Background: We previously published that insulin pump initiation immediately after IV insulin therapy was associated with improved post-surgical glycemic outcomes compared to multiple daily injections (MDI) in pediatric patients following total pancreatectomy with islet autotransplantation (TPIAT) . We investigated metabolic outcomes of this population at 1 year post-TPIAT to assess if the improved outcomes in the early pump group were sustained over time.

Methods: We retrospectively reviewed 40 patients post-TPIAT previously studied at days post-surgery (14 used MDI and 26 used pump therapy immediately post-ICU, all discharged on pump therapy) . Data analyzed included: demographics, islet equivalents per kilogram (IEQ/kg) transplanted, exogenous insulin use, and baseline vs. 1 year (via mixed meal testing) HgA1C, fasting glucose, insulinogenic index and area under the curve (AUC) for insulin and c-peptide.

Results: More patients were off insulin at 1 year in the early pump group compared to the MDI group (43% vs. 14%, p=0.08) . Of all patients off insulin, 100% of the early pump users weaned off by 6 months post-TPIAT compared to 30% of the MDI users. Two known variables associated with favorable insulin outcomes, lower age and higher IEQ/kg were not significantly different between groups. Fasting glucose was lower in the early pump group compared to the MDI group (median 97.5 vs. 122 mg/dL, p=0.01) while AUC c-peptide was greater in early pump users at 1 year post-TPIAT but did not reach significance (median 65.3 vs. 49.5 ng/mL x minutes, p=0.07) . Other metabolic outcomes did not differ between groups.

Conclusion: Despite lower median age and higher IEQ/kg in the MDI group, the early pump group had a lower fasting glucose. Though not significant, the early pump group had a higher proportion off insulin and greater AUC c-peptide secretion. This study was limited by sample size. Early pump use may have long term benefits in post-TPIAT endogenous insulin secretion.


S.E.Tellez: None. L.Hornung: None. J.D.Courter: None. M.Abu-el-haija: None. J.D.Nathan: None. S.Lawson: None. D.A.Elder: None.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at