33 ± 39.57 | 0.02* |
| CRP a | 12.8 ± 20.52 | 14.89 ± 14.18 | 0.071 |
| Severe mitral regurgitation b | 9 (18.4%) | 15 (30.6%) | 0.159 |
| Severe tricuspid regurgitation b | 8 (16.3%) | 14 (28.6%) | 0.146 |
Abbreviations: BMI, body mass index; CD146, cluster of differentiation 146; CKD, chronic kidney disease; CRP, C‐reactive protein; DM, diabetes mellitus; EDV, left‐ventricular end‐diastolic volume; EF, ejection fraction; eGFR, estimated glomerular filtration rate; GLS, global longitudinal strain; HR, heart rate; IHD, ischemic heart disease; IVC, inferior vena cava diameter; NT proBNP, N‐terminal pro–B‐type natriuretic peptide; SBP, systolic blood pressure; SD, standard deviation; SPAP, systolic pulmonary artery pressure; TAPSE, tricuspid annular plane systolic excursion.
Mean ± SD; non‐parametric Mann–Whitney test.
Counts (percentage); asymptotic chi‐square test or Fisher's exact test.
Mean ± SD; median (inter‐quartile range with Tukey's hinges); non‐parametric Mann–Whitney test.
P < 0.05.
P < 0.01.
Efficacy and safety associated with SGLT2i added to structured intravenous furosemide therapy
The mean weight loss during hospitalization was not significantly different in patients treated with dapagliflozin compared with those receiving only standard intravenous diuretic therapy (concrete values in Table 2 ). The combination of dapagliflozin and diuretic intravenous therapy did not result in significant changes in the EVEREST score, NT‐proBNP, serum CD146 or inferior vena cava diameter compared with standard diuretic therapy. However, a substantial reduction in lung ultrasound B‐lines was observed at discharge and at 1‐month assessment in the dapagliflozin group (actual values in Table 2 ).
Table 2.
Primary efficacy and safety endpoints
| SGLT2i group (N = 49) | Control group (N = 49) | P‐value a , c , d , e | |
|---|---|---|---|
| Weight change compared with baseline (hospital admission) | |||
| Baseline weight (kg) a | 86.55 ± 20.35 | 85.68 ± 25.33 | 0.363 |
| Discharge weight (kg) a | 80.82 ± 18.58 | 80.96 ± 25.16 | 0.522 |
| Discharge change in weight (kg) a | −5.73 ± 5.88 | −4.72 ± 4.91 | 0.108 |
| Adjusted mean change in weight at discharge (kg) e |
−4.9 ± 0.93 (−6.74 to −3.06) |
−4.28 ± 0.81 (−5.89 to −2.66) |
0.622 |
| P‐value b | <0.001** | <0.001** | |
| Congestion relief compared with baseline (hospital admission) | |||
| Discharge B‐lines values a | 11.29 ± 5.1 | 11.33 ± 4.96 | 0.83 |
| Discharge change in B‐lines a | −23.80 ± 14.03 | −13.96 ± 10.55 | <0.001** |
| Adjusted mean change in B‐lines at discharge e |
−19.93 ± 0.87 (−21.66 to −18.20) |
−18.64 ± 0.79 (−20.21 to −17.10) |
0.227 |
| 1‐month B‐lines values a | 11.39 ± 6.8 | 14.49 ± 10.31 | 0.058 # |
| 1‐month change in B‐lines a | −23.80 ± 14.04 | −13.96 ± 10.55 [N = 47] | <0.001** |
| Adjusted mean change in 1‐month B‐lines e |
−19.65 ± 1.54 (−22.71 to −16.6) |
−14.82 ± 1.43 [N = 47] (−17.67 to −11.97) |
0.012* |
| Discharge change in EVEREST score a | −7.55 ± 3.30 | −8.39 ± 3.30 | 0.329 |
| Discharge change in IVC a | −0.652 ± 0.464 | −0.695 ± 0.487 | 0.669 |
| Discharge change in NTproBNP c |
−6116.42 ± 5850.75 −4427 (−8441 to −2039) |
−8107.53 ± 6853.25 5911 (−10 972 to −2285) |
0.127 |
| Discharge change in CD146 c |
−199.31 ± 527.72 −186.4 (−526 to −30) |
−221.67 ± 206.52 −200.8 (−333 to −41) |
0.702 |
| Treatment safety: change from baseline | |||
|
Discharge change in eGFR (mL/min/1.73 m2) c |
−3.88 ± 17.59 −6 (−17 to +10) |
−4.23 ± 14.6 [N = 47] −4 (−13.5 to +2) |
0.82 |
| Discharge change in serum creatinine (mg/dL) a | 0.13 ± 0.27 | 0.05 ± 0.18 [N = 47] | 0.032* |
| Adjusted mean change in serum creatinine at discharge (mg/dL) e |
0.154 ± 0.043 (0.068 to 0.239) |
