{
  "id": "02",
  "title": "Glucosamine and hexosamine stress",
  "summary": "Naturally elevated glucosamine may disrupt nerve metabolism enough to cause terminal loss.",
  "source_title": "Excess glucosamine and hexosamine-associated metabolic stress",
  "page": "02_glucosamine_hexosamine.html",
  "source_markdown": "sources/02_glucosamine_hexosamine.md",
  "structured_data": "data/02_glucosamine_hexosamine.json",
  "snapshot_date": "2026-09-14",
  "source_review_date": "2026-09-14",
  "scope_label": "Early type 2 diabetes",
  "scope_exception": false,
  "source_headline_markdown": "**About 5% probability; uncertainty range (sensitivity) 0.1–38%.** The central product is 5.4%. Experimental glucosamine exposure can damage peripheral nerves. The principal uncertainty is whether naturally occurring free glucosamine in early human diabetes reaches a concentration and duration that produces material terminal loss. Literature reviewed 14 September 2026.",
  "probability_percent": {
    "skeptical": 0.13125,
    "base": 5.445,
    "favorable": 38.08
  },
  "probability_meaning": "Subjective belief in the complete causal proposition. Scenario endpoints are sensitivity products, not confidence limits.",
  "causal_proposition_markdown": "In adults with type 2 diabetes diagnosed within five years, without confirmed distal polyneuropathy and with initially preserved distal-leg skin innervation, an increase in **free D-glucosamine** in sensory neurons or their connected Remak Schwann cells contributes to terminal loss. Keeping that exposure at its matched nondiabetic concentration, starting before prospective loss, would avert at least 20% of five-year net peptidergic C-fiber terminal loss in a baseline subgroup containing at least 10% of incident DPN cases under reference care.\n\nFree glucosamine, glucosamine-6-phosphate and UDP-N-acetylglucosamine are different molecular species. This proposition concerns free glucosamine and the consequences of changing it. It does not assert that every increase in hexosamine-pathway activity is harmful, or that elevated O-GlcNAc protein modification is required. ATP restriction, competition with glucose metabolism and downstream protein modification are possible explanations for a glucosamine effect, not compulsory serial requirements. The total causal effect of the specified exposure intervention is the priced quantity.\n\nThe ideal intervention buffers free glucosamine to the matched physiological concentration in the specified neuron or Remak-cell compartment. It preserves extracellular glucose, insulin, lipids and other initiating exposures; glucosamine-derived metabolites and downstream cell functions may change. It does not abolish OGT, hexokinase or essential basal glycosylation. Buffering a named small molecule is an ideal biochemical capability, not a claim that a current inhibitor selectively does this. A separate restoration of the original glucosamine trajectory and a second selective manipulation distinguish the exposure effect from artifacts of buffering.\n\nThe allowed baseline groups are the whole eligible population, HbA1c ≥7.5%, or local free glucosamine above the matched nondiabetic 95th percentile in the specified compartment. Membership is measured before manipulation, without selecting future responders. An allowed group must have concentrations above that reference for at least 90 cumulative days during the first year, beginning before prospective loss. The same group and connected nerve unit must satisfy the complete claim. These concentration, duration and subgroup rules bound the question rather than establish observed toxicity thresholds.\n\nCount terminal disappearance and successful replacement in a fixed distal-skin territory, retaining stable peptidergic identity rather than counting peptide staining alone. Net loss is disappearance minus replacement over five years, divided by baseline terminal number. Define B = (mean reference loss − mean intervention loss)/mean reference loss, with a positive reference mean. Materiality requires B ≥0.20. Coverage C is the fraction of new DPN cases under reference care belonging to the qualifying baseline group, requiring C ≥0.10. New DPN means bilateral length-dependent sensory signs and objective small-fiber loss or abnormal nerve conduction, without another neuropathy. These chosen definitions are distinct from supplement effects, type 1 diabetes, or rescue of established disease.",
  "claims": [
    {
      "id": "R1",
      "source_id": "R1",
      "claim": "At least one allowed group has the defined persistent pre-loss elevation of free glucosamine in a connected neuronal or Remak-cell compartment.",
      "claim_markdown": "At least one allowed group has the defined persistent pre-loss elevation of free glucosamine in a connected neuronal or Remak-cell compartment.",
      "conditional_percent": {
        "skeptical": 25.0,
        "base": 55.0,
        "favorable": 80.0
      },
      "cumulative_base_percent": 55.0,
      "source_cumulative_display": null,
      "reason_markdown": "Diabetic mouse nerve metabolomics supports natural accumulation, but only as relative pooled signals. Human absolute concentration, localization and early duration are missing.",
      "question": "Is free glucosamine persistently elevated in sensory neurons or their Remak Schwann cells before terminal loss?"
