Skip to main content
MyDeficiencies

Duloxetine

SNRI

Also sold as Cymbalta.

Analyze another medication Medication directory

Findings for this medication

Research about this medication, graded for how strongly each finding can be attributed to it. Nothing here describes your own levels.

Duloxetine has no classical nutrient-depletion machinery. It does not chelate minerals in the gut, does not induce hepatic enzymes that accelerate vitamin catabolism, and its elimination is overwhelmingly hepatic rather than renal, so it does not waste electrolytes through a damaged nephron the way a diuretic does. What duloxetine does carry, and carries specifically, is a labeled risk of hyponatremia mediated by the syndrome of inappropriate antidiuretic hormone secretion (SIADH): the kidney retains free water in response to an inappropriate antidiuretic-hormone signal, diluting serum sodium. This is a disturbance of water handling, not a sign that the patient is not eating enough salt, and it must never be framed as a dietary sodium deficiency, exactly as with the sodium entries in this series for lisinopril and hydrochlorothiazide. The at-risk population is concentrated: geriatric patients and those already volume-depleted or on diuretics.

SNRI antidepressant · ATC N06AX21 · also sold as Cymbalta · reviewed September 6, 2026, expanded September 10, 2026. This describes research about the medication. It does not say anything about your own levels or health.

How to read the evidence grades

These colours describe how strongly a finding can be attributed to the medication in the research. They are not a severity score and they say nothing about your own health.

Grades describe how strongly a finding can be attributed to Duloxetine in the research — not how common it is, how serious it would be, and not anything about you personally.

Red — Severe
severe: 2+ independent Tier A sources, this drug in the studied cohort, direct measurement, consistent direction, clinically meaningful, named mechanism, no contradicting Tier A evidence.
Orange — Moderate
moderate: 1+ Tier A source with this drug in the cohort showing a real but bounded effect (exposure window, defined subgroup, or functional rather than general-population deficit).
Yellow — Low
low/theoretical: mechanistic plausibility only, class-level extrapolation without this drug distinguishable, or a reverse-direction interaction worth flagging.
Not graded — disclosed evidence gap
Nothing has been measured here, so this finding cannot be graded. That does not establish either safety or harm, and it is not the same as a White result.Source definition: Not gradeable under the rules above because nothing has been measured. Published as a documented gap, never as reassurance that there is no effect.

What the research found

Sodium

Orange — Moderate

See study details

Duloxetine can trigger SIADH, causing the kidney to retain free water and dilute serum sodium; this is a water-handling disturbance concentrated in older and volume-depleted patients, not evidence that dietary sodium intake is inadequate.

Full detail, limits and 2 references

What the grade means here: moderate: 1+ Tier A source with this drug in the cohort showing a real but bounded effect (exposure window, defined subgroup, or functional rather than general-population deficit). disturbance, not depletion

Limits on what this shows

  • Key negative/null findings: Hyponatremia does not appear with a quantified incidence in duloxetine's own placebo-controlled clinical-trial adverse-reaction tables (Section 6.1 of the label); it is documented through case reports and postmarketing/real-world data rather than captured at meaningful rates in short pivotal RCTs, consistent with a rare event concentrated in specific risk groups rather than a general-population signal.
  • At-risk subgroups: Geriatric patients (named explicitly in the label); patients on diuretics or otherwise volume-depleted (named explicitly in the label); at least nine published case reports place duloxetine-associated hyponatremia disproportionately in older adults, including one case reported in a child with acute leukemia on multiple concurrent medications, illustrating that severe comorbidity and polypharmacy, not age alone, can also drive individual cases.

Regulatory status

FDA-labeled warning (Warnings and Precautions 5.13), quoted in full above, naming duloxetine directly and citing cases of serum sodium below 110 mmol/L that reversed on discontinuation.

Population evidence

Mo H, Channa Y, Ferrara TM, et al., "Hyponatremia Associated with the Use of Common Antidepressants in the All of Us Research Program," Clin Pharmacol Ther, 2025 (PMID 39540435, DOI 10.1002/cpt.3484): a real-world cohort of 17,439 individuals on common antidepressants. Duloxetine and escitalopram carried the highest hyponatremia risk relative to a sertraline reference; cumulative incidence across the antidepressant cohort was reported at 0.87% within 30 days and 10.5% over 3 years, with findings stated to hold after adjustment for comorbidities and concurrent medications. Duloxetine is a named, distinguishable arm in this cohort, satisfying the cohort test for Tier A. Exact hazard ratio and confidence interval for duloxetine specifically could not be confirmed from the accessible abstract text and are not quoted here.

Mechanism

SIADH: non-osmotic, likely serotonergic-mediated release of antidiuretic hormone causes renal free-water retention, diluting plasma sodium. This is mechanism class F (dilutional/hormonal disturbance), not renal sodium wasting and not reduced dietary intake or absorption of sodium.

Counter-evidence held on record

Confounding by indication and by age is active here: duloxetine is prescribed across major depressive disorder, generalized anxiety disorder, diabetic peripheral neuropathic pain, and fibromyalgia, populations with very different age distributions and comorbidity burdens. Diabetic neuropathic pain patients skew older and are more likely to be on diuretics or have renal impairment independent of duloxetine, which can inflate apparent hyponatremia association in that indication. The All of Us cohort study used sertraline, not placebo, as the comparator, so it establishes relative risk among antidepressants rather than risk against no treatment at all. This is a dilutional, hormonally-mediated water-retention disturbance (SIADH), not a dietary sodium deficiency. It must never be presented as evidence that the patient should increase salt intake or take a sodium supplement; the appropriate response to suspected duloxetine-associated hyponatremia is clinical evaluation and, if warranted, discontinuation under medical supervision, not dietary or supplemental sodium loading.

CYP1A2 / CYP2D6 interactions

Yellow — Low

Direction: reverse direction (as labelled in the source)

Duloxetine's own exposure rises sharply when combined with CYP1A2 inhibitors, and duloxetine itself raises the exposure of CYP2D6-substrate drugs; both directions are pharmacokinetic interactions between drugs, not a nutrient interaction, and are listed here only so the distinction is explicit and on record.

