Dexmedetomidine and postoperative cardiac surgical delirium: a reply

We thank Liu et al. [1] for their careful reading of our article [2] and for their thoughtful comments. We are pleased to address the points raised and provide clarification. We acknowledge that node-split plots and the corresponding statistical outputs were not included in the published version of our article. This omission was not due to a lack of analysis and was instead to ensure only the most appropriate and relevant data and figures were included in the final version. We would be delighted to share any such data with anyone who requests it from the corresponding author. We agree that consistency testing is an important aspect of any network meta-analysis. However, in this network, performing and reporting these results would not alter our main conclusions or affect the certainty ratings from the grading framework, which already classified most outcomes as low or very low certainty. Given the limited evidence base, we considered that further emphasis on consistency plots would add length without influencing interpretation materially. We nonetheless appreciate the reminder that such reporting enhances completeness and transparency. We acknowledge that our discussion of publication bias could have been more wide-ranging. The Egger's test indicated a small-study effect, and we noted this limitation, but we agree that more explanation of possible sources and implications would have strengthened the discussion. On the interpretation of the most effective intervention, the statement that the combination of dexmedetomidine and melatonin was the most effective intervention reflects the statistical output of surface area under the cumulative ranking curve and p score rankings, not a subjective assertion. We were careful to describe our findings with appropriate caution, noting the limited number of randomised trials and the low certainty of evidence. On the apparent synergy between dexmedetomidine and melatonin, we appreciate the opportunity to clarify this point. Our discussion proposed potential synergism based on shared anti-inflammatory, neuroprotective and sedative mechanisms. However, we agree that further mechanistic exploration is warranted. The current literature provides only limited data to support specific pharmacodynamic interactions between these drugs. On the diagnostic tools for delirium, most of the included trials indeed used the confusion assessment method (CAM) or CAM-ICU criteria. Unfortunately, several studies did not report their diagnostic tools clearly, which made subgroup analysis based on diagnostic methodology unfeasible without introducing further bias from missing data. We thank Liu et al. for their comments, which contribute to an important discussion on peri-operative neurocognitive protection. We hope that our clarifications address their concerns and reaffirm the transparency and methodological rigour of our work.