Pediatric Sedation Newsletter
(A newsletter for all who provide sedation to children)
Departments of Anesthesiology and Pediatrics, Dartmouth Hitchcock Medical Center,
Editors: Joseph P. Cravero, MD (firstname.lastname@example.org)
George T. Blike, MD (email@example.com )
Circulation = 1887
Our apologies to all who were looking for a March newsletter. Personal and professional
deadlines conspired to make this a bimonthly newsletter – at least this time. In fact we
will evaluate your response to this decreased frequency and decide if this timing is
actually preferable to a monthly mailing. We begin this month’s issue with some
comments on issues brought up by our readers followed by a review of the current status
of propofol sedation in the ICU. We conclude with review of a sedation overdose case.
Letters: Chloral Hydrate and Remifentanil
We receive many questions and comments from readers – two of which we would like to
put out for our readership to comment on.
1. An ED physician from Chicago states that his anesthesiology department is trying
to push through a requirement for IV access mandatory whenever Chloral Hydrate
sedation is given. Their current practice has been to give oral Chloral Hydrate and
monitor the patient with a pulse oximeter and a skilled observer. He asks if we
know of any guidelines that mandate IV access for this type of sedation.
This is a really interesting question that is not as easy to answer as we initially thought.
The ASA guidelines for sedation can be accessed through the ASA website without a
password and the AAP guidelines quoted below can be found on the AAP website. The
ASA guidelines for sedation by non-anesthesiologists are somewhat ambiguous on the
requirement for IV access. I have quoted the “recommendations” section which (to our
reading) really leaves the decision up to the provider:
“In patients receiving intravenous medications for sedation/analgesia, vascular access
should be maintained throughout the procedure and until the patient is no longer at risk
for cardiorespiratory depression. In patients who have received sedation/analgesia by
nonintravenous routes or whose intravenous line has become dislodged or blocked,
practitioners should determine the advisability of establishing or reestablishing
intravenous access on a case-by-case basis. In all instances, an individual with the skills
to establish intravenous access should be immediately available.”
The AAP guidelines (which are due to be revised soon) currently separate patients into
“conscious” and “deep” sedation groups. The description of conscious sedation is
“Conscious sedation is a medically controlled state of depressed consciousness that (1)
allows protective reflexes to be maintained; (2) retains the patient's ability to maintain a
patent airway independently and continuously; and (3) permits appropriate response by
the patient to physical stimulation and/or verbal command, eg, "open your eyes." A
minimally depressed level of consciousness should be used for the very young or
handicapped child incapable of the usually expected verbal responses.”
Deep sedation is also described:
“Deep sedation is a medically controlled state of depressed consciousness or
unconsciousness from which the patient is not easily aroused. Deep sedation may be
accompanied by a partial or complete loss of protective reflexes, including the inability
to maintain a patent airway independently and to respond purposefully to physical
stimulation or to verbal command. The state and risks of deep sedation may be
indistinguishable from those of general anesthesia.”
In our experience patients receiving 75mg/kg of Chloral Hydrate do not neatly fit into
either of these categories. In any case, the AAP guidelines go on to advise:
“Patients receiving deep sedation should have an intravenous line in place or have
immediately available a person skilled in establishing vascular access in pediatric
patients.” There is no mention of IV access for conscious sedation patients.
We would answer this reader’s question by saying there is no clear requirement (in either
of these guidelines) for IV access in oral chloral hydrate sedation cases. Without doing
any survey we would hazard a guess that 98% of institutions that provide oral chloral
hydrate sedation in this country do not require IV access. In spite of this, there is still a
need to evaluate each case – in those patients where a very deep level of sedation is
achieved by this dose of chloral hydrate, the option for ready IV access should be
available. In point of fact, it is probably much more important to assure the immediate
availability of an appropriate oral airway, mask, and bag for ventilation than the IV
Finally, in examining the practices at our own institution, we have found it interesting
that many or our CT scan patients (who receive oral Chloral Hydrate) require IV’s for
contrast instillation – yet our radiology department does not use that route for sedation
administration because of the protocols in our institution.
Your comments on chloral hydrate and specific protocols concerning IV’s or discharge
criteria are more than welcome.