Full detail and limits

What the grade means here: low/theoretical: mechanistic plausibility only, class-level extrapolation without this drug distinguishable, or a reverse-direction interaction worth flagging. reverse-direction / drug-affects-drug

Limits on what this shows

  • Key negative/null findings: No CYP450 induction by duloxetine itself has been identified in the label; duloxetine is a substrate and a CYP2D6 inhibitor, not an enzyme inducer, so mechanism class A (accelerated catabolism via induction) does not apply.
  • At-risk subgroups: Patients co-prescribed potent CYP1A2 inhibitors (e.g., fluvoxamine, ciprofloxacin per the label's interaction class) for duloxetine's own exposure; patients co-prescribed CYP2D6 substrates with a narrow therapeutic index (e.g., certain tricyclic antidepressants, some antiarrhythmics) for the reverse direction.

Regulatory status

Labeled interaction (Drug Interactions 7.1, 7.2; Clinical Pharmacology 12.3). "Co-administration of CYMBALTA with potent CYP1A2 inhibitors should be avoided." When duloxetine 60 mg was co-administered with fluvoxamine 100 mg (a potent CYP1A2 inhibitor), duloxetine AUC increased approximately 6-fold and Cmax approximately 2.5-fold. In the other direction, duloxetine 60 mg twice daily increased the AUC of desipramine (a CYP2D6 substrate) approximately 3-fold. Duloxetine's own bioavailability is also reduced by about one-third in smokers, consistent with CYP1A2 induction by smoking.

Population evidence

Label-derived pharmacokinetic interaction studies (healthy-subject, single- or multiple-dose designs), not population outcome studies. No nutrient was measured in these studies; they characterize drug exposure only.

Mechanism

CYP1A2 and CYP2D6 are the enzymes responsible for duloxetine's own metabolism (mechanism relevant to duloxetine as a substrate); duloxetine is separately a moderate inhibitor of CYP2D6, affecting other drugs that rely on that enzyme (mechanism class B, but running from duloxetine outward to other drugs, not from another agent onto a nutrient).

Counter-evidence held on record

None of these interactions involve a nutrient as either the affected or affecting agent; they are included solely to document that the direction was checked (nutrient-affects-drug vs. drug-affects-drug) and correctly excluded from the depletion framework, per the requirement to check interaction direction before any nutrient claim is filed. Do not present this as a nutrient interaction. It is a drug-drug pharmacokinetic interaction and should be flagged to prescribers/pharmacists managing co-medication, not to a consumer nutrient tracker.

Bone mineral density / fracture risk

Yellow — Low

See study details

SSRIs and SNRIs as a class have been studied for a possible bone mineral density and fracture signal, but a direct search naming duloxetine alongside bone mineral density or fracture returned no results; the mechanism is plausible but unverified for this specific molecule in humans.

Full detail, limits and 3 references

What the grade means here: low/theoretical: mechanistic plausibility only, class-level extrapolation without this drug distinguishable, or a reverse-direction interaction worth flagging. class extrapolation

Limits on what this shows

  • Key negative/null findings: The zero-result direct search for duloxetine is itself the key negative finding: the class-level hypothesis has not been tested against duloxetine as a named arm in the literature surveyed for this review.
  • At-risk subgroups: If the class effect applies, older adults and postmenopausal women would be the plausible highest-risk subgroup, consistent with general osteoporosis risk factors rather than anything duloxetine-specific.

Regulatory status

No bone density or fracture language appears anywhere in the CYMBALTA label. The label's "Falls" precaution states "Falls with serious consequences including fractures and hospitalizations have been reported," but this is a fall-risk/sedation outcome statement, not a bone-mineral-density or nutrient claim, and is not filed as one here.

Population evidence

A PubMed search combining "duloxetine," "bone mineral density," and "fracture" returned zero results on the date of this review, confirming duloxetine has not been isolated as a distinguishable arm in this literature. Class-level studies exist and study serotonergic antidepressants broadly: Moura et al., Osteoporos Int 2014 (PMID 24566587), and Lanteigne et al., CNS Drugs 2015 (PMID 25708711), both analyze SSRI/SNRI fracture risk without naming individual SNRI agents in the titles/abstracts available to this review; full text of both was not independently opened and no effect sizes from them are cited here. One class-member animal study was identified with a named SNRI: Erdem et al., Injury 2023 (DOI 10.1016/j.injury.2023.03.034), a rat calvarial defect model comparing fluoxetine (SSRI) and venlafaxine (SNRI) on bone healing; duloxetine itself was not the SNRI studied, and effect sizes were not accessible from the retrievable metadata.

Mechanism

Serotonin transporter and receptors are expressed on osteoblasts and osteoclasts, giving SERT-inhibiting drugs (SSRIs, and by extension SNRIs like duloxetine) a biologically plausible route to affect bone remodeling. This is established pharmacology at the class level; it has not been confirmed as operative for duloxetine specifically in a human bone-outcome study.

Counter-evidence held on record

Confounding by indication applies directly: depression and chronic pain conditions are independently associated with reduced physical activity, poorer diet, and higher fall risk, all of which affect bone health and fracture rates without any drug-specific bone mechanism. Fracture is a clinical outcome, not a nutrient-status measurement; no study reviewed here measured a bone-relevant nutrient (calcium, vitamin D, or a bone turnover marker) in duloxetine-treated patients, so this entry is deliberately kept at the "outcome under investigation" level rather than recast as a calcium or vitamin D deficiency claim. Never state or imply that duloxetine causes calcium or vitamin D deficiency, or that bone density loss has been demonstrated for this drug. Fracture and BMD data at the class level, where they exist, do not identify duloxetine and must not be attributed to it by name.

The five body-system areas

Each finding is counted once, in a single area. Where a study is mentioned in a second area it is context only.

Mitochondrial Burden

Each finding below is counted once, in this area only.

Mitochondrial function

Not graded — disclosed evidence gap

See study details

No duloxetine-specific Tier A, B, or C evidence of an effect on mitochondrial function or biogenesis was confirmed in this review, and no circulating consumer claim linking duloxetine to mitochondrial harm was identified to rebut; this is a genuine gap, not a clean bill of health.

