Initiation of chloramphenicol therapy in the newborn infant

Parenteral antibiotics for the treatment of serious neonatal bacterial infections in developing country settings

BACKGROUND

A number of special issues must be considered when selecting simple, safe, inexpensive, and effective antimicrobial regimens for treatment of neonatal sepsis in developing country community settings.

METHODS

We reviewed available data regarding pharmacologic profiles of parenteral antibiotics with specific attention to properties relevant to their use in the treatment of neonatal infections in developing country communities.

RESULTS

For community-based management of neonatal infections, particularly attractive properties include efficacy and safety of extended-interval, intramuscular dosing regimens. The penicillins and cephalosporins have relatively favorable efficacy and safety profiles. Although the aminoglycosides have narrow therapeutic indices, when used appropriately, they are safe and effective. Although inexpensive and effective, the potential for significant life-threatening toxicity among neonates associated with chloramphenicol makes it the least preferred of the parenteral agents for empiric therapy.

CONCLUSIONS

The preferred parenteral regimens for community and first-level facility use are a combination of procaine penicillin G and gentamicin, or ceftriaxone given alone, which are safe and retain efficacy when dosed at extended intervals (> or =24 hours) by intramuscular administration.

Antibiotics in neonatal infections: a review

The bacteria most commonly responsible for early-onset (materno-fetal) infections in neonates are group B streptococci, enterococci, Enterobacteriaceae and Listeria monocytogenes. Coagulase-negative staphylococci, particularly Staphylococcus epidermidis, are the main pathogens in late-onset (nosocomial) infections, especially in high-risk patients such as those with very low birthweight, umbilical or central venous catheters or undergoing prolonged ventilation. The primary objective of the paediatrician is to identity all potential cases of bacterial disease quickly and begin antibacterial treatment immediately after the appropriate cultures have been obtained. Combination therapy is recommended for initial empirical treatment in the neonate. In early-onset infections, an effective first-line empirical therapy is ampicillin plus an aminoglycoside (duration of treatment 10 days). An alternative is ampicillin plus a third-generation cephalosporin such as cefotaxime, a combination particularly useful in neonatal meningitis (mean duration of treatment 14 to 21 days), in patients at risk of nephrotoxicity and/or when therapeutic monitoring of aminoglycosides is not possible. Another potential substitute for the aminoglycoside is aztreonam. Triple combination therapy (such as amoxicillin plus cefotaxime and an aminoglycoside) could also be used for the first 2 to 3 days of life, followed by dual therapy after the microbiological results. In late-onset infections the combination oxacillin plus an aminoglycoside is widely recommended. However, vancomycin plus ceftazidime (+/- an aminoglycoside for the first 2 to 3 days) may be a better choice. Teicoplanin may be a substitute for vancomycin. However, the initial approach should always be modified by knowledge of the local bacterial epidemiology. After the microbiological results, treatment should be switched to narrower spectrum agents if a specific organism has been identified, and should be discontinued if cultures are negative and the neonate is in good clinical condition. Penicillins and third-generation cephalosporins are generally well tolerated in neonates. There is controversy regarding whether therapeutic drug monitoring of aminoglycosides will decrease toxicity (particularly renal damage) in neonates, and on the efficacy and safety of a single daily dose versus multiple daily doses of these drugs. Toxic effects caused by vancomycin are uncommon, but debate still exists over the need for therapeutic drug monitoring of this agent. When antibacterials are used in neonates, accurate determination of dosage is required, particularly for compounds with a low therapeutic index and in patients with renal failure. Very low birthweight infants are also particularly prone to antibacterial-induced toxicity.

Chloramphenicol in children: dose, plasma levels and clinical effects

Despite concerns about adverse effects, chloramphenicol (CMC) continues to be used in certain situations and, due to its low therapeutic index and variable pharmacokinetics, therapeutic drug monitoring (TDM) is often recommended. At our centre, CMC finds applications in typhoid and meningitis and TDM is routinely performed. Elsewhere in Malaysia, however, CMC is used without TDM. We therefore decided to evaluate our TDM for CMC in relation to its roles in CMC therapy in children, who constitute most of our patients. Our objective was also to develop strategies to improve our TDM for CMC use. Data were collected from 168 children given CMC for various indications and monitored by the TDM service. Plasma CMC was determined by HPLC and used to adjust doses to maintain concentrations within a range of 10-25 micrograms/ml. Outcomes measured included daily temperatures and haematological indices. Daily doses and plasma CMC varied greatly. Doses averaged 40.5 mg/kg for neonates and 75.5 for older children. Average peak concentrations were therapeutic in 60% and trough in 42%. Average duration of fever was 6.3 days and it was unaffected by plasma CMC. Typhoid was eradicated in 97% but nine children with other diagnoses died. Side-effects were confined to mild reversible haematological abnormalities which developed in 11% of children at plasma concentrations which tended to be high. We conclude that CMC remains useful in children with typhoid. Its use for other indications, however, should be reviewed. Routine TDM for CMC is probably not warranted, at least until a clearer role is defined by well designed prospective studies.

Determinants of drug disposition in man

This article reviews some of the important determinants of variation in drug disposition such as age, gender, body weight, diet, environmental influences, drug - protein interactions, compliance, drug - drug interactions, endogenous substances, disease states, circadian variation and genetics.

Use of chloramphenicol palmitate in neonates

The absorption and disposition of orally administered chloramphenicol palmitate (chloramphenicol-P) was studied in seven neonates (four preterm, three term). The highest measured chloramphenicol serum concentrations occurred greater than or equal to 4 hours after the dose, and ranged from 5.5 to 23 micrograms/ml after doses of chloramphenicol-P 50 mg/kg/day orally. The dosage had to be increased in all preterm neonates from 25 mg/kg/day to 50 mg/kg/day to obtain adequate serum levels during therapy. In four neonates the apparent half-life could not be estimated, because there was no decline in serum concentrations. The apparent half-life was 3 and 6 hours, respectively, in two neonates in whom the serum concentration declined during the dosing interval. Urinary excretion of chloramphenicol and the glucoronide ester in three neonates varied from 24% to 55% of the total dose administered. These preliminary data suggest considerable variability in serum chloramphenicol levels when chloramphenicol-P is administered orally in neonates. The delay in achieving the maximum serum concentration, nondeclining serum curve, and low renal recovery is indicative of incomplete, prolonged, and erratic absorption, possibly related to delayed gastric emptying or decreased intraluminal hydrolysis of the palmitate ester.

Chloramphenicol toxicity in neonates: its incidence and prevention

The incidence of dose related chloramphenicol toxicity was determined in 64 neonates from 12 hospitals. Ten of the 64 exhibited symptoms attributed clinically to chloramphenicol toxicity. Nine received the dose prescribed and one an overdose. Symptoms of the grey baby syndrome were observed in five of the 10 babies; four babies suffered reversible haematological reactions; and one baby was described as very grey. Peak serum chloramphenicol concentrations in these 10 babies ranged from 28 to 180 mg/l and trough concentrations from 19 to 47 mg/l. Serum chloramphenicol concentrations above the therapeutic range (15-25 mg/l) were observed in a further 27 neonates (two had received a 10-fold overdose), none of whom showed signs of toxicity. Serious toxicity was associated with either prescription of dosages greater than that recommended or overdosage of chloramphenicol. High concentrations in young neonates may be avoided by prescribing and giving the recommended dose and then careful monitoring; concentrations should be maintained between 15 and 25 mg/l. No babies with concentrations within this range showed clinical signs of toxicity.