Physiological pattern differences in non-intubated vs. intubated thoracic anesthesia based on time series analysis.

BACKGROUND: It is still weak with evidence of the targeted anesthesia management for non-intubated thoracic anesthesia, mainly based on the professional experience and information on the anesthesia sheet. Our study elucidates the differences in dynamic physiological parameters, including arterial oxygen saturation (SpO2), end-tidal carbon dioxide (ETCO2), bispectral index (BIS), mean arterial pressure (MAP), and heart rate (HR), as revealed by the intraoperative real-time database during one-lung ventilation with reduced anesthetics in non-intubated compared with intubated thoracic anesthesia.

METHODS: A prospective, observational, controlled study was conducted at a single national center for respiratory medicine, involving 131 non-intubated and 155 intubated patients undergoing video-assisted thoracoscopic surgery (VATS) [including sublobectomy (SubL) or lobectomy (Lob)]. Propensity score matching (PSM) resulted in 78 matched pairs for SubL; however, PSM was not applied for Lob. Clinical characteristics, anesthetic use, adjuvant drug use, and postoperative outcomes were compared between the two groups. Physiological parameters were recorded every 10 seconds using a lung-brain-heart-anesthesia platform and analyzed through descriptive statistics, visualization, feature extraction, and slope analysis.

RESULTS: The protocol of both non-intubated SubL and Lob reduced the use of sevoflurane, sufentanil, remifentanil, and cisatracurium (P<0.01). In the process of non-intubated thoracic anesthesia, statistical analysis of interquartile range (IQR) and the percentage of monitoring parameters values within the normal range showed significant results: a higher BIS percentage (72.2-78.9%), a similar SpO2 percentage (88.5-96.1%), an initial drop to 0 mmHg followed by higher and fluctuating ETCO2 percentages (46-55.2%) in the upper range, a lower percentage of MAP (69.4-70.1%), and a higher percentage of HR (82.1-82.8%). The fitted trends of both ETCO2 and HR increased (P<0.001), indicating a higher ETCO2 level related to increased HR during non-intubated anesthesia, although HR remained within the safe normal range. Notably, the slope of Aldrete scores improved significantly under non-intubated anesthesia for Lob patients [0.088 (0.074-0.100)] vs. intubated anesthesia [0.066 (0.050-0.085); P=0.03] during postanesthesia care unit (PACU) stay. Additionally, PACU stay duration was shorter in the non-intubated group {72.5 [60-82] minutes} compared to the intubated group {85 [70-106] minutes; P=0.04}.

CONCLUSIONS: Using a real-time database, the non-intubated thoracic anesthesia management pattern is firstly clearly demonstrated, which differs from the intubated anesthesia with the machine-controlled pattern. Reducing intravenous anesthetics with regional nerve blocks could maintain spontaneous ventilation (SV), ensure adequate oxygen supplementation, an appropriate depth of anesthesia and HR. This approach, tolerating mild hypercapnia, permissive lower mean artery pressure, notably results in postoperative outcomes comparable to intubated anesthesia. Non-intubated anesthesia accelerates postoperative recovery in the PACU, particularly for Lob patients.