In an era when rehospitalization of patients with heart failure for episodes of acute decompensation has become a top target for reduction in the U.S. health care system via Medicare’s Readmissions Reduction Program CardioMEMS looks like the tool that every U.S. health care system and medical center has dreamed of having.
A wireless and battery-less implanted device, CardioMEMS guides management of stage III heart failure patients by allowing daily monitoring of patients’ pulmonary-artery pressure (PAP). Using these measures to fine-tune patient treatment with diuretics and vasodilators led to a statistically significant and clinically meaningful 37% relative reduction in heart failure rehospitalizations over an average 15-month follow-up, based on results from CardioMEMS’ pivotal U.S. trial as well as from subsequent secondary analyses. All-cause mortality was reduced and patients’ quality of life improved. And device-driven therapy yielded these benefits equally well in patients with either preserved or reduced left ventricular function as well as in a variety of other subgroups including patients with cardiac resynchronization devices or implantable cardioverter defibrillators, chronic obstructive pulmonary disease, chronic kidney disease, atrial fibrillation, pulmonary hypertension, or a history of myocardial infarction.
Yet in the first year after CardioMEMS received Food and Drug Administration approval and came onto the U.S. market, the device has taken a strikingly slow path into routine heart failure practice, according to heart failure cardiologists at several U.S. centers. A representative of St. Jude, the company that markets CardioMEMS, said in July that sales of the device during the first half of 2015 had involved more than 200 U.S. customers and slightly surpassed earlier expectations, and the company now anticipated full 2015 sales to run roughly 25% ahead of projections made at the start of this year. Despite that, the company’s second quarter report acknowledged the challenges that CardioMEMS faced during its first year on the U.S. market.
Numbers show a slow rollout
“I belong to a consortium of academic heart failure physicians who come from many of the major U.S. academic medical centers, and a recent straw poll of the members showed that close to no one was using it [CardioMEMS] on a regular basis, and the majority said they were not yet using it at all or in the process of starting their program,” said Dr. Javed Butler , a heart failure specialist who is professor of medicine, chief of cardiology, and codirector of the Stony Brook (N.Y.) University Heart Center.
At Stony Brook, no heart failure patient had received the device as of July 2015, although Dr. Butler said that his program’s use of CardioMEMS will probably start soon. Cardiologists at a handful of other U.S. centers report similarly slow starts.
At Massachusetts General Hospital (MGH) in Boston not a single patient has received CardioMEMS, though the heart failure staff there hopes to soon launch pilot use in 10 patients, said Dr. Kimberly A. Parks, associate director of the resynchronization and advanced cardiac therapeutics program. At the University of Colorado Hospital in Aurora, three patients have received the device since their first implant in the late winter, and the program is now on track to place devices in another one or two patients each month, said Dr. Natasha L. Altman , a cardiologist who heads the center’s CardioMEMS program. At Brigham and Women’s Hospital in Boston, which participated in the CHAMPION (CardioMEMS Heart Sensor Allows Monitoring of Pressure to Improve Outcomes in NYHA Class III Heart Failure Patients) pivotal trial that led to CardioMEMS approval, so far a “handful, fewer than 10 patients” received the device since marketing began, said Dr. Akshay S. Desai , associate director of Brigham Cardiovascular Consultants. Even at Ohio State’s Wexner Medical Center in Columbus, the program run by Dr. William T. Abraham , coprincipal investigator for CHAMPION, a relatively modest number of roughly 50 patients had received CardioMEMS in routine practice since the first patient received a device there in June 2014 following the FDA’s approval, Dr. Abraham said in an interview.
Contrasting with this level of use is the substantial number of patients who meet the CHAMPION enrollment criteria of New York Heart Association class III heart failure and a heart failure hospitalization within the prior year. Experts estimate that throughout the United States this group must number at least a half-million patients, and at the level of individual medical centers that provide advanced heart failure care, the numbers likely reach several hundred patients at each site.
Several challenges slow CardioMEMS implantation
Why such a slow start for what seems to be such an attractive device that clinicians uniformly praise for having a strong evidence base in CHAMPION and a compelling medical rationale? The answer seems multifactorial, including the need to convince skeptical hospital administrators and insurance payers to provide and pay for the device and follow-up care, a requirement to ramp up the patient-monitoring and management infrastructure, and the challenge of transitioning a relatively complex therapeutic formula from a successful clinical trial model into routine care.
“If this was a pill that you could prescribe to patients at hospital discharge, people would jump on it. But CardioMEMS is not as simple as prescribing a pill. There are a lot of logistical issues that make it very difficult,” said Stony Brook’s Dr. Butler in an interview.
This sentiment was shared not only by other cardiologists but also by St. Jude itself, as a company spokesman itemized several challenges the company encountered once it began trying to sell CardioMEMS commercially. “Developing the market for CardioMEMS will continue to take time,” said the St. Jude staffer during a July webcast on the company’s second-quarter 2015 performance. CardioMEMS’ sales were affected by the need to educate multiple constituencies, satisfy new-technology review committees, address reimbursement, access capital budgets, and create consensus among disparate stakeholders, the webcast said. In addition, the early St. Jude experience selling CardioMEMS showed that once a new customer signs a contract, “we find that customers tend to introduce CardioMEMS … [on] a pilot basis to gain experience with the technology and the reimbursement process.”
