Among the most common operations in otolaryngology, laryngeal microsurgery (LMS) requires administration of general anesthesia and endotracheal tube (ETT) intubation typically. In addition to providing stable gas exchange, ETT intubation can protect airways by preventing secretions from entering the lower respiratory tract. Parameters monitoring such as end-tidal CO2 and tidal volume through ETT is also crucial.

Another option is nonintubated anesthesia applied in combination with transnasal humidified rapid-insufflation ventilatory exchange or high-flow nasal oxygen (HFNO) for the estimated LMS time <30 minutes.1 Its use in LMS can prevent the complications caused by ETT intubation such as oral soft tissue trauma, tracheal injuries, and dental injuries,2 as well as enable comprehensive vocal cord inspection and treatment by providing a loosen clearer surgical field.3 Currently, the practice of nonintubated anesthesia in LMS involves the administration of a muscle relaxant, which helps prevent bucking during the procedure. However, the muscle relaxant administration can lead to apnea and hypercapnia which may have negative effect on hemodynamics. The superior laryngeal nerve block (SLNB) is an efficacious intervention that mitigates laryngeal reflex responsiveness and sustains hemodynamic stability during laryngeal manipulations. As evidenced in the literature, administering 1 to 2 mL of 2% lidocaine in the region inferior and lateral to the greater horn of the hyoid bone yields proficient analgesic effects, lasting from 60 to 180 minutes post-injection.4 The utilization of SLNB is noted for providing superior analgesic effects with extended duration. Therefore, this study investigates the effectiveness and safety of HFNO in conjunction with SLNB for the maintenance of spontaneous ventilation in patients undergoing nonintubated LMS.

In this prospective cohort study, the efficacy of LMS with and without intubated general anesthesia was examined. Furthermore, we wanted to compare the safety and practicality of LMS with nonintubated general anesthesia using two different ventilation strategies. Ethics approval for this study was provided by the Institutional Review Board of Kaohsiung Veterans General Hospital (No. KSVGH20-CT9-09). Additionally, our study has been registered with ClinicalTrials.gov under the registration number NCT05420649. Written informed consent was obtained from all trial participants. The Declaration of Helsinki was followed in the conduct of this investigation. Enrollment in the study was open to patients between the ages of 20 and 80 who underwent LMS with intravenous general anesthesia at Kaohsiung Veterans General Hospital between October 2020 and December 2022. The exclusion criteria were as follows: presence of severe airway obstruction and severe airway disease, American Society of Anesthesiologists (ASA) physical state > III, pregnancy, or body mass index ≥40 kg/m2. Patients requiring LMS under intravenous general anesthesia were recruited and stratified into three cohorts: the intubated general anesthesia group (ITGA), which involved endotracheal intubation; the neuromuscular blocking group (NMB) with HFNO, which did not involve intubation but included the administration of muscle relaxants; and the superior laryngeal nerve block group (NB) with HFNO, which proceeded without either intubation or neuromuscular blockade, utilizing SLNB to facilitate the surgical process. Peripheral intravenous access monitoring was performed for all patients preoperatively, including peripheral oxygen saturation (SpO2), electrocardiogram, noninvasive blood pressure (BP), and GE Entropy EasyFit Sensor/bispectral (BIS) index (Medctronic, Minneapolis, MN) monitoring; train-of-four (TOF) monitoring was performed only in ITGA and NMB group. Glycopyrolate (0.2 mg), hydrocortisone (100 mg), and midazolam (1-5 mg) were administered intravenously as premedications before induction. Intravenous propofol (Schneider model effect-site target-controlled infusion [TCI] = 2.5-5 μg/mL), remifentanil (Minto model effect-site TCI = 0.5-3 ng/mL), and lidocaine (1-1.5 mg/kg) were administered at the start of induction. The target concentration of TCI propofol will be adjusted to maintain the value of Entropy or BIS around 40 to 60 during operation. Rocuronium (0.6 mg/kg) or cisatracurium (0.15 mg/kg) was administered at loss of consciousness in the ITGA group, and TOF monitoring helped maintain deep-to-moderate neuromuscular blocks for ETT intubation (5.0-5.5, I.D. mm) and for the entire LMS period. Rocuronium was administered before LMS initiation in the NMB group, and a deep-to-moderate neuromuscular block was maintained for the entire LMS period. A face mask with 100% O2 at a flow rate of 6 L/min before induction was used for preoxygenation in the ITGA group. After intubation, the ventilator was set to operate in volume control mode, in which the tidal volume was set to 6 to 8 mL/kg, 50% O2 was supplied at a flow rate of 2 L/min, and end-tidal CO2 was maintained at <40 mmHg. For HFNO delivery in the two nonintubated groups, an Optiflow device (Fisher & Paykel Healthcare, Auckland, New Zealand) was used to perform preoxygenation with 100% O2 at a flow rate of 20 L/min for at least 5 minutes before induction, followed by increasing the flow rate to 50 L/min after induction and for the entire LMS period. In the study, we conducted laser surgery by using a diode laser in the few patients in the ITGA, NMB, and NB groups. For the patients in the NMB and NB groups, we supplied HFNO by substituting 100% O2 with pure air maintained at a flow rate of 50 L/min for 2 to 3 minutes before using the laser. During this period, we prepared 20 mL of normal saline in the syringe as a fire extinguisher. The oxygen supply with pure air could be turned off immediately at any time.5 The HFNO was restored to 100% O2 following the conclusion of the laser application. All diode laser operations were completed successfully without airway ignition.