0.06 ± 0.038 [N = 47] (−0.02 to 0.13) |
0.048* |
| In‐hospital WRF d , f | 15 (30.6%) | 6 (12.8%) [N = 47] | 0.048* |
| Discharge change in serum Na c |
−0.02 ± 3.18 0 (−2 to +2) |
−24.89 ± 173.89 0 (−2 to +1) |
0.769 |
|
In‐hospital hyponatremia (Na levels < 135) d |
4 (8.2%) | 5 (10.2) | 1 |
| Discharge change in serum K c |
0.07 ± 0.63 0.03 (−0.32 to +0.4) |
−0.13 ± 0.68 0 (−0.4 to +0.2) |
0.199 |
|
In‐hospital hypokalaemia (K levels < 3.5) d |
2 (4.1%) | 4 (8.2%) | 0.678 |
| Discharge change in SBP c |
−25.92 ± 23.16 −25 (−37 to −10) |
−28.08 ± 19.79 −25 (−37 to −20) |
0.509 |
| Discharge FE Na c |
1.62 ± 1.48 1.3 (0.6 to 2) |
2.32 ± 1.88 2.1 (1.3 to 2.9) |
0.001** |
For B‐lines, eGFR and serum creatinine in the control group, two records did not have complete data (the actual number of values is shown between square brackets).
Abbreviations: ANCOVA, analysis of covariance, CD146, cluster of differentiation 146; CI, confidence interval; CKD, chronic kidney disease; eGFR, estimated glomerular filtration rate; FE Na, fractional excretion of sodium; IVC, inferior vena cava diameter; NTproBNP, N‐terminal pro–B‐type natriuretic peptide; SBP, systolic blood pressure; SD, standard deviation; WRF, worsening renal function.
Mean ± SD; non‐parametric Mann–Whitney test for the difference between the SGLT2i and control groups.
In each group, non‐parametric Wilcoxon signed rank test for the difference between discharge and baseline values.
Mean ± SD; median (inter‐quartile range with Tukey's hinges); non‐parametric Mann–Whitney test.
Counts (percentage); asymptotic chi‐square test or Fisher's exact test.
Adjusted mean ± standard error (95%CI); ANCOVA with type III sums of squares, baseline values as a covariate, dapagliflozin group as a fixed factor and CKD as a random factor.
WRF was defined as an absolute increase in creatinine ≥0.3 mg/dL, or ≥25% increase from baseline level.
P < 0.1.
P < 0.05.
P < 0.01.
Figure 2 shows the follow‐up values and trends in lung B‐lines, body weight, creatinine and eGFR. Figure 3 shows the changes in lung B‐lines. Both unadjusted and adjusted mean changes, with corresponding confidence intervals for the latter, are presented in full in Table 2 . Furthermore, the supplementary material comprises the ANCOVA analysis of changes in B‐lines.
Figure 2.
Follow‐up of efficacy and safety indicators with significant differences between the two study groups. At each time point, the standard errors are depicted. Numerical values are in Table 1 and Table 2 .
Figure 3.
Changes in lung B‐lines along the time landmarks. The box‐plots depict the median values with their respective inter‐quartile intervals (IQRs). The whiskers are proportional to 1.5*IQR (or trimmed to the minimum or maximum values) and the bullets are outliers. Numerical values are in Table 2 .
There were no significant differences in the occurrence of hyponatremia, hypokalaemia or hypotension between the treatment groups (values in Table 2 ), but there was a higher proportion of cases with WRF in the dapagliflozin group (30.6 vs. 12.8%, P = 0.045).
Diuresis and natriuresis outcomes
Among the 98 patients with complete 24‐h urine collections, there was no statistically significant difference in mean total urine volume between the two groups at 3 days post‐randomization (values in Table 3 ). However, dapagliflozin plus intravenous diuretic therapy resulted in increased 24‐h natriuresis compared with intravenous diuretic therapy alone (Table 3 ). The mean of total dose of intravenous furosemide administered during hospitalization adjusted for the CKD status was only slightly lower in the dapagliflozin group compared with the control group (470 mg vs. 510 mg, P = 0.702) (values in Table 3 ).
Table 3.
