[Previous Months][Date Index][Thread Index][Join - Register][Login]
[Message Prev][Message Next][Thread Prev][Thread Next]

[IP] Nightime CSII Revisited (Editorial Diabetes Care)

VOLUME 23 NUMBER 5 - Page  571 


Nighttime Continuous Subcutaneous Insulin Infusion Revisited

A strategy for improving insulin delivery

It has been 20 years since the demonstration that intensive insulin therapy 
(IIT) can result in near-normoglycemic levels (1–4). Despite this evidence, 
IIT has been used infrequently in children and adolescents. The majority of 
the early studies of type 1 diabetic children and adolescents used continuous 
subcutaneous insulin infusion (CSII) as the mode of therapy (5–10). It has 
also been shown that, in adolescents, a regimen of premeal regular insulin 
plus NPH at bedtime was almost as effective as CSII (11,12). Another strategy 
that combined nighttime CSII with daytime multiple daily insulin injections 
(MDI) was also shown to be as successful as CSII or MDI alone in achieving 
near normal HbA1c levels (13). Despite these encouraging results, widespread 
efforts to improve blood glucose control were put on hold for a number of 
years, because clinicians and patients were awaiting the results of the 
Diabetes Control and Complications Trial (DCCT). In 1993, the DCCT group 
published its first manuscript, which showed that IIT with MDI and/or CSII 
was much more effective than conventional therapy (CT) in decreasing HbA1c 
levels (14). Even though near normoglycemia could not be maintained over 
prolonged periods of time with either method, the improvement of HbA1c levels 
observed during the 7.4 years of the trial was good enough to substantially 
reduce the risk of progression of diabetic microangiopathy. These results, 
combined with the availability of smaller more reliable insulin pumps, 
insulin pens, and infusion sets, encouraged diabetologists around the world 
to renew their efforts to use MDI and CSII in a larger number of patients of 
all ages (15–17).

In this issue, Kaufman et al. (18) report the results of a randomized 
cross-over trial that compared the use of nighttime insulin pump therapy 
beginning at dinner (nighttime CSII) combined with a mixture of NPH plus 
lispro before breakfast with a regimen of 3 injections/day in a group of 10 
children <10 years of age with a history of nighttime hypo- or hyperglycemia. 
Nighttime CSII resulted in significantly lower mean average, breakfast, and 
bedtime blood glucose levels; lower mean fructosamine levels; and an increase 
in the percentage of blood glucose values within the target range of 70–150 
mg/dl. This is an important study for pediatric diabetologists. There have 
been very few reports on the feasibility of administering CSII to children 
(7,10). In particular, this is the first study that has used nighttime CSII 
only in children. Nighttime CSII combined with daytime MDI was first 
implemented 20 years ago in adolescents and young adults for up to 10 months 
of therapy (13). This strategy was devised when it was noted that individual 
fasting blood glucose values were more variable with MDI than with CSII (12). 
Recently, it has been shown that, in young adults, nighttime CSII can improve 
counterregulatory responses and warning symptoms in patients with 
hypoglycemia unawareness (19). Kaufman et al. concluded that the nighttime 
CSII group had less hypoglycemia, although this conclusion is not very 
evident given the data shown. Because the study was very short, it remains to 
be seen if nighttime CSII is clearly superior to thrice-daily injections over 
the long term. It is difficult to exclude the possibility that increased 
parental involvement and adherence led to the improvement of blood glucose 
control in the nighttime CSII group; parental concerns about their children's 
safety may have been heightened by the use of a new treatment. One wonders if 
increased familiarity with the pump over time could lead to lapses in 
supervision. A mechanical problem that goes unnoticed could result in severe 
hyperglycemia, because the use of lispro provides shorter insulin coverage 
than the use of regular insulin. In fact, ketoacidosis was a fairly frequent 
complication reported in the early studies of young patients, even when 
regular insulin was used. In the DCCT, adolescent and adult subjects on CSII 
had significantly lower HbA1c values than subjects on MDI (6.8 vs. 7%), and 
they also had slightly higher frequencies of diabetic ketoacidosis and 
hypoglycemia with coma or seizure than subjects on MDI (20). A recent report 
of a 2-year prospective study of children and adolescents aged 1–18 years who 
were treated with MDI (21) showed levels of HbA1c comparable with those of 
the adolescents in the DCCT and with those of the children in Kaufman et 
al.'s study; however, these patients showed lower frequencies of hypoglycemia 
(1.17–1.43 events/patient-year) than those of patients observed in the DCCT. 
Although it is difficult to extrapolate what the long-term frequency of 
hypoglycemia would have been in Kaufman et al.'s study, the results show the 
potential of nighttime CSII to improve blood glucose control and to, perhaps, 
reduce the frequency of nocturnal hypoglycemia in children, a complication 
that has been shown to be associated with abnormal counterregulation (22) and 
possibly with an increased risk of memory impairment (23). The data obtained 
by Kanc et al. (19) from a study of adult patients who used nighttime CSII 
are very encouraging, and there are no obvious reasons why this form of 
therapy will not be as effective in young children.