The bilateral transverse-approach ultrasound-guided SLNB was executed subsequent to induction, utilizing a FUJIFILM SonoSite“X-Porte linear probe within the frequency range of 6 to 15 Hz, as delineated in the preceding literature.6 The sonographic landmark, specifically the greater horn of the hyoid bone, was pinpointed under a transverse orientation of the cervical ultrasonography. Subsequently, a quantity of 1 to 2 mL of 2% lidocaine was meticulously infiltrated around the identified landmark using a 25-gauge needle for local anesthetic purposes.

On the basis of surgical plethysmographic index (SPI) values of 30 to 50, TCIs of remifentanil were administered to reduce the effect of LMS-induced stimulation and pain on vital signs. At the end of the LMS procedure, intravenous infusions ceased in the NB group, and the patients were allowed to breathe spontaneously as they emerged from general anesthesia. In the ITGA and NMB groups, intravenous sugammadex or neostigmine was administrated as reversal agent for rocuronium or cisatracurium. After extubation and recovery to spontaneous breathing in the ITGA group, and after face mask ventilation and HFNO cessation in the NMB group, the patients in both groups were transferred to the postanesthesia care unit (PACU).

Pre-LMS period measurements were obtained after the patient entered the operating room, and all monitors were set every 5 minutes three times. LMS period measurements were conducted every 5 minutes from the start of the LMS until the end of the LMS. PACU measurements were performed every 5 minutes after patients arrived at the PACU and were breathing with low-flow supplemental oxygen (2-4 L/min).

Arterial blood gas (ABG) measurements were performed at three-time points to assess the partial pressure of arterial carbon dioxide (PaCO2), the partial pressure of arterial oxygen (PaO2), and the pH in each group. These time points are described as follows: the first ABG measurement was performed before intubation in the ITGA group, before muscle relaxant administration in the NMB group, and before bilateral SLNB execution in the NB group (pre-LMS). The second ABG measurement was performed immediately after the end of the LMS (post-LMS). Finally, the third measurement was performed 15 minutes after the patient arrived in the PACU.

EtCO2 was measured and maintained below 40 mmHg only in the ITGA group by using an ETT. Heart rate (HR) and noninvasive blood pressure (NBP) measurements were conducted every 5 minutes during the entire procedure.

The primary outcome was PaCO2 measured immediately post-LMS (ie, the second ABG measurement time point). Moreover, the secondary outcomes were as follows: (1) pH and PaO2 determined at the second ABG measurement time point; (2) PaCO2, pH, PaO2, and bicarbonate determined at the first and third ABG measurement time points; and (3) HR, systolic BP (BPs), diastolic BP (BPd), mean arterial pressure (MAP), and SpO2 determined at the three measurement periods (pre-LMS period, LMS period, and PACU period).

