Brigham and Women's Hospitals

Respiratory

Updated: May 19, 2020

Acute Lung Injury (ALI) and Acute Respiratory Distress Syndrome (ARDS)

Pathophysiology

  1. Histology of COVID-19 associated lung disease shows bilateral diffuse alveolar damage with cellular fibromyxoid exudates, desquamation of pneumocytes, pulmonary edema, and hyaline membrane formation. There is also some evidence of direct viral injury to lung tissue, not just inflammatory sequelae. (Xu et al, Lancet Respir Med, 2020).
  2. Anecdotally, patients with COVID-related lung disease have significantly higher compliance than is typical for their shunt fraction, indicating this may be a very different phenotype than typical ARDS. The explanation remains unclear, with pulmonary perfusion dysregulation posited as one possible explanation. (Gattinoni, AJRCCM, 2020)
  3. The SARS-CoV-2 virus binds to the ACE2 receptor as its target receptor for cell entry which may be an explanation for many of the pathophysiological manifestations of infection. The ACE2 receptor is expressed by select populations of cells including the pulmonary endothelium, Alveolar Type 2 cells, proximal renal tubule cells, gastrointestinal epithelial cells, and many other others. The cells that express ACE2 may be the cell populations injured by infection or targeted by the immune response. Hypothesized manifestations include the following:
  1. In the pulmonary endothelium, ACE2 acts to downregulate angiotensin II (a potent vasoconstrictor) while increasing levels of angiotensin (1,7) (a vasodilator). Infection with SARS-CoV-2 may impair this pathway or may lead to imbalances in the ACE/ACE2 pathways with uncertain effects on hypoxic vasoconstriction mechanisms. Impairment of this pathway may induce increased shunt and resultant hypoxia.
  2. Alveolar Type 2 cells secrete surfactant. Their selective injury caused by SARS-CoV-2 infection may lead to selective loss of surfactant in a subgroup of patients resulting in increased derecruitment at low opening pressures and a highly PEEP sensitivity to maintaining lung function. These patients may experience prolonged recovery periods dependent on repopulation of Alveolar Type 2 cells and reconstitution of surfactant.

Definition of Acute Respiratory Distress Syndrome (ARDS)

  1. Most patients with COVID-19 who require ICU level of care will develop ARDS.
  2. The Berlin definition of ARDS requires the following four criteria:
  1. Acute (onset over 1 week or less)
  2. Bilateral opacities detected on CT or chest radiograph
  3. PF ratio <300mmHg with a minimum of 5 cmH20 PEEP (or CPAP)
  4. Must not be fully explained by cardiac failure or fluid overload

Severity

PaO2/FiO2 (on PEEP/CPAP >5)

Mortality (all cause, cohort)

Mild

200-300

27%

Moderate

100-200

32%

Severe

<100

45%

Time course

  1. Anecdotally, many report that progression of hypoxemic respiratory failure occurs rapidly (within ~12-24 hours).
  2. From onset of symptoms, the median time to:
  1. Development of ARDS: 8-12 days (Wang et al, JAMA, 2020; Zhou et al, Lancet, 2020; Huang et al, Lancet, 2020)
  2. Mechanical ventilation: 10.5-14.5 days (Huang et al, Lancet, 2020; Zhou et al, Lancet, 2020)

Hypoxemia Management

Supplemental Oxygen Support

  1. Goals of therapy:
  1. Maintain target SpO2 92-96%
  1. Target SpO2 88-94% in patients with oxygen-dependent COPD
  1. Maintain stable work of breathing
  1. Goal respiratory rate < 24
  2. Target normal respiratory effort (no signs of accessory muscle use or obvious increased respiratory work)
  1. Supplemental oxygen support:
  1. Initial oxygen delivery should be humidified nasal cannula (NC) titrated from 1 to 6 LPM to meet goals of therapy.
  2. If goals of therapy are not met at 6 LPM NC then advance to either:
  1. Oxymizer mustache:
  1. Initiate at 6 LPM
  2. Titrate to maximum of 12 LPM to meet goals of therapy
  1. Venturi mask
  1. Initiate at FiO2 40% (check device’s instructions to determine minimum flow rate needed to achieve 40%)
  2. Titrate to maximum of FiO2 60% to meet goals of therapy (some devices are not able to achieve Fi02 of 60%, please discuss with RT)
  1. Considerations during oxygen support escalation:
  1. Clarify goals of care and appropriateness of ICU hospitalization prior to escalating to ICU transfer and pursuing intubation
  2. Consider awake self-proning in selected patients
  3. Consider the rate of change of oxygen escalation as well as pre-existing cardiopulmonary disease in determining threshold for ICU transfer (such as COPD patient with pre-existing supplement oxygen use at baseline)
  4. Consider Pulmonary Consult if there are concerns for other etiologies of hypoxia
  1. If appropriate, consult ICU for triage and evaluation:
  1. If SpO2 < 92% (<88% in COPD) or unstable work of breathing at
  1. Oxymizer at 10 LPM or Venturi mask at FiO2 50%
  1. ICU triage (pager #39999)

Intubation

  1. If appropriate, call COVID airway team for intubation:
  1. If SpO2 < 92% or unstable work of breathing at:
  1. Oxymizer at 12 LPM or Venturi mask at FiO2 60%
  1. COVID anesthesia airway team (pager #39265)
  1. Page STAT line if code/ emergent (pager #26555)
  1. Notify team if you anticipate needing arterial line or central access, as they may be able to place the line while in full intubating PPE (see “Bedside procedures)
  2. Notify team of any known prior difficult intubations, prior head or neck surgery or radiation therapy, or know airway abnormalities
  1. Pre-oxygenation for patients on advanced supplemental oxygen support:
  1. Increase Venturi mask to FiO2 100% (consult RT if not possible with your device) or Oxymizer to 15 LPM prior to planned intubated
  1. Avoid NIPPV or HFNC to stave off intubation (see discussion above)
  1. For patients already on NIPPV/HFNC, transition to Venturi mask or non-rebreather mask if possible, ideally 45 minutes prior to intubation
  1. Rapid Sequence Induction(RSI) should be performed by the most experienced airway provider without bag-valve masking and using a video laryngoscope (SCCM COVID19 Guidelines)(APSF Considerations for Airway Manipulation, 3/20/2020).
  1. For more detailed instructions, see Intubation
  1. Intubations outside the ICU should be attended by the Resource RT, who can facilitate early and appropriate ventilator settings
  2. After intubation, see “Initial Mechanical Ventilation for next steps

Non-invasive Positive Pressure Ventilation (NIPPV) and High Flow Nasal Cannula (HFNC)