Secondary endpoints
| SGLT2i group (N = 49) | Control group (N = 49) | P‐value a , b , c | |
|---|---|---|---|
| Diuresis and natriuresis | |||
| 3‐day unadjusted mean diuresis a | 7230.78 ± 2309.22 | 7241.67 ± 2110.11 | 0.725 |
| 3‐day adjusted mean diuresis c |
7536.55 ± 392.3 (6757.74 to 8315.36) |
7422 ± 343.76 (6739.55 to 8104.45) |
0.802 |
| 24‐h urinary Na a | 253.35 ± 158.49 | 177.97 ± 99.99 | 0.012* |
| Adjusted mean of 24‐h urinary Na c |
248.03 ± 23.69 (200.99 to 295.07) |
173.83 ± 20.76 (133.62 to 216.05) |
0.009** |
| Total furosemide per episode‐of‐hospitalization (mg) a | 496.94 ± 600.22 | 525.92 ± 394.81 | 0.034* |
| Adjusted mean of total furosemide per episode‐of‐hospitalization (mg) c |
469.12 ± 90.78 (288.9 to 649.34) |
509.51 ± 79.55 (351.59 to 667.43) |
0.702 |
| Hospital care indicators | |||
| Hospital length of stay (days) a | 6.22 ± 2.15 | 7.57 ± 3.52 | 0.056 # |
| In‐hospital death b | – | – | – |
| 30‐day rehospitalization b | 15 (30.6%) | 15 (30.6%) | 1 |
| 30‐day death b | 1 (2%) | 2 (4.1%) | 1 |
Abbreviations: ANOVA, analysis of variance, CI, confidence interval; CKD, chronic kidney disease; SD, standard deviation.
Mean ± SD; non‐parametric Mann–Whitney test.
Counts (percentage); asymptotic chi‐square test or Fisher's exact test.
Adjusted mean ± standard error (95% CI); mixed‐effects ANOVA with type III sums of squares; dapagliflozin group as a fixed factor, and CKD as a random factor.
P < 0.1.
P < 0.05.
P < 0.001.
Hospitalization indicators
A slight reduction in the length of hospital stay was noted for the dapagliflozin group (i.e., more than one‐day reduction in average) and higher variability in the control group, although these differences did not reach statistical significance (values in Table 3 ). No deaths were reported during hospitalization. At 1‐month follow‐up, the rates of hospital readmissions were the same in both groups, with one death in the dapagliflozin group and two deaths in the control group (also shown in Table 3 ).
Exploratory analysis of echocardiographic parameters
Table 4 synthesizes the 1‐month changes in echo parameters of heart function. The median increase in ejection fraction at 1 month in the dapagliflozin group compared with the control group was statistically significant (5% vs. 0%, P = 0.003). The dapagliflozin group also experienced a median decrease in left ventricular end‐diastolic volume of 6.5 mL, whereas the control group showed a median increase of 4 mL (P = 0.03). Moreover, the median reduction in SPAP was consistently greater in the dapagliflozin group, although this difference was only marginally significant. Figure 4 depicts the echo parameters with significant improvement at 1‐month assessment.
Table 4.
One‐month change in echo parameters of heart failure function
| One‐month change from baseline | SGLT2i group (N = 48) | Control group (N = 47) | P‐value a |
|---|---|---|---|
| Change in EF (%) a |
5.99 ± 10.07 5 (0.35 to 11.5) |
1.46 ± 7.47 0 (−1 to +5) |
0.003** |
| Change in EDV a |
−12.54 ± 34.81 −6.5 (−27.5 to +3) |
−4.21 ± 30.91 4 (–11.5 to +10) |
0.033* |
| Change in SPAP a |
–12.29 ± 14.05 –13.5 (–20 to –6) |
–8.64 ± 9.07 −7 (−11 to −1) |
0.052 # |
| Change in GLS a |
1.28 ± 3.97 1.45 (−1 to +3.5) |
– | – |
| Change in RV strain a |
1.59 ± 7.54 2 (−3.45 to +5.25) |
– | – |
Abbreviations: EDV, left‐ventricular end‐diastolic volume; EF, ejection fraction; GLS, global longitudinal strain; RV, right ventricle; SD, standard deviation; SPAP, systolic pulmonary artery pressure.
Mean ± SD; median (inter‐quartile range with Tukey's hinges); non‐parametric Mann–Whitney test.
P < 0.1.
P < 0.05.
P < 0.001.
Figure 4.
One‐month changes in echo parameters of heart function. The box‐plots depict the median values with their respective inter‐quartile intervals (IQRs). The whiskers are proportional to 1.5*IQR (or trimmed to the minimum or maximum values) and the bullets are outliers. The significant asymmetry in values' distributions is apparent. Numerical values are in Table 4 .
