Rifampin

Limits on what this shows

• Null findings: None found. No human mitochondrial endpoint (lactate, respirometry in blood cells, mtDNA) was located.
• At-risk subgroups: Not defined for mitochondrial effects.

Regulatory status

No mitochondrial statement in the label sections retrieved; the label carries a hepatotoxicity warning [1], which is a toxicity outcome, not a mitochondrial deficiency.

Evidence (Tier B, in vitro and animal)

HepG2 cells [27]: rifampicin reduced ATP dose- and time-dependently (24-hour IC50 about 0.5 mM versus 70 mM for isoniazid and 84 mM for pyrazinamide); drug combinations produced synergistic inhibition of complex I; reductions in complex I and III activity, NAD+ and membrane potential with higher lactate were reported for the drugs collectively, not split by drug (abstract and repository page). QSG7701 hepatocytes, 72 hours [28] (access via a summarizing fetch tool; authors and title not retrieved): rifampicin lowered mitochondrial membrane potential and raised mitochondrial ROS and permeability more than isoniazid, and increased Drp1 and its activated form p-Drp1 S616 on mitochondria. Rats given isoniazid, rifampicin and pyrazinamide for 45 days (n=6 per group) [29]: swollen mitochondria on electron microscopy in the combination group (conference abstract; no rifampicin-only arm).

Mechanism

Named in vitro: complex I and III inhibition, membrane potential loss, Drp1 fission. The relation of the 0.5 mM HepG2 concentration to concentrations reached in people was not established from the sources opened.

Counter-evidence held on record

Counter-evidence Cell lines at high concentration, combination-only animal data, no human data. Framing caution Mechanistic hypothesis only. Hepatotoxicity is a drug toxicity, not evidence of a nutrient need, and this entry supports no antioxidant or mitochondrial supplement.

Limits on what this shows

• Null and contradicting findings: The two human cohorts show improvement, not worsening, of oxidant markers on rifampin-containing therapy. Persistently low vitamin C and E after treatment in [30] reflects disease or diet and is not attributable to rifampin (Hard Rule 6).

Regulatory status

No oxidative stress statement in the label sections retrieved [1].

Evidence (Tier B)

Rats, triple combination for 45 days (n=6 per group) [29]: malondialdehyde rose (P<0.001) and fell with silymarin; no rifampicin-only arm; conference abstract. QSG7701 hepatocytes [28]: mitochondrial ROS rose more with rifampicin than isoniazid.

Human cohorts (Tier A for rifampin in regimen, attribution not separable)

Nepal [30]: 56 newly diagnosed patients and 50 controls; 35 re-sampled after 2 months of isoniazid, rifampicin, pyrazinamide, ethambutol: MDA 8.37 +/- 1.55 to 3.84 +/- 1.28 nmol/mL (P<0.0001); nitrite 49.58 to 27.71 umol/L; vitamin C 0.83 to 1.19 mg/dL (P<0.0001); vitamin E 0.85 to 0.94 mg/dL (not significant); vitamin C and E stayed below controls. Iraq [31]: 46 patients on a fixed-dose tablet containing isoniazid, rifampicin 150 mg, pyrazinamide and ethambutol plus vitamin B6 10 mg daily for 2 months: MDA 2.06 +/- 0.21 to 1.64 +/- 0.13 umol/L and total antioxidant status 1.11 +/- 0.19 to 1.37 +/- 0.12 mmol/L (both P<0.0001; controls MDA 1.10, TAS 1.86). Neither has an untreated tuberculosis comparison.

Mechanism

Mitochondrial ROS generation in vitro; not demonstrated in people.

Counter-evidence held on record

Counter-evidence Disease inflammation resolving is the leading explanation for the human marker changes. Hepatotoxicity is a toxicity outcome, not an oxidative-stress deficiency. Framing caution Do not read this as a rationale for antioxidant supplements. It is a mechanistic flag only.

Limits on what this shows

• Null findings: A PubMed search of rifampicin with iron absorption, serum iron, ferritin or iron deficiency (45 records; first 10 viewed) found no human study measuring iron status as an effect of rifampicin.

Regulatory status

No iron statement in the label sections retrieved. The label records that rifampin induces delta-aminolevulinic acid synthetase and notes isolated reports of porphyria exacerbation [1]; hemolysis is covered in the oxygen delivery entry.

Evidence (Tier B)

Cell and mouse study [32] (PubMed-listing summary; full text not opened): rifampicin inhibited HSPA8, enhanced autophagy-driven ferritin degradation and raised free iron, with ferroptosis and liver toxicity relieved by blocking HSPA8 or autophagy. hPXR-transgenic mice given rifampicin and isoniazid [33] (listing summary): liver proteins in iron-sulfur cluster assembly and heme biosynthesis were altered.

