Cefprozil

Limits on what this shows

• Negative or null findings: None specific to cefprozil. No human mitochondrial marker (lactate, mtDNA, respirometry) study with cefprozil was found.
• At-risk subgroups: Not defined.
• Counter-evidence on record: The class study used cultured cells and mice, not people; doses and which beta-lactams were tested are unverified.

Direction

Potential direct drug effect on mitochondria (toxicity), not a nutrient shortfall. Class extrapolation; cefprozil absent from the record read.

Regulatory status

No mitochondrial text in the labels read. [1][2][3][4]

Population evidence

Kalghatgi 2013 [22] (cell culture and mice; abstract read as an OpenAlex record through a summarizing fetch tool): quinolones, aminoglycosides and beta-lactams caused mitochondrial dysfunction and ROS overproduction in mammalian cells and raised oxidative stress markers in mouse blood and tissue; the damage was reduced by N-acetyl-L-cysteine and avoided with bacteriostatic antibiotics. Ampicillin, cephalexin and cefprozil are not named in the record; the PMC full text returned a bot check, so the specific beta-lactams tested are unknown. Treated as cefprozil NOT in the cohort (Tier C). PubMed listing "cefprozil AND (oxidative OR mitochondria OR mitochondrial OR ATP OR reactive oxygen)" returned 4 records: a spectrophotometric assay, two polymer partitioning methods papers and an ABCB11 cholestasis case report; none measured mitochondrial function with cefprozil [28].

Mechanism

Proposed in the class study: the ROS generation that bactericidal antibiotics cause in bacteria is reported to occur as mitochondrial dysfunction and ROS overproduction in mammalian cells. Not tested for cefprozil.

Counter-evidence held on record

Boundary and why not higher Yellow is the ceiling for class extrapolation with cefprozil not confirmed in the cohort. Framing caution A theoretical class signal on toxicity, not a deficiency. No supplement (for example coenzyme Q10 or carnitine) is implied.

Limits on what this shows

• Negative or null findings: None specific to cefprozil.
• At-risk subgroups: Not defined.
• Counter-evidence on record: Same limits as the mitochondrial entry. The cefepime study points the other way for a different cephalosporin in infected rats, which shows how drug and infection effects are entangled.

Direction

Potential drug-induced oxidant load (toxicity), not a nutrient shortfall. Class extrapolation.

Regulatory status

No antioxidant or redox text in the labels read. [1][2][3][4]

Population evidence

Kalghatgi 2013 [22], as in the mitochondrial entry: mouse blood and tissue oxidative stress markers rose and antioxidant defense genes were upregulated with quinolones, aminoglycosides and beta-lactams; cefprozil not named; Tier C. The PubMed listing for oxidative and mitochondrial terms (4 records) held no cefprozil redox study [28]. An Egyptian rat study of cefepime (a different cephalosporin) in E. coli infection, which reported that cefepime ameliorated infection-raised malondialdehyde and lowered glutathione, superoxide dismutase and catalase, surfaced in the web search and is not cefprozil evidence [29].

Mechanism

Proposed in the class study; untested for cefprozil. A directly opposing signal (an antioxidant effect of a cephalosporin through infection control) is shown only for cefepime and is not carried over.

Counter-evidence held on record

Boundary and why not higher Yellow is the ceiling for class extrapolation with cefprozil not confirmed in the cohort. Framing caution No antioxidant supplement is implied; a theoretical class signal, not a deficiency.

Limits on what this shows

• Negative or null findings: The label lists no mineral chelation interaction for cefprozil. Cefprozil is about 95% absorbed in fasting subjects, which would leave little unabsorbed drug in the lumen to bind minerals (an inference, not a measurement) [1].
• At-risk subgroups: Not defined.
• Counter-evidence on record: Bench synthesis of metal complexes in a chemistry journal does not show a nutritional effect.

Direction

Chemical plausibility of metal binding only; direction in the body undetermined. The label's "copper reduction tests" text is a laboratory artifact (false-positive urine glucose), not a copper finding.

