Effects of Glucagon-like Peptide-1 Receptor Agonists on Heart, Kidney, and Metabolic Health in Romanian Patients with Type 2 Diabetes

This observational, retrospective study enrolled, consecutively, 150 patients (79 women and 71 men) diagnosed with T2D and treated with GLP-1 RA who were hospitalized in the Department of Diabetes, Nutrition and Metabolic Diseases of the "Pius Brînzeu" Emergency County Hospital in Timișoara, Romania and then monitored for a period of 1 year, with the assessment of biological parameters at the initiation of treatment, at 6 months, and at 12 months. All of the 150 patients completed the 12 months of follow-up without any discontinuations of GLP-1 RA treatment and without missing data. The protocol, designm and informed consent of this study were approved by the Ethics Committee of the hospital (no. 422/28.02.2024). All patients provided written informed consent, while the research was carried out according to the principles of the Declaration of Helsinki. A previous diagnosis of T2D and an age > 18 years represented the inclusion criteria, while the exclusion criteria were an estimated glomerular filtration rate (eGFR) < 30 mL/min/1.73 m² or the inability to provide an adequate medical history or informed consent. The patients were treated with either semaglutide administered subcutaneously, once weekly (0.5 or 1 mg); semaglutide administered orally, once daily (7 or 14 mg); or dulaglutide administered subcutaneously, 1.5 mg, once weekly.

The patients' gender, age, age at the diagnosis of T2D, height, weight, associated comorbidities, and the presence of diabetes complications were collected from the medical charts. Weight was measured using standardized equipment and procedures. The body mass index (BMI) was assessed as weight (kg)/height² (m²). Hypertension was defined as systolic blood pressure > 140 mmHg, diastolic blood pressure > 90 mmHg, or treatment with antihypertensive agents, while the measurements were performed according to the European Society of Cardiology guideline recommendations [15]. eGFR was calculated using the CKD-EPI creatinine equation, while the presence and stage of CKD were established depending on the eGFR and urinary albumin/creatinine ratio (UACR) [16]. Additionally, fasting blood glucose (FBG), hemoglobin A1c (HbA1c), total cholesterol (TC), low-density lipoprotein cholesterol (LDLc), triglycerides (TG), high-density lipoprotein cholesterol (HDLc), uric acid (UA), creatinine, urinary albumin, and urinary creatinine were assessed using standardized methods in the laboratory of the "Pius Brînzeu" Emergency County Hospital in Timișoara, after at least 12 h of fasting, and postprandial blood glucose was determined using a glucose meter. In order to assess insulin resistance, validated surrogate markers were used, namely the triglyceride/high-density lipoprotein cholesterol (TG/HDLc) ratio and the triglyceride–glucose index (TyG) calculated as ln [fasting triglycerides (mg/dL) × fasting plasma glucose (mg/dL)/2] [17].

MedCalc^® Statistical Software version 20.210 was used for the statistical analysis of the collected data (MedCalc Software Ltd., Ostend, Belgium; https://www.medcalc.org↗ (accessed on 17 January 2025); 2022). The continuous variables with Gaussian distribution are presented as mean ± standard deviation; continuous variables with non-parametric distribution as median, [interquartile range], and (average rank); and categorical variables as frequencies and percentages. The significance of differences between two groups was assessed with the Mann–Whitney U test in non-parametric populations, Student's t-test in Gaussian populations, and with the chi-squared test in categorical populations, while the Friedman test was used to determine the significance of differences between 3 groups. Additionally, the significance of the median difference between two paired samples was assessed with the paired-samples Wilcoxon test. The sample size was calculated before the inclusion of subjects, considering data found in the literature, in order to obtain a confidence level of 95% and a statistical power higher than 80%. The p-value threshold below which the statistical significance was considered is 0.05.

This research revealed statistically significant differences between male and female participants in terms of initial weight, serum uric acid, and prevalence of peripheral artery disease (higher in men) and HDLc (higher in women), results similar to those of previous studies [18,19].

Regarding the potential impact of treatment with GLP-1 RAs, 59.3% of patients had a weight reduction of more than 5% during the entire study duration, and 24.7% had a weight decrease of 10% after 12 months of treatment, effects similar to those described in the PIONEER 5 clinical trial, although it had a radically different design [20]. The absolute weight loss differences between month 6 and the initial dose (an average of 4 kg), month 12 vs. month 6 (an average of 2 kg), and between month 12 and the initial dose (an average of 6 kg) indicate a sustainable reduction in body weight, with significant early beneficial results and with the main weight loss being measured in the first months of the treatment. These results are similar to those observed in the SUSTAIN-1 clinical trial, in which the weight loss was of approximately 3.73 kg with injectable semaglutide (0.5 mg/week) and 4.53 kg with injectable semaglutide 1 mg/week, and those observed in the AWARD-3 clinical trial, which highlighted a decrease of 3.1 kg in 12 months of treatment with dulaglutide (1.5 mg/week) [21,22].

