Article Text
PDF
Statistics from Altmetric.com
As a result of the collaboration between scientists, physicians, the pharmaceutical and medical device industry, and overseeing regulatory bodies such as the Food and Drug Administration (FDA), the diagnosis and treatment of many diseases has advanced dramatically in recent decades. This partnership and process has helped to progress—and in some cases revolutionize—the neurointerventional (NI) care of those with devastating diseases such as complex or giant cerebral aneurysms and emergent large vessel occlusive acute ischemic stroke.1 Recent advances within our dynamic space is undeniably tied to medical device innovation. The cost for this progress in terms of time and capital is considerable. It comprises research and development at the front end, regulatory review and approval in the middle and manufacture and marketing at the tail end. This editorial comment is intended to briefly review the regulatory pathways and processes for NI device approval and monitoring that are overseen by the FDA, and also seeks to highlight the ever more apparent need for a more rigorous post-market device reporting system.
The vital partnership between clinicians, industry and the FDA has been at the core of advances in healthcare, which in turn have directly benefited thousands of patients throughout the world. As in most endeavors, it is important to periodically review these regulatory and monitoring pathways as opportunities for process improvement are potentially available. The voluntary recall of Penumbra’s Jet7 reperfusion catheter with Xtra Flex technology (Penumbra, Alameda, California, USA) has provoked debate regarding the current post-market device monitoring systems in terms of patient safety and device effectiveness, a topic of historic interest in this journal.2–5
Several regulatory pathways exist to bring a medical device to market and they are largely defined by a risk-based approach.6 These pre-market approval processes are overseen by the Center for Devices and Radiological Health (CDRH), which is a branch of the FDA. When a novel device is submitted to the FDA, the CDRH will determine the level of regulatory control necessary for a medical device based on its likely risks—that is, probable harms or discomforts—to a patient or user. The device will then be classified in terms of increasing level of patient or operator risk as class I, II or III. The regulatory standards or controls that are required to be met by the device naturally increase commensurate with its denoted class. Regulatory controls or standards are either general and apply to all devices (eg, good manufacturing, reporting of adverse events, device registration and listing with the FDA, labeling, adulteration and misbranding) or special/device-specific (guidelines and pre-market data, post-market surveillance and registries, special labeling requirements, training). Devices may come to market by pre-market notification (510 k), pre-market approval (PMA) or the humanitarian device exemption (HDE) pathway. A fourth pathway is via an Investigational Device Exemption (IDE), which is required to evaluate investigational devices in a clinical study. Data accrued from an IDE may be used by a manufacturer to evaluate the safety and effectiveness of a device occasionally prior to a 510 k or, more often, a PMA submission.
Post-market device surveillance and event reporting
The FDA may order manufacturers of class II or III devices to set up a system whereby an individual device can be tracked to the specific patient in which it was used. This system can permit the FDA and manufacturer to identify and rapidly remove devices from the market that are seen to be dangerous or defective in post-market reporting. The FDA can also require holders of a PMA or HDE to perform a 522 post-market surveillance study (522PMS). The FDA can require this type of study for any class II or III devices that meet certain criteria. The Comaneci embolization assist device is an example of a NI device with a 522PMS requirement (PS190001).7 8
For class II and III devices not covered by mandated tracking or a 522PMS, post-market reporting of adverse events involving medical devices falls under the responsibility of the healthcare facility and the manufacturer. Medical device reporting regulations require the manufacturer to report to the FDA any device-related patient deaths, serious injuries or device malfunctions that may have resulted in death or serious injury. Healthcare facilities are required to report patient deaths that are suspected to be device-related to both the manufacturer and FDA. Users are only required to report serious injuries to the manufacturer. Time points dictate by when a manufacturer or healthcare facility must report these types of events. There are two post-market surveillance systems currently in use to monitor adverse event reports:
Medical Device Reporting (MDR): The MDR system is used to monitor device-related adverse events including deaths, serious injuries, and device malfunctions. The system is intended to provide the FDA with significant medical device adverse event information from manufacturers, importers, and user facilities.
MedWatch: MedWatch is the process by which both mandatory and voluntary adverse events are reported. Healthcare facilities use the MedWatch form to report events. Healthcare practitioners and patients may also use the form to voluntarily report adverse events.
Medical device reports are maintained in the Manufacturer and User Facility Device Experience (MAUDE) database.9 This database consists of voluntary, user facility, distributor, and manufacturer reports, and the MDR database.
Established efforts to improve safety and outcome reporting
The approval process is destined to walk the tight rope between appropriately thorough scrutiny of novel devices that have a real potential to cause patient harm and, at the same time, expediting the availability of valuable technologies to physicians in service of their patients. The recent recall of the Jet7 Xtra Flex aspiration catheter highlights potential shortcomings in the present system. Devices that are fast tracked through the 510 k process by definition do not need to provide prospective scientific clinical evidence confirming safety and efficacy, except in special circumstances. This can lead to situations where the safety and effectiveness of a novel device is assumed to be the same as its predicate. The Jet7 with Xtra Flex technology is not the first recall of a 510 k approved product following patient harm or death. Another high-profile device recall was initiated in 2010 for an orthopedic implant. In 2005, the ASR XL Acetabular System (Depuy, Johnson & Johnson, Warsaw, Indiana, USA) was introduced to the US market. The metal on metal acetabular cup design was borrowed from a second device, the ASR hip resurfacing system, and fitted onto a predicate implant. It was deemed to be substantially equivalent to the prior implant and approved for use in 2005 via the 510 k process. However, despite initial experience showing benefit, longitudinal outcomes showed the device to have significant design flaws that led to high rates of revision surgeries (up to 60% within 5 years compared with the 15% industry standard) secondary to wear, adverse local tissue reactions and elevated serum chromium levels.10
Mandating pivotal trials to demonstrate the safety and effectiveness of all devices arguably could avert patient harm in some cases while almost certainly hurting other patients by denying them access to novel therapies and lifesaving interventions. The emergency use authorization (EUA) pathway and recent approval of novel COVID-19 vaccines by the FDA acts as an important reminder of the value of an expedited approval process for appropriately triaged interventions.11–18 An alternative and potentially superior mechanism for 510 k and HDE products would be the creation of a robust and mandatory post-market device surveillance process, specifically tracking safety and effectiveness outcomes for newly approved devices. Such a mechanism would help to identify safety issues early and would expedite the removal of that product from circulation if appropriate.
