Clancy J. Clark, MD, assistant professor of Surgical Oncology at Wake Forest School of Medicine, discusses "Utility of Consumer-based Activity Monitors in Evaluating the Frailty of the GI Cancer Patient."
Submit a question to Dr. Clancy Clark
View Doctor Profile CLANCY J. CLARK: And I just want to highlight, here, which everyone knows, is that pancreas cancer is a disease of the elderly. If you can see here, the median age is 71. Most of our patients who undergo this operation are actually in their sixties. We are skewing ourselves to the younger patient. And if you look here, there's a large majority of patients who are over 80 who have a diagnosis of pancreatic cancer. And some of the work that we've done has demonstrated that the normal pancreas cancer patient has significant impairment in their activities of daily living. We were looking at a group of over 300 pancreas cancer patients, and 41% of those patients had some impairment in their activities of daily living. That impairment was not due to their pancreas cancer. We were actually looking at pancreas cancer patient impairments in the years before their diagnosis. So we're talking about limitations in doing daily activities even before they get their cancer diagnosis. And what those impairments lead to are limitations in treatment options. And we learned about that earlier today-- that patients who have impairments in their daily living do worse when it comes to chemotherapy. And that's for obvious reasons, that they probably are bringing much more to the table than a patient has no impairments. And we also know that patients who have limitations in activities of daily living have a slower post-operative recovery, more complications, and more issues regarding getting back to their good quality of life that they had before the diagnosis. And part of that extension is, we looked at patients' activities of daily living before even know they had pancreas cancer, and asked, if you could put all that diagnosis aside-- if you could separate out their stage, and separate out their advanced stage, of separate out their comorbidities, and everything else they bring to the table-- do you just their limitations of activities of daily living actually predict their survival? And, in fact, it does. And this is very similar to some of the chemotherapy curves that you've seen this morning. It's not a dramatic difference, but there's a significant difference in survival when a patient has impairments of daily living within two years before their diagnosis. So a patient who comes to your clinic, new diagnosis pancreas cancer, new diagnosis colon cancer, who is elderly. And we just heard about this definition of what a frail patient is, and I'll go through the five characteristics of that frail patient. One is, they're shrinking, literally. They have less muscle mass. If you do a CAT scan and compare their muscle mass, their psoas muscle is smaller than an age match control. They're weaker. Their grip strength-- they're not able to grab and hold onto things as well as a patient who's not a frail cancer patient. General exhaustion, day-to-day, and everyone knows this when you talk to these patients. They are slow-- slow to get up from a chair, slow to get up on the exam table, slow to get around their house. Similarly, their overall activity level is diminished. They're burning fewer calories per week. They're staying in bed more often. That, in itself, all five of these come together to define what a frail cancer patients is. And, when you understand what a frail patient is, then you can understand how the parts of frailty can outweigh, maybe, the comorbidities that they bring. So in these, sort of, graphical images, here, you can see a patient has normal function, has no geriatric-like syndromes, normal cognition, normal nutrition, but has significant comorbidities. They actually may do better in their surgical outcomes, whereas a patient who has low commorbidities-- actually pretty healthy, they have no medical history significant-- but that poor functional status, cognitive impairment, some geriatric syndromes, or poor nutrition, they're going to have a worse outcome. And we know that, from numerous studies done, that if you have a frail patient, they will stay in the hospital longer. Their 30 day readmission rate is higher-- this is after surgery. Their complication rates are higher. And the likelihood that they're discharged to a skilled nursing facility is higher. And this graphic tries to get to the meat of what is exactly going on, and I'll use an example of the pancreas cancer patient. As their disease progresses, it causes a cyclical process where they'll have muscle wasting, which decreases their strength and power, their overall ability to walk, ambulate, activity decreases, their total expenditures decrease. They may have some other neuroendocrine effects of aging, or their disease, which then leads to malnourished state, which cycles back to worsening sarcopenia, or muscle loss. And this vicious cycle vicious cycle continues as the patient progresses along in their disease. And what we face, right now, is really poor insight into understanding the frail patient. Most of our evaluations are clinic-based studies. None of them are longitudinal. We will test a patient today, maybe wait a couple months and test the patient again. But they're usually in clinic. They are static images of that patient. Many of them are resource intensive. And some are difficult to administer. As an example, cardiopulmonary exercise testing, which most hospitals have, is a test that takes time and effort and expertise. Patients can do daily diaries, and say how well they're doing at home, but you have to rely on the patient to complete that form. There's also questionnaires, and there's many, many questionnaires that patients can have administered. Many of them are complex, many of them need to have a nurse administer them. And the question is, do we have better things out there today that can be simple tools to help understand this patient population? And that's where consumer-based activity monitors come into play. And we have a $9 pedometer you can get at the grocery store, to $200 complex pedometers to your cell phone, which is $400, $600, $800, which will measure steps, will measure your movement and activity. And some can be very complex. I'll just highlight, here, these are all the major commercial brands. There are, in fact, devices now-- this device, here, is called Adafruit. It's a little device that's about $6, and you can connect your own pedometer devices to it for another $2, and you could build your own pedometer. And they can mail you a kit for about $15. You can have your homemade one which you can program, and have something that is as good as iPhone watch, or