Home Oxygen Concentrator Recalls: 20 Class I Actions and the Fire-and-Purity Risk

Executive Summary

Home oxygen concentrators are among the few medical devices that run continuously in patients' homes while enriching the surrounding atmosphere with oxygen — a combination that makes fire the defining product-risk theme. Our analysis of the complete FDA enforcement, recall, and MAUDE record for the oxygen-concentrator product-code family finds a category whose post-market signature is dominated by two failure modes:

• 44 FDA enforcement actions match oxygen-concentrator devices, of which 20 are Class I — an unusually high Class I share (45%) for a Class II device family, driven by the fire and loss-of-oxygen hazards
• Fire and thermal faults are the leading root cause (13 of 44 actions), followed by low oxygen purity or output failures (9), and electrical and power faults (8) — together accounting for the majority of the enforcement record
• Nidek Medical drives the largest share of the most-serious actions: of the 20 Class I recalls, 9 list Nidek Medical Products as the recalling firm, almost all citing "capacitor failure may result in a fire hazard and loss of supplemental oxygen"; Baxter (4), Getinge (3), and Medical Depot/Drive DeVilbiss (1) account for most of the remainder
• The MAUDE record is the largest of any respiratory home-care device category: 53,418 device reports under the concentrator family product codes, including 605 reports recording a patient death and 1,398 injury reports, with Invacare alone named on 46,137 reports

For quality, regulatory, and home-medical-equipment teams, the takeaway is that concentrator recall risk is overwhelmingly a thermal-electrical and oxygen-delivery-purity problem, concentrated in a small group of legacy manufacturers (Invacare, Nidek, Caire, and DeVilbiss) whose installed base defines the category's post-market footprint.

Why Oxygen Concentrators Carry a Distinct Risk Profile

A home oxygen concentrator separates nitrogen from room air using a molecular-sieve (zeolite) bed, delivering up to 95% oxygen to a patient who is often on therapy continuously. Two physics facts shape the device's entire post-market record: oxygen vigorously supports combustion, and a patient who depends on supplemental oxygen can be harmed by both fire and by a silent drop in delivered purity.

The fire risk is not theoretical. The National Fire Protection Association estimates that during 2017–2021, oxygen-administration equipment was involved in an average of 228 home fires per year reported to local fire departments, resulting in an average of 96 civilian deaths and 106 civilian injuries per year — and roughly one in every three of those home fires was fatal (NFPA, "Fires and Burns Involving Home Medical Oxygen"). These are fire-department-reported events that capture the downstream consequence of an ignition in an oxygen-enriched home; they are a separate population from the device-fault reports in MAUDE, but together they bracket the same risk. The clinical literature on home-oxygen fires reports comparable severity, with a case series mortality around 10% and elderly smokers the over-represented group (Home oxygen and domestic fires, PMC).

The product code at the center of this analysis is CAW — Generator, Oxygen, Portable, a Class II device under the Anesthesiology medical-specialty panel. The family also includes the oxygen conserver (NFB), portable liquid-oxygen units (BYJ), and medevac oxygen generators (OLF).

The concentrator (CAW) is the focal product code and accounts for the overwhelming majority of the records below.

Data Source and Method

• Sources: FDA device enforcement report database (reason-for-recall text and Class I/II/III classification), FDA Medical Device Recall database (recall initiation records by product code), FDA MAUDE adverse-event database (device reports by product code), and the FDA 510(k) database (clearances by product code)
• Analysis sample: 44 enforcement-report records matching oxygen-concentrator devices; 56 recall-initiation records under the four concentrator-family product codes; 53,418 MAUDE device reports under the same codes; 413 510(k) clearances under the same codes
• Run date: 2026-06-14
• Method: All counts were computed by MedDeviceGuide from the public FDA enforcement, recall, MAUDE, and 510(k) extracts. Enforcement records do not carry a populated product-code field, so concentrator records were identified by text matching of the device description and reason-for-recall fields; firm names are reproduced as recorded by FDA. Root-cause themes were tallied from reason-for-recall text and are not mutually exclusive. MAUDE death and injury counts are as-filed report counts and do not by themselves establish that the device caused the outcome.

The Enforcement Footprint: A High Class I Share

The 44 oxygen-concentrator enforcement actions carry a classification split that is unusually weighted toward the most serious tier:

A 45% Class I share is high for a Class II device family and reflects the dual fire-and-suffocation hazard: a concentrator that ignites, or one that silently stops producing therapeutic oxygen, can both cause death in a home setting where the patient may be alone and where there is no clinical staff to intervene.

Class I: A Nidek-Dominated Fire Record

The Class I record is the core of the concentrator safety story. Of the 20 Class I actions, the recalling firms are:

Nidek's nine Class I actions share an almost identical reason statement: "capacitor failure may result in a fire hazard and loss of supplemental oxygen." That single component-failure mode — a capacitor that both ignites and cuts oxygen delivery — is the most concentrated Class I pattern in the entire concentrator record. The recent Class I actions extend the fire theme to newer entrants and importers:

• Jiangsu Jumao X-Care TruAire-5 (Class I, correction letter November 2024; FDA-posted February 2025) — the concentrator "may melt or catch on fire during use," with all units in a defined serial-number range (Model O2C5L, JA2311000001–JA2312000740) ordered removed from service and returned to Compass Health Brands (Health Exec, 2025; FDA/MEHCA, 2025)
• Medical Depot (Drive DeVilbiss) iGo2 portable concentrator car adapter (Class I, May 2025) — "potential for DC Power Supply housing to become hot to the touch and deform the plastic housing," with affected charger cords ordered destroyed (Healthcare Brew, 2025)

These recent actions fit the established pattern: the ignition source varies (a capacitor, a power-supply housing, an unknown spontaneous-combustion cause), but the oxygen-enriched operating environment is what turns a routine electrical fault into a Class I event.

