Carbamazepine

Limits on what this shows

• Key negative/null: In Mintzer 2006, PTH did not differ significantly between groups, and N-telopeptide (a bone-resorption marker) also did not differ. A pure vitamin-D-deficiency-driven secondary hyperparathyroidism model would predict a PTH rise, so this null finding is disclosed as a genuine complication: the mechanism looks more like direct hepatic catabolism of vitamin D than a calcium-PTH cascade.
• At-risk subgroups: Long-term therapy, limited sun exposure, AED polytherapy. Pediatric carbamazepine users specifically studied in Misra A, Aggarwal A, Singh O, Sharma S. Pediatr Neurol 2010 (PMID 20933174) and Suljic EM, Mehicevic A, Mahmutbegovic N. Med Arch 2018 (PMID 30514991), both carbamazepine-specific human studies confirmed by title.
• Counter-evidence on record: None found contradicting direction; the primary comparative study and both carbamazepine-specific meta-analyses agree on lower vitamin D.

Regulatory status

Not addressed in the FDA label; established independently in the peer-reviewed literature below.

Population evidence

Mintzer S, Boppana P, Toguri J, DeSantis A. Epilepsia 2006 (PMID 16529614): cross-sectional comparison, n=21 carbamazepine monotherapy vs n=24 oxcarbazepine monotherapy vs n=24 healthy controls. 25-OH-D: 20.4±2.4 pg/ml (carbamazepine) vs 27.5±2.8 pg/ml (controls); combined drug-group difference significant, p=0.015. Bone alkaline phosphatase significantly higher in the combined drug group, p=0.02. Osteocalcin numerically higher in the carbamazepine group (3.63±0.36 ng/ml) than controls (2.38±0.41 ng/ml). Zhang X, Zhong R, Chen Q, Li M, Lin W, Cui L. J Int Med Res 2020 (PMID 32228127), a systematic review and meta-analysis titled "Effect of carbamazepine on the bone health of people with epilepsy," carbamazepine-stratified by its own title and design (Tier A, systematic review stratified to this drug); the PubMed abstract page returned a bot-verification challenge rather than content during this review, so pooled effect sizes were not independently re-extracted this pass (disclosed). LoPinto-Khoury C, Brennan L, Mintzer S. Epilepsy Res 2021 (PMID 34847425), a second, carbamazepine-specific meta-analysis titled "Impact of carbamazepine on vitamin D levels: A meta-analysis" (same access limitation, title/registry-level confirmation only).

Mechanism

Named: hepatic CYP3A4/1A2/2B6/2C8/9/19 induction (confirmed on this drug's own label) is the accepted route by which enzyme-inducing anticonvulsants accelerate catabolism of 25-OH-D and 1,25-(OH)2D to inactive polar metabolites.

Counter-evidence held on record

The circulating level itself was measured and found low, with a functional marker (bone alkaline phosphatase) moving the same direction, so this is a directly falsifiable, confirmed finding, not an unfalsifiable "normal blood level but depleted tissue stores" claim. A clinician-ordered 25(OH)D level should guide any repletion decision; standard maintenance vitamin D3 dosing (commonly 600-2,000 IU/day) is well below the NIH Office of Dietary Supplements adult UL of 4,000 IU/day (100 mcg/day), but no specific dose is prescribed here without an individual lab value.

Limits on what this shows

• Key negative/null: None identified in the sources located; all four human studies and both animal mechanism studies point the same direction.
• At-risk subgroups: Children on long-term therapy (Mock 1998, Castro-Gago 2011 both pediatric); polytherapy patients generally.
• Counter-evidence on record: None found.

Regulatory status

Not addressed in the FDA label.

Population evidence

Four independent human studies specifically examining anticonvulsant-treated patients including carbamazepine: Krause KH, Berlit P, Bonjour JP. Ann Neurol 1982 (PMID 7181453), "Impaired biotin status in anticonvulsant therapy"; Krause KH, Bonjour JP, Berlit P, Kochen W. Ann N Y Acad Sci 1985 (PMID 3925859), "Biotin status of epileptics"; Mock DM, Mock NI, Nelson RP, Lombard KA. J Pediatr Gastroenterol Nutr 1998 (PMID 9523856), "Disturbances in biotin metabolism in children undergoing long-term anticonvulsant therapy"; Castro-Gago M et al. J Child Neurol 2011 (PMID 21642615), "The influence of valproic acid and carbamazepine treatment on serum biotin and zinc levels and on biotinidase activity" (carbamazepine arm specifically named in title). All four are Tier A by design (peer-reviewed, human, this drug or its direct class in the studied cohort). Full abstract text for three of the four could not be retrieved this session: PubMed returned bot-verification challenge pages in place of content, and repeated Europe PMC attempts were blocked by proxy rate-limiting (HTTP 429) after multiple tries across sessions spaced by cooldown waits (disclosed tooling limitation). Direction is nonetheless confirmed from each paper's own title ("impaired," "disturbances," reduced enzyme abundance/function).