Regulatory status

No copper nutrient text in the labels read; urine glucose false-positive with copper reduction tests (Benedict's, Fehling's, Clinitest), enzyme tests not affected, ferricyanide blood glucose false-negative. [1][3][4]

Population evidence

El-Zawawy 2023 [19]: synthesis and characterization of cefprozil complexes with Co(II), Ni(II), Zn(II), Mn(II), Cr(III), Fe(III) and Cu(II) (IR, UV-Vis, magnetic susceptibility, ESR, elemental and thermal analysis, DFT); Fe(III) and Cu(II) bind through the amide and carboxylate groups with octahedral geometry; some complexes showed antimicrobial activity. Abstract read on the Springer page through a summarizing fetch tool; full text paywalled. No human or animal copper measurement. PubMed listing "cefprozil copper" and the broader mineral listing returned fetch client errors and are not counted [28].

Mechanism

Named in outline: chelation of Cu(II) by the amide and carboxylate groups. Whether this occurs at gut or plasma concentrations is not shown.

Counter-evidence held on record

Boundary and why not higher Tier B in vitro chemistry only; copper is held as its own entry (Hard Rule 7) and shares its source with the iron entry. Framing caution This is not a copper deficiency claim, and it does not support copper supplementation.

Limits on what this shows

• Negative or null findings: None found in either direction; the hemoglobin and Coombs text in the oxygen entry is not iron evidence.
• At-risk subgroups: Not defined.
• Counter-evidence on record: As for copper: bench chemistry, no clinical measurement.

Direction

Chemical plausibility of metal binding only; direction undetermined. Not an iron deficiency and not a hemoglobin finding.

Regulatory status

No iron interaction text in the labels read. Iron oxide appears only as an inactive tablet ink ingredient in one tablet label and is not listed in the suspension label. [1][2][3]

Population evidence

El-Zawawy 2023 [19]: the Fe(III) complex was synthesized and characterized (see copper entry). A 2002 analytical paper [20] used oxidation of cefprozil with Ce(IV) or Fe(III) in acid medium as a spectrophotometric assay; that is assay chemistry, not physiology, and is not counted as evidence of an interaction. PubMed listings "cefprozil iron" and the combined iron, copper and zinc listing returned fetch client errors and were not read [28].

Mechanism

Named in outline: Fe(III) binding by amide and carboxylate groups; not shown in vivo.

Counter-evidence held on record

Boundary and why not higher Tier B in vitro chemistry only; shares its source with the copper entry (see the overlap map). Framing caution Not an iron deficiency claim; no iron supplement or separation interval is implied by this entry.

Limits on what this shows

• Negative or null findings: Lode: effect on flora described as limited and fully reversed in 4 days [10]. Raymond: Simpson diversity change not significant, and levels returned to baseline at day 90 in most participants (two did not return to their baseline enterotype cluster) [9]. A 1992 Japanese study of 4 children aged 9 months to 6 years (cefprozil 10.0 to 11.6 mg/kg three times daily for 4 to 14 days) reported no significant change in major aerobes and anaerobes, no fungi bloom and undetectable fecal drug (listing-level summary only, abstract not opened in full) [13]. A pediatric otitis media randomized trial of 1,009 children with a cefprozil arm showed resistant gram-negative bacilli colonization falling after all regimens, with no cefprozil-specific result in the abstract [15]. Webb 2020 (Intermountain, hospital cohort) and Baxter 2008 (Northern California) do not name cefprozil in the text read and are not used [16][17].
• At-risk subgroups: Older adults in long-term care (Brown) [8]; the general adult claims population (Miller) [7]. The labels do not define a higher-risk group. Brown excluded residents with recent hospital stays or multiple antibiotics, so it does not speak to those groups.
• Counter-evidence on record: Both CDI estimates are observational and confounded by indication and healthcare exposure. In Miller the comparison is implicitly against subjects not exposed to cefprozil in the window (the reference group is not stated), not against another antibiotic, and cefprozil sits close to first-generation cephalosporins [7]. Brown has only 11 cefprozil cases, hence the wide interval (lower bound 1.13), and the paper does not explicitly discuss confounding by indication (per the summarizing tool); residents with combinations of antibiotics were excluded [8]. Effect sizes differ with design and window (OR 2.97 at 30 days in claims, relative risk 1.89 at 90 days in nursing homes). The flora studies are small (8 to 18 treated), mostly uncontrolled and in healthy young adults.

