Early versus late tracheostomy in critically ill COVID-19 patients.

BACKGROUND

The role of early tracheostomy as an intervention for critically ill COVID-19 patients is unclear. Previous reports have described prolonged intensive care stays and difficulty weaning from mechanical ventilation in critically ill COVID-19 patients, particularly in those developing acute respiratory distress syndrome. Pre-pandemic evidence on the benefits of early tracheostomy is conflicting but suggests shorter hospital stays and lower mortality rates compared to late tracheostomy.

OBJECTIVES

To assess the benefits and harms of early tracheostomy compared to late tracheostomy in critically ill COVID-19 patients.

SEARCH METHODS

We searched the Cochrane COVID-19 Study Register, which comprises CENTRAL, PubMed, Embase, ClinicalTrials.gov, WHO International Clinical Trials Registry Platform, and medRxiv, as well as Web of Science (Science Citation Index Expanded and Emerging Sources Citation Index) and WHO COVID-19 Global literature on coronavirus disease to identify completed and ongoing studies without language restrictions. We conducted the searches on 14 June 2022.

SELECTION CRITERIA

We followed standard Cochrane methodology. We included randomized controlled trials (RCTs) and non-randomized studies of interventions (NRSI) evaluating early tracheostomy compared to late tracheostomy during SARS-CoV-2 infection in critically ill adults irrespective of gender, ethnicity, or setting.

DATA COLLECTION AND ANALYSIS

We followed standard Cochrane methodology. To assess risk of bias in included studies, we used the Cochrane RoB 2 tool for RCTs and the ROBINS-I tool for NRSIs. We used the GRADE approach to assess the certainty of evidence for outcomes of our prioritized categories: mortality, clinical status, and intensive care unit (ICU) length of stay. As the timing of tracheostomy was very heterogeneous among the included studies, we applied GRADE only to studies that defined early tracheostomy as 10 days or less, which was chosen according to clinical relevance.

MAIN RESULTS

We included one RCT with 150 participants diagnosed with SARS-CoV-2 infection and 24 NRSIs with 6372 participants diagnosed with SARS-CoV-2 infection. All participants were admitted to the ICU, orally intubated and mechanically ventilated. The RCT was a multicenter, parallel, single-blinded study conducted in Sweden. Of the 24 NRSIs, which were mostly conducted in high- and middle-income countries, eight had a prospective design and 16 a retrospective design. We did not find any ongoing studies. RCT-based evidence We judged risk of bias for the RCT to be of low or some concerns regarding randomization and measurement of the outcome. Early tracheostomy may result in little to no difference in overall mortality (RR 0.82, 95% CI 0.52 to 1.29; RD 67 fewer per 1000, 95% CI 178 fewer to 108 more; 1 study, 150 participants; low-certainty evidence). As an indicator of improvement of clinical status, early tracheostomy may result in little to no difference in duration to liberation from invasive mechanical ventilation (MD 1.50 days fewer, 95%, CI 5.74 days fewer to 2.74 days more; 1 study, 150 participants; low-certainty evidence). As an indicator of worsening clinical status, early tracheostomy may result in little to no difference in the incidence of adverse events of any grade (RR 0.94, 95% CI 0.79 to 1.13; RD 47 fewer per 1000, 95% CI 164 fewer to 102 more; 1 study, 150 participants; low-certainty evidence); little to no difference in the incidence of ventilator-associated pneumonia (RR 1.08, 95% CI 0.23 to 5.20; RD 3 more per 1000, 95% CI 30 fewer to 162 more; 1 study, 150 participants; low-certainty evidence). None of the studies reported need for renal replacement therapy. Early tracheostomy may result in little benefit to no difference in ICU length of stay (MD 0.5 days fewer, 95% CI 5.34 days fewer to 4.34 days more; 1 study, 150 participants; low-certainty evidence). NRSI-based evidence We considered risk of bias for NRSIs to be critical because of possible confounding, study participant enrollment into the studies, intervention classification and potentially systematic errors in the measurement of outcomes. We are uncertain whether early tracheostomy (≤ 10 days) increases or decreases overall mortality (RR 1.47, 95% CI 0.43 to 5.00; RD 143 more per 1000, 95% CI 174 less to 1218 more; I = 79%; 2 studies, 719 participants) or duration to liberation from mechanical ventilation (MD 1.98 days fewer, 95% CI 0.16 days fewer to 4.12 more; 1 study, 50 participants), because we graded the certainty of evidence as very low. Three NRSIs reported ICU length of stay for 519 patients with early tracheostomy (≤ 10 days) as a median value, which we could not include in the meta-analyses. We are uncertain whether early tracheostomy (≤ 10 days) increases or decreases the ICU length of stay, because we graded the certainty of evidence as very low.

AUTHORS' CONCLUSIONS: We found low-certainty evidence that early tracheostomy may result in little to no difference in overall mortality in critically ill COVID-19 patients requiring prolonged mechanical ventilation compared with late tracheostomy. In terms of clinical improvement, early tracheostomy may result in little to no difference in duration to liberation from mechanical ventilation compared with late tracheostomy. We are not certain about the impact of early tracheostomy on clinical worsening in terms of the incidence of adverse events, need for renal replacement therapy, ventilator-associated pneumonia, or the length of stay in the ICU. Future RCTs should provide additional data on the benefits and harms of early tracheostomy for defined main outcomes of COVID-19 research, as well as of comparable diseases, especially for different population subgroups to reduce clinical heterogeneity, and report a longer observation period. Then it would be possible to draw conclusions regarding which patient groups might benefit from early intervention. Furthermore, validated scoring systems for more accurate predictions of the need for prolonged mechanical ventilation should be developed and used in new RCTs to ensure safer indication and patient safety. High-quality (prospectively registered) NRSIs should be conducted in the future to provide valuable answers to clinical questions. This could enable us to draw more reliable conclusions about the potential benefits and harms of early tracheostomy in critically ill COVID-19 patients.