Journal of Clinical Medicine Research, ISSN 1918-3003 print, 1918-3011 online, Open Access
Article copyright, the authors; Journal compilation copyright, J Clin Med Res and Elmer Press Inc
Journal website https://jocmr.elmerjournals.com

Original Article

Volume 17, Number 3, March 2025, pages 164-173


Predictive Factors for Diagnosing Diabetic Ketoacidosis or Simple Hyperglycemia in Adults With High Blood Glucose: The “1-DKA Alert” Study

Figures

↓  Figure 1. Study flow diagram.
Figure 1.
↓  Figure 2. Area under the receiver operating characteristic (AuROC) curve. CI: confidence interval.
Figure 2.
↓  Figure 3. Calibration plot for predicted and observed probabilities of diabetic ketoacidosis. CI: confidence interval.
Figure 3.
↓  Figure 4. Decision curve analysis. DKA: diabetic ketoacidosis.
Figure 4.
↓  Figure 5. Risk curve showing predicted probability of diabetic ketoacidosis across linear combination values.
Figure 5.

Tables

↓  Table 1. Baseline Characteristics of Patients With Diabetic Ketoacidosis and Simple Hyperglycemia
 
Baseline characteristics Missing Diabetic ketoacidosis (n = 56), mean ± SD Simple hyperglycemia (n = 218), mean ± SD P value
SD: standard deviation; bpm: beats per minute; SpO2: peripheral capillary oxygen saturation; Cr: Creatinine; eGFR: estimated glomerular filtration rate; CKD: chronic kidney disease.
Female, n (%) 0 31 (55.4) 114 (52.3) 0.765
Age (years) 0 52.9 ± 18.5 60.4 ± 13.7 0.001
Body mass index (kg/m2) 4 22.4 ± 4.6 23.4 ± 5.0 0.209
Symptoms, n (%) 0 54 (96.4) 124 (56.9) < 0.001
  Dyspnea 0 42 (75.0) 74 (33.9) < 0.001
  Gastrointestinal/hepatobiliary 0 23 (41.1) 55 (25.2) 0.030
  Fever 0 18 (32.1) 55 (25.2) 0.312
Vital signs
  Temperature (°C) 0 36.7 ± 1.0 36.8 ± 1.1 0.613
  Respiratory rate 0 27.4 ± 8.7 22.1 ± 5.6 < 0.001
  Systolic blood pressure (mm Hg) 0 124.8 ± 34.3 138.3 ± 31.9 0.006
  Diastolic blood pressure (mm Hg) 0 74.7 ± 20.5 78.4 ± 18.6 0.199
  Mean arterial pressure (mm Hg) 0 91.4 ± 24.1 98.3 ± 21.3 0.035
  Pulse rate (bpm) 0 112.4 ± 22.3 95.6 ± 21.8 < 0.001
  SpO2 (%) 0 96.5 ± 6.0 97.1 ± 3.5 0.348
Initial capillary blood glucose (mg/dL) 0 495.3 ± 117.0 420.7 ± 108.5 < 0.001
Initial blood glucose (mg/dL) 0 594.8 ± 299.5 431.3 ± 147.7 < 0.001
Hemoglobin A1c (prior visit) (%) 95 10.1 ± 3.4 9.3 ± 3.0 0.173
Hemoglobin A1c (latest visit) (%) 115 12.2 ± 3.1 10.5 ± 2.9 0.001
Cr (prior visit) (mg/dL) median (IQR) 46 0.87 (0.70, 1.28) 0.94 (0.7, 1.43) 0.330
eGFR (prior visit) (mL/min/1.73 m2), median (IQR) 46 87.26 (54.86, 107.21) 78.76 (46.75, 101.26) 0.135
Cr (latest visit) (mg/dL), median (IQR) 0 1.35 (0.67, 2.76) 1.11 (0.75, 1.70) 0.263
eGFR (latest visit) (mL/min/1.73 m2), median (IQR) 0 56.24 (21.60, 104.55) 63.64 (41.13, 94.71) 0.604
Urine ketone, median (IQR) 0 3 (2, 4) 0 (0, 0) < 0.001
Suspected infection, n (%) 0 46 (82.1) 104 (47.7) < 0.001
Presumed source of infection, n (%)
  No infection 0 11 (19.6) 113 (51.8) < 0.001
  Pneumonia 0 15 (26.8) 24 (11.0)
  Urinary tract infection 0 13 (23.2) 31 (14.2)
  Septicemia 0 9 (16.1) 16 (7.3)
  Gastrointestinal/hepatobiliary infection 0 1 (1.8) 5 (2.4)
  Infected wound 0 0 (0) 16 (7.3)
  Other 0 7 (12.5) 13 (6.0)
First diagnosis: hyperglycemia, n (%) 0 2 (3.6) 22 (10.1) 0.183
Previous diabetic ketoacidosis, n (%) 0 14 (25.0) 9 (4.1) < 0.001
Diabetes mellitus, n (%) 0 54 (96.4) 196 (89.9) 0.183
Type of diabetes mellitus (DM), n (%)
  Undiagnosed DM 0 2 (3.6) 22 (10.1) < 0.001
  Type 1 DM 0 8 (14.3) 3 (1.4)
  Type 2 DM 0 46 (82.1) 193 (88.5)
CKD stages 3 - 5, n (%) 0 18 (32.1) 81 (37.2) 0.535
Hypertension, n (%) 0 33 (58.9) 154 (70.6) 0.108
Dyslipidemia, n (%) 0 24 (42.9) 120 (55.1) 0.133
Insulin usage, n (%) 0 39 (69.6) 64 (29.4) < 0.001
Type of insulin, n (%)
  Multiple dose regular insulin 169 6 (15.4) 12 (18.2) 0.794
  Isophane insulin 170 9 (23.1) 14 (21.5) 1
  Mixtard insulin 170 31 (79.5) 41 (63.1) 0.124
Oral hypoglycemic agent used, n (%) 0 25 (44.6) 138 (63.3) 0.014
Type of oral hypoglycemic agents, n (%)
  Sulfonylureas (glipizide) 110 9 (36.0) 69 (49.6) 0.277
  Biguanides (metformin) 110 24 (96.0) 112 (80.6) 0.081
  Thiazolidinediones (pioglitazone) 110 13 (52.0) 45 (32.4) 0.071
Diet control alone, n (%) 0 2 (3.6) 21 (9.6) 0.183
Poor compliance, n (%) 0 42 (75.0) 97 (44.5) < 0.001

