We obtained written informed consent from the patient. A 66-year-old man with a history of Type II diabetes mellitus, hypertension, and hyperlipidemia, was presented to our hospital with a one-day history of acute shortness of breath, cough, and fever. Further, he had a recent travel history to Batam, Indonesia.

His vitals were as follows: temperature 38.0°C, blood pressure 130/70 mmHg, pulse 140 bpm, respiratory rate of 25 per minute. His oxygen saturation was 92% on room air, and he was promptly placed on a 40% Venturi mask. Physical examination revealed bilateral crepitations and a 20 × 20 cm back carbuncle.

Blood investigations revealed a white blood cell count of 31000/µl, a hemoglobin level of 11. 5 g/dl, a lactate level of 5.92 mmol/L, and a pro-brain natriuretic peptide level of 731 pg/ml. Arterial blood gas in supplemental oxygen showed a pH level of 7.42, pCO₂ of 35.1 mmHg and pO₂ of 91.9 mmHg with a PaO₂/FiO₂ ratio of 230. A chest X-ray (CXR) revealed bilateral diffuse pulmonary infiltrates.

After discussion, the anesthesia and the surgical teams concluded that the patient required saucerization of the carbuncle for sepsis control, and surgery could not be delayed till the COVID-19 swab results were revealed as the patient was acutely septic. Initially, our hospital conducted COVID-19 diagnostic tests only twice a day, and it would have taken another 6 hours to determine the results of this patient's swab test. Since his recent travel history, infective symptoms, and CXR made him a possible COVID-19 suspect, additional infection control precautions had to be taken.

Prior to the transfer, a team huddle was performed to ensure that all members were aware of the sequences and their roles in the transportation of the patient to the operating theatre (OT). The patient was transported on a plastic-covered trolley while receiving supplemental oxygen through a face mask (Fig. 1A). Signages were posted to specify the transportation route, and security personnel were deployed to divert human traffic during patient transfer. Upon entering the OT, the plastic cover on the trolley was removed and discarded into a biohazard bag.

An airway trolley was placed in the OT, and a C-Mac video laryngoscope with disposable blade was chosen as the first-line airway equipment. Before the surgery was started, the anesthetist assembled all the drugs and equipment required for the procedure on a tray to minimize contact with the drug trolley during the surgery. Whenever additional drugs were required, hand cleansing and glove changing were performed before handling the drug trolley.

An adequate preoxygenation of 8 vital capacity over 1 minute, achieving an FiO₂ of 0.80, and modified rapid sequence induction was our preferred induction technique. Induction agents including 2–3 mg/kg of propofol, 1–2 mg/kg of succinylcholine, and 1–2 µg/kg fentanyl were administered, and mask ventilation was avoided. We ensured that sufficient muscle paralysis was achieved after loss of consciousness before intubation. Thereafter, the airway was secured by the most senior anesthetist using a video laryngoscope, as it was believed that this process had a higher probability of success in the first attempt. Moreover, it avoided repeated instrumentation of the airway. General anesthesia was maintained with desflurane, and mechanical ventilation with a tidal volume of 7 ml/kg was instituted as the patient's lung compliance and oxygenation were normal. As the patient was in respiratory distress on presentation, we decided to keep him intubated to avoid re-intubation.

Fortunately, the patient was tested negative for COVID-19.

During a global pandemic, the importance of minimizing nosocomial transmission cannot be overemphasized. Medical workers have accounted for 3.8% of the total number of COVID-19 cases in mainland China [8]. Anesthetists and other perioperative care providers are particularly at risk during airway management and other procedures performed on patients with COVID-19. Hence, it is crucial to design disease outbreak response measures in conjunction with the surgical, nursing, and allied health staff.

For confirmed or high suspicion cases, identified by the presence of symptoms with CXR changes, the minimum personal protective equipment (PPE) should be powered air-purifying respirator (PAPR). Personnel such as surgeons, anesthetists, and their assistants who are likely to be in contact with droplets, aerosol or bodily fluids, should wear an additional green cape to protect their neck and ears (Fig. 1B). For patients with low suspicion, i.e. asymptomatic but positive contact history, N95 respirators and face shields as well as disposable gowns, are necessary. Within the OT, the number of staff should be limited to a maximum of two senior anesthetists, two anesthesia unit nurses, two surgeons, and two scrub nurses.

