Medical Device Recalls in Neurosurgery: A United States Regulatory Analysis

Abstract

BACKGROUND AND OBJECTIVES:

The neurosurgery landscape has witnessed substantial advancements in recent years, with an increasing reliance on medical devices to enhance surgical safety and efficacy. However, alongside this progress, there has been a rise in the number of device recalls, prompting a need for comprehensive analysis to understand the underlying trends and factors contributing to these recalls. This quantitative review aimed to analyze 2 decades of U.S. Food and Drug Administration (FDA) recall data within neurosurgery to identify patterns and trends in device recalls. By examining the frequency, timing, and reason for recalls, we sought to identify the specific types of neurosurgical devices that are most often recalled.

METHODS:

Using the openFDA device recalls database, we queried recall data for neurosurgical devices from January 2003 to December 2022. Recalls were categorized based on device specialty and reason, with validation by neurosurgical experts by the FDA medical device recalls database. The analysis focused on identifying peaks and trends in recall frequencies over the study period.

RESULTS:

A total of 412 unique neurosurgical device recalls were identified from 2003 to 2022. Recalls increased until 2013, then declined through 2022. General neurosurgery devices accounted for the largest proportion (168, 40.8%), followed by stereotaxic (94, 22.8%) and functional (79, 19.2%) devices, consistently making up the largest share each year. Most recalls were class II (362, 87.9%), followed by class I (28) and class III (22). The most common cause was process errors (67, 14.8%), followed by issues with components/materials (56, 13.6%), device design (61, 14.8%), and labeling (50, 12.1%). Marketing-related issues were least frequent (16, 3.9%).

CONCLUSION:

General neurosurgery devices accounted for the highest recall rates, warranting closer evaluation of contributing factors. Process errors were the leading cause, with most being class II recalls. Understanding recall patterns can help guide strategies to improve device safety and reduce patient risk.

ABBREVIATIONS:

FDA

U.S. Food and Drug Administration

MR

magnetic resonance

PMA

premarket approval

TBD

to be determined.

In the 21st century, the field of neurosurgery has evolved rapidly, leveraging novel technologies to improve procedural safety and expand treatment options for neurological conditions. Certain subspecialities, such as functional, endovascular, and spine neurosurgery are driven by devices, and rely on innovation to enhance our surgical capabilities. As technology advances in general, particularly in software and manufacturing, the rate of medical device innovation has accelerated.¹ Consequently, as the device industry has aggressively grown, the number of serious adverse events related to these devices has increased.^2-4 This trend has led to a rise in device recalls by the U.S. Food and Drug Administration (FDA) in recent years.³

The FDA categorizes medical devices into 3 main classes based on risk: class I (low risk), class II (moderate risk), and class III (high risk).⁵ The approval pathways for these devices vary according to their risk classification, intended use, and the presence of a predicate device. The most common pathway is the Premarket Notification (510(k)) process for class II devices, which allows manufacturers to introduce incremental updates to existing devices without requiring full FDA submission.⁶ Other pathways (Table 1) include premarket approval (PMA) for high risk (class III) devices, which typically requires clinical evidence that a device is safe and effective; the humanitarian device exemption pathway, which is reserved for devices used in the treatment of rare diseases; and the breakthrough device program, for investigational devices that have the potential to provide a more effective treatment or diagnosis for life-threatening or irreversibly debilitating conditions.⁷

TABLE 1.

FDA Device Approval Pathways, Use Cases and Their Requirements

Pathway	Regulatory basis	Intended use/device class	Key requirements	Typical review time
510(k) Clearance	Section 510(k) of FD&C Act	Moderate-risk (Class II), some Class I	Demonstrate substantial equivalence to a predicate device	∼90 d
PMA (Premarket Approval)	Section 515	High-risk (Class III)	Clinical evidence of safety & effectiveness; manufacturing inspection	∼180-320 d
HDE (Humanitarian Device Exemption)	Section 520(m)	Rare diseases (<8000 pts/y)	Probable benefit outweighs risk; no requirement for full effectiveness	∼75 d
Breakthrough Devices Program	21st Century Cures Act	Serious/irreversible conditions	Priority review & iterative FDA feedback; final approval by PMA/510(k)	Variable

HDE, humanitarian device exemption; FDA, U.S. Food and Drug Administration; PMA, premarket approval.