Full detail, limits and 1 references

What the grade means here: Nothing has been measured here, so this finding cannot be graded. That does not establish either safety or harm, and it is not the same as a White result. ungraded

Limits on what this shows

  • Key negative/null findings: A separate, unrelated case report, Cozart B, et al., Clin Neuropharmacol 2018, describes duloxetine used to treat depression in a patient with pre-existing MELAS mitochondrial encephalopathy. This documents duloxetine's use in a patient whose mitochondrial disease already existed, not duloxetine causing mitochondrial harm, and is explicitly excluded from being read as a harm signal.
  • At-risk subgroups: Cannot be characterized from the evidence located.
  • Dose/duration: Not applicable. No dose is recommended anywhere in this dossier.

Regulatory status

No mitochondrial, cellular-energy, or CoQ10-related language appears anywhere in the CYMBALTA label.

Population evidence

A duloxetine-plus-mitochondria literature search located no confirmed duloxetine-specific mitochondrial-function or biogenesis study. Two potentially relevant leads were identified but could not be independently verified after repeated attempts: Wernicke J, et al., Curr Drug Saf 2008, "Hepatic effects of duloxetine-I: non-clinical and clinical trial data" (PMID 18690897), whose search-result summary referenced toxicology "mitochondrial effects," and Nabekura T, Ishikawa S, Tanase M, Okumura T, Kawasaki T, Curr Res Toxicol 2022, "Antidepressants induce toxicity in human placental BeWo cells," whose search-result summary referenced an effect on mitochondrial membrane potential among tested antidepressants. Whether duloxetine specifically, as opposed to other drugs in either study, showed the referenced effect could not be confirmed by directly opening either source; both were attempted multiple times and both attempts returned no retrievable content. Neither is used as counted evidence for this entry.

Mechanism

Mechanism class E (mitochondrial interference or conjugation-driven loss) was screened for; no duloxetine-specific evidence confirming or excluding this mechanism was located.

Counter-evidence held on record

Confounding by indication: depression and chronic pain are independently associated with altered mitochondrial markers in the psychiatric and pain literature; any future duloxetine-specific mitochondrial finding would need to rule this out before being attributed to the drug. This gap must never be presented as reassurance that no mitochondrial effect exists, or converted into a soft claim in either direction. It is printed here as an open question, not an answer.

Oxidative Stress

Each finding below is counted once, in this area only.

Oxidative stress markers

Yellow — Low

Direction: reverse direction (as labelled in the source)

Every duloxetine-specific animal and cell study located that measured oxidative stress showed the drug reducing oxidative stress markers in injury and disease models; no duloxetine-specific evidence was found showing the drug increases oxidative stress or depletes antioxidant capacity in humans.

Full detail and limits

What the grade means here: low/theoretical: mechanistic plausibility only, class-level extrapolation without this drug distinguishable, or a reverse-direction interaction worth flagging. mechanistic, reverse direction

Limits on what this shows

  • Key negative/null findings: No duloxetine-specific study was located reporting duloxetine increasing any oxidative stress marker (malondialdehyde, protein carbonyls, reactive oxygen species) in any model. The absence of a harm-direction signal across five independent duloxetine-specific studies is itself informative and is recorded as a negative finding for the depletion hypothesis.
  • At-risk subgroups: Not applicable in the depletion sense. The literature located concerns duloxetine tested as a candidate protective agent in neuropathic-injury, ischemia, and stress models, populations that overlap with duloxetine's own approved indications rather than a distinct at-risk group.
  • Dose/duration: Not applicable. No dose-response oxidative-stress data in humans was located, and no dose is recommended anywhere in this dossier.

Regulatory status

No oxidative stress or antioxidant-status language appears anywhere in the CYMBALTA label.

Population evidence

No Tier A human study measuring an oxidative stress marker in duloxetine-treated patients was located. Five duloxetine-specific Tier B (animal or human-cell) studies were located, all reporting the same direction: Engel DF, et al., Neurochem Res 2018, human neuroblastoma cells, duloxetine reduced oxidative-stress-induced cell death via Akt/Nrf-2/HO-1 signaling; Demirdaş A, Nazıroğlu M, Övey İS, Mol Neurobiol 2017, rat hippocampus and dorsal root ganglion, duloxetine reduced oxidative stress, apoptosis, and calcium entry; Akpinar A, Uğuz AC, Nazıroğlu M, J Membr Biol 2014, PC12 cells, duloxetine alone and combined with agomelatine reduced oxidative stress; Lee TK, et al., J Neurol Sci 2016, rat cerebral ischemia-reperfusion model, duloxetine pretreatment reduced oxidative stress markers; Meejuru GF, et al., J Pharm Pharmacol 2021, mouse chronic immobilization-stress model, duloxetine showed protective, antioxidant-consistent effects. Specific PMID/DOI identifiers for these five could not be independently confirmed in this review and are not stated as verified.

Mechanism

The proposed mechanism across these studies is Nrf2/HO-1 and Akt-pathway antioxidant signaling activation, an injury-model pharmacological property of duloxetine, not a match to any of the nine depletion mechanism classes screened in the pharmacokinetic and mechanism screen above. This is mechanism running the opposite direction: a plausible antioxidant action, not evidence of oxidative-stress induction.

Counter-evidence held on record

Confounding by indication is a major risk here specifically: major depressive disorder, generalized anxiety disorder, and chronic pain conditions (diabetic neuropathy, fibromyalgia) are independently associated with elevated oxidative stress markers and altered mitochondrial markers in the psychiatric and pain literature, independent of any drug treatment. None of the five Tier B studies located measured oxidative stress in the context of duloxetine's approved human indications; they are heterologous cell-injury and animal-stress models testing duloxetine as a candidate protective agent, not characterizations of duloxetine's own physiological burden in treated patients. Do not present this entry as evidence that duloxetine causes oxidative stress or depletes antioxidant capacity. Equally, do not present it as evidence that duloxetine protects patients against oxidative stress in humans; the underlying evidence is preclinical injury-model data, not a human antioxidant-status finding, and depression and chronic pain themselves carry an independent oxidative-stress signal that has nothing to do with duloxetine.

