Economic analysis, level III; therapeutic study, level IV.

BACKGROUND: Pharmacologic agents are frequently utilized for management of intensive care unit (ICU) delirium, yet prescribing patterns and impact of medication choices on patient outcomes are poorly described. We sought to describe prescribing practices for management of ICU delirium and investigate the independent association of medication choice on key in-hospital outcomes including delirium resolution, in-hospital mortality, and days alive and free of the ICU or hospital. METHODS: A retrospective study of delirious adult ICU patients at a tertiary academic medical center. Data were obtained regarding daily mental status (normal, delirious, and comatose), pharmacologic treatment, hospital course, and survival via electronic health record. Daily transition models were constructed to assess the independent association of previous day mental status and medication administration on mental status the following day and in-hospital mortality, after adjusting for prespecified covariates. Linear regression models investigated the association of medication administration on days alive and free of the ICU or the hospital during the first 30 days after ICU admission. RESULTS: We identified 8591 encounters of ICU delirium. Half (45.6%) of patients received pharmacologic treatment for delirium, including 45.4% receiving antipsychotics, 2.2% guanfacine, and 0.84% valproic acid. Median highest Richmond Agitation-Sedation Scale (RASS) score was 1 (0, 1) in patients initiated on medications and 0 (−1, 0) for nonrecipients. Haloperidol, olanzapine, and quetiapine comprised >97% of antipsychotics utilized with 48% receiving 2 or more and 20.6% continued on antipsychotic medications at hospital discharge. Haloperidol and olanzapine were associated with greater odds of continued delirium (odds ratio [OR], 1.48; 95% confidence interval [95% CI], 1.30-1.65; P < .001 and OR, 1.37; 95% CI, 1.20-1.56; P = .003, respectively) and increased hazard of in-hospital mortality (hazard ratio [HR], 1.46; 95% CI, 1.10-1.93; P = .01 and HR, 1.67; 95% CI, 1.14-2.45; P = .01, respectively) while quetiapine showed a decreased hazard of in-hospital mortality (HR, 0.58; 95% CI, 0.40-0.84; P = .01). Haloperidol, olanzapine, and quetiapine were associated with fewer days alive and free of hospitalization (all P < .001). There was no significant association of any antipsychotic medication with days alive and free of the ICU. Neither guanfacine nor valproic acid were associated with in-hospital outcomes examined. CONCLUSIONS: Pharmacologic interventions for management of ICU delirium are common, most often with antipsychotics, and frequently continued at hospital discharge. These medications may not portend benefit, may introduce additional harm, and should be used with caution for delirium management. Continuation of these medications through hospitalization and discharge draws into question their safety and role in patient recovery.

In critically ill surgical patients with cIAI, a short duration of antimicrobial therapy after source control resulted in similar outcomes to previously published studies, providing support for the safety of this approach in critically ill patients.

Daptomycin therapy appeared to cause a false and substantial INR elevation in a patient who had been receiving warfarin. Results of TEG suggested that the INR elevation was an artifact of a drug-laboratory interaction and did not represent an anticoagulated state. The patient's INR normalized after linezolid was substituted for daptomycin.

Glucose metabolism in humans is exceedingly complex. At baseline, it is controlled by elaborate signaling mechanisms, and these mechanisms are profoundly altered by the surge of catecholamines and cytokines associated with acute postsurgical and post-traumatic stress. These alterations in signaling mechanisms result in hyperglycemia; although this hyperglycemia can start very rapidly after the traumatic or surgical insult, it can persist during the entire period of critical illness and even afterward. Numerous randomized clinical trials have been conducted to determine if hyperglycemia is associated with increased mortality in surgical patients. These studies have had different conclusions that are difficult to interpret in light of differences in study methodology, but there is certainly ample evidence that inadequately controlled hyperglycemia causes harm due to increased infectious morbidity, and possibly increased mortality. As we have become more proficient in controlling hyperglycemia, the concept of insulin resistance, determined as the amount of insulin required to achieve hyperglycemia, has come to the fore. Insulin resistance is not a static concept, and may change before significant events such as infection. Patients with elevated and persistent insulin resistance have been demonstrated to suffer increased infectious morbidity and mortality, albeit in nonrandomized studies. Along with insulin resistance, the concept of glycemic variability, the amount of variation in serum blood glucose over time, has also become relevant; increased variability has been associated with hypoglycemia and mortality. Both of these risks can result from aggressive insulin therapy, and glycemic control protocols must be appropriately planned and implemented to avoid hypoglycemia and excessive externally induced variability. Computer-assisted protocols may be of significant benefit in optimizing glycemic control. The most recent recommendations available are to keep serum blood glucose levels below 150 mg/dL and to avoid hypoglycemia.