Post market, the FDA continues to monitor device performance. Companies can initiate recalls to address defects that could jeopardize patient safety. The FDA classifies recalls into 3 categories (Table 2): class I (serious health consequences or death), class II (remote probability of serious health consequences), and class III (unlikely to cause adverse health consequences).^3,4,8,9 The FDA also provides recall reasons along with classification to further categorize each recall.⁹ Previous studies have examined the reasons for medical device recalls broadly.^4,6,10-14We analyze 2 decades of FDA recall data to investigate the specific reasons for recalls in neurosurgery. This analysis aimed to inform the key stakeholders about recent recall trends among neurosurgical devices and provide guidance for industry and surgical innovators to prevent common pitfalls in device design and manufacturing. We hypothesized that recall rates likely shared a relationship with implementation of quality-system reforms.

TABLE 2.

FDA Recall Classifications by Class With Definition, Severity and Example

Recall class	Definition (FDA)	Severity of risk	Example
Class I	Reasonable probability of serious adverse health consequences or death	Highest	Stealthstation S8
Class II	Temporary or reversible adverse events; remote probability of serious harm	Moderate	Codman Certas Programmable Valve
Class III	Not likely to cause adverse health consequences	Lowest	Viasys Surface Electrodes

FDA, U.S. Food and Drug Administration.

METHODS

The openFDA device recalls database (https://open.fda.gov/data/device-recall/) was queried in March 2023 for all recalls with a medical specialty description of “neurology” from January 2003 to December 2022. Devices meeting these criteria were extracted, and their recall classification was determined using the medical device recalls database (https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfres/res.cfm) as the recall class was not available in the openFDA data. The openFDA device application programming interface (https://open.fda.gov/apis/) was queried for all 510(k) and PMA approvals with an advisory committee of “neurology” from January 2003 to December 2022.

To avoid data duplication, products with the same recall initiation day, recalling firm, and similar reason for recall were treated as a single device recall. This avoided duplication of similar products (eg, screws of varying sizes) and products that were recalled in lots (eg, all products in a dissection tool kit because of labeling issues). The devices were categorized by the authors into 6 main categories: general (eg, surgical frames, bone drills, etc.), neurocritical care (eg, intracranial pressure monitors, external ventricular drains, etc.), neurology (eg, electroencephalogram or nerve conduction study equipment, etc.), functional (eg, deep brain stimulators, vagal nerve stimulators, etc.), endovascular (eg, flow diverting stents, thrombectomy catheters, etc.), and stereotaxic devices (eg, stereotactic aspiration systems, laser ablation systems, etc.). Two authors (L.S.B. and T.H.) independently sorted the data into the various categories. All groupings, including any discrepancies, were validated by an attending neurosurgeon (K.Y.). Any device with exclusively orthopedic indications and devices related to spine were excluded because of a large overlap with the specialty description “orthopedic surgery”.