Copper and Iron Balance

Copper and iron are assessed separately. Nothing here is counted twice.

Copper status

Not graded — disclosed evidence gap

See study details

No duloxetine-specific evidence of an effect on copper status was located at any tier, and no circulating consumer claim linking duloxetine to copper deficiency was identified to rebut.

Full detail and limits

What the grade means here: Nothing has been measured here, so this finding cannot be graded. That does not establish either safety or harm, and it is not the same as a White result. ungraded

Limits on what this shows

  • Key negative/null findings: None located; this is an absence of study, not a study with a null result.
  • At-risk subgroups: Cannot be characterized from the evidence located.
  • Dose/duration: Not applicable. No dose is recommended anywhere in this dossier.

Regulatory status

No copper, ceruloplasmin, or trace-mineral language appears anywhere in the CYMBALTA label.

Population evidence

A dedicated literature search combining "duloxetine" and "copper" returned no biologically relevant results; every result concerned copper-catalyzed synthetic chemistry used to manufacture duloxetine and related molecules in the laboratory, not copper physiology in patients taking the drug.

Mechanism

Mechanism classes A through I were each screened against copper specifically; none is documented for duloxetine. The one theoretically available route, hepatic synthesis of the copper-carrier protein ceruloplasmin being secondarily affected by duloxetine's labeled hepatotoxicity risk (ALT >3x ULN in 1.25% of duloxetine-treated patients vs. 0.45% on placebo, CYMBALTA label, Hepatotoxicity 5.2 and Adverse Reactions 6.1), is a generic fact about liver physiology, not a duloxetine-specific mechanistic study measuring copper or ceruloplasmin in hepatotoxicity cases. It does not meet the Tier B bar and is not counted as evidence; it is documented here only to show the possibility was actively considered and found unsupported by any actual measurement.

Counter-evidence held on record

Not evaluable in the absence of any signal to confound. Do not convert this gap into a claim that duloxetine is confirmed safe for copper status. Absence of evidence is not evidence of absence.

Iron status

Not graded — disclosed evidence gap

See study details

No duloxetine-specific evidence of an effect on iron status was confirmed at any tier, and no circulating consumer claim linking duloxetine to iron deficiency was identified to rebut.

Full detail and limits

What the grade means here: Nothing has been measured here, so this finding cannot be graded. That does not establish either safety or harm, and it is not the same as a White result. ungraded

Limits on what this shows

  • Key negative/null findings: The label's own adverse-reaction and hematology data contain no anemia or iron-parameter signal for duloxetine.
  • At-risk subgroups: Cannot be characterized from the evidence located.
  • Dose/duration: Not applicable. No dose is recommended anywhere in this dossier.

Regulatory status

No iron, ferritin, hemoglobin, or anemia language appears anywhere in the CYMBALTA label; the label's adverse-reaction tables do not list anemia or any iron-related term among reported events.

Population evidence

A dedicated literature search combining "duloxetine" and "iron" located one incidental case report, Yau AH, Xiong W, Ko HH, BMJ Case Rep 2015, describing a patient on duloxetine, among other medications, diagnosed with collagenous enterocolitis presenting as watery diarrhea and iron-deficiency anemia. The case report's full text could not be independently opened in this review, so whether duloxetine was implicated as a contributing cause, or was simply a concurrent medication in a patient whose anemia is attributable to a distinct GI disease process, could not be confirmed. It is not counted as evidence of a duloxetine-iron relationship.

Mechanism

Mechanism classes A through I were each screened against iron specifically; none is documented for duloxetine. Serotonin-iron literature exists in the opposite direction (iron as a cofactor for tryptophan hydroxylase in serotonin synthesis, a nutrient-affects-drug-target relationship), which is excluded from this entry per the interaction-direction check required before any nutrient claim is filed.

Counter-evidence held on record

The one case report located involves an independent GI disease (collagenous enterocolitis) with its own well-established iron-loss mechanism (chronic diarrhea, malabsorption); this is a plausible full explanation for that patient's anemia without invoking duloxetine at all, illustrating exactly the kind of confounding this review is required to consider before filing a nutrient claim. Do not convert this gap into a claim that duloxetine is confirmed safe for iron status. Do not cite the BMJ case report as evidence of duloxetine-induced anemia; it does not support that conclusion and was not independently verified in this review.

Gut Microbiome Risk

Each finding below is counted once, in this area only.

Gut microbiome composition

Yellow — Low

See study details

Duloxetine carries a labeled, drug-specific, high-frequency GI symptom burden (nausea, diarrhea, constipation), but that is a symptom profile, not a measured change in gut microbial composition; duloxetine-specific microbiome composition data was not located, and the antidepressant-class literature that does study gut bacteria could not be confirmed to include duloxetine as a named, tested drug.

Full detail, limits and 1 references

What the grade means here: low/theoretical: mechanistic plausibility only, class-level extrapolation without this drug distinguishable, or a reverse-direction interaction worth flagging. symptom vs. composition, class extrapolation

Limits on what this shows

  • Key negative/null findings: The label's own adverse-reaction data characterizes duloxetine's GI effects entirely in symptom terms; it does not report or imply any microbiome or dysbiosis finding. A drug-specific, Tier A GI-symptom dataset exists and does not extend to a composition claim, which is recorded here as the key negative finding.
  • At-risk subgroups: Patients in early treatment, since nausea and GI tolerability effects are most pronounced at initiation and titration for SNRIs generally; patients with pre-existing GI disease, given duloxetine's use in fibromyalgia and diabetic neuropathy populations that may carry independent GI comorbidity.
  • Dose/duration: Not applicable. No dose-response composition data exists to cite, and no dose is recommended anywhere in this dossier.

Regulatory status

CYMBALTA label, Adverse Reactions 6.1: nausea 23% (duloxetine) vs. 8% (placebo); constipation 9% vs. 4%; diarrhea 9% vs. 6% (adult placebo-controlled trials). No microbiome, dysbiosis, or gut-bacteria language appears anywhere in the label.

