key: cord-0879452-5wzd45vd
authors: Serpa Neto, Ary; Checkley, William; Sivakorn, Chaisith; Hashmi, Madiha; Papali, Alfred; Schultz, Marcus J.
title: Pragmatic Recommendations for the Management of Acute Respiratory Failure and Mechanical Ventilation in Patients with COVID-19 in Low- and Middle-Income Countries
date: 2021-01-13
journal: Am J Trop Med Hyg
DOI: 10.4269/ajtmh.20-0796
sha: ab684813f2537cef11fafb491831e6fbc0145c82
doc_id: 879452
cord_uid: 5wzd45vd

Management of patients with severe or critical COVID-19 is mainly modeled after care for patients with severe pneumonia or acute respiratory distress syndrome (ARDS) from other causes, and these recommendations are based on evidence that often originates from investigations in resource-rich intensive care units located in high-income countries. Often, it is impractical to apply these recommendations to resource-restricted settings, particularly in low- and middle-income countries (LMICs). We report on a set of pragmatic recommendations for acute respiratory failure and mechanical ventilation management in patients with severe/critical COVID-19 in LMICs. We suggest starting supplementary oxygen when SpO(2) is persistently lower than 94%. We recommend supplemental oxygen to keep SpO(2) at 88–95% and suggest higher targets in settings where continuous pulse oximetry is not available but intermittent pulse oximetry is. We suggest a trial of awake prone positioning in patients who remain hypoxemic; however, this requires close monitoring, and clear failure and escalation criteria. In places with an adequate number and trained staff, the strategy seems safe. We recommend to intubate based on signs of respiratory distress more than on refractory hypoxemia alone, and we recommend close monitoring for respiratory worsening and early intubation if worsening occurs. We recommend low–tidal volume ventilation combined with FiO(2) and positive end-expiratory pressure (PEEP) management based on a high FiO(2)/low PEEP table. We recommend against using routine recruitment maneuvers, unless as a rescue therapy in refractory hypoxemia, and we recommend using prone positioning for 12–16 hours in case of refractory hypoxemia (PaO(2)/FiO(2) < 150 mmHg, FiO(2) ≥ 0.6 and PEEP ≥ 10 cmH(2)O) in intubated patients as standard in ARDS patients. We also recommend against sharing one ventilator for multiple patients. We recommend daily assessments for readiness for weaning by a low-level pressure support and recommend against using a T-piece trial because of aerosolization risk.

Management of patients with severe or critical COVID-19 is mainly based on care for patients with severe pneumonia or acute respiratory distress syndrome (ARDS) from other causes, although some aspects of this new disease may demand a different approach. Recommendations for treatment of severe pneumonia and ARDS management have been gathered mainly from investigations in resource-rich intensive care units (ICUs), mostly located in high-income countries (HICs).

It may not be practical to apply these recommendations to resource-restricted settings, particularly in low-and middleincome countries (LMICs). Indeed, high-dependency units and ICUs in LMICs are frequently restricted to availability of infrastructure, equipment, medications, skilled, nurses and doctors. An international task force comprising members from LMICs and HICs, all with direct experience in various LMIC settings, critically appraised a list of questions regarding acute respiratory failure and mechanical ventilation for patients with severe/critical COVID-19.

We provide a list of recommendations and suggestions after pragmatic, experience-based appraisal. A summary of the recommendations is shown in Table 1 .

A full description of the methods is provided in the Appendix. In brief, we formulated a set of clearly defined questions regarding acute respiratory failure and mechanical ventilation for patients with suspected or confirmed severe/critical COVID-19 in LMICs. The list of questions was reviewed for content and clarity by other members of the COVID-LMIC Task Force, the full membership of which can be found in the Appendix. After approval, the subgroup assigned one member to search the literature for evidence to answer each question. The literature search was performed in a minimum of one general database (i.e., MEDLINE and Embase) and the Cochrane libraries. Furthermore, we identified investigations from LMICs and searched for unpublished study results. We selected relevant publications, appraised the evidence, and classified the quality of evidence as high, moderate, low, or very low. Recommendations were rated as strong or weak, depending on the quality of evidence and several other factors such as availability, affordability, and feasibility in LMICs. A strong recommendation was worded as "we recommend. . ." and a weak recommendation as "we suggest. . .," followed by the quality of evidence. A number of recommendations could remain "ungraded," when, in the opinion of the subgroup members, such recommendations were not conducive for the process described previously (Appendix Table 2 ). The recommendations were reviewed by the subgroup in an iterative process and were later reviewed by the entire Task Force in two rounds. How to supplement oxygen in COVID-19 patients in LMICs? Background. Supplementary oxygen is a first and essential therapeutic step in the care of hypoxemic COVID-19 patients and may be the dominant focus of care for these patients in LMICs.

Search results. We searched the databases of PubMed and Embase through Ovid until November 1, 2020. The following terms were used, either as medical subject headings (MeSH) terms or as free text words: "COVID-19," "coronavirus," "novel corona," "SARS-CoV-2" and "oxygen," "pneumonia," "oxygen target," "liberal," "conventional" in oxygen initiation and targets; "oxygen therapy," "aerosol," "high-flow nasal oxygen (HFNO)," "non-invasive ventilation (NIV)" in oxygen delivery interfaces; "fever control," "cough suppressants," "sedation" in fever control, use of cough suppressants in case of severe coughing, and light sedation; and "intubation criteria," "hypoxemia," "silent hypoxemia" in initiation of invasive ventilation. From the search, it was found five randomized control trials, six systematic reviews, four narrative reviews, and one observational study, and none of the studies were from LMICs.