  1. Avoid HFNC and NIPPV (CPAP/BIPAP) in most PUI and COVID patients
  1. Reason 1: Aerosolization remains a concern
  1. NIPPV: General consensus suggests that NIPPV increases the risk of viral transmission, but the degree of aerosolization is poorly understood and data on this is lacking. A systematic review on SARS found that NIPPV was associated with increased risk of viral transmission to healthcare workers (n=2 studies), but HFNC was not (n=1) (Tran et al, PLoS One, 2012)
  1. “Helmet” NIPPV may pose less risk than face mask NIPPV, but again data is limited
  1. HFNC: Data on aerosol risk is unclear. One post-hoc analysis that found no secondary infections in medical staff from patients with influenza H1N1 treated with HFNC (but n=20) (Rello et al, J Crit Care, 2012). A recent non-peer reviewed pre-print (Iwashyna et al, medRXiv, 2020) tested aerosol levels in healthy volunteers and concluded that there was no variation in aerosol level among room air, 6L/min NC, 15 L/min NRB, 30L/min HFNC and 60 L/min HFNC regardless of coughing. Similarly, a randomized, controlled crossover trial of HFNC versus conventional oxygen mask found no increase in air or contact surface contamination by bacteria (Leung et al, J Hosp Infect, 2020)
  1. Reason 2: The benefit is unclear
  1. Given the rapid progression of disease in most patients, we do not anticipate many patients would avoid intubation using NIPPV/HFNC, but this remains unknown and may change as we gain more experience
  1. Generally, NIPPV is thought to stave off intubation only in early ARDS and the data is inconsistent (Rochberg et al, ERJ, 2016).
  2. Case reports from China suggest high failure rates for non-invasive ventilation, including high-flow nasal oxygen (Zuo et al, Chin Med Sci J, 2020), though there are some patients who may recover on HFNC.
  1. Exceptions:
  1. Obstructive Sleep Apnea or Tracheobronchomalacia:
  1. Patients on home nocturnal CPAP or BiPAP should continue nocturnal NIPPV, but only following the instructions below.
  1. Some causes of rapidly-reversible hypoxemia:
  1. Consider NIPPV on a case-by-case basis for rapidly reversible causes of hypoxemia, such as flash pulmonary edema or COPD exacerbation.
  2. For cases where it is unclear whether NIPPV should be used, consult the SP-ICU triage pager (#39999) and/or biothreats pager to discuss.
  1. If NIPPV is used:
  1. Use BWH NIPPV machine with dual limb with a HEPA filter and BWH mask without anti-asphyxia valve.
  1. Patients may NOT use a home NIPPV mask or nasal pillow or single-limb machine due to increased aerosol risk.
  1. Use under strict airborne precautions, including N95s, strict isolation, and a negative pressure room.
  2. Ensure masks/devices fit well and there is minimal air leak
  1. Measured exhaled air distances are minimally increased with CPAP pressures up to 20 cm H2O and HFNC up to 60 LPM; importantly device/interface leaks cause significant lateral air travel (Hui et al, Eur Respir J, 2019)
  1. If HFNC is used:
  1. We generally do not recommend use. However, if it is used, patients should wear surgical masks and flow rate should be limited to < 30 L/min

Self-proning

  1. Potential benefits of self-proning:
  1. Proning is thought to provide physiologic benefits for patients with COVID infection: it improves recruitment of alveoli in dependent areas of the lungs and it may improve perfusion to ventilated areas, improving V/Q mismatching. Typically proning is used in ventilated ICU patients, however the same benefits may accrue to non-ventilated patients.
  2. Intubated proning: Proning is one of the mainstays of ARDS therapy for intubated patients, showing both 28 day and 90 day mortality benefit in the PROSEVA 2013 trial. In intubated ARDS patients with P/F <150
  3. Self-proning (non-intubated) in non-Covid-19 patient cohorts: ARDS, after lung transplant, and post-surgery. These small studies showed that self-proning was associated with lab, radiographic, or clinical improvement
  1. In one observational study, (Scaravilli et al, J of Critical Care, 2015) 15 patients with pneumonia underwent a total of 43 self-proning procedures Of the 43 procedures, 24 were performed with O2 mask, 1 with HFNC, 11 with helmet CPAP, 7 with NIV., with an average of 3 hours, range 2-8 hours. They found improvement in P/F ratio Pre P/F 127 +/- 49 --> Prone 186 +/- 72 --> Post 141 +/- 64 (p < 0.05). and PaO2 without complications. (“displacement of indwelling catheters, facial edema, pressure sores, pressure neuropathies, compression of nerves, and retinal vessels or vomiting”)
  2. In another limited study of 20 patients (Ding et al, Critical Care, 2020) with ARDS with P/F < 200 requiring NIV or HFNC of at least PEEP of 5 and FiO2 of 0.5 who underwent self-proning for at least 30 minutes, many fewer (45%) required intubation than would have been expected based on previous data (75%)
  1. Use in COVID-19:
  1. In one study (N=24), Covid-19 patients tolerated self-proning for 1 hr (17%, all required intubation within 72 hours), 1 to 3 hrs (21%) or more than 3 hrs (63%). Only 1 of the patients who tolerated proning for more than 1 hr was intubated in 10d follow-up. The only patients with a significant increase in PaO2 were the patients who self-proned for at least 3 hours (74 mmHg to 95 mmHg). This increase in PaO2 was not sustained after re-supination. Back pain was a limiting factor to self-proning (Elharrar et al, JAMA, 2020)
  1. Patient Selection:
  1. Eligibility:
  1. Self-proning can be used on stable patients (on room air or on supplemental oxygen) and as a “rescue” for those who have escalating supplemental O2 requirements.
  2. Multidisciplinary discussion is required with provider, RT, and RN.
  3. Patient must be able to move independently and have the cognitive and physical status to supinate themselves if they become uncomfortable. This includes the ability to prone/supinate while safely managing their supplemental oxygen, IV tubing, SpO2 monitor and other leads and attachments (within reason).
  1. Contraindications:
  1. Absolute:
  1. Inability to independently supinate or pronate safely (see above)
  2. Imminent risk of intubation (see “when to stop self-proning”)
  3. Spinal instability
  4. Facial or pelvic fractures
  5. Open chest or unstable chest wall
  6. Open abdomen
  1. Relative contraindications:
  1. Altered mental status
  2. Nausea or vomiting
  3. NIPPV
  1. Protocol:
  1. RN documentation
  1. RN should document prone/supine in the EPIC flowsheet under “Daily Cares” with every vitals sign check
  2. In addition, RN documents just prior to proning, 1 hr after proning, and then at termination of proning:
  1. Supine/Prone position, patients subjective assessment of their breathing, Sp02 (ABG not required), oxygen device and flow rates, respiratory rate.
  2. The goal is to identify which patients respond to this treatment using objective and subjective criteria
  1. Monitoring
  1. Continuous O2 monitoring is required. If telemetry indicated, EKG leads can remain on anterior chest wall for continuous monitoring; attempt to avoid pressure
  1. Prior to proning
  1. Make plans in advance for toileting, call bell, entertainment, and cellular phone
  2. If possible, place the bed in reverse Trendelenburg (head above feet, 10 degrees) to help reduce intraocular pressure.
  3. Have patient empty bladder
  4. Educate the patient. Explain the procedure and rationale of the intervention to the patient. “Lying on your stomach is what we call prone position. This position can improve your breathing, helping your lungs to expand to get oxygen to the rest of your body. It may help you feel better.”
  5. Arrange tubing to travel towards the top of the bed, not across the patient, to minimize risk of dislodging. Ensure support devices are well-secured to the patient. (Ex. Sleeve over IV access site, position urinary catheter)
  6. Assess pressure areas to avoid skin breakdown and use skin protective devices as needed
  1. Prone position
  1. The patient should lay on their abdomen (arms at sides or in “swimmer” position).
  2. If a patient is unable to tolerate, they may rotate to lateral decubitus or partially prop to the side (in between proning and lateral decubitus) using pillows or waffle cushioning as needed. Ideally the patient should be fully proned rather than on the side as there is currently no data about whether side positioning is beneficial.
  1. Time spent proning
  1. Patient should try proning every 4 hrs, and stay proned as long as tolerated. Proning is often limited by patient discomfort, but they should be encourage to reach achievable goals, like 1-2 hours (or as long as possible).
  2. Patient should attempt to prone at night as tolerated
  3. Our ideal goal is 16 hrs per 24 hours (e.g., 4 times for 4 hours each session) based on common interpretations of the PROSEVA trial (proning of intubated patients, Guerin et al, NEJM, 2013). However, we realize that few (if any) patients will tolerate 16 hrs of proning per 24 hrs.
  1. When to stop awake proning
  1. Patient may choose to stop self-proning at any time
  2. If intubation is within a patient’s goals of care, consider ceasing proning if the patient has an escalating oxygen requirement leading to concern for intubation (e.g. an escalating Venturi mask requirement in 40% - 60% range).
  3. Self-proning can be done while on Venturi mask or Non-Rebreathing Mask (NRB) with multi-disciplinary discussion for safety
  4. For patients who are DNI, it would be reasonable to continue Self-proning up to maximal levels of supplemental O2.

