| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2001;71:S204-S209
© 2001 The Society of Thoracic Surgeons
a Division of Cardiothoracic Surgery, Department of Surgery, The Milton S. Hershey Medical Center, Hershey, Pennsylvania, USA
b Presbyterian University Hospital, Pittsburgh, Pennsylvania, USA
c Johns Hopkins Hospital, Baltimore, Maryland, USA
d Division of Cardiovascular and Respiratory Devices, US Food and Drug Administration, Rockville, Maryland, USA
Address reprint requests to Dr Pae, Division of Cardiothoracic Surgery, MC H165, The Milton S. Hershey Medical Center, 500 University Dr, Hershey, PA 17033
e-mail: wpae{at}psu.edu
Presented at the Fifth International Conference on Circulatory Support Devices for Severe Cardiac Failure, New York, NY, Sept 15–17, 2000.
Abstract
Progress in the field of ventricular assist devices requires a more rigorous and systematic method of collecting outcomes data. A worldwide registry of device implants and results is proposed. With widespread participation, data from this registry would improve the identification of risk factors and complications, and allow for the creation of predictive models that would enhance patient selection. Professional societies should lead the development of a registry in close partnership with government and industry. Data collection using the Web, with rigorous security measures to protect patient privacy, would offer numerous advantages in efficiency and immediacy of communication for all participants.
In mid-1985 it was proposed by the International Society for Heart Transplantation (ISHLT) and the American Society for Artificial Organs that a combined registry for the clinical use of mechanical ventricular assist devices (VAD) and the total artificial heart (TAH) be established [1]. The registry was to be modeled after the transplantation registry then maintained by the ISHLT. At that time there had been an exponential increase in the use of temporary mechanical support for postcardiotomy cardiogenic shock. Scattered reports in the literature were encouraging in terms of both short- and long-term survival in patients who would otherwise be expected to die. A number of devices were in use, but the overall results achieved with these various support systems were unknown and comparisons were impossible. Overall cardiac and noncardiac disability, and device- and nondevice-related complications, were likewise difficult to interpret. Furthermore, during that same time, there was a great resurgence of interest in heart transplantation. Donor organ availability was becoming a problem. The initial encouraging results obtained with temporary ventricular support in the postcardiotomy cardiogenic shock group of patients led to the idea that temporary ventricular support may be complementary to heart transplantation. Again, it was unlikely that one institution could accumulate the number of patients necessary to provide data necessary for solid analysis. The first official report was in 1986 and contained data on only 83 patients [2]. The last report in 1995 analyzed data on more than 2,000 patients [3]. Inadequate funding of the endeavor led to its demise and an unsuccessful attempt was made to incorporate the data into The Society of Thoracic Surgeons (STS) National Database as a separate module. Since then left ventricular assist devices have proved valuable not only in sustaining life in the patient awaiting orthotopic heart transplantation, but also in alleviating the clinical syndrome of heart failure in many recipients of these devices. Although much experience has been gained with the use of left ventricular assist devices as a bridge to heart transplantation in patients with the most advanced heart failure, optimal patient selection and timing, linearized complication rates (device and nondevice related), and competitive technologies are still not completely known. A number of approved devices are available, but routine postmarket surveillance is insufficient to solve these issues.
Because of the epidemiologically significant demands on cardiac support technologies [4–6], in June 2000 a multidisciplinary conference jointly sponsored by the American College of Cardiology and a number of major national and international medical and surgical societies brought together physicians, scientists, the US Food and Drug Administration (FDA), and industry representatives to discuss the future role of mechanical circulatory support and to agree on the design principles of future trials [7]. A broad consensus was reached that there should be a mandatory registry for all implantable mechanical circulatory support devices. Issues that led to this conclusion included the lack of a information pool from which to determine:
Progress in this field requires establishment and maintenance of a mandatory registry that includes all implantable devices, both before and after approval. The combined effort of the various stakeholders is required to address issues of funding, data format and management, compliance, and access. The responsibility to support such registries should be allocated between industry and governmental agencies, and supportive surgical and medical societies representing the physicians involved in the use of these devices. It is rational to expect the two major medical societies with the largest stake of its membership in this area to take the lead once again. Thus the ISHLT, a nonprofit medical subspecialty society in existence since 1981 dedicated to the science and treatment of end-stage heart and lung disease with approximately 2,300 individual members worldwide, recently formed a scientific council to consider, evaluate, and address issues related to mechanical circulatory support. One of the charges for this Council was to oversee the creation and management of a database of demographic and medical data for all recipients of mechanical circulatory support devices. Therefore, proposals for the creation and management of this database were recently solicited based on a recent request for proposals (RFP). Certainly the stage has been set for development of a meaningful registry and a number of responses have been received. The STS is the largest thoracic surgical organization in the world, with more than 4,100 members. The STS has been involved in all issues of concern to thoracic surgeons, including the development of a National Database for Cardiac and Thoracic Surgery. The RFP is in concert with their objectives. Specifically, the purposes of a mechanical circulatory support device registry as outlined in this RFP would be:
Additionally, a number of logistical issues were identified in the RFP and stated explicitly, including:
Web-based systems
For health care applications, the Internet offers a dynamic mechanism for facilitating geographically separated collaborators on projects such as clinical trials and data registries. Rapid proliferation of Internet availability and usage over the last few years has strengthened this hope, and diverse potential participants in a device registry—surgeons, professional societies, industry, and government—could be brought together around this common medium to share in the process of data collection, analysis, and distribution. Internet technologies might help foster the collaboration that will be essential for a comprehensive device surveillance initiative.
