The site of the Media Sphera Publishers contains materials intended solely for healthcare professionals.
By closing this message, you confirm that you are a certified medical professional or a student of a medical educational institution.

A.G. Farshatov

Kirov Military Medical Academy

A.D. Khalikov

St. Petersburg City Clinical Oncology Center

E.N. Ershov

Kirov Military Medical Academy

N.G. Marova

North-Western State Medical University named after I.I. Mechnikov

V.A. Panafidina

Pavlov First St. Petersburg State Medical University

V.V. Mordovin

St. Petersburg City Clinical Oncology Center

A.V. Shchegolev

Kirov Military Medical Academy

Efficacy of high-flow oxygen insufflation during one-lung ventilation


A.G. Farshatov, A.D. Khalikov, E.N. Ershov, N.G. Marova, V.A. Panafidina, V.V. Mordovin, A.V. Shchegolev

More about the authors

Views: 3303

Downloaded: 203

To cite this article:

Farshatov AG, Khalikov AD, Ershov EN, Marova NG, Panafidina VA, Mordovin VV, Shchegolev AV. Efficacy of high-flow oxygen insufflation during one-lung ventilation. Russian Journal of Anesthesiology and Reanimatology. 2022;(5):18‑22. (In Russ., In Engl.)


One-lung ventilation (OLV) is the "gold standard" for thoracoscopic surgery [1, 2]. Lung collapse provides optimal surgical approach [1, 3]. The generally accepted technique for OLV is intubation with a double-lumen endotracheal tube [3]. However, isolation of one lung changes ventilation-perfusion relationship that results significant hypoxemia in 5-10% of cases [4]. Despite small incidence, hypoxemia can be a significant problem affecting the results of surgery and postoperative morbidity [4-6]. OLV and subsequent decrease in oxygen delivery may be followed by damage to vital organs including myocardium. Particularly critical consequences can occur in patients with coronary artery disease [4].

To solve this problem, various authors proposed several methods improving ventilation-perfusion relationship. High-frequency ventilation (HFV) can be used as an alternative to total collapse [7]. In case of high-frequency jet ventilation, intermittent oxygen flow is delivered under certain pressure and frequency that improves oxygenation during OLV. However, this technique also has certain drawbacks. HFV has adverse effect in patients with obstructive ventilation disorders and also increases the risk of intraoperative barotrauma [8].

As an alternative method of ventilation support, oxygen can be delivered with gradual increase of pressure. Independent lung is gradually inflated by oxygen flow and creates constant positive pressure in the airways. In English literature, this method is known as CPAP (Continuous Positive Airway Pressure). This technique is widely used in many hospitals, but there is still no unequivocal opinion on its effectiveness. Some authors propose continuous positive pressure in independent lung over a wide pressure range (5-20 cm H2O) [9]. According to other authors, effectiveness of high continuous positive pressure of 15-20 cm H2O in independent lung regarding improvement of oxygenation does not exceed effectiveness of low pressure (5-10 cm H2O) [10]. Moreover, positive pressure of 15-20 cm H2O leads to active inflation of independent lung and affects local hemodynamics. Inflated alveoli compress intraalveolar capillaries, increase vascular resistance in independent lung and cause significant hemodynamic disorders. In addition, inflated lung fills the pleural cavity and complicates surgical manipulations.

Alternative technique for correction of hypoxemia following OLV is high-flow oxygen insufflation. According to national and foreign scientists, high-flow oxygen insufflation is effective for pneumonia complicated by respiratory failure and weaning from long-term respiratory support. This method also reduces the risk of re-intubation [11-13]. In February 2019, Duwat A. et al. [14] described high-flow oxygen insufflation in 11 patients for correction of hypoxemia following OLV (Journal of Anesthesia and Clinical Research) [14].

Importantly, there are no literature data on various flows of high-flow oxygen insufflation during OLV, as well as their effect on blood gases.

The purpose of the study was to evaluate the effectiveness of high-flow oxygen insufflation in unventilated lung during OLV.

Material and methods

The ethics committee of the Kirov Military Medical Academy approved the study (protocol No. 246 dated December 22, 2020). The study was performed at the St. Petersburg City Clinical Oncology Center. Age of patients was 44-74 years (59.0±9.2). There were 437 thoracoscopic lobectomies in patients with peripheral lung cancer (T1-2N0M0). We analyzed ASA class, electrocardiography data and respiratory function in all candidates.