Initial Mechanical Ventilation

Checklist following intubation

  1. Set the initial ventilator settings:
  1. Initiate ARDS ventilation as described below
  2. Determine PEEP and mechanics as described below
  3. Assure adequate sedation as described below
  1. Obtain STAT portable CXR to confirm endotracheal tube location
  1. Prioritize CXR and vent settings over procedures (such as central venous catheter placement) if possible.
  1. Complete the “Mechanical Ventilation with Sedation” orderset in EPIC
  2. Obtain an ABG (preferred) or a VBG within 30 minutes
  1. Calculate P/F ratio from initial post-intubation ABG. Adjust oxygenation as described below
  2. Goal pH 7.25 to 7.45 adjust ventilation as described below

Initial ARDS Ventilation Settings

  1. Set mode to volume control (AC/VC)
  2. Set Initial tidal volume (Vt):
  1. Vt = 6 ml/kg (based on ideal body weight [IBW] from ARDSnet table, see table below)
  1. IBW men (kg) = 50 + 2.3 (height in inches – 60)
  2. IBW women (kg) = 45.5 + 2.3 (height in inches – 60)

  1. Set Initial respiratory rate
  1. Typical starting rates will be 16-24 titrated to goal minute ventilation of 5-8 L/min
  2. Consider starting rates of 24-28 titrated to goal minute ventilation of 8-12 L/min in setting of acidosis (pH < 7.25) pre-intubation
  1. Set an Initial PEEP based on BMI (empirically chosen targets):
  1. BMI < 40: PEEP 5
  2. BMI ≥ 40: PEEP 10
  1. Initial FiO2: 100% on intubation then rapidly wean to SpO2 92-96% (Barrot et al, N Engl J Med, 2020)

Determining PEEP and mechanics

  1. Titrate FiO2 and PEEP for oxygenation
  1. Initiate PEEP based on BMI, per above, and then titrate PEEP and FiO2 to target oxygenation SpO2 92-96% as per the following guidelines:
  1. BMI < 40: titrate PEEP and FiO2 as per the ARDSnet LOW PEEP table

  1. BMI ≥ 40: titrate PEEP and FiO2 as per the ARDSnet HIGH PEEP table

  1. If SpO2 < 92% or > 96% then titrate PEEP and FiO2 according to the ARDSnet table as per BMI
  2. Special consideration: anecdotal reports of COVID-19 patients describe a compliant, highly PEEP dependent phenotype in which PEEP management may not strictly adhere to specified ARDSnet tables (e.g., FiO2 0.4 - 0.5 but does not tolerate PEEP <10)
  3. Avoid elevated plateau pressures (with goal ≤ 30), particularly if using the higher PEEP table. Special cases (e.g., morbid obesity, burns) may need extra diagnostics, such as esophageal balloons, which we do not recommend for routine use given limited resources and infection risk.
  1. Obtain respiratory mechanics:
  1. Plateau pressure (with goal ≤ 30, management below)
  2. Static compliance

Sedation and Ventilator Synchrony

Pain, Agitation, Delirium model

  1. Pharmacologic therapy is chosen to target pain, agitation, and delirium in that order (both assessment and treatment).
  1. Please see the chart below for details, and the BWH Guidelines for Pain Agitation and Delirium in Mechanically Ventilated Patients for full detail (Partners login required)
  1. Strategies to minimize shortages and improve patient care:
  1. Boluses of benzodiazepines and opioids are preferred to continuous infusions.
  1. If continuous infusions are used, boluses should be administered prior to starting the infusion as well as when infusions are up-titrated. Bolus doses are typically 50-100% of the hourly infusion dose.
  1. Use the lowest dose that can achieve the desired effect
  2. Change IV medications to enteral medications if appropriate, especially as patients are weaning
  1. Typical regimen in a ventilated ARDS patient:
  1. Generally these patients require a period of continuous IV sedation and analgesia to establish ventilator synchrony. Specific regimens vary per the instructions below but the most common recommended agents are:
  1. Pain: Low-dose opioids are recommended as the initial analgesic agents, in bolus or enteric form whenever possible
  2. Agitation: Propofol is preferred for most patients, with adjunct agents (see below) to help dose-reduce
  1. Dexmedetomidine is generally reserved for patients approaching ventilation liberation; tachyphylaxis may occur with prolonged use
  2. Benzodiazepines can be used, but carry a high risk of delirium, so see further details below
  1. Delirium: If patients are agitated on SAT and CAM-ICU positive or receiving continuous sedation for >48hrs (with goal of weaning) we recommend an antipsychotic, generally quetiapine

PAD Bundle

Pain

Agitation

Delirium

Assess and Document

Able to self report → Numeric Rating Scale (0-10) (NRS)

Unable to Self Report → Critical Care Pain Observation Tool (0-8) (CPOT)

RASS (-5 to +4)

BIS in patients receiving Neuromuscular Blockade

CAM ICU-modified (+ or -)

Preferred Tools

Frequency

At least every 8 hours on all patients

If receiving intermittent or continuous analgesia or sedation, every 2 hours

At least every 8 hours on all patients

If receiving intermittent or continuous analgesia or sedation, every 2 hours

At least every 8 hours on all patients

Interpretation

Patient is in significant pain if:

Sedation/agitation depth defined according to RASS scale

Usual goal RASS is 0 to -1

Delirium present if: CAM-modified is positive

Achieve ventilator synchrony

  1. Start analgesia and sedation immediately:
  1. Prior to intubation, ensure analgesia/sedation infusion is available at bedside. The sedative boluses used during intubation will wear off long before the paralytic and the patient must be sedated while paralyzed (assume 60 minutes of sustained neuromuscular blockade for rocuronium and cisatracurium, and 10 minutes for succinylcholine).
  1. Assess patient synchrony with the ventilator:
  1. After neuromuscular blockade has worn off, assess synchrony (e.g., signs of breath-stacking, double triggering, other ventilator alarms)
  1. If synchronous, lighten sedation to the lowest level that maintains synchrony, ideally RASS score 0 to -1.
  2. If not synchronous:
  1. First adjust ventilator settings (including flow settings, trigger settings, and modest liberalization of settings within ARDSnet criteria [Vt 4-8cc/kg]).
  2. Then escalate sedation as needed to achieve synchrony
  1. If dyssynchronous despite deep sedation (RASS -4 to -5):
  1. First discuss additional ventilator changes with respiratory therapy:
  1. Options include tidal volume liberalization (within ARDSnet criteria) and mode adjustment (to PRVC or pressure control with careful Vt assessment)
  1. If this fails, consider neuromuscular blockade:
  1. Titrate neuromuscular blockade to ventilator synchrony
  1. Monitor degree of paralysis using “Train of Four” and assessment of ventilator synchrony
  1. Target sedation to RASS - 4 to -5 before dosing neuromuscular blocker.
  1. For intermittent vent dyssynchrony or to facilitate supination/pronation, bolus neuromuscular blockade may be given. In this setting, maintain sedation at levels that achieved RASS -4 to -5 until paralysis resolves (generally 20-60mins)
  1. Maintain Target BIS of 60 after initiation of continuous neuromuscular blockade.
  1. BIS monitoring is not reliable before neuromuscular blockade is initiated due to facial muscle activity
  2. BIS monitoring can be falsely low in the setting of edema or hypotension
  3. BIS monitoring may be falsely high in setting of ketamine administration (see below)

