This research was conducted in accordance with the ethical principles set forth in the Declaration of Helsinki. This research employed a retrospective cohort study design, focusing on patients with HFpEF who received treatment at Shandong Provincial Hospital Affiliated to Shandong First Medical University between June 2022 and April 2024. Of these, 105 patients underwent standard treatment and were classified into the routine group, while 110 patients were treated with a combination of standard therapy and dapagliflozin and were classified into the dapagliflozin group. Data on patient demographics, such as gender, age, and underlying conditions, as well as medication usage and adverse reactions, were extracted from the hospital’s medical record system (Fig. 1).

Fig. 1.

Flowchart depicting participant selection and inclusion in this study. Abbreviations: HF, heart failure; LVEF, left ventricular ejection fraction; NYHA, New York Heart Association.

The study received approval from the Ethics Committee of Shandong Provincial Hospital Affiliated to Shandong First Medical University (Approval No. 2024-427). Due to the study’s reliance solely on de-identified patient data, informed consent was waived, as no potential harm or impact on patient care was anticipated.

Participants were eligible for the study if they were over 75 years old, had a clinical diagnosis of HFpEF [18] with a left ventricular ejection fraction (LVEF) greater than 50%, and were classified under the New York Heart Association (NYHA) functional classes II to IV. Additional inclusion criteria include stable history of heart failure (defined as no hospitalization for heart failure within the last three months prior to enrollment), evidence of structural heart disease, and elevated levels of natriuretic peptides, specifically N-terminal pro-B-type natriuretic peptide (NT-proBNP) $\ge$225 pg/mL or B-type natriuretic peptide (BNP) $\ge$75 pg/mL [19]. For those with atrial fibrillation, the required levels were NT-proBNP $\ge$375 pg/mL or BNP $\ge$100 pg/mL.

Participants were excluded if they had been hospitalized for decompensated heart failure within the past seven days; had an estimated glomerular filtration rate (eGFR) of less than 20 mL/min/1.73 m² at admission; had type 1 diabetes, multiple organ dysfunction, or allergies to any medications used in the study; or had malignant tumors. Additionally, patients with incomplete clinical records or missing key data points were also excluded.

Standard care was given to each patient of both groups to address their coexisting conditions. Treatment regimens included: 5 mg once daily enalapril tablets (angiotensin-converting enzyme [ACE] inhibitor; China Resources Shuanghe Limin Pharmaceutical Co., Ltd., Jinan, China, Approval No. H20066730), 100 mg once daily losartan potassium tablets (angiotensin II receptor blocker [ARB]; Ningbo Korcomer Novartis Pharmaceutical Co., Ltd., Ningbo, China, Approval No. H20213818), 50 mg once daily Entresto (angiotensin receptor-neprilysin inhibitor [ARNI]; Beijing Novartis Pharmaceutical Co., Ltd., Beijing, China, Approval No. J20190001), 10 mg once daily propranolol ($\beta$-blocker; Shenyang First Pharmaceutical Company of Dongbei Pharmaceutical Group, Shenyang, China, Approval No. H21021826), and 20 mg once daily furosemide tablets (diuretic; Tianjin Lisheng Pharmaceutical Co., Ltd., Tianjin, China, Approval No. H12020163). Spironolactone tablets (20–40 mg once daily), mineralocorticoid receptor antagonist [MRA]; Hangzhou Minsheng Pharmaceutical Group Co., Ltd., Hangzhou, China, Approval No. H33020070) were also administered. These treatments were aimed at controlling blood pressure, managing blood glucose levels, addressing fluid retention, and enhancing heart function. Specifically, ACE inhibitors and ARBs reduce systemic vascular resistance and improve cardiac remodeling; ARNIs combine the benefits of an ARB with a neprilysin inhibitor, which increases natriuretic peptides that promote vasodilation and natriuresis; $\beta$-blockers help reduce heart rate and myocardial oxygen consumption, improving overall cardiac function; and diuretics are administered to manage fluid overload and alleviate symptoms of heart failure. Additionally, some patients utilized pacemakers (Discovery, Model 1272, Guidant, Indianapolis, IN, USA) or internal cardiac defibrillators (ICDs) (Maximo, Model 7232, Medtronic, Dublin, Ireland) to manage arrhythmias and prevent sudden cardiac death. The entire treatment period lasted for 12 weeks, during which patients were closely monitored for any adverse effects, and adjustments were made as necessary.

