Extubation success without using high-ow nasal cannula and non-invasive ventilation for patients with coronavirus disease 2019 (COVID-19): A retrospective observational study

Background In March 2020, the World Health Organization declared coronavirus disease 2019 (COVID-19) to be a global pandemic. It was also reported that approximately 20%-23% of patients with COVID-19 admitted to the hospital required mechanical ventilation (MV). Reintubation rates ranged from 13–19%, leading to increased mortality. Moreover, exposure of health care workers is a serious problem and intubation is a high-risk procedure because of aerosol generation. To reduce reintubation, preventive high-ow nasal cannula (HFNC) and non-invasive ventilation (NIV) after extubation are recommended. However, using HFNC and NIV in patients with COVID-19 may lead to the spread of infection. Without using HFNC and NIV, conventional spontaneous breathing trial (SBT) may overestimate a patient’s respiratory status. A highly demanding SBT accurately assesses a patient’s reserve capacity and may reduce reintubation. We did not use preventive HFNC or NPPV and extubated with a highly demanding SBT protocol. This study aimed to evaluate whether our protocol can be performed safely. Methods This single-centre study was conducted at the Kawasaki Municipal Hospital from February 2020 to June 2020. Patients with COVID-19 requiring MV were screened for their eligibility for the weaning procedure every day. Patients were extubated after administering methylprednisolone if SBT was well tolerated. SBT was performed with 3 cmH 2 O pressure support, 3 cmH 2 O positive end-expiratory pressure, and ≤ 0.30 fraction of inspired oxygen. In extubated patients, the reintubation rate, duration of MV, length of intensive care unit (ICU) and hospital stay, and rate of ventilator-associated pneumonia (VAP) were analyzed. Results Ten patients received MV. Two of 10 patients had no improvement in their respiratory status and did not qualify for the extubation criteria and died. Eight patients were extubated. No reintubations were performed (0%). The median durations

Reintubation would re-expose health care workers to the virus, all efforts should be made to avoid reintubating patients with COVID-19.
In the previous study, it was reported that preventive high-ow nasal cannula (HFNC) therapy and noninvasive ventilation (NIV) after extubation reduce the risk of reintubation for patients at a high risk of extubation failure who have been receiving MV for more than 24 h and who have passed a spontaneous breathing trial (SBT) [12][13].
The American College of Chest Physicians/American Thoracic Society (ACCP/ATS) guidelines [14] recommend preventive non-invasive ventilation after extubation for patients at a high risk for extubation failure who are on MV for more than 24 h and have passed a SBT.However, in patients with COVID-19, the infection-related risks of HFNC therapy and NIV are not well known.Some reports suggest that the use of HFNC and NIV in patients with COVID-19 should be discouraged because of the risk of aerosol generation and exposure of health care workers [15][16].Although no studies have directly evaluated the risk of aerosol generation or infection transmission associated with HFNC and NIV use in patients with COVID-19 [17][18][19][20][21][22][23][24], the use of NPPV and HFNC aggravates this risk through continuous open circuit viral aerosolization.If the patient is not in a negative pressure room, viral shedding through aerosol generation can extend beyond the room, thereby contaminating the hallways and beyond.Therefore, the use of NPPV and HFNC for patients with COVID-19 should be minimized and restricted [16].
The ACCP/ATS guidelines [14] recommend that the initial SBT be conducted with inspiratory pressure augmentation (5-8 cmH 2 O) rather than without inspiratory pressure augmentation (T-piece or continuous positive airway pressure [CPAP]) in acutely hospitalized patients ventilated for more than 24 h.This recommendation places a high value on reducing the duration of MV and maximizing the probability of extubation success.These reports are based on using HFNC and NPPV after extubation.Without the use of HFNC or NPPV, conventional SBT may overestimate a patient's respiratory status.When we extubate patients with COVID-19, a more accurate assessment of the respiratory status is needed.Some studies reported that a pressure support (PS) of 5 cmH 2 O reduces the respiratory workload by 31%-38% [25][26].
Other study reported that a positive end-expiratory pressure (PEEP) of 5 cmH 2 O reduces the respiratory workload by up to 40% in some patients [27][28][29].
SBT at PS 5 cmH 2 O and PEEP 5 cmH 2 O may overestimate a patient's reserve capacity.Physiologically minimal PS and PEEP may be lower than conventional SBT.
A meta-analysis that assessed the effort to breathe using various SBT techniques suggested that highly demanding SBT techniques, such as those with low PS and PEEP, or T-piece accurately re ected a patient's reserve capacity after extubation [30].
A highly demanding SBT accurately assesses a patient's reserve capacity and may reduce reintubation rates in patients with COVID-19 without use of HFNC or NPPV.We did not use preventive HFNC or NPPV and extubated patients with COVID-19 with a highly demanding SBT protocol.This study aimed to evaluate whether extubation with a highly demanding SBT protocol can be performed safely.

