Short-Term Predictors of Near-Normoglycemic Remission inAfrican-American Subjects with Ketosis Prone Diabetes Mellitus Open Access

Smiley, Dawn (2008)

Permanent URL:


Purpose: Patients that present with diabetic ketoacidosis (DKA) but have clinical features of type 2 diabetes are referred to as having ketosis-prone type 2 diabetes (KPDM) and most patients are able to discontinue insulin (near-normoglycemic remission) within 12 weeks of therapy. Most patients, however, experience a hyperglycemic relapse within a year. The aims of this study were 1) to identify clinical, metabolic, and immunogenetic variables that are markers of remission and 2) to correlate measures of pancreatic β-cell function and insulin sensitivity with the ability to achieve remission.

Null Hypotheses: Measured variables, measures of β-cell function nor measures of insulin sensitivity correlate with the ability to achieve remission.

Design and Methods. In a prospective approach, obese, African-American subjects with new onset diabetes presenting with either DKA (glucose ≥ 250 mg/dL, ph < 7.3 and bicarbonate ≤ 18 mmol/L) or severe hyperglycemia (HG, glucose ≥ 400mg/dL) were enrolled in the study and treated with pre-mixed insulin for up to 12 weeks. Within a week of diagnosis, subjects underwent glutamic acid decarboxylase (GAD) antibody testing, β-cell function (glucagon stimulation test, GST) and insulin sensitivity assessments (IV glucose tolerance test, IVGTT). At 1 week following discontinuation of insulin or at 12 weeks of insulin therapy (no remission), β-cell function and insulin sensitivity assessments were repeated.

Results. Of 39 enrolled subjects, 17 had DKA (12M/5F) and 22 had HG (12M/10F). Nearly 70% went into remission within 12 weeks. Except metabolic acidosis in the DKA cohort, there were no baseline clinical differences between groups. Weight loss at presentation trended with the ability to attain remission. There were no group differences in β-cell function or insulin sensitivity at presentation; however, subjects that achieved remission at ≤ 12 weeks of insulin therapy had a greater acute insulin response to glucagon compared to subjects that failed remission.

Our preliminary data did not find significant differences between the study cohorts at presentation nor establish a consistent prognostic model for remission. The initial data suggests that longitudinal studies may be more appropriate to establish a model to predict outcomes, particularly as it relates to response to remission and medical treatment.

Table of Contents

Table of Contents

Introduction...1 Background and Significance Methods...2 Methods

Specific Aim...6 Null Hypothesis...6 Study Design...6 Patient Selection...8 Study Protocol...9

Experiment A and Experiment B...9

Detailed Methods...10 Sample Size Determination and Power...12 Statistical Analysis...13


Study Population...14 Autoantibody Testing...15 Assessment of β-cell Function by Glucagon Stimulation Testing...15 Assessment of Insulin Sensitivity by FSIVGTT...16 Variable Differences between Subjects that Achieve Remission versus Non-remission...19



Conclusions...24 References...25 Tables

Table 1. Clinical Characteristics Presentation...28

Table 2. Clinical Characteristics of Subjects at ≤ 12 weeks Follow-up...29 Table 3. Clinical Characteristics in Subjects with Remission versus Subjects without Remission...30 Table 4. Glucagon Stimulation Test C-peptide Response @ 3 minutes at Presentation...31 Table 5. Indices of Insulin Sensitivity and β-cell insulin Secretion at Presentation...32 Table 6. Minimal Model Indices @ presentation based on Group...32 Table 7. Minimal Model Indices @ presentation based on Outcome...33 Table 8. Minimal Model Indices @ 12 weeks based on Group...33 Table 9. Minimal Model Indices @ presentation based on Outcome...34 Table 10. Putative Factors for Remission in Subjects with New Onset KPDM...34 Table 11. Dichotomization of Continuous Predictors...35 Table 12. Frequency of Predictors for Remission versus Non-remission...35 Table 13. Measure of Association between Potential Predictors and Outcome...36


Figure 1. Plasma C-peptide levels before and after stimulation with IV glucagon (1 mg) in obese subjects with DKA and with HG...31

About this Dissertation

Rights statement
  • Permission granted by the author to include this thesis or dissertation in this repository. All rights reserved by the author. Please contact the author for information regarding the reproduction and use of this thesis or dissertation.
  • English
Research field
Committee Chair / Thesis Advisor
Committee Members
Last modified

Primary PDF

Supplemental Files