Predicting ICU Outcomes; Post-Stroke Brain Stimulation

TTHealthWatch is a weekly podcast from Texas Tech. In it, Elizabeth Tracey, director of electronic media for Johns Hopkins Medicine in Baltimore, and Rick Lange, MD, president of the Texas Tech University Health Sciences Center in El Paso, look at the top medical stories of the week.

This week’s topics include predicting outcomes for older adults in the ICU, use of aspirin for secondary prevention, an implant for stroke rehabilitation, and frailty and blood pressure control.

Program notes:

0:48 Predicting how older adults will do in the ICU

1:48 Top ten risk predictors

2:48 Fewer biases

3:56 Effect of frailty on intensive blood pressure control

4:56 Intensive treatment was beneficial

5:56 How about patient satisfaction?

6:11 An implant for stroke rehabilitation

7:11 Larger clinical trials needed

8:12 Proof of concept

8:25 Aspirin for secondary prevention worldwide

9:25 Only 40% received low-dose aspirin

10:35 End

Transcript:

Elizabeth: Can we predict outcomes for older adults in the ICU?

Rick: Are people who would benefit from aspirin actually getting it?

Elizabeth: Can we use a device to stimulate the cerebellum and improve motor function in people who’ve had strokes?

Rick: And should frailty affect the management of blood pressure?

Elizabeth: That’s what we’re talking about this week on TTHealthWatch, your weekly look at the medical headlines from Texas Tech University Health Sciences Center in El Paso. I’m Elizabeth Tracey, a Baltimore-based medical journalist.

Rick: And I’m Rick Lange, president of Texas Tech University Health Sciences Center in El Paso, where I’m also dean of the Paul L. Foster School of Medicine.

Elizabeth: Rick, if it’s okay with you, I’d like to turn first to The Lancet. This is an issue that faces me front and center in the chaplain role. It’s this question of, can we figure out how older adults are going to fare when they’re in the ICU?

This is a study that’s called ELDER-ICU. It’s an international, multicenter study and they also have subgroup evaluations in this. It’s an attempt to develop a model that would be able to predict early and accurate illness severity assessment so that appropriate decision-making for clinicians would be possible in providing care to these patients.

There are some factors that we know already are really important. Comorbidities, frailty, and decreased cognitive function do lead to a higher risk of death for those persons older than 65 years of age during acute medical events that may result in an ICU stay.

They developed this ELDER-ICU, which is a machine-learning model, to assess the illness severity of older adults in the ICU. They initially used data from 14 U.S. hospitals and then validated that in 171 hospitals from the U.S. and the Netherlands. Basically, what they found was that the top 10 risk predictors were the Glasgow Coma Score, urine output, respiratory rate, mechanical ventilation use, best state of activity, the Charlson Comorbidity Index score, the Geriatric Nutritional Risk Index, code status, age, and maximum blood urea nitrogen. When they put all those things together, they’re able to develop this model that can reliably predict the risk of in-hospital mortality. They tout this model as being helpful and accretive to this whole prediction because it uses routinely collected clinical features.

Rick: The value of this particular study is the large number of patients in whom the data set was derived — over 50,000. Prior to this, the largest study of elderly individuals predicting ICU mortality was only about 300 individuals. Some of the machine learning that was previously done actually unfortunately had some inherent biases that this particular one doesn’t have. It doesn’t have it because it’s not a single-center study, they used a lot more variables to look at, and it was an international study.

The other thing I would mention is, these are all things that are measured when the person first came into the ICU. This is the first day. That information on the first day — not the second or third or fourth or fifth — can actually provide information about who may not do well. We may be able to take it a step further and say, “OK, maybe they don’t have these risk factors early on, but maybe they trend towards them.” One of the things the study needs to do is it needs to look also what happens during the hospitalization, not only on the first day.

Elizabeth: Let’s note that this thing is called the XGBoost model, and they also compared it against other machine learning algorithms and clinical scores. I, for one, will welcome it because I think that there are many times when futility seems to be pretty apparent and if it’s possible for us to be able to identify that early, I think that that could really reduce a lot of suffering.

Rick: It can. Two purposes. One is to identify individuals that are not going to do well and also those that may need greater or earlier attention in the ICU.

Elizabeth: Right. Which of your two would you like to turn to?

Rick: Since we’re already talking about elderly individuals, let’s talk about the effect of frailty on the efficacy and safety of intensive blood pressure control. We’ve talked before about the SPRINT trial, and that was a trial that looked at trying to get the blood pressure below 130 systolic in older individuals. It was thought it could be too dangerous to do so and it wasn’t sure there was any benefit. In fact, they did show a benefit in terms of reducing cardiovascular outcomes, albeit there were some increased risks as well.

Well, in this post-hoc analysis they asked whether frailty should affect the intensity of blood pressure control. They had 9,300 patients aged 68, and about a fourth of them had frailty, i.e. they had some problem with global cognition, some self-related health or depression symptoms, some laboratory measurements, comorbidities, and decreased mobility.

