Stolen Medical Devices Hit Pharmacies

A doctor holding a miniature shopping cart filled with various medication packs

When a high‑risk medical device like a continuous glucose monitor leaves the controlled supply chain and re‑enters the market through theft, the problem is no longer just about product defects — it is about traceability, trust, and whether patients can rely on the data keeping them alive.

Key Points

Two specific lots of Dexcom G7 glucose sensors — 1725204004 and 1725069002 — were intentionally scrapped, stolen before destruction, and then resold through an unauthorized wholesaler.
One lot was rejected for incomplete sterilization, raising infection risk; the other for an elevated failure rate and potential loss or inaccuracy of glucose readings.
The FDA and Dexcom classify the resulting risks as serious, even potentially life‑threatening, despite reporting no severe adverse events to date.
Patients can and should verify lot or serial numbers and obtain free replacements, while understanding the broader pattern of supply‑chain diversion now affecting high‑value medical devices.

What Actually Went Wrong With These Glucose Monitors

The core facts of this episode are not speculative; they are laid out in Dexcom’s own press materials and in the FDA’s recall database. Dexcom’s quality control identified two lots of G7 continuous glucose monitoring (CGM) sensors — lot 1725204004 and lot 1725069002 — that did not meet release standards and were formally designated as scrap. Lot 1725204004 contained sensors that were not properly sterilized, creating an increased risk of skin infection at the insertion site if they were used. CGM sensors sit under the skin for days at a time; incomplete sterilization is not a cosmetic defect, it is a direct pathway for local infection and, in vulnerable users, potentially systemic illness.

Lot 1725069002 failed at a different point in Dexcom’s internal checks. It showed an elevated failure rate during testing, meaning a higher likelihood that the sensor would either provide no readings at all or inaccurate readings. For a technology whose clinical value is continuous, trend‑level glucose data — allowing people with diabetes to catch hypoglycemia and hyperglycemia in time to intervene — losing data or trusting bad data is a concrete safety risk. The FDA’s recall entry explicitly cites the potential for missed detection of hyperglycemic events or incorrect treatment decisions, with the possibility of injury requiring medical intervention to prevent permanent harm.

Under normal circumstances, those scrap lots would never reach a patient. As Dexcom describes, scrap sensors are routinely sent to a third‑party vendor for destruction and recycling, part of standard quality control practice. In this case, however, the company discovered through quality and accounting reviews that the lots were stolen during the destruction process, diverted into the market, and ultimately traced to Pharmsource, LLC, an unauthorized distributor supplying independent pharmacies and durable medical equipment (DME) outlets. That chain of events is what transformed internal quality findings into a public health issue.

How an Internal Quality Scrap Became a Public Recall

Once Dexcom confirmed that scrapped sensors had been stolen and sold, the company triggered what FDA terminology calls a “medical device correction” rather than a full physical removal. In practical terms, that meant a press release, direct notifications, and an “URGENT: MEDICAL DEVICE CORRECTION” notice to customers explaining the problem and the specific lot numbers involved. The FDA’s recall record identifies the case as a Class 2 device recall — a category used when a device poses temporary or medically reversible health risks, with a low but real chance of more serious outcomes.

The distinction between correction and removal matters. A correction is performed where the product sits: the manufacturer issues warnings, instructions, and sometimes software updates, but it does not necessarily retrieve every unit. Removal implies physically pulling devices from shelves or homes. Here, because the affected sensors were never part of Dexcom’s controlled distribution, there was no centralized stock to seize; the best available tool was communication and replacement. Dexcom created an online “theft‑check” tool that lets users enter sensor serial numbers to see whether they match the affected lots, and urged anyone with suspect sensors to stop using them immediately and request replacements.

Regulators outside the United States followed suit once the diversion became clear. Safety agencies in New Zealand, Australia, and Canada issued aligned alerts or recalls, each specifying that lots 1725204004 and 1725069002 had been earmarked for destruction in the US due to incomplete sterilization and deployment or testing failures, then stolen and resold through unauthorized channels. All share the same risk language: skin infections, inaccurate or lost readings, and missed detection of low or high blood sugar that could lead to serious health consequences.

Risk, Evidence, and the Argument That “No One Has Been Hurt”

A natural question is whether these theoretical risks have translated into real harm. As of the latest public statements, both Dexcom and the FDA report that no severe adverse events have been linked to the stolen sensors. That absence anchors much of the informal counter‑narrative: if no one has been seriously injured, some argue, perhaps the sensors are safe enough and the warning is overblown — especially for users who obtained devices at lower cost via gray‑market channels.

That argument misses two key points. First, the original decision to scrap the lots did not rest on clinical complaints but on pre‑market quality control — incomplete sterilization and elevated failure rates documented in Dexcom’s own testing systems. These are upstream indicators designed to prevent harm before it occurs; waiting for infections or mis‑treatment injuries to show up in the adverse event database would defeat the purpose of quality assurance. Second, infections at sensor sites and episodes of mismanaged glucose excursions are not always reported back to manufacturers in a way that can be easily linked to a specific lot number. Without a targeted epidemiological study following people who unknowingly used these stolen sensors, the absence of severe reports is reassuring but not decisive.

