Abstract
We want to greatly thank Tran and colleagues 1 for taking the time to read and respond to our recent article. 2 The only way that academic medicine works is if we all stay engaged and show a willingness to communicate with and criticize our colleagues.
Their letter brings up several excellent points that we would like to address. The historical practice at our institution was to use bronchial blockers for elective lobectomy in small children. Then, a practice change was made in 2017 given multiple reports, at both the American Pediatric Surgical Association annual meeting and the American Academy of Pediatrics Surgical Section meeting, from other institutions who have abandoned the use of blockers and achieved greater success with endobronchial intubation. So this was clearly not a case of our anesthesiologists only placing one to two blockers each year.
If anything, our anesthesiologists should have been relatively inexperienced with endobronchial intubation, as that modality was used very infrequently before the 2017 practice change was instituted. Despite that relative inexperience, our anecdotal outcomes, as well as our documented outcomes in this study, greatly improved. When our article was originally submitted for consideration, our discussion simply stated that we achieved better outcomes with endobronchial intubation at our institution.
However, 2 of our reviewers criticized that statement and said that our conclusion was overly weak given our findings. Rather, they suggested that we change our discussion to state that perhaps the benefits of blockers are outweighed by the downsides that come along with their use. Since this criticism was offered by 2 separate reviewers, we did indeed choose to adopt their suggestion into our discussion.
We also had extensive discussions with 2 independent statisticians at our institution. Although they disagreed on several points, they both agreed that adopting a Bonferroni correction was not applicable to our study. Their argument was that use of Bonferroni is only applicable when the events being analyzed are completely unrelated to one another. For instance, when looking at table 1, one would certainly expect there to be a correlation between a higher end tidal CO2 (ET CO2) after intubation, with higher ET CO2 during surgery and higher ET CO2 before extubation.
Furthermore, one would also expect that higher blood loss to be associated with an increased rate of packed red blood cell transfusion and, similarly, conversion to open to be associated with increased length of stay. For these reasons, our statisticians were adamant that Bonferroni correction would not apply.
The point that the writers made about it being extremely difficult to get accurate estimates of intraoperative blood loss is well taken and we completely agree with it. However, that would not apply to operative times or time under anesthesia, which are well documented in our electronic medical record.
Finally, the entire reason we wanted to publish our study was to make it clear that different institutions can arrive at different decisions on how to best care for their patients. It is conclusively clear that at Mount Sinai, we achieve better clinical outcomes when using endobronchial intubation for pulmonary lobectomy in small children. Those clinical improvements have been well documented in multiple reviews and are agreed upon both by our institutions' pediatric anesthesiologists and pediatric general surgeons.
However, we can certainly acknowledge that other institutions may feel they achieve better outcomes when using bronchial blockers for similar cases. We certainly do not all have to arrive at the same decision on how to best treat the children who present to our respective hospitals. Rather, we feel every institution should look critically at their own outcomes to determine whether they are using the best techniques to take care of their patients or whether a practice change in warranted. We made such a change in 2017 and our patients are now better served by that change.