Understanding the airflow within the OT is crucial in minimizing the risk of infection. As all our operating theatres were designed to only have a positive pressure, we designated an OT located at a separate bank that utilized a separate filter system from the rest of the OT complex. Access from the emergency department was possible via a dedicated route. The pressure in the scrub room was designed to be more positive than the OT and, the workflow was adapted to maintain human traffic in a single direction (Fig. 1C). For example, if a team member was to experience PAPR malfunction during the surgical procedure, he would go through Route 1 (Fig. 1C) to the doff and perform hand hygiene in the doffing room, before re-entering the donning room. Air exchange was increased from the default setting of 16–18 to a maximum of 20–23 air exchanges per hour.

Since COVID-19 can be spread by contact, surface cleansing of the anesthesia workplace is particularly important. In general, coronaviruses can survive on surfaces for up to nine days [9]; but, is susceptible to killing by 62–71% alcohol, 0.5% hydrogen peroxide, or 0.1% sodium hypochlorite [9]. However, routine cleaning of surfaces and interior storage areas of anesthesia machines and carts is very challenging [9,10]. Hence, plastic covers were placed over 'high touch surfaces' such as the anesthesia machine, patient monitor, computer keyboard, mouse, and touch screens, as these have been shown to reduce the bioburden [11]. In addition, the same OT and anesthesia machine have been designated for use for confirmed/suspected COVID-19 cases throughout the pandemic. Each anesthesia circuit is equipped with a high-efficiency particulate air filter. Both the heat and moisture exchanger filters and soda-lime will be changed after each case. The anesthesia machine, patient monitor surfaces, and patient cart were thoroughly cleaned after use. Equipment that had been dedicated to the isolation OT were labeled to avoid inadvertent exchange and cross-contamination.

As the procedure entailed the complete saucerization of a large hyperemic area, the use of diathermy was inevitable. However, the generator was maintained at the lowest necessary setting in order to reduce the amount of plume generated. In addition, the scrub nurse provided close and constant suctioning using a smoke evacuation device to minimize contamination of the OT air. Given the size of the carbuncle, saucerization was performed in quadrants, starting with a cruciate incision across the center of the affected area. As the infected tissue in each quadrant was excised, hemostasis was maintained, and the operative area was packed with gauze. This reduced the aerosolization of the blood in contact with the heated diathermy. Similarly, additional precautions were undertaken during laparoscopic procedures as a part of our institution's COVID-19 workflow; but did not pertain to this case of open surgery.

Aerosols produced during airway management are particularly hazardous to anesthesia providers [12,13]. A systematic review showed that compared to healthcare workers who did not perform aerosol-generating procedures, those who performed tracheal intubation were at a higher risk of contracting SARS (odds ratio: 6.6), as were those who performed non-invasive ventilation (odds ratio: 3.1), tracheotomy (odds ratio: 4.2), and manual ventilation before intubation (odds ratio: 2.8) [13]. In general, coughing and assisted mask ventilation can generate aerosols and hence, were avoided. A C-Mac video laryngoscope was chosen as it not only allows direct visualization like a direct laryngoscope but also allows the assistant to visualize the airway and facilitate the procedure. Adjuncts like these are useful because visibility is known to be reduced with PPE, and they are crucial to achieve success in intubation in the first attempt since most patients are already in acute hypoxemic respiratory failure with minimal to no respiratory reserve [14]. In addition, such equipment also allows more distance between the operator and the patient's airway and reduces the risk of airborne transmission.

Since we anticipated further respiratory deterioration in the patient, he was kept intubated postoperatively to avoid the risks of re-intubation. The need for emergent intubation limits the time available for donning PPE and increases the likelihood of mistakes in the procedure in increased aerosol exposure. Patients who can be extubated are recovered in the OT till standard Post Anesthetic Discharge Scoring System discharge criteria are achieved. Subsequently, terminal cleaning of the OT is performed and then allowed to ventilate for at least an hour [15].

Since this was the first case experienced by the team, there were several challenges despite having well-designed protocols and processes. Hence, in our opinion all involved healthcare workers should attend briefings and simulations to familiarize themselves with the workflow and avoid inadvertent contamination.

Communication is difficult within PAPR, especially while using a telephone to communicate with people outside the OT. Thus, to circumvent the problem of handwriting misinterpretation, we instituted a 'write and write back' system, similar to the read-back system in a closed-loop communication. Hence, a writing board can be used to communicate with colleagues outside the OT to minimize contact and miscommunication.

In conclusion, organizing surgery for COVID-19 patients requires the involvement, and therefore potential exposure of various healthcare workers. Detailed planning and coordination between departments are required to minimize the risk of disease transmission. It is imperative for the containment measures to be effective yet practical, without hindering patient care, especially during a surgical emergency.