Special Considerations for Resuscitation of Older Patients

February 19, 2020 – An editorial written by DCRI faculty discusses resuscitation, which may not always be the best option for older adults because of poor survival rates.

Resuscitation after out-of-hospital cardiac arrest should not be administered using a one-size-fits-all approach, especially when treating older patients, according to an editorial in Journal of the American Geriatrics Society.

Monique Starks, MDThe editorial, written by the DCRI’s Monique Starks, MD, MHS (pictured left), and Karen Alexander, MD (pictured right), accompanies a published study led by Patrick Druwe, MD, of Ghent University Hospital in Belgium. Druwe’s study, which surveyed clinicians about their perceptions of recent resuscitations of patients over 80 years old, found that more than half believed the cardiopulmonary resuscitation (CPR) was appropriate, despite very low survival rates for this population.

Patients older than 80 tend to be frailer and have more comorbidities that put them at higher risk for negative outcomes, Starks and Alexander write. Although they acknowledge the study may have been influenced by response bias and missing data, they point out that other studies have also shown poor survival rates for older adults after out-of-hospital cardiac arrest. Beyond survival, other factors should be important considerations in deciding whether to administer CPR, such as maintaining meaningful cognition and physical function for these patients.

Although “the time‐sensitive nature of out-of-hospital cardiac arrest makes it extremely difficult to create a solution that can be nuanced and patient oriented in the field,” the authors suggest Karen Alexander, MDthree considerations that could help in decision-making and reduce harm to older patients:

  • Better physician‐patient communication about death and CPR;
  • Increased use of out‐of‐hospital do‐not‐resuscitate orders; and
  • “Starting rules” for CPR to avoid futile attempts at resuscitation in adults older than 80, such as in the event of unwitnessed arrests (when no one sees the patient go into cardiac arrest, thereby making it difficult to determine how long the patient has been in arrest), nonshockable rhythms (when the heart is not pumping), or in nursing home settings.

Starks was also interviewed by The New York Times on the topic, along with Druwe. “As a resuscitation researcher, I want to save everybody,” Starks told The New York Times. “But I think we’ve entered this zone where we’re trying to escape ordinary death.”


DCRI Researchers Weigh in on Design of New Coronavirus Trials

February 17, 2020 – The DCRI’s Michael Pencina, PhD, and Sheng Luo, PhD, discussed two new trials beginning in China.

DCRI researchers recently discussed the design of two new Phase 3 clinical trials testing a potential treatment for coronavirus.

The DCRI’s Sheng Luo, PhD, explained that remdesivir, the treatment being tested in these two trials, was originally developed to treat the Ebola virus and has already gone through Phase 1 and Phase 2 trials.

Enrollment for the two trials is based on the severity of the disease. One of the trials is enrolling 308 patients with mild to moderate cases of coronavirus, while the other is enrolling 452 patients with severe cases. The primary outcome for the mild/moderate trial is complete recovery, while the primary outcome for the severe trial is recovery on a six-point scale.

Both trials are randomized and placebo-controlled. The randomization is important, said the DCRI’s Michael Pencina, PhD, to ensure that the treatment actually works. Without randomization, he explained, it might happen that the treatment is given to patients with less severe cases, leading to skewed results about the treatment’s effectiveness.

The study is also blinded, meaning that neither the patients nor the investigators know which patients are receiving the treatment and which are receiving the placebo. This is critical, Pencina said, to avoid bias within the study.

The trials are slated to run until April, but Pencina pointed out that an interim analysis, conducted before the trial is over, would be important, as time is of the essence in determining results. Investigators cannot be involved in analyzing the data as it accumulates because it could result in bias; therefore, an independent data monitoring committee would need to oversee such analysis. If the treatment looks promising at the interim analysis, he said, the study could be terminated at that point for overwhelming efficacy.

More than 1,000 deaths have already been attributed to coronavirus, a novel virus which emerged in Wuhan, China in December. The World Health Organization declared a global health emergency on Jan. 30. For Luo, the crisis is personal. He earned both his undergraduate and his master’s degree in Wuhan from Huazhong University of Science and Technology, and his in-laws still live in Wuhan.

