ACCEL: American College of Cardiology Extended Learning
ACCEL interviews and topical summaries of cardiology’s most interesting research areas

CardioSource WorldNews | The Conundrum of Cost-effective but Unaffordable Care: The Plight of High-tech New Interventional Therapies

We’re not #1! At least when it comes to health expenditures, the U.S. per capita rate of $9,146 is third (third!!) to Norway’s lead at $9,715 and Switzerland’s per capita rate of $9,276. (The World Bank lists the data [in U.S. dollars] from the World Health Organization Global Health Expenditure database: ) Granted, these are much smaller countries.

If you’re feeling competitive or miss no longer being first in per capita health care expenditures, then you will be happy to know that among major countries we remain #1 in total health expenditure as a percent of gross domestic product (GDP) (Table). However, we miss being #1 among all nations, beat out by tiny Tuvalu (formerly known as the Ellice Islands), a Polynesian island nation located midway between Hawaii and Australia. There you will find health care expenditures that are 19.7% of their GDP.

You probably have seen older graphics showing the U.S. as a resounding #1 in both categories, which certainly was the case. Our descent to #3 is a recent phenomenon; as of 2010, the U.S. was spending more per capita than either Norway or Switzerland—or anyone else, for that matter. And that had been the case since about 1980.

Also, you likely have seen the trends in deaths considered amenable to health care in people younger than 75 years. In an analysis of the U.S. and 18 other industrialized countries, investigators reported such deaths account, on average, for 23% of total mortality in this age group among males and 32% among females. The decline in amenable mortality in all countries averaged 16% between 1997-98 and 2002-03. The U.S. was an outlier, with a decline of only 4%. Had the U.S. reduced amenable mortality to the average rate achieved in the three top-performing countries (France, Japan, and Australia), then the U.S. would have realized 101,000 fewer deaths per year by the end of the study period.


This brings us to what has been called the current crisis in technology: cost-effective (based on historical measures) yet unaffordable care. Here are some numbers: if ICDs were used in patients shown to benefit in MADIT-II, the price-tag would be $15 billion per year. For left atrial appendage occlusion (LAAO), based on PROTECT-AF, the applicable annual cost for expanding use would be $13 billion. Throw in more patients receiving drug-eluting stents (DES; an extra $2.4 billion based on SIRIUS) and a wider use of transcatheter aortic valve replacement (PARTNER data and an additional cost of $3 billion), then these four interventional therapies would add $33.4 billion to annual healthcare costs.

These numbers apply to expanding established interventional technologies, but this problem is not confined to high-tech devices. Consider the new lipid-lowering agents, known as PCSK9 inhibitors: with approximately 2.6 million U.S. individuals who could potentially receive a PCSK9 inhibitor over the next 5 years, the total budgetary impact over that time period would be $19 billion (for those with familial hypercholesterolemia), $15 billion (for those who have CVD but are statin-intolerant), and $74 billion (if used for individuals with CVD but not at their low-density lipoprotein cholesterol target).

According to David J. Cohen, MD, director of cardiovascular research at Saint Luke’s Mid America Heart Institute, Kansas City, KS, there is already informal rationing in cardiovascular care, including limiting use of LV assist devices, carotid stenting, and transcatheter heart valves. Coming soon, he said, you might see limits placed on the use of PCI in stable coronary artery disease, renal stenting, LAAO, and perhaps others.

From a public health standpoint, there are data to support further expansion of spending on health care over many other areas, but there is a need for continued education of the public regarding the true “value” of medical technology. Dr. Cohen also noted that even the current economic environment will continue to support innovation over iteration: technologies that provide substantial benefit and fill truly unmet clinical needs are most likely to be covered and reimbursed.

He added that study designs should emphasize clinical benefit and focus on identification of optimal populations. Also, there should be a demonstration of economic value through “real world” studies that focus on outcomes that are relevant to patients and payers (survival, QOL, and lower costs of care).

Dr. Cohen added that treatments are not ‘cost effective’ unless they are truly effective. And for truly transformative technologies, the true value may not be immediately apparent.


  1. Nolte E, McKee CM. Health Aff (Millwood). 2008;27:58-71.

Messages from the DAPT Study
What Do You Need to Know?