An exploratory analysis of the changes in CD146 was also conducted. While the aggregate statistics did not reveal a significant difference between the two treatment groups (values in Table 2 ), further exploration of outliers indicated CD146 as a potential congestion marker. A synthetic analysis is presented in the Supporting Information.
Discussion
This single‐centre randomized clinical research investigated the efficacy and safety of initiating dapagliflozin (i.e., SGLT2i) within 24 h of hospital admission, in addition to the standard intravenous diuretic therapy in patients with AHF. Our findings highlight several key observations: (i) early introduction of dapagliflozin resulted in more efficient lung decongestion, as indicated by the lung B‐lines reduction at discharge and 1‐month assessment, along with a significant increase in 24‐h natriuresis; (ii) early initiation of dapagliflozin was found to be safe, with no major adverse events reported in AHF patients; (iii) initiation of dapagliflozin within the first 24 h of admission led to improved heart function at 1‐month follow‐up, as assessed through echocardiography.
In an AHF episode, clinicians face the dual challenge of effectively managing congestion while mitigating the risk of hemodynamic instability and electrolyte disturbances—all within the context of early initiation of optimal GDMT. 1 , 2 , 3 This approach aims to enhance prognosis and decrease the likelihood of early readmissions and mortality, in line with heart failure guidelines. 1 , 2 , 3
The combination of dapagliflozin with standard intravenous diuretics surpassed the standard intravenous diuretics alone in the lung B‐lines reduction and provided clinical advantages in effective decongestion. These findings hold significant clinical implications, for residual pulmonary congestion as assessed by lung ultrasound at discharge is recognized as a strong prognostic indicator for post‐discharge outcomes. 18 In a recent study by Perillo et al., the presence of pulmonary congestion at discharge assessed through lung ultrasound scores predicted worse short‐term outcomes in patients with AHF (e.g., re‐hospitalization, worsening of dyspnoea or cardiovascular death). 13 Recent studies have demonstrated that lung ultrasound‐guided therapy in AHF patients significantly reduces subclinical congestion at discharge. 19 In another investigation, Yeoh and colleagues found that the mean decrease of lung B‐lines over four days in AHF patients was comparable between those assigned to dapagliflozin versus metolazone. 12 Our study showed that dapagliflozin improved 24‐h natriuresis but did not enhance the total urinary output in these patients compared with the control group. After adjustment for CKD status, the mean of total furosemide per episode of hospitalization proved to be comparable in the two study groups (only slightly better for patients on dapagliflozin). Overall, these results reflect clinical benefits in congestion reduction. Our findings align with the DICTATE‐AHF trial, which demonstrated reduced loop diuretic dosages and improved median 24‐h natriuresis for AHF patients receiving dapagliflozin. 7 The EMPA‐RESPONSE‐AHF trial reported similar loop diuretic dosages in empagliflozin and placebo groups; however, a higher urinary output was noted in the empagliflozin group without a proportional increase in fractional excretion of sodium. 6 In another study on patients with AHF and diuretic resistance, dapagliflozin did not show any superiority at relieving congestion in comparison to the distal tubule diuretic, that is, metalozone and urinary output were the same in both groups. 12 Also, total furosemide dose was higher in the group assigned to dapagliflozin. 12 On the other hand, the EMPULSE trial (the largest trial on the use of SGLT2 inhibitors in AHF patients) reported a net clinical benefit in favour of patients receiving SGLT2 inhibitors, by assessing a composite symptom score. 20 The proposed decongestive properties of SGLT2 inhibitors responsible for alleviating the clinical outcomes in heart failure are osmotic and diuretic effect, restoration of glomerular filtration and correction of hyperfiltration, interstitial drainage, intracardiac pressure reduction, haemoconcentration and intravascular volume contraction. 21
Despite discrepancies regarding hospitalization outcomes and patient responses in numerous studies, the broader implications of these findings necessitate further exploration to clarify the influence of SGLT2 inhibitors on diuresis, natriuresis and decongestion. AHF is often characterized by sympathetic hyperactivity, which enhances SGLT2 expression at the nephron's proximal tubule. 22 , 23 Inhibition of SGLT2, where most glucose is reabsorbed, leads to a glycosuric‐dependent osmotic diuresis. The resultant increase in tubular osmolality induces proximal sodium dilution, fostering a sodium leak into the lumen, thereby increasing urine volume and natriuresis. 24 However, the inhibition of sodium reabsorption in the proximal tubule through glycosuria‐dependent osmotic diuresis is relatively insignificant; thus, sodium fractional excretion must also be understood in the context of the sodium‐hydrogen exchanger (NH3 exchanger). 25 , 26 , 27 The NH3 exchanger is closely linked to the SGLT2 receptor and accounts for approximately two‐thirds of sodium reabsorption in the proximal tubule. 