Background

Obesity is expressed as body mass index (BMI) which is the weight in kilograms divided by the height in square meters (kg/m2) of more than 30. Obesity is associated with multiple comorbidities; one should pay attention to respiratory complications. It affects lung volumes causing restrictive pattern and decreases the functional residual capacity (FRC) to the point that is less than the closing volume resulting in atelectasis and hypoxemia. Obesity also increase the minute ventilation and hence work of breathing, decreases lung compliance, and increases airway resistance. It can result in obstructive sleep apnea (OSA) (Hines and Marschall 2018).

Obesity combined with postoperative respiratory muscle dysfunction may lead to respiratory failure. As a result, obesity is associated with a higher risk of postoperative hypoxemia (Stéphan et al. 2017).

Baltieri et al. (2016) reported a 37% prevalence of atelectasis in obese patients after bariatric surgery in a retrospective observational study. Respiratory complications, on the other hand, are not uncommon in the general surgical population and have been demonstrated to lengthen hospital stays and increase death (Fulton et al. 2018).

High-flow nasal oxygen (HFNO2) therapy provides warmed humidified oxygen and low-level, flow-dependent positive airways pressure, and may be more tolerable than the continuous positive airway pressure (CPAP) or non-invasive ventilation; also, HFNO2 improves washout of nasopharyngeal dead space, resulting in improved oxygenation. In giving prophylactic support to preterm newborns after extubation, HFNO2 has been demonstrated to be both safe and non-inferior to standard CPAP (Zochios et al. 2018).

Aim of the study

The aim of this study is to compare the clinical outcome of treating postoperative laparoscopic sleeve gastrectomy surgery patients having atelectasis by using high-flow nasal oxygen therapy versus Venturi mask oxygen therapy.

Methods

After the approval of our institutional ethics committee, number FMASU M D 239/2019 and FMASU M D 239a/2019/2020, this prospective randomized, controlled, unblinded, single-center clinical trial was conducted over 110 patients (55 patients in each group) for 18 months from December 2019 to June 2021. Written informed consent was obtained from the patients or the first kin relative. The study protocol was explained to the patients before taking their informed consent.

Inclusion criteria

All patients had BMI above 40 kg/m2 from both genders, aged 18–60 years old, with preoperative Physical Status ASA III, underwent laparoscopic sleeve gastrectomy who had postoperative atelectasis confirmed clinically, by chest X-ray (CXR) and lung ultrasound in the ICU. All patients were fully conscious upon ICU admission.

Exclusion criteria

Patients were excluded due to refusal of the intervention or participation in the study, age below 18 years old and above 60 years old, pregnancy or lactation, psychiatric illness, or known comorbidities such as chronic pulmonary diseases and cardiac diseases.

All patients were admitted to the ICU for postoperative care and were randomized using a randomization table created by a computer software program, allocated by the holder of the sequence who was situated off-site and assigned in a 1:1 ratio to one of the following two groups:

  • Group A: HFNO2 therapy group: patients who were randomized to high-flow nasal oxygen therapy.

  • Group B: VMO2 therapy group: patients who were randomized to venturi mask oxygen therapy.

In the operating theater, all patients were premedicated with intravenous (IV) 8 mg ondansetron and 8 mg dexamethasone. Standard monitors were applied, an ECG, pulse oximeter, non-invasive arterial blood pressure monitor, and the arterial line was inserted. Preoxygenation was carried out for 3 min by means of a face mask with 100% oxygen. Anesthesia was classically induced by IV fentanyl 1–2 μg/kg, propofol 2 mg/kg, and atracurium 0.5 mg/kg. After endotracheal intubation, capnography was applied. The lungs were ventilated with a tidal volume of 6–8 ml/kg, FiO2 0.6, and the respiratory rate was adjusted to maintain ETCO2 between 35 and 40 mmHg. Maintenance was done with 1.5 MAC isoflurane and top-up doses of atracurium every 20 min. At the end of the surgery, the muscle relaxant was reversed and all patients were extubated and sent to the intensive care unit (ICU). All patients received 5 mg IV nalbuphine for analgesia when the pain was present (postoperative pain was assessed using a visual analog scale (VAS)) donated by VAS 4–10.