The reasons for device recalls were classified into 10 categories based on FDA root cause descriptions, similar to Sarkissian et al²: (1) device design, (2) component/materials, (3) processing, (4) packaging, (5) software, (6) marketing, (7) labeling, (8) to be determined (TBD), (9) miscellaneous, and (10) other. Owing to overlap, categories 9 and 10 were combined. Device design issues involve flaws in engineering, such as the Medtronic Duet External Drainage and Monitoring System catheter (Medtronic), which risked disconnection and serious harm resulting in a class I recall.¹⁵ Component/material issues arise from defective parts, like DuraMatrix (Stryker) membranes with unsealed pouches compromising sterility.¹⁶ Processing problems occur postmanufacturing, exemplified by DePuy Synthes' improperly programmed Codman Certas Plus shunt (Integra LifeSciences) resulting in over and under drainage of cerebrospinal fluid. The failure mechanism centered on mechanical tolerance issues within the magnetic rotor assembly, which could cause the displayed setting to diverge from the actual valve resistance—particularly at very low or very high pressure levels.¹⁷ Packaging errors, such as nonsterile packaging in Stryker Total Performance System Burs (Stryker), pose infection risks.¹⁸ Software issues affect device functionality, as with Medtronic's Synergy and StealthStation S7 Cranial software (Medtronic) causing navigation inaccuracies with possibly devastating consequences. The error stemmed from a misalignment in the image registration algorithm that could display the surgical tool several millimeters from its true anatomic position. This posed a risk of unintended bone removal, neural injury, or suboptimal screw placement. The recall required a mandatory software patch and user notification, along with updated instructions emphasizing verification of anatomic landmarks before use.¹⁹ Marketing problems involve improper promotion and instructions for use, like the Integra Camino catheter (Integra LifeSciences) lacking magnetic resonance (MR) safety labeling.²⁰ Labeling errors include inaccurate instructions, as seen in the Elekta Leksell (Elekta) system following MR sequence changes.²¹ The difference between this and the marketing example is that the MR sequence changes occurred after approval, resulting in proper marketing but incorrect labeling. TBD applies to recalls still under FDA investigation, such as the Natus Cranial Access Kit (Natus Medical Incorporated).²² Miscellaneous/Other covers uncategorized recalls, including Codman's Holter Rickham Reservoir (Integra LifeSciences) with absent drainage tips.²³ This detailed categorization helps in identifying specific areas for improvement and implementing targeted corrective actions to prevent future recalls. Sorting recalls into these categories was performed independently by 2 authors (L.S.B. and J.C.G.) and validated by the attending neurosurgeon (K.Y.).

Descriptive statistics were used to summarize device recalls by year, recall class, device category, and root cause. To evaluate trends in recall frequency over time, simple linear regression was performed with year as the independent variable and the annual number of recalls as the dependent variable. The slope, P-value, and coefficient of determination (R²) were used to evaluate the presence and strength of a statistically significant trend. A statistically significant positive slope was interpreted as evidence of an increasing trend. Differences in recall class and root cause across device categories were described using frequency distributions and proportions. A P-value <.05 was considered statistically significant for regression analysis. All statistical analysis and figure generation was conducted using GraphPad Prism version 10 (GraphPad Software). Statistical analysis was repeated for recalls as a percentage of approvals for that year.

RESULTS

A gross total of 1147 device recalls were identified from January 2003 to December 2022 in neurosurgery, resulting in a total of 412 unique device recalls (Figure). Peaks in recalls occurred in 2010 (30 recalls) and 2013 (35 recalls). Over the entire study period, recalls as a percentage of approvals showed a statistically significant increasing trend (slope, 95% CI: 0.065-1.175; P = .03), although the relationship was weak (R² = 0.23). Closer examination reveals a significant increase from 2003 to 2013 (95% CI: 1.54, 3.35; R² = 0.81; P = .0002), followed by a significant, steady decrease until 2022 (95% CI: −2.50, −0.58; R² = 0.63; P = .003; Figure A). The overall trend in recalls as a percent of approvals depicts a similar trend as shown by gross recalls. Recalls as a percentage of approvals increased significantly from 2003 to 2013 (95% CI: 0.011, 0.031; R² = 0.71; P = .001), then decreased significantly from 2013 to 2022 (95% CI: −0.027, −0.005; R² = 0.60; P = .009). Despite these opposing trends, the overall recall rate relative to device approvals remained stable across the full study period (2003-2022) (Figure D). The general category had the largest proportion of recalls (168, 40.8%), followed by stereotactic (94, 22.8%) and functional devices (79, 19.2%). General neurosurgery devices generally made up the largest proportion of devices each year. The number of device recalls ranged from 35 total in 2013 to 8 in 2003. Most recalls were class II (362, 87.9%), with class I (28) and class III (22) following. The highest number of class I recalls occurred in 2018 and 2019 with 4 recalls each year.

