Use of dapagliflozin as an add-on to insulin therapy in patients with suboptimally controlled type 2 diabetes

Dear Editor,

Type 2 diabetes is a chronic and progressive disease that necessitates treatment with oral hypoglycaemic agents (OHAs), which are usually dependent on either insulin secretion or improvement in insulin resistance and sensitivity. In the later stages, OHAs fail to achieve satisfactory glycaemic targets and patients require insulin therapy, which requires frequent titration. Although up-titrating insulin is a reasonable treatment option, it may not always lead to optimal glycaemic control and the glycaemic effectiveness of high doses of insulin is not well established in some patients.^1,2 In addition, higher doses of insulin may lead to unwanted weight gain and fluid retention, and pose an increased risk of hypoglycaemia.^3,4 This may limit the long-term practicability and acceptability of increasing insulin doses over time.

Dapagliflozin, a competitive and selective inhibitor of sodium-glucose co-transporter 2 (SGLT2), induces urinary excretion of excessive blood glucose by blocking the re-uptake of glucose in the kidney, resulting in improved glycaemic control. ⁵ The magnitude of the glucose-lowering effect correlates with plasma glucose concentrations and is independent of β cell function and insulin sensitivity. ⁶ As dapagliflozin acts independently of insulin, it seems reasonable to co-administer it with insulin regimes to improve glycaemic control.

We carried out a retrospective analysis of 36 patients who attended our diabetes clinics from June 2015 to December 2016. These patients had type 2 diabetes with suboptimal control (HbA1c > 58 mmol/mol; 7.5%) and were on insulin with doses >30 units/day (either as basal or multiple daily injections) in combination with up two OHAs (metformin, sulfonylureas or gliptins). They were started on dapagliflozin 10 mg once daily by the attending clinician during their initial visit. Their demographic and biochemical data were collected at baseline and at 6–9 months, and relevant nursing records were reviewed to study the contacts these patients had with diabetes specialist nurses either via telephone or direct visit to the diabetes centre for titration of insulin doses as per blood glucose profile. These patients did not have any other titration in their OHA doses.

In our patients, the mean duration of insulin therapy was approximately 6 years with mean daily insulin dose of 93.7 ± 63.2 units. With the introduction of dapagliflozin, the mean HbA1c dropped by −1.8% [95% confidence interval (CI) −1.5 to −2.1%; p < 0.0001] and body weight showed a mean change of −2.8 kg (95% CI −1.8 to −3.8 kg; p = 0.001) after 6–9 months of treatment. In addition, the mean daily insulin dose fell by −10.8 units (95% CI 0 to −21.6 units; p = 0.008). Individually, 72.2% of patients had a reduction in their daily insulin dose, with a mean change of −19.9 units (95% CI −10.9 to −28.9 units; p < 0.0001), whereas 27.8% had their insulin dose increased by a mean of 9.4 units (95% CI 9.1−9.7 units; p = 0.002). We did not note any major changes in blood pressure, kidney function, lipid profile or urine albumin excretion. Only five patients had experienced a single or more episode of hypoglycaemia and three patients had single episodes of genitourinary infection that responded to routine antibiotic treatment.

SGLT2 inhibitors like dapagliflozin have added a new leaf in the treatment paradigm for type 2 diabetes owing to their insulin-independent mechanism of action. Our observational study confirmed that dapagliflozin, when added to pre-existing background insulin therapy, effectively reduces HbA1c and body weight in a difficult-to-treat population who are older, have long-duration of diabetes with poor glycaemic control and have limited treatment options left as they are already receiving large doses of insulin. Dapagliflozin reduces body weight by fluid loss (by osmotic diuresis) and by calorie loss in the form of urinary glucose, but actual reduction in fat mass has also been reported. ⁷ The reduction in insulin dose may be a reflection of a dose-sparing effect or an improvement in insulin sensitivity,^8,9 which have also been shown by other authors in their long-term studies using different doses of dapagliflozin (2.5 mg/5 mg/10 mg).^6,10 Some studies have also demonstrated clinically meaningful reduction in systolic and diastolic blood pressure, lipid profile but worsening of microalbuminuria.^6,11

We therefore conclude that dapagliflozin is a reliable option as an add-on to insulin treatment in suboptimally controlled type 2 diabetes as it improves HbA1c, lowers weight, and also reduces insulin dosing and requirement without causing additional risk of hypoglycaemia or other complications.