Population evidence

No Tier A human study measuring gut microbial composition (sequencing, metagenomics, or culture-based composition change) in duloxetine-treated patients was located. A duloxetine-titled narrative review, Rejili M, et al., Naunyn Schmiedebergs Arch Pharmacol 2026, "Duloxetine as a gut-brain axis modulator in irritable bowel syndrome: neuroimmune, microbiota, and pharmacogenomic perspectives," was identified by title; its full text/abstract could not be independently opened in this review, so whether it presents original duloxetine-specific composition data or only a class-level narrative could not be confirmed, and it is not used to support any specific claim here. Class-level in vitro and animal literature on antidepressants and gut bacteria was also identified: Rukavishnikov G, et al., Front Behav Neurosci 2023 (in vitro antimicrobial activity of antidepressants on gut microbiota); Ding P, Lu J, Lei T, et al., Gut Microbes 2025 (antidepressant drugs and gut-bacterial plasmid transfer, mouse and human microbiota); Lukić I, et al., Transl Psychiatry 2019 (antidepressants and mouse gut microbiota, linked to depressive-like behavior). None of the three could be independently opened in this review to confirm whether duloxetine specifically, as opposed to other antidepressants, was among the tested drugs; they are treated as Tier C class-level evidence at best.

Mechanism

Mechanism class G (microbiome disruption affecting endogenous synthesis) is plausible in principle for any drug with a substantial GI-transit effect, and duloxetine's own labeled GI adverse-effect profile establishes altered GI symptomatology; however, altered transit or symptom frequency is not itself evidence of altered bacterial composition, and no duloxetine-specific mechanistic link from symptom to composition change was located.

Counter-evidence held on record

Confounding by indication: irritable bowel syndrome and functional GI symptoms are themselves associated with depression, anxiety, and fibromyalgia independent of any drug treatment, so a patient's GI complaints while on duloxetine may reflect the underlying indication rather than a drug effect on gut bacteria. Nausea and GI symptoms are also typically most pronounced at treatment initiation and diminish with continued use, consistent with a transient GI-tolerability effect rather than a sustained microbiome disruption. Do not present duloxetine's documented nausea, diarrhea, or constipation rates as evidence of gut microbiome disruption. A symptom is not a composition measurement. Do not present the Rejili 2026 title, or the class-level antidepressant-microbiome literature, as duloxetine-specific composition evidence; none of it was independently confirmed to include duloxetine as a tested, named drug in this review.

Oxygen and ATP Output

Oxygen delivery and ATP production are assessed separately. Nothing here is counted twice.

Oxygen delivery and utilization

Not graded — disclosed evidence gap

See study details

No duloxetine-specific evidence of an effect on oxygen delivery or tissue oxygen utilization was located, and no circulating consumer claim on this specific point was identified to rebut; literature access limits during this review prevented a full search, and that limitation is disclosed rather than treated as a completed null result.

Full detail and limits

What the grade means here: Nothing has been measured here, so this finding cannot be graded. That does not establish either safety or harm, and it is not the same as a White result. ungraded

Limits on what this shows

  • Key negative/null findings: None confirmed; the underlying search could not be completed.
  • At-risk subgroups: Cannot be characterized from the evidence located.
  • Dose/duration: Not applicable. No dose is recommended anywhere in this dossier.

Regulatory status

No hypoxia, oxygen-carrying capacity, hemoglobin, or tissue-oxygenation language appears anywhere in the CYMBALTA label; no anemia signal appears in the label's hematology or adverse-reaction data (see Iron status entry above), which would be the most direct route to an oxygen-delivery effect.

Population evidence

A dedicated literature search for duloxetine combined with hypoxia, oxygen delivery, or oxygen consumption could not be completed in this review; the search request was blocked by a literature-database access and rate limit before results could be retrieved, and repeated retries were unsuccessful. This is disclosed as an incomplete search, not a confirmed absence of evidence.

Mechanism

No mechanism class from the nine screened points specifically toward oxygen delivery or utilization for duloxetine. Duloxetine's known serotonergic and noradrenergic vascular effects are the drug's own labeled mechanism of action and adverse-effect profile (for example, blood pressure changes) and are not filed here as a nutrient-oxygen finding, per the requirement not to file a drug's own known mechanism as a nutrient deficiency.

Counter-evidence held on record

Not evaluable in the absence of any signal to confound. This entry must not be read as reassurance that duloxetine has no oxygen-delivery effect. The search for this category was incomplete because of access constraints during this review and should be revisited with a completed literature search before any grade is assigned.

ATP production

Not graded — disclosed evidence gap

See study details

No duloxetine-specific evidence of an effect on cellular ATP production was confirmed at any tier, and no circulating consumer claim on this specific point was identified to rebut.

Full detail and limits

What the grade means here: Nothing has been measured here, so this finding cannot be graded. That does not establish either safety or harm, and it is not the same as a White result. ungraded

Limits on what this shows

  • Key negative/null findings: None confirmed; the one identified lead could not be verified.
  • At-risk subgroups: Cannot be characterized from the evidence located.
  • Dose/duration: Not applicable. No dose is recommended anywhere in this dossier.

Regulatory status

No ATP, cellular energy, or bioenergetic language appears anywhere in the CYMBALTA label.

Population evidence

A literature search combining duloxetine and ATP production located one potentially relevant lead, Nabekura T, Ishikawa S, Tanase M, Okumura T, Kawasaki T, Curr Res Toxicol 2022, "Antidepressants induce toxicity in human placental BeWo cells," whose search-result summary referenced an effect on mitochondrial membrane potential among the antidepressants tested. This is the same source flagged, and not counted, under Mitochondrial function above; it is not double-counted as independent evidence here. Repeated attempts to open the source directly and confirm whether duloxetine specifically was the drug associated with that finding, and what if any ATP measurement was reported, returned no retrievable content, so it is not used as evidence for this entry.

Mechanism

Mechanism class E (mitochondrial interference) is the plausible route to an ATP-production effect; no duloxetine-specific confirmation or exclusion was located.

Counter-evidence held on record

Not evaluable in the absence of any verified signal to confound. This gap must not be converted into reassurance. It is printed as an open question, pending independent verification of the one lead identified (Nabekura et al., 2022) in a future review.

