Does the Risk Scale Predict Emergence Agitation in Children?

    loading  Checking for direct PDF access through Ovid


I read with interest the study by Hino et al,1 who developed a risk scale for emergence agitation (EA) in children. Pediatric anesthesiologists may erroneously assume that the authors accomplished their objective publishing a scale (Table 3) that predicts the risk of emergence delirium (ED) (referred to as EA by the authors),1 but I believe their scale is not only invalid but further serves to confuse the literature on ED in children. To investigate and develop a risk scale for ED, only children who are pain free should be studied because pain shares many of the characteristics of ED, undermining the veracity of the diagnosis of ED. The children used to develop this risk scale demonstrated incontrovertible evidence that pain was present in at least 50% in both groups (median Children's Hospital of Eastern Ontario pain scale [CHEOPS] scores of 8–10).2 With so many children exhibiting pain, the diagnosis of ED in that study is suspect, precluding the inclusion of those children in the current study to develop a risk scale for ED. In developing the Pediatric Anesthesia Emergence Delirium (PAED) scale, we administered dimenhydrinate, a nonanalgesic drug, to children who exhibited signs consistent with ED: we diagnosed ED only in those who responded to dimenhydrinate with resolution of their symptoms.3
The authors included several other factors in their risk development for ED. They identified “operative procedures” as a risk factor for ED.1 However, the only mechanism by which surgery could contribute to ED is if it causes pain (eg, adenotonsillectomy). Surgery in the absence of sevoflurane does not cause ED; otherwise surgery during propofol anesthesia would cause ED, which it does not. Although the authors cite studies in which certain surgeries were associated with ED, those studies also conflated pain and delirium. They also asserted that the duration of anesthesia was associated with the incidence of ED in children. A predominantly sevoflurane-based anesthetic introduces a 10%–30% risk of ED in children in the vulnerable age range, independent of the duration of anesthesia. There is no evidence that duration of anesthesia is associated with ED. Preoperative behavior has been correlated with ED in a previous study.4 However, those authors treated the PAED score as a continuous variable and identified a correlation between preoperative behavior and the PAED score, although scores <10 (or other cutoff values) hold no clinical significance or meaning. Indeed, preoperative behavior has never been confirmed as a predictor of ED. I submit that creating a risk scale for ED based to a large extent on factors that are not known to be associated with ED (eg, type of surgery, duration of anesthesia, preoperative behavior score and pain) may yield internally consistent results but holds no external validity or predictive value for the true risk of ED in children anesthetized with sevoflurane.
    loading  Loading Related Articles