Michael Pencina, PhD, and Sheng Luo, PhD, biostatistics professors from the Duke Clinical Research Institute, discuss two new Phase 3 clinical trials that are starting in China to test a potential treatment for coronavirus. Chinese and English subtitles are available for this video.

System-Wide Change Needed to Conduct More Randomized Clinical Trials at Lower Cost

February 12, 2020 – A focus on building a nationwide system of real-world data could provide the opportunity for more randomized clinical trials to generate needed evidence to inform health care decisions.

System-wide changes in the ecosystem of randomized clinical trials are needed to conduct trials more efficiently and at a lower cost, according to a recent paper in JACC written by current and former representatives of the DCRI.

Renato Lopes, MD, PhDThe assertions in the paper build on and respond to findings published in JAMA in March 2019 by a writing group that included the JACC paper’s co-authors, DCRI’s Renato Lopes, MD, PhD, former DCRI faculty member Robert Califf, MD, and former DCRI fellow Alexander Fanaroff, MD, MHS (now of the University of Pennsylvania). The March paper found that 90 percent of guidelines used to manage care for heart patients are not based on randomized clinical trial data, the gold standard for determining the efficacy of medical practices. In their more recent paper, Lopes, Califf, and Fanaroff explore barriers to conducting randomized clinical trials, and suggest solutions to overcome them.

A major challenge is that trials are typically funded by a company seeking data to support approval for a new medication or device. This structure creates “chasms in the evidence base” because it does not enable certain types of trials to be conducted, “including trials that compare treatment strategies, health services interventions, or the relative efficacy of approved medications and devices, evaluations of combinations of treatments, or trials intended to reduce the use, duration, or dose of a therapy,” the authors write.

The paper points to innovations that have emerged in the past decade in an effort to reduce costs, from adaptive designs that require smaller sample sizes, to virtual trials that can alleviate both cost and complexity. However, these advances have not moved the needle on the proportion of guideline recommendations that draw from randomized clinical trial data. “To move toward a world in which most clinical decisions are supported by high-quality evidence will require transformative, simultaneous structural changes in clinical trials and clinical care ecosystems,” the authors write.

The authors suggest that building randomized clinical trials within a robust real-world data system would create an environment in which multiple trials could use the renewable resource of data collected during routine patient care. Currently, limitations hinder the creation of such a system, including data inoperability challenges among health care systems and regulatory constraints.

In the U.S., the authors write, a culture change is also needed: Healthcare systems, hospitals, and healthcare providers must place more value on research participation and encourage their patients to participate in randomized clinical trials.

These changes will require collaboration and both public and private investment. The authors note that pharmaceutical companies would benefit from investing in building a system that will ultimately lead to reduced costs. The government should also participate in the creation of such a system, the authors write, because “the lack of high-quality evidence from RCTs to support most patient care decisions is an urgent public health issue that deserves an appropriately strong response.”

More Flexible Path Needed for Diabetes Drug Approval

February 11, 2019 – While cardiovascular safety remains an important consideration, evidence generated over the last decade shows that medications used for diabetes are relatively safe and often reduce cardiovascular risk, and they need an easier approval process in order to provide the most benefit for patients with diabetes.

While assuring cardiovascular safety of medications used to treat diabetes remains important, current FDA requirements for proving this safety could be revisited in light of evidence that has emerged over the last decade, say the authors of a recent paper published in Circulation.

Christopher Granger, MDThe paper shares takeaways from a February 2018 DCRI Think Tank, which convened representatives from academia, industry, regulatory agencies, and the NIH to consider guidance issued by the FDA in 2008 on glucose-lowering therapies. Under this guidance, glucose-lowering therapies cannot receive approval until evidence is provided that the therapies do not increase risk of major adverse cardiovascular events. This evidence is typically generated through cardiovascular outcomes trials, many of which have been completed since the guidance was issued. Attendees to the DCRI Think Tank, which was chaired by the DCRI’s Christopher Granger, MD (pictured left), and the BHF Glasgow Cardiovascular Research Centre’s John McMurray, MD, considered the guidance and its current relevance in light of the findings from many of these trials.

