Research evidence for ventilatory support recommendations in sepsis and septic shock management has been mainly gathered from investigations in resource-rich settings. Often, it is not practical to directly translate this evidence to resource-limited settings. Indeed, resource-limited intensive care units (ICUs) are frequently restricted in the availability of equipment, laboratory support and skilled staff. We report on a set of pragmatic recommendations for ventilatory support in sepsis and septic shock management in resource-limited settings, built upon two previous sets of guidelines for sepsis management, the most recent Surviving Sepsis Campaign guidelines [1] and the recommendations for sepsis management by the Global Intensive Care Working Group of the ESICM [2], as well as upon an updated literature search and expert opinions [3].

An international team of physicians from resource-rich and -limited settings with hands-on experience in resource-limited ICUs critically appraised a list of questions regarding ventilatory support by partly using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) tools [4] and by reviewing the literature for any additional evidence from resource-limited settings. The quality of evidence was scored from very high (A) to very low (D), and the strength of recommendations was strong (1) or weak (2) considering the indirectness of the evidence, the magnitude of effects, availability and feasibility in resource-limited ICUs, and safety. In the absence of evidence from resource-limited ICUs, evidence from resource-rich ICUs was adopted after pragmatic, experience-based appraisal. Full scoring details are available in the online supplement.

Results and recommendations for ventilatory support in resource-limited ICUs

Although ventilatory support is generally seen as a simple and potentially life-saving intervention, there is increasing evidence for its possible harmful effects. Therefore, besides grading the evidence for general ventilatory settings (ventilation mode, oxygen concentration, tidal volume, positive end-expiratory pressure (PEEP) and recruitment maneuvers), we weighted strategies that might obviate the need for invasive ventilation, or shorten the duration of ventilation through accelerated weaning. The literature search for additional evidence from resource-limited settings identified several investigations that could be used to grade the evidence for use in resource-limited ICUs, amounting to one randomized controlled trial [5] and eight observational studies [612]. Key recommendations are provided in Table 1 and more detailed information on the literature search and grading of recommendations is included in the online supplement. In cases of strong recommendations, we use the wording ‘we recommend …’; in cases of less strong recommendations, we use the wording ‘we suggest …’.

Table 1 Recommendations and suggestions on ventilatory support in patients with sepsis or septic shock in resource-limited settings (with grading)

We suggest identifying patients with ARDS diagnosis in the ICUs where this is feasible, as this may improve proper selection of additional ventilator strategies—we do recognize that this is not the case in the majority of ICUs with limited resources and therefore our overall strength of recommendation for this matter is downgraded to a suggestion; in settings where blood gas analyzers are unavailable, the SpO2 relative to delivered oxygen concentration (S/F) could be used for decision-making and continuous monitoring; in the absence of chest radiography, use of ultrasound exam of lungs and heart could be helpful in narrowing down the diagnosis of non-cardiogenic pulmonary edema, where feasible; regardless of the feasibility of diagnosing ARDS, septic patients with acute respiratory failure should be managed by employing the principles of lung-protective ventilation; we recommend that the elevated head-of-bed position should be maintained in all ventilated patients to decrease the risk of aspiration; we recommend the use of invasive mechanical ventilation in cases of severe hemodynamic disturbance or severe hypoxemia, and suggest a trial of non-invasive ventilation only in patients with minor hemodynamic and oxygenation disturbances; and in these patients, we recommend close monitoring and frequent reassessments regarding the need for intubation to assure that mechanical ventilation is instituted without the delay; we suggest using a minimum level of PEEP of 5 cm H2O in all patients with sepsis or septic shock with acute respiratory failure, and suggest using higher levels of PEEP only in patients with moderate or severe ARDS; we suggest a preference for a low FiO2 aiming at low oxygenation goals [i.e., 8 kPa (60 mmHg) and/or SpO2 (88–95 %)], and we suggest using PEEP/FiO2 tables to find the best PEEP–FiO2 combination; we suggest applying recruitment maneuvers only in patients with moderate or severe ARDS with refractory hypoxia, and only when the staff are trained and experienced in performing these maneuvers; patients who need higher levels of PEEP and recruitment maneuvers are preferably closely monitored, preferably by using an arterial line, as hypotension and circulatory depression may develop; we recommend using low tidal volumes in patients with ARDS (and to avoid tidal volumes larger than 10 ml/kg predicted body weight, and to consider tidal volumes between 5 and 7 ml/kg predicted body weight in all patients); where feasible, end-tidal CO2 monitoring could be helpful in timely recognition of dislodgement of the endotracheal tube and under- or overventilation; ‘volume-controlled’ modes could be safer than ‘pressure-controlled’ modes as minute ventilation and tidal volume size is guaranteed with volume-controlled modes; we cannot recommend a preference for ‘support’ modes of ventilation over ‘control’ modes, but do recommend checking regularly whether a patient tolerates ‘support’ mode; we also suggest performing spontaneous breathing trials to timely identify patients who are ready for extubation, but also to plan extubating patients when sufficient staff are around to guarantee safe re-intubation, if necessary.


We provide a set of simple, readily available and affordable recommendations for the ‘safe’ ventilatory support in patients with sepsis or septic shock in resource-limited ICUs. Most evidence has come from resource-rich settings; therefore, there is an urgent need for related studies in resource-limited settings. Given the immense variability and range of capabilities, preparedness, and staffing of ICUs in resource-limited ICUs, each ICU practitioner will have to rationally and practically further adapt the guidelines based on their site-specific circumstances.