Evidence. Oxygen initiation and targets. The decision to initiate supplemental oxygen in adults with COVID-19 has not been studied, but indirect evidence from studies in pneumonia from another cause suggests that an oxygen saturation (SpO 2 ) less than 90% is associated with an increase in mortality and duration of hospitalization. 1 Silent hypoxia is described as low oxygen levels with normal/hypercapnic respiratory failure and presenting without dyspnea. Potential mechanisms of mismatch between the severity of hypoxemia and the relatively mild respiratory discomfort may be a consequence of hypoxemia-driven hyperventilation and/or leftward shift of the oxyhemoglobin dissociation curve. This shift increases the alveolar oxygen tension and systemic saturation after reducing alveolar carbon dioxide tension, the Bohr effect. PaO 2 and oxygen delivery can be optimized by modulating blood pH, PaCO 2 , hemoglobin concentration, cardiac output, and arterial content of oxygen. These factors mean close attention is warranted when implementing lung-protective strategies, particularly when using low oxygen targets (55-70 mmHg) and permissive hypercapnia. However, inappropriate use of supplementary oxygen in patients at risk of hypercapnic respiratory failure can result in life-threatening hypercapnia. 2, 3 Therefore, these patients need to be identified before prescribing and administering supplementary oxygen. Finally, a really important point to address is that frequently arterial blood gas analyses are not widely available as in HICs.

Ten randomized controlled trials (1,458 participants) reported an increase in the number of serious adverse events with higher fractions or oxygenation targets. 4 One metaanalysis of 25 randomized clinical trials in critically ill patients before the COVID-19 pandemic shows that liberal oxygen therapy increases mortality without improving other patientimportant outcomes and that supplemental oxygen might result in harm when provided if SpO 2 > 97%. 5 Recent randomized clinical trials showed no benefit or harm of a conservative oxygen therapy, when compared with a strategy that used usual oxygen targets, 6 or a more liberal strategy. 7 COVID-19 pneumonia may be more susceptible to potentially harmful effects of oxygen because adaptive responses are impaired by local inflammation and damage. 8 In addition, COVID-19 patients tolerate hypoxemia pretty well, even when SpO 2 decreases to < 88% probably because of a large intrapulmonary shunt with well-preserved lung compliance, which is unlike many other cases of ARDS by other causes. 9 Targeting permissive hypoxemia (PaO 2 55-70%, SpO 2 88-95%) in COVID-19 patients without pre-existing conditions that may make more severe hypoxemia unsafe is appealing (e.g., stroke, new myocardial infarction without revascularization, and neurotrauma). 10 Of note, this approach may not be different from that practiced in HICs.

Oxygen delivery interfaces. A non-rebreather mask is a safe interface in supplemental oxygen therapy in COVID-19 patients as these masks help limit the dispersion of droplets. Human mannequin studies show that the maximum exhaled dispersion distance at 10 L/minute is < 10 cm, suggesting that this modality generates the least dispersed aerosols. 11 Nonrebreather masks can reach FiO 2 of > 85% at flow rates approaching 15 L/minute. 12 To prevent hypercapnia, the reservoir bag must remain inflated at all times; this requires flow rates of at least 8-10 L/minute. 13 The use of nonrebreather masks usually requires oxygen via a piped system within the hospital.

High flow nasal oxygen, characterized as devices providing oxygen with flows from 20 to 70 L/minute, and NIV may reduce the need for intubation 14, 15 and could improve outcome in acute hypoxemic respiratory failure. 16 In LMICs where the options for respiratory support can be severely limited, especially the first interface could be the best alternative. Several practical considerations need to be mentioned, though. First, patients should be closely monitored so that they can be intubated quickly when gas exchange worsens, or in case of increased work of breathing. 17 Delayed intubation was associated with adverse outcomes and high failure rates in patients with the Middle East respiratory syndrome. 18, 19 Second, there may be more aerosol formation, 20,21 which could increase the risk of viral transmission, although this may be limited when correct interfaces are used. 22, 23 Simulation studies suggest that surgical masks placed over the HFNO interface may reduce exhaled air dispersion ( Figure 1 ). 24, 25 Also, during NIV, strict control of any air leak is recommended, and if a single limb circuit is used, a viral filter should be placed on the exhalation port of the mask (Figure 1 ). 26 To date, no article comparing different interfaces for oxygen delivery in COVID-19 in LMICs is available.

Fever control, use of cough suppressants in case of severe coughing, and light sedation. COVID-19 patients at times have high to very high minute volumes, which eventually leads to respiratory exhaustion and the need for endotracheal intubation. Because minute volume increases with body temperature, adequate and strict fever control could help preventing exhaustion, and thus, there is the need for invasive ventilation. One frequent symptom of COVID-19 is a persistent and severe cough, during which severe hypoxemia could develop. Cough control may be helpful, for example, by giving codeine. Light sedation may also reduce minute volume and also reduce coughing. In the absence of any literature on fever control, cough control, and light sedation in COVID-19 patients, this bundle, often combined with prone positioning, is currently practiced in non-intubated COVID-19 patients in the centers of the authors of this set of suggestions and recommendations.

Availability, affordability, safety, and feasibility. Improper use of supplementary oxygen occurs often in LMICs, frequently resulting in low pulse oximetry readings. 27 As our appreciation of the risks of hypoxia, combined with the need to optimize the use of a scarce resource, we suggest starting supplementary oxygen when SpO 2 is persistently lower than 94% for COVID-19 patients.

