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

Tables

Table 1. Baseline Characteristics of Patients With Diabetic Ketoacidosis and Simple Hyperglycemia
 
Baseline characteristicsMissingDiabetic ketoacidosis (n = 56), mean ± SDSimple hyperglycemia (n = 218), mean ± SDP 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 (%)031 (55.4)114 (52.3)0.765
Age (years)052.9 ± 18.560.4 ± 13.70.001
Body mass index (kg/m2)422.4 ± 4.623.4 ± 5.00.209
Symptoms, n (%)054 (96.4)124 (56.9)< 0.001
  Dyspnea042 (75.0)74 (33.9)< 0.001
  Gastrointestinal/hepatobiliary023 (41.1)55 (25.2)0.030
  Fever018 (32.1)55 (25.2)0.312
Vital signs
  Temperature (°C)036.7 ± 1.036.8 ± 1.10.613
  Respiratory rate027.4 ± 8.722.1 ± 5.6< 0.001
  Systolic blood pressure (mm Hg)0124.8 ± 34.3138.3 ± 31.90.006
  Diastolic blood pressure (mm Hg)074.7 ± 20.578.4 ± 18.60.199
  Mean arterial pressure (mm Hg)091.4 ± 24.198.3 ± 21.30.035
  Pulse rate (bpm)0112.4 ± 22.395.6 ± 21.8< 0.001
  SpO2 (%)096.5 ± 6.097.1 ± 3.50.348
Initial capillary blood glucose (mg/dL)0495.3 ± 117.0420.7 ± 108.5< 0.001
Initial blood glucose (mg/dL)0594.8 ± 299.5431.3 ± 147.7< 0.001
Hemoglobin A1c (prior visit) (%)9510.1 ± 3.49.3 ± 3.00.173
Hemoglobin A1c (latest visit) (%)11512.2 ± 3.110.5 ± 2.90.001
Cr (prior visit) (mg/dL) median (IQR)460.87 (0.70, 1.28)0.94 (0.7, 1.43)0.330
eGFR (prior visit) (mL/min/1.73 m2), median (IQR)4687.26 (54.86, 107.21)78.76 (46.75, 101.26)0.135
Cr (latest visit) (mg/dL), median (IQR)01.35 (0.67, 2.76)1.11 (0.75, 1.70)0.263
eGFR (latest visit) (mL/min/1.73 m2), median (IQR)056.24 (21.60, 104.55)63.64 (41.13, 94.71)0.604
Urine ketone, median (IQR)03 (2, 4)0 (0, 0)< 0.001
Suspected infection, n (%)046 (82.1)104 (47.7)< 0.001
Presumed source of infection, n (%)
  No infection011 (19.6)113 (51.8)< 0.001
  Pneumonia015 (26.8)24 (11.0)
  Urinary tract infection013 (23.2)31 (14.2)
  Septicemia09 (16.1)16 (7.3)
  Gastrointestinal/hepatobiliary infection01 (1.8)5 (2.4)
  Infected wound00 (0)16 (7.3)
  Other07 (12.5)13 (6.0)
First diagnosis: hyperglycemia, n (%)02 (3.6)22 (10.1)0.183
Previous diabetic ketoacidosis, n (%)014 (25.0)9 (4.1)< 0.001
Diabetes mellitus, n (%)054 (96.4)196 (89.9)0.183
Type of diabetes mellitus (DM), n (%)
  Undiagnosed DM02 (3.6)22 (10.1)< 0.001
  Type 1 DM08 (14.3)3 (1.4)
  Type 2 DM046 (82.1)193 (88.5)
CKD stages 3 - 5, n (%)018 (32.1)81 (37.2)0.535
Hypertension, n (%)033 (58.9)154 (70.6)0.108
Dyslipidemia, n (%)024 (42.9)120 (55.1)0.133
Insulin usage, n (%)039 (69.6)64 (29.4)< 0.001
Type of insulin, n (%)
  Multiple dose regular insulin1696 (15.4)12 (18.2)0.794
  Isophane insulin1709 (23.1)14 (21.5)1
  Mixtard insulin17031 (79.5)41 (63.1)0.124
Oral hypoglycemic agent used, n (%)025 (44.6)138 (63.3)0.014
Type of oral hypoglycemic agents, n (%)
  Sulfonylureas (glipizide)1109 (36.0)69 (49.6)0.277
  Biguanides (metformin)11024 (96.0)112 (80.6)0.081
  Thiazolidinediones (pioglitazone)11013 (52.0)45 (32.4)0.071
Diet control alone, n (%)02 (3.6)21 (9.6)0.183
Poor compliance, n (%)042 (75.0)97 (44.5)< 0.001

 

Table 2. Final multivariable risk ratio regression analysis for the predictive model
 
PredictorsDiabetic ketoacidosis
mRR95% CIP value
mRR: multivariable risk ratio; DM: diabetes mellitus; CI: confidence interval.
Initial capillary blood glucose (mg/dL)11.00, 1.000.043
History of DMReference
  Type 1 DM2.931.00, 8.520.049
  Type 2 DM2.140.83, 5.530.117
Insulin usage2.791.63, 4.77< 0.001
Poor compliance2.431.45, 4.050.001
Respiratory rate1.041.01, 1.060.001
Suspected infection2.511.32, 4.780.005

 

Table 3. Internal Validation Using the Bootstrap Resampling Method
 
ParametersApparent performanceBootstrap performance
AuROC: area under the receiver operating characteristic curve; CITL: calibration-in-the-large; E:O ratio: expected-to-observed outcomes ratio.
AuROC0.8777 (0.8294, 0.9259)0.8770 (0.8330, 0.9310)
Slope1.000 (0.730, 1.270)0.991 (0.743, 1.285)
E:O ratio1.0001.004
CITL0.000 (-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-pointDKA (n = 56)Simple hyperglycemia (n = 218)TotalSensitivitySpecificityPPVNPV
DKA: diabetic ketoacidosis; PPV: positive predictive value; NPV: negative predictive value.
Probability > 15%50 (18.2%)67 (24.5%)11789.3%78.1%42.7%96.2%
Probability ≤ 15%6 (2.2%)151 (55.1%)157
Probability > 19.4%43 (15.7%)42 (15.3%)8576.8%80.7%50.6%93.1%
Probability ≤ 19.4%13 (4.7%)176 (64.3%)189
Probability > 25%38 (13.9%)25 (9.1%)6367.9%88.5%60.3%91.5%
Probability ≤ 25%18 (6.6%)193 (70.4%)211