 

↓  Table 2. Final multivariable risk ratio regression analysis for the predictive model
 
Predictors Diabetic ketoacidosis
mRR 95% CI P value
mRR: multivariable risk ratio; DM: diabetes mellitus; CI: confidence interval.
Initial capillary blood glucose (mg/dL) 1 1.00, 1.00 0.043
History of DM Reference
  Type 1 DM 2.93 1.00, 8.52 0.049
  Type 2 DM 2.14 0.83, 5.53 0.117
Insulin usage 2.79 1.63, 4.77 < 0.001
Poor compliance 2.43 1.45, 4.05 0.001
Respiratory rate 1.04 1.01, 1.06 0.001
Suspected infection 2.51 1.32, 4.78 0.005

 

↓  Table 3. Internal Validation Using the Bootstrap Resampling Method
 
Parameters Apparent performance Bootstrap performance
AuROC: area under the receiver operating characteristic curve; CITL: calibration-in-the-large; E:O ratio: expected-to-observed outcomes ratio.
AuROC 0.8777 (0.8294, 0.9259) 0.8770 (0.8330, 0.9310)
Slope 1.000 (0.730, 1.270) 0.991 (0.743, 1.285)
E:O ratio 1.000 1.004
CITL 0.000 (-0.367, 0.367) 0.002 (-0.386, 0.425)
Bootstrap shrinkage - 0.991

 

↓  Table 4. Accuracy and Diagnostic Indices at Different Probability Cut-Points for Predicting DKA
 
Cut-point DKA (n = 56) Simple hyperglycemia (n = 218) Total Sensitivity Specificity PPV NPV
DKA: diabetic ketoacidosis; PPV: positive predictive value; NPV: negative predictive value.
Probability > 15% 50 (18.2%) 67 (24.5%) 117 89.3% 78.1% 42.7% 96.2%
Probability ≤ 15% 6 (2.2%) 151 (55.1%) 157
Probability > 19.4% 43 (15.7%) 42 (15.3%) 85 76.8% 80.7% 50.6% 93.1%
Probability ≤ 19.4% 13 (4.7%) 176 (64.3%) 189
Probability > 25% 38 (13.9%) 25 (9.1%) 63 67.9% 88.5% 60.3% 91.5%
Probability ≤ 25% 18 (6.6%) 193 (70.4%) 211