First draft prepared by
    Dr R. Fuchs
    Institute for Medical Research and Occupational Health
    University of Zagreb, Zagreb, Croatia


         Chloramphenicol (CAP) is a broad spectrum antibiotic used in cattle,
    swine and poultry in dose ranges of 22-66 mg/kg bw. Absorption of the
    drug by the oral or parental route of administration is rapid, with
    maximum blood concentrations reached in 1 to 5 hours. The major route of
    excretion in pigs and cattle is in the urine.

         CAP had been previously evaluated at the twelfth and thirty-second
    meetings of the Committee (Annex 1, references 17 and 80). At the
    thirty-second meeting the Committee was unable to establish an ADI
    because it was not possible to give an assurance that residues in foods
    of animal origin would be safe for human consumption; it was concluded
    that human exposure to CAP can cause aplastic anaemia.

         This monograph addendum summarizes the information on CAP that has
    become available since the previous evaluation.


    2.1  Biochemical aspects

    2.1.1  Biotransformation

         Isolated liver cells from rats and rainbow trout were used to study
    biotransformation of labelled CAP. After 2 hours in suspension, 85 and
    25% of the dose was biotransformed by rat and trout hepatocytes,
    respectively, and in both species primarily by glucuronidation. Three
    major phase I metabolites, identified as CAP-oxamic acid, CAP base, and
    CAP alcohol, were found in the hepatocyte suspensions. In the rat the
    pattern of metabolites produced  in vitro was significantly different
    from the metabolites reported  in vivo, CAP-arylamine and -arylamide
    being identified in rat urine. The authors concluded that these
    metabolites resulted from the action of the intestinal microflora and,
    to a lesser extent, from the activity of tissue nitroreductase. In the
    trout, no CAP-oxamic acid was detected in urine. The gills are a more
    likely route of elimination of this metabolite in the trout (Cravedi &
    Baradat, 1991).

         Quantitative analysis of goat urinary metabolites of CAP indicates
    that the glucuronide prevails (36.5%). The sulfate (22.5%) and phosphate
    (7.9%) contribution to the detoxification of CAP are also important (Wal
     et al., 1988).

         In human patients with normal hepatic function, approximately 90% of
    a CAP dose was conjugated to the glucuronide in the liver, and excreted
    by the kidneys. Only about 5-15% was excreted by glomerular filtration
    as unchanged CAP in urine. Minor metabolites have also been identified.
    The mean elimination half-life was approximately 4 hours in adults and
    children, but can be 9-12 hours in newborns. In patients with hepatic
    dysfunction or renal insufficiency conjugation of CAP and elimination of
    glucuronide conjugates were slower. Renal impairment did not modify the
    elimination rate of the active and potentially toxic free drug (Goodman
    & Gilman, 1991).

         Several drug interactions have been reported involving enzyme
    induction by phenobarbital, phenytoin and rifampicin. Inhibition of CAP
    glucuronidation by paracetamol in neonates was suspected, but was later
    refuted in a large study (Paap & Nahata, 1991). No significant
    alteration of CAP half-life, area under the curve or peak concentration
    was observed in adult patients after paracetamol treatment (Stein  et
     al., 1989).

         A study performed in 225 children showed that large variations in
    CAP metabolism and rate of elimination can lead to under/overdosage,
    especially in neonates and young infants . The authors concluded that
    CAP concentrations should be monitored every 48 hours in babies under 1
    year of age (Rhodes & Henry, 1992)

         No new information was available on metabolites produced by
    intestinal bacteria, but it was stated that nitroreduction derivatives
    may play an important role in haematological toxicity. (See "Special
    studies on haematotoxical effects", sec. 2.2.2). There is no clear
    evidence that CAP nitroreduction takes place  in vivo in animals or
    humans without the presence of intestinal bacteria. Whether this
    metabolism occurs in the bone marrow, especially in susceptible
    subjects, or whether toxic metabolites may find their way to the bone
    marrow is also unknown at this time (Jimenez  et al., 1990).

         New data on CAP pharmacokinetics in food-producing animals were
    available. The authors concluded that meat and offal from treated
    animals contain CAP and non-genotoxic metabolites (Milhaud, 1993a).

         The plasma concentration of CAP was followed in 8 mature cats, after
    ocular application of 1% CAP ophthalmic ointment at eight-hour intervals
    for 21 days. It was estimated that the daily dose applied was 2.7
    mg/cat/day. On day 21 of treatment the plasma concentration of CAP in
    these treated cats was measured at 0.09 µg/ml (Conner & Gupta, 1973).

    2.2  Toxicological studies

    2.2.1  Special studies on genotoxicity

         The results of recent genotoxicity assays with CAP are summarized in
    Table 1.