First is the challenge of selling a hospital’s administrative leadership on making an upfront capital investment in CardioMEMS equipment, giving the green light to performing procedures that just about break even relative to reimbursement, and then waiting to recoup the initial expenditure and perhaps make some money in the long term by avoiding readmissions and cutting lengths of stay. According to an analysis run by Dr. Parks of MGH, based on the CHAMPION results, for every 10 patients managed using CardioMEMS for 6 months, a center could expect to prevent nearly 15 patient-days in the hospital.
“Our administration is in support, but skeptical; I think that’s why it’s been slow to start,” said Dr. Parks. “The biggest limitation is the upfront cost of the device, and it’s not clear that the reimbursement will allow you to break even” when putting in devices, she said in an interview. “You could justify this by saying you’ll reduce hospitalizations, but the first impression from our administrators was that we were already doing a pretty good job limiting rehospitalizations so why do we need to add this?” The MGH leadership and clinicians eventually agreed on a plan to start the program with 10 implants and then analyze the results to decide if it makes sense to continue. Dr. Parks said she and her colleagues hope to have their first 10 patients implanted with a CardioMEMS before the end of this year.
Another hurdle at MGH was setting up the infrastructure so that a nurse could monitor patients and set in motion the alerts and treatment changes designed to normalize PAP normalized when it falls out of the target range. “It’s a lot of work to put the system in place to manage the devices,” Dr. Parks noted.
Dr. Butler echoed both these challenges. “You need to convince the hospital administrators and make a case based on the cost savings [later on during ongoing management] rather than positive revenue when you do the procedure. If you can expect future cost savings it’s a viable case to make, but a more difficult case to make,” he said. “You also take on the liability of monitoring patients” long term. “If you can follow several hundred patients there may be enough [follow-up] interventions to pay the salaries of staff ” who monitor the patients, but it is very difficult if you have a nurse who is monitoring five patients,” he said. Another issue complicating the economics is that the physicians who supervise monitoring are mostly not the same ones who performed the CardioMEMS placement procedure and received the procedure’s reimbursement. “These are the system barriers that are out there,” Dr. Butler said.
Dr. Altman in Colorado faced a different challenge. “We had good buy-in from our administration. Everyone is interested in reducing rehospitalizations so the administrators were very supportive. The major roadblock has been insurers. Medicare covers it , but so far Colorado Medicaid and several private insurers do not,” Dr. Altman said in an interview. The inconsistent pattern of insurance coverage has already meant that some heart failure patients in her program who were good clinical candidates for CardioMEMS could not receive the device. “I’ve had at least six or seven good candidates, but only three received the device because of insurance reimbursement issues,” she said.
But Dr. Altman expressed optimism that the coverage situation would improve as more programs start using CarioMEMS and insurers grow more familiar with daily PAP monitoring of heart failure patients. She noted that a new CardioMEMS program will soon start at a second Denver-based medical center, and she expressed confidence that ongoing pressure from physicians and administrators at both institutions will change the mind of officials at Colorado’s Medicaid program to provide reimbursement, and once that happens she expects the private insurers will change their policies as well.
To Dr. Abraham at Ohio State, one of the developers of the concept of using an implanted PAP monitor to guide management of heart failure patients, CardioMEMS slow take off is not surprising. “It provides physicians with daily information on a patient’s hemodynamics, which is something they never had before except in the catheterization laboratory, intensive care unit or cardiac care unit. Managing patients based on hemodynamics even in the absence of worsening signs and symptoms is a paradigm shift. It takes time to adopt new things.” He noted that most hospitals and health systems already have in place case managers or nurse navigators who run systems that have relied on the insensitive parameters of signs, symptoms, and patients’ weights. The same infrastructure should be able to fairly easily switch to focusing instead on PAP, said Dr. Abraham, professor and director of cardiovascular medicine at Wexner Medical Center.
He acknowledged that the upfront capital cost required to start a CardioMEMS program can poses a financial barrier at many U.S. centers, but once that start-up price is paid the actual implantation into each patient comes close to breaking even with existing reimbursements and the system should eventually result in a return on the investment in the form of reduced readmissions and keeping patients stable, he said. Dr. Abraham suggested that the possibility also exists that maintaining better stability in patients with class III heart failure and preventing episodes of acute decompensation through better-titrated fluid-volume control could produce a long-term change in the natural history of these patients, whose disease historically has been marked by progression to class IV heart failure and the eventual need for a left ventricular assist device, heart transplant, and death. Although this potential impact of refined treatment based on daily PAP monitoring remains to be proven, a secondary analysis from CHAMPION showing a 57% relative reduction in all-cause mortality over an average of 17 months of follow-up that Dr. Abraham reported at the American College of Cardiology annual meeting in March
The hurdle of routine practice
Transitioning CardioMEMS from its successful research track record to everyday clinical practice may pose the trickiest barrier of all. The consensus among heart failure experts seems to be that the best approach to do this successfully is to start slow, focus on the most rational patients within the broad enrollment criteria used in CHAMPION, and then gradually expand from that presuming the first wave of results from routine use at a particular site look similar to those from the trial.