The Web offers a number of advantages as a platform for data collection. Its ubiquitous availability allows participants to enter data from a wide range of clinical or office locations. Standardization of Web browser software across multiple computing environments reduces the complexity inherent in software distribution, allowing any user with Web access to have the necessary software for participation. The Web browser–user interface is both simple and widely familiar, thereby reducing the burden of user training. Direct interaction of the Web user with server-based software in front of the database offers an opportunity for real-time validity checking as data are entered by participants. And finally, the combined effect of these benefits should be to lower the overall costs of administrating the registry.
The Internet was designed to facilitate information exchange among mutually trusted entities, and although security provisions have evolved to meet the demands of e-commerce, solutions that are suitable for the financial industry may not be adequate for health care data. There is enormous sensitivity to clinical information, and patient safety is at risk if data become corrupted, transmitted incorrectly, or unavailable in emergencies. The low tolerance for data mistakes in health care make it a particularly challenging environment for implementing secure Internet applications.
Addressing the issue of security as the key to enhancing health care usage of the Internet, a recent report by a committee from the National Research Council [9] identified six elements of security:
Of these security elements, encryption methods are the best developed and widely used on the Internet. Secure socket layer technology is encryption methodology that has been embedded in Web browsers for several years, and has been used increasingly to transmit sensitive information such as credit card numbers. Strong authentication mechanisms are less widely available or standardized, and the most rigorous solutions (bimetric measurements such as fingerprints or voice analysis, or smart-card password generators) introduced increased costs and decreased convenience for users. The Intel Corporation and the American Medical Association recently announced an initiative to provide physicians with an authentication mechanism known as digital certificates, which would enable professionals to verify their identities when transmitting such items as prescriptions or test results. Widespread deployment of an authentication mechanism will be essential for Internet-based health care, and the registry project must select one of a number of mechanisms from an evolving array of technologies.
Precise legal requirements for security are also unclear at this time. The Health Insurance Portability and Accountability Act of 1996 directed the Department of Health and Human Services to promulgate privacy regulations if the US Congress failed to enact legislation. The final rule from the Department of Health and Human Services is still pending, but provisions are under consideration for permitting use of electronic health care data for a variety of research and surveillance initiatives. To what extent these provisions will facilitate or complicate the proposed device registry remains to be seen.
Despite these significant concerns about security, the transition of registries from paper-based and stand-alone software systems to Internet-based systems is highly likely over the next few years, and planning for this evolution would be wise. The considerable gains in efficiency and immediacy of communication for all participants are likely to outweigh the extra work required to implement rigorous security mechanisms.
Databases and mechanical support of the failing heart: the perspective of the Food and Drug Administration
The Food and Drug Administration’s role in medical device regulation
The FDA is charged with approving new devices only after demonstration in an appropriately designed clinical study of both safety and effectiveness for the intended function when the device used in accordance with the labeled indications. It is also within the FDA’s purview to monitor the performance of marketed devices. The FDA has two avenues for fulfilling its regulatory mandate to protect public health. These are the premarket approval (PMA) process and postmarket surveillance.
Premarket approval of mechanical heart and ventricular assist devices
Mechanical support for end-stage cardiac disease is a unique and invasive departure from the accepted treatment of this condition. The stakes for patients participating in research evaluating therapies for end-stage cardiac failure are extremely high. Mechanical ventricular assist usurps the function of the failing heart to ensure adequate circulation and thereby maintain homeostasis. Medical therapy, by contrast, attempts to enhance the output of the impaired heart by directly improving its function or reducing its workload. The established therapy for end-stage cardiac failure with proven efficacy is allograft cardiac replacement, which is constrained, however, to treating only about 2,000 of the estimated 20,000 candidates annually due to lack of donor availability. Furthermore, a larger population of patients exists with end-stage cardiac disease who are not candidates for heart transplantation and have no other effective treatment options and might benefit from permanent cardiac assist. Whereas devices such as intraaortic balloon pumps provide a limited degree of temporary circulatory support, clinical equipoise dictates that only implanted devices that have proved capable of providing ambulatory circulatory support should be used for destination therapy of end-stage heart failure.
The FDA has therefore required PMA of mechanical support for end-stage cardiac failure, as definitive treatment is considered only when transplantation is not an option. An application for this intended use must be supported with data from a study performed under an investigational device exemption controlled to maximum medical therapy, the present standard of care. Acknowledging the continuous change that is occurring in medical therapy, the FDA requires that approval for VADs demands a concurrent control for the clinical trial.