         In a recent literature review of the genetic and genotoxic potential
    of CAP, the author concluded that reports of studies for many important
    genetic endpoints, for example, gene mutations or the induction of
    unscheduled DNA synthesis in mammalian cells could not be found despite
    widespread use of CAP in human medicine. The only consistent findings
    were of chromosomal effects on somatic cells. These effects may reflect
    the CAP-induced blockade of the cell cycle and of macromolecule
    synthesis (Rosenkranz, 1988).

         The results of a battery of genotoxicity studies with CAP in
    mammalian cell systems showed that at concentrations (2-4 mM) markedly
    higher than those (0.08-0.15 mM) reported in humans following exposure
    to a maximal therapeutic dose, CAP produced a minimal amount of DNA
    fragmentation in both V79 cells and metabolically competent rat cells. A
    level of DNA-repair synthesis indicative of a weak but positive response
    was detected in primary cultures of liver cells obtained from 2 of 3
    human donors, and a borderline degree was present in those prepared from
    rats. The promutagenic character of CAP-induced DNA lesions was
    confirmed by a low but significant increase in the frequency of
    6-thioguanine resistant clones of V79 cells, which were absent in the
    presence of co-cultured rat hepatocytes. Administration of a single oral

    dose of 1250 mg/kg bw CAP to rats (1/2 the LD50) failed to induce an
    increased incidence of either micronucleated hepatocytes or
    polychromatic erythrocytes. The authors concluded that CAP should be
    considered intrinsically capable of producing a very weak genotoxic
    effect, but only at concentrations about 25 times higher than those
    occurring in patients treated with the maximum therapeutic dose
    (Martelli  et al., 1991).

         Chromosomal aberrations and sister-chromatid exchanges (SCE) in
    human lymphocyte cultures treated with CAP, in bone marrow cells of
    treated mice, and in Chinese hamster cells (V79), were studied. No
    aberrations were induced by short-term treatment of human lymphocytes
    exposed in the G1 and G2 phases. A high frequency of aberrations,
    exclusively of the chromatid type, were induced during a whole cell
    cycle. Aberrant metaphases were detected at the end and a few hours
    after the end of treatment; at later times, aberrant cells reached
    control values. Doses producing chromosomal aberrations slightly
    increased SCEs both in human lymphocytes and in V79 cells. In mouse bone
    marrow cells, CAP induced a high mitotic delay and few structural
    aberrations. Intrachromosomal vacuoles were observed, which indicated a
    possible effect on chromosome condensation (Sbrana  et al., 1991).

         The genotoxic potential of different CAP concentrations (5, 20, 40,
    and 60 µg/ml) was investigated in bovine fibroblast primary lines by SCE
    assay. All doses caused a significant increase in the number of SCEs, in
    relation to the control dose (Arruga  et al., 1992). 

         Cytotoxic and genotoxic effects of CAP and six metabolites were
    investigated in human peripheral blood lymphocytes (PBL). Raji lymphoma
    cells were included in the study to assess the effectiveness of a nitro
    reduction step in the toxicological action of CAP and its derivatives.
    As a model of target cells involved in aplastic anaemia, an immortalized
    lymphoblastoid cell line originating from human bone marrow (RiBM) was
    used. The metabolites were nitroso-CAP, dehydro-CAP, dehydro-CAP base,
    CAP base, CAP glucuronide, and CAP alcohol. CAP and its metabolites were
    assayed in a large range of concentrations (1 x 10-5 M to 4.10-3 M)
    on the 3 cell types. The cytotoxic effect was determined by inhibition
    of 3H-thymidine incorporation in DNA. The genotoxic effect was
    evaluated by the induction of DNA single strand breaks detected by the
    method of alkaline elution.

         Three CAP metabolites, nitroso-CAP, dehydro-CAP, and dehydro-CAP
    base presented significant cytotoxic and genotoxic effects in the 3
    cellular models. CAP and the 3 metabolites were devoid of cytotoxic and
    genotoxic effect in all 3 models, up to a concentration of 1 x 10-3 M.
    At higher concentrations a cytotoxic effect was observed for CAP and
    CAP-base in human lymphocytes and RiBM cells, and for CAP only in Raji
    cells. No DNA damage could be detected with these compounds.

    Table 1. Results of genotoxicity assays on chloramphenicol

    Test system        Test object      Concentration Results   Reference

    DNA fragmentation  V79 cells        4mM           weak      Martelli  et al.,
                                                      positive  1991

    DNA fragmentation  Rat hepatocytes  2 mM          weak      Martelli  et
                                                      positive   al., 1991

    DNA repair assay   Cultured human   2 mM          weak      Martelli  et
                       and rat          positive                 al., 1991

    Mutation to TGr    V79 cells        2 mM          weak      Martelli  et
                                                      positive   al., 1991

    Micronucleus test  Rat liver bone   1250 mg/kg    negative  Martelli  et
                       marrow cells                              al., 1991

