200 Locke PGR Co2 Covid ARDS
200.1 Summary
- How permissive, and for how long.
- Case
- PCO2?
- Another Patient
- What, if anything, should be done about this?
- First Principles: What causes high CO2?
- PaCO2 k VCO2 / VE (1-Vd/Vt)
- Dead space in ARDS
- Dead space in ARDS: Is COVID different?
- Deadspace trajectory through ARDS
200.2 Slide outline
200.2.1 Slide 1
- How permissive, and for how long.
- Brian Locke MD
- Fellow
- C
- 2 ### Slide 2
- Case
- 62 unvaccinated M was transferred from OSH after presenting with COVID-19 and DKA.
- He was initially treated with meropenem->CTX, azithromycin, and vancomycin. For COVID-19, he received Remdesevir and Dexamethasone
- 1 week into his hospitalization at the OSH, he was intubated and transferred to U of U
- After transfer, he had the following sequence of chest imaging studies ### Slide 3
- TODO: No text extracted from this slide. ### Slide 4
- TODO: No text extracted from this slide. ### Slide 5
- TODO: No text extracted from this slide. ### Slide 6
- TODO: No text extracted from this slide. ### Slide 7
- TODO: No text extracted from this slide. ### Slide 8
- TODO: No text extracted from this slide. ### Slide 9
- TODO: No text extracted from this slide. ### Slide 10
- PCO2? ### Slide 11
- PCO2?
- 2.2g CO2 1mol/44 g CO2 0.05 mol 50 mmol
- 12 oz 355 mL
- 50 mmol / 0.355 L 140 mmol/L
- Cola Line ### Slide 12
- TODO: No text extracted from this slide. ### Slide 13
- Another Patient ### Slide 14
- What, if anything, should be done about this?
- Review of CO2 kinetics
- Why does this happen? Dead space fraction in COVID-ARDS
- Tolerating Hypercapnia: Pros and Cons
- Review of control of ventilation related to hypercapnia and acidosis ### Slide 15
- First Principles: What causes high CO2?
- PaCO2 k VCO2 / VA
- PaCO2 k VCO2 / VE (1-Vd/Vt)
- If you hold VCO2 and Vd/Vt constant, you get the metabolic hyperbola (solid line) which relates PaCO2 to Ve ### Slide 16
- PaCO2 k VCO2 / VE (1-Vd/Vt)
- VCO2: doesn’t vary all that much in covid
- VE: we are often (mostly) in control in this in the situation of COVID ARDS. In health, the body regulates this incredibly tightly.
- Vd/Vt: is this uniquely bad in COVID-ARDS versus other ARDS?
- Vd/Vt (PaCO2 - PECO2) / PaCO2 ### Slide 17
- Dead space in ARDS
- MIGET studies (10-15cc/kg days): Vd/Vt ~0.60 in severe ARDS
- Estimate using PECO2 (~EtCO2) to PaCO2 ratio (requires blood gasses)
- VD/VT (PaCO2-PECO2)/PaCO2
- Ventilatory ratio (VR): [minute ventilation (ml/min) × PaCO2 (mm Hg)]/[predicted body weight × 100 (ml/min) × 37.5 (mm Hg)]
- VCO2 estimated based on body size.
- Index of impaired ventilatory efficiency that correlates with Vd/Vt ### Slide 18
- What, if anything, should be done about this?
- Review of CO2 kinetics
- Why does this happen? Dead space fraction in COVID-ARDS
- Tolerating Hypercapnia: Pros and Cons
- Review of control of ventilation related to hypercapnia and acidosis ### Slide 19
- Dead space in ARDS: Is COVID different?
- Beloncle et al 2021: n112 COVID-19 ARDS matched to n198 non-COVID pulmonary ARDS
- Matched on age, SAPS2 score, P:F, PEEP
- Found Crs was no different. VR was lower in COVID-19 at day 1 compared to non-COVID, but increased to similar degree by day 7.