Direction

Disturbance and infection risk (an outcome entry with a direct composition arm). Not a nutrient deficiency, and no probiotic or supplement is implied. The named mechanism is loss of colonization resistance from antibiotic exposure of the gut flora.

Regulatory status

Tablet and suspension labels: "Clostridium difficile associated diarrhea (CDAD) has been reported with use of nearly all antibacterial agents," ranging from mild diarrhea to fatal colitis and reported "over two months after the administration of antibacterial agents"; "Prolonged use of cefprozil may result in the overgrowth of nonsusceptible organisms." Adverse reactions: diarrhea 2.9%, nausea 3.5%, vomiting 1%, diaper rash and superinfection 1.5%, genital pruritus and vaginitis 1.6%. The Canadian monograph says antibacterial treatment "may alter the normal flora of the colon and may permit overgrowth of Clostridium difficile" and gives diarrhea 2.7% (n=4227) but no CDAD incidence. [1][2][3][4]

Population evidence

Miller 2023 [7]: matched case-control (5 controls per case), Merative MarketScan commercial, Medicare and multi-state Medicaid claims, 159,404 community-associated CDI cases and 797,020 controls, exposure in the 30 days before the index date. Cefprozil row (Table 2, read twice through a summarizing fetch tool on the OUP and PMC pages): 244 exposed cases, 289 exposed controls, unadjusted OR 4.23, adjusted OR 2.97 (95% CI 2.4 to 3.68). Adjusted for Elixhauser comorbidities, PPI receipt, number of healthcare visits and other antibiotics (matching handled age, sex, insurance and enrollment period). The text calls cefprozil a "later-generation cephalosporin" with risk "comparable to first-generation cephalosporins" (the table footnote labels it third-generation, which conflicts with its ATC class J01DC and the second-generation description in [9]). Brown 2021 [8]: longitudinal case-cohort study of Ontario nursing-home residents aged 66 or older, 2012 to 2017, linked ICES health-administrative data, 82,592 residents and 1,708 CDI cases, resident-day unit with a 0.1% sample of control-days. Cefprozil: 352 thousand resident-days (0.3% of the cohort), 11 CDI cases, 3.13 per 100,000 resident-days (95% CI 1.56 to 5.59); for a 7-day course, 90-day unadjusted relative risk 2.61 (95% CI 1.49 to 3.89) and adjusted relative risk 1.89 (95% CI 1.13 to 3.02) versus no antibiotic in the prior 90 days; same table for scale: cefixime 4.26 (2.41 to 7.42), cefuroxime 2.46 (1.61 to 3.45), cephalexin 1.85 (1.54 to 2.15). Fourteen covariates including acid suppressants and Charlson index. Read through a summarizing fetch tool (OUP article page). Raymond 2016 [9]: healthy adults aged 21 to 35, cefprozil 500 mg orally twice daily for 7 days, 18 treated and 6 untreated controls, 72 stool samples (before, day 7, day 90), shotgun metagenomics. Lachnoclostridium bolteae rose in 16 of 18 (about 20-fold on average) and returned to baseline at day 90; assembled metagenome size fell 17.3% (about 118 to 97 million nucleotides, P=0.01); six families fell significantly (Bifidobacteriaceae, Coriobacteriaceae, Eubacteriaceae, Oxalobacteraceae, Pasteurellaceae, Veillonellaceae); Simpson diversity fell 7.6% (P=0.42, not significant); Enterobacter cloacae complex rose from about 0.0001% to 0.1% in six participants; resistance-gene content shifted (blaCfxA-6 allele changes). C. difficile was not measured. Abstract read on PubMed and OpenAlex records; numbers from the nature.com article page read through a summarizing fetch tool. Lode 1992 [10]: randomized double-blind placebo-controlled, 12 volunteers (8 cefprozil, 4 placebo), 500 mg every 12 hours for 8 days: moderate decrease in enterobacteria, slight increases in enterococci, staphylococci and bacteroides, all normalized 4 days after dosing; C. difficile not reported in the record read. Motohiro 1990 [11]: 8 healthy men, BMY-28100 (the development code indexed to cefprozil) 250 mg three times daily for 7 days, no control group: transient rises in total aerobes, E. coli and yeast-like organisms; transient falls in Veillonellaceae, Peptococcaceae and Eubacterium; total anaerobes constant; C. difficile in 1 subject and D-1 toxin in 5 although stools looked normal; no adverse effects. Abstract only. Jarmusch 2020 [12]: 14 healthy adults, 500 mg twice daily for 7 days, 16S V4 sequencing, self-controlled: observed OTUs significantly lower on study days 5, 7, 8 and 9 than day 1 and recovered by day 15; interindividual variability increased.

