Delirium impact in a chronic ventilatory care unit

Background

Delirium is a common but underdiagnosed and treated problem in the ICU. It has been associated with poor hospital outcomes, including increased morbidity and mortality, prolonged length of stay and functional decline. Recently published clinical practice guidelines from the Society of Critical Care Medicine recommended monitoring for the presence of delirium in all mechanically ventilated patients.

Objectives

To compare the usual clinical assessment for delirium and the Confusion Assessment Method for the Intensive Care Unit (CAM-ICU), to describe its characteristics in chronically ventilated patients and to evaluate the incidence, associated clinical conditions, use of antipsychotic drugs and late mortality.

Patients and methods

A prospective observational study where delirium was evaluated on a daily basis and followed by a group of physicians and a nurse, previously trained. Thirty-five consecutive patients were admitted to the Ventilatory Care Unit (VCU) during a period of 9 mouths. Thirteen tracheotomized, mechanically ventilated, awake and cooperative subjects were included in the study protocol. Three of them returned to the VCU after discharge and were re-included, resulting in a total of 16 patients studied. The CAM-ICU tool was applied 5 days a week at same time in the afternoon and its data was compared with the physicians' and nurse's evaluation. The results are expressed as the mean ± SD. For statistical analysis we used the chi-square test for evaluated difference of proportion, and considered statistical significance as P < 0.05.

Results

Five females and 11 males were studied. The mean age of the study population was 66.46 ± 14.98 and the mean APACHE II score 18 ± 4.6. Delirium occurred in seven patients (43.8%) during the VCU stay and the CAM-ICU tool detected 68% of it. The mean onset was 11.17 ± 8.91 days and the mean duration was 1.62 ± 0.94 days. New infection was associated with delirium in 30.7% of cases; however, using the CAM-ICU tool this association increased to 56% (P = 0.16). Comparing the mean titulated C reactive protein (CRP-t) level and insulin dose used in patients diagnosed with delirium (D group) and no delirium (ND group) was not significant (Table 1). Sensorial deficit was not present in any patient without delirium. Mortality after 30 days was not significant between both groups. Olanzapin and bromazepan were the most prescripted drugs.

Table 1 (abstract P443)

Conclusion

The incidence of delirium in this study was less than expected. CAM-ICU demonstrated inferior sensibility than that described in the literature. We could not demonstrate any difference between groups. The main limitation of this study was the low number of patients enrolled.