In places where pulse oximetry is available, targeting a low SpO 2 seems a safe strategy. Pulse oximeters are affordable and widely available and may be shared between COVID-19 patients in places with scarcity. Supplemental oxygen with lowest possible FiO 2 helps prioritization of oxygen flow to the most severely ill patients within hospitals. However, in settings where continuous pulse oximetry is not available, it could be safer to use slightly higher targets for SpO 2 (SpO 2 at 95-97%).

Low-flow supplemental oxygen via nasal prongs, a spongetipped catheter, or a face mask is widely available and affordable. 10 High-flow systems need piped oxygen systems, which is often not available in LMIC hospitals. Oxygen concentrators may be a solution, but these are not widely available, expensive, and may be unsafe in settings with electricity fluctuations or power outages. High flow nasal oxygen and NIV are both feasible in LMICs, but come at additional costs for the machines and interfaces, have technical challenges and practical concerns, demand close monitoring, and depend on a stable oxygen supply. In settings with electricity fluctuations or blackouts, safety could improve when the devices used for HFNO and NIV have battery backup options. High flow nasal oxygen and NIV generate aerosols and should therefore be used in combination with a surgical mask or, preferably, in an isolation room. In addition, in settings where oxygen supply is limited, HFNO is not a good option as it has a high need for oxygen.

For fever control and cough suppression, paracetamol and codeine are widely available and cheap, and their use is without additional risk for patients. Nonsteroidal antiinflammatory drugs are suggested to be harmful, 28-30 which could outweigh a potential benefit on symptom control for respiratory tract infection. 31 For light sedation, benzodiazepines and morphine are widely available and cheap. Light sedation could be a challenge in resource-limited settings, although when there is lack of monitoring, training, or experience. Light sedation with small boluses of benzodiazepine, like midazolam, if needed combined with small boluses of morphine, is safe in experienced hands. It should be remembered that benzodiazepines could worsen or even induce agitation, or delirium. In those cases, it should be stopped. To improve tolerance to the NIV interface, or in case of delirium, dexmedetomidine, if available, is preferred over haloperidol because of significant QT interval prolongations. [32] [33] [34] Recommendations and suggestions ( 4. In hypoxemic COVID-19 patients in LMICs, we recommend using low-flow oxygen systems, for example, nasal prongs, a sponge-tipped catheter, or a facial mask for supplementary oxygen, and, where available, a non-rebreathing mask (strong recommendation, low quality of evidence). 5. In COVID-19 patients in LMICs who remain hypoxemic despite use of low-flow oxygen therapy, we suggest the use of HFNO or NIV with monitoring and isolation considerations (especially for NIV), as discussed in the text (weak recommendation, low quality of evidence). 6. In COVID-19 patients in LMICs, we suggest fever control with paracetamol, using cough-suppressants, in case of severe coughing, and light sedation (weak recommendation, low quality of evidence).

Should prone positioning be used in COVID-19 in LMICs? Rationale. Prone positioning is associated with improvement in oxygenation and survival in patients with ARDS under invasive mechanical ventilation. The use of prone positioning before and after intubation is being discussed in patients with COVID-19 under oxygen therapy.

Search results. We searched the databases of PubMed and Embase through Ovid until November 1, 2020. The following terms were used, either as MeSH terms or as free text words: "COVID-19," "coronavirus," "novel corona," "SARS-CoV-2," "prone," and "prone positioning." No randomized clinical trial on the topic was found. All but one study was conducted in HICs.

Evidence. A multicenter study conducted in patients with acute respiratory failure and no COVID-19 reported prone positioning in non-intubated patients as a feasible and wellaccepted intervention that was associated with improvement in oxygenation. 35 In another multicenter study in nonintubated moderate to severe ARDS patients under NIV or HFNO, it was suggested that early application of prone positioning with HFNO, especially in patients with moderate ARDS and baseline SpO 2 > 95%, may help avoid intubation. 36 Specifically in COVID-19 patients, in one study involving 24 patients, 63% of patients were able to tolerate prone positioning for more than 3 hours, but only 25% had an improvement in oxygenation Also, the study was not powered to assess clinical outcomes, but 21% of the patients required intubation within 10 days. 37 Another case series of 15 COVID-19 patients showed that the combination of prone positioning and NIV on general wards was feasible and resulted in a decrease in respiratory rates and increase in oxygenation. 38 The only case series in an LMIC (Iran) was conducted in 10 patients and reported improvement in SpO 2 and a decrease in dyspnea, with no intubation in the group. 39 Additional case series confirmed the feasibility of the maneuver and its association with improvement in oxygenation and dyspnea. [40] [41] [42] The impact of prone positioning in non-intubated patients on patientcentered outcomes remains unknown because no randomized clinical trial is still available.

In patients receiving mechanical ventilation, there is strong agreement to use prone positioning in severe cases of ARDS, 43 which is supported by evidence from a randomized controlled trial. [43] [44] [45] Based on our personal experience, we suggest a strategy of prone positioning in non-intubated patients receiving oxygen, HFNO, or NIV and without signs of severe respiratory distress. Patients should be kept in prone positioning as long as possible but aiming at least 3 hours per session. Of note, this approach may not be different from that practiced in HICs. In intubated patients with PaO 2 /FiO 2 < 150 mmHg, prone positioning for 12-16 hours should be implemented.

Availability, affordability, safety, and feasibility. In places with an adequate number and trained staff, trying a strategy of prone positioning in patients with hypoxemia seems safe. Close monitoring, and a clear failure and escalation criteria are needed. The use of this strategy can help avoid some intubations and help the overall system at-risk of shortage of equipment. However, among the limitations of prone positioning is the workload for healthcare professionals; however, this is only a relative contraindication for prone positioning.