Avoid prolonged sedation

  1. We recommend a daily SAT This involves a large reduction in continuous sedation to achieve wakefulness (RASS 0), followed by re-titration to RASS / vent synchrony goals. unless contraindicated E.g. neuromuscular blockade, severe hemodynamic instability, patient near limits of mechanical ventilatory support
  2. Wean quickly, while monitoring for withdrawal:
  1. Patients who have been on continuous sedation for less than 7 days can be weaned rapidly with minimal concern for withdrawal
  2. Wean sedation and analgesics by at least 20% per day in patients that have been on sedation for greater than 7 days once patients have stabilized from a ventilatory and hemodynamic perspective
  1. Faster weaning is frequently possible due to drug accumulation in tissues
  2. Consult unit pharmacist if concerns for withdrawal
  1. Adjuncts including alpha-2 blockers, antipsychotics, and enteral agents can facilitate weaning (see below)
  1. Other non-pharmacologic strategies for delirium prevention and treatment can be found in the Delirium section.
  1. Try ventilator adjustments as described above to facilitate ventilator synchrony

Pain: Analgesic agents

  1. Goal: CPOT < 3 and NRS < 4 (RASS 0 to -1 may be utilized in addition)
  1. Use adjunctive therapies such as acetaminophen systemically and heat packs or ice / lidocaine patches / local anesthetic injections for focal sources of pain as appropriate. If the patient’s pain has a neuropathic component, consider adding gabapentin or pregabalin.
  2. Bolus doses of opioids should always be ordered prior to initiation of infusions, and should be available as needed at approximately 50-100% of the hourly infusion rate
  1. 1st line – Fentanyl or hydromorphone
  1. Fentanyl is most helpful for short term use or in severe renal dysfunction
  1. Use with caution in liver dysfunction, concern for serotonin syndrome, or obesity (lipophilic agent)
  2. Consider switching to hydromorphone in patients on longer term fentanyl infusions (over 2-3 days) and/or higher fentanyl infusion doses (roughly over 250 mcg/hr)
  1. Hydromorphone is preferred in ARDS, longer term use, liver dysfunction, obesity, ECMO
  1. Use with caution in severe renal dysfunction since the half-life is prolonged in these patients
  1. 2nd line - Morphine
  1. Use with caution in hemodynamic instability, as boluses do have venodilatory properties
  2. Avoid in severe renal dysfunction due to accumulation of toxic metabolites
  1. Enteral options:
  1. The use of enteral medications can optimize symptom management and help wean continuous infusions
  1. Generally, patients should be hemodynamically stable (minimal pressor use) and tolerating other enteral meds
  2. Systematically decrease continuous infusion opioid dose by at least 20% per day, after initiation of enteral agents
  1. Consult unit-based pharmacist for case review as warranted
  1. Opioid Analgesics
  1. Consider rotating to other medications within the same class (e.g. IV hydromorphone to enteral methadone or oxycodone) due to tachyphylaxis
  1. Typically, dose should be reduced by ~ 25% due to incomplete cross tolerance with conversion. Detailed instructions on cross-titration and conversion, including equivalency tables, is available in the opioid section of the palliative care chapter.
  1. Methadone
  1. May be useful in helping to wean IV opioids (opioid agonist / NMDA antagonist)
  2. Monitor QTc daily
  3. Starting dose 10 mg PO every 6 to 8 hours
  1. Titrate up total daily dose by 20 to 40 mg every 48 hours as needed
  2. Long half-life decreases, but does not eliminate, chance for withdrawal after prolonged use. If used >2 weeks taper dose by 25% every 3 days
  1. Atypical/Neuropathic analgesics (gabapentin, pregabalin)
  1. Strongly consider use in pts previously on these agents
  2. Low-moderate doses can sometimes decrease the need for analgesics; however, these can be sedating themselves, including respiratory depression.
  3. Dose-adjustment needed for renal disease

Agitation: Sedative agents

  1. Assessment and monitoring
  1. RASS goal
  1. Patient not on neuromuscular blockade: ideally 0 to -1, but may target -2 to -3 (or lower) if needed for ventilator synchrony
  2. Patient on neuromuscular blockade: sedation targets BIS 60 (RASS does not apply)
  1. 1st line continuous infusion sedatives – Propofol and/or Dexmedetomidine
  1. Propofol is preferred in renal and liver dysfunction, ARDS
  1. Use with caution in prolonged sedation, obesity, bradycardia
  2. Monitoring: Triglycerides, CK and Lipase (if triglycerides elevated > 400) every 24-48 hours
  1. If Triglycerides > 500, consider dose-reducing strategies prior to discontinuation (unless concern for pancreatitis)
  2. Discontinue if triglycerides > 1000 or signs of pancreatitis
  1. Dexmedetomidine is preferred for light sedation, nearing extubation, spontaneous modes of ventilation; it may have some analgesic effect.
  1. Dexmedetomidine can be used in combination with propofol for propofol dose reduction; bradycardia and hypotension may be limiting
  2. Use with caution in bradycardia, liver dysfunction
  3. Tachyphylaxis may be seen with prolonged use
  1. Benzodiazepines
  1. IV Benzodiazepines are second-line agents due to both deliriogenic properties and current shortages.
  1. Boluses are preferred in most cases for intermittent agitation or vent dyssynchrony.
  1. However, if a continuous infusion is needed, benzodiazepine boluses should be ordered prior to initiation, as well as PRN at 50-100% of infusion rate
  1. Use with caution in obesity, renal disease, and liver dysfunction
  1. PO benzodiazepines should be considered for weaning in patients previously on high dose benzodiazepines (PO or IV)
  1. Consider rotating to other medications within the same class (i.e. IV midazolam to PO lorazepam) due to tachyphylaxis. Dose reduce by ~ 25% due to incomplete cross tolerance with conversion
  2. Please see here for dose conversion calculator
  1. Phenobarbital
  1. Phenobarbital is a long-acting agonist at GABA receptors. It is most useful as a benzodiazepine-sparing agent, consider adding to help wean/interrupt benzodiazepines and other gabaergic sedatives
  1. It has a unique binding site different than benzodiazepines and works independent of endogenous GABA concentrations
  2. It is also an NMDA receptor antagonist, so may help with agitation control
  1. Dosing
  1. Consider loading dose of IV phenobarbital 4 to 8 mg/kg (ideal body weight) followed by 1-2 mg/kg twice daily
  2. Dose titrations of 1-2 mg/kg/day every 48 to 72 hours as needed
  1. Caution in patients with liver dysfunction as phenobarbital may accumulate
  2. May cause bradycardia and hypotension
  3. Half-life is about 70 hours
  1. Ketamine
  1. Ketamine has anesthetic and analgesic effects, and can reduce requirements for other sedating agents (propofol, opioids).
  1. Safety concerns and the risk of severe psychotomimetic (hallucinations, nightmares) effects argue against the routine use of ketamine in the ICU.
  1. Pharmacy should be involved for dosing by any non-Anesthesia team.
  2. Ketamine can cause a dissociative experience that may be mitigated with a bolus of midazolam before ketamine initiation.
  1. It has several physiological unique features that influence its use:
  1. It may increase respiratory rate and act as respiratory stimulant.
  2. It increases oral and respiratory secretions
  3. It has bronchodilatory properties
  4. It increases sympathetic tone (increases BP, HR, CO), which may be beneficial hemodynamically unless there is concern for ischemic heart disease or severe cardiac dysfunction. It is contraindicated in patients with significantly elevated PA pressure or ICP increases.
  5. It can cause a spurious increase in BIS levels which decreases the ability to monitor a patient’s level of sedation while receiving neuromuscular blockade
  1. Enteral sedating agents:
  1. These agents can be used as:
  1. alternatives to antipsychotics for sedation weaning and delirium in patients with prolonged QTc
  2. adjuncts in addition to antipsychotics in severe agitated delirium
  3. enteral conversions to facilitate weaning and discontinuation of IV infusions of agents in the same class
  1. Valproic Acid
  1. Starting dose 250mg q8 to q12 hours
  2. Caution in patients with liver dysfunction as valproic acid may accumulate. No need to monitor serum levels unless concerned for toxicity
  1. Phenobarbital
  1. IV:PO conversion is 1:1, side effects and half-life are same as IV
  2. No need to monitor serum levels
  3. Despite long half-life, occasional need for short taper when discontinued to avoid withdrawal
  1. Clonidine
  1. Most frequently useful for patients having improvement in agitation while on dexmedetomidine to facilitate dexmedetomidine weaning
  2. Wean / interrupt dexmedetomidine 1 to 2 hours after clonidine initiation
  3. Clonidine dosing (PO or patch can be used)
  1. Initiate at 0.1 to 0.2 mg every 6 to 8 hours
  2. Increase by 0.1 mg every two days (up to 0.3 mg every 6 hours) if patient still requiring dexmetetomidine