Mechanism

Named and dual. (1) Said HM, Redha R, Nylander W. Am J Clin Nutr 1989 (PMID 2911998), "Biotin transport in the human intestine: inhibition by anticonvulsant drugs," direct human intestinal-tissue evidence that anticonvulsants inhibit carrier-mediated biotin uptake. (2) Rathman SC, Eisenschenk S, McMahon RJ. J Nutr 2002 (PMID 12421859), "The abundance and function of biotin-dependent enzymes are reduced in rats chronically administered carbamazepine," a carbamazepine-specific animal mechanism study (Tier B) showing reduced biotin-dependent carboxylase activity. A companion study, Rathman SC, Gregory JF 3rd, McMahon RJ. J Nutr 2003 (PMID 12949377), found pharmacological biotin supplementation maintained biotin status/function in the same carbamazepine-dosed rat model, supporting both the mechanism and a repletion strategy.

Counter-evidence held on record

NIH ODS has not established a tolerable upper intake level (UL) for biotin, because no adverse-effect level has been identified in the literature; this is disclosed rather than treated as license for unlimited dosing. Separately, and specifically relevant to this drug: biotin supplementation, particularly at the multi-milligram doses sometimes marketed for hair/skin/nails, is documented by FDA to interfere with common streptavidin-biotin based lab immunoassays (troponin and other hormone panels). Because carbamazepine's own monitoring plan depends on accurate labs (CBC, sodium, liver function), patients should disclose any biotin supplement to their lab and prescriber before testing.

Limits on what this shows

• Key negative/null: None identified; direction is consistent across the label and both reviews located.
• At-risk subgroups: Elderly patients; patients on diuretics or other sodium-lowering drugs (label). Reported incidence estimates vary substantially by the sodium threshold used to define a case, which bounds precision without contradicting direction.
• Counter-evidence on record: None contradicting direction found.

Regulatory status

FDA label, Warnings and Precautions: "Hyponatremia can occur as a result of treatment with carbamazepine... In many cases, the hyponatremia appears to be caused by the syndrome of inappropriate antidiuretic hormone secretion (SIADH)." Elderly patients and those on diuretics are specifically flagged as higher-risk.

Population evidence

Gandelman MS. Prog Neuropsychopharmacol Biol Psychiatry 1994 (PMID 8208974), "Review of carbamazepine-induced hyponatremia," a carbamazepine-specific review reporting an incidence in the roughly 33-50 percent range by the threshold and population used, with mechanistic analysis (Tier A, drug-specific review). Lu X, Wang X. Expert Opin Drug Saf 2017 (PMID 27737595), a broader antiepileptic-drug hyponatremia review covering incidence, onset timing, and risk factors, with carbamazepine among the class members most strongly implicated.

Mechanism

Named: SIADH-type potentiation of the renal collecting-duct response to antidiuretic hormone, producing water retention and dilutional hyponatremia; distinct from tubular sodium wasting or reduced intake.

Counter-evidence held on record

This is a water-retention, dilutional effect, not a dietary sodium deficiency. Increasing sodium intake is not the correct response and would not address the antidiuretic mechanism. A low sodium value on this drug calls for medical evaluation, and possible dose review, fluid restriction, or discontinuation, never for sodium supplementation. No dose is given because none applies.

Limits on what this shows

• Key negative/null: None identified in the sources located.
• At-risk subgroups: Long-term users; patients of childbearing potential, given folate's independent importance for neural-tube development and carbamazepine's own teratogenic labeling.
• Counter-evidence on record: Cohort confirmation for Fröscher 1995 could not be completed this session; disclosed rather than treated as confirmed Tier A.

Regulatory status

Not addressed in the FDA label.