Population and study setting
This single-centre study was conducted at the Kawasaki Municipal Hospital, which is an emergency and critical care centre in Kawasaki City, located in the south of Kawasaki City in Kanagawa Prefecture.
Kawasaki City is a small city with an area of 144 km 2 and a population of 1,500,000.Kawasaki City has one of the highest population densities among all cities in Japan.We engaged in the treatment of patients with COVID-19 in the Diamond Princess at the port of Yokohama in Kanagawa.From February 6, 2020, our hospital had 10 beds for treating severe COVID-19.

Study protocol
Inclusion criteria included patients who were screened and diagnosed with COVID-19 between February 6, 2020 and June 30, 2020, who required more than 24 h of MV.Exclusion criteria was age < 18 years.COVID-19 was diagnosed in accordance with the WHO guidelines [1].
The ow chart of the extubation protocol is shown in Fig. 1.Patient eligibility for the extubation procedure was identi ed by daily screening and SBT.Daily screening was deemed to start immediately after admission.Daily screening was as follows: fraction of inspired oxygen (FiO 2 ) < 40%; PEEP ≤ 5 cmH 2 0; with no or small sedation; and responding to simple orders.Patients who met the daily screening criteria were assessed using SBT.From an infection control standpoint, we avoided the use of T-piece and SBT was performed under the following conditions: 3 cmH 2 O pressure support ventilation, 3 cmH 2 O PEEP, and FiO 2 ≤ 0.30.SBT was successful when the patient could breathe spontaneously for 30 min without clinical intolerance.
Clinical intolerance was de ned as a partial pressure of arterial oxygen (PaO 2 ) ≥ 60 mmHg; spontaneous tidal volume ≥ 4 ml/kg predicted body weight; respiratory rate ≤ 35/min; pH ≥ 7.3; and no respiratory distress (distress = 2 or more).Respiratory distress was de ned as heart rate > 120% of the baseline; marked accessory muscle use; abdominal paradox; diaphoresis; and marked dyspnea.These clinical intolerances were referenced based on the ARDS Net recommendation [31].Patients were extubated if they passed SBT.We administered 20 mg of methylprednisolone every 4 h starting 12 h before extubation [32].From an infection control standpoint, we did not enforce cuff leak test.If SBT was not tolerated, the patient was returned to their prior ventilator settings and screened the day after.Successful discontinuation of MV was de ned as continuous independence from ventilator support for a period of at least 48 h.Reintubation was performed if any of the following signs were noted: suspected airway stenosis or PaO 2 < 60 mmHg with O 2 approximately 5 L (Fig. 1).

Data collection
The medical records of patients were analyzed by the COVID-19 research team from the Kawasaki Municipal Hospital.Data of patients who were deemed appropriate for extubation according to the protocol were collected.Data on age, sex, nationality; comorbidities; signs and symptom at the time of onset; laboratory ndings; complications, including acute respiratory distress syndrome (ARDS) and ventilator-associated pneumonia (VAP); treatments, including antiviral drugs, antibacterial drugs, corticosteroid therapy, and renal replacement therapy; and outcomes, were obtained from electronic medical records using data collection forms.
Sequential Organ Failure Assessment (SOFA) scores and Acute Physiology and Chronic Health Evaluation II (APACHE II) scores were assessed on the day of ICU admission.Partial pressure of arterial oxygen/fraction of inspired oxygen (P/F ratio) immediately before extubation was determined.ARDS was de ned and classi ed according to the Berlin de nition [33].VAP was de ned as the initiation of antibiotics and ful lment of at least 2 of the following criteria: positive protected bronchoscopy cultures; fever; or increasing leucocyte count [34].

Presenting characteristics
Of all patients admitted to our hospital for COVID-19 between February 1, 2020 and June 30, 2020, 10 patients received MV.Two of the 10 patients had no improvement in their respiratory status and did not qualify for the extubation criteria and died.Eight patients were extubated (Fig. 2).Patient characteristics are given in Table 1.The mean age of the patients was 53.6 ± 11.9 years, and 8 patients (100%) were Japanese males.Seven patients (87.5%) had at least one comorbidity.The most common comorbidity was hypertension in 3 cases (37.5%), followed by diabetes mellitus in 2 cases (25%), asthma in 2 cases (25%), chronic kidney disease in 2 cases (25%), and cardiovascular disease in 1 case (12.5%).The most common symptoms at the onset were fever in 5 cases (62.5%) and respiratory distress in 5 cases (62.5%), dysgeusia in 2 cases (25%), and cough in one case (12.5%).At the time of data collection, all the patients were either discharged or deceased.The results of laboratory tests are given in Table 2.The mean leukocyte count was 6.9 ± 2.0 × 10 3 /µL, and the lymphocyte count was reduced in all 8 (100%) cases.D-dimer, aspartate aminotransferase, alanine aminotransferase, lactate dehydrogenase (LDH), ferritin, and C-reactive protein (CRP) levels were elevated in 8 patients (100%).The mean APACHE II and SOFA scores (SD) were 4.1 ± 1.2 and 10.6 ± 2.4, respectively.The mean (SD) value of the P/F ratio immediately before extubation was 266.9 ± 47.7.Data are presented as mean ± standard deviation.ICU: intensive care unit, P/F ratio: ratio of partial pressure of arterial oxygen to the fraction of inspired oxygen