None of the cardiovascular outcomes trials completed since 2008 have demonstrated increased risk in traditional cardiovascular outcomes (a combination of cardiovascular death, non-fatal heart attacks, and non-fatal strokes) leading to discussion about whether the FDA requirements currently in place for glucose-lowering therapies may be modified for future studies. The cost of these trials can be prohibitive, which has resulted in “concern from the pharmaceutical industry that this enormous investment might act as a disincentive to the development of new treatments and thus lead to fewer options for patients with diabetes,” the paper’s authors write.

However, without the mandates for cardiovascular outcomes trials, the field would not have had the investment that has led to discovery that some of the newer glucose-lowering therapies are not only safe, but they have clear and compelling cardiovascular benefit for patients. These discoveries have already begun to change clinical practice and, in addition, have led to new trials studying potential benefits of these drugs in patients with heart failure and with advanced chronic kidney disease. One example of this kind of trial is EMPA-Kidney, for which the DCRI’s Jennifer Green, MD, serves as a principal investigator.

The paper’s authors strike a balance between these considerations by recommending greater flexibility in the FDA requirements in certain situations. Their recommendations include requiring at least one cardiovascular safety outcomes trial, conducting trials for longer durations, considering studying glucose-lowering therapies as first-line management of type 2 diabetes, and evaluating active drugs in head-to-head studies trials.

The paper also provides other recommendations, such as using pragmatic approaches to reduce the costs of cardiovascular outcomes trials, or gathering evidence through other methods like registries, which can ensure longer follow-up and may be more reflective of real-world clinical practice. Greater diversity is also needed in these trials, which could be achieved through mandating a certain percentage of underrepresented populations, or through expanding inclusion criteria for trials.

“As we approached a decade of implementation of this FDA guidance, it was important for us to gather stakeholders with diverse perspectives to discuss whether this guidance’s applicability now that we have generated more evidence,” McMurray said. “We made great headway in our discussion and produced recommendations that will help guide a more individualized approach in determining the safety and efficacy of each new glucose-lowering therapy.”

“The cardiovascular outcomes trials that have been completed over the last 10 years have added a lot of exciting new evidence to the field, which provided for a robust discussion at our think tank,” Granger said. “In addition to re-considering what the FDA guidance should require and how it should be applied to new glucose-lowering therapies, we need to implement these new findings into clinical practice to improve patient outcomes.”

Abhinav Sharma, MD, PhD“As we move forward from this discussion, another thing to consider will be the use of electronic health records and other technologies to help reduce the burden of cardiovascular outcomes trials in this space,” said former DCRI fellow Abhinav Sharma, MD, PhD (pictured right) (currently an assistant professor of medicine at McGill University, Montreal, Canada), who served as lead author for the paper. “While these methods need further research to validate their reliability, they do provide exciting potential to capture outcomes in a cost-effective manner.”

Other DCRI think tank attendees and contributors to the Circulation paper include Neha Pagidipati, MD, MPH; Jennifer Green, MD; Eric Peterson, MD, MPH; and Robert Califf, MD and Matthew Roe, MD, MHS (both formerly of the DCRI).

Abnormal Electrocardiography Despite Normal Stress Echocardiography Could be Warning Sign

February 5, 2020 – A DCRI study stratified patients based on their exercise ECG and stress echo results and examined patient outcomes to determine whether discordant results could be used as an indicator for higher risk of adverse outcomes.

Findings from a DCRI-led study recently published in JAMA Internal Medicine suggest new prognostic implications for patients who have abnormal exercise electrocardiography (ECG) but normal stress echocardiography (Echo) results.

Melissa Daubert, MDIn clinical practice, exercise ECG is combined with stress Echo imaging for the evaluation of coronary artery disease. Although it is not uncommon for patients to have discordant results (abnormal ECGs and normal Echos), it was previously unclear whether this had any prognostic significance for these patients. An observational study led by DCRI’s Melissa Daubert, MD, sought to answer this question.