    },
    {
      "id": "R2",
      "source_id": "R2",
      "claim": "In a group satisfying R1, normalizing free glucosamine averts at least 5% of five-year net human peptidergic terminal loss.",
      "claim_markdown": "In a group satisfying R1, normalizing free glucosamine averts at least 5% of five-year net human peptidergic terminal loss.",
      "conditional_percent": {
        "skeptical": 7.0,
        "base": 30.0,
        "favorable": 70.0
      },
      "cumulative_base_percent": 16.5,
      "source_cumulative_display": null,
      "reason_markdown": "Imposed exposure produces large mouse fiber loss and adult rat neuron injury. Naturally elevated human exposure could nevertheless be far below the active dose; adaptation and support-cell transfer are unresolved.",
      "question": "Would restoring local free glucosamine to normal concentrations avert at least 5% of five-year net terminal loss?"
    },
    {
      "id": "R3",
      "source_id": "R3",
      "claim": "In a group satisfying R1–R2, preservation reaches B ≥0.20.",
      "claim_markdown": "In a group satisfying R1–R2, preservation reaches B ≥0.20.",
      "conditional_percent": {
        "skeptical": 25.0,
        "base": 55.0,
        "favorable": 80.0
      },
      "cumulative_base_percent": 9.075,
      "source_cumulative_display": null,
      "reason_markdown": "Given a real structural contribution, a large experimental effect supports possible materiality. It provides no human estimate of competing causes or the fraction attributable to this exposure.",
      "question": "Would restoring local free glucosamine to normal concentrations avert at least 20% of five-year net terminal loss?"
    },
    {
      "id": "R4",
      "source_id": "R4",
      "claim": "At least one group satisfying R1–R3 contains C ≥0.10 of reference incident cases.",
      "claim_markdown": "At least one group satisfying R1–R3 contains C ≥0.10 of reference incident cases.",
      "conditional_percent": {
        "skeptical": 30.0,
        "base": 60.0,
        "favorable": 85.0
      },
      "cumulative_base_percent": 5.445,
      "source_cumulative_display": null,
      "reason_markdown": "Hyperglycemia is common; substantial local free-glucosamine accumulation in the materially affected group may not be. No representative case-coverage measurement exists.",
      "question": "Does the benefiting subgroup account for at least 10% of incident DPN cases under reference care?"
    }
  ],
  "sections": [
    {
      "heading": "The causal proposition",
      "markdown": "In adults with type 2 diabetes diagnosed within five years, without confirmed distal polyneuropathy and with initially preserved distal-leg skin innervation, an increase in **free D-glucosamine** in sensory neurons or their connected Remak Schwann cells contributes to terminal loss. Keeping that exposure at its matched nondiabetic concentration, starting before prospective loss, would avert at least 20% of five-year net peptidergic C-fiber terminal loss in a baseline subgroup containing at least 10% of incident DPN cases under reference care.\n\nFree glucosamine, glucosamine-6-phosphate and UDP-N-acetylglucosamine are different molecular species. This proposition concerns free glucosamine and the consequences of changing it. It does not assert that every increase in hexosamine-pathway activity is harmful, or that elevated O-GlcNAc protein modification is required. ATP restriction, competition with glucose metabolism and downstream protein modification are possible explanations for a glucosamine effect, not compulsory serial requirements. The total causal effect of the specified exposure intervention is the priced quantity.\n\nThe ideal intervention buffers free glucosamine to the matched physiological concentration in the specified neuron or Remak-cell compartment. It preserves extracellular glucose, insulin, lipids and other initiating exposures; glucosamine-derived metabolites and downstream cell functions may change. It does not abolish OGT, hexokinase or essential basal glycosylation. Buffering a named small molecule is an ideal biochemical capability, not a claim that a current inhibitor selectively does this. A separate restoration of the original glucosamine trajectory and a second selective manipulation distinguish the exposure effect from artifacts of buffering.\n\nThe allowed baseline groups are the whole eligible population, HbA1c ≥7.5%, or local free glucosamine above the matched nondiabetic 95th percentile in the specified compartment. Membership is measured before manipulation, without selecting future responders. An allowed group must have concentrations above that reference for at least 90 cumulative days during the first year, beginning before prospective loss. The same group and connected nerve unit must satisfy the complete claim. These concentration, duration and subgroup rules bound the question rather than establish observed toxicity thresholds.\n\nCount terminal disappearance and successful replacement in a fixed distal-skin territory, retaining stable peptidergic identity rather than counting peptide staining alone. Net loss is disappearance minus replacement over five years, divided by baseline terminal number. Define B = (mean reference loss − mean intervention loss)/mean reference loss, with a positive reference mean. Materiality requires B ≥0.20. Coverage C is the fraction of new DPN cases under reference care belonging to the qualifying baseline group, requiring C ≥0.10. New DPN means bilateral length-dependent sensory signs and objective small-fiber loss or abnormal nerve conduction, without another neuropathy. These chosen definitions are distinct from supplement effects, type 1 diabetes, or rescue of established disease."