2. A reader asked how we use the combination of propofol and remifentanyl for
sedation/analgesia. Specifically he wanted to know how we mix it up – doses we
use etc. We realize that this is a question that will be of interest primarily to
anesthesiologists, but other providers may be interested in knowing how we use
these drugs. We are interested in hearing comments from those who use the
For those not familiar with remifentanil, it is an ultra short acting opioid that has been
aggressively marketed for use as part of a balanced anesthetic technique and for selected
sedation cases. It has been shown to be an effective agent in many situations although
published data on pediatric sedation is very slim at this point. The use of this agent varies
greatly from one institution to another. (The manufacturer states that sales are steadily
increasing). Problems with the drug include the fact that it must be diluted in a fairly
complex manner prior to use, and it is relatively expensive. The short half-life of the drug
is excellent for bringing patients back to their baseline consciousness quickly, but also
leaves them without effective pain control if a major procedure has been performed.
Like most anesthesiologists who frequently give sedation to children, we have
historically used a lot of propofol. We (like most others) find that for non-painful
procedures this drug is wonderful – allowing a controlled state of sedation and resulting
in quick emergence with minimal side effects. Unfortunately, for procedures that involve
significant painful stimuli, we have found that a relatively large dose of propofol is
required to gain movement control sufficient to satisfy most operators and families. This
large dose leads to significantly prolonged recovery times in some cases. In short, since
propofol is primarily a sedative it is not necessarily ideal for very short but painful
We are not alone - over the years practitioners who use propofol have added
analgesics to this sedative to “smooth out” sedation/anesthesia when pain is involved.
(There are many papers detailing the use of ketamine with propofol for instance).
Remifentanil has been proposed as a logical complement to propofol because it has a
short half-life and can be titrated in an infusion like propofol.
Here is the controversial part: How does one add remifentanil to a propofol sedation?
Are two separate infusions required? Can one add remifentanil directly to the propofol
and use it as one infusion? Of course this is not an FDA approved usage of the drug and
there is some emerging data on how the drugs separate out over time etc. Still, as is often
the case with new drugs and combinations, many practitioners are trying different
mixtures of propofol and remifentanil in advance of any really good studies to tell us
what works. We have had various providers from all over the country suggest different
fixed combinations of these drugs for sedation. Our personal favorite is a combination of
propofol 10mg/ml with remifentanil added at a concentration of 10mcg/cc. We dose this
combination at 25-100mcg/kg/min (propofol component) depending on desired depth of
sedation. Results have been excellent for painful procedures with smooth and rapid
emergence from sedation. Unfortunately, central apnea – which in our experience is
relatively uncommon with propofol alone at these doses - is much more common with
this combination. Note: This is truly “anesthesia” sedation. Please do not consider this
comment an endorsement of this preparation of propofol-remifentanil or these dosing
guidelines! We are just describing our experience with the hope of sparking others to do
the same. Large trials are needed before we could comment on the safety or efficacy of
these drugs used in combination in this way.
One caveat: since the effects of remifentanil are quite marked and are very sensitive to
infusion rate changes, it is best to have the infusion enter very close to the intravenous
catheter (right near the hub) otherwise changes in intravenous fluid flow rates can result
in marked differences in remifentanil effect – see apnea comment above.
We would love to hear how others are making use of remifentanil for pediatric
sedation, with or without propofol.
FDA Propofol Warning
Thanks to another reader, David Fisher MD, who pointed out to us the following letter
from AstrZeneca concerning the use of propofol in the ICU for sedation of acutely ill
children. The following has appeared on Medwatch:
“The FDA has determined that there may be important safety concerns when propofol,
marketed by AstraZeneca as DIPRIVAN (propofol) Injectable Emulsion, is used for
sedation in the Intensive Care Unit (ICU) in pediatric patients. Healthcare professionals
are reminded that propofol is not approved in the U.S. for sedation in pediatric ICU
The AstraZeneca "Dear Healthcare Provider" letter is available on
The MedWatch website as an Acrobat pdf file. Go to the MedWatch 2001
Safety Information page at
This letter will have some impact on the usage of propofol although there is no doubt it
will continue to be used in PICU’s for selected patients. Several reports have come out
over the last 8 years concerning propofol and the development of acidosis and
cardiovascular collapse – I will review one case report and an accompanying editorial
from a 1998 issue of Critical Care Medicine. A list of references is also included. With
the medical-legal landscape now slightly altered by this letter of warning, all practitioners
who use propofol for prolonged sedation will need to be aware of this literature.
