A 73-year-old female was hospitalized because of exacerbation of chronic obstructive pulmonary disease (COPD) with acute-on-chronic respiratory failure. She was cachectic (Body mass index: 18 kg/m2), an ex-smoker (with a history of one-hundred-pack-years), on inhalers at home, and was using long term oxygen therapy (LTOT) for the last 3 years, but suboptimal (3 h/day on average). She had hypertension and cardiac arrhythmia with moderate functional limitation (i.e. short of breath with light activities and unable to perform two or more activities of daily living).

On arrival to emergency department; she was started on noninvasive ventilation (NIV) because of her increased respiratory distress (accessory muscle use with RR: 30 bpm, HR: 120 bpm, Systolic/diastolic BP: 150/125 mmHg) and arterial blood gas result, revealing acute hypercapnic respiratory failure (AHRF) (pH: 7.29, PaCO2: 69 mmHg, PaO2: 54 mmHg, HCO3: 28 mmol/L, SatO2: 86% on room air). She was transferred to the respiratory ward to continue her management regarding COPD exacerbation and AHRF.

On admission day 2, the patient's clinical and gas-exchange status significantly improved (pH: 7.35, PaCO2: 61.6 mmHg, PaO2: 88.7 mmHg, HCO3: 29.6 mmol/L, SatO2: 97.7% on IPAP: 20/EPAP: 6 cmH2O, inspired oxygen fraction (FiO2): 0.28). We decided to wean the patient gradually off NIV.

Noninvasive ventilation is effective in the management of different etiologies of acute hypercapnic respiratory failure (AHRF), especially for exacerbation and post-extubation weaning phase of COPD patients, as well as for palliative purposes.1,2 The use of NIV in non-COPD causes of AHRF, such as acute decompensation of the obesity-hypoventilation syndrome (OHS) or neuromuscular diseases (NMD), has also been incorporated into clinical practice, although this has been based on mainly cohort studies.2-4 On the other hand, NIV can also be utilized for the management of acute hypoxemic respiratory failure; including immune-compromised patients, post-operatively, patients with acute cardiogenic pulmonary edema, or Covid-19 pneumonia.1,5,6

Although NIV implementation for the management of AHRF has been framed well by the guidelines,1,2 the strategies for weaning from NIV have not been well-defined yet. Most of the time, clinicians decide on NIV weaning intuitively according to their expertise and daily clinical practice. Unnecessarily prolonged use of NIV can lead to increased morbidity and even mortality, due to complications, including facial skin lesions or NIV-related pneumonia, as well as to the increased duration of hospitalization and cost.7-12 Attempting extubating a patient as soon as he/she recovers from the primary disorder leading to respiratory failure is recommended to decrease complications associated with intubation13; similarly, NIV should be stopped as soon as the acute episode is has passed.2 On the other hand, premature discontinuation of NIV can also result in worsening of respiratory status, with possible reinstitution of NIV. This risk might be extremely likely, especially in patients with a high risk of nocturnal hypoventilation (i.e. partial arterial pressure of carbon dioxide increase during sleep >10 mmHg), such as severe COPD, OHS, NMD, or chest wall disorders.4,14,15 Keeping all of these important physiological points in mind, here is the critical question: When is the best time for weaning from NIV for whom?

Despite the wide range of studies available on weaning from invasive mechanical ventilation;13,16,17 few numbers of studies have evaluated the NIV weaning strategies in patients with AHRF, especially with COPD.18-27 Once the patient's general condition improves, NIV can be discontinued either by the patient (e.g. NIV intolerance) or by the physician (based on clinical findings, such as normalization of pH as well as respiratory status) or by protocol. In comparison to physician-directed weaning, protocol-directed discontinuation of NIV can reduce the duration of NIV and ICU stay as well as the mortality.18,19,28 The strategies for weaning can be currently grouped into three, including a gradual decrease in the level of ventilator support and/or duration of NIV as well as abrupt discontinuation.2,20-24,29 The 2016 BTS/ICS guidelines state that NIV can be discontinued in AHRF due to exacerbation of COPD, by reducing the daytime periods of ventilation in 2 to 3 days, based on clinical criteria, before being discontinued overnight.2 Stepwise decrease in NIV duration, as recommended,2 was shown to have no difference in NIV withdrawal success rates compared to abrupt discontinuation of NIV,21,22 except for a longer intermediate care unit stay.22 Similarly, in a recent RCT performed on 90 COPD patients, the success rate of weaning was similar between abrupt discontinuation, stepwise decrease in ventilator support level, and stepwise decrease in NIV duration arms; although the duration of NIV use and length of hospital stay were lower in first two groups.23 Therefore, abrupt removal of NIV can be an acceptable option for spontaneously breathing patients of COPD, whereas the clinical status and disease severity can alter the physician's choice.

For non-COPD patients recovering from an episode of AHRF, a referral to a home ventilation service for assessment of ‘domiciliary NIV use’ has been recommended, while continuing ‘nocturnal NIV’ till to the accomplishment of the evaluation.2 The presence of ongoing respiratory failure, in-hospital stability of NIV, and local care pathways can affect the decision on the location and timing of this assessment.30 Recently, hospital discharge with positive airway pressure management was shown to reduce mortality in patients with OHS or suspected of having OHS.31

Patient-derived factors [such as etiology of AHRF (such as COPD vs. non-COPD causes), disease severity, history of prior NIV use, clinical status (for example rapid shallow breathing index (RSBI), bicarbonate level or pH<7.35 just before weaning] can help to predict NIV weaning outcome and eventually to choose the best method for each individual.23,24,26,27 In the sole multicenter RCT, including both COPD and non-COPD patients, all three tested strategies were similarly effective in terms of NIV weaning success; while the presence of restrictive respiratory disorder due to obesity as an underlying lung disease was a predictor of failure, in contrast to COPD as a predictor of success (unpublished data under review).24 In the future, additional research is required on weaning strategies/processes in non-COPD patients.