The MAUDE database provides good data on individual device malfunctions and events but is unable to determine event rates—which is a critical component in identifying outliers. This limitation is explicitly stated on the front page of the MAUDE database:
MDR data alone cannot be used to establish rates of events, evaluate a change in event rates over time or compare event rates between devices. The number of reports cannot be interpreted or used in isolation to reach conclusions about the existence, severity, or frequency of problems associated with devices. 9
The MAUDE database is also somewhat limited by a lack of standardized reporting metrics, which can prevent a reviewer from determining with certainty that the reported event was directly caused by the device in question.
In 2013, the FDA rolled out a 6–8-year process whereby all medical devices are assigned a Unique Device Identifier (UDI). This UDI allows the device to be tracked throughout its entire life cycle and has the potential to be a powerful tool for identifying event rates.19 As of 2018, all devices used in neurothrombectomy or clot retrieval have been assigned a UDI and manufacturers are now required to label medical devices with UDIs.20 These essentially function as a digital fingerprint containing information on model and production characteristics allowing a rapid and definitive way to identify a given device. Manufacturers also must submit key product information to the FDA’s global UDI database—that is, primary device identifier, UDI issuing agency, company name, branch, version/model number, device description, packaging information, sterility information, MRI safety/compatibility information. This database is accessible to both patients and providers. The present issue is that neither healthcare systems nor individual providers are required to implement and integrate UDI systems into billing, inventory, and electronic health record (EHR) systems. While EHR systems have been mandated to include fields for UDIs, healthcare systems are not incentivized to fill them out, markedly limiting the trackability of an individual device to researchers and regulators alike.
A further FDA strategy to address the issue of device surveillance is the creation of Coordinated Registry Networks (CRNs) through the National Evaluation System for Health Technologies (NEST). These are intended to gather real-world evidence by capturing and linking data from EHRs, claims data, industry and society-based registries, patient reported outcomes and clinical trial data. These individual 'super registries' are dedicated to a particular medical field and link devices, clinical outcomes and cost data together and are run through a partnership between physician specialty societies, leading clinical researchers, manufacturers and the federal government. One such NEST CRN called the Devices used for Acute Ischemic Stroke Intervention (DAISI) is dedicated to thrombectomy devices and was launched in 2017.21 However, at present DAISI remains in its infancy with limited capacity to effectively track safety and outcome data across the breadth of cases in which thrombectomy devices are now used, let alone the remaining growing list of devices and interventions offered in NI.
A potential solution
The question becomes: How can we improve device safety and effectiveness tracking beyond the current MDR and UDI system specifically in NI? An ideal solution might be to create a nationwide database of device usage combined with high quality data tracking of individual devices. Such a database would catalog selected adverse events and device malfunction rates as well as key procedural outcomes when specific devices are employed. This would, in effect, represent a universal 522PMS for any newly approved class II or III device. The number of units tracked and evaluated could be tailored to the class of device in question. Potentially, the first 1500 class II and 500 class III devices used or implanted following FDA approval could be closely monitored for both short- and long-term safety and effectiveness outcomes. The duration of follow-up might also be tailored by device type, and as an example with mandatory follow-up out to 5 years for those that are implanted or permanent. Any system proposed for mandatory post-market analysis would require independent oversight with strict regulations and guidelines ensuring submitted data to be fair, timely and accurate. Such a system will inevitably incur financial costs associated with its initiation, supervision, data collection and review. This could lead to a shift in the way funds are allocated for event sponsorships or post-market registries by manufacturers and possibly the creation of an industry-wide support scheme bolstered by supplementary FDA contributions. The author acknowledges the increase in administrative burden to both manufacturers and healthcare professionals associated with a mandatory post-market tracking system. However, such a system would add new robust protections for patients while ensuring ongoing device innovation, development and corporate profitability.
Conclusion
Recognizing that the current system of FDA clearance has led to healthcare improvement in millions of patients’ lives, the recent voluntary recall of Penumbra’s Jet7 Xtra Flex reperfusion catheter highlights the challenge of ensuring the timely availability of novel devices while also following an appropriately rigorous regulatory evaluation process that optimizes safety and effectiveness. Initiating a widespread overhaul of the 510 k and HDE pathways and/or mandating extensive prospective safety and effectiveness data prior to approval for all devices would lead to unacceptable delays in device clearances which would be detrimental to patient care. An approach designed to evaluate the safety and effectiveness of a novel device based on comparison with an 'equivalent' device predicate has been shown to be hazardous. A robust post-market surveillance mechanism for 510 k devices and humanitarian use devices (HUDs)would potentially help to define safety and effectiveness outcomes and, importantly, provide more definitive event rates allowing for suspect devices to be flagged and removed from circulation more rapidly. The author of this comment readily acknowledges that the proposal may cause challenges for an already overburdened healthcare system and add cost; however, it is hoped that such a database would improve patient safety and potentially stimulate further device development.
References
Footnotes
Twitter @rdeleacymd
Contributors RDL is the sole contributor to this commentary.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Patient consent for publication Not required.
Provenance and peer review Commissioned; internally peer reviewed.