Delirium: Non-Pharmacologic and Antipsychotic agents

  1. Non-pharmacologic strategies for delirium prevention and treatment can be found in the “Delirium” section.
  2. Antipsychotic
  1. Antipsychotics should be strongly considered as adjunct sedatives for pts who are:
  1. Agitated and delirious (CAM-ICU positive)
  2. Receiving continuous sedation for >48hrs with agitation on SAT and goal of weaning
  1. Monitor QTc daily
  2. Agent selection
  1. Quetiapine may be preferred due to shorter duration of action
  1. Typical starting dose 25 to 50 mg every 6 to 12 hours
  2. Titrate up to effect by 50 mg – 100 mg/day. Max dose: 600-800 mg/day
  1. Olanzapine
  1. Typical starting dose 2.5 to 5 mg daily
  2. May be given IV push if concern PO absorption
  3. Max dose: 20 mg/day. Based on onset (6h), PO/SL olanzapine should not be used PRN for agitated delirium
  1. Haloperidol
  1. Enteral formulation not recommended
  2. May use IV for breakthrough agitation
  3. IV 1-5 mg q4h PRN agitation

Neuromuscular Blockade

  1. Safety and Monitoring
  1. Use to achieve ventilatory synchrony
  1. Neuromuscular blockade is used solely to achieve ventilator synchrony in patients requiring Assist Control or Mandatory ventilation to reduce lung injury after alternative approaches to achieving ventilator synchrony have been pursued (see Ventilator Synchrony above)
  1. Proning can be achieved in most (but not all) patients with deep sedation and without the need for neuromuscular blockade; in some patients bolus neuromuscular blockade to facilitate pronation/supination may be useful
  1. Always use analgesia and sedation:
  1. Neuromuscular blockers have no sedative or analgesic properties. Use of a BIS monitor and deep sedation with a continuous sedative infusion is required (RASS -4 or -5 prior to initiation of neuromuscular blockade and BIS goal 60 after initiation, see ventilator synchrony)
  2. Full recovery time (above) should be allowed and “Train of Four” should be 4/4 before sedation weaning
  1. Use the lowest effective dose for the shortest possible time:
  1. Prolonged neuromuscular blockade may contribute to weakness, prolonged weaning, and delayed recovery.
  1. If using “Train of Four” – goal ≥ 2 twitches and vent synchrony. Consider decreasing dose of neuromuscular blockade if consistently 0-1 twitch out of four and synchronous with vent. Considerable variability in TOF monitoring exists.
  1. Concomitant corticosteroids may increase risk of severe myopathy
  2. Trial of cessation of neuromuscular blockade should be performed after 48 hours unless the patient is near the limits of mechanical ventilatory support and should be reassessed daily thereafter
  1. Pharmacotherapy:
  1. Bolus dosing strategy:
  1. We suggest a trial of prn/intermittent boluses of neuromuscular blockers to facilitate pronation/supination or lung protective ventilation in patients with intermittent ventilator dyssynchrony.
  2. Boluses may be administered by ICU nurses in appropriately sedated patients (RASS -4 to -5).
  1. In the event of persistent ventilator dyssynchrony requiring >3 bolus doses in 2 hours, we suggest using a continuous infusion with re-evaluation every 24-48 hours.
  1. 1st line – Rocuronium.
  1. Preferred in renal dysfunction; its half-life can be modestly prolonged with significant hepatic impairment.
  1. Dosing - Intermittent Bolus: 0.6-1.2 mg/kg Infusion: 0-20 mcg/kg/min Start infusion at 3 to 5 mcg/kg/min
  1. Cisatracurium is in severely short supply – contact pharmacy if there is a patient with severe hepatic and renal dysfunction that might require it
  1. Dosing - Intermittent Bolus: 0.1-0.2 mg/kg Infusion: 0-5 mcg/kg/min
  1. If concerns for tachyphylaxis:
  1. Consider rotating to an alternative agent in this order of preference: Vecuronium (Bolus: 0.08-0.1 mg/kg Infusion: 0.8-1.7 mcg/kg/min Start infusion at 0.5 mcg/kg/min; Highly dependent on renal and hepatic elimination, duration may be prolonged with organ dysfunction), then Atracurium. Please see chart below for comparison of pharmacokinetics

Variable

Cisatracurium

Atracurium

Vecuronium

Rocuronium

Duration/Recovery (min)

80-180

20-40

30-60

20-30

Renal excretion (%)

Hoffman Elimination

Hoffman Elimination

50

20-30

Effect renal failure

No change

No change

Increased, especially metabolites

Minimal

Hepatic excretion (%)

Hoffman Elimination

Hoffman Elimination

35-50

< 75

Effect hepatic failure

No change

No change

Variable, mild

Moderate

Histamine release

No

Dose dependent

No

No

Ventilator Adjustments and Daily Management

General management of ventilated patients

  1. Consider whether patient requires daily CXR:
  1. CXR clearly indicated for:
  1. Clinical change
  2. Concern for displaced ET tube:
  1. Sudden increase in peak inspiratory pressure or resistance
  2. Decreased, unilateral breath sounds (usually on the right)
  3. RN or RT concern for change in depth of ET tube at teeth
  1. COVID-19 ICU Bundle:
  1. Ventilated patients should all have a daily ICU “Bundle” of best practices. See here for our COVID-19 ICU Bundle.
  1. Ventilator consults:
  1. If you need additional assistance managing ventilator choices, you can request a pulmonary phone/in-person consult (p11957)

Changing ventilation parameters (respiratory rate and tidal volume)

