Cautionary Tales

You Are Who You Are, Not What You Eat (or when you ate it)

Case submitted by Joseph Cravero, MD
Commentary by Patricia Scherrer, MD

“A 3 year old male is having an LP and bone marrow biopsy performed in the procedural suite.  He has ALL and in the consolidation phase of his chemotherapy.  The history is significant only for the fact that he was not feeling well this morning according to his mother.  He is tired and somewhat pale.  Mother thought he might be a little nauseated, but he has difficulty expressing this and she thinks he is OK now.  He is NPO for solids for 12 hours and for clear liquids for 3 hours.  His counts are low, but his ANC is acceptable for the procedure and his platelets are 100K.  

The patient is given a 3mg/kg bolus of propofol after which he is unconscious and responds to painful stimulus.  He is given mask O2.  He receives another 1mg/kg bolus prior to the local anesthetic for the procedure.  He moves but is very controllable with holding.  As the LP is completed, the sedation provider notes bilious emesis in the mouth.  The patient is suctioned and emerges from sedation in about 10 minutes.  At this point when O2 is removed his O2 sats sink to about 85%.  His CXR shows a non-distinct infiltrate in the R lower lobe.  

Patient is observed for four hours and at which point he no longer requires oxygen.  He is admitted for overnight observation and does well.  He is discharged in the next morning.”

Fortunately, vomiting with aspiration of gastric contents is a rare complication associated with procedural sedation, occurring in 0.007% of patient encounters recorded in the Pediatric Sedation Research Consortium database.1  However, in considering the risk of vomiting and aspiration, important questions arise.  What is the association between vomiting/aspiration and NPO times?  How were the elective NPO times we all use in our clinical practice derived?  What other factors beyond NPO times are associated with vomiting and aspiration risk?  And, does every episode of vomiting with aspiration result in significant patient injury?

Dr. Cote’s pediatric anesthesiology text provides a superb overview of the history of the formulation of current NPO recommendations.2  Since the 1940s, patients have been fasted before sedation and anesthesia to minimize the risk of vomiting with pulmonary aspiration of gastric contents.  After a number of reports of maternal deaths during childbirth related to pulmonary aspiration during the induction of anesthesia in the 1950s, NPO times were evaluated more closely. 

The half-life to empty clear liquids from the stomach is approximately 15 minutes, as detailed in a 1954 report in the Journal of Physiology.  Clinical studies in children have demonstrated that a two hour fasting period for clear liquids was associated with no increased risk of aspiration/pneumonitis versus longer NPO times for clears.3  Although gastric emptying time for breast milk often mirrors clear liquids, breast milk can have variable (and at times quite high) fat content therefore delaying half-life emptying times to an average of 50 minutes versus clear liquids.4  Similarly, infant formula can have an emptying time half-life of 75 minutes or more. 

Based on this information, the American Society of Anesthesiologists formulated their preoperative fasting guidelines with two hours for clear liquids, four hours for breast milk, six hours for infant formula, and 8 hours for solids such as fatty or fried foods.5  Some centers allow dry toast up to six hours before induction.

What underlying factors are associated with aspiration risk in children during anesthesia for surgical procedures?  A study reported in Anesthesiology examined risk factors for pulmonary aspiration of gastric contents from 63,180 general anesthetic procedures performed in children <18 years of age at the Mayo Clinic between 1985 and 1997.6  Pulmonary aspiration occurred in 24 patients, or 1:2632 anesthetics.  Risk of aspiration was significantly increased in emergency procedures (1:373).  Of the children who aspirated, 15/24 had no sequelae.  Of the remaining nine patients, five required respiratory support including supplemental oxygen, and three required mechanical ventilation for >48 hours.  No child died from sequelae of pulmonary aspiration. 

The majority of children who aspirated had a bowel obstruction or ileus.  Nearly all gagged or coughed during airway manipulation or during induction of anesthesia.  The three children with significant sequelae included an infant with multiple GI malformations who aspirated during laryngoscopy, a two year old with a mesenteric tear and bowel ischemia who aspirated during rapid sequence induction, and a 16 year old with a full stomach, high EtOH level, and severe head trauma who also aspirated during laryngoscopy.  There was no significant morbidity in >55,000 ASA I and II patients undergoing elective procedures. 

