What is delirium and how frequent is it thought to be in patients with advanced cancer?
Delirium is a terribly distressing syndrome of acute confusion. We often see it in the emergency department in older patients and in those with multiple medical problems. They may be acting strangely at home and concerned loved ones bring them to the emergency department.
_Page_18_Image_0001.jpg)
© MD Anderson
The difficult thing about delirium is that it's often intermittent. It may be present at one time, but not present at another. This makes it very difficult to detect, particularly for emergency physicians because it may express itself one minute, but be less apparent the next.
That led us in emergency medicine to think about how we can routinely, and in a standard or explicit fashion, recognize delirium when it's occurring.
In terms of how common it is, it really depends on the population one is dealing with. We recognize it most often in older patients.
I think delirium became most apparent to physicians initially in the Intensive Care Unit, where there are many disturbances such as bells ringing, lights being on at all times of day and sleep or wake disturbances that really unmask underlying delirium.
From there, we've come to recognize it in other places, particularly in the emergency department. For the past 10 years, there's been a concerted effort to better understand delirium and how early detection of delirium in the emergency department can help us prevent delirium from lasting a long time and try to correct underlying causes of delirium where we can.
Why have most delirium studies in cancer patients previously been limited to the palliative care setting?
Delirium is a common occurrence at the end of life. Those who practice palliative care see a lot of delirium and come to expect it. It’s a matter of being primed.
Palliative care physicians who work in hospice or in palliative care teams, know to expect delirium and, for that reason, they more commonly find it. The palliative care community, particularly those associated with cancer centers, helps oncologists recognize how common delirium is among patients, not only at the end of life but when they are experiencing tremendous disturbances due to cancer related pain that can be a cause of delirium, and in particular, the medications to treat pain can often contribute to delirium.
Palliative care has done a wonderful job of helping educate the oncology community about the importance of recognizing delirium and treating it.
Can you please outline your recent study which looked at the frequency of delirium in patients with cancer presenting to the emergency department?
Having spent my career in general emergency departments, I had a wonderful opportunity at MD Anderson in 2010, to establish the first academic Department of Emergency Medicine in a comprehensive cancer center.
_Page_23_Image_0001.jpg)
© MD Anderson
One of the early faculty members hired was a physician who'd been trained in palliative care. He ran the inpatient palliative care unit at MD Anderson and was a wonderful addition to our team. I'm speaking of Dr. Ahmed Elsayem, the first author of this paper.
When Ahmed came to join the faculty, we talked about these interests. He had been particularly interested in the distress that delirium causes patients. I got my interest in delirium from the general emergency medicine community and we recognized that no one had done a study of delirium in the emergency department, specifically in cancer patients.
MD Anderson has the largest emergency department dedicated to cancer in the world. We see approximately 24,000 patient visits a year and it was an ideal place to study emergent presentations in cancer. We obtained a small grant from MD Anderson and began planning the study.
As is usually the case, you run into many problems trying to do a study like this. We realized that we were studying a population that is vulnerable and often hard to enroll in clinical trials. Even though this was an observational, low risk study, it required a number of administrative and bureaucratic hurdles to pass in order to begin it.
We intended to examine the presence of delirium in all patients with cancer who were coming to the emergency department, realizing that many of these studies are really only carried out among older patients. It was an ideal place to do the study.
Which methods did you use to assess patients?
We routinely canvassed the emergency department with research assistants after training them in the use of the Confusion Assessment Method (CAM) tool.
We used one that had been validated in other emergency departments and, on a routine basis, our research staff visited the emergency department to involve patients.
We enrolled approximately 250 patients and found that if one performs an explicit standardized detection user assessment instrument, about 10% of our patients were actually delirious in the emergency department and that this was pretty much true, regardless of age.
Our younger patients with advanced cancer were presenting with delirium in proportions that were similar to those of older patients, which was a surprising finding.
