Delirium Associated With Unplanned Extubations

Author: Thomas Rosenthal

Anesthesiology News

As in previous studies, unplanned extubation (UE) has been found in new research to be associated with delirium, a greater reintubation rate, and longer stays in the ICU and hospital. The researchers called for increased exploration into factors involved in UE.

“The most important finding is that unplanned extubation was associated with delirium as determined by confusion assessment method for ICU [CAM-ICU],” said Nicholas Watson, MD, an anesthesiologist in Grand Rapids who is associated with Michigan State University College of Human Medicine.

“Other studies have found agitation and sedation factors associated with UE, which suggest that delirium is also [a] factor. However, no study to my knowledge has specifically identified a link between UE and delirium as determined by a validated instrument,” said Dr. Watson, the lead study author.

Of the 122 UE patients, 94 qualified for 2:1 case-control matching. Patients were then matched on age, sex, ICU subtype and date of ICU admission. The two groups also had similar severity of disease by APACHE II scores and the number of days intubated at the time of extubation, Dr. Watson said.

Delirium found on any CAM-ICU in the 24 hours preceding extubation, including the last CAM-ICU before extubation, was more common in the UE patients than the control group, Dr. Watson said.

Association Versus Causality

The original data presented at the SCCM meeting were preliminary and showed UE patients had higher rates of delirium than controls, but the difference was not statistically significant, according to Dr. Watson. The 50 UE patients were matched 1:1 to controls, he said. Following the SCCM meeting, the researchers expanded their study to 94 UE patients, who were matched 1:2 to controls, he said. By expanding both the UE sample and doubling the matched controls, the larger sample better reflected the actual UE and control populations, he said. “The new data still show a higher rate of delirium in UE patients, but the correlation between UE and delirium changed from not significant to statistically significant once the sample sizes became large enough,” Dr. Watson said.

“While the resulting increase in ICU length of stay and the apparent lack of any increase in mortality have been reported previously, the observed relationship between the use of dexmedetomidine and UE may be new and appears counterintuitive, as dexmedetomidine in prospective randomized controlled clinical trials has been associated with a lower risk of delirium, which at least in theory should lower the risk of UE,” they wrote.

They also believe that the large number of patients who did not require subsequent intubation after extubations shows a need for more research.

“The fact that 60% of patients in the UE group did not require reintubation indicates that our ability of predicting success of extubation remains poor,” Drs. Walz and Heard noted. “Further study of this topic is needed to enhance our understanding of factors contributing to UE, establish causality for delirium as a risk factor for UE, and to articulate management strategies for prevention of UE.”

When asked whether a patient who had an UE had been on mechanical ventilation for too long, Dr. Watson said it was difficult to know. Based on his study’s data, “it appears that patients who experience UE generally are less fit for liberation from mechanical ventilation than their controls,” he said.

“However, most critical care providers have had the experience of a patient self-extubating at a moment when planned extubation was not being considered, and the patient subsequently remains free of mechanical ventilation,” Dr. Watson said.

UE is associated with escalated and prolonged care, which increases costs and risk for complications over time, Dr. Watson said. “Based specifically on our current study, it seems logical that we could reduce UE by preventing or optimally managing delirium.”

He also said, “Caution is indicated here because it is not entirely clear whether these factors contribute to UE or are simply consequences of the pattern of care we typically provide to patients prone to UE.”