P23 Evolution of inpatients with new hyperglycemia (ambiglycemia)

Abstract

Introduction: Hyperglycemia in the inpatients is associated to poor clinical outcome, although there is controversy about the magnitude of this effect on diabetic or not diabetic patients.

Objectives: To differentiate two populations of inpatients with new hyperglycemia: (A)- previously unknown Diabetes Mellitus or debut, and (B)-unexpected Hyperglycemia; and to explore the associations between metabolic variables at admission, and clinical outcomes. We propose the term ambiglycemia (AmbiG) for this ambiguous glycemia (severity vs stress), with the same value but different meaning.

Methods: Prospective cohort study with 2 years follow-up, watching the evolution of inpatients with new hyperglycemia (>140mg/dl confirmed), comparing two groups according to basal HbA1c at hospitalization (A) ≥6.5% or (B) <6.5%, and reclassifying through a new measurement of HbA1c at least 3 months after discharge. Statistics: SPSS v20.

Results: Analyzing our data, we select a total of 100 cases, 45 in group A and 55 in group B. When comparing group A vs B, were found male gender 57% vs 44%, mean age 57 vs 55 years, mean weight 95.7 vs 83.7 kg, mean body mass index (BMI) 34.1 vs 31.1 kg/m2, admission mean glycemia 243 vs 162 mg/dl and mean HbA1c 8.5% vs 5.8%, this last two with P<0.0001 by t-test. The most important comorbidities were obesity 75%
vs 51%. The frequency of infection was 71% vs 60%, and sepsis 26% vs 27%. Metabolic parameters significantly associated by t-test with group A vs B were admission HbA1c% (DS): 8.26 (2.2) vs 5.8 (0.4); admission glycemia mg/dl (DS): 236.6 (104.8) vs 164.2 (49.3); maximum glycemia mg/dl (DS): 291.5 (92.3) vs 215.5 (58.4), time in glycemic range (TIR%) (DS): 49.3 (29.4) vs 82.5 (21.6), P<0.0001 for all of them. Cases were reclassified according to the HbA1C as follows: 47 cases of unknown DM (unkDM) with HbA1c ≥ 6.5%, 31 cases of pre Diabetes (preDM) with HbA1c ≥5.7-6.5%, and 22 cases of stress Hyperglyucemia (SHG) with HbA1c <5.7%. The Pearson´s correlation analysis showed that admission HbA1c% (DS): 7.81 (2.3) vs 6.1 (0.8); maximum glycemia mg/dl (DS): 290.0 (87.9) vs 215.4 (63.9); and TIR% (DS): 52.9 (31.2) vs 80.0 (23.1), P<0.001, were associated with development of Diabetes mellitus, P<0.001 for all of them.

Conclusions: In our study we were able to differentiate two inpatients populations with
new hyperglycemia, describing the relationship between starting metabolic variables, glycemic goals and development of Diabetes mellitus. We could reveal the meaning of this ambiglycemia (AmbiG) at the hospital admission.