Roxana Mehran, MD, FACC, professor of medicine and director of interventional cardiovascular research and clinical trials at Mount Sinai School of Medicine, New York, NY, puts the problem this way: “This is a daily question: When can I stop dual antiplatelet therapy [DAPT] safely after I’ve put in a stent?”

It’s been more than a decade since the FDA approval of the first drug-eluting stents (DES): sirolimus-eluting in 2003 and paclitaxel-eluting in 2004. Original recommendations for patients receiving first-generation DES advised 3 to 6 months of DAPT with aspirin plus a thienopyridine. After a warning of late risk in 2004 and alarming additional evidence in 2006, it appeared that the reduction in restenosis associated with DES was achieved at the cost of late—and potentially fatal—stent thrombosis.

In 2011, the ACC/American Heart Association (AHA) guidelines for percutaneous coronary intervention recommended a minimum DAPT duration of at least 12 months after DES, irrespective of clinical presentation.1 Yet, with evidence suggesting that DES use is associated with a late increased risk of catastrophic stent thrombosis at a rate significantly higher than with bare-metal stents (BMS), it was speculated that DES may require protracted and possibly indefinite clopidogrel therapy.

Indeed, at the 2014 Cardiovascular Research Technologies annual meeting, investigators from Thomas Jefferson University Hospital reported “very, very late stent thrombosis” occurring more than 5 years after DES placement. They found seven patients in their center with definite late stent thrombosis from 5.6 to 7.1 years after DES placement. None of the patients were taking clopidogrel and only two were taking aspirin at the time of their very, very late stent thrombosis. The interval between clopidogrel discontinuation and stent thrombosis ranged from days to months to years. The authors concluded: “There appears to be no definable ‘safe’ time when antiplatelet therapy may be discontinued without potential stent thrombosis.”

Drawing Lessons from DAPT

Kirk N. Garratt, MD, FACC, is the associate medical director of the Center for Heart & Vascular Health at Christiana Care Health System in Delaware. He was one of the coauthors of the Dual Antiplatelet Therapy (DAPT) Study,2 the only randomized controlled trial (RCT) sufficiently powered to assess rates of stent thrombosis and major adverse cardiac and cerebrovascular events (MACCE) after stenting. The study found that continuing DAPT from 12 to 30 months after DES placement provided important reductions in in-stent thrombosis and MACCE.

During the primary-analysis period (month 12 to month 30), all-cause mortality was 2.0% in the group that continued to receive thienopyridine and 1.5% in the placebo group, with a p value of 0.05. During the secondary-analysis period (month 12 to month 33), all-cause mortality was 2.3% versus 1.8% (HR: 1.36; p = 0.04) and the rate of death from noncardiovascular causes was 1.1% versus 0.6% (HR: 1.80; p = 0.01).

Yet a subsequent meta-analysis of 10 trials,3 published shortly after the DAPT results were announced, found that all-cause mortality was increased with longer DAPT despite the fact that stent thrombosis and MI were significantly reduced with this strategy. Nevertheless, this reduction did not result in a decrease in cardiac mortality with longer DAPT.

When all data are considered, Dr. Garratt said a cogent argument can be made for using just 3 to 6 months DAPT in patients treated with contemporary second-generation DES when the goal of treatment is to avoid stent thrombosis. Longer therapy should be recommended for patients treated with first-generation drug-eluting stents, for whom a persisting signal of risk is apparent, and for patients with low risk for bleeding who wish to minimize their risk of atherothrombotic events, both related and unrelated to DES.

Now, to assist clinical decision-making, the ACC/AHA have released a systematic review of DAPT for the 2016 PCI guideline update.4 The evidence is based on an analysis of 11 RCTs enrolling a total of 33,051 patients (mostly receiving newer-generation DES).

Six RCTs compared 18 to 48 months with 6 to 12 months of DAPT and the analysis of data found reductions in MI and stent thrombosis, no difference in major adverse cardiac events (MACE), an increase in major hemorrhage, and no change in mortality in the primary analyses. That might seem to be a positive finding, compared with the previously mentioned meta-analysis; however, secondary analyses of trials stratified by enrollment demonstrated “weak evidence” of increased mortality with prolonged DAPT in RCTs that successfully achieved their predefined enrollment targets.