1. Cray S, Robinson B, Cox P. Lactic acidemia and bradyarrhythmia in a child
sedated with propofol. Critical Care Medicine, Vol 26(12) Dec 1998, pp 2087-92
This case report is somewhat typical of the type of report that has led to the FDA
warning. The paper is generally well written and describes a previously well 10 month
old boy who developed respiratory distress because of an aspirated foreign body. He
required intubation and mechanical ventilation as part of his management. Propofol was
administered as sedation during this period of time at a mean rate of 10 mg/kg/hr for 50.5
hours. He was noted to have green urine and marked hyperlipidemia – followed by
severe lactic acidemia and bradyarrhythmias unresponsive to conventional treatment.
These abnormalities ultimately resolved after institution of CVVH and prolonged
vasopresser support. He developed encephalopathy and liver/muscle necrosis consistent
with a toxic insult. A subtype of cytochrome oxidase was found to be deficient in the
muscle biopsy of this patient – but not in the skin fibroblasts (thus ruling out an inborn
error of metabolism involving this enzyme system). There was no evidence of infection
or underlying metabolic disease.
Comment: The authors conclude by recommending against the prolonged use of propofol
for ICU sedation. They site several other similar case reports (10 deaths) from the
literature and highlight the many similarities between this case and several others in the
literature. These features include a lack of severe cardiomyopathy on echocardiogram, no
evidence of metabolic disease in the child in spite of an extensive work up, and a severe
lactic acidosis only responsive to hemofiltration.
A hypothesis is offered that suggests that propofol or one of its metabolites affected
cellular metabolism leading to increased lactate production and wide-spread organ
dysfunction. The fact that cytochrome oxidase was deficient in the muscle tissue of this
patient (but not in skin fibroblasts) is taken as an indication that this could be the
metabolic insult that leads to difficulties in this patient population. Improvement with
CVVH suggests that the toxic agent is water soluble.
This report is well put together and quite readable. The discussion is thought
provoking although many of the conclusions that the authors make are based on quite a
bit of conjecture. The fact is we know of no known “toxic metabolite”, and there is no
clear causal relationship between propofol and this patient’s deterioration. The lack of
any convincing prospective data on this entity still leaves doubt as to its exact nature and
the extent of the problem.
2. Susla G. Propofol toxicity in critically ill pediatric patients: Show us the proof.
Critical Care Medicine. Volume 26(12) December 1998, pp 1959-1960
This editorial was printed along with the case report reviewed above and takes a
skeptical view of the literature concerning ICU propofol use and cardiovascular collapse.
The author outlines the lack of clear cause and effect presented in the 11 cases of
propofol toxicity presented in the literature at the time of this printing. He also reviews
previous studies from Macrae et al. and Reed et al. both of whom evaluated over one
hundred pediatric patients receiving propofol without significant difficulties. He also sites
a study by Martin which carefully evaluated the hemodynamic and metabolic effects of
long term propofol infusions (48 hours) in 9 children. Other than a significant rise in
triglycerides, no significant problems were identified. Finally Susla mentions a
retrospective study by Pepperman and Macrae that compared pediatric ICU patients
receiving either propofol or “other” sedation – primarily midazolam. Of the 192 patients
evaluated there was no clear trend toward acidosis in the propofol group. In addition in
the few patients in the propofol group that develop acidosis, none were lipemic.
The editorial concludes by quoting a 1992 FDA statement that stated there was not
sufficient evidence to conclude that there is any direct link between propofol and the
morbidity and mortality that was being associated with its use. Susla maintains this is still
the case and that the advantages of propofol sedation may outweigh any possible risks.
Comment: The argument put forth in this editorial is important although we are not sure
we are convinced about the conclusions. There is still no good prospective data
comparing propofol to other sedatives in a consistent group of ICU patients. Work needs
to continue in an effort to determine who (if) there is a specific group of patients who are
at greatest risk (i.e. upper respiratory tract infection patients). Unfortunately one effect of
the recent FDA ruling may be to make further study of propofol in the ICU setting very
difficult - as institutional review boards are made aware of this warning.