Additionally, the absolute decreases in BMI between month 6 and the initial dose (an average of 1.1 kg/m²), month 12 vs. month 6 (an average of 0.8 kg/m²), and between month 12 and the initial dose (an average of 2 kg/m²) are consistent with those of the weight loss per se, suggesting a significant effect in reducing the metabolic risk of patients treated with GLP-1 RAs, effects already noticed in the first half of the trial duration. These findings can also explain the beneficial effects that treatment with GLP-1 RA had on insulin resistance, with a statistically significant reduction in TG/HDLc ratio (with an average of 1.04) and in TyG index (with an average of 0.16), both surrogate markers of insulin resistance, suggesting improved insulin sensitivity as also shown in multiple other trials and meta-analyses [23,24,25].

From the point of view of HbA1c, the results are similar, with the most significant reduction being noticed in the first 6 months of treatment, 0.7%, with a total average decrease of 0.8% in the 12 months of study, indicating a sustainable and significant glycemic effect without increasing the risk of hypoglycemia, similarly to what was identified in the PIONNER 5 clinical trial [20]. Moreover, though at the beginning of the treatment with GLP-1 RA, only 10% of the patients had reached the HbA1c target of <7%, after 12 months of treatment, 27.3% reached the target without any additional interventions, results underlying the favorable impact of using this class of antidiabetic drugs. These effects have been also noticed regarding the values of FBG, which decreased statistically significantly in the first 6 months, with approximately 15 mg/dL; in the following 6 months, with 11.5 mg/dL; and during the entire study duration, with 38.5 mg/dL, without an additional hypoglycemia burden. This has occurred with a marked decrease in the first 6 months of treatment, a decrease that continued in the following 6 months, with a slower downward slope. The results of this study showed a more significant reduction in fasting blood glucose levels compared to the AWARD-3 trial, where the decrease was approximately 30 mg/dL [20,21].

Treatment with GLP-1 RAs was also associated with improvements in the lipid profile of the included patients, with a statistically significant difference between the initiation of the medication and 12 months later, as follows: a lower LDLc with an average of 8 mg/dL, a lower TG with an average of 19 mg/dL, a lower nonHDLc with an average of 3.8 mg/dL, and a higher HDLc with an average of 4 mg/dL, or, summed up, a cardioprotective improvement in atherogenic dyslipidemia, as also shown in the SELECT trial published in November 2023, although the causal effect cannot be assessed without a comparator arm [26]. Other favorable results concern other components of metabolic syndrome, namely blood pressure values, both systolic and diastolic, which decreased significantly in the 12 months of treatment with GLP-1 RAs, and also serum uric acid levels, which were significantly reduced, similarly to the results of the same SELECT randomized controlled trial and also to the results of several meta-analyses with this class of antihyperglycemic agents, although each was carried out with a completely different design not directly comparable with this real-world sample [26,27,28,29].

Lastly, although UACR did not change significantly in this study, the eGFR decrease was slower than the one expected for a similar population without a GLP-1 RA in their treatment regimen, results comparable to the ones of the FLOW trial, which highlighted a less steep decline in renal function in patients treated with injectable semaglutide [30]. However, considering that our study did not include a control group, it can only lead to descriptive associations, in contrast to the FLOW trial, where nephroprotection of injectable semaglutide can be affirmed.

This study analyzes the impact of treatment with GLP-1 RAs on glycemic values, HbA1c, weight, insulin resistance, and other cardio-reno-metabolic risk factors in patients with T2D and offers insights regarding a less-studied specific population, namely Romanian individuals (Central–Eastern European population) with this metabolic condition. Considering that T2D is a complex disease in which multiple genetic, environmental, and behavioral risk factors influence individuals' susceptibility, this study focusing on a specific population offers valuable insights that can be further used in daily medical practice [31].

The weaknesses of this research include the retrospective design of the study and the absence of a control group, which both limit the possibility of identifying causal relationships between treatment with GLP-1 RA and suggested beneficial effects in this population, since residual confounding cannot be ruled out. Another limitation is caused by the scarce information regarding some of these potential confounding factors, including the other types of medication used by the subjects such as antihyperglycemic, antihypertensive, or lipid-lowering therapies, details that will be included in future studies. Regarding the statistical analysis of the data, small effect sizes can raise questions about clinical versus statistical significance, questions that could be resolved through future studies including a larger population observed for a longer duration of time. Additionally, information regarding lifestyle changes (such as dietary interventions or physical activity) and details regarding hospitalization, certain comorbidities, drug dosage, adherence, adverse events, tolerability, and treatment discontinuation rates were not consistently available; nevertheless, they will be included in future research.

This study highlights the importance of early initiation of treatment with GLP-1 RAs in patients with T2D in order to improve their cardio–reno–metabolic risk. Furthermore, it emphasizes the need for an early and multifactorial approach for these patients in order to prevent, to detect in a timely manner, and to treat possible chronic degenerative complications and comorbidities frequently encountered in such a setting, choosing, according to each case, an antihyperglycemic agent with proven beneficial pleiotropic effects. Future research that could assess in more detail the impact of GLP-1 RAs should include a greater number of subjects, with or without T2D but with excessive weight, while also including multiple insulin resistance, inflammation, and hepatic biomarkers, with an additional focus on body composition, quality of life, and mental health evaluations.