Claims checked and not supported

These circulate online for this medication. They are listed here only so they can be answered, never as findings, and they are not graded.

A claim with no supporting evidence is not necessarily disproven. Where evidence actually argues against a claim, that is stated below.

  • Magnesium

    Where it comes from: Generic "antidepressants deplete magnesium" content applied to the SNRI/SSRI class broadly, sometimes extended to duloxetine by name on consumer depletion lists.

    Why it fails: No mechanism (no renal magnesium wasting, no GI chelation, no altered magnesium transporter activity) is documented for duloxetine. A 2026 trial identified in a duloxetine-plus-magnesium search evaluated magnesium as an adjunct treatment given to patients already on antidepressants to test whether it improves depression outcomes, the reverse of a depletion claim; full text was not independently opened and no numbers from it are cited here. No Tier A or Tier B duloxetine-specific depletion evidence exists.

  • Vitamin D

    Where it comes from: Fibromyalgia-focused nutrition content that pairs duloxetine (a fibromyalgia drug) with vitamin D deficiency, and general antidepressant-and-vitamin-D commentary.

    Why it fails: Vitamin D deficiency prevalence in fibromyalgia patients reflects confounding by indication (chronic pain, reduced sun exposure and activity), not a duloxetine effect. The one duloxetine-specific vitamin D study located, Khan et al., J Biochem Mol Toxicol 2025 (DOI 10.1002/jbt.70473), is a rat study testing vitamin D as an add-on to enhance duloxetine's antidepressant effect, the opposite direction of a depletion claim, and provides no evidence duloxetine lowers vitamin D status.

  • Vitamin B12

    Where it comes from: Occasional inclusion on broad "psychiatric medication depletion" lists, likely carried over by association with other drug classes (metformin, acid suppressants) that do have real B12 mechanisms.

    Why it fails: No absorption, transport, or catabolic mechanism connects duloxetine to B12 status; duloxetine does not suppress gastric acid, does not compete with intrinsic-factor-mediated absorption, and no duloxetine-specific B12 study was located in the literature surveyed.

What the research says is worth watching

General information from the medication label and the studies above. It is not personal medical advice, and any change belongs with your prescriber.

  • Sodium

    Baseline and early-treatment sodium check in higher-risk patients

    Serum sodium is worth checking at baseline and again in the first weeks of duloxetine therapy in patients who are elderly, on diuretics, or otherwise volume-depleted, since that is where labeled and case-report signal concentrates. Watch for confusion, headache, or lethargy, the clinical presentation of hyponatremia, rather than assuming any low-sodium reading reflects diet.

  • Drug interactions

    Flag CYP1A2 and CYP2D6 co-medications to the prescriber/pharmacist

    Potent CYP1A2 inhibitors (e.g., fluvoxamine, ciprofloxacin per the label) raise duloxetine's own exposure substantially and should be avoided or managed by a prescriber. Conversely, duloxetine raises exposure of co-administered CYP2D6-substrate drugs; this is a medication-management issue, not a nutrient one.

  • Renal / hepatic status

    Confirm renal and hepatic function before and during use

    The label recommends avoiding duloxetine in severe renal impairment (GFR <30 mL/min) and in chronic liver disease or cirrhosis, given large exposure increases documented in both settings. This affects drug clearance, not nutrient status, but belongs in the same monitoring conversation.

  • Fall risk

    Track falls as a labeled outcome, separate from any bone-nutrient claim

    The label notes falls with fractures and hospitalizations have been reported. This is worth monitoring as a sedation/fall-risk outcome in older adults, but it is not evidence of calcium or vitamin D depletion and should not be recorded as one.