Mechanism

Named: suppression of susceptible commensals (enterobacteria, several anaerobic families, Veillonellaceae, Bifidobacteriaceae) with bloom of tolerant taxa (Lachnoclostridium bolteae, E. coli, yeast-like organisms, Enterobacter cloacae in a subgroup) and loss of colonization resistance. About 95% of a fasting dose is absorbed [1], so the amount reaching the colon is small; the Motohiro abstract reports no active metabolites detected in feces [11], the 4-child study found fecal cefprozil undetectable [13] and the Lode record gives no fecal drug level [10], so the route by which flora is changed (biliary excretion or residual unabsorbed drug) is not established by these records.

Why red

Re-tested against the five conditions. (1) Two independent Tier A sources with cefprozil in the cohort: Miller (US commercial and public claims, adults, 2001 to 2021, case-control) and Brown (Ontario nursing homes, ICES linked data, 2012 to 2017, case-cohort) differ in country, data system, population, design and outcome definition. (2) The category is directly measured: stool flora was cultured or sequenced in adults given cefprozil in four studies [9][10][11][12], and C. difficile infection is the category's own clinical endpoint, so the entry is an outcome-plus-composition entry labeled as such. (3) Direction is consistent: both CDI estimates have lower bounds above 1 (2.4 and 1.13) and every adult flora study shows shifts away from susceptible commensals. (4) Size is meaningful for a labeled adverse reaction (adjusted OR 2.97; relative risk 1.89), and the mechanism is named. (5) No substantial contradicting Tier A evidence: the null-leaning findings are a 4-child listing-level study [13], a non-significant Simpson index change inside a study that otherwise showed significant compositional shifts [9], and a moderate, reversible change in 8 treated volunteers [10]. The grade attributes a disturbance and infection risk to the drug; it is not a deficiency claim. It is at the lower edge: one estimate rests on 11 cases and both rest on summarized reads of the key tables. Re-verification 2026-10-10: the cefprozil rows were read again on the PMC page of Miller (244 exposed cases, 289 exposed controls, unadjusted OR 4.23, adjusted OR 2.97, 95% CI 2.4 to 3.68; the reference is implicitly unexposed controls) and on the OUP page of Brown (Table 2: 11 cases, 352 thousand resident-days, 3.13 per 100,000 with CI 1.56 to 5.59; Table 3: 7-day course, unadjusted relative risk 2.61 with CI 1.49 to 3.89, adjusted 1.89 with CI 1.13 to 3.02, adjusted 90-day incidence 1.07 versus 0.57 per 1000 for no antibiotic, 14 covariates). The Raymond figures (118 to 97 million nucleotides, P=0.01; Simpson 7.6%, P=0.42; 16 of 18) were re-read on the nature.com page, and the Lode and Jarmusch abstracts on OpenAlex records. All figures matched; each read was still through a summarizing tool, so the tables were not seen raw. Held at red because the listed conditions are met on those numbers; the confounding-by-indication presumption is addressed for the composition arm by the placebo-controlled [10] and untreated-control [9] volunteer designs.

