Device uptake and development have progressed over the last decade, but few quantitative data exist examining the overall operating characteristics and temporal trends of these clinical trials. We performed a systematic analysis of all cardiovascular device clinical trials from 2001 to 2012 published in medical and cardiovascular journals with the 8 highest impact factors. Of the 1,224 identified cardiovascular clinical trials, 299 (24.4%) focused specifically on devices. Each trial included a median of 335 patients (162 to 745) recruited from a median of 14 sites (3 to 38) over a median enrollment duration of 1.9 years (1.2 to 3.3). Median enrollment rate was 1.1 patients/site/month (0.5 to 4.2). Most device trials targeted coronary artery disease (55.2%), followed by arrhythmias (17.4%). Most were industry sponsored (53.6%) and included mortality as a primary end point (69.6%). The median number of patients (225 to 499, p <0.001 for trend) and enrolling sites (11 to 19, p = 0.07 for trend) increased from 2001 to 2012. During the study period, multinational enrollment grew and approached 50% (p = 0.03), whereas trials enrolling in North America exclusively decreased from 30% to 17% (p = 0.10 for trend). Approximately 70% of device trials met their primary end points; this rate did not significantly change over time. In conclusion, this descriptive study of the contemporary cardiovascular device clinical trials highlights recent trends toward larger, more international trial programs. These aggregate data may help inform future cardiovascular device development.
Cardiovascular devices have an established role in ameliorating symptom burden, improving health-related quality of life, and reducing mortality and hospitalizations in a wide spectrum of cardiovascular diseases. Unlike contemporaneously developed pharmaceutical agents, cardiovascular devices have less well-established pathways of development and regulatory approval. Randomized clinical trials of new devices face unique challenges related to difficulties in blinding, biases associated with operator- and site-related learning curves, narrow eligibility windows limiting enrollment, rapidly evolving technologies, and global variation in regulatory approval and surveillance processes. These factors create major hurdles to expeditious and efficient completion of cardiovascular device clinical trials. Overall, approximately 1/3 of contemporary clinical trials fail to meet intended enrollment targets and another 1/3 require enrollment extension, curtailing the full realization of these novel devices. Device uptake and development have progressed over the last decade, but few quantitative data exist examining the overall operating characteristics and temporal trends of these clinical trials. Thus, the aim of the present analysis was to provide an overview of cardiovascular device clinical trials from 2001 to 2012 published in major medical and cardiovascular journals.
Methods
A 2-tiered search strategy was used to capture device-related randomized controlled trials with large potential influence on general cardiovascular clinical practice. Journals with the 8 highest impact factors in the subject categories, “General and Internal Medicine” and “Cardiology,” according to the 2013 edition of Journal Citation Reports, were screened for cardiovascular trials. These journals included the New England Journal of Medicine , Lancet , Journal of the American Medical Association , British Medical Journal , Annals of Internal Medicine , Journal of the American College of Cardiology , Circulation , and European Heart Journal . Articles were identified using (1) electronic search of the PubMed database with the terms “trial*” and “random*” in the aforementioned high-impact journals; and (2) manual search of each individual journal edition from the first issue in January 2001 to the last issue in December 2012. Publications of the primary results of phase II to IV cardiovascular clinical trials were reviewed. This initial screen excluded trials involving only pediatric patients, those focused on resuscitation, and trials with hospitals as units of intervention.
Cardiovascular trials evaluating diagnostic and therapeutic devices were then manually screened and selected based on the definition the Food and Drug Administration (FDA) uses for regulatory evaluation, which is included in section 201(h) of the Federal Food Drug and Cosmetic Act. Relevant interventional trials focusing on the following were excluded: (1) medications; (2) exercise, behavioral, or other lifestyle-based programs; (3) specific diets or supplements; (4) transitions of care or integration of health care; (5) cardiovascular surgeries; (6) gene therapy or intracoronary stem cells; and (7) single time-point diagnostic testing/imaging (without ongoing monitoring capabilities). The exclusion of clinical trials involving cardiovascular surgery in either arm limited the number of device studies related to valvular heart disease in the study.
The following data were abstracted from each cardiovascular trial publication: (1) journal, (2) year of publication, (3) disease state investigated, (4) study duration (based on trial start and end dates), (5) follow-up duration, (6) total number of patients enrolled, (7) total number of sites, (8) number of participating countries, (9) funding sources, (10) primary end points, (11) whether the trial met its primary end point, and (12) regions including enrolling sites. Duplicate trials were identified using the registration number. When data were not reported in the primary trial publication, supplementary data and ClinicalTrials.gov entries were further queried.