The study, which included 15,077 patients without known coronary artery disease who underwent stress exercise testing at Duke University Medical Center, stratified patients into three groups based on their test results:

  • Negative ECG and negative Echo: This was the most common combination of results, with 85.5 percent of the study cohort.
  • Positive ECG and negative Echo: 8.5 percent of patients fell into this group.
  • Positive Echo: This group represented 6 percent of patients.

The median time each patient within the study was followed was 7.3 years. Researchers measured how many patients experienced the primary outcome of death, heart attack, hospitalization for unstable angina, or coronary revascularization. In the positive ECG-negative Echo group, 14.6 percent of patients experienced one of the primary outcomes, compared to 8.5 percent of patients with negative ECG-negative Echo.

When researchers drilled down into specific outcomes, they found that patients with positive ECG were more likely to experience death (5.9 percent versus 4.8 percent in the negative ECG group), as well as a heart attack (3.6 percent versus 2.2 percent).

The study team concluded that patients with positive ECG had a higher rate of adverse events, and that considering positive ECG may help to further risk stratify patients with normal Echo results.

“This study helped us to identify a new at-risk population,” Daubert said. “Although we often see patients with this combination of results, little was known about whether this had an effect on patient outcomes. Now, we know that patients with positive ECG and normal Echo are at higher risk for adverse events.”

Approval Rates for PCSK9 Inhibitors Remain Low Even After Positive Trial Results

January 29, 2020 – Neither prescription volume nor approval rates for PCSK9 inhibitors increased following the release of results from two large outcomes trials that demonstrated safety and effectiveness of this novel therapy.

A new study from the DCRI demonstrates continued challenges for the adoption of PCSK9 inhibitors for patients who need additional lipid-lowering therapy beyond statins.

The study team, which recently published its findings in Circulation: Cardiovascular Quality and Outcomes, analyzed electronic health record and pharmacy claims data from the Decisions Resources Group database. The study included more than 60,000 patients who had at least one prescription for a PCSK9 inhibitor between August 2015 and April 2018.

Of these more than 60,000 prescriptions, nearly 54 percent were approved within 60 days. Prescriptions approvals were more likely to occur for women, as well as patients who had government insurance, were seen by a cardiologist, or had a secondary prevention indication.

Prior to the publication of results from FOURIER and ODYSSEY, two major trials that involved PCSK9 inhibitors, many providers and payers claimed to be waiting for these findings before approving and prescribing PCSK9is in larger volumes. Although results from both trials supported use of PCSK9 inhibitors, the study team found that neither prescription volume nor approval rates for increased following the trial results, suggesting that “lack of outcomes data was neither a primary driver of clinician treatment inertia nor payer utilization management policies,” the paper’s authors write.

The authors also consider other reasons for the lack of uptick in prescription and approval rates. They suggest that payers may need more time to adjust their payer contracts and benefit plans in response to trial data, or that perhaps the results from the two trials were not impactful enough to lead to shifts in provider and payer behaviors.

Ann Marie Navar, MD, PhD“Recent updates to guidelines from the American College of Cardiology and American Heart Association emphasize the role PCSK9 inhibitors in secondary prevention, but to fully realize the benefits we’ll need to see greater uptake,” said Ann Marie Navar, MD, PhD, the lead author of the paper. “I’m optimistic that recent price drops and changes in utilization management practices will help drive broader access to these effective therapies.”

In addition to Navar, the DCRI’s Eric Peterson, MD, MPH, contributed to this study. The DCRI’s Hillary Mulder, MS, and Daniel Wojdyla provided support for the statistical analysis.

Improvements Needed for Assessing Treatment Effectiveness Using Real-World Data

January 28, 2020 – In a recent editorial, DCRI faculty call for improved heart failure trial enrollment, as well as modifications to the infrastructure used to evaluate whether treatments work in real-world settings.

A recent editorial in JACC: Heart Failure written by the DCRI’s Adam DeVore, MD, and Adrian Hernandez, MD, MHS, calls for improvements in enrollment for heart failure trials, as well as advanced methods for assessing effectiveness of heart failure therapies in real-world settings.