The patients’ characteristics are presented as absolute and relative percentages for categorical variables and as means and SD for continuous variables. We performed a one-way analysis of variance and Bonferroni post hoc multiple comparisons to compare the mean values of continuous data between the three groups. A chi-square test was used to compare the proportion of categorical variables among the three groups. A p < 0.05 was considered statistically significant. All statistical analyses were performed using SPSS (version 20; IBM, Armonk, NY). On the basis of several studies, we determined the sample size required for this study by using tools provided on ClinCalc.com (https://clincalc.com/stats/samplesize.aspx) to calculate the normal mean range for PaCO2 (50 ± 5 mmHg) and the mean value of the lowest acceptable normal value of PaCO2 (55 mmHg), and alpha is set at 0.05, beta is set at 0.2, and the power is set at 0.8. The minimum required sample size for each group is 16.7,8

From October 2020 to December 2022, a total of 63 patients successfully completed the study protocol. Among them, 18 were assigned to the ITGA group, 21 to the NMB group, and 24 to the NB group (Fig. 1). Two cases from the NB group were excluded due to missing ABG data. There were no significant differences observed between the groups in terms of most patient characteristics (Table 1). Within the three study groups, the number of patients who underwent laser surgery varied: in the ITGA group, there were seven patients; in the NMB group, four patients; and in the NB group, a total of eight patients received laser surgery. Table 2 provides information on the patients’ surgical status and average operation duration times. The NMB group had the highest PaCO2 values of the ABG measurements during LMS, while the ITGA group had the lowest (p < 0.001; Table 3). At the same time point, the NMB group had the significantly lowest mean pH value (p < 0.001; Table 3). The ITGA group had the lowest mean PaO2 value of the ABG measurements during LMS (Table 3). There were no significant between-group differences in the mean SaO2 values of the ABG measurements during LMS (p = 0.055; Table 3). No significant differences of the ABG measurement parameters were observed between pre-LMS and PACU.

Study protocol. ITGA = intubated general anesthesia; NB = nerve block; NMB = neuromuscular blocking.

We also found no significant differences in hemodynamic parameters, including HR, BPs, BPd, and MAP, between the pre-LMS and PACU measurements (Table 4). During the LMS period, the NMB group had the highest mean HR (p < 0.001; Table 4). Similarly, the NMB group had a higher mean BPs value compared to the ITGA and NB groups (p < 0.001; Table 4). However, the ITGA group had a higher mean BPd value than the NMB and NB groups (p = 0.003; Table 4). The mean MAP values during LMS did not differ significantly between the ITGA and NMB groups, but the NB group had a significantly lower MAP value (p < 0.05; Table 4).

In comparing the outcomes of LMS with and without laser surgery within the NMB and NB groups, notable differences emerged. Specifically, significant variations were observed in the PaCO2 and SO2 levels of the ABG measurements during LMS in the NMB group (Supplementary Table S1, https://links.lww.com/JCMA/A231). However, no significant differences were found in the ABG measurements within the NB group. Furthermore, in the realm of hemodynamic measurements during LMS, the NMB group exhibited significant differences in NBPs and SpO2 (Supplementary Table S2, https://links.lww.com/JCMA/A231). In addition, there were marked differences in NBPs, noninvasive blood pressure diastolic (NBPd), and MAP during LMS in the NB group.

The primary aim of this investigation was to assess the effectiveness and safety of HFNO when used in conjunction with SLNB in the context of nonintubated LMS, by examining relevant anesthetic parameters. The analysis indicated that the PaCO2, pH values, and hemodynamic measures during the LMS procedure in the NB group were comparable to those observed in the ITGA group, thereby validating the efficacy and security of SLNB in this surgical context.

Nonintubated LMS combined with HFNO and muscle relaxant was reported in previous study.3 Yang et al reported HFNO with 100% O2 at the flow rate of 50 L/min could maintain adequate oxygenation in apneic patients. But no further information about retention of CO2 was noted in previous study; hence, we planned to find out by arranged ABG measurement right after the end of the LMS in our study. Our ABG measurement results reveal that the degree of CO2 accumulation during LMS was higher in the NMB group than in the ITGA and NB groups. These results are consistent with those of a previous study on HFNO and apnea.9 Booth et al reported a mean (SD) PaCO2 of 89 (16.5) mmHg in the patients with apnea after 30 minutes of GA with HFNO composed of 100% O2 supplied at a flow rate of 70 L/min. In our study, the mean (SD) PaCO2 of the NMB group was 97.5 (24.9) mmHg at the end of the LMS, which is higher than that reported by the aforementioned study; our apnea group also had a shorter apnea time (mean: 15.1 minutes). We speculate that the HFNO flow rate has a certain degree of correlation with the CO2 removal efficiency in patients with apnea. In addition to a shorter operation time, the use of a sufficiently high-flow rate and effective humidity–temperature maintenance may be an effective ventilation strategy for patients with apnea.10,11

Previous trial revealed that the infusion of intraoperative remi