Boundary and next step

Applies to adults (claims population and nursing-home residents) for CDI and to healthy adult volunteers on 7 to 8 day courses for flora; not to infants or young children (the 4-child study found no significant change). Advance to only after an editor opens the full texts to confirm the cefprozil rows (Miller Table 2, Brown Tables 2 and 3) and the Raymond cohort numbers. customerVisible is never assigned here. The labels direct that diarrhea after antibiotic use be considered for C. difficile.

Counter-evidence held on record

Framing caution This is an infection and disturbance risk, not a nutrient deficiency. No probiotic or supplement recommendation is made or implied.

Limits on what this shows

• Negative or null findings: No study of cefprozil measuring arterial oxygenation, methemoglobin, red-cell mass or tissue oxygen delivery was found (PubMed listing with iron, copper, zinc, ferritin, hemoglobin and methemoglobin terms was refused or returned a client error and is not counted) [28].
• At-risk subgroups: People with a history of cephalosporin-associated hemolytic anemia (Canadian monograph) [4]; not defined further by evidence.
• Counter-evidence on record: The label statements are class-wide and carry no cefprozil denominator or comparator, so they cannot be read as a cefprozil rate.

Direction

Adverse drug reaction and laboratory change (immune red-cell destruction). Not a nutrient deficiency, not iron deficiency and not a drug-intended effect.

Regulatory status

Tablet and suspension labels: "Positive direct Coombs' tests have been reported during treatment with cephalosporin antibiotics"; class-level reactions listed include hemolytic anemia, aplastic anemia, hemorrhage, pancytopenia, neutropenia and agranulocytosis; thrombocytopenia in postmarketing reports. "Decreased hemoglobin" does not appear in the tablet label text read, and the Canadian monograph says immune-mediated hemolytic anemia has been observed with cephalosporins and the drug should not be used in patients with a history of cephalosporin-associated hemolytic anemia. No incidence is given for either. [1][2][3][4]

Population evidence

None found with cefprozil. PubMed listing "cefprozil AND (hemolytic OR hemolysis OR Coombs OR anemia)" returned 25 records (10 shown), all efficacy trials, reviews or in vitro studies of group A beta-hemolytic streptococci; none of the 10 shown reported hemolysis, hemoglobin or red-cell mass in people given cefprozil [28]. No cefprozil case report of drug-induced immune hemolytic anemia was found in the opened records; a case report that surfaced in a web search (cefoperazone/sulbactam [31]) is not cefprozil evidence.

Mechanism

Named in outline for cephalosporins as a class: drug-dependent antibody binding to drug-coated red cells. Not demonstrated for cefprozil in any record opened.

Counter-evidence held on record

Boundary and why not higher Orange would need a cefprozil hemoglobin change quantified against a comparator, or an incidence for immune hemolysis with cefprozil. Neither was found. Framing caution A laboratory and allergic-type adverse reaction flag for clinicians. It says nothing about iron, B12 or folate status, and implies no supplement.

Limits on what this shows

• Negative or null findings: In the same class, amoxicillin (an aminopenicillin, a different beta-lactam subclass) had no significant effect on mitochondrial respiration in permeabilized human embryonic kidney cells in Sailer 2025 [35] (read as a summary on the Bioblast page; the PMC article returned a bot check and was not opened). In [33] cephalexin did not change succinate uptake after in vivo exposure. No human ATP, lactate, muscle energy or respirometry study with any cephalosporin was found, and no cefprozil study of any kind.
• At-risk subgroups: Not defined.
• Counter-evidence on record: The lowest ceftriaxone concentration tested (25 ug/mL) is roughly twice the cefprozil mean peak serum concentration of 11.5 mg/L (11.5 ug/mL) in volunteers given 500 mg [10], and ceftriaxone is a third-generation cephalosporin with different side chains and far longer exposure than a 7 to 10 day cefprozil course, so the class data may not apply at cefprozil exposures. The studies are isolated-organelle and rodent work, not people. Oxidative phosphorylation capacity is a surrogate: no ATP content was measured.

Direction

Potential direct drug effect on mitochondrial energy conversion (toxicity), not a nutrient shortfall. Class extrapolation; cefprozil absent from every record read. Held as its own entry (Hard Rule 7); it does not borrow the Kalghatgi study used for the mitochondrial and oxidative entries.

Regulatory status

No mitochondrial, energy or ATP text in the labels read. [1][2][3][4]

Population evidence

CLASS EXTRAPOLATION, Tier C, cefprozil NOT in the cohort of any source below; no human data. Barath 2022 [32] (isolated liver mitochondria from male Sprague-Dawley rats, 1 hour incubation, high-resolution respirometry, n=12; summarizing tool read of the MDPI page): ceftriaxone at 25, 250 and 2500 ug/mL gave oxidative phosphorylation capacity of 76.0 (SD 2.5), 66.5 (SD 5) and 63.5 (SD 4.2) pmol/s/mL versus 90.8 (SD 2.7) for control, significant at P under 0.05 per the paper (exact P values not stated); respiratory control ratio fell from 5.2 to 4.2 and hydrogen peroxide production rose from 1.9 to 3.8 at the 250 ug/mL level the authors call therapeutic; in the same study's rats, tissue oxidative phosphorylation was lower after ceftriaxone (245.9 and 236.5 versus 279.0 pmol/s/mL in the ceftriaxone and combined-treatment groups; tissue not specified in the read) and colon respiration was reduced while ileum was not significantly changed. ATP content and membrane potential were not measured. Older cephalosporin work: a J Pharmacol Exp Ther study [33] reported that cephaloglycin reduced net succinate uptake by about 70% in isolated mitochondria after in vivo and in vitro exposure, and that cephalexin inhibited succinate uptake only after in vitro exposure (no effect after in vivo exposure); ADP uptake was affected less, ATP was not reported, concentrations were not in the excerpt, and the title, authors and year were not shown on the page read. A related paper [34] (PMID 2133431; abstract truncated on the page read) states that cephaloridine, cephaloglycin and imipenem caused injury to renal mitochondrial anionic substrate uptake and respiration after 1 to 2 hours in vivo, that in vitro toxicity was nearly identical but immediate, and that it was reversed by washing the mitochondria or raising substrate concentration. Cefprozil is not mentioned in any of these. Searches for cefprozil-specific work: a PubMed listing for cefprozil with mitochondrial, membrane potential, respiration or bioenergetics terms was refused (HTTP 429) and is not counted; the earlier combined listing (4 records) held no ATP measurement [28]; the web search returned only the class studies above.

Mechanism

Described for the class, not shown for cefprozil: cephalosporin-induced loss of mitochondrial anionic substrate uptake (reported as acylation of membrane transport proteins in the discussion of [32], citing earlier work) and increased mitochondria-derived reactive oxygen species, which would lower oxidative phosphorylation capacity. Whether any of this applies to cefprozil is untested; cephalexin was the milder agent in [33].

Counter-evidence held on record

Boundary and why not higher Yellow is the ceiling for class extrapolation with cefprozil not in the cohort. Orange would need a cefprozil measurement of ATP, respiration or membrane potential, or a human energy-metabolism study that includes cefprozil. Framing caution A theoretical class signal on mitochondrial toxicity, not a deficiency and not a finding of low energy in people taking cefprozil. No supplement (for example coenzyme Q10 or carnitine) is implied.