26 Consequently, SGLT2 inhibition modulates the functionality of this exchanger, leading to increased natriuresis. 26 While these mechanisms of SGLT2 inhibition could theoretically result in substantial loss of water and sodium, with subsequent volume depletion and risks of ketoacidosis, 28 most trials involving SGLT2i in heart failure have demonstrated only short‐term, non‐sustained increases in urinary output alongside modest sodium excretion. 29 This phenomenon can be partially attributed to the activation of counterbalancing renal tubular mechanisms in response to SGLT2 inhibition, such as increased distal tubular sodium avidity due to enhanced downstream chloride delivery and activation of tubuloglomerular feedback, which mitigates sodium and water excretion. 29 Therefore, the observed short‐term increases in diuresis and natriuresis are primarily due to the glycosuric osmotic effect. However, this effect tends to be amplified in patients with diabetes and diminished in those with CKD. 30 These factors contribute to the inconsistent performance of SGLT2i in achieving sustained diuretic and decongestant effects in AHF patients. Given the complex pathophysiology of congestion in AHF, diuretics alone cannot achieve efficient and sustained euvolemia at discharge and post‐discharge; therefore, an early neuro‐hormonal blockade during admission (i.e., by adding SGLT2 inhibitors together with neprilysin inhibitors and mineralocorticoid inhibitors) would attenuate the sodium avidity and lead to better outcomes and congestion status. 31
While our study did not provide substantial evidence of benefit for the 1‐month outcome (hospital readmissions and mortality) when comparing the two groups, we did observe reduced left ventricular volumes and pulmonary systolic pressure, along with increased ejection fraction, in the dapagliflozin group. This is particularly noteworthy, as previous randomized clinical research have often reported reductions in ventricular volumes without concurrent improvements in ejection fraction. 32 , 33 Whether these favourable cardiac remodelling and functional effects are attributable to diuretic action or to distinct mechanisms requires further investigation.
CD146 has emerged as a biomarker associated with congestion and cardiovascular diseases, including heart failure. 34 In our study, SGLT2i appeared to affect CD146 levels variably across different patient subgroups, suggesting that individual responses to SGLT2i could be influenced by comorbidities. Previous clinical studies have indicated that CD146 levels are significantly elevated in patients with type 2 diabetes and CKD, correlating with microalbuminuria, thus highlighting its potential role as a biomarker for renal and vascular response to injury. 35 Nonetheless, careful consideration and additional analyses are needed for clinical interpretation of these observations.
Strengths and limitations
This study's strengths lie in its focus on the effects of very early dapagliflozin initiation (within 24 h of admission), a thorough investigation of various congestion markers, and the inclusion of 1‐month echocardiographic follow‐up to assess cardiac remodelling and long‐term functional changes. While other studies provided crucial data on broader clinical outcomes, this study offers a nuanced understanding of dapagliflozin's physiological effects on congestion and cardiac function in the acute setting, potentially revealing specific therapeutic windows and mechanisms of action not fully captured in larger, more broadly focused research.
On the other hand, the conclusions of this single‐centre, open‐label study should be interpreted cautiously due to several limitations: (a) the single‐centre design restricts generalizability; (b) the lack of blinding introduces potential bias; (c) the 1‐month follow‐up may be insufficient to fully assess long‐term effects; (d) the unconstrained randomization and subsequent imbalance between the two treatment groups in regard with CKD status and lung B‐lines is a potentially confounding factor; (e) changes at 1 month might be substantially influenced by the long‐term therapy, rather than the early administration of dapagliflozin; (f) although a robust analysis, the assumptions of ANCOVA, such as bivariate normal distribution and identical variance–covariance matrices in both groups, were not fully met, which might impair the adjustments by the baseline values; (g) the protocol and sample size planning did not include multivariate analysis to capture a comprehensive view of SGLT2i efficacy and safety in AHF patients. In addition, in the case of exploratory analysis of the CD146 changes, there might have been insufficient statistical power for detection of meaningful associations, for which the mechanisms should be further explored.
Conclusions
This study provides evidence that early initiation of dapagliflozin in AHF patients, in addition to standard intravenous loop diuretic therapy, resulted in (i) improved pulmonary decongestion; (ii) enhanced natriuresis; (iii) improved cardiac function at 1 month; (iv) an acceptable safety profile, with no increase in major adverse events. These findings suggest a potential benefit of early dapagliflozin use in optimizing decongestion and improving outcomes in AHF.
Conflict of interest
There are no conflicts of interests.
Funding
This research received no external funding.