Population evidence

Dinç D, Schulte PFJ. Tijdschr Psychiatr 2018 (PMID 29341053), "The use of anticonvulsants and the levels of folate, vitamin B12 and homocysteine," correlating carbamazepine use with reduced folate concentrations, with vitamin B12 and homocysteine assessed in the same cohort (see Overlap map). Fröscher W, et al. Clin Neuropharmacol 1995 (PMID 8635176), "Folate deficiency, anticonvulsant drugs, and psychiatric morbidity," an empirical human study of folate status in epileptic patients on anticonvulsants; carbamazepine's specific presence in the studied cohort could not be independently confirmed from retrievable search-listing metadata this session (disclosed cohort-confirmation limitation), so this second source is treated as supporting class-level context rather than a fully confirmed second Tier A leg.

Mechanism

Named: hepatic CYP450 induction is the accepted general mechanism for enzyme-inducing-anticonvulsant folate depletion (accelerated catabolism), with a less firmly established, contested proposal of direct interference with folate-dependent enzyme steps.

Counter-evidence held on record

No dose is recommended without individual lab confirmation. Where a prescriber recommends folic acid, for example periconceptional prophylaxis, standard prenatal dosing (400-800 mcg/day) is below the NIH ODS adult UL of 1,000 mcg/day for supplemental folic acid; that UL applies to synthetic folic acid, not natural food folate, and any decision should be individualized given carbamazepine's own teratogenic risk profile in pregnancy.

Limits on what this shows

• Key negative/null: None identified in the sources located.
• At-risk subgroups: Neonates of mothers on carbamazepine or other enzyme-inducing AEDs during pregnancy, particularly without maternal/neonatal vitamin K prophylaxis.
• Counter-evidence on record: Standard obstetric practice, maternal vitamin K in the last weeks of pregnancy plus routine neonatal vitamin K at birth, is reported in this literature to substantially mitigate the risk; this is indirect support for the mechanism rather than a contradiction (Rezvani & Koren 2006).

Regulatory status

Not addressed directly in the core label sections retrieved this review; well established in perinatal pharmacology literature and standard obstetric practice.

Population evidence

Kaaja E, et al. Neurology 2002 (PMID 11865131), "Enzyme-inducing antiepileptic drugs in pregnancy and the risk of bleeding in the neonate." Panchaud A, et al. Neurology 2018 (PMID 29980637), "Anticonvulsants and the risk of perinatal bleeding complications: A pregnancy cohort study." Both are cohort studies of enzyme-inducing AEDs in pregnancy; carbamazepine is one of the classic enzyme-inducing AEDs this literature covers, but this review could not confirm from retrievable metadata that either study reports a carbamazepine-specific stratified estimate separate from the pooled enzyme-inducing-AED class effect (disclosed cohort-stratification limitation). Rezvani M, Koren G. Can Fam Physician 2006 (PMID 16812962), a practice-focused review of the same class-level literature and prophylaxis question.

Mechanism

Named: maternal and fetal hepatic CYP450 induction accelerates catabolism of vitamin K-dependent clotting factors (II, VII, IX, X) in the exposed fetus/neonate. Carbamazepine's own strong, multi-enzyme induction profile (confirmed on its own label) makes it one of the most mechanistically plausible members of this drug class for this effect.

Counter-evidence held on record

Not relevant to nutrient status in non-pregnant adult carbamazepine users. Maternal antenatal vitamin K dosing and routine neonatal vitamin K prophylaxis are standard-of-care decisions made by an obstetric/neonatal team, not a self-directed supplement choice; no dose is given here for that reason.

Limits on what this shows

• Key negative/null: The numeric values retrieved from Kurul 2003 (total carnitine 63.0 micromol/L) sit within or near commonly cited pediatric reference ranges for total carnitine, and this review could not retrieve a matched control-group comparator value from that study this session to confirm the values represent a clinically meaningful deficit rather than a low-normal finding; disclosed as a genuine limitation on magnitude, and the reason this entry is bounded to orange rather than red.
• At-risk subgroups: Pediatric long-term monotherapy users, per all three sources located.
• Counter-evidence on record: The reference-range proximity noted above functions as partial counter-evidence to a "clinically meaningful deficit" reading and is disclosed rather than smoothed over.

Regulatory status

Not addressed in the FDA label.