All references used above
  1. CYMBALTA (duloxetine hydrochloride) delayed-release capsules, FDA-approved prescribing information, Eli Lilly and Company — DailyMed, Warnings and Precautions 5.13 (Hyponatremia), Drug Interactions 7.1-7.2, Clinical Pharmacology 12.3, Use in Specific Populations 8.6-8.7, Adverse Reactions 6.
  2. Mo H, Channa Y, Ferrara TM, et al. Hyponatremia Associated with the Use of Common Antidepressants in the All of Us Research Program. Clin Pharmacol Ther. 2025. — Real-world cohort, n=17,439; duloxetine and escitalopram highest relative risk vs. sertraline; cumulative incidence 0.87% at 30 days, 10.5% at 3 years. PMID 39540435, DOI 10.1002/cpt.3484.
  3. Kulkarni M. Duloxetine induced hyponatremia. Indian J Nephrol. 2015. — Case report, duloxetine named specifically. PMID 26199483.
  4. Wang D, Lai J, Lu S, Huang M, Hu S, Xu Y. Rapid-onset hyponatremia and delirium following duloxetine treatment for postherpetic neuralgia. Medicine (Baltimore). 2018. — Case report. PMID 30431592.
  5. Yoshida K, Aburakawa Y, Suzuki Y, Kuroda K, Kimura T. Acute Hyponatremia Resulting from Duloxetine-induced Syndrome of Inappropriate Antidiuretic Hormone Secretion. Intern Med. 2019. — Case report. PMID 30799365.
  6. Sun CF, Chen YL, Li YH, Kumaraswamy M, Lo YC, Chen YT. Duloxetine-Induced Hyponatremia in an Elderly Male Patient with Treatment-Refractory Major Depressive Disorder. Case Rep Psychiatry. 2019. — Case report, elderly patient. PMID 31214374.
  7. Oliver WD, D'Angelo R, Gonzales J, Wilson T, Millstein LS. Acute Severe Hyponatremia Induced by a Duloxetine Overdose in an Elderly Woman. Cureus. 2020. — Case report, elderly patient, overdose context. PMID 33052279.
  8. Nagamine T. Duloxetine-induced Hyponatremia Can Occur in Not Only Syndrome of Inappropriate Secretion of Antidiuretic Hormone but Also Cerebral Salt Wasting Syndrome. Intern Med. 2022. — Case report distinguishing SIADH from cerebral salt wasting. PMID 34897153.
  9. Baig MU, Madden K, Moody K. Hyponatremia with Antidepressant: A Rare Side Effect from Duloxetine in a Child with Acute Leukemia. J Palliat Med. 2022. — Case report, pediatric, multiple concurrent medications and severe underlying illness. PMID 35537479.
  10. Moura C, Bernatsky S, Abrahamowicz M, et al. Antidepressant use and 10-year incident fracture risk: the population-based Canadian Multicentre Osteoporosis Study (CaMoS). Osteoporos Int. 2014. — Class-level SSRI/SNRI cohort analysis. PMID 24566587. Full text not independently opened; individual SNRI agents, including whether duloxetine was represented in the cohort, could not be confirmed. No effect sizes cited from this source.Full text not independently opened; individual SNRI agents, including whether duloxetine was represented in the cohort, could not be confirmed. No effect sizes cited from this source.
  11. Lanteigne A, Sheehy O, Bérard A. Serotonin-norepinephrine reuptake inhibitor and selective serotonin reuptake inhibitor use and risk of fractures: a new user cohort study. CNS Drugs. 2015. — Class-level SNRI/SSRI cohort analysis. PMID 25708711. Full text not independently opened; whether duloxetine was distinguishable from venlafaxine within the SNRI arm could not be confirmed. No effect sizes cited from this source.Full text not independently opened; whether duloxetine was distinguishable from venlafaxine within the SNRI arm could not be confirmed. No effect sizes cited from this source.
  12. Erdem MN, et al. Comparison of the effects of fluoxetine and venlafaxine on bone healing in a rat calvarial defect model. Injury. 2023. — Animal (rat) study; venlafaxine is the SNRI studied, not duloxetine. DOI 10.1016/j.injury.2023.03.034. Effect sizes not accessible from retrievable metadata; cited for existence of class-member animal data only.Effect sizes not accessible from retrievable metadata; cited for existence of class-member animal data only.
  13. PubMed search: "duloxetine" AND "bone mineral density" AND "fracture", performed September 6, 2026. — Zero results returned, confirming no duloxetine-specific bone mineral density/fracture study was indexed at time of review.
  14. Khan N, Ullah S, Ilyas M, Khan MW. Investigating the Role of Vitamin D in Enhancing Duloxetine and Fluoxetine Efficacy: Mechanistic Insights Into NRF2 Activation, Oxidative Stress, and Neurochemical Modulation. J Biochem Mol Toxicol. 2025. — Animal (Wistar rat) study; vitamin D given as adjunct to improve antidepressant efficacy, not a depletion study. PMID 40901743, DOI 10.1002/jbt.70473.
  15. Siddique A, Naeem U, Khokhar MM, Irshad M, Farooq T, Kanwal S. Clinical evaluation of adjunctive magnesium therapy in patients receiving antidepressants for depression. Naunyn Schmiedebergs Arch Pharmacol. 2026. — Magnesium given as adjunct treatment to patients already on antidepressants; not a depletion study. PMID 42283837. Full abstract could not be independently retrieved; whether duloxetine specifically was among the antidepressants used could not be confirmed. No numbers cited from this source.Full abstract could not be independently retrieved; whether duloxetine specifically was among the antidepressants used could not be confirmed. No numbers cited from this source.
  16. WHO Collaborating Centre for Drug Statistics Methodology, ATC/DDD Index — N06AX21 duloxetine; cross-referenced against DrugBank DB00476. — Identifier/classification reference only, not a clinical claim.
  17. CYMBALTA (duloxetine hydrochloride) delayed-release capsules, FDA-approved prescribing information, Eli Lilly and Company. — Same document as source 1, cited again for Hepatotoxicity 5.2 and Adverse Reactions 6.1: hepatic failure reports; ALT >3x ULN in 1.25% (144/11,496) of duloxetine-treated patients vs. 0.45% (39/8716) on placebo; nausea 23% vs. 8%, constipation 9% vs. 4%, diarrhea 9% vs. 6%. No anemia, iron, copper, or hematologic language appears in the label.
  18. Engel DF, et al. Duloxetine Protects Human Neuroblastoma Cells from Oxidative Stress-Induced Cell Death. Neurochem Res. 2018. — Human cell (neuroblastoma) study, duloxetine-specific; reported protective effect via Akt/Nrf-2/HO-1 signaling. PMID/DOI could not be independently confirmed in this review; title and journal confirmed via PubMed search-listing only, full abstract not opened, no statistic quoted.PMID/DOI could not be independently confirmed in this review; title and journal confirmed via PubMed search-listing only, full abstract not opened, no statistic quoted.
  19. Demirdaş A, Nazıroğlu M, Övey İS. Duloxetine Reduces Oxidative Stress, Apoptosis, and Ca2+ Entry. Mol Neurobiol. 2017. — Rat hippocampus and dorsal root ganglion study, duloxetine-specific. PMID/DOI could not be independently confirmed in this review; title and journal confirmed via PubMed search-listing only, full abstract not opened, no statistic quoted.PMID/DOI could not be independently confirmed in this review; title and journal confirmed via PubMed search-listing only, full abstract not opened, no statistic quoted.
  20. Akpinar A, Uğuz AC, Nazıroğlu M. J Membr Biol. 2014. — PC12 cell study; duloxetine alone and combined with agomelatine reported to reduce oxidative stress. Full title, PMID/DOI could not be independently confirmed in this review; journal and topic confirmed via PubMed search-listing only, full abstract not opened, no statistic quoted.Full title, PMID/DOI could not be independently confirmed in this review; journal and topic confirmed via PubMed search-listing only, full abstract not opened, no statistic quoted.