Human context (not drug-attributable)

Tuberculosis patients had lower serum iron, zinc, copper and selenium than healthy controls in an Iranian cohort of 180 patients and 180 controls [36]; the drugs were not named in the abstract and no values were given.

Mechanism

Present in models only. No evidence that iron absorption or stores fall on rifampin.

Counter-evidence held on record

Counter-evidence Models are hepatic-toxicity studies. Iron indices in tuberculosis are driven by inflammation (Hard Rule 6). Framing caution Not an iron deficiency claim, and the orange discoloration of urine and sweat caused by rifampin is a pigment effect, not iron or blood loss. Do not recommend iron from this entry.

Limits on what this shows

• Null findings: A PubMed search of rifampicin with copper or ceruloplasmin (55 records; first 10 viewed) returned only materials chemistry, nanoparticle and in vitro microbiology papers; none measured copper as a biological analyte in people on rifampicin.

Regulatory status

No copper statement in the label sections retrieved [1].

Population evidence (Tier C for this drug)

Sepehri 2018 [36]: 180 pulmonary tuberculosis patients and 180 healthy controls in Sistan, Iran, sampled at treatment start and at months 2, 4 and 6; serum Cu, Zn, Se and Fe lower than in controls (P<0.001); Cu/Se and Cu/Zn ratios differed (P<0.001). No concentrations in the abstract, and the abstract names no drug, so under the cohort test rifampin cannot be confirmed in the studied regimen. Abstract only (PubMed summary).

Class cohort test

CLASS EXTRAPOLATION. Rifampin was not named in the cohort. Capped at yellow. The finding cannot be attributed to rifampin.

Mechanism

Absent for rifampin. Serum copper rises with inflammation, so its pattern in tuberculosis is a disease signal. The copper chelation mechanism cited for neuropathy in a first-line drug review belongs to ethambutol, a different drug (search record, not opened).

Counter-evidence held on record

Counter-evidence Confounding by indication is presumed and unresolved: the comparison is to healthy controls, not to untreated patients. Framing caution Do not display as a rifampin copper effect. Copper remains its own entry and is not merged with iron.

Limits on what this shows

• Null and contradicting findings: Shannon diversity was preserved on treatment in [24]; differences between latent tuberculosis participants and healthy volunteers were not significant in [22] because of size; active tuberculosis itself lowers diversity [25], so disease is a confounder of every four-drug cohort; healthy volunteers in [22] were stable but not diet-controlled; taxon-level directions are not consistent across cohorts (Coprococcus rose in [22] and fell more than 200-fold in [24]). No functional or clinical consequence (infection, nutrient status) was measured for rifampin.
• At-risk subgroups: Not defined. All human cohorts are small and HIV-negative where stated [24].

Regulatory status

No microbiome statement in the label sections retrieved. The label does record that rifampin is eliminated rapidly in bile and undergoes enterohepatic circulation [1], which exposes the gut lumen.

Population evidence (Tier A, rifampin in cohort)

Rifampin alone [22][23]: prospective cohort, Florida clinics; 6 latent tuberculosis participants (4 on daily rifampicin, the 4R regimen, and 2 on weekly rifapentine plus isoniazid) and 6 healthy volunteers, stool 16S sequencing at baseline, during and about 2 months after therapy. Rifamycin therapy explained 17.0% of weighted UniFrac variance (F=4.5, P=0.001); the 4R (rifampin) subgroup alone 6.5% (F=8.2, P=0.001). Clostridium XIVa, Romboutsia and Roseburia fell; Coprococcus, Lachnospiraceae, Ruminococcaceae and Parasutterella rose. Alpha diversity fell and rose again after therapy but stayed below pre-treatment. The 4R subgroup size (4) comes from the companion data note [23]; no 4R-specific alpha diversity numbers are given in the text read. The Scientific Reports paper and the BMC Research Notes data note describe the same cohort and count as one source. Four-drug regimen (isoniazid, rifampin, pyrazinamide, ethambutol) [24]: Haiti, cross-sectional; 19 on treatment (mean 3.4 months) versus 25 age-matched latent tuberculosis controls and 50 uninfected controls. Observed OTUs about 150 versus 230 (P=0.0074) and 229 (P=0.0122); Shannon diversity unchanged; composition differed (PERMANOVA P=0.023). Lactobacillus and Coprococcus fell more than 200-fold, Ruminococcus 675-fold, Bifidobacterium about 20-fold, Blautia about 10-fold. Cured patients (n=19, mean 1.2 years after therapy) still clustered apart from controls (P=0.007). Second four-drug cohort [25]: China, 16S plus fecal metabolomics: untreated active tuberculosis (n=41) had lower alpha diversity than latent infection (n=30) or healthy volunteers (n=49); on isoniazid, rifampin, pyrazinamide, ethambutol for 2 months then isoniazid and rifampin, evenness and Simpson indices rose relative to untreated disease. Treated-group sample sizes were not in the page text read.