Supporting information
Figure S1. Relationship of the one‐month changes in lung B‐lines as being linearly dependent on the baseline values. The ANCOVA model assumes equal regression slopes for the two groups (left graph), but unadjusted fitted lines display unequal regression slopes (right graph).
Figure S2. Relationship of the one‐month changes in CD146 as being dependent on the baseline values. The two regression lines were fitted on the raw data, with no adjustments. In the two treatment groups, the regression coefficients are the same and the coefficients of determination R‐squared are large.
Figure S3. Exploratory analysis of CD 146 changes at discharge: the sex makes no difference in changes of CD146, but the comorbidities lead to a wider range in SGLT2i group (−1559 to 1177) compared to the control group (−766 to 120).
Acknowledgements
Preliminary results of this research have been communicated by Dr. Radu I. Lala at the World Congress of Acute Heart Failure 2024, 11–14 May 2024, Lisbon, Portugal, and is indexed on Web of Science as WOS 001290635503178: IIR Lala, D Mocanu, L Pilat, M Puschita, A Pop‐Moldovan, D Darabantiu. Management of acute and residual congestion in patients with acute heart failure by adding dapagliflozin to intravenous loop diuretic therapy‐ENDORSE‐HF. European Journal of Heart Failure 26 (Suppl S2), 430–431.
Open access publishing facilitated by Anelis Plus (the official name of “Asociatia Universitatilor, a Institutelor de Cercetare – Dezvoltare si a Bibliotecilor Centrale Universitare din Romania”), as part of the Wiley ‐ Anelis Plus agreement.
Mocan, D. , Puschita, M. , Lungeanu, D. , Pop‐Moldovan, A. , Pilat, L. , Darabantiu, D. , Jipa, R. , and Lala, R. I. (2025) Pulmonary congestion relief by adding dapagliflozin to intravenous loop diuretic in acute heart failure patients. ESC Heart Failure, 12: 3320–3332. 10.1002/ehf2.15356.
References
- 1. McDonagh TA, Metra M, Adamo M, Gardner RS, Baumbach A, Böhm M, et al. 2021 ESC guidelines for the diagnosis and treatment of acute and chronic heart failure. Eur Heart J 2021;42:3599–3726. doi: 10.1093/eurheartj/ehab368 [DOI] [PubMed] [Google Scholar]
- 2. McDonagh TA, Metra M, Adamo M, Gardner RS, Baumbach A, Böhm M, et al. 2023 focused update of the 2021 ESC guidelines for the diagnosis and treatment of acute and chronic heart failure. Eur Heart J 2023;44:3627–3639. doi: 10.1093/eurheartj/ehad195 [DOI] [PubMed] [Google Scholar]
- 3. Correction to: 2023 focused update of the 2021 ESC guidelines for the diagnosis and treatment of acute and chronic heart failure: developed by the task force for the diagnosis and treatment of acute and chronic heart failure of the European Society of Cardiology (ESC) with the special contribution of the heart failure association (HFA) of the ESC. Eur Heart J 2024;45:53. doi: 10.1093/eurheartj/ehad613 [DOI] [PubMed] [Google Scholar]
- 4. Felker GM, Ellison DH, Mullens W, Cox ZL, Testani JM. Diuretic therapy for patients with heart failure. J Am Coll Cardiol 2020;75:1178–1195. doi: 10.1016/j.jacc.2019.12.059 [DOI] [PubMed] [Google Scholar]
- 5. Wilcox CS. Antihypertensive and renal mechanisms of SGLT2 (sodium‐glucose linked transporter 2) inhibitors. Hypertension 2020;75:894–901. doi: 10.1161/HYPERTENSIONAHA.119.11684 [DOI] [PubMed] [Google Scholar]
- 6. Damman K, Beusekamp JC, Boorsma EM, Swart HP, Smilde TDJ, Elvan A, et al. Randomized, double‐blind, placebo‐controlled, multicentre pilot study on the effects of empagliflozin on clinical outcomes in patients with acute decompensated heart failure (EMPA‐RESPONSE‐AHF). Eur J Heart Fail 2020;22:713–722. doi: 10.1002/ejhf.1713 [DOI] [PubMed] [Google Scholar]
- 7. Cox ZL, Collins SP, Hernandez GA, McRae AT, Davidson BT, Adams K, et al. Efficacy and safety of dapagliflozin in patients with acute heart failure. J Am Coll Cardiol 2024;83:1295–1306. doi: 10.1016/j.jacc.2024.02.009 [DOI] [PubMed] [Google Scholar]
