This study evaluated the impact of Fio2 during the ARM on postoperative atelectasis development, using LUS. The postoperative modified LUS score was higher in the high-Fio2 group than in the low-Fio2 group, indicating more severe loss of aeration. In addition, postoperative consolidation was more frequently observed in the high-Fio2 group, while no significant difference was observed in preoperative modified LUS score. In terms of oxygenation, there was no significant difference between groups at any time-point. These observations were consistent with our hypothesis that, using high Fio2 (1.0) during the ARM would not benefit oxygenation and would cause more postoperative atelectasis than using low Fio2 (0.4).
High Fio2 has been considered responsible for development of absorption atelectasis during general anesthesia (22, 23). To the best of our knowledge, however, the impact of high Fio2 temporarily applied during ARMs on atelectasis development has not been investigated previously. In the present study, patients assigned to the high-Fio2 group received ARM with Fio2 1.0, whereas those in the low-Fio2 group received ARM with Fio2 0.4. The Fio2 was uniformly maintained at 0.4 with 5-cmH2O PEEP during post-ARM mechanical ventilation in both groups. A high oxygen concentration in the alveoli during ARM was predicted to cause more absorption atelectasis. Consequently, the postoperative modified LUS score was markedly better in the low-Fio2 than in the high-Fio2 group and the difference persisted in the PACU.
Rothen and others demonstrated the progression of absorption atelectasis over time, using computed tomography, after ARMs in 12 patients, using Fio2 of 0.4 or 1.0 during ARM and thereafter (24). While absorption atelectasis developed within 5 minutes in the Fio2 1.0 group, it developed after 40 minutes in the Fio2 0.4 group. Although the impact of oxygen concentration was obvious, this previous and the present study differed in that the previous study applied the designated Fio2 not only during ARM, but also during the rest of the study period. Additionally, Song and colleagues studied absorption atelectasis according to Fio2 during mechanical ventilation was evaluated by LUS in children (25). Although Fio2 had no impact on the incidence of significant atelectasis (consolidation score ≥ 2), high Fio2 caused higher consolidation and B-line scores. That study compared Fio2 0.3 and 0.6, which is a relatively small difference, and did not included laparoscopic surgeries in the Trendelenburg position, which may explain the discrepancy with our results.
We observed no significant difference in the Pao2 to Fio2 ratio at any time point. Recruitment of collapsed alveoli with high oxygen concentrations led to a rapid re-collapse of the inflated alveoli, rather than benefiting oxygenation. In clinical practice, Fio2 may be raised during the ARM to improve Spo2 immediately, in case of desaturation during surgery. Nonetheless, we found that high Fio2 during ARM does not actually improve oxygenation, despite a transient, rapid increase in Spo2. A recent study of 32 patients undergoing laparoscopic cholecystectomy compared Pao2 levels after 2 intraoperative ARMs, with Fio2 0.3 or Fio2 1.0 (26). Intraoperative Pao2 did not differ between the 2 groups, but Pao2 was significantly better in the Fio2 0.3 group, by postoperative blood gas analysis, which differed from our findings. This may be due to differences in mean operation time and position between the studies. Atelectasis may be more affected by Fio2 than by other factors in surgery using a sitting position, such as laparoscopic cholecystectomy than in surgery using a Trendelenburg position.
The overall intraoperative desaturation incidence was markedly lower in our study than in the study of Monastesse and others, which was assumed to be mainly due to repetitive ARMs [defined as Spo2 < 95% vs. Spo2 < 94%; 5/90 (5.6%) vs. 4/29 (13.8%); excluding a case of endobronchial intubation] (20). Spo2 never fell below 90% in either group in our study and no patients required rescue by a change in Fio2 or PEEP. Furthermore, in-hospital pulmonary complications were absent in both groups. Inclusion of only patients with low risk of pulmonary complications may account for this complication-free recovery in pulmonary function, along with repeated ARMs during mechanical ventilation. Postoperative fever (≥ 37.5℃) developed in a considerable number of patients in both groups. Length of hospital stay was non-significantly longer in the high-Fio2 group.
The postoperative modified LUS scores in the present study were in accordance with those of Monastesse and colleagues (20). The PACU score of the low-Fio2 group was lower and that of the high-Fio2 group was higher than in the previous study, although the mean values of both studies were similar. We also analyzed the incidence of significant atelectasis additionally; it was observed in > 80% of patients in the high-Fio2 group. A higher score as well as consolidation were mainly observed in the posterior (dependent) part of the thorax. This high significant atelectasis incidence may be attributed to pneumoperitoneum and the Trendelenburg position. As all patients showed at least a single, small, subpleural consolidation after pneumoperitoneum in Monastesse and colleagues’ study (20), the incidence of significant atelectasis is likely to be acceptable. Nonetheless, the significant atelectasis observed in our study did not alter the clinical outcome.
Our study had several limitations. Firstly, since only patients with a low risk of pulmonary complications were included, our results may not extend to patients with lung disease. Moreover, clinical consequences of the observed atelectasis may not have been observed for the same reason. Second, the anesthesiologist who performed the ARM was not blinded. However, the anesthesiologist performing LUS for outcome measurement was blinded to the Fio2 used for the ARM. Third, there is a possibility of incomplete intraoperative recruitment with the opening pressure obtained in the supine state before surgical incision. The opening pressure was used since access to the dependent part of the thorax was limited during the surgery. Nevertheless, it was considered to be sufficiently effective because ARMs were mostly performed at a high pressure of ≥ 30 cmH2O. Fourth, the definition of significant atelectasis was not validated by previous studies. Although previous literature have used LUS as a diagnostic tool for atelectasis (14, 18, 20, 27–29), criteria for significant atelectasis need to be established.
In conclusion, for patients undergoing laparoscopic surgery in the Trendelenburg position, absorption atelectasis occurred more frequently when ARMs were performed with high Fio2 than when it was performed with low Fio2. Based on the findings of our study, high Fio2 during the ARM yields no oxygenation benefit and may result in more atelectasis than low Fio2.