Walsh et al. published “Derivation of Candidate Clinical Decision Rules to Identify Infants at Risk for Central Apnea.” in Pediatrics in November, which attempted to derive several CDRs and compare them for identifying risk of central apnea in pediatric patients with respiratory illness. Of course, for an outcome as rare as central apnea in a population that usually does very well, almost any set of criteria you apply to patients will leave you with a rule that has a very high NPV — so what did they find?
The group analyzed 990 ED visits for 892 infants. Central apnea subsequently occurred in the hospital in 41 (5%) patients. Three candidate CDRs were generated by different techniques, and the results were analyzed and yielded the following risk factors: Parental report of apnea, previous history of apnea, congenital heart disease, birth weight ≤2.5 kg, lower weight, and age ≤6 weeks all identified a group at high risk for subsequent central apnea. All CDRs and RFs were 100% sensitive (95% confidence interval [CI] 91%-100%) and had a negative predictive value of 100% (95% CI 99%-100%) for the subsequent apnea.
The third tool, not shown above, is a computationally-intensive algorithm that used a Random Forest method to generate a risk stratification. Much like the recently-published work on sepsis using Big Data strategies, this had a better AOC than either of the above two, which are much simpler and can by applied by clinicians. This rule and others like it may have a future in the form of electronic heath record-embedded decision support, but are less amenable to being remembered and applied by the physician at the point of care when making a disposition decision.
It is important to note some caveats about this and the results — particularly the prevalence of apnea in this population, which accepted parental reports of apneic events as part of the numerator (i.e. not just monitored and captured events), but it still underscores the idea that parental concern should be your concern until proven otherwise.
Anyway, all this to say, bronchiolitis-related apnea is a terrible outcome but a very rare outcome. Admission for observation may be considered in high-risk patients, and should be discussed with parents. If a hospital doesn’t have apnea monitoring, is it still reasonable to admit these kids for observation? That’s not really germane to the studies published here, but came up recently on one of my rotations — I guess if a respiratory arrest happened, it would be better to be in a setting where a response could occur swiftly and with full capabilities, but I don’t know that such an admission is better than sending the kid home with parents who will likely be steadfast bedside observers of the child’s respiratory status throughout the night. That question will have to wait for another study, I suppose.