Recommendations and suggestions (Table 1 ).

despite use of oxygen, HFNO or NIV and who do not exhibit clear signs of respiratory distress, we suggest a trial of prone positioning; this however requires close monitoring, and clear failure and escalation criteria (weak recommendation, low quality of evidence). 2. In intubated patients with COVID-19 in LMICs, we recommend using prone positioning for 12-16 hours in case of refractory hypoxemia (PaO 2 /FiO 2 < 150 mmHg) (strong recommendation, low quality of evidence).

When to intubate for invasive ventilation in COVID-19 patients in LMICs? Rationale. The decision and timing of intubation is crucial in the management of patients with COVID-19. The decision to intubate these patients, especially in LMICs, must weigh the risks of premature intubation against the risk of delayed intubation and its consequences, like respiratory arrest and an increase of risk of poor outcome.

Search results. We searched the databases of PubMed and Embase through Ovid until November 1, 2020. The following terms were used, either as MeSH terms or as free text words: "COVID-19," "coronavirus," "novel corona," "SARS-CoV-2," "intubation criteria," "hypoxemia," "silent hypoxemia," and "initiation of invasive ventilation."

Evidence. "Silent hypoxemia," characterized by severe desaturations with no or only some dyspnea, is frequently seen in COVID-19 patients. 46, 47 It seems that dyspnea in COVID-19 patients, if this develops, does not result from hypoxemia alone but is due to other causes, such as airflow obstruction, dead space ventilation, respiratory muscle dysfunction, and anxiety. 46, 48 Thus, hypoxemia alone is not necessarily an indication for intubation in COVID-19 patients. In addition, the respiratory rate is an important parameter to be followed up in these patients, and tachypnea is one of the signs that could suggest that intubation is needed. This concept was implemented by the Chinese Society of Anesthesiology, which recommends to proceed with endotracheal intubation in COVID-19 patients showing no improvement in respiratory distress or tachypnea (respiratory rate > 30/ minute) in combination with worsening oxygenation (PaO 2 / FiO 2 < 150 mmHg) after applying 2-hour HFNO or NIV. 47 This measure emphasizes more the worsening of respiratory symptoms, rather than focusing on hypoxemia per se. It is important to emphasize that with the evolving of the pandemic, a more liberal approach is being used, with sustained use of noninvasive support as long as the patient does not present signs of discomfort or clinical deterioration. In addition, the ROX, an index combining the respiratory rate and oxygenation (using the SpO 2 /FiO 2 ) to predict outcome outperformed the diagnostic accuracy of the two variables separately, but was only validated in patients undergoing HFNO. The ROX index was also recently validated in COVID-19 patients, but again only in those receiving HFNO. Yet, the prognostic value in limited-resource settings and beyond the HFNO domain and potential diagnostic properties have to be proven. Whenever invasive ventilation is needed, it should be timely and effectively provided. Waiting to intubate hypoxemic patients until they require FiO 2 of close to 1.0 may make intubation a challenge. Despite no formal evidence available for the best timing of intubation in COVID-19 patients, this recommendation is in line with previous guidelines of respiratory support of patients in LMICs. 49, 50 Availability, affordability, safety, and feasibility. Initiating invasive ventilation based on clinical signs like fatigue and exhaustion, in combination with refractory hypoxemia, helps prioritize intubation in COVID-19 patients to the ones who really need it. However, close monitoring for worsening signs of respiratory symptoms could be a challenge in LMICs, in particular when less experienced healthcare workers have to take care of these patients in the surge of patients who need hospitalization. [51] [52] [53] When arterial blood gases are not available, the PaO 2 /FiO 2 ratio could be replaced by the SpO 2 /FiO 2 according to well-defined formulas. [54] [55] [56] [57] When available, intubation teams formed by skilled professionals and with predefined protocols for airway management should be considered.

Recommendations and suggestions (Table 1 ). 1. In hypoxemic COVID-19 patients in LMICs, we recommend to intubate and initiate invasive ventilation based on signs of fatigue and respiratory distress more than on refractory hypoxemia alone (weak recommendation, low quality of evidence). 2. In hypoxemic COVID-19 patients in LMICs, we recommend close monitoring for respiratory worsening and early intubation if worsening occurs (strong recommendation, low quality of evidence). 3. In hypoxemic COVID-19 patients in LMICs, we recommend staff members to be readily available to closely monitoring a patient when an early intubation was not performed, and staff should be trained in timely recognition of respiratory distress and sign of worsening of respiratory distress (weak recommendation, low quality of evidence).

How to set the ventilator in COVID-19 patients in LMICs? Rationale. Data to support evidence-based management of invasive ventilation in COVID-19 patients remain scarce. However, an adequate ventilatory strategy is frequently associated with better outcomes in patients receiving mechanical ventilation, and this should not be different in COVID-19 patients.

