Early percutaneous tracheostomy for patients with COVID-19.

The COVID‐19 pandemic has led to unprecedented demand for critical care. Early tracheostomy may reduce the duration of mechanical ventilation, intensive care unit length of stay and sedative requirements and facilitate resource management during the pandemic. Although standard UK practice is to consider tracheostomy after 7–10 days, COVID‐19 guidelines advise balancing the complications of prolonged tracheal intubation vs. the risks of viral exposure for staff and waiting for 14 days after tracheal intubation and a negative SARS‐CoV‐2 antigen test [1]. However, the rate of failed tracheal extubation observed in patients with COVID‐19, with possible associated negative outcomes, may favour weaning via earlier tracheostomy.

We believe that, with precautions, early percutaneous tracheostomy can be safe for patients with COVID‐19 and for staff, and describe our experience of 29 percutaneous tracheostomies between 19 March and 14 April 2020 at our Trust. Baseline characteristics are presented in Table 1. Overall, staff safety was maintained, there were no complications during any of the procedures, and sedative and vasopressor requirements decreased substantially following tracheostomy.

Table 1

Characteristics of 29 patients who received a tracheostomy. Values are number (proportion) or median (IQR [range]).

Sex; male	25 (86%)
Black, Asian or Minority Ethnic	4 (14%)
Age; years	66 (59–70 [36–82])
BMI; kg.m−2	27.6 (24.6–32.0 [21.2–54.0])
Days of ventilation before tracheostomy	4 (2–8 [1–13])
Co‐existing conditions
Any co‐existing condition	22 (76%)
Hypertension	12 (41%)
Diabetes	6 (20%)
Asthma	6 (20%)
Chronic obstructive pulmonary disorder	0
Obesity	12 (41%)

Potential for transmission was minimised by using full personal protective equipment for aerosol‐generating procedures and a modified percutaneous technique. Aerosol generation was reduced by switching the ventilator to standby at key points and performing tracheostomy with minimal delay under apnoeic conditions. The same experienced operator performed every tracheostomy, and received a negative SARS‐CoV‐2 antibody test result 2 months after the last tracheostomy. No other member of the treating team developed COVID‐19 symptoms or received a positive test result within 2 weeks of a tracheostomy. Reassuringly, this is echoed in nationwide data, with no instances of COVID‐19 infection among tracheostomy operators (Hamilton et al., preprint: https://www.medrxiv.org/content/10.1101/2020.05.22.20104679v1).

Further, we found early tracheostomies can be performed safely in patients with COVID‐19. Our median time to tracheostomy was 4 days after tracheal intubation, and there were no peri‐operative complications in our cohort. It is our usual practice to use a percutaneous technique, which allowed procedures to take place at the patient bed‐side in an area already designated for COVID‐19. A percutaneous technique is also associated with a lower complication rates than surgical techniques in both COVID‐19 and non‐COVID‐19 patients [Hamilton et al., preprint: https://www.medrxiv.org/content/10.1101/2020.05.22.20104679v1, 3]. Despite initial concerns, we actually found it easier to prone patients who had a tracheostomy, rather than a tracheal tube, when we used prone head supports, chest and pelvic rolls. All 29 tracheostomies were secured with sutures in addition to a tracheostomy‐securing device.

Finally, there was a substantial decrease in sedation and vasopressor requirements following tracheostomy. Sedation decreased from a median of more than 10 mg.h−1 of both morphine and midazolam, or equivalent sedatives, while intubated to less than 5 mg.h−1 during each of the 5 days after tracheostomy, with 48% of patients requiring zero sedation. In addition, noradrenaline requirements decreased from 5 to 0 μg.min−1 over 5 days. In our experience, without tracheostomy these patients would have remained on higher doses of both for a longer period. At our peak, we experienced shortages of sedatives, vasopressors and syringe drivers as we ventilated patients at 660% of our usual level 3 capacity. Early tracheostomy may preserve resources for other current and future patients during a pandemic without compromising patient or staff safety. It also facilitated earlier rehabilitative physiotherapy. With our local neurological rehabilitation centre, we established a step‐down ward to treat patients with tracheostomies, although ultimately only one patient was not decannulated before step‐down.

Waiting for 14 days may avoid tracheostomy for subsequent non‐survivors but it could negate the benefits outlined above. In addition, increased duration of mechanical ventilation, particularly if using anaesthetic machines, may have associated complications, such as delirium and longer‐term psychological sequelae.

In summary, we found that early tracheostomy can be performed in a way that is safe and potentially beneficial for patients and healthcare staff. We suggest that decisions regarding tracheostomy for patients with COVID‐19 should be based on the best interests of the patient on a case‐by‐case basis, whereas maintaining the safety of the healthcare team; an approach we aspire to for all of our intensive care patients.