Postmarket surveillance of approved devices
The FDA PMA provides a snapshot of device performance. The study is undertaken in a relatively circumscribed research environment and, of necessity, over a relatively short period of observation. So for assessment of device safety and effectiveness in the real world of clinical practice, reliance is placed on the second regulatory arm available to the FDA in executing its function as a monitor of health care, namely postmarket surveillance. The tools available to the agency for postmarket surveillance are considerably less robust than are those accessible for the PMA process. Although the agency can request postmarket studies as a condition of PMA approval, considerable problems are encountered in enrolling the adjudicated sample size, completing the required study duration, and designing a study of sufficiently robust design once commercialization of the device is permitted. The agency must perforce rely for postmarket surveillance on the medical device reporting (MDR) and MedWatch programs for monitoring long-term device safety. These systems use passive reporting of the occurrence of device adverse effects. They are completely devoid of the clinical depth necessary for assessing clinical utility, for suggesting beneficial ancillary treatment, or of supporting any modifications to labeling, particularly the indications for use.
Notification by sponsors of adverse events related to their devices is a regulatory requirement that has excellent compliance. However, this practice is dependent on the responsible reporting of incidents by clinicians, an activity notably lacking in practitioners and a weak link in the reporting chain that significantly compromises both the MDR and MedWatch programs. The FDA has come to rely in considerable measure on peer-reviewed literature and anecdotal reports by clinicians for the earliest evidence of problems with approved devices. This was well exemplified by the recent recall of a cardiac valve prosthesis.
Registries as a postmarket surveillance instrument
Registries, depending on design, have the propensity for providing reliable data on device performance. They can be an open-ended source of data on the outcomes of interventions and can deliver these data in real time. Professional societies have been in the forefront of registry development. The STS National Database for Cardiac and Thoracic Surgery provides probably the best example of the effectiveness and benefits accruing from a well-crafted database established for the monitoring of surgical procedures. Various government agencies have also initiated a registry approach for perhaps less altruistic clinical imperatives, such as establishing surgical performance standards for reimbursement or accreditation purposes. The FDA’s Center for Devices has been interested in investigating the application of registries for postmarket surveillance and recently cosponsored a workshop on the subject. Indeed, the Center for Devices has entered into a limited arrangement with the STS for its database to assess some aspects of transmyocardial revascularization procedures.
Registry design requirements
The design of a registry can be flexible in terms of both inclusion criteria and the type of data collected. Participation can be encouraged by serving as a requisite for professional society membership or more coercively enforced by government fiat. Ventricular assist devices are possibly at the stage that coronary artery bypass grafting was in the 1970s. Having demonstrated the feasibility of the latter in the seminal CASS study, clinical experience began refining the specific role for the procedure. Now a variety of government agencies have instituted mandatory reporting to state-operated registries on coronary artery bypass grafting outcomes. This paradigm must be viewed as imminently reproducible, particularly for devices such as VADs and TAHs, devices carrying the burden of high visibility, high associated risk, and high economic cost. The preemptive introduction of a database by a professional clinical society that will address the legitimate needs of all who are party to the use of VADs is a preferred route to one that focuses narrowly on a single area of interest. The FDA believes that such registries could serve as a peerless source of data for postmarket surveillance.
The registry design that the FDA deems necessary for postmarket surveillance must generate data that are perhaps of lesser concern to other entities. Unlike third party payees or accreditation organizations for example, the agency is interested in device-specific information and long-term adverse event rates impacting on quality of life. Conversely, the FDA is constrained from any decision based on even peripheral economic factors, such as length of hospital stay, a matter of considerable concern to third party payees. The clinical fraternity’s interests relate more to registry data that allows fine tuning and perhaps expansion of indications for use, the determination of the role for appropriate adjunctive therapy, and developing criteria for and mode of possibly weaning patients from the device. The manufacturers of devices seek to learn about many of the same issues from a registry as does the Agency and the clinician. They can also hope that a suitably designed registry could support supplements to their PMA for approval of incremental technical improvements and for expanding labeling. It may additionally alleviate some of the administrative burden of MDR regulations or PMA conditional studies.
Although the use of registries holds promise of advantages for all parties involved with VADs, significant hurdles must be negotiated before these benefits can be achieved. Cooperation is required from all segments of the health care system, not the least being that between device manufacturers. The registry design should be such as to ensure capture of an accurate denominator forthe variables recorded. Patient privacy issues will need to be addressed as well as the defrayal of costs for the database.
The role of VADs in general and for management of end-stage cardiac failure in particular is probably now at the stage that coronary artery operation had reached in the early 1970s and prosthetic cardiac valve replacement had reached in the 1960s. As with these earlier innovations, well-crafted trials must demonstrate an acceptable role before extrapolation of the intervention can be permitted for general use, however circumscribed. Subsequent information, generally gleaned from observational studies once the procedure becomes a widely accepted therapy, may permit refinement of devices or treatments and may modify indications for use. Well-designed registries can facilitate this progression. The FDA believes that data from a suitably designed registry can provide a valuable asset to its regulatory activities.
Failure of concerned clinical entities to develop such registries for VADs and TAHs may leave parties with other than clinical concerns obligated to correct the deficiency in accordance with their agenda. In this event, the resulting registry design may provide databases of lesser clinical content.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