FIGURE.

Trends in neurosurgical device recalls, 2003–2022. A, Total recalls by subspecialty category. B, Total recalls by FDA recall class. C, Total recalls by root cause of recall. D, Annual recalls expressed as a percentage of total device approvals. FDA, U.S. Food and Drug Administration; TBD, to be determined.

The FDA root cause description was used to determine the reason for device recall (Table 3). The most common recall reason was process errors (67, 14.8%), followed by device design (61, 14.8%), components/materials (56, 13.6%), and labeling (50, 12.1%). Marketing issues were the least common cause (16, 3.9%).

TABLE 3.

The Total Number of Devices Recalled During the Study Period Separated Into Subspecialty Category and Recall Class

Recall devices by subspecialty and class
	Total number of recalls	Class 1 recalls	Class 2 recalls	Class 3 recalls	Top reasons for recalls		Example
					Reason	N (%)
Endovascular	20	2 (10.0)	15 (75)	3 (15.0)	Process	7 (35)	Micrus Endovascular Microcoil Delivery System
					Label	4 (20)
					Other, TBD	3 (15)
Functional	79	5 (6.3)	66 (83.5)	8 (10.1)	Device	13 (16.5)	Integra Ojemann Cortical Stimulator
					Other	12 (15.2)
					Label	10 (12.7)
General	168	11 (6.5)	150 (89.3)	7 (4.2)	Process	33 (19.6)	Codman Certas Programmable Valve
					Label	26 (15.5)
					Component/material	24 (14.3)
Neurocritical	23	1 (4.3)	21 (91.3)	1 (4.3)	TBD	6 (26.1)	Natus Cranial Access Kit
					Device	5 (21.7)
					Component/material	3 (13.0)
Neurology	28	0 (0)	26 (92.9)	2 (7.1)	Component/material, Software	6 (21.4)	Integra Camino 110-4 ICP Monitoring Catheters
					Marketing/Other	5 (17.9)
					Device	3 (10.7)
Stereotaxic	94	9 (9.6)	84 (89.4)	1 (1.1)	Software	25 (26.6)	Elekta Leksell Stereotactic System
					Process, Device	17 (18.1)
					Component/Material	14 (14.9)
Total	412	28	362	22	Process	67 (16.3)
					Device	61 (14.8)
					Component/Material	56 (13.6)

ICP, intracranial pressure, TBD, to be determined.

Table 4 shows each device category and the number of devices per recall class and the top reasons for recalls. Most of the class I recalls occurred in the general group with no recalls from the neurology group. Endovascular and stereotactic devices had the highest proportion of device recalls with class I classification (10% and 9.6% of all recalls, respectively). Process errors, such as shipment of products with missing parts or deviation from manufacturing protocol, were the most common cause of recalls for the endovascular and general groups. Device problems were the most common reason for recall in the functional category with software issues the most common for stereotaxic devices, component/material for neurology, and TBD issues, including “Under investigation by firm,” “Pending,” and “Other,” most common for neurocritical care.

TABLE 4.

The Total Number of Devices Recalled During the Study Period Separated Into Reason for Recall and Recall Class

