Clinical and metabolic outcomes of artificial pancreas systems in patients with type 1 diabetes: A real-world study

Introduction

Type 1 diabetes mellitus (T1DM) is a chronic autoimmune disease characterized by insulin deficiency and persistent hyperglycemia. Despite advances in insulin therapy, maintaining optimal glycemic control remains challenging for many patients. Poor glycemic control is associated with the development of microvascular and macrovascular complications, including nephropathy, retinopathy, and cardiovascular disease.

Artificial pancreas systems, also known as closed-loop insulin delivery systems, combine continuous glucose monitoring with automated insulin delivery algorithms^1,2,3. These systems represent a major technological advancement in diabetes management and are designed to improve glycemic control while reducing the burden of self-management^4,5,6.

Recent studies have demonstrated that artificial pancreas systems improve time in range and reduce HbA1c levels in patients with T1DM^7,8. However, real-world data evaluating their effects on broader cardiometabolic parameters remain limited. In particular, their potential effects on lipid profiles and renal markers such as proteinuria have not been extensively investigated.

Therefore, the aim of this study was to evaluate the real-world clinical and metabolic outcomes of artificial pancreas systems in patients with T1DM, focusing on glycemic control, cardiometabolic parameters, and proteinuria.

Materials and methods

This retrospective study included patients with T1DM who initiated artificial pancreas therapy at our center. A total of 32 patients were included in the analysis. The mean age of the patients was 31.19 ± 7.69 years (range 19–50), and the mean duration of diabetes was 17.44 ± 7.41 years. Of the participants, 22 (68.8%) were female and 10 (31.2%) were male. Clinical and laboratory data were retrospectively obtained from patient records (Table 1).

OPEN IN VIEWER

Table 1.

Demographic characteristics of the study population.

Variable	n (%)	Mean ± SD	Min–Max
Age (years)	—	31.19 ± 7.69	19–50
Diabetes duration (years)	—	17.44 ± 7.41	3–33
Sex
Female	22 (68.8)	—	—
Male	10 (31.2)	—	—

Glycemic parameters including HbA1c, fasting blood glucose (FBG), and postprandial blood glucose (PBG) were recorded before pump initiation and at 3 months, 6 months, and 1 year after therapy. Lipid parameters (LDL cholesterol, HDL cholesterol, triglycerides, and total cholesterol) and spot urine proteinuria levels were evaluated before pump therapy and at the 1-year follow-up. The proportion of patients achieving HbA1c target levels was also evaluated during follow-up.

Statistical analyses were performed using IBM SPSS software. Continuous variables were expressed as mean ± standard deviation, while categorical variables were presented as numbers and percentages. Repeated measures ANOVA and paired t-tests were used for comparisons of continuous variables across time points. Cochran’s Q and McNemar tests were used to evaluate changes in categorical variables. A p-value <0.05 was considered statistically significant.

Results

A significant reduction in HbA1c levels was observed during follow-up. The mean HbA1c decreased from 8.74 ± 1.21% before pump initiation to 7.92 ± 0.98% at 3 months, 7.61 ± 0.87% at 6 months, and 7.48 ± 0.83% at 1 year (p < 0.001). Similarly, fasting blood glucose levels decreased from 186.3 ± 52.4 mg/dL before pump therapy to 149.9 ± 35.6 mg/dL at 1 year (p < 0.001). Postprandial glucose levels also significantly decreased from 228.7 ± 61.5 mg/dL to 181.5 ± 41.2 mg/dL during follow-up (p < 0.001) (Table 2).

OPEN IN VIEWER

Table 2.

Comparison of metabolic parameters before and after artificial pancreas therapy.

Variable	Pre-pump	3 months	6 months	1 year	p
HbA1c (%)	8.74 ± 1.21	7.92 ± 0.98	7.61 ± 0.87	7.48 ± 0.83	<0.001
FBG (mg/dL)	186.3 ± 52.4	162.8 ± 41.7	154.6 ± 38.2	149.9 ± 35.6	<0.001
PBG (mg/dL)	228.7 ± 61.5	196.4 ± 49.3	187.8 ± 44.6	181.5 ± 41.2	<0.001

Lipid profile analysis showed a significant decrease in LDL cholesterol from 112.6 ± 29.8 mg/dL to 104.2 ± 26.1 mg/dL at 1 year (p = 0.029). Total cholesterol levels also decreased significantly (p = 0.006). HDL cholesterol levels increased significantly from 49.7 ± 11.2 mg/dL to 52.3 ± 10.4 mg/dL (p = 0.022). Although triglyceride levels decreased from 146.8 ± 58.3 mg/dL to 132.4 ± 49.6 mg/dL, this change was not statistically significant (p = 0.069). Proteinuria levels decreased significantly from 84.6 ± 42.3 mg/g before pump therapy to 61.8 ± 35.7 mg/g at 1 year (p = 0.011) (Table 3).

OPEN IN VIEWER

Table 3.

Cardiometabolic parameters before pump therapy and at 1 year.