Trials were divided into four 3-year periods on the basis of publication date (2001 to 2003, 2004 to 2006, 2007 to 2009, and 2010 to 2012). Disease states were broadly categorized into: (1) arrhythmia, (2) coronary artery disease (CAD) including acute coronary syndromes and chronic CAD, (3) heart failure (HF)/cardiomyopathy including acute and chronic HF in inpatient and outpatient settings, and (4) hypertension/vascular/venous thromboembolism including systemic hypertension, pulmonary hypertension, stroke, peripheral vascular diseases, and venous thromboembolism. Primary funding mechanisms were categorized based on definitions specified in ClinicalTrials.gov: (1) industry, (2) government, (3) university or other nongovernment organizations, and (4) mixed if ≥2 of the earlier mentioned sources were identified. A “positive” trial was defined by the ability to reject the null hypothesis and support the alternative hypothesis for the primary end point (i.e., that intervention was either superior or equivalent/noninferior according to the primary hypothesis). Trials were divided into those conducted: (1) exclusively in North America (NA), (2) exclusively in Western Europe (WE), (3) exclusively outside NA and WE, and (4) mixed/multiregional.
Enrollment rate was calculated for each trial and expressed as number of patients enrolled per site per month. All categorical variables were expressed as n (%) and compared across groups using the chi-squared testing. All continuous variables were expressed as median (interquartile range) and compared across groups using Kruskal-Wallis test, 1-way analysis of variance tests, and Bonferroni-adjusted post hoc pairwise comparisons to maintain the family-wise error α = 0.05. Temporal trends across the predefined trial time frames were performed using nonparametric tests for trend. All statistical analyses were performed with STATA 12.0 (StataCorp, College Station, Texas).
Results
Of the 1,224 identified cardiovascular clinical trials, 299 (24.4%) focused specifically on cardiovascular medical devices, as defined by the FDA, enrolling a total of 224,401 patients. Table 1 categorizes the major types of devices by medical indication. Table 2 summarizes the major operating characteristics of these selected device trials. Variability was observed in the number of device trials conducted over time without a clear trend (p = 0.01). Device trials targeting CAD (55.2%) were significantly more common than other disease entities (p <0.001). Most device trials were funded through industry sponsorship (53.6%), whereas a minority were funded through government sources (9.2%, p <0.001). Mortality was the primary end point in most device clinical trials (37.8% as a single end point and 31.8% included in a composite end point). Approximately 70% of device trials met prespecified primary end points. Trials exclusively conducted in WE (44.5%) and those with mixed enrollment across regions (28.3%) were most common (p <0.001) during the 12-year period.
| Arrhythmia | Pacemakers |
| Implantable Cardioverter-Defibrillators | |
| Remote Arrhythmia Monitoring | |
| Coronary Artery Disease / Acute Coronary Syndrome | Coronary Stenting ∗ |
| Refractory Angina Devices | |
| Aspiration or Rotational Thrombectomy | |
| Ischemic Preconditioning | |
| Distal Coronary Embolic Protection | |
| Heart Failure / Cardiomyopathy | Cardiac Resynchronization Therapy |
| Implantable Cardioverter-Defibrillators | |
| Mechanical Circulatory Support | |
| Invasive and Non-invasive Hemodynamic Monitoring | |
| Ultrafiltration | |
| Non-invasive Positive Pressure Ventilation | |
| Hypertension / Vascular / Venous Thromboembolism | Peripheral Vascular Disease Stenting † |
| Pneumatic Compression Devices | |
| Thrombectomy in Ischemic Stroke | |
| Patent Foramen Ovale Closure | |
| Remote Blood Pressure Monitoring | |
| Non-invasive Positive Pressure Ventilation for Obstructive Sleep Apnea and Hypertension | |
| Distal Carotid Embolic Protection | |
| Mechanical or Ultrasound-Assisted Thrombolysis |
∗ Includes percutaneous coronary interventions with stent implantation with or without adjunctive pharmacotherapies, imaging techniques, or intracoronary interventions and trials focused on timing and associated on-site support of intervention.
† Includes stenting of carotid, aortic, renal, or peripheral vessels.