    },
    {
      "heading": "What the chemistry does and does not imply",
      "markdown": "Glucosamine entering a cell can be phosphorylated by hexokinase, consuming ATP. By contrast, the canonical hexosamine biosynthetic pathway starts with fructose-6-phosphate and glutamine to form glucosamine-6-phosphate. Greater flux through that reaction does not automatically create a large pool of free glucosamine. Pool size depends on production, uptake, phosphorylation, deamination and export, and is not itself a flux measurement.\n\nA large glucosamine load could compete with glucose uptake/phosphorylation, consume energy or change mitochondrial hexokinase association. Lower ATP could limit ion gradients and maintenance. Yet increased hexokinase expression alone does not prove greater ATP production or wasteful ATP consumption; it can be a response to stress. Inosine is a metabolizable purine, not a selective replacement for missing neuronal ATP. Its rescue therefore supports an energy interpretation without uniquely establishing it.\n\nUDP-GlcNAc supplies O-GlcNAc transferase (OGT), which adds a sugar modification to intracellular proteins; O-GlcNAcase removes it. Those modifications regulate transport, signaling and survival. Some are protective. A free-glucosamine toxicity result does not justify suppressing all O-GlcNAcylation. Nor must a glucosamine effect disappear merely because one attempt to increase total O-GlcNAcylation failed to worsen it."
    },
    {
      "heading": "Conditional probabilities",
      "markdown": "Each probability assumes all preceding requirements hold in at least one common baseline group and connected compartment. Each cumulative event retains only groups that satisfy every preceding requirement. ATP mediation is not an additional hidden requirement: R2 estimates the structural consequence of the defined glucosamine intervention, however that consequence is mediated.\n\n| Claim | Additional proposition | Best probability | Uncertainty range (sensitivity) | Reason for estimate and breadth |\n|---|---|---:|---:|---|\n| R1 | At least one allowed group has the defined persistent pre-loss elevation of free glucosamine in a connected neuronal or Remak-cell compartment. | 55% | 25–80% | Diabetic mouse nerve metabolomics supports natural accumulation, but only as relative pooled signals. Human absolute concentration, localization and early duration are missing. |\n| R2 | In a group satisfying R1, normalizing free glucosamine averts at least 5% of five-year net human peptidergic terminal loss. | 30% | 7–70% | Imposed exposure produces large mouse fiber loss and adult rat neuron injury. Naturally elevated human exposure could nevertheless be far below the active dose; adaptation and support-cell transfer are unresolved. |\n| R3 | In a group satisfying R1–R2, preservation reaches B ≥0.20. | 55% | 25–80% | Given a real structural contribution, a large experimental effect supports possible materiality. It provides no human estimate of competing causes or the fraction attributable to this exposure. |\n| R4 | At least one group satisfying R1–R3 contains C ≥0.10 of reference incident cases. | 60% | 30–85% | Hyperglycemia is common; substantial local free-glucosamine accumulation in the materially affected group may not be. No representative case-coverage measurement exists. |\n\nThe 5% structural boundary distinguishes a small effect from material preservation and is nested inside 20%. A 30% reference loss versus 28% intervention loss gives 6.7% preservation: R2 YES, R3 NO. Identifying injury at a deliberately high laboratory dose does not settle R2, which assumes R1's natural human concentration and timing.\n\nBase [55,30,55,60]% gives **5.445%**. Coordinated skeptical [25,7,25,30]% gives **0.131%**; favorable [80,70,80,85]% gives **38.08%**. Merging structural causation and materiality gives a conditional 16.5%, preserving the same product. A separate whole-claim assessment is “unlikely but biologically possible”: an imposed-exposure structural anchor exists, but its actual dose and the human exposure bridge are unresolved. That supports a few-percent judgment, without distinguishing 5% from 7% confidently."