    Chromosomal        Human            2.4-4.8       positive1 Sbrana  et al.,
    aberrations        lymphocytes      mg/ml                   1991

    aberrations        Mouse bone       50 & 100      positive  Sbrana  et al.,
                       marrow cells     mg/kg                   1991

    SCE                Human            2.4-3.2       weak      Sbrana  et al.,
                       lymphocytes      mg/ml         positive  1991

    SCE                V79 cells        3-12 mg/ml    weak      Sbrana  et al.,
                                                      positive  1991

    SCE                Bovine           5-60 µg/ml    positive  Arruga  et al.,
                       fibroblasts                              1992

    DNA single stand   PBL2, Raji       >1 x 10-4M    positive4 Decloitre  et al.,
    breaks             lymphoma c.                              1993

    1    Negative in G1 and G2 phases.
    2    Human peripheral blood lymphocytes.
    3    Immortalized lymphoblastoid cell line originated from human bone
    4    Only for 3 CAP metabolites: nitroso-CAP, dehydro-CAP and
         dehydro-CAP base.
         The authors concluded that nitroso-CAP, dehydro-CAP, and dehydro-CAP
    base were able to induce a cytotoxic effect at concent-ration superior
    to 1 x 10-6 M and a genotoxic effect at 1 x 10-4 (no effect
    concentration 2 x 10-5) in human cells. Human bone marrow-derived
    cells were less sensitive than human peripheral lymphocytes and Raji
    cells (Decloitre  et al., 1993).

         In a literature review and analysis of  in vitro studies of the
    toxic effects of CAP and its metabolites the authors concluded that the
    essential problem is the evaluation of the lowest concentration capable
    of inducing genetic lesions. In their view, CAP has a very low or
    genotoxic capacity and, for those of its metabolites which are more
    active, 3 to 4 logs of concentration separate the possible genotoxic
    concentrations in serum from the concentrations that could result from
    the ingestion of residues in food of animal origin (Frayssinet  et al.,

    2.2.2  Special studies on haematological effects

         CAP was administered intravenously for 8-17 days to 5 newborn calves
    at a daily dosage of 100 mg/kg bw High levels of serum CAP were observed
    throughout the study, despite a marked increase in elimination rate seen
    with increasing age. Adverse effects included severe hypotension
    following rapid i.v. administration and severe gastrointestinal
    dysfunction. Minor haematological effects were observed in only one
    animal, which included a slight decrease in packed cell volume and
    haemoglobin without any related changes in the bone marrow. Based on
    these results the authors concluded that it was unlikely that
    haematological effects would be an important limiting factor in CAP use
    in cattle (Burrows  et al., 1988).

         CAP, tetracycline and gentamicin have been shown in vitro to have
    detrimental effects on bovine polymorphonuclear lymphocytes (PMNL). To
    evaluate their effects on bovine PMNL  in vivo, three Holstein dairy
    cows received intramammary infusions of one of these antibiotics into
    two quarters and an infusion of phosphate buffered saline into one of
    the remaining untreated quarters. A fourth cow received intramammary
    infusions of phosphate buffered saline solution (10 ml/treatment) only.
    Dosages administered were 5 g/infusion of CAP and 500 mg/infusion of
    gentamicin and tetracycline. Each dose was dissolved in 10 ml of
    phosphate buffered saline. Each cow was treated once a month for four
    consecutive months. CAP at a dose of 5 g per treatment failed to show
    stronger effects than tetracycline on the basis of PMNL lymphocyte
    morphologic features or phagocytosis ability (Paape et al., 1990a).

         The activity of CAP against bovine neutrophils was assessed in an
     in vitro model This activity was compared to florphenicol and
    thiamphenicol, two analogs of CAP in which the p-nitro group is replaced
    by a methyl sulfonyl group. All drugs altered neutrophil morphology

    (lack of pseudopodia), but only CAP depressed phagocytosis and
    completely blocked respiratory burst activity of neutrophils at
    concentrations of 2000 and 4000 µg/ml, but not at 10 µg/ml. At
    concentration of 4000 µg/ml CAP also blocked chemolumin-escence
    activity. (Paape  et al., 1990b).

         Sequence analysis of the 3' end of the 16S rRNA gene of
    mitochondrial DNA revealed a single base change (Guanine to Adenine) at
    position 3010, in a patient who had recovered from CAP-induced aplastic
    anaemia. A link between this mutation and CAP-induced aplastic anaemia
    was ruled out by investigating three other similar patients that died,
    none of whom had the mutation. The authors concluded that the relatively
    high frequency of this mutation in the population (14% in Europeans),
    made its role as a potential determinant of a person's susceptibility to
    CAP-induced aplastic anaemia unlikely (Mehta  et al., 1989).

