inpatient / pulmonology and critical care

Acute Respiratory Distress Syndrome (ARDS)

PDF coming soon!

If You Remember Nothing Else

ARDS Berlin criteria include obilateral opacities within 1 week of insult, not due to cardiogenic pulmonary edema with P:F <300. The main way to help people with ARDS is to treat the underlying etiology while allowing the lungs to heal while keeping the patient alive and protecting their lungs; cross your fingers that severe fibrosis does not make the process irreversible. We do this by maintaining PaO2 55-80 or SpO2 88-94% with pH >7.2 to 7.25 via tidal volume 4-6cc/kg of predicted body weight, keeping Pplat <30, and Driving pressure <15. Keep the lungs dry and do not over oxygenate. Proning and early steroids work, pulmonary vasodilators and continuous paralysis probably have less impact on important clinical outcomes. Throughout the patient's care, consider whether they are a candidate for VV ECMO or lung transplantation.

Clinical Pearls

• ARDS is diffuse, immune-mediated lung injury causing pulmonary capillary and alveolar epithelial damage leading to increased vascular permeability, impaired gas exchange, and decreased lung compliance
• Exudation of neutrophils and protein-rich fluid leads to the formation of alveolar hyaline membranes which leads to impaired gas exchange and hypoxemia - damage to type I and type II pneumocytes which leads to a decrease in surfactant and alveolar collapse which causes intrapulmonary shunting; type II pneumocytes proliferate and lead to infiltration of fibroblasts causing interstitial fibrosis
• ARDS occurs in ~25% of intubated patients but only a subset (~2/3) have diffuse alveolar damage
• Most patients that improve will do so within 1-3 weeks with symptoms fully resolved, however mortality for ARDS is incredibly high (26%, 32%, 45% for mild, mod, and severe)
• The main way to help people with ARDS is to treat underlying etiology while allowing the lungs to heal while keeping the patient alive and protecting their lungs; cross your fingers that severe fibrosis from diffuse alveolar damage does not make the process irreversible
• For contextualizing P:F ratio, it's helpful to think what a perfectly healthy persons P:F is --> 100/0.21 since best PaO2 is ~100 and roon air is about 21% FiO2 - which comes out to a P:F of ~476
• Lung-protective ventilation = low tidal volume ventilation (LTVV) = low stretch protocol - the overall goal is to prevent ventilator-induced lung injury (VILI) or barotrauma 
• Permissive hypercapnia is the idea that we can allow CO2 to be higher (and thus pH to be lower) in service of pursuing lung-protective ventilation via low Vt - if pH is low, increase RR up to 35/min (but watch for auto-PEEP at higher RR); if issues with acidosis, look to treat underlying metabolic acidosis and consider IV bicarbonate
• Driving pressure is Pplat - PEEP and ideally should be less than 15
• Proning decreases VQ mismatch by decreasing compressive atelectasis from the heart and diaphragm and leads to more alveolar recruitment - PROSEVA Trial; goal to prone 16+ hours - if make supine and P:F >150 after 2 hours, can remain supine
• PEEP helps recruit alveoli, but also decreases preload and can lead to barotrauma
• Paralysis maximizes oxygenation by decreasing vent dyssynchrony and chest wall compliance, however, it does not show a survival benefit in routine, continuous use for P:F <150  - better to utilize if c/f dyssynchrony
• Pulmonary Vasodilators work by selectively dilating vessels that perfuse well-ventilated lung thus reducing V/Q mismatch; should see ~20% increase in PaO2; otherwise, stop therapy - inhaled nitric oxide does result in transient improvement in oxygenation but does not reduce mortality and may be harmful, especially for renal function
• Hyperoxia leads to increased mortality in ICU patients, so in general SpO2 <94% is okay with goal 94-98% - if FiO2 needs persistently above 0.6, should try to tweak PEEP
• VV ECMO - essentially replaces lungs - to treat hypoxemic resp failure; relies on the native ability to produce cardiac output - contraindicated if severe right or left HF, non-recoverable disease, arrest with CPR >30 mins without ROSC, active bleeding, contraindication to AC, recent NSGY procedure or brain bleed; relative contraindications are age >70, pHTN, GVHD, active malignancy, immunosuppression, DIC, patients ventilated >7 days
• Not clear if ECMO itself improves outcomes in ARDS or if just the referral to an ECMO-capable facility and receiving treatment there led to the better outcomes
• Pancreatitis can cause a systemic inflammatory response that leads to ARDS

Trials and Literature

• ARDS Review (NEJM, 2017)
• Low tidal volume ventilation decreases mortality (31.0% vs 39.8%) and increases vent free days vs 12 cc/kg with goal Pplat <50; (NEJM 2000);  showed NTT 10 for mortality 28 days in the hospital, 12 for mortality at the time of discharge with no clear harms - these results are mostly based on the NEJM stud7 (Cochrane Analysis)
• PROSEVA Trial - prolonged proning decreased 28d and 90d mortality in those with severe ARDS (16% vs 32.8% and 23.6% vs 41.0%); however in Cochrane Analysis, there was no mortality benefit to proning and proning led to pressure sores and tracheal tube obstruction - PROSEVA is only one of 9 trials as of 2013 that showed a positive outcome (NEJM, 2013)
• FACCT Trial - conservative vs liberal strategy of fluid management (diuresis and fluid restriction to achieve CVP<4 vs CVP <10-14) showed no change in death at 60 days (25.5% vs 28.4%) but did increase the number of ventilator free days and days not in the ICU and did not increase the incidence or prevalence of shock or need for dialysis (10% vs 14%) (NEJM, 2016)
• High (~13) vs Low (~8) PEEP did not improve rates of death (27.5% vs 24.9%) in ARDS (NEJM, 2013)
• ROSE Trial - early, continuous neuromuscular blockade with cisatracurium and deep sedation goal did not reduce death at 90 days in those with mod-severe ARDS (P:F <150) (42.5% vs 42.8%) but it reduces physical activity and leads to more CV events
• DEXA-ARDS - early dexamethasone in ARDS improved ventilator free days and mortality at 60 days (21% vs 36%); however, giving methylpred likely increased mortality if given for ARDS after they’ve already had the disease for 14 days (NEJM 2006); Cochrane Analysis from earlier studies in the 80’s suggests no benefit and NNH for infection was 3
• RECOVERY Trial - in those hospitalized with COVID, dexamethasone resulted in lower 28-day mortality if intubated or O2 needs (22.9% vs 25.7% overall, 29.3% vs 41.14% if intubated, 23.3% s 26.2% if just O2 needs) (NEJM, 2021)
• CESAR Trial - 2009 - severe ARDS - referral to ECMO center increased survival without disability at 6 month (63% vs 47%) - note only 75% of those referred actually got ECMO
• EOLIA Trial - severe ARDS ECMO within 7 days leads to more days without renal failure, decreases ischemic stroke, but didn’t change 60 day mortality (35% vs 46% but p=0.09 based on sample size); ECMO leads to more bleeding, thrombocytopenia
• Oxygen-ICU Randomized Trial - conservative (94-98%) vs conventional (97-100%) O2 goals in ICU patients resulted in lower ICU mortality  (11.6% vs 20.2%) - reason the bundle includes a task for avoiding hyperoxia
• Lower driving pressures associated with improved survival in ARDS - observational study

Other Resources

• MD Calc - Murray Score
• Internet Book of Critical Care - ARDS - has an amazing list at the bottom of the page listing of all of the seminal studies in ARDS that were summarized by The Bottom Line