For the group receiving routine treatment plus dapagliflozin, dapagliflozin tablets (10 mg per tablet, packaged as 14 tablets per pack, AstraZeneca Pharmaceuticals Co., Ltd., Shanghai, China, Approval No. H20234463) were additionally administered. The dosage was set at 10 mg once daily over the 12-week treatment period [20].

The Kansas City Cardiomyopathy Questionnaire (KCCQ) was employed to assess heart failure-related symptoms, specifically their frequency, severity, and recent changes, along with evaluations of physical function, quality of life, and social function, both before and after treatment [21]. Within the KCCQ, the Total Symptom Score (TSS) measures the frequency and severity of symptoms, while the Physical Limitation Score (PLS) assesses physical function. The Clinical Summary Score (CSS) encompasses both the symptom and physical function domains. Meanwhile, the Overall Summary Score (OSS) integrates all major domains, including symptom frequency and severity, physical function, quality of life, and social function. Each domain’s validity, reproducibility, responsiveness, and interpretability have been independently confirmed. Scores were scaled from 0 to 100, with higher scores indicating better health status. The instrument demonstrates high reliability, with a Cronbach’s alpha of 0.97 and an intraclass correlation coefficient of 0.95 [22].

Fasting venous blood samples (5 mL) were collected from patients both before and after treatment. These samples were treated with ethylenediaminetetraacetic acid (EDTA) as an anticoagulant and centrifuged at 2800 rpm for 15 minutes to separate the serum. The serum was then analyzed for NT-proBNP, BNP, eGFR, and glycated hemoglobin (HbA1c) using an automated biochemical analyzer (BS-280, Mindray, Shenzhen, China).

The Six-Minute Walk Test (6MWT) was administered indoors on a flat 30-meter corridor, with patients instructed to walk as far as possible within 6 minutes, in accordance with the American Thoracic Society guidelines [23]. The total distance covered during the 6-minute period was recorded in meters.

Color Doppler echocardiography (Vivid E95, GE HealthCare, Chicago, IL USA) was employed to assess cardiac function using a 3.0 MHz probe frequency both before and after treatment. Subjects were positioned in the left lateral decubitus position, and the apical four-chamber view was used for pulsed Doppler measurements. The left ventricular end-diastolic dimension (LVEDD), left ventricular end-systolic dimension (LVESD), and left ventricular end-diastolic volume (LVEDV) were measured from the parasternal long-axis view. The left ventricular mass (LVM) was calculated using the area-length method, from which the left ventricular mass index (LVMI) was derived. Additionally, the left ventricular remodeling index (LVRI) was determined by dividing LVM by LVEDV.

Adverse events were defined as any untoward medical occurrence in a patient receiving treatment, temporally associated with the use of the study medication, whether or not considered related to the medication. Adverse events were monitored and recorded throughout the 12-week treatment period through regular clinical follow-up visits and review of medical records. All adverse events were documented in the electronic medical record system, including their severity, duration, and any actions taken. Serious adverse events were defined as those resulting in death, life-threatening conditions, hospitalization, or significant disability. The primary adverse events of interest included hypoglycemia, diabetic ketoacidosis, acute kidney injury, myocardial infarction, stroke, and lower limb amputations.

A hospital-developed questionnaire was utilized to evaluate patients’ satisfaction with the treatments administered in both groups. The questionnaire was scored on a scale from 0 to 100 points. The total scores obtained were categorized as follows: very satisfied (90–100 points), satisfied (60–89 points), and dissatisfied (less than 60 points). Higher scores indicated greater levels of satisfaction among participants.

Data were analyzed using the SPSS version 29.0 statistical software (SPSS Inc., Chicago, IL, USA). Categorical data are presented as frequencies (n) and percentages (%). The chi-square test was applied using the basic formula when the sample size was $\ge$40 and the expected frequency (T) was $\ge$5. When the sample size was $\ge$40 but 1 $\le$ T 5, the corrected chi-square formula was used. For sample sizes 40 or when T 1, Fisher’s exact test was employed. Continuous variables were assessed for normal distribution using the Shapiro–Wilk test. Normally distributed data are expressed as mean $\pm$ standard deviation (SD). Independent sample t-tests were used for inter-group comparisons, and paired sample t-tests were applied for intra-group comparisons. A p-value 0.05 was considered statistically significant.

