Background As an important part of diabetes treatment, insulin can be administered either by multiple daily injections (MDI) or continuous subcutaneous insulin infusion (CSII) via an insulin pump, but there are still controversies around the effectiveness of insulin administered by both methods in type 2 diabetes mellitus (T2DM) among older adults.
Objective To evaluate the effectiveness of CSII in T2DM in older adults using meta-analysis, and to test the validity of the analysis results using trial sequential analysis (TSA) .
Methods Cochrane Library, PubMed, Embase, Medline, Scopus, Web of Science, CNKI, Wanfang Data, CQVIP and SinoMed were retrieved for randomised controlled trials (RCTs) from inception to December 2021 regarding the effectiveness of CSII via an external insulin pump compared with that of MDI of insulin for older adults with T2DM. The primary outcomes included fasting plasma glucose (FPG), 2-hour postprandial plasma glucose (2 hPG), HbA1c and incidence of hypoglycaemia. The secondary outcomes included the mean amplitude of glycemic excursions (MAGE), total daily dose of insulin, and the time to achieve glycemic target. Two researchers independently carried out literature enrolment, quality evaluation and data extraction. RevMan 5.3 was adopted for meta-analysis. TSA v0.9 (Copenhagen Trial Unit) was adopted for TSA.
Results Sixteen RCTs were included in total. Meta-analysis demonstrated that CSII via an external insulin pump outperformed MDI of insulin in improving FPG〔MD=-0.82, 95%CI (-1.09, -0.54), P<0.05〕, 2 hPG〔MD=-0.76, 95%CI (-1.39, -0.14), P<0.05〕, HbA1c〔SMD=-1.23, 95%CI (-2.23, -0.23), P<0.05〕, incidence of severe hypoglycaemia〔RD=-0.10, 95%CI (-0.17, -0.03), P<0.05〕, total daily dose of insulin〔MD=-9.63, 95%CI (-12.35, -6.92), P<0.05〕, and MAGE〔MD=-1.19, 95%CI (-1.40, -0.97), P<0.05〕. Better improved primary outcomes by CSII via an external insulin pump were also found by a TSA, namely more significantly reduced FPG, 2 hPG, and HbA1c.
Conclusion CSII via an external insulin pump has been shown to be more effective in improving glycaemic control, reducing the incidence of hypoglycaemia and MAGE among older patients with T2DM.