    },
    {
      "heading": "Evidence that determines the judgment",
      "markdown": "**Natural diabetic exposure is observed in mice, not quantitatively bridged to humans.** Mizukami compared wild-type and aldose-reductase-knockout mice after 12 weeks of STZ diabetes. Each metabolomics observation pooled bilateral sciatic nerves from four mice; there were three independent pools per group. Free glucosamine appeared in the diabetic groups. The supplementary table reports an increase without a usable numerical fold change or P value for that metabolite; the figure gives relative peak areas, not molar concentrations. UDP-GlcNAc ratios were 1.0 in both diabetic genotypes (P=.87 and .59). These data support a candidate exposure while providing no absolute toxic-dose match or inference n larger than three pools/group. [Original study and supplement](https://doi.org/10.1093/braincomms/fcaa168).\n\n**There is substantial imposed-exposure injury.** In the same study, 12-week glucosamine infusion reduced skin-fiber density. All twenty plotted Figure 9B points give approximate means **26.6 versus 10.3 fibers/mm** in untreated versus exposed wild-type mice and **26.1 versus 13.0** in knockouts, five mice/group. That is approximately 61% and 50% lower density. The wild-type difference is approximately −16.3 fibers/mm, unadjusted Welch 95% interval −19.0 to −13.6, before image-reading error. This is a large mixed-fiber endpoint at imposed exposure, not a selective diabetic rescue. [Figure 9](https://pmc.ncbi.nlm.nih.gov/articles/PMC7713992/); [reproducible quantitative support](../../../findings/damage_calibration_glucosamine_quantitative.md).\n\nThe chronic dose descriptions conflict. Methods specify 4.8 mg/mL at 0.15 μL/hour, implying **0.576 mg/kg/day** for a 30 g mouse. Figure 8 instead states **170 mg/kg/day**, about 295 times greater, and a different pump rate. The methods' additional equation of 4.8 mg/mL with 1 M is also inconsistent. Supplementary methods do not resolve this. The observed structural effect survives as evidence of an imposed exposure, but a precise dose-response or human-equivalent exposure cannot be reconstructed from the printed protocol.\n\n**The energy mechanism has intervention support but incomplete specificity.** Immortalized mouse Schwann cells received 1–10 mM glucosamine; 10 mM over 24 hours produced roughly 40% Annexin/propidium-positive cells and reduced ATP. Hexokinase-I knockdown and inosine reduced injury markers. PUGNAc 100 μM increased O-GlcNAcylation without further worsening viability, and antioxidant or ER-stress interventions did not reproduce all rescue effects. Adult female rat DRG cultures exposed to 1 or 10 mM for up to six days also lost viable neurons and neurite growth. There were three to four culture experiments, with no basis to count individual neurites as independent donors. Those observations favor an energy-related account over a mandatory total-O-GlcNAc toxicity model, without establishing a unique mediator. [Primary methods, Figures 3–7 and supplement](https://doi.org/10.1093/braincomms/fcaa168).\n\n**Available human dose measurements create an objection, not a nerve concentration bound.** In twelve healthy volunteers, Persiani measured endogenous plasma glucosamine at 10.4–204 ng/mL, approximately **0.058–1.14 μM**. Repeated 1,500 mg oral glucosamine sulfate produced peaks near **10 μM**. The 1–10 mM nerve-culture exposures are 100–1,000 times that supplemented plasma peak. Intracellular nerve production and compartment trapping could differ, so these measurements do not exclude a local diabetic accumulation. They do show why oral dosing, free plasma exposure and local hexosamine metabolism cannot be interchanged. [Primary pharmacokinetic study](https://pubmed.ncbi.nlm.nih.gov/16168682/).\n\nA prospective UK Biobank analysis included 21,171 people with type 2 diabetes free of recorded microvascular complications, with 831 incident coded neuropathy cases over median 12.3 years. Habitual glucosamine use had adjusted neuropathy HR **0.88, 95% CI 0.71–1.08**. This does not suggest a large excess risk among supplement users, but exposure was self-reported without nerve concentrations, dose or randomized allocation. Healthy-user differences and coded rather than fiber-specific outcomes limit causal interpretation. It neither demonstrates protection nor excludes an endogenous intracellular route. Only published public aggregates are used here. [Cheng 2025](https://pmc.ncbi.nlm.nih.gov/articles/PMC11961739/).\n\n**Normal O-GlcNAc function supports nerve survival.** Adult-onset sensory-neuron OGT deletion at seven weeks of age caused approximately 60% fewer epidermal fibers four weeks later, with five mice/condition at that endpoint. The use of adult induction makes this more than a developmental phenotype. It is a strong reason not to treat total OGT inhibition as selective prevention of glucosamine toxicity. [Su and Schwarz 2017, Figure 7](https://pubmed.ncbi.nlm.nih.gov/28115479/). Schwann-cell OGT deletion likewise produced myelin abnormalities and progressive axonal degeneration; structural quantification used small groups, including three mice/genotype. That is supporting-cell dependence in a largely myelinated system, not proof of a free-glucosamine cause of diabetic C-fiber loss. [Kim 2016](https://pmc.ncbi.nlm.nih.gov/articles/PMC5039245/).\n\n**There are context-specific effects in human cells.