- Luca Grieco et al 2021: n30 COVID-19 matched to n30 non-COVID ARDS
- Matched on P:F, FiO2, PEEP, tidal volume
- Ventilatory ratio (2.1 [1.7–2.3] vs. 1.6 [1.4–2.1], p 0.032) slightly higher in COVID-19 group
- Conclusion: mixed but underwhelming differences
- COVID-19 patients are often otherwise healthy, present with single organ failure
- Non-COVID ARDS matched for the degree is likely to either die or be cured (antibiotics). Bias due to Survivorship and illness duration?
- Possible COVID-ARDS evolves differently than other lung injury ### Slide 20
- Deadspace trajectory through ARDS
- https://twitter.com/danleisman/status/1468689855370702850?s11 ### Slide 21
- Why is a normal PaCO2 33ish mmHg (SLC)?
- Ventilation is metabolically expensive
- work of breathing 2% of O2 at rest, but increases hugely with exercise or pathology
- The same pH can be achieved at any PaCO2 by adjusting the bicarbonate: pH 6.1 + log [ HCO3 / (0.03 pCO2) ]
- The existence of ‘compensation’ suggests pH is the defended variable
- Why doesn’t the body operate with with a PaCO2 of 70? (and an HCO3 of 44 pH 7.42) This would be metabolically advantageous.
- Especially in states where the work (power) of breathing is higher - such as lung disease
- PaCO2 k VCO2 / VE (1-Vd/Vt)
- VE (1-Vd/Vt) k VCO2 / PaCO2 ### Slide 22
- Davenport Diagram: pH 6.1 + log [ HCO3 / (0.03 pCO2) ]
- Hypoventilation
- Renal Compensation
- Why not live here? ### Slide 23
- What actually happens in this process:
- 3-5 days to reach steady state metabolic compensation in dogs; unknown time in humans
- Acute: 0.1 mEq/L per 1 mmHg – modified some by pre-existing degree of HCO3. Not dependent on kidney
- Chronic: 1960s – 0.35 to 0.4 mEq/L per 1mmHg Co2 increase; 0.48 - 0.51 mEq/L per 1mmHg more recently; depends on renal function.
- Summarized in doi: 10.1053/ j.ajkd.2019.05.029 ### Slide 24
- How much CO2 is ‘sequestered’? From https://doi.org/10.1152/jappl.2000.88.1.257 ### Slide 25
- In a world without supplemental oxygen…
- PAO2 (Patm – PH2O) FiO2 – (PaCO2 / RQ) + FiO2 PaCO2 (1-RQ)/RQ
- (Patm – PH2O) FiO2 represents O2 arriving at the alveoli, which becomes saturated with H2O through the airways
- PaCO2 / RQ represents the volume of O2 in the alveoli removed into the blood
- FiO2 PaCO2 (1-RQ)/RQ represents the passive flow that must occur to maintain steady state when RQ / 1, because every O2 molecule leaving the alveoli is NOT replaced by 1 molecule of CO2.
- Hypoventilation causes hypoxemia;
- evolutionarily advantageous to have an oxygenation buffer?
- severe hypercapnia is only possible with supplemental oxygen (Dalton’s Law) ### Slide 26
- PAO2 (Patm – PH2O) FiO2 – (PaCO2 / RQ) + FiO2 PaCO2 (1-RQ)/RQ
- Normal A-a gradient? 4 + (0.25 Age)
- Critical PAO2 paO2 + (Normal A-a gradient) ### Slide 27
- When we control the alveolar gasses…
- At 100% FiO2, hypercapnia will almost never cause hypoxemia
- PAO2 (Patm – PH2O) FiO2 – (PaCO2 / RQ) + FiO2 PaCO2 (1-RQ)/RQ
- PaCO2 40: PAO2 (647 – 47) 1 – (40 / 0.8) + 1 40 (1-.8)/.8 556 mmHg
- PaCO2 120: PAO2 (647 – 47) 1 – (120 / 0.8) + 1 120 (1-.8)/.8 534 mmHg
- True effect might be different:
- Moves O2-Dissociation curve to the right (could cause hypoxia)
- Worsened V/Q matching (vasoconstriction in high VA zones) ### Slide 28
- What, if anything, should be done about this?
- Review of CO2 kinetics
- Why does this happen? Dead space fraction in COVID-ARDS
- Tolerating Hypercapnia: Pros and Cons
- Review of control of ventilation related to hypercapnia and acidosis ### Slide 29
- Argument for allowing hypercapnia
- Could you ventilate at <6 cc/kg if you allow PaCO2 to increase?