Other propofol references:
1. Noorslet J, Wahlgreen C. Propofol infusion for sedation of children. Crit Care
Med 1990;18: 890-892
2. Macrae D, Blumer JL. Propofol infusion in children BJM 1992;305;953
3. Martin PH, Murthy BVS, Petros AJ. Metabolic, biochemical and haemodynamic
effects of infusion of propofol for long-term sedation of children undergoing
intensive care. Br J Anaesth. 1997; 79:276-279
4. Pepperman ML, Macrae D. A comparison of propofol and other sedative use in
paediatric intensive care in the United Kingdom. Paediatric Anaesthesia
5. Parke TJ, Stevens JE, Rice ASC et. al. Metabolic a cidosis and fatal myocardial
failure after propofol infusion in children: Five Case Reports BMJ 1992;305:613-
6. Barclay K, Williams AJ, Major E. Propofol infusion in children. BMJ
7. Strickland RA, Murray MJ. Fatal metabolic acidosis in a pediatric patient
receiving an infusion of propofol in the intensive care unit: Is there a relationship?
Crit Care Med. 1995;23:405-409.
8. Strickland Ra, Murray MJ. Propofol Bashing: the time to stop is now! Crit Care
9. Nathan N, Debord J, Narcisse F et at. Pharmacokinetics of propofol and its
conjugates after continuous infusion in normal and renal failure patients: A
preliminary study. Acta Anaesth Belg 1993;44:77-85
Close Call and Critical Incident Reports
We invite those receiving the newsletter to submit cases to this "Close Calls" section
aimed at the key safety issues associated with providing pediatric procedural sedation.
The cases have been "sanitized" such that the date, patient identifiers, institutional
identifiers, provider information, etc. will not violate the privacy of the patient or care
A 2-year-old child presented to a hospital Emergency Department (ED) after falling
down a staircase. The child seemed fine to the parents, but a forehead bruise and the
desire to “play it safe” prompted a trip to the local ED. A head CT was ordered with IV
sedation provided by the RN caring for the child. The hospital did not have specific
training for sedation providers and routinely used primary nurses to give sedation “under
the direction of a physician.” After a phone conversation between the RN and the
Emergency Medicine Physician, doses were calculated for Versed and Fentanyl.( This
nurse had never used Versed-Fentanyl combination in a child.) The RN performed
sedation for the CT which was completed without incident (the physician was busy and
not present, but was readily available). The child was then returned to the ED and was
placed in a room with intermittent pulse oximetry monitoring (the child fussed too much
to keep the oximeter on continuously, and it alarmed too much so the probe was taken
off). The staffing in the ED on the night of this event had this nurse managing 2 other
“low acuity” patients (nursing shortages, a staff sick call and an busy waiting room
resulted in this ratio) About 30min after the procedure, the nurse went into the room to
check on the child and found him apneic and cyanotic (although the parents were with the
child they did not recognize the problem because the room had been darkened to let the
child sleep. Their ethnicity made cyanosis difficult to observe). Resuscitation ensued
immediately and appropriately upon discovery of the child - resolving the hypoxia and
bradycardia within minutes. Unfortunately, the child suffered severe anoxic brain injury
and life support was withdrawn several days later. The head CT was negative. The doses
of Versed and Fentanyl were considered “appropriate.”
This case has three major issues we are going to focus on 1) sedation with drug
combinations; 2) recovery monitoring; and 3) strategic factors in management that impact
on front-line workers.
1) Drug combinations:
In this case, the goal of sedation is to have this 2 year old child be still for about 10
minutes allowing the head CT to be performed. The imaging study is not associated with
pain and an exhausted child at night will sleep through positioning and the study.
However, if contrast is to be used, the IV and dye injection can hurt and typically
awakens a 2 year old causing crying and blurry films. In addition, the dye cannot be
given repeatedly (max dose to avoid renal problems). If a child moves during the 2-3min
after dye is given, you may need to wait 24hrs and do it all over again. These factors
result in sedation being used commonly in the two year old, even for a short study like a
Many sedation options exist. In this case Versed and Fentanyl were used in combination.
The major point to make is that combinations of drugs produce pharmacodynamic effects
that are less predictable that use of either drug alone. If given over several minutes,
Versed (midazolam) can be given in doses of 0.5mg/kg before apnea is produced (5mg in
a 10kg child). Similarly, fentanyl 3mcg/kg can be given without apnea (30mcg in a 10kg
child). However, when given together one needs to reduce the doses by more than 50%.