Search results. We searched the databases of PubMed and Embase through Ovid until November 1, 2020. The following terms were used, either as MeSH terms or as free text words: "COVID-19" and "mechanical ventilation,"

Evidence. All major studies assessing the mechanical ventilation strategy were conducted in HICs. The evidence-based principles underpinning invasive ventilation for patients with ARDS should also be considered for COVID-19 patients: provide ventilation to protect the lungs using low tidal volumes and keep airway pressures low (both plateau [< 30 cmH 2 O] and driving pressures [< 15 cmH 2 O]). All articles that discussed ventilation strategies recommended the use of lowtidal volume ventilation with tidal volumes anywhere between 4 and 8 mL/kg of predicted body weight (PBW) according to recommendation by the ARDS Network, derived from HICs. 45, 46, [58] [59] [60] [61] [62] [63] [64] [65] [66] Among articles suggesting severe COVID-19 pneumonia phenotypes, at least one suggested using tidal volumes on the higher end of the spectrum (6-8 mL/kg PBW) to attenuate dyspnea. 58 However, this recommendation is based on small case series of low quality and on phenotypes not externally validated. Thus, this strategy should not be considered as a standard of ventilation strategy for COVID-19 patients.

Among the studies that discussed phenotypes of patients with severe COVID-19 pneumonia requiring invasive mechanical ventilation, 45,46,58-65 a common theme was the identification of two classes of patients: patients with a preserved to near normal (> 50 mL/cmH 2 O) respiratory system compliance or those with a low respiratory systemic compliance (< 40 mL/cmH 2 O). However, no validation or ventilation strategy tailored by these phenotypes was tested. At least five studies suggested using low to moderate levels of positive end-expiratory pressure (PEEP) (8-10 cmH 2 O) for those with preserved compliance and higher levels of PEEP (> 10 cmH 2 O) among those with low compliance or a more typical presentation of ARDS. 45, [58] [59] [60] [61] [62] However, these recommendations were not validated in larger and well-performed multicenter studies or clinical trials. One study suggested overall poor alveolar recruitability in severe COVID-19 pneumonia. 62 There was no specific recommendation on how to manage FiO 2 relative to setting PEEP in any of the identified articles. One article discussed new evidence to aim for higher PaO 2 or SpO 2 based on recent evidence compared with the recommendation of aiming for PaO 2 55-80 mmHg or an SpO 2 88-92%. 46 There is no evidence that higher PEEP is more beneficial than lower PEEP regarding survival in patients with ARDS, with exception for a more severe group. [67] [68] [69] In addition, there is no evidence that PEEP titrated according to the lower driving pressure results in better outcomes than the protocol-driven lower PEEP/ higher FiO 2 tables. 70 In fact, a recent randomized controlled trial of a strategy with lung recruitment and titrated PEEP compared with low PEEP increased 28-day all-cause mortality. 71 Early experience with COVID-19 pneumonia is that many patients have remarkable mismatches between severity of hypoxemia and stiffness of the respiratory system. Indeed, static compliance is higher than what is usually found in ARDS from other etiologies. In such patients, high PEEP may result more in hyperinflation of normally aerated lung zones than in recruitment of collapsed regions.

In settings where there is limited availability of mechanical ventilators, unconventional approaches have been suggested, such as connecting multiple patients to a single ventilator. 44, [72] [73] [74] Specifically, when using parallel ventilation, tidal volumes could not be controlled and ranged between 257 and 621 mL, even in patients with good compliance. 72 Three identified articles cautioned against or did not endorse the practice but allowed room for consideration as a rescue strategy. [73] [74] [75] When using invasive ventilation, there are clear recommendations about keeping the patient ventilator circuit as a closed system to avoid risk of aerosolization. Specifically, in one article, it was recommended clamping the endotracheal tube before any disconnection from the ventilator and placing the ventilator on standby before any circuit disconnection.

Availability, affordability, safety, and feasibility. Lack of mechanical ventilators poses a major management challenge in LMICs, especially during the COVID-19 pandemic, where anywhere between 2.3% and 33.1% of hospitalized patients may require mechanical ventilation. 76 The number of ventilators per 100.000 inhabitants varies enormously in LMICs. In addition, in general, the outcomes of ventilated patients in LMICs, especially when the numbers of ventilators are low, are poor. This may be mainly related to safety issues. In settings where there is limited availability of mechanical ventilators, unconventional approaches have been suggested, such as connecting multiple patients to a single ventilator. [72] [73] [74] However, there is a recommendation against this practice. 72 Recommendations and suggestions (Table 1 ). 1. In intubated patients with COVID-19 in LMICs, we recommend using a low tidal volume (4-8 mL/kg PBW), and whenever possible, a tidal volume £ 6 mL/kg PBW should be pursued (strong recommendation, high quality of evidence). 2. In intubated patients with COVID-19 in LMICs, we recommend using a "low PEEP/high FiO 2 table" (moderate recommendation, low quality of evidence). 3. In intubated patients with COVID-19 in LMICs, we recommend against using routine recruitment maneuvers, unless as a rescue therapy in refractory hypoxemia (moderate recommendation, low quality of evidence). 4. In intubated patients with COVID-19 in LMICs, we recommend against sharing one ventilator for multiple patients (strong recommendation, low quality of evidence).

How to wean a COVID-19 patient from the ventilator in LMICs? Rationale. The COVID-19 pandemic has resulted in the implementation of rapidly changing protocols and guidelines related to the indications and protocols for extubation. Both are aerosol-generating procedures, commonly associated with coughing, which necessitates proximity of the physician to the patient.

Search results. We searched the databases of PubMed and Embase through Ovid until November 01, 2020. The following terms were used, either as MeSH terms or as free text words: "COVID-19," "coronavirus," "novel corona," "SARS-CoV-2," "weaning," "spontaneous breathing trial," "tracheostomy," and "extubation." From the search, three randomized control trials, one systematic review, and four narrative reviews were found, and none of the studies were from LMICs. The search did not result in any studies from LMICs that directly answered the question of interest. We therefore discuss several pertinent studies from HICs.

