Tracheal intubation in COVID-19 patients: update on recommendations. Response to Br J Anaesth 2020; 125: e28–37

Editor—We thank Sethi and Sethi¹ for their interest and comments on our paper² that combined a report of experiences in Wuhan, China, early in the pandemic accompanied by consensus recommendations made by international experts. This unusual approach was adopted to meet the needs of the rapidly emerging global crisis by simultaneously reporting data and providing guidance.² We now provide some updates to our recommendations based on further developments.

The recommendation to consider a fluid bolus before induction of anaesthesia came from several co-authors. Early in this pandemic, a restrictive fluid strategy was common as part of a strategy to minimise hypoxaemia in patients with acute lung injury. However, restrictive fluid strategies were modified in the light of increased recognition of severe dehydration and high rates of acute renal failure in patients with COVID-19. Indeed, supplies for renal replacement therapy were often a greater limitation to critical care delivery than was availability of ventilators. A more liberal resuscitative approach is now adopted by many. Jaber and colleagues³ showed a decrease in life-threatening complications in the ICU after the implementation of a tracheal intubation management protocol that included fluid loading before tracheal intubation unless contraindicated.

Regarding ‘prophylactic use of vasopressors’, we advocate ‘immediate availability and appropriate use of prophylactic cardiovascular-stimulating agents’. This is in line with guidelines for avoidance of cardiovascular collapse in the critically ill.⁴ We do not agree that comparing results between two hospitals in our retrospective and observational study provides value for guiding clinical practice because of the potential for multiple confounding factors. Of note, the four cardiac arrests occurred in Hospital B, where prophylactic vasopressors were not administered.

Hypotension and cardiac arrest are common during induction of general anaesthesia in critically ill patients. Cardiac arrest during tracheal intubation is associated with poor outcome.⁵ The cases in our series support this: all the four cardiac arrests in Hospital B occurred during induction of anaesthesia, and each was followed by immediate advanced life support and successful resuscitation, without cardiac defibrillation. However, all four patients subsequently died from multiple organ failure. Anaesthetic drug choice, fluid administration, and use of cardiovascular-stimulating drugs whether during or after induction of anaesthesia all contribute to safe emergency tracheal intubation. The use of these interventions may be affected by available time and efforts for pre-induction patient preparation, availability of induction agents (such as etomidate or ketamine), and clinical judgement. Considering the observed high incidence of cardiovascular collapse in our series, we believe we provided a balanced approach with our recommendation for the prophylactic use of fluids and vasopressors and the use of alternative anaesthetic induction agents. Optimising patient condition before induction of anaesthesia and minimising hypoxaemia with timely and prompt tracheal intubation at the first attempt may also increase safety.

About one in eight (12.9%) of the patients in our series was unconscious. Causes included profound hypotension, severe hypoxaemia, carbon dioxide retention, electrolyte disturbance, and acute encephalomyelitis. This illustrates the severity of illness amongst patients referred to critical care at that stage of the pandemic.

Regarding ‘rapid sequence induction’ (RSI), it has been widely taught for many years that RSI does not include mask ventilation. This has recently been challenged, and whilst ‘modified RSI’ is a loose and undefined term, we suggest that most would regard mask ventilation during RSI to be a modification of classical teaching.⁶ Casey and colleagues⁷ recently confirmed that mask ventilation after induction of anaesthesia and before tracheal intubation could ameliorate severe hypoxaemia in critically ill patients. We agree that infection control issues suggest that mask ventilation should be avoided when possible. We advocate the use of preventative measures, such as apnoeic oxygenation. Further, we propose that it is not controversial to recommend the use of mask ventilation when severe hypoxaemia supervenes.⁸ We recommend using two-hand two-person mask ventilation with a ‘VE’ grip to assure a tight seal, combined with the use of a viral filter. The airway team should be dressed in airborne-precaution personal protective equipment (PPE) throughout this aerosol-generating medical procedure. Supervised donning and doffing of PPE remain a critical step in the avoidance of cross-infection in healthcare workers.

Declarations of interest

HW is a consultant of Well Lead Medical Company, Guangzhou, Guangdong, China. The other authors have no relevant conflicts to declare.