** In H9 stem-cell-derived sympathetic neurons, changing medium from 5 to 30 mM glucose for five days increased activity and O-GlcNAcylation. Low-dose Ac-5SGlcNAc, 12.5 μM, reduced hyperactivity, oxidant signal and mitochondrial depolarization. Key assays used three to five differentiations from one stem-cell line, not three to five independent donors. Stronger or differently timed suppression impaired development/survival. This supports context-sensitive O-GlcNAc biology in human autonomic neurons, while leaving free-glucosamine concentration and sensory-terminal preservation unanswered. [Human sympathetic-neuron study, 2023](https://pmc.ncbi.nlm.nih.gov/articles/PMC10203903/).\n\nIn C. elegans, OGT loss and PHGDH perturbation interact strongly in axotomy regeneration, but the ATP reporter is intestinal and bulk ATP is whole-animal. That cannot identify sensory-axon energy supply or establish a human glucosamine mediator. [Yadav 2024](https://doi.org/10.7554/eLife.86478). The strongest relevant human structural prevention comparison and absolute early diabetic nerve glucosamine measurements were not recovered in searches through the review date. More general glucose or sorbitol measurements do not fill those gaps."
    },
    {
      "heading": "Ideal experiments that would resolve each claim",
      "markdown": "These define truth conditions, not proposed work. Assume noninjurious compartment-specific concentration measurement, selective free-glucosamine buffering/restoration, and five-year tracking of identified human peptidergic terminals and replacements. Sample independent people representatively, use fixed baseline groups and randomize the concentration intervention. Keep other initiating exposures and diabetes care comparable while allowing downstream ATP, glycosylation, glial support and axonal behavior to respond. Verify specificity by measuring neighboring metabolites and preserving basal physiological functions. NO requires excluding every allowed surviving group/compartment. The same setup is shared, but each row has its own readout and boundary.\n\n**R1 — persistent natural exposure.** With no predecessor assumed, repeatedly measure free D-glucosamine in mol/L of cell water separately in axons and connected Remak cells through the first year, with matched nondiabetic reference distributions and concurrent terminal tracking. Resolve free glucosamine from its phosphate and acetylated derivatives; use isotope tracing where needed to separate local production from uptake. YES is at least one baseline group above the matched 95th percentile for ≥90 cumulative days beginning before prospective loss. NO means every group fails the concentration/duration/timing requirement. Unknown molecular identity, bulk-only measurements, missing pre-loss observations or uncertainty across the 90-day boundary remains unresolved. Mouse pooled metabolomics supplies a relative natural exposure observation, not this human absolute trajectory. This claim says nothing about whether the concentration is injurious.\n\n**R2 — at least a small structural contribution.** Assuming R1, randomize five-year buffering of free glucosamine to its reference concentration versus intact exposure in each surviving compartment/group. Verify concentration trajectories and cell selectivity; reintroduce the original exposure in a separate arm and use a second independent manipulation to test intervention artifacts. Count actual disappearance and successful replacement with stable subtype identity. YES is B ≥0.05; NO is B <0.05 for all surviving groups, including harm or no benefit. Thirty percent reference loss versus 28% intervention loss passes with adequate precision. Intervals crossing 5%, incomplete buffering, a vanishing reference-loss denominator or staining changes without tracked structures remain unresolved. The mouse infusion and adult rat culture data show causal capability at selected exposure; neither supplies a selective contrast at the natural early human dose. ATP measurements can explain a positive result but are not substitutes for it.\n\n**R3 — material preservation.** Assuming R1–R2, compare the same structural effect with B=0.20. Thirty percent reference loss versus 24% or less passes; 30% versus 28% fails while retaining a small causal contribution. YES is B ≥0.20 in a surviving group, NO is B <0.20 in all. Finite uncertainty spanning 20% remains unresolved even if a treatment effect differs from zero. This step adds magnitude, not another discount for whether cells respond to glucosamine. Neither imposed mouse injury nor supplement associations measures this human prevention fraction.\n\n**R4 — reference-case coverage.** Assuming R1–R3, determine baseline membership and five-year incident DPN in a representative reference-care population. YES is at least one materially protected group's share of reference cases C ≥0.10; NO is C <0.10 for every surviving group. For example, 1,500 of 10,000 reference incident cases passes, while 500 fails, given precision excluding the boundary. Do not sum overlapping groups unless their union also satisfies the preceding effect criteria. Selected concentration-high samples, incomplete incidence follow-up or bounds crossing 10% remain unresolved. No existing exposure study supplies this case fraction; the large supplement cohort measures a different exposure and cannot substitute for it."