  1. Follow ARDSnet ventilation where possible:
  1. Starting tidal volume of 6 cc/kg (Tidal volumes should be 4-8 cc/kg using IBW (see table above) to excessive pressures and ventilator injury).
  1. Minute ventilation (respiratory rate x tidal volume) typically drives pH and PCO2:
  1. Titrate ventilatory parameters to pH and not PCO2.
  1. To achieve low tidal volumes will tolerate hypercapnia (functionally no limitation unless clinical sequelae) and acidemia (pH > 7.2).
  2. Because tidal volumes are low, the respiratory rate often has to be high to accommodate; typical RR is 20-35 breaths/minute.
  1. pH goal is normally 7.25-7.45:
  1. If pH > 7.45, decrease respiratory rate
  2. If pH 7.15-7.30, then increase respiratory rate until pH > 7.30, or PaCO2 < 25 (maximum RR= 35 breaths/minute)
  3. If pH < 7.15, then increase respiratory rate to 35 breaths/minute while monitoring for full exhalation between breaths
  4. If pH still < 7.15, then perform the following:
  1. Tidal volume may be increased by 1 mL/kg until pH > 7.15 (until plateau pressure reaches 30 cm H2O or tidal volume reaches 8 cc/kg)
  2. Deep sedation advancing to RASS -5 if needed
  3. If still no improvement, initiate prone ventilation (may improve V/Q matching and better ventilation)

Changing oxygenation parameters

  1. Minimize oxygen toxicity: PEEP and Fi02 drive oxygenation
  1. The goal is to deliver a partial pressure of oxygen to perfuse tissues (PaO2 ≥ 65, Sp02 ≥ 92%) Setting the lower oxygen limits: There is debate on the proper PaO2 goal, and our rationale relies on evidence for lack of benefit from conservative PaO2 goals in clinical trials (i.e., PaO2 > 55) and past association between lower PaO2 and cognitive impairment, although the evidence is certainly not definitive (mean PaO2 71 [IQR 67-80] for cognitively impaired survivors versus mean PaO2 86 [IQR, 70-98] in non-impaired survivors of ARDS (Mikkelsen et al, Am J Respir Crit Care Med, 2012). In the LOCO2 multi-center, randomized clinical trial, patients with ARDS were randomized to their PaO2 55-70, SpO2 88-92%; or PaO2 90-105, SpO2 >95%); the trial was stopped after enrollment of 205 patients due to futility and safety concerns (44% mortality in conservative oxygen group versus 30%; (Barrot et al, New Eng J Med, 2020). while limiting lung injury from high distending pressures (Ppl ≤ 30) and hyperoxia (FiO2 ≤ 60%, SpO2 ≤ 96%). Avoiding hyperoxia: Extensive mammalian animal data demonstrates that hyperoxic injury occurs at an FiO2 ≥ 75% (at sea level) with the rate of injury increasing as FiO2 exceeds that threshold. In multiple mammalian models, an FiO2 of 100% for 48 to 72 hours is associated with nearly 100% mortality rate. In these models, FiO2 < 0.75 appears to be a key threshold for injury. In these guidelines, we strive for an even lower goal, FiO2 < 0.60, but we wish to focus particular interest in the ARDS pathway when FiO2 >= 0.75 (i.e., increased sedation, paralysis, proning, inhaled vasodilator and ECMO consultation). For a review of hyperoxic acute lung injury, see Kallet and Matthay, Respir Care, 2013.
  1. Lower limit goals for PaO2 / SpO2 are widely debated; PaO2 > 55 and SpO2 >88% are also commonly used at BWH.
  1. PEEP Optimization:
  1. COVID-specific data:
  1. Preliminary anecdotal reports suggest a common phenotype of high compliance with PEEP-sensitive hypoxia. The pathophysiology of this phenotype has yet to be determined but it may reduce the efficacy of the ARDSNET PEEP tables to guide FiO2 and PEEP management.
  1. PEEP should be set and titrated as explained above using the ARDSNET PEEP tables to guide FiO2 and PEEP determination
  2. Optimal PEEP methods: significant efforts to determine a physiologically optimal PEEP are described in the literature but no specific method has demonstrated improved outcomes in large studies. In the setting of persistent hypoxemia or deviation from the ARDSNET PEEP tables, there are several methods employed at BWH for determining optimal PEEP.
  1. Best PEEP: BWH employs a “Best PEEP” protocol to optimize PEEP in selected patients in which the RT iterates changes in PEEP and compliance measurements to determine the physiologically optimal PEEP.
  1. However, to minimize demands on RT time, we will avoid routine use of “Best PEEP” protocol for COVID-19 ICUs.
  1. PV Tool: For patients on the Hamilton G5 ventilator, the Pressure-Volume (PV) tool may be used to determine the optimal PEEP as described below:
  1. Set the Pstart = 0 and Pmax = 40 and Pstop = prior PEEP
  2. Set the time step at 2 seconds with time hold at 0 seconds
  3. On completion of the maneuver, a PV loop is displayed demonstrating the inspiratory and expiratory limbs of the hysteretic loop
  1. The optimal PEEP is selected as slightly greater (1 to 2 cm H2O) above the lower inflection point (considered to reflect alveolar collapse and risk of “atelectrauma”)
  1. Note: test is optimally performed when patients are not making voluntary respiratory effort (eg deeply sedated or paralyzed).
  1. Esophageal balloon: Use of esophageal balloons to measure transpulmonary pressure will not be routinely performed on COVID patients due to infectious risk to staff.
  1. In other contexts, some patients in severe, fibrotic stage ARDS require very low PEEP (even <5 occasionally). Anecdotally, this very low compliance phenotype may be less common in COVID-19, but should not be missed (e.g., by tracking respiratory mechanics).
  1. Adjust FiO2:
  1. If using the ARSNET PEEP tables, titrate the FiO2 and PEEP to achieve the target SpO2 ≥ 92%
  2. If using Optimal PEEP methods above then adjust Fi02 after determining an optimal PEEP
  3. Goal FiO2 ≤ 60%; if FiO2 >60%; patient requires ventilator optimization. If you need assistance, pulmonary consultation is available (pager 11957)
  1. It is reasonable to put a desaturating patient temporarily on 100% FiO2, but remember to wean oxygen as rapidly as possible
  2. An FiO2 ≥ 75% is a crucial limit, based on animal studies of hyperoxia.
  1. Check plateau pressure:
  1. Check plateau pressure with every change in tidal volume, PEEP, or clinical deterioration (worsening oxygenation) but not as part of routine practice
  2. If plateau pressure is >30 cm H20, then decrease tidal volume by 1 mL/kg (minimum 4 mL/kg)
  3. If plateau pressure is < 25 H20 and tidal volume < 6 mL/kg, then increase tidal volume by 1 mL/kg until plateau pressure is > 25 cm H2O or tidal volume = 6 mL/kg
  4. If plateau pressure is < 30 cm H20 and patient is breath stacking or dyssynchronous, then increase tidal volume in mL/kg increments to 7 mL/kg or 8 mL/kg while plateau pressure is < 30 cm H20

Refractory Hypoxemia

  1. Refractory Hypoxemia pathway:
  1. If patient is hypoxic (PaO2 <75) despite PEEP optimization as above); and FiO2 >= 0.6 or PaO2 / FiO2 ratio < 150 then perform the following in this order:
  1. Optimize volume status by diuresing or RRT if possible; if no improvement then:
  2. Deep sedation, advancing to RASS -5 if needed; if no improvement then:
  3. Initiate continuous paralysis; if no improvement then:
  4. Initiate prone ventilation early: Discuss proning when PaO2/FiO2 < 150.

