Sedating Children in an Adult ICU

Abstract & Commentary

Synopsis: Over a 10-year period, 295 procedures were performed in 127 children using the monitoring and clinical management resources of an adult ICU. Sedation using intravenous midazolam and ketamine administered by an adult intensivist resulted in a 3% complication rate and no deaths.

Source: Slonim AD, Ognibene FP. Crit Care Med 1998;26:1900-1904.

Slonim and ognibene report their experience of providing a sedation service in an adult ICU for children with cancer undergoing painful or frightening procedures. They used intravenous midazolam and ketamine for 295 procedures performed in 127 children over a 10-year period. Ketamine was chosen for its analgesic and amnestic effects and the fact that airway reflexes are preserved at sedative levels. Midazolam reduces the frequency of dysphoric reactions to the ketamine. The sedation team consisted of a senior adult intensivist, an ICU nurse, and a critical care fellow. The clinician performing the procedure was not part of the sedation team. Complications were identified from notes in the medical record. Data for this study were collected retrospectively by chart review.

Children were kept nothing by mouth (NPO) for six hours prior to the procedure, all had an intravenous line in place, and all were monitored with continuous ECG and pulse oximetry, and intermittent blood pressure determinations. Oxygen was provided by flowby in all patients. The average age of the patients was 7.7 years, with a range of 1-18 years. Procedures included bone marrow aspiration (43%), lumbar puncture (27%), bone biopsy (13%), and central venous line placement (9%). Ketamine dose was 2.0 ± 1.1 mg/kg, with one patient receiving 10 mg/kg. Midazolam dose was 0.05 ± 0.06 mg/kg. The average duration of the procedure was 144 ± 95 min.

Nine patients experienced complications attributable to sedation (3%). These included wheezing requiring treatment in a 2-year-old; two cutaneous rashes; prolonged oxygen desaturation in one child; hypotension requiring volume resuscitation; hypertension and tachycardia (two patients); and emergence reaction (two patients). The procedure was prolonged in the children experiencing a complication (264 ± 98 min), mostly due to a longer monitoring period following the complication. No patient was believed to have significant, persistent morbidity from any complication.


The significance of this paper is several-fold: the demand for sedation and anesthesia outside the operating room is expanding, and anesthesiologists will probably not routinely be available to provide it due to a growing national shortage. Using ICU space and personnel may be a viable alternative to taking the patient to the operating room. While the use of ketamine and midazolam was the thrust of this paper, the important issue of the qualifications of the person performing sedation was incompletely discussed. An adult intensivist has many important skills, such as monitoring gas exchange and vital signs. However, airway management and intubation of the child requires additional experience. Although none of the children in this series appeared to require intubation, if this were to occur, a system to provide this level of skill must be available. This problem could be addressed by having an agreement with the operating room to provide backup.

Was the outcome of sedation in this group as good as Slonim and Ognibene suggest? What is not reported is how many children were unable to be adequately sedated to be able to perform the required procedure. No follow-up visits were reported to assess the patients’ and their parents’ satisfaction with the sedation. Prolonged and delayed reactions to ketamine are frequent. A 3% complication rate seems high, although most complications were minor. The two respiratory complications—wheezing and prolonged desaturation—could have been unrecognized aspiration. The consequence of limiting the choice to sedation without an airway rather than including the option of general anesthesia with endotracheal intubation always includes the risk of pulmonary aspiration during the procedure. A 1% incidence of aspiration during an elective procedure is unacceptable.

The complication rate reported in this study probably underestimates the actual frequency of problems. Since it was a retrospective chart review, only those occurrences that reached a high level of concern would have been written down. No continuous records were kept and there was no requirement for reporting specific events, such as airway obstruction or desaturation.

This paper correctly stresses the important point that the person sedating and monitoring the patient cannot be the person performing the procedure. The model suggested by Slonim and Ognibene involves three additional individuals: an ICU attending, a critical care fellow, and an ICU nurse. If all three individuals provide continuous attendance to the patient, this is an expensive model to adopt. I suspect that the nurse and the critical care fellow stayed at the child’s side while the ICU attending participated at the start and end of the procedure and was available in close proximity at other times. This model for sedation and anesthesia should not be considered a less expensive alternative to having an anesthesiologist provide care, although using an ICU location rather than the operating room may generate institutional savings. Other models for sedation of adults or children might include an experienced respiratory therapist and a nurse instead of the intensivist. The importance of this paper is to stimulate ICUs and hospitals to look at underused resources to improve care to poorly or underserved patients in the hospital system.

Ketamine is an excellent drug:

    a. for intubating children in respiratory failure.
    b. for sedating children for painful procedures.
    c. to restore circulation following cardiac arrest.
    d. to prevent withdrawal in crack cocaine abusers.
    e. to achieve muscle relaxation during general anesthesia.