Like many sedative and analgesic medications, these drugs are synergistic in producing
respiratory depression when given together. As anesthesiologists who use this
combination commonly, we find ourselves giving one of the drugs (typically the opioid at
1-2mcg/kg) and then titrate the second drug in very small doses until the desired effect is
reached. When 3 drugs are used, the inability to predict the effect by dosing becomes
even more of an issue.
2) Recovery from sedation and monitoring:
The lack of predictability associated with poly-pharmacy is true for the effect of the drug,
the onset time, the side effect profile, the effective halflife, etc. In this case, the nurse
underestimated the duration of risk for airway obstruction. Although imperfect, many
would argue for a more provocative test of airway obstruction risk after sedation (in
which you lift the head forward to cause airway obstruction and see if the child un-
obstructs themselves (try this in a sleeping child and you will see he/she will struggle and
awaken rather than remain obstructed and become hypoxic). Seeing a child progress from
the inability to correct obstruction to regaining this capability is important for providers
as they determine the level of monitoring that is required during the recovery phase of a
Another key concept in recovery is the choice of continuous vs semi-continuous or
intermittent monitoring. Pulse oximetry, capnometry, and respirometry can all be
considered apnea monitors. Pulse oximetry lags behind capnometry and respirometry
(desaturation typically occurs 30-60 seconds after breathing stops when on room air). In
this case, the nurse involved wanted to use pulse oximetry continuously, but movement
made this impractical. Given the rapidity with which apnea can progress to brain damage
if not diagnosed and managed, intermittent apnea monitoring in a young child is doomed
to be too late for successful resuscitation (without complications) to be accomplished. It
may be that the nurse was considering the parents to be monitors, however, this case
illustrates that we often underestimate the skills associated with medical practice when
we think a lay-person can perform such a duty.
3) Management/organizational factors:
While it is easy to identify the nurse and physician in this case as the “culprits” that
caused this event, more important are the systemic factors that created an environment
that created an “accident waiting to happen.” Economic pressures combined with a
nursing shortage have led management in many institutions to adopt policies that are
unsafe. Goal conflicts, responsibility-authority double binds, and situations in which
demands exceed resources are common for front-line workers. For example, this nurse
had not had formal training in sedation and had never given Versed-Fentanyl to a child,
yet she was expected to do so during a busy night. Hospital policies supported this
practice because they provided no other resources or alternatives. Peaks and valleys in
workload are to be expected in an ED, however few hospitals staff for the worse case
patient volumes. Clinicians are simply expected to do their best when faced with heavy
workloads. The nurse had a supervisor and physician that assigned duties. Although
responsible, they had no authority or formal mechanism to off-load work, without
overloading another nurse. This hospital (like most) had a mechanism for calling in
additional personnel for disasters, but it is not intended for the normal peaks and valleys
experienced in the ED. We hear colleagues make comments such as, “we were slammed
last night”…, “I was running around like a chicken with my head cut off…”, “I didn’t
have time to pee”…). These statements are markers for excessive work load on the front-
line staff. Imagine instead a hospital (and some do exist) in which clinicians have sound
methods for mobilizing additional resources. Comments about a busy night might
instead be, “…we had to call in three additional people”, “we sent 10 patients to another
hospital”…, “I had to send that 2 year old child I sedated to the OR recovery room for
recovery so I could attend to my other patient”…, etc.)
Sedation care strategies that may impact on this type of scenario:
Avoid poly-pharmacy, titrate secondary drugs to effect. We all recognize that
patients do not need the exact same amount of medication for a given sedation.
Patients that need little medication should receive only a little and patients that
need large doses should receive large doses.
Test for the potential for airway obstruction during the recovery from sedation.
Provocative testing (such as flexing the head) should be performed before going
from continuous to intermittent monitoring. Consider borrowing a philosophy of
early vs. late recover from the anesthesia domain (termed phase I vs phase II
recovery) to mark a logical transition in staffing and vigilance that correlates with
a change in risk profile.
Work with management to develop policies and procedures that acknowledge
rather than ignore the peaks and valleys in workload that occur in healthcare.
Respect the challenge facing hospital management trying to provide adequate
resources to front-line workers, without going broke.
Once again, there are no trivial answers to these complicated issues. For a great review
of the latent failure model regarding large complex system failures, read “Operating at
the Sharp End: The Complexity of Human Error,” by R Cook and D Woods, in Human
Error in Medicine; M Bogner editor, Lawrence Erlbaum Associates (1994). Thanks for
the reports that are being submitted, please keep them coming so that we can all learn