Effect of high‐flow nasal oxygen on postoperative oxygenation in obese patients: A randomized controlled trial

Abstract Background and Aim Postoperative hypoxemia is common after general anesthesia in obese patients. We investigated if early application of high‐flow nasal oxygen (HFNO) improved postoperative oxygenation in obese patients compared with standard oxygen therapy following general anesthesia for laparoscopic bariatric surgery. Methods This was an open labeled randomized controlled trial conducted at a university hospital in Sweden between October 23, 2018 and February 11, 2020. The study was performed as a substudy within a previously published trial. After ethics committee approval and written informed consent, 40 obese patients (body mass index [BMI] ≥ 35 kg m−2) scheduled for laparoscopic bariatric surgery were randomized to receive oxygen using a standard low‐flow nasal cannula (NC group) or HFNO at 40 L min−1 (HF group) immediately upon arrival to the post‐anesthesia care unit. Flow rate (NC group) or FiO2 (HF group) was titrated to reach an initial SpO2 of 95%–98% after which settings were left unchanged. The primary outcome was PaO2 at 60 min following postoperative baseline values. Secondary outcomes included PaCO2, SpO2, hemodynamic variables, and patient self‐assessed discomfort. Results Thirty‐four patients were available for analysis. PaO2 was similar between groups at postoperative baseline. After 60 min, PaO2 had increased to 12.6 ± 2.8 kPa in the NC group (n = 15) and 14.0 ± 2.7 kPa in the HF group (n = 19); (mean difference 1.4 kPa, 95% confidence interval −0.6 to 3.3; p = 0.16). There were no differences in PaCO2, hemodynamic variables, or self‐assessed discomfort between groups after 60 min. Conclusion In obese patients, HFNO did not improve postoperative short‐term oxygenation compared with standard low‐flow oxygen following general anesthesia for laparoscopic bariatric surgery.


| INTRODUCTION
After general anesthesia, disconnection from the ventilator circuit results in loss of positive airway pressure that may lead to lung derecruitment and atelectasis-related shunt, 1,2 especially in obese patients. 3,4 This is a major cause of postoperative hypoxemia and the cause of considerable postoperative morbidity 5,6 ; hence there is a need for preventive strategies. 7 In patients with hypoxemic respiratory failure after abdominal surgery, the application of continuous positive airway pressure (CPAP) reduces the need for intubation and mechanical ventilation. 8 Although high-flow nasal oxygen (HFNO) likewise increases airway pressure and therefore may prevent atelectasis formation, early application of this technique in patients with body mass index (BMI) < 35 kg m −2 failed to reduce the incidence of postoperative hypoxemia after major abdominal surgery compared with standard low-flow oxygen. 9 Furthermore, after cardiothoracic surgery, HFNO did not improve atelectasis formation on chest X-ray compared with standard nasal cannula. 10 In obese patients undergoing laparoscopic bariatric surgery, CPAP improves postoperative oxygenation, and immediate application after extubation maintains respiratory function better than application in the post-anesthesia care unit (PACU). 11,12 Contrary to the findings in studies of patients with normal BMI, HFNO has been reported to reduce the risk of postoperative hypoxemia in obese patients compared with standard oxygen supplementation when HFNO was applied in the operating room before extubation and maintained during transport to the PACU. 13 Although this strategy eliminates time without positive airway pressure and presumably reduces the risk of lung derecruitment, the transport of patients with HFNO remains challenging in clinical practice. The effect of HFNO on postoperative oxygenation in obese patients when applied early instead of immediately after extubation has not been well characterized.
We hypothesized that HFNO would improve postoperative oxygenation in obese patients compared with standard low flow oxygen therapy when applied at arrival to the PACU. The primary aim of this trial was to compare oxygenation using these two strategies during the first postoperative hour in obese patients recovering from general anesthesia for laparoscopic bariatric surgery.

| Participants
Adult patients, aged 18-60 years, scheduled for elective laparoscopic bariatric surgery were eligible for study inclusion if they had a BMI ≥ 35 kg m −2 . Patients were excluded if they had an ASA Class > 2 (not assessed by BMI), chronic obstructive pulmonary disease or asthma causing limitations in everyday activities, heart failure with clinical symptoms equivalent to New York Heart Association functional classification > 2, restrictive lung disease with > 20% reduction in total lung capacity, allergy to anesthetic agents used in the study, or if they could not comprehend oral or written information.

| Study prerequisites, randomization, and masking
This trial was performed as a sub-study within a previously published trial 14 in which patients were randomized to receive preoxygenation with a FiO 2 of 1.0 using face mask and positive end-expiratory pressure (PEEP) of 7 cmH 2 O or HFNO with 70 L min −1 before induction of anesthesia. Patients randomized to face-mask with PEEP were also allocated to postoperative oxygen supplementation with a standard nasal cannula (NC group), whereas patients randomized to preoxygenation with HFNO were allocated to receive postoperative oxygen supplementation with HFNO (HF group).
Randomization was performed before induction of anesthesia in an allocation ratio of 1:1 and a block size of two by opening sealed opaque sequentially numbered envelopes. There was no blinding. Flow rate (NC group) or FiO 2 (HF group) was titrated to reach a SpO 2 of 95%-98% and settings were thereafter left unchanged during the study period. The PACU was located near the operation room.