Several prospective studies have shown that, in children and adolescents, 
severe hypoglycemia is a very frequent event with both CT and IIT; ~50% of 
the total daily events occur nocturnally (24,25). During sleep, the autonomic 
symptoms may not always wake the patient, and unrecognized nocturnal 
hypoglycemia may result in hypoglycemia unawareness, which would predispose 
the patient to hypoglycemia during the day (26). Excessive counterregulation, 
if present, and/or food intake may result in hyperglycemia, which may induce 
overinsulinization and further hypoglycemia. The incidence of nocturnal 
hypoglycemia during nighttime CSII could be decreased, as previously 
demonstrated in a study on 20 patients on CSII (27). Patients were randomized 
to receive their usual bedtime snack or to receive extra food 1.5 h after the 
bedtime snack if blood glucose levels were <120 mg/dl. Results showed that 
the probability of nighttime hypoglycemia and morning hyperglycemia could be 
minimized to 20% at bedtime blood glucose values of ~140 mg/dl. Because there 
is currently no clear evidence of the long-term effects of hypoglycemia on 
the cognitive function of young children, avoidance of hypoglycemia in this 
age-group is paramount. In addition to the potential risk of cognitive 
impairment, severe hypoglycemia may induce adverse psychological reactions in 
patients and caretakers. A common response to such reactions is to maintain 
high blood glucose levels, particularly at bedtime. Thus, repeated episodes 
of hypoglycemia may induce poor blood glucose control for long periods of 

Nighttime CSII has several advantages. It provides the delivery of insulin 
with more physiological pharmacokinetics. It results in a square-wave insulin 
profile (28), avoiding the peak and waning effect of NPH that causes large 
blood glucose nadirs in the middle of the night and hyperglycemia in the 
early morning. With CSII, the infusion rate can be adjusted according to the 
patient's requirements. In young patients, nightttime CSII offers a flexible 
lifestyle because the pump can be connected before or after dinner, depending 
on the child's activities, and thereby avoids the inconvenience of carrying 
the pump during the day. During holidays and sleep-overs, nighttime CSII can 
be replaced by NPH. The disadvantages of nighttime CSII include the time 
spent in the preparation and the technical skills required to safely handle 
the infusion and costs incurred by families, which, for uninsured patients, 
could be prohibitive. It remains to be seen if the use of lispro insulin will 
not result in an increased ketosis or ketoacidosis in the case of pump 

Nighttime CSII in children is a simple strategy that can improve glycemic 
control as long as it is administered under careful parental supervision. 
However, it should be emphasized that the success of diabetes therapy in this 
population does not depend only on the number of insulin injections, the use 
of CSII, the frequency of blood glucose monitoring, or the access to a 
diabetes specialist. Both the presence of a stable family environment that is 
capable of providing support to the patient and access to a specialized 
multidisciplinary health care team that is available for guidance, education, 
and maintaining motivation also play a role in determining the outcome of 
diabetes treatment.

Alicia Schiffrin, MD
for HELP or to subscribe/unsubscribe, contact: HELP@insulin-pumpers.org
send a DONATION http://www.Insulin-Pumpers.org/donate.shtml