It is essential to decide the timing and methodology of NIV discontinuation after recovering from AHRF in adults, to improve patient outcomes. In this context, we aimed to describe and discuss the course of weaning and various weaning protocols, based on the most recent available literature and our experience both on COPD and non-COPD populations. Thereby, a practical guide for daily practice was intended.

The process of liberating patients from NIV generally referred to as weaning, should be planned as soon as the patient is initiated on NIV. The optimal timing for initiation of weaning is as important as the methods of NIV discontinuation. These can differ based on the underlying etiology of AHRF (COPD vs. non-COPD) and the severity of the disease.

The 2016 BTS/ICS guideline states that discontinuation from NIV can be planned in AHRF due to acute exacerbation of COPD, in whom NIV is successful (pH ≥7.35 achieved, resolution of underlying cause and symptoms, respiratory rate normalized) following the first 24 h or longer duration on NIV2. Patients can be screened daily by the physician or respiratory therapist for the clinical criteria to be met before the weaning attempt, as summarized in Table 1.22-24,29 If patients pass the baseline screening criteria, they can be discontinued from NIV onto nasal/Venturi oxygen at the minimal level (maximum of 5 L/min) to achieve the same oxygenation targets of NIV. If they also pass the re-assessment criteria after 1-4 h of spontaneous breathing with supplementary oxygen, the weaning process can be initiated. While pH>7.35 was mostly used as criteria, two of the studies used pH>7.30 as cut-off for screening.18,26 However, a recent study27 pointed pH<7.35 before weaning as a predictor of weaning failure. The physicians should keep in mind that more liberal criteria can increase the chance of earlier weaning with reduced rate of NIV complications and the duration of NIV and hospital stay; while stricter criteria can lead to a greater rate of withdrawal success, as in the case of extubation.31

Chronic obstructive pulmonary disease patients with persistent hypercapnia after recovering from AHRF have a poor prognosis with high rates of readmission and mortality within one year.32 Actually, a relevant proportion (25%) of COPD patients in GOLD stage 3 and 4 exhibits chronic hypercapnia.14 The long-term domiciliary NIV treatment can improve daytime hypercapnia and admission-free survival with no to little benefit in quality of life of those patients.33 ‘Normalization of PaCO2’ has been recommended previously, for the COPD patient to be weaned off NIV,2 indirectly increasing possibility of initiation of long-term NIV in hospital. However, based on emerging evidence, currently, reassessment for ‘long-term NIV’ has been suggested at 2–4 weeks after resolution, since nearly one fifth of the patients are no longer hypercapnic by then.34-36

Once non-COPD patients (such as exacerbation of NMD or OHS) or COPD patients with suspected or known sleep apnea syndrome have achieved clinical stability and the underlying cause of exacerbation has resolved, consideration can be given to in or outpatient assessment of OHS or overlap syndrome as well as domiciliary use of NIV, which can reduce mortality.30,34 These patients cannot be weaned off NIV and may require in-hospital continuation and transition to long-term NIV, as suggested by home ventilation services.2,34

Although NIV implementations for AHRF due to different etiologies have been described in detail,1,2 the weaning methodology has not been well-defined yet by the guidelines. There are different weaning protocols recommended for invasive mechanical ventilation;13,16,17 however, there have been no definitive recommendations for NIV weaning published in global guidelines yet,1,2 except for Spanish and Indian guidelines.29,37 Monitoring PaCO2 on and off NIV can be a useful guide for the withdrawal process,2 in addition to other clinical parameters, including vitals, pH, and SatO2 levels. Patients can be considered as ‘weaning failure’; if NIV reinstitution or intubation criteria19,38,39 (Table 2) are met during the weaning phase/ right after NIV discontinuation, or if weaning is not possible. The time to define whether weaning is successful or not varies among studies from 2 to 8 days, the most common threshold being 3-5 days (Table 1).

NIV has generally been withdrawn depending on either by decision of the physician based on clinical findings with gas exchange status, or by the patient's demand (such as in NIV intolerance). On the other hand, the weaning can be protocolized. Duan et al compared those two strategies in 73 patients with COPD using NIV>24 h for AHRF.18 In the protocol-directed arm, patients passing daily screening for NIV weaning criteria were discontinued by respiratory therapists from NIV directly onto oxygen with a nasal cannula till the patient was discharged. If the patient's clinical condition worsened as given in the protocol, NIV was reinstituted with daily screening to wean. By this method, the authors showed that NIV duration (-1.8 days) and length of ICU stay (-2.3 days) decreased as compared to the physician-directed method. Reduction in 28-day mortality was also shown as a benefit of protocols.19 Accordingly, NIV weaning using a standardized protocol can provide better outcomes, with the prevention of prolonged NIV use due to an unplanned approach.28

It is important to know the timing and methods of weaning from NIV for patients recovering from AHRF, since inappropriate discontinuation can result in increased morbidity and mortality. Clinical status and disease severity may play a part in choosing the best method. None of the methods have yet been clearly established to be superior to the others in terms of relapse of AHRF in COPD patients after liberation from NIV, as well as mortality and long-term ventilator dependence. Therefore, abrupt removal of NIV can be a better option for uncomplicated cases of COPD to decrease the rate of NIV complications, while any of the methods can be adopted based on the patient's characteristics. Last but not least; we need more robust evidence for alternative NIV weaning strategies in different patient subgroups with AHRF.