Evidence. Weaning and spontaneous breathing trials. A meta-analysis of 10 randomized control trials including 3,165 patients showed no significant difference in the successful extubation, reintubation rate, ICU mortality, and ICU length of stay between spontaneous breathing trials (SBTs) using low levels of pressure support and a T-piece. 77 Three wellperformed trials in resource-rich ICUs clearly showed benefit from early SBTs. [78] [79] [80] However, because of much aerosol generating from a T-piece trial, a low-level pressure support is the preferred method for SBTs in COVID-19 patients. Anecdotal evidence demonstrated longer intubation periods, higher volume of secretions, and airway edema in mechanical ventilated COVID-19 patients, 81 all of these factors increase the risk of post-extubation respiratory failure. Therefore, we suggest a higher degree of readiness for weaning and direct supplementation of low-flow oxygen after extubation in COVID-19 patients.

Extubation. The consensus statement published by an "Australian and New Zealand Intensive Care Society" 7 and the "Difficult Airway Society, the Association of Anesthetists the Intensive Care Society, the Faculty of Intensive Care Medicine and the Royal College of Anesthetists," 82 and two additional opinions 83, 84 suggest that safe, simple, familiar, reliable, and robust practices should be adopted for all episodes of airway management for patients with COVID-19. Staff members should wear personal protective equipment (PPE) during extubation. Efforts should be made to minimize coughing and exposure to infected secretions at extubation time by previous suctioning of secretions. Before extubation, all necessary equipment should be checked. A regular surgical mask should then be placed on the patient's chin, and nasal prongs with low-flow oxygen should be applied after extubation. After extubation, ensure the patient immediately wears a face mask and their oxygen mask or nasal cannula, when practical. 82 When available, a high-flow nasal cannula or NIV should be used after extubation to prevent reintubation. 85 Information on tracheostomy is available in another article.

Availability, feasibility, affordability, and safety. Spontaneous breathing trials are available and affordable in LMICs. 86 It is safer to perform SBTs using a low-level pressure support technique, rather than SBTs using a T-piece in COVID-19 patients; because with the former, a ventilatory support is guaranteed and little aerosol is generated. We recommend using the technique with low-level pressure support in all ventilated COVID-19 patients in resource-limited ICUs.

To reduce the risk of reintubation, we prefer a higher degree of extubation readiness in COVID-19 patients. This practice should include higher criteria for passing SBTs, or promote SBTs to last longer, or repeated for confirmation. Nurses and physicians should develop local SBT protocols, and oversedation should be recognized and avoided at the moment the decision is taken to proceed without invasive ventilation. Lowand middle-income countries have insufficient negative pressure airborne infectious isolation rooms in general. Then, at every extubation, one of the main aims should be the best possible protection of the healthcare workers. Personal protective equipment should be used and changed every time, and the extubating sequences should be safe, simple, familiar, and reliable according to available resources. A Canadian group restricted the aerosolization and droplet spraying during extubation with plastic drapes applied over the head and the endotracheal tube. 87 Finally, whenever available, the physiotherapist should be involved in the weaning and extubation process.

Recommendations and suggestions (Table 1 ). 1. In COVID-19 patients in LMICs, we recommend daily assessments for readiness for weaning by a low-level pressure support (strong recommendation, low quality of evidence). 2. In COVID-19 patients in LMICs, we recommend against using a T-piece trial because of contamination risk (weak recommendation, low quality of evidence). 3. In COVID-19 patients in LMICs who are ready to extubate, we recommend using familiar, local extubation protocols to minimize infection risk among healthcare workers (weak recommendation, low quality of evidence).

prognostic modeling were formulated. These were reviewed for content and clarity by the subgroup members and heads from the other subgroups. After approval by the subgroup members and heads from the other subgroups, the subgroup members split up, each seeking evidence for recommendations regarding three or four of the specific questions posed, seeking help from the subgroup members in identifying relevant publications, where necessary. During this process, questions could be combined, so the subgroup heads were finally left with four major questions. The subgroup heads summarized the evidence in the Appendix and formulated a set of recommendations and suggestions after online discussions. These were communicated among the subgroup members. After their approval, the subgroup heads summarized the evidence in a report, which was sent for approval by all members of the Task Force. Search techniques. The literature search followed the same techniques as previously described. 1 In case a question was identical to one in those recommendations, the subgroup members only searched for additional articles, specifically new investigations or meta-analyses related to the questions, in a minimum of one general database (i.e., MEDLINE and Embase) and the Cochrane libraries. Furthermore, the subgroup members identified investigations from LMICs and also searched for unpublished study results.

Grading of Recommendations. The subgroup members classified quality of evidence as high or low and recommendations as strong or weak. The factors influencing this classification are presented in Table A1 .

The subgroup members paid extensive attention to several other factors as used before, but now focusing on LMICs, that is, availability and feasibility in LMICs, and safety matters in LMICs. A strong recommendation was worded as "we recommend" and a weak recommendation as "we suggest." A number of recommendations could remain "ungraded,", when, in the opinion of the subgroup members, such recommendations were not conducive for the process described above (Table A2) .

Reporting. The report was edited for style and form by Alfred Papali or Marcus Schultz, with final approval by subgroup heads and then by the entire "COVID-LMIC Task Force." A final document was submitted to the "American Journal of Tropical Medicine and Hygiene" for potential publication as a 1,000-word article and made open access.

Financial support. Open access fees for this manuscript, and all nine others in the series, were supported by the Wellcome Trust of Great Britain. Factors that may decrease strength of evidence include high likelihood of bias; inconsistency of results, including problems with subgroup analyses; indirectness of evidence (other population, intervention, control, outcomes, and comparison); imprecision of findings; and likelihood of reporting bias.