Upon ICU admission, medical history and examination of all patients were done; standard monitors were attached including continuous electrocardiogram, non-invasive blood pressure, and pulse oximeter; the arterial line was inserted in the operating room and was used for sampling (Vygon Leadercath Arterial PE - UK); arterial blood gas was obtained, CXR (13) (Parke et al. 2014); and lung ultrasound (15, Table 1) (Lee 2016, Mongodi et al. 2017) was done for assessment of atelectasis. Patients were monitored at least every 1h (for monitoring but not all readings were analyzed).

Table 1 Lung ultrasound score calculation
Full size table

In the HFNO2 group, the HFNO2 cannula (AIRcon gen Respiratory Humidifier WILAmed, WILAmed high-flow AIR/O2 blender with flowmeter and Oxi. Plus nasal high flow kit) was applied at a flow of 30 L/min at FiO2 of 0.6 at a temperature of 36 oC. Flow and FiO2 were adjusted according to the attached HFNO2 therapy Algorithm, Fig. 1, Table 2. Weaning was also be adjusted according to it. For the VMO2 group, the venturi mask with FiO2 0.6 was applied. The FiO2 was adjusted according to the attached venturi mask oxygen therapy algorithm, Fig. 2.

Fig. 1
figure 1

High-flow nasal oxygen therapy algorithm. *1: Thoraco-abdominal asynchrony and/or persistent auxiliary muscle use. *2: The table shows escalation and de-escalation of oxygen in the HFNO2 group (Table 2). *3: Respiratory rate <25 breath/min or SpO2 ≥90%, no increased work of breathing. ABG, no respiratory acidosis (PH>7.35), no CO2 rise for > or = 10 mmHg; HFNO2, high-flow oxygen therapy; VMO2, venturi mask oxygen therapy; NIPPV, non-invasive positive pressure ventilation; IPPV, invasive positive pressure ventilation; SpO2, oxygen saturation

Full size image
Table 2 The table shows escalation and de-escalation of oxygen in the HFNO2 group
Full size table
Fig. 2
figure 2

Venturi mask oxygen therapy algorithm. NIPPV, non-invasive positive pressure ventilation; IPPV, invasive positive pressure ventilation; SpO2, oxygen saturation

Full size image

Data collection, measurements, and outcome

Upon ICU admission age, weight, BMI, sex, and duration of surgery were recorded. During ICU stay, the respiratory rate was recorded on admission, 1, 2, 3, 4, 8, 12, 16, 20, and 24 h, and ABGs and PaO2/FiO2 were recorded on admission, 2, 4, 8, 12, and 24 h. Chest X-ray (CXR) was done on admission assessed by Modified RAS score (13) (Parke et al. 2014). Lung ultrasound (Mindray M5 Diagnostic Ultrasound System (China)) was done on admission (using both 3C5s and L14-6s probes) as well assessed by lung U/S score for regional atelectasis (15, Table 1) (Mongodi et al. 2017). The primary outcome of this study was to compare PaO2/FiO2 during 24 h of oxygen therapy in both groups, and the secondary outcome was to compare the respiratory rate during 24 h and the length of ICU stay.

Statistical method

Data were collected, coded, tabulated, and then analyzed using the SPSS software package (SPSS for Windows®, Version 16.0. Chicago, SPSS Inc.). Numerical variables were presented as mean (standard deviation), and categorical variables were presented as frequency (%). Between-group comparisons were done using unpaired t test and Fisher’s exact test, for numerical variables and categorical variables, respectively. Repeated-measured variables were analyzed using repeated-measures ANOVA. Sphericity assumption of repeated-measures ANOVA was tested using Mauchly’s test of sphericity. A greenhouse-Geisser correction was applied whenever a lack of sphericity was evident. Error bars represent a 95% confidence interval. Any difference with a p-value < 0.05 was considered statistically significant.