Tier B support

Mouse model [26] (3 to 5 mice per group depending on the experiment): in single-drug arms rifampin alone reduced diversity to a degree similar to the full three-drug regimen, and the community stayed distinct from naive mice for at least 3 months after therapy. Ethambutol was not used in that model.

Mechanism

Present: broad antibacterial activity with biliary excretion and enterohepatic cycling [1]; rifamycin or metabolite was still detectable in stool of 3 of 6 participants 2 months after therapy [22], which the authors note may confound recovery estimates.

Why orange, not red

Only 4 participants received rifampin without other agents; the larger cohorts combine four drugs, so attribution to rifampin rests on [22] and the mouse data [26]. No clinical consequence is demonstrated. Boundary: small cohorts, therapy windows of 2 to 6 months, incomplete recovery.

Counter-evidence held on record

Counter-evidence Disease-driven dysbiosis, small samples, non-randomized controls and cross-sectional designs in the two four-drug cohorts. Results are composition changes; none shows loss of a nutrient. Framing caution This is a signal of microbial composition change on rifamycin therapy, not evidence of vitamin or short-chain fatty acid deficiency, and not a reason to add probiotics or supplements.

Limits on what this shows

• Null findings: None for continuous daily therapy were retrieved; the hemoglobin effect of ordinary daily dosing was not measured in any opened source. Tuberculosis itself lowers hemoglobin through inflammation, a confounder not tested here.
• At-risk subgroups: People on intermittent dosing, high-dose intermittent regimens, or restarting after a drug-free interval [1][34][35].

Regulatory status

Label: hematologic reactions include thrombocytopenia (mainly with high-dose intermittent therapy), leukopenia, hemolytic anemia, decreased hemoglobin, bleeding and, rarely, agranulocytosis and DIC; flu-like syndrome (fever, chills, headache, dizziness, bone pain) with intermittent dosing or irregular dosing [1]. Read through a summarizing fetch tool; incidence figures were not in the text retrieved.

Population evidence

Poole 1971 [35]: 49 patients moved from daily to high-dose twice-weekly rifampicin; rifampicin antibodies in 16 (33%); side effects in 9 of 16 with antibodies versus 2 of 33 without (P<0.001); thrombocytopenia in 3 of 49 (6%); the abstract reports no hemolysis. Sanwal 2020 [34] (single case, Tier B): 49-year-old with latent tuberculosis, hemolysis within 4 days of restarting rifampin after a 2-week gap; hemoglobin 13.4 to 8.5 g/dL, direct Coombs test positive, LDH 450 units/L, hemoglobin casts on renal biopsy; 8 dialysis sessions over 17 days; recovered after withdrawal. No cohort incidence of hemolysis on continuous daily dosing was retrieved.

Mechanism

Consistent with drug-dependent antibody hemolysis: rifampicin antibodies are common on intermittent regimens [35] and the case had a positive direct Coombs test [34]. This is an immune reaction, not an effect on iron, B12 or folate.

Hard Rule 2

Hemolytic anemia and low hemoglobin are adverse-reaction outcomes. They are not filed as iron deficiency, and no nutrient is implied. Nothing in this entry supports supplementation.

Counter-evidence held on record

Counter-evidence Rare event; the only detailed human report is a single case; the opened cohort abstract [35] did not report hemolysis. Framing caution An adverse drug reaction to watch for with irregular or restarted dosing. It is not a nutrient finding and not a reason to take an iron or other supplement.

Limits on what this shows

• Null findings: None found; no human ATP surrogate (lactate, phosphorus magnetic resonance, red-cell ATP) located.

Regulatory status

No ATP or energy-metabolism statement in the label sections retrieved [1].

Evidence (Tier B)

HepG2 [27]: ATP decreases were dose- and time-dependent over 4, 24 and 48 hours; IC50 at 24 hours about 0.5 mM for rifampicin, about 140 times lower than isoniazid (70 mM) and 170 times lower than pyrazinamide (84 mM). QSG7701 [28]: both rifampicin and isoniazid reduced ATP production to a similar degree at 72 hours.

Mechanism

Downstream of the respiratory chain effects in the mitochondrial burden entry; the same sources, so not independent.

Counter-evidence held on record

Counter-evidence In vitro only, supraclinical concentration not excluded. Framing caution Not a claim of low energy or fatigue from rifampin, and no basis for energy supplements.