Recall devices by reason for recall and class
FDA cause	Count (%)	Aggregate label	Count (%)	Example
Package design/selection	7 (1.70)	Packaging	23 (5.58)	Stryker 5190 Total Performance System Series Burs
Packaging	8 (1.94)
Packaging change control	2 (0.49)
Packaging process control	6 (1.46)
Component change control	6 (1.46)	Component/Materials	56 (13.59)	DuraMatrix Collagen Dura Substitute Membranes
Component design/selection	11 (2.67)
Material/component contamination	1 (0.24)
Mixed-up of materials/components	2 (0.49)
Nonconforming material/component	33 (8.01)
Release of material/component before receiving test results	2 (0.49)
Manufacturing material removal	1 (0.24)
Device design	59 (14.32)	Device	61 (14.81)	Medtronic Duet External Drainage and Monitoring System
Finished device change control	2 (0.49)
Process change control	13 (3.16)	Process	67 (16.26)	Codman Certas Programmable Valve
Process control	43 (10.44)
Process design	11 (2.67)
Software change control	2 (0.49)	Software	40 (9.71)	Synergy Cranial and StealthStation S7 Cranial Software
Software design	27 (6.55)
Software design (manufacturing process)	1 (0.24)
Software design change	2 (0.49)
Software in the use environment	4 (0.97)
Software manufacturing/software deployment	4 (0.97)
Labeling change control	12 (2.91)	Labeling	50 (12.14)	Elekta Leksell Stereotactic System
Labeling design	8 (1.94)
Labeling false and misleading	9 (2.18)
Labeling mix-ups	11 (2.67)
Error in labeling	10 (2.43)
No marketing application	14 (3.40)	Marketing	16 (3.88)	Integra Camino 110-4 ICP Monitoring Catheters
Premarket approval	2 (0.49)
Under investigation by firm	24 (5.83)	TBD	26 (6.31)	Natus Cranial Access Kit
Pending	2 (0.49)
Other	53 (12.86)	Other	53 (12.86)	Codman Holter Salmon Rickham Reservoir with Ventricular Catheter
Employee error	10 (2.43)	Miscellaneous	20 (4.85)
Environmental control	3 (0.73)
Equipment maintenance	1 (0.24)
Incorrect or no expiration date	2 (0.49)
Unknown/undetermined by firm	2 (0.49)
Storage	1 (0.24)
Use error	1 (0.24)
Total	412 (100.00)

DISCUSSION

According to publicly available databases, there was a significant increase in U.S. FDA neurosurgical device recalls between 2003 and 2013, followed by a steady decline through 2022. This pattern may be attributed to heightened regulatory scrutiny during the earlier period, which identified more device-related issues, subsequently prompting manufacturers to implement improved quality control and risk management practices. The 2016 update of ISO 13485 introduced major revisions to quality management requirements, including life-cycle risk management, enhanced supplier oversight, strengthened postmarket surveillance, and expanded software validation obligations. These changes increased manufacturer reporting fidelity and may have contributed to the rise in recalls observed in the second decade of our study period, particularly for software-driven neurosurgical technologies. The subsequent decline in recalls after 2018 may reflect manufacturers' maturation into these updated standards. Alternatively, there may have been an increase in device quality in the latter years, reflected in a fewer recalls after 2013. Shifts in administrative oversight may influence inspection frequency, enforcement priorities, and recall practices. Although our Data set cannot directly link recall fluctuations to political cycles, this contextual factor is likely relevant and warrants future study.

In this study, class II recalls accounted for 87.9% of neurosurgical device recalls, closely mirroring the overall rate for all medical devices (90.7%).²⁴ Most neurosurgical devices fall into class II and III categories (moderate and high risk, respectively), with relatively few class I devices.²⁵ Notably, class III devices, which undergo the more stringent PMA approval process, are associated with a higher proportion of severe class I recalls compared to class II devices cleared through the 510(k) pathway. Thus, a possible explanation for the prevalence of class II recalls is that the risk profile of class II devices results in decreased likelihood of severe or minor consequences corresponding with class I or class III recalls. Recalls do not imply that patient harm occurred; many recalls are preventative. The FDA database does not reliably link recall events to clinical adverse outcomes, limiting our ability to quantify harm.

There was a high concentration of device recalls in the general, stereotaxic, and functional specialties, with general devices accounting for the most class I recalls. Across specialties, recalls most often stemmed from issues with processing, device design, components/materials, or labeling. Design flaws can directly affect function and outcomes, whereas component and material choices determine biocompatibility, durability, and toxicity. Processing recalls typically arose from manufacturing, assembly, or consistency errors, reflecting the vulnerability of multistep production.⁴ Even well-made devices must be accurately labeled; many labeling recalls involved mix-ups or errors, underscoring the need for stricter oversight. Notably, software-related recalls—more frequent in neurology and stereotaxic devices—were a major contributor to class I recalls, as these systems depend on reliable integration, control, and communication to prevent serious adverse outcomes.^13,22 Recent FDA guidance, such as the 2023 cybersecurity updates, emphasizes protection against unauthorized access to implantable electronic devices, but specific requirements for neurosurgical implants remain limited. This represents a future opportunity for regulatory expansion as device connectivity grows.