Adam DeVore, MDFindings from recent clinical trials have added to the number of therapies available for treating heart failure. However, because clinical trial populations are not always representative of clinical care populations, it is necessary to conduct additional studies to ensure the effectiveness of therapies in real-world settings. DeVore (pictured left) and Hernandez’s (pictured right) editorial accompanied results from one such observational study, which used claims data and was conducted by the Mayo Clinic’s Nicholas Tan, MD, MS, and colleagues.

DeVore and Hernandez commend the study, which found improved outcomes for patients treated with sacubitril-valsartan over those treated withAdrian Hernandez, MD, MHS ACEIs/ARBs. However, these improved outcomes did not extend to black patients, opening an opportunity for further investigation. “Resolving effectiveness in black Americans should be a top research priority,” DeVore and Hernandez write.

In the editorial, they suggest this anomaly could be a result of limited sample sizes, residual confounding, or selection bias, and they call for better ways to address these challenges, as well as improvements to the infrastructure for evaluating therapies in real-world settings. Current observational studies rely primarily on claims data, like Tan’s study, which lacks important information needed for phenotyping, or on registries, which can be expensive and challenging to maintain. Moving forward, DeVore and Hernandez write, the heart failure community must identify ways to leverage the electronic health record, which would provide in-depth information about heart failure treatment in a real-world setting.

The editorial also recommends an effort to improve enrollment so that future heart failure trials will be more representative of real-world populations. Although this solution is complex, engagement with patients, clinicians, health systems, and other stakeholders will be paramount.

Longitudinal Study Links Variable Blood Pressure in Young Adults to Increased Risk for Cardiovascular Disease

January 22, 2020 – Two DCRI researchers contributed to a study that points to a potential early warning sign for increased risk of cardiovascular disease.

A new study conducted by Duke and DCRI researchers reveals that young adults who have variable blood pressure readings may be at higher risk of cardiovascular disease by middle age.

Study results were published today in JAMA Cardiology. Lead author and Duke faculty member Yuichiro Yano, MD, PhD, said the findings suggest that current practice, which averages fluctuating blood pressure readings to determine whether medications are necessary, could be overlooking a potential early warning sign.

“If a patient comes in with one reading in December and a significantly lower reading in January, the average might be within the range that would appear normal,” Yano said in a news release issued by Duke Health News. “But is that difference associated with health outcomes in later life? That’s the question we sought to answer in this study, and it turns out the answer is yes.”

Yano and colleagues, including the DCRI’s Ann Marie Navar, MD, PhD, and Eric Peterson, MD, MPH, analyzed 30 years of data from a large, diverse cohort of 3,394 young people enrolled in the Coronary Artery Risk Development in Young Adults study between March 1985 and June 1986.

The main reading the study examined was the systolic blood pressure level, the upper number in the equation that measures the pressure in the blood vessels when the heart pumps. A systolic blood pressure reading over 130 is considered hypertensive and has long been a major risk factor for cardiovascular disease. The study team was able to identify which young people had variations in systolic blood pressure by the age of 35 and then follow them for 20 years to determine whether there appeared to be a correlating increase in cardiovascular disease.

The researchers found that each 3.6-mm spike in systolic blood pressure during young adulthood was associated with a 15 percent higher risk for heart disease events, independent of the averaged blood pressure levels during young adulthood and any single systolic blood pressure measurement in midlife.

Eric Peterson, MD, MPH“Studies like this one indicate that we still have a lot to learn about the world’s most common and modifiable cardiac risk factor, blood pressure,” Peterson said. “These data show that continuously monitoring blood pressure and looking for spikes and variations can add important prognostic information and potentially affect treatment decisions.”

In addition to Yano, Navar, and Peterson, study authors included Jared Reis, Cora Lewis, Stephen Sidney, Mark Pletcher, Kirsten Bibbins-Domingo, Michael Bancks, Hiroshi Kanegae, Samuel Gidding, Paul Muntner and Donald Lloyd-Jones.

The study received funding support from the National Heart, Lung, and Blood Institute (HHSN268201800005I, HHSN268201800007I, HHSN268201800003I, HHSN268201800006I, HHSN268201800004I, R01 HL144773-01, T32HL069771, K01HL133416).

