A Testing For Ketone Bodies In The Breath, Blood And Urine: Diagnostic And Therapeutic Implications For Diabetic Ketosis

Background： Diabetic ketosis (DK) is the most common acutecomplications of diabetes, which can be fatal if not identifiedand promptly treated. Ketone bodies including acetoacetate, β-hydroxybutyric acid and acetone. Urine ketone testing mainly reflect thelevel of acetoacetate but not acetone in urine. Blood ketone bodiestesting mostly reflect the level of β-hydroxybutyric in blood, and thebreath acetone mainly reflect the level of acetone. All the methods abovecan be used to test ketone bodies level, but the clinical significanceof each is different.Objective： To test acetoacetate, β-hydroxybutyric acid, acetoneand evaluate the significance for the diagnosis and treatment ofdiabetic ketosis. To explore the advantages of various detection methodsand evaluate the value of breath ketone test.Methods： According to the WHO diabetic diagnostic criteria in1999，99diabetes patients (including49male and50female, aged from11to85years) were recruited from the Second Hospital of Jilin Universityfrom March2012to May2013. And the patients were divided into groupsof diabetic ketosis (DK) and non-diabetic ketosis group (NDK). Bloodketone bodies greater than the reference value (0.5mmol/L) is positive，and urine ketone with+above (including+) is positive.In these patientsthe concentration of acetone in the patient breathing gas were test bygas chromatography/mass spectrometry (GC/MS) method, exclude thatfingertip blood ketone, fasting plasma glucose, HbA1c, blood lipids,liver and kidney function assay were also detected.We used SPSS17.0 statistical software for analysis,P<0.05was considered statisticallysignificant.Results:1.According to the qualitative detection of urine ketone thepatients were divided into five groups, group1（-，n=19）, group2（+-，n=16）、group3（+，n=25）, group4（++，n=19）, group5（+++，n=20）. Fastingblood glucose in group5was much higher compare with other four groups,and had Significant difference(P<0.001, P=0.005, P=0.029, P=0.008respectively).The fasting glucose of between group one to four didn’thave significant difference. The patients in group5were significantlyyounger than the group one to three（P=0.005, P=0.001, P=0.001）, and thepatients in group4were significantly younger than group2（P=0.037）.Agebetween the other groups didn’t have statistically significant.2.Blood ketone of group4was significantly higher than group1, group2and group3（P=0.003, P=0.008, P=0.023）, and blood ketone of group5was significantly higher than other four groups（P<0.001，P<0.001,P<0.001, P<0.001respectivelly）. Breath ketone in Group4wassignificantly higher than group1and group3（P=0.028, P=0.035）. Breathketone in Group5was significantly higher than other four groups（P<0.001,P<0.001, P<0.001, P=0.002）.The positive rates of blood ketone in group1-5were6.7%，14.29%，43.48%，71.43%，89.5%, respectively. The positiverates of breath ketone in group1-5were6.67%，14.29%，60%，80%，92.86%,respectively.3.The level of breath ketone was significantly related to FBG value（r=0.428；P<0.001）, blood ketone（r=0.817；P<0.001）, urine ketone（r=0.581；P<0.001）, TG（r=0.219；P=0.09）, LDL（r=0.255；P=0.047）, CRE（r=0.385；P=0.002）and BUN（r=0.362；P=0.003）.4.Receiver operating characteristic (ROC) curve for exhaled acetone inthe diagnosis of diabetic ketosis show that the area under the curve(AUC)is0.902(P<0.001).The concentration of breath ketone as the best critical point is1.185ppm to predict diabetic ketoacidosis, which sensitivity is90.6%and specificity is77.1%.Conclusion： FBG and age are important influence factors of DKA. Thelevel of breath ketone was significantly related to FBG, blood ketone,urine ketone, TG,LDL-C，CRE and BUN. The sensitivity and specificity ofbreath ketone were higher values. We conclude that the exhaled breathacetone concentration can be used as a good biomarker to facilitate thenon-invasive diagnosis and monitoring of the status of diabetes.