Root Causes: Fire and Purity Lead

Tallying reason-for-recall text across all 44 enforcement actions produces a hierarchy that mirrors the device's underlying physics:

The fire/thermal and electrical/power buckets together (21 actions) describe the ignition pathway; the low-purity bucket (9 actions) describes the silent-suffocation pathway. A representative purity-failure reason comes from Ohio Medical: "erroneous calibration values, potentially resulting in: as the device ages, oxygen purity will gradually decrease" — a defect that produces no fire and no alarm, only a slow loss of delivered oxygen that a home patient may not detect until symptomatic.

The MAUDE Record: 53,000 Reports

The MAUDE adverse-event database shows the concentrator category's post-market scale. Under the four family product codes, we count 53,418 device reports, including 605 reports in which a patient death was recorded, 1,398 injury reports, and 51,143 malfunction reports. (MAUDE reports are filed voluntarily and by manufacturers under MDR obligations; a death report does not by itself establish device causation, and oxygen-therapy patients are often seriously ill, but the volume is the relevant benchmark for category-level surveillance.)

Annual report volume has risen steadily through the post-pandemic period:

The brand-level distribution is dominated by Invacare's home-concentrator platforms:

By manufacturer, Invacare Florida Operations alone is named on 46,137 of the 53,418 reports — roughly 86% of the family's MAUDE footprint — followed by Respironics (936), Caire (373), and Inogen (359). The PerfectO2 and Platinum lines are Invacare's flagship home concentrators; their dominance of the MAUDE record is a function of a very large installed base rather than necessarily a higher per-unit failure rate, but it means that Invacare's product decisions and recall history effectively set the category's safety baseline.

The Recall Record: Concentration and Recurrence

The FDA recall-initiation database records 56 concentrator-family recalls, with the recalling firms again concentrated in the legacy manufacturers:

Recall activity clusters in two periods — nine recalls initiated in 2012 and eight in 2022, with a further five in 2024, three in 2025, and one already in 2026 — suggesting a cyclical rather than monotonically declining pattern. The same firm names recur across decades: Nidek and Caire appear in both the 2012 and the 2022–2025 clusters, indicating that some root causes (capacitor reliability, sieve-bed purity drift) have persisted across product generations.

The 510(k) Entry Record: A Mature, Concentrated Market

We count 413 510(k) clearances under the concentrator family codes, the largest clearance record of any respiratory home-care device category we have analyzed. The clearance pathway is overwhelmingly Traditional (372 Traditional, 29 Special, 12 Abbreviated). The applicant field is dominated by the same firms that appear in the recall and MAUDE records:

Clearance activity has picked up recently — eight in 2023, 13 in 2025, and three in the first half of 2026 — partly reflecting new portable-concentrator entrants and partly replacement clearances as firms refresh platforms after the 2021 Philips Respironics recall reshaped the respiratory-device market. The combination of a large, decades-old clearance base and recurring fire-and-purity recalls means that concentrator post-market risk is effectively the post-market risk of a small number of long-running platforms from Invacare, Nidek, Caire, and DeVilbiss.

What This Means for Quality, Regulatory, and Home-Medical-Equipment Teams

The data supports four practical takeaways:

1. Design for the oxygen-enriched environment first. Fire and thermal faults are the leading root cause and the leading Class I driver; capacitor and power-supply component qualification, flame-retardant housings, and thermal cutoffs are where the highest-severity post-market exposure is created or avoided for this family.
2. Treat oxygen-purity monitoring as a life-critical function, not a comfort feature. The low-purity failure mode is silent — no fire, often no alarm — and the calibration-drift pattern means periodic verification of delivered oxygen concentration is a core risk control, especially for aging units in the field.
3. Track the installed base, not just the current catalog. Because Invacare, Caire, Nidek, and DeVilbiss recall and MAUDE records span decades of platforms, a home-medical-equipment provider's effective concentrator safety exposure is defined by its service and rental fleet, including legacy units still in patient homes.
4. Benchmark against the recent baseline. MAUDE report volume has roughly quadrupled from 2018 to 2025, and Class I actions continue into 2025; a long-run average understates current concentrator recall risk.

Home oxygen concentrators will remain a focus of FDA enforcement for as long as they enrich the atmosphere around a patient while running unattended. The public data — 44 enforcement actions, 20 of them Class I, and 53,000 MAUDE reports — is the evidence base, and the consistent answer is fire, electrical faults, and silent purity loss.

Data sources: FDA device enforcement report database, FDA Medical Device Recall database, FDA MAUDE adverse-event database, and FDA 510(k) database; analysis by MedDeviceGuide, run date 2026-06-14. Root-cause themes are derived from FDA reason-for-recall text and are not mutually exclusive. MAUDE death and injury counts are as-filed report counts and do not establish device causation. This article is educational and is not regulatory, quality-system, or clinical advice for a specific product.