| Anesthesia management
All patients were premedicated with Paracetamol 1 g and Oxycodone

| Outcomes
The primary outcome was PaO 2 60 min following completed titration of allocated oxygen supplementation in the PACU. Secondary outcomes were SpO 2 , PaCO 2 , pH, noninvasive blood pressure (NIBP), heart rate at 60 min. Further secondary outcomes were incidence of SpO 2 < 90% and the level of comfort using either intervention.

| Measurements
Arterial blood was sampled 10 min after completed titration of allocated oxygen therapy in the PACU, and at 30 and 60 min thereafter. Blood samples were analyzed with a point of care device (Abbot i-STAT ® 1, Abbott Laboratories). SpO 2 , NIBP, and heart rate were recorded with a standard bedside monitor (Dräger Infinity Delta ® monitor, Dräger Medical System Inc.) at the same time points as arterial blood sampling. Duration and incidence of SpO 2 < 90% and information on opioid requirements were extracted from the electronic medical record (Metavision ® ). The level of discomfort in regard to the interventions was assessed using a four-step ordinal scale defined as none, mild, moderate, or severe.

| Sample size calculation
The sample size calculation was performed for the original trial. 19 The sample size for the present study may therefore be considered a convenience sample. However, to confirm adequate power in the current investigation, a post hoc estimation of the minimal detectable effect was performed. Considering the sample size and the dispersion estimates of the final analysis, the present study had an 80% power to detect a 20% difference in PaO 2 at 60 min after intervention initiation at α of 0.05.

| RESULTS
Of 40 randomized patients, 15 patients were available for final analysis in the NC group and 19 patients in the HF group ( Figure 1).
In the NC group, two subjects withdrew consent and arterial access could not be achieved in two additional subjects. One additional patient in each group was excluded due to a protocol violation.
Patients in the HF group were a mean of 5 years older compared with controls, and five (26%) patients in the HF group were active or previous smokers compared with 2 (13%) in the NC group. There were furthermore patients with asthma in the HF group, however, all had mild disease and were asymptomatic on the day of the study.

Subject characteristics and intraoperative respiratory measurements
were otherwise similar between groups (Table 1).

| Primary outcome
At baseline, 10 min after completed titration of the assigned postoperative oxygen therapy PaO 2 was 11.9 ± 3.2 kPa in the NC group and 11.9 ± 2.3 kPa in the HF group. After 60 min, PaO 2 had increased to 12.6 ± 2.8 kPa in the NC group and 14.0 ± 2.7 kPa in the HF group (Figure 2), which was not statistically significant (mean

| Adverse events
Five (15%) patients had SpO 2 lower than 90% during the study period. Two patients in the NC and HF groups respectively had one shorter (less than 2 min) episode of desaturation below 90%. None of these patients had a SpO 2 of less than 85% at any timepoint. One patient in the HF group had 10 desaturation episodes below 90% and four of these were less than 85%.

| DISCUSSION
The main finding of this study was that early application of postoperative HFNO after general anesthesia for laparoscopic bariatric surgery failed to improve oxygenation compared with standard nasal cannula in patients with BMI ≥ 35 kg m −2 .
HFNO generates positive airway pressure directly correlated to flow rate and inversely related to mouth opening. [20][21][22] Furthermore, HFNO improves respiratory mechanics and gas exchange due to expansion of the functional residual capacity. 23,24 These physiological findings suggest that HFNO could be beneficial for postoperative oxygen supplementation in obese patients. However, the airway pressures generated by HFNO are moderate at best 22 and if patients F I G U R E 1 Consolidated Standards of Reporting Trials (CONSORT) flow chart of enrolled, randomized, and analyzed participants. BMI, body mass index; HFNO, high-flow nasal oxygen are breathing through the mouth, the PEEP effect is nearly nullified. 20 This may in part explain the neutral findings of our study.
Obese patients have reduced FRC at baseline compared with non-obese, and are at risk of further reduction during anesthesia which may lead to atelectasis development, decreased lung compliance, and oxygenation difficulties. [25][26][27] Although an intraoperative open lung approach, which may include lung recruitment maneuvers and PEEP titration, improves respiratory mechanics during anesthesia, positive airway pressures are lost following extubation and the beneficial effects from intraoperative ventilation approach on respiratory mechanics are attenuated. 28    half-life 30 that facilitates rapid recovery and may be more in line with current practice.
Corroborating the results of our study, a recently published investigation reported that HFNO failed to improve postoperative oxygenation compared with conventional oxygen therapy up to 6 h following bariatric surgery. 31  although not related to this study.

DISCLOSURES
The corresponding author affirms that this manuscript is an honest, accurate, and transparent account of the study being reported; that no important aspects of the study have been omitted; and that any discrepancies from the study as planned (and, if relevant, registered) have been explained

TRANSPARENCY STATEMENT
We confirm that this manuscript is an honest, accurate, and transparent account of the study being reported; that no important aspects of the study have been omitted; and that any discrepancies from the study as planned (and, if relevant, registered) have been explained.

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.