I will give a caveat here. We could only really enroll those who could consent, or consent by proxy and it's likely that the 10% figure is a low estimate of the true prevalence of delirium in the emergency department for those with advanced cancer. You can imagine the difficult barriers to enrolling patients who are confused.
We also excluded a number of patients who had been in the emergency department a good time before we were able to capture them for an interview, in case those patients had developed delirium after coming to the emergency department. Thus the 10% should therefore be thought of as a minimal estimate of delirium.
Were you surprised by the number of patients with advanced cancer that showed evidence of delirium?
I'm not surprised at the proportion of older patients who presented with delirium, but we were surprised at the prevalence of delirium in younger patients.
Among geriatricians who studied those with cancer, there's an idea that having cancer, particularly advanced cancer, in some senses makes you appear to be an older person. Cancer itself ages one. It takes away a number of the neural safety nets and impacts on so many different body systems. It's easy for one of them to fail.
One can look at a middle aged person with advanced cancer and yet consider them as more of a geriatric patient, in whom many subtle things might be going on such as a disease state, syndrome, or symptoms. That is the result of multiple different parts of the body not working quite as they should.
We're saying that the cause of delirium in this case is really multi-factorial; it's hard to pin down to a single issue. This is a very common issue in older patients, where multiple things may be going wrong that contribute to distress or ills, but the same thing appears to be happening in younger patients with cancer.
What impact do you think your findings will have?
There are a number of questions here. Certainly, we've demonstrated it's feasible to perform standardized screening for delirium among those with cancer.
One can implement such a screening tool in an emergency department among a general population and among those who you might sense are at a higher risk such as cancer patients or the elderly.
We really did not answer the questions about how identifying delirium specifically allows us to intervene or prevent worsening of delirium.
One of the findings that the press might pick up on is that delirium was often not identified by the treating physician in the emergency department. In our study at MD Anderson, only a little more than half of patients with delirium were actually identified as such by the emergency physicians treating them.
The idea that, had these emergency physicians detected delirium, they may have changed their management strategies in such a way to improve the patient's outcome is really an open question.
It could be that our physicians are doing what they need to do: changing drugs, treating underlying causes, treating infection and achieving the same outcome that they might achieve even if delirium wasn't explicitly defined.
We often make the assumption that making this diagnosis is terribly important, but it may be that we're doing all the things that might improve on the condition, without actually identifying that it exists.
There is one future avenue here: the implementation of standard delirium screening in a given population in the emergency department to determine if making that diagnosis improves outcome.
Simply making a diagnosis doesn't necessarily lead to an improved outcome, and, in fact, if there's no way to improve that outcome or improve on the patient's quality of health, one can ask if it's necessary to make the diagnosis in the first place.
I think most of us in the field think it's important to do that, but we would like the explicit proof that it is the case.
What can be done to reduce delirium in advanced cancer patients?
There are many things that one can identify and treat such as sepsis or underlying electrolyte abnormalities. One of the things we found is that the presence of pain and the use of certain pain medications, opioids, might be contributing to delirium in many cases.
It wasn't uncommon that withdrawing the opioid and replacing one opioid with another opioid would improve delirium. Certainly, once having identified with delirium, one can start looking at certain factors that might contribute and take pains to identify those and correct them.
I think there may be a larger lesson here, which is that the emergency department can serve as an indicator of quality care prior to the patient arriving at the department.
In any health care system, once these patients who are routinely coming to the emergency department with a specific delirium syndrome associated with use of their medications, that is an opportunity for the emergency department to move upstream and educate the oncologists and others who are caring for those patients that this might be a problem.
The emergency department can say “You might want to look at this, you might want to be monitoring your patients more carefully before they reach this crisis point of having to come to the emergency department.”
That's part of a more system-based approach to oncologic care that incorporates the emergency department as a quality monitoring system to the extent that we can find patterns in treatment or in presentation and then educate our colleagues on how they might have prevented that visit.
No one wants to go to an emergency department. Going to the emergency department is generally a sign that something bad is happening.