         Burst-promoting activity (BPA) measured in the sera from 31 children
    with aplastic anaemia was significantly higher when compared with
    healthy children. BPA elevation is a biological compensation for the
    haematopoietic disorder, and the authors stated that its measurement in
    patients with aplastic anaemia may be helpful in evaluating the
    haematopoietic condition. However, none of the three patients suffering
    from CAP-induced aplastic anaemia exibited increased BPA (Wang  et al.,

         Three CAP metabolites known to be produced by intestinal flora,
    dehydro-CAP, nitroso-CAP, and nitrophenyl-CAP, are much more toxic to
    bone marrow  in vitro than CAP itself. The author concluded that the
    p-nitro group of CAP is the structural feature underlying aplastic
    anaemia. In predisposed subjects this group undergoes nitroreduction,
    leading to the production of toxic intermediates (nitroso,
    hydroxylamine), resulting in stem cell damage. Nitroso-CAP in
    macromolecular concentrations inhibits myeloid growth irreversibily and
    arrests cells in the G2 phase of the cycle. The fact that the parent
    compound thiamphenicol appears free of irreversible toxicity may be
    related to the lack of p-nitro group in the molecular structure (Yunis,

         Colony stimulating factors (CSFs) completely reversed the inhibitory
    effects of CAP on different cell strains (CFU-GM, KG-1, HL-60). In
    contrast, inhibition by dehydro-CAP and nitroso-CAP was not affected by
    CSFs and CSFs were inhibited by dehydro-CAP and nitroso-CAP, but not by
    CAP or nitrophenyl-CAP. The authors suggested that the dual
    toxic-inhibitory effect of some intestinal metabolites of CAP on
    haematopoietic growth and on CSF production render them very potent as
    potential mediators of CAP-induced aplastic anaemia (Jimenez et al.,

    2.3  Observations in humans

         The most serious adverse effect of CAP in humans are its ability to
    cause a reversible dose-related bone marrow suppression or serious,
    generally irreversible, aplastic anaemia, which is not considered to be
    dose-related. The aplasia usually develops after a latency period, and
    the affected people are considered to have some biochemical
    predisposition. The incidence of the disease varies significantly and is
    rather low, 1 in approximately 30 000 or more courses of therapy, but
    the fatality rate is high when bone-marrow aplasia is complete. There is
    a possible higher risk of acute leukemia in those who recover (Goodman &
    Gilman, 1992).

         Other manifestations of CAP toxicity have been reported. Prolonged
    oral administration of therapeutic doses of CAP may induce bleeding,
    either by bone marrow depression or inhibition of vitamin K synthesis
    due to the reduction of intestinal flora that produce vitamin K. At dose
    levels exceeding 25 mg/kg bw/day, "the grey syndrome" may occur in
    premature and newborn infants, and also in adults and older children
    given very high doses. The "grey syndrome" is characterized by abdominal
    distention, vomiting, ashen colour, hypothermia, progressive pallid
    cyanosis, irregular respiration, and circulatory collapse followed by
    death in a few hours or days (Martindale, 1989). Haemolytic anaemia has
    occurred in some patients with a genetic deficiency of
    glucose-6-phosphate dehydrogenase activity. In 20 children with
    glucose-6-phosphate dehydrogenase deficency who developed intravascular
    haemolysis, CAP was involved in 5 cases (Choudry  et al, 1990).

         In a study of 45 pediatric patients who received CAP for the
    treatment of central nervous system infections, anaemia occurred in 10,
    leukopenia and neutropenia in 4, and eosinophilia in 16. The mean
    cumulative dose of CAP ranged from 1.2 to 1.8 g/kg bw in patients with
    adverse effects in comparison with 0.9 to 1.1 g/kg bw in those without.
    This suggests that a high cumulative dose may be an important factor in
    predisposing some patients to reversible haematological toxicity of CAP
    (Nahata, 1989).

         Contact sensitivity to CAP is rare in humans, most cases being
    elicited by eye drops. The systemic administration of CAP is capable of
    eliciting a reaction at sites previously exposed and sensitized (Urratia
     et al., 1992).

         Documentation on the ophthalmic use of CAP provides no evidence that
    this route of administration is associated with the same toxicity risk
    as CAP administered parenterally. In a study on 300 patients with
    ophthalmic complaints treated topically with fusidic acid or CAP, the
    incidence of side-effects was similar in both groups (Kairys & Smith,

         A case of aplastic anaemia was reported in a patient treated with
    i.v. CAP and cimetidine. The patient died 19 days after the initiation
    of therapy. In an evaluation of nine cases of aplastic anaemia following
    parenteral administration of CAP, the interval from the start of the
    treatment to the onset of aplasia varied from 7 to 270 days. The
    interval between the initial dose and death ranged from 18 days to 4.8
    years. The authors concluded that the potential for aplastic anaemia
    must be considered whenever CAP is used, regardless of the route of
    administration (West  et al, 1988).