DCRI Study Reveals Opportunities to Improve Resuscitation Care for Patients on Dialysis

January 21, 2020 – Despite gaps in resuscitation care between patients on dialysis and patients not receiving dialysis, study results showed similar survival rates for both groups.

A recent study led by the DCRI leveraged real-world data to shed new light on resuscitation for patients on maintenance dialysis when they experience in-hospital cardiac arrest.

Rates of survival for in-hospital cardiac arrest are low across the board at around 22 percent, and patients maintained on dialysis are at exceptionally high risk. Previous studies have suggested that patients receiving dialysis have lower survival rates compared to patients who do not receive dialysis, and some have questioned whether lower quality resuscitation care could help explain worse outcomes for this group.

Monique Anderson-Starks, MDThe most recent study, led by the DCRI’s Monique Starks, MD (pictured left), and Patrick Pun, MD, (pictured right), and published in the Clinical Journal of the American Society of Nephrology, showed similar rates of survival to hospital discharge between dialysis and non-dialysis patients. However, results did indicate opportunities for improvement in the quality of resuscitation care for patients on dialysis, as they were less likely to receive defibrillation within two minutes and had lower composite scores for resuscitation quality.

The study included 31,144 patients from 372 sites drawn from the Get With The Guidelines-Resuscitation registry, and 27 percent of the study cohort received maintenance dialysis. This data was also linked to Medicare and Medicaid data to enable additional analysis.

The finding of similar survival between dialysis and non-dialysis patients receiving resuscitation was Patrick Punsurprising and differed from previous literature, Pun said. This group was also more likely to have better neurological function at time of discharge than its counterpart. These differences in results could be attributable to the fact that this study uses a different data source—registry data rather than billing codes, which can lack specificity. The differences could also be explained by the study team’s efforts to avoid confounding—for example, matching patients from the different groups on several factors, including by year of event and by hospital.

“Although many factors should be considered when providers and patients discuss options in the event of a cardiac arrest, our study suggests that CPR is not a futile intervention for patients on dialysis,” Pun said. “Our findings also present the opportunity to further improve resuscitation outcomes in patients on maintenance dialysis by improving patient monitoring and resuscitation response times.”

Other DCRI faculty contributors to this publication include Eric Peterson, MD, MPH; and Roland Matsouaka, PhD. Staff statistical support was provided by Judith Stafford, MS and Jingjing Wu, MS (formerly of the DCRI).

Evidence from DCRI Study Supports Effectiveness of Digital Treatment for ADHD

January 16, 2020 – The study found that a video game-based treatment was effective both for children receiving ADHD medication along with the digital therapeutic and for children receiving only the digital therapeutic.

Evidence from a recently completed DCRI-led clinical trial supports the hypothesis that an investigational digital therapeutic may be an effective treatment for children with ADHD.

Daniel Laskowitz, MDAkili, the creator of the video game-based treatment called AKL-T01, announced yesterday results from the trial. The study, led by the DCRI’s Daniel Laskowitz, MD, (pictured left) with faculty support and expertise from the DCRI’s Scott Kollins, PhD, (pictured right) included 206 participants between the ages of 8 and 14 with a diagnosis of ADHD. Participants were assessed using the ADHD Impairment Rating Scale three times: at study launch before using AKL-T01, after using AKL-T01 for one month, and after using AKL-T01 for three months. Rates on the scale increased at the one-month mark, and additional increases were observed at the study’s end, suggesting that longer use of the therapy leads to further improvement.

The DCRI also led a prior study examining the efficacy of AKL-T01, but the most recent study was the first to study the effects of the treatment in both patients receiving a Scott Kollins, PhDcombination of AKL-T01 and ADHD medications and in patients receiving only AKL-T01. Both groups experienced similar improvements.

Akili is currently seeking FDA clearance for AKL-T01.

“As someone who has studied ADHD for my entire career, I am excited to see clinical evidence that supports the effectiveness of a novel treatment option for children with ADHD,” Kollins said. “Although there are already products on the market that claim to increase attention, AKL-T01 is the only product of its kind that has undergone the rigorous testing of a clinical trial. The results from our study provide clear evidence that users of AKL-T01 will actually benefit.”