A similar evaluation from the Karolinska Hospital from 1967 to 1983 noted that risk factors of neurologic or esophagogastric abnormality, emergency surgery, ASA-PS status ≥3, intestinal obstruction, increased intracranial pressure, increased abdominal pressure, obesity, and the experience of the anesthesiologist were all associated with increased incidence of aspiration, but most aspirations were not associated with significant sequelae.7

What is the data with regard to pediatric procedural sedation?  In the 2000s, several pediatric emergency medicine programs reported their experiences with NPO times and incidence of adverse events.  In 2001, of the 1014 children requiring procedural sedation in the emergency department at Boston Children’s, only 15 had procedure-associated emesis and there were no episodes of aspiration.8  Eleven of the 15 patients who vomited received ketamine as the primary sedative/analgesic for fracture reductions or laceration repairs.  Of these patients, only 44% were fully NPO as per the ASA elective guidelines, with an average fasting duration for solids in the non-NPO group of 5.2 hours.  NPO times were not correlated with the occurrence of adverse events. 

Similarly, Roback et al reported on a cohort of 2085 children requiring parenteral procedural sedation in the emergency department at the Children’s Hospital of Colorado between 1996 and 2003.9  More than 1/3 of these patients were fasted ≤4 hours prior to their procedure.  Vomiting occurred in 156 (7.5%), but there were no episodes of clinically apparent aspiration.  As in the previous study, approximately ¾ of these patients received ketamine as part of their sedative regimen.  There was no correlation between NPO times and risk of vomiting. 

However, neither of these studies was adequately powered to detect significant differences in rate of emesis with or without aspiration.  In contrast, the recent report from the PSRC reviewed 139,142 procedural sedation encounters recorded between 2007 and 2011.1  From this group, there were 10 episodes of aspiration, eight of which occurred in appropriately NPO patients and two in non-NPO patients.  There was no apparent association between NPO status and major complication, and “there were too few aspirations to perform a multivariate analysis for this outcome.” 

However, very few of the patients sedated were NPO <4 hours and almost none were NPO for <2 hours.  Details of the 10 patients who aspirated can be found in the report; several had underlying oncologic diagnoses, and four underwent bronchoscopy or endoscopy.  The two non-NPO patients were both NPO for solids, liquids, and clears for six hours.  It should be noted that it is unlikely that the PSRC database captured grossly violated NPO times such as a full meal <2 hours prior to sedation, since most providers would either postpone the sedation or would transfer the child’s care to an anesthesiology colleague for more definitive airway protection.10   

So, in summary, aspiration is a very rare event in the context of pediatric procedural sedation and its occurrence does not seem to correlate with the timing of NPO.  Fortunately, most aspiration episodes do not lead to significant sequelae or long standing injury.  It can occur in spite of appropriate NPO times; however, we have little to no data on children who have REALLY full stomachs, and therefore we should not conclude that NPO guidelines are irrelevant to procedural sedation care.  The vast majority of reported aspiration events occur in children with underlying problems that put them at risk.  For our patients, who they are at baseline seems to matter more to the risk of aspiration than what they ate or when they ate it.  To quote Dr. Cravero, patient pathology trumps NPO time – every time!


  1. Beach ML et al.  Major adverse events and relationship to Nil per os status in pediatric sedation/anesthesia outside the operating room.  Anesthesiology 2015; 124:80-88.
  2. Ghazal EA, Mason LJ, Cote CJ.  Perioperative evaluation, premedication, and induction of anesthesia.  In:  Cote CJ, ed.  A Practice of Anesthesia for Infants and Children.  Philadelphia: Elsevier Saunders; 2013: 31-63.
  3. Brady M et al.  Preoperative fasting for preventing perioperative complications in children.  Cochrane Database Syst Rev 2005; CD005285.
  4. Litman RS et al.  Gastric volume and pH in infants fed clear liquids and breast milk prior to surgery.  Anesth Analg 1994; 79:482-485.
  5. Warner MA et al.  Practice guidelines for preoperative fasting and the use of pharmacologic agents to reduce the risk of pulmonary aspiration: application to healthy patients undergoing elective procedures.  A report by the American Society of Anesthesiologists Task Force on Preoperative Fasting.  Anesthesiology 1999; 90:896-905.
  6. Warner MA et al.  Perioperative pulmonary aspiration in infants and children.  Anesthesiology 1999; 90:66-71.
  7. Olsson GL et al.  Aspiration during anesthesia: a computer-aided study of 185,358 anaesthetics.  Acta Anaesthesiol Scand 1986; 30:84-92.
  8. Agrawal D et al.  Preprocedural fasting state and adverse events in children undergoing procedural sedation and analgesia in a pediatric emergency department.  Ann Emerg Med 2003; 42:636-646.
  9. Roback MG et al.  Preprocedural fasting and adverse events in procedural sedation and analgesia in a pediatric emergency department: are they related?  Ann Emerg Med 2004; 44:454-459.
  10. Cravero JP, Kamat PP.  Complications of procedural sedation.  In: Tobias JD, ed.  Procedural Sedation for Infants, Children, and Adolescents.  Elk Grove Village, IL: American Academy of Pediatrics; 2016: 115-135.

Back to top