The “response” of Drs. Glass and Shellhaas to my earlier Commentary in the Journal reiterates most of the points that they made in the original report concerning Early Discontinuation of Antiseizure Medication (ASM) in Neonatal Seizures (Glass HG, et al., JAMA Neurology, 2021). They have clarified some of the terminology that was not entirely clear in the original paper. I am pleased that they agree with my recommendation to proceed with caution in the decision concerning discontinuation of ASM in the neonatal period. However, their statement that “we... stand by our conclusions that routine discontinuation of ASM after resolution of acute provoked neonatal seizures and prior to hospital discharge is safe and warranted in most cases” remains confusing to me. At the risk of enmeshing ourselves in semantic silliness, if discontinuation of ASM is acceptable to Drs. Glass and Shellhaas “in most cases”, then why use the term “routine”? Their conclusion that the recommendation for discontinuation applies to “most” but presumably not all cases indicates to me that the recommendation is not “routine”, nor should it be.
The remainder of the Response of Drs. Glass and Shellhaas principally repeats some of the findings in their original study. Although the repetition is useful, their otherwise excellent study is insufficiently powered to draw decisive conclusions about certain etiologies, e.g., arterial stroke. In their study fully 72% of the infants with stroke were maintained on ASM after discharge. Only 22 infants with stroke had ASM discontinued. As stated in my Commentary, this small number renders the comparative date concerning outcomes in stroke seriously underpowered. Moreover, their “three simple questions” for the clinician to ask before discontinuation of ASM do not include mention of the results of the neurological examination. Surely Drs. Glass and Shellhaas do not make such an important decision as ASM discontinuation without carefully examining the infant. As with stroke, only approximately 20% (n = 20) of the infants with an abnormal neurological examination at discharge had ASM discontinued, thus again rendering the conclusion that this assessment is not necessary in the decision-making underpowered. Thus, I consider their “three simple questions” for the clinician to address re: discontinuation of ASM to be too “simple”.
The final sentence of their Response indicates that Drs. Glass and Shellhaas have dug in their heels and will retain the ambiguity associated with use of the terms “routine” and “for most neonates” in the same sentence. I conclude, then, from that closing statement that some newborns with acute provoked seizures should not have their ASM discontinued prior to hospital discharge. I agree. The absolute number of such newborns is likely quite small. Moreover, it is that small group for whom I suggested follow-up in 1– 3 months. I did not suggest and do not suggest “1– 3 months” of ASM treatment for more than that small group, as implied incorrectly by Glass and Shellhaas in their Response. Finally, I agree with the statement made by Glass et al. in the original paper that “larger, longer-term studies are needed”. Clearly, we need more data to identify more conclusively the small group for whom discontinuation of ASM in the neonatal period is not optimal.