First, Dr Lerman states, “In order to investigate and develop a risk-scale for emerging delirium, only children who are pain-free should be studied.” We agree with the opinion that pain-free patients should be included to diagnose delirium accurately. In our study, the negative behavior of children emerging from anesthesia, such as disorientation, tantrums, or agitation, is referred to as emergence agitation (EA), and we aimed to develop the risk scale for predicting postoperative EA. As we mentioned in the Discussion section, the major causes of EA were postoperative delirium or pain, but we could not distinguish between these causes of EA. The pediatric anesthesia emergence delirium scale2 we used could not distinguish between pain and delirium because “it is possible that the pediatric anesthesia emergence delirium scale items ‘The child is inconsolable’ and ‘The child is restless’ may reflect pain and emergence delirium.”2 We could have diagnosed the cause of EA as delirium only if pain-free patients were included in the study. However, we did not restrict our inclusion criteria because the aim of our study was not to diagnose the cause of EA, but to predict postoperative EA. Therefore, the EA risk scale we developed could only be used for predicting agitation, and not delirium. We have discussed this issue in the “Limitations” part of the discussion.
Second, Dr Lerman pointed out that the factors included in the EA risk scale (ie, type of surgery, duration of anesthesia, and preoperative behavior) are not known to be associated with EA. We included these factors as candidates of predictive factors in the initial model based on previous studies and on a clinical perspective. These factors remained in the final regression model owing to a strong association to EA in our development cohort.3 Type of surgery4 and preoperative behavior5 have previously been reported to be associated with EA, but duration of anesthesia was not. We speculated that duration of anesthesia (ie, duration of surgery) was selected as an independent predictor because it indirectly reflects the difficulty and complexity of the procedures.
Third, Dr Lerman states that the EA risk scale may yield internally consistent results but holds no external validity. We agree with this point. We could only confirm the “internal validity”; the cohort used in both development and the validation phase was from the same hospital. The “external validity,” or the validity of the EA risk scale for other patient cohorts in other hospitals, should be confirmed in further research.
We sincerely appreciate Dr Lerman’s concerns and suggestions on these issues in our study. We hope that our discussion will promote future studies in this field.