PURPOSE: To investigate the effect of blood sugar control on mortality in critically-ill-septic patients and to evaluate if hypoglycemia is associated with worse outcome.
METHODS: Retrospective chart review of 115-critically-ill-medical/ surgical septic-patients who stayed in ICU>24hrs were randomly selected over one-year.Those with admission diagnosis of DKA or Hyperosmolar-nonketotic-comma were excluded.Charts were reviewed for demographics,admission diagnoses,co-morbidities,blood sugar(finger stick)on the day of admission and every 6-hour daily until the day of discharge from ICU.Outcome was evaluated depending on mean blood sugar(MBS)and on the basis of survival-rate,length of stay,MODS&SOFA-scores,number of ventilator-days,episodes of hypoglycemia,and death.Patients were divided into 3-groups.Patients with MBS<150(group-I),150-200(group-II),and>200(group-III).Statistics were done using Chi-square,ANOVA&T-test.
RESULTS: 115-patients with sepsis,65patients(56%)had MBS<150,36patients(35%)had MBS=150-200,and 14patients(14%)had MBS>200.No statistically-signficant-difference between the three-groups in terms of sex,race,average-insulin used or steroids-use during the ICU-stay.The mortality rates:group-I=49%,group-II=47%&group-III=50%(P-value=0.97).The ICU&hospital mean-length of stay for G-I,G-II&G-III were=12.26,9.4&14(P=0.4)and 19.48,22.9&19(P=0.7).The mean-ventilator-days for:G-I=6.9,G-II=7.0&G-III=6.9(P=0.9).No significant difference in the rate of positive-blood-culture between the three-groups(P=0.38).The mean-SOFA-scores at discharge and mean-MODS at discharge were not significantly-different between the three groups(P=0.67&0.23,respectively).G-I&G-III had more episodes of hypoglycemia compared to G-II(P=0.015&0.004).The mortality rate for those with hypoglycemia was 80%compared to 42% without-hypoglycemia,(P=0.002).Mean-ventilator-days for those with hypoglycemia was 5.5 compared to 13.8 for those with no-hypoglycemia(T=3.725).However,ICU and hospital length of stay were not different between the two groups(T=1.88&0.38,respectively).
CONCLUSIONS: Glycemic control has no affect on mortality&morbidity of critically-ill-septic-medical/surgical patients.We found that MBS<150 or >200 was associated with increased-risk of hypoglycemia,and hypoglycemia was associated with increased mortality and higher number of ventilator-dependent-days.
CLINICAL IMPLICATIONS: We should re-evaluate our strategies of glycemic-control since aggressive-approaches may increase morbidity&mortality by causing more hypoglycemic-episodes.
DISCLOSURE: The following authors have nothing to disclose: Lylla Shahab, Shadi Soufi, Suzzanne Al-Hamad, Muhammad Shibli
No Product/Research Disclosure Information