- k VCO2 / 40 mmHg vs k VCO2 /120 mmHg > 3x less VA; without the (probable) evolutionary reason we don’t do this (O2)
- Direct Evidence: (Permissive Hypercapnia)
- LTVV ARDS trials allow pH < 7.30 if RR 35+ or Pplat 25-30
- What is attributable to the PHC, and what is the LTVV? Is PHC beneficial outside these indications? What about hypercapnia without acidosis? ### Slide 30
- Is it the hypercapnia, or the severity of ARDS that leads to mortality?
- Nin et al, 2017: 1998-2010, 1899 patients with ARDS
- Maximum PaCO2 within 48h <30, 30-9, 40-9, 50-9, 60-9, 70+ mmHg
- Multivariable Logistic regression: age, SAPS2, P/F, PEEP, RR, Acidosis, vent stg
- Causal? Many other pertinent confounders (baseline CO2, lung disease, other management) ### Slide 31
- Is it the hypercapnia, or the severity of ARDS that leads to mortality?
- Muthu et al 2017: n415 with ARDS 2001-2016 at a single center
- Multivariable logistic regression: APACHE2, pH, severity of ARDS
- PaCO2 not independently associated with mortality (but pH is) ### Slide 32
- TODO: No text extracted from this slide. ### Slide 33
- CO2 ‘Narcosis’?
- Less lipid soluble than N2;
- CNS effects mediated by changes in intracellular pH and cerebral blood flow
- Decreased LOC occurs when CO2 rises above 90-120 mmHg (Nunn’s and Lambs)
- Likely contributes to encephalopathy in ICU patients
- Adaptation almost certainly occurs
- Healthy astronauts inhaling 8% CO2 (60 mmHg) experience > ### Slide 34
- What, if anything, should be done about this?
- Review of CO2 kinetics
- Why does this happen? Dead space fraction in COVID-ARDS
- Tolerating Hypercapnia: Pros and Cons
- Review of control of ventilation related to hypercapnia and acidosis ### Slide 35
- Sedation vacation:.
- Will this patient feel more, less, or the same degree of dyspnea compared to a hypothetical patient with normal acid/base status? ### Slide 36
- Ventilator Dyssynchrony
- Anxiolysis, Analgesia, and ‘Respirolysis’ are only weakly related when assessed by P 0.1
- https://journals.lww.com/ccmjournal/fulltext/2021/12000/managingpatientventilatordyssynchrony.18.aspx
- https://journals.lww.com/ccmjournal/fulltext/2021/12000/discordancebetweenrespiratorydriveandsedation.9.aspx ### Slide 37
- Haldane, Submarines, and Space Shuttles
- What happens if you keep increasing FiCO2?
- FiCO2 0 to 1.5%: VE increases, kidneys compensate, PaCO2 stable.
- FICO2 1.5 to 3-6%: Partial renal compensation, VE increases to defend CO2
- Bicarbonate excretion from kidneys 0, max renal effort (1963 - Operation Hideout).
- FiCO2 3-6%+: somehow, the respiratory control system realizes it would be too costly to defend a given CO2 and PaCO2 levels are allowed to rise
- This occurs before MVV
- ’Comroe’s Law’ ### Slide 38
- Drive to breath in 1 minute:
- 4 sources:
- Chemical drive – related to changes in arterial PO2 (NOT oxygen saturation), arterial and brain PCO2 and pH. Mediated by central and peripheral chemoreceptors.
- Metabolic drive – related to metabolic rate and mediated by unknown mechanisms (e.g. how the Ve increases in CPET before blood gases change)
- Wakefulness drive – disappears during sleep & anesthesia, a minor PaCO2 drop in presence of normoxia produces a central apnea
- Voluntary / Supratentorial factors (anxiety hyperventilation, anticipation) ### Slide 39
- Ventilatory Control
- Central Chemoreceptors
- Ventral surface of medulla
- Responds to H+ in ECF of CNS – this equilibrates with CO2 in blood (no charge) more than pH in blood. CSF HCO3 lessens response. No O2 sensing. Slower response
- Peripheral Chemoreceptors
- Carotid and Aortic bodies.