- 8. Biegus J, Voors AA, Collins SP, Kosiborod MN, Teerlink JR, Angermann CE, et al. Impact of empagliflozin on decongestion in acute heart failure: the EMPULSE trial. Eur Heart J 2023;44:41–50. doi: 10.1093/eurheartj/ehac530 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Schulz KF, Altman DG, Moher D. CONSORT 2010 statement: updated guidelines for reporting parallel group randomised trials. Trials 2010;11:32. doi: 10.1186/1745-6215-11-32 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Ambrosy AP, Pang PS, Khan S, Konstam MA, Fonarow GC, Traver B, et al. Clinical course and predictive value of congestion during hospitalization in patients admitted for worsening signs and symptoms of heart failure with reduced ejection fraction: findings from the EVEREST trial. Eur Heart J 2013;34:835–843. doi: 10.1093/eurheartj/ehs444 [DOI] [PubMed] [Google Scholar]
- 11. Cohen J. A power primer. Psychol Bull 1992;112:155–159. doi: 10.1037/0033-2909.112.1.155 [DOI] [PubMed] [Google Scholar]
- 12. Yeoh SE, Osmanska J, Petrie MC, Brooksbank KJM, Clark AL, Docherty KF, et al. Dapagliflozin vs. metolazone in heart failure resistant to loop diuretics. Eur Heart J 2023;44:2966–2977. doi: 10.1093/eurheartj/ehad341 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Perillo A, Basile C, Fucile I, Rozza F, De Luca N, Mancusi C. Lung ultrasound at discharge predicts outcomes in heart failure: a pilot study. J Cardiovasc Med 2024. doi: 10.2459/JCM.0000000000001613 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Stewart WC, Jackson AL, Jenkins JN. Dropout rates for intent‐to‐treat and per protocol analyses. Am J Ophthalmol 2004;137:639–645. doi: 10.1016/j.ajo.2003.11.028 [DOI] [PubMed] [Google Scholar]
- 15. Shieh G. Power and sample size calculations for contrast analysis in ANCOVA. Multivar Behav Res 2017;52:1–11. doi: 10.1080/00273171.2016.1219841 [DOI] [PubMed] [Google Scholar]
- 16. Schwarz W. The ANCOVA model for comparing two groups: a tutorial emphasizing statistical distribution theory. Front Psychol 2025;16. doi: 10.3389/fpsyg.2025.1600764 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Vickers AJ, Altman DG. Statistics notes: analysing controlled trials with baseline and follow up measurements. BMJ 2001;323:1123–1124. doi: 10.1136/bmj.323.7321.1123 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Coiro S, Rossignol P, Ambrosio G, Carluccio E, Alunni G, Murrone A, Tritto I, Zannad F, Girerd N. Prognostic value of residual pulmonary congestion at discharge assessed by lung ultrasound imaging in heart failure. Eur J Heart Fail 2015;17:1172–1181. doi: 10.1002/ejhf.344 [DOI] [PubMed] [Google Scholar]
- 19. Burgos LM, Baro Vila RC, Ballari FN, Goyeneche A, Costabel JP, Muñoz F, et al. Inferior vena CAVA and lung ultrasound‐guided therapy in acute heart failure: a randomized pilot study (CAVAL US‐AHF study). Am Heart J 2024;277:47–57. doi: 10.1016/j.ahj.2024.07.015 [DOI] [PubMed] [Google Scholar]
- 20. Voors AA, Angermann CE, Teerlink JR, Collins SP, Kosiborod M, Biegus J, et al. The SGLT2 inhibitor empagliflozin in patients hospitalized for acute heart failure: a multinational randomized trial. Nat Med 2022;28:568–574. doi: 10.1038/s41591-021-01659-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Biegus J, Fudim M, Salah HM, Heerspink HJL, Voors AA, Ponikowski P. Sodium–glucose cotransporter‐2 inhibitors in heart failure: potential decongestive mechanisms and current clinical studies. Eur J Heart Fail 2023;25:1526–1536. doi: 10.1002/ejhf.2967 [DOI] [PubMed] [Google Scholar]
- 22. Katsurada K, Nandi SS, Sharma NM, Patel KP. Enhanced expression and function of renal SGLT2 (sodium‐glucose cotransporter 2) in heart failure: role of renal nerves. Circ Heart Fail 2021;14. doi: 10.1161/CIRCHEARTFAILURE.121.008365 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Rafiq K, Fujisawa Y, Sherajee SJ, Rahman A, Sufiun A, Kobori H, Koepsell H, Mogi M, Horiuchi M, Nishiyama A. Role of the renal sympathetic nerve in renal glucose metabolism during the development of type 2 diabetes in rats. Diabetologia 2015;58:2885–2898. doi: 10.1007/s00125-015-3771-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Layton AT, Vallon V. SGLT2 inhibition in a kidney with reduced nephron number: modeling and analysis of solute transport and metabolism. Am J Physiol‐Ren Physiol 2018;314:F969–F984. doi: 10.1152/ajprenal.00551.2017 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Schultheis PJ, Clarke LL, Meneton P, Miller ML, Soleimani M, Gawenis LR, et al. Renal and intestinal absorptive defects in mice lacking the NHE3 Na+/H+ exchanger. Nat Genet 1998;19:282–285. doi: 10.1038/969 [DOI] [PubMed] [Google Scholar]