  21. Lee TK, et al. J Neurol Sci. 2016. — Rat cerebral ischemia-reperfusion model; duloxetine pretreatment reported neuroprotective via decreased oxidative stress. Full title, PMID/DOI could not be independently confirmed in this review; journal and topic confirmed via PubMed search-listing only, full abstract not opened, no statistic quoted.Full title, PMID/DOI could not be independently confirmed in this review; journal and topic confirmed via PubMed search-listing only, full abstract not opened, no statistic quoted.
  22. Meejuru GF, et al. J Pharm Pharmacol. 2021. — Mouse chronic immobilization-stress model; duloxetine reported protective, antioxidant-consistent effects. Full title, PMID/DOI could not be independently confirmed in this review; journal and topic confirmed via PubMed search-listing only, full abstract not opened, no statistic quoted.Full title, PMID/DOI could not be independently confirmed in this review; journal and topic confirmed via PubMed search-listing only, full abstract not opened, no statistic quoted.
  23. Wernicke J, et al. Hepatic effects of duloxetine-I: non-clinical and clinical trial data. Curr Drug Saf. 2008. — Duloxetine-specific hepatic toxicology and clinical-trial data paper. PMID 18690897. Flagged as inaccessible: repeated attempts to open this source directly returned no retrievable content. A search-result summary referenced toxicology "mitochondrial effects," but this detail could not be independently verified in the source text and is not cited as confirmed.Flagged as inaccessible: repeated attempts to open this source directly returned no retrievable content. A search-result summary referenced toxicology "mitochondrial effects," but this detail could not be independently verified in the source text and is not cited as confirmed.
  24. Cozart B, et al. Treatment of Depression With Duloxetine in Mitochondrial Encephalopathy. Clin Neuropharmacol. 2018. — Case report; duloxetine used to treat depression in a patient with pre-existing MELAS mitochondrial disease. Cited to clarify this is not evidence of duloxetine-induced mitochondrial harm. PMID/DOI could not be independently confirmed in this review; title, journal, and topic confirmed via PubMed search-listing only.PMID/DOI could not be independently confirmed in this review; title, journal, and topic confirmed via PubMed search-listing only.
  25. Nabekura T, Ishikawa S, Tanase M, Okumura T, Kawasaki T. Antidepressants induce toxicity in human placental BeWo cells. Curr Res Toxicol. 2022. — Human placental cell (BeWo) toxicity study of antidepressants; search-result summary referenced an effect on mitochondrial membrane potential among the antidepressants tested. Flagged as inaccessible: repeated attempts to open this source directly returned no retrievable content. Whether duloxetine specifically, as opposed to other antidepressants in the study, showed the referenced effect could not be confirmed, so this source is not counted as evidence anywhere in this dossier.Flagged as inaccessible: repeated attempts to open this source directly returned no retrievable content. Whether duloxetine specifically, as opposed to other antidepressants in the study, showed the referenced effect could not be confirmed, so this source is not counted as evidence anywhere in this dossier.
  26. Rejili M, et al. Duloxetine as a gut-brain axis modulator in irritable bowel syndrome: neuroimmune, microbiota, and pharmacogenomic perspectives. Naunyn Schmiedebergs Arch Pharmacol. 2026. — Duloxetine-titled narrative review connecting duloxetine to IBS gut-brain axis and microbiota. Full text/abstract could not be independently opened in this review despite repeated attempts; whether it presents original duloxetine-specific microbiome data or a class-level narrative only could not be confirmed. Not used to support any specific claim; cited for existence only.Full text/abstract could not be independently opened in this review despite repeated attempts; whether it presents original duloxetine-specific microbiome data or a class-level narrative only could not be confirmed. Not used to support any specific claim; cited for existence only.
  27. Ding P, Lu J, Lei T, et al. Antidepressant drugs promote the spread of broad-host-range plasmid in mouse and human gut microbiota. Gut Microbes. 2025. — Class-level antidepressant/gut-microbiota plasmid-transfer study. Full text not independently opened; whether duloxetine specifically was among the antidepressants tested could not be confirmed. No statistic cited from this source.Full text not independently opened; whether duloxetine specifically was among the antidepressants tested could not be confirmed. No statistic cited from this source.
  28. Rukavishnikov G, Leonova L, Kasyanov E, et al. Antimicrobial activity of antidepressants on normal gut microbiota: Results of the in vitro study. Front Behav Neurosci. 2023. — Class-level in vitro antidepressant/gut-bacteria study. Full text not independently opened despite repeated attempts; whether duloxetine specifically was among the antidepressants tested could not be confirmed. No statistic cited from this source.Full text not independently opened despite repeated attempts; whether duloxetine specifically was among the antidepressants tested could not be confirmed. No statistic cited from this source.
  29. Lukić I, Getselter D, Ziv O, et al. Antidepressants affect gut microbiota and Ruminococcus flavefaciens is able to abolish their effects on depressive-like behavior. Transl Psychiatry. 2019. — Class-level mouse antidepressant/gut-microbiota study. Full text not independently opened; whether duloxetine specifically was among the antidepressants tested could not be confirmed. No statistic cited from this source.Full text not independently opened; whether duloxetine specifically was among the antidepressants tested could not be confirmed. No statistic cited from this source.
  30. Yau AH, Xiong W, Ko HH. Collagenous enterocolitis manifesting as watery diarrhoea and iron-deficiency anaemia. BMJ Case Rep. 2015. — Case report; patient on duloxetine among other medications, anemia attributed to collagenous enterocolitis. Full text could not be independently opened in this review despite repeated attempts. Whether duloxetine contributed causally, or was merely a concurrent medication, could not be confirmed; not counted as evidence of a duloxetine-iron relationship.Full text could not be independently opened in this review despite repeated attempts. Whether duloxetine contributed causally, or was merely a concurrent medication, could not be confirmed; not counted as evidence of a duloxetine-iron relationship.
  31. PubMed search: "duloxetine" AND "copper", performed September 10, 2026. — All returned results concerned copper-catalyzed synthetic chemistry used in duloxetine manufacturing, not copper physiology; no biologically relevant result was returned.
  32. Europe PMC / PubMed search: "duloxetine" AND "mitochondria" / "mitochondrial", performed September 10, 2026. — Reviewed; no duloxetine-specific mitochondrial function or biogenesis study was confirmed among the results returned.
  33. Literature search: duloxetine AND (hypoxia OR "oxygen delivery" OR "oxygen consumption"), attempted September 10, 2026. — Search could not be completed; the request was blocked by repeated literature-database access and rate limits despite multiple retries across two source hosts. Disclosed here as an incomplete search, not a confirmed absence of evidence. This source is flagged as inaccessible for this review rather than cited as a completed null-result search.This source is flagged as inaccessible for this review rather than cited as a completed null-result search.