(We prone earlier than typical in non-COVID-19 ARDS)

  1. Prone within 12 hours of FiO2 > 75%.
  2. Strongly consider early in severe ARDS (<36 hrs from ARDS onset) if no improvement then:
  1. Initiate continuous inhaled epoprostenol (veletri) or inhaled nitric oxide (see “pulmonary vasodilators” below) if no improvement then:
  2. Consider ECMO consultation (see below) if, despite the above steps:
  1. Persistent PaO2 < 75 requiring FiO2 > 0.75
  2. Plateau pressure >30
  3. Refractory hypercapnia and pH < 7.2
  4. Absence of contra-indications (see ECMO section)

Prone Ventilation

  1. Prone early:
  1. We recommend early proning in severe ARDS and prefer to initiate proning prior to use of continuous paralytics (except to achieve patient-ventilator synchrony) or inhaled pulmonary vasodilators
  2. We initiate proning when a patient requires an FiO2 ≥ 60% to achieve an SpO2 ≥ 92% (or PaO2 ≥ 65) with a P:F ≤ 150 prior to vasodilator trial (Guérin et al, N Engl J Med, 2013).
  1. Eligibility criteria for proning:
  1. The only absolute contraindications are spinal cord injury, open chest, and unstable airway; BMI and patient size are not contra-indications
  2. For tracheostomy, we recommend that patients have their tracheostomy replaced by oral endotracheal intubation (ETT) while recognizing that some institutions prone patients with a tracheostomy. In the setting of COVID-19, decannulating a tracheostomy and placing an ETT is higher risk and the ICU team and anesthesiology should carefully discuss the risks.
  3. RRT can be performed while proned (e.g, typically via femoral vein catheter due to frequent neck turns while proned) but should be discussed with renal consultation prior to proning
  4. If the RN staff in the SP-ICU is unfamiliar with proning, the SP-ICU charge nurse should consult with the MICU charge nurse.
  1. Managing a proned patient:
  1. If you are needing assistance, page the Prone Team (pager #34433, available 24/7). This team consists of 3+ physical therapists who can help physically prone patients and provide expertise in positioning and wound prevention.
  2. Please see BWH MD MICU proning protocol
  1. Detailed instructions can be found in the BWH Nursing MICU proning protocol (Partners log-in required)
  2. This NEJM video provides brief instruction
  1. Prone ≥16 hrs per 24 hrs. Supine ≥ 4 hrs per 24 hrs.
  2. 1 hour post-initiation of prone ventilation:
  1. Adjust oxygen parameters: re-assess lung mechanics (plateau pressure and re-optimize PEEP, see above)
  2. Assess tidal volume and adjust ventilation parameters as in section 6
  1. Preferred tidal volume is 6 ml/kg (range 4-8)
  2. Goal plateau pressure ≤ 30
  1. If patient demonstrates improvement on proning then recommend:
  1. Discontinue neuromuscular blockade if initiated for patient-ventilator dyssynchrony and re-assess.
  2. Consider discontinuing further proning if patient meets these goals after supine for >4 hrs:
  1. PaO2 / FiO2 ratio > 150
  2. Ppl < 30
  3. pH > 7.25
  4. FiO2 < 60% to meet an SpO2 ≥ 92% (or PaO2 ≥ 65)
  1. If patients do not meet criteria for supine ventilation then recommend ongoing prone ventilation. There is no time limit for maintaining prone ventilation and it should be continued while beneficial.

Pulmonary Vasodilators

  1. There is no evidence of survival benefit of inhaled vasodilators in ARDS, and it can demand significant respiratory therapist resources (Fuller et al, Chest, 2015; Gebistorf et al, Cochrane Database Syst Rev, 2016; Afshari et al, Cochrane Database Syst Rev, 2017).
  2. Limited in vitro data notes that iNO at high doses inhibits replication of SARS-CoV, but this has not been studied in vivo (Akerstrom et al, J Virol, 2005; Gebistorf et al, Cochrane Database Syst Rev, 2016).
  3. Instructions for use:
  1. Inhaled epoprostenol
  1. Exclude contraindications: Alveolar hemorrhage (epo has mild antiplatelet effect), LV systolic or diastolic CHF (vasodilators cause ↑ pulm blood flow → ↑ LV filling pressure → ↑ pulmonary edema & ↓ PaO2 → consider CHF if pt gets worse after starting).
  2. Measure baseline ABG for PaO2
  3. Start continuous nebulization at 0.05 mcg/kg/min based on IBW (MDcalc online calculator).
  4. Do not change ventilator settings, sedation, paralysis, patient position or other care that could affect oxygenation.
  5. Re-check ABG 2 hrs after initiation of inhaled epoprostenol.
  6. If PaO2 increased by >10% from baseline, continue inhaled epoprostenol.
  7. If less than 10% improvement, consider inhaled nitric oxide (iNO); it is not required to attempt iNO.
  8. Avoid abrupt discontinuation if using for RV failure
  1. Inhaled nitric oxide (iNO) (at BWH, iNO is supplied by Respiratory Dept via INO Vent Device):
  1. Exclude contraindications: LV systolic or diastolic CHF (see epoprostenol above).
  2. Measure baseline ABG for PaO2
  3. Initiate iNO at 20ppm.
  4. Do not change ventilator settings, sedation, paralysis, patient position or other care that could affect oxygenation.
  5. Re-check ABG 2 hrs after initiation of iNO.
  6. If PaO2 increased by >10% from baseline, continue iNO.
  7. If PaO2 has not increased by >10%, increase iNO to 80ppm.
  8. Re-check ABG 2 hrs later; if PaO2 increased >10% from baseline, continue iNO at 80ppm. If PaO2 increased less than 10% from baseline, wean iNO to off.
  9. NO can cause methemoglobinemia (↑ risk if on nitroglycerin or nitroprusside). Monitor methemoglobin on ABG Q6 hr for the first 24 hr, Q12 hr for the second 24 hr, daily thereafter, or PRN if new clinical deterioration or SpO2-PaO2 dissociation.
  1. Weaning protocol for inhaled epoprostenol and iNO:
  1. Attempt to wean off after 24-48 hrs and daily afterwards.
  2. Wean inhaled epoprostenol by decreasing 0.01mcg/kg/min every hour. Wean inhaled nitric oxide to off by 20ppm per 30 mins or more slowly at 20ppm per 2 hrs if concerned for right heart failure or pulmonary hypertension. Monitor SpO2 and hemodynamics.
  3. Re-check ABG 2 hrs after weaned off. If PaO2 worsened by >10%, restarted inhaled epoprostenol or inhaled nitric oxide.

Airway pressure release ventilation (APRV) and alternative modes of mechanical ventilatory support

  1. There exists significant practice variation around the use of bilevel ventilatory modes which includes indications for use and role of deep sedation and paralysis and is highly dependent on expertise and usage at specific centers
  2. BWH current practice is to trial use of APRV or bilevel ventilation either as salvage therapy in patients with persistent hypoxemia not responsive to advanced conventional therapy who are also not ECMO candidates or in patients with persistent ventilator dyssynchrony that is impairing weaning and lightening of sedation.
  3. There may be a role for bilevel-type ventilatory support for patients with COVID who require high PEEP pressures but at present, our preference is to use bulk flow ventilation methods based on local experience and lack of definitive evidence of superiority of bilevel methods for general use

ECMO consultation

  1. BWH ECMO guidelines
  1. BWH ECMO consult pager is 35010.
  2. BWH ECMO guidelines are linked here (BWH log-in required).
  3. BWH participates in the New England ECMO consortium to discuss regionally ECMO availability and policies during COVID-19.
  4. The information below conveys general principles used by many medical centers and is NOT meant to reflect current BWH specific policies (linked above).
  1. Indications:
  1. Persistent PaO2 / FiO2 ratio < 75 mmHg despite optimized ARDS management (optimized PEEP, neuromuscular blockade, proning, inhaled vasodilator).
  2. Plateau pressure > 30 cm H2O on ARDSnet ventilation.
  3. pH < 7.2
  4. No potentially reversible causes (e.g., pulmonary edema, mucus plug, abdominal compartment syndrome)
  1. Contra-indications: Each patient is assessed on a case-by-case basis.