Regulators treat this kind of scenario as a risk‑management problem, not a courtroom proof exercise. Under FDA recall guidance, manufacturers are expected to act when a device “may be” violative or presents a risk to health, not only when that risk has already materialized. Here, both the nature of the defects and the additional uncertainty introduced by unknown storage and handling conditions during unauthorized distribution made the risk unacceptable to leave unaddressed. Criminal diversion breaks the chain of custody: devices may have been stored at improper temperatures, exposed to humidity, or handled in ways that compromise sterility or sensor chemistry. Even if the exact mechanism of any added risk is not spelled out in public documents, regulators reasonably treat this as compounding the original defects.

The Broader Pattern: Supply-Chain Diversion in Medical Devices

What happened with these G7 sensors is not a bizarre one‑off; it fits a broader pattern that has become more visible as medical devices grow more complex and more valuable. Research reviewing device recalls in the US market from 2020 to 2023 found that the number of recalls rose from 33 to 61 over that period, with inaccurate results, software failures, and manufacturing defects among the most frequent root causes. High‑value, small‑form devices — insulin pumps, implantable cardiac hardware, and CGMs — are particularly susceptible to both production issues and illicit redistribution because they command high prices and can be moved easily.

The specific mechanism here — quality‑rejected units stolen in transit to destruction and then sold through an unauthorized wholesaler — is simply one variant of a larger supply‑chain integrity problem. Once any step in that chain is weak, rejected units, counterfeit products, or devices outside their validated storage parameters can quietly re‑enter clinical use. The user sees a legitimate brand name and model number and has no easy way to know whether the unit in their hand followed the validated path from factory to pharmacy.

Regulatory frameworks are only starting to catch up. While drug supply has long been subject to pedigree and serialization requirements, device tracking has lagged. The Dexcom case underscores why stronger device traceability, tighter controls on scrap handling, and oversight of secondary wholesalers are not bureaucratic niceties but patient‑safety measures. Every untracked diversion undermines the assumption that a cleared device in the field still matches the risk profile regulators evaluated.

What Patients and Clinicians Should Do, Practically

For individual users, the practical implications are straightforward even if the underlying system issues are complex. Anyone using Dexcom G7 sensors should check lot numbers printed on the box or on individual sensor packaging; the affected lots are 1725204004 and 1725069002. If the packaging is unclear, the serial number look‑up tool described by Dexcom — accessible through its “theft‑check” site — provides another way to verify whether a sensor belongs to the stolen cohorts.

If a sensor matches those lots, the recommended course is not hedged or conditional. Users are advised to discontinue the sensor immediately, dispose of it, and either replace it with an unaffected sensor or revert to a conventional blood glucose meter until a safe replacement is available. Dexcom and its regional partners are offering free replacement sensors to affected users, which serves both safety and liability interests but ultimately benefits the user by eliminating the defective lot from their regimen.

Clinicians — endocrinologists, diabetes educators, and primary care physicians — should incorporate a simple question into visits with CGM users: where did you obtain your sensors, and have you checked your lot numbers? Because Pharmsource and other secondary sellers supply independent pharmacies and DME distributors, not every exposure will involve obvious online gray‑market purchases. A patient who bought sensors from a small local pharmacy may never realize that the device bypassed Dexcom’s authorized channel.

Disentangling This Recall From Other Dexcom Issues

One complication in public understanding has been the conflation of this stolen‑sensor recall with other Dexcom recalls and FDA actions, particularly those involving G6 and G7 receivers and mobile apps. Separate Class 1 recalls have addressed receiver speaker malfunctions that could cause missed alerts for dangerous blood sugar levels, and FDA has issued corrections for G7 and ONE+ apps due to software issues affecting data display and alarms. These are distinct from the stolen sensor lots, even though they share the same brand name and, in the case of receivers, the same general product line.

For users, the practical rule is to treat each notice as specific. A hardware receiver recall does not imply that all G7 sensors are unsafe; a stolen‑sensor warning does not mean the app update on your phone is compromised. The current case is tightly scoped to two lot numbers of G7 sensors, both of which were never cleared for normal distribution and were only used because they were stolen.

Why This Episode Matters Beyond Dexcom

In one sense, this is a contained story: identified lots, clear defects, a known unauthorized distributor, and an established replacement pathway. But in another sense, it is a stress test of how modern health systems handle the intersection of manufacturing quality, criminal diversion, and patient reliance on continuous data. CGMs are not lifestyle gadgets for most users; they are safety nets built into daily life, influencing insulin dosing, nutrition decisions, and alarm‑driven responses to low or high glucose.

When the integrity of that safety net is compromised — whether by incomplete sterilization, unreliable readings, or an opaque supply chain — the margin for error shrinks. Regulators and manufacturers are right to act aggressively before harm shows up in adverse event statistics. For the informed patient and clinician, the lesson is not to panic about every device but to understand where the real vulnerabilities lie: in lot‑specific defects, in unauthorized channels, and in the quiet assumption that any labeled device is automatically safe.