Variable	Pre-pump	1 year	p
LDL (mg/dL)	112.6 ± 29.8	104.2 ± 26.1	0.029
HDL (mg/dL)	49.7 ± 11.2	52.3 ± 10.4	0.022
Triglycerides (mg/dL)	146.8 ± 58.3	132.4 ± 49.6	0.069
Total cholesterol (mg/dL)	189.5 ± 34.7	178.1 ± 31.2	0.006
Proteinuria (mg/g)	84.6 ± 42.3	61.8 ± 35.7	0.011

The proportion of patients achieving HbA1c targets increased substantially over time (Table 4). While only 6 patients (18.8%) were within target before pump therapy, this proportion increased to 14 patients (43.8%) at 3 months, 18 patients (56.3%) at 6 months, and 20 patients (62.5%) at 1 year. The increase in target achievement was statistically significant (p < 0.001), with post-hoc analysis indicating that the difference between baseline and 1-year follow-up was particularly significant (Figure 1).

OPEN IN VIEWER

Table 4.

Comparison of HbA1c target achievement before and after artificial pancreas therapy.

Variable	Pre-pump	Post-pump 3 months	Post-pump 6 months	Post-pump 1 year	p (Pre-pump vs 3 months)	p (Pre-pump vs 6 months)	p (Pre-pump vs 1 year)
Target achieved, n (%)	6 (18.8)	14 (43.8)	18 (56.3)	20 (62.5)	1.000	0.074	<0.001
Target not achieved, n (%)	26 (81.2)	18 (56.2)	14 (43.7)	12 (37.5)

OPEN IN VIEWER

Figure 1.

Change in HbA1c levels after initiation of artificial pancreas therapy.

Discussion

Our study demonstrates that artificial pancreas systems significantly improve glycemic control in patients with type 1 diabetes in a real-world clinical setting. The progressive reduction in HbA1c observed during follow-up is consistent with findings from randomized trials evaluating closed-loop insulin delivery systems ⁹ .

Previous large clinical trials have shown that artificial pancreas systems significantly reduce HbA1c and increase time spent within the target glucose range compared with conventional insulin therapy. For example, Brown et al. demonstrated significant improvements in glycemic control with closed-loop systems during a 6-month randomized trial. Similarly, Bergenstal et al. reported improved glycemic outcomes with hybrid closed-loop therapy in patients with type 1 diabetes^1,2.

Beyond improvements in glycemic control, our study also demonstrated favorable changes in cardiometabolic parameters. The observed reduction in LDL cholesterol and total cholesterol levels, along with the increase in HDL cholesterol, may reflect the metabolic benefits associated with improved glycemic regulation. Previous studies have shown that improved glucose control is associated with reductions in cardiovascular risk factors in patients with diabetes^10,11.

Another important finding of our study was the significant reduction in proteinuria levels during follow-up. Diabetic nephropathy remains one of the most important microvascular complications of type 1 diabetes, and persistent hyperglycemia plays a central role in its pathogenesis. Improved glycemic control achieved with artificial pancreas systems may contribute to renal protection by reducing glomerular hyperfiltration and metabolic stress^7,10,12.

Additionally, the proportion of patients achieving HbA1c targets increased substantially during follow-up, highlighting the clinical effectiveness of artificial pancreas therapy in routine clinical practice^13,14.

Strengths of the study

Several strengths should be noted in our study. First, this study provides real-world clinical data on artificial pancreas therapy in patients with type 1 diabetes, reflecting outcomes observed in routine clinical practice rather than controlled trial environments. Second, the study evaluated multiple cardiometabolic parameters, including lipid profile and proteinuria, in addition to glycemic control. These outcomes are less frequently reported in artificial pancreas studies and provide a broader understanding of the metabolic impact of automated insulin delivery systems. Third, the study included longitudinal follow-up up to 1 year, allowing the evaluation of sustained metabolic effects after initiation of artificial pancreas therapy.

Limitations of the study

This study has several limitations that should be considered when interpreting the results. First, the study was conducted in a single center with a relatively small sample size, which may limit the generalizability of the findings. Second, the retrospective design may introduce potential selection bias and limits the ability to establish causal relationships. Third, the study lacked a control group using conventional insulin therapy, which would have allowed a more direct comparison of treatment effects. Finally, although improvements in proteinuria and lipid parameters were observed, the follow-up duration was limited to 1 year, and longer-term studies are needed to evaluate the sustained renal and cardiovascular effects of artificial pancreas therapy.

Clinical implications

Artificial pancreas systems may provide benefits beyond glycemic control, potentially improving cardiometabolic risk factors and early markers of diabetic nephropathy. These findings support the broader clinical use of automated insulin delivery systems in the management of patients with type 1 diabetes.

Conclusion

Artificial pancreas systems significantly improve glycemic control in patients with type 1 diabetes and markedly increase the proportion of patients achieving HbA1c targets. Beyond improvements in glycemic parameters, the favorable changes observed in lipid profile and the significant reduction in proteinuria suggest that artificial pancreas therapy may confer beneficial cardiometabolic and renal effects. Considering that proteinuria typically increases with longer diabetes duration due to the progression of diabetic nephropathy, the reduction observed in our cohort highlights the potential role of improved glycemic regulation in modifying early markers of diabetic complications. These findings emphasize that artificial pancreas systems may provide benefits extending beyond glucose control and support their growing role in routine clinical practice for improving metabolic outcomes in patients with type 1 diabetes. Larger prospective studies with longer follow-up are warranted to confirm the long-term cardiometabolic and renal benefits of artificial pancreas therapy.