As to the question about using extending therapy in patients more than 1 to 3 years after MI, this new meta-analysis, conducted for the guidelines, showed a significant reduction in MACE but an increase in major hemorrhage. Moreover, the evidence suggests that the net benefit of extending DAPT is not static but dynamic as a function of the bleeding and thrombotic propensity for each patient being treated.

Overall, the new analysis of RCTs suggests that patients undergoing safer, newer-generation DES implantation may be treated with a minimum DAPT duration of 3 to 6 months to prevent early and largely stent-related thrombotic events, but extension of DAPT beyond 12 months entails a tradeoff. The declining risk of late stent thrombosis with newer-generation DES and the inability to predict life-threatening bleeding limits the appeal of 18 to 48 months of DAPT over 6 to 12 months of therapy. In contrast, patients with prior MI at high risk of atherothrombosis experience fewer ischemic events with prolonged DAPT at a cost of increased bleeding events.

The systematic review has now been published,4 as has a new ACC/AHA PCI guidelines update.5 To see a review of the overriding concepts and updated recommendations for DAPT use and duration, please see the table below.


  1. Levine GN, Bates ER, Blankenship JC, et al. 2011 J Am Coll Cardiol. 2011;58:e44-122.
  2. Kereiakes DJ, Yeh RW, Massaro JM, et al. Lancet. 2015;385:2371-82.
  3. Bittl JA, Baber U, Bradley SM, Wijeysundera DN. J Am Coll Cardiol. 2016 [Epub ahead of print].
  4. Levine GN, Bates ER, Bittl JA, et al. J Am Coll Cardiol. 2016 [Epub ahead of print].

Prevalence of Resistant Hypertension is Increasing (or Not)

High levels of blood pressure (BP) are associated with premature death, stroke, cardiovascular events, and renal failure. Antihypertensive therapy reduces this risk, but some patients seem resistant to therapy.

According to 2005 to 2008 National Health and Nutrition Examination Survey data, 11.8% of U.S. adults with hypertension met the criteria for resistant hypertension: systolic/diastolic BP ≥ 140/90 mm Hg despite the use of antihypertensive medications from three different drug classes or drugs from four or more antihypertensive drug classes regardless of BP.1 This represents a doubling in prevalence from 5.5% in 1998 to 1994 and a nearly 40% increase from that reported in 1999 to 2004 (8.5%).

Michael A. Weber, MD, FACC, professor of medicine, State University of New York, Downstate Medical Center, recently offered advice to clinicians facing a patient with apparent resistant hypertension.2 In brief:

  • Is the patient taking the prescribed medications? Poor treatment adherence, he said, accounts for about half of all failure in BP control.
  • Is the treatment failure real? He said at least one-third of “office resistant hypertension” is actually white-coat hypertension that can be confirmed by ambulatory BP monitoring.3
  • Is an optimal treatment regimen being prescribed? Guidelines suggest that in the absence of a concomitant condition requiring particular drug classes or an established contraindication, patients with resistant hypertension should be on all three of these agents: 1) renin–angiotensin system blocker (angiotensin-converting enzyme inhibitor or angiotensin receptor blocker); 2) calcium channel blockers; and 3) thiazide diuretics.
  • Is the patient taking conflicting drugs? NSAIDs, cold remedies, weight loss medications, and antidepressants are among the drugs that can interfere with treatment and raise BP, sometimes sharply (Table).4
  • Does the patient have a secondary form of hypertension (e.g., chronic kidney disease, sleep apnea, aldosterone excess)? Studies have suggested that the prevalence of idiopathic hyperaldosteronism ranges from 17% to 22% of individuals with resistant hypertension.
  • Could an additional drug help (spironolactone can be a good first step)? Other options: beta-blockers; centrally acting drugs like clonidine; alpha blockers; or direct vasodilators, such as hydralazine.

To give you an idea of what you might expect from using these individual pieces of advice in a population of patients, Dr. Weber points to one study of 375 patients referred for specialist care due to uncontrolled hypertension despite being on three drugs.5 Of these:

  • 267 fell out of the study by maximizing doses and excluding white-coat hypertension;
  • 15 had secondary hypertension;
  • 17 were controlled on four drugs;
  • 40 were nonadherent (based on urinary analysis), with 30% taking none of their meds at all and the rest taking about half of what was prescribed. (At least there was consistency: lack of adherence was almost evenly distributed among different classes of antihypertensive drugs.); and
  • 36 were true resistant hypertensives (i.e., just 10% of the 375 referred patients).