“The next big question for PAP monitoring is can we replicate the trial’s success in routine practice? Can this be scaled up?” said Dr. Desai from Brigham and Women’s. “We’re still learning how to identify the right patients, the ones who’ll benefit,”
“The two types of patients we are primarily implanting now are those with classic class III heart failure who were hospitalized during the past 12 months and now are hospitalized again and are right in front of us. They are the lowest-hanging fruit because they are in the hospital today,” said Dr. Abraham. “The second group are the patients who come into the clinic complaining about their symptoms. What we are not yet doing is calling in all patients” with stage III heart failure and a history of at least one hospitalization.”
Other groups are taking a much more selective approach. Dr. Altman and her colleagues in Denver are only targeting patients who have been hospitalized at least twice within the past year, and more specifically patients recently rehospitalized within 30 days of their prior hospitalization. They are also focusing on patients expected to have a high level of compliance with the daily data-collection demands of PAP monitoring and the need to regularly adjust their medication dosages after receiving call backs from their clinicians. This criteria means that they are ruling out patients with a history of substance abuse, she said.
She also noted the need to tailor the target PAP to the specific clinical status of individual patients. Patients with mitral regurgitation, for example, will have a higher “normal” diastolic PAP and hence require a somewhat different target for stability maintenance. “You need to understand each patient’s baseline pressures and adjust their medications based on that,” Dr. Altman said.
“In real life patients tend to be older and sicker, so their benefit may be even greater than what was seen in CHAMPION, but perhaps the results will also be diluted because of comorbidities like renal failure or chronic obstructive pulmonary disease,” said Dr. Butler, although post hoc subanalyses of CHAMPION data showed that these comorbidities did not blunt the positive impact of PAP-guided treatment. “I don’t know if I’ll be successful in selecting the right patients, and whether my interventions in real life will produce the same good outcomes” seen in CHAMPION. “We’ll select patients who are close to the CHAMPION criteria, but not patients on dialysis [who were excluded from CHAMPION]; we’ll select patients with a modest degree of comorbidity and reasonable expected survival,” Dr. Butler said.
“If you had a drug that reduced rehospitalization rates it would be pretty clear how to proceed, but in this case it is not the device that reduces readmissions, it just gives you data and you need to act on the data,” Dr. Butler added. He acknowledged that the CHAMPION investigators used a relatively simple, straightforward management algorithm for patients whose PAP fell out of the target range, such as a diastolic PAP greater than 20 mm Hg or less than 8 mm Hg. The first management step is to raise or reduce the patient’s diuretic dosage, and if that fails to quickly normalize PAP the next step is to adjust the vasodilator dosage.
Managing patients in CHAMPION was “not totally an art” but neither was it “totally a science,” Dr. Butler said. Managing patients based on daily data collected using a CardioMEMS device will “require a little finesse,” he said. “It’s not straight science.”
Programs that have already used PAP monitoring on a routine basis, like those in Denver and at Brigham and Women’s, consider their experience too small in size and short in duration to draw any substantive conclusions regarding their success compared with the CHAMPION results. But Dr. Abraham, whose program has now placed CardioMEMS routinely in a few dozen patients and followed them for as long as a year, said that in his center’s patients the device and management system has produced outcomes that roughly match what he saw in the pivotal trial.
“The ongoing concern is that what was achieved in the CHAMPION trial may not be what is achieved in routine practice,” said Dr. Desai. The care that patients received in the trial is “a little of a black box,” he said, and during the FDA’s approval process some reviewers raised concerns that patients in the active-treatment arm simply received more intensive surveillance, although this concern eventually resolved. “There is not a lot of skepticism” among clinicians about this treatment, but there is uncertainly about exactly whom are the best patients to treat and whether their responses be as good as in CHAMPION, Dr. Desai said in an interview.
“The proof [of efficacy] will be in each individual center’s experience” using CardioMEMS, Dr. Altman said.
CHAMPION was initially funded by CardioMEMS, which then was acquired by St. Jude which is the company marketing the CardioMEMS device and associated hardware. Dr. Butler said that he had no current relevant disclosures, although in the past he had been a consultant to CardioMEMS (prior to its acquisition by St. Jude). Dr. Parks and Dr. Altman had no disclosures. Dr. Desai has been a consultant to and has received honoraria as a speaker on behalf of CardioMEMS and St. Jude. Dr. Abraham has been a consultant to CardioMEMS and St. Jude, was co-principal investigator on the CHAMPION trial, and is the principal investigator on a new St. Jude-funded trial studying CardioMEMS.
On Twitter @mitchelzoler
