M. Tehranisharif1; S.I.N.A. Jahan2; S. Jamshidi2
Introduction
The aim of this study was to validate available human PBGMs in Iran's market for use in pet practice.
Methods
Fifty-five dogs were included in this study. The study was performed from January to February 2013. Four PBGMs, Bionime GM110® (Switzerland), Bionime GM300® (Switzerland), Accu-Chek Active® (Germany), GLUCOCARD® (Japan), were evaluated in this study. The sampling area (pinna) was pricked by a lancet and marginal bloods were evaluated with PBGMs; 3 ml of whole fresh blood was drawn from the cephalic vein of each animal and transferred to plain tubes. Immediately after blood collection, serum was harvested and transferred on ice to the laboratory and kept at -20°C until laboratory evaluation. The glucose concentrations were determined by enzymatic method (Pars Azmoon®, Iran) with semiautomated photometer (RIELE 5010, Germany). To evaluate the repeatability, precision and within-run imprecision, one sample (low, normal, high control) was analyzed 15 times, consequently with a lag time of 15 sec, and coefficient of variance (CV) was determined. In order to evaluate the correlation between glucose and PBGMs, Pearson correlation (r) was determined. To validate PBGMs, Bland-Altman plot and mean difference plot were performed. Agreement between methods was verified using the Bland-Altman difference plot. In order to validate PBGMs, at various blood levels, high- and low-level glucose controls were prepared.
Results
Table 1. Descriptive statistic of various methods (For peer review)
|
|
Mean ± SE (mg/dl)
|
Bias %
|
Correlation
coefficient
|
Coefficient of
determinant (R2)
|
Intercept
|
Slope
|
95% limit of agreement
|
|
Laboratory
method
|
115.5 ± 9.6
|
-
|
-
|
-
|
-
|
-
|
-
|
|
Bionime GM110/
venous sample
|
110.6 ± 9.41
|
-4.7%
|
0.98
|
0.96
|
-1.65
|
1.07
|
-8.3 to 1.1%
|
|
Bionime GM300/
venous sample
|
108.5 ± 9.8
|
-6%
|
0.98
|
0.97
|
-1.85
|
1.09
|
-18.9 to 6.9%
|
|
AccuCheck Active/
venous sample
|
99.1 ± 8.39
|
-15.3%
|
0.98
|
0.96
|
-11.3
|
1.06
|
-58.1% to 27.2%
|
|
Glucocard/
venous sample
|
105.1 ± 11.1
|
-15.2%
|
0.97
|
0.95
|
-24.5
|
1.12
|
-40.8% to 10.4%
|
|
Bionime GM110/
capillary sample
|
111.1 ± 9.43
|
-4.2%
|
0.98
|
0.97
|
-1.67
|
1.1
|
-7.9% to 0.5%
|
|
Bionime GM300/
capillary sample
|
110.3 ± 9.31
|
-4.9%
|
0.98
|
0.96
|
-1.67
|
1.11
|
-9.3 to 11.7%
|
|
AccuCheck Active/
capillary sample
|
108.7 ± 10.87
|
-10.4%
|
0.97
|
0.95
|
-18.3
|
1.1
|
-35.1% to 11.4%
|
Table 2. Coefficient of variation of PBGMs for the low, normal and high control solutions (For peer review)
|
PBGMs
|
Low control
CV%
|
Normal control
CV%
|
High control
CV%
|
|
Bionime GM110
|
4/1
|
4/1
|
3/2
|
|
Bionime GM300
|
4/3
|
4/3
|
4/3
|
|
AccuCheck Active
|
5/2
|
4/7
|
4/6
|
|
Glucocard
|
4/9
|
4/4
|
4/2
|
Discussion
The results of this study indicated that human-based PBGMs are useful for determining glucose levels in emergency situations and monitoring of diabetic dogs.