This is not an isolated study’s findings. In one recent trial evaluating patients before being sent for renal denervation (RDN), top tier French specialists searched through 1,416 referred resistant patients to find 106 eligible subjects who truly had resistant hypertension (7.5%).6

Once you subtract all those factors that get a patient classified as having resistant hypertension, Dr. Weber suspects the prevalence of resistant hypertension in those being treated for high blood pressure is, in reality, about 1%. “It’s a pretty rare diagnosis,” he said.

Remember RDN?

While catheter-based RDN failed spectacularly in a sham-controlled study, the approach may be down but not out. Several reasons for the variability in response to renal artery denervation in humans have been posited, particularly limitations with the anatomic targets used for radiofrequency ablation in clinical trials.

Recently in JACC, Mahfoud et al. reported that increasing the number of radiofrequency lesions in the main renal artery was not sufficient to yield a clear dose-response relationship. However, targeted treatment of the renal artery branches or distal segment of the main renal artery resulted in markedly less variability of response and significantly greater reduction of both norepinephrine and axon density than conventional treatment targeting only the main renal artery for RDN.7

The investigators postulate that the shorter distance of nerve endings to the arterial lumen in the distal segment of the renal artery may account for the improved treatment efficacy. In an accompanying editorial comment,8 coauthored by Deepak Bhatt, MD, MPH, FACC, (first author the sham-controlled trial that brought an inglorious end to SYMPLICITY HTN-39), “It appears that the addition of distal lesion targets may be the most efficient approach for improving success in renal denervation, despite fewer nerves surrounding the distal vessel.”

Bhatt and his coauthor Neal N. Sawlani, MD, added that ongoing studies in humans are currently underway using new approaches with multielectrode catheters, incorporating an ongoing accrual of knowledge (NCT02439775, NCT02439749, and NCT02392351). They concluded, “With new methodological standards and novel preclinical studies, catheter-based renal denervation is poised to undergo significant innovation in device application. Renal denervation appears to have reached a new branch point in its development.”


  1. Mozaffarian D, Benjamin EJ, Go AS, et al. Circulation. 2016;133:e38-e360.
  2. Weber MA. Trends in Cardiovasc Med. 2015;25:755-6.
  3. de la Sierra A, Segura J, Banegas JR, et al. Hypertension. 2011;57:898-902.
  4. Calhoun DA, Jones D, Textor S, et al. Hypertension. 2008;51:1403-19.
  5. Jung O, Gechter JL, Wunder C, et al. J Hypertens. 2013;31:766-74.
  6. Azizi M, Sapoval M, Gosse P, et al. Lancet. 2015;385:1957-65.
  7. Mahfoud F, Tunev S, Ewen S, et al. J Am Coll Cardiol. 2015;66:1766-75.
  8. Sawlani NN, Bhatt DL. J Am Coll Cardiol. 2015;66:1776-1778.
  9. Bhatt DL, Kandzari DE, O’Neill WW, et al. N Engl J Med. 2014;370:1393-401.
  10. Weber MA, Kirtane A, Mauri L, Townsend RR, Kandzari DE, Leon MB. J Clin Hypertens (Greenwich). 2015;17:743-50.

The ‘Metastatic Cancer of Electrophysiology’
For long-standing persistent AF, how about empirical LAA isolation?

Longstanding persistent atrial fibrillation (AF) is the most challenging type of AF to treat with catheter ablation. During 5-year follow-up, Tilz and colleagues noted that of 202 such patients treated with circumferential pulmonary vein isolation (PVI), single- and multiple ablation procedure success was 20% and 45%, respectively.1 Compare that to the single procedure success rate seen in patients with paroxysmal AF (40% at 1 year and 30% at 5 years) and for multiple procedures (> 80% and > 60% at 5 years).2,3

In an editorial accompanying the paper by Tilz et al., longstanding persistent AF was referred to as ‘the metastatic cancer of electrophysiology’: It is one of the most difficult problems to treat and until recently, with no options, patients were expected to learn to live with their burden.4 The authors of the commentary, all from St. David’s Medical Center, Austin, wrote: “At least two-thirds of the population improved after a long follow-up. This is not an inconsequential number and would be seen as a major victory in cancer medicine.”

