BOSTON -- Insulin pumps may improve but often don't control glycemic levels in type 1 diabetes, a retrospective U.K. study showed.
Fully 38% of patients maintained a hemoglobin A1c over 8.5% on continuous subcutaneous insulin infusion and 11% were over 10% A1c, Lalantha Leelarathna, MBBS, PhD, of the University of Manchester, England, and colleagues found.
That was despite an average 0.6% A1c improvement after staring pump therapy, with greater declines in patients starting at higher levels, they reported here at the American Diabetes Association meeting.
To see such high rates of uncontrolled HbA1c levels in adults was surprising, commentedClifford J. Bailey, PhD, of Aston University in Birmingham, England.
"I would have expected only a few patients to be over 8.5% unless they were adolescents," he told MedPage Today.
Leelarathna agreed and suggested that adherence might have been one explanation and that limited use with continuous glucose monitoring, which is not routinely funded in the U.K. national healthcare system, in the cohort could have been another factor.
"Management of type 1 diabetes is still very challenging," he told MedPage Today. "Even with pump therapy many patients don't achieve targets. This is real life."
"You always hope the pump will improve glycemic control, but it depends on whether the patient does the necessary things," commented Cynthia Young, RN, of Maine Medical Center in Portland, Maine.
"A pump is only as smart as the user," agreed Erica Page, RN, CDE, of Mercy Hospital also in Portland, Maine.
The study included all 442 patients on a pump for at least 12 months at a single tertiary care center and with glycated hemoglobin averaged over a 30 month period through November 2014.
The cohort ranged in age from 31 to 50 and a median 20 years of experience with the disease at initiation of pump therapy.
Age, duration of diabetes, and length of follow-up were not significant factors in change in glycemic control after getting the insulin pump.
Overall, the proportion of patients in control of their diabetes with an A1c under 7.5% improved from 19% before getting the insulin pump to 32% afterward, while the proportion in the over 8.5% HbA1c group declined from 52% to 40%.
The brand of pump also was not significantly linked to improvement, although traditional pumps were more often associated with at least 0.5% improvement in A1c compared with the patch-style OmniPod.
Limitations included the retrospective single-center design, lack of data on hypoglycemia, and possible selection bias.
Generalizability to other practice settings, like U.S. centers, wasn't clear, Leelarathna cautioned.
"Use of pumps is less than in the U.S.," Bailey noted. "It is reserved for more difficult to treat individuals, usually ones that are uncontrolled in both directions [hypoglycemia and hyperglycemia], but our major concern is to avoid severe hypos."