JOHNNY LUDVIGSSON, MD dwelling time of the ... Thus, although trying to reduce injection pain is an im- ... Pain sensitivity might cause patients to take fewer ...
R E P O R T
Metabolic Control Is Not Altered When Using Indwelling Catheters for Insulin Inactions S. RAGNAR HANAS, MD JOHNNY LUDVIGSSON, MD
OBJECTIVE — To determine if the use of indwelling catheters for insulin injections affects the long- and short-term metabolic control of insulin-dependent diabetes mellitus (IDDM). RESEARCH DESIGN A N D M E T H O D S — Sixteen children and adolescents 9-20 years of age were included in a randomized 10-week crossover study using indwelling catheters (Insuflon, Pharma-Plast, Lynge, Denmark; CHRONIMED, Minnetonka, Minnesota) for insulin injections. Their diabetes duration was 7.5 ± 3.3 years (range 2-14), and they used multiple injection therapy with 4-5 doses/day. C-peptide was 0.15 nM fasting and/or >0.30 nM postprandial, radioimmunoassay, according to L. Heding) and one because she did not complete the laboratory tests. The remaining 16 patients, 9 boys and 7 girls, had a mean age of 14.9 ± 3 . 1 years (range 9-20). Their diabetes duration was
DIABETES CARE, VOLUME 17,
NUMBER 7, JULY
1994
Hanas and Ludvigsson
Rubber membrane Insulin pen car syringe
Hard plastic ^ tube "H"
Soft teflon catheter
Subcutaneous tissue Figure 1—The Insuflon Cannula (Pharma-Plast) is an indwelling catheter consisting of a plastic tube
with a rubber membrane and a Teflon catheter. The insulin is deposited in the subcutaneous tissue. Both insulin pens and syringes can be used for injections. It is replaced at home by the patient or the parents. The average indwelling time is 4-5 days.
7.5 ± 3.3 years (range 2-14). All patients used multiple injections with 3-4 doses of short-acting insulin (Actrapid, NovoNordisk, Bagsvaerd, Denmark) before meals and intermediate NPH (Insulatard, Novo-Nordisk, n = 11) or long-acting lente (Ultratard, Novo-Nordisk, n = 5) insulin at bedtime. Because lente and soluble (shortacting) insulins do not mix well from a pharmacokinetic point of view (3), only the patients using NPH insulin used Insuflon for their bedtime injection. After a 10-week run-in period with their regular therapy, the patients were randomly allocated to use either indwelling catheters or normal injections. After 10 weeks they changed injection method, being their own controls. At the end of each treatment period, we measured HbAlc (standard high-performance liquid chromatography method, reference 3.4-6.0%), and the patients were hospitalized to measure 24-h profiles of blood glucose and free insulin (radioimmunoassay, according to L. Heding), insulin antibodies (PEG method), C-peptide in urine and blood, glucosuria, and ketonuria. Body weight, insulin antibodies, and insulin doses were also recorded. The patients checked for glucosuria at home three times daily twice a week and determined blood glucose seven times daily every fortnight throughout the
DIABETES CARE, VOLUME 17,
NUMBER 7, JULY
1994
study (including the run-in period). The protocol was approved by the ethical committee of the Faculty of Health Sciences at the University of Linkoping, Sweden.
Statistical analysis Two-sided Student's t test, Wilcoxon's rank-sum test and the Mann-Whitney U test were used. Statistical calculations were made with the MULREG statistical program.
RESULTS — There were no significant differences in HbA ]c , weight, insulin doses, or levels of insulin antibodies (Table 1) with or without Insuflon. Twentyfour-hour profiles of blood glucose and free insulin (Fig. 2) did not differ significantly; glucosuria and ketonuria samples also did not differ during 24 h at the hospital. Home testing for blood glucose and glucosuria showed no significant differences. The frequency of low blood glucose readings (