| Device Trials 2001-2012 (n=299) | p | |
|---|---|---|
| # of Patients per Trial | 335 (162-745) | |
| # of Sites per Trial | 14 (3-38) | |
| Enrollment Duration (years) | 1.9 (1.2-3.3) | |
| Enrollment Rate (patients/site/month) | 1.1 (0.5-4.2) | |
| Follow-up (months) | 8 (6-12) | |
| Journal | <0.001 | |
| Annals of Internal Medicine | 4 (1.3%) | |
| British Medical Journal | 2 (0.7%) | |
| Circulation | 62 (20.7%) | |
| European Heart Journal | 30 (10%) | |
| Journal of the American College of Cardiology | 75 (25.1%) | |
| Journal of the American Medical Association | 30 (10%) | |
| Lancet | 31 (10.4%) | |
| New England Journal of Medicine | 65 (21.7%) | |
| Year | 0.01 | |
| 2001-2003 | 71 (23.8%) | |
| 2004-2006 | 90 (30.1%) | |
| 2007-2009 | 57 (19.1%) | |
| 2010-2012 | 81 (27.1%) | |
| Type of Device | <0.001 | |
| Arrhythmia | 52 (17.4%) | |
| Coronary Artery Disease | 165 (55.2%) | |
| Heart Failure | 30 (13.4%) | |
| Hypertension / Vascular | 42 (14.1%) | |
| Funding Source | <0.001 | |
| Government | 23 (9.2%) | |
| Industry | 134 (53.6%) | |
| University/Organization | 78 (31.2%) | |
| Mixed | 15 (6%) | |
| Mortality as Primary Endpoint | 113 (37.8%) | |
| Mortality included in Composite Endpoint | 95 (31.8%) | |
| Positive Primary Endpoint | 207 (69.2%) | |
| Multinational | 108 (39%) | |
| Number of Countries | 1 (1-3) | |
| Region | <0.001 | |
| North America | 63 (23.2%) | |
| Western Europe | 121 (44.5%) | |
| Outside North America and Western Europe | 11 (4%) | |
| Multi-regional | 77 (28.3%) |
Each trial included a median of 335 patients (162 to 745) recruited from a median of 14 sites (3 to 38) over a median enrollment duration of 1.9 years (1.2 to 3.3). Table 3 provides a detailed analysis of enrollment trends and patterns across a number of device trial subsets. Industry- or mixed-funded trials enrolled greater number of patients (p = 0.03). Device trials with mortality as a primary end point (p <0.001) or included as a composite end point (p <0.001) enrolled more than double the number of patients as those that did not. Studies that met their primary end point included 298 patients (146 to 715) compared with 431 (239 to 828) in those that did not (p = 0.02). Device trials with multiregional involvement recruited the highest number of patients (p = 0.02).
| Number of Enrolled Patients | p | Enrollment Rate ∗ | p | |
|---|---|---|---|---|
| Journal | <0.001 | 0.50 | ||
| Annals of Internal Medicine | 888 (151-2355) | 3.7 (1.4-7.8) | ||
| British Medical Journal | 448 (340-555) | 1.0 (1.0-1.0) | ||
| Circulation | 281 (140-500) | 1.2 (0.4-4.1) | ||
| European Heart Journal | 202 (120-500) | 1.7 (0.5-7.1) | ||
| Journal of the American College of Cardiology | 206 (123-480) | 1.4 (0.7-5.4) | ||
| Journal of the American Medical Association | 390 (300-745) | 1.0 (0.5-2.5) | ||
| Lancet | 600 (231-1800) | 0.8 (0.4-5.6) | ||
| Annals of Internal Medicine | 715 (224-1572) | 0.8 (0.3-2.4) | ||
| Year | <0.001 | 0.19 | ||
| 2001-2003 | 225 (120-453) | 1.2 (0.5-2.5) | ||
| 2004-2006 | 283 (145-571) | 1.4 (0.5-5.1) | ||
| 2007-2009 | 426 (201-1005) | 1.0 (0.4-4.3) | ||
| 2010-2012 | 499 (200-1450) | 1.0 (0.5-4.1) | ||
| Type of Device | 0.26 | <0.001 | ||
| Arrhythmia | 188 (104-617) | 1.0 (0.4-1.8) | ||
| Coronary Artery Disease | 416 (202-1002) | 2.0 (0.9-8.4) | ||
| Heart Failure | 305 (149-660) | 0.4 (0.3-0.5) | ||
| Hypertension / Vascular | 238 (121-555) | 0.7 (0.3-2.2) | ||
| Funding Source | 0.03 | <0.001 | ||
| Government | 212 (126-500) | 0.6 (0.2-1.8) | ||
| Industry | 427 (178-909) | 0.8 (0.4-2.0) | ||
| University/Organization | 336 (180-778) | 4.2 (0.6-13.1) | ||
| Mixed | 1572 (96-2645) | 1.2 (0.8-2.5) | ||
| Mortality as Primary Endpoint † | 674 (335-1572) | <0.001 | 0.9 (0.4-3.7) | 0.15 |
| Mortality included in Composite Endpoint † | 625 (325-1798) | <0.001 | 1.2 (0.4-5.6) | 0.74 |
| Positive Primary Endpoint † | 298 (146-715) | 0.02 | 1.5 (0.6-4.9) | <0.001 |
| Multinational † | 467 (203-1199) | 0.003 | 0.7 (0.3-1.3) | <0.001 |
| Region | 0.02 | <0.001 | ||
| North America | 426 (219-1002) | 0.5 (0.3-1.4) | ||
| Western Europe | 257 (160-605) | 4.1 (1.0-13.5) | ||
| Outside North America and Western Europe | 292 (99-850) | 1.1 (0.6-6.5) | ||
| Multi-regional | 571 (269-1415) | 0.7 (0.3-1.2) |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