Absolute or relative contra-indications can include:

  1. Advanced age
  2. Active malignancy
  3. Severe shock; high cardiac output state
  4. Multi-system organ failure
  5. Prolonged ventilation or ARDS with poor chance of pulmonary recovery or severe chronic lung disease.
  6. Severe neurologic injury or intracranial hemorrhage
  7. Overall poor life expectancy (e.g., < 6 months); poor functional status at baseline; poor potential to recover functional status.
  8. Active hemorrhage or inability to anticoagulate
  9. Thrombocytopenia (plt < 50)
  10. Neutropenia (ANC < 500)
  11. BMI > 40 / total body weight > 180 kg

Ventilator Weaning and Extubation

  1. Clinical goal is to liberate patients from mechanical ventilation as soon as safe and feasible.
  1. Prolonged intubation is associated with ventilator-associated pneumonia (VAP) with median-time to VAP onset of 8 days in retrospective study of 191 COVID patients in Wuhan (Zhou et al, Lancet, 2020).
  1. All patients with improving or stable respiratory disease should be considered for weaning from sedation and mechanical ventilation when they meet the following criteria:
  1. Improving or stable respiratory disease
  2. FiO2 ≤ 50%, PEEP ≤ 10 for BMI ≤ 40 (or ≤ 16 for BMI > 40) with SpO2 ≥ 92%
  3. Hemodynamically stable (defined as HR < 120, MAP > 65, and vasopressor requirement of levophed gtt < 10 mcg/min)
  1. Assess patient readiness for weaning at least once daily
  1. A daily spontaneous awakening trial (SAT), consisting of temporary cessation of sedatives until a RASS of 0 is achieved, is be considered for all patients who meet the following criteria:
  1. Patients are in supine position
  2. Continuous paralytics discontinued for a minimum of 6 hours prior to SAT and has evidence of spontaneous motor activity and/or train of fours is 4/4 for neurostimulator test
  3. Hemodynamically stable (defined as HR < 120, MAP > 65, and vasopressor requirement of levophed gtt < 10 mcg/min)
  4. SpO2 ≥ 92% or PaO2 > 65 with an FiO2 ≤ 50% and PEEP ≤ 10 for for BMI ≤ 40 (or ≤ 16 for BMI > 40)
  1. A daily spontaneous breathing trial (SBT) is considered for all patients who meet the requirements for a daily SAT
  1. SBT consists of Pressure Support ventilation mode with a PS = 5 and PEEP = 5
  1. Consider higher PEEP for BMI > 40
  1. SBT discontinued if the patient develops
  1. Evidence of increased work of breathing with RR > 30
  2. Hypoxia (SpO2 < 92%)
  3. Hemodynamic instability
  4. Rapid shallow breathing index (RSBI) = RR/TV > 105
  1. Terminate all SBTs after 30 minutes and return to prior VC settings if patients are deemed not ready to extubate
  1. Extubation readiness:
  1. Extubation should be considered if patients meet the following criteria
  1. Breathing spontaneously
  2. RASS 0 to -1
  3. Able to follow commands
  4. Intact cough and able to protect airway
  5. Requiring airway suctioning for secretion < q2h
  1. Other considerations include:
  1. FiO2 < 40% at the time of extubation
  2. Optimization of volume status prior to extubation
  1. Weaning can fail in the setting of the following conditions (address appropriately) from Boles et al, ERJ, 2007:
  1. Respiratory factors:
  1. Ongoing pneumonia or pulmonary inflammation
  2. Bronchoconstriction
  3. Glottic and airway edema, sputum production, impaired cough
  1. Cardiac factors:
  1. Cardiac dysfunction or shock
  1. Neuromuscular factors
  1. Weakness and prolonged immobility
  2. Effects of steroids or neuromuscular blockade
  1. Neuropsychological factors
  1. Delirium
  2. Sedating medications
  1. Metabolic factors
  1. Malnutrition
  2. Electrolyte disturbances (hypophosphatemia, etc)
  1. Place NG or Dobhoff tube prior to extubation for patients intubated for >48h
  1. Patients should have an NG or Dobhoff tube placed prior to extubation given the frequency of swallowing issues post-extubation in these patients and delayed clearance for swallowing due to challenges obtaining video swallow / FEES (fiber-optic) in COVID patients
  1. NGT or Dobhoff placement is higher risk for clinicians on floor services (since the patient no longer on a closed ventilator circuit)
  2. Exceptions (e.g. in young patients who are A/Ox3) must be discussed by attending
  1. Prior to extubation, remove OGT replace with an NGT or Dobhoff
  1. Regular NG tubes have the advantage of being able to be put to suction and be used for bolused feeds.
  2. If the team anticipates long-term enteral access, strongly consider small bore feeding tube (more challenging to place given institutional practice for two-step chest x-ray or bronchoscopy to confirm placement). If Doboff is to be placed, strongly consider bridling.
  1. Place SLP consult at time of extubation

ICU Extubation

  1. Confirm patient meets criteria for extubation
  2. Don appropriate PPE
  3. Minimize staff: Only respiratory therapist and/or provider should be in the room
  4. Place patient on 1.0 FiO2 on the ventilator and ensure non-rebreather mask ready with flow “OFF”
  5. Place bed pad or towel on patient chest and ensure yankauer suction “ON” and readily available. Consider placing a plastic drape on top of patient to prevent exposure to any coughing that may occur.
  6. Secure NGT or Doboff feeding tube to nose.
  7. Suction mouth and loosen tape securing ETT to patient
  8. Turn all gas flows to “OFF” (may still have some O2 flow as safety mechanism for most machines) and extubate the patient
  9. Immediately discard of ETT, chuck or towel, and drape
  10. Immediately place non-rebreather on patient, then turn oxygen flow to 10-15L/min. Ensure patient is oxygenating and ventilating
  11. All providers will sanitize/change gloves while maintaining base layer PPE. Do not allow anyone into the room for at least 47 minutes after extubation to facilitate 99% of aerosolized virus removal by negative pressure room (assumes ACH of 6/hr)

Tracheostomy Management

  1. For tracheostomy procedure, see “Tracheostomy” section in surgery
  2. This section is in development

Alternative Ventilator Options

Use of Anesthesia Machine for Prolonged ICU Ventilation

  1. In the event of shortage of ICU ventilators, anesthesia machines may be used for prolonged ICU ventilation(ASA/ASPF Ventilator Guidance)
  2. A quick reference sheet and hotline to set up and monitor a repurposed anesthesia machine are provided(ASA/APSF Quick Setup Instructions)(1-800-224-1001)
  3. Draeger and GE have provided specific guidance for their anesthesia machines(GE Guidance)(Draeger Guidance)

Use of Single Ventilator Multiple Patients

  1. The ASA, SCCM, APSF, AARC, AACN, and CHEST societies have issued a joint consensus statement against using single ventilator for multiple patients (Joint Statement On Multiple Patients Single Ventilator). Splitting of ventilators comes with many risks, including infection transfer between patients, difficulty positioning of essential equipment, difficulty adjusting set respiratory parameters to meet individual patients’ needs and different clinical courses, difficulty controlling for sensed parameters (e.g. spontaneous respiration), alarm failures, measurement error in ventilator self-checks, and ethical dilemmas in prioritizing different patients’ treatment plans.