         A compilation of 576 cases of blood dyscrasia related to CAP
    administration showed that aplastic anaemia was the most common type
    reported, accounting for about 70% of the cases. The outcome was
    apparently unrelated to the dose of CAP taken. However, the longer the
    interval between the last dose and the appearance of first sign of the
    blood dyscrasia, the greater was the mortality rate; nearly all patients
    in whom this interval was longer than 2 months died (Goodman & Gillman,

         The incidence of aplastic anaemia is particularly low in Hong Kong,
    despite extensive use of CAP. Sales of CAP are between about 11 and
    440-fold greater in Hong Kong than in several western countries and
    Australia. In contrast, the certified death rate from aplastic anaemia
    in Hong Kong is only 0.4 per 1000 deaths compared with 1.0 per 1000 in
    England and Wales. The authors concluded that this low incidence may be
    due to either genetic factors, inter-ethnic differences in the flora of
    the gut, or relatively short treatment periods in Hong Kong (5 days in
    most cases) (Kumana  et al., 1988a).

         The literature on the occurrence of medullary hypoplasia associated
    with topical application of CAP in eye treatment is extremely limited,
    despite the prevalence of this particular use. Based upon the analysis
    of all known reported cases (4 cases reported from 1965-1982) the author
    concluded that, even if usual  per os doses of CAP could cause
    medullary aplasia and if the prolonged administration of lower doses was
    determined to be an added risk, an association between ocular use and
    occurrence of blood dyscrasias could not be proven on the basis of cases
    reported in literature (Ascari, 1989).

         In a critical study of the risk of developing acute bone marrow
    aplasia associated with the use of CAP, the author of the paper
    referenced in this paragraph concluded that CAP, even administered in
    small doses per os (1 g/day for 3-6 days) can cause bone marrow aplasia
    and induce as side effects, malignant myelo-dysplasia or paroxystic
    nocturnal haemoglobinuria. The repetition of treatments or the
    chronicity of small doses is an additional risk. However, the potential
    risk associated with small doses (of the eye drop type), cannot be
    determined because a test case has not been documented (Najean, 1988).

         During the past 30 years numerous epidemiological studies have been
    performed to assess the incidence of bone marrow aplasia in populations
    and to determine whether an association exists between the use of CAP
    and the etiology of aplastic anaemia. The author of the paper referenced
    in this paragraph concluded that in the past decade a remarkable decline
    in the reported incidence of aplastic anaemia occurred when careful
    diagnostic criteria have been applied to population surveys and the
    proper clinical data have been collected for statistical evaluation. In
    these studies, the incidence of CAP as an associated etiology has
    decreased to an unmeasurable level. The widespread use of topical ocular
    CAP throughout Europe in the past decade suggests that this mode of
    application is associated with minimal if no cases of aplastic anaemia
    (Gardner, 1990).

         A multicentric prospective study in France between 1984 and 1987
    recorded 250 cases of aplastic anaemia, with an annual incidence of 1.5
    per million inhabitants. This is similar to the incidence in other
    European regions, but lower than those published for the United States.
    The fatality rate was estimated to be 17% at 3 months and 34% at 1 year
    after diagnosis. As to the etiology of the disease 74% were declared
    idiopathic, 13% presumably associated to drug toxicity, and 5% related
    to hepatitis. The authors concluded that retrospective studies lead to a
    much higher incidence rate than prospective studies. In their view, the
    most accurate method to assess the incidence of aplastic anaemia in a
    well defined population is by prospective identification of cases with
    confirmation by follow-up (Mary  et al., 1990).

         The results of a large case-control study on the etiological factors
    of aplastic anaemia conducted in France between 1984 and 1988 failed to
    show any association between aplastic anaemia in humans and exposure to
    CAP through ophthalmic use. However, the authors stated that the
    frequency of the consumption of CAP made the quantification of its risk
    impossible to estimate accurately using case-control methods. For
    ophthalmic use of CAP, the ability of an epidemiological survey to
    accurately assess an association with aplastic anaemia is unclear
    (Baumelou  et al., 1993).

         In a report evaluating correlations between the incidence of
    aplastic anaemia in different countries and the available information
    about CAP use for medical, ophthalmic, or veterinary purposes, the
    authors observed that no relationship could be demonstrated between
    ophthalmic and veterinary uses of CAP and incidence of aplastic anaemia
    (Mary & Baumelou 1993).

         Based on the previous three reports and the status of knowledge on
    aplastic anaemia available from the literature, an independent expert
    concluded that there was no evidence that humans may be at risk of
    developing aplastic anaemia when exposed to doses of CAP associated with
    ophthalmic use or levels found as residues in food resulting from
    veterinary uses (Mary, 1993).