    },
    {
      "heading": "Interpretation of uncertainty and evidence updates",
      "markdown": "The skeptical scenario accepts laboratory toxicity but places the human free-glucosamine increase near physiological micromolar levels, with little structural consequence and uncommon material exposure. The favorable scenario assumes human cell-specific accumulation substantially exceeds plasma and reproduces a large portion of the experimental nerve effect in a common metabolic subgroup. The absolute-dose gap and inconsistent pump protocol permit both interpretations. Sampling uncertainty in five-mouse structural groups is much smaller than the uncertainty in transferring their exposure and endpoint to humans.\n\nThe numbers are literature-informed subjective judgments, not empirically calibrated forecasts. The stated range describes coordinated interpretations of the same proposition, without independent row distributions or a manufactured variance. Changing the free-glucosamine scope, subgroup rules, duration, or 20%/10% thresholds changes the question. A generic O-GlcNAc-only hypothesis would require its own defined exposure and intervention; it cannot inherit this probability.\n\nHypothetically, a selective human comparison at measured natural concentrations could raise R2 from 30% to 70%, taking the product to **12.7%** if other inputs stayed fixed. Precisely excluding a 5% structural benefit despite verified buffering could lower R2 to 5%, taking the product to **0.9%**. Actual evidence might also change exposure, magnitude and coverage judgments. A larger O-GlcNAc staining difference alone would not justify the upward update.\n\nThis exposure may converge on [bioenergetic insufficiency](07_bioenergetic_insufficiency.md), [ER stress](23_er_stress.md), [transport](26_axonal_transport.md) or [Schwann-cell support](17_schwann_metabolic_support.md). No single one is a necessary mediator of the total glucosamine effect priced here, so their probabilities are not multiplied into this one."
    }
  ],
  "source_claim_table_markdown": "| Claim | Additional proposition | Best probability | Uncertainty range (sensitivity) | Reason for estimate and breadth |\n|---|---|---:|---:|---|\n| R1 | At least one allowed group has the defined persistent pre-loss elevation of free glucosamine in a connected neuronal or Remak-cell compartment. | 55% | 25–80% | Diabetic mouse nerve metabolomics supports natural accumulation, but only as relative pooled signals. Human absolute concentration, localization and early duration are missing. |\n| R2 | In a group satisfying R1, normalizing free glucosamine averts at least 5% of five-year net human peptidergic terminal loss. | 30% | 7–70% | Imposed exposure produces large mouse fiber loss and adult rat neuron injury. Naturally elevated human exposure could nevertheless be far below the active dose; adaptation and support-cell transfer are unresolved. |\n| R3 | In a group satisfying R1–R2, preservation reaches B ≥0.20. | 55% | 25–80% | Given a real structural contribution, a large experimental effect supports possible materiality. It provides no human estimate of competing causes or the fraction attributable to this exposure. |\n| R4 | At least one group satisfying R1–R3 contains C ≥0.10 of reference incident cases. | 60% | 30–85% | Hyperglycemia is common; substantial local free-glucosamine accumulation in the materially affected group may not be. No representative case-coverage measurement exists. |"
}