Complications and treatments
Complication and treatment details are given Table 3. Complications included ARDS in 8 patients (100%) and VAP in 3 patients (37.5%).Seven patients had moderate ARDS and 1 had severe ARDS.All patients (100%) received antibacterial therapy, and 1 patient (12.5%) received antiviral therapy (favipiravir).Two patients (25%) received glucocorticoid therapy; both these patients were diagnosed as having asthma and were treated with glucocorticoid therapy.One patient (12.5%) was treated with continuous renal replacement therapy.

Discussion
In our study, all the patients were male and 87.5% of them had comorbidities, all patients had ARDS complications, and blood samples showed reduced lymphocyte and elevated CRP, LDH, and D-dimer levels.Previous studies have reported that older age, being male, comorbidities, ARDS, reduced lymphocyte, and elevated CRP, LDH, and D-dimer levels correlated with the disease severity in patients with COVID-19 [35][36][37][38][39][40].
In the present study, we performed highly demanding SBT with 3 cmH 2 O PEEP, 3 cmH 2 O PS, and FiO 2 0.3 and administered steroids to prevent laryngeal edema.All patients were extubated without the use of HFNC or NIV.Previous studies have reported a reintubation rate of 13%-19% [4][5][6] while in this study reintubation requirement was nil (0%).Highly demanding SBT possibly reduces reintubation.However, the duration of MV in our study tended to be the same or longer than that in previous studies.In a retrospective case series of 1000 patients in New York, the median duration of MV in patients who survived was 9 (6.5-12) days [3], and another multicentre cohort study of 2215 patients in New York reported that the median duration of MV in patients who survived was 13 (9)(10)(11)(12)(13)(14) days [41].The median duration of MV for patients in the current study was 13 (IQR, 10-16) days.Some previous studies have reported the length of ICU stay to be approximately 10-20 days [42][43], and the length of ICU stay in the current study was not different from these studies.Although the length of ICU stay was not prolonged, the duration of MV was tended to prolonge.The protocol used in this study may lead to a prolonged duration of MV.
In a study involving limited patients, the rate of VAP among patients with COVID-19 was approximately 12% [44].In our study, the incidence rate of VAP was 37.5%.
The incidence of VAP is reported to vary over time ranging from approximately 3% per day in the rst week of mechanical ventilation, 2% per day in the second week to 1% per day in the third week and beyond [45].There are concerns that our protocol led to a prolonged duration of MV and an increase in the rate of VAP.
It is possible that a highly demanding SBT may lead to reduction in reintubation rates in COVID-19 patients when HFNC and NPPV cannot be used.We successfully extubated all patients without any increase in reintubation rates; however, the use of our protocol may have resulted in prolonged duration of MV and an increased VAP rate.
This study had some limitations.First, only 8 patients were included.This study had a small sample size.
Second, this study used a retrospective observational and non-interventional design, and it did not have a comparator.Third, this study was conducted at a single centre in Kawasaki, Japan.Hence, the results might not be generalizable to other institutions.Fourth, because this study is not a comparative study and does not compare to PEEP/PS used in other studies, a larger number of patients and comparators are needed to examine whether the PS and PEEP used in this study are optimal.Fifth, the incidence of VAP may be inaccurate because the diagnosis was made by individual clinicians.A multicentre clinical trial is needed to demonstrate the bene ts of a comprehensive protocol for ventilator weaning and extubation.

Conclusions
This study suggests that a highly demanding SBT protocol may be safe for extubation without using HFNC and NIV for patients with COVID-19.However, because there is insu cient evidence concerning extubation of patients with COVID-19, further studies are needed to clarify the criteria for extubation.

List Of Abbreviations
This study was approved by Institutional Review Board and Research Ethics Committee of Kawasaki Municipal Hospital (Approval Number: 2020-17).The requirement for written informed consent was waived.Outcomes Outcomes assessed were the rate of reintubation, overall duration of MV, ICU length of stay, hospital length of stay, and overall incidence of VAP.Statistical analysis Categorical variables are presented as percentages, and continuous variables are presented as mean ± standard deviation (SD) or median and 25% and 75% percentile values.All statistical analyses were undertaken with R statistical software, version 3.3.3(R Foundation for Statistical Computing Vienna, Austria).

Table 1
Demographics and clinical characteristics of the patients with COVID-19