| Objective Analysis the characteristics of ICU patients’ blood glucose mean andglucose variability, assess their influence to ICU patients’s mortality.Method observational study, convenience sample. All patients admitted to the ICU ofGeneral Hospital of Ningxia Medical University, between march and september of2011, withstress high glucose and have glucose moniter are admitted. Grouping patients in2ways:(1)depending on glucose variability (GV), group into1,2,3and4.(2) depending on glucosemean (GM), group into A，B，C，D. Study about the difference of different prognosis markamong those branch groups, and conclusion the character of critical ill patietns’ glucosevariation..Results514patietns are admited, mean age is55.15±18.00（4~90）, the mean APACHEII score is15.19±6.73（5~41），the total mortality is4.86%.(1) GV groups:: With thegrowing of the GV level the mortality rises up obviously, the mortality of the four branchgroup is0%,4.10%,6.62%and13.23%（P=0.000）; so does the glucose mean (GM) and theAPACHE II score.(2)GM groups: About APACHE II score, GM groups have the sametendency as GV groups. The reletionship between GM and ICU mortality is like a J curve.The lowest point is group B (7.65mmol/l＜GM≤8.44mmol/l，the mortality is2.88%), whichhas the similar glucose level as the target glucose (6.1~8.3mmol/l), the highest mortality lie ingroupD（GM＞9.77mmol/l, the mortality is12.40%）, which also has the higest GV level.There’s a strong connection between GV and GM（Pearson=0.583，P=0.000）, and they allconnect with the ICU mortality, but the AUC score are difference, GV is more sensitive andspecifitic than GM(AUC area:0.745vs.0.633). Conclusion ICU patients’ blood glucose mean and variability are associated with ICUmortality,and the blood glucose variability is more sensitive and sepcifitic than glucose mean.When we give a ICU patients tight glucose control, we should not only concerned on theirglucose level, pay more patients on their glucose variability. |
PDF Full Text Request