Cognitive Dysfunction in ICU Patients

Every month a group of #SLPeeps blogs about research. For formation, you can find more out about it here.

For my previous post, I talked about mild TBI resulting from electrical injury. That was an article focusing on a general disorder resulting from a specific injury. This month, I found an article that was something of an extension of this idea: general disorder(s) resulting from general illness. I spend a good deal of time in the intensive care unit, and my current rotation is focused on medicine patients. I work in both the medical ICU as well as the medicine sub-acute units. I work every day with a population who we don't typically think of as being our target audience.

One of the challenges of acute care, as I've mentioned previously, is that the intensity of patients' illness prevents doing any meaningful therapy. This isn't a bad thing; we're taught in school to start therapy immediately, but I've learned that there's a caveat to this. Starting therapy early is helpful only once a patient is medically stable. Without that last component, therapy is bound to be anything but effective.

The article I selected this month is not a study itself, but rather a "narrative review". The authors noted that unlike other clinical outomes, "cognitive function in critical care survivors has not been deeply studied."

The Study

The primary medical diagnosis discussed in the study was acute respiratory distress syndrome (ARDS). ARDS can be associated with a number of other diagnoses, such chronic obstructive pulmonary disease (COPD) or sepsis, but the requirement of mechanical ventilation was emphasized in the review.

The authors wanted to compare cognitive impairment at time of hospital discharge and then again later on, all in the context of acute illness (i.e. other than neurological injury, which are more widely studied). "At hospital discharge, 70% to 100% of patients were determined to have cognitive impairment... At 1- and 2-year follow-up, the prevalence of cognitive impairment was 46% to 78% and 25% to 47%, respectively." Though there is a decline observed in prevalence, it's interesting that there is cognitive impairment at all; these are patients we often receive consults for as primarily swallowing patients due to prolonged intubation. What's even more interesting: "the domains of cognitive function most commonly affected were attention and concentration, memory, and executive function." The caveat is that the severity of deficits rated across studies, but it is interesting to note that these are all areas in which speech pathologists work with other patients.

Factors Associated with Cognitive Impairment

The authors note that pre-existing cognitive impairment is difficult to rule out, and that a certain amount of pre-existing cognitive impairment may be prevalent especially in elderly ICU patients. Conditions such as genetic predisposition to Alzheimer's dementia (apolipoprotein E4, or APOE4) demonstrate a "stronger association with duration of elirium than age, severity of illness score, sepsis, or benzodiazepine use". Other factors to consider are pre-existing psychiatric impairment, such as depression. The authors also note that 10-58% of survivors of critical illness suffer from depression.

The above risk factors may indicate an increased likelihood for developing cognitive impairment. "The pathogenesis of cognitive impairment following critical illness is not fully understood but may represent an accelerated neurodegenerative process tht develops in in vulnerable hosts." Delirium is becoming more widely understood to not simply be "confusion" related to illness, but rather something that can have more significant and long-lasting effects. "[P]atients who suffered a longer duration of delirium had greater overall brain atrophy and ventricular enlargement as well as smaller superior frontal lobes and hippocampal volumes 3 months following hospital discharge." There were further findings of loss of white matter in the corpus collosum and internal capsule.

Clinical Variables

The first variable listed is hypoxia, which is not foreign to SLPs. We often consider hypoxic injuries relevant, and that consideration transcends area of expertise. An additional component the others discuss is hypotension; while we so often recognize hypertension as a risk factor for such things as stroke or aneurysm (heck, hypertension even earns itself the casually tossed about HTN shorthand), hypotension could be seen as a red flag for possible anoxic injury.

Other variables noted included sepsis, dysglycemia, delirium, and sleep efficiency. Patients with sepsis were observed to have "cognitive and functional decline... [and had] deficits in verbal learning and memory and were seen to have significant reductions in left hippocampal volume compared with healthy controls." Even EEG results showed changes, reflecting more low-frequency activity, "indicating a nonspecific brain dysfunction." Dysglycemia refers to fluctuations in blood sugar levels. The authors noted that "patients with a highest blood glucose level (>153.5 mg/dL) and those with with greater fluctuations in blood glucose had three times the odds of being cognitively impaired at 1 year compared with patients who did not experience either glycemic condition." Associated impairment with dysglycemia and hyperglycemia included deficits in visuo-spatial skills.

As I mentioned above, delirium is becoming more widely acknowledged as a relevant condition that can have lasting effects. I've been noticing it being discussed more frequently on rounds among more and more medical teams. Deliriums is defined as "an acute change in mental status that is characterized by inattention and a fluctuating course... [and] it is associated with longer lengths of stay, increased duration of mechanical ventilation, and higher risk of death." The link between delirium, which was once considered temporary (and still is, in some ways), and cognitive impairment is "hypothesized to be mediated directly or indirectly through a systemic inflammatory response," leading to chronic neuroinflammation and neurotoxicity.

Finally, sleep is a large component of cognitive well-being. A common theme among many of my patients is how hard it can be to get good sleep. It's easy to see why this is: patients are frequently woken for vital signs throughout the night, carted away for tests, or simply get restless (being stuck in bed for days, weeks or even months on end is astonishingly hard on the body). Though the studies found didn't report cognitive outcomes, they did note that "sleep fragmentation (quantified by actigraphy) was associated with a nearly 1.5-fold increased risk of incident Alzheimer disease after controlling for demographics, total daily rest time, chronic medical conditions, and the use of medications."

Moving Forward

Though we often concern ourselves initially with swallowing safety when we first begin working with ICU patients, it would behoove us to not look too lightly at cognitive status. Early on, I developed a habit of monitoring cognition from the moment I start working with a patient so that I could monitor their progress. I used to think of it as a way to really see (and document) how my patients are doing, but now I have even more reason to do so. Sure, I might shrug off confusion as "par for the course", but I will be watching closely to see how long it takes patients to really clear.

I attended a lecture recently that discussed delirium, and one of the most interesting points discussed was how easing sedation, spontaneous breathing trials daily, and early mobilization were helpful in reducing length of time for mechanical ventilation as well as reducing delirium. If these ideas pan out as they're hypothesized to, perhaps cognitive function may be more spared. The authors also point out that improving sleep efficiency can impact recovery. With that in mind, I found myself this week advocating for a patient with sleep apnea to be able to use his CPAP (no order had been written, and though he had his machine, it had not yet been cleared for use); he was so exhausted he kept falling asleep during my evaluation. Not only do patients perform better with rest; it may also help prevent further cognitive deficits down the road.

As I learn more about delirium and its long-term effects, I see potential for how SLPs can be assets to multidisciplinary teams. I'm learning to see how duration of delirum is as important, if not more important, than the "severity" of the delirium. This new information may begin to shed light on why we encounter patients who present with cognitive deficits (especially "frequent flyer" patients) with unclear etiology; we may be seeing early markers of cognitive decline resulting from chronic illnesses that lend themselves to chronic delirium.

Article Citation

Wilcox, M. Elizabeth, MD, MPH, et al. (September 2013.) Cognitive Dysfunction in ICU Patients: Risk Factors, Predictors, and Rehabilitation Interventions. Critical Care Medicine, vol. 41, #9, S81-S98.