- Aortic body contributes to sympathetic and cardiovascular response
- Respond (rapidly) to PaO2, PaCO2, and pH
- Experiments injecting acids into CSF of animals leads to increase in VA
- When PaCO2 and pH differ? (DKA) Mediated by metabolic drive to breath and peripheral chemoreceptor. ### Slide 40
- Sensitivity of the controller system to CO2:
- The PCO2/Ventilation Response ‘Curve’: straight/dashed line.
- The amount that the respiratory system will respond to an increase in PaCO2
- slope change in Ve / change in PaCO2
- Steeper slope more sensitive, flatter less sensitive ### Slide 41
- Might HCO3 buffering.. reduce dyspnea?
- Brain curve (line): Ve that would occur if the respiratory system could do what the brain wants
- Ventilation curve (line): Actual Ve
- Dissociation of brain and ventilation curves -> dyspnea ### Slide 42
- Returning to our patient:
- Do you maximize VE within the parameters of ARMA to slowly correct? (RR 35, Plat 25-30)
- Do you give acetazolamide?
- Do you tolerate (or even encourage) hypercapnia to continue to minimize ventilator trauma and hope that eventually dead space improves? ### Slide 43
- The lung: an ultra-kidney for volatile acids.
- ‘CO2 Clearance’: Analagous to creatinine kinetics
- Urine volume : VE (exhaled volume)
- [Ucr] : FExhaled CO2
- Serum creatinine : PaCO2
- GFR : CO
- Implication:
- Further down on the ventilation axis, right on the PaCO2 axis, or R shifted the curve (increased VCO2 or Vd/Vt)
- Larger the change in CO2 that will occur from a given change in Ve
- Analogy: sCr of 6 is not much different than 5; PaCO2 70 not much different than 80
- Yet, impacts of CO2 on hypoxemia, CNS, CV system, and drive to breathe may be linear. ### Slide 44
- Does the PaCO2 ‘set-point’ change?
- For example, in OHS:
- Loop gain terminology primarily describes tendency toward stability of ventilation. ‘Gain’ of the feedback loop.
- Controller gain: ventilatory response to PO2 and PCO2 )
- Plant gain: changes in pulm capillary blood PO2 and PCO2 in response to changes in vent
- Mixing gain: effective circulation time – time which cap blood changes are sensed by controller
- Many changes are known to influence chemoreflex sensitivity
- Ventillatory long-term facilitation is the process that should keep things on track
- Ventilatory long-term facilitation is the term for adaptation to chronic stimulus (ie. sustained hypercapnia should increase loop gain to re-establish a baseline, and if this goes wrong you get a ‘wandering baseline’) ### Slide 45
- Acetazolamide?
- No effect in this RCT – OHS and COPD
- https://pubmed.ncbi.nlm.nih.gov/28286047/ ### Slide 46
- Post-script: meta-analysis
- https://link.springer.com/article/10.1007/s00134-022-06640-1
- Found conflicting results
- Positive results from protective ventilation
- Negative from effects on pulmonary vascular dysfxn ### Slide 47
- https://annalsofintensivecare.springeropen.com/articles/10.1186/s13613-022-01006-8
- COPD – ECCO2R
- Faster normalization of blood gasses (CO2, pH) -> resolution of dyspnea. ### Slide 48
- Editorial for https://link.springer.com/article/10.1007/s00134-022-06640-1
- https://doi.org/10.1007/s00134-022-06696-z
- They object to “induced” vs “permissive” hypercapnia done in the study
200.3 Learning objectives
- How permissive, and for how long.
- Case
- PCO2?
- Another Patient
- What, if anything, should be done about this?
200.4 Bottom line / summary
- How permissive, and for how long.
- Case
- PCO2?
- Another Patient
- What, if anything, should be done about this?
200.5 Approach
- TODO: Outline the initial assessment or decision point.
- TODO: Outline the next diagnostic or management step.
- TODO: Outline follow-up or escalation criteria.
200.6 Red flags / when to escalate
- TODO: List red flags that require urgent escalation.
200.7 Common pitfalls
- TODO: Capture common errors or missed steps.
200.8 References
TODO: Add landmark references or guideline citations.