Sample size

The sample size was calculated using the STATA program, setting the type-1 error (α) at 0.05 and the power (1-β) at 0.8. The result from a previous study (Testa et al. 2014) showed that the mean PaO2/FiO2 was 140±90 among conventional oxygen therapy compared to 190±100 among the HFNO2 group. Calculation according to these values produced a minimal sample size of 52 cases per group approximated to 55 per group (total 110).

Results

Between December 2019 and June 2021, 122 patients underwent laparoscopic sleeve gastrectomy surgery with successful extubation and 12 were excluded as shown in the CONSORT flow chart (Fig. 3 and Table 3).

Fig. 3
figure 3

Flow chart for patient enrollment

Full size image
Table 3 Baseline characteristics of the patients participating in the study presented as mean ± SD or frequency (%) as appropriate
Full size table

PaO2 was found to be significantly higher 131.764 (95% CI 124.562–138.965) in the HFNO2 group versus 106.767 (95% CI 99.565–113.968) in the VMO2 group (p-value < 0.001). PaO2/FiO2 was found to be significantly higher 321.81 (95% CI 307.486–336.153) in the HFNO2 group versus 276.767 (95% CI 262.433–291.100) in the VMO2 group (p-value < 0.001). The PaO2/FiO2 was maintained in both groups till the 8 h reading. That was followed by a comparable gradual increase in PaO2/FiO2 in both groups (p-value > 0.05), but higher in the HFNO2 group than in the VMO2 group (p-value < 0.05) as shown in Fig. 4 and Table 4.

Fig. 4
figure 4

PaO2/FiO2 throughout the study time is significantly higher in the HFNO2 group. The red marks mean that the p-value was found to be significant

Full size image
Table 4 PaO2/FiO2 is significantly higher in the HFNO2 group
Full size table

The respiratory rate was found to be significantly lower 20.778 (95% CI 20.172–21.385) in the HFNO2 group versus 24.047 (95% CI 23.441–24.654) in the VMO2 group, noted from the first hour (p-value < 0.001). That was followed by a continuous drop till 12 h after initiation. Then, the respiratory rate was maintained till the end of the study in both groups as shown in Fig. 5 and Table 5.

Fig. 5
figure 5

The respiratory rate is significantly lower in the HFNO2 group. The ICU stay is statistically signi. The red marks mean that the p-value was found to be significant

Full size image
Table 5 The respiratory rate is significantly lower in the HFNO2 group. The ICU stay is statistically significantly
Full size table

The length of the ICU stay was 1.09 ± .29 days in the HFNO2 group when compared to 1.00 ± .000 day in the VMO2 group (p-value 0.002), but this difference is about 2 h and 10 min as shown in Table 5.

Discussion

In our study, postoperative laparoscopic sleeve gastrectomy surgery patients with confirmed atelectasis by chest auscultation, mRAS score, and lung ultrasound examination (lung US score) were randomly assigned to either the HFNO2 group or VMO2 group. The study showed that HFNO2 resulted in a significantly higher oxygenation represented by PaO2, PaO2/FiO2, and effectively decrease the respiratory rate.

Comparing both groups across the first 24 h postoperatively, they showed a statistically significant increase in PaO2/FiO2 along the 2-h, 4-h, and 24-h values with p-value < 0.05 as shown in Table 4 and Fig. 4. They also showed a statistically significant decrease in respiratory rate along the first 12 h with a p-value < 0.05 as shown in Table 5 and Fig. 5.

Maurizio et al. study showed that oxygenation for the same set FiO2 was improved by HFNO2 therapy when compared to the venturi mask in patients with acute respiratory failure in the post-extubation period, and PaO2/FiO2 was 287.2 ± 74.3 versus 247.4 ± 80.6 mmHg (p-value 0.03). In addition to that, the HFNO2 therapy decreased the respiratory rate with a mean difference of 4 ± 1 breaths/minute (Maurizio et al. 2014).