Factors that may increase strength of evidence: large magnitude of effect (direct evidence, relative risk > 2 with no plausible confounders); very large magnitude of effect with relative risk > 5 and no threats to validity (by two levels); and dose-response gradient.

Adapted 

The higher the quality of evidence, the more likely a strong recommendation Certainty about the balance of benefits vs. harms and burdens

The larger/smaller the difference between the desirable and undesirable consequences and the certainty around that difference, the more likely a strong/weak recommendation Certainty in or similar values

The more certainty or similarity in values and preferences, the more likely a strong recommendation Resource implications

The lower/higher the cost of an intervention compared to the alternative the more likely a strong/weak recommendation Availability and feasibility in LMICs

The less available, the more likely a weak recommendation Affordability for LMICs

The less affordable, the more likely a weak recommendation Safety of the intervention in LMICs The less safe in an LMIC, the more likely a weak recommendation * In case of a strong recommendation, we use "we recommend. . ."; in case of a weak recommendation, we use "we suggest. . ."

Adapted from Dondorp AM, Dünser MW, Schultz MJ, eds., 2019. Sepsis Management in Resource-limited Settings. Springer. doi.org/10.1007/978-3-030-03143-5.

Heads: Alfred Papali (Atrium Health

Interdepartmental Division of Critical Care Medicine

Subgroup members, in alphabetical order: Andrew Achilleos (Sunnybrook Health Sciences Centre

Sopheakmoniroth Chea (Calmette Hospital

The Johns Hopkins University School of Medicine and The Johns Hopkins University Center for Global Health

Interdepartmental Division of Critical Care Medicine, Sunnybrook Health Sciences Centre

University of Pittsburgh School of Medicine

and Department of Intensive Care

Critical Care Department, NIH, Bethesda, MD; Division of Pulmonary, Allergy and Critical Care

São Paulo, Brazil; Australian and New Zealand Intensive Care Research Centre (ANZIC-RC)

Interdepartmental Division of Critical Care Medicine, Sunnybrook Health Sciences Centre, University of Toronto

Department of Anesthesia and Intensive Care, Miulli Regional Hospital

Division of Pulmonary, Critical Care, and Sleep Medicine

Amsterdam

Tribhuvan University Teaching Hospital

Interdepartmental Division of Critical Care Medicine, Sunnybrook Health Sciences Centre

Department of Intensive Care, Amsterdam University Medical Centers, Location "'AMC

Oxygen saturations less than 92% are associated with major adverse events in outpatients with pneumonia: a population-based cohort study

British thoracic society guideline for oxygen use in adults in healthcare and emergency settings

National Chronic Obstructive Pulmonary Disease Resources and Outcomes Project Implementation Group

Higher versus lower fraction of inspired oxygen or targets of arterial oxygenation for adults admitted to the intensive care unit

Mortality and morbidity in acutely ill adults treated with liberal versus conservative oxygen therapy (IOTA): a systematic review and meta-analysis

Conservative oxygen therapy during mechanical ventilation in the ICU

Liberal or conservative oxygen therapy for acute respiratory distress syndrome

Hyperoxic exposure in humans: effects of 50 percent oxygen on alveolar macrophage leukotriene B4 synthesis

Differentiating COVID-19 pneumonia from acute respiratory distress syndrome (ARDS) and high-altitude pulmonary edema (HAPE): therapeutic implications

A low-pressure oxygen storage system for oxygen supply in low-resource settings

Aerosol dispersion during various respiratory therapies: a risk assessment model of nosocomial infection to health care workers

Delivery of high inspired oxygen by face mask

Carbon dioxide narcosis due to inappropriate oxygen delivery: a case report

Noninvasive ventilation in acute hypoxemic nonhypercapnic respiratory failure: a systematic review and meta-analysis

High-flow nasal cannula oxygen therapy is superior to conventional oxygen therapy but not to noninvasive mechanical ventilation on intubation rate: a systematic review and meta-analysis

High-flow oxygen through nasal cannula in acute hypoxemic respiratory failure

Failure of high-flow nasal cannula therapy may delay intubation and increase mortality

Noninvasive ventilation in critically ill patients with the Middle East respiratory syndrome

Clinical course and outcomes of critically ill patients with Middle East respiratory syndrome coronavirus infection

Aerosol generating procedures and risk of transmission of acute respiratory infections to healthcare workers: a systematic review

Protecting healthcare workers from SARS-CoV-2 infection: practical indications

Association between administered oxygen, arterial partial oxygen pressure and mortality in mechanically ventilated intensive care unit patients

Association between arterial hyperoxia following resuscitation from cardiac arrest and in-hospital mortality

High-flow nasal cannula for COVID-19 patients: low risk of bio-aerosol dispersion

Exhaled air dispersion during coughing with and without wearing a surgical or N95 mask

NHS Guidance on Use of NIV in Adults with Coronavirus

Oxygen and pulse oximetry in childhood pneumonia: a survey of healthcare providers in re-source-limited settings

Non-steroidal anti-inflammatory drugs may worsen the course of community-acquired pneumonia: a cohort study

A multicentre case-control study of nonsteroidal anti-inflammatory drugs as a risk factor for severe sepsis and septic shock

Risks of nonsteroidal antiinflammatory drugs in undiagnosed intensive care unit pneumococcal pneumonia: younger and more severely affected patients

Nonsteroidal anti-inflammatory drugs for the common cold

Dexmedetomidine versus haloperidol for prevention of delirium during non-invasive mechanical ventilation