I think work like this can be very useful. One of my colleagues, Amy Abernethy, talks about how we can use the data flow to a complex treatment system like this, to understand patient experience and feed back to earlier stages in the progression of a disease and thus prevent downstream complications. That's really our goal.
Having founded this department, I'm very interested in how emergency departments and the oncology community can work together. Early in my career, I was taking care of patients at Emory in the department.
One patient who had come to the department was in tremendous pain. I worked with her and her daughter to understand the cause of her pain from her underlying pelvic malignancy. I talked with her and tried to understand exactly how she got to this point.
One of the things that surprised me is that she had just come from her oncologist's office, which was across the street from the emergency department. I said "What did your oncologist say about the pain you're having?" The woman replied "I didn't tell my oncologist about my pain."
I thought well, why would that be? There's this interesting thing that happens in the clinic, in the office setting. I think patients go well prepared to their sessions with oncologists or another continuity provider, but often, they present themselves in the best light.
When one goes to the oncologist while experiencing a problem such as pain or something else, there's a tendency to cover that up. They cover up anything that might create more awareness that this is not going well and might be a sign that their cancer is progressing.
However, in the emergency department, those things tend to come out. People are driven to the emergency department by something that they can't ignore. There may be symptoms or other issues that lead to patient distress becoming apparent in the emergency department, whereas they might not become apparent in an office setting.
I think those different angles in terms of observing patient experience are one way that emergency medicine can really add to a healthcare system.
We've certainly seen that progress at MD Anderson with a new department within an existing cancer center. I think that's a rich area for discussion and for continued refinement of how we best deliver superior oncology care.
Where can readers find more information?
- Ahmed F. Elsayem et al. "Delirium Frequency Among Advanced Cancer Patients Presenting to an Emergency Department: A Prospective Randomized Observational Study." CANCER; Published Online: July 25, 2016 (DOI: 10.1002/cncr.30133).
- Knox H. Todd et al. “An Inflection Point in the Evolution of Oncologic Emergency Medicine” Annals of Emergency Medicine; Published Online: March 30, 2016 DOI: http://dx.doi.org/10.1016/j.annemergmed.2016.03.008
- Lawlor, P. G. (2016), Cancer patients with delirium in the emergency department: A frequent and distressing problem that calls for better assessment. Cancer. doi:10.1002/cncr.30132
About Dr Knox Todd
Dr. Todd received his medical degree from the University of Texas Southwestern Medical Center and completed his emergency medicine residency and research fellowship at UCLA.
After serving as Medical Director of the Grady Memorial Hospital Emergency Department and founding Vice Chair of the Department of Emergency Medicine at Emory University, he joined the faculty of the Albert Einstein College of Medicine and established the Pain and Emergency Medicine Institute in New York.
In 2010, he became founding Chair of the Department of Emergency Medicine at The University of Texas MD Anderson Cancer Center, the first academic department of emergency medicine within a comprehensive cancer center.
He currently resides in Mendoza, Argentina, and directs EMLine.org, a practice improvement initiative targeting pain management and palliative care in emergency medicine.
Dr. Todd conducted the first studies of ethnic disparities in analgesic use and has developed a network of collaborators in emergency medicine, pain medicine and palliative care who work with him to promote practice excellence in these areas.
He has an extensive research funding record and served as Associate Editor for the Annals of Emergency Medicine for 20 years. He has held past Board of Director positions for both the American Pain Society and the American Chronic Pain Association; served on the FDA Anesthetic and Analgesic Drug Product Advisory Committee (AADPAC); and is currently a member of the Executive Committee on Acute Pain Taxonomy for the Acute Pain Taxonomy, Analgesic, Anesthetic, and Addiction Clinical Trial Translations, Innovations, Opportunities, and Networks (ACTTION).
He has edited two textbooks: Culture, Brain and Analgesia, published by Oxford University Press in 2012, and Oncologic Emergency Medicine: Principles and Practice, released by Springer in June, 2016.
Triple-negative breast cancer patients with high immune cell levels have lower relapse risk after surgery