Timing to Intubation COVID-19 Patients: Can We Put It Off until Tomorrow?

Júlio César Garcia de Alencar, Juliana Martes Sternlicht, Alicia Dudy Müller Veiga, Júlio Flávio Meirelles Marchini, Juliana Carvalho Ferreira, Carlos Roberto Ribeiro de Carvalho, Izabel Marcílio, Katia Regina da Silva, Vilson Cobello, Marcelo Consorti Felix, Luz Marina Gómez Gómez, Heraldo Possolo de Souza, Denis Deratani Mauá, Emergency Usp Covid Group, Hcfmusp Covid- Study GroupHealthcare 2022 · DOI

TL;DR

This study investigated the impact of delaying endotracheal intubation on mortality in COVID-19 patients requiring mechanical ventilation. They found that each day of delay in intubation after symptom onset was significantly associated with increased in-hospital mortality, suggesting that earlier intubation might be beneficial for these critically ill patients.

What Problem Does This Paper Address?

The management of severe COVID-19 patients, particularly those developing Acute Respiratory Distress Syndrome (ARDS), presents a significant challenge. A critical decision point is when to transition from non-invasive respiratory support to invasive mechanical ventilation via endotracheal intubation. While early intubation has been advocated by some guidelines to prevent further lung damage from spontaneous breathing efforts, there's also a recognition of the risks associated with invasive ventilation. This paper addresses the uncertainty surrounding the optimal timing of intubation for COVID-19 patients, aiming to quantify the association between the delay in intubation and patient mortality.

Methods

The researchers conducted a unicentric retrospective cohort study. They analyzed data from consecutive adult patients admitted with confirmed COVID-19 who required invasive mechanical ventilation. The primary outcome measured was all-cause mortality within 28 days after intubation. To evaluate the association between the timing of intubation and mortality, they employed a Cox proportional hazards model. This model was used to assess the effect of the time from symptom onset to intubation on mortality. They also performed multivariable Cox models, adjusting for various clinical factors and a validated score (4C score) to account for patient baseline characteristics and severity.

Datasets

The study utilized data from the electronic medical records of patients admitted to the Hospital das Clínicas da Universidade de São Paulo (HC-FMUSP) between March 2020 and August 2020. The dataset comprised 3020 COVID-19 patients, of whom 592 required invasive mechanical ventilation and were included in the analysis. Specific data points collected included patient demographics, COVID-19 RT-PCR results, admission dates, symptom onset dates, oxygen saturation (SpO2), and outcomes (death or discharge).

Key Insights and Ideas

  • The "wait-and-see" approach might be detrimental: Despite the inherent risks of intubation, the study suggests that delaying this intervention in COVID-19 patients with respiratory failure could lead to worse outcomes.
  • Timing matters, even after accounting for severity: The association between delayed intubation and increased mortality persisted even after adjusting for multiple clinical variables and a severity score (4C score), strengthening the finding.
  • Symptom onset is a crucial reference point: Using symptom onset as the starting point for calculating the delay, rather than admission date, provides a more accurate reflection of the disease progression and the impact of delayed intervention.
  • Pandemic context influenced decisions: The authors acknowledge that during the early stages of the pandemic, uncertainty about optimal management and fear of intubation-related complications might have contributed to delayed intubation decisions.

Key Results

  • A total of 592 patients (20% of those admitted with COVID-19) were intubated.
  • Out of these, 310 patients died within 28 days of intubation.
  • The median time from symptom onset to intubation was 9 days (IQR 7–14).
  • For each additional day of delay in intubation after symptom onset, the unadjusted hazard ratio for death within the follow-up period was 1.006.
  • After multivariable adjustment (including the 4C score), the hazard ratio for death was 1.018 for each additional day of delay in intubation. This indicates a statistically significant association between delayed intubation and increased mortality.

Limitations and Future Work

The authors acknowledge several limitations:

  • Retrospective design: This study design makes it difficult to definitively establish causality between delayed intubation and increased mortality. It's possible that sicker patients were more likely to have their intubation delayed.
  • Inability to determine factors influencing clinician decisions: The study could not identify the specific reasons behind clinicians' choices regarding the timing of intubation.
  • Limited granularity of pharmacologic data: The retrospective data did not allow for a detailed examination of the impact of pharmacological therapies administered before or after intubation.

Future work could focus on prospective studies to establish causality and explore the specific factors that influence intubation timing decisions. Further research into optimal timing strategies, potentially incorporating dynamic risk assessment, would also be valuable.

Summary updated

Abstract

The study protocol was approved by the local Ethics Committee (opinion number 3.990.817; CAAE: 30417520.0.0000.0068).
BibTeX
@inproceedings{alencar2022timing,
  author = {Júlio César Garcia de Alencar and Juliana Martes Sternlicht and Alicia Dudy Müller Veiga and Júlio Flávio Meirelles Marchini and Juliana Carvalho Ferreira and Carlos Roberto Ribeiro de Carvalho and Izabel Marcílio and Katia Regina da Silva and Vilson Cobello and Marcelo Consorti Felix and Luz Marina Gómez Gómez and Heraldo Possolo de Souza and Denis Deratani Mauá and Emergency USP COVID Group and HCFMUSP Covid- Study Group},
  title = {{Timing to Intubation COVID-19 Patients: Can We Put It Off until Tomorrow?}},
  booktitle = {Healthcare},
  year = {2022},
  doi = {10.3390/healthcare10020206},
}
Figure