Dexmedetomidine vs. haloperidol in delirious, agitated, intubated patients: a randomised open-label trial

Sedation of mechanically ventilated COVID-19 patients: challenges and special considerations

Prone positioning improves oxygenation in spontaneously breathing nonintubated patients with hypoxemic acute respiratory failure: a retrospective study

Efficacy and safety of early prone positioning combined with HFNC or NIV in moderate to severe ARDS: a multi-center prospective cohort study

Use of prone positioning in nonintubated patients with COVID-19 and hypoxemic acute respiratory failure

Respiratory parameters in patients with COVID-19 after using noninvasive ventilation in the prone position outside the intensive care unit

Prone positioning in management of COVID-19 hospitalized patients

Awake prone positioning for nonintubated oxygen dependent COVID-19 pneumonia patients

Prone positioning combined with high-flow nasal or conventional oxygen therapy in severe COVID-19 patients

Early awake prone position combined with high-flow nasal oxygen therapy in severe COVID-19: a case series

Prone positioning in severe acute respiratory distress syndrome

Distinct phenotypes require distinct respiratory management strategies in severe COVID-19

Management of respiratory failure due to COVID-19

Basing respiratory management of coronavirus on physiological principles

Critical care crisis and some recommendations during the COVID-19 epidemic in China

Treating dyspnea: is oxygen therapy the best option for all patients?

Expert recommendations for tracheal intubation in critically ill patients with noval coronavirus disease 2019

Ventilatory support of patients with sepsis or septic shock in resource-limited settings

Respiratory support in COVID-19 patients, with a focus on resource-limited settings

New issues in nursing management during the COVID-19 pandemic in Italy

Vital signs directed therapy for the critically ill: improved adherence to the treatment protocol two years after implementation in an intensive care unit in Tanzania

Vital signs directed therapy: improving care in an intensive care unit in a low-income country

Derivation and validation of SpO2/FiO2 ratio to impute for PaO2/FiO2 ratio in the respiratory component of the sequential organ failure assessment (SOFA) score

Hospital incidence and outcomes of the acute respiratory distress syndrome using the Kigali modification of the Berlin definition

External confirmation and exploration of the Kigali modification for diagnosing moderate or severe ARDS

Acute respiratory failure in COVID-19: is it "typical

COVID-19 pneumonia: ARDS or not?

COVID-19 does not lead to a "typical" acute respiratory distress syndrome

Advanced respiratory monitoring in COVID-19 patients: use less PEEP!

Lung recruitability in COVID-19-associated acute respiratory distress syndrome: a single-center observational study

Critical care management of adults with community-acquired severe respiratory viral infection

Intubation and ventilation amid the COVID-19 outbreak: Wuhan's experience

COVID-19 pandemic: overview of protective-ventilation strategy in ARDS patients

Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome

Liberal or conservative oxygen therapy for acute respiratory distress syndrome

Higher vs lower positive end-expiratory pressure in patients with acute lung injury and acute respiratory distress syndrome: systematic review and meta-analysis

An official American thoracic society/European society of intensive care medicine/society of critical care medicine clinical practice guideline: mechanical ventilation in adult patients with acute respiratory distress syndrome

Fifty years of research in ARDS. Setting positive end-expiratory pressure in acute respiratory distress syndrome

Writing Group for the Alveolar Recruitment for Acute Respiratory Distress Syndrome Trial (ART) Investigators, 2017. Effect of lung recruitment and titrated positive end-expiratory pressure (PEEP) vs low PEEP on mortality in patients with acute respiratory distress syndrome: a randomized clinical trial

Unconventional approaches to mechanical ventilation-step-bystep through the COVID-19 crisis

One ventilator for two patients: feasibility and considerations of a last resort solution in case of equipment shortage

Two for one with split-or co-ventilation at the peak of the COVID-19 tsunami: is there any role for communal care when the resources for personalised medicine are exhausted

Shared ventilation in the era of COVID-19: a theoretical consideration of the dangers and potential solutions

Mechanical ventilation in COVID-19: interpreting the current epidemiology

Comparison of T-piece and pressure support ventilation as spontaneous breathing trials in critically ill patients: a systematic review and meta-analysis

Effect on the duration of mechanical ventilation of identifying patients capable of breathing spontaneously

Efficacy and safety of a paired sedation and ventilator weaning protocol for mechanically ventilated patients in intensive care (awakening and breathing controlled trial): a randomised controlled trial

Daily interruption of sedative infusions in critically ill patients undergoing mechanical ventilation

Baseline characteristics and outcomes of 1591 patients infected with SARS-CoV-2 admitted to ICUs of the Lombardy region

Consensus guidelines for managing the airway in patients with COVID-19: guidelines from the difficult airway society, the association of anaesthetists the intensive care society, the faculty of intensive care medicine and the royal college of anaesthetists

Extubation of patients with COVID-19

Safer intubation and extubation of patients with COVID-19

High-flow nasal cannula compared with conventional oxygen therapy or noninvasive ventilation immediately postextubation: a systematic review and meta-analysis

Sepsis Management in Resource-Limited Settings

Chaisith Sivakorn, Madiha Hashmi, Marcus Schultz, and Alfred Papali were assigned to this subgroup based on their specific expertise and interest in acute respiratory failure and mechanical ventilation. Meetings. An initial Internet subgroup heads meeting was held to establish the procedures for literature review and drafting tables for evidence analysis. The subgroup heads continued work via the Internet. Several meetings occurred through teleconferences and electronic-based discussions among the subgroup heads and with members of other subgroups