         A case report of acute myeloblastic leukemia associated with
    previous CAP use and the onset of malignancy two years later, apparently
    induced by oral CAP treatment (250 mg every six hours for 12 days), has
    been described (Shah, 1988). Two more cases from India were reported
    where the authors claimed that treatment with CAP in two children
    resulted in hypoplastic anaemia which later turned into acute leukaemia
    (Mitra, 1989). Another case of acute myeloid leukaemia with a rapid,
    fatal outcome was reported from India in a 65-year old man following
    treatment with 6 hourly doses of 500 mg of CAP for 8 days (Mappilacherry
    & Chandra, 1990). Similar observations available from China (Shu  et al.,
    1988) have been debated by other authors (Kumana  et al., 1988b),
    leading to inconclusive evidence. 

         The use of CAP is strictly regulated in different parts of the
    world. It is specifically prohibited in all food-producing animals in
    the USA and Australia (Page, 1991). There are no data to implicate the
    presence of residues of CAP in foods consumed by humans as a cause of
    aplastic anaemia (Woodward, 1991).


         Additional genotoxicity data together with new epidemiological data
    addressing the occurrence of aplastic anaemia in humans were available.
    In addition, the Committee re-evaluated previous data on CAP, which were
    summarized in the toxicological monograph published after the
    thirty-second meeting (Annex 1, reference 81). 

         Rapid and extensive absorption of CAP occurred after oral
    administration to laboratory animals and humans. It is distributed to
    all major organs and tissues. In contrast to oral absorption, the
    systemic uptake of the drug from ophthalmic application appeared to be
    poor. The major route of excretion in animals and humans is urinary (up
    to 90% of the administered dose, of which 15% is excreted as parent
    compound and the rest in the form of metabolites). A number of
    metabolites are formed, the major one being the glucuronide.

         Single i.v. doses of CAP were moderately toxic to mice (LD50
    1300-1800 mg/kg bw). There were no repeat-dose studies available to the

         The Committee concluded that adequate carcinogenicity studies were
    not available, and an evaluation report was not presented (Annex 1,
    reference 104, section 2.3). IARC came to the same conclusion in 1982,
    1987, and 1990 (IARC, 1990).

         The original genotoxicity studies and the new data suggested that
    CAP and its metabolites were genotoxic in a number of  in vitro test
    systems, and in an  in vivo study for chromosomal aberrations in mice.
    The only negative study was a rat micronucleus test.

         In rabbits given CAP orally at doses of 0, 500, 1000, or 2000 mg/kg
    bw/day over days 6-15, 6-9, or 8-11 of gestation, high incidences of
    fetal deaths occurred. There was no evidence of a teratogenic effect.
    The NOEL was 500 mg/kg bw/day. In a series of studies in the rat,
    embryolethality occurred even at the lowest dose tested, 500 mg/kg
    bw/day. When mice were given oral doses of 500-2000 mg/kg bw/day CAP,
    embryotoxicity occurred at all doses, but there was no evidence of a
    teratogenic effect.

         Adequate reproduction studies were not available, and an evaluation
    report was not presented to the Committee (Annex 1, reference 104,
    section 2.3).

         The original epidemiological evidence reviewed by the Committee
    suggested that treatment of humans with CAP was associated with the
    induction of blood dyscrasias, particularly aplastic anaemia. The
    Committee considered the new epidemiological data on aplastic anaemia,
    which showed that the total incidence was of the order of 1.5 cases per

    million people per year. Only about 15% of the total number of cases was
    associated with drug treatment and among these CAP was not a major
    contributor. These data gave an overall incidence of CAP-associated
    aplastic anaemia in humans of less than one case per 10 million per
    year. In considering epidemiological data derived from the ophthalmic
    use of CAP, the Committee concluded that systemic exposure from this
    form of treatment was not associated with the induction of aplastic
    anaemia. However, it was not possible to quantify the actual systemic
    exposure from ophthalmic use.

         The Committee noted the extremely low overall incidence of aplastic
    anaemia and the lack of association between the ophthalmic use of CAP
    and aplastic anaemia. It concluded that human exposure to CAP residues
    in food of the same order as exposure resulting from systemic uptake
    after ophthalmic use would not cause any demonst-rable alteration in the
    incidence of the disorder.

         Toxicological considerations overshadowed any level of concern for
    microbiological effects.


         The Committee was unable to establish an ADI for CAP due to the lack
    of information to assess its carcinogenicity and effects on reproduction
    and because the compound was genotoxic in a number of  in vitro and  in
     vivo test systems.


    ARRUGA, M.V., CATALAN, J. & MORENO, C. (1992). Effect of chloramphenicol
    on sister chromatide exchange in bovine fibroblasts.  Res. Vet. Sci.,
    52: 256-259.

    ASCARI, A. (1989). Literature analysis of the risk of medullary aplasia
    following topical applications of low doses of chloramphenicol.
    Unpublished report from Medical Clinic II University of Pavia. Submitted
    to WHO by Chloramphenicol Joint Scientific Program (CJSP),
    Magny-en-Vexin, France.

