engleski

Diabetic patients in intensive care unit: increased risk and some specificites

Diabetes mellitus (DM) with its chronic and acute complications puts patients with the disease at higher risk in case of acute illness. Although patients with diabetes usually are recognized as patients with increased risk when admitted to intensive care units, this risk is not addressed in any of the scoring systems commonly used in intensive care medicine. Most commonly used score that accounts for chronic health is APACHE II score, but it does not score any points for diabetes. Some complications of diabetes, primarily chronic renal failure, may score some points in most of the scores, but there are many other complications such as cardiomyopathy, microvascular and macrovascular disease that certainly put the patient with diabetes at additional risk. On the other hand, patients with diabetes and good glycaemic control have fewer chronic complications and should be at lesser risk. The influence of hyperglycaemia during acute illness has been proven to be harmful, which has led to strict glucose control, which has become one of the recommendations in the Surviving Sepsis Campaign [2]. The influence of medium-term glycaemic control which may alter immunologic and inflammatory responses has not been investigated. We have analysed patients with sepsis and diabetes in medical ICU to determine the additional risk that could be attributed to the disease. Also, we have compared the course and outcome of sepsis according to different levels of HbA1c as a measure of medium-term glycaemic control. To evaluate diabetes as a risk factor, data from a medical ICU in a university hospital during the time of five years were analysed. Patients with sepsis as the primary admission diagnosis were included and split to diabetes and nondiabetes groups. The diagnosis of diabetes mellitus had to be established prior to the admission to the hospital, according to the ADA or WHO criteria [3]. Patients with newly diagnosed diabetes during the ICU stay were not included in any of the groups. Hyperglycaemia in non-diabetics during the ICU stay did not exclude patients from the non-diabetes group if DM was not confirmed before discharge. APACHE II and SOFA score were calculated for all patients. HbA1c was measured for all patients with diabetes. ICU and hospital mortality and length of stay (LOS) were the outcome measures ; incidence of organ failure was a measure of disease course. MedCalc statistical software was used for data analyses. Non-parametric tests were used regardless of distribution type for all comparisons. Sepsis was the most common primary admission diagnosis in our ICU during the analysed period: of total 2205 admissions, there were 356 (16.1%) patients with sepsis, 78 of them (21.9%) with diabetes mellitus diagnosed prior to admission. There was no statistically significant difference between the groups either in age, sex distribution, APACHE II score or SOFA score at admission. Mortality in the ICU was 34.7% ; median ICU LOS was 8 (95% CI 7–9.3) days. Patient with DM, compared to non-diabetics had higher mortality (38.9% vs. 34.1% ; P = 0.60) and longer ICU LOS (median 6 vs. 10 days ; P < 0.001). In the subgroup of patients who developed severe sepsis, those with DM had higher number of failing organs (median 2 vs. 1 ; P = 0.024). The most common organ failure in both diabetes and non-diabetes groups was renal failure, followed by respiratory failure. In a logistic regression, DM was found to be related to lethal outcome, together with admission APACHE II and SOFA scores. In the diabetes group, surviving patients had significantly lower HbA1c levels (6.6% vs. 9.6% ; P = 0.001). In a multiple regression model HbA1c was found to relate to LOS together with SOFA score and age. HbA1c was found to be independently related to ICU outcome together with SOFA score. Although some chronic effects of diabetes mellitus can be included in multi-parameter scoring systems such as APACHE II score, the disease itself is not scored. We have shown on patients with sepsis, the most common diagnose in ICU, that diabetes mellitus is an independent predictor of mortality and LOS and that it has significantly higher incidence of organ failure. Patients with DM should be given appropriate attention as high risk patients in the ICU. HbA1c was shown to be predictive of mortality and hospital LOS of patients with sepsis and a history of DM. Chronic hyperglycaemia and consequent increased formation of glycation products in blood and tissues may influence inflammatory and immune response and thus be responsible for increased risk. Proper glycoregulation in diabetic patients could reduce the risks in the event of infection.

Diabetes mellitus; ICU; sepsis; risk factors

nije evidentirano