Global trends in post-market surveillance of high-risk medical devices: An empirical analysis based on regulatory data

Materials & Methods

This study was undertaken by the department of Neuroscience, National University of Ireland Galway, in collaboration with the department of Chemistry, SNS College of Technology, Coimbatore, India with additional contributions from industry research in biomedical engineering. The study period was from March 2024 to February 2025. The analysis was based entirely on secondary data from publicly accessible regulatory databases (EUDAMED, MAUDE, BfArM, and ANSM).

Results

The distribution of risks associated with medical devices across product categories and manufacturers is shown in figure 1. Manufacturer 4 had the most significant frequency of risks involved, especially in surgical/medical equipment and implanted devices, while Manufacturers 1 through 5 exhibited differing degrees of impact. There were fewer issues reported in general-use gadgets, and other manufacturers were spread more evenly throughout categories.

Close

Fig. 1.

This figure visualizes the affected product categories within medical device classes (Class IIa, IIb, III) and their respective risk counts, illustrating the variability of risk distribution across device types.

Export to PPT

The impaired product categories for Class IIa, IIb, and III medical device classes are also depicted in Fig.1. With six occurrences, Class III cardiac monitoring and implanted devices had the most significant risk percentages. Class IIb has greater vulnerability due to the prevalence of infusion pumps/system issues and various implanted and specialised devices. Class IIa exhibits significant risks in surgical/medical devices as well as diagnostic and monitoring devices. A unbalanced breakdown of risks is shown in the graph, with Class III devices being more commonly implied because of their more advanced complication and vital role.

The five clusters of medical device hazards hardware/mechanical failures, calibration/accuracy concerns, software problems, contamination risks, and patient safety risks are shown in figure 2. With 25 incidents, hardware/mechanical issues were the most common among them.

Close

Fig. 2.

Distribution of identified risk types across five major device failure categories.

Export to PPT

A summary of hardware failures as the main risk across different medical device categories is shown in Table. The most common problems, such as implant corrosion as well as premature wear on equipment like hip implants, are reported by orthopaedic and implantable devices, underscoring serious material fragility issues. The significance of dependable energy sources is highlighted by problems like declining battery life and broken blood pumps that follow cardiac monitoring and implanted devices. Risks during procedural usage are indicated by the major hardware issues with invasive and diagnostic devices, such as stapler misfiring and catheter breakage. While many implantable devices indicate issues including electrode deterioration and device migration, diagnostic and monitoring tools reveal moisture-related device malfunctions, highlighting the need for meticulous quality assurance among all categories.

Field modifications accounted for the largest proportion (46%), followed by software updates (26%) and recalls (22%) (Supplementary Fig. 1). As shown in supplementary table II and supplementary figure 2, the most common user actions included device replacement, performance monitoring, and software updates. Regulatory notifications were primarily directed to the FDA and relevant national health authorities.

Discussion

This study analysed incident reporting and Field Safety Corrective Actions (FSCAs) for Class IIb and III medical devices across four major regulatory databases: EUDAMED, MAUDE, BfArM, and ANSM. The findings revealed that hardware failures remain the most frequent category of post-market incidents, particularly in orthopaedic implants and cardiac devices, while software-driven devices such as insulin pumps demonstrated high recurrence rates despite FSCA interventions. Field modifications were the most frequently implemented corrective action, and although generally effective in reducing recurrence, variability in FSCA outcomes highlighted gaps in current regulatory strategies. Regional discrepancies between databases also underscored the lack of harmonisation in incident reporting and transparency across jurisdictions^1,10,14

These observations are consistent with prior literature documenting challenges in high-risk medical device safety and the growing complexity of managing failures in both hardware- and software-based systems^1,14–16. Studies on orthopaedic implants and cardiac monitoring devices have similarly reported durability concerns, infections, and device-related mortality risks^15,16. The persistence of complications in software-based tools aligns with reports on Software as a Medical Device (SaMD), where calibration issues and algorithmic limitations reduce FSCA effectiveness^11,17. The findings also reflect broader regulatory discussions emphasising proactive surveillance, as outlined in EU MDR Articles 83 and 88 and related MDCG guidance, which stress trend reporting and continuous monitoring as essential to patient safety^9,18. This analysis highlights the significance of customised regulatory actions for high-risk manufacturers and illustrates how specific manufacturing practices can directly influence device dependability^19,20. These findings are consistent with prior reports emphasising the role of manufacturing quality and process control in minimising post-market device failures.

A key strength of this study lies in its comparative use of multiple international databases, offering a cross-regional perspective on incident trends and regulatory responses. However, limitations must be acknowledged. First, reporting practices varied significantly across EUDAMED, MAUDE, BfArM, and ANSM, which may affect data consistency and comparability. Second, EUDAMED was not fully operational during the study period, restricting comprehensive EU-wide analyses. Third, anonymisation and inconsistent terminology in some databases may have impacted device classification. Finally, as this study relied on publicly available secondary data, it is observational in nature and does not permit causal inference regarding the effectiveness of specific FSCAs²¹.

These findings hold practical implications for both clinical practice and regulatory policy. For clinicians, awareness of recurring risks such as battery depletion in pacemakers or failures in orthopedic implants can guide monitoring and patient follow-up strategies to mitigate adverse outcomes. For regulators, the study highlights the need for harmonised reporting frameworks and differentiated approaches to hardware- and software-based device regulation²². The Indian context is particularly relevant: while the Materiovigilance Programme of India (MvPI) provides a foundation, it lacks the comprehensiveness of platforms like EUDAMED^13,23. Establishing a centralised national registry, supported by digital health initiatives and regulatory reforms, could enhance early detection, FSCA coordination, and international alignment²⁴. Overall, this study demonstrates that strengthening post-market surveillance, standardising data transparency, and addressing the unique challenges of software-driven devices are critical steps toward reducing patient harm and improving medical device governance worldwide^9,17,25.