Chloramphenicol is a bacteriostatic man-made antibiotic that was discovered in 1947. Initially designed for the treatment of typhoid fever, it has fallen out of favor due to the ubiquity of antibiotic-resistant Salmonella typhi. It was also historically used for the empiric treatment of pediatric patients presenting with petechial rash and fever for its excellent coverage of meningococcal sepsis and rickettsial disease. Due to its low-cost, wide spectrum of coverage, and low incidence of toxicity, chloramphenicol has been added to the World Health Organization’s List of Essential Medicines, and the growing problem of antimicrobial resistance to current broad-spectrum antibiotics has brought back interest in its use worldwide. Twelve years after its discovery, the first case report of a potentially fatal adverse reaction to chloramphenicol was discovered in neonates, with a predilection towards preterm infants. Neonates born at less than 37 weeks gestation were given chloramphenicol in an intravenous or oral formulation within two days of birth when they began to develop abdominal distention, vomiting, hypothermia, cyanosis, and cardiovascular instability. Vasomotor collapse resulting in mottling of skin and eventual ashen-gray skin discoloration led to the naming of this reaction as "gray-baby syndrome."
Elevated levels of chloramphenicol circulating in the plasma result from two distinct pathophysiologic processes. A normally functioning liver will metabolize the chloramphenicol parent molecule (primarily by glucuronidation). The immature neonatal liver is unable to synthesize and recycle the UDP-glucuronyltransferase enzyme efficiently. Similarly, the neonatal kidneys are unable to excrete chloramphenicol and its metabolites efficiently. These two deficiencies result in elevated serum levels of chloramphenicol. The chloramphenicol molecule displaces unconjugated bilirubin from albumin, leading to kernicterus and eventually death if untreated.
Premature infants and neonates are at the highest risk of the gray-baby syndrome from chloramphenicol exposure due to their decreased hepatic and renal function. Case reports of chloramphenicol toxicity have also been reported in children and adolescents. Various weight-based dosage adjustment has been suggested for newborns younger than 15 days, infants between 2 to 4 weeks days, and children older than one month.
Gray-baby syndrome usually begins between 2 to 9 days after chloramphenicol therapy is initiated. The relative hepatic and renal dysfunction in neonates (especially in premature infants) result in elevated serum levels of chloramphenicol. This impairs electron transport within the mitochondrial and consequently cellular respiration, leading to direct cellular toxicity. The chloramphenicol parent molecule also displaces unconjugated bilirubin from albumin, giving way to kernicterus and eventually death or permanent neurological sequelae if left untreated.
Serum concentrations after a single oral or intravenous dose of chloramphenicol peak 1 to 2 hours after ingestion; chloramphenicol has excellent absorption in the gastrointestinal (GI) tract. Intramuscular chloramphenicol has variable absorption with serum concentrations reaching only 5% to 65% the concentration of the equivalent intravenous or oral dose. Roughly half of serum chloramphenicol is bound to albumin and other plasma proteins. Elimination happens primarily in the liver through O-glucuronidation, which puts neonates with immature hepatic metabolism at risk for the gray-baby syndrome. This syndrome has been seen in patients who were given doses greater than 200 mg daily. Urinary excretion of the parent chloramphenicol compound is approximately 20% in children and 10% to 12% in adults; the rest is excreted as the glucuronidated metabolite.
The presentation of the gray-baby syndrome will vary depending on the level of toxicity from chloramphenicol. Ideally, chloramphenicol exposure will be provided in the history from the caregiver. Poor feeding, fussiness, and vomiting are often elicited in the history. Exposure from maternal use has also been observed. Chloramphenicol has been assigned Pregnancy Category C (risk not ruled out) and is contraindicated during lactation (as it passes readily into breast milk). A physical exam may reveal altered mental status from lethargy to obtundation, ashen-gray cyanosis, pallor, and abdominal distention/tenderness.
In the undifferentiated sick neonate presenting with cyanosis, a broad differential diagnosis must be considered, including but not limited to neonatal sepsis, non-accidental trauma, midgut volvulus, congenital heart disease, and inborn errors of metabolism. Blood work should include glucose, complete blood count with differential, complete metabolic panel, blood gas analysis, serum ammonia, serum lactic acid, serum ketones, and consideration for cardiac biomarkers including troponin and brain natriuretic peptide. In the setting of chloramphenicol toxicity, serum chloramphenicol levels may be drawn. Radiologic studies should include chest and abdominal films, and CT head or abdominal ultrasound (depending on the history). An electrocardiogram should also be obtained.
Management of chloramphenicol toxicity centers primarily around supportive care. The general approach to the ashen-gray hemodynamically unstable neonate starts with aggressive resuscitation and an early call to the pediatric intensive care unit or extracorporeal life support team, as some of these patients may be ideal candidates. These patients should be hemodynamically stabilized, appropriately oxygenated and ventilated, and intubated early. Checking a core temperature is critical as hypothermia is common in the gray neonate. Aggressive rewarming should be considered. A point-of-care glucose should also be checked, and hypoglycemia should be reversed if present. The differential diagnosis for an ashen-gray, cyanotic neonate should include chloramphenicol toxicity, congenital heart disease, adrenal insufficiency/hypothyroidism, inborn errors of metabolism, trauma, seizures, and of course, sepsis. Empiric administration of broad-spectrum antibiotics such as vancomycin, ampicillin (targeting Listeria), and a third-generation cephalosporin such as ceftriaxone or cefotaxime is recommended. Additional consideration should also be given to empiric prostaglandin administration in gray/cyanotic neonates, especially if a duct-dependent congenital cardiac lesion is present.
Modalities that have been used for the treatment of gray-baby syndrome are primarily aimed towards direct removal of the parent chloramphenicol molecule. This has been achieved through charcoal hemoperfusion and exchange transfusion. There have also been reports of phenobarbital being used for induction of the UDP-glucuronyltransferase enzyme. Consideration for cardiopulmonary bypass including extracorporeal membrane oxygenation may also be considered.
If the condition is diagnosed immediately and the drug is discontinued the prognosis is good but if the infant has developed organ dysfunction, the prognosis is guarded.
Once Gray baby syndrome has been diagnosed, consultation with a pediatrician and an infectious disease expert is recommended.
Patients with gray-baby syndrome should be admitted to a closely monitored telemetry setting, ideally in the intensive care unit. Changes in the dosing regimen (lower doses given at longer intervals) and careful monitoring of drug levels can help prevent this fatal complication of chloramphenicol administration in neonates and premature infants.
Gray baby syndrome is a preventable problem. With the number of antibiotics available today, there should be no valid reason to use this agent for the management of infection in babies. All healthcare workers should be aware of the serious complications of this antibiotic and the onus is on them not to start treatment unless there is no other choice. The pharmacist should fully question the prescription of this antibiotic for neonates and infants and should recommend other agents. Data suggest that when the drug is discontinued immediately after initiation of therapy, then the outcomes are good. But once the syndrome has developed, the prognosis is guarded. Deaths were once commonly reported in neonates when started on chloramphenicol.