- 26. Pessoa TD, Campos LCG, Carraro‐Lacroix L, Girardi ACC, Malnic G. Functional role of glucose metabolism, osmotic stress, and sodium‐glucose cotransporter isoform‐mediated transport on Na+/H+ exchanger isoform 3 activity in the renal proximal tubule. J Am Soc Nephrol 2014;25:2028–2039. doi: 10.1681/ASN.2013060588 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Coady MJ, El Tarazi A, Santer R, Bissonnette P, Sasseville LJ, Calado J, Lussier Y, Dumayne C, Bichet DG, Lapointe J‐Y. MAP 17 is a necessary activator of renal Na+/glucose cotransporter SGLT2. J Am Soc Nephrol 2017;28:85–93. doi: 10.1681/ASN.2015111282 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Ledoussal C, Lorenz JN, Nieman ML, Soleimani M, Schultheis PJ, Shull GE. Renal salt wasting in mice lacking NHE3 Na +/H + exchanger but not in mice lacking NHE2. Am J Physiol‐Ren Physiol 2001;281:F718–F727. doi: 10.1152/ajprenal.2001.281.4.F718 [DOI] [PubMed] [Google Scholar]
- 29. Packer M, Wilcox CS, Testani JM. Critical analysis of the effects of SGLT2 inhibitors on renal tubular sodium, water and chloride homeostasis and their role in influencing heart failure outcomes. Circulation 2023;148:354–372. doi: 10.1161/CIRCULATIONAHA.123.064346 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Hu S, Lin C, Cai X, Zhu X, Lv F, Nie L, Ji L. The urinary glucose excretion by sodium–glucose cotransporter 2 inhibitor in patients with different levels of renal function: a systematic review and meta‐analysis. Front Endocrinol (Lausanne) 2022;12. doi: 10.3389/fendo.2021.814074 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Biegus J, Cotter G, Metra M, Ponikowski P. Decongestion in acute heart failure: is it time to change diuretic‐centred paradigm? Eur J Heart Fail 2024;26:2094–2106. doi: 10.1002/ejhf.3423 [DOI] [PubMed] [Google Scholar]
- 32. Packer M, Cleland JGF, Bauersachs J. Great debate: SGLT2 inhibitors should be first‐line treatment in heart failure with reduced ejection fraction. Eur Heart J 2024;45:2186–2196. doi: 10.1093/eurheartj/ehae300 [DOI] [PubMed] [Google Scholar]
- 33. Omar M, Jensen J, Ali M, Frederiksen PH, Kistorp C, Videbæk L, et al. Associations of Empagliflozin with left ventricular volumes, mass, and function in patients with heart failure and reduced ejection fraction. JAMA Cardiol 2021;6:836. doi: 10.1001/jamacardio.2020.6827 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Gayat E, Caillard A, Laribi S, Mueller C, Sadoune M, Seronde M‐F, et al. Soluble CD146, a new endothelial biomarker of acutely decompensated heart failure. Int J Cardiol 2015;199:241–247. doi: 10.1016/j.ijcard.2015.07.039 [DOI] [PubMed] [Google Scholar]
- 35. Boutin L, Roger E, Gayat E, Depret F, Blot‐Chabaud M, Chadjichristos CE. The role of CD146 in renal disease: from experimental nephropathy to clinics. J Mol Med 2024;102:11–21. doi: 10.1007/s00109-023-02392-7 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Figure S1. Relationship of the one‐month changes in lung B‐lines as being linearly dependent on the baseline values. The ANCOVA model assumes equal regression slopes for the two groups (left graph), but unadjusted fitted lines display unequal regression slopes (right graph).
Figure S2. Relationship of the one‐month changes in CD146 as being dependent on the baseline values. The two regression lines were fitted on the raw data, with no adjustments. In the two treatment groups, the regression coefficients are the same and the coefficients of determination R‐squared are large.
Figure S3. Exploratory analysis of CD 146 changes at discharge: the sex makes no difference in changes of CD146, but the comorbidities lead to a wider range in SGLT2i group (−1559 to 1177) compared to the control group (−766 to 120).