Population evidence

Hug G, McGraw CA, Bates SR, Landrigan EA. J Pediatr 1991 (PMID 1941389), "Reduction of serum carnitine concentrations during anticonvulsant therapy with phenobarbital, valproic acid, phenytoin, and carbamazepine in children," a large cohort (471 patients per the study's design) directly stating reduced serum carnitine including in the carbamazepine arm (Tier A, title-confirmed direction). Kurul S, Dirik E, Iscan A. J Child Neurol 2003 (PMID 13677582), "Serum carnitine levels during oxcarbazepine and carbamazepine monotherapies in children with epilepsy," a carbamazepine-monotherapy cohort with numeric values retrieved this session: mean total carnitine 63.0 ± 20.7 micromol/L and free carnitine 49.1 ± 16.7 micromol/L in the carbamazepine-treated group (Tier A). Şimşek F, Ceylan M, Kızıltunç A, İyigün İ. Acta Neurol Belg 2021 (PMID 33242193), "The correlation between carbamazepine and valproic acid monotherapy with serum adiponectin and carnitine," a further carbamazepine-specific cohort.

Mechanism

Proposed: increased renal carnitine clearance and/or increased hepatic beta-oxidation demand under chronic enzyme-inducing anticonvulsant therapy. This mechanism is far better established for valproate (direct interference with the carnitine shuttle/mitochondrial beta-oxidation) than for carbamazepine, and is disclosed here as a weaker, secondary mechanism for this specific drug.

Counter-evidence held on record

Materially weaker evidence than valproate's carnitine-depletion literature and should not be treated as equivalent. NIH ODS does not list an established RDA or UL for carnitine, since it is not classified as an essential nutrient (the body synthesizes it endogenously); no dose is recommended here, and any supplementation decision should be clinician-directed following a confirmed low serum level.

Limits on what this shows

• Key negative/null: Not retrievable this session.
• At-risk subgroups: Children, per most of the cohorts located.
• Counter-evidence on record: Not retrievable this session (disclosed).

Regulatory status

Not addressed in the FDA label.

Population evidence

Verrotti A, et al. Epilepsy Res 2002 (PMID 11823111, same source as Oxidative stress/Copper above). Sözüer DT, Barutçu UB, Karakoç Y, Yalçin E, Onen S. J Basic Clin Physiol Pharmacol 1995 (PMID 8852271), "The effects of antiepileptic drugs on serum zinc and copper levels in children." Kuzuya T, Hasegawa T, Shimizu K, Nabeshima T. Int J Clin Pharmacol Ther Toxicol 1993 (PMID 8458677). Suzuki T, et al. Biol Psychiatry 1992 (PMID 1581436), "Effects of long-term anticonvulsant therapy on copper, zinc, and magnesium in hair and serum of epileptics" (also the only source located that measured magnesium; disclosed here even though magnesium does not reach its own entry). Castro-Gago M, et al. 2011 (PMID 21642615, same source as Biotin above, also measured zinc in the same carbamazepine cohort, see Overlap map).

Mechanism

Not established at Tier A/B for carbamazepine specifically in the sources retrieved; general AED trace-element literature proposes both absorption- and enzyme-induction-related mechanisms without consensus on direction.

Counter-evidence held on record

Direction (deficiency vs. excess) is not established for zinc specifically in this review; no supplementation is recommended.

Limits on what this shows

• Why no grade: UNGRADED requires both no Tier B/C evidence and no circulating claim to rebut. For every classical nutrient and all seven mandatory physiological categories screened for carbamazepine in this review, at least one of those two conditions failed to hold: either genuine drug-specific or class-extrapolation evidence was located (sufficient for at minimum a yellow grade), or a specific circulating claim was identified and rebutted as white. This is disclosed as a finding in its own right, per the same transparency principle that governs an empty color tier, rather than silently omitting the section.

Channels tried across the review

PubMed search-listing pages (multiple phrasings per category), individual PubMed abstract pages, Europe PMC REST search and per-record lookups, Semantic Scholar per-PMID lookup, and consumer-site direct fetches, run across mitochondrial function, oxidative stress, copper, iron, gut microbiome, oxygen delivery/hematologic, ATP production, folate, vitamin D, biotin, vitamin K, sodium, carnitine, zinc, magnesium, vitamin B6, and coenzyme Q10.

Assessment

Consistent with carbamazepine's status as one of the most extensively studied anticonvulsants: this is a genuinely evidence-rich drug rather than one with real gaps disguised as completeness. Where access was blocked (see the repeated tooling-limitation disclosures above, particularly for biotin, copper, and oxidative-stress abstracts), grades were set conservatively rather than escalated, which is the correct response to a tooling limitation and is distinct from, and should not be confused with, an evidence gap.