The primary aim of our pilot study was to formally and prospectively compare the effectiveness of moderate sedation with ketamine for gastrointestinal endoscopy in children with that of more commonly used regimens combining benzodiazepines and opioids. To a great extent, this study was undertaken with the expectation that piloting ketamine in our endoscopy unit with an eye to quality and safety would reveal ketamine to be superior to our current standard of midazolam and fentanyl. We accepted a patient-selection bias, and indeed our results show that independent of age, pediatric patients who received ketamine were less likely to vocalize distress than those administered midazolam/fentanyl. On the other hand, we also found that ketamine was associated with increased (not decreased) body movement and the same need for restraint by clinical staff as children receiving midazolam/fentanyl. Such results are useful for clarifying a discussion of the comparative effectiveness of these two sedation regimens for endoscopy.
Patients in our study were administered a particular sedative based on physician judgement. In general, we found that physicians were more likely to refer younger children for ketamine, perhaps based on a perception that midazolam/fentanyl does not adequately provide sedation in this age group and that ketamine would be more effective. We believe such selection biases should have favored the effects of ketamine. Instead, we found that patients who received ketamine required restraint for equal percentages of procedure time as their peers who received midazolam/fentanyl.
A secondary aim of our study was to examine sedation safety, at least in terms of intraprocedural events. In our small study, there were two patients who experienced laryngospasm in the ketamine group versus no patients with adverse events in the midazolam/fentanyl group. Although ketamine has been deemed safe based on its cardiovascular and respiratory protective effect [8
], it is also associated with increased risk of laryngospasm [7
]. Both episodes of laryngospasm in our study occurred during upper endoscopic procedures, where stimulation of the posterior pharynx may increase risk of airway complications.
Our pilot work objectively assesses and compares effectiveness of sedation regimens for children undergoing endoscopic procedures using independent monitoring and a standardized, continuous, behavioral rating scale. Previous reports in dental and oncology populations have used global ratings by clinicians to state that ketamine reduces behavioral distress [22
], increases patient cooperation [23
], and reduces crying and body movements [24
]. In contrast, the independently obtained observations in our study are consistent with the fact that ketamine sedation increases muscular hypertonicity and stereotypical “random movements” in children [26
], leading to use of restraint. The discrepancies between our investigation and prior reports are most likely due to a previous lack of standardized and objective behavioral scoring methods in head-to-head comparisons of the effectiveness of sedation regimens for gastrointestinal endoscopy. We believe the use of continuous measures over the duration of a procedure may ultimately allow a clearer definition of effectiveness as a descriptor of sedation regimens.
The Ohio State University Behavioral Scale (OSUBS) used in this study has been previously shown to be a reliable method for evaluating behaviour in sedated children undergoing dental procedures [16
], and has been validated by our group in children undergoing gastrointestinal procedures [13
]. In turn, effectiveness of sedation can be operationalized according to the OSUBRS. In particular, it may be plausible to define optimally effective sedation as rendering a patient “quiet, still, and unrestrained”; while a patient who is “vocalizing distress, moving, and restrained” is least effectively sedated. The use of the OSUBRS also allows measurement of the tendency for patients to move back and forth between behavioral states, as the effects of the medications, their underlying anxiety, and the stimulus of the procedure vary over time.
There are a number of limitations to the present study that should be considered, in addition to the most important concern that the study was not randomized. In particular, the pilot nature of this study precluded adequate sample size calculations, the sample sizes analyzed were small, and study groups were unequal with regards to patient age and procedure type. Patients receiving ketamine were overall younger (median age of 5 years) and more likely to be undergoing colonoscopy compared to the fentanyl/midazolam group (median age of 12.5 years). The small sample size also precludes a true discussion of safety. Our pilot data may be useful for planning larger studies required to determine whether laryngospasm or other adverse events occurs more frequently with ketamine than has been reported previously.
Another possible limitation of this study was that it used relatively broad-based definitions of “moving” and “needing restraint”. Moving in the present study included both intentional (combative) and unintentional (drug-induced) movement. Nonetheless, regardless of reason for restraint or forcefulness of restraint, the number of staff needed for the procedure to be safely conducted remained unchanged. Ketamine, therefore, may not afford advantages for pediatric endoscopy in terms of reducing staff or resources.
A final aspect not considered here, but worthy of investigation are more long-term adverse events associated with ketamine, as a dissociative and hallucinogenic agent. Ketamine has been observed to induce a “locked-in” state, which may prevent patients from vocalizing distress, but allow them to witness their procedure. Future studies might consider sending patients home with a diary to document their experiences after their procedure under ketamine.
In conclusion, fentanyl/midazolam is a standard sedative regimen commonly used for GI procedures that is not totally reliable at rendering patients quiet and nonmoving. More recently, a number of studies have suggested that ketamine may provide superior and equally safe pediatric sedation [8
]. Although patients in our prospective study were less likely to vocalize distress, they were still likely to move and require restraint during procedures. In addition, our study found ketamine sedation to be more associated with laryngospasm during pediatric endoscopy, which has currently limited our interest in using it further in our unit. Our study demonstrates that effective sedation relies on numerous factors and may not be adequately measured by global postprocedural ratings. With the use of objective and continuous behavioral measurements, key differences in the effectiveness of different sedation regimens for gastrointestinal endoscopy will become clearer and guidelines for improvements in pediatric sedation more likely.