ATSDR 2005 — Toxicological Profile for Tin and Tin Compounds

The August 2005 ATSDR Toxicological Profile for Tin and Tin Compounds (Profile No. 55, “tp55”) is the comprehensive U.S. toxicology synthesis for elemental tin, inorganic tin compounds (stannous and stannic ions, principally entering the diet through tinplated cans), and organotin compounds (mono-, di-, and tributyltin; triphenyltin; trialkyltins), prepared under CERCLA Section 104(i) authority and superseding the September 2003 draft for public comment. The profile maintains an enforced inorganic-vs-organotin species distinction throughout: derived oral Minimal Risk Levels (MRLs) span three orders of magnitude between inorganic tin (0.3 mg Sn/kg/day, intermediate) and tributyltin oxide (0.0003 mg/kg/day, intermediate and chronic), reflecting the fundamentally different toxicological profiles of the two classes.

Key numbers

Minimal Risk Levels (oral exposure)

No inhalation MRLs were derived for either inorganic or organic tin compounds on adequacy-of-data grounds. No oral MRL was derived for triphenyltin, triethyltin, trimethyltin, dioctyltin, or for inorganic tin at acute or chronic duration (Section 2.3, pp. 16-22). The thousand-fold gap between the inorganic-tin intermediate MRL and the tributyltin oxide MRL is the central toxicological fact in the profile and is what makes inorganic-vs-organotin speciation non-substitutable.

Tin levels in food (Biégo et al. 1999, French total diet study; Table 6-3, pp. 270-271)

Summary aggregations elsewhere in the profile (Section 6.4.4, p. 275) restate: “Foods in lacquered cans generally were found to contain tin concentrations below 10 mg/kg, and tin concentrations ranged from 24 to 156 mg/kg in food from unlacquered cans.” A 1994 U.K. total diet study (Ysart et al. 1999) reported mean tin in canned vegetables and fruit products at 44 and 17 mg/kg fresh weight, respectively. Mean tin in evaporated, concentrated, and sweetened condensed canned milk products (Lithuania, 1990-1992; Ramonaityte 2001) was 85, 89, and 40 mg/kg, respectively, compared with 0.22 mg/kg in raw milk. The Public Health Statement (Section 1.3, p. 3) gives the consumer-facing summary: foods in lacquered cans contain “less than 25 ppm” tin and foods in unlacquered cans “up to 100 ppm,” and “>90% of tin-lined cans used for food today are lacquered” (Can Manufacturers Institute, CMI 1988). The Biégo Table 6-3 distribution above provides the underlying empirical basis.

Tributyltin (TBT) in seafood (selected geographies; Table 6-4, pp. 272-274)

Eight worldwide market surveys (Keithly et al. 1999) report mean TBT concentrations in seafood as ng TBT/g wet weight. Across categories, bivalves, pelagic fish, pelagic invertebrates, and flatfish averaged 40, 16, 7.4, and 4.6 µg TBT/kg, respectively (Section 6.4.4, p. 275). Selected market means:

A separate six-market U.S. summer/winter survey (Cardwell et al. 1999b) detected TBT in 35% of samples; summer means by category were 0.98-7.1 µg/kg for bottom fish, 1.0-2.4 for crustaceans, and 2.0-13 for mollusks. A Japanese market-basket study (Tsuda et al. 1995) found TBT and triphenyltin only in the fish-mollusks-crustaceans group (5.2 µg TBT/kg, 0.4 µg TPT/kg) and the vegetables-and-seaweeds group (0.2 and 0 µg/kg). Sediment is the major TBT sink in coastal environments (Section 6.4.3, pp. 265-269); Coddington Cove, Rhode Island sediments measured 0.032-0.372 mg Sn/kg dry weight (Wade et al. 2004), and intertidal sites in Portland and Boothbay Harbor, Maine ranged 0.024-12.4 mg TBT/kg dry weight (Page et al. 1996).

