Flexible Intensive Versus Conventional Insulin Therapy in Insulin-Naive Adults With Type 2 Diabetes: An open-label, randomized, controlled, crossover clinical trial of metabolic control and patient preference

We tested the hypothesis that FIT and CIT in insulin-naive adults with type 2 diabetes are equally effective in regard to metabolic outcomes. We hypothesized that younger participants, in employment, would prefer FIT.

The trial was designed as a clinical, prospective, randomized, open-label, single-center, crossover study. The primary end point was glycosylated hemoglobin A_1c (GHb); secondary end points were mild and severe hypoglycemia, insulin dosage, blood pressure, BMI, and therapy preference. Participants started insulin therapy either with CIT (group A) or FIT (group B), randomly. Individual insulin dosage and carbohydrate intake were determined over a 4-week run-in period with weekly visits. All visits were the same in both groups and were held in the study center. In CIT, daily blood glucose self-control was performed before breakfast and dinner; in FIT, this was done before main meals. Oral antidiabetes drugs were not permitted. The participants completed an outpatient Diabetes Treatment and Teaching Program with five lessons (90–120 min) during run-in. The run-in was followed by an 8-week study sequence until crossover. At crossover, participants were given one structured teaching session for refreshing and switched from CIT to FIT or from FIT to CIT. After a 1-week run-in period for insulin dose adjustment, participants completed the second 8-week study sequence. At trial end, therapeutic preference was investigated with a structured interview.

Eligible participants who failed therapeutic goals under oral antidiabetes therapy were consecutively recruited (outpatient clinic, Friedrich-Schiller-University). Exclusion criteria were not having type 2 diabetes, diabetes duration <2years, not insulin naive, ineffective oral antidiabetes therapy <3 months, GHb <7 or >11%, and age <40 or >65 years.

GHb was measured using high-performance liquid chromatography (normal range 1.6–5.9%; mean 5.2 ± 0.33%; TOSOH-Glykohämoglobin-Analyzer-HLC-723-GHbV; Tosoh, Tokyo, Japan). Mild hypoglycemia was defined as symptomatic neuroglycopenia or blood glucose readings <3.3 mmol/l. Severe hypoglycemia required intravenous glucose or subcutaneous/intramuscular glucagon injection.

To have a 90% chance of detecting as significant (at the two-sided 5% level), a 0.5% difference in GHb between the two groups with an assumed SD of 0.8%, 38 participants were required. Intent-to-treat analysis was carried out according to a preestablished analysis plan. Means and SDs were calculated. The t test for paired and nonpaired samples was used where appropriate. For statistical analysis of associations of GHb and participant's characteristics, linear mixed-effects models were used. P < 0.05 was regarded as statistically significant. Statistical analysis was performed with SPSS (version 15; SPSS, Chicago, IL).

The study was approved by the local ethics committee and was performed according to the principles of the Declaration of Helsinki. Written informed consent was obtained before participants took part. The trial was registered with clinicaltrials.gov (NCT00440284).

Baseline clinical data between groups were not significantly different (Table 1, group A [first CIT, n = 20] vs. group B [first FIT, n = 19]): age 56.6 ± 7.3 vs. 54.7 ± 6.6 years, diabetes duration 7.5 ± 4 vs. 8.2 ± 4.2 years, total cholesterol 5.3 ± 1.1 vs. 5.3 ± 1.1 mmol/l, creatinine 82 ± 18.4 vs. 77 ± 16.3 μmol/l, peripheral neuropathy 9 vs. 4 participants, and number of participants employed 12 vs. 15, respectively.

All participants completed both study sequences except for one person who refused to switch. After initiation of insulin therapy, GHb declined from 8.9 ± 1.7 to 7.3 ± 0.9% (P < 0.0001), and BMI increased from 29.4 ± 4.4 to 30 ± 4.2 kg/m² (P < 0.01). Blood pressure and lipid profiles remained unchanged. There was no significant difference between CIT and FIT regarding GHb, BMI, blood pressue, insulin dosage, and hypoglycemia (Table 1). In linear mixed-effects models, GHb after CIT and GHb after FIT were not associated with age, sex, diabetes duration, initial GHb, blood pressure, lipid profiles, comorbidity, and occupation.

Twenty participants preferred to continue CIT, whereas 18 opted for FIT. Reasons to opt for FIT were therapy flexibility (n = 9), easier therapy (n = 1), and metabolic control (n = 1). Arguments in favor of CIT were easier therapy (n = 8), fewer injections (n = 8), and metabolic control (n = 3). In a binary logistic regression model, 88% of therapy decisions were explained by the last therapy option. The model was not improved when additional variables like age, sex, diabetes duration, GHb after CIT or FIT, initial GHb, blood pressure, lipid profiles, and occupation were added.

Initiation of insulin therapy in type 2 diabetes was safe and effective. Metabolic control improved during the first study sequence in FIT and CIT but did not further improve in the second sequence. Participants did not reach GHb levels below 7%. After having practiced CIT and FIT for 8 weeks each, participants preferred their last therapy. This indicates that clinical advantages of CIT or FIT were of minor importance for participants irrespective of age or whether they were employed. Interestingly, in contrast to patients with type 1 diabetes, gaining more dietary freedom seems not to be a prevalent motive in type 2 diabetes (6).

There are several limitations to consider. The sample size of this randomized controlled trial was too small and the study period too short to detect minor differences. Carry-over effects of the crossover design can reduce the effect of the second trial sequence. The primary outcome measure of this trial (GHb) was a surrogate parameter. Comparison of side effects may be limited by different definitions of mild and severe hypoglycemia. Participants were young and had early manifestation of type 2 diabetes compared with the general population (9). Older patients with impaired cognitive function might be unable to practice effective diabetes self-management using insulin therapy or might be disinterested in the clinical advantages of FIT and CIT (10).