The groups were well-matched in terms of age (80.15 $\pm$ 1.47 vs 80.34 $\pm$ 1.59 years, p = 0.362), gender distribution (51.43% vs 56.36% male, p = 0.468), and body mass index (BMI) (29.78 $\pm$ 2.34 vs 29.27 $\pm$ 2.12 kg/m², p = 0.096). Hypertension prevalence was high in both groups (87.62% vs 85.45%, p = 0.642). No significant differences were observed in other comorbidities including smoking history (23.81% vs 28.18%, p = 0.465) and drinking history (32.38% vs 30.00%, p = 0.706), LVEF (55.83 $\pm$ 2.14% vs 56.16 $\pm$ 2.28%, p = 0.275), heart failure duration, New York Heart Association (NYHA) classification, previous heart failure hospitalizations, or vital signs (all p $>$ 0.05). There were no statistically significant differences in any baseline characteristics between the two groups (all p $>$ 0.05) (Table 1).

The use of ACE inhibitors was similar between the two groups, with 33.33% in the routine group and 31.82% in the dapagliflozin group (p = 0.813) (Table 2). ARBs were utilized by 37.14% of the routine group and 35.45% of the dapagliflozin group (p = 0.797). The use of ARNIs was low in both groups, at 3.81% and 2.73%, respectively (p = 0.950). $\beta$-blocker use was high in both groups, with 78.10% in the routine group and 75.45% in the dapagliflozin group (p = 0.647). MRAs were used by 37.14% in the routine group compared to 41.82% in the dapagliflozin group (p = 0.483). Similarly, the use of loop diuretics was prevalent in the routine and dapagliflozin groups, accounting for 79.05% and 82.73% patients, respectively (p = 0.492). Pacemaker implantation was observed in 15.24% of the routine group and 12.73% of the dapagliflozin group (p = 0.595), and the utilization of an internal cardiac defibrillator (ICD) was rare in both groups (p = 1.000). Overall, usage of different treatment modalities was comparable across the two groups, an indication of comparable baseline management of heart failure.

Both groups had similar baseline levels of NT-proBNP and BNP, with no significant differences prior to treatment (p $>$ 0.05) (Table 3). After treatment, the dapagliflozin group showed a significant reduction in NT-proBNP levels, compared to the routine group (619.04 $\pm$ 145.47 vs 659.34 $\pm$ 148.75 pg/mL, p = 0.046). Similarly, BNP levels exhibited significantly greater decrease in the dapagliflozin group compared to the routine group (132.36 $\pm$ 20.43 vs 138.29 $\pm$ 21.38 pg/mL, p = 0.039). The eGFR was comparable between the two groups before treatment (p = 0.716), but post-treatment, the dapagliflozin group had a more pronounced reduction than the routine group (50.13 $\pm$ 10.36 vs 53.11 $\pm$ 10.24 mL/min/1.73 m², p = 0.035). HbA1c levels before treatment were similar (p = 0.565), while after treatment, the dapagliflozin group experienced a significantly greater reduction compared to the routine group (5.75 $\pm$ 1.04% vs 6.12 $\pm$ 1.02%, p = 0.009). These findings suggest that the addition of dapagliflozin may enhance biochemical outcomes in this patient population, particularly those concerning heart failure and glycemic control markers. Both groups showed reduced NT-proBNP and BNP levels post-treatment. Dapagliflozin administration also improved HbA1c and eGFR significantly, indicating enhanced cardiac and renal functions following treatment.

