A Single-Center Long-Term Continuous Subcutaneous Insulin Infusion (CSII) Experience: Higher Fractional Use Is Associated With Less Diabetes Complications

In the past decades, type 1 diabetes (T1DM) intensive therapy has reduced the incidence of cardiovascular diseases and overall and cardiovascular mortality.^1,2

Continuous subcutaneous insulin infusion (CSII) is more effective than multiple daily injections (MDI) ³ achieving metabolic targets and is associated with lower cardiovascular mortality, ⁴ but there are few data published on mortality and complications in long-term pump users.

In this study we report mortality, cardiovascular events, end stage retinopathy and renal failure rate and CSII attrition (rate, causes) in a cohort of 141 patients who started CSII before 2005 with over ten years of follow-up.

The indications to CSII were poor metabolic control (70%) and/or recurrent severe hypoglycemia (30%). Mean age at diabetes onset was 23 ± 14 while the mean age at CSII start was 38 ± 14 years. Patients were followed for a median follow-up of 13 (11–16) years (mean 14 ± 4 years).

At last follow-up, mean HbA1c was 75.6 ± 13 mmol/mol (36-119 mmol/mol).

In regard to mortality, 5 (3.5%) deaths occurred: 1 due to kidney transplantation complication, 1 to HIV infection, 2 to cancer, and 1 to heart failure. A Kaplan-Meier survival curve describes 95.8% (90.0–98.2%) survival (IC) at 15 years; no significant gender difference was found.

In regard to complications, 15 patients (11%) experienced at least one complication, 10 (7%) had a macrovascular event, less than in the Sweden Registry (10.5% after 7 years), ² 5 (4.2%) a microvascular one.

The incidence of complications was related to the time percentage (years of CSII) of the overall duration of diabetes (fractional use, FU ). Patients were grouped in quartiles according to FU (I: <39%, II: 39-55%, III: 55-75%, IV: >75%). ANOVA was used to check differences among quartiles. A Cox univariate regression model was used for complication analysis. FU analysis demonstrated that the more is the time spent in CSII, the more is the freedom from complications (P = .026) (Figure 1).

Figure 1.

Freedom from complications according to quartiles of time spent in CSII.

The quartile with the lowest FU of CSII was excluded from the analysis for its longest duration of disease and oldest age at therapy start. The other three quartiles shared similar CSII starting age and recent HbA1c levels. Insulin pump users who spent a longer time (>75% of their disease) in CSII had less events while patients treated for a shorter time (<55%) are 1.6-times more likely to experience a complication (HR = 1.64, IC [1.33-20.00], P < .05). These results are consistent with observations on children and adolescents. ⁴

For CSII attrition, 11 (7.8%) patients stopped CSII after 13 ± 5 years: 8 for psychological-social problems, 2 for loss of self-management ability (post-stroke and dementia) and 1 for diabetes remission (following successful pancreas transplantation). The attrition rate was 7.8% versus 10% as reported in literature. ⁵ This was likely due to the patients selection and high motivation and the health care team experience utilizing a structured educational programs. ⁶

In conclusion, long-term CSII therapy is associated with low mortality and complication rate, high adherence and good glycemic control. Higher FU of CSII is associated with less complications.

Insulin pump therapy represents an efficient tool for T1DM for long-term management in adults and seem effective to prevent diabetes related complications.