Tin and organotin in human tissues and fluids (Table 6-5, p. 277)

Inorganic tin and butyltin distributions differ sharply, reflecting their differing absorption and partitioning. NHANES III (1988-1994; n=500; Paschal et al. 1998) detected tin in 89% of urine samples (detection limit 0.1 µg/L) with a 95th-percentile concentration of 20.1 µg/L. Adipose-tissue tin in the 1982 National Human Adipose Tissue Survey (n=9; Stanley 1986) ranged 4.6-15 µg/g. Spanish autopsy series (Llobet et al. 1998, n=20; García et al. 2001, n=78) found tin highest in bone (means of 6.18 and 0.47 µg/g wet weight, respectively) and lowest in brain (0.98 and 0.27 µg/g wet weight). Human butyltin blood and liver concentrations are reported on the order of single-digit to low-tens ng/g wet weight: Michigan blood (n=32; Kannan et al. 1999) mean monobutyltin / dibutyltin / tributyltin 8.17 / 4.94 / 8.18 ng/mL; Polish liver (n=9; Kannan and Falandysz 1997) total butyltin range 2.4-11.0 ng/g wet weight; Japanese liver (n=4; Takahashi et al. 1999) mean monobutyltin 18 ng/g, dibutyltin 66 ng/g, tributyltin <2 ng/g; Danish liver (n=18; Nielsen and Strand 2002) tributyltin and triphenyltin both below detection (<0.3 and <3 ng/g), with mean dibutyltin 9 ng/g and >25-fold inter-individual variability.

NPL and exposure summary

Tin and organotin compounds had been identified at 214 and 8, respectively, of the 1,662 hazardous-waste sites then proposed for the EPA National Priorities List (HazDat 2004; Section 6.1, p. 249). The average daily tin intake of a U.S. adult was estimated at 4.003 mg (4 mg from food, 0.003 mg from air; drinking water undetectable; Section 2.1, p. 11). Tinplated cans are the dominant dietary route for inorganic tin (Biégo et al. 1999, WHO 2003); seafood from coastal waters and contact with PVC/polyurethane consumer products carrying alkyltin stabilizers are the principal routes for organotin exposure (Section 6.2, pp. 253-256; Section 6.5, p. 282).

Methods

The profile is a literature synthesis prepared in accordance with the ATSDR-EPA guidelines published in the Federal Register on April 17, 1987. The text covers chemical and physical properties (Chapter 4), production and use (Chapter 5), environmental fate and exposure (Chapter 6), analytical methods for biological materials and environmental samples (Chapter 7), and regulations and advisories (Chapter 8). MRL derivations follow the standard ATSDR framework documented in Appendix A: identification of the most sensitive NOAEL or LOAEL across the available studies for each route-duration cell, then division by an uncertainty factor that combines animal-to-human extrapolation (10), human inter-individual variability (10), and where applicable an extra LOAEL-to-NOAEL factor (10) and species variability factor (10). External CERCLA-mandated peer review was conducted by Michael Aschner (Wake Forest University School of Medicine), Olen Brown (University of Missouri-Columbia), and Bruce Jarnot (American Petroleum Institute).

Inorganic-tin and organotin mechanisms

Inorganic tin’s principal toxic effect after ingestion is local irritation of the gastrointestinal mucous membrane producing acute nausea, vomiting, and diarrhea, with symptomatic thresholds in single-dose human challenges reported on the order of thousands of ppm tin in acidic canned juices. Systemic absorption is limited (more than 90% of an ingested dose is excreted in feces; bone is the primary depot for the small absorbed fraction). Reported repeat-dose effects from inorganic tin include hematological changes (hemoglobin and hematocrit reductions, reversible with iron or copper enrichment of the diet), and at substantially higher doses liver, kidney, and pancreatic effects in rodent studies.

