HbA1c values and hospital admissions in children and adolescents receiving continuous subcutaneous insulin infusion therapy.

AIMS: To assess HbA1c values and hospitalization rates before, during and after continuous subcutaneous insulin infusion (CSII) therapy. METHODS: Demographic and hospitalization data were extracted from 161 individuals with Type 1 diabetes who received continuous subcutaneous insulin infusion (CSII) therapy between 2002 and 2013 at the Leeds Children and Young People's Diabetes Service for those aged < 20 years. The median (range) age at CSII start was 11.9 (1.1-17.6) years. The median (range) follow-up time was 2.3 (0-8.1) years. Random intercept models were used to compare HbA1c values before and during CSII initiation (and after CSII for those who discontinued it). Hospitalization rates were calculated for diabetic ketoacidosis and severe hypoglycaemia. RESULTS: The mean HbA1c concentration decreased by 7 mmol/mol [95% CI 6-8; 0.6% (95% CI 0.5-0.7%)]. For the discontinued group (n=30), mean HbA1c decreased by 5 mmol/mol [95% CI 2-8; 0.4% (95% CI 0.2-0.7%)]. HbA1c returned to pre-CSII start levels at the end of this therapy. Diabetic ketoacidosis admissions increased threefold during CSII compared with before CSII start [2.2 per 100 person-years (95% CI 1.3 to 3.6) vs 7.4 per 100 person-years (95% CI 5.1 to 10.8)] and was highest during the first year of CSII. No difference in severe hypoglycaemia incidence rate was found during CSII compared with the pre-CSII period. CONCLUSIONS: Despite significant reductions in HbA1c levels for individuals treated with CSII, improvements are needed to reduce diabetic ketoacidosis hospitalizations for those new to the therapy.

Mortality and acute complications in children and young adults diagnosed with Type 1 diabetes in Yorkshire, UK: a cohort study.

AIMS: To examine all-cause and cause-specific mortality in a population-based cohort of people with early and late onset of Type 1 diabetes. METHODS: The Yorkshire Register of Diabetes in Children and Young People includes individuals with early (0-14 years) and late (15-29 years) Type 1 diabetes onset, diagnosed between 1978 and 2013. This register was linked to death certification data from the Office for National Statistics to calculate standardized mortality ratios, cumulative mortality curves using Kaplan-Meier survival estimates, and Cox regression modelling. Ethnicity was derived using Onomap. Deprivation status was classified using the Townsend index. The underlying cause of death in each case was clinically verified. RESULTS: There were 229 deaths in 5498 individuals with 100 959 person-years of follow-up. The overall standardized mortality ratio was 4.3 (95% CI 3.8 to 4.9). There were no significant differences in standardized mortality ratios according to age of onset, sex or deprivation status. The standardized mortality ratios were significantly higher for people of white ethnic origin [8.1 (95% CI 6.9 to 9.4)] than for those of South-Asian ethnic origin [3.4 (95% CI 1.7 to 6.4)]. The mortality risk was lower in those diagnosed in later years (2002 to 2013 for the early-onset and 2006 to 2013 for the late-onset group) compared with earlier years (1991 to 1997 for the early-onset and 1991 to 1997 for the late-onset group) for both onset groups [hazard ratio 0.13 (95% CI 0.05 to 0.33) vs 0.24 (95% CI 0.07 to 0.81)]. Mortality risk improved over time for chronic complications in the early-onset group only, but there was no improvement in either onset group with regard to acute complications. CONCLUSIONS: An excess of deaths in the population with Type 1 diabetes remains. Although the all-cause mortality risk has fallen over time, no improvement has been found in the mortality risk associated with acute complications.