| Objective: Diabetes is a common metabolic disorder in dogs.In recent years,as the number of pet dogs and obese dogs has increased,the incidence rate has gradually increased.Diagnostic indexes for fasting blood glucose in diabetic dogs are based on clinical experience without systematic and rigorous statistical analysis.The aim of this study was to explore the reasonable cut-off points for fasting blood glucose diagnosing for diabetic dogs by using scientific and rigorous data collection and analysis.METHODS: The dogs were from Jilin University Animal Hospital of Jilin University,Jilin Province,China Agricultural University Veterinary Teaching Hospital,Dr.Sun Animal Hospital,Animal Hospital of Yunnan Agricultural University,Nanchang Police Dog Base of Ministry of Public Security,Changchun Police Dog Base of Ministry of Public Security in December 2016 to December 2017.All of the dogs were outpatient dogs and physical examination dogs.A total of 151 valid data were available,of which 77 were males and 74 were females.The average age of diabetic dogs was 5.9500±2.57 years old,and the average age of non-diabetic dogs was 3.3347±1.32 years old.All test dogs were received fasting blood glucose,standard OGTT test.Follow-up visits were conducted.Diabetic and non-diabetic groups weredivided according to the test indicators and clinical diagnosis of doctors.Among them,there were 50 patients in the diabetic group(25 females and 25 males)and 101 patients in non-diabetic groups(49 females and 52 males).All test dogs met the inclusion criteria and were fasted for 8 to 14 hours before the experiment.2ml of sputum venous blood was collected in the early morning of the next day to test fasting blood glucose.The standard OGTT test was performed and the OGTT test was performed for 2 hours later.The blood was submitted and the venous blood was tested on a veterinary automatic analyzer(glucose oxidase method was used for detection of glucose).The data were detailed record.SPSS23.0 statistical software was used for data analysis to explore reasonable cut-off point for fasting blood glucose screening for diabetes.Results:(1)Statistical analysis was conducted on the gender and age composition of the tested dogs.The gender and age of the tested dogs in the diabetic group and non-diabetic group were analyzed by the t test with P was more 0.05.It meant the gender and age of these two groups were comparable.The age distribution of diabetic dogs was analyzed.Among them,dogs aged 5 to 6 years and 7 to 8 years old accounted for 36% and 30% of diabetic dogs,respectively.Diabetes mellitus was frequently found in dogs aged 5 to 8 years old.Analysis of the body type and breed of the diabetic group showed that the prevalence of the dogs was highest,and the prevalence rates of Teddy,Bichon,Bianmu,Jinmao,Labrador and Chinese pastoral dogs were all higher;(2)The comparison of fasting blood glucose and frequency distribution between the diabetic group and the non-diabetic group: the value of fasting bloodglucose in the diabetic dogs and non-diabetic dogs 18.56±3.82,5.10±0.64,and P<0.05,indicating that the FPG values of the diabetic group were significantly higher than those in non-diabetic group and the difference was statistically significant.Frequency analysis showed that 50% of the subjects in the diabetic group were with the FPG <15.1 mmol/L,24% of tested dogs were with FPG <11.1mmol/L;In non-diabetic groups,FPG in 88.1% dogs was less than 10.7 mmol/L In the non-diabetic group,indicating that the diagnosis of diabetes with FPG ≥ 11.1 mmol/L had a missed rate of 24% and was not sensitive enough;(3)FPG and PG2 h in the diabetic group,the non-diabetic group and all the subjects were with significant correlation(r=0.789),P<0.05;(4)The regression analysis of FPG and PG2 h showed that when PG2h=12.95 mmom/L,FPG=10.7 mmol/L;(5)The results of ROC curves of fasting blood glucose showedthat the area under the ROC curve was maximal when the fasting blood glucose value is 10.7 mmol/L(0.935,P<0.05).Additional,the sensitivity was 0.840,the specificity was 0.980,and the Udon index was 0.820.This results showed it may be a reasonable cut-off for canine diabetes screening when FPG was 10.7 mmol/L;(6)When FPG was 10.7 mmol / L,the sensitivity for diagnosis of diabetes was 84.00%,specificity was 98.00%,and the missed rate was 16%,the misdiagnosis rate was 2%.Compared with FPG was 11.1 mmol/L,the missed diagnosis rate was reduced by 8.0%.Therefore,set the 11.1mmol/L as the fasting blood glucose cut-off points for screening canine diabetes was not with enough sensitivity.The positive predictive value was 95.47% when set the FPG 10.7 mmol/L as the cut-off points,and the negative predictive was 62.4%.when set the value of FPG was 10.7 mmol/L as the diagnostic criterion,the probability of misdiagnosi-nosing diabetes to non-diabeticwas only 4.53%.In summary,it was more reliably to regard the value of FPG was 10.7 mmol/L as the cut-off point to diagnosis for diabetes in dogs. |
PDF Full Text Request