Baseline scores were similar across all KCCQ subscales, with no significant differences observed before treatment (p $>$ 0.05) (Table 4). After treatment, the KCCQ OSS showed a significant increase in the dapagliflozin group compared to the routine group (72.48 $\pm$ 12.49 vs 68.59 $\pm$ 13.11, p = 0.027). The CSS also improved significantly in the dapagliflozin group versus the routine group (74.87 $\pm$ 11.49 vs 71.61 $\pm$ 10.68, p = 0.033). Similarly, the TSS exhibited significant increase in the dapagliflozin cohort compared to the routine cohort (80.06 $\pm$ 20.26 vs 74.28 $\pm$ 20.13, p = 0.037). The PLS was significantly higher in the dapagliflozin group than in the routine group (67.67 $\pm$ 3.26 vs 66.39 $\pm$ 3.45, p = 0.006). These results highlight the potential of dapagliflozin in improving clinical outcomes and quality of life in elderly patients with HFpEF. The improvement in KCCQ scores was not only statistically significant but also clinically meaningful, as indicated by the substantial increase in the OSS and PLS. These improvements suggest that dapagliflozin may enhance both symptom relief and physical functioning in the elderly HFpEF patients. In the routine group, the PLS was the only KCCQ subscale showing significant improvement post-treatment (p 0.001). Significant improvements in the KCCQ OSS, CSS, TSS, and PLS subscales were observed in the dapagliflozin group post-treatment (all p 0.05).

Before treatment, the distances were similar between the routine group and the dapagliflozin group (244.58 $\pm$ 65.49 vs 244.26 $\pm$ 61.24 m, p = 0.970) (Table 5). After treatment, however, there was a marked increase in the 6MWT distance in the dapagliflozin group compared to the routine group (358.47 $\pm$ 28.62 vs 311.69 $\pm$ 30.26 m, p 0.001). This suggests that the addition of dapagliflozin to the standard treatment regimen substantially enhances exercise capacity in elderly patients with HFpEF. The increase in 6MWT distance highlights the potential impact on quality of life through improved exercise capacity. Walking distance increased significantly after treatment in both groups.

Baseline measurements of LVESD, LVEDD, LVMI and LVRI were similar between the routine group and the dapagliflozin group (p $>$ 0.05) (Table 6). Compared to the routine group, the dapagliflozin group demonstrated a significantly lower LVESD (36.95 $\pm$ 4.16 vs 38.51 $\pm$ 3.12 mm, p = 0.002) and LVEDD (52.32 $\pm$ 5.27 vs 54.12 $\pm$ 4.23 mm, p = 0.006) post-treatment. The LVMI also decreased significantly in the dapagliflozin group compared to the routine group (104.23 $\pm$ 12.24 vs 108.34 $\pm$ 13.15 g/m², p = 0.019). Additionally, the LVRI showed a greater reduction in the dapagliflozin group compared to the routine group (1.37 $\pm$ 0.12 vs 1.43 $\pm$ 0.14 g/mL, p 0.001). These findings indicate that dapagliflozin treatment was associated with favorable cardiac remodeling and improved echocardiographic parameters in this elderly patient cohort. In both groups, significant reductions in LVESD, LVEDD, LVMI, and LVRI were noted post-treatment (all p 0.001) (Table 6). Additionally, imaging assessments showed that left ventricular structure and remodeling indices were significantly improved in both groups of patients after treatment (Fig. 2), the left ventricular cavity diameter decreased, and the wall tended to restore to normal thickness.

Fig. 2.

Echocardiographic images before and after treatment. (A,B) Representative images of patients from the routine group before (A) and after treatment (B). (C,D) Representative images of patients from the dapagliflozin group before (C) and after treatment (D).

There were no occurrences of lower limb amputations, diabetic ketoacidosis (DKA), or severe hypoglycemia in either group (p = 1.000) (Table 7). A single instance of nonfatal myocardial infarction (MI) and stroke was reported in the routine treatment group, while there were no cases in the dapagliflozin group, but these differences were not significant (p = 0.488 for both). Acute kidney injury occurred in 4.76% of the routine group and 4.55% of the dapagliflozin group, with no statistically significant difference (p = 1.000). These findings suggest that adding dapagliflozin to the treatment regimen does not increase the risk of these adverse events in the studied population.

In the dapagliflozin group, 30.91% of patients expressed high level of treatment satisfaction (labeled as “very satisfied”) compared to 19.05% of patients in the routine group (Table 8). The proportion of patients who were “satisfied” with the treatments administered was similar across the two groups, with 64.76% in the routine group and 61.82% in the dapagliflozin group. However, the percentage of “dissatisfied” patients was lower in the dapagliflozin group (7.27%) compared to the routine group (16.19%). Overall, the satisfaction rate was significantly higher in the dapagliflozin group than in the routine group (92.73% vs 83.81%, p = 0.041). These findings suggest that elderly HFpEF patients over 75 years old are generally satisfied with the addition of dapagliflozin to the standard treatment regimen.