Several studies have shown that, in addition to pulmonary vein (PV) isolation, other areas may be the source of initiation and maintenance of AF in patients. The most common non-PV sites are the superior vena cava, the ligament of Marshall, the coronary sinus, the crista terminalis, the left atrial posterior wall, and the left atrial appendage (LAA).

Luigi Di Biase, MD, PhD, FACC, and colleagues think that the latter is an under-recognized trigger site of AF. In one study of nearly 4,000 patients, they reported that LAA appears to be responsible for arrhythmias in 27% of patients presenting for repeat procedures.5 They concluded that isolation of the LAA “could achieve freedom from atrial fibrillation in patients presenting for a repeat procedure when arrhythmias initiating from this structure are demonstrated.”

Empirical LAA Isolation

Dr. Di Biase is director of arrhythmia services, section head of electrophysiology, and associate professor of medicine (cardiology) at Einstein/Montefiore in New York City. He and colleagues recently reported their results treating patients with longstanding persistent AF using empirical electrical isolation of the LAA plus extensive PV antrum and non-PV trigger ablation (n = 85; group 1) versus extensive ablation alone (n = 88; group 2) in a multicenter randomized trial. (The effects of LAA isolation in addition to PV isolation has not been investigated before in a prospective randomized fashion.)

Empirical isolation of the LAA improved the long-term freedom from AF without increasing complications. Specifically, at 12-month follow-up, freedom from recurrence after a single procedure (and without anti-arrhythmic drug therapy) was seen in 48 (56%) of the patients with empirical LAA isolation versus 25 (28%) in the ablation-only group (p = 0.001). In group 2, about one-third of patients showed firing from LAA during isoproterenol testing but a sustained arrhythmia was observed in only eight of these patients and LAA was isolated in these patients.

Sixty-two patients (27 group 1 and 35 group 2) underwent a second procedure and LAA isolation was performed in all of these patients undergoing repeat ablation. After an average of 1.3 procedures, success at 24-month follow-up was 65 (76%) in group 1 and 49 (56%) in group 2 (p = 0.003).

After adjusting for age, sex, and left atrial diameter, LAA isolation plus standard ablation was associated with a 55% reduction in overall recurrence (HR: 0.45; p = 0.004).

The mean radiofrequency time was significantly longer with empirical LAA isolation (93.1 ± 26.2 minutes versus 77.4 ± 29.9 minutes; p < 0.001). But there were no significant differences in safety endpoints when LAA was empirically added to extensive ablation.

At ESC 2015, where the data were presented, the discussant for the trial was Professor Gerhard Hindricks, director of the department of electrophysiology at Leipzig University Heart Center, Germany. The results, he said, “are interesting and important as they add new information about the potential role of LAA triggers in patients with longstanding persistent atrial fibrillation.”

However, overall he considers the trial hypothesis generating rather than fully conclusive. Further studies are necessary, he said, before LAA isolation can be recommended as an integral part of catheter ablation of longstanding persistent AF.


  1. Tilz R, Rillig A, Thum A, et al. J Am Coll Cardiol. 2012;60:1921-9.
  2. Weerasooriya R, Khairy P, Litalien J. J Am Coll Cardiol. 2011;57:160-6.
  3. Medi C, Sparks PB, Morton JB, et al. J Cardiovasc Electrophysiol. 2011;22:137-41.
  4. Burkhardt J, Di Biase L, Natale A. J Am Coll Cardiol. 2012;60:1930-32.
  5. Di Biase L, Burkhardt JD, Mohanty P, et al. Circulation. 2010;122:109-18.
Read the full May issue of CardioSource WorldNews at

Clinical Topics: Arrhythmias and Clinical EP, Invasive Cardiovascular Angiography and Intervention, SCD/Ventricular Arrhythmias, Atrial Fibrillation/Supraventricular Arrhythmias

Keywords: CardioSource WorldNews, Aspirin, Atrial Fibrillation, Catheter Ablation, Drug-Eluting Stents, Health Expenditures, Pulmonary Veins, Pyridines, Sirolimus, Thienopyridines, Thrombosis, World Health Organization

< Back to Listings