A possible advantage of continuous subcutaneous insulin infusion (CSII) therapy in type 1 diabetes is the decrease of glucose excursions with respect to multiple daily injections (MDI) of insulin. Few studies investigated this aspect of insulin treatment using continuous glucose monitoring systems (1).
The aim of our study was to evaluate whether CSII reduces glucose variability with respect to MDI, in patients with comparable A1C levels. The analysis was conducted in 36 type 1 diabetic patients, treated with CSII (15 male and 21 female subjects, aged 35 ± 12 years, duration of diabetes 16 ± 11 years, A1C 8.3 ± 1.5%) and 77 patients treated with MDI (35 male and 42 female subject, aged 40 ± 15 years, duration of diabetes 17 ± 12 years, A1C 8.5 ± 1.4%). All patients used insulin analogs. To evaluate glucose variability, interstitial glucose concentration was measured continuously over 72 h (by Continuous Glucose Monitoring System Gold, Medtronic).
Glucose variability was analyzed by calculating mean amplitude of glycemic excursions, coefficient of variation (CV), continuous overall net glycemic action (CONGA)2 and CONGA4, and mean of daily differences (2,–4). CONGA2, CONGA4, and CV were also evaluated in the temporal division between day (0700–2300 h) and night (2300–0700 h).
To analyze the effect of average blood glucose control on these measures of variability, patients were divided in quartiles based on distribution of A1C (1st A1C ≤7.5%, 2nd A1C >7.5 and ≤8.3%, 3rd A1C >8.3 and ≤9.2%, 4th A1C >9.2%). Data are expressed as means ± SD.
The differences between groups were compared using ANOVA with Bonferroni correction post test.
RESULTS
The baseline clinical data of the two groups were not significantly different. The patients in the CSII quartile with the best metabolic control (A1C ≤7.5%) showed a significantly lower glucose variability than the MDI group (CV: 0.36 ± 0.07 vs. 0.42 ± 0.09, P < 0.05; CV/day: 0.35 ± 0.08 vs. 0.42 ± 0.09, P < 0.05; CONGA4/day: 73.2 ± 30.1 vs. 98.8 ± 38.1 mg/dl, P < 0.05; and CONGA4/night: 64.5 ± 50.3 vs. 81.1 ± 29.8 mg/dl, P < 0.05). The indexes of glucose variability of the two groups were similar in the 2nd and 3rd quartile. In the quartile with the worst metabolic control (A1C >9.2%), the CSII group showed higher glucose variability than the MDI group (CONGA4/day: 127.7 ± 31.5 vs. 89.9 ± 11.1 mg/dl, P < 0.05; CONGA4/night: 115 ± 31.6 vs. 83 ± 41 mg/dl, P < 0.05).
CONCLUSIONS
In our population of type 1 diabetic patients, glucose variability was lower in the CSII group than in the MDI group only when glucose control was good (A1C ≤7.5%). This data confirms the result of a crossover study that compared CSII and MDI with glargine in 32 type 1 diabetic patients with good metabolic control (baseline A1C 7.6%) (5). The advantage of CSII with regard to glucose variability disappeared when A1C was >7.5%. Indeed, glucose variability was even worse in CSII-treated patients when A1C was >9.2%. This is likely the consequence of a more frequent use of correction boluses in patients of the CSII group than in those of the MDI group (3.2 ± 2.1 vs. 1.2 ± 1.3 correction boluses/day, P < 0.005). Our data suggest that CSII is particularly advantageous in decreasing glucose variability when strict metabolic control is difficult to maintain with MDI without disabling hypoglycemia or disturbing glucose fluctuations.
Acknowledgments
No potential conflicts of interest relevant to this article were reported.