    BAUMELOU, E., GUIGUET, M. & MARY, J.Y. (1993). Epidemiology of aplastic
    anaemia in France: A case control study. I: Medical history and
    medication use. Blood (in press).

    BURROWS, G.E., TYLER, R.D., SANGHIA, S. & KEETON R.D. (1988).
    Experimental chloramphenicol intoxication in neonatal calves:
    intravenous administration.  Res. Vet. Sci., 45: 101-106.

    GHANI, R. (1990). Drug-induced haemolysis and renal failure in children
    with glucose-6-phosphate dehydrogenase deficiency in Afghanistan.  Ann.
     Tropic. Pediatr., 10: 335-338.

    CONNER, G. H., AND GUPTA, B.N. (1973). Bone marrow, blood and assay
    levels following medication of cats with chloramphenicol ophthalmic
    ointment.  Vet. Med. SAC, 885-899.

    CRAVEDI, J.P. & BARADAT, M. (1991). Comparative metabolic profiling of
    chloramphenicol by isolated hepatocytes from rat and trout
    ( Oncorhynchus mykiss ). Comp. Biochem. Physiol. 100C: 649-652.

    C. (1993). Study of cellular effects of chloramphenicol and its
    metabolites: cytotoxic, genotoxic effects and dose-response
    relationship. Unpublished report from Institut de Recherches
    Scientifiques sur le Cancer, Paris, France. Submitted to WHO by
    Chloramphenicol Joint Scientific Program (CJSP), 

    Literature review and analysis: Studies  in vitro of the toxic effects
    of chloramphenicol and its metabolites. Unpublished report from Institut
    de Recherches Scientifiques sur le Cancer, Paris, France. Submitted to
    WHO by Chloramphenicol Joint Scientific Program (CJSP), 

    GARDNER, F.H. (1990). The risk of using chloramphenicol as a possible
    cause of aplastic anaemia. Unpublished report from the Division of
    Haematology-Oncology, John Sealy Hospital, University of Texas, Austin,
    TX, USA. Submitted to WHO by Chloramphenicol Joint Scientific Program
    (CJSP), Magny-en-Vexin, France.

    GOODMAN AND GILMAN (1992). The pharmacological basis of therapeutics.
    8th edition, Pergamon press, New York, pp. 1125-1130.

    IARC (1990) IARC Monographs on the Evaluation of Carcinogenic Risks to
    Humans, Vol. 50, Chloramphenicol, Lyon, pp. 169-193.

    JAMES, M.R., BROGAN, R. & CAREW-McCOLL, M. (1991). A study to compare
    the use of fusidic acid viscous eye drops and chloramphenicol eye
    ointment in an accident and emergency department.  Arch. Emerg. Med.,
    8: 125-129.

    Interaction of chloramphenicol and metabolites with colony stimulating
    factors: possible role in chloramphenicol-induced bone marrow injury.
     Am. J. Med. Sci., 300: 350-355.

    KAIRYS, D. & SMITH, M.B. (1990). Topical ocular chloramphenicol:
    clinical pharmacology and toxicity in optometric practice. J. Am.
    Optometric Assoc., 61: 14-17.

    KUMANA, C.R., LI, K.Y. & CHAU P.Y. (1988a). Worldwide variation in
    chloramphenicol utilization: should it cause concern?  J. Clin.
     Pharmacol., 28: 1071-1075.

    KUMANA, C.R., CHAU, R.Y. & CHAN, T.K. (1988b). Chloramphenicol use and
    childhood leukaemia (letter).  Lancet, I: 476-477.

    MAPPILACHERRY, T.K. & CHANDRA, S. (1990). Chloramphenicol and
    preleukemia/leukemia.  J. Assoc. Phys. India, 38: 670-671.

    BRAMBILLA, G. (1991). Genotoxicity testing of chloramphenicol in rodent
    and human cells.  Mut. Res., 260: 65-72.

    MARTINDALE (1989). The extra pharmacopoeia. Antimicrobial agents:
    Chloramphenicol. 29th edition, pp. 186-192.

    MARY, J.Y., BAUMELOU, E. & GUIGUET, M. (1990). Epidemiology of aplastic
    anaemia in France: A prospective multicentric study. Blood, 75:

    MARY, J.Y. & BAUMELOU, E. (1993). Aplastic anaemia and chloramphenicol
    use: objective analysis of available data in different countries.
    Unpublished report from INSERM U263, Centre de Bioinformatique,
    Université Paris 7, France. Submitted to WHO by Chloramphenicol Joint
    Scientific Program (CJSP), 

    MARY, J.Y. (1993). Conclusion about epidemiological evidence of the risk
    of aplastic anaemia from low dose exposure to chloramphenicol in man.
    Unpublished report from INSERM U263, Centre de Bioinformatique,
    Université Paris 7, France. Submitted to WHO by Chloramphenicol Joint
    Scientific Program (CJSP),

    new genetic polymorphism in the 16S ribosomal RNA gene of human
    mitochondrial DNA.  Ann. Hum. Genet., 53: 303-310.