Inclusion criteria: age over 18 years, thoracoscopic lobectomy, oxygen saturation <92% during OLV for more than 15 min. Non-inclusion criteria: age over 75 years, ASA class > 2 including decompensated preoperative respiratory failure (FEV1/FVC <25%), patient refusal to participate in the study.

Exclusion criteria: extended surgery (bilobectomy/pneumonectomy), hemodynamic instability (mean blood pressure (MBP) <60 mm Hg), forced two-lung ventilation (desaturation after 15 minutes despite additional oxygenation).

We observed desaturation <92% during OLV in 42 patients. Two patients were excluded due to extended surgery and bleeding. Thus, 40 patients were finally enrolled.

Left-sided upper lobectomy was performed in 16 (40%) cases, left-sided lower lobectomy — 8 (20%) cases, right-sided upper lobectomy — 8 (20%) cases, right-sided middle lobectomy — 4 (10%) cases, right-sided lower lobectomy — 4 (10%) cases.

Peripheral venous access was provided in all patients. Radial artery was catheterized for analysis of arterial blood gases and invasive blood pressure measurement. Epidural space (ThV-ThVI) was catheterized too. Catheter was inserted by 4-5 cm in cranial direction. Induction of anesthesia included intravenous administration of propofol 2 mg/kg, fentanyl 3.0 µg/kg and rocuronium bromide 0.6 mg/kg. Direct sequential laryngoscopy and tracheal intubation were performed (insertion of a double-lumen tube with verification of correct position by bronchoscopy). Double-lumen tube size was selected according to the recommendations provided by the manufacturer: women under 160 cm tall — 35 Fr, above 160 cm — 37 Fr, men under 170 cm tall — 39 Fr, above 170 cm — 41 Fr. Maintenance of anesthesia was ensured by sevoflurane at a dose of one minimum alveolar concentration. Mechanical ventilation was performed in volume mode (Primus, Dräger Medical GmbH, Germany). Tidal volume was established at 4-6 ml/kg, appropriate respiratory rate to reach PetCO2 35-45 mm Hg, positive end-expiratory pressure — 5 cm H2O. Baseline FiO2 in air mixture was 0.5. Maintenance of analgesia was provided by prolonged epidural blockade with ropivacaine (12-20 mg/h). The study was divided into three stages (I, II, III). At the 1st stage, we used OLV with FiO2 of 1.0 and collapse of contralateral lung. At the 2nd stage, we continued OLV with FiO2 of 1.0 and initiated high-flow oxygen insufflation into contralateral lung (40 l/min, FiO2 0.5) for 15 min. At the 3rd stage, high-flow oxygen insufflation rate was increased up to 50 l/min with the same other parameters (FiO2 0.5, 15 min).

High-flow oxygen insufflation was performed using AirVo 2 device (Fisher & Paykel, New Zealand) into appropriate lumen of intubation tube through the OPTIFLOW tracheostomy guidewire (Figure). After high-flow oxygen insufflation for 15 minutes, we sampled arterial blood for analysis of gases, recorded ventilation and monitoring data.

High-flow oxygen insufflation through a double lumen tube for OLV using a tracheostomy adapter.

The following parameters were recorded at all stages: arterial oxygen tension (PaO2), arterial carbon dioxide tension (PaCO2), arterial oxygen saturation (SaO2), pulse oximetry parameters (SpO2), MBP. These parameters were recorded at the 15th minute of OLV and 15th minute of high-flow oxygen insufflation with different flow rates (40 and 50 l/min, FiO2 0.5).

Statistical analysis was performed using the SPSS 23 software. Distribution normality was assessed using the Kolmogorov-Smirnov test. In case of normal distribution, we compared dependent samples between various stages using two-way analysis of variance (ANOVA). Friedman test was used for repeated measurements with abnormal distribution or inequality of variances. Wilcoxon test was used to identify significant differences. Data are presented as median and interquartile range — Me (Q1; Q3). Differences were considered significant at p-value <0.05.