A large portion of recalls were because of process errors in manufacturing, assembly, packaging, and product consistency. Although FDA inspection guidelines exist for production and process control, manufacturers are ultimately responsible for enforcing strict quality control to reduce recalls. Although the FDA's current quality system regulation (21 Code of Federal Regulations 820) has been in place for over 20 years, a proposed amendment was published in February 2022 to align with the international ISO 13485 standard for medical device quality management.

Device design-related recalls can be reduced through rigorous premarket testing that includes edge-case and real-world clinical scenarios. Human factors engineering, such as usability testing with target end users, helps identify interface or workflow flaws early. Digital simulations, including digital twin modeling, can replicate device performance under diverse conditions. Involving clinicians in early testing and postmarket feedback enables real-world data to drive iterative design improvements.

Component and material-related recalls could be reduced by adopting standardized stress-testing protocols that replicate dynamic surgical environments such as temperature, pressure, and biochemical variability. Integrating predictive material modeling, allows forecasting of degradation or failure mechanisms under physiological stress before clinical deployment.²⁶

Despite existing FDA guidelines, the high number of labeling recalls highlights the need for better quality control by manufacturers to ensure labels match components and are clear to end users. Manufacturers should adopt automated label verification systems, such as barcode or QR-code scanning tied to centralized databases, to ensure label-content accuracy and alignment with device components. Similarly, packaging errors from incorrect or damaged materials could be reduced through stricter quality control and delivery checks.

Software recalls were because of errors in third-party system integration, device control, and data processing. The FDA currently practices oversight of software that meets the definition of a device in section 201(h) of the FD&C Act, with its latest guidance document with updated recommendations published September of 2022. Because software applications become increasingly intricate with systems control, complex analysis, artificial intelligence integration, and augmented and virtual reality applications, frequent FDA regulatory updates will be required.^13,27 For practitioners, understanding the training and testing conditions, common errors, and continued validation during use can help prevent medical errors.

Marketing recalls were mostly because of discrepancies in indications for use and regulatory approval. Clinicians should verify a device's intended use before adopting new technologies. Other recalls stemmed from environmental controls, equipment maintenance, storage, and user errors, often overlapping with process and labeling issues. Addressing these requires stronger quality systems, regulatory compliance, and comprehensive training. Improved device training and stricter adherence to maintenance protocols could help reduce these errors and enhance device safety.

Our neurosurgical recall trends follow similar device recall patterns in general and plastic surgery.²⁸ Plastics and general surgery device recalls peaked between 2012 and 2014 before declining—a pattern mirrored in our data, with neurosurgical recalls peaking in 2013 and steadily decreasing through 2022. This alignment suggests that broader regulatory shifts or industry-wide quality improvements may have impacted both neurosurgical, general and plastic surgery device markets. Although we did not directly stratify our recall data by approval pathway, the prevalence of class II recalls correlates with a similar dominance and recall pattern of the 510(k) pathway in recalled general and plastic surgery devices. The parallel decline in both Data sets may reflect enhanced manufacturing practices, updated FDA guidance, or increased scrutiny of postmarket performance during the past decade.

The number of device recalls as a proportion of device approvals has remained overall flat between 2003 and 2022 (Figure D), indicating that changes in recall number year over year may reflect the changes in the number of device approvals rather than changes in policy or device quality. The proportion did significantly peak between 2009 and 2013, but the overall trend between the 2 decades was flat. This may suggest an overall consistent regulatory approach from the FDA, and it is yet to be seen whether further recalls as a percentage of overall device clearance continue to decrease or remain constant at around 0.1%-0.2%.