Organotin mechanisms are species-dependent. Tributyltin and dibutyltin are immune toxicants and the rodent thymus is the most-sensitive organ in repeat-dose dietary studies, supporting the immune-system-endpoint derivations of both organotin MRLs (Vos et al. 1990; Seinen et al. 1977b). Triethyltin and trimethyltin are central nervous system toxicants producing ataxia, paralysis, and self-mutilating behavior at sub-mg/kg doses (Magee et al. 1957; Bouldin et al. 1981; Brown et al. 1984). Some organotins are reproductive or developmental toxicants in rodents (Ema and Harazono 2000; Ema et al. 1997b; Adeeko et al. 2003), and a few are tumorigenic in chronic-duration animal studies. Organotins generally do not penetrate intact human skin in significant amounts at environmental exposure levels, but skin and eye irritation have been reported after acute high-concentration occupational contact.

Provenance notes

The manual-fetch PDF in the June 3 Folder is byte-identical (SHA-256 de21f125...e313f4, 7.7 MB) to the earlier raw/Digest/Bookshelf_NBK599935.pdf retrieval from the NCBI Bookshelf record for tp55. The 2005 final profile supersedes the September 2003 draft for public comment. ATSDR has not finalized a post-2005 update for tin as of the access date; the 2005 profile remains the operative U.S. agency reference. Profile authored by Carolyn Harper, Fernando Llados, Gary Diamond, and Lara L. Chappell with external peer review by Aschner, Brown, and Jarnot. License us-government-work; no copyright restriction applies. Canonical agency URL: atsdr.cdc.gov/toxprofiles/tp55.pdf; secondary host: NCBI Bookshelf NBK599935.

Verification notes

This page was merge-enhanced on 2026-06-04 from a pre-2026-05-14 revision. Corrections made on this pass against the source PDF (tp55, August 2005):

• “Dibutyltin chloride” in the prior MRL table was corrected to “Dibutyltin dichloride” — the source consistently names the test article as “dibutyltin dichloride” (Section 2.3, p. 19; Seinen et al. 1977b).
• The prior “Selected tin levels in foods” paragraph cited “ATSDR Tables 6-2 and 6-3” but Table 6-2 is “Organotin Levels in Sediment” (p. 267-268), not food. The correct table is 6-3 alone. The numeric values in the prior paragraph (“meats around 0.6-2.9 mg/kg,” “fishes 6.0 mg/kg,” “canned alcoholic beverages around 4.5 mg/kg”) do not appear in Table 6-3 and were removed; the actual Table 6-3 values are now reported in the Key numbers food block above (e.g., lacquered-can meats mean 4.5 mg/kg n=4 range 1.1-9.4; lacquered-can fishes mean 0.7 mg/kg n=4 range 0.3-0.9; alcoholic beverages and canned alcoholic beverages both reported <0.003 mg/kg in fresh foods).
• The tributyltin oxide MRL row now records the actual point of departure (NOAEL 0.025 mg/kg/day) rather than the bare “NOAEL” placeholder of the prior table.
• Legacy headings ## Summary and ## Wiki pages updated on ingest were removed per current Part 6 template. The Wiki-pages section is superseded by the routing audit, which fans this source out to product, ingredient, and metal pages automatically from the frontmatter.
• raw_handle, raw_sha256, and near_duplicates fields were added; raw_path updated to the manual-fetch PDF (byte-identical to the prior Bookshelf retrieval). access_url updated to the canonical ATSDR URL; NCBI Bookshelf retained as secondary host in the provenance note.
• source_type normalized from government_report (4 instances corpus-wide) to gov-report (77 instances; current convention).
• Audit subagent (2026-06-04) verdict PROMOTE; one minor ⚠️ concern applied — fresh-foods row label in Table 6-3 block previously named “carrots, cabbage, salad, potatoes, etc.” with upper anchor 0.13 mg/kg, but 0.13 is the lentils mean (range 0.09-0.17) and potatoes range to 0.2 mg/kg. Row rewritten to anchor the upper bound on lentils and note the potatoes maximum, with alcoholic beverages added to the <0.003 row per Table 6-3.

Page history

The five most recent substantive edits to this page. The full version history lives in git; when DOI minting comes online (see schema docs), each entry below will also link to a version-pinned DataCite DOI.