DETERMINATION OF FREQUENCY OF A NOVEL DIAGNOSTIC CATEGORY OF HYPERGLYCEMIA I.E. IMPAIRED RANDOM GLUCOSE IN PREVIOUSLY UNKNOWN PATIENTS OF PREDIABETES AND DIABETES MELLITUS

Abstract

Objective: To determine the frequency of normoglycemia (NG), prediabetes and diabetes mellitus (DM) among
patients having impaired random glucose (IRG) and establish the optimum cutoff of impaired random glucose
for the diagnosis of diabetes and prediabetes.

Study Design: Cross sectional study.

Place and Duration of Study: Armed Forces Institute of Pathology Rawalpindi and Combined Military Hospital
Quetta, from Nov 2016 to Nov 2018.

Methodology: Healthy individuals of both genders undergoing investigation for diabetes mellitus were enrolled through non probability consecutive sampling while patients with diabetes, pregnant, hospitalized, using concomitant medications (corticosteroids, immunosuppressive, chemotherapy) and those unable to complete oral glucose tolerance test (OGTT) were excluded. Random plasma glucose (RPG) was determined and patients having impaired random glucose underwent oral glucose tolerance test, analysis being done on ADVIA1800®using hexokinase methodology.

Results: Among the 280 study participants, majority were female {156 (57.5%) vs 124 (44.3%) male}. Mean age in male and female patients was 33.625 ± 3.34 years vs 35.150 ± 2.79 years with p-value 0.50 while mean IRG was 7.12 ± 1.47 vs 6.90 ± 1.17 mmol/L respectively with p-value 0.16 (statistically insignificant). Oral glucose tolerance test results showed NG in 61.8% (173), prediabetes in 24.6% (69) while diabetes was found in 38 (13.6%) patients. Optimal impaired random glucose cutoff value for diagnosis of diabetes was found 7.45 mmol/L (AUC 0.956 [CI 0.927-0.984], p<0.001, sensitivity 94.7%, specificity 74.4%), however, same cut-off value showed impaired random glucose as fair diagnostic test for prediabetes (AUC 0.771 [CI 0.717-0.825], p<0.001, sensitivity 72.5%, specificity 77.3%).......