Do the people with frailty respond just as well and does it in fact have the desired outcome in terms of preventing cardiovascular disease? Secondly, are they more likely to suffer side effects? Now, what they discovered is, patients with frailty are more likely to have a bad outcome. Nevertheless, the intensive treatment was beneficial.

Elizabeth: Let’s mention that this is in Circulation. I think that we’re working with these models as we have more people successfully aging into their 80s and beyond of, “What should we do with folks when they’re frail?” Many times that results in retrenchment and, especially with regard to this blood pressure issue, we see an awful lot about, “Gosh, should we continue with this intensive treatment in somebody who doesn’t look like they may necessarily benefit?” So I welcome this study.

Rick: Yeah. I think we’re all tempered because we recognize that people with frailty are more likely to develop cardiovascular events. However, we really weren’t sure whether they would receive benefit from intensive blood pressure lowering or whether they have unnecessarily increased risk of serious adverse events. Now we know that they actually do benefit and the risks of serious adverse events really aren’t higher.

Elizabeth: I’m just wondering. I guess in this study there was no assessment of how the patient felt.

Rick: Elizabeth, you’re right. This did not assess either patient satisfaction. If patients didn’t tolerate blood pressure medication because of a side effect or some serious adverse event, they were discontinued from the medication.

Elizabeth: You’re going to stay on your blood pressure meds whether you like it or not it sounds like.

Now, let’s turn to Nature Medicine. This was a hopeful study, even though it was extremely small. It is a look at cerebellar deep brain stimulation for chronic post-stroke motor rehabilitation. It’s just a phase I trial, but I was uplifted by this.

Upper extremity impairment after a stroke is a major therapeutic challenge. In this case, they applied deep brain stimulation into a part of the cerebellum that’s called the dentate nucleus, combined with renewed physical rehabilitation, to see if they could promote functional reorganization of the ipsilateral, the cortex on the same side where the injury occurred, in 12 individuals.

In these folks, what I thought was incredibly persuasive was that this was 1 to 3 years after their stroke that they enrolled these people and implanted this device. They were able to show that, sure enough, they were able to see robust functional gains. They are saying, “Wow, what we really need to do now is take a look at this strategy in larger clinical trials.”

Rick: This is this whole idea of what’s called neuroplasticity. It’s that there is either spontaneous or therapy-driven improvements in motor function after someone has had a stroke. In this particular case, as you said, they implant the device in the cerebellum on the same side of the stroke; those fibers actually cross over to the other side of the brain.

As you mentioned, these are individuals that had a stroke 1 to 3 years ago. The fact that they’re still having improvement is really pretty amazing. It seems that individuals that have some residual function at baseline seem to have the most improvement. The interesting thing is, Elizabeth, to get this small number of patients… and how many were involved in this?

Elizabeth: Fifteen individuals enrolled. Only 12 ended up staying.

Rick: Right. They actually screened over 11,000 patients. Some of them didn’t have sufficient information. Some of them didn’t have the inclusion criteria. They contacted 82; 67 declined. There were only 15 of the original 11,000 patients. So proof of concept; we’ll hear a lot more about it, but it may not be applicable to a large portion of the individuals that have had a stroke, unfortunately.

Elizabeth: We’ll be watching. Finally, let’s turn to your last one; that’s taking a look at aspirin. That’s in JAMA.

Rick: We know that aspirin is beneficial for what’s called secondary prevention of cardiovascular disease — people that have already had some evidence of cardiovascular disease. They’ve had a bypass surgery or had a stent placed or a heart attack or a stroke.

We know that low-dose aspirin can prevent a second event. That’s why it’s called secondary prevention. Aspirin is also routinely available. It’s incredibly inexpensive. The question is, how often do people that would benefit from it actually receive it?

This study looked at 51 low-income, middle-income, and high-income countries, surveys that were done in people that had cardiovascular disease. These surveys were conducted between 2013 and 2020: how many individuals that should be taking aspirin were actually receiving it?

There were over 124,000 individuals in this study, the mean age 52 years. About half of them were women. Almost 11,000 had a self-reported history of cardiovascular disease. Unfortunately, in that group only 40% were receiving low-dose aspirin. If you’re in a low-income country, only 17%. If you’re in a middle-income country, it’s between 24% and 50%. In a high-income country, about 65%.

This tells us that worldwide, aspirin is underused in secondary prevention, with less than half of people receiving it that should, and more importantly, that disparity is really significant in low-income countries where really only about 1 in 6 patients is actually receiving aspirin when all of them should be.

Elizabeth: Of course, we’ve got no excuses here. This is not a patented medication and it’s incredibly inexpensive.

Rick: Absolutely. It speaks a lot about the health systems in low- and middle-income countries and the education of individuals as well.

Elizabeth: More to address. On that note then, that’s a look at this week’s medical headlines from Texas Tech. I’m Elizabeth Tracey.

Rick: And I’m Rick Lange. Y’all listen up and make healthy choices.