Fitbit, or something of that sort. And what we're trying to do is, can we use these devices, through their perioperative period? Can we have a patient monitored prior to surgery, when they're diagnosed, and during a possible pre-habilitation phase, during their hospital stay-- after they recover both in the hospital, and after hospital-- and then, during their chemotherapy or recovers survivorship phase. Can these devices enhance, and help monitor, that period of time? And the concept behind this is that, a person is wearing some device, which is by Best Buy, that connects to the cloud, that then connects to a nurse-observed dashboard type system, where they can see a patient-- whether they're in the hospital, or out hospital, at home, in a retirement home, wherever. You can know how many steps they did that day, how long they slept that day, how many times they were up out of bed that day, whether they were wearing the monitor or not. And that can trigger feedback to the patient about, how are you doing today? You haven't been out of bed all day? Or, are you wearing your device, it hasn't been used for the last five days and the battery's dying. What's going on? And you can have easy interventions from a nursing level, or a nurse practitioner level, to feedback to the patient about, maybe, identifying patients who are at risk for complications from a procedure, chemotherapy, et cetera. Maybe you can use it as an early warning system. Maybe you can decrease post-operative complications. Maybe you can also decrease readmission. These devices are not FDA approved for monitoring this mechanism, and, nor do we know how they are used in an elderly population. And, if you talk to an accelerometry expert, they would argue that these devices are designed, and the algorithms that they're based on, are for normal middle-aged-- or, actually, 20-year-old-- adults. When we talked about an elderly population, their gait is different, they have a different sway. They might be expending more calories per step than a 20-year-old. So their step equivalents, being 2,000 or 4,000, might be the same number of kilocalories per day as a younger person might have for 5,000 steps. We don't quite understand that process. And so we've been doing a study that's a prospective feasibility study of, can we use these devices? And we're enrolling 38 patients. They're specifically older cancer patients-- 60 to 90-year-old. They have no impairment in ambulation or mobility, undergoing major abdominal cancer operations. And they have a battery of assessments, those classic in-clinic assessments. And then they're also monitored using a commercially-available Fitbit. And they also wear a Kenz Lifecorder, which is a more widely-accepted pedometer for scientists who focus on accelerometry data. From this, we've enrolled 19 patients so far. What we've learned is that patients are willing to wear these devices both pre-operative, in the hospital, and post operative. We can capture steps in elderly patients, which many people have questioned the ability to do that. We have also been able to complete many of the assessments and pre-operative/post-operative, and we're following these patients up to 90 days after their operation. And so far from this data we've learned that, comparing their pre-operative daily steps, there's a dramatic, as you expect, drop from what they were doing before their operation, to what they're doing in the hospital. And then, that persists, as we would expect, post-operatively. And, actually, with a very inexpensive device, monitor and measure changes in their activity. And I'll just give you a couple examples of patients, two examples. Both had no complications after their operation. And this is their step data from when they started wearing the device. And, you can see, some days, 2000 steps, to 3,000 steps, and high variability-- but only 2,000 to 3,000 steps. Now, if anyone knows anything about the news, and pedometers, and stuff, we're supposed to be a 10,000 steps. I would say, all the patients I've looked at their data in this study. None of them are over 5,000 steps. So we need to ask a question of is, we're dealing with a different population than what our pediatrist would like us to see. When the patient entered the hospital, absolutely no movement for the entire hospital stay. And we've asked patients, and they do get up, and they may ambulate to the bathroom, back to their bed, to the bathroom. Nurses will say, oh, they're walking the halls. But really they're not. And then, as soon as they leave the hospital-- and this patient's good example-- he barely resumed his mobility and activity. And still, weeks out, had not returned back to his baseline. And when we look at his other measures of mobility and activity, and a self-reported activity, he had yet, by three weeks, had not returned back to his baseline. And, if you ask, he had absolutely no complications from his operation. He recovered well, but just had a severe change in his overall activity as a result of the operation. Here's another patient, and we have less preoperative monitoring. We tried to achieve four days of monitoring prior to the operation. He was doing 2,000 to 3,000 steps per day. As soon as his operation he was hospitalized. He had absolutely no movement within the hospital. But he rapidly, over a several week period of time, restored his activity back to his baseline of where it was prior surgery. And this is Fitbit data. This is not a research-based accelerometer, but easily demonstrates that gradual rise in increased activity after surgery. And what we hope is to be able to map these expected recovery trends to those who have a delayed recovery. And can we detect differences? Can we identify patients that are more at risk for complications? And so, what we've learned from this study-- and it's an ongoing study-- is that consumer-based activity monitors, such as the Fitbit Zip, can measure both in-hospital and out-of-hospital activity, that these activities monitors are well suited for integration into our surgical oncology program. They're very affordable, easy to use. And, as a connected device, we have an opportunity to promote activity before surgery and after surgery. Many of these devices can give self feedback to the patient. Currently, the patients are blinded to the steps. They have no idea what they're doing day-to-day. But many of the devices that are out there, an integral part of that device is to be able to come back and see what you have personally done. And I think it's an exciting time, where we can use devices that don't cost thousands of dollars, and are not complex, and provide a more longitudinal picture of recovery after surgery. And this should be a point of intervention for future trials. Thank you very much.