Yu et al. underwent a multicenter randomized interventional trial showed that the application of HFNO2 therapy to patients who underwent thoracoscopic lobectomy after the extubation could decrease the risk of hypoxemia (29.62% with conventional oxygen therapy, 12.51% with HFNO2) and reintubation as well as improve oxygenation represented by PaO2, PaO2/FiO2, and SaO2/FiO2 when compared to conventional oxygen (p-value < 0.05) (Yu et al. 2017).

Testa et al. found that the HFNO2 therapy was not able to affect the partial pressure of carbon dioxide in the arterial blood (PaCO2) in the pediatric population following open-heart surgeries. PaCO2 was used as a primary endpoint as the patients had both cyanotic and cyanotic heart diseases; however, HFNO2 was found to improve PaO2 levels in both categories, and PaO2 was significantly higher in HFNO2 (p-value 0.01). PaO2/FiO2 was found to be also statistically significant (p-value < 0.001) (Testa et al. 2014).

According to Corley’s study, using the HFNO2 therapy resulted in a decrease in respiratory rate by 3.4 breaths/min (95% CI 1.7–5.2) and improved oxygenation. Thus, HFNO2 therapy may be a useful treatment option for patients experiencing respiratory dysfunction post-cardiac surgery, especially those who cannot tolerate non-invasive ventilation and those with BMI ≥ 30 kg/m2 (Corley et al. 2011).

In contradiction to the previous studies, in patients with a BMI ≥ 30 kg/m2, direct extubation onto HFNO2 therapy following cardiac surgery did not improve atelectasis, oxygenation (PaO2/FiO2 in the 24 h post-extubation with a mean of 227.9 for HFNO2 and 253.3 for the control group with p-value 0.08), and respiratory rate (mean was 16.7 in control the group and 17.24 in HFNO2 with a difference of 0.54 and p-value 0.17) when compared to standard oxygen therapy, or reduce the need for escalation of respiratory support, a result found by Corley et al. They suggested further research clarifying the role of HFNO2 therapy (Corley et al. 2015).

As regards the length of the ICU stay in our study, there was the significantly longer length of ICU stay (1.09 ± .29 days) in the HFNO2 group when compared to (1.00 ± .000 day) in the VMO2 group, yet it is clinically insignificant as the increased duration in HFNO2 group can be attributed to the longer weaning of high-flow nasal oxygen therapy applied in our study protocol.

According to Xiang et al., when compared to conventional oxygen therapy, HFNO2 therapy reduced the rate of intubation or non-invasive ventilation for respiratory failure in postoperative patients at high risk of pulmonary complications, but did not reduce the length of stay in the hospital or ICU, the rate of oxygen requirement after discontinuation, or hypoxemia. Although HFNO2 does not reduce mortality, there was little harm associated with its usage, suggesting that it could be a better alternative to conventional oxygen therapy in postoperative patients at high risk of pulmonary complications. Future studies should concentrate on determining which subgroups of postoperative patients are most likely to benefit from HFNO2 therapy (Xiang et al. 2021).

Study limitations

Our study has some limitations, it has a relatively short duration and was limited to 24 h. PaCO2 was not analyzed and the 2-h difference in the ICU stay may not be attributable to the interventions being practiced, but due to logistic issues or the longer weaning hours of the HFNO2 as mentioned in its algorism, the pain assessment was not analyzed and patient satisfaction was not studied, data about complications was not collected, follow-up with CXR and lung U/S were not recorded and analyzed, and hospital stay was not studied; these data can be considered in future studies.

Conclusions

In conclusion, our study has shown that high-flow nasal oxygen therapy in postoperative laparoscopic sleeve gastrectomy patients with atelectasis-maintained oxygenation represented as PaO2 and PaO2/FiO2 higher than the venturi mask and significantly decreased the respiratory rate but did not decrease the length of the ICU stay when compared to venturi mask oxygen therapy.