To enhance device quality and lower risk of recalls, we suggest a multifactorial approach which involves encouraging higher standards from device manufacturers, following the most recent regulations from the FDA, and highlighting due diligence from medical institutions and providers. More stringent testing of device function, limits, and longevity would help decrease recalls because of device design and material/components. The FDA is already moving to update their quality systems regulations, which could help prevent process, labeling, and packaging errors. FDA regulations which stay current with software advances and rigorous edge case and robustness testing are crucial to preventing program errors and malfunctions.¹² In addition, looking to other industries for their standardized testing protocols may better inform premarket testing. Finally, medical institutions should verify a new device's testing history, safety profile, and performance in their intended use at each point in its cycle to proactively catch any errors in function. These steps can help minimize adverse events and maximize the positive impact of the rapid progress in neurosurgical devices.

Limitations

Limitations of this study include inherent discrepancies when comparing different types of devices with varying FDA classifications and recall classes. There is no specific specialty description for “neurosurgery” further illustrating inherent limitations of the FDA databases. In addition, expanded use of devices from other fields such as orthopedics and cardiology, which include spinal and endovascular devices, were not accounted for in the data presented. It is challenging to quantify neurosurgical recall as a proportion of all recalls because of these limitations. Finally, the total number of devices submitted for FDA approval during this time frame is inaccessible, as opposed to the number of devices approved. It is therefore difficult to fully ascertain that more recalls was due to increased safety issues rather than a product of increased submissions during this time frame. In addition, inconsistent reporting in the database makes analysis of recall initiation (FDA vs manufacture) challenging, but will be a point of future study. Direct comparison to international regulatory bodies was not feasible because of lack of publicly indexed recall databases; however, the CE Mark and Japan Pharmaceutical and Medical Devices Agency follow differing postmarket surveillance frameworks, and future comparative analyses may reveal differing recall patterns.

CONCLUSION

We analyzed FDA recall data for neurosurgical devices from the publicly available FDA database to highlight trends and better understand common safety errors.^6,11-13,29 Between 2003 and 2022 there was a significant increase in the number of recalls overall, although a decreasing trend over the last 10 years, suggesting improvements in quality control by manufacturers and the FDA. To continue to improve safety, manufacturers should adopt rigorous testing and quality control measures, whereas updated FDA regulations and guidance on emerging technologies are essential. Collaboration between regulators, manufacturers, and healthcare providers is key to ensuring that innovation in neurosurgery continues to enhance patient care while minimizing risks. Because new devices are developed, it will be imperative to follow stringent safety measures to provide the best care for neurosurgical patients.

Funding

This study did not receive any funding or financial support.

Disclosures

The authors have no personal, financial, or institutional interest in any of the drugs, materials, or devices described in this article.

COMMENTS

The authors review patterns, root causes, and trends in FDA neurosurgical device recalls over a 20-year period using the open FDA application programming interface (API), which searches the Manufacturer and User Facility Device Experience (MAUDE) raw database.

The authors found a significant increase in FDA neurosurgical device recalls from 2003 to 2013, followed by a decline through 2022, but it was difficult to attribute this to any specific device or regulatory factors. The authors suggest this represents improvement in quality control oversight and cooperation between device manufacturers and regulators. They conclude that collaboration between industry, regulators, and healthcare providers can help facilitate innovation that improves patient care and minimizes risks.

The authors acknowledge limitations including that direct comparison to foreign and international regulatory bodies was not feasible due to lack of public data for those agencies and different postmarket surveillance protocols. They suggest future comparative analyses to elucidate different recall patterns and practices, opening the possibility for international cooperation. Perhaps future technologies linked to artificial intelligence large language models (LLM) can contribute to regulatory oversight by sorting and classifying large quantities of heterogeneous data such as these.

It should also be noted that the FDA database does not link recall events to clinical outcomes. As the authors note, many recalls are preventative, without established causality. The FDA itself notes that reports resulting in database inclusion may be incomplete, inaccurate, unverified, or biased. It should be clear that the FDA database is a tool for identifying patterns, examining trends, qualitative analysis, and hypothesis generation rather than hypothesis testing.

Christopher Carr

Augusta, Georgia, USA