    MILHAUD, G. (1993a). Metabolic study discussion on chloramphenicol.
    Unpublished report from Dept. of Pharmacy & Toxicology, National
    Veterinary College, Maisons-Alfort, France. Submitted to WHO by
    Chloramphenicol Joint Scientific Program (CJSP),

    MILHAUD, G. (1993b). Overall conclusions of CJSP programme - Expert
    report. Unpublished report from Dept. of Pharmacy & Toxicology, National
    Veterinary College, Maisons-Alfort, France. Submitted to WHO by
    Chloramphenicol Joint Scientific Program (CJSP), 

    MITRA, A. (1989). Chloramphenicol and preleukemia/leukaemia (letter).
     J. Assoc. Phys. India, 37: 613-614.

    NAHATA, M.C. (1989). Lack of predictability of chloramphenicol toxicity
    in pediatric patients.  J. Clin. Pharm. Therap., 14: 297-303.

    NAJEAN, Y. (1988). Critical study of the risk of using chloramphenicol
    in small doses as possible cause of bone marrow aplasia. Unpublished
    report from Hopital Saint-Louis Université, Paris, France. Submitted to
    WHO by Chloramphenicol Joint Scientific Program (CJSP), Magny-en Vexin,

    PAAP, C.M. & NAHATA, M.C. (1990). Clinical pharmacokinetics of
    antibacterial drugs in neonates.  Clin. Pharmacokinet., 19: 280-318.

    PAAPE, M.J., NICKERSON, S.C. & ZIV, G. (1990a).  In vivo effects of
    chloramphenicol, tetracycline, and gentamicin on bovine neutrophil
    function and morphologic features.  Am. J. Vet. Res., 51: 1055-1061.

    PAAPE, M.J., MILLER, R.H. & ZIV, G. (1990b). Effects of florfenicol,
    chloramphenicol, and thiamphenicol on phagocytosis, chemiluminescsnce,
    and morphology of bovine polynuclear neutrophil leucocytes.  J. Dairy
     Sci., 73: 1734-1744.

    PAGE, S.W. (1991). Chloramphenicol: hazards of use and the current
    regulatory environment.  Australian Vet. J., 68: 1-2.

    RHODES, K.H. & HENRY, N.K. (1992). Antibiotic therapy for severe
    infections in infants and children.  Mayo Clin. Proc., 67: 59-68.

    ROSENKRANZ, H.S. (1988). Chloramphenicol: magic bullet or double-edge
    sword?  Mut. Res., 196: 1-16.

    SBRANA, I., CARETTO, S., RAINALDI, G. & LOPRIENO, N. (1991).  Mut. Res.,
    248: 145-153.

    SHAH, B.M. (1988). Chloramphenicol and preleukemia/leukemia.  J. Assoc.
     Phys. India, 36: 531.

    (1988). Chloramphenicol use and childhood leukemia (letter).  Lancet, 
    I: 476-477.

    STEIN, C.M., THORNHILL, D.P., NEILL, P. & NYAZEMA, N.Z. (1989). Lack of
    effect of paracetamol on the pharmacokinetics of chloramphenicol.  Br.
     J. Clin. Pharmac., 27: 262-264.

    DE CORRES L.F. (1992). Senzitization to chloramphenicol.  Contact
     Dermatitis, 26: 66-67.

    WAL, J-M., PELERAN, J-C., PERDU, E., RAO, D. & BORIES, G. (1988). New
    aspects of metabolic conjugation of chloramphenicol.  Drug Metab. Disp.,
    16: 635-639.

    WANG, W., SHANG, S. & TAN, H. (1990). The changes of BPE level in 31
    cases of children with aplastic anaemia and its clinical significance.
     Cell Tissue Kinet. 23, 391-400.

    WEST, B.C., DeVAULT, G.A., CLEMENT J.C. & WILLIAMS D.M. (1988). Aplastic
    anaemia associated with parenteral chloramphenicol: review of 10 cases,
    including the second case of possible increased risk with cimetidine.
     Rev. Infect. Dis., 10: 1048-1051.

    WOODWARD, K.M. (1991). Hypersensitivity in humans and exposure to
    veterinary drugs.  Vet. Hum. Toxicol., 33: 168-172.

    YUNIS, A.A. (1988). Chloramphenicol: relation of structure to activity
    and toxicity.  Ann. Rev. Toxicol., 28: 83-100.

    See Also:
       Toxicological Abbreviations
       Chloramphenicol (WHO Food Additives Series 53)
       Chloramphenicol (WHO Food Additives Series 23)
       Chloramphenicol (IARC Summary & Evaluation, Volume 50, 1990)