Anesthesia Safety for Mothers and Children: What does the FDA warning mean?

  • Agarwal, Rita, MD, FAAP, FASA
| Jan 23, 2017

Agarwal-RitaThe FDA issued a “Drug Safety Communication” on December 14, warning that repeated or lengthy use of general anesthetic and sedation drugs during surgeries or procedures in children younger than three years, or in pregnant women during their third trimester, “may affect the development of children’s brains.” The agency is requiring warning labels on all anesthetic agents and sedatives, including propofol, midazolam and all volatile anesthetic agents.

This warning is based on more than a decade’s worth of overwhelming and abundant animal data, and on conflicting data in humans. It has caused an uproar in the pediatric and obstetric communities. Many of us are scrambling to figure out what this means to our practices, and to the children and pregnant patients we care for. All sedatives and anesthetics have been implicated, although the FDA goes on to say that one brief anesthetic does not seem to be associated with developmental problems in children.

What is the evidence?

evidenceThere are plentiful and compelling animal data showing increased neuronal cell death, apoptosis, and development problems with learning and memory in all types of young animals, including non-human primates, who are exposed to anesthetics. Ketamine, nitrous oxide, diazepam, propofol, midazolam, lorazepam and volatile agents have been studied and implicated. Opioids do not seem to cause apoptosis, and dexmedetomidine may not either.

The animal studies show that cell death and its consequences are more pronounced at certain developmental and neuronal ages. For example, anesthesia on postnatal day seven in mice has a more profound effect than anesthetics at other times. This time period correlates with the period of maximal synaptogenesis in mice.

Many theories exist as to how or why anesthetics disrupt normal cell growth and development. It is possible that interference with normal inhibitory and excitatory input to the nervous system may be responsible, or there may be alterations in connectivity of synapses and white matter tracts. Animals of a developmental age that is outside the critical time frame may still have cell death, but do not seem to develop learning issues.

How do these data relate to humans?

Our period of maximum synaptogenesis is much longer (mid 2nd trimester to three years of age). Most children or pregnant women are receiving anesthesia or sedation for a surgical operation or other procedure. These patients NEED their procedures. The neurocognitive effects of surgery with and without anesthesia or sedative medications have not been well studied in animals for humane reasons.

humansUntreated pain in both adult and juvenile animals lead to increases in stress, inflammatory responses and a host of other complications. Human babies typically receive lavish attention and affection after a procedure, unlike animals who may be left alone to recover. Human growth and behavior are complex; siblings in the same family may have vastly differing degrees of achievement, learning difficulties and education. How do we evaluate the effect of one event amongst the many that make up early childhood?

Not surprisingly, the results of studies in children are mixed. Most are retrospective reviews of large databases. Most of these databases were collected for reasons unrelated to the anesthesia or surgery, but contain enough information that results can be extrapolated.

The only prospective study is the “GAS” study (general anesthesia vs. spinal), led by Andrew Davidson, MBBS, and Mary Ellen McCann MD, MPH, which compares the effects of spinal/caudal anesthesia versus general anesthesia in infants undergoing inguinal hernia repair. GAS has published early data that show no differences at age two, but definitive results won’t be available until the children are older.  The Pediatric Anesthesia Neurodevelopment Assessment (PANDA), led by Lena Sun, MD, shows similar results.

Children exposed to one brief (less than 80-minute) anesthetic for a simple operation do not appear to be at increased risk of neurocognitive problems. “SMARTTOTS”, a private-public partnership between the FDA and the International Anesthesia Research Society (IARS), was developed to help answer questions about anesthesia safety in children. Abundant research is underway to help define the problem (if it exists), and to investigate potentially less neurotoxic and/or more protective drugs.

What’s next?

nextThe ASA, the American Academy of Pediatrics’ Section on Anesthesiology (AAP SOA) and Committee on Drugs (COD), and the Society for Pediatric Anesthesia (SPA) responded to the recent ruling, cautioning parents and surgeons not to put off needed surgery, but to reconsider the risk-benefit ratio given the current concerns. The American College of Obstetricians and Gynecologists (ACOG) has issued its own statement, citing the lack of evidence and of any studies showing that babies exposed in utero to an anesthetic may be at increased risk of neurocognitive problems.

One of the FDA’s intentions with this ruling was to alert other physicians and health care professionals that all sedative medications have been implicated in animal studies. We don’t know if a brief exposure to ketamine or midazolam in the emergency department is either more or less toxic than a brief, well-conducted anesthetic in the OR.

Combinations of medications may be more problematic, but are often unavoidable. Repeated anesthetics or exposure to sedative medications and prolonged anesthetics are usually reserved for young children with significant medical problems, and the impact of the underlying medical comorbidities cannot be easily separated from the effects of the medications.

The findings cited in the warning are not new. They have been discussed by previous FDA advisory committees since 2007. However, not all practitioners using these medications for sedation or anesthesia in children are aware of these findings, and many incorrectly think that just “general anesthesia” is the problem. The AAP is leading a coordinated response to create information for families and health care professionals to reduce confusion and fear. There are many resources available through the SMARTTOTS, AAP, and SPA websites, for those interested in exploring the issues. In the meantime, avoid hypotension and hypoxia (as though we need this advice!), and consider the urgency and need for each operation or procedure requiring sedation or anesthesia in children under three.

Suggested Reading

  1. Patel P, Sun L. Update on Neonatal Anesthetic Neurotoxicity: Insights into Molecular Mechanism and Relevance to Humans.  Anesthesiology 4 2009, Vol.110, 703-708.
  2. Psaty BM, Platt R, Altman RS. Neurotoxicity of Generic Anesthesia Agents in Infants and Children. JAMA 313, 2015, Vol. 15, 1515-1516.
  3. Sall J. Anesthesia Kills Brain Cells, But What Does That Mean? Anesthesiology 12 2016, Vol.125, 1090-1091.
  4. AAP response to the FDA Drug Safety Communication,
  5. FDA Drug Safety Communication,


Leave a comment
  1. Rita Agarwal | Jan 24, 2017
    There are 2 even more recent studies that are showing no effects from either one or repeated anesthesia in children < 2. It underscores the complexity of human neurodevelopment.
  2. George Gregory | Jan 23, 2017
    Rita is correct.  We do not have enough good data in humans.  It is appropriate to be cautious, but we should not harm children in process by not doing needed surgery.

    Leave a comment