Fig. 1The process of minimally invasive interstitial fluid (ISF) extraction technology from pretreatment with a microneedle array to ISF glucose accumulation. Accumulated glucose corresponds to glucose area under the curve (AUC) during 8 hours.
Fig. 2Correlation between glucose area under the curve (AUC) measured by continuous glucose monitoring (CGM) and interstitial fluid glucose AUC measured by minimally invasive interstitial fluid extraction technology (MIET) examined at (A) daytime (y=x, r=0.76) and (B) nighttime (y=x, r=0.82).
Fig. 3The area under the receiver operating characteristics curves showed a positive discrimination threshold for interstitial fluid glucose of minimally invasive interstitial fluid extraction technology (MIET) area under the curve (AUC) of glucose level over 200 mg/dL by continuous glucose monitoring in all patients. The optimal cutoff value of MIET AUC to predict over 200 mg/dL of blood glucose level was 1,067.3 mg·hr/dL (sensitivity of 88.2% and specificity of 81.5%).
Fig. 4No correlation between serum Na+ level and ratio of interstitial fluid glucose area under the curve (AUC) measured by minimally invasive interstitial fluid extraction technology (MIET) to glucose AUC measured by continuous glucose monitoring (CGM) in all patients.
Fig. 5Response of the patients to a questionnaire about minimally invasive interstitial fluid extraction technology in all patients. (A) Pain at stamping. (B) Impression at stamped area.
Table 1Clinical characteristic of diabetic patients
Characteristic |
Patients examined at daytime |
Patients examined at nighttime |
All patients |
Number |
7 |
26 |
33 |
Sex, male/female |
5/2 |
11/15 |
16/17 |
Age, yr |
65.9±13.0 |
58.5±15.8 |
60.0±15.4 |
Body mass index, kg/m2
|
26.0±2.4 |
24.7±5.4 |
24.9±4.9 |
Duration, yr |
12.4±10.7 |
12.3±10.7 |
12.3±10.6 |
Fasting blood glucose, mg/dL |
165.4±61.9 |
159.4±69.2 |
160.7±66.8 |
HbA1c, % |
10.1±2.3 |
9.7±2.6 |
9.8±2.5 |
eGFR, mL/min/1.73 m2
|
66.0±20.4 |
73.4±29.7 |
71.9±27.9 |
C-peptide, ng/mL |
2.1±1.2 |
2.3±2.3 |
2.3±2.1 |
No. of cases of <60 mL/min/1.73 m2 of eGFR |
1 |
6 |
7 |
No. of cases of <30 mL/min/1.73 m2 of eGFR |
0 |
3 |
3 |
Type of diabetes |
|
|
|
Type 1 |
0 |
3 |
3 |
Type 2 |
7 |
22 |
29 |
Others |
0 |
1 |
1 |
Medications |
|
|
|
DPP-4 inhibitor |
2 |
11 |
13 |
Sulfonylureas |
1 |
6 |
7 |
α-Glucosidase inhibitors |
1 |
5 |
6 |
Glinides |
0 |
3 |
3 |
Biguanide |
0 |
1 |
1 |
Thiazolidine |
0 |
2 |
2 |
GLP-1 receptor agonist |
0 |
1 |
1 |
Insulin |
6 |
18 |
24 |
Table 2CGM parameters and MIET-derived AUC in diabetic patients
Parameter |
Patients examined at daytime |
Patients examined at nighttime |
All patients |
Number |
7 |
26 |
33 |
Parameters with CGM data |
|
|
|
Mean glucose level, mg/dL |
181.7±20.7 |
139.4±37.2 |
148.4±38.3 |
SD, mg/dL |
44.3 (22.5–54.9) |
18.3 (12.7–38.9) |
22.5 (13.4–43.9) |
CV, % |
24.7 (13.6–28.5) |
13.4 (8.5–27.7) |
13.8 (10.0–28.0) |
Highest blood glucose level, mg/dL |
261.3±28.4 |
190.7±48.9 |
205.6±53.7 |
Lowest blood glucose level, mg/dL |
115.1±26.4 |
99.8±35.5 |
103.0±34.0 |
MAGE, mg/dL |
71.0 (49.5–91.0) |
40.2 (23.6–70.8) |
48.0 (30.0–83.5) |
M value |
22.2 (15.2–30.7) |
5.7 (1.3–13.2) |
8.1 (1.6–18.9) |
J-index |
49.7±12.4 |
28.6±14.2 |
33.1±16.2 |
Glucose AUC of CGM, mg·hr/dL |
1,469.0±170.4 |
1,118.0±301.6 |
1,192.4±312.6 |
Glucose AUC of MIET, mg·hr/dL |
1,644.9±306.0 |
1,113.5±338.2 |
1,226.3±394.4 |
Table 3Linear regression analysis between MIET-derived AUC and CGM parameters in all diabetic patients
Variable |
r
|
P value |
Mean glucose level, mg/dL |
0.865 |
<0.0001 |
Log-SD |
0.350 |
0.046 |
Log-CV |
0.014 |
0.936 |
Highest glucose level, mg/dL |
0.768 |
<0.0001 |
Lowest glucose level, mg/dL |
0.603 |
<0.0001 |
Log-MAGE |
0.409 |
0.018 |
Log-M value |
0.844 |
<0.0001 |
J-index |
0.848 |
<0.0001 |