TL;DR — The mechanism is photographed. The magnitude is unpublished. Electron microscopy has caught floss filaments snapped after a single use. The damage showed most clearly in a flat, compacted floss. Genuinely twisted threads showed only minor damage. Yet no study has ever weighed or counted what floss actually releases. Domestic laundry, by contrast, is measured again and again at 9.6-1,240 mg of microfiber per kilogram of textile per wash. Used floss goes to landfill rather than down the drain. So the wastewater pathway that dominates textile microfiber pollution barely applies to it.
No one has ever weighed the microplastic shed by a strand of dental floss. Not one published study reports a number. Not in milligrams, and not in particle counts. Laundry, by contrast, has been measured over and over. Consumer apparel releases 9.6 to 1,240 mg of microfiber per kilogram of textile per wash. Polyester garments average 161 ± 173 mg/kg. That asymmetry is the honest summary of floss and microplastics. One side measured, the other blank.
Figure 1 — Measured microfiber shedding per wash — and the missing floss bar
The mechanism is not in doubt. Floss is a polymer thread pulled under tension across hard enamel edges. And polymer threads pulled under tension shed fragments. But mechanism is not measurement. Most writing on this topic quietly swaps one for the other.
Here is every claim in the debate, ranked by how much evidence actually stands behind it.
The evidence, ranked
| Rank | Claim | Type of evidence | Quantified? |
|---|---|---|---|
| 1 | Filaments break during normal use | Direct electron microscopy of used floss | Imaged, never weighed |
| 2 | Oral-care products generate particles that can be swallowed | Mechanistic review of the product category | No floss-specific figure |
| 3 | Floss is a plausible microplastic source | Systematic review of dentistry | Explicitly "sparse" |
| 4 | Floss contributes measurably to wastewater microplastic | Inference only | No data, and the wrong disposal route |
| 5 | Floss microplastic causes disease | None | No direct evidence |
Tier 1: the fibers do break, and it has been photographed
The strongest evidence here is also the least dramatic. A 2022 study imaged four commercial flosses before and after real use (Stavrakis et al., Materials). The sample with clearly fractured filaments after a single use was a flat, compacted multifilament floss. The two genuinely twisted multifilament threads showed only minor visible damage. A flat tape came through with its rectangular profile intact. Some of its fibers twisted, almost none broke. The damaged sample went from roughly 50 % elongation at break when new to about 150 % after use. Its internal structure was wrecked. Yet its breaking force fell by only about 10 N.
Figure 2 — Four floss materials: structure and surface roughness
That detail matters. Fraying is not surface polishing. It is filaments breaking. A broken filament end is, by definition, a fiber fragment.
Two limits, stated plainly. The paper identifies products by brand, not by polymer. Its own results also refuse the tidy sentence "twisted sheds, tape does not". The worst-damaged sample was the flat, compacted one. And the paper never collected or weighed the debris.
Structure still sets a ceiling on what can break. A 2023 comparison of four floss materials describes PTFE floss as a broad, sheet-like, layered structure rather than twisted filaments. Nylon, silk and UHMWPE are all "many filaments wound into thread" (Huang, Broadbent & Choi, Biomaterial Investigations in Dentistry). A true sheet has no filaments to snap. But the 2022 images show that pressing filaments flat is no guarantee they survive. The same study measured surface roughness at Ra 0.0478 ± 0.0034 µm for PTFE against 0.3035 ± 0.0254 µm for silk. That is a sixfold spread in how much the thread grips as it passes.
The same paper also carries the sharpest counter-evidence in the file. It ran a 16-person split-mouth trial. A large majority of responses disagreed that any of the four flosses frayed or shredded during use. Whatever happens at the micrometer scale, most people never see it.
Tier 2: the swallowing pathway is described, not measured
A review of micro- and nanoplastics from oral-care products sets out the release mechanism. Particles come off toothpaste, brushes and orthodontic materials through mechanical friction and temperature swings. Those released particles can be ingested (Saha et al., Ecotoxicology and Environmental Safety). Friction against a sharp enamel edge is exactly that mechanism. The review supplies no figure for floss specifically. No such figure exists.
Tier 3: the review that says the quiet part
The most useful single source on this topic is a 2025 systematic review of micro- and nanoplastics in dentistry. It confirms that floss is made from nylon, polyethylene or PTFE. That, the review says, makes it a plausible source of microplastic in the mouth. It then states that the quantitative data are sparse. It also states that no direct evidence links microplastic of dental origin to systemic disease (Šimunović, Bačić & Meštrović, Materials).
Treat any article that hands you a confident particles-per-use figure as fabricated until it shows a method.
The reasons that number is missing
Three of them, and none will be solved soon.
- The debris is unrecoverable. Fragments end up in saliva, in the sink, on a tissue, or swallowed. There is no closed system to filter here. A washing machine drum, by contrast, can be filtered.
- No standard test exists. The nearest established laboratory protocol is an enamel abrasion test. In the 2023 four-material comparison it found no significant difference between PTFE, nylon, silk and UHMWPE. That test measures damage to the tooth, not loss from the thread.
- The masses are tiny. Gravimetric microfiber work on textiles operates on kilograms of fabric per run. A single used strand sits orders of magnitude below that. The method would have to be rebuilt around particle counting rather than weighing.
Tier 4: wastewater is mostly the wrong pathway
Wastewater is where the large microplastic numbers live. It is tempting to file floss there. It does not belong there, for a mundane reason: used floss goes in the trash. A US municipal utility lists floss among the things never to flush. Its grounds are that floss is not biodegradable and catches on everything in the sewer. The same utility places floss under landfill waste (San Francisco Public Utilities Commission).
For whatever fraction does get flushed, scale is worth understanding. A systematic review of 15 treatment plants found removal efficiencies from 35 % to 99.9 %. Primary treatment did most of the work (78–98 %). Secondary treatment added 7–20 %. Tertiary treatment added no significant further removal. Even a well-performing plant in Glasgow still discharges roughly 65 million microplastic particles a day. Much of what is captured ends up in sludge spread on farmland (Cristaldi et al., IJERPH).
None of that is floss data. It is context for why a high removal percentage still leaves enormous absolute numbers. It also shows why the pathway a product takes matters more than the polymer it is made from. Laundry has a direct, continuous water route. Floss has an indirect and largely closed one.
Tier 5: the health claim has nothing behind it — in either direction
One short section, because that is all the literature supports. The same 2025 systematic review grants floss plausibility as a source. It also states that no direct evidence links microplastic of dental origin to systemic disease. That absence reflects an absence of studies, not a clean bill of health. And it cuts equally against alarm and against reassurance. Anyone asserting harm, or asserting safety, is ahead of the data.
The textile comparison, with one side blank
This is the comparison people actually want.
Textiles, measured. Domestic washing of consumer apparel sheds 9.6–1,240 mg of microfiber per kilogram of textile per wash. Mechanically-treated polyester averaged 161 ± 173 mg/kg. Woven nylon averaged 27 ± 14 mg/kg. Cotton and wool averaged 165 ± 44 mg/kg (Vassilenko et al., PLOS ONE). At the polyester average, a single 6 kg load lands on the order of one gram of shed fiber. That is our arithmetic on their figure. And that gram goes straight down the drain, every wash, every week.
Floss, unmeasured. The dental association recommends about 18 inches per session, once a day (ADA MouthHealthy). That works out to 0.457 m × 365 = 167 meters a year. The figure assumes daily use and no waste. The total mass of thread involved is small. But "small" is not a measurement. No published study has weighed a used strand. And the fraction leaving as fragments is unknown.
These two figures cannot be added, subtracted or ranked against one another. They are not the same kind of number. Here is what can be said. Only one household microplastic source has been measured in grams, on a weekly cycle, with a direct route to water. That source is the washing machine, not the bathroom cabinet.
Plastic-free thread does not mean fragment-free
Changing the polymer changes the end-of-life question far more than the shedding question. A thread that snaps filaments still snaps filaments, whatever it is made of.
And "biodegradable" only means something with its conditions attached. Melt-spun PLA fibers disintegrated solely under industrial composting conditions. That meant three weeks at 58 °C under ISO 20200. Samples held at 28 °C under home-composting conditions showed no disintegration at all after 14 weeks (Naeimirad et al., Scientific Reports). US federal guidance is blunter than most marketing. An unqualified degradable claim is deceptive if the item does not completely decompose within one year after customary disposal. Items entering landfills or incinerators do not (16 CFR §260.8, FTC Green Guides).
Used floss customarily goes to landfill. So a bare "biodegradable floss" claim tells you something about the seller rather than about the fiber.
Verdict
The five claims from the table, in the same order.
- Filaments break during use. Photographed, after one use, on real products. The damage showed most clearly on a flat, compacted floss. Twisted threads showed only minor damage. Solid.
- Fragments can be swallowed. The mechanism is established for oral care broadly. There is no floss-specific number.
- Floss is a plausible source. The dentistry review confirms it. That same review calls the quantitative data sparse. The total quantity released is unknown. Not "negligible" — unknown.
- Wastewater contribution. Almost certainly minor, because the disposal route is landfill rather than the drain.
- Health effect. No direct evidence, in either direction.
And outside the ranking, there is the comparison that started this article. Laundry is the measured, gram-scale, water-connected microplastic source in the same household. Floss is not in the same league of measurement. It is not even in the same league of mass.
The defensible position is neither alarm nor dismissal. Floss is a small, real, unquantified plastic input. Nobody disputes its mechanism. Nobody has published its magnitude. Anyone offering a precise figure is filling that gap with marketing.
And anything involving bleeding, pain or persistent gum symptoms is a conversation with a dentist. Not with an article about polymers.