Final results are presented in the Table. High-flow oxygen insufflation significantly increased SpO2 at the 2nd and 3rd stages compared to OLV alone (p<0.05). No significant differences were established between the 2nd and 3rd stages. High-flow oxygen insufflation increased PaO2. We found significant differences in this variable between the 2nd and 3rd stages (p<0.05). PaO2 was higher at the 3rd stage.

Laboratory parameters during one-lung ventilation and high-flow oxygen insufflation



High-flow insufflation (40 l/min)

High-flow insufflation (50 l/min)


T; Z

SpO2, %

91.5 (90; 92)* ^

98.0 (98.0; 99.0)*

98.0 (98.0; 99.0)^


210; –3.95

PaO2, mm Hg

67.7 (63.1; 81.2)* ^

98.7 (91.1; 130.1)* #

105.1 (98.9; 141.3)# ^


210; –3.92

SaO2, %

90.4 (89.2; 91.2)* ^

96.8 (96.3; 97.6)* #

97.3 (96.7; 98.1)# ^


210; –3.92

PaCO2, mm Hg

42.5 (40.5; 46.0)

41.5 (40.0; 45.6)

42.2 (40.0; 46.8)


126; –0.78

MBP, mm Hg

67.0 (64.3; 69.0)

69.0 (67.0; 70.0)

69.0 (65.5; 71.8)


126; –1.77

Note. Data are presented as median and interquartile range — Me (Q1; Q3). * — p<0.05 compared to stages I and II; # — p<0.05 compared to stages II and III; ^ — p<0.05 compared to stages I and III.

Comparison of SaO2 revealed significant differences between the 2nd and 3rd stages. This value was the highest at the 3rd stage. SaO2 significantly differed between the 2nd and 3rd stages (p<0.05). Nevertheless, SaO2 was within the reference values. PaCO2 was similar at all stages (p>0.05). Comparison of MBP at all stages revealed no significant differences (p>0.05).


High-flow oxygen insufflation into an unventilated lung during thoracic surgery is effective to maintain pulmonary gas exchange in patients experiencing hypoxemia. Insufflation circuit is connected via the OPTIFLOW adapter (Figure).

A major advantage of high-flow oxygen insufflation is delivery of moistened and heated oxygen-air mixture with low oxygen concentration (0.21-0.50). According to some data, insufflation of 100% oxygen leads to oxygen intoxication, increases oxidative stress and results abnormal changes similar to acute lung injury syndrome [1, 2]. Safe FiO2 is still unclear. Nevertheless, high-flow oxygen insufflation is preferred for critical hypoxemia. One of the fundamental factors of high-flow oxygen insufflation is reduction of dead space volume due to high insufflation flow [15].

We found positive effect of high-flow oxygen insufflation on oxygenation at the 2nd and 3rd stages compared to traditional OLV. There was a significant increase in SpO2, SaO2 and PaO2. No effect on PaCO2 and MBP was found.

Despite significant difference, these values are within reference range at different flows. Perhaps, this difference is associated with higher airway pressure at higher flow. Lampland A.L. et al. [16] found that high-flow oxygen insufflation through the nasal cannulas increased airway pressure at higher flow. Ritchie J.E. et al. [17] used high-flow oxygen insufflation through the nasal cannulas in volunteers (FiO2 0.6 and flow rates of 10, 20, 30, 40 and 50 l/min) with closed and open mouth. Thus, the researchers created open and semi-open circuits. As flow increased, positive airway pressure increased in volunteers with closed mouths. At the same time, positive airway pressure was the same in people with open mouth. We can assume that pressure in semi-open circuit, such as tracheal port of double-lumen tube, correlates with closed-mouth model of the patient and creates mild positive pressure in the independent lung.

Chatila W. et al. [18] found no changes of PaCO2 in patients with low-flow and high-flow insufflation. We also observed that PaCO2 does not change depending on flow rate.


High-flow oxygen insufflation with a flow rate of 40 and 50 L/min during OLV ensures effective correction of hypoxemia during video-assisted thoracoscopic surgery.

The authors declare no conflicts of interest.

Email Confirmation

An email was sent to with a confirmation link. Follow the link from the letter to complete the registration on the site.

Email Confirmation

We use cооkies to improve the performance